' * ,o ■ r-. '..-'^^.-O ,^' ,. ' ■*^' ^^ * a , ^ * v"^^ o"^ - ^ c \' ^/": / \,# >^' -V. ; ^^- %^^ xV -/>, .,v\^ 'p \'^ ^ " Cr :r> .s<>. - >,*" ''^•^ %'' \^- ^^ ', 'i<". ''^^. '^"ste ^.-: ^ ^^' •-. ' ') .-, >. ^ • ■y -^ f x^^.. ■X'- \ V^^ x^^ '>\ , ^;^ ■■ i/: vV ■->.. '«.'^c .0 o ^•'%inj'^Sf ^0 0^ <- V .;^' ^-^^' STUDIES IN HISTORY ECONOMICS AND PUBLIC LAW EDITED BY THE FACULTY OF POLITICAL SCIENCE OF COLUMBIA UNIVERSITY VOLUME ELEVENTH COLUMBIA UNIVERSITY StXfa port 1898-1899 CONTENTS PAGE The Growth of Cities — Adna Ferrin Weber, Ph. D. . . . 495 ^ ■ ^^ 1. England and Wales 2. Scotland 3. Australia (7 colonies) 4. Belgium 5. Saxony 6. Netherlands 7. Turkey in Europe 8. China 9. Uruguay 10 Prussia 11. Germany 12. Argentina 13. United States 15. France 16. Denmark 17. Spain 18. Italy ig. Bavaria .20, Iceland "21. Canada 22. Chile 23. Norway 24. Switzerland 27. Austria 28. Hungary 29. Egypt 30. Ecuador 31. Venezuela 33. Roumania 34. Greece 35. Sweden 36. Central America 37. Japan 39. Mexico 41. Transvaal 43. Portugal 45. Bulgaria 47. Brazil 49. Russia 52. British India 56. Servia Diagram (based on Table CXII, pp. 143-4) showing the percentage of population dwelling in cities at the latest censuses. Legenb. I Percentage of total population dwelling in cities of 100,000 + . , ^ " " " " 20,000-100,000, 3 " *' 10,000-20,000. The entire length of the bars therefore represents the percentage of city dwellers in the total population of the countries named. Broken ends indicate lack of satisfactory statistics for exact measurements. STUDIES IN HISTORY, ECONOMICS AND PUBLIC LAW EDITED BY THE FACULTY OF POLITICAL SCIENCE OF COLUMBIA UNIVERSITY Volume XI THE GROWTH OF CITIES IN THE NINETEENTH CENTURY A STUDY IN STATISTICS BY > ADNA FERRTN WEBER, Ph.D. SOMETIME UNIVEKSITY FELLOW IN ECONOJnCS AND SOCIAL SCIENCE IN COLUMBIA UNIVEKSITY DEPUTY COMMISSIONEE OF LABOK STATISTICS OF NEW YORK STATB PUBLISHED FOR Columbia xantvetsitp BY THE MACMIliT.AN COMPANY, NEW YORK P. S. KING AND SON, LONDON 1899 .*^*^ v^ COPYRIGHT, 1899 By the Columbia Studies in History, Economics and Public Law " The proportion between the rural and town population of a country is an important fact in its interior economy and condition. It determines, in a great degree, its capacity for manufactures, the extent of its commerce and the amount of its wealth. The growth of cities commonly marks the progress of intelligence and the arts, measures the sum of social enjoyment, and always implies excessive mental activity, which is sometimes healthy and useful, sometimes distempered and pernicious. If these congregations of men diminish some of the comforts of life, they augment others; if they are less favorable to health than the country, they also provide better defense against disease and better means of cure. From causes both political and moral, they are less favorable to the multiplication of the species. In the eyes of the moralist, cities afford a wider field both for virtue and vice; and they are more prone to innovation, whether for good or evil. The love of civil liberty is, perhaps, both stronger and more constant m the country than the town; and if it is guarded in the cities by a keener vigilance and a more far-sighted jealousy, yet law, order and security are also, in them, more exposed to danger, from the greater facility with which intrigue and ambi- tion can there operate on ignorance and want. What ever may be the good or evil tendencies of populous cities, they are the result to which all countries that are at once fertile, free and intelligent, inevitably tend." — George Tucker, Progress of the United States in Population and Wealth in Fifty Years, p. 127. (iii) ^0 JACOB GOULD SCHURMAN SCHOLAR, ADMINISTRATOR, STATESMAN THIS ESSAY IS DEDICATED WITH THE ESTEEM AND GRATITUDE OFTHB AUTHOR. PREKACK. The present essay embraces the results of a statistical in- vestigation of the growth of cities during the nineteenth century which was originally undertaken for a doctor's dis- sertation. In preparing the essay for a wider circle of readers than the specialists to whom the doctor's thesis primarily appeals, the author realized the difficulty of reconciling the aims of a scientific treatise, wherein sharply-defined technical terms and simple, abstract statements lend conciseness to style, with the requisites of a popular work; and in explain- ing technical terms and illustrating the propositions laid down or deductions drawn, he has had to expand the essay beyond his original intention. While its value may thereby have been somewhat impaired for the specialist, the subject itself is so important to present-day students that it will lend interest to almost any attempt to present the facts with clearness and impartiality. The assistance rendered to the author by instructors, librarians and friends has been so generous that full acknowledgment cannot be given in this place. He is under especial obligation, however, to Dr. E. Blenck, director of the Royal Prussian Statistical Bureau in Berlin, through whose courtesy he was enabled to collect most of his statis- tical data in the unrivalled statistical library of the Bureau ; to Professors Johannes Conrad, of the University of Halle, Max Sering, of the University of Berlin, and John B. Clark, of Columbia University, for helpful suggestions ; and most of all to Professor Walter F. Willcox, of Cornell University, for stimulating criticism and sound advice during almost the entire period of the preparation of the essay. Acknowledg- ments are also due to Professors Edwin R. A. Seligman and Richmond Mayo-Smith, of Columbia University, who read the proof. All responsibility for errors, however, rests upon the writer alone. While he has exercised due care in the copying and compilation of statistics, he does not hope for absolute accuracy, and begs the indulgence of his readers in their judgment of arithmetical errors. Albany, May 15, 1899. (vii) ABBREVIATIONS AUg. St. Ar. — Allgemeines Statistisches Archiv. Bleicher — Beitrage zur Statistik der Stadt Frankfurt am Main, Neue Folge, Erstes Heft: Statistische Beschreibung der Stadt und ihrer Bevolkerung, II. Theil: Die innere Gliederung der BevSlkerung. Bearbeitet von dem Vorsteher des Statistischen Amtes, Dr. H. Bleicher, 1895. Conrad's Hdwbh. — Handworterbuch der Staatswissenschaften, edited by Conrad, Elster, Lexis and Loening. First edition, 6 vols. -|- 2 sup. vols., Jena, 1890—97. J. of St. Soc. — Journal of the Royal Statistical Society, London. Levasseur — La Population Frangaise, 3 vols. St. Mon, — Statistische Monatschrift (pub. by Austrian Statistical Bureau). Ilth Cen., Agr. — Eleventh Census of the United States (1890). Report on the Statistics of Agriculture, nth Cen., Mfs. — Eleventh Census of the United States (1890). Report on Manufacturing Industries. nth Cen., Pop. — Eleventh Census of the United States (1890). Report on Population, nth Cen., Stat, of Cities — Eleventh Census of the United States (1890). Report on the Social Statistics of Cities (by Dr. J. S. Billings).. For the full titles of the works cited under the following abbreviations, the reader is referred to the bibliographical note at the end of the volume : (p. 476) Harper. Kuczynski. Hassel, 1809, Legoyt. Hassel, 1822. Supan. Kolb. Worcester, (viii) CONTENTS PAGE Diagram Illustrative of the Concentration of Population — Front- ispiece. Preface vii Abbreviations viii CHAPTER I Introduction Comparison of the distribution of population in the United States in 1790 and in Australia in 1891 I Methods of studying the distribution of population 2 Density of population 4 Agglomeration, or community life : In antiquity 5 In the middle ages 6 In modern times 7 Statistical measurements of agglomeration : French, Italian and English methods 9 Supan's method lO Legal definitions of urban centre 12 Definitions of ofHcial statisticians 14 Final classification of dwelling-places 16 Comparability of urban statistics : Limitations arising from territorial extent 17 Limitations arising from annexation of new territory 18 CHAPTER II The History and Statistics of Urban Growth. I. The United States 20 II. The United Kingdom of Great Britain ana Ireland: 1. England and Wales 40 2. Scotland 57 3. Ireland 64 (ix) X CONTENTS PAGE III. France 67 IV. Germany 80 V. Austria 94 VI, Hungary 100 VII. Russia 105 VIII. Sweden 109 IX. Norway iii X. Denmark 112 XL The Netherlands 1 14 XII. Belgium 115 XIII. Switzerland 117 XIV. Italy • • 117 XV. Other European Countries : Spain 119 Portugal 120 Greece I20 Turkey in Europe 120 Bosnia and Herzogowina 121 Servia 121 Bulgaria 121 Roumania 122 XVI. Asiatic Countries : 1. Asiatic Turkey 122 2. Persia 123 3. British India 123 4. Philippine Islands 128 5. China , 129 6. Japan 129 XVII. American Conntries : 1. Canada 130 2. Mexico 132 3- Brazil 133 4. Argentina 134 5- Chile 13s 6. Other American Countries , . 135 XVIII. African Countries 137 XIX. Australasia 138 XX. Sumi7iary and Conclusions 142 CONTENTS XI PAGE CHAPTER III Causes of the Concentration of Population I. Introductory 155 The evolution of industrial society , 158 II. The divorce of men from the soil: Primitive agriculture the all-embracing industry 160 Agriculture and ancient cities 163 Agricultural progress in the eighteenth and nineteenth centuries . . 164 III. The growth of commercial centres: Authorities 169 The origin of towns 171 The location of towns 172 The differentiation of society into town and country 175 The town economy 176 The national economy .... 177 Commerce requires large cities 181 Biological analogy i S3 IV. The growth of industrial centres: Authorities 184 The evolution of industry 185 The factory system and the decay of villages 187 Connection of centralized industry with steam 192 The division of labor a permanent advantage of centralized industry . 195 Effect on the distribution of population 196 Relation of transportation methods to geographical specialization . . 197 Railway tariff policy in the United States and Europe 199 Recent decentralizing tendencies 202 Relative advantages of city and country for production 204 V. Secondary, or individual causes : 1. Economic 210 2. Political 213 3. Social 218 VI. Conclusions: Effect of the different industries upon the distribution of population . 223 The law of diminishing returns 225 Settlement of new lands 226 Changes in consumption 227 Forecast , 228 xi CONTENTS PACK CHAPTER IV Urban Growth and Internal Migration Migration cityward not of recent origin : References in the literature of four- teenth and subsequent centuries 230 London's immigrants in 1580 and 1880 compared 232 The rapid urban growth of the nineteenth century a result of the reduction of mortality 233 The course of evolution as shown in Swedish statistics 237 Modern European cities 239 Natural increase and immigration compared, for {a) Great cities, (^) Urban population 240 Critical review 246 Statistics of birth-place 247 The volume of internal migration 248 Migration in different countries compared 249 Interstate migration decreasing in the United States 25 1 Internal migration increasing in Continental Europe 252 Nature of the migratory movement : (i) Predominantly for short distances 255 (2) Cities the centres of attraction 255 (3) Magnitude of the city affects the average distance traversed by migrants , 259 (4) The larger the city, the greater its proportion of outsiders .... 260 American conditions 263 No progression through places of increasing magnitude 267 Mobility of population, {a) rural, {b) urban, (r) large cities 273 Character of migrants : (i) Sex 276 (2) Age . , • 280 (3) Length of residence 282 CHAPTER V THE STRUCTURE OF CITY POPULATIONS I. Sex — The cities contain a larger proportion of women than do the rural populations 285 Causes, immigration of females ? 289 emigration of males ? 290 ratio of the sexes at birth ? 294 heavy infant mortality ? 295 adult male mortality ? 296 CONTENTS xiii PACK II. Age — middle-age classes predominate in city populations 300 Immigration the cause 301 Results of age grouping 302 III. Race and Nationality: Foreigners not increasing in American cities 304 Negroes in American cities — Influx and mortality 310 IV. Occupation and Social Rank 314 CHAPTER VI The Natural Movement of Population in City and in Country I. Marriages: Higher marriage-rate in the cities 318 How explained? By age-distribution? By employment of young women ? . > 320 Conjugal condition ; cities contain the more unmarried 322 Contradictory statistics in Massachusetts 324 Result of late marriages 326 Influence of migration 328 Immigrants to cities marry early 328 Divorces : More numerous in cities 329 Reasons 329 II. Fecundity : Crude birth-rates 330 Refined birth-rates 332 Prussia, Massachusetts, Saxony, Denmark, etc t^t^t, Conflicting results 335 Average size of families 336 The theory of population 338 Occupation and social rank influence marriage and fecundity 341 III. Deaths: Crude and refined rates favor the rural districts 343 Relation of mortality to density of population 344 Mortality at various ages 345 Duration of life in city and country 346 Causes of the high urban mortality 348 Sanitary improvements needed 349 The question of model tenements 353 Other suggested remedies 354 Progress as indicated in reduced death-rates 355 xiv CONTENTS PAGE German and English experience * 356 Occupational mortality 359 Infant mortality 360 Influence of migration 365 CHAPTER VII The Physical and Moral Health of City and Country Are cities the cause of race deterioration ? 368 Hansen's plea for the peasantry and indictment of cities examined 370 (i) Do the city-born reside in the poorest quarters of the city, the country-born in the wealthiest ?"•..•• 37 1 (2) Do the city-born predominate in the unskilled occupations and lowest social ranks ? 373 (3) Do the cities contribute disproportionately to the ranks of paupers and degenerates? 383 (4) Would cities die out if the current of migration were to cease ? . . 386 (5) Is the intellectual aristocracy incapable of self-continuance ? . . . 387 The cities as the instruments of natural selection 388 Hansen's Bevolkerungsstrom 388 Rise of the immigrants in city industrial ranks 389 Physical infirmities 392 Stature 393 Girth of chest 393 Military efficiency 395 Conclusions as to physical vigor , 396 Moral conditions , 397 Intelligence 397 Religion and morality 399 Suicide 401 Crime 403 Conclusions as to urban morality 407 CHAPTER VIII General Effects of the Concentration of Population Economic conditions in city and country 410 Statistics of income 41 1 Wages of unskilled labor 41 1 Cost of living 412 City rents 413 Overcrowding 414 CONTENTS XV FAGB I. General economic effects of agglomeration 417 On the production of wealth 417 Agriculture 418 Distribution of wealth 419 Organization of labor » 419 Has the movement gone too far ? 420 Unemployment in cities 420 Scarcity of labor on the farm 422 Conclusion 424 II. Political effects 425 National wealth and power 425 National stability 426 Tenancy in the United States • . 426 Internal politics 427 The problem of municipal government 428 III. Social effects 43I 1. Extreme individualism in cities threatens social solidarity 432 Counter-tendencies 434 2. Rural and village decadence 437 Rural education 437 3. Cities are the centres of free thought and liberalism, of civilization and progress 439 Need of cities in the Southern States 440 4. Role of the cities in the process of natural selection 441 CHAPTER IX TENDENCIES AND REMEDIES Concentration and centralization of population all but universal 446 Obscured in England and the United States 446 Great cities in history 448 Growth of modern great cities 449 Limits to the growth of a metropolis 45^ Discussion of remedies in the past 454 Recent propositions 454 Fees for settlement 455 Agricultural improvements 455 Village attractions 45^ Administrative decentralization 45^ City improvements discontinued 457 Suburban development 45^ The process of " city-building " 459 Xvi CONTENTS PACK Transportation and the density of population 469 Suburban travel in America 470 Importance of rapid transit 471 Advantages of suburbs for manufacturing 473 The " rise of the suburbs," the hope of the future 474 Bibliographical Note 476 Index of Authors 479 Index of Subjects 483 CHAPTER I. INTRODUCTION. 1790. Popiilation of the United States 3,929,214 Population of cities of 10,000 and more 123,551 Proportion living in cities of 10,000 and more >3- 14 per cent. Population of the 7 colonies of Australia 3,809,895 Population of cities of 10,000 or more 1,264,283 Proportion living in cities of 10,000 or more 33.20 per cent. That the most remarkable social phenomenon of the present century is the concentration of population in cities is a common observation, to which point is given by the foregoing comparison of two typical countries of different centuries. The Australia of to-day has the population of the America of 1790; it is peopled by men of the same race ; it is liberal and progressive and practical ; it is a virgin country with undeveloped resources ; it is, to an equal extent, politically and socially independent of Euro- pean influence. But Australia is of the nineteenth, rather than of the eighteenth century ; and that is the vital fact which explains the striking difference in the distribution of population brought out by the introductory comparison. What is true of the Australia of 1891 is, in a greater or less degree, true of the other countries in the civilized world. The tendency towards concentration or agglomeration is all but universal in the Western world. What are the forces that have produced such a shifting of population? Are they enduring? What is to be the ultimate result? (O 2 THE GROWTH OF CITIES What are the economic, moral, political and social conse- quences of the re-distribution of population? What is to be the attitude of the publicist, the statesman, the teacher toward the movement? These are some of the questions to be answered, so far as may be, in the course of the present investigation. They are not questions capable of off-hand answers, for they are parts of a great problem. As Mackenzie says, " the growth of large cities constitutes perhaps the greatest of all the problems of modern civilization."^ It is the problem of dwindling dis- trict schools, of city labor disputes, of the tenement house, of municipal transit, of agrarian reforms, of the "destitute" country village, of the "submerged tenth" and the physical wastes of civilization, — in short, it touches or underlies most of the practical questions of the day. " The social problem that confronts practical people is in a very great degree the problem of the city." ^ It is, therefore, of prime importance to ascertain the extent of the movement and its probable direction in the future ; the forces that may be presumed to cause it ; the more immediate as well as the ultimate conse- quences; and the possible remedies. To a certain extent the distribution of the inhabitants of the earth is determined by man's physical environment.3 Nature's mandate it is that explains why the arctic have fewer inhabitants than the temperate zones, why mountain- ous regions are not so densely settled as valleys. To study the distribution of population, geographers and statisticians calculate the density of population, the number of inhabi- tants to the square mile or acre, and then compare variations in density with variations in climate, soil, earth formation, 1 Introduction to Social Philosophy, p. loi. ' Gilman, Socialism and the American Spirit, p. 30. *Cf. Ratzel, Antkropo-Geographie. INTRODUCTION 3 political institutions, etc., in order to ascertain the causes that determine the distribution. But the distribution of population is only partially ex- plained by natural causes. With the same physical environ- ment, the American people are differently distributed from the native Indians. The latter lived in tribes and congre- gated only in villages, because means of communication were too undeveloped and the population too small to permit the division of labor between city and country. Primitive peoples probably did not live in scattered dwellings, for man is a social being ; at the present time, at least, the lowest races like the Australians and Terra del Fuegians dwell in small family groups. The group can never be very large as long as it derives its sustenance from the land it occupies. With the growth of transport facilities and the development of trade, the community may obtain its food-supply from outside sources in exchange for its own products. Then arises a differentiation of dweUing-centres and their func- tions, which increases pari passu with the development of methods of communication, and very noticeably affects the customs and modes of life of the inhabitants. That the townsman is different from the countryman has long been recognized in politics, law and social science. The names "pagan" and "heathen" originally designated countrymen, while the abjective " urbane " and the nouns " citizen " and "politics" are derived from the Latin and Greek terms for city. In modern German '^ kleinstddtisch" is a term of re- proach, while in nearly all languages there exists a strong antithesis between "citified" and "countrified." As our study proceeds, we shall discover fundamental dif- ferences in the structure of city and rural populations, which underlie and explain the ordinary manifestations of disagree- ment just noted. But the first step must be the determina- tion of a method of measurement that can be used with some A THE GROWTH OF CITIES degree of refinement. In other words, when does a dweUing centre cease to be rural and become urban? One method of studying the spatial relations of men and communities to one another is by measuring the density of population ; the more human beings to the square mile, the closer together must be their habitations. The Hmitations upon the use of this method are, however, very considerable. Given two districts of equal population and territorial extent, there will be equal density; but in one case, the population may be scattered in small communities, and in the other congregated in a few large centres. In the latter case the average density will not be a true average ; for example, to say that the average number of persons to the square kilo- meter in the province of Brandenburg, including the city of Berlin, is 112, is to give a ratio that ' is true neither of the rural part of the province (70) nor of Berlin itself (26,456). And when it is said that in California there are 7.78 persons to the square mile, in New Hampshire 41.81,^ it does not follow that in California the people are scattered and in New Hampshire agglomerated. On the contrary, 41 per cent, of the Californians dwell in towns of 10,000 or more inhabitants, and in New Hampshire only 25 per cent.- Or compare Eng- ' Cf. Willcox, " Density and Distribution of Population in the United States at the Eleventh Census," Economic Studies of Amer. Econ. Ass'n., ii, 395. ^ Table CXV, infra. Objection may be made to the use of California in such comparisons, since its average density is not a true average. But no such objec- tion can be raised against Iowa, which, as Prof. Willcox says, {op. cii., 418), "is perhaps the most evenly settled State in the Union as measured by the mean variation of the densities of population of its counties from the State average." Vermont and Mississippi also have small percentages of variation, yet how differ- ent the distribution of population ! Table I. Per cent, of population Density. in cities of 10.000 +. Missouri 38.98 25.59 Vermont 36.39 7.93 Iowa 34.47 13.62 Mississippi 27.83 2.64 United States I7>99 27>59 INTR OD UCTION 5 land and Bengal. In density the two countries are as nearly- equal as can be (Table CXIII), but in Bengal 4.8 per cent, of the population is urban, and in England 61.7 per cent. (Tables XCI and CXII). The fact is, that in order to show the proximity of human habitations by density figures, the unit of territory must be too small for use in statistical studies. Only by means of maps and cartograms can the average density be made to portray the conditions of residence, considered with relation, not to the land, but to the people themselves. It is this fact, doubtless, that leads Professor Mayo-Smith in Statistics and Sociology to treat of density of population in the chapter on "Physical Environment;" of concentration of population in the chapter on " Social Environment." Density is far more dependent upon natural conditions than is agglomera- tion. But it must be admitted that the study of agglomeration by means of percentages of the total population dwelling in centres of a specified size offers some difficulties that are escaped when the comparison is limited to density. What, for instance, is the real significance of the terms " urban population," "rural population?" Does urban population include the dwellers in villages and small towns as well as those in cities? What is the line of division between urban and rural districts? In ancient times the distinction between urban and rural populations was easily drawn, because the term "urban" was suited only to the few large cities. There existed, indeed, smaller centres of population like our towns or villages, but these bore few or none of the characteristic marks of the city ; they were closely identified with the scattered popula- tion) which was devoted to agriculture. The development of the arts and sciences, the prosecution of industry, and politi- cal activity — all the social forces going to make up civiliza- 5 THE GROWTH OF CITIES tion — ^were phenomena of the great capitals like Memphis, Thebes, Nineveh, Babylon. In classic antiquity, indeed, the identification of city and civilization becomes complete ; the Greek republics were city states, and when Aristotle wishes to characterize man as a social or gregarious animal he says man is by nature a citizen of the city {■Kokiq^. The essential identification of the city with all the higher interests of humanity by the Greeks and Romans is to be observed at the present day in the English word "city" and "civiliza- tion," both of which are derived from the Latin *' civis." The tremendous influence of the classic city on the life of society has since been equaled by the mediaeval city re- publics in Italy (Venice, Florence, Genoa) and Germany (the Hansa towns and free imperial cities). Society then entered upon a new phase of development, and it is only with the prodigious growth of the great centres of population and industry in the last half of the present century that the city has come once more to have something like the dom- inating influence that it exercised in antiquity. The ancient city was a walled town and hence was easily distinguished from the surrounding rural districts. Similarly, in the middle ages the only places of collective residence were the enclosed towns which were absolutely cut off from the scattered population in the rest of the country. In such circumstances there could be but one distinction between city and country. This distinction, moreover, was recognized by the law, which by royal charter conferred certain privi- leges upon the towns as compared with the open country. The basis of the distinction was the pursuit of industry and commerce, /. e., the cities were manufacturing or market places. Hence it was that the difTerentiation of population into town and country came to signify a contrast between manufacturing industry and commerce on the one hand, and agriculture on the other, and this distinction was the one INTRODUCTION y made by scientific writers in Germany up to the most recent times. ^ In the eighteenth century the line between town and country was indeed a sharp one, and the opposition of their interests was clearly marked. The era of steam and machin- ery broke it down in England early in the present century, but on the Continent this influence has worked more slowly. An English writer in the middle of the century noted that the distinction was still sharp even in constitutional France and Belgium." Town and country manifested a spirit of hostility toward each other rather than a desire for friendly intercourse ; the cities maintained their walls, levied local taxes or duties {octroi) on goods brought in, and carried out a searching examination of every peasant's cart that was driven through the city's gate. The towns, with their special privileges, lived an isolated life and exerted little influence on the country population. Thus the movements following the February Revolution (1848) were confined to a few town populations like Hamburg, Berlin, Munich, Dresden, Frank- fort. The cities were the "oases of civilization;" "the people outside were in the same condition as they had been for ages." 3 But in the last half-century all the agencies of modern civilization have worked together to abolish this rural isola- tion ; the cities have torn down their fortifications, which separated them from the open country ; while the railways, the newspaper press, freedom of migration and settlement, ^Cf. Siissmilch, Die gottliche Ordnun^, Berlin, 1740; Wappaus, Allgemeine Bev'olkerungsstatistik, Leipzig, 1 861, vol. ii, who says that " it is rather the nature of the occupation and not the mere place of dwelling " that explains the difference in the vital statistics of city and country. * Laing's Observations on the State of the European People, 1848-9, p. 273. ' It is worthy of note that this mediasval separation of town and country still exists in Sweden and Norway, where all places of collective residence are still called " cities," though some of them have fewer than 500 inhabitants. Wagner- Supan, Bevolkerun^ dtr Erde, ix, 50-51. 3 • THE GROWTH OF CITIES etc., cause the spread of the ideas originating in the cities and Hft the people of the rural districts out of their state of mental stagnation. Industry is also carried on outside of the cities,^ so that the mediaeval distinction between town and country has lost its meaning in the advanced countries. In Hungary, which is a relatively backward country, there were in 1890 thirteen legal "cities" having less than 3,000 inhabitants each, while 38 other places that had more than 10,000 inhabitants each had not attained the dignity of "cities." The old distinction between town and country is still preserved in the Prussian statistics, and in the census of 1895, 192 places of from 5,000 to 50,000 or more inhabi- tants with an aggregate population of 1,800,000 were in- cluded in the legal rural population.^ In the light of such facts, the absurdity of holding to the mediaeval classification of dwelling-centres long since became patent to statisticians, and they have been seeking some other method. That there is a difference in the conditions of life of a city-dweller and a farmer is very evident, but on what basis are we going to separate the two? Hitherto each governmental statistical bureau has framed its own definition. The Russian government in one of its official documents,^ affirms that " in Russia the urban population forms 12.8 per cent of the total, as compared with 29 per cent, in the United States." But the fact is not there noted that, in the Russian estimate, towns of 2,000+ are rated as urban, while in the United States only places of at least 8,000 are called urban. On the 2,000 basis the comparison would be 37.7 and 12.8 per cent, on the 8,000 basis 29.2 and about 9 per cent. ^ See especially the Englisli censuses, and Lommatzsch, Die Bewegung des Bevolkerungsstandes im K'dnigreich Sachsen, 14 ff., and Losch,'Z'»V Entwicklung der Bevolkerung Wiirtembergs von iSyi-iSgo, in fViirtt. yahrbuchfiir Stat, und Landeskunde, 1894. * See infra, ch. ii, sec. 4. ' Th^ Industries of Russia (published for the Chicago Exposition), iii, 42. INTROD UCTION 9 One of the modern methods of distinguishing between dwelling-places is to divide the population into agglomerated and scattered. The agglomerated population includes all persons living in houses immediately contiguous to one another or separated only by parks, streets, etc., while the remainder of the population, generally speaking, is agricul- tural. But with the increasing density of population, ag- plomeration must naturally increase, and it becomes increas- ingly difficult to determine the distance which must separate houses in order to count their inhabitants in the " scattered " population. Italy and France have classified their popula- tions as agglomerated and scattered, with these results : France.^ Italy.* 1872 60.7 1871 74.3 1876 60.4 1881 72,7 1881 60.1 1886 61.0 1891 60.5 England expressed somewhat the same idea by giving the average distance between houses at various censuses, but has latterly abandoned the method. The mere fact of ag- glomeration, however, is probably less significant than some of the European statisticians would have us beheve, and it seems to attract less attention in the census bureau than it formerly did ; nor is it very important. One may well doubt if there exists any considerable difference between the rural population of France, parts of Germany and some other European countries, on the one hand, and the rural popula- tion of America on the other, that can be traced to the fact that in the one case peasants live in hamlets, and in the other case on their own farms; yet, in the former case, the per- ^ Hisultats statistiqites du denombrement de i8gi, p. 61. * Rauchberg, Art. Bevolkerungswesen, in Conrad's Handwbh., ii, 431, 10 THE GROWTH OF CITIES centage of agglomerated population would be much larger than that of the latter. It is thus evident that agglomeration alone does not furnish a true means of distinguishing urban population from rural population according to the meaning attached to those terms in common parlance. An urban population must indeed be agglomerated ; but it must also exceed a certain number of inhabitants or it will remain practically rural in character. A German Dorf or village containing, say, 300 peasants, is just as properly called rural as an American township con- taining 300 farmers living perhaps half a mile apart. Thus it is that modern statisticians have agreed upon a numerical boundary line between city and country. Many objections to this conclusion can of course be raised. It is rightly said to ignore the individuality of communities. Here is a sub- urban village in close contact with a manufacturing city and possessing most of the characteristics of city life ; if its population does not reach the arbitrarily chosen number selected as the qualification for entrance to the urban group, it must be put in the rural class along with the ver- iest farming community. Here again is the townsman with a country residence ; he is in the country but not of it. Nevertheless he must be counted in the rural population. The young manufacturing place, a harbor, a thousand-year old village, a residence town, all places of the same size must be thrown into one class if we follow numerical distinc- tions. Then again, as population becomes more dense one naturally expects to find the average size of the town or village increasing; the town of 5,000 now plays the role in the formation of social judgments formerly taken by the village of 1,000. Hence the geographical statistician Supan favors a sliding line of division, varying in different countries with the density of population. In his Ortsstatistik he has included all places with over 1,000 inhabitants in thinly INTRODUCTION I j settled countries like Australia, all over 2,000 in countries with a somewhat denser population (United States, Russia, Peru, Greece and the Balkan countries), and in thickly- populated countries Hke most of the older European states only such places as have 5,000 or more inhabitants.* These objections to the numerical line between city and country, however, do not outweigh its advantages.^ For statistical purposes no other distinction is so available ; hence this distinction has been sanctioned both in theory and in practice. But no such agreement has been reached as regards the determination of the numerical boundary. It is not alto- gether easy to define the distinguishing characteristics of a city, but in a general way the student will observe that, when a community attains a certain size, new needs and purposes manifest themselves. The close association of a large body of people alters even the material conditions of life. The artesian well and cistern must give way to a common water supply brought from distant springs ; a sewerage system must be introduced, likewise street light- ing, and rapid transit between the home and the workshop. The liberty of the individual to do his own sweet pleasure must be curtailed for the common benefit ; the streets may not be used as depositories of materials for new buildings ; noises must be abated, such as music practice with open ' Petermann's Mideilungett, Erganzungsheft, No. 107 : Die Bevolkerung der Erde, ix, Wagner und Supan. * Some statisticians still identify rural and agricultural populations. The States- man's Year Book (1897, p. 678), for example, takes such a position in the follow- ing: "In Northern Italy the population is scattered over the country, and there are few centres. In Southern Italy, and in the islands, the country people live in the towns, coming and going to cultivate their own plots of land; consequently there are many populous centres where, if numbers alone were considered, the population would be regarded as urban, though it is in truth almost exclusively rural." 12 THE GROWTH OF CITIES windows during the sleeping hours; nuisances are pro- hibited, and the like. In response to the new wants of the community, the framework of the local government under- goes alteration, the law itself recognizing the difference be- tween urban and rural populations. Not only are new wants to be satisfied,, but they must be satisfied by new methods. In the country village, where every citizen knows every other citizen, the town meeting, or primary assembly in its pure form, is the ideal governing body, but with every in- crease in the size of the town, representation must be given fuller play. Officials are multiplied by the score and hun- dred, and must be appointed rather than elected, since the voters are unable to inform themselves concerning the merits of so many candidates. In the United States, then, the law usually provides for three forms of local government : ( i ) for the township, the primary civil division, (2) for the village, the smallest agglomeration, whose charter of incorporation is granted by the administration in accordance with general laws, (3) the city, whose charter of incorporation is usually a special act of legislation. But as regards the requirements for these grades there is no uniformity of practice. In some of the Western States almost any town may aspire to the dignity of city. Thus North Dakota contained four places which in 1890 severally had a larger population than 2,000 and all were "cities," namely, Bismarck, 2,186; Fargo, 5,664; Grand Forks, 4,979; Jamestown, 2,296.^ Kansas has no villages or towns ; every place sufficiently large to be in- corporated is dubbed "a city." Kingman county has three "cities" : Norwich with a population in 1890 of 301 ; Spivey 205, and Kingman 2,390." The climax of absur- ^ Compendium of the iiih Census, vol. i, population by minor civil divisions. Pembina " city " with its three wards had a population of 670. * Op. ciL, i, 167. INTR OD UCTION 13 dity is reached in the case of the "city" of Mullinville, Kiowa county, with 79 inhabitants ! It is greatly to be re- gretted that the word " city " is thus degraded, for it is thus coming to lose its peculiar significance. In the East, how- ever, the meaning has been preserved and no one calls a place a "city" unless it possesses claims to the superior in- fluence that accompanies a large population. In Massa- chusetts, all communities and areas are under the town (township) government until they attain a population of 12,000, when they are incorporated as cities; that is to say, the town meeting is then replaced with a repre- sentative legislature and executive. In New York, there is no rigid limit, but in practice a population of about 10,000 is required for incorporation as a city. There arc now six cities in New York that had a popula- tion of 8,000-10,000 in 1892. But taking the United States as a whole, it must be admitted that the legal city is much less populous than the statistical unit (8,000), for there are 1,623 incorporated cities in the United States, and only 448 towns exceeding 8,000 population,'' In England, the municipality corresponding to our city is the borough; the term "city," when used at all, referring by tradition to a borough that happens to be the residence of a bishop. But for the exercise of the important and numer- ous functions connected with sanitation the country is divided into urban and rural sanitary districts. When the local government board finds an unusually high death-rate prevailing in a country district, it may by order create an urban sanitary district, which is endowed with greater powers in regard to the preservation of health. And it is the urban sanitary districts that contain the urban population of Eng- land. There are few districts with a smaller population than ^ Gannett, The Building of a Nation, 32. 14 THE GROWTH OF CITIES 3,000, but a large number under 10,000 and even under 5,000/ In France, all communes are governed in accordance with the provisions of one municipal code, excepting only the cities of Paris and Lyons. The law thus makes no distinc- tion between urban and rural populations. In Germany and Austria-Hungary, as already observed, the law regarding municipal corporations has not developed in recent years and largely preserves mediaeval distinctions, with the result of producing anomalies. The conclusion is forced upon us that the legal definition of urban population lacks uniformity. The law-givers do indeed recognize a vital distinction between urban and rural populations, but they do not help us to draw the line. What, now, is the opinion of statistical scientists on the question? The disagreement is not so great here as in the preceding cases, for of late years the numerical boundary (2,000) chosen by France in 1846 has made its way through most of Continental Europe, and its adoption in 1887 by the In- ternational Institute of Statistics '^ makes it reasonably cer- tain that it will be the generally accepted line of division for many years to come. If we hold in mind the distinction commonly made in America between city and town, we shall see that the difference in the meaning attached to the word "urban" by European and American statisticians rests simply on this fact: that in America the "town" 3 is re- garded as rural and in Europe as urban. The question of ' See further ch. ii, sec. 2, Introduction. ' See Bulletin de Vlnstitut International de Statistique (1887), ii, 366. ' In American usage generally, the town is something between village and city, a kind of inferior or incomplete city. The thing which the town lacks, as compared with the complete city, is ... . municipal government." Fiske, Civil Government in the United States, p. 103. But Fiske says the town is urban rather than rural. INTR OD UCTION 15 classification is always more or less of an empty one, but in determining the position of a town it may be well to note that the authority of one of the most eminent statisticians of this century is practically on the American side. Gustav Riimelin in discussing the Landstadt, which, like the Ameri- can town, occupies a middle position between the country and the great city, says that as a rule it has more of the characteristics of the Z>^r/" (hamlet or village) than of the Grossstadt (large city)/ And on consideration one must incline to the view that the peculiar marks of a city as described in a foregoing paragraph do not pertain to the village or town. The American legal practice of making broad the distinction be- tween village and city, rather than that between village and rural district, is the sound one. In England, too, the limit 10,000 is important; boroughs containing not less than 10,000 inhabitants may themselves regulate matters of local concern which in other cases are attended to by the county council.^ While, then, a population of 10,000, will in the comparative tables be accepted as the minimum limit of an urban agglomeration, in studying the several countries it will be convenient to follow the ofificial definition. Germany, in particular, does not recognize the line of 10,000, but divides the dwelling-centres into these four groups :3 Landstadte 2,000-5,000 inhabitants Kleinstadte 5,000-20,000 " Mittelstadte . 20,000-100,000 " Grossstadte more than 100,000 " ' Stadt und Land, in Reden und Aufsdtze, i, 352 : " Die kleine Landstadt liegt in der Kegel von der Grossstadt noch viel weiter als vom Dorf .... Die kleinen Stadte sind die Vermittlungskanale fiir den Wechselverkehr von Stadt und Land," etc. As will appear later on, the age distribution in the town differs considerably from that of the city; this fact, almost entirely a result of emigration or immigra- tion, is of vast importance in determining the social character of a community. ' Goodnow, Comparative Administrative Law, i, 244. ' For the basis of these distinctions, see Statistik des Deutscken Reichs, Neue Folge, Bd. 32, p. 29.* 1 6 THE GROWTH OF CITIES In order to avail ourselves of the best statistics it will therefore be necessary at times to regard as urban popula- tion all the inhabitants of places of 2,000 or more. As the foregoing table indicates, Germany divides the urban population into classes. The practice is a common one among official statisticians, but they seldom agree on the lines of division. In most cases 10,000 appears as one limit; France and Austria add 50,000, etc. But in nearly every instance, a separate class is made of cities that exceed 100,000 souls in population. Such cities are rightly called great-cities {Grossstcidte, les grandes villes). They differ from smaller cities in that their influence extends not merely beyond their county, but beyond the commonwealth or province, becoming national or even international, Hence the grande ville has become recognized not only in the official statistics, and in the writings of savants, but also in the legislation of many modern states. The Institut Inter- national de Statisque made it the only sub-class in the urban portion of the population.^ The discussion on classification of population according to the size of dwelling-place may be summarized in the following manner : f Scattered. Rural Population. \ ^^ j I . Hamlets and villages (less than 2,000 pop.) . L 2. Towns (from 2,000 to 10,000 pop.). Urban Population. | 3- Cities (more than 10,000 pop.). I a. Great cities (more than 100,000 pop.). One point still calls for notice regarding the comparability of urban statistics, namely, the area that constitutes the urban unit. In American usage, outside of New England, it is the incorporated village or city within the township;' * Bulletin, ii, 366. ^ " Township " is the common American name of the primary political division. But there are numerous variations, e. g., " parish " in Louisiana, " precinct " in INTRODUCTION 17 but in New England, where there are no other incorporated communities than the city, the township is the statistical unit. Obviously, this variation in practice destroys the comparability of urban statistics, since a township is large enough to contain several villages and a large number of scattered dwellings besides. The census returns should designate communities within the township. The percentage of error in such cases will of course vary directly with the extent of territory and inversely with the number of inhabitants. Given a large township, there still may not be a single community or dwelling-centre of 2,000 population, although the entire township may contain 10,000 people. This is indeed an extreme case, though it can doubtless be parallelled in the statistics of Spain, where the dwelling centres are never separately returned. In Italy, where both the agglomerated population, or inhabitants of communities, and the total communal or township popula- tion are returned, important differences are found. Brescia township contained, in 1881, 60,630 inhabitants; Brescia, Texas, " district " in Virginia, etc. In New England the term " town " is un- fortunately substituted for township. Historically, this latter usage is incorrect, as well as confusing. In primitive Anglo-Saxon times, " township " (tunscip) was, without much doubt, regularly used to designate the municipality in its en- tirety (i. e.f the whole area within the hedge or walls) ; and " town " (tun) meant the settled portion (Cf. W. F. Allen, "Town, Township and Tithing" in Essays and Monographs). In the course of time, however, the term "town" usurped the name " township," being applied to the whole area : and in this sig- nificance was brought to America by the Pilgrims. Consequently, in New Eng- land, the word means the entire municipality, while the settled portion is variously designated as the " village," the " middle of the town," etc. The use of the word " town " for " township," is, moreover, productive of confusion, because " town " is also frequently used to designate a dwelling-centre of a size intermediate be- tween the village and the city. On the other hand, the English use it constantly in the place of the word " city;" with them London is always a "town." But the American practice is far more logical and convenient and conforms more closely with historic usage; for " city " has always signified a town of high rank and dignity, as appears in the classic phrase, " free and imperial city." J 8 THE GROWTH OF CITIES the community, 43,354. Chiari township, 10,414; Chiari, the town, 6,000.' The average area of the Italian township is 35.3 square kilometers, as compared with 54 for the Spanish township. In Servia, where the township is also large (38.2 sq. km.), the population in townships of more than 2,000 inhabitants constituted, in 1890, 36.5 per cent, of the population ; but the population in cities, towns and villages of 2,000+, only 13.25 per cent. In Germany where the Gemeinde (township) is very small,=^ the respective figures in 1890 were 47 and 42.5 per cent. France has a method of her own ; holding to the commune or township as the unit, she counts as urban communes only such as pos- sess an agglomerated population exceeding 2,000. Theo- retically, this is indefensible because it adds villages and isolated farmers to the urban population ; but it is better than the German or New England method of counting as urban those townships {Gemeinden) whose total population — without regard to the fact of agglomeration — exceeds the boundary limit adopted. This objection, however, has much less weight when the minimum limit of an urban commune is placed at 10,000, for against so large a number, the scattered population or even a hamlet or two counts rela- tively little. Finally, one has to consider whether the population of a single city, or of a number of cities, shall be employed as it is returned at each census, or whether the population dwell- ing on a given territory, thus diffusing over a long period the increase that actually comes by annexations. And in order to ascertain the growth of an urban population, shall we use the contemporaneous figures of each census, or settle upon a fixed number of cities for the whole series of cen- suses, thus discounting the effect of new cities arising from ' Supan, 58. 'Only 7 square kilometers; cf. table in ch. ii, sec 20, infra. INTR OD UC TION 19 the villages and towns? As will be pointed out in the course of the investigation, both methods have their advant- ages. For studying the gradual growth of urban population it is usually better to take a fixed territory and a definite number of cities and avoid accidental additions.^ But to compare the distribution of population at long intervals such refinement would be both superfluous and incorrect. The relative growth of different cities is one thing ; the concen- tration of population is another.^ As the first condition of an analytical study is, of course, an exact knowledge of facts, our plan will be to sketch the movement toward the concentration of population during the present century in the leading industrial states of the world and correlate the stages in this movement with in- dustrial changes ; then to present the statistics of other countries in less detail, and finally to bring the results together in comparative tables. Not till then may general causes be intelligently discussed. After that, some consid- eration will be given to the structure of city populations (their peculiarities as distinguished from the rural popula- tions), and the consequences of the movement from country to city. In conclusion, remedies will be discussed and some attempt made at forecasting the distribution of population in the near future. ' The statistical difficulties raised by the adventitious extension of the munici- pality or the political city have been pointed out anew by Prof. E. J, James in a brief article upon " The Growth of Great Cities " in the Annals of the American Academy of Political and Social Science (Jan., 1899), xiii, I seq. ' On the question of method it will suffice to refer the reader to the bibliog- raphies appended to Sections 24-27 of G. von Mayr's BevolkerungsstatisHk (^Statistik und Geselhchaftslehre, vol. ii; Freiburg i B. 1897), ^^ addition to the references already given. CHAPTER II. HISTORY AND STATISTICS OF URBAN GROWTH. I. THE UNITED STATES. In a new country the rapid growth of cities is both natural and necessary, for no efficient industrial organization of a new settlement is possible without industrial centres to carry on the necessary work of assembling and distributing goods. A Mississippi Valley empire rising suddenly into being without its Chicago and its smaller centres of distribution is almost inconceivable to the nineteenth century economist. That America is the " land of mushroom cities" is therefore not at all surprising. But, on the other hand, it is astonishing that the develop- ment of the cities in a new country should outstrip that of the rural districts which they serve. The natural presump- tion would be that so long as land remains open to settle- ment, the superfluous population of the older States or of Europe would seek the fundamental, or food-producing, industry of agriculture, and build up cities only in a corres- ponding degree. Yet in the great cereal regions of the West, the cities have grown entirely out of proportion to the rural parts, resulting there, as in the East and in Europe, in an increasing concentration of the population. The only States ' where the urban population has in recent years pro- portionately diminished or remained stationary are Louis- iana, South Carolina, Vermont, Mississippi, and one or two * Cf. the historical diagram illustrating the proportion of urban to total popula- tion by States and Territories in the nth Cen., Pop., pt. i, p. Ixv. (20) STATISTICS OF URBAN GROWTH 2 1 others. These are the commonwealths where industry is less progressive and up-to-date than elsewhere ; the popula- tion is not economically organized, or there would be a more pronounced growth of centres of industry and commerce. In tracing the historical causes of the concentration of population in the United States, it is needful to remember that internal migration here has not been of one kind ex- clusively, as in Europe. We have not only the migration cityward, but also the migration westward. Hence in a period when the western movement is particularly strong, the growth of cities is likely to diminish relatively. The urban population is defined in the United States census reports as the population of cities or towns having at least 8,000 inhabitants. Table II, column 3, shows the growth of the urban populatipn from 210,873 persons resid- ing in six cities in 1800, to 18,284,385 inhabitants of 448 cities in 1890. That is, the urban population has increased eighty-seven fold in the century, while the population of the entire country has increased only twelve-fold. In 1800 the city population of this country was grouped as follows : ^ PhUadelphia 69,403 New York (county) 60,489 Baltimore 26,114 Boston 24,937 Charleston 20,473 Salem 9,457 Total 210,873 City life was practically unknown to the fathers of the Republic ; their largest city held a smaller population than 1 Census 0/18^0, p. lii. These are the figures used in the summaries of subse- quent censuses. But in the table of cities, uth Cen. Pop., i, 371, the population of Philadelphia is given as 41,220. The population of Philadelphia county or the present city, was 81,009. 22 THE GROWTH OF CITIES * tJ- lo ii On d dNOOONO •4m m" woo O rO N "1 "^00 NO lO •* u-> 00 lO On li^NO i^NO "^ *> On f^i- OnOnCOm d i^t^'^' NO NO c. u^ 1-1 voOO On fOOO ON PI 1-1 r>.NO N NO TTNO NO NO 00 On "* N •^ « ir)NO_ N f^ "^ 00 iH^ i-J" i-" •^ lO On •^nO 10 On i^ >-c t^ ti- t^ r^NO >-> r^ 00 M ro tJ-nO On lOOO NO t^ "*00 w pT t? no On <-0 ON-1-'^ i-c ro ■* vovo t^ " . Tl" i-c rOOO NO PfJNO q_ Ojoo K? "1; *^ "^ '^^ "^ t^oo iocfNprdiioO>^"?r^ p) PONO rl-OO "INO On On ■I P< P<1 ■>a- t^ i-c PTJOOnOnO^ i-T pT fo •. t^ 1-1 00 11 P) ro Tj-00 -^OO f^ PJ, « pT 1000 "00 1-1 IM NO NO M PONO tJ- "1 1-1 NO NO 00 w « p« TJ-OO Tt N 00 -e- 1-1 N Pt -^ Th PO 11 P4 O PONO 1-1 1-1 PO o Moooo N p) "U>.p) r^oo "1 pj^ Ti^oo 00 q^ '*oo, ro PO i^ p) OnOO On PONO" CTn i-T POOO lO pT PJ O PO PONO NO On Tt- "1 "1 P> On PO PJ^NO^OO q^'^'*^'^"^ PO 10 1^ oi pT t^ PO i-Too o pT 1-1 >H p) PO PO "iNO OQOOOOOQOOO On O ii P< P<1 ■* "InO t-»00 On t^OO 000000000000000000 t) V • n ■S-S-.g.X' « - ^ K* "d «•- " c C « M C *^ U CO u „ c ° o c o rt C C5 n - 3 3 ii B ^ O.T3 B •0 B rt rtrS rt 8 >J2 in B r In B 3 B •2 !?o « 3 a. Oi ,^ rt 3 p. -"0 a rt .J) B B C7> 8 13 •0 j3-a p. B < t^,^ ^ u >> B .B 4) .Q in ^ S t». c; e q X) ^ e ■3 > 8 ■ri 3 B n pi Id >, ■(5 Tif. ^t) '0 •s = ■3 •s-s ^ U rt .S.S-S •a H f-n S A £s: f rt J2 JiC B c nS-S-S •S:° s B _- 1 S"? 1? ^ rt 3 '^w •0 i^ s m 13 STATISTICS OF URBAN GROWTH 23 Cambridge, Massachusetts, in 1890, and but slightly larger than Atlanta. To-day it would rank forty-second among American cities. Not only have the cities increased in size and number, but they have absorbed a vastly larger propor- tion of the population. In 1790, out of 100 Americans only 3.35 were city dwellers; in 1890, the percentage was 29.2. Table II also shows that this process of concentration has not been uniform in point of time. While the rural popula- tion has suffered a steady decline in the rate of increase from 35 per cent, in the decade 1800-10 to 15 per cent, in 1860-70 arid 1880-90, the rate of growth of the urban population fluctuated enormously. In 1810-20 it was 33 per cent.; in 1 840-50, 99 per cent, or three times as great. The causes of these fluctuations are to be sought in the economic conditions of the country. Before 1820 the phenomenon of concentration of popula- tion was not to be found in the United States as a whole. In Maryland and Massachusetts, indeed, the urban popula- tion was gaining slightly upon the rural population, but in the other commonwealths, including New York, Pennsyl- vania, Rhode Island, where the largest proportions of urban residents were to be found, there was no such increase. In fact, the decade 1810-20 showed a relative decline of the cities in nearly all the States, and the urban population of the whole country held its own and no more. This was a consequence chiefly of the destruction wrought to American commerce by the War of 1812, the resulting stagnation of the commercial cities and a movement toward agricultural pursuits. But early in the next decade there opened the era of canals, followed closely by the era of railways, which not only built up great commercial centres, but stimulated the industrial cities by immensely extending their market. The rate of increase goes up from 33, the country's average in 24 THE GROWTH OF CITIES 1 810-20, to 82, two and a half times the country's average in 1820-30. In order to indicate clearly the rate of increase of the urban population as compared with the general rate, the following table is presented.^ It is constructed from Table II by regarding the percentage rates of increase of the population of the United States as a standard, or 100, and calculating the ratio which the other decennial rates bear thereto; thus for the decade 1800-10, 35.10 is to 60 as 100 is to 190, etc. Table III. Urban popu- 124 large cities Rural popu- lation, of 1890. lation. 1800-10 100 190 140 91 1810-20 100 100 no 99 1820-30 100- 244 155 93 1830-40 100 -■ 208 188 92 1840-50 100 -- 276 226 84 1850-60 100 213 171 85 1860-70 100 - 261 207 66 1870-80 . 100 134 129 90 1880-90 100 V 246 196 60 Whether one considers the total urban population with the annual additions to the number of cities, or the fixed number of cities, one finds the periods of maximum concentration, in order, to be 1840-50, 1860-70, and 1880-90, while 1870- 80 and 1850-60 are in both cases minima. The connection with the country's industrial development may be traced out as follows : The immediate result of the opening of the Erie canal in 1821 was the rapid expansion of commercial centres ^ The urban population increases by the growth of villages into towns and cities as well as by the increase of city populations, thus bringing in a possible factor of disturbance much like large annexations to a single city. In order to avoid violent fluctuations that might be caused by large additions to the number of cities at any particular census, the writer has summarized in Table II the popu- lation at each census of all the cities (124) which in 1890 contained not less than 25,000 inhabitants, thus securing a comparatively fixed territory for the entire period. STATISTICS OF URBAN GROWTH 25 in New York State. New York City started on its amazing development, and before 1830 had wrested from Philadelphia the position of metropolis of the New World. Buffalo and Rochester in the same decade passed from the village to the urban stage, soon to be followed by Syracuse, Rome, Utica and other towns along the route of the canal. Simultane- ously began the expansion of manufactures in the New Eng- land States, causing a considerable concentration of popula- tion there. In the following decade 1830-40, the work begun by the canals was continued by the railways ^ and the cities grew apace,^ attaining their maximum rate in 1840-50. Thus far the railways had been confined chiefly to the Eastern States, and by opening up large markets for Eastern manufactures had caused a concentration of population which appears in all the figures of Table II. But after 1850 the railway system was extended into the West, where it became rather an instrument for the dispersion of population by permitting the settlement of the Western lands and furn- ishing an outlet for their products. It was in the decade of 1850-60 that the Mississippi Valley was peopled ;3 and as this was an agricultural or rural movement, the urban popu- lation of the country increased less rapidly than in 1840-50. ^The railway mileage in the United States (^iitk Cen., Trans., i, 6) was: 1830 39.8 1840 2,755.2 1850 8,571.5 i860 29,919.8 1870 49.168.3 1880 87,724.1 1890 163,562.1 ' According to Table II, the urban population had a larger relative increase in 1820-30 than in 1830-40. This may be due to a larger addition to the number of cities in the former period. Between 1820-30 the number of cities doubled; between 1830-40 they increased from 26 to 44. If only a definite number of cities be considered, the larger increase is in 1830-40 (Table III). ' Cf . the Census Reports ; also the standard histories. 26 THE GROWTH OF CITIES But the next decade 1 860-70 showed another large urban increase, second only to that of 1840-50. This may be partly accounted for by the defects of the census of 1870, which admittedly left uncounted thousands of negroes be- longing to the rural population of the South ; but the chief explanation is the check imposed by the Civil War upon the settlement of the West, and the large numbers of the popu- lation that devoted themselves to manufactures during and following the war. The following table gives the rates of increase per cent, in the sections indicated:^ Table IV. North Central South Central Western States. States. States. 1840-50 61.23 42.24 115.12 1850-60 68.35 34-05 246.15 1860-70 42.70 1 1-54 60.02 1870-80 35-76 38.62 78.46 1880-90 28.78 23.02 71-27 This shows how the great movement toward settlement in 1850-60 was followed by a slackening in 1860-70, and by a renewal of the migration westward in 1 870-80, this bringing about a large increase in the rural population and a corres- ponding decrease in the urban percentage of growth as ap- pears in Table III. The fact is even more distinctly brought into light by the following figures respecting the increase in the number of farms ^ and in cereal crops: 3 Table V. Farms in United States. Production of Cereals. Increase Total bushels Increase Per percent. (000,000 omitted), percent, capita. 1840 .. 616 .. 36.1 1850 i,449»o73 •• 867 41 37,4 i860 2,044,077 41 1,239 43 39-4 1870 2,659,985 25 1,387 12 35.9 1880 4,008,907 51 2,698 94 53.8 1890 4.564.641 14 3.519 30 56-2 ^ nth Cen. Pop., i, 4, 5. * jjth Cen., Agr,, p. i. ^ Ibid., p. 6. STATISTICS OF URBAN GROWTH 27 The remarkable increase of the agricultural population in 1870-80, as set forth in this table, connected with the in- dustrial paralysis after the panic of 1873,^ fully explains why the urban growth was relatively smaller than it had been in any decade since 1820.'' In the latest decade, 1880-90, the development of the United States was industrial rather than agricultural, and the migration was cityward instead of westward. While the number of farms and the cereal production have increased but little, the manufacturing interests have prospered as never before. Table VI. Manufactures; Percentage of Increase.' Average number Net value of Capita . of employees. product. 1850-60 89.38 37.01 84.1 1 1860-70 67.80 56.64 63.31 1870-80 64.10 3149 40.01 1880-90 120.78 65.77 106.59 It is thus clearly shown that the decade 1880-90 saw an exceptionally rapid development of the manufacturing in- dustries. The result is that the number of towns of 8,ooo-l- population rose from 286 to 448, an increase of 162, as com- pared with an increase of 60 in 1870-80; and the increase ^ The principal manufacturing States, Rhode Island, Massachusetts, New Jersey, Pennsylvania, as well as Ohio, Maryland and other Eastern States, show a marked decrease in urban growth; and Missouri, Nebraska, Louisiana, Utah, etc., show an actual decline in the urban population. * Nevertheless, it is probably true that the census of 1880 gives too small a num- ber for the urban population. In New England, where the local unit, the town- ship, includes a rural as well as urban population, the Tenth Census attempts to eliminate the rural population, whereas other censuses count the entire popula- tion of the Itownship as urban. In Massachusetts, the Tenth Census finds an urban population of r,042,039 in 2,2) cities; with the methods of the Eleventh Census there would be 36 towns with a population of 1,098,004. For all New England the difference would be about 100,000. * nth Cen., Mfs., i, 4. 23 THE GROWTH OF CITIES in the urban population very nearly reached the maximum of 1840-50. Considering the close connection thus far shown to exist between manufactures and concentration of population, one would expect to find the bulk of city-dwellers in the North, as is indeed the case:^ Table VII. Divisions. Urban population, 1890. Proportion in each division. North Atlantic 9,015.383 49-31 South Atlantic i ,419,964 7.76 North Central 5.793»896 31.69 South Central 1,147,089 6.27 Western 908,053 4.97 Total 18,284,385 100.00 One-half the entire urban population of the United States is in the North Atlantic States and four- fifths in the territory north of the Ohio and Missouri rivers,'' a fact of consider- able social, poHtical and economic significance, and one that will help to explain the results of election contests and legis- lative battles where the economic interests of different com- munities come into conflict. More than one-half of the urban population, again, is concentrated in five common- wealths, as follows : New York 3o99.877 Pennsylvania 2,152,051 Massachusetts i,564>93i Illinois 1,485,955 Ohio 1,159,342 Total 9,962,156 Missouri, which ranks above Massachusetts in total popu- lation, has less than half the number of urban dwellers ' ijth Cen., Pop., pt. i, p. Ixv. * Approximately but not absolutely true, as parts of Missouri and South Dakota and all of Kansas and Nebraska are south of the Missouri river. STATISTICS OF URBAN GROWTH 29 (703,743), and in the present list follows after the little com- monwealth of New Jersey (780,912 urban population). After Missouri, Michigan is the only State that has more than half a million city dwellers, although California lacks very little (about 5,000) of reaching the mark. Thus far, the definition of urban population has been the one formulated by the census reports, the requirement being a population of 8,000. If, however, the line be drawn higher, the Eastern States will have a still larger proportion of the urban population ; and vice versa. But the difference is not so great as might be expected. Table VIII.i Percentage of population in the specified divisions for places Over 8,000 to 4,000 to 2,500 to 1,000 to Total over 25,000. 25,000. 8,000. 4,000. 2,500. 1,000. North Atlantic 51.03 43.70 44.15 39.23 38.04 46.61 South Atlantic . . 7.68 8.05 5-34 9.20 6.9s 7-53 North Central . . 30.38 35-95 37.80 34.80 37.96 33.32 South Central . . 5-9° 7.48 7.64 10.43 II. 14 7-34 Western 5.01 100. 4.82 5-07 6.34 5-91 5.20 United States 100. ICO. 100. 100. 100. It appears from this table that the North Atlantic States have, in their urban population, a large proportion of the great-city dwellers and fewer villagers. The Western States are about equally represented in all the groups, while the re- maining commonwealths are much stronger in villages and small towns than in large cities. The contrast between the North Atlantic and the North Central States is noteworthy, as the two sections contain the bulk of the urban population. In the North Atlantic States the urban population is essen- tially of the large-city type, and in the North Central States of the town or small-city type.- ^ Censtts Bulletin, No. 165. * These facts are made the basis of an interesting study of the " Distribution of our Urban Population," which appeared in the American Statistical Association's 20 THE GROWTH OF CITIES While the eastern and northern States contain the vast majority of the townsmen of the country, it does not follow that they contain the greatest proportionate number of city dwellers. On the contrary, the large urban population of the Mississippi Valley is counterbalanced by a large rural population, while the Western States have a comparatively small rural population. The result is that in the Western States a larger percentage of the people dwell in cities than do the people of the North Central States, as will be seen in the following table:' Table X. Percentage of population residing in towns of io,ooo+. 8,ooo + . i,ooo + . North Atlantic division 48.68 51.58 70.0 South " " 15.05 16.04 22.2 North Central " 24.59 25.90 38.9 South " " 9-82 10.45 1 7-5 Western " 29.71 29.74 44.8 United States 27.59 29.20 41.69 It is somewhat surprising thus to find so large a propor- tionate number of town-dwellers in the far West, as com- pared with the middle West. The probable explanation is Publications (Iv, 11 3-6), and showed how the large percentages of urban popu- lation in the East are due to the presence of great cities. New York and New Jersey afford a striking contrast, as do Missouri and Indiana. In the one case, the vast majority of the urban population is found in the commercial centres of New York City and St. Louis, while in the second case the urban population is scattered in manufacturing towns : Table IX. Percentage of total population New York. New Jersey. Missouri. Indiana. In all places of 1,000-+- ... 68.79 64.58 36.69 32.12 " places of 1,000-25,000 .. 16.21 21.66 12.86 21.98 " " " 25,000-100,000. 7.37 19.06 1.99 5.30 " « " 1 00,000 -|- 45.20 23.86 21.82 4.84 "^ nth Cen., Soc. Stat, of Cities, pp. i, 2; Pop., pt. i, p. Ixv; Census Bulletin, No. 165, pp. 2, 3. STATISTICS OF URBAN GROWTH 31 the difference in the physical features of the two regions. While in the central States agriculture is the staple industry and is prosecuted by small farmers, in the Pacific States commerce brings men together in towns, and agriculture does not call for a dense population, as it is carried on so largely on large estates or ranches ; the average size of a .farm in the Western States, according to the census of 1890, being 324 acres as compared with 133 acres in the North Central States and 144 acres in the South Central States. The following table arranges the States and Territories by groups according to the proportion of their population living in cities of 10,000 and upward : ^ Table XI. CLASS I. MORE THAN ONE-HALF URBAN. 1. District of Columbia 88.10 2. Massachusetts 65.88 3. Rhode Island 57-91 4. New York 57-66 5. New Jersey 5°'9* CLASS II- MORE THAN ONE-QUARTER URBAN. 6. Maryland 43-87 7. Connecticut 41.86 8. California 40-98 9. Pennsylvania 39'^° 10. Illinois 38.08 11. Colorado 37-07 12. Delaware 36-46 13. Ohio 30-15 14. Utah 28.73 15. Washington 28.27 16. Minnesota 27.69 UNITED STATES 27.59 17. Missouri 25.59 CLASS III. MORE THAN ONE-TENTH URBAN. 18. New Hampshire 24-7^ 19. Michigan 23.90 1 iiih Cen., Stat, of Cities, pp. i, 2. 32 THE GROWTH OF CITIES 20. Louisiana 23.65 21. Wisconsin 22.46 22. Nebraska 22.15 23. Wyoming - 19.26 24. Montana 18.58 25. Oregon 18.14 26. Maine 17.16 27. Indiana 16.68 28. Kentucky 13.87 29. Iowa 13.62 30. Virginia 1 2.85 31. Florida 12.02 32. Tennessee 10.65 CLASS IV, LESS THAN ONE-TENTH URBAN. 33. Georgia 9.91 34. Kansas 9.73 35. Texas 9.71 36. Vermont 7.93 37. South Carolina 6.1 1 38. West Virginia 5.85 39. Alabama 5.23 40. North Carolina 3.37 41. Arkansas 3.30 42. South Dakota 3.10 43. Mississippi ' 2.64 CLASS V. NO URBAN POPULATION. 44. North Dakota. 45. Idaho. 46. Nevada. 47. Arizona. 48. New Mexico. 49. Oklahoma. 50. Indian Territory. The cartogram prepared by Dr. Billings, the census agent, to illustrate these figures emphasizes one or two facts of ex- ceptional interest. First, all the Northern States {i. e. the North Atlantic and North Central States together with Deleware and Maryland) have more than 15 per cent, of urban population, except the Dakotas, Kansas, Iowa, and Vermont. The Dakotas are recently settled agricultural STATISTICS OF URBAN GROWTH 33 States and cannot be expected to have cities ; Kansas is in the cattle raising belt, with Texas and Oklahoma ; Vermont has no sea-board and no manufacturing centres, and so falls below Maine and New Hampshire ; but the position of Iowa is peculiar. Iowa stands in a position of isolation in the Middle West; Minnesota has 27.7 per cent, urban popula- tion, Missouri 25.6, Illinois 38.1, Wisconsin 22.5 and Nebraska 22.15, but Iowa has only 13.62 per cent. The Iowa people believe that the growth of their cities has been checked by railway discriminations which have favored Chicago, St. Paul, St. Louis and Omaha, as against their own commercial centres, and this is indeed the only adequate explanation. Among the Southern States, on the other hand, Louisiana (23.65 per cent.) stands in isolation, for it is the only com- monwealth whose urban population exceeds fifteen per cent. It is commerce rather than manufacturing that explains this, for Georgia is far more of a manufacturing State than Louisiana. The presence of a great commercial centre, New Orleans, is what gives Louisiana its prominence. The Western States seem to go to extremes. Four of these are in the second group, and three in the third, while the remainder are in the fifth. None of them are in the fourth group, which is almost entirely Southern. The second group, it may be noticed, contains six States west and six States east of the Mississippi ; but the Western States (except California and Colorado) fall in the second half of the group, while most of the Eastern States are in the first half. The striking feature is that California should out-rank Pennsyl- vania. The commercial importance of San Francisco, fruit- growing and agriculture carried on in large estates, must be regarded as the causes of the high proportion of Californians who live in cities. It was earlier shown that the historical tendency in the 34 THE GROWTH OF CITIES distribution of population in the United States is for an ever- increasing proportion of the people to dwell in towns. A question of considerable importance is whether the tendency is toward great or small cities. The only careful investiga- tion of this question is a paper by Mr. Carl Boyd in the Publications of the American Statistical Association.^ His classification of the cities rests upon the mean population of 1880 and 1890; thus Milwaukee, which had a population of 204,468 in 1890, is not included in the class of 200,000 and upwards because its population in 1880 was 115,587 and the mean would be 160,000, The result of the analysis is as follows : Table XII. Percentage increase of No. of Cities. population, 1880-90. 200,000 or more 12 36.87 60,000-200,000 28 63.07 30,000-60,000 40 52.45 17,000-30,000 81 53-72 11,000-17,000 93 53.74 8,000-11,000 102 47'30 356 This apparently demonstrates that the great cities of America are growing less rapidly than the smaller ones. But the evidence is inconclusive, for the class of cities 200,000 and upwards happens to include the cities of the settled East where population is increasing less rapidly than in the West; only two of the 12 cities in this group (St. Louis and San Francisco) are west of the Mississippi. On the other hand the second group, which shows the largest increase of all, includes the most rapidly growing cities of the new West; Minneapolis with an increase of 251 per cent; Omaha, 360 per cent; St. Paul 221 per cent; Kansas City 138 per cent; Denver 200 per cent; etc. The only ^ " Growth of Cities in the United States during the Decade 1880-90," iii, 416^. STATISTICS OF URBAN GROWTH 35 inference to be drawn from these figures is that in an era of rapid colonization the growth of new centres of distribution may overshadow the development of the older commercial centres. We are still left in doubt as to the relative growth of large and small cities in settled communities. It is un- fortunate that the American census authorities do not emu- late the European ofificial statisticians in making useful tabulations of the facts they gather ; it is too much to ask of the individual investigator to classify the cities and compute their population for all the censuses. In Table II, however, the 124 cities of 25,000+ in 1890 have been divided into two classes, the line being drawn at 100,000, and the aggregate population of the four principal cities has also been com- puted. A comparison of the rates of growth of the three groups with the rate for the United States entire is as fol- lows, the latter rate being represented as 100 as in Table III : Table XIII. New York, Chicago, Phila- Cities of 100,000+ Cities 25,000-100,000 delphia and Brooklyn. in 1890. in 1890. 1800-10 130 169 84 1810-20 80 120 80 1820-30 145 148 175 1830-40 165 195 173 1840-50 200 227 238 1850-60 186 170 173 1860-70 143 194 228 1870-80 117 125 137 1880-90 173 180 236 In only three of the nine decades (1800-10, 1810-20, 1830-40) did the great cities grow as rapidly as the middle- sized cities, while the four great centres were distanced in every decade except 1850-60. Since 1840, at least, it would appear as if there had been no movement toward centralization, such as exists in France with its one great metropolis ever distancing the other cities in growth. But 36 THE GROWTH OF CITIES some individual cities do, nevertheless, exert something of the power of attraction possessed by Paris. Take for example the cities of New York State. In 1890 there were 13 of 25,000 or more inhabitants; three of which, New York, Booklyn and Long Island City, belong to the metro- poHs. The population of these three cities in 1890 was 3.8 times as great as in 1850; the population of the other ten cities 3.6 times as great, thus indicating the more rapid growth of New York city than the interior cities, some of which, however, notably Buffalo and Yonkers, outstripped New York. But Yonkers is itself a New York suburb, and if we were to add the parts of industrial New York that lie in New Jersey, we should find that the metropolis heads the list. While our evidence is by no means conclusive, it points to the inference that New York like Paris, grows at a more rapid rate than the smaller cities in the settled parts of the country. Of course. New York has been outdone by the commercial centres of the new West, and will continue to be outdone by them until their dependent territory is in a measure filled up. The great difficulty in the statistical method here lies in its failure to take account of the great city's growth outside its own boundaries. The movement toward the suburbs, which is stronger in America than elsewhere with the excep- tion of Australia, not only necessitates frequent annexations of territory, but even then baffles the statistician. This may be illustrated by studying historically the distribution of population in Massachusetts. And first as to annexations. In 1890 Massachusetts contained sixteen cities whose population individually exceeded 25,000. Computing their aggregate population at each census and calculating the rates of increase, the following result is reached:' ^ nth Cen. Pop., i, 370-3. STATISTICS OF URBAN GROWTH ^y Table XIV. Decennial rates of increase Proportionate RATES OF INCREASE. Boston. t Mass. 16 cities. Boston. Mass. 16 cities. a. *. i8co-io.. ..11.63 27.8 33-24 100 238 283 344 1810-20. . ..10.83 ^S'9 30.22 100 147 280 250 1820-30.. ..16.68 55.6 41-79 100 332 250 228 1830-40.. ..20.85 55-7 52.11 100 268 250 240 1840-50.. ..34.81 63-9 46.58 100 183 134 149 1850-60.. ..23.79 33-0 29.92 100 139 126 147 1860-70.. ..18.38 39.6 40.87 100 215 222 98 1870-80.. ••22.35 42.1 44-83 100 189 201 107 1880-90.. ••2S^57 32.9 23.60 ICO 129 92 90 The historical changes in the rate are not difficult of ex- planation. The impetus to city growth in 1820-30 was given by the cotton trade; Lowell was founded in 1826 and appears in the census of 1830 with a population of 6,474. The high rate of increase of the sixteen cities in 1840-50 is without doubt due to the fact that Lawrence, Somerville and Holyoke appeared for the first time in the census of 1850, completing the list. Since then the rate, especially as com- pared with the average for Massachusetts, has steadily de- clined. But this is also true of Boston, so that one feels warranted in drawing the conclusion that the abnormal growth of cities is a matter of the past rather than the future. This inference would seem to be more strongly justified, moreover, when one considers the rate of increase for the Boston of to-day, as given in the last column of the table. Here the disturbing influence of the annexations of 1 867-1 873 =" have been removed and it is now seen not only ' Column a corresponds to the rates given above, and is based on the popula- tion of Boston as returned at each census; column 6 is based on the population, at each census, of the present area of Boston, thus including all annexed territory. *The annexed districts with population at federal census of 1870 were : 1867, Roxbury 34.753 1869, Dorchester 12,261 1873, West Roxbury 8,686 1873, Brighton 4.967 1873, Charlestown 28,323 Cf. Census of Mass., 1895, J. 221. 38 THE GROWTH OF CITIES that Boston's proportionate rate of growth has been dimin- ishing throughout the century, but that, since 1820, it has fallen below that of the sixteen cities, and in 1860-70 and 1880-90 below the average of Massachusetts. But all such calculations are upset by the lack of agree- ment between the urban centre proper (the economic city) and the legal city or municipality of the time being. The recent State census of Massachusetts shows that the most rapidly growing towns are the suburbs of Boston. As follows was the percentage increase of population in 1885-95/ the suburban cities being indicated thus* : ♦Everett 218.85 *Malden 81.07 *Somerville 74- 1 7 Fitchburg 7i'77 "'Quincy 70*54 New Bedford 65.46 *Medford 60.08 Brockton 59-58 Fall River 56.85 North Adams 52.59 Massachusetts 28.73 Boston 27.29 " (^within 8 miles of the State House) 37'i9 «« «« 12 " " " " " 37-26 All incorporated cities (32) 38-05 Rural remainder I4-I5 While, therefore, the capital city with its environs is con- tinuing to absorb the population of the state, so that to-day the " Greater Boston " contains 39.4 per cent.^ of the popu- lation of Massachusetts as against 5.9 percent, in the Boston of 1800 and 13.7 per cent, in the Boston of 1850, its relative ' Census of i8g^, i, 46-9. The list includes all cities whose rate of increase in the decade exceeded 50 per cent. *The so-called Metropolitan district includes virtually all towns within a radius of ten miles of the State house. With a twelve-mile radius, the percentage in Boston would be 40.17. Cf. op. cit., i, 47. STATISTICS OF URBAN GROWTH 39 growth is less rapid than it was during the first half of the century. The proportion 28.75 : 37.26=100: ;tr yields 130, whereas the proportionate rate did not fall below 147 until 1870. But even at this rate the cities would gain upon the popu- lation of the country at large, and one may look for still more profound changes in the distribution of population than those indicated in the following table, which is based on Table XVI : Table XV. Percentage of population of the United States residing in specified cities in the years: 1800. 1850. 1890. loOjOOO-f o 6.0 15.5 20,000-100,000 3.8 3.8 8.3 10,000-20,000 O 2.2 3.8 Total 10,000+ 3.8 12.0 27.6 The present indications are that before 1910 one-half of the American people will be residents of cities of 8,000, This condition would be reached about 1920 if the rate of concentration from 1820 to 1890 should prevail; in about 191 5 if the rate of 1850-90 should be maintained ; and soon after 1905 if the rate of the most recent decade, 1880-90. Changes in the position of the different commonwealths with reference to the proportion of urban population may be expected. Concentration began in Massachusetts and Rhode Island, Connecticut and Maryland.' The era of rail- ways with the accompanying expansion of commerce on the one hand and of the iron industry on the other, caused a rapid growth of cities in New York and Pennsylvania. In 1830 and again in i860. New York and Massachusetts had an equal proportion of urban population ; but now Massa- chusetts has a percentage of 70 per cent, of its population in towns of 8,000+ and New York 60 per cent. New Jersey ^ Cf. Diagram in nth Cm., Pop., pt. i, p. Ixv. 4.0 THE GROWTH OF CITIES and Connecticut are overtaking New York, as a result of the development of small manufacturing cities. Both passed Maryland and Pennsylvania in the decade 1860-70, when the war gave life to so many machine industries. Illinois, now the ninth State on the census chart showing proportion of urban to total population, betrays the most uniform and consistent increase of any of the States. With the continual development of manufactures in Illinois and the continued growth of Chicago, there can be no doubt that by 1900 Illinois will have passed Pennsylvania, California and Mary- land and have taken place close after New Jersey and Con- necticut. A summary of the urban statistics of the United States is given in Table XVI. 1800. 1850. 1890. Classes of cities. No. Population. No. Population. No. Population. 100,000-1- .. 6 1.393.338 28 9,697,960 20,000-100,000. 5 201,416 24 878,342 137 5,202,007 10,000-20,000. . 5 36 66 495,190 2,766,870 180 345 2,380,110 Total io,ooo-1-. 201,416 17,280,077 " 2,000 -f. . 18% est. 1,916 23.593.605 Authorities. For 1800, see p. 21. For 1850, the Census of 1850, p. Hi and full tables. To the 22 cities of 20,000-100,000 in the lists on p. Hi are added Allegheny City (21,262) and Chicago (29,963). Only 27 cities of the third group are given in the lists of principal cities (p. Hi), their aggregate population being 372,584; to this number may be added Manchester, N. H., New Bedford and Charlestown, Mass., Nor- wich, Conn., Oswego, N. Y., New Brunswick, N. J., Norfolk and Petersburg, Va., and Lafayette, Ind., whose population was found in the tables for the separate States. The estimate of towns of 2,000+ is derived from a comparison of Tucker {Progress of the United States, 132), who givei the total for 1840 (2,321,527, or 13.6 per cent, of the total population) with the census figures of 1840 and i860 for towns of 8,000+. For 1890, the Census Report on Social Statistics of Cities, p. 1, gives the total of cities xo,ooo+. The population of towns of 2,000+ is derived from Census Bulletin, "Ho. 165, and the other totals are summaries from the Report on Population, pt. i, pages 370-3. II. THE UNITED KINGDOM. §1. England and Wales. — From many points of view, England offers superior advantages for the study of the dis- tribution of population and the causes affecting the same. STATISTICS OF URBAN GROWTH 41 England was the pioneer in the modern industrial move- ment and is even now the typical industrial country. For while the aggregate output of machine or factory-made pro- ducts in the United States exceeds that of England, it does not constitute so large a proportion of the entire national product. A smaller percentage of Englishmen and Scotch- men are devoted to agricultural pursuits than of any other nation of the world. The latter being the only workers who are of necessity resident in scattered habitations, it will be worth while to ascertain under what conditions the remainder of the population has dwelt. Unfortunately, the English statistics present serious diffi- culties to the classification of dwelling-centres. This is chiefly the fault of the historical English method of local government, distributing the various functions to a variety of independent authorities over different areas and thus pro- ducing a chaos of boundaries and officials. In 1871, the 938^ dwelling-centres which were taken to represent the urban population of England and Wales were thus classified : Municipal boroughs, comprising cities and towns to which charters of incorporation had been granted and were later governed by the Municipal Corporations Reform Act, 224; local board districts established either under the Public Health Act of 1 848 or under the Local Government Act of 1858, 721 (including 146 municipal boroughs); places which had improvement, paving, lighting or other commis- sions under (special) Local Acts, 88 (including 37 municipal boroughs); other "towns" of some 21,000 population, 96. But these " towns " not under a regular municipal authority had no recognized boundaries, and " the Superintendent Registrar of the District in each case distinguished the houses which in his opinion might properly be considered within the limits of the town." ^ Or 946, if London be counted in its parts instead of a unit. 42 THE GROWTH OF CITIES Such was the condition of affairs in 187 1. In the earlier censuses there were still fewer places which had definite and recognized boundaries ; such being the boroughs alone. But a borough might be either municipal or parliamentary. And sometimes there was a vast difference between the limits of the two ; for example, Wolverhampton, the munici- pal borough, had a population of 49,985 in 185 1, but a population of 119,748 dwelt within its parliamentary- boundaries. All of which tended to confuse the statistician. In 1872, however, legislation simplified matters consider- ably. By the Public Health Act of 1 872 ', it was enacted that all municipal boroughs, local board districts and towns with improvement commissions, should henceforth be termed " urban sanitary districts," and to these authorities were transferred all the powers and duties previously exercised by any other authority in the districts under the provisions of acts relating to local government, the utilization of sewage, the removal of nuisances, the regulation of common lodging houses, baths and wash houses and the prevention of disease. These powers connected with sanitation, which have since been augmented, are so important and, in fact, so essential a part of city government, that the English urban sanitary district may well be considered the typical urban com- munity. To the original urban sanitary districts of 1872, others have been added from year to year, being carved out of the great area comprised in the rural sanitary districts as rapidly as an agglomeration of people becomes of sufficient magnitude to require urban sanitary regulations. The num- ber of urban sanitary districts in the Census of 1891 was 1,011, of which only 194 had less than 3,000 inhabitants. Their aggregate population constituted 71.7 per cent, of the entire population of England and Wales; only 1.3 per cent. ^ 35 and 36 Vict., cap. 79. STATISTICS OF URBAN GROWTH 43 of the population dwelt in urban districts smaller than 3,000.^ With these explanations, the following summary table is put forward to show the degree in which the population was concentrated at the different periods in the present century : Table XVII. 1801. 1851. 1891. Classes of cities. No. Population. No. Population. No. Population. Over 20,000 15 1,506,176 63 6,265,011 185 15,563,834 10,000-20,000 .... 31 389,624 60 800,000 175 2,362,376 5,000-10,000 60 418,715 140 963,000 262 1,837,054 Total 5,0004- •• 106 2,314,515 263 8,028,011 622 19,763,264 " vinder 5,000. ... 6,578,021 ... 9,899,598 ... 9,239,261 Grand total 8,892,536 ...17,927,609 ... 29,002,525 Sources. The population of cities of 20,000 and upwards is derived from Table XVIII, where the sources are named. For the smaller cities and towns, the data were derived as follows: For 1801, a list of cities and towns containing upwards of 5,000 inhabitants in the Annual Register for 1801, pp. 171-3. A compilation from the data in the Census of 1841 (Introductory Volume, p. 10) gives approximately equal results, e. g., 46 cities of 10,000 with total population of 1,882,667 in the Annual Register. For 1851, only the aggregate urban and rural populations are given in the census volume (I, p. xlvi, Table XXIII). The summaries of cities above 20,000 are exact, and are based on table 42 (p. cxxvi of the iB^i Census, vol. I); of the smaller cities they are only approximately correct (Table VII, p. ccivff). For 1891, Census o/iSqi, iv, 10, General Report. The numbers relate to urban sanitary dis- tricts, and the administrative county of London is reckoned as one district. In the ninety years covered by the table, over twenty millions of people were added to the population of England and Wales ; but while the rural inhabitants (those dwelling in places of less than 5,000) increased from 6,600,000 to 9,200,000, the town-dwellers increased from 2,300,000 to 19,800,000. That is, of the total increase of twenty millions, about 17,400,000, or 80 per cent., fell to the towns and cities. It is, moreover, noticeable that the increase of the rural population took place entirely in the first half of the century, and later turned into an actual decrease. Some 800,000 more people are classed as belonging to the rural ' That some rural sanitary districts contain a population of 50,000 or more signifies nothing, since this is a scattered population contained within a large area. 44 THE GROWTH OF CITIES population in 185 1 than in 1891 ; so that while in some countries much is said of a rapid growth of cities, causing the rural population to suffer a relative decline, in England there is a decrease in the absolute numbers of the rural in- habitants. The decline began in 1 861, as may be seen in the following percentages of decennial increase or decrease cal- culated from the figures of Table XVIII : Urban. Rural. 1851-61 21.9 -f-1.88 1861-71 28.1 —5.86 1871-81 25.6 — 3.84 1881-91 18.5 — 2.76 It does not follow from this that there has been a rural depopulation in England in the strict sense of the term; for the aggregate rural population of a country may diminish either as a result of emigration or as the result of the growth of hamlets and villages into towns and cities. The increase in the mere number of cities is remarkable. In 1891 Eng- land and Wales contained 360 urban sanitary districts, or towns, of 10,000 and more inhabitants; in 1881 there were 123 ; in 1 801, 45, and in 1377, with a population nearly one- fourth as great as that of 1801, only two cities, London with almost 35,000 inhabitants and York with about 11,000. Nine towns are believed to have had a population of 5,000 or more, and 18 a population of not less than 3,000^ in lieu of the present number of 817 in a population about fourteen times as large. In 1377 the population of these 18 towns, formed eight per cent, of the entire population^. In 1688 the towns, according to Gregory Kings, contained one- ^ Based on an enumeration for the poll tax. For the list of cities, see Jas. Lowe, The Present State of England, London, 1822, appendix, p. 74. Cf. also the economic histories. *The ratio of urban to rural population was as i : 12,34. — Rogers, Six Cen- turies of Work and Wages, 129. * National and Political Observations upon the State and Condition of England, 36. STATISTICS OF URBAN GROWTH 45 fourth of the population. But such estimates are extremely crude, as appears from Arthur Young's observation in 1770 that half the population was urban ^ ; whereas Table XVII shows that not more than one-third of the population could properly be called urban even in 1801. The question of rural depopulation in England has been frequently discussed by the English statisticians and will not be entered into here, at any length, since this paper is con- cerned with the relative increase or decrease of the country as compared with the city. Dr. Ogle's investigation' covered the seventeen registration counties in which more than ten per cent, of the population were devoted to agri- cultural pursuits, with the exception of the two mining counties of Cornwall and Shropshire. Defining as the rural population the inhabitants of all districts in these counties except urban sanitary districts of 10,000 and upwards, he found that it aggregated 2,381,104 in 1851 and 2,358,303 in 1 88 1. The loss is scarcely perceptible. And even if the rural population be restricted to sanitary districts of less than 5, 000, it shows a decrease of only 2.1 per cent, in the entire period. There are fifteen English counties which reached their maximum population in 1841 or at some other census prior to 1 89 1. Their aggregate loss from the year of maximum to 1891 is 133,600, and of this loss 46,570 or over one- third is found in Cornwall, where the cause is not agricul- tural depression, but failure of the tin mines.3 All this does not imply a rural depopulation, which connotes a great 1 Travels in France (2d ed.), i, 480. Cf. Toynbee's criticism, Industrial Rev- olution (Humboldt ed.), p. 37, foot-note 3. * Ogle, " The Alleged Depopulation of the Rural Districts," in Jour, of Stat. Soc. (1889), 52 : 205/: Cf. also Longstaff, loc. cit. (1893), 56 : 380, and Census of jSgr, iv, 12 (^Parliamentary Papers, 1893, cvi). » Cf. Longstaff, Studies in Statistics, ch. v, where similar results are given for the census of 188 1. 46 THE GROWTH OF CITIES Q O -^ . O M " -^ ■* Ti- \0 vO ^ t^OO On '^^O 0\ "^ cT •? cT 0\^0 OOii-iM'Oi-'t^O 00 M ooq_q;q\q_t^o;q^ od"""""'""" Aggregate rural pop- ulation. . O ^ M «^ " . O N VO 0\ N • c» M 00^ r;- q_ • mo r- m O • On I-" ^ ro 1- • 00 cfvOO 00 00 Aggregate urban population. Towns. Population. • OsOO rl->J3 Tl- • O On O S" O • 00_ OS 'JnO vo ■ cT 0^ i-To lo . 0\vD Tf ro On , On 0\ O ^ 00_ • 00 O ■J-t -^00 N NO o fO t^ Tf N Tj- M ITNOO P) NO fO m ^^ *^ ^ ^ '-^ ''2°^ i-Toio'foio'^eri-r-ifoN ■«*■"" t^o Onn mr^ij-)0 nOnOOO c^no NnOnO l^ro i-T i-T cT pT ro ■- T?0\to>^r^prpO^-^Pr NO O P) r~»r-~0 O to fO m 00 o PI '^00 '^•^ ^°^ ^ i-Ti-rt-ri-rprprrornTf »> O " w_l| * mOO-S C V c c " V S o o ?fo „-.S .j3 a " u E ? S ^■S'S 1^ 2 S u 2 o-B.ai->^ « s S S "> u CJ3 o « s y 5 «'r5_, a j=n3_oJ'«cW5Ma « „ <<2 5j^.§ ° X S ! - a . " ."^ c_ « >< u _e a-" n C.2 °^ o-g •- ■g 3 M a ,„-, T3 u u « 2 0.-9-= 2-5 o S n^TS a M T-1 u u 11 ci o.-^ -tv :i -a o lu " C n ui O " >>J3 -- Co a ! 6". 3 U |- n.S'^ ^<_ ■"•S-- C^ •5 3 "•OJi (!i u°° '^S !>^ a S "u *^ M-s o^i— *^ "* I ■a^.S|^£S|i2S-5,d -a a^°° §3 « " » g-- 2 a • iS M "'B o-c " - 2 o S ^■a g a B-g a o a^ ^.ojs ^ ° 2 2^ o c^5 g'f-S o -.2 „ a " cS «, a 3-3 o..a S nJ='~' CO 0)3.5 E a« 4j . ali ^^uo S a-g (1.-5 «.a S Sfoo o -a 2 •_ « Mf. 2 a*^LH^rt^ -O-Q^-Ma 4; C ^* .. "* ^ . fli 1> C3 4J cS *-• CO S rn ^* '♦ja 3 t^ (Xn fi.S a ♦^ o o STATISTICS OF URBAN GROWTH 47 scarcity of farm labor. Nevertheless, it is somewhat start- ling to read in the preliminary census report of 1891 (p. vi) that in the preceding decade there was a decline of popu- lation in 271 out of the 632 registration districts in England and Wales; and that in 202 of the 271 there had also been a decHne in the decade 1 871-81. And even Dr. Ogle admits that hard times, the use of labor-saving machinery and the conversion of arable into pasture land have caused the farmers to reduce their labor force. But the decrease of the rural population in England is less significant than the great urban increase already indicated in Table XVII and set forth in greater detail in Table XVIII. Looking at the movement of population as a whole it is seen what a considerable change in the conditions of life has taken place during the century. The population of cities of 20,000 or more inhabitants has multiplied tenfold, while the remainder of the population has not quite doubled. A table of percentages will present the facts embodied in Table XVIII more graphically : Table XIX Percentage of population of England and Wales in London. Other great Cities 20,- All cities Urban Rural cities. 000-100,000. 20,000 + . districts. districts 1801... 9-73 0. 7.21 16.94 I8II... 9-93 2.08 6.10 18.II I82I... 10.20 3-27 7-35 20.82 I83I... 10.64 5-71 8.70 25-05 I84I . . . "•75 6.52 10.63 28.90 . .. I85I... 13.18 9.40 12.42 35-00 50.08 49.92 I86I... »3-97 11.02 13.22 38.21 54.60 44.40 I87I... 14-33 11.50 16.20 42.00 61.80 38.20 I88I... 14.69 14.91 18.40 48.00 67.90 32.10 I89I... 14.52 17-30 21.76 53-58' 72.05 27.95 In 185 1 the urban population constituted 50 per cent, of the whole; in 1891, 72 per cent. This increase, moreover, was not in the smaller cities, since the cities of 20,000 and 48 THE GROWTH OF CITIES over account for 18.5 of the 22 per cent, increase. The in- crease has not been divided equally among the three classes of cities over 20,000 shown in the table, for London has gained only 1.34 per cent, while the other " great cities " have gained 7.9 per cent, and the middle-sized cities 9.34 per cent. London, fn fact, grew no more rapidly than the small cities, i. e. those having between 2,000 and 20,000 in- habitants. While, then, there has been a concentration of population in cities of at least 20,000 population, it is not a form of concentration carried to the extreme ; for London's population is barely holding its own in the general growth of population, and is dropping behind the population of provincial and middle-sized cities. These latter cities are indeed constantly recruited from the next lower order of cities, but even when such manner of growth is excluded and the comparison confined to a fixed number of towns, it will not be found that population is concentrating in one great metropoUs very rapidly. An English statistician, Mr. R. Price Williams, has, happily, summarized the population in a certain number of towns at each census from 1801 to 1871, using, in nearly all cases, the same territorial limits. While many considerable cities of 1871 were mere villages in 1801, and therefore a part of the rural population, for present pur- poses this fact may be neglected. Taking the rate of in- crease as given in Mr. Williams's brilliant paper in the Journal of the Royal Statistical Society^, and comparing them with the rate of increase for the entire country as the average or standard (100), the following interesting results are obtained : >Vol. xliii (1880), 462-496. STATISTICS OF URBAN GROWTH 49 Table XX. Proportionate Decennial Increase. Other All cities of Total Total Pop. of Eng- London. great ao,ooo+ exc. Small urban rural land and cities. London. towns, population, population. Wales. 180I-II ... 131 150 140 91 122 84 100 181I-2I ... 117 163 147 105 126 81 100 1821-31 ... 126 239 190 95 143 66 100 1831-41 ... 122 202 182 86 138 67 100 1841-51 ... 167 209 195 82 158 46 100 185I-61 ... 157 172 165 61 137 61 100 1861-7I ... 121 131 152 83 132 64 100 Explanations. London — the registration district. " Other great cities " — the 16 cities which, in addition to London, had over 100,000 inhabitants in 1871. " Small towns " — places which contained, in 1871, a population of 2,000-20,000. Urban population — the population at the various censuses of all towns with 2,000 or more in- habitants in 1871. Rural population — the remainder. Some of these figures agree with the calculations of Sir Rawson W. Rawson in the J. of Stat. 5c£-.,43:soo. It will be noticed that the urban population of Mr. Wil- liams's calculations has uniformily exceeded in its rate of increase the average of the entire country ; while the rural population, being the complement of the urban population in the general total, has of course fallen below the average. The smaller towns, it is worth noting, have also grown less rapidly than the population of England and Wales in its entirety, except in the single decade 1811-21. The 98 cities other than London which had a population in excess of 20,000 in 1 87 1, have far outstripped the small towns and rural districts ; but even their rapid growth is inferior to the expansion of the sixteen " great cities," in every decade ex- cept the last. There is therefore a regular progression in the rate of growth, beginning with the villages and scattered population in the country districts and proceeding through the several classes of towns to the largest cities. LevasseurS observing a similar phenomenon in France, formulated the law that " the force of attraction in human 1 Following Legoyt, Cf. La France et V Stranger (1865), pp. 50, 262, etc. so THE GROWTH OF CITIES groups, like that of matter, is in general proportionate to the mass."' If London be classed with the other English cities cff 100,000, Levasseur's rule will hold good for England. But London's population is properly a class by itself and is so treated in the tables. London might be expected to grow more rapidly than the provincial cities, just as Paris does. But such is not the case in a single decade of the century. Moreover, in the last decade represented by Mr. Williams's figures, the great cities were outstripped by the middle-sized cities. This tendency has become even more manifest in the subsequent periods, as appears in the follow- ing table ^ for 188 1-9 1 : Table XXI. Urban Sanitary Districts. Mean percentage of in- Population. No. crease of population. London i 10.4 250,000-600,000 5 7.2 100,000-250,000 18 19.9 50,000-100,000 38 22.8 20,000-50,000 123 22.1 10,000-20,000 17s 18.9 5,000-10,000 262 1 1.5 3,000-5,000 ^95 ^'^ Under 3,000 194 3.6 Total urban 1,01 1 15.4 «• rural 3.0 It thus appears that the largest growth is in the cities of from 20,000 to 100,000 inhabitants, with the classes 100,000 to 250,000 and 10,000-20,000 in close company. The urban districts under 3,000 have gained little more than the rural districts and the small towns (3,000-10,000) and the six great cities have increased less than the general population. That the villages should be falling behind is not surprising, * La population franfaise, ii, 355. ' Census ofjSgi, iv, 10. STATISTICS OF URBAN GROWTH 51 but it at first seems strange that the very largest cities should manifest so slow a growth, The six cities are : Increase or decrease per cent London 10.4 Liverpool — 6.2^ Manchester 9.3 Birmingham 9.4 Leeds 18.9 Sheffield 14.0 Leeds and Sheffield alone rise above the general rate (11.65) fo'' England, while Liverpool has actually lost! Yet nobody believes that Liverpool is decaying ; the expla- nation of the matter is simple enough : the growing business of the city requires the transformation of dwellings into stores and the dispossessed persons move away from the centre of business. As there is little more room within the municipal limits most of these people live in the environs, but are no longer counted in Liverpool.^ This process is going on in nearly all the great cities, as will appear later. But as it is comparatively recent that such cities as London have reached "the point of saturation," it does not affect the conclusion that in England the largest urban aggrega- tion has exerted a weaker power of attraction than the class of " great cities." The concentration of population has not been carried to its utmost point in the England of the nine- teenth century. There now arises the question of the causes that in- fluenced the distribution of population in England and Wales in the direction shown by the preceding analysis. The tables of Mr. Price Williams 3 show that the maximum urban increase was in 181 1-2 1 ; but this was balanced by a ^ Decrease. " Cf . E. Cannan, " Growth of Manchester and Liverpool " in Economit Jour- nalfiv, 111-114. ' y. of St. Soe., 43 : 462, seq. 52 THE GROWTH OF CITIES similarly large increase in the rural districts, so that the con- centration was not so great then as later. Table XIX shows that the entire period 1821 to 185 1 was a period of concen- tration in cities, and in this period the two decades 182 1-3 1 and 1 841-5 1 are especially marked. The earlier decade saw the rise of the sixteen "great cities;" the second decade, the expansion of London. The charts accompanying the article of Price Williams already referred to clearly show that most of the great English cities attained their maximum rate of growth in 1 82 1-3 1, which was in some cases phenomenal. Brighton's increase, for example, was 69.7 per cent., Bradford's 65.5, Salford's 55.9, Leeds's 47.3, Liverpool's 45.8, Manchester's 44.9, Birmingham's 41.5, Sheffield's 40.5. Thus eight of the twelve^ "great cities" of 1871 owe their largest decennial increase to the causes at work in 1 821 -31. With the ex- ception of Bristol, all the eight cities are in the manufactur- ing district in the North ; and Bristol is a port in an adjacent county. It may therefore be conjectured that the period was marked by a great expansion of the manufacturing in- dustries ; and it was indeed at this time that the cotton trade began to assume large dimensions. The number of pounds of cotton imported'' was in — 1781 5.198,775 1785 18,400,384 1792 34,907,497 1813 51,000,000 1832 287,800,000 1841 489,900,000 "Though 1800 marks the beginning of a continuous ex- pansion in both cotton and woollen manufactures, it was not until about 1817, when the new motor had established itself ^ The four other cities were Hull, Nottingham, London, and its suburb West Ham. * Hobson, Evolution of Modern Capitalism, 60. STATISTICS OF URBAN GROWTH 53 generally in the large centers of industry and the energy of the nation was called back to the arts of peace that the new forces began fully to manifest their power."' It must there- fore be clear to every mind that the decade under discussion ( 1 821-31) presents in England a typical instance of the effect which the growth of manufactures and the develop- ment of the factory system, or system of centralized industry, has upon the distribution of population.^ The marked concentration of population in 182 1-3 1 was, then, produced by the industrial changes affecting the cities of northern England. In the succeeding decade the concen- tration continued, at a somewhat diminished rate, under the same influences. But in 1 841-51 was reached the most notable period of concentration in the century. Table XX shows that London's rate of increase reached its maximum at this time, but that alone does not fully account for the effect. A more detailed analysis, as follows, shows that the ^ Hobson, op. cit., 64. ' The influence of the above-mentioned forces on the movement of population in a small area may be studied with profit in Lancashire, the seat of the cotton trade. Price Williams {loc. cit., p. 476^) gives these figures : Table XXII. Decennial Increase OF Population in Rural Urban Small towns. Large towns. Lancashire. districts. districts. (2,000-20,000.) (20,000+.) I80I-II. .. 23.02 20.44 25.07 19.40 26.37 I8II-2I. .. 27.09 20.20 32.37 25.44 33-08 1821-31 . .. 26.97 13.29 36.46 19.44 39-95 I83I-4I • .. 24.70 12.51 31-73 19.07 33-94 1841-51. .. 21.84 12.67 26.37 14.35 28.23 I85I-6I. .. 19.61 20.46 19.23 20.77 19.01 I86I-7I. . . 16.06 14.86 16.58 19.91 16.12 The noteveorthy feature of Lancashire's growth is the concentration of popula- tion in the large cities which was going on throughout the period 1 801-51, and especially in the decade 1821-31, when the great expansion of the textile industry took place. 54 THE GROWTH OF CITIES middle-sized cities were even more influential in their action on the general result : Average Eng. Other great 82 cities Towns 2,000- Total and Wales. London, cities (ioo,ooo-f )• 20,000-100,000. 20,000. urbaji. 1821-31 .... 100 126 239 150 95 143 184I-51 .... 100 167 209 182 82 158 The counties in which the urban population {i. e., cities of 20,000 and upwards) increased most in 1841-51 are as follows : Per cent. Monmouth 78.7 Bedford 65.0 Chester 48.8 Glamorgan 45.3 Lincoln 44.7 Average of the 82 cities 23.2 As to the influences on the growth of the city population of Monmouth and Glamorgan counties there can be no doubt. Glamorganshire contains the great coal-fields and iron-mines of South Wales, and it was the iron industry that built up its three largest cities, Merthyr Tydvil, Cardiff, Swansea, and the iron-smelting seat Newport in Monmouth. It is to be regretted that the development of the iron in- dustry in the North cannot also be traced ; but the iron de- posits and the smelting-centres there are mainly in York- shire and Lancashire, where the other factors of previously established manufactures complicate the study. Outside the iron districts the cities that showed the highest rates of increase in this period were ports. Thus in Lincolnshire, the growth was mainly in Grimsby, a rival of Hull ^ ; Cheshire's high rate is due almost wholly to Birken- head, a port on the Mersey opposite Liverpool. Of the ten ^ Chisholm (^Handbook of Commercial Geography), speaking of Hull, says that "since the introduction of railways it has had a growing rival in Grimsby." (P-231.; STATISTICS OF URBAN GROWTH 55 cities that grew most rapidly in 1 841-51, seven were seaports.^ Now commercial statistics show that English trade ex- perienced a very considerable impetus about this time ^, and the stimulus came from the opening of railways, which was of course accompanied by a great expansion of the iron industry. The first railroad, the Liverpool and Manchester line built by George Stephenson, was opened in September 1830. Its effect on the distribution of population was not immediate, owing to the slowness with the system was de- veloped. But by 1840 the United Kingdom possessed 800 miles of railways, and construction was then going on so rapidly that by 1850 the number of miles had risen to 1 Birkenhead (port), Hanley in Staffordshire, Torquay (p), Grimsby (p), Cardiff (p), Newport (p), Southport (p), Luton in Bedfordshire, Hythe (p), Merthyr Tydvil. Bradford is excluded as being one of the sixteen " great cities." ''The following are the statistics of imports and exports in million pounds sterling : Imports. Exports. 1785 14.27 13.66 1795 20.10 22.23 1810 39.30 43-57 1825 44-21 56.32 1830 46.30 69.70 1835 49-03 91-16 1840 67.49 116.48 1845 85-30 150-88 1850 100.47 197-31 1855 123.60 116.70 i860 210.53 164.52 These figures are not to be depended on absolutely; and in 1854 the method of valuation was changed, which probably had some effect. Up to 1800 the data refer to Great Britain; thereafter to the United Kingdom. But making allow- ance for all uncertainties, it would still appear that a great expansion of trade be- gan about 1835-40. Between 1835 ^°^ ^850 the imports doubled in value, and the exports more than doubled; and this was in a period of falling prices. The repeal of the Com Laws and adoption of a free trade policy in 1846, favored a larger foreign trade.. 56 THE GROWTH OF CITIES 6,600.^ The census of 185 1 is therefore the first one to show the effects of the new transportation methods on the distri- bution of population. Contemporaneously with the begin- nings of railway enterprise occurred the establishment of the iron industry, the leading events of which were the substitu- tion of hot for cold blast in 1829 and the adoption of raw coal in place of coke in 1833^, so that it was not much before 1840 that the production of iron assumed large di- mensions.3 The conclusion is therefore unavoidable that it was the opening of railways, with the concomitant development of the iron industry, and expansion of domestic and foreign commerce under free trade, which occasioned the great con- centration of population in the decade 1 841 -51, in the sea- ports and iron producing districts. This explains why London, Wolverhampton and Portsmouth, alone of the seventeen great cities attained a higher rate of growth in 1 84 1 -5 1 than in 1 821-31. Wolverhampton was a centre of the StafTordshire iron manufacturing district, and the three other cities were great seaports. It has already been shown that the middle-sized cities whose growth contributed so much to the high degree cf concentration in 1 841-51 were either seaports or iron-centrcc. ^ Hobson, op. cit., 82. * Ibid., 65. * The following are the statistics of the pig iron production in Great Britain and Ireland : 1790 68,000 Tons 1800 158,000 " 1810 305,000 « 1820 400,000 " 1830 700,000 " 1840 1,396,000 " 1850 • 2,250,000 " i860 3,827,000 " *The Census of 1831 classified 212 cities, and computed their rate of growth STATISTICS OF URBAN GROWTH 57 Since 185 1 the process of concentration has sensibly diminished. As already indicated, this is due in part to the overflow of municipal boundaries in the greater cities where a small area was already filled. This movement and the growth of small suburban towns account for the recent tendency of city-growth to center in the smaller cities. Cf. Table XXI. §2. Scotland. — Like England and Wales, Scotland is pre- eminently a manufacturing and commercial country. A mountainous country, with comparatively little arable land, Scotland could not maintain its population with the products of its own soil, and, as in the case of England, the agricul- tural population attained its maximum number some years ago, in 1 86 1, as appears from Table XXIV. Since then the entire addition to the number of inhabi- tants, amounting to the number of one million souls, has found shelter in the towns, where by pursuing commerce or industry it obtains the wealth needed to import food-pro- ducts for its own sustenance. The rural population has lost some 50,000 persons since 1861 and indeed is now fairly equal to the number in 185 1, the earliest year for which we have an official classification of the people into urban (towns of 2,000-f-) and rural populations. But the tendency toward concentration may be perceived at a much earlier date; during the half century 1801-51. It is interesting to note that the highest rate was that of the watering places, as appears below {op. cit., vol. i, p, xlix) : Table XXIII. No. Class. Increase per cent., 1801-51. 15 Watering places 254.1 51 Manufacturing places 224.2 28 Mining and hardware places 217.3 26 Seaports (exc. London) 195-6 I London 146.4 99 County towns (exc. London) 122.1 212 Cities 176.1 58 THE GROWTH OF CITIES MvO NOOOO •^OMi r^'* •Tj-oo lO ro i-c r~ M O '^^ od" lo ►h' •? c5"co cT cT I'i >^ O O 0\vO N 00 vo vo to N vS 00 O fO^OO o f^ t^ O ►H'^H'tTcTcTcTrOfOrO'^ CO O Tj- w 0\ vo •-' fO t^ <0 i-r>o 00 00 •^ OXTf Tf N 00 t~ w O "^ 0\ 0\ rr> t^oo Cj^ On i-i "^ O to > 5 N rOOO t-« rj- >- r^ N rO i-i Tj- T^oo M u-irorovO "^oo VO r^ to to o^ "^^"^ Q, '^ prt^cToo'Hroo"t^'^i-»"N t^kO O tOOO N ■* >-" Tl- 11 c< to uivo t^ On 0_ to r^ 0_ O 00 l^ t^OO O NO 00 t^vO \rv-0 to M C) On N N tJ-OO oooooo>^o_toTj'^t^q_ 00 ti^vO^t~»0NVO"^^ ^NO OOOt^NffOt^O M M M M N O to ^0vO^^t^O^O^l-lVO OnO H M M C4 § N M 00 NO o N to N On ^^ On vonO no •. t^ t^ '^'^ S, o" N CnnO O \r ►H tooo U o N 1- ■-< 04 M to to M Tj-U-) ir> tr> to ir^NO t>. t^NO M to Jl „-,&*(g-S| .1.5=2 3'>No iS n' « V, s "« -e.s " r'" O ■" NO <— >, S "S NO ».<-; i^T S'o c 3 t* g «5'S:3 ^ S E-0-- bill's ssts^ J .H rb-fi " n c " S S Er^ s Sg-2ji.&a a t- a o o 0,S^ (!.« ;S'G o.S-c ■C-.2 T^ to •^nO On lO N t^ "i- N 00 N 00 OnOO — vo t^ N t^to«-i '00^'-^O^tO tod'ONONONcTfOONCf O "^ N OnOO no On »o O 1-1 M fOtOTfiot^O N wNejNNNtOTf'*' >>6u« S .. «> C — ^ 3 .00 ■" Mt; - . H s E ^ .— .«J<'!t>°a>E'nO< < "SI'S SS-SSS^-SS"^ S'SC?«'3,,„?S— ! «■" i « S>t: ---c n S « H -"S ^ " H o 3t-i o " 5 3 ? »» o-'^!-: « 3 c, S s V u u og-SS' _ -a "".3 . s 8.2--S+Ks-Sv'E.S t^1So5-£a«"f •S&3i8s'^"S"?'=« ii STATISTICS OF URBAN GROWTH 59 Table XXIV indeed shows that whereas the population in towns of 10,000+ doubled in the forty years 1851-91, it nearly tripled in the forty years 181 1-5 1. And there was a long period in which the statisticians of the British govern- ment announced in each census that the " tendency toward agglomeration in towns is even stronger in Scotland than in England." This appears to be true, however, only of the larger cities and especially of Glasgow, as appears from the following comparisons : Table XXV. Percentage of i>opuIation constituted by : 1801. 1851. i8gi. (a) the cities of 10,000+. England 21.3 39.5 61.7 Scotland 1 7.0 32.2 49.9 (b) the towns of 2,000+. England 50.1 72.1 Scotland 51.8 65.4 (c) the metropolis of England 9.7 13,2 14.5 Scotland 5.1 11.5 19.4 From (b) it is clear that since the middle of the century the tendency toward concentration has been far less pro- nounced in Scotland than in England; indeed, in 1851 Scotland had a relatively larger urban population than Eng- land, but by 1 89 1 there was a considerable difference in England's favor. The preceding comparison (a) shows that between 1801 and 185 1 the process of concentration went forward at an equal rate in the two countries ; it is even pos- sible that Scotland's increase is here made to appear too small as a result of the necessity of counting in the popula- tion of an entire parish, or township, in 1801, when only the agglomeration should be reckoned. This would make 17 per cent, an unduly large percentage for 1801. But when the comparison is turned toward London and Glasgow (c), it is easy to realize why the statisticians so 6o THE GROWTH OF CITIES often spoke of the " stronger tendency of the Scotch people toward agglomeration in large towns." lu 1801 the English metropolis coutained nearly one-tenth of the population of England and Wales ; the Scotch metropolis little more than one-twentieth of the population of Scotland. In 185 1 the two were on almost equal footing, and in 1861 Glasgow had taken the lead in the movement toward concentration, so that in 1891 it counted for considerably more in Scotland, in point of numbers, than London did in England. The explanation of Glasgow's exceptional growth lies in the extent and variety of its natural resources. It occupies the position of a great commercial centre like London ; it has a climate peculiarly favorable to the textile industry like Manchester; and it is situated in the midst of a great coal and iron district like Birmingham. The city's growth in numbers is shown in the following table, which also in- cludes the aggregate population of the seven next largest cities (Edinburgh, Dundee, Aberdeen, Leith, Paisley, Greenock and Perth; Govan, a suburb of Glasgow, being excluded) : Table XXVI. Glasgow. Seven Cities. Parliamentary. City and Suburbs. Ratio to pop- Percent- Popula- Percentage Popula- Percentage ulation of Popula- age in- tion. increase. tion. increase. Scotland. tion. crease. 1801 77.058 .... 81,048 .... 5.1 194,428 .... 1811 103,224 34 108,788 34.2 6.0 231,965 19.3 1821 140,432 36 147,043 35-2 7'0 290,316 as.l 1831 193,030 37.6 202,426 37.7 8.6 356,981 23.0 1841 261,004 35.2 274,533 3S.6 10.0 396.930 i3t.a 1851 329,097 26 344,986 25.7 ii.s 450,601 13.5 1861 394,864 20 436,432 23.6 14.2 480,725 6.7 1871 477,156 20.8 547,538 25.5 16.3 579,315 20.S 1881 487,985 2.3 674,09s 23.1 18.3 680,097 17-4 1891 564,981 15-7 782,445 16.1 19.4 763,012 i2.a Sources. — Census of Great Britain, jBsj, vol. i, p. cxxvii, and tables of Section II, for 1801-51; for 1861-1881, the Parliamentary Return 0/18S3, op. cit.; Census of Scotland, i8<)l , 172. The population of suburban Glasgow as given here for 1891 is based on the area assumed by the census in i88r. The parliamentary limits serve for the seven cities throughout. STATISTICS OF URBAN GROWTH 6 1 In the earlier half of the century Glasgow's rate of growth closely agrees with Lancashire's (Table XXII supra) ^ both being due to the development of the cotton trade after 1817. But since 1841, Glasgow has felt the influence of a vast ex- pansion of the iron industry % and of commerce also ^, so that the falling off in the rate of increase is much less marked than it is in the case of Lancashire. Indeed, when com- pared with the rate of increase for the entire country, whose population began to be seriously diminished by the emigra- tion in 18503, Glasgow is shown to have exercised greater powers of attraction in 1851-61 than in any other decade. This is brought out in the following table, constructed on the same lines with Table XX, so as to permit of comparison with the English cities : ' The production of pig iron in Scotland, according to the Encyclopoedia Britan- nica, Art, " Scotland," was as follows : 1796 18,640 tons 1830 37.500 " 1840 241,000 " 1845 475.000 " 1865 1,164,000 " 1870 1,206,000 " (Maximum) 1884 988,000 " In 1859 Scotland produced one-third of the entire British output. The in- fluence of railway building may be noted here as in the case of England. ' According to the same article, the total exports and imports of Scotland amounted to about eleven million pounds sterling in 1825, fourteen million in 1 85 1, and fifty million in 1874. * Before 1850, the number of Scotch immigrants to the United States did not, in any single decade, reach 4,000 (in 1841-50 it was 3,712); but in 1851-60, the number leaped to 38,331. 62 THE GROWTH OF CITIES Table XXVII. Decennial increase per Proportionate Decennial Increase cent. of the popula- Scot- Seven Glasgow, in- The eig tion of Scotland. land. cities. cluding suburbs. cities.' 1801-II .. . .. 12.27 100 158 278 191 181I-2I .. .. 15.82 100 158 222 180 182I-31 .. ,.. 13.04 100 176 289 211 1831-41 .. .. 10.82 100 103 329 181 I 841-5 I .. . .. 10.25 100 132 249 181 1851-61 .. . . . 6.00 100 112 340 205 1861-71 .. . . . 9.72 100 ill 263 212 1871-81 .. ... II. 18 100 148 207 170 1881-91 .. ... 7.77 100 157 208 167 Of the eight cities here considered, only four (Glasgow, Edinburgh, Dundee, and Aberdeen) have a population in excess of 100,000. But the others exceed 60,000 except Perth (30,000). The influence of Glasgow on the aggregate is seen particularly in the decade 1 851-61. The maximum in 1861-71 is to be explained by the vast increase of com- merce consequent upon the annihilation of the American carrying-trade in the Civil War ; Glasgow, it may be surmised, here lagged behind the other cities chiefly because of the cotton famine, which did not affect them. In general there is an agreement with the facts brought out in Table XX for the English cities, except that the falling off in the rate of concentration occurs later. This is undoubtedly due to the fact that the boundaries of the Scotch cities except Glasgow have been so broad that population has not found it necessary to cross the city limits as business encroached upon the resident districts. While in the last two decades the increase of the urban population as a whole (20.2 per cent, in 187 1-8 1 and 14.1 per cent, in 1 88 1-9 1, in towns of 20,000+) exceeded the increase of the ^ In this summary, Glasgow's population is that of the parliamentary borough, except in 1881 and 1891, where the area is that adopted for the parliamentary borough after Nov. i, 1891; the numbers being 577,419 and 658,198 respect- ively {Census of i8gi, i, 149, 152, 155). STATISTICS OF URBAN GROWTH 63 seven cities considered, this is caused partly by the accession of new towns from among the villages and partly by Glas- gow's higher rate (including suburbs). Speaking broadly, the force of attraction among the population centres of Scot- land is proportional to the mass, i. e. in rapidity of growth the order is (i) Glasgow, (2) the other larger cities, (3) urban population, (4) rural population, which is declining. The absolute and relative decrease of the rural population presented in Table XXIV is due almost entirely to a diminu- tion of the agriculturists, caused by deep-seated industrial movements which have brought more fertile land into com- petion with the old-world soil. In Scotland much complaint is made of the enclosure of land in forest preserves, ^ but it is obvious that this is merely the immediate occasion of a re-distribution of population caused by changes in the world's industry. That it is the scattered agricultural popu- lation and not the village population that is declining, is wit- nessed by the following percentages of increase and de- crease : Table XXVIII. 1861-71. 1871-81. 1881-91. Towns (2,0004-) -{- 20.76 +18.20 -f- 14.06 Villages (300-2,000) +I3-90 + 15-73 +4-OI Rural districts (300-) — 7.69 — 3.96 — 5.33 In the last decade the village population failed to hold its own, denoting either an exceptional accession of villages to the rank of towns, or a real falling ofif in the growth of villages. In 1861 the village population constituted ii.i percent, of the population of Scotland; in 1871, 11. 5 per cent; in 1881, 12 per cent; in 1891, only 11. 6 per cent. Taking a wider view of the distribution of population in Scotland, it will be observed that the smaller cities have only maintained their ground while the great increase in tha * Cf. Longstaff's article on "Depopulation " in Y^grvi€% Dictionary of Politi- ■cal Economy. 64 THE GROWTH OF CITIES urban population has been concentrated in the largest cities, thus: Table XXIX. Percentage of population resident in 1801. 1851. iSgi. Glasgow 5.1 11.5 19.4 Other " great cities " o. 5.4 10.4 Cities 20,000- 1 cxj.ocx) 8,8 10.8 12.6 " 10,000-20,000 3.1 4.5 7,5 Total io,ooo-|- 1 7.0 32.2 49.9 Towns 2,000-10,000 19,6 15.5 Total 2,000 -f 51.8 65.4 §3. Ireland. — Ireland is the one country of the Western world that has suffered a decline in population in the present century; but it nevertheless shows the general tendency toward concentration. The following table shows a steady increase down to 1851, followed by as steady a decrease: Table XXX.^ Population of Ireland. Urban Population. 1801 S>2i6,33i 1811 5,956,460 1821 6,801,827 1831 7.767.401 1841 8,196,597 1,143.674 1851 6,574,278 1,226,661 1861 5.798,967 1,140,771 1871 5,412,377 1,201,344 1881 5,174,836 1,245.503 1891 4,704,750 i,244,"3 In the decade 1 841-51 occurred the great potato famine (1846) which carried off by death over 700,000 individuals and caused nearly a million more to depart from the country. Now it appears that this entire loss fell on the •The first census of Ireland was taken in 1821, and the earlier figures are sub- stantially estimates. The later figures agree with those given in the latest censli* (1891, pt. ii, Gene7-al Report, p. 107), except in the years 1841 and 1851, which here include the military, and therefore exceed those of the 1891 census by about 20,000. The urban population in the last three or four censuses has included the population of all places exceeding 2,000; the data have been found in the varioot census reports, detailed reference to which is unnecessary. STATISTICS OF URBAN GROWTH 65 rural districts, for the towns not only did not lose in popula- tion, but made gains in many cases, amounting in the aggre- gate to 83,000. Further, the Irish population has been de- creasing ever since the famine, but the urban population remains practically stationary ; the result being that we have in Ireland the universal phenomenon of relative growth of the cities (as compared with the country). Thus the pro- portion of urban and rural population was : 1841. 1851. 1861. 1871. 1881. 1891 Urban.. 13.9 18.7 19.7 22.2 24.1 26.4 Rural 86.1 81,3 80.3 77.8 75.9 73.6 In the last half-century the urban population has nearly doubled its proportion of the aggregate, but it is to be noted that this increase is not general among all the places counted as " urban ■' by the census. The villages and small towns have, in fact, decayed in sympathy with the emigra- tion of the rural population ; but this loss is fully made up by the rapid growth of Dublin and Belfast. Comparing the aggregate population of the nineteen cities of 10,000 and upwards in 1841 and 1891, it will be found that they gained about 237,000; but Dublin (inclusive of suburbs) and Bel- fast in the same period gained 257,000, thus showing a net loss for the remaining urban population."^ Broadly speak- ^The following table, based on statistics furnished by Dr. Longstaff in his article on "Rural Depopulation" (y. of St. Soc, 58:429), includes the cities which had a population of io,ooo-{- in either 1841 or 1891 : Table XXXI. Population (ik thousands). 1841. 19 cities (lOjOoo-j-) 671. i Dublin City 232.7 " suburbs 33.6 Total Dublin 266.4 Belfast 75.3 Dublin and Belfast 341-7 4 cities with 20,000 to 100,000 in 1891 167.5 13 remaining cities having less than 20,000 161.9 I89I. Increase or Decrease per cent. 908.5 +35 245- +5 97-3 + 189 342.3 +28 255-9 +240 598-3 + 75 166.5 — 0.6 143.6 — 12. 66 THE GROWTH OF CITIES ing, the growth of population in Ireland is confined to Bel- fast and the suburbs of Dublin ; Dublin itself attained its maximum numbers in 185 1 and has since steadily declined. Even in the most recent decade no other cities showed any substantial progress, although Londonderry, Lisburn and Lurgan and Dundalk increased slightly.'' Hence in Table XXXII the smaller cities are shown to have lost population in 1 881-91 ; the class of cities between 20,000 and 100,000 betrays a small gain, which, however, is due to the inclusion of the larger Dublin suburbs. The gain in Dublin and Bel- fast is 43,000, which would be increased to 62,000 with the inclusion of the former's suburban population. Table XXXII. Cities. 1800. No. Pop. 1851. No. Pop. 1881. No. Pop. 1891. No. Pop. I 167,900 4 186,000 5 64,500 10 418,400 1 261,700 6 313,060 7 96,930 2 457.724 6 217,897 12 149,314 2 500.951 6 218,617 10 124,981 Total, lo,ooo-j- 14 671,690 20 824,935 18 844,549 Sources. This summary is compiled from a variety of sources, the official documents not always being at hand. For 1891, the Census of Ireland, Part II, General Report, p. 327-9. For 1881, the Cen- sus 0/ 1881. For 1851, the Census (?/" /S^y/, supplemented by British Almanac, 1S52, and Harper's Statistical Gazetteer, 1855. For 1800 there are no official statistics. Those used in this table are mainly derived from Hassel, 1809, and relate to the period 1800-09. The actual figures follow: 1800. 1851. 1. Dublin 167,900 261,700 2. Belfast 50,000 99,660 3. Cork 67,000 86,485 4. Limerick 39,000 55.268 5. Waterford 30,000 26,667 6. Galway 12,000 24,697 7. Kilkenny 16,500 20,283 1800. I 85 I. 8. Londonderry 11,000 19,604 9. Drogheda 10,000 16,876 10. Newry 15,000 i3>227 11. Wexford 12,471 12. Clonmel 12,386 13. Sligo ir,4ii 14. Carlow 10,995 ^ Cf . Supan, ix, 49^ STATISTICS OF URBAN GROWTH 67 Probably no other country in Europe exhibits so clearly as does Ireland the influence which modern economic changes have brought about in the distribution of popula- tion. The re-organization of industry founded on interna- tional specialization has almost destroyed Irish agriculture and depopulated the rural districts and small towns of Ireland ; on the other hand it has built up the one industry wherein Irish producers work advantageously — the linen manufacture — and has moreover concentrated this industry in one city. The result is that the growth of Belfast in this century is equaled by very few cities in Europe ; while Dublin, a manufacturing and commercial city, and London- derry, a seaport, are the only other cities of importance that have grown at all since 1841. III. FRANCE. French statistics of urban population are exceptionally in- structive for two reasons : ( i ) they have been kept for a longer period and with greater scientific accuracy than those of other European nations, and (2 ) they concern the one great country of the nineteenth century whose population has reached a stationary condition. Elsewhere in Europe the rural populations, by reason of their continuous increase, produce a surplus which must migrate either to the cities or to foreign lands ; but in France this rural excess has been very small. That the cities of France have nevertheless enjoyed a rapid growth demonstrates the existence of world- wide influences upon the distribution of population. Since 1846 the French census reports have grouped to- gether as urban communes all the communes which contain a population of 2,000 or more, living in contiguous houses. ^ A commune might not contain any one community of 1,000 or even 500 people; but so long as it included within ^ An urban commune — " celle dont la population agglomeree depasse 2,000 habitants;" agglomeree — " celle qui se groupe immediatement autour'du clocher." 68 THE GROWTH CF CITIES its borders 2,000 persons living in communities or groups and not entirely scattered, it would be classed among the urban communes, and its total population — the scattered as well as the agglomerated — would be added to the urban population. Table XXXIII.i 1846 185 1 1856 1 861 1866 1872 1876 1881 1886 1891 Population. France. 35,400,486'^ 36,139.364 37.386,313 38,067,064 36,102,921* 36,905,788 37,672,048 38,218,903 38,343.192 Rural. 26,753.743 26,647,711 26,244,536 26,596,547 26,471,716 24,868,022 24,928,392 24.575,506 24,452,395 24,031,900 Urban. 8,646,743 9,135.459 9,844,828 10,789,766 ".595.348 11,234,899 ",977,396 13,096,542 13.766,508 14,311,292 Percentage of urban in total population. 24.4 25-5 27.3 28.9 30-5 31-1 32.4 34.8 35-9 37-4 The table shows the rapid increase in the proportion of the French population classed as urban, starting from 24.4 per cent, in 1846 and reaching 37.4 per cent, in 1891. This is an increase of 50 per cent., which is as much as England's increase in the same period. It is calculated that in 1920 one-half of the population of France will be urban.* In France, the decline of the rural population is not only relative but absolute, thus being in effect a real rural de- population, except in so far as the rural loss is caused by the growth of villages into towns which thereby pass from the rural to the urban group. In 1846 the rural population ' Risultats staiistiques du dinombrenient de i8gj. Paris, 1894, p. 32 and p. 65. 'As finally corrected, 35,401,761. 'Decrease due to loss of Alsace-Lorraine in the war of 1870-1. * Op. ciL, p. 66. STATISTICS OF URBAN GROWTH 69 amounted to 26,753,743 ; in 1891 to 24,031,900. The de- crease is noticeable in all but two inter-censal periods, 1856-61 and 1872-76. The former exception is accounted for by the annexation of Nice and Savoy in 1859, with a population of about 669,000. The latter increase, which is only 60,000, is very probably due to the dislocation of popu- lation in 1872 (concentration of troops), as a result of the war with Germany. Since 1876, the rural depopulation has continued unceasingly and has given the French sociologists a real problem to investigate. That the growth of the urban population and the decline of the rural population are not caused by differences in the natural movement of population is clearly shown by the vital statistics :^ Table XXXIV. 1881-86. 1886-91. Rural. Urban. Rural. Urban. Births 2,587,437 1,543,112 2,609,250 1,692,719 Deaths 2,254,994 1,499,447 2,417,506 1,693,848 Natural increase .. -1-332,443 +43,665 -1-191,744 —1,129 Total increase or decrease — 123,111 +669,966 — 393,479 +517,768 Gain or loss by mi- gration —455.554 +626,301 —585,223 +518,897 In the five-year period 1886-91, the urban population in- creased by 544,784 persons. Of this increase 27,014 were added by the growth of rural into urban communes ; but all of the remainder (517,768) and 1,129 persons in addition (the deficit of births) came from the rural districts which had an emigratfon of 585,223. This number is thrice as large as the excess of births over deaths. The increase of the urban population is due, not to a high birth-rate, but to 1 Op. ciL, p. 72. The influence of villages growing into cities is here dis- counted by considering the same number of cities in 1886 and 1891, thus giving different results from those in Table CXX. 7© THE GROWTH OF CITIES migration from the country. Perhaps this rural depopula- tion is most forcibly recalled by the fact that the number of small communes has been constantly increasing. As people leave the rural villages, these latter sink into the grade of hamlets ; thus the number of communes containing less than 500 inhabitants has increased as follows : 1851 15,684 1861 16,547 1876 16,442' 1891 17,590 1896 18,054 A question of interest and importance now arises : Is the urban increase a village growth or a large-city growth? And Table XXXV gives the answer. Throughout the century the proportion of Frenchmen residing in villages (2,000-10,000) has scarcely varied from 1 1 per cent. Nor has the proportion in the smaller cities (10,000-20,000) appreciably increased. The middle-sized cities, and more particularly the great cities, are the places that have really gained from the re-distribution of the population. It is also interesting to find that the class of large cities increases not only by the continual recruitment from below but also by its own rapidity of growth. Thus, Levasseur tabulated for 1801 the cities which in 1881 had a population of at least 20,000.^* His results may be arranged thus : Increase per cent. 1801-1881. France 40. Cities of 20,0004- (excluding Paris) 145. 9 cities of 100,000+ (excluding Paris) 185. Paris 314- ^ 1,689 communes lost in the war with Prussia. * Les Populations O'rbaines, pages II-14; reproduced in La Population Fran- (aise. STATISTICS OF URBAN GROWTH ri ■o 1 o n fi" « 8 il 3 ^ * <^ n » y ** I? ooosoooog td ^ m w o s a ~ o a ESft •» •* 3 R-o W C 8 ^ 5-^ » ih. ♦^'^ ^ *< ^ g 8 . "V B • S- 00 ^ EToi 3. "13 o i« re ^ »-» M X t^ £■<§ HiS. »' ^ • n o- H fT g- (t p X « X X tN ^'JJ H 8 i- 1 — 1 1 — 1 8 1 — 1 \o M OJ N * 1 1 1 1 o 1 1 <^ 00 VO 00 1 — 1 1 — 1 10 1 — 1 >H OJ ■P' OJ H 00 Ol 00 1 1 1 1 o 1—1 M " 1^ o\ OS :2 - 4>> OJ a. -^ oo o* tn Ul " 4^ 00 o OS " OS 00 OJ M 8 4>- OS \o M 00 ^ VO '-' o N -P" ^ '-' 10 g^ ^4 »4 ^ •f» \o M !■ 00 a> •^ Ul NO 00 " O > H H > O I-" Z S O Tl X X •t: 3 f% THE GROWTH OF CITIES Taking the first 80 years of the present century, it is thus clearly demonstrated that the most rapid growth has been in the large cities ; and while in England London's rate of increase was invariably lower than that of the other great cities, or even the middle-sized cities, in France the capital leads the other cities by a wide margin. Table XXXVI will help to carry out the analysis more thoroughly; it is based on the principle of a fixed area or number of cities, which does away with the disturbing element of new cities arising from towns. And as the cities considered throughout are those of the France of to-day, we must take the population of France on its present territory in order to exclude Strasburg and other cities of Alsace- Lorraine. The table relates to four classes of cities. In de- fault of better statistics, the chefs-lietix or chief towns of the several arrondissemenis, which of course remain fixed in number, ^ will represent the smaller cities. To show how misleading it would be in this analysis to classify the cities at each census, column 1 1 has been added, which may be compared with column 9. The latter is concerned through- out with 12 cities, while the former starts with 2 and ends with 12. Obviously, any deductions as to the rate of in- crease of an average large city would be misleading if based on the high rate of column 11. ^In this case no deduction is made for Alsace-Lorraine; but as there was no change until 1 87 1, the only years affected are 1886 and 1891, whose ratios are somewhat lower than they should be. The average population of the chefs-lieux in 1 80 1 was 10,618, and they were graded as follows: 1801. 1836. Under 5,000 population 153 121 From 5,000 to 10,000 124 136 From 10,000 to 20,000 52 65 From 20,000 to 100,000 31 38 Over 100,000 3 3 363 363 STATISTICS OF URBAN GROWTH n 3> '^ S 2 005 Of?P M B 3 o X'' ^ 3Ift 3 '^ !4 S V- "'^s: " 3 ^ alp O D " •O I 3 <> ^ 3 a. 3? •s ftXi - ^ 00 ^2 1 » ^ s ia-f :-ft.3 -.3 5 3' ft. oo» G. *■ 3.*" Ol *-• '■^ » *-\ ^ * "3' "3. ooooooooooooooooooooooooooooooooQO VO vo 00 00^1 ■^ On 0\V« t^^4>.CK>oj M 1-1 O OJOJOJUlOJOJMOJOJOJUtOjOJU) to * ^^pO^pc^ On O CT^^.^ (.n Ji .^i OJ M ii VO . I.M oj M "oN^b ^ +k "00^ vb Oa i^ ><» Cj "bo • M4k i-.va o OvO-&^a 04k 0<^ 00.^ • ■;'VXOnNO«-i! Ui MUj OOOOwVO m.;^ OOtn .fk vO O On . 3.S >il H 2 2 3 .333 On « M 00 On 0j<-" 00 P M 00 vb Co u> VO \0 --I fk 4» f>. 4!>- OJ >i3 ^oj ^ 00 Oi i-i M ONCn ^ vp J1 Oj 4^ "bv^ "b ^ To OOOj to I- O 4i. ONVO O M Ov-f o .v|<.flU> "•^(^ik'w^k'b^ 00"^ Ol ^ lo ^ vo \0 On'O ^-r OJ Oj ^J vO 00<-n Oj +» Ln M 00 00 OJ'^^PP^l^OjJ^^vpOJ^^J OOU) "I 4i. -fk M oovo "Jo -tj "(0 osb IXS^IJI To i^i "b To 1X3 \0 ►HVOf.n OsO'-n O 00-(^ W O O ON 00 i-c OJ w ^J O OO^O -fi. -vl M NO On~.J Oj t/i N N .t» 00 m H 3" oa ^ !n» < <-n-UCo MNO OOOOOn"-" O OnOvO^J^* ONLn OJ 4^ -f>- On OOtn NNO».J<-n<^OJ O^J >-> lO^i oovb ijT "b 00-tj To "^i u) To 00 To ^ oj vb "on-^ OJtnV/i to 0'0«~l4!>-4>' O^iO On to OJ ONOo Ln 4^>»I OtK» ONtO^^ M OnW^ "-I OnOOONONOn *A -P'^k-f^OJOjOJOJOJOOU) to to to w » OJ to tONO^i4^Lr>OJ 00-f^ " "^ T S OoiOONtOtooiio».JO>->OONOOto " n to to 1-4 HH M M IH HH w HH s-O Ol tk> ONtn Ji. to i-t w 2 5* 0^ w i-i^I OONO totn Q ?< 5i* to to ^J NO On Wn OOw 4>. *^ • to • to w M °Q • • -^ (0 f5' • 4^ . On • 00 • ON " + -P^ U> OJ OJ OtOIOtOMWWl-llHl-lh -t ^ „ 3,32. 00 Oni^ I 1NOVO ONi-c^j^jtyi^kOJ 1 tn 00 OOOJ C » 00 ^ OnOj 4>. -^ « I 3 oc K>^ ^ i « OOta to 00*a tn 'jy * M M •m^S js. .&. 4^4>.0 oOJOJCotOiii-ii-ii-ii-cfc ■4 „ Sf JNUI 00 vo NO ^J On4^ ^s^g^-^^ J OOto On.^ Oj Oj >-i "^ >-< I ■) 3000000000C D (^Lnv/1 4a k4k4a.ootOK>IOi-ii-iMi- w M SS89 VO VI to (J 1 to 0^1 0011 0»~ItnOJ t> 1 Q H iS8'' VO tn 11 ^ tvO OOOOMtn Otn^J^vi c \oo g J^ ONVf > c On4^ 1- " t. i V£ >^ INC > w 1. >^ > to ^ .«y.Oo 74 THE GROWTH OF CITIES The table shows at once that the French cities have been growing at a rate proportionate to their size ; Paris leads off and is followed by the twelve large cities, the middle-sized cities, the smaller cities, and finally by France. And it is to be remembered that each of the lower rates gets the benefit of the higher rate, inasmuch as Paris is included among the large cities, the large cities in the class of cities above 20,000 population, and so on. Thus there seems to be no exception to the rule that the larger the city the faster the growth.^ A graphical presentation of the facts of Table XXXVI would show that in the decade 1846-56 the great cities outstripped Paris; but in 1859 Paris annexed all the suburbs as far as the fortifications, and her curve of growth therefore takes a sharp upward bend in 1856-61. Since then Paris has on the whole kept ahead of the great-cities' class, although in certain periods, e. g., 1866—72, she fell somewhat behind. The third class of cities follow at a slower rate, although since 1840 they have distanced the chefs- lieux. The curve for the general population of France drags along at the bottom. Recent figures tend to show that the most rapid growth is in the middle-sized cities, as in England. 3 Thus the follow- ing table shows an increase of ^7 per cent, for all cities of ^ The proof will be strengthened by the following figures giving the sum of all the chefs-lieux and of other cities with more than 10,000 inhabitants (a class midway between the chefs-lieux and cities of 20,000 — columns 7 and 8) :^ 1801 3,894,000 1,000 1836 S.i35»ooo 1.319 1846 6,179,057 1,587 1851 6,406,557 1,645 1861 7.099.975 1.^23 1866 7.769,906 1,995 1886 10,202,000 2,620 1891 11,055,401 2,838 'Cf. Block, op. cit., i, 57, and Levasseur, ii, 343. ' Supra, Table XX. STATISTICS OF URBAN GROWTH 75 more than 10,000 population: the great cities («'. e., those with 100,000+) show precisely the same gain; the small cities (10,000-20,000) fall considerably below, and the middle-sized cities (20,000-100,000) rank somewhat above the average : Table XXXVII. Groups of Communes haviug in- ^ j ,,iio„ ;„ thousands. Pe-«ntage habitants agglomeres. i-v»i.ui«i,.v-ioo,ooo 6 10.77 30,ooo-5o,cxx) 14 10.67 20,000-30,000 20 11.60 10,000-20,000 76 13-77 5,000-10,000 273 6.61 Total 392 10.02 The inferences from these figures are supported by the diagrams of Levasseur and Meuriot, which show a falling off in the rate of increase in the great cities ; but with the in- corporation of their rapidly growing suburbs, which had contributed to the high rate for the smaller cities, the great cities again distanced the small ones. Our figures incon- trovertibly establish the fact that over a long period ( 1 801-81), the French cities have grown in this order: 1. Paris. 2. The great cities. 3. Middle-sized cities (20,000+). And, though the data cannot claim entire accuracy, it is reasonably certain that the small cities rank below the above classes in rate of increase. The connection of urban growth with industrial develop- ment in France is very close and may be easily shown with the aid of the commercial statistics at hand. M. Levasseur has graphically indicated in several diagrams in La Popula- tion Frangaise^ the progress of the great cities in France. Their growth does not fairly begin until 183 1 ; in 185 1 it is redoubled and continued until the war of 1870— i, since which, as a general rule, it has fallen off very considerably. 1 Vol. ii, 347, 348. STATISTICS OF URBAN GROWTH 77 The following percentages, based on Table XXXVI, illus- trate the course of development already portrayed by M. Levasseur : Table XL. Percentage increase of population in France. Twelve great cities. Ratio I. 3. a: I. 1801-21 10.95 21'2 193* 1821-31 6ut3 9-5 148. 1831-41 5-1 17-7 349- 1841-51 4-5 13.9 310- 1851-61 2.7 51.9 1880. 1861-72 i.o 1 1.3 1 130. 1872-81 4.4 18.7 425. 1881-91 1.8 9.7 535. • The vast increase of population in the cities in 1851-61 is in part due to large annexations by Paris (about 500,000 in 1859), Lyons (1852), Lille (1858) and Havre. But if this annexed territory be included in the earlier census there will remain an unprecedented increase in the years 1851-72; thus the corrected rate of increase and the ratio to the rate of entire France are : 1831-41 22.0 432 1841-51 19.0 424 1851-61 33.0 1220 1861-72 lo.o 1000 The Industrial Revolution, the efifects of which on the dis- tribution of population culminated in England in the decade 1 82 1-3 1, began shortly after the Revolution of 1830 in France and was only fairly under way when Stephenson's invention of the railway locomotive revolutionized systems of communication. The full influence of both these revolu- tionary changes in industry did not make itself felt in France until the middle of the century, as appears from statistics re- specting the consumption of coal and cotton, the number of steam engines and the production of iron : 78 THE GROWTH OF CITIES Table XLI. 1789. 1802. 1810. 1811. 1815. 1816. 1817. 1820. 1825. 1830. 1835- 1840. 1845. 1850. »8SS- i860. 1865. 1869. 1870. 1871. 1875. 1880. 1885. 1890. Cotton consumed * (million kilos). 12 13 29 53 Stationary steam engines "^ (horse- power) , 59 123 132 34,350 66,642 180,555 320.447 544.152 863,007 Coal consumed ' (thousand tons). 450 935 864 1,112 1,348 1.994 2,494 3,288 4,257 6.343 7,225 12,294 14,270 18,522 21,400 18,880 24,658 28,846 30,035 32.318 Production of pig and bar iron * (thousand tons). 69 199 266 295 348 439 416 849 898 1,204 1,381 1,448 1.725 1,631 The consumption of cotton is a fair index of the state of the textile industry, and it will be noticed that the first marked increase occurs in the decade 1830-40. The figures for 1850 are hardly representative of the later decade, since industry had not then recovered from the destruction and dislocation produced by the Revolution of 1848; but it is clear that the period 1 840-60 was one of great growth in the cotton trade. Steam-power was evidently utilized to only a small extent in France before 1 840. Its rapid increase in 1 840-60 will be perceived. ' Block, in Lalor's Cyclopedia of Foliiical Science, ii, 274. * A. de Foville, La France Economique, annie 1889, p. 194. 'Foville, o/. «V., 208. * Ibid,, 214, and Block, Statistique de la France, ii, 200. STATISTICS OF URBAN GROWTH 79 The consumption of coal does not take on imposing dimensions before 1825 ; and once more the period 1840-60 (together with 1872-85) is seen to be one of unusually- rapid expansion. Inferences of a similar character may be drawn from the statistics of iron production. Taken all together, they demonstrate the importance of the period 1840-60 in the industrial history of France, Hand in hand with this indus- trial development went a commercial development. A diagram '' representing the total foreign commerce (imports and exports, excluding goods in transitu) would show that steady growth began about 1834 and continued slowly until 1846. But after the February Revolution the curve rises more rapidly and reaches maxima in 1856, 1864 and 1872. The internal commerce passed through similar stages of growth. As is generally known, the era of railway building opened in France several years later than it did in England and the United States ; the legislation of 1 842 may fairly be said to denote a serious purpose on the part of the country to develop the new system of transportation.'' Hence, it ^ Based on the statistics given by Block, op. cit, ii, 286, and Foville, op.cii, 271-2. ^QA.'R^^^y, Railroad Transportation, 190. The following table (compiled from Block, ii, 322, and Foville, 307) indicates its subsequent progress : Railways Total length of rail- Railways Total length of rail- opened, way in operation. opened. way in operation. (Kilometers). (Kilometers). 1835 149 1853 190 1842 27 593 1854 596 1843 229 1855 886 5,527 1844 2 1856 664 1845 52 1857 1,262 1846 438 1858 1,222 1847 509 1859 - 393 1848 390 i860 365 9,433 X849 628 1865 515 13,585 1850 151 3,002 187s 21,770 1851 544 1880 26,190 1852 316 1885 32,497 8o THE GROWTH OF CITIES was not until the middle of the century that the changes in the distribution of population began on a large scale in France, while they took place in the two Anglo-Saxon countries in 1840-50. The merchandise trafific of the French railways amounted to 200,000 tons (metric) in 1847; in 1854, it had reached a million tons; in 1857, two million tons; in i860, three million tons, and thereafter in- creased at the rate of about one million tons each three years, aggregating in 1883, eleven million tons, and since then increasing very slowly/ The connection between industry and distribution of population is thus seen to be too direct in France to demand extended exposition. IV. GERMANY. The German Empire has done much for the improvement of statistics, and its own statistical publications are of the highest rank. But as the Empire itself is a comparatively recent creation, its statistics do not suffice for a study of the concentration of population in its historical development. For the greater part of the century, therefore, the investiga- tion will be based on the statistics of Prussia, the predomi- nating and typical state of the Empire. Saxony is more advanced industrially than Prussia, but is too small to serve as the typical German state. Table XLII shows the urban and rural populations at the several censuses. The definition of urban population, it should be remarked, is here a legal rather than a statistical one, and includes only such places as possess certain rights and privileges. The legal distinction between Stadtgemeinden and Landgemeinden had a real basis in fact earlier in the ^ Album de statisiique graphique de iSgj (published by the Bureau de la Statii* tique generale de la France), Table 20. STATISTICS OF URBAN GROWTH gl century, but it is now lost. Many Landgemeinden have be- come cities in every sense of the word except the legal one. Thus in 1895, according to the preliminary census returns, there were 52 towns classed in the rural population although they contained a population of more than 10,000 in each case, some of them being large cities. The distribution of the rural population according to the size of the dwelling place was as follows : Places containing No. Aggregate population. 50,ooo-(- inhabitants 2 122,615 20,000 to 50,000 " 6 172,788 10,000 to 20,000 " 44 598,413 5,000-10,000 " 140 922,038 2,000-5,000 " 700 2,048,191 Total 2,000+ 892 3,864,045 Less than 2,000 5i>570 i5>03Ij976 18,896,021 Here is a population of 3,864,045 which should be reck- oned as urban ; on the other hand, there were 317 urban Gemeinden containing less than 2,000 inhabitants with an aggregate of 435,761. If the population be distributed on the usual basis of division we shall have this contrast : Urban. Per cent. Rural. Per cent. Legal 12,953,774 = 40-66 18,896,021 = 59.34 Statistical 16,382,058 = 51.41 15.467,737 = 48.59 Prussia's population is now more than half urban accord- ing to the statistical definition, whereas only 40.7 per cent, is urban according to the legal distinction. But in 1867, the earliest year in which the statistical distinction between urban and rural population can be drawn, the difference was 82 THE GROWTH OF CITIES Table XLII. growth of population in prussia. Cities of 100,000 + No. Pop. Per cent. No. AU Cities. Pop. Per cent. Rural. Pop. Per cent. Total Pop. of Prussia. 1816 2 2 2 3 3 3 3 4 4 6 7 12 16 18 197,817= i.gi 257,336= 1.90 265,394= 1.88 311,491= 2.09 333,990= 2.16 495,995= 3-o8 505.376= 3-09 535,990= 3-i6 648,415= 3.78 677,443= 3-84 776,679= 4.28 883,377= 4-6i 1,105,831= 4.61 1,275,663= 5.18 1,673,728= 6.50 2,049,136= 7.51 2,880,293^=10.17 3,979,886=13.29 4,632,731=14.55 935 972 972 973 979 980 980 981 986 987 992 994 1273 1290 1286 1287 1287 1263 1266 2,627,655=25.46 3,464,587=25.64 3,639,983=25.82 3,861,017=25.87 4,059,840=26.25 4,308,065=26.73 4,370,863=26.76 4,438.377=27-38 4,750,144=27-71 5,038,812=28.53 5,351,219=29.41 5,717,586=29.84 7,456,160=31.10 8,000,931=32.46 8,780,267=34.18 9,707,802—35.58 10,554,596=37-27 11,786,061=39.38 12,953,774=40.66 7,438,460=73.50 9,825,256=72.73 10,244,353=72.67 10,863,337=72.77 ii,2o8,376=:72.45 1 1 ,6o3,990=;72.02 11,714,275=71-73 12,120,211=71.57 I2,l8l,39i:=7I.ll 12,436,610=70.41 12,810,719=69.28 13,191,943=68.85 16,515,177=68.90 16,605,253=67.54 16,913,367=65.82 17,571,309=64.42 17.763,874=62.73 18,169,220=60.62 18,896,021=59.34 10,349,031 13,507,999 14,098,125 14,928,503 15,471,084 16,112,938 16,331,187 16,935.420 17,202,013 17.739,913 18,491,220 19.255,139 23.971,337 24,641,539 25.693,634 27,279,111 28,318,470 29,955,281 31,849,79s 1846 1858 I86I 1864 1867 1880 1885 1890 189s Sources. — From 1816 to 1864, inclusive. Dr. H. Schwabe " Ueber die Quellen fur das Wachs- tum der grossen Stadten im preussischen Staate," Berliner Stadt und Gemeinde-Kalendar tind Statistisches Jahrbuchfiir iBby, Erster Jahrgang. From 1871 to 1895, inclusive, Vorldufige Ergebnisse der Volkszdhl-un^ votn s Dezember, l8()S% im Konigreich Preussen ; and Jannasch, " Das Wachsthum und die Concentration der Bevolkerung des preussischen Staates," Zeitschrift des konigltch preussischen statistischen bureaus, 1878, xviii, 263-284. The official Volkszahlungen. Up to 1864, the military population is excluded from the urban and rural classification while appearing in the general total for Prussia; hence the sum of the urban and rural is not equal to the total in the last column. In 1871 there Ytfere 35,355 troops in France, which are similarly treated; i8t6, similarly 29,038; 1849,34,704. While the passing of towns from one category to another by their growth has important effects on the percentages, it plays no considerable part in the rural population ; the latter lost thus only some 17,000 inhabitants in the period 1840-55. The lack of figures antecedent to 1816 is of no consequence, as the fifteen years of war left the country practicsdly where it was in the first year of the century. The figures are not absolutely exact owing to frequent annexations, and to the varying treatment of the military hy the different authorities. In the general total, the aim has been to include the whole population even when some Prussian troops were on foreign soil. The authority mainly relied upon for this was the Statistisches Handbuchfur den preussischen Staat, 1893 (Vol. ii) ._ Another set of figures was given by Dr. Engel, the director of the Prussian bureau of statistics, in the bureau's Zeitschrift for 1861, i, 25: Military Cities. 1816 2,881,533 1822 3,167,933 1831 3,599.63s 1840 4,066,266 1849 4,582,198 1858 5,235,999 Total. in city. m country 10,319,993 150,094 11,664,133 169,960 13.038,970 258,215 14,928,503 205,247 16,296,483 257,385 3,451 17,672,609 195,966 4,273 Country. 7,838,460 8,496,200 9,439,335 10,862,237 11,714,285 12,436,610 In these statements are excluded 29,038 Prussian troops stationed in France in 1816; 34,704 military in 1849; 67,304 persons in 1858, most of whom were residents of the principality of Hohenzollem annexed in 1849. As between this table and Table XLII, the principal difference that exists relates to the urban population, and is due to the inclusion of the military in the former case and its exclusion in the latter. The military constituted the following proportion of every 100 inhabitants in the years 1834. 1837. 1840. 1843. 1846. 1849. 1852. 1855. 1858. 1861. 1864. 1.63 1. 51 1.36 1.30 1.25 1. 51 1.05 1.18 1.06 1. 31 1. 31 STATISTICS OF URBAN GROWTH 83 less important % and it is safe to neglect it altogether for the first half of the century. Down to 1852 the concentration of population can scarcely be observed in Prussia. Table XLII shows, indeed, a very small and gradual increase in the percentage of the popula- tion reckoned as urban, but this would almost disappear if the military were added to the urban population — ^where it mostly belongs.' The figures of Dr. Engel therefore indi- cate that the urban increase was actually slower than the rural increase down to 1843. His figures "= end with 1858, but the ratios for 1895 have been appended: i8i6. 1822. 1831. 1840. 1849. 1858. 1895. Prussia.... i96i 1865-67 332,742 1868-70 339*637 Poor harvests in 1867 and 1868 also had some effect on the movement of pop- ulation. STATISTICS OF URBAN GROWTH gcj While the rate of increase of the whole German population was greatly reduced in 1867-71, Berlin nearly maintained its high rate of 1858-64 and contributed largely to the concen- tration noticeable in Table XLV. The table shows the effects of the crisis of ''j'j in checking agglomeration. The industrial depression following this crisis was even more severe in Germany than in the United States, as it brought to an end the period of unprecedented speculation following the successful termination of the French war and the pay- ment of the French idemnity. All this time emigration was declining '^ as a result of the hard times in the United States after 1873, thus depressing the general rate of increase. Since 1880 Germany has been rapidly advancing along healthy lines in both commerce and manufactures, and her city populations have increased amazingly. Since 1880 the rural population has apparently been declining (Table XLVI), but this is in no sense a rural depopulation. Be- tween 1880 and 1890 the number of towns having at least 2,000 inhabitants increased from 2,707 to 2,891, thus involv- ing a minimum rural decrease of 368,000. But the total rural decrease in 1880-90 was only 328,290. As may be inferred from Tables XLIV and XLV, the tendency toward agglomeration in Germany has been a cen- tralizing one, that is, a concentration in the largest cities. The great cities have increased more rapidly than the urban ^The total immigration to all foreign countries was as follows (^Ibid., i, 1019) : 1871 75.912 1872 128,152 1873 110,438 1874 47.671 1875 32,329 1876 29,644 1877 22,898 1878 25,627 1879 35.888 1880 : 117,097 90 THE GROWTH OF CITIES t^ro O »^ ^ 00 tC ■^f trioo O t;.N 00 -"f Tj- ij- rf •«^ -^ M 00 M t^H u^OO ro N •^ - t^w t^OO N O voro "" SO vO ^ vO vO W N N N N 00 N O l^O\ O t^O 00 fO O^ IT) N t^ Tf "fvO 00 O fO M w w « N fO« ^00 ■'l- 00 lOfovfT t^ vo vO t^ "* 0\ OjvO f^ 'T ON M cf ro >if lO t^oo 00 t^o N »^0 00 N t^ t^ t^ i-< On T^00 N t^N to CO ■«?•«? ^ '^ "* vr> ON fO NO ij- N N r^ 00 ^ w T? iJ 00 N t^voOO 2 o. On .»0 fO . 2 d. M "il-t^ NO O w ro t~ o o. f*JfOTt Tj- ^ IS. •o A 00 00 r^ w w M M O r^ON o # CO ir> t>. t^oo !z; 2 N N N N N 2 . 00 N 6 S !2;2 00 N MJ-«\0 ■I N PI iS 2 CU r^N ON OnOO "S o p. t^oooooo On ■^ ^ B wiOO N vO > t«»oo O '-' ll O 1 • 2 o. N ONO On w M ci N ci fo M M M M l-l On 11 M to • N ON- t^ On ►" NO f^ojoo o VO" C?nOO" lOl'N 00 t^ ij- O fO O fO *^00 On 1-1 ir>Q vr>0 r^ t^oo 00 ON 00 00 00 00 00 :z;2 ^vo t;*-*© N c5 p5 t>i ei VO t^O i-" »^00 On On 11 voo »oo t^r^oo 00 ON 00 00 00 00 00 g *4 g ~ o u Co S-O "^ C « to ^"O ><> C f « « »< ^ 3S *^ - « !«•§ Is* §iS3B t^ 2 <«-2 -o cuo'S ^ i« N V « (fl «5 q. -s. 1 i 1 00 s "o •«f ^ ;« (>. J{ S ^ B i. a « •a 0* 1 § r B av c^ bSb h 0. •J! :« {> •«- ^ u u t « .s ^ -• 3 lO 1 n in ry > 2 < iJCO S sx B <:w « ST A TISTICS OF URBAN GR WTH 9 j population, and Berlin faster than either. This conclusion appears with overwhelming force in the following statistics from the German census of 1885 : Table XLVII. ANNUAL INCREASE PER THOUSAND OF THE MEAN POPULATION: 1867-71. 1871-75. 1875-80. 1880-85. 1867-85. Gemeinden of less than 2,000 in '85. 0.6 i.i 6.0 — 0.2 2.0 " 2,000-5,000.... " " 6.4 12.4 12.4 8.4 9.9 " 5,000-20,000... " " 12.4 25.5 19.7 16.4 18.3 " 20,000-100,000. " " 22.0 30.3 22.9 21.7 23.6 " ioo,ooo-t- " " 26.9 33.2 25.3 24.1 26.6 Germany 5.9 10.0 11.4 7.0 8.6 In this table, the irregularities occasioned by the passing of towns from one group to another between censuses are avoided by classifying all the towns as returned in the census of 1885 and then computing their population at the previous censuses. The result is singularly striking. In each inter- censal period the percentage of growth increases with the size of the agglomeration. If the several rates be compared with Germany's for the entire period 1867-85, the following result will be obtained : Germany 100 Towns of less than 2,000 23 " " 2,000-5,000 115 " "5,000-20,000 213 " " 20,000-100,000 275 " " 100,000 and more 310 And as has already been shown (in Table XLIV), the class of great cities has itself been outstripped by Berlin. In 1819, of the residents of the 25 cities under consideration, one in six was a Berliner, but in 1890, one out of every four. If now we take account of the increase in population in cities not only through the tendency of people to migrate thither, but also through additions to the number of cities, 9'2 THE GROWTH OF CITIES we shall see the concentration of population in cities consid- erably intensified. Table XLVI shows that the smaller towns and cities have held their own in the general increase, while the middle-sized cities (20,000-100,000) made only a moderate gain. Almost the entire increase of population has been absorbed by the great cities. To indicate the development at wider intervals, it is necessary to recur to the data for Prussia. Tables XLII and XLIII yield the following percentages : * Table XLVIII. Percentage of population of Prussia in Towns of 1816. 1849. 1890. ioo,cxx5-|- 1.8 3.1 12.9 20,000-100,000 4.2 4.7 lO.I 10,000-20,000 1.25 2.83 7.0 Total io,ooo-f 7.25 10.63 3°'° 2,000-10,000 19-25 I7'67 18.5 Total 2,000+ 26.50 28.30 48.5 Saxony's urban growth has surpassed that of Prussia, the reason being that Saxony is more exclusively a manufactur- ing state. For Saxony the following summary is presented : * As previously remarked, the legal definition of " Stadtgemeinde " corresponds closely to the statistical definition down to about 1850, and is therefore used in these totals for 1816 and 1849. There are no statistics earlier than 1816, but it has been found that few changes took place between 1800 and 1816, the gains in population having been wiped out in the Napoleonic wars. That the distribution of population in Prussia may be regarded as fairly typical of Germany may be shown by comparing Tables XLVII and XI.VIII : Prussia. Germany. Cities of 100,000+ 12.9 I2.X " "20,000-100,000 lO.I 9.8 " " 2,000-20,000 25.5 25.1 Total 48.5 47.0 STATISTICS OF URBAN GROWTH 93 Table XLIX. 1834. No. Pop. 1849. No. Pop. 1890. No. Pop. I'- 3 792,318 22.6 3 132,072 I 11,279 3 187,219 6 69,667 9 256,886 (92 354,984) 9 259,559 165,461 1,217,338 (1,043,445) 7-4 4-7 4 143,351 (72 281,729) 34-7 76 425,080 1,595,668 loi 611,870 1,894,431 2,260,783 3,502,684 64.53 100. Authorities. 1815. No. Pop. Cities of 100,000+ o " " 20,000-100,000.. 2 88,700 " " 10,000-20,000... I 13,623 Cities of 10,000+ 3 102,300 Towns of 2,000-10,000 (by subtraction) Total 2,000+ Saxony 1,148,802 Statistische Mittheilungen aus dem Konigreich Sachsen, i, 1851, for period 1834-49. " Stadt und Land im Konigreich Sachsen von 1834 bis 1875," in Zeitschrift des K'dnigl, saclis sfaiis. Bureaus, xxii, 296-306 (1876). G. Lommatzch, i?ze Bewegung des Bevolkerungsstandes im Konigreich Sachsen 1871-90. Dresden, 1894. Hauptergebnisse der sachsischen Statistik. A. Bevolkerungsstatistik, 1834-90, in Zeit- schrift, e.\.z., 36:51-63 (1890). Results of the census of 1890 in Zeitschrift, etc., 37: 51 (1891). Cf. also Repertorium der in sammtlichen Puhlicationen des Kong, sacks, siatis. Bureaus von 1831-86 behandelten Gegenstdnden. Bavaria has a smaller urban population, especially as re- gards great cities. The earlier statistics are wanting : Table L. 1818. 1852. 1890. Per Cities. No. Pop. No. Pop. Per cent. No. Pop. cent. ioo,ooo-f- o I 109,574 2.4 2 493.184 8.8 20,000-100,000 4 136,800 5 169,318 3.72 10 393.938 7-0 10,000-20,000 17 240,184 4.7 Total 10,000+ 29 1,127,306 20.5 Urban (2,000+) 1,782,463 31.9 Urban (official) 52 611,122 13.4 52 1,287,704 20.5 Bavaria 3.707.966 • • 4,559,452 5.594.982 Authorities. Beitrdge zur Statistik des KSnigreichs Bayerns, Heftl. (1850); Heftiz (1865); Hefte 28, 31, 32 (1871); Hefte 36, 42 (1875); Heft 46 (1880); Heft 58 (1890). Also Zeitschrift des KSnigl. bdyerischen statist. Bureaus, vols. 8, g, 13, 14, 22, 23, 24. And the Bavarian statistical Jahrbuch, ii (1895). For index of Bavarian official statistical documents cf. Geschichte und Einrichtung der amtlichen Statistik im K'dnigr. Bayern, Miinchen, 1895 (p. 309.^-) • The "official" urban population consists of 52 towns (see note i, p. 94). The census of 1890 showed that the concentration was in the direction of the larger cities : 94 THE GROWTH OF CITIES Table LI. Increase per cent. 1840-52. 1853-71. 1871-80. 1880-qo. 1840-90. Bavaria 4.31 6.42 8.92 5.87 28. Rural 3.32 2.52 5.59 .78 12.71 Urban^ 11.24 3i-6o 25.67 27.41 134.39 12 cities of 20,000 -|- in 1890 12.97 4542 30-03 3440 187.09 Munich 14.55 55.14 19.60 43.28 204.52 V. AUSTRIA. The Austrian census of 1890 is a model work. A com bination of American ingenuity in the way of electrical tabulating machines, and of German thoroughness and com- pleteness in working up the results, produced a statistical document far and away superior to any other census. It is especially valuable for its classification of dwelling-centres, with the presentation of all the essential data in accordance with this classification ; thus encouraging investigations into the structure and composition of town-populations of vary- ing size. At the present time there exists in Austria a strong tendency toward agglomeration. Thus, between 1880 and 1890 the percentage increase was as follows'' in Places of less than 500 inhabitants 4.83 " " 500-2,000 " 2,30 " " 2,000-5,000 " 7.34 " " 5,000-10,000 " 6.93 " " more than 10,000 " 33'06 Austria 7.91 The Austrian population increased 7.91 per cent, in the decade, but none of the towns under 10,000 population reached this rate, while the exceptionally high rate of in- ' Urban : the " unmittelbaren Stadte " 41 in number, and the 1 1 larger towns in the Pfalz. ^Rauchberg, " Der Zug nachder Stadt,"in Statistische Monatschrift (Vienna), xix, 127 (1893). STATISTICS OF URBAN GROWTH 95 >H *■ U 4> .S S £i s fOOCOO^Ot^t-'-. , fOOCO G^ O t^w Ot^H in\o vO moo C-* t^ ^3- a a t^ d^ eT w" M IN. '^^^ m 00 rt rt » t^ o -* oo H O M O CT* O o H 1/ico moo o »o m m m ON "i- ■*»o NO o M M vo « r^ M^vo^oo^oo o oo^ o^ « -^ m ^NOmoNfoo>«NO'^ t-T ««««ciNroeoco>«- m ONVO O bs « "* C) On t^ vo moo NO rooo ->*■ « ,. ll^OCOtN.OOON'^ CO CO CO CO CO -Tl- '^ -^VO l>. t^ CO • • M 00 ■«• m moo CO • • m in\o 00 0> N m : :ll INI II II II II O 1^ • . ►^ tN w M M M CO ' '• M 8 : : " ■>!■ lO §QOho^N«C .^ 1. " R -S 5 "J; '->a.y s^^ § Km ^■§:s-^f »^;~|.>< " ■" ^ ■* P J! H 00 «oo as S - c ^ n 5 V u S ■♦■Sj3 - ^"oo *^ 00*O rt C M 01 J3.S 3<» te TlOO o «'j]_a °° ^ ^ o " M 11 "1 S !3 c c rt -a s- « c o, «sr -.S '5 § ° " ^ ^ u'3 M 3-C O O V « 6, S in-a -g ii c.S H S ta 3 > o ,, O o «~ - 01 3 S-^ {SO o<-2H "2 «« « S; 5 _c -*- 05O0 S,e « P 3; M t: ?^ c H 2 g _ t!^ •o a> 3 §|« « 2 Jug) 96 THE GROWTH OF CITIES crease in the urban population indicates a widespread move- ment from the country to the city. How long this move- ment has been in progress it is not easy to determine, owing to the scantiness of Austrian statistics prior to the recent model census. But from a multiplicity of sources of varying value, Table LII has been constructed. The civil popula- tion in the present territory of the kingdom of Austria is there given at different periods together with some data con- cerning the population of cities. ' The figures for towns of 10,000 population and upwards are not thoroughly trust- worthy guides, but, as the following table will indicate, they are in agreement with other figures which are more accurate although less complete : Table LIII. Proportionate rates of increase. Cities of Vienna and 17 leading Austria. 10,000+. Vienna. suburbs. cities. = 1800-21 . . 100 106 112 1821-37 . . . 100 232 186 1837-43 • • . 100 298 192 1830-40 262 272 1843-46 . . . 100 456 290 1840-50 .... 402 377 1846-57 . . 100 1,071 480 1850-57 .... 403 388 1857-69 . . . 100 206 253 i8e7— 6q. ....... .... "XAX 28s *"j/ "y ••••••• Of 1869-80 . . . 100 563 183 1869-80 •••• 339 298 1880-90 . . . 100 425 167 1880-90 .... 258 318 It is to be observed, first, that in all four classes of urban population in this table the rate of increase, compared with that for the entire country, is low during the earlier half of the century ; but it is constantly increasing and in the period 1846—57 reaches its maximum. A variety of causes con- tributed to this result. The great famine of 1847 and 1848 carried off thousands of Austrians, and, combined with the Hungarian revolt of 1848 and the political troubles at home, resulted in a serious loss of population, as shown by the ^ Chiefly from Rauchberg's article, loc. cit.f xix, I38-4I* ''Cf.TableLII, col. 8. STATISTICS OF URBAN GROWTH 97 census of 1850.^ The famine naturally afifected most seri- ously the scattered population of the rural districts away from the transportation routes, while the shifting of popula- tion caused by military movements doubtless drew additional persons from the rural districts. But both these factors may be discounted, for the table also shows that the years 1850-57, which exclude them, have a maximum rate for Greater Vienna and 17 of the principal cities. It is beyond question the influence of railway building which accounts for the shifting of population made manifest in 1857. The railway era fairly opened in 1840, but for some years brought advantage only, or chiefly, to Vienna. The de- velopment is set forth in the column headed " Vienna and suburbs," which is practically the Vienna created by law December 19, 1890. Compared with the general rate of in- crease for Austria, Vienna attained its highest rate in 1840—50 and 1850-57, since which it has regularly declined. Vienna's high rate in 1840-50 acts decisively on that of the 17 leading cities, but in the last decade they manifest an in- dependent growth, and one larger than Vienna's. This again may be referred to the influence of railways, for it was in the decade 1865-75 that the largest extensions were carried ouf ^ Cf. Table LII, first and last columns; also Rauchberg, Die Bevolkerung Oesterreichs, p. 27, and diagram. "^ The following figures are from Neumann-Spallart's Uebersichten der Weltwirtk- schaft,2Xid Conrad's Hdwbh., iii, 214: Length of railways (kilometers) in Austro-Hungary. 1830 121 1840 475 1845 ^'°58 1850 2,240 1855 2,829 i860 , 5)i6o 1865 6,397 1870 9>76i 1875 16,766 1880 18,476 1885 22,341 1889 26,501 98 THE GROWTH OF CITIES The apparent decrease of agglomeration in 1857-69 must be explained by a visitation of the cholera and the war with Prussia in 1866. Nearly all the cities of Bohemia, where the war was fought, show an unusually low rate of increase in 1857-69.^ The effect of the different modes of reckoning in Table LIII appears in the proportionate rates for 1880-90 of the class of cities of 10,000+ and of a definite number of cities (17). Still, both show a very considerable tendency toward agglomeration in the last decade. But does agglo- meration in Austria mean concentration in a few great cities ? Table LIV. Showing the distribution of population. Number and population of Gemeinden. Proportions of total population. 1800. ^843. 1890. 1800. 1843. 1880. 1890. Austria 12,600,000 47,438 17,073,231 58,891 23,895,413 100. 100. 100. 100. Under 2,000 46,71313,852,766 57,57816,128,205 ... 81. i 70.4 67.5 Over 2,000 725 3,220,465 1,313 7>767,2o8 ... 18.9 29.6 32.5 2,000-10,000 697 2,235,865 1,212 3,977,843 ... 13.1 16.8 16.7 10,000-30,000 . 8,102,000 21 264,054 69 919,106 0.81 1.6 I 9 o $ 3-^ 20,000-100,000. 5,217,000 5 235,606 27 962,836 0.93 1.4 ' '4. 100,000-1- 1,232,000 2 484,942 5 1,907,423 2.63 2.8 4.6 8. AirrHORiTiEs. Compiled chiefly from Rauchberg's article already cited (Table LII.). The figures for 1800 are gathered from various statistical hand-books (Staatenkunde) of the early part of the century. It should be noticed that the figures for 1880-90 refer to the " actually present" (prtsanwesend) instead of the civil population, (cf. Table LII, Explanations.) In 1846 the city republic of Cracow was annexed, adding some 50,000 to the middle-sized cities It will be seen from Table LIV that the urban popu- lation (towns of 2,ooo-|-) increased from 18.9 per cent, of the total population in 1843 to 32.5 per cent, in 1890, and that while the increase was divided among the several classes of cities, a disproportionately large part fell to the " great ^Cf. St. Mon.y-xxm, 138. It is possible that the urban population of 1857 is estimated too high in the tables, for in the census of 1857 the population of towrns is given on the basis of the entire township. If this is so, the rate for 1850-57 would be somewhat reduced, and that for 1S57-69 increased. STATISTICS OF URBAN GROWTH 99 21. 16. 23- 16.1 13.2 i»3X3 2.S" 12. 14. 13- 7- 33. 1.317 2.173 29. to. 25 23. 18. 1,460 2,781 19. 14- 20. 14- 17- 1.340 2,448 22. SI- 43 38. 22. 1,680 4,846 45- 66. 66 66. 26. 3,238 13,260 32. 67. 43 "5. 63. 3.442 6. 5-7 9 2 8. 7-9 1^172 1,508 cities." While, however, the great cities are absorbing an ever increasing proportion of the population, it will be in- teresting to know whether the average great city grows more rapidly than the average small city, — whether Levas- seur's hypothesis regarding the attractive power of cities ac- cording to size holds good of Austria. The following table ^ shows roughly the percentages of increase in several groups of cities classified according to their population in 1 890 : Table LV. Increase from 1,000. 1831-40. 1841-50. 1851-57. 1858-69. 1870-80. 1880-90. 1870-90. 1831-90. Vienna 12.4 Four " great cities " . . . . 15. Four " large cities " 5. Above nine chief cities. . . 12. Eight small cities 16. Vienna suburbs — 16 57. Prague suburbs — 9 Austria 6.2 Throughout the entire sixty years the eight small cities have grown more rapidly than the chief cities ; and of the nine chief cities, the four great cities (those of 100,000+ in 1 890) have been outstripped by the smaller ones. Vienna's growth is exceeded by that of the small cities and middle- sized cities. But account has to be taken of extra-municipal or suburban growth, and the table shows that of the classes of towns, the suburbs of Vienna and Prague have had the most rapid growth. It is therefore necessary to compute the population of the city and suburbs as an industrial unit. In Table LIII, it has already been seen that the enlarged * Based on Table LII. The percentages of 185 1-7 and 1858-69 have, for the sake of comparison, been reduced to ten year periods, .... The suburbs of Prague increased in population from 18,000 in 1840 to 172,000 in 1890 The ratios for Vienna are based on the ancient limits of the city; making the present limits the basis for 1830 and 1869, as well as 1890, the ratios in the last two columns would be 1,550 and 3,728 instead of 1,313 and 2,512. LcFC. lOO THE GROWTH OF CITIES Vienna outstripped the other cities in every decade except 1830-40 and 1880-90.^ A more complete analysis is made in Table LVI. 1857. i8go. Increase per cent. Vienna (present limits) 593.000 1,341,900 126 Eight cities so.ooo-f in 1890 .... 481,000 785,700 63 Twenty cities, 20,000-50,000 in 1890 295,000 634,600 115 Three suburbs of Prague 20,000-)-. 11,000 108,000 873 Total cities of 20,000-)- in 1890. 1,380,000 2,870,200 108 Eight large cities including suburbs 513,000 937>900 83 Eight selected small cities 71,000 181,300 154 These figures show that the eight smaller cities heretofore considered are not quite typical of their class, for in the period 1857-90 they increased 154 per cent., and the class 20,000-50,000 (averaging about the same in population) only 115 per cent. The rank of the cities as regards growth is now: Vienna, 126; middle-sized cities, 115; large cities, 83 ; Austria 30. In the last decade, as we have seen in Table LV, the four great cities in addition to Vienna increased at a somewhat higher rate than the eight small cities which in turn have grown more rapidly than their class ; so that the present tendency in Austria seems to be toward centraliza- tion, although not so marked as in France. VI. HUNGARY. In Hungary the mediaeval distinction between town and country, based chiefly on political status, still holds good, and the statistical distinction would in any event be invali- dated for the smaller classes of towns on account of their unusual extent of territory. In Hungary the township or primary political unit has an average area of 22 square • The diminished rate of increase in the Vienna suburbs in 1880-90 is due to the spread of population into new territory; and indeed these figures include only about two-thirds of all the suburban inhabitants taken into the city in 1890. STATISTICS OF URBAN GROWTH lOI |JL2 O O Q ^O u-ivO >- 00 i^ N Oft ^ oov5ioo\T)-u->a\ . . .T^^T w 1-1 i-i N fO •* .Svx O O O fO M t>« fOO ■^ '^ "^^^ t '^ ^ ^ O" ■>? od' «>r i^ CO pT r^ C) i/^ O OnnO vO ON'O N N ro ro ■^ tovo 00 ■* ^ i-i •* N t^t^ .S 3)2 2 c 3 w 00 00 t^ O «^ t^ N tJ- 1-1 N ro CT\ " i-H d" ■^oo" J>i rf •* n^vO O 00 d" CT 1-^ •- N t^OO •"^ fO t^ Tfoo N vni-i U-) ro \ri\0 fO •* •*• "* O 11 r^oo ON o 1-1 OS t^ f^ N r^ O w ON N vO vO -"f t^OO O cf^ooNo" fT i-T ^O 00 rf w ro vO t^ ON >- fO 8 O t^ rrj VO N Tj- O 5 t^OO Tt N ON ^ '^ ^< '^ " ^^ "t ONCTO'^r^'-odr^ VO Tf VO io\0 NO « fO oq_roi*voo "^*^'-' d^i-TiJiH'errof^vo •a-g CI & bin 3 o ro fO t>.N 00 VOHH N O ON vovo ro >^ f^ MOO ti cT On ONvO w ^ ^ ro fO VO VO t^ O fO N On OnOO VO N Tl-00 f^ fO d\ o\ -«i-N w d r>.oo NO 11 '^ ^ ^ ^ 'i- W Ht d I O •-■ O »^OnO o I M fO VO lovo 00 On I 00 00 00 00 00 00 00 O "-* O *^ On O M ro VO iO<0 00 A i i i ■ Q M fO lO VO' _ ^ 00 00 00 00 00 OQ oq ;J o § ft^ yis « « ^!^ £ "^ Sj'-a s o »!^a •C s^s - „ J, O h; !^ 3h5i S g « § g g M .= "^ •111 'O 3 in " g 8" j: a J, « a ^ . rt 3 8 • r:« 3 H o e *- 3 " J,V •:= o'M 3 c« w CO ^S M « 3 5,2 •a o 5 « ^ S l-l' £"•5 o ^ 3°; £?S SP3gg II .Su :.S* -S-'o 3=2 3^-^ S 1 c: VO en .3 4>.3fH 3- N w 102 THE GROWTH OF CITIES kilometres, while in Austria it is 1 1 and in Germany 7. Obviously, such a township might contain a population of more than 2,000, which should be so scattered as to deprive it of any urban character. The Hungarian statistics, there- fore, regard as urban the population of 131 towns and cities possessing special political privileges.^ Twenty-five of the cities are known as "towns with municipal charters" {Stddte mit Municipium), and contain in each case upwards of 10,000 inhabitants, their average population being, in 1880, 44,513, and in 1890, 53,243. They appear in Table LVII as " free cities " and will represent the middle-sized cities. The other class of cities embraced in the urban population consists of the towns with magistrate appointed by the crown {Stadte mit geordnetem Magistral), numbering 106 in 1890 and 118 in 1880; their average population in 1880 was 8,743, and in 1890, 10,550. They will represent the class of smaller cities and appear in the table under that heading. These two groups together may be taken as constituting the urban population of Hungary. In the forty years 1850-90, they have increased much more rapidly than the rural remainder. Indeed, in one period (1869-80), which includes years of severe famine, the rural population actually declined. In the other periods its rate was subject to great fluctuations, while the cities progressed at a more uniform rate. As the periods covered by the percentages in Table LVII are not of equal length, it is necessary to compute the percentage increase per annum, thus : Table LVIII. Yearly average increase. Hungary. 131 cities. 1850-57 63 2.075 1858-69 98 1.235 1870-80 II 1.044 1881-90 68 1-447 * The justification of this classification is stated in the census of 1890, Part I., p. 66* (Cf. Table LVII for full title of the census). STATISTICS OF URBAN GROWTH 103 It is here shown that the most rapid urban growth took place in the years 1850-7, as is the case in Austria. And while the figures are not at hand to prove it, it may well be surmised that this period of urban expansion began, as in Austria, in 1840-50. Thus the first real impetus to the growth of the capital city Budapest came in 1840; ^ between 1830 and 1840 its population was stationary, so that be- tween 1840 and 1850 its growth was fully 50 per cent., as may be observed in Table LVII, and this is a higher rate than has since obtained. But Budapest's development then was exceptional, and it apparently had little influence on the rate of growth of the seven leading cities (those now exceed- ing 50,000 population), in which the annual average (geometrical) increase was 1800-20 72 ca 1820-31 ' 1 .96 ca 1831-50 I.2I ca 1850-57 2.446 1857-69 1.617 1869-80 1.878 1880-90 2.239 These figures are not of course entirely trustworthy, but they indicate that the growth of cities, while increasing from year to year, has not reached the high rate attained in 1850-7, which may be attributed in part to political and social causes, but chiefly to the establishment of railway communication between Budapest and the country popula- tion. Considerable interest attaches to the relative rates of in- crease of large and small cities. In Hungary there is not the slightest doubt that the movement has been toward con- centration and centralization. Even France gave no more ' Korosi, Die Hauptstadt Budapest in 1881, Heft I. I04 THE GROWTH OF CITIES Striking example than the following, which indicates the total increase per cent, in 1850-90: 1. Buda-Pest 214-33 2. Seven large cities I20.cx3 3. Free cities 98.78 4. Small cities 50-07 5. Rural districts 25.10 Hungary 30.88 6. Urban (cols. 4 and 5) 73-13 An examination of the table shows that this is the order of growth in every period, without exception, since \%^o. And as the number of cities is in each case nearly constant, the evidence is conclusive that places are growing at a rate proportionate to their size. But the cities of Hungary are still too small for these in- dications of concentration to excite alarm. Budapest, the only city that has reached the 100,000+ class, constitutes but 3.2 per cent, of the Hungarian population. And the entire urban population is relatively small, notwithstanding the inclusion of towns which have a considerable population residing in extended territory under rural rather than urban conditions, as shown by Table LIX. Growth of cities in Hungary. 1800-1808. 1850. 1890. Per Per Per No. Pop. Cent. No. Pop. Cent. No. Pop. Cent. 100,000+ o o I 156,506 1.35 I 491.938 3-23 20,000-100,000. 6 228,000 2.31 12 369,996 3.2 37 1,218,000 8.03 10,000-20,000.. 24 300,000 3,04 .. (525,000 4.5) 67 961,520 6.34 Total 10,000+ 30 528,0005.35 ..(1,050,000 9.1) 1052,671,458 17.6 Total 2,000+ 4357.413.38749- Authorities. As in Table LVII for 1850 and 1890. The cities of 10,000-20,000 are estimated, as it is impossi- ble to distinguish what is urban population in townships the size of those in Hungary. If the urban papulation be defined as that resident in towns of 2,000+, Hungary would have, in 1890, a percentage of 49, which is absurd. The real urban population of Hungary is comprised in the 131 cities of Table LVII, and amounts to 16.1 per cent, of the entire population. The figures above for 1880-08 are derived from various hand-books of " Staatenkunde." STATISTICS OF URBAN GROWTH 105 VII. RUSSIA. The giant-nations that will struggle for world-supremacy in the twentieth century are the United States, the British Empire and Russia. In population, Russia ranks second, and if the vigor of an army depends upon a numerous peasantry, Russia may lay claim to the first place ; for scarcely twelve per cent, of the Russian population dwell in "men-consuming" cities. It is open to doubt, however, whether Russia with its immense numbers of agriculturists, or the United States and Britain, with their centres of enlightenment and business enterprise, would be the more formidable military power. Russia has made more progress in the manufacturing in- dustries than is commonly supposed. It now imports but a small percentage of its woolen and cotton goods, and in all the fundamental industries except iron and steel is nearly as self-containing as the United States.^ The factory industries are concentrated mainly in Poland and the region about Moscow, and in those districts a considerable urban popula- tion is found, which is rapidly growing. Perhaps the most extreme instance is the Polish city of Lodz, the " Manchester of Russia," which in i860 had 31,500 inhabitants and in 1897, 314,780, a growth that would be regarded as phenom- enal even in Artierica. But the factory system has to struggle against many dis- advantages in Russia. Its great competitor, household or domestic industry, has almost entirely succumbed in the countries of the Western world ; but in Russia, the winters "are so long and severe that the agricultural population is largely left without work and obliged to follow other pur- suits. Most of the peasants carry on manufacturing industry in their homes, but in some of the larger villages and pro- 'Cf. The Industries of Russia (5 vols., St. Petersburg, 1893), * series of gov- emmental publications for the World's Fair. Translated by John M. Crawford. I06 ^^^ GROWTH OF CITIES vincial cities, they go into factories, which are thus either shut during the summer or kept in operation with a limited number of workmen. It is estimated that " at least half of the urban population are peasants coming and going, seek- ing and losing occupation." ^ In 1886, the Russian Empire contained 555,990 settle- ments, of which 1,281 had municipal institutions. Their aggregate population is taken as the urban population ; this mediaeval distinction appearing to be substantially equiva- lent to the statistical limit of 2,000 population.^ The pre- liminary results of the census of 18973 show the distribution of the urban population to be as follows : Table LX. Total Percentage of urban population. In towns. pop. in total. Russia proper 94,215,415 11,830,546 12.5 Poland 9.455»943 2,059,340 21.8 Caucasus 9,248,695 996,248 10.8 Siberia 5,727,090 462,182 8.1 Central Asia 7,721,684 932,662 12.1 Total 126,368,827 16,280,978 12.9 Finland 2,527,801 est. [250,100] 10. The Empire 1 28,896,628 1 6,5 3 1 ,000 12.8 Thus it appears that even in the most densely populated province of the Empire (Poland), the urban population con- stitutes only 21.8 per cent, of the total. Considering the fact that serfdom was abolished only in 1 861, and that the era of industrialism scarcely opened be- fore the seventies, it is to be expected that the growth of the urban population has been comparatively recent. The sta- tistics are scanty and not thoroughly trustworthy, for it was not until February 9, 1897, that a general census was taken of ^ Crawford, op. cit., iii, 59. * While over 300 of the towns contained less than 2,000 inhabitants, their aggregate population is almost infinitesimal compared with the total. Cf . Tables LXIII, LXI. » For title, see Table LXI. STATISTICS OF URBAN GROWTH 107 the Empire, in the modern sense of the word. The earlier statistics of population rest on partial enumerations of males for military and fiscal purposes, combined with the records of births and deaths. Nevertheless, these estimates have some value, as is shown by the fact that the estimated popu- lation in 1895 was 129,545,000, as compared with an enum- erated population at the beginning of 1897 of 129,000,000.^ The growth of population as calculated by the Central Commission in the " Revisions " and subsequent estimates, and the aggregate population of towns as computed by the writer from the best authorities, are shown in Table LXI. Year. Authority. Pop. of Russia. Cities. Authority. Percentage in cities. 1722.... ist Revision ' 14,000,000 2,279,412 2,850,926 3,521,052 4.745,632 5,684,000'' 9,064,039 13,947,825 13.972,643 16,280,978 Storch Herman v. Olberg .... Koppen v. Olberg .... Jannasch .... Annuaire .... Kalendar .... P. R 6.5 7- 7- 8.75 8.96 10.66 1796. . . . 1811 1815.... 5th Revision 6th Revision 7th Revision 36,000,000 41,000,000 45,000,000 183s-... 1838.... 1856.... 1870 8th Revision 59,000,000 59,042,866 K 71,243,616 \ 63,862,000'' 85,938.504 \ 85,018,082 s 108,787,235 115,989,443 126,368,827 Official tables I Mitteilungen '82 . . 1885.... 1890 1897.... 12 8 P R i2.g ^ The several " Revisions " may be found in Schnitzler, L' Empire des Tsars, ii, 107. The other authorities mentioned are as follows: Annuaire Statistique de la Russie, 1884-5. Publication du'Comite Central de Statistique, Ministere de I'Interieur. St. Petersburg, 1887 (in Russian and French!. Statistical Tables o/the Russian Empire, 1856. (Pub. in Russian, by the Statistical Central Com., Min. of Interior, St. Petersburg), 1858. German trans, of above by E. v. Olberg, Berlin, 1859. St. Petersburger Kalendar (German I. Statistische und andere wissenschaftliche Mitteilungen aus Russlana, 1868-83. St. Peters- burg. ¥. yon ']Lo^\)en, Russland's Gesamtntbevdlkerungim yahre xZ-i^. H. Storch, Hist, statistische Gemalde des russ. Reiches. Riga-Leipzig, i793-i8o3._ 8 vols. Thaddaus Bulgarius, Russland: Statistik. German trans, by von Brackel. Riga-Leipzig. 1839. Fr. Fred. W. von Reden, Das Kaiserreick Russland. Berlin, Posen and Bromberg, 1843. A. von Buschen, Bevolkerung des russ. Kaiserreichs. Gotha, 1862. Renseignemefits sur la population de Finlande. Helsingfors, 1869. (By the chief of the Bureau of Statistics.) Jannasch, in Zeitsckrift des konigl. preus. statis. Bureaus, 1878, p. 283. Premier Recensement general de la population de F Empire de Russie, 1897. Livraison i, pp. 27-29. The foregoing figures exclude Finland. In Livraison 2, the urban population was corrected to read 16,504,086. ''■ Exclusive of Poland, Finland and Turkestan. *The provisional results of the 1897 census may be found in Statesman's Year Book and Almanac de Gotha, 1898, and J. of St. Soc. 60: 774-5 (Dec, 1897). io8 THE GROWTH OF CITIES The process of agglomeration in, Russia is still slow, even though it has been quickened in the most recent decades, The apparent decline in 1885-90 is, however, likely to be due to some deficiency in the statistics. While Russia has important and rapidly growing " great cities," the rural popu- lation also increases so rapidly as to neutralize urban growth. This will appear from the following statistics of population of the thirteen cities that in 1885 ranked as " great cities " ( 100,- 000-f-) ; the figures, which are of course little better than esti- mates, are based chiefly on the authorities of Table LXI: Table LXII. Percent- Per cent. Percent- Per cent Russian age in- of total French age in- of total Year. Cities. crease. pop. Year. Cities. 1 crease. pop. 1815-25. 827,000 .. 1.8 182I . 1,423,082 .. 4.6 1856.... • 1,583.300 94 2.2 1856.... . 2,494,089 75 6.9 1870 2,383,000 SO 2.8 1872 . 3,531,701 53 9.8 1885.... • 3,541,865 68 3-2 1886.... . 4,190,958 10 12.0 1897.... • 4,641,395 31 3.6 1896.... • 4,793,491 16 12.4 While the Russian cities have grown rapidly, more so than the French, they have not greatly distanced the rural popu- lation, and consequently the part they play in the national life is relatively small. In the period 1885-97 the increase of the great cities was 31, that of Russia 19 per cent., the ratio of the former to the latter being 160 to 100. While this is a somewhat lower rate of concentration than prevails in some other countries, it shows that Russia has entered the circle of " Capitalism." The general result for the cen- tury is given in ^The 12 principal cities (ioo,ooo-|-) in 1891. STATISTICS OF URBAN GROWTH Table LXIII. 109 Towns. 1820. 1856. 1885. No- Pop. ?«„■;, No. Pop. lZ^_ No. Pop. l^^ 100,000+ 3 S9SiOoo 1-4 4 1,123,698 1.6 13 3.541.865 3.2 30,000 — 100,000 17 493,000 i.o 42 1,407,266 1.9 116 4,301,508 4.0 10,000—20,000 45 571,000 1.39 41, 250,178 1.8 164 2,293,344 2.1 Total.. 10,000+.. 65 1,659,000 3.7 140 3,781,142 5.3 293 10,136,717 9.3 a,ooo — 10,000 656 3,215,878 3.0 Total... 2,000+ 949 13,352,59s 13.3 Russia 45,000,000 100 71,200,000 100 108,800,000 100 So far as these figures have value, they indicate that the tendency in Russia is towards the growth of middle-sized cities (20,000-100,000) rather than great cities. But too much weight should not be attached to so small percentages.^ VIII. SWEDEN. The following percentages, being the proportion of the Swedish population residing, at the various censuses, in the 19 cities that had more than 10,000 inhabitants each in 1890, show that the movement toward the cities did not begin until about 1850: 1805 6.3 1810 6.1 1820 6.4 1830 6.4 1840 6.3 1850 6.7 i860 7.6 1870 9.05 1880 10.7 1890 13.7 The same inference may be made from Table LXIV, where the term urban is used in the legal sense, over 40 "towns " having in 1890 less than 2,000 inhabitants. Stock- ^ According to the preliminary results of the census of 1897, Russia contained 19 cities of the first class (100,0004-). with an aggregate population of 5,718,738, or 4.43 per cent, of the total population of the Empire (exclusive of Finland). no THE GROWTH OF CITIES holm since 1850 has been gaining on the rest of the popula- tion. Table LXIV, Proportion No. of Stockholm Year. Population. Rural. Urban. Rural. Urban. Towns. Percentage 1805.. . 2,412,772 2,180,715 232,057 90.39 9.61 86 3.01 1810... • 2,377»85i 2,155,116 222,735 90.63 9.37 86 2.75 1815.. . 2,465,066 2,223,894 241,172 90.18 9.82 .. 2.96 1820.. . 2,584,690 2,330,798 253.892 90.18 9.82 86 2.93 1830... . 2,888,082 2,607,124 280,958 90.27 9-73 2.79 1840.. . 3.138,887 2,835,204 303.683 90.33 9.67 2.68 1850... • 3.482,541 3,131,463 35^078 89.91 10.09 2.67 i860.. • 3.859.728 3,425,209 434,519 88.74 11.26 2.90 1870.. . 4,168,525 3,628,876 539.649 87.05 12.95 3.27 1880.. . 4,565.668 3.875.237 690,431 84.88 15.12 90 3-70 1890... . 4,784,981 3,885,283 899,698 81.20 18.80 92 5-15 It is worth while noting that the urban population failed to maintain its own for several decades early in the century, and that Stockholm itself participated in the relative decline. Since 1840, the urban population has had a rapid growth, as Stockholm has had since 1850. The difference between the legal or official urban popula- tion, and the statistical one is brought out in the following table : 1805. No. Pop. 100,000+ o 30,000-100,000 I 73)652 10,000-20,000 2 23,043 Total 10,000 + 3 95,695 3,000-10,000 30 74,371 Total 2,000+ 23 170,066 " urban (legal) . . 86 232,057 Table LXV. 1 185 D. 1890. Per cent. Per cent. Per cent. of total. No. Pop. of total. No, Pop. of total. 3 351,111 7.3 3.0 3 "9.IS4 3.4 6 167,348 3-S 0.9 3 44,100 1-3 11 142,638 2.9 3-9 5 163,254 4.7 19 661,097 ^3-74 3.1 41 199.358 4-3I 7.0 60 860,455 18. 9.6 92 899,698 18.8 ^ Most of the necessary data for the study uf the subject may be found in the census of 1890; Bidrag till Sveriges officiele Staiistik. Befolknings-Statistik. Nyfolgd I., 2, etc. STATISTICS OF URBAN GROWTH III IX. NORWAY. The urban and rural growth in Norway appears in the fol- lowing decimal rates of increase : ^ Table LXVI. Proportion of Urban in Rural. Urban. Christiania. loo inhabitants. 180I-15 0.3 I.I 12.3 180I 10.7 1815-25 17.7 25.9 52.8 1815 10.7 1825-35 13.7 13.0 32.2 1825 II.3 1835-45 9.9 21.7 35.8 1835 II.3 1845-55 10.5 24.1 25.0 1845 12.3 1855-65 II.6 31.0 39.0 1855 13.6 1865-75 3.2 24.8 36.0 1865 15.6 1875-91 3.1 42.6 96.0'' 1875 18.3 1891 23.7 1801-91 93.4 406.7 1170.0 Table LXVII. Year. Norway. Rural. Urban. Christiania. 1 801 883,038 789*469 93.569 11.923 1815 886,374 791.741 94.633 13.586 1825 1,051,318 932,219 119.099 20,759 1835 1,194,827 1,060,282 134.545 24,445 1845 1,328,471 1,164,745 163,726 33,177 1855... . 1,490,047 1,286,782 203,265 41,266 1865 1,701,756 1,435,464 266,292 57.382 1875 1,813,424 1,481,026 332,398 76,866 1891 2,000,917 1,526,788 474.129 151.239* Thus even in agricultural Norway the cities are growing rapidly, while the rural population has in the last quarter- century come almost to a standstill. The concentration since 1865 has been enormous. That the urban growth is 1 Calculated on statistics given in Statisiik Aarbogfor Kongeriget Norge, 1893. The distinction between urban and rural is still the mediaeval one; in 1801 out of an " urban " population of 93,569, but 28,854 lived in towns of 10,000 or more, and 70,472 in towns of at least 2,000. In 1890, however, out of an urban popu- lation of 474,129, all but 30,570 lived in towns of 2,000 and more. ' Due to annexation of suburbs. 112 THE GROWTH OF CITIES due mainly to the larger towns and cities, the following table will demonstrate : Table LXVIII.i 1801. 1845. 1891. Per Per Per No. Pop. cent. No. Pop. cent. No. Pop. cent. 100,000+ o o I 151,239 7'S6 30,000-100,000 o 2 55.518 4.2 4 124,335 6.22 10,000-20,000 2 28,854 3-27 I 14,778 I.I 5 58,123 2,90 2,000-10,000 10 41,618 4.7 17 68,154 5.1 24 109,862 S.49 Under 2,000 40 23,097 2.7 36 25,276 1.8 27 30,570 1.53 Total urban 52 93,569 10.7 56 163,726 12.3 61 474,129 23.70 2,000+ . 12 70,472 7.9 20 138,450 10.4 34 H43,559 22.17 10,000+ 2 28,854 3.27 3 70,296 5.3 10 333,697 16.68 X. DENMARK. The growth of the various categories of population is dis- played in the following table. In order to show the effect on classification of the modern growth of suburbs, two col- umns are given for the urban and rural classes ; in the first column the urban population includes Copenhagen with all its suburbs and also the Handehplader (seaports) ; in the second column the Handehplader are included in the rural population and also the suburbs of Copenhagen, except Frederiksburg, which is included in " Provincial towns." Table LXIX. Average annual increase in 10,000 of mean population.'' 1834-40. 1850-60. 1870-80. 1880-90. Population 1890. Denmark 84 84 iii iii 102 102 99 99 2,172,380 Rural 85 86 92 97 68 77 21 38 1,434,230 1,508,814 Urban 8i 79 181 162 197 180 273255 738,150 663,566 Copenhagen 51 46 156126 247209 323291 375,719 312,859 Provincial Towns, in 112 201 192 155157 223224 362431 350,707 ^ For index to the Norwegian censuses, cf. Fortegnehe over Norges officielle Statistik mit Flere Statistiske Vaerker, 1828-30. June, 1889. The following have been most used : N'orges officielle Statistik, Aeldre Raekke C, No. I (cen- suses of 1865 and 1875, *"^ *^^° ^^ 1801-25); Folkemaengdens Bevaegelse i Aarene 1856-65. Christiania, 1868-90. ^Cf. Danish censuses of 1890 and 1840. STATISTICS OF URBAN GROWTH 113 From this it appears that early in the century suburban growth was relatively unimportant ; the inclusion or exclu- sion of suburbs does not greatly effect the rural or urban rate, as may be seen by comparing the town columns for the period 1834-40. But in 1880-90, it makes a considerable difference where the suburbs are placed; without them, Copenhagen has an annual increase of 2.91 per cent., with them 3.23. In the period of 1834-40, the rural increase was larger than the urban, which was depressed by the low rate for Copenhagen ; the other towns and cities had the maximum rate. In 1850-60 the urban rate exceeded the rural, but the provincial towns still lead Copenhagen, In 1 870-80 and 1880-90 there has been a great falling off in the rural rate of growth, while the urban rate has been correspondingly increased, and Copenhagen has forged way ahead of the other cities. Additional information necessary to this study is given in Tables LXX, LXXI, LXXII. Table LXX. Urban population. Copenhagen^ Denmark. Actual. ^ of total, per cent, of total. 1801 929,cxx) 194,431 20.9 10.9 1834 1,223,797 251,502 20.5 9.7 1840 1,289,075 266,822 20.7 9.6 i860 1,608,362 381,662 23.7 10.4 1870 1,784,741 450,241 25.2 II.5 1880 1,969,039 563,930 28.6 13.9 1890 2,172,380 738,150 34-0 17.3 Table LXXI. Number of Towns and Proportion of Urban Population in bach Class. 1801. 1840. i860. 1880. 1890. i, f ic ^ ^, Copenhagen i 5I>9 i 46.1 i 44.0 i 48.5 i 50.9 10,000-40,000 . . . . o o o o 3 14.1 5 15.2 7 18.7 2,000-10,000 9 18.5 24 34.8 31 32.6 39 28.8 42 25.6 Under 2,000 62 29.6 46 19,1 39 9.3 29 7.5 23 4.8 Total 72100.0 71 loo.o 74 loo.o 74 loo.o 73 loo.o 114 THE GROWTH OF CITIES Table LXXII. 1801. 1840. 1890. 100,000+ I 100,975 10.9 I 123,123 9.6 I 375.719 17-3 20,000-100,000 . . ... 2 63,574 2.9 10,000-20,000 . . . ... 5 74,528 3-4 Total 1 0,000 -f- . . • I 100,975 10.9 I 123,123 9.6 8 513.821 23.6 Total 2,000 H 10 136,967 14.8 25 215,962 16.8 50 702,679 32.4 Total "Urban". • •• 194.431 20.9 .. 266,822 20.7 •• 738.150 34.0 Authorities. The principal source of information for all the tables is the Danish census of 1890: Danmarks Statistik, Statistik Tabelvaerk, fjerde Raekke, Litra ^, Nr 8a, Hovedresultaterne af Folketaellingen i Kongeriget Danmark den iste Feb., i8go. Udgivet of det statistiske Bureau, Kjobenhavn, 1894. The census of 1840 ( Tabelvaerkf Siette Haefte) has also been used to a considerable extent. XI. THE NETHERLANDS. The tendency toward agglomeration in the Netherlands is first noticeable about 1850; up to that time the rural popu- lation increased more rapidly than the urban. City growth has increased since 1850, and in the last decade was unpre- cedented, as shown in Table LXXIII. Increase per 10,000 by decades.^ Increase of Population from 1829.1 TotaL Urban.2 Rural. Ratio of urban to General. Urban. * Rural. general increase. 1829. . . . 100. 100. 100. 1839. . . . 109.46 109.02 109.60 945 908 962 ii 1849. . .. 116.97 » 15-42 117.50 686 591 728 86 1859. . .. 126.62 126.84 126.54 825 995 751 120 1869. ... 136.97 140.40 135-76 817 947 761 116 1879- ... 153.54 167.52 148.69 1210 1725 979 143 1889. . .. 172.62 209.82 159.73 1240 2400 193 ^ Overzicht von de Uitkomsten bewerkt door de Centrale Commissie voor de Statistik, 1893. ^As urban is regarded the population at each census of the 21 Gemeenten (communes) whose population exceeded 20,000 in 1889. ' Uitkomsten derzerde tienjarige Volkstelling in het Konigrijk der Nederlanden, 31 Dec, 1879. 's. Gravenhage, 1881; Resume statistique pour la Royaume des Pays-Bas 1850-83, published by La Societe de Statistique des Pays Bas, Le Haye, 1884. * As urban is here regarded the population of 34 communes which severally had 10,000 4- population in 1859. STATISTICS OF URBAN GROWTH 1 1 5 The rapid increase in the urban population since 1869 is very largely due to the growth of the great cities, Amster- dam, Rotterdam and The Hague. Hence, in the Netherlands there exists a considerable concentration of the agglomerated population. The changes that have taken place in the dis- tribution of population are shown in Table LXXIV. A dif- ficulty with Dutch urban statistics is that they usually adopt as the unit the township {Gemeente), which includes a con- siderable rural population, owing to its territorial extent. The difference between the township population and the true urban population is brought out in the last two columns of Table LXXIV. Population of the Cities. 1795. 1829. 1849. 1889. 1889. 100,000+ 217,024 202,364 224,035 766,728 750,763 20,000-100,000.. 242,910 324,168 438,980 644,856 565,700 10,000-20,000... 94,986 151,517 221,919 527,899 187,400 Total 10,000+. 554,920 678,049 884,938 1,939.483 1,503.900 Holland 1,880,463 2,613,487 3,056,879 4,511,415 4,474,461 Number of cities and percentages in total population. 100,000+ I II.5 I 7.7 I 7.3 3 17.0 3 16.6 20,000-100,000... 9 13.0 9 12.4 II 14.4 18 14.3 i6 12.7 10,000-20,000... 7 5.0 n 5.9 16 7.3 .. II. 7 15 4.2 Total 10,000 -I — 17 29.5 21 26.0 28 29.0 .. 43.0 ^^ 33.5 Sources. — Compiled from the authorities of Table LXIII, except the last column, which is based on Supan for the cities, and for Holland the figures denote the " actually present" popula- tion. XII. BELGIUM. In Belgium the line between urban and rural population is drawn at 5,000. The growth of the urban population since 1846 is shown in Il6 THE GROWTH OF en lES Table LXXV.' Urban. Belgium. Rural. No. of Per cent, places. PoP-'at'on- of total. 1846 4.337.196 2,921,329 112 1,415,867 32.6 1856 4.529.461 2,952,079 1 18 1,577,382 34.8 1866 4.827,833 3.046460 131 1.781,373 36-9 1880 5.520,009 3,143.232 166 2,376,777 43.1 1890 6,069,321 3.174.627 191 2,894,694 47.7 In Belgium the tendency toward agglomeration has ex- isted from the earliest date of the records, but it is more marked in recent years. The urban population, moreover, is largely concentrated in a few great cities and their sub- urbs. Table LXXVI indicates the importance of the sub- urbs of Brussels and Antwerp in any classification of the urban population. While the totals remain about the same, the distribution between large and small cities is vastly dif- ferent when the suburbs of Brussels, Antwerp and Ghent are separately treated : Table LXXVI. 1802-15. 1846. 1890. Estimates. Exclusive Inclusive Exclusive Inclusive of suburbs. of suburbs. Brussels 66,297 123,874 188,458 176,138 465,517 Antwerp 56,318 88,487 97>94S 224,012 268,397 Ghent 55>i6z 102,297 105,894 148,739 171,927 100,000+ o 2 226,851 2 294,352 4 696,539 •• 1,053,501 20,000-100,000 5 261,408 XI 415,875 II 425,336 25 826,210 .. 525,684 10,000-20,000 II 145,135 18 234,224 14 181,131 44 573,687 .. 526,251 Total 10,000 -\ 16 406,543 31 876,950 27 900,819 73 2,096,436 61 2,105,436 Percentage of population of Belgium. 100,000+ o 5.2 6.8 11.4 17.4 30,000-100,000 8.7 9.6 9.8 13.7 8.7 10,000-30,000' 4.8 5.4 4.2 9.4 8-7 Total 10,000 + 13.5 20.2 20.8 34.5 34.8 Sources — Recensement de i8qo, and Supan. For 1800 the estimates are compiled from Hassel, 1809; the population of Belgium was then estimated at 3,411,082, but it included parts of Luxem- burg and Limburg. afterward added to Holland; without them the population may be roughly estimated at 3,000,000. "^ Statistique de la Belgique, Ricensentetit general du ji Dec., i8go, Tome I.. p. XV. Brussels, 1893. For 1846 and 1856 the population is defait; for 1866, 1880, 1890, de droit. STATISTICS OF URBAN GROWTH 117 XIII. SWITZERLAND. Even in the little agricultural republic there has been in the present century a very noticeable tendency toward ag- glomeration. Compared with the entire population, the ag- gregate population at each census of the fifteen principal towns of 1888 (/. e., those of 1 0,000+ )has increased very rapidly : Switzerland. Percentage in 15 towns. 1850 2,392,740 9.4 i860 2,510,494 II.4 1870 2,655,001 12.8 1880 2,831,787 14.5 1888 2,917,754 16.5 The periods 1850-60 and 1 880-1 888 show large gains for the towns. The urban growth is indeed for the most part in the larger towns, as appears trom Table LXXVII. 1822. 1850. x888. Per Per Per cent. cent. cent. No. Pop. of total. No. Pop. of total. No. Pop. of total. 20,000-100,000.. I 24,600 1.3 4 125,080 5.2 8 384,360 13.2 10,000-20,000... 4 54,300 3.0 4 51,048 2.1 7 96,028 3.3 Total lOjOOoH — 5 78,900 4.3 8 176,128 7.3 15 480,388 16.5 Switzerland 1,855,300 Total 5,000+ •• • II 120,000 6.5 28301,538 12.7 52 726,060 24.7 Sources. — Sckweizerische Statistik, Die Ergebnisse der Eidgetidssischen Volkszahlung vom I Dez. l8S8. (Esp. Lieferungen 84, 88, 97) ; M. Wirth, Statistik der Sckweiz, Zurich, 1871-3; S. Franscini, Neue Statistik der Sckweiz, Bern, 1848; Hassel, 1822. XIV. ITALY. Like Germany, Italy is a new nation with few statistics of the national dominion, and it is very difficult to obtain sta- tistics concerning the urban population prior to 1861. No census has been taken since 1881. In Italy the mass of the population dwells in small towns ; even the agriculturists dwell in villages and go out to their 1 1 8 THE GEO WTH OF CITIES work in the fields. The agglomerated population is there- fore comparatively large. Hence, it is customary in Italy to reckon with the urban population only the centri of 6,000 or more inhabitants ; while the smaller centri, the casali or vil- lages, and the scattered population are all combined in the rural population. The percentages since 1861 were as fol- lows: Table LXXVIII.» 1861. 1871. 1881. Urban ..(Centri of 6,000 -|-) 25.17 24.93 27.02 ( Centri of 6,000 — 42.84 49*37 45-68 "'^ ' \ Scattered pop 31-99 25.70 27.30 100. 100. 100. Italy's urban percentage therefore appears to be about 25, which is also the percentage of population dwelling in capo- luoghi, or head communes of provinces and districts {circon- dari) in 1881 — 7,082, 163 in a total population of 28,459,628. The urban population is apparently increasing but slowly ; in the larger cities however there is a constant growth : Table LXXIX. Agglomerated Population. Towns. 1S71. 1881. Under 2,000 43«59 40.25 2,000-6,000 22.86 22.59 6,000-8,000 5.12 5.30 8,000-20,000 13-13 13-56 20,000-100,000 7.23 8.76 100,000-1- 8.07 9,54 100.00 100.00 And throughout the century there can be traced a tendency to concentrate in the great cities : ' Censimento della Populazione del Regno d'' Italia al ji dicembre l88l ; same 1871 and 1861. STATISTICS OF URBAN GROWTH 119 Table LXXX.i Number and Population of Italian Cities. 1800. 1847-8. Italy 18,124,000 100 23,617,000 100 100,000 + (4 800,000 4.4) (8 1,425,000 6.) 30,000-100,000 .. 10,000-20,000 < Total 10,000+ Total 2,000+ 28,459,628 9 1,974,394 57 1,811,188 149 2,084,806 100 6.9 6.4 7-3 215 5,870,388 ... 12,358,430 20.6 43-43 XV. OTHER EUROPEAN COUNTRIES. The more important countries of Europe have been treated at length, but, in order to complete the study, statis- tical tables setting forth the development of urban popula- tions in the remaining European countries are appended. Trustworthy data are usually lacking for these countries ex- cept in recent years, but the best authorities have been relied upon. Table LXXXI. Spain. 1800-10. 1820 ca. 1857. Increase 1887. Increase 1800-57. 1857-87. Spain 10,836,000 11,411,924 15,464,430 42.5 17,565,632 13.6 15 cities 888,850 1,340,326 50.9 1,928,691 43. 5 great cities 495,332 778,214 56.5 1,190,725 53, Madrid 156,670 167,607 281,170 79.0 470,283 62. 1820 ca. 1857. 1887. 5^ of 5$ of ^ of No. Pop. total. No. Pop. total. No. Pop. total. Total 100,000+ I 167,607 1.45 4 683,921 4.4 5 1,190,725 6.8 20,000-100,000 24 945,270 8.3 23 805,767 5.2 56 1,975,423 112 20,000+ 25 1,112,877 9-75 27 1,489,688 g.6 6i» 3,166,148 18. 10,000-20,000 (estimated) (36 476,530) .... 72 (1,080,000) ... 140 (2,100,000) io,ooo+(estiinated).... 61 (1,600,000 14.) 99 (2,570,000 16.2) 201 (5,200,000 29.6) Note. — The five great cities are Madrid, Barcelona, Valencia, Sevilla, Malaga. The fifteen cities include these five and ten others exceeding 50,000 in population in 1887. The authority for 1800-10 is Hassel 1809; for 1820, Hassel 1823; for 1857, Censo de la Poblacion de Espana, on 21 de Mayo de i8s7, and Kolb, 1868; for 1887, the Census. The statistics relate to the commune or township, which in Spain is exceedingly large and contains a rural population. The estimates for towns 10,000-20,000 are arithmetical (multiplying 15,000 by 72 and 140); but they probably approximate actual conditions. According to the Nomenclater de Espagna, 1888, the urban population was 4,851,903, and the rural 12,713,369, or 27.89 and 72.11 per cent, respectively, * Estimates are enclosed in parent? eses. I20 THE GROWTH OF CITIES Table LXXXII. Portugal. iSoi. 1857 Increase 1878. Increase 1890. Increase per cent. per cent. per cent. Portugal 3,661,809 3,908,861 6.5 4iS5o,699 16.4 5,082,247 11. i Eight cities 115,600 195,600 64.0 308,099 6.2 253,050 12.6 Lisbon 350,000 275,286 21.4a 246,343 10.5a 307,661 12.4 1857 5$ of ^ of 5J of No. Pop. total. No. Pop. total. No. Pop. total. 100,000+ I 350,000 9.5 I 275,286 7.2 2 447,517 8.8 20,000-100,000. . I 30,000 .8 3 140,000 3.5 I 23,089 .4 10,000-20,000... 7 85,600 2.4 6 85,600 2.2 12 178,329 3.5 10,000+ 9 465,600 12.7 10 500,800 12.9 15 648,935 12.7 a = decrease. Note. — The eight cities arc Oporto, Braga, Funchal, Coimbra, Setubal, Evora, Angra, Elvas, being the large cities of 1800. As regards the sources, the data for 1890 are from the Statesman's Year Book, 1897; ^°^ ^^878, from Censo No. 1 de jfaneiro, 1878, Populacao, Lisbon, 1881; for 1801 and 1857 ^^ population of Portugal is given in Block, Bevolkerutig Spaniens und Por- tugals (1861), p. 53, while the population of the cities in 1800 is from Hassel, 1809, and other early hand-books; in 1857, from Kolb, i860, etc. Additional references are Balbi, Essai Statis- tiqrte sur le Roya-unte de Portugal, 2 vols., Paris, 1822; Minutoli, Portugal und seine Colonien tm Jahre 1854, 2 vols., Stuttgart, 1855. The earlier figures are regarded as very inaccurate. Table LXXXIII. Greece. 1852. Pop. i> of Greece 1,002,112 total Athens 3i»i25 3.1 Other cities of 20,000+ Other cities of 10,000- 20,000 Total 10,000+ Sources. — FonSjg, Almanack de Gotha,\%%s'< tl^e figures include parts of Thessaly an- nexed in 1881. For 1889, Supan. For 1852, Kolb, i860. Table LXXXIV. Turkey in Europe. 1885. No. Turkey Constantinople Salonica. ... , Cities 20,000-100,000 4 No. 1879. Pop. i.979ii47 i. No. 1889. Pop. 2,187,208 1'- I 3 63,374 67,794 3.2 3-4 I 3 107,251 89,960 4-9 4.1 4 55,991 2.8 8 109,128 5.0 8 187,159 9-S 12 306,339 14.0 Per cent. Population. of total. 4,786,64s 100 873,565 18.3 150,000 3-1 166,000 3.5 Cities 20,000 -|- 6 1,189,600 24.9 Sources. — Census of 1885, in Statesman's Year Book for 1897, p. 1018. The territory of Tur- key has been so frequently altered that not even the old estimates can be g^ven for comparison. Constantinople is said to have had a population of 597,600 (Hassel, 1833) at the beginning of the century, and 700,000 at its middle (Kolb, i860). It has apparently gained in population in recent years, for so late as 1885 iht Almanack de Gotha credited it with only 6-700,000. STATISTICS OF URBAN GROWTH I2I Table LXXXV. Bosnia and Herzogowina. Census, May 4, 1883. Entire country 1,336,091 Sarajewo 26,268 1.90 Mostar 12,665 '95 Banjaluka iIj357 -95 Cities 10,000 -|- 3 50,290 3.8% Source. — Von Asboth, Bosnia utid Herzogowina, Wien, 1888. Table LXXXVI. Servia. 1800-10. 1854. 1874. 1890. Pop. 5?. Pop. f. Pop. f. No. Pop. !». Servia 960,000 100. 985,000 100. 1,353,890 100. 2,161,961 100. Belgrade 30,000 3.1 16,723 1.7 27,605 2.0 i 54,249 2.5 Cities 2,000-20,000 4 55.812 2.6 Total 10,000 -f- 5 110,061 5.1 " 2,ooo-|- 104 286,46613.35 Source. — Statistigue du Royautne de Serbie, Belgrade, 1892-3. Servian communes are unusually large (38.2 sq. km. on the average), but in this table only the agglomerated population is counted. If the unit taken were the commune instead of the dwelling- centre, the result would be considerably different (1890) : No. Pop. Per cent. Total 10,000 + 7 131,534 6.1 " 2,000 277 787,492 36.S The statistics of 1800-10, from Hassel, 1809, are little more than guesses, but, even when dis- counted, show a large city population. Table LXXXVII. Bulgaria.^ 1850. 1888. Entire country 3»i54.375 loo-o Philoppel 40,000 33,032 i.o Sofia 30,000 30,428 1.0 Other cities 20,ooo-|- 4- 96,504 3.1 Cities 10,000-20,000 15-189,203 6.0 Total 10,000 + 21-349,167 II. I ' Including East Roumelia. The authority is ResuUais du Ricensement de la Population, 1888 (Sofia 1888) and Supan (jOrtssiatistik') p. 73-5; for 1850, Kolb i860. According to estimates at the first of the century, Sofia, the capital, had a population of 46,000 (Hassel 1809), But recently Sofia has grown rapidly and at the census of 1893 ^^^ ^ population of 47,000 (^St. Yr. Bk. 1897, p. 1034). 122 THE GROWTH OF CITIES Table LXXXVIII. Roumania. 1800-10. 1859-60. 1889-90. Roumania i,370,5cx> ... 4424,961 ... .. 5,038,342 ... Bucharest 42,000 3.1 100,000 2.4 .. 220,000 4.4 20,000-100,000 8 316,152 6.3 10,000-20,000 13 174,385 3.5 Total 10,000 -(- 22 710,50014.2 Total "urban" (legal definition) 885,70017.6 Authorities. — For 1800-10, Hassel 1809; for 1859-60 Almanack de Gotha; for 1889-90, Supan, except Bucharest, which was returned at 194,633. This inaccuracy shows that the attempt in 1890 to repair the defects of 1889 were unsuccessful. The census of 1859-60 is also of doubtful value. The earlier figures are, of course, only estimates. It should be added that the unit of city populations is the territorial subdivision. XVI. ASIATIC COUNTRIES. Statistics of Asiatic Russia have already been presented. Additional statistics for this ancient grand division are not worth much space on account of the untrustworthiness of mere estimates. The modern periodical census so familiar in the Western world has scarcely been introduced in Asia. Even progressive Japan has not thoroughly learned the art of numbering the people. Hence the only really valuable data for present purposes are in the English census of India. § I . Asiatic Turkey covers a vast extent of territory and contains numerous cities. But only the larger cities are known to the statisticians, and these imperfectly. The esti- mated population in 1885 : Table LXXXIX.' Population. Per cent. Entire country 21,608,000 100 Cities 100,0004- (4) 700,000 3.2 17 cities, 25,000-100,000 691,000 3.2 21 cities, 25,000-]- 1,391,000 6.4 Asiatic Turkey includes Asia Minor, Armenia and Khun- distan, Mesopotamia, Syria and Arabia. The four great cities are Smyrna (200,000), Damascus (200,000), Bagdad * Statesman's Year Book, 1894. STATISTICS OF URBAN GROWTH 123 (180,000), Aleppo (120,000). No statistics of growth can be given, beyond noting that in 1823 Hassel published these estimates : Damascus, 1 30,000 ; Bagdad, 96,000. The Oriental cities are, it appears, not stationary in population. § 2. Persia. — The latest estimates of the population of Persia are as follows : ^ Table XC. Population. Percentages. Inhabitants of cities 1,963,800 25.6 Wandering tribes 1,909,800 24.9 Villages and country districts 3,780,000 49.5 Total 7,653,600 100.0 Teheran 210,000 % Tabriz 180,000 j ^' Cities 25,000-100,000 (11) 440,000 5.7 §3. British India. — The official statistics at the present time can be compared in accuracy and trustworthiness with those of the Western nations. And they are especially instructive as presenting the distribution of population in a country which is as densely populated as Europe, and has therefore emerged from the barbaric state and even attained a con- siderable degree of civilization. But the industrial organiza- tion of India is totally different from the European. Industry in India is mainly carried on in local, autonomous and self- sufficing communities. Between these communities there is little commerce, for each devotes nearly its entire population to supplying its own wants.^ This is essentially true of the ^St. Yr. Bk., 1897,811. * The Census of i8gi {^General Report, p. 94), shows that nine-tenths of the population are engaged in local industries : Percentage : Total pop. Rural pop. Engaged in primitive occupations 84.84 88.26 Engaged in supplementary (semi-rural) " 5.47 o. Engaged in other " 9.69 ii'74 100 100 124 THE GROWTH OF CITIES rural population, amounting to nearly nine-tenths of the whole: ^ Table XCI. Percentage of urban pop. in Urban population. towns oi Per cent, of 10,000- Under Total pop. Actual, total, pop. 50,000 + 50,000 10,000. Bengal 71,346,987 3,443>876 4-82 44 46 10 Bombay 18,901,123 3,502,678 18.51 43 35 22 Madras 35,630,440 3,406,105 9.56 29 46 25 N.-W. Provinces .. . 46,905,085 5,314,328 11.33 Other Provinces ... . 48,389,317 4,724,142 9.80 Total Provinces. . 221,172,952 20,391,129 9.22 Feudatory States.... 66,050,479 6,860,047 10.38 Total India 287,223,431 27,251,176 9.48 35 36 29 The definition of a town in this table is not a statistical one, as there are many towns included whose population is under 2,000, and some villages excluded whose population exceeds 20,000. The census regards as " towns " all places " established as municipalities or brought under similar regu- lations for police and sanitary purposes," and secondly, all places wherein at least one-half of the population is non- agricultural; in the latter case the numerical standard of 5,000 was prescribed/ As regards the aggregate results, the two "tests" of urban population are tantamount to a statisti- cal limit of 5,000, as will appear later. (Table XCII.) While, then, nine-tenths of the people of British India are rural, and dwell in village communities, it might be expected that the introduction of railways would have developed a considerable migratory movement. But it is said that at least four-fifths of the people in the villages belong to the classes that composed the original village community ; 3 and the new-comers occupy an inferior position. India's econ- omic organization is therefore of an antiquated character, such as prevailed in Europe two or more centuries ago, and the distribution of the population bears about the same anti- quated relation to that of the present Western world, as will appear from Table XCII : ' Op. cit., p. 43, and appendix p. xv. * Op. cit., p. 42. ' Op. cit., p. 48, STATISTICS OF URBAN GROWTH 125 a; I "2. 2" Not registered British In ( 3" 2^ t 3 a. ■t t •0 I 8 s 8 ioo,ooo-|- ... 20,000-100,00 2 0. 3" 3' 3 ~i H 8- (T VI s • S" re ^o !Z5 s ; M •^ ON 4». ^ 1 H 2 Cm VO W VO U) "^ ;i tn -P- ■^ •*>■ ^J ^ 8 i *> to ^N «) 7 •*a ^J VO tn ^ Ov s &• #; K> 4^ U> 4^ 4>. Cl M ^ U\ OJ U> Ov 00 -t»- ^ J? 2. u vN tn ^ \o 00 ^ •«J OJ •§ !^ re' s. »< 00 To ON OJ vb "10 w c. in ^J tn tn Ov VO 00 00 to »* > 0\ 0^ VO •^ -»>. Oo 00 OJ i •0 re s e-. »< * • • <' ^I • • to *• 1 5?: c OJ Ov VO -t ^ 3 5* re < o* s i (0 „ f VO Ov »H hd M -y •8 ^ ► 14 U) "^ ^I -F^ OJ 1 & •^ to 00 Ov tn *H '8 »0 ^ t CTn 00 vp V 4k so *^ K l k i- CTn M 00 N »H « n" k »-J 00 ^ VO 00 5' re tn t 00 tn -^ s S ^ S. •»a 1 ; ; • ^J tn to (S i i Ox • • • S' a> ^ \ 3 1 1 d* 1 1 ■s "O »n tn 1 » § 10 to a 1 E. 1 J 00 4^ to to + ■(>' > ■4 ■^ *• a. 00 ( Jv ^ 1 V • Ov ^ f t "O c J^ OJ •> 4 On Ui ^ ^ u> H M* 8 V <« p OJ < JN h i k U> vb vb "^ 1-4 4>. » 4 1 OJ t J h Ch> OJ to 00 VO ON » §■ N -v i "-1 t •*k vO "vi VO ^* VO s 1 68 5- 1 J? i? M 00 K- B a. M V D p i 3 f- 00 ^ to to t s- 5- 8 •*>■ D - b to 4^ u> -(^ bo I 1 5" 5 1 1 E I ^ i 126 THE GROWTH OF CITIES It would be interesting to know whether a tendency to- ward concentration of population in India has yet mani- fested itself. The following statistics indicate the rate of increase between 1881 and 1891, and answer the question in the affirmative : Table XCIII.' Classes of Cities. Principal Cities. 100,000 -(- 10.58 Bombay 6.28 75,000 — 6.54 Calcutta 8.25 50,000 — 13.60 Madras 11.50 35,000 — 9.48 Haidrabad 16.92 20,000 — 11.58 Lucknow 4.49 10,000 — 10.66 Benares 2.19 5,000 — 7.54 Delhi 11.06 3,000 — 1.54 Under 3,000 86 Total 9.40 The larger cities are evidently growing considerably more rapidly than the smaller places. The variations among the groups of large cities is undoubtedly due to the character of the towns, — whether commercial and manufacturing, or military or religious centres. This is shown in the second column where the half dozen principal cities are named. Benares is purely a religious centre of ancient renown, and is nearly stationary in population. Lucknow and Haidrabad are capitals of native states, but the latter has a large in- crease. The slow growth of Bombay and Calcutta, the great commercial centres, is explained by an overflow to the suburbs as in the case of London. As a general rule the largest increase is shown^ to have taken place in the case of the industrial cites, e. g., Hubli, 43.4; Karachi, 43.01; Ajmer, 41.26. The only military station or capital to ap- proach such a rate is Rawal Pindi, 39.30. Many of the feudatory state capitals show an actual decrease, and nearly all of them are on the wane. ^ Op. cit., 79, 81. * op. cit., p. 81. STATISTICS OF URBAN GROWTH 127 An attempt has been made to discover the relative rates of increase of the population in the larger provinces and their chief cities, but no valuable results have been obtained. It is difficult to get any data earlier than 187 1-2, the date of the first general census in India, and the frequent annexation of new territory, together with the prevalence of local famines or diseases, has broken in upon any uniformity that might otherwise have been discovered. The results obtained follow : Table XCIV.^ Percentage of Increase (-t ) or Decrease (— ). 1850-71. 1871-81. 1881-91. Bombay 19 18 14 Four chief cities ''■ — i 25 8.8 Northwest Provinces 29. 6.2 4.6 Six chief cities * 8.6 24.3 10.3 Bengal — .165 -I-6.7 Calcutta,* Patna —.826 -|-8. Madras — 1,5 15. Madras city 3.0 12. Punjab 7. 10.7 Three chief cities ^ 24. 5. Oude 1.3 II. Lucknow — 8.2 4.5 In almost every instance there is one period in which the province had the larger increase and another period in which the cities grew more rapidly. During the last decade, the cities fell behind in Madras, Bombay, Punjab, and Oude, while they surpassed the general rate of growth in the Northwest Provinces, Bengal, and the feudatory state, ^The authorities are the Census of i8qi, the Census of England, 1871, iv, 294, and, for 1850, Kolb, i860, and Harper'' s Gazetteer, 1855, ^ Bombay, Puna, Ahmadabad, Surat. •'' Benares, Cawnpore, Allahabad, Agra, Bareli, Meerut. * Calcutta, inclusive of Howrah and suburbs. '" Delhi, Lahore, Amrilsar. 128 THE GROWTH OF CITIES Haidrabad. It is therefore by no means certain that there is a tendency toward concentration in India, except such as results from the passing of cities from a lower to a higher group. The increase per cent, of the 26 " great cities " of 1 89 1, whose population is given in 1 881, Ms 10, while all India increased by ii per cent., even without reckoning the annexations. This is due to the decay of many of the great religious centres or native capitals. Madras, for example, was re- ported to have a population of 817,000 in 1823 (Hassel) and 720,000 in i860 (Kolb), but the census of 1881 gave it 405,848; Benares, reported 580,000 in 1823, 185,984 in i860; Delhi, 400,000 in 1823, 152,406 in i860; Calcutta, 900,000 in 1823, 794,193 in 1871 ; Surat, 450,000 in 1823, 109,844 in 1 88 1. No doubt many of these early estimates egregiously exaggerated the population, but there has cer- tainly been a decline in some of the cities almost sufificient to balance the gain in others ; so that the cities seem to be just about maintaining the same proportion in the general population. § 4. Of the other Asiatic countries, the only one whose sta- tistics are worth attention is Japan. The Philippine Islands, indeed, are reported in the Spanish census of 1887, but, as in Spain, the local statistics are based on the township, a territorial subdivision, rather than upon the dwelling-centre, an agglomeration of people : ^ 1887. Philippine Islands 7,000,000 100. Manila i 154,062 2.2 Towns 20,000-100,000 21 526,152 7.5 Total 20,ooo-|- 22 680,214 9.7 ^ There were 28 cities in this class in 1891, counting Calcutta and three suburbs as one city; but Maridalay and Shringar were not given in 188 1. ' Supan, 87. STATISTICS OF URBAN GROWTH 129 § 5. China, — The Chinese statistics are, of course, the crud- est estimates, and do not claim even approximate accuracy. Still, some idea of the Chinaman's tendency towards agglom- eration may be gathered from the statement that the aggre- gate population of the 52 cities in China which Supan estimates as above the 100,000 limit, is 88,336,000. This enormous " great-city" population forms 22 per cent, of the 402,680,000 persons accredited to China in the most recent estimates.^ § 6. Japan. — Even the Japanese statistics are of doubtful value ; thus, in the official statement of population of cities of 30,000+ which appears in the annual Resume Statistique and the Annuaire Statistique de V Empire du jfapon, the popula- tion of the city of Sendai fluctuates in the following manner:" January, 1884 SS'32i December, 1886 9i>709 December, 1887 7i»5^7 December, 1889 90,231 December, 1890 66,310 The following table shows the distribution of the urban population 1 887-90 : ^ Table XCV. No. Pop. Per cent. Japan 40,453,461 100. 100,000+ 6 2,353,807 5.84 20,000-100,000 49 1,829,601 4.53 10,000-20,000 80 i>099,389 2.72 Total 10,000+ 135 S.282,797 13.09 '^Statesman's Yr. Bk., 1897. * Cf. Supan, p. 83. 3 The RSsumi Statistique de P Empire du Japan; 6\h\t&x, 1892 (pp. 10, 16), gives the population of cities over 30,000 for Dec. 31, 1890; the smaller cities were found in Supan, the population being for 1887. The figures are all careful estimates {Berechnungen), not enumerations. In 1890 there were 141 towns of 10,000.+ 130 THE GROWTH OF CITIES There has naturally been a rapid urban growth in Japan in recent years. While a few cities, like Kanazawa, have lost in population, several of the commercial cities have had an as^ tounding growth if the Japanese statistics may be believed. Thus Kobe had a population of 20,579 in 1881, and 136,968 in 1890; Yokahama, 63,048 in 1881, and 127,987 in 1890. Taking the six cities which severally contained 100,000+ in 1890, and the 11 cities of 50,000-100,000, and comparing their aggregates with 1881, the following percentages of in- crease are obtained : ^ Japan i i.o 6 great cities 51.0 1 1 other cities 16.0 There is, therefore, a strong tendency toward concentra- tion in the great cities. In 1881 there dwelt in the "great cities " of Japan 4.4 in every hundred of the population, as compared with 5.84 in 1890. XVII. AMERICAN COUNTRIES.^ § I . Canada. — ^The Dominion of Canada was established in 1867 and the statistics of urban population since then are as follows : Table XCVI.» Urban Total Pop. Urban Pop. Percentage. 1871 3,635,024 689,019 18.8 1881 4,324,810 912,934 21. 1 1891 4»833.239 1,390,910 28.7 The urban population is officially defined as the aggre- gate population of towns of 1,500 and upwards. This is a lower limit than elsewhere prevails, but as appears in Table ^For 1890 as above, for 1881 Almanack de Gotha, 1885. ^ For the United States see Sec. I. of this chapter. ^Statistical Year Book of Canada^ 1895, p. ^^T* STATISTICS OF URBAN GROWTH 131 XCVIII, there is no great change in the percentage if the line be drawn at towns of 2,000. The census of 1891 shows an exceedingly large urban increase ; in fact, four-fifths of the entire increase in Canada in the last decade was in the towns. The official statisti- cians, in explanation, say that the phenomenon is " caused to a considerable extent by the growth of a number of places which had not attained a population of 1,500 in 1881." ^ But as there were only 29 such places, it is hardly possible that this element could have contributed more than 45,000, or less than one-tenth of the entire urban increase. The comparative rates of increase of large and small towns are herewith shown : Table XCVII.» 1851-71. 1871-81. 1881-91. Canada 30. 18.97 1 1.76 1 1 small cities 71. 62. 7 middle-sized cities 33. 21. 2 great cities 85. 54. 58. 20 cities i 46. 43. Thus even with a fixed number of cities, the urban popu- lation is increasing three times as rapidly as the general pop- ulation. The large percentages for the small cities is partly due to the fact that in 1851 only eight of the 11 existed as separate municipalities whose population could be ascer- tained; in 1 87 1, one of the 11 is still unrepresented. Mak- ing allowance for these, it is probable that the rate would be lower than that of the middle-sized cities. The tendency toward concentration is more adequately shown in the following table : ^ Statistical Year Book of Canada, 1895, P* ^^7» ' The classification follows that in Table XCVIII., where the population of each group is given for 1 89 1 and of Canada for 185 1. 132 THE GROWTH OF CITIES Table XCVIII. 1851. 1891. No. Pop. Per cent. No. Pop. Per cent. Canada •. 2.375.597 100. •• 4.833.239 100. Cities 100,000+. .... .. 2,397.870 8.2 " 20-100,000 5 175,287 7-4 7 291,578 6.0 " 10-20,000.. 2 25,697 I.I II 139,938 2.9 Total 10,000+ • 7 200,894 8.5 20 829,386 17.I " 2,000+ .. .. .• •• 1,319,060 27-3 Sources. — For 1891, summarized from Census of Canada, 1891, iv, 400: for 1851, The Census of iSsi for Upper and Lower Canada (Ontario and Quebec), sppplemented by Statistical Year Book of Canada, 1896, p. 7, and Harper's Gazetteer, 1855. The population of Prince Edward Island (62,678) included in the total of 1851 is for 1848. It is interesting to note the provinces that have contributed most to the urban increase. The proportion of urban (towns 1,500+) to total population: Table XCIX.i 1871. Ontario 194 Quebec 19.5 Nova Scotia 14.0 New Brunswick 24.3 Manitoba 1.2 British Columbia 8.9 Prince Edward Island 1 1.5 The Territories That Ontario and Quebec should contain increasing pro- portions of town populations is natural, but it is surprising to find a relative decrease in New Brunswick," and a wonder- ful increase in British Columbia and Manitoba. § 2. Mexico. — The Mexican statistics of city populations are too untrustworthy to demand serious study. In 1895, indeed, a fairly accurate census was taken, but the earlier ^ Census of i8g J, vf, ^01. ' The New Brunswick towns have been losing in population, while the province itself is stationary. I88I. 1891. 22.8 33.2 22.8 29.2 13-6 21.2 22.3 194 I2.I 22.5 II.9 42.5 I4.I I3-0 ... 5-6 STATISTICS OF URBAN GROWTH 13 j data are absurdly inaccurate, as a few comparisons will indi- cate: Veracruz. Guanajuato. Puebla. Almanack de Gotha 1880 56,1 12 75>ooo Supan = 1889 24,cx)o 52,000 110,000 St. Year Book {Q.^w%M&) 1895 88,993 39.337 9I.9I7 The best available statistics are the following : Table C.^ 1889. No. Pop. Per cent. Mexico . 11,632,924 100. Mexico City I 329,535 2.8 Cities 20-100,000 . . . 20 730,261 6,3 Cities 10-20,000 • . • • 30 400,156 3.4 Cities 10,000+ . . . • 51 1.459,952 12.5 1895- No. Pop. Per cent. 12,570,195 100. 344.377 2.74 18 892,052 7. The Mexican cities seem to be growing no faster than the rest of the country. A few commercial cities like Veracruz are indeed developing rapidly, but to counterbalance this is the slow growth or even decline of many ancient capitals : Mexico Capitals of provinces. 25 1850.2 7,661,919 703,186 = = 9-2% 29 1880.^ 9,787,629 946,886 : ^9-7/^ § 3. Brazil. — At the middle of the century Brazil had about the same population as Mexico, but now has about twenty- five per cent more, the census of 1890 giving a total of 16,330,216. The only available urban statistics are the offi- cial estimates of 1888: ^For 1889, Bureazi of American Republics: Bulletin N'o. jo,^. 166; for 1895, St. Yr. Bk., 1897, P- 739- I*^ is probable that the popvilation of cities in 1889 is here underestimated; Supan's figures for the same year give a total of 795,200 for cities 20-100,000, or 6.8%. ^Harper, 1855. '^ Almanack de Goiha, \%%t^. 134 THE GROWTH OF CITIES Table CI. No. Population. Percentages.. Brazil 14,002,335 100 Rio de Janeiro Si5»5S9 3-7 Bahia 162,065 i.i Pernambuco 130,000 .9 Cities 100,000+ 3 807,600 5.7 " 20,000-100,000 II 420,000 3.0 " 10,000-20,000 17 206,000 1,5 Cities 10,0004- 31 i!433>6oo 10.2 According to the census of 1856, Rio de Janeiro contained 300,000 inhabitants in a total population of 7,677,800'' — a percentage of 3.9. Apparently the metropolis has not grown as rapidly as the rest of the country, but there is a large suburban population not counted in the figures of 1888. It is stated that the city with suburbs now (1898) has a popu- lation of about 1,000,000. § 4. Argentina. — The noteworthy thing about this progres- sive republic is the remarkable concentration of population in one large city, the metropolis, Buenos Ayres. In 1869^ at the time of the first authentic census, it had a population of 177,787, or 9.8 per cent, of a total population in Argentina of 1,812,490.3 In 1887, suburbs containing 28,000 inhabi- tants were annexed to a population of 404,000. At the end of 1896 its estimated population was 71 2,095. ^ The results of the enumeration of 1895 are not entirely obtainable, and in the table below, the official estimates of 1890 are also given.s * Amer. Repub., Bui. No. jo, p. 64. »Kolb, i860, p. 369. ' Almanack de Gotha, 1885. * Annuaire statistique de la Ville de Buenos Ayres, 1896. ' According to Kolb, i860, Buenos Ayres and suburbs had a population in 1856 of 130,000, which is 8.7 per cent, of the estimated population of Argentina (1,500,000;. Harper's (1855) gave the city in 1852 120,000, and the Republic 829,400 (excluding aborigines) ; i. e., 14.5 per cent. STATISTICS OF URBAN GROWTH 135 Table CII. 1890.1 1893.2 No. Population. Per cent. No. Population. Per cent. Argentina 3,456,000 loo. 3j952,990 100. Buenos Ayres i 561,160 16.2 677,786 17.1 Cities 20,000-100,000 5 227,000 6.6 7 289,043 7.3 " 10,000-20,000. 13 173,000 5.0 Total io,ooo-j- .. . 19 961,800 27.8 § 5. Chile. — In Chile there apparently exists a tendency toward concentration, although the statistics are untrust- worthy; for example, the fifth census (1875) warns its readers that an addition of 207,597 should be made for omissions, thus recognizing an error of 10 per cent. Table CIII, 1850. 1875. 1885. Per cent. Pop. Per cent. No. Pop. Per cent. Chile 1,600,000 ... 2,075,971 100. .. 2,527,320 100. Rural pop i,350.48i 65.3 ..1,464,776 58. Urban " 725,490 34.7 .. 1,062,544 42. Santiago 65,000 \ f 189,332 Valparaiso 30,000 i »- 104,952 Cities 100,0004- 2 294,284 1 1.6 20,000-100,000 3 69,000 2.7 10,000-20,000 5 69,000 2.7 Total io,ooo-f ID 432,300 17.1 Sources. — For 1850, Harper's Gazeiieer, 1S55; for 1875, A Itnanack de Gotha, 1885; for 1885, St. Yr. Bk., 1897, and Supan. The list of cities in Bur. Am. Repub. Bui., 50, is evidently based on a territorial unit, as there are 46 " cities" of 20,000-100,000 with an aggregate popula- tion of 2,035,000! The Chilean definition of urban population is unknown to the writer, but it obviously includes very small towns or else is based on a large territorial subdivision as the unit. § 6. The remaining American states are too unimportant to detain us long. They are either small and thinly popu- lated or else their population consists largely of uncivilized Indians and half breeds. They are grouped in Table CIV : ^ Supan, pp. 118-119. The list of cities in the Bui. of Amer. Repub., does not seem to be complete for cities 10,000-20,000, but its figures for the population of Buenos Ayres and Argentina are here adopted. ' St. Yr. Bk., 1897, pp. 322-3. 136 THE GROWTH OF CITIES m to 1^ O^OcT *^00 t^ lO ^00 O « O On fO CO COM M M « ts. « b-vo ts.H -^o ro«^o ■* CI O ^ "^ ■■ OOO CO '^ " \o ^^ H ir> tN.«o •* « Q^O « 3O0COCOC0CO0OO0CO t^ M CO O OO OMN ^ vo -"J* ro O E « 6 Eg* SS _«.§_« = S c fi IE • • • <5 • « p S .' ^ -^^ rt iZ rt ^-i* J ^ M\\0 tN.00 o> O CO O^ O On OO CO 00 00 OS oAkIz; ^•a c ,c ^ rt „ " u o ■•'O « c S 73 S £ .S " I. STATISTICS OF URBAN GROWTH 137 XVIII. AFRICAN COUNTRIES. But a small proportion of the inhabitants of Africa have been enumerated. The most ancient country of African civilization, Egypt, has the largest population and the great- est cities. The census of 1882 yields the following results : Table CV.' Egypt Cairo Alexandria. Population. Per cent. 6,817,265 IOC. 374,838 213,010 587,848 8.6 179,756 2.6 303,295 4-5 Cities 100,000-+- 2 " 20,000-100,000 6 " 10,000-20,000 22 « 10,000-f 30 1,070,899 15.7 At the middle of the century, Cairo and Alexandria con- tained respectively 250,000 and 60,000 inhabitants,'' or 6.9 per cent, of the entire population. Their growth since has been at a more rapid rate than the population of Egypt in its entirety. The other countries are summarized in Table CVI. Orange Free State, as well as Abyssinia, is noteworthy as having scarcely any towns of more than 5,000 population, and none reaching 10,000; but Bornu, a native state, has one large city and several smaller ones. Table CVI. Source. Date. Population. Cities 20,000-100,000 10,000-20,000 10,000+ 1. Algiers Census 1891 4,124,732 5 248,690 6.0 6 94,401 2.3 11 343,091 8.3 2. Cape Colony " 1891 1,529,224 3 103,235 6.8 2 20,976 1.4 s 124,211 8.2 3. Natal " 1891 S43>9i3 ° •• 230,2375.6 2 30,237 5-6 4. Orange Free State " 1890 207,503 o o o o o 5. Transvaal .. off. est. 1896 790,000 (Johannesburg, the sole city,) = i 102,71412.9 6. Abyssinia. . . est. 1896 3,500,000 o o o o 7. Bornu (Sou- dan) " 1896 5,000,000 I 55,000 II.O I Authorities. Supan for i, 2, 3, 4. St. Yr. Bk., 1897, for 5, 6, 7. It is to be noted that the population of Johannesburg includes the district within a three miles radius. ^ Supan, pp. 90-91. * Harper, 1855. The population of Egypt at the census of 1846 was 4,463,244 {St. Yr. Bk., 1897.) 138 THE GROWTH 0I< CITIES XIX. AUSTRALASIA. The most remarkable concentration, or rather centraliza- tion, of population occurs in that newest product of civiliza- tion, Australia, where nearly one-third of the entire popula- tion is settled in and about capital cities. The following^ table gives the absolute numbers and ratios for 1891 and the ratios for 1881 : Table CVII.^ Colony. Capital. New South Wales Sydney ... Victoria Melbourne Queensland Brisbane • South Australia Adelaide . Western Australia Perth .... Tasmania Hobart • . . New Zealand Wellington i8qi Pop. of Ratio of capital Ratio Colony. 1,132,234 1,140,405 393>7i8 320,431 49,782 146,667 626,658 Capital. to colony. 383,386 490,902 101,564 133.252 8,447 33,450 33,224 34-27 43-09 25,80 41.58 16.97 22.81 5-3 1881. 28.79 32.14 13-70 36.27 19.36 17-75 4.10 3,809,895 1,184,225 31. 1 25,10 The population of the capital city formed a larger percent- age of the entire population in 1891 than in 1881 in all of the colonies except Western Australia. Melbourne and Adelaide contain over two-fifths of the whole population of Victoria and South Australia. When it is remembered that there are only seven States in the American Union in which all the cities of 8,000 inhabitants and upward contain two- fifths of the population, the conditions in Australia can be better understood. There are several countries in the world that contain a larger proportion of urban population, but in none of them is it so massed in a few centres.'^ The rate at ^T. A, Coghlan (Government statistician of New South Wales), A Statistical Account of the Seven Colonies of Australia, Sydney, 1892, pp. 335, 352. ^ Except in individual States of the American Union, e. g.. New York City con- tains fully 50 per cent, of the inhabitants of the State, and metropolitan Boston 40,17 per cent, of Massachusetts's population. STATISTICS OF URBAN GROWTH 139 which this centralization has been going on appears in the following figures : Table CVIIL' Total population of the Percentage of total population living Seven colonies. Their capitals. in the capitals. 180I 6,508 1821 C 3S,6lO 1831 79>3o6 1841 21 1,095 43»76i 20.7 1851 430.596 84,503 19.6 1861 c 1,252,994 276,960 22.1 1871 c 1,924,770 431,533 22.4 1881 c 2,742,550 689,634 25.1 189IC 3,809,895 1^184,225 31.1^ ^ ^ ^ ^^ \ 835,888 21.92 Up to 1 87 1 the cities did not grow very much more rapidly than the rest of the population, but in the last two decades the difference has been marked. On the whole, these figures may be fairly taken as repre- senting Australia'a urban population, but in one way they exaggerate the concentration, or congestion of population, inasmuch as they include large suburban districts. Thus, Sydney includes 35 suburbs, and its area is larger than any city in the United States except Chicago, being 150 square miles, or 96,000 acres. Chicago has 103,000 acres, Philadel- phia, 83,000, and London, 75,000; the other great cities are smaller, Paris having an area of 19,000 acres. New York (before 1898) 25,000, and Berlin only 16,000.3 Mel- bourne's acreage, 163,942, is still greater than Sydney's. If now the AustraUan suburbs be counted in with the great cities only when they form industrial parts thereof,-* and ^ Coghlan, op. cit., 334, 352. " C " indicates census years. * With and without the suburbs, respectively. ' iitk Cen., Soc. Stat, of Cities, 13. * The writer here follows Supan. I40 THE GROWTH OF CITIES 0+ IN 189I. Per cent. Population. 'fi of colony. in i8si. 381,444 33.6 28,2 525,632 46.1 30. 64,455 16.3 0. 90,786 28.3 28. ca 0. 0. 46,487 32.0 150,479 24.0 0. Otherwise classed as separate municipalities, we shall have the following figures : Table CIX. CiTlBS OF I( No. New South Wales 10 Victoria 7 Queensland 2 South Australia 2 Western Australia o Tasmania 2 New Zealand 4 Total 27 1,264,283 33.2 There are now 27 cities that exceed the limit 10,000, whereas, if the official statistics of population be accepted many of these cities, together with smaller towns, would be assigned to Sydney and Melbourne, which would indeed reduce the number of cities, but increase their population. Following the official grouping, this classification will result thus: Table CX.^ Number and popuuvtion of cities in 1891. 20,000- 10,000- Total 10,000+. 100,000 -p. 100,000. 20,000. No. Pop. fb. New South Wales i 383,386 i 51,561 452,596 6 487,543 43.1 Victoria i 490,902 3 107,481 o 4 598,383 52.4 Queensland i 101,564 o i 13,380 2 114,944 29,2 South Australia i 133,252 o i 15,976 2 149,228 46.7 Western Australia ...-o o o o Tasmania o i 33,450 i 17,208 2 50,658 34.5 New Zealand o 4 178,062 o 4 178,062 28.5 Total 4 1,109,104 9 370,554 7 99,160 20 1,578,818 41.4 Percentage of Austral- asian population ... 29.1 9.7 2.6 41.4 The historical development of urban population may be best studied in New South Wales,^ the original colony from 1 Coghlan, op. cit., 353. '' Census of i8gi. Statistician's Report, p. 120. STATISTICS OF URBAN GROWTH 141 which the other colonies of the mainland have been sepa- rated. The censuses proper begin with 1861, but before that year Sydney virtually represented the urban population, as New Castle (the next largest city) only reached 7,810 population in 186 1. In 1790, at the time of the first "mus- ter," all of the 591 European inhabitants of the colony lived at Sydney; in 1799 half of the 5,088 inhabitants lived there; in 1811,4,895 out of 10,025 or 48.8 per cent.; in 1821, 13,401 out of 29,662, or 45.2 per cent.; in 1831, 16,232 out of 60,794, or 26.7 per cent; in 1841, 29,973 out of 116,631, or 25,7 per cent. This was the smallest percentage ever reached for the city. In 185 1 Sydney contained 53,924 out of 191,099 inhabitants, or 28.2 per cent. The movement toward the occupation of lands has stopped, and the popula- lation flocks to the great city. In 1 861-71, as appears from the following table,^ the outside districts gained two new settlers to one gained by Sydney; but in 1 881-91, the con- dition was more than reversed, Sydney having an increase of 160,000 to 67,000 increase for the rural districts: Table CXI. 1861. 1871. 1881. 1891. Sydney i 95,789 i 137,776 i 224,939 i 383.283 Other cities 5,000+ . . 3 19,081 4 32,987 6 58,481 8 120,753 Towns 2,-5,000 4 14,623 7 20,564 20 54,608 41 117,587 Total urban 8 129,493 12 191,327 27 338,028 50 621,623 Villages 29 30,341 44 43,486 85 88,910 108 108,396 Rural 189,116 .. 266,766 .. 321,303 ... 388,231 Total 348,950 •• 501,579 •• 748,241 ••• 1,118,250 Shipping, etc 1,910 .. 2,402 .. 3,227 ... 5,704 Aborigines 8,280 Grand total 350,860 .. 503,981 .. 751,468 ... 1,132,234 Percentages. Sydney 27.45 27.47 30.06 34.27 Other towns 9.65 10.67 i5-io 21.32 Villages 8.70 8.67 1 1.90 9.69 Rural 54.20 53.19 42.94 34.72 Total 100. 100. 100. 100. ^ Census of i8gi. Statistician's Report, p. 126. 142 THE GROWTH OF CITIES In this classification "village" includes all municipalities with a population under 2,000; as will be seen, they average about 1,000 each. The purely rural or agricultural popula- tion has declined relatively from 54.20 per cent, in 1861 to 34.72 in 1 89 1. The villages have done little more than hold their own, while Sydney and the other cities have grown rapidly. XX. SUMMARY AND CONCLUSIONS. In order to bring statistics of urban population in different countries into comparison, the author has aimed to secure, as the town unit, an actual agglomeration of people, and not a ter- ritorial unit or political subdivision. The distinction between the two modes of procedure has been discussed in Chapter I ; here it is necessary only to call attention to the exceptions ncessarily made in a table of comparisons. Where the local unit is not a territorial subdivision {i. ^., where it is the vil- lage, town or city, in the United States ; the urban sanitary district in England ; the Centri in Italy ; and the German Gemeinde, virtually coinciding with the Ort or Wohnplatz in Saxony), it is unnecessary to pay any attention to the area; elsewhere the size of the territorial unit may appreciably affect urban percentages, because it will often contain a scat- tered or rural, as well as an agglomerated, population. The average size of the territorial unit in square kilometres is as follows : Massachusetts, town or city 61.4 Spain, Ayuntamento 54. Netherlands, Gemeente 28.7 Hungary, Gemeinde 22. France, commune 14.62 Switzerland, commune 1 2.6 Belgium, commune 1 1.4 Austria, Gemeinde 10.6 Germany " 7. Prussia " 6.34 STATISTICS OF URBAN GROWTH ^43 It will be perceived that the danger of reckoning isolated residents among urban dwellers is at the minimum in the small Gemeinde of Germany, which, in extent, fairly conforms with the incorporated village in America. In France the commune is not much larger, and in ascertaining the urban population only those communes are included which con- tain an agglomerated population of 2,000 and upwards, although to all such the scattered population is then added. In the Netherlands, the territorial unit is still larger, and it affects the percentage of urban population considerably, as the note to the table shows. The Spanish Ayuntamento (also in Cuba and the Philippines) is so large as to render the comparison worthless for the smaller towns, and these have therefore been enclosed in brackets. The New Eng- land township, again, must obviously include a rural popu- lation ; but in the more careful computations of urban popu- lation there, it is customary to treat the incorporated cities (approximately those towns of 12,000+ agglomerated popu- lation) separately as constituting the urban population. With the reservations here made, Table CXII is presented as a summary of results obtained in the statistical investiga- tions of the present chapter. In Table CXII the countries are arranged in the order of percentage of urban population (/. e., population in towns of 10,000 or more inhabitants). In some cases where this population could not be ascertained, positions have been assigned after comparing the other percentages, and at the same time keeping in mind the different conditions. China's position, however, is an arbitrary one, as only one percentage is given, and that is of slight value. In this table, the official statistics of Australia, regarding the territorial extent of the great cities, have been followed ; had the limits been drawn with Supan, at the actual municipality, the percentage for the seven colonies (10,000+) would be 33.2 instead of 144 THE GROWTH OF CITIES toi-cm M i : >o : 2 ►^ ^ t^ S : \ ** I ) ^ d d 00 t^ t^vd vd NO U^ -^ CO « CO c C* C4 Ct C4 « n w « dicd tC. « CO d d d^vd 00 vd d»N nvd t>. ■♦^d c^^d covd,*>'*oo «M M H MM MM '— ' W CO W M O O O^ \r,\0 OOOOOt**wO i OOO «NO O O OS Ch On 0\ OnQO 00 0» O* CT* O^ On O^oO On 0*oO 00 On On 0*00 OnOO On O^ On On oooooooocooooooooooooooooooooooooooooooooooooooooocooooo 5 + oooooo * p- • rtS >,*in ,*■* in« 00 o-* •♦ i.s ;■«■=;>>; >>cw . :^ : : ' : : : •jso : : ^ ! =-s S5 5:a S s 5 Ef b-§ 2 S g.^ ^g §:s J-g-H j; S » 1 m 'i- iA»o t^fio ov d M M c*> ■* vjvo t^eo o*'^ ■• " ^^.^•'^ -i^ 1 •♦ >ovo f-oo o> o M « ro ■«• wo rNOO STATISTICS OF URBAN GROWTH W«>:5«Ot^O^HS£H(S(£(J:m££c5«^c3l^4^i g^S.S.S,ffi3;S^&^2oi«r> « ««-WSro^S5'S.S^i;^??¥^5:^o;o M « ^ j^>d ^^d; lo lo vo m io m lo 146 THE GROWTH OF CITIES 41.4. But that would still leave Australia near the head of the list. One is impressed with the extent of the variations in the percentage of urban population in the different countries of the world. On the one hand, England with 62 per cent, of its population city-dwellers ; on the other hand, several Balkan states with only five city people out of every hundred, and the Orange Free State with no real urbanites at all. Of the causes of such extensive variations, that which most readily suggests itself is density of population. Given two countries of equal area, it would naturally be expected that the more populous country would contain the larger number of cities. Thus it would seem impossible that peo- ple could be crowded together as they are in Belgium or England, without living in such close proximity as to con- stitute agglomerations. But such is the case. Bengal, for example, has as many inhabitants as the United States in a territory scarcely larger than Great Britain and Ireland ; and the density of population in Bengal is exceeded only slightly by that of Belgium and England. Nevertheless, the per- centage of urban population in Bengal is only 4.8 as com- pared with 47.7 in Belgium.^ In order to compare the relation of density of population to its concentration, the following table showing the number of inhabitants to each square kilometer of territory has been compiled : "^ ^ In both cases urban population= towns of 5,000 -f-. * Statistik des Deutschen Reiches, Neue Folge, Bd, 68, p. 6*. A few countries have been added from Almanack de Gotha, and are enclosed in parenthesis marks. The data refer to the censuses of 1889-91. STATISTICS OF URBAN GROWTH 147 Table CXIII. Saxony 234. Belgium 206. England and Wales 192. Bengal 181.8 North- West Provinces 168.5 Netherlands 138.7 Italy (1893) 107. Japan 106.5 Madras 97.4 Germany 91.5 (China-proper) 87. Prussia 86. Austria 79.6 Bavaria 74. Switzerland 73.3 Punjab 72.8 France 72.5 British India 71. i Bombay 58.3 Ireland 57.6 Denmark 55.1 Hungary 54.2 Scotland 52.2 Portugal (1881) 51. Servia 44.5 North Atlantic States 41.5 Roumania 38.5 Spain (1887) 34-8 Greece 34. '^Chinese Empire) 32. Bulgaria 31.8 Bosnia and Herzogovina 26. (European Russia. 1897) 19. (Cuba) 13.7 Sweden 10.8 United States 8.2 (Mexico) 64 Norway 6.2 (Russia, 1897) 5-8 Victoria, Australia 5. (Colombia) 4.8 (Chile) 3.8 (Peru) 2.8 (Cape Colony) 2.7 (Brazil) 1.7 New South Wales 1.4 (Argentine) 1.2 Canada 0.6 India, Italy and Japan are densely populated countries; but they have relatively small urban populations. On the other hand, the United States and Australia are thinly popu- lated and still have relatively large urban populations. Scot- land and Argentina do not occupy parallel positions. Evi- dently there are other factors in producing agglomerations than mere populousness. A more probable explanation of large urban populations is the organization of industry on a modern scale. It appears, indeed, that nearly all of the more advanced industrial nations are included among the first fifteen countries in Table CXII, while none of the countries in the second half of the list, with the exception perhaps of Japan, can be said to be in the fore- front of modern industry. 148 THE GROWTH OF CITIES It cannot be said that manufacturing or machine industry alone causes the concentration of population. Ranking the leading nations by the amount of steam power per lOO in- habitants, for example, does not yield the same order as that in Table CXII. Thus, the countries that utilize steam to the extent of at least 20 horse-power per 100 inhabitants are the United States, England and Scotland ; more than 10 and less than 20 — Belgium, Germany, France ; more than 7 and less than 10 — Netherlands, Denmark, Scandinavia, Ireland; more than 3 and less than 7 — Russia, Austria, Hungary, Switzer- land, Italy, Spain ; less than 3 — Portugal and the Balkan States, including Greece and Turkey.' The United States should follow England and Scotland if manufactures alone determined the percentage of urban dwellers ; while the Netherlands, Turkey, etc., would occupy positions much lower in the list. But Holland is a great commercial country, carrying on a larger commerce per capita than any other nation in the world ; its large urban population is chiefly to be attributed to that fact. The same applies, to a less degree, to Turkey. Constantinople contains by far the larger portion of Turkey's urban population, its percentage being 18.3, while for all cities of 20,000-f- it is only 24. In this case some influence may be attributed to politics as a cause of concentration ; but it still remains true that it is Constantinople's commercial advantages which have made the city the seat of government. It is of course true that back of density of population and industrial organization are the physical features of a country and its comparative natural advantages for different indus- tries. Nature has perhaps determined that in Uruguay a very large percentage of the population shall be centered in the city and department of Montevideo. Nature has also ' Hobson, Evolution of Capitalism, 85-6. STATISTICS OF URBAN GROWTH 149 favored the building of a great commercial city at Buenos Ayres and discouraged the dispersion of the population by providing better advantages for grazing than for cultivation of the soil. Such is emphatically the case in Australia, while in India a rich soil entices to an extensive cultivation, and supports a large agricultural population, which in the very nature of things cannot be brought together in great agglomerations. But after all, Nature has been subjected to man's commands, and if the English people and the East Indians were to exchange places, it is altogether likely that India would become a land of great cities and England an agricultural country with a scattered population. Hence in Australia, it is not a sufficient explanation to say that the physical features of the country (few harbors, few rivers, vast plains suitable for grazing) are the determining factor. It is rather the alertness with which the progressive Austral- ian democracy has adjusted itself to the requirements of the modern industrial organization with its international and local division of labor. Australia has no anciently estab- lished manufactures like those of old England, nor even the vigorous "infant industries" of New England. On the con- trary, Australia has vast tracts of unoccupied lands tempting men to agriculture. The main reasons why the Australians prefer to remain in the seaboard cities rather than settle the interior is that nineteenth century industry requires few workers on the land. In European countries the process of agglomeration proceeds more slowly because the superfluous agriculturists have been brought up on the farm, and have to overcome the inertia of their position in order to find their true place in the industrial organism ; it requires a distress- ful agricultural depression like the one that has prevailed since 1893 to bring home to the agriculturist the conviction that his labor is not wanted on the farm. But in Australia the mass of the population has been in the seaboard cities. 150 THE GROWTH OF CITIES where the emigrants land, and consequently has no such inertia to overcome. Australia is therefore the representa- tive of the new order of things, toward which the modern world is advancing. It is thus in the dynamic rather than the static aspect that the true significance of the agglomeration of population manifests itself. The reasons why the distribution of popula- tion in England is so different from that in India are clearly seen when one studies the causes of the movement which has made the England of to-day so dififerent, as regards the dis- tribution of population, from the England of 1800. Then it will appear that the physical features of, say, England and India, count for less as a factor in the problem than the qual- ities of the race and its progress in material civilization. It is not to be denied that even the material civilization of a country depends upon its natural advantages to a certain ex- tent, but the principal consideration after all is the use to which such advantages are put by their possessors. China is known to be rich in coal and iron — the fundamental ele- ments of machine industry — but China has not become a great industrial nation like England. While, therefore, the topography and the resources of the country and also the density of its population do sometimes influence the distribu- tion of the population (notably Australia, Turkey, Uruguay, Argentina), in the majority of cases it is economic organiza- tion that constitutes the decisive influence. If now the percentages of urban population in the different countries given in Table CXII, be compared for the years 1800, 1850, and 1890, as in the accompanying diagram, it will be found that the urban growth has very generally taken place since 1850. The exceptions are England and Scot- land, the United States, and in a smaller degree, Belgium, Saxony and France. In the two former, the process of con- centration wrought greater changes in 1800-50 than in STATISTICS OF URBAN GROWTH O rT I/O lis- iSi 49- Russia census years. (Based on Table CXII.) NoxK.-Saxonys first percentage should be 8.9 for xSr^, instead of ,.5 for x8oo; that of ^ stead 'France 95 in- 152 THE GROWTH OF CITIES 1850-90. But in many other countries the movement had scarcely begun in 1850. In Denmark and Holland, indeed, the urban percentage actually diminished during the first half century. In Portugal there has been a continual diminu- tion, lasting, it appears, down to the present time. The following table showing roughly the principal periods of rapid concentration serves to make clearer the analysis : TABLE CXIV. England 1820-30, 1840-50 Prussia 1871-80, 1880-90 United States 1840-50, 1860-70, 1880-90 France 1850-60, 1860-70 Austria 1846-57, 1 880-90 Hungary 1850-57, 188090 Russia 1870-97 Sweden 1880-90, 1860-70 Norway 1875-91, 1865-75 . Denmark 1870-90 Netherlands 1880-90, 1870-80 Belgium 1866-80, 1880-90 Switzerland 1850-60, 1880-88 Canada 1881-91, 1871-S1 Australia 1881-91, 1871-&1 This amounts to a demonstration that the Industrial Revo- lution and the era of railways, both of which opened earliest in England and the United States, have been the transform- ing agents in the re-distribution of population. They are the elementary forces in the bringing about of Modern Capital- ism. And the effects of their introduction into the conti- nental countries of Europe are to be observed at the present time. The re-distribution of population is accomplished not only by a movement from the fields to the cities, but also by migration across the seas. This is a factor of prime import- ance, for example, in the Scandinavian countries, whence issues an emigration second only to that from Ireland.^ ^ Thus the emigration in the last 70 years from the countries specified bore the following relations to the total population of those countries in 1890-91 : Immigrants Total pop. in 1890-91 From to the UnitedStates. in millions. Ratio. Ireland 3,481,074 4,7 74. Norway and Sweden .... 925,031 6,8 13.5 Germany 4,504,128 49, 9.2 England and Wales 1,637,065 29, 8.6 (^iith Cen., Pop., i, p. Ixxxi.) STATISTICS OF URBAN GROWTH 153 In conclusion, it may prove of interest to place in com- parison with the European countries of large urban popula- tions, some of the American commonwealths. Of the four great Western nations, Great Britain very considerably out- ranks the others, which are very close together. France comes fourth, while Germany (Prussia) and the United States are very nearly equal. But the German urban statis- tics, as have been shown, are based on the Gemeinde (town- ship) which gives it an advantage, and it is to be noted that if the comparison be restricted to cities of 20,000+ or 100,- 000+, the United States clearly ranks above both Germany and France. The only considerable portion of Germany that has a larger urban population than the United States is Saxony, a country somewhat more than half as large as Massachusetts,^ which has an urban percentage of 66 to Saxony's 35. Table CXV.^ Percentage of population in towns of 10,000 -i. Massachusetts 65.9 Uruguay 30.4 Eng. and Wales 61.7 Ohio 30.2 Rhode Island 57.9 Prussia 30.0 New York 57.7 Utah _. 28.7 New Jersey 50.9 Washington 28.3 Scotland 49.9 Argentina 27.8 Maryland 43.9 Minnesota 27.7 Connecticut 41.9 UNITED STATES 27.6 (Australia 414) '^ France 25.9 California 41 .0 Missouri 25.6 Pennsylvania 39.1 New Hampshire 24.8 Illinois 38.1 Michigan 23.9 Colorado 37.1 Louisiana 23.7 Delaware 36.5 Denmark 23.6 Belgium 34.8 Wisconsin 22.5 Saxony 34.7 Nebraska 22.2 Netherlands 33.5 Italy 20.6 (Australia 32.2)' Bavaria 20.5 'Area of Saxony, 5,787 sq. miles; Massachusetts, 8,315. 2 Based on Tables XI and CXII. * Varies with the inclusion or exclusion of suburbs of Sydney, etc. 154 THE GROWTH OF CITIES This table shows that while the United States as a whole has a smaller urban percentage than several other countries, it contains States as large as those foreign countries, with, larger urban percentages. Scotland, Belgium, Saxony, Hol- land and Uruguay are all small countries as compared even with American commonwealths. Even England and Wales embrace an area of only ^'j^'jQ^ square miles, which is almost precisely equal to the combined territory of New York, Mas- sachusetts and Rhode Island. And 60 per cent, of the aggre- gate population of these commonwealths live in cities of 1 0,000-1-, while in Massachusetts alone the percentage rises to 65.9. Nor is this large percentage due to the large terri- torial extent of the Massachusetts town, for in 1895 the ag- gregate population of the State living under city government formed 65.43 per cent, of the total; and the smallest of the cities (Beverly) had a population of 1 1,806.^ Twelve Amer- ican commonwealths rank above Prussia in urban popula- tion and contain 26 million people to Prussia's 30 million. These facts should be remembered when it comes to com- paring the continental United States with the small countries of Europe. ' Census of Mass., i8g_5, i., 48-49. CHAPTER III. CAUSES OF THE CONCENTRATION OF POPULATION. I. INTRODUCTORY. In seeking the causes of the remarkable concentration of population that has taken place in most of the civilized countries of the globe during the last fifty or one hundred years, careful discrimination must be made between this phenomenon and the growth of cities as a mere accompani- ment to the general increase of population. The real checks upon the growth of population in previous centuries were war, famine, pestilence, and insanitary cities involving par- ticularly high infantile mortality. Hence, as soon as the progress of medical and sanitary science, transportation methods, industrialism and the other factors of modern civiHzation had mitigated the " scourges of mankind," there followed a period of unprecendented increase of population in all Western countries. In this general increase, the cities have naturally participated — and have even outrun smaller communities and scattered populations. For it was in crowded centres that modern science was most needed to render the conditions of life healthful. Throughout the middle ages and the earlier centuries of modern times, the cities of Europe depended almost entirely upon the influx of country people for their growth ; the mortality was so high that the deaths annually equalled or exceeded in number the births. London was no worse off than other European cities, Paris and one or two other places possibly excepted ; and yet London's birth-rate never exceeded its death-rate, (I5S) 1^6 THE GROWTH OF CITIES for any considerable period, until the very beginning of the nineteenth century. It is a patent fact that the rapid growth of London in the present century has been in part due to its ability to confine the mortality within reasonable bounds and thus secure a natural increase (an excess of births over deaths) ; the natural increase has of course steadily aug- mented with the constant improvements in municipal admin- istration and the application of the discoveries of medical science. Nevertheless, the transformation of a deficiency into an excess of births is not the essential reason of city-growth, which must rather be sought in economic conditions. Com- paring, for example, the demographical statistics of the French cities for the decade 1881-1891 it will be found that Lyons, Marseilles and Bordeaux had fewer births than deaths ; and yet they surpassed the other large cities of France (Roubaix and Lille alone excepted) in their rate of growth.^ The " mushroom " growth of American cities, which has been the subject of considerable comment, should occasion no surprise, since it is mainly due to the settlement of un- cultivated territory. It is only when the growth of a city has proceeded more rapidly than the development of its contributory territory, that one needs to study other under- lying causes. For then one is face to face with the problem ' Cf. Statistisches yahrbuch der Stadt Berlin, xix (1892), pp. 94-5 : Natural increase Total or decrease. increase. Paris 4-2.36 8.54 Lyons — 1.23 iP-73 Marseilles — 2.56 12.34 Lille 4-7.08 17.86 Bordeaux — 0.85 I4'I9 Roubaix +13.38 24,40 The Italian statistics show even more striking instances, but owing to the ab- sence of a regular census since 1881, they cannot be trusted. CA USES 157 of the concentration of population, — an increasing proportion of the population collected in cities. While, then, it is generally true that the unprecedented increase of population during the present century has been a condition of the rapid growth of cities, it has not necessa- rily been a positive cause of their relatively rapid growth as compared with the remainder of the population — a cause, that is, of the phenomenon of concentration. Positive forces may exist to drive a larger proportion of people into the rural districts notwithstanding an all-round increase in popu- lation. Such has been the actual result in Portugal, where, if statistics are not at fault, a smaller percentage of the in- habitants is living in cities to-day than in the middle of the century.^ On the other hand, the cities of France have been enjoying a rapid growth all the time that the population of the country has as a whole has been virtually at a standstill, and those of Ireland have likewise grown while the popula- tion in general has declined. It is now clear that the growth of cities must be studied as a part of the question of distribution of population, which is always dependent upon the economic organization of society — upon the constant striving to maintain as many people as possible upon a given area. The ever-present problem is so to distribute and organize the masses of men that they can render such services as favor the maintenance of the nation and thereby accomplish their own preservation. Population follows the line of least resistance in its distribu- tion, and will consequently be affected by changes in the methods of production. When the industrial organization demands the presence of laborers in particular localities in order to increase its efficiency, laborers will be found there ; the means of attraction will have been "better living" — in other words, an appeal to the motive of self-interest. Econ- 1 Cf. Table LXXXII. 158 THE GROWTH OF CITIES omic forces are therefore the principal cause of concentration of population in cities ; but there are other motives exhibited in the " Drift to the Cities," and these will also receive con- sideration. What, now, are the economic forces that have caused the massing of people in large communities? The business man's answer would probably be short and trenchant, ^^ " Steam." Literary critics and dilettantes in political econ- omy pronounce the present era of great cities a result of the " centripetal tendencies of steam," and congratulate their readers upon the dawning of a new era wherein the " centri- fugal powers of electricity" will disperse the population of crowded tenements. Steam and machinery have certainly been among the most important influences tending toward the concentration of pop- ulation ; but neither steam nor machinery was used by the ancient Egyptians, Medes, Phoenicians, Greeks, or Romans, who nevertheless built great cities. The fact is that no one human instrument can be held accountable for such an im- portant social phenomenon, and one cannot make clear to oneself the true causes in their true relation without viewing the social body in its entirety. That is to say, a successful investigation of the causes of the city growth set forth in the preceding chapter must begin with a study of social, or more strictly speaking, economic evolution. According to Herbert Spencer, evolution consists of two distinct processes — a dififerentiation and an integration.' By differentiation, he means increase of heterogeneity out of originally homogeneous conditions. By integration, he *" Evolution is an integration of matter and concomitant dissipation of motion, during which the matter passes from an indefinite, incoherent homogeneity to a definite, coherent heterogeneity; and during which the retained motion undergoes a parallel transformation." — Spencer, First Principles, Sec. 145. CA USES 159 means a growing inter-relation and inter-dependence of parts. A simple biological analogy will serve to make clear the process. The lowest type of organism is simply an agglo- meration of like cells ; the creature is all stomach, all mouth, all hands and feet, so to speak. It has no special organs ; one part of the body is just like the other parts. The first rude diflferentiation is into two layers, which develop later into a sustaining and a regulating system. A third stage is found in the formation of organs — the heart, stomach, eye, ear, etc. — each of which assumes a single function and performs it for the entire body. But in order to effect this distribution or specialization of functions, in which each organ relieves all the others of its special work, a system of complete and intimate communication between the parts must first be de- veloped. That is to say, with differentiation, specialization, or division of labor, as we choose to call it, there must always go integration or combination. Now, without stretching the analogy, we may liken indus- trial society of to-day — embracing all countries within the cir- cle of exchange of products — to a great organism composed of heterogeneous parts. This organism, however, is the pro- duct of ages of slow growth. Originally, in place of the one all-embracing social organism, there were myriads of small social units, each complete in itself and independent of the others, if not positively hostile to them. The history of civ- ilization is simply the narrative description of the breaking down of the barriers that separated the primitive social units — the original family group, clan, patriarchal family, the en- larged village community or the manorial group. And the most conspicuous and influential role in the process was played by the trader, working upon men's desires for what they did not possess or produce. Neither war (conquest) nor religion has been of so vital and far-reaching influence in the integration and amalgamation of isolated social groups as trade and commerce. l60 THE GROWTH OF CITIES When, therefore, it is pointed out that towns owe their origin to trade, that the commercial metropolis of to-day is the successor of the primitive market-place established beside the boundary stone between hostile but avaricious tribal groups, that the extension of the market means the en- largement of the market-centre — then one will readily per- ceive the connection of the growth of industrial society to its present world-wide dimensions with our problem of the con- centration of population. The relations of transportation systems and means of communication to commerce and commercial centres will therefore form one of the subjects of discussion. The other side of the process, differentiation, involves, as we have seen, extensive changes in the units themselves. The results of territorial specialization, or the geographical division of labor, upon manufacturing and other industries will therefore require consideration. Special attention should also be given to the internal structure of industry, or the form of business organization in the various stages of evolu- tion, as bearing directly upon the problem of concentration. But first it is necessary to consider the negative side of the subject — that is, how the forces making for the dispersion of population have steadily lost ground in the evolution of soci- ety. The diminishing importance of agriculture will there- fore constitute the first topic of discussion. n. THE DIVORCE OF MEN FROM THE SOIL. If men were like other animals and had no further wants than bodily appetites and passions, there would be no large aggregations of people ; for in order to produce food, men must live either in scattered habitations like American farm- ers, or in hamlets like the ancient family or tribal group, the village community, the Russian mir, and the modern agri- cultural village of Continental Europe. Even with a com- CAUSES l6l paratively high grade of wants, men may live in these small groups, each of which is economically autonomous and self- sufficing, producing for itself and buying and selling little if anything. It is the period of the Naturalwirthschaft, in which all payments are in kind. The principle of division of labor finally led to the disruption of the village'community, but its triumph was long delayed. The principle was of course grasped only imperfectly by primitive man. At first the only division of labor was that based on sex, age, muscular power, or relation to the governing head of the group ; in other respects there was no assignment of special tasks to particular individuals. Very gradually men discovered among themselves differences of natural aptitude. The members of a community at length realized that it was more economical to have their flour made in a village mill by one member who should give all his time to that particular work, than to have it made by bits in a score of individual mills. One by one other industries have followed the mill — have departed from the separate households and taken up their abode in a central establishment. Clothing ceased to be made at home ; there arose a village weaver and a village shoemaker. To this process of development there is almost no conceivable end. Only a few years ago the American farmer not only raised his own food, but furnished his own fuel and sometimes made his own clothing. Now, however, he is a specialist, and thinks nothing of going to the market even for table supplies. Formerly, the farmer made his own tools ; now he buys implements made in factories. But yes- terday, and the men who reaped the fields of ripe grain were bound to the soil and compelled to dwell in isolated homes or small communities ; to-day these men live in cities and make machinery to reap the grain. Thus, it appears that agriculture, the industry that dis- perses men, has ever narrowed its scope. Formerly, when 1 62 THE GROWTH OF CITIES men's wants were few and simple, agriculture was the all- embracing occupation. The agriculturist produced the neces- sary sustenance, and in his idle moments made whatever else he needed. But human wants have greatly multiplied and can no longer be satiated with food-products alone. More- over, the business of providing for the new wants has been separated from agriculture. The total result is that the pro- portion of people who must devote themselves to the satis- faction of the elementary wants of society has vastly dimin- ished and is still diminishing. And this result is attained not only by the diminishing importance of bread and butter in the realm of human wants, but also by the increased per capita product which a special- ized body of workers can win from the soil. By the use of fertilizers, by highly scientific methods of cultivation, by labor-saving machinery, and by the construction of transpor- tation systems to open up distant and virgin fields, the pres- ent century has immensely reduced the relative number of workers who must remain attached to the soil to provide society's food-supply. These facts are of fundamental importance in seeking the causes of urban growth. For cities are made up of persons who do not cultivate the soil ; their existence presupposes a surplus food-supply, which in turn premises either great fertility of the soil or an advanced stage of the agricultural arts, and in either case convenient means of transportation. All three conditions were present in the river valleys of the Nile and Euphrates when the first great cities of history arose — Thebes, Memphis, Babylon, and Nineveh. No accu- rate estimates of the number of their inhabitants exist ; but as the Greeks, who had cities of their own with at least loo,- ooo inhabitants, regarded the ancient cities with wonder, it may safely be said that they were great cities. Similar con- ditions were present during the period of Roman city civili- CAUSES 163 zation. The high perfection to which the arts of agriculture had been brought (only recently approached, perhaps, in the modern era), permitted the existence of a large number of Oriental cities with a population of 100,000 or more in the first century before Christ. Two cities, Rome and Alexandria, probably attained a population of half a million souls each — a number reached by no other cities until the end of the seventeenth century (London and Paris)/ The Italian peninsula could not furnish sufificient bread- stuffs for the growing population of Rome, and had to import them from other Mediterranean countries. The difficulties of transportation, while doubtless great,^ were not to be com- pared with the difficulties of acquiring the grain by the method of legal appropriation. No very large portion of the supply was ever bought and paid for with the products of industry, for Rome was never an industrial city. The grain was regarded as one of the fruits of conquest, and was seized by the governors of the Provinces as legitimate tribute. Had Rome contained a manufacturing population, able to pay for its own food-supply, it is more than likely that the difficulties of transportation would have been over- come. It is only in this sense, if at all, that there is any truth in the contention of those writers who maintain that the fall of Rome was due to lack of means of communication with the provinces. ^ On the population of ancient cities, see Beloch, Die Bevolkerung der griechisch-rdmischen Welt. The same writer gave an excellent survey of the development of the great cities of Europe from the earliest times, in an address before the eighth International Congress of Hygiene and Demography at Budapest in 1894; cf. the Report of the Proceedings, vii, 55-61. See also his article "Zur Bevolkerungsgeschichte des Altertums " in yahrbilcher filr National- okonomie und Statistik (1897), ^^ • 37^ ■^^?' * Augustus is said once to have been on the verge of suicide out of fear lest his overdue corn-ships should not arrive. — Hume, Essay on the Populousness oj Ancient Nations, 1 64 THE GROWTH OF CITIES Within the last century the difficulties of transportation that troubled Rome so much have been solved by England, which likewise had to look to foreign countries for its food- supply. Not only that, but science has become so useful a handmaid of agriculture that the farmer, by using a little more capital, secures a larger product without adding to his labor force. The net result of progress in these two direc- tions is the removal of a surplus rural population to the manufacturing and commercial districts (cities), for "when the rural population has once become sufficiently numerous to carry on cultivation in the most profitable way, all further growth becomes disadvantageous ; whereas the materials with which the varied manufactures deal are practieally unlimited in amount and there is no other check to possible growth to such industries than the difficulty of finding markets for their products. " The improvements that have taken place in agricultural methods within the last one hundred and fifty years have been almost unprecedented ; but they have been so over- shadowed in the minds of economists and statisticians by the revolutionary changes in manufactures and transportation that they have hardly received their due share of attention. We hear often enough of the Industrial Revolution in Eng- land in the second half of the eighteenth century, but we seldom hear of the Agrarian Revolution that took place at about the same time. English agriculture had been progressing throughout the eighteenth century ; while population had doubled itself, the number of persons engaged in agriculture had decreased not only relatively, but down to 1770, actually.' But the revo- lution in English agriculture, whereby unscientific methods were replaced with scientific methods, is connected with the ^ Prothero, The Pioneers and Progress of English Farming, p. 38. CAUSES 1 5^ enclosure of the common lands, which began to be carried out on a large scale in the reign of George III, beginning in 1760.^ The results of the enclosures were the extension of arable cultivation to inferior and waste lands, the destruction of the antiquated common-field system, whereby one-third of the land lay fallow each year, the consolidation of small farms into large and the consequent introduction of the prin- ciple of rotation of crops, of roots and artificial grasses, and other improved methods.^ The common-field system, ac- cording to Arthur Young, yielded 17-18 bushels of wheat per acre, the new system of large farms 26; the fleece of sheep pastured on common fields weighed only 3^ pounds as compared with 9 pounds on enclosures,^ Bakewell ( 1 725-1 794), whom a French writer pronounces "un homme de genie qui a fait autant pour la richesse de son pays que ses contemporains Arkwright et Watt,"-* began scientific stock-breeding ; as a result of his efforts and the enclosure of the common pasture land, "during whose existence the cattle were stunted if not starved," the average size of cattle was greatly increased with scarcely any increase in expen- diture.s At the Smithfield Market the average weight of animals was In Beeves. 1710 370 lbs. 1795 800 " The consequence of all these improvements was to set free a vast number of men who found employment in the new manufacturing industries which were supplying the markets ^ Cf. the table in Prothero, op. cit., p. 257. *Prothero; cf. also Toynbee, ThelndustrialRevolution, 88-9 (Humboldt ed.). ' Gibbins, Industrial History of England, 117. * Prothero, op. cit., 49. ''Ibid.,t,i. Calves. Sheep. Lambs. 5° 28 18 148 80 50 1 66 THE GROWTH OF CITIES of the world. "The enclosures drove the laborers off the land because it became impossible for them to exist without their rights of pasturage for sheep and geese on common lands." ' At the same time the consolidation of farms re- duced the number of farmers.^ Thus, while the acreage under production and the aggregate product have greatly increased since 1750, the agricultural population has declined relatively to the industrial population. In 1770, Arthur Young estimated the total population of England at eight and one-half millions, of which three and one-half millions were agricultural and three millions manufacturing. Although the total numbers are too large, the proportions may be accepted as approximately correct.^ Comparing these pro- portions with the results of census returns in the present century, we have the following table : ^ Proportion of agr. pop. to entire pop. 1770 42 per cent. 1811 34 " 1821 32 1831 28 184I 22 " The figures since 1841 may be omitted, inasmuch as a decline in the aggregate production of cereals then began,^ ^ Toynbee, op. cit., 89. * Cobbett, writing in 1826, mentioned a single fanner who held " the lands that the now living remember to have formed fourteen farms, bringing up in a respect- able way fourteen families." (Quoted by Toynbee, op. cit., 89.) This, however, cannot be considered a typical case. Arthur Young found in i8oi that of thirty- seven enclosed parishes in Norfolk, the population had increased in twenty-four, diminished in eight, and remained stationary in five. (Prothero, op. cit., 72.) There could have been no such extensive rural depopulation as took place in the period of the sixteenth century enclosures, because now the object of the change was not more pasturage, but increased tillage. ' Gibbins, Ind, Hist, of Eng., p. 152. * Prothero, op. cit., in. ^Following upon the repeal of the corn laws in 1846. CAUSES 167 and there would naturally be a decrease in the agricultural population. But up to the middle of the century England had grown the bulk of her own wheat, and the importations were mostly confined to the years of poor crops. The table therefore shows that while in 1770 42 per cent, of the popu- lation was attached to the soil in order to produce the nation's food-supply, in 1841 but 22 per cent, of the population was required for this purpose, thus leaving 78 per cent, to settle wherever other industries determined. The increase in agricultural production might be followed out in other countries, but it would be unprofitable, inasmuch as England's progress is typical. A distinguished French authority wrote about the middle of this century that the total produce of French agriculture had doubled since the Revolution.^ But it is very improbable that the agricultural population of France increased by more than 20 per cent, in that period. In the United States, where machinery has been so exten- sively applied to agriculture, there is no lack of evidence as to the increased production per cultivator. Consider the application to agriculture of some of the most ingenious machinery invented by a nation of especial mechanical talent. Consider also the results of farming on a large scale by capitalistic methods. We hear of immense farms in the West Hke that of Dr. Glinn in California, who has 45,000 acres under wheat ; and Mulhall estimates that one farmer like Dr. Glinn, with a field of wheat covering a hundred square miles, can raise as much grain with 400 farm servants as can 5,000 peasant proprietors in France.^ It is no easy matter to estimate the saving in labor force which American methods have accomplished on the farm, but Mr. Mulhall may not be altogether out of the way. The special agent of ^ L. de Lavergne, &conomie rurale de la France depuis lySg, 2d ed., p. 59. * Progress of the World (1880), p. 499. 1 68 THE GROWTH OF CITIES the Tenth Census, who reported on Interchangeable Mechan- ism, made a somewhat more conservative estimate. Mr. Fitch said : " It is estimated by careful men, thoroughly- conversant with the changes that have taken place, that in the improvement made in agricultural tools, the average farmer can, with sufficient horse-power, do with three men the work of fifteen men forty years ago, and do it better." ^ Nor is this all of America's contribution ; her improvements in the raising of grain and production of meats have been matched by her improvements in the handling and marketing of the same supplies ; by cutting down the wastes of distribu- tion, such improvements reduce the amount of agricultural labor needed. The importance of transportation in bringing new lands into the area of cultivation for the world market has already been mentioned. But transportation, in connec- tion with the other modern mechanics of exchange, econo- mizes the amount of agricultural labor by diminishing the need of over-production, and thus equalizing the supply in space and time. The auxiliaries of transportation are char- acteristically American methods of centralization.^ Instead of the multitude of small buyers who conduct the grain busi- ness in Europe, America has concentrated the trade in the hands of a smaller number of large firms who have built the great grain-elevators and developed the system of handling grain in bulk, which, as much as anything else, has enabled American grain to enter the markets of the world. In no other field have American business ability and capacity for organization produced greater economies; for along with the improvement in the technical means of handling grain, there has been developed a more efficient system of credit, of buying and selling (produce exchanges ! ), etc.3 The grand ^ zoth Cen.y Mfs. (Section on Agricultural Implements), p. 76. * See Sering, Die Landwirthschaftliche Konkurrenz Nordamerikas, 491 seq. ' For example, the grading of wheat whereby it is bought and sold without the use of samples. CAUSES 169 result is an equal distribution of the food supply to all coun- tries within the area of exchange. Local surpluses are abolished. Local scarcities are also abolished and therewith the fear of famine, which necessitated the maintenance of stores of grain. Both the surpluses and the stores denoted over-production, or the occupation of more laborers in till- ing the soil than society really needed. They are now free to settle where they choose. III. THE GROWTH OF COMMERCIAL CENTRES. [Authorities. — Easily the best survey of economic evolution is Professor Karl Biicher's essay on Die Entstehung der Volkswirthsckaft (Tubingen, 1893), a brilliant piece of work which ought long since to have been translated into English. Professor Schmoller's interpretation of economic history differs little from Professor Biicher's, and one of his studies of " Die Wirth- schaftspolitik Friedrichs des Grossen," which contains his views, has been published in English as one of Professor Ashley's " Economic Classics " under the title of The Mercantile System. With this should be read Herbert Spencer's treatment of " Industrial Institutions " in the third volume of his Principles of Sociology. The general features of economic organization are described in such works as Cunningham's Growth of English Industry and Commerce (2 vols., Cambridge, 1890-1), and Ashley's Intro- duction to English Economic History (2 vols., New York, 1888-93). The town economy will be understood by readers of Mrs. J. R. Green's Town Life in the Fifteenth Century (2 vols.. New York, 1894). The patriarchal family is familiar to Americans from the works of Sir Henry Maine {Ancient Law and Village Cotnmunities), Heam, (The Aryan Household) , de Cou- langes (The Ancient City), and W. W. Fowler (The City State of the Greeks and Romans). For the abundant literature on the mark and the manor, the reader is referred to the bibliographies in Ashley. The origin and location of towns is treated in a philosophic manner by Roscher in the introduc- tory chapter of his work on commerce and industry ; Systejn der Volkswirthschaft, dritter Band, Die Nationaldkonomik des Handels und Gewerbefleisses, " Einleitung — Aus der Natur- lehre des Stadtewesens im Allgeimenen." Roscher further develops the theory of the location of cities in an essay " Ueber die geographische Lage der grossen Stadte," published in his A nsichten der Volkswirthschaft, vol. i. Cf. the work by J. C. Kohl, Die geographische Lage der Haupi- stadten Europas,a.iiA. the latter's standard work, Der Verkehr und die Ansiedelungen der Menschen in ihrer Abhdngigkeit von der Gestaltung der Erdoberflciche, 1843. This was written before the era of railways, however, and is in some respects superseded by the later works: E. Sax, Die Verkekrsmittel in Volks- und Staats7uirthschaft,'V\enxia., 1878; A. de Foville, De la Tratisfortnation des Moyens de Transport et ses Consequences economiques et sociales, Paris, 1880; the most recent treatment is by a young American, Dr. C. H. Cooley, whose book (The Theory of Transportation, in Publications of the American Economic Association) will interest the general reader as well as the trained economist; it is by far the best study of the sub- ject in English, but, like the others, is written from the standpoint of transportation. Sir James Stewart devotes a chapter of his Inquiry into the Principles of Political Economy to this sub- ject (" What are the Principles which regulate the Distribution of Inhabitants into Farms, Vil- lages, Hamlets, Towns and Cities ") Chap, ix in vol. i of the Works of Stewart, edited by Gen. Sir J. Stewart, London, 1805). More recently the subject has been approached from the side of geography, e. g., Ratzel's Anthropo-geographie, vol. ii, §§ 12-14, in which connection may also be mentioned a paper by Prof. W. Z. Ripley on " Geography and Sociology " (with bibliography) in the Pol. Sc. Quar., x, 636-53.] I^O ^-^-^ GROWTH OF CITIES In point of time, the earliest economic force working for the concentration of the released agricultural population was trade. Without trade and commerce, indeed, no towns or cities could come into existence ; for they presuppose a non- agricultural population which buys its food-supply. When a portion of the tillers of the soil abandon agriculture and obtain their food-supplies by the exchange of other products of industry, the foundation of trade is laid ; the specializa- tion of functions, or the division of labor in its broad sense, has then begun its course, which tends to an ever-widening circle of exchange and enlargement of the commercial centres. The economic historians, in tracing the develop- ment of industrial society, are accustomed to distinguish three or four periods, in which the predominating types of organization were, respectively, (i) the household or village economy, (2) the town economy, (3) the national economy, (4) the international economy, toward which we are ap- proaching as the ultimate goal. The first period — repre- sented by such diverse human groups as the patriarchal family, the village community, mark or inir — is much the longest, and endured in Europe to about 1000 A. D.^ The characteristics of the village economy are generally familiar since the time of Sir Henry Maine. " The village is an economic and commercial system complete in itself and closed against the outside world." Salt, iron, and in the later stages, tar and millstones, were the only commodities brought into the village from the outside.^ ^ That is, in the advanced countries. It lasted until quite recently in Russia. Cf. Hourwich, Economics of (he Russian Village (Columbia Studies). * The Mercantile System, 5, 6. " It is hardly possible for [a villager] to come into closer intercourse with outsiders; for to remove any of the products, whether they may be derived directly or indirectly from the common land, is forbidden." And Prof. Thorold Rogers says : " The trader did not exist in the villages. In most villages he hardly existed at the beginning of the present century. In my native village the first shop was opened, for general trade, about 60 years ago." (^Six Centuries of Work and Wages, p. 147.) CA USES 171 Trade was in fact impossible so long as villages were hostile to one another ; but even then, as Cunningham notes,^ its advantages were so clearly felt that the boundary- place between two or more townships came to be recognized as a neutral territory, where men might occasionally meet for mutual benefit. The boundary-stone was the prede- cessor of the market-cross and the neutral area around it of the market-place. Trade was also promoted by religious assemblies. From very early times, says Cunningham,'' men have gathered to celebrate the memory of some hero by funeral games, and this has given the occasion for meeting and trading ; so that fairs were held annually at places of burial. Mediaeval towns grew up around shrines and monasteries built at the graves of early martyrs.^ Other towns grew up around forts or the castles of feudal lords. It is remarked that many of the earliest English towns (leav- ing out of account the cities of the Roman period), were founded by the Danes, who were noted traders.* During the invasion of the Northmen, the garrisons of both Danes and English became nuclei of towns. But the most common origin of American and English towns is the primitive agricultural village, or a coalescence of several villages. Thus the latter origin is attested in numerous cities by the survival of several agricultural functionaries like haywards, pinders, molecatchers, etc.s And since facilities for trade were the primary cause of the develop- ment of village communities into towns, it follows that those villages which were situated at fording places, in the midst of a fertile plain, or on good trade routes, would be favored ^Vol. i, p. 76. * Ibid., i, 90. * On the origin of ancient towns, see E. Kuhn, Ueber die Entstehung det Stadte der Allen, Leipzig, 1878. * Cunningham, op. cit., i, 88. * Ibid., i, 23. 172 THE GROWTH OF CITIES in growth. We are not especially concerned in this essay with the principles of city-location, and it will suffice merely to indicate the fundamental theory. It may, then, be stated with some confidence that while certain cities derived their location in former ages from proximity to a fort or a religious establishment ; while many modern cities have had their location determined by political reasons {e.g., Washington and many of our State capitals) ; while numer- ous cities in all periods have arisen in the vicinity of mines or other riches of the earth which furnished natural advan- tages for production — yet, nevertheless, the prevailing influ- ences in determining the location of cities are facilities for transportation. The greatness of an inland city will depend on the size of the plain for which it is the natural centre of distribution, and in a second degree on the fertility of soil, which determines the number of inhabitants in the plain.^ The factor of chief importance in the location of cities is a break in transportation. A mere transfer of goods will require considerable machinery; and so we find commercial centres at the confluence of rivers, head of navigation, fords, meeting-point of hill and plain, and other places where the ^ Paris became the metropolis of France because it was the centre of a great plain including more than one-half of France; modern transportation has perpet- uated its natural advantages. In new countries, however, transportation tends to emancipate cities from this condition of dependence. Botzow (" Bodenbeschaf- fenheit und Bevolkerung in Preussen," Zeitschrift des preussischen statistischen Bureaus, xxi, 287-91), aims to show that density of population is largely deter- mined by the fertility of soil, even in the industrial stage. The number of in- habitants to each 100 square kilometres of territory (excluding water surface) for the rich and the poor soils was : Good soils. Increase per 100 inhabitants. 1819. 1849. 1819-49. 11849-75. Urban... 1,259 1,891 150 170 Rural ... 4,313 6,184 143 125 Prussia.. 5,572 8,076 145 135 2,891 4,221 146 117 Poor : SOILS. Increase per 100 inhabitants. I8I9. 1849. 1819-49. 1849-75. 170 260 153 145 2,723 3.961 145 "5 CAUSES 173 physical configuration requires a change of vehicle. But the greatest centres will be those where the physical transfer of goods is accompanied with a change of ownership ; there is then added to the mechanical apparatus of temporary storage and transfer, the complex mechanism of commercial ex- change. Importers and exporters, merchants and money- changers accumulate vast wealth and require the presence of other classes to satisfy their wants, and population will grow rapidly. It is therefore easy to understand why so many of the large cities of the world are commercial centres, if not actual seaports. Every great city owes its eminence to commerce, and even in the United States, where the rail- ways are popularly supposed to be the real city-makers, all but two of the cities of 100,000 or more inhabitants are situ- ated upon navigable waters;^ the most rapidly growing cities of their class in the country are the lake ports, Chicago, Buf- alo, Cleveland, Detroit, Milwaukee, etc.^ One would expect that, in a country where cities have been located since the advent of railways, the matter of water communication would be of minor importance. That this is not the case shows the one-sidedness of the reasoning that steam alone is the funda- mental cause of the concentration of population.3 In Europe, ^ The exceptions are Indianapolis and Denver. In its early days, however, In- dianapolis was the seat of a considerable river trade. Denver, which is situated on the south fork of the small Platte river, is a great railway and distributing centre, reaching both the mining regions of the mountains and the agricultural districts of the plains. It is at once interesting and instructive to study a topo- graphical map showing, along with the physical features of the coimtry, the loca- tion of the principal cities. (Such maps may be found in the Statistical Atlas of the United States, the Census Reports on Transportation, and Cramps Universal Atlas, 1897, p. ZZ^ ' ''■ In his suggestive study of the " Density and Distribution of Population in the United States at the Eleventh Census " (^Economic Studies of the American Eco- nomic Assn., 11,448), Prof. Willcox emphasizes the fact that the most rapidly growing cities are the commercial centres, and especially the lake ports. ' There are a number of cities in central New York — Rochester, Utica, Syra- 174 THE GROWTH OF CITIES too, the traveller is struck with the frequency with which the blue strips of water appear on the street maps of cities in his guide books. That not all the large cities of Europe are situated on navigable rivers, lakes or seas, is due not to their recent foundation as railway centres, but to their original foundation for political or military reasons. New York's primacy depends upon her location at the junction of land and water transportation ; in New York occurs the change of ownership and transfer of goods in the commerce be- tween Europe and the United States. If the water route could be extended inland to Chicago by means of a ship canal, Chicago would become the terminus of European commerce, and in the course of time would with scarcely any doubt take from New York the rank of commercial and financial centre of the New World, and prospectively, of the globe. But, to return to the history of the growth of commerce, the primitive trading point at a fort, monastery, ford, castle or harbor, might long have remained a mere market-place or fair-ground, if the trading class had not attracted to it other industries, thus bringing about the division of labor. The earUest realization of this principle was in ancient times when the paterfamilias abandoned his farm and a portion of his dependents, and betook himself with another portion of his cuse, etc. — that owe their early growth to the Erie canal; the steam engine has but perpetuated them. And indeed the influence of steam on transportation methods is frequently exaggerated. Ocean commerce expanded rapidly before the application of steam to freight carriers. It is only two decades back that the marine engine was perfected; prior to 1875 ^^ ocean steamship had not been a formidable competitor of the sailing vessel. Even to-day sailing vessels constitute a large part of the shipping of the United States, especially in the coastwise trade. According to the Eleventh Census, 41 million tons of freight were carried in steamers along the Atlantic coast and Gulf of Mexico in 1889, and 40 million tons in sailing vessels. {^Compendium, \\\,OiOt^^ Finally, it should not be forgotten that the Roman world witnessed the foundation of great cities^without the aid of steam, thus testifying to the possibilities of water transportation routes. CAUSES 175 slaves or dependents to the city. The Athenian lived upon food sent to him from his own Attic estate outside the city walls, and had almost everything else made for him by his own slaves. In the later Roman times the division of labor had been so far developed that some 150 different occupa- tions were carried on by the slaves of a single family/ In more recent times, when the original social unit was the vil- lage community of freemen, it was the artisan who abandoned the soil and went to live in the town. It is hardly necessary in this place to trace the process by which the members of the village group became differentiated into farmers and artisans. One member of the household after another, pos- sessing some particular gift of skill or strength, found it more advantageous to devote his whole time to one kind of work than to take his turn at the various tasks of the household. One man perhaps would begin by making all the shoes for the household ; having by long practice acquired more than the usual skill at shoemaking, he would offer his services to other households, travelHng about like a modern umbrella- mender or the village dress-maker. At length, instead of carrying a few tools about with him he would learn that he could do better by settling down in one place and augment- ^ Prof. Biicher {op .cit., 25) quotes a Dutch work of the seventeenth century as enumerating 146 different occupations, and adds that modern scholarship has re- sulted in additions. The progress made in the separation of employment since then may be seen in the following figures giving the number of trades practiced at Frankfort, Germany, in the Middle Ages, compared vcith some ancient and modern figures in other lands : Rome 10—20 Greece, 337 A. D 35 Frankfort, 1387 148 " 1440 ■ 191 " 1500 300 China, 1890 350 Germany, 1882 4>785 (SchmoUer, " Die Thatsachen der Arbeitsteilung," Jahrbuch fm- Gesetzgebung, Verwaltwtg und Volkszvirscka/t,-xm, 1045.) 176 THE GROWTH OF CITIES ing the number of his tools. And thus the shoemaker, the smith, the weaver, the dyer, the brewer, the bricklayer, the carpenter, accumulated capital and settled in the town. Such is the process of development of the free handicraftsman of the middle ages. The differentiation between town and country is an im- portant event in economic history, for it marks the separa- tion of industry from agriculture.' Adam Smith could say that " the great commerce of every civilized society is that carried on between the inhabitants of the town and those of the country."' But while the towns thus carry on trade with the surrounding agriculturists, each town is as tightly closed against other similar groups as was the original house- hold or village. This self-sufficing character of the mediaeval town is not easily realized by the American of to-day, unless he has traveled in some backward country of Europe where the towns close their gates at nightfall and levy octroi duties on incoming merchandise. Perhaps Prof. Schmoller's de- scription of town policy will aid in the realization of the sep- arateness of mediaeval communities : " Except during a fair, the foreigner was excluded from all retail trade, allowed to remain only a certain time, and prohibited from lending money to or entering into partnership with a burgess. He was burdened with heavier dues or fees for setting up a stall, for having his goods weighed, and for the services of brokers and exchangers. ... In short, the town market formed a complete system of currency, credit, trade, tolls and finance, shut up in itself, and managed as a united whole and on a settled plan ; a system which found its centre of gravity ex- clusively in its local interests, which carried on the struggle ' Karl Marx says that " the foundation of every division of labor that is well de- veloped and brought about by the exchange of commodities is the separation be- tween town and country." Capital, p. 212 (Humboldt edition). So also Adam Smith, Bk. iii, ch. i. CAUSES lyj for economic advantages with its collective forces, and which prospered in proportion as the reins were firmly held in the council by prudent and energetic merchants and patricians able to grasp the whole situation." ^ Nor was the solicitude of the mediaeval town devoted to its market alone ; it was equally concerned in preventing the contributory territory from trading with other towns. The cattle and dairy products must be sent to the town-market in order to keep down the cost of living of the townspeople ; it would not do to have their produce go to other towns. Not only must the farmers not buy goods from any other town, but they must not make them for themselves ; just as in the i8th century the colonial system expressed the pohcy of the mother country to do all the manufacturing and to con- fine the colonists to the production of raw materials, so the town economy of the middle ages aimed to take away from the country every competing industry that it possibly could. In Germany, when the towns were not curbed by a royal hand, they forbade the countrymen to brew beer, etc. Now it is sufficiently obvious that, in an age when com- merce was chiefly confined to the trade between a town and the surrounding country, the commercial centres would not attain large dimensions ; and such was the character of the commerce that existed down to circa 1500 A. D., a fact apt to be forgotten when one thinks of the glory of Venice and the other Italian cities, which rested in the main upon foreign commerce. Nevertheless, this foreign commerce was con- cerned with draperies, silks, spices, gems, and other luxuries, and was really small in comparison with the annual com- merce between the towns and the surrounding country. The transition from the town economy to the national economy is often lost sight of in contemplation of the con- ^ The Mercantile Systevi,-^. ii. Cf. Ashley, Introd. to Eng. Econ. Hist., vol. ii, bk. ii, ch. i, on the " Supremacy of the Towns." 178 THE GROWTH OF CITIES temporaneous political changes, and it is forgotten that the central monarchy was but the outward expression in the political sphere of the triumph of new economic forces. The towns had been acquiring wealth with considerable rapidity before the Crusades began, but during the Crusades they absorbed a large share of the riches of feudalism, as a result of the wiUingness of the nobles to part with their estates and furniture in exchange for ready money with which to pay the expenses of their trips to the Holy Land. The cities then bought charters of freedom, and finally, through their alliance with the king, gave the death-stroke to feudalism. This happened in France in the reign of Louis XL (1461- 1483), and in England at the conclusion of the Wars of the Roses and the accession of Henry Tudor in 1485. Industrial progress in the towns was preparing the way for that national unity which began to be realized with the dis- appearance of feudal sovereignty, and attained its object under the Mercantile system of Cromwell, Colbert and Frederick the Great. As far as England was concerned, a national economy might have existed in the fourteenth cen- tury. A national Parliament arose under Edward L and soon accomplished national legislation that broke down local privileges ; but there was no real national economic devel- opment until the end of the fifteenth century, and until then industry did not need a wider organization than the town community.^ The growth of business enterprise and mer- cantile wealth conditioned the progress of the arts of produc- tion. But the entrepreneur, undertaker, adventurer, had made his appearance in English agriculture before the end of the fourteenth century, after the Black Plague (1348) had abolished serfdom and substituted money payments. The same spirit made its appearance in commerce. The Cru- sades not only transferred vast wealth from the nobility, who ^ Ashley, op. cit., ii, 89. CA USES J 70 seldom used it productively, to the burgesses of the towns, with whom it became capital, but stimulated commerce by the creation of new wants and tastes for Oriental luxuries. The annual visit of the Venetian fleet, laden with Oriental tapestries, jewels and spices, was a great event for England and the other nations of western Europe. These commodi- ties of foreign commerce were, to be sure, mainly luxuries. The arts of transportation at that time were adequate only for articles that were at once light, durable, valuable, and, therefore, purchasable by the few alone. But with the growth of commercial enterprise and wealth, trade began to embrace other goods, until it no longer remained true .hat the commodities used by the mass of the people were pro- duced near the spot where they were consumed.^ Already in the fourteenth century geographical specialization took place in wool ; for, after the Black Plague, England became more and more devoted to sheep-raising, and sent her wool to Flanders to be made into cloth.^ Not only did Flanders thus early become noted for its woollen fabrics, but Strass- burg and other Alsatian towns in the same period drove the clothmakers out of Basel and other neighboring cities by their competition.3 Between 1450 and 1550 the cloth manu- facture of Germany was concentrated in places peculiarly fitted for it.'^ Gradually the weekly town market, in which the wares sold were almost entirely local products, gave way to the great annual fairs, which brought together the pro- ducts of many cities and countries.^ ' Cf. SchmoUer's essay, " Der moderne Verkehr iai Verhaltniss zum wirthschaf t- lichen, socialen und sittlichen Fortschritt " (1870), in Ziir Social- unci Gewerbe- politik del Gegemvai-f. ^Gibbins, Industrial History of England, 48. ^ SchmoUer, " Die Thatsachen der Arbeitsteilung," in his Jahrbuch, 13 : 1066. * Schmoller, The Mercantile System, 32. ^ The weekly local market began to give place to the fair in Germany about 1506 (SchmoUer's yaki-buch, 13: 1066), and the culminating point of the great l80 THE GROWTH OF CITIES But as the towns thus came into closer connection, their rivalries led to innumerable petty conflicts which could be reconciled only by a superior authority. Town leagues were tried, and after their failure, the current set strongly toward nationalism.^ Mercantilism is simply the attitude of this movement toward foreigners ; the great Mercantilists were also engaged in sweeping away the internal barriers to nationalization. Colbert, whose name is so prominently identified with the mercantile policy that Colbertism has be- come a synonym of Mercantilism, built roads and canals, promoted technical and artistic education, and worked for a uniform customs system by reforming the river tolls, etc., within France, his aim all the while being to break down municipal and provincial autarchy and make of the French people a united nation.^ Frankfort fair was reached later in the same century (Biicher, Entstehutig der Volkswirthschaft, 72). Until the era of these interstate or international fairs, the mediaeval town, instead of importing goods from rival towns, had pursued the policy of importing skilled workmen, who thus became, according to Prof. Biicher, greater migrants than the artisans of these days of railways. (Essay on " Die in- neren Wanderungen und das Stadtewesen in ihrer entwicklungsgeschichtlichen Bedeutung," op. cii., 302, ff. Cf. also Schmoller, op. cit., xiii, 1066.) ^ " With the transformation and enlargement of commerce, the growth of the spirit of union and the consciousnesss of interests common to whole districts; with the augmented difficulties in the way of a proper organization of economic life on the basis merely of town and village interests, and the increasing hopeless- ness of victory over the anarchy of endless petty conflicts, efforts and tendencies everywhere made their appearance toward some larger grouping of economic forces." Schmoller, The Alercanlile System, 13. ^Cf. Smoller, op. cit., 54-5 : "The great laws of Colbert .... more important than the tariffs of 1664 and 1667 .... founded the legal as well as economic unity of France." Prof. Schmoller was one of the first investigators to seize upon the nation-building movement as the essence of Mercantilism; see his " Studien iiber die wirthschaftliche Politik Friedrichs des Grossen," in his Jahrbuch, 1884-8. Prof. Cunningham's great work, Growth of English Induitry and Coi/imerce, takes a similar view of the mercantile policy, as does also that brilliant historical essay of the late Prof. Seeley, The Expansion of England. Cf. also Biicher's Entstehimg der Volkswirthschaft, 70 ff. CA USES 1 8 1 The nation, however, did not remain a self-sufficing indus- trial society. Mercantilism fulfilled its mission as soon as it unified the nation and gave national production a good start. When once it had produced a sufficient diversification of industries, it was permitted to lapse, and the nation gave itself to the production of those commodities for which it was particularly suited. Protectionism, the child of Mercantil- ism, has had a strong influence upon the industry of the United States ; but no American statesmen have entertained the ambition of making the country absolutely self-sufficing. While protecting the infant manufactures of the United States against the competition of the established industries of Europe, they have kept the door wide open to the admission of tropical products. It may therefore be stated as a general truth that the whole world forms to-day a single industrial society. The effect of this gradual enlargement of the economic society upon the growth of cities must be clear to every student. How different is London, a local market for the agriculturists of Essex and Middlesex, from London, the world's financial and commercial centre ! Or take the Dutch as a representative commercial nation, and compare their cities with those of their neighbors, the Belgians. Belgium is a great manufacturing country, consuming more coal and and iron per capita than any other European country except Great Britain. Holland has few manufactures, but carries on an extensive commerce.^ It is also less densely populated ^Of the Belgians, 57 per cent, are engaged in manufacturing industry, the per- centage in England being at the same date (1881) only 55 (SchmoUer, " Die Thatsachen der Arbeitsteilung," in his Jahrbuch, 13: 1073). But statistics of occupation are difficult to compare, and for the Netherlands none comparable are at hand. The estimated annual value of manufactures in Belgium is 102 millions sterhng, in Holland 35, being a rario of 3:1, while the population of the two countries stands at the ratio of 4 : 3. (Hobson, op. cit., 87.) On the other hand the Netherlands have the largest foreign trade in proportion to population of any 1 82 THE GROWTH OF CITIES than Belgium.^ Those who maintain that cities are the pro- duct of manufactures would expect to find a much larger urban population in Belgium than in the Netherlands. The actual percentages are as follows : Netherlands. Belgium. Townships of 2,000+ ca 85. ca 75. " 10,000-1- 43. 34.8 " " 20,000-}- 31.3 26.1 The disparity of the percentages in favor of Holland might in the case of the townships having only 2,000 inhabitants, be due to the larger size of the Dutchtownship, which would thus include more of the scattered population, not really urban.^ But in the case of the larger cities, the scat- tered population is too small to be a factor of any influ- ence. The plain inference from the comparison is that commerce is one of the main causes of the concentration of population in large cities. The analysis of the growth of English cities after the opening up of railways also confirms the hypothesis. It has been shown above (pp. 54-56), that nearly all of the cities which attained their maximum rate of increase in 1 841-51, the decade of railway building, were seaports; the manufacturing centres having reached their highest rate of growth at the time when steam was applied to stationary machinery. country in the world, averaging in the eighties over $200 per annum for each in- habitant, while the Belgian average is about ^90. (Neuman-Spallart, Uebersich- ten der Weltwirthschaft, 1 883-4, p. 549.) Hobson {op. cit., 116) gives a map of the foreign trade of European nations; Holland is in a class by itself, while Great Britain, Belgium and Switzerland fall in the second class. ^ The average size of the Dutch Gemeente is 28.7 square kilometres, of the Bel- gian commune, 11.4 square kilometers. (P. 142.) Population to Population. Area. i sq. km. ■^Belgium (1890) 6,069,321 29,456 sq. km. 206. Holland (1889) 4.511.41S 32,538 " 138.7 — Statistik des Deutschen Heiches, Neue Folge, Bd. 68, p. 6*. CA USES 183 As has been said, the growth of commerce is an accompan- iment of the speciaHzation of functions, or territorial division of labor. Recurrence to the biological analogy will serve to make clear this process. Primitive society consists of the household group, which may be likened to the lowest type of organism ; each is a composite of like cells. The differentiation of the organism into an inner and outer layer is parallelled by the first differ- entiation of society into town and country. Finally, there comes the formation of special organs and a distributive sys- tem — in other words, the agglomeration of population in centres of collection and distribution, and a system of trans- portation. To a highly developed society such centres of mass and force are as necessary as are its organs to any organism of the higher type. Economy of force, which is always the aim of specialization, is impossible of attainment without concen- tration. In the economic organism, transportation is the distributive system ; differentiation is therefore impossible without improvements in this system. The better the trans- portation, the higher the specialization and the greater the concentration. The close connection between transportation facilities and the territorial division of labor has led some writers to regard transportation as the sole agent in determining the distribu- tiou of population, its concentration in cities and the location tion of cities.^ The truth doubtless is that transportation is one of the conditions, one of the external causes. The development of transportation proceeds pari passu with, the evolution of the social body ; as the differentiation progresses, transportation, like the distributive system in an organism, must be able to assume increased burdens. The growth of ' " The whole matter of the distribution of population, wealth and industries over the face of the earth is, in one of its aspects, a matter of transportation." Cooley, Theory of Transportation, 73. Cf. also Ratzel, Auikro-geogrophie. 1 84 THE GROWTH OF CITIES commerce in the eighteenth century, due in large part to the growth of the American market, opened unlimited opportuni- ties to the English manufacturer and brought out inventions that revolutionized the textile industry. The new machinery required for its highest efficiency a more regular power than that afforded by mill streams, and Watt invented the steam engine. Production therefore increased hundred-fold, and laid upon the canals and turnpikes impossible tasks ; it called for improved means of transportation, and the railway came into being. As Hobson justly remarks, the history of the textile inventions does a good deal to dispel the " heroic" theory of invention — that of an idea flashing suddenly from the brain of a single genius and effecting a rapid revolution in trade. It should rather be said that inventions result from the inner " pressure of industrial circumstances which direct the intelligence of many minds towards the comprehension of some single central point of difficulty."^ IV. THE GROWTH OF INDUSTRIAL CENTRES. [Authorities. On the evolution of industry, Biicher is again the foremost authority, and no serious student of economics can afford to neglect his essay on " Die gewerblichen Betriebssysteme in ihrer geschichtlichen Entwicklung" (in Enstehung der Volkswirthschaft, Tubingen, 1893), and the article " Gewerbe," in Conrad's H a>idwdrterbuch . Cf., also Held, Zwei Bucher zur sozialen Geschichte Englands, 541 and Erster Anhang (entitled "Handwerk und Grossindustrie), and SchmoUer's Klehigewerbe; while the Domestic System is portrayed in Report from the {H. C.) Co7nniittee on the Woollen Manufacture of England, 1806, pp. 3, 6, 8, 9, 10. The evolu- tion of industries and towns in the United States may be studied in Weeden's Economic and Social History of New Etigland, 2 vols., Boston and New York, 1891, esp., pp. 73, 271, 304, 305. (The development of Lynn as a typical iactory town may be followed on pp. 308, 682, 735, etc.) The Factory System and Centralized Industry, or production on a large scale, are treated in the standard works on political economy; see especially. Mill, Pri7tciples of Politieal Econ- iJ^^y, Bk. I., chs. 8, 9; MarshaW, Principles of Economics ; WaW^x's Political Economy ; Had- ley's Economics ; the two American writers giving particular prominence to the undertaker or entrepreneur. The historical side is treated by Carroll D. Wright in his essay on the Factory System in Rep. on the Mfs. of the U. S. at the Tenth Census, and also in his little book en- titled The Industrial Evolution of the United States; lucwasscnr, VOuvj'ier Americain, 2 vols., Paris, 1898; Stieda, Art. " Fabrik" in Conrad's Handworterbuch ; Marx, Capital ; Hob- son, The Evolution of Modern Capitalism ; Schulze-Gavernitz, Grossbetrieb or The Cotton Trade; V^eS^s, Recent Economic Changes; Atkinson, The Distribution of Products ; Schon- hof, The Economy of High Wages, and The Indjtstrial Sitjtation : Mataja, Grossmagazin und Kleinhandel; SchmoUer, " Wesen und Verfassung der grossen Unterjiehmungen" in Zur '^Evolution of Capitalism, 57. Cf. Spencer, Social Statics (1893), p. 72; Schulze-Gavernitz, Social Peace, 15, CAUSES 185 Social- und Gewerbepolitik der Gegenwart ; also " Die Entwicklung des Grossbetriebs und die sociale Klassenbildung," in Pretissiscke Jahrbiicher, LXIX, (1892), Heft 4; Sinzheimer, Die Grenzen der Weiterbilditng des fabrikmcissigen Grossbetriebs, a monograph in Brentano's Munchener Volkswirthschaftliche Studien ; an interesting popular exposition of the manage- ment of large industries was given in a series of articles on " The Conduct of Great Businesses," in Scribner's Magazine, 1897. The most luminous treatment of the principle of the division of labor is Biicher's "Arbeitsteilung und sociale Klassenbildung," one of the essays appearing in Die E7itstehung der Volkswirth- schaft : Prof. SchmoUer has also made noteworthy contributions to the subject, the most import- ant being two articles in his journal, ( Jahrbuchfilr Gesetzgebung, Verwaltung -und Volkswirih- schaft, vols. XIII and XIV), entitled respectively " Die Thatsachen der Arbeitsteilung," and " Das Wesen der Arbeitsteilung;" see also Hackel, Arbeitsteilung in Menschen und Thier- leben ; Spencer, Princij>les of Sociology, es^. Part VIII, ch. II; Bagehot, Physics and Politics; Schaffle, Bau und Leberi des Socialen Korpers, as well as his text-book of political economy. Das gesellschaftliche System der menschlichen Wirthschaft ; Simmel, Ueber sociale Differ- encirung ; Durkheim, Xa Division du Travail ; T^osch, JVationale Production und ftatiofiale Bent/sgliedernng ; and the leading text-books — Mill, Walker, etc.] In the foregoing section it was shown how economic devel- opment, or the integration of isolated social and economic groups, demands the concentration of a portion of the popu- lation in commercial cities. Similarly, it may be shown how the enlargement of the market, which is one aspect of the process of growth of industrial society from the village econ- omy to the world economy, has brought about centralization in the manufacturing industries and enforced the concentra- tion of another portion of the population in industrial, or perchance commercial, cities. A brief review of industrial evolution will suffice to show the importance of the size of the market with regard to the business organization or inter- nal structure of industry. In the evolution of industry, four principal stages may be clearly distinguished : ( i ) the household or family system, (2) the guild or handicraft system, (3) the domestic or cot- tage system, (4) the factory system or centralized industry. Under the household regime, each family manufactures its own supplies, and there is no buying and selling. The village community and manorial group are but enlarged families, and we have already pointed out their industrial autonomy. The advent of the trader and the introduction of money caused the disintegration of the village by dififerentiating its 1 86 THE GROWTH OF CITIES members. The family shoemaker, instead of helping on the farm and making shoes for the other members of the family at odd moments, began to travel about working for other families like the itinerant village dressmaker still to be met with in many parts of the country. At length he would have accumulated sufficient capital to buy his own materials, and would then set up a shop and make shoes to order. In locating his shop he would naturally choose the most cen- tral site, and would therefore in all likelihood become a neighbor of the trader in the town or a large village. Thus the handicraft system of industry and the town economy are different aspects of one stage of economic evolution. Both lasted to about the middle of the fifteenth century. In the course of time the primitive shoemaker's business would have increased so much by reason of the growth of population and the extension of the market, that he would feel himself impelled to employ assistants or apprentices. But before enlarging the capacity of his own shop, which would probably require considerably more capital than he could command in those early times, he would be more likely to place orders with some of his fellow-craftsmen who lacked his business talent and enterprise in attracting custo- mers or in accumulating capital. This new division of labor, involving the differentiation of employing or wage-earning classes (the capitalists, undertakers, business men on the one hand, and the laborers or " hands" on the other), is the beginning of modern Capitalism. Logically, it is only a short step hence to the factory system, wherein the employer brings together the workers under one roof. But it required immense improvements in means of communication to replace the local market, in which the handicraftsmen made the vast majority of things to order, with a wider market wherein the master (eventually the capitalist) sold his ready- made products to whatever buyers appeared. Historically CA USES 187 it required nearly three centuries to make a modern capital- istic employer out of the mediaeval master handicraftsman employing one or two journeymen and apprentices. The first middleman in manufactures was the mediaeval clothier, who owned no buildings, but bought the raw materials, dis- tributed them among the weavers, and sold the cloth in the market. The factory system was made possible by the gradual enlargement of the market;^ its triumph was assured by the invention of power-machinery in the eighteenth century and the development of the modern systems of transportation and communication in the nineteenth century. The coun- tries most energetic in introducing the new improvements in means of communication are the countries that have carried the factory system to its highest development. The tendency toward production on a large scale is too familiar a fact to the American to demand statistical proof ; it is sufficiently illustrated in the statement that the average number of em- ployees to an establishment in the textile industries has increased as follows:^ 1850 48 i860 64 1870 58 1880 95 1890 124 In the surrender of the small producer to the corporation and trust lies, of course, the explanation of the decay of vil- lage or local industries carried on under the antiquated ^ " Communication is the outer vehicle, commerce the inner soul which gave the impetus to centralized industry." Schmoller, in Prtussische yahrbucher, voU Ixix, pt. 4. '^ nth Cen., AJfs., Part III, p. 3. Cf. ibid., Part I, p. 4. While in 1850 the average amount of capital to each manufacturing establishment reported was ^4,300; in i860 it was $7,100; in 1880,^x1,000; in 1890,^19,000. The aver- age number of employees rose from 8 in 1S50 to 14 in 1890. 1 88 THE GROWTH OF CITIES handicraft regime. The story of the decHne of the villages in the United States has been told by Mr. H. F. Fletcher in an article entitled " The Doom of the Small Town," and pub- lished in the Forum, April, 1895. The number of village saw-mills, flour and grist-mills, establishments devoted to furniture and cabinet-making, and the manufacture of agri- cultural implements, brick and tile, etc., has perceptibly declined. Mr. Fletcher also investigated the statistics of population of the villages along the Chicago, Rock Island and Pacific and the Michigan Central railways from Des Moines to Detroit, a line of 500 miles, running through a flourishing agricultural region. The only villages along these two lines of railway that showed a gain of population at the last census were those immediately adjacent to Chicago ; all the other small communities have steadily lost population. Mr. Fletcher attributes the decline to discriminating railway rates, favoring the great cities ; but the more general cause is the substitution of production on a large scale for local industries. The transformation of industry in Germany has been going on at a very rapid rate in the last decade, and the industrial census of 1895 showed that the old Handwerker, or master artisans, are disappearing before the advance of the factory system, thus : Percentage increase or decrease, 1882-1895.* Persons working on their own account — 5.3 " " in establishments of 5 or fewer workmen. +23.0 « 6-50 " . 4-76.3 " " " " " more than 50 " . +88.7 Similar statistics may be given of village industries in England.^ Thus, in the agricultural county of Huntingdon ^Cf. summary yahrbilcher fiir N.-O. tend Statistik, Ixx, 665. The question is discussed by Sinzheimer, op. cit. '^ Ogle, "The Alleged Depopulation of the Rural Districts," in Jour, of Stat. Soc, Hi. (1889), pp. 219, 226, 228, 230. The subject is treated at some length by Graham, The Rural Exodus, chapter iv. CA USES 189 the number of handicraftsmen, tradesmen and other classes was as follows : 1851. 1861. 1871. 1881. Building trades Iji40 1,050 1,092 997 Milliners and seamstresses 741 889 815 830 Lace makers 1,022 708 678 389 Shoemakers 700 669 499 364 Paper- makers 1 60 230 264 305 Total of II trades 4,932 4,6n 4,307 3,704 Shopkeepers 1,338 1,370 1,513 1,444 Drink trade 319 324 415 268 Professional classes, teachers ... . 332 369 408 421 do clerical, medical, etc. 246 255 243 234 Personal services 2,308 3,165 3,638 3,293 Agriculture 12,256 12,173 11,819 10,161 In England the decline of agriculture is a cause of the decay of villages which can scarcely be said to exist in the case of the Western villages embraced in Mr. Fletcher's investigation. But agricultural depression evidently cannot account for the remarkable decrease in the number of shoe- makers and lacemakers, the cause of which is indisputably the centralization of industry. The foregoing figures of Dr. Ogle's suggest other conse- quences of the modern transformation. Thus the decreasing number of persons engaged in the drink trade results from the decline of travel on the highways since the era of rail- ways.^ The professional classes, omitting teachers, are de- clining in number as a natural result of the modern ten- dency toward specialization, and the ability of the rich cities to attract to themselves the leading specialists. The village doctor cannot compete against the specialists and hospitals — both city institutions. The small decrease in the number of shopkeepers is worthy of comment. The storekeeper ' But the bicycle era, upon which we have already entered, promises to re- habilitate the country inn. igQ THE GROWTH OF CITIES would naturally be among the last of the villagers to feel the competition of centralized industry. At first, indeed, the railway benefited the village tradesman, for it enabled him to supply his customers with the most approved goods from the centres of fashion. Thus the village storekeeper pros- pered long after the downfall of the village manufacturer ; but in these later days he is fast succumbing to city competition. The mail order system perfected by the city department stores has drawn much of the village trade to the city, and the cheapening of both passenger fares and express rates enables ever-increasing numbers of villagers to do their shop- ping in the city and have their purchases delivered free. The department store, or Universal Provider, now found in every large American city, affords one of the best illustra- tions of the development of centralized industry, although it belongs properly to the topic of commerce. The evolution of retail trade maybe presented in three stages: (i) A single store meets the demands of the entire community ; its stock consists of nearly everything wanted, from a needle to an anchor. This is the old-time " village store" which still exists in small communities. But its inefficient organization prevents it from keeping pace with the growing demands of a rising community, and so (2) specialized or exclusive stores spring up, each of which in its own line outdoes the general store. But after a time the growth of capital and of business ability, and the expansion of the market, enable some saga- cious man to unite several of the specialized businesses under one roof and management in (3) the department store, which is like the village general store in outward form, but very different in internal organization,' for it adheres to the spe- ciahzation developed in the second stage. Its large capital enables it to offer greater variety in each line of goods than the village store and lower prices than the specialized store. 1 Each department has its own head. CAUSES ipi Manufacturing industry has not yet reached the third stage, represented in retail trade by the department store. The tendency has been toward greater speciaHzation of the pro- cesses ; in the woolen industry, for example, the highest development (Bradford, England) has resulted in separate establishments for scouring, carding and combing, spinning, weavmg, dyeing and finishing, packing. Some experts re- gard this as the final development.^ But there is already one establishment in Germany (Krupp's gun works) which not only carries on all the purely manufacturing processes, but also works up its own materials. In the United States the tendency seems to be to distinctly toward the consolida- tion of the various processes ; thus the great steel " barons" are acquiring control of iron mines and transportation lines, and the Standard Oil Company, which began as refiners, have become carriers and are now buying up the control of the crude oil product.^ Similarly, there exists a tendency toward the consolidation of allied manufactures, the successful bicy- cle maker applying his capital and business methods to the production of a general line of sporting goods, and finding his market already secured by the advertising he gave to his original product. The advantage of conducting manufac- turing on such a " department store" plan lies, of course, in the greater steadiness of the business. Competition in any one line can be met by a reduction of prices there, to be made up by the profits on other lines, and at a time of de- pression in one trade the idle workmen may be set to work in other departments. The effect of centralized industry — production for the world market — upon the distribution of population has already been noticed, but it is an interesting question to con- ' Col. North in iilh Qn., Mfs.. Part iii, i8. * It is related that one of the great New York hotels grows its own celery and raises its own poultry on Pennsylvania farms. 192 THE GROWTH OF CITIES sider future prospects. No one can expect that future de- velopment will be in the direction of production on a small scale and for the local market. But while the unit of capital will in all probability increase as time goes on, may it not be dispersed in several small establishments, rather than concen- trated in a single large establishment? One of the most characteristic achievements of the modern corporation is the conduct of such a business as that of a London bread com- pany with large capital, almost entire control of the market, and a highly centralized management, but with branch stores scattered all over the great city. Something of the same phenomenon is to be observed in the United States, where manufacturing establishments scattered from east to west and north to south, unite in syndicates and trusts with large cap- ital and central control. Why may not this process of cen- tralization and decentralization go on indefinitely and even extend itself to separate establishments as soon as the per- fection of small electric motors permits the diffusion of motive power in small shops and the homes of workingmen ? In other words, what is to prevent a return to the cottage system, or domestic industries, with the discovery of cheap methods of distributing water power, as foreshadowed already by the harnessing of Niagara Falls? In the first place, it is to be observed that factories came into existence long before the age of steam. Even England, whose writers are prone to assume the origin of the factory system in the inventions of Watt and Arkwright, possessed large enterprises in the sixteenth century. Tradition tells of " the famous and worthy clothier of England, Jack of New- bury," or John Winchcombe, who kept a hundred looms at work in his own house, and marched to Flodden Field at the head of one hundred of his journeymen.^ The " Weavers' Act" of 1555 suggests a movement toward factories, and the ^ Cf. Ashley, ii, 229. CAUSES 193 tendency reappeared in the seventeenth and eighteenth centu- ries before the introduction of power-machinery. The large manufacturer had the advantage of a greater division of labor and greater rapidity of production. We need not, therefore, be surprised at Professor Ashley's concluding " it is certain that in the sixteenth century it was not at all impossible that the large manufactory might become an important — if not dominant — feature in the woollen trade of England. The prevention of such a development was due primarily to legis- lative action."' On the continent of Europe there was a more considerable development of factory methods antece- dent to the invention of the steam engine.^ The chief reason why the mediaeval clothier concentrated his workers in a factory instead of allowing them to continue working at home was economic in its character and accorded with a fundamental tendency of evolution ; it was the division and combination of labor. That this was the main advan- tage of concentration will appear upon summarizing the ad- vantages of production on a large scale, giving rise to the ^ That is, by limiting the number of journeymen that a master might employ and the numerous guild restrictions. ■^ In the second half of the sixteenth century, a great Basel merchant, Ryff, visited Geneva, and was astonished to see " gigantic six-story houses " in which spinning was done on a colossal scale (Stieda, art. " Fabrik " in Conrad's HandworterbucJi) . He found similar enterprises in Venice engaged in making sail-cloth. In France mention is made of a woollen mill at Abbeville, opened in 1669 with 500 Dutch workmen. And so it was in other countries. The most convincing proof that steam favors the factory system only because it promotes the division of labor, is found in the existence of centralized industry in the ancient cities. Alexandria was the great industrial city of the Roman world, and " here we meet centralized-industry magnates of the most modern pattern like Firma, who, in the reign of Aurelian, even stretched out his hand toward the crown, a baron of industry who made such profits from his paper mills alone that he boasted of his ability to maintain an army on- papyrus." (Pohlmann, Die Uebervolkerung der antiken Grossstddte, p. 31. For the factory system in Rome, see Bliimner, Die geiverbliche Thdtigkeit der Volker des classischen Aiterthums, H2ff._) 194 THE GROWTH OF CITIES law of increasing returns in manufactures : ( i ) Economy in motive power and in the erection and maintenance of the plant. One large mill costs much less than two small ones. (2) Economy in machinery. Modern industry is essentially machine production, and machinery is constantly becoming more complicated and expensive. Improvements and new inventions are so frequent that only the concerns with large capital can keep abreast of the times and survive fierce com- petition. (3) Saving in wages, by securing the most exten- sive division of labor,^ and employing the most highly spe- cialized ability. It is only the ** big concern" that can afford to employ a superintendent of the highest ability ; a small establishment would not have work enough to engage his utmost energy and talent. So, too, only the large establish- ment can maintain its own staff of inventors and experts to experiment and investigate suggestions. (4) Economy in the utilization of by-products, which become profitable to handle only in large quantities. The four advantages already mentioned may be regarded as belonging to production proper ; a fifth class of economies belongs to the commer- * A great deal has been added to the theory of the division of labor since Adam Smith's first attempt, which was very inadequate. Smith suggested three advan- tages : (a) Increased dexterity, because " practice makes perfect." (b) Close at- tention to a single process encourages the invention of machinery to take over automatic operations, (c) Saving of time in going from one operation to an- other — of very minor importance, since the time lost by the all-round artisan is fully compensated by the stimulus of variety and change, (d) The most import- ant advantage of all was overlooked by Smith — the gradation of labor. The sep- aration of a process into its simplest elements permits the use of cheap labor for the heavy, mechanical work, and the concentration of the skilled and expensive labor exclusively upon the finer tasks. In the French silk industry of the middle ages each artisan understood every one of the 100 operations in his trade. (SchmoUer, "Die Thatsachen der Arbeitsteilung," in his Jahrbuch, 13: 1047.) To-day such an all-round education is worse than useless, because many of the operations can be performed by the ordinary street gamin with the shortest pos- sible training, (e) Saving in capital, because the several workmen no longer re- quire full outfits. CAUSES IQ5 cial side of the business, and consists of (5) special facilities for buying, selling, shipping and advertising. The enter- prise backed by large capital can buy its raw materials in the cheapest market and at the most favorable time, and store them until needed ; it can also hold its finished product until the most favorable opportunity for sale arrives. It se- cures discounts by buying in large quantities, and low freight rates by shipping in large quantities. It can maintain its own "drummers," or commercial travellers, and in other ways advertise on a large scale.' In short, the motto, " Large sales and small profits" explains why the " big con- cern" is the fittest to survive in the economic world. Of all these advantages of the "big concern," the most important without doubt are those of the fifth class, which are conected simply with large aggregations of capital. But next to those in importance are the economies connected with machinery and the division of labor. Machinery on the whole tends to become more complicated, uniting more operations and requiring the co-operation of more laborers in tending each machine. Its rapid evolution has almost always and everywhere favored the growth of the large estab- lisment. Similarly, the efficient division of labor makes for centralization, on account of saving in wages of superintend- ence. As Marx says, " Laborers as a general rule cannot co-operate without being brought together." An additional influence favoring the substitution of the factory system for cottage industries in those trades like that of ready-made clothing, where the division of labor is not particularly advantageous, is the necessity of public inspec- tion of industries. From purely selfish motives, society can- not afford to permit the sale of garments made in rooms ' A New York department store finds it profitable to expend ^300,000 a year for advertising. The immense expenditures of bicycle-makers and manufacturers of proprietary articles for the same purpose are well-known facts. igQ THE GROWTH OF CITIES poisoned with disease, and it has all but decided to abolish the " sweat-shop" on account of the difficulty of supervision. While, then, cottage industries have revived fn some parts of the world (notably in Ireland) we must not be sanguine of any general revival, even should transmission of power become an established fact. Motive power is but one factor in the triumph of the factory system, as history shows, and those who look for fundamental changes in the structure of industry, as soon as electric motors shall have superseded steam engines, are doomed to disappointment.^ We may be sure that the factory system has come to stay. But how does it affect the distribution of population ? First, by de- stroying family industry prosecuted in the farm houses, it diminishes the number of agriculturists, as pointed out in Section II ; secondly, by destroying industries in the handi- craft stage (village shoemaking, milling, etc.), it removes population from the villages. The entire effect on the dis- tribution of population is therefore centralizing. But it re- mains a question whether it favors the growth of large cities, as commerce does, or of small cities and towns. It is a ques- tion of the advantageous location of the large factory. If local facilities for transportation preponderate among the natural advantages, then the factory will go to the great city ; while, if local facilities for production determine the site, the factory is likely to go to the small city. Let us consider the prob- lem.'' ^The German professors, in their anxiety to disprove Marx's thesis regarding the concentration of capital, are unduly hopeful of such changes as those men- tioned above. The v^riter recalls the statement of a Berlin professor, lecturing on Unternehmungsformen, that the baker's trade among others could never be cen- tralized, since the area that could be served by one shop is so small. The possi- bilities of telephone and delivery service escaped him, but even on his assumption that each block must have its own bakery, there is nothing to prevent all these small shops from becoming branches in a large concern, as is actually the case in London. ^ In addition to the literature given in the preceding section, the following ar- CA USES 197 In the first place, it may be observed that where the divi- sion of labor is undeveloped, industry is carried on in prox- imity to consumers, and with reference to the advantages of consumption alone. Such is the case in the household economy, and for the most part in the town economy. Nat- ural advantages for production are ignored, and it is in fact the very purpose of utilizing these that gives rise to the division of labor. This leads to the improvement of ways of communication, which in time alters the conditions of pro- duction. With regard to agriculture, mining, and the cruder manufactures, such improvement emphasized and intensified every natural advantage possessed by one locality over an- other; without easy transportation the fertile land of North Dakota would not compete with the rocky soil of Massachu- setts in raising food for the industrial population of the eastern commonwealths. It is cheap transportation that has transferred the Irish and German peasants to the western plains of America, and that, by making profitable the work- ing of iron mines lying at a distance from coal fields, has scattered the Cornish miners all over the world. By making available every natural advantage, transportation disperses the agricultural and mining population. But what is true of agriculture and mining does not neces- sarily hold of manufacturing, which is far less dependent upon special qualities of the soil. In manufacturing, the raw material that comes from the soil is but a single factor in pro- duction ; other factors are capital, labor, rent, taxes, market for the sale, and facilities for the shipment, of the manufac- tured products. Now the effect of improved transportation tides may be referred to : Ross, " Location of Industries," in Quarterly Journal of Economics, vol. x, (1895-96); E. Laspeyres, "Standort der Gewerbe," (in the United States), Berlin Vierteljahrschrift fu rVolkstuirthschaft, 1870, Nos. II and III, 1871, p. I; Roscher, "Studien liber die Naturgesetze die den zweckmassig- sten Standort der Industriezweige bestimmen," in Ansichten der Volkswirthschaft, 3d ed., vol. U. 198 THE GROWTH OF CITIES is to cheapen the raw materials, and as a consequence diminish their relative importance as a factor of production. The very fact therefore that railways have so cheapened transportation as to permit the shipment of bulky and heavy commodities (/. e., raw materials) has diminished the impor- tance of local natural advantages and increased the impor- tance of the non-natural or artificial advantages for produc- tion. The crude manufactures {e.g., lumber mills, tanneries) are still located near the source of supply of raw materials,, but not the finer manufactures, and the question arises. Is their location determined by other conditions of production^ or by the conditions of consumption, /. e., facilities for mar- keting, and cheap shipment to consumers? If nearness to consumers is the most important advantage in a manufactur- ing site, then it might be expected that the great commercial centres would also be the manufacturing centres, for they not only contain a rich and numerous body of consumers, but apparently afford superior facilities for distributing goods to the remaining consuming population. The tendency in man- ufacturing would then be toward centralization, and the great cities would grow at an enormous rate.' Such, indeed, has ' This is the argument and conclusion of Dr. Cooley in his authoritative work on The Theory of Transportation. He says (p. 88) : "Natural facilities for trans- portation in many, if not most cases, determine the seats of manufacturing indus- tries and of the population associated. Convenience of transportation becomes itself, in all advanced conditions of industry, the most important of local facilities for production; only cruder processes (sawing lumber, smelting ore, etc."), need take place near the source of raw material. Those of a finer sort, in which the cost of moving the raw materials is relatively less important, tend to seek the large centres of the collection, distribution and exchange of products. The vicinity of cities, wherever they may be located, will always be the chief seat of the finer manufactures, on account of the convenience that cities offer for selling and ship- ping goods." The important influence that transportation facilities have upon the location of industries appears from the following facts, which also illustrate the intricacies of railway tariffs. It is said on good authority that the principal reason that potter- ies thrive in Staffordshire is because the Liverpool ships, carrying iron from the CAUSES igg been the actual tendency. In former times, the manufacturer located his plant chiefly with regard to two advantages, water power and nearness to raw materials. Steam applied to sta- tionary engines has made him independent of water power ; applied to engines of locomotion it has made him all but in- dependent of the source of his raw materials. Cheap trans- portation may put the great city on a level with the small town adjacent to the raw materials, i. e., one tendency of modern improvements is to make the commercial centre also a manufacturing centre. The centralizing influence of railways is particularly strong in countries where competition has had full sway ; the com- petitive points, enjoying lower rates than rival towns on a single line, absorb all the growth of a region. This fact has been a matter of every-day observation in many parts of the United States, and it is confirmed by all railway authorities.^ It is asserted that " the entire net increase of the population from 1870 to 1890 in Illinois, Wisconsin, Iowa and Minne- sota, except in the new sections, was in cities and towns pos- sessing competitive rates, while those having non-competi- tive rates decreased in population;"" and in Iowa it is the adjacent counties, want some bulky but light goods to fill up the cargoes. Pitts- burg in times past received a low rate on rough goods from Cleveland because they could be transported in the empty cars that had taken coal to Cleveland. A slight decrease in the railway rates on wheat, or increase on flour, from the West to the East, would probably transfer the milling industry from Mmneapolis to eastern cities. ^ See the Second Annual Report of the Interstate Conwierce Commission, p. 30; also the Aintk Annual Report, p. 16 : " The effect of these disproportionate charges aids the building up of large cities and the concentration of great numbers of peo- ple at a few central places." The Massachusetts Board of Railway Commissioners reported (1884) that the short-haul law " has helped to save small industries and small places from being crushed out of existence; it has checked the tendency toward consolidation which would build up one place or a few places at the cost of local enterprise." (Quoted in Pol. Sc. Quar., v, 426.) ' Stickney, The Railway Problem, 62. 200 THE GROWTH OF CITIES general belief that the absence of large cities is due to the earlier policy of the railways in giving Chicago discriminat- ing rates/ If in the United States excessive competition among the railways has concentrated population in a few competitive points, the absence of competition in France has produced a similar effect. The French government supervised the con- struction of railways very closely and never permitted the waste involved in building more than one line between the same two points ; each road therefore had a monopoly in its own district, and as it could earn a higher rate of profit on through business than on the local traffic, it neglected the latter.'' Local branches remained unbuilt until subsidies were forthcoming from the central government and local authori- ties (1865). Ten years later an attempt was made to divert the local roads from their true purpose, and by building con- necting links bring them into competition with the old roads for the through traffic; general insolvency of the local roads resulted, which was followed by a new monopoly. The state's guarantee of dividends (1883) undoubtedly prevents the French railways from building many new lines to develop new business ; but in America this has been overdone, and it is probable that the smaller places are better served in France than they are in the United States — at least in the West. In Germany, too, there was a tendency for manufactures to settle in the cities upon the opening up of railway commu- nication. A careful statistical study of the effect of railways upon the growth of places of different size was made several years ago by Dr. Schwabe, of the Berlin Statistical Bureau. He calculated the rates of growth of 125 cities for a certain 'Dixon, State Railroad Control, pp. 204, 151. * Hadley, Railroad Transportation, 192-9, CAUSES 20 1 period (ranging from six to twenty-one years) before and after the formation of railway connections.^ The result was as follows: (A) Of 80 cities having less than 10,000 inhabi- tants each, the increase in population was greater in 23 and less in 57 after the opening of the railway. (B) Of 37 cities with populations 10,000-50,000, 18 showed an increased rate of growth, 19 a decreased rate. (C) Of 8 cities of more than 50,000 inhabitants, all but one (Cologne) grew more rapidly after the introduction of railways than before. It thus appears that the railways stopped the increase of population in the smaller cities, except those of an industrial character, and hastened the growth of the large cities. The railways concentrate transportation in a few channels, and the termini get the benefit. Investigations similar to those of Dr. Schwabe have since been undertaken by the impe- rial German statistical office, and while the conclusions are less positive they tend to confirm his results regarding large cities.^ The statistics of manufactures furnished by the United States government are not altogether trustworthy, but they at least show that in the period 1860-90 the movement was a centralizing one, toward the larger cities. In i860 the annual production. of manufactures per capita was $60 for the United States as a whole, $193.50 for ten cities having a population of 50,000 or more, $424 for ten cities under 50,000, and $44 for the rural districts.3 Thus the per capita production was ^ "Statistik des preussischen Stadtevvesens," in (Hildebrand's) yahrbvcher fiif Naiionalokonomie und Staiislik, vii, (l866), pp. I— 32. '•' " The railways do not hasten the growth of the smaller cities; their absence does not hinder the development of small places in comparison with those of the same size that are provided with railway connections." October Heft of Monatshefte zur Statistik des Deutschen Reiches fiir das Jahr 1878, or Statistik des Deutschen Reichesy xxx, Theil II, p. 14. A second study appeared in the Monatshefte of 1884, (Mai) V, 9. ' E. Laspeyres, " Die Gruppirung der Industrie innerhalb der Nordamerikan- ischen Union," in Vierteljahrsch-ift fiir Volksiuirthschaft und Kultu7-geschichte, xxxiv, 17. 202 'i^HE GROWTH OF CITIES at that time largest in the smaller cities. In 1890, however, the per capita product of manufactures was $455 in the 28 great cities, $355 in the 137 cities of 20,000-100,000 popu- lation, and $58 for the remainder of the country.^ The supe- riority of the smaller cities in 1 860 had in 1 890 given way to that of the great cities. But it is probably safe to afifirm that the centralization of manufacturing industry has reached its Hmit. A reaction toward decentralization began when manufacturers located their mills in the suburbs of large cities in order to escape the high city rents and still avail themselves of the city's superior shipping facilities. Suburban enterprises have in the last decade become increasingly familiar phenomena, not only in the United States, but also in England and Germany, and have brought hope to social philanthropists disheartened with the poverty and misery of congested cities. The statis- tics of manufactures do not portray the tendency, because it is comparatively recent ; but the Twelfth Census, to be taken next year, will show how rapidly manufacturing industries are leaving the larger cities. To give one example : the writer was informed by William A. Perrine, of the Ironmoulders' Conference Board of New York, that of some 65 iron foun- dries in New York City fourteen years ago, only fifteen now remain. Some have gone out of existence ; but most of the remaining establishments have removed to Brooklyn, or sub- urban towns on the Hudson or in New Jersey. In recent years the decentralizing movement has taken a still more favorable turn, largely as a result of continued im- provements in transportation methods and a more enlight- 1 Computations based on the returns of the iilh Census, The statistics refer to gross values, and present some incongruities when individual cities are compared. When, for example, the raw materials constitute so large a proportion of the gross value of the product as do the cattle and hogs of the Kansas City and Chicago stockyards, a small establishment will be credited with a large gross product. But such differences vanish in general averages of a whole class of cities. CA USES ened policy on the part of railway managers, who i. learned that the factor of distance is of minor importance i the expense account as compared with the additions to the revenue that result from a judicious encouragement of industries in small cities along their lines. To-day, practi- cally every shipping point in New England enjoys precisely the same freight rates to points south and west of New York city as does the metropolis itself.^ This is in effect the zone tarifif system, which has been fully developed in Hungary; it gives one and the same rate to all points within each zone. Many influences favor the adoption of a single, uniform rate, as in the postal system, for territories of moderate extent. A uniform rate for the whole United States is of course imprac- ticable ; so large a country would have to be divided into zones. But in a country like Belgium the distances are so short that a single rate seems feasible. Students of railway tariffs are familiar with the financial basis of the cheap " long- haul" rate. The English railways, for example, have found it profitable to carry fish to London as cheaply from Scot land as from English ports halfway to Scotland ; it is busi- ness that more than covers the actual cost of movement, thus contributing something to the permanent, fixed charges, and at the same time it is business that could be secured in no other way. The railways entering New York have carried 'Although the writer has not the tariffs of the English railways at hand, it is evident that they are pursuing somewhat of the same policy as the American roads, from the fact that decentralization has set in among the cotton factories of Lancashire. The new factories are established neither in the great city nor in its suburbs, but in small localities outside. Both Manchester and its environs are being abandoned as manufacturing seats. (Schulze-Gavernitz, The Cotton Trade, 74.) Professor Marshall notes that although " Manchester, Leeds, Lyons are still the chief centres of the trade in cotton, woollen and silk stuffs, they do not produce any great part of the goods to which they owe their chief fame." (^Op. cit., 354.) The decentralizing movement is also in progress in German manufactures; cf. Jannasch, FAiropdische Bautti-wollindnstrie, 11, 12, " Auswanderung der In- dustrie nach dem flache Lande." THE GROWTH OF CITIES ji to the metropolis at a single rate from all points within radius of lOO miles at first, and now nearly 400 miles. George R. Blanchard, of the Joint Traffic Association, testi- fied before the Interstate Commerce Commission that the single rate could profitably be extended to distances up to 1,000 miles, and the counsel of the Delaware, Lackawanna and Western railway favored the extension to any distance within which it is possible to transport milk without injury. Distances, except they be transcontinental, are in fact losing much of their importance for the modern railway, inasmuch as the cost of service in no wise corresponds to the length of haul. The following diagram, taken from the work of a practical railway manager,^ will show how slight a factor is distance in the rates between New York and Chicago : Rates on sugar, per cwt., in car-load lots. Distance in From New York to miles. Cents. Graphic comparison. Harrisburg, Pa 200 1 5 Altoona, Pa 326 15 Pittsburg 444 15 Bucyrus, Ohio 640 21 * Hamlet, Ind 840 25 Chicago 942 25 The effect of a single, uniform railway rate, if it is ever realized (and we have seen that it is, to a certain extent, already a reality for New England manufacturers), would be to eliminate the factor of transportation facilities from the advantages or disadvantages of particular localities for pro- duction. The great city would then distribute its products no more cheaply than the small city. Of the remaining facilities for production, there is no pre- ponderance on the side of the great commercial centres. The important functions of buying and selling, and the securing of capital and credit, which formerly determined the ^ E. P. Alexander, Railway Practice, p. 16. CAUSES 205 location of many enterprises in the commercial centres, can now all be accomplished by means of a city office ; there is not the slightest need of bringing the factory itself to the city. On the other hand, the small town has the great ad- vantage of much lower rents and taxes, which in most cases will be decisive, especially if the town offers freedom from taxation and sufficient land for a building site as an induce- ment, a poHcy that has been the making of many a small city in Michigan, New Jersey and other commonwealths.^ As regards the supply of labor, the relative advantages of city and country differ according to the kind of occupations. As a rule, the wages of skilled workmen are higher in the city than in the village, largely because of the greater effi- ciency of labor organizations. Even where wages rule the same, many employers have abandoned the great city to escape other exactions of the labor unions. The typogra- phers have one of the strongest of trade unions, and their aggressiveness has already caused the removal to suburbs or small cities of the printing houses of several New York and Boston publishers. It is difficult to say how far this move- ment will extend ; it is opposed now with all the strength of the trade unions in the city, and on the other hand, improved means of communication may in the course of time permit the formation of labor organizations in the country that will be as strong and efficient as those in the city, where large numbers who can meet together on short notice render a powerful association easier of formation. When that time arrives, the small town loses one of its attractions for the manufacturer. With regard to unskilled labor, the case is somewhat differ- ent. The great city contains a large population that is ^On the other hand, it often happens that a firm of manufacturers build a factory in the open country, and start a land speculation of their own. The re- sult in either case is a small manufacturing town. 2o6 ' 1HE GROWTH OF CITIES uneducated, unskilled and poverty-stricken. Incapable of organization, it sells its energies to the bidder at starvation wages. Its standard of living is that of the " submerged tenth" of London, or the slum population of New York and Chicago. Although rent and the necessaries of life are higher in the great city than in the rural districts, the middleman who runs the " sweat-shop" finds cheap city labor more sub- missive and profitable. In England such industries as glove- making, hand-made lace, etc., are pursued successfully in the rural districts, where female labor is to be secured cheaply. The disadvantage of such labor, however, is its irregularity, which has prevented its employment in this country. On the whole, the great city seems now to be at a disad- vantage in manufacturing, except in the case of cheap and unskilled labor, such as that engaged in the clothing trade.' The existence of other manufacturing enterprises in the metropolis may probably be set down to one of the follow- ing causes : ( i ) Certain old establishments started on the outskirts of the city in an earlier period, and now loath to remove, when the city's growth has enclosed them. (2) Certain industries requiring either traditionally skilled labor, which is not yet to be found outside the original seat, or a high development of technique and art. (3) Many indus- tries whose product is chiefly for local consumption. The number of these is large, since the cities contain so large a proportion of scientific and mechanical contrivances of the age. New York and Chicago together probably possess a larger number of the modern web-perfecting printing presses than all the rest of the United States. Putting together all the paraphernalia of a great commercial city, vehicles of all kinds, vaults and safes, elevated railway apparatus, etc., one will see the necessity of the existence in the great city of a large num- 1 The clothing manufacture is the principal industry of New York City. CA USES 207 ber of mechanical industries. To these must be added the enterprises that cater to the wants of the rich consumers of a commercial city — furniture, table-ware, carriages, etc. Some of the articles might, indeed, be made outside the city, but there is considerable advantage in "being on the ground." (4) Certain industries whose raw materials come equally by land and water routes. In this case the point of intersection — a commercial centre — will be the most economical place of assemblage. An instance in point is the iron and steel industry of Chicago. That local consumption, unlimited supply of cheap labor, and other considerations just mentioned, rather than natural advantages, determine the location of manufacturing indus- tries in the great city, plainly appears in the Census Statistics of Manufactures {Yzxt III, p. xxxvii) which show that the six leading industries of New York according to net value of product in 1890, were — 1. Men's clothing — factory product. 2. Newspapers and periodicals. 3. Women's clothing — factory product. 4. Tobacco, cigars and cigarettes. 5. Malt liquors. 6. Book and job printing and publishing. With all the advantages for manufacturing industry pos- sessed by the village or small city, it may look as if the country were destined to be covered with industrial villages built up around one or two immense factories. But there are many forces to oppose this tendency. In the first place, one large modern factory alone gives employment to hun- dreds of operatives and tends to attract other industries, for it is a well understood fact that place-specialization is ex- tended not to a single trade but to a group of allied trades.^ Hence, when the benefits of specialization cause a manu- ^ Marshall, Principles of Eco7iomics, 353. 2o8 THE GROWTH OF CITIES facturer to confine himself to a single process in any in- dustry, say weaving in the textile trade, it is natural for other firms carrying on the processes of carding, dyeing, etc., to locate their establishments in the vicinity. Auxiliary trades and repair shops also attach themselves to the group. Further, there are by-products to be utilized ; thus it hap- pened that the erection of a large tannery in a western New York village was shortly followed by that of a glue-factory. Finally, in a factory town where the labor of one sex is ex- clusively employed, other industries will frequently spring up to utilize the labor of the opposite sex. Thus one of the earliest factories in the city of New Britain, Conn., was devoted to the manufacture of carpenters' rules and levels, and employed male labor alone ; it was not long before a cotton factory, in which the cheap and abundant labor of women and children could be used to advantage, was planted in the town. A similar tendency' has caused the location of textile factories in mining, metal and machine towns in England. These are some of the reasons why an industrial village soon becomes a large town. But the process does not stop there. New factories are apt to seek the neighborhood of old establishments in the same industry on account of the " initial difficulties " (familiar in the "infant industry" tariff argument) which attend the upbuilding of an industry in an entirely new atmosphere. The advantages of inherited skill and traditions favorable to the genesis of improvements, created by friction among the followers of the same skilled trade in one place, have been well described by Professor Marshall : " The mysteries of the trade become no mys- teries ; but are as it were in the air, and children learn of them unconsciously. Good work is rightly appreciated, inventions and improvements in machinery, in processes, ^ Marshall, Principles of Economics, 353. CAUSES 209 and the general organization of the business have their merits promptly discussed ; if one man starts a new idea, it is taken up by others and combined with suggestions of their own ; and thus it becomes the source of new ideas. And presently subsidiary trades grow up in the neighbor- hood, supplying it with implements and materials, organiz- ing its traffic and in many ways conducing to the economy of its material." ' In former periods the costs of transporta- tion afforded some protection to " infant industries," but the scaling down of these costs allows the established business in a distant city to compete with the local industry on equal terms in the local market. As we have seen, discriminating or dififerential rates in favor of competitive points are alleged as the reason for the decay of Western villages, V. SECONDARY OR INDIVIDUAL CAUSES. At bottom the question of the distribution of population is a question of economic organization, of the play of eco- nomic forces which we have been studying in the preceding sections. These economic forces, however, act upon men in various ways to produce the necessary shifting of population ; they play upon their motives to draw them where their pro- ductive power will be greatest. Legislation is a necessary part of the movement ; sometimes legislators aid the action of economic forces blindly, at other times consciously and deliberately. In many cases the political and social move- ment seems to be independent of, and even antecedent to, the economic movement. There is no doubt that what was at first an effect of economic causes, has in its turn become a cause. The tene- ment house classes, for instance, came to the city in order to better their condition — that was an economic cause. But once settled in the city their love of society becomes so great ^ op. cit„ 352. See also Schulze-Gavernitz, The Cotton Trade, 82. 210 THE GROWTH OF CITIES that they will not leave the city for the country even when they might thereby greatly improve their material condition. The effect has now become cause : these working classes furnish the manufacturer with an abundant labor supply and induce new industries to settle in the city. There is thus to be seen a continued interaction of cause and effect; and although we hold the growth of cities to be a matter of eco- nomic organization, we must not neglect what may be called secondary, or in contradistinction to the general causes already discussed, individual causes of the movement. For convenience of discussion, these may be classed as economic, political and social. § I. Economic Causes. — Obviously, if the efficient organiza- tion of the industrial powers of society requires a transfer- ence of productive power from agriculture to manufactures, the transfer will be obtained by elevating the condition of the men in the latter industry or depressing the condition of the agriculturists. Such is the significance of the general agri- cultural depression that has been felt in all the older agri- cultural countries for twenty years. The introduction of machinery and the opening up of virgin fields in Argentina and the American West have rendered unnecessary and un- profitable much of the agricultural labor in Germany, France, England,^ and the Eastern States of America. Amid such circumstances a considerable rural emigration is to be ex- pected, and it has everywhere taken place. Without going into European statistics,^ we may measure the extent of the ^"Of 1,995 ex-metropolitan sub-districts in England and Wales dealt with in the census report, 945 show a decline of population; and, roughly speaking, these localities constitute the farm land of the country." Graham, litiral Exodus, 11. Additional statistics will be found elsewhere in this essay. "■' The subject of rural depopulation has given rise to considerable literature. In England, where the agricultural depression has been most severely felt, Parliament- ary commissions have accumulated an enormous amount of evidence. The final report of the Royal Agricultural Commission of 1897 brings a long inquiry to an CA USES 2 1 1 movement in the United States from the data in the Eleventh Census. A glance at the " map showing gain or loss of rural population between 1880 and 1890" will give a good idea of the extent of the movement/ The area in which the rural population ^ declined includes most of Maine, New Hamp- shire, Vermont, New York, Ohio, Illinois and a large part of Indiana and Iowa. The States in which this area of declin- ing rural population embraced more than half of the total area are the following : Per cent. Per cent. 1 Nevada 90.50 6 New Hampshire ... . 63.10 2 New York 82.66 7 Ohio 61.39 3 Vermont 77.20 8 Connecticut 60.85 4 Illinois 65.73 9 Marylapd 54.1 1 5 Maine 64.96 end. A popular treatment of the subject is given in two volumes of Methuen's Social Science series : IVie Rural Exodus, by P. A. Graham, and Back to the Land, by H. E. Moore, and in Dr. Longstaff' s article on " Depopulation," in Palgrave's Dictionary of Political Economy ; while the statistics for England ap- pear in Dr. Ogle's article on "The Alleged Rural Depopulation," J. of St, Soc, 1S89, vol. lii. Dr. Longstaff presented additional statistics for England and other countries in the same yoiirnal'va. 1893, ^o^- ^^i- Agricultural conditions in Germany were thoroughly investigated in 1892-3 by the Verein fiir Socialpolitik {cf. the Verein's publications, vols, liii-lvi, Verhalt- nisse der Landarbeiter ; an excellent summary of which is given by Drage in the British Royal Commission on Labor's series of Foreign Repo7-ts, 1893'). See further the essay by A. Wirminghaus, " Stadt und Land " in Jahrbucher fiir Nationalokonomie und Statistik, vol. Ixiv; and one by Vicomte de Beaumaire in y, of St. Soc, xlix, 450. For France the subject was thoroughly treated some twenty-five years ago by Legoyt, E>u Progres des Agglomeratiojts tirbaines et de F Emigration rurale, and has since been continuously discussed in periodical literature, to which references will be found in Lavasseur, La Population Franfaise. The causes of " Agricultural Discontent " are analyzed by Dr. Emerick in the Pol. Science Quar., xi, while the whole subject is systematically and scientifically discussed in Buchenberger's Agrarpolitik. ^ nth Cen. Pop., pt. i, p. Ixx, map 4; also in the Statistical Atlas of the United States. ^ The method employed in these calculations is as follows : " From the total population of each county in 1890 has been subtracted the population of all cities 212 THE GROWTH OF CITIES The actual net loss in the States whose rural population decreased in the decade 1 880-1 890 is as follows: 1 New York . 163,176 7 Nevada I3f085 2 Illinois 66,741 8 Connecticut 11,964 3 Maine 24,391 9 New Hampshire ... 8,575 4 Vermont 18,944 ^o Massachusetts .... 6,522 5 Maryland 1 7,220 1 1 Rhode Island 508 6 Ohio 13,274 In many other States the rural population increased so slightly as to be virtually stationary ; thus the gain in Dela- ware was but 40, in Indiana 8,073. The States mainly affected are the New England States, New York, Ohio, Indi- ana and Illinois, whose agriculture has been languishing under the competition of Western farms. Nevada's decrease may be ascribed to other causes.^ The rural emigrants from Europe and the Eastern States of America do not all settle upon the farming lands of the West. On the contrary, many of them go to the nearest city or town. Such internal migration we know to be on the in- crease in Europe, while the extraordinarily rapid growth of or other compact bodies of population which number looo or more. From the population of the same counties in 1880 has been subtracted the population of the same places at that time, and the remainders, which are assumed to be the rural population are compared for increase or decrease." — Op. cit. p. Ixix. ' Rural depopulation in five great agricultural States is strikingly brought out in the following table : Number of townships which were Stationary Gained Lost in population 1880-90. Total. Ohio 32 529 755 1,316 Indiana 16 496 482 994 Illinois 45 579 800 1,424 Iowa 29 893 691 1,613 Michigan 22 506 416 944 Total 144 3,003 3,144 6,291 — Fletcher, "The Doom of the Small Town," /^7r«w, xix, 215. CA C/SES 213 American cities indicates the existence of similar condi- tion here ; and it has been statistically proved that interstate migration is declining in the United States.^ From all but three ^ of the eleven States given in the list above, emigra- tion decreased in the last census decade. What motives induce the farmer's boy and the village lad to go to the city? At bottom it is undoubtedly the eco- nomic motive, although it may seldom resolve itself into a matter of dollars and cents, of higher wages pure and simple. Agricultural laborers in England are leaving the farms whether they have low wages, as in Wiltshire, or high wages, as in Northumberland.3 The skilled mechanic, indeed, often moves from the town to the city in order to obtain better wages ; but with the mass of the young men who go to the city, the magnet is the superior field for ambition which modern industrial organization has rendered the city. In former times a larger proportion of the prizes of life could be attained by the villager or countryman; to-day his only chance for leadership is in politics. In the trades and pro- fessions the great prizes must now be sought in centres of wealth, while in business there are no rewards at all for iirst-class ability outside the cities. Every young man is optimistic as regards his prospective achievements in life, and longs to compete for leadership ; to enter the fray, to rise in the world, to make his mark, he must go to one of the great cities, which " afford such extraordinary facilities for the division and for the combination of labor, for the exer- cise of the arts and for the practice of all the professions." § 2. Political Causes. — As a political influence in favor of city growth must be reckoned all the measures of the state ^ Willcox, " The Decrease of Interstate Migration " in Political Science Quarterly, vol. X (1895). '^ Illinois, Maine, Nevada. ' Graham, Rural Exodus, 9. 214 THE GROWTH OF CITIES that promote commerce and manufactures. The political unit and economic unit were identical for so many centuries that political growth meant the enlargement of the economic territory and vice versa. When the village economy and the town economy gave way to a national economy, Mercantil- ism was one of the causes. To-day, Protectionism discrimi- nates in favor of manufacturing industry; at the same time, other forms of taxation (by the commonwealths) discrimi- nate against the farmer, so that there is some ground for saying that legislation is one of the causes of agricultural depression. It would be tedious to enumerate the various legislative acts which have influenced the distribution of population ; they may be summarized under a few heads as follows : (i) Legislation promoting freedom of trade. This acts in the same way as improved means of communication, by enlarging the market. As regards internal commerce, per- fect freedom is now virtually realized by measures adopted in many of the European countries only in the present century. America, however, has not had any important restrictions on the internal movement of goods since 1789. (2) Legislation promoting freedom of migration. This policy also favors the growth of cities by giving them greater opportunities for securing laborers from the superfluous rural population. Mediaeval restrictions had to be swept away to secure freedom of migration and of domicile. It was only in 1795 that England modified the infamous law of settlement, which permitted the local authorities to drive any newcomer out of the parish under the pretence that he might become chargeable to the local poor rates. It was now enacted that a person should not be removed on the ground that he was likely to become chargeable to the parish, but only when he had " become actually chargeable."^ In Germany, freedom ^ Aschrott and Preston-Thomas, The English Poor-Law Systevi, i8. CA USES 2 1 5 of movement (^Freizitgigkeii) has been more recently secured. The right to leave a community was of course first recog- nized in Prussia with the abolition of serfdom in 1808; the right to take up residence in another community than that of birth was long denied from fear of the responsibility of poor relief, and it was not till 1 842 that the Prussian law of settle- ment followed the English act of 1795. The mobility of labor is now an accomplished fact in all civilized countries, as it has always been in the United States. (3) Centralized Administration. The tendency toward administrative centralization is undeniable both in the United States and in England.^ It exists not only in the transfer- ence of various duties from the local to the central authori- ties, but also in the consolidation of municipalities like New York and London. The transference of governmental ma- chinery from the country to the city affects directly and indi- rectly considerable numbers of the population.^ Especially is this true of military states like France and Germany. For- merly garrisons were much more scattered than they are now, when strategical reasons (railways!) require concentration. The young recruits from the country, after a compulsory residence of three years in the great city, yield to its fascina- tions and remain there under almost any conditions of life. (4) Land Tenure. Exaggerated importance has often been attached to the form of land tenure as a cause of the ' The development of central administrative control in England is admirably set forth in Dr. M. R. Maltbie's English Local Government of To-Day (Colum- bia University Studies.) ^ Indirectly by contributing to the dullness of country life. The removal of local business to the State or national capital restricts by so much the range of local ambition and endeavor. The movement has proceeded further in England than in this country, but recent legislation in New York (excise, State insane asylums, etc.) manifests the tendency toward centralization, Cf. H. C. Stephens, Parochial Self- Government in Rural Districts (London, 1893), Parti, ch, v, " The Parish and Rural Depopulation." 2i6 THE GROWTH OF CITIES migration from the country to the city.^ And yet it has been shown that the rural emigration is as great from the districts of small holdings in southern and western Germany as from the north-eastern districts of large holdings;^ from rural France with its peasant proprietors as from rural England, with its latifundia. Peasant proprietorship in England does not stop the drift to the cities.^ " The movement is confined to no one locality, but is to be observed in every district that lies remote from towns. "'^ The existence of communal lands, the Allemende, in Switzerland, has been thought by some to have had a decentralizing influence.^ But our sta- tistics (Table LXXVII) show that the concentration of popu- lation has been going on in Switzerland at an extremely rapid rate. If the system of land tenure has anywhere had a real ^ Agrarian agitators who repudiate Henry George's policy of land nationaliza- tion, still hold to the necessity of Allemenden (commons) in each village, which will enable the agricultural laborer and smaller cultivators to pasture their cattle or sheep, if the drift to the cities is to be stopped. Cf. H. Sohnrey, Der Zug vom Lande und die Sociale Revolution ; von der Goltz, Die Idndlichen Arbeiter- classen imd die Stadt. *Cf. books cited in foot-note, p. 21 1, especially Sering, Z??V innere Kolonisation im ostlichen Deutschland (vol. Ivi of the publications of Verein fiir Social-politik) . In Mecklenburg, the system of small holdings has been established, and " the conditions of labor are especially favorable;" but the children are not willing to follow the life of the agriculturist, and the emigration from Mecklenburg is ex- ceeded in volume by no other province of Prussia except East Prussia, which in 1 885-90 lost more inhabitants through emigration than it gained through the ex- cess of births over deaths, and hence actually declined in population (c/ cit., p. 6.). The statistical tables in vols. Ivi and Iviii (p. 55) of the Verein's publications clearly show that agriculturists prosper in Germany in proportion to their nearness to industrial cities, rather than in consequence of any particular form of land tenure. ^ In 1890 Lincolnshire had 1,000 more small holdings (under 50 acres) than any other county in Great Britain ; yet " the people are scurrying out of Lincoln- shire faster than out of any other rural district in Great Britain." — Graham, Rural Exodus, 137. * Graham, op. cit., 9. ^ Laveleye, Primitive Property, 80. CAUSES 217 influence upon the distribution of population, it is in Austra- lia, where the squatter system of free land-grabbing has con- centrated land ownership in a few hands and kept the people from the soil ; but geographical and cHmatic conditions have also had a most important influence, by making sheep-rais- ing (which requires few laborers) more profitable than agri- culture.^ In Victoria they have been pursuing a policy favorable to the taking up of land;^ a progressive land tax since 1877 has discouraged large estates; the Act of 1884 contains stringent regulations against owners who are not bond fide cultivators, and there has been an import duty on cereals. Notwithstanding these endeavors and the fact that agricultural production has increased in a larger ratio than population,3 the proportion of the population outside the cities has been steadily decreasing.* It is still too early to judge of the ultimate efifects of the radical land legislation in New Zealand. (5) Miscellaneous. Various special acts of the legislature have at different times contributed toward the concentration of population. Examples are the Enclosure Acts in Eng- land,5 so numerous during the reign of George III, and the modern creation of deer forests.^ The disbandment of the great Union armies at the close of the Civil War sent to the 'The 1891 census report of New South Wales (p. 128) says that the concen- tration of population in seaboard cities is the " only possible mode of develop- ment in Australia because there are no great rivers with leagues of navigable waterway stretching into the heart of the country, far remote from the seaports. Communication with the outer world has begun and ended with a good roadstead for shipping." ^ Epps, Land Systems of Australia., 79-83. ^ Ibid., 2^. *From 45 per cent, in 1 881 to 41 per cent, in 1 891. ^Toynbee, Industrial Revolution (Humboldt ed.j, p. 89. * Longstaff, Art. " Depopulation " in Palgrave's Dictionary of Political Econ- omy, vol. i. 2i8 THE GROWTH OF CITIES cities many hundreds of men who found their places at home occupied ; although the West absorbed perhaps even larger numbers. Considered as a secondary cause or reflector of economic causes, politics is a considerable factor in the dis- tribution of population. § 3. Social Causes. — To enumerate the social advantages that the cities possess as compared with the country would demand too much space, but most of them will be found to be embraced in the following classification : (i) Educational. The city alone must be the residence of those who study art, medicine, music, etc. Even in the matter of primary education, city advantages are superior to those of the rural districts, though not to those of the vil- lages. Where, as in New York State, there are 3,000 school districts with an average daily attendance of fewer than ten pupils,^ facilities are wanting for thorough instruction accord- ing to modern standards. (2) Amusements. The opera, philharmonic concerts, art exhibits, etc., may be classed as educational advantages or mere amusements, but there are many other forms of recre- ation afforded by the city and not by the country, which come under the head of amusements alone. (3) The standard of living. The desire for a higher standard of life, for purely material comforts and luxuries, brings many people to the city. Food is to be procured at prices almost as low as in the country, and in vastly 'Cf. Report of Superintendent of Public Instruction, 1896, vol. i, p. x; the exact figure is 2,983, while there were 7,529 districts in which the average attend- ance varied from one to twenty. Superintendent Skinner says that the cities have sucked the life out of the country schools; since 1S60 the rural school population of the state having decreased 30 per cent. On the other hand it is to be remem- bered that the largest rural emigration in England is from the districts with the best educational advantages (^Census of i8gi, iv, 44). There are many objectors to popular education in England ; but popular education is not so much at fault as the particular kind of instruction. CA USES 2 1 9 greater variety; while everything else is cheaper. The buyer enjoys a larger consumer's rent, as the economists say; that is, he can buy at prices much below those he would be willing to give if pressed, thus deriving a surplus of enjoyment. Then there are conveniences to be had in the city which in many cases could not be obtained in the country, on account of the small numbers to bear the heavy expenses. Such for example are establishments that bring light and fuel to one's door, furnish protection against fire (water works and fire departments), sewerage, rapid transit, etc. The field of municipal activity has been con- stantly widening, until now the city furnishes its residents not only parks and playgrounds, but museums, libraries and art galleries ;^ not only hospitals, but baths and washhouses, municipal lodging houses and model tenements. In order to guarantee the purity of ^food supplies the city has its abattoirs and market stalls, its public analysts and milk inspectors.^ This movement is not transitory ; it promises to continue all over the world, notwithstanding the cry of " Socialism." The advantages of collective action here appear at their best. But there will still be left a large field for pri- vate associations, whose activities have already added to the comforts of city life. Consider the conveniences at the dis- posal of iYi&finde siecle city housewife : a house with a good part of the old-fashioned portable furniture built into it, e. g., china cabinets, refrigerators, ward-robes, sideboards, cheval glasses, bath tubs, etc. ; electric lights, telephones and elec- tric buttons in every room, automatic burglar alarms, etc. ^ Melbourne employs a city organist, who gives free concerts on the fine organ in the city hall. Boston is experimenting in the same direction. '^ Cf. Shaw, Municipal Government in Great Britain, ch. vii, " Social Activities of British Towns." An exhaustive study of municipal undertakings has been made by Dr. M. R. Maltbie in Municipal Functions, constituting the December, 1898, number of Municipal Affairs. 220 THE GROWTH OF CITIES " The laundryman long ago joined the letter carrier, butcher, milkman, grocer and baker in their periodical visits to the basement door ; and whenever madame shuts up her house — all barred and bolted and chained as it has been by the builders — she turns it over to a sort of care-taking or watch- man's company. If she moves out of a house, there are companies to send packers who will bundle up her belong- ings with professional skill, and that will store them for her by carrying them in padded vans to lire-proof warehouses. Her rugs and carpets are now beaten by machinery, and she may hire her house cleaning done precisely as she gives out her washing. Before she rents a house she may order it in- spected by a private company that will report upon the char- acter of its construction and plumbing, and this company also offers to proceed at law against all nuisances in otherwise nice neighborhoods. Thus has vanished the necessity for drawing water, hewing wood, keeping a cow, churning, laun- dering clothes, cleaning house, beating carpets, and very much of the rest of the onerous duties of housekeeping, as our mothers knew it."^ (4) Intellectual Associations. The village is dull not only to the man pursuing light amusements, but to him who seeks cultivated associations, for in these days the cities are the centres of intellect as of wealth. Even the college town with its intellectual atmosphere is to many high-minded people less stimulating than the city, where intellectual ability is so much more varied.^ (5) Such are some of the advantages of city life; some ' Julian Ralph. Cf. also Salmon, Domestic Service. ^ The decay of the small town caused by the emigration of the best minds to the city long ago gave rise to a religious problem which has been considerably discussed, namely, the religious destitution of villages. See the chapter on this subject in J. H. Crooker, Problems in American Society, which contains references to periodical literature. On the general subject of village deterioration, see Fletcher, " Decay of the Small Town," Forum, xix, 237. CAUSES 221 of them are modern, and some are as old as civilization. Not the least important factor in city growth is gregarious- ness or the social instinct itself, which appears to be stronger than ever before in these days of restlessness. English investigators have noticed an increased objection among agricultural laborers to isolation.^ " The isolation of the farm home ; no provision for satisfying the cravings of the youngpeoplefor having good social times" are reasons given for discontent with rural life by farmers of New York to a committee of the New York Association for Improving the Condition of the Poor.^" Another thing to be reckoned with is the passion for " the crowd, the hum, the shock of men," among those who have once lived in the city. One of the trying difficulties of social workers in their efforts to improve the housing conditions of the tenement population is the strong desire of these poor people to be among their associ- ates, and their absolute refusal to settle in more comfortable homes in the country or in the suburbs. The story is told of a kind lady who found a widow with a large family of chil- dren living in the depths of poverty and filth in the city. She moved them to a comfortable country home, where, with a moderate amount of exertion, they were sure of a living. Some six months later, her agent reported the disappearance of the family, and going back to their old haunts in the city tenement district, she found the family living there again. In great surprise she asked the widow how they could leave their comfortable home in the country for such squalid quar- ters in the city, and received the reply, " Folks is more com- pany nor sthoomps, anyhow."^ ' Cf. W. C. Little's report for the British Labor Commission, Fifth and Final Report. ■^ Cf. Leaflet No. i, An Inquiry into the Causes of Agricultural Depression in New York State, p. 9. 'Kingsbury (President's Address at 1895 meeting of the American Social Science Association), "The Tendency of Men to Live in Cities^' in Journal of Social Science, XXXIII, 8. 222 THE GROWTH OF CITIES (6) Finally, we have to take into consideration the forces which in recent times have spread a knowledge of the advan- tages of city life among all classes of the community. Edu- cation has a great deal to do with it, especially the half education which prevails in the rural districts and gives the farmers' boys a glimpse of a more attractive life, without teaching them how to attain such a life at home.^ Then the newspaper comes in to complete the enchantment, with its gibes against the " hayseed" and " country bumpkin." Thus the spread of information, made possible by nineteenth cen- tury improvements in communication, creates a distaste for country life, and more especially for rural life ; while easier travel enables young men lightly to abandon the distasteful life. VI. CONCLUSIONS. To what practical conclusions regarding the future distri- bution of population do the principles deduced in the pres- ent chapter lead? Are the rural districts and villages to continue pouring out streams of migration, which will flow toward the great cities? Or is the migratory movement from country to city but a temporary event, a transitional phenomenon.^ The questions deserve at least an attempt to answer. ' Mr. Lecky seems to regard this, with an " increased restlessness of character and much stronger appetite for amusement and excitement," as the principal cause of agricultural depopulation. He affirms that national (popular) education produces " among the poor a disdain for mere manual labor and for the humbler forms of menial service." (^Deinocracy and Liberty, ii, 477). But Mr. Lecky is not a Liberal in politics. Mr. Pearson has a much more pleasant way of expressing the same fact. He says that state education is raising the poorest classes to the level of the higher class with its taste and ambitions, and they are able to com- pete with it in commerce. "The cleverest boys of the village schools do not care to remain ploughboys." (^National life and Character, p. 145). * This is the opinion of Prof. Karl Bucher, who says it is due to the transition from the town economy to the national economy. The features of the latter per- iod, in his judgment, are similar to those of the movement toward the towns in the nineteenth century. — Entstehitng der Volkswirthschaft, 303. CAUSES 223 The industries of the human race may be conveniently grouped thus : (i) extractive, including agriculture, mining; (2) distributive, including commerce, wholesale and retail trade, transportation, communication, and all the media of exchange; (3) manufacturing; (4) services and free in- comes, including domestic servants, government oflficials, professional men and women, students, etc. The extractive industries generally require the dispersion of the persons engaged therein,^ In particular, agriculture, the principal extractive industry, cannot be prosecuted by persons residing in large groups. It is conceivable that transportation methods might be so perfected as to permit the cultivator of the soil to reside in a city, but it is very unlikely. On the contrary, the improvements heretofore made in transportation have, as we have seen in Sec. II, only strengthened the dispersion of the agricultural population by permitting uninhabited parts of the earth's surface to be settled and brought into cultivation. This will probably be the development of the future as far as human eyes can see. The distributive industries, on the other hand, are dis- tinctly centralizing in their efifects upon the distribution of the population engaged in them. As methods of distribution have been improved and the distributive area enlarged, the tendency toward concentration has increased. The consoH- dation of two railway lines transfers employees from the junction to the terminal city. Every improvement in the mechanism of exchange favors the commercial centre. Of even greater importance is the fact that the production of ^ In mining districts, it is true, the population is oftener than not quite dense. Nevertheless, it is seldom concentrated in great cities, the Transvaal being an ex- ception to the general rule. At present about one-fourth of the total white popu- lation of the South African Republic is to be found in the Rand, {i. e., in the vicinity of Johannesburg), and Mr. Bryce (^hnpressions of South Africa, p. 467), thinks that ten years hence the Rand may contain 500,000 persons, or about one- half the total white population. 224 THE GROWTH OF CITIES wealth is increasing at leaps and bounds ; every year there is vastly greater wealth to distribute, and the process of dis- tribution will require a growing percentage of all the work- ers for its efficient action. Hence, the more the social organ- ism grows, and the higher its evolution, so much greater will the commercial centres become. Manufacturing industries also tend toward the concentra- tion of population, and up to recent years manufacturing centres were coincident with the commercial centres, i. e., the great cities. Recently the equalization of transportation facilities and the excessive rents of great cities have caused the managers of a good many industries to abandon them as sites in favor of the suburb or small town. The reason that this movement does not make for complete decentralization is that production on a large scale is the goal toward which all industries are tending with enlarging and more regular markets, and more convenient means of communication ; and production on a large scale requires, as a rule, the large fac- tory and the grouping of allied trades. Other obstacles to decentralization are the presence in the large city of a supply of cheap, unskilled labor ; of the best knowledge of art and technique ; and especially of numerous industries whose pro- ducts are intended for local consumption. The remainder of the population will in the main follow where the preceding classes lead. Those engaged in the professions or the rendering of personal service must reside near the consumers of their products, that is, where people are numerous and money is plenty. Wealth is always con- centrated in commercial centres, which therefore attract those employments that Adam Smith called " unproductive." To be sure, commercial cities do not always patronize music, painting and the other fine arts, but that is the general rule. Thus it appears that the efficient industrial organization of a nation on modern lines requires the concentration of pop- CA USES 225 ulation in virtually all the industries except agriculture ; and since this industry, for several decades, has been able to deliver its product by employing a continually smaller pro- portion of the total population,^ it follows that the proportion in the centres of population has been increasing. This is the simple but philosophical explanation of the movement known by the popular phrase, " The Drift to the Cities." Could it be known that the law of diminishing returns in agriculture would not come in force again, there would be some certainty in predicting a continuance of the movement toward concentration of the population ; until it does reappear there will be no movement " back to the land." The reason why practical men deny the existence of the abstract law of diminishing returns — historically considered — depends upon the counteracting forces, which may be discussed in two groups. In the first place, science and invention have come to the aid of the farmers with the tender of fertilizers, improved processes, such as the rotation of crops, labor-saving machin- ery, etc., and have thus enabled them to increase their pro- duction without increasing the amount of labor. But the real counter-agent, without which the per capita product must inevitably have declined in spite of this increased produc- tion, is the opening up of new territory. It is easy to see that twenty men will not be able to produce twenty times as much garden truck as one man, if their energies are confined to the piece of land that the one man has been using ; but if his lot is surrounded with unoccupied land enough to give ^ Mr. Hobson (in Evolution of Modern Capitalisni) admits that the proportion of the population engaged in agriculture in England has decreased, but argues that somewhere on the globe there must have been a growth of the rural population to furnish means of subsistence for the large agglomerations in industrial states. Now the United States has usually had an annual surplus of breadstuffs sufficient to cover England's deficiency ; the two countries together may therefore be re- garded as a self-sufficing economy. Nevertheless, the rural population in each has been proportionally diminishing for a hundred years. And in the other coun- tries which export breadstuffs, there is also an increasing concentration. 226 "J^HE GROWTH OF CITIES the twenty men full employment, the aggregate product may be multiplied twenty-fold or even more. This is essentially what has been happening in the world at large since the discovery of America : the per capita product of the cultivators of the soil has been kept up and even increased by the occupation of virgin land as fast as the old farms became crowded. But the amount of available and unoccupied land on the globe is not unlimited. The United States, east of the Mis- sissippi, is now pretty densely settled, and it is improbable that additional labor force would augment the per capita agricultural product very considerably. The West is fast approaching the East in this respect, and the amount of arable land still unoccupied is so small that ten, or at most twenty years, will find it all brought under cultivation. Cap- ital may aid the farmer in reclaiming barren lands, but sooner or later the time must come when capital will find more lucrative employment in manufacturing industries than in irrigation works on some desert plain of Arizona or New Mexico, or in fertilizers for Eastern farms. When capital thus ceases to replace labor in agriculture, the per capita product will diminish and, if population increases, there will result a movement "back to the land." But there are two contingencies which may postpone the necessity. One is the importation of breadstuffs from the as yet unopened lands of Canada, Australia, Russia, South America and Africa. The amount of arable and unoccupied land in these countries is of course imperfectly known. Optimists think that the extent of this territory is large enough to last the race indefinitely, but more careful statis- ticians, like Mr. Giffen, who have observed the rapid pace at which colonization is proceeding, are more conservative in their estimates.^ In any event, the present generation is ^ Cf. the luminous essay, " The Utility of Common Statistics," in Giff en's Essays in Finance, 2d series. CAUSES 227 not likely to see such a condition of "world-crowding" as to draw a larger proportion of Americans into agricultural pursuits.^ A second means of postponing the return to the fields, even with stationary arts of agriculture, consists in changes of consumption, a theme so suggestively treated by Pro- fessor Patten. The diversification of consumption is a remedy which men have in their own hands. If those classes of people who marry early and have large families to support with incomes that barely sufhce to buy bread and the other necessaries of life, would exercise more prudence and self-control and strive to attain a higher standard of life, they would probably consume a smaller quantity of the domestic staples and a larger quantity of luxuries. But a di- versification of consunlption even in the direction of economy would lessen the pressure toward diminishing returns ; for example, should rye or corn come into use as food, a great deal of land unsuited to wheat-growing would be profitably cultivated.^ Finally, some mention should be made of the possibilities of the ocean as a food-producer, which President Andrews discussed a few years since in the North American Review. Considering these possibilities and the more speculative possibilities of physical and chemical science in aiding agri- cultural production or even substituting chemical food-pro- ducts, it does not seem irrational to regard the law of diminishing returns as a very remote contingency. But the ^ The writer therefore disagrees with Sir William Crookes, President of the British Association for the Advancement of Science in 1898, Mr. C. Wood Davis, who writes for the Forum, and ex- Governor John W. Bookmaker, of Ohio, all of whom predict an early scarcity in the wheat supply, which will considerably raise the price of wheat; ergo, produce an exodus from the cities to the fields. ' It is unnecessary to pursue the subject further, as it has been so well devel- oped in Prof. Patten's works. Vide especially the Dynamics of Consumption, and The Premises of Political Economy. 228 THE GROWTH OF CITIES law of increasing returns, at present manifested in agricul- ture, would, if superseded, be followed first by the law of constant returns, which would require a permanently con- stant proportion of the population to be engaged in agricul- ture. When this happens, cities will cease to grow more rapidly than the rural districts. Only as population in- creases in density would its concentration take place ; but this itself would disturb the equilibrium and would therefore cause a movement away from the cities. The occurrence of these conditions is too remote to be predicted.^ In the immediate future, we may expect to see a continua- tion of the centralizing movement. While many manu- facturers are locating their factories in the small cities and towns, there are other industries that prosper most in the great cities. Commerce, moreover, emphatically favors the great centres, rather than the small or intermediate centres. And since, with ever-increasing production flowing from improved methods, commerce and trade are constantly ex- panding and absorbing an increasing proportion of the population, while manufacturing in a country where it has reached the stage of self-sufficiency employs a constant or even declining proportion of the population,^ the prospect is ^ Even should the law of diminishing returns require a larger proportion of the population to be engaged in raising food, there is no reason for pessimism re- garding the per capita wealth and prosperity of society. If, for example, the present distribution of the population of the United States (two-fifths in agri- culture, three-fifths in other industries) should be reversed a century hence, and three-fifths needed in agriculture, the remaining two-fifths, as a result of the law of increasing returns, would produce more and better form-utilities than do three- fifths at present. (Clark, Philosophy of Wealth, loi.) " Occupation statistics are still very imperfect, but the inferences of the follow- ing English figures are confirmed by the French statistics extending back to 1851 : Percentages of the Population. Agriculture and Trade and mining. Mfc. Transport. Various. 184I 22.7 27.1 12.9 37.3 1851 24.9 32.7 I5.I 27.3 1861 22.5 33.0 15.7 29.5 1871 18.7 31.6 18.7 31.0 1881 16.3 30.7 20.1 32.9 CAUSES 229 that the larger cities, including of course their suburbs, will continue to absorb the superfluous population of the rural districts and villages; Greater London, New York, Paris, Berlin and Chicago show no signs of falling behind the smaller cities in rate of growth. The occupations classed as " various " include building trades, civil service, army, navy, professions and non-working classes. These, together with the com- mercial industries, are increasing proportionally, and, as already pointed out, tend toward the larger cities. The progress of employments is discussed in chap- ter viii of Hobson's Evolntioit of Modern Capita lis7>i, whence the foregoing sta- tistics are derived (p. 230). The American statistics are incomplete for the earlier years of the century, but since 1840, at least, the agricultural population has been decreasing in relative numbers, while, of course, the manufacturing population has increased even to the present time, although not to the same extent as com- merce and transportation. In 1820, the percentage of the total population en- gaged in agriculture was 21.49; i^^ 1840, 21.79; in 1870, 15.43; in 1890, 13.68. The similar percentages for manufactures were 1820, 3.63; 1840, 4.64; 1870, 6.36; 1890, 8.13. (^Bulletin of the Dtpartjue'al of Labor, July, 1897, pp. 398-9.) The distribution of the " workers" (persons of 10 years old or over engaged in gainful occupations) at the last three censuses was as follows {Ibid., 397) : 1870. 1880. 1890. Agriculture, fishing, mining 49-1 1 46.03 39.6$ Manufacturing and mechanical industries. 19.61 19-63 22.39 Domestic and personal services 18.48 20.14 19.18 Trade and transportation 9.83 10.73 14-63 Professional services 2.97 3.47 4.15 ICO. 100. 100. CHAPTER IV. URBAN GROWTH AND INTERNAL MIGRATION. The enormous and unprecedented growth of cities during the nineteenth century is often regarded as the result of a great migratory movement from the farm to the town ; the process appears in full light when one studies the growth of the Lancashire district in the thirties, or of Chicago with its vast throng of oversea immigrants. But before the recent growth of cities can be attributed solely to the factor of im- migration from country districts at home or abroad, it must be shown that such immigration is of recent orgin, coinciding with the recent concentration of population. Such a demonstration will not be at once forthcoming. The fact is that migration cityward is not an economic phenomenon peculiar to the nineteenth century. The com- plaints of the Physiocrats, the first economists, about the scarcity of labor in the rural districts should be generally familiar. Quesnay, in his celebrated article Fermiers in VEncylopedie, noted that the most energetic and intelli- gent countrymen migrated to the cities, and attributed it to the expenditure of money in Paris and other large towns by the courtiers and nobles. The Physiocrats were in agree- ment as to the existence of a migration cityward, which they called depopulation of the rural districts, and declared was of long standing in France. It certainly dates back to the mercantilist and industrial policy of Colbert in the seven- teenth century, and Legoyt quotes a writer of the fourteenth century, who complained of the increased difficulty of 230] INTERNAL MIGRATION 231 obtaining farm labor at remunerative rates, as confirmatory evidence of a rural emigration.^ In France the official re- ports from the provinces to the Hats generaux recommended restrictive measures in order to keep a large supply of labor on the farms. But in England, where a similar migratory movement was at this time in evidence, the governmental point of view was the city instead of the country. Hence both Elizabeth and James I issued proclamations forbidding migration into London, whose population was swelling to portentous dimensions. In Germany, too, the evidence points to a large internal migration in the late middle ages, although it was in large part between the towns themselves.^ Biicher, indeed, does not hesitate to compare the migratory movements of the fourteenth and fifteenth centuries with those of the nineteenth, the underlying cause in each case being the transition from one stage of industry to another.' The most conclusive evidence of a large migration from the fields to the toWns, however, is afiforded by the bills of mortality begun in several cities in the sixteenth and seven- teenth centuries. These death reports formed the first material of the new science of demography or population statistics, at first known as "Political Arithmetic." Now these early bills of mortality almost uniformly showed more deaths than births each year; the natural result of which would be the decadence of the city. But on the contrary, ^Des Agglomerations Urbaines, p. 7 : " Leopold Delille, {Etude sur la classe agricole en Normandie au moyen-age^, raconte que les chanoines de Mondaie, en Normandie, se plaignaient en 1388 que 'Ton ne peu trouver serviteur pour culti- ver et labourer les terres qui ne veuille plus gaigner que six serviteurs ne faisent au commencement du siecle.' " ^ Cf. Bucher, " Die inneren Wanderungen und das Stadtewesen in ihrer ent- wicklungsgeschichtlichen Bedeutung," one of the brilliant essays in the collection entitled Die Entstehung der Volkswirthschaft. ' Ibid., 285, 295, 303. In the middle ages the self-sufficing town was more likely to import artisans than merchandise. 232 THE GROWTH OF CITIES the city grew; its population even increased more rapidly than the rural population. The simple explanation of such a state of affairs was a large emigration from the rural dis- tricts to the cities. And this was the conclusion of Captain John Graunt, the founder of the new science.' He estimated the annual immigration to London to be 6,000 persons.' While this number is purely conjectural, it raises a very strong presumption that migration to the metropolis was re- latively greater 250 years ago than it is to-day. For, between 1871 and 1881, with a population nine or ten times as large as in 1650, London's net immigration amounted to less than 11,000 per annum.3 ^ Natu7-al and Political Observations Made upon the Bills of Mortality, 4th Impression, Oxford, 1665, pp. 81-84 (ch. vii) : "The next Observation is, That in the said Bills there are far more Burials than Christenings. This is plain, de- pending only upon arithmetical computation, for in 40 years, from the year 1 603 to the year 1644, exclusive of both years, there have been set down (as hapning within the same ground, space or Parishes, although differently numbered and divided), 363,935 Burials and but 330,747 Christenings From this single Observation it will follow. That London should have decreased in its people; the contrary whereof we see by its daily increase of Buildings upon new Foundations, and by the turning of great palacious Houses into small Tenements. It is there- fore certain that London is supplied with people from out of the country, whereby not only to supply the overplus or difference of Burials above-mentioned, but like- wise to increase its Inhabitants, according to the said increase of housing." London's growth might also be seen in the increasing number of christenings (p. 72 : " The Decrease and Increase of People is to be reckoned chiefly by Christ- enings, because few bear Children in London but Inhabitants, the others die there.") Graunt's table of christenings in London (pp. 1 74-5) shows the fol- lowing increase : 1604-1 1 52,190 161 2-19 60,316 1620-27 62,124 1628-35 75»774 1636-43 80,443 London's population increased in the ratio from 2 to 5 in 54 years, while it took a typical rural district 200 years to double its population (p. 143). ^Ibid., 131 ff. 'In 1580 there were said to be 5,060 foreigners resident in London, which then had a population of about 150,000 — a larger proportion than now obtains of foreigners and colonials together. INTERNAL MIGRA TION 233 The reason why the stream of emigration did not make the cities grow so rapidly in former centuries as in the nineteenth is the excessively high death-rate then prevalent. Not only did poor sanitation exact a heavy tribute from the infantile population year in and year out, but it also favored periodi- cal visitations of the plague, which naturally wrought fearful havoc. Hence the difificulty of ascertaining a regular, uni- form rate of death or migration. It appears that while migration to the cities was large, it did little more than fill the vacant places caused by death. And Captain Graunt was probably right in saying that no matter how great the number of deaths caused by the plague, the city would be quickly re-peopled ; the influx of strangers would in the second year fill all the vacant places.^ But economy in the organization of industry has steadily demanded an increase in the number of city dwellers, and the cities have thus been able to absorb the migrants from the rural districts at the same time that they have found use for the net increases of their own populations, which have grown to large proportions as a result of the decline of death rates. Thus, statistics show that the migration into Berlin is now but slightly larger than it was in the first half of the century ; but Berlin is now growing about twice as rapidly as it was then. That is because the excess of births over deaths is now large,^ whereas in earlier times it was ' Ibid., p. 75 : " The next Observation we shall offer is the time wherein the City hath been Re-peopled after a great Plague; which we affirm to be by the second year. For in 1627, the Christenings (which are our standard in this case) were 8,408, which in 1624, next preceding the Plague-year 1625 (that had swept away above 54,000), were but 8,299; and the Christenings of 1626 (which were but 6,701) mounted in one year to the said 8,408, Now the Cause hereof, for- as-much as it cannot be a supply by Procreations; Ergo it must be by new Affluxes to London out of the Country." ^The following data were compiled by Kuczynski, Zug nach der Stadt, p. 252, and indicate the annual increase as a percentage of the mean population : 234 THE GROWTH OF CITIES very small or else vanished into an absolute deficiency of births. It would make an interesting piece of investigation to trace the diminution of the city death rate from the sixteenth to the nineteenth centuries, as set forth in the works of Graunt's successors, Petty, Halley, Siissmilch, Deparcieux, and Wappaus. But it might be tedious for the reader, and is at any rate unnecessary for present purposes. A few ex- amples will suffice to show the general tendency. A German student who investigated the church record of baptisms and burials in several German cities came to the conclusion that on the average there were 80 or 90 births to 100 deaths in the period 15 50-1 750. In the last fifty years (1700-50) of this period the number of births fluctuated between 66 and 96;'' but in 1877-82 the ratio of births to Excess of births. Net Total Number. Percentage. Immigration. Increase. I711-1815 — 31,310 — 0.2 1.4 I.I 1816-37 23,505 0.5 1.3 1.8 1838-58 55>5i3 0.7 1.6 2.3 1858-75 95.460 0.8 3.2 4.0 1875-95 289,240 I.I 1.6 2.7 The period just previous to the Franco-Prussian war includes the heaviest mi- gration to Berlin. (^Supra, Table XLV.) This period was exceptional, and as appears from the foregoing percentages of net immigration, the present move- ment toward Berlin is not greatly in excess of that in the earlier periods of the century. And the statistics of Fremdgeborenen in Berlin do not indicate that the percentage of outsiders is now perceptibly larger than it was in 1875 : Bom outside Berlin, to each 1,000 inhabitants. 1864 520.9 1871 563.7 1875 586.6 1880 566.3 1885 576.0 1890 593.0 Kuczynski also shows (pp. 262-270) that the age-grouping of the Berlin pop- ulation nas not greatly changed since the beginning of the century, indicating that a large immigration then as now filled the middle age periods. ' J. Wernicke, Das Verhdltniss zwiscken Geborenen und Gestorbenen in histor- ischer Entwicklung (Conrad's series of dissertations) , pp. 57, 90. INTERNAL MIGRA TION 235 deaths in 173 German cities was 147 to 100. Sir William Petty, writing in 1681, estimated the ratio of births to deaths in London at 5 : 8 or62j4 to 100, while in all England he said it was 125 : 100.^ Now the date at which the cities succeeded in turning the excess of deaths into an excess of births naturally varies according to country and circumstance. The very stream of immigration which was to maintain the population of a city was one circumstance ; for it brought strangers born outside the city to die in the city. Graunt noticed it but casually, saying that the 6,000 strangers who annually came to London added 200 to the burials every year. Edmund Halley was the first to emphasize the influence of immigra- tion upon the death rate in the cities : " Both London and Dublin by reason of the great and casual accession of stran- gers who die therein (as appeared in both, by the great excess of the funerals above the births) rendered them inca- pable of being standards for this purpose, which requires if it were possible, that the people we treat of should not at all be changed, but die where they were born, without any adventitious increase from abroad or decay by migration elsewhere."^ In Breslau, with its small migratory move- ment, Halley found a small surplus of births over deaths. The disturbing effect of migration upon the relation be- tween births and deaths, thus first emphasized by Halley, has been discussed by all subsequent writers without exhausting the subject. Deparcieux demonstrated that while migration into the city might increase the number of deaths as com- pared with the number of births, it diminished the ratio of ^ Several Essays in Political Arithmetic, 4th ed., London, 1755, p. 36: "Ob- servations upon the Dublin Bills of Mortality," 1681. ^ Philosophioal Tra7isactions of the Royal Society, vol. xvii, for the year 1693, No. 196 : E. Halley, " An estimate of the degrees of mortality of mankind, drawn from curious tables of the births and funerals at the city of Breslau; with an attempt to ascertain the price of annuities upon lives." 236 THE GROWTH OF CITIES deaths to the living, since the migrants were persons in the active, healthful years of life, with a death rate lower than the average.^ Hence as respects France the death rate was found more favorable in Paris than in the small towns ; but one reason why Paris was the first great city to establish a clear preponderance of births over deaths was the French practice of sending infants to the coutry to be nursed. Their births were recorded in Paris ; their deaths in the country. While Paris could show a small natural increase in the eighteenth century,^' London did not succeed in doing so until the beginning of the nineteenth ; a result achieved more by the diminution of deaths than by the increase of births.3 Berlin first attained a similar permanent status after iSiOj-^ ^ Essai sur les probabilitiis de la durie de la vie htimaine ; d'ott I'on dedtiit la mani^re de determiner les rentes viagh-es tant simples qu' en tontines, Paris, 1746. * The annual average number of births in excess of deaths in Paris (according to Levasseur, ii, 395) was: 1750-59 323 1 780-89 27 1799-1808 —668 1809-16 373 1817-30 3,177 ^ The data for London are as follows, the capital letter " P " designating the visitation of the plague or an epidemic (^Encycl. Brit., Art. " London ") : Year. Deaths. Births. Excess of deaths. 1593 17.844 4.021 13,823 1603 42,042? 4,789 37.253 1625 54.265 P 6,783 47.482 1636 22,359 9,522 13,837 1665 97.306 P 9.967 87,339 (Annual average.) I700-1710 21,461 15,623 5,838 174O-175O 25,352 14,457 10,895 1790-1800 24,270 22,605 1.665 ' The excess of deaths over births in Berlin fluctuated very considerably, and at times yielded to an excess of births (Kuczynski, 252; see also supra, foot-note p. 234) : 1709 428 1770 —10,372 1720 +512 1780 — 2,708 1730 —3,581 1790 +2,793 1747 —447 1800 —7.089 1755 .....—12,334 1810 +1,296 1763 +592 INTERNAL MIGRATION 237 while Leipzig reached that point in 1821-30, and Frank- fort in 1840/ But the clearest picture of the whole move- ment is furnished by the statistics of Sweden,'' where scien- tific carefulness and accuracy were early devoted by the government to the collection of population statistics : Table CXVI. Births per 1,000 population: 1816-40. 1841-50. 1851-60. 1861-70. 1871-80. i88i-go. Rural population 33-34 3i-3o 32.82 31.20 30.21 28.65 Urban population 3o-95 29.29 32.53 32.95 32.13 31-07 Stockholm 33.03 32.59 35.49 34.56 31.75 32.39 Other cities 30.11 28.07 3^-5° 32-38 32.25 30.59 Deaths per 1,000 population: Rural population 22.26 iQ-70 20.57 i9-33 17-32 16.36 Urban population 34-44 28.73 31.20 26.17 24.05 19-74 Stockholm 45-09 38.11 41-51 32-25 30.28 22.60 Other cities 30.10 25.24 27.60 24.01 21.96 18.71 Natural increase per 1,000 population: Rural population 11.08 11.60 12.25 11.87 12.89 12.29 Urban population — 3.49 0.56 1.33 6.78 8.08 11.33 Cities of less than 10,000 7.1 7.7 9.7 Cities over 10,000 pop 6.7 8.2 ii.g Stockholm — 12.06 — 5.52 ^-6.02 2.31 1.47 9.79 Other cities 9.3 12.0 13.2 Cities without Stockholm o.oi 2.83 3.90 8.37 10.29 11.88 Gain or loss ( — ) by migration per 1,000 population: Rural population — 1.45 — 1.67 — 3.22 — 6.19 — 6.35 — 12.03 Urban population 12.62 14.12 20.28 14-49 16.32 14.89 Cities under 10,000 pop 6.7 13.7 7.1 Cities over 10,000 17.9 17.4 17.9 Stockholm 17-70 15-61 25.49 ^5-99 19.66 26.95 _ Other cities 19.2 i6.t 12.6 Cities without Stockholm 10.55 13-56 18.45 12.95 15.21 10.55 Total increase per 1,000 population: Rural population 9.63 9.93 9.03 5.68 8.54 0.26 Urban population 9.13 14.68 21.61 21.27 24.40 26.22 Cities under 10,000 13.8 21.4 16.8 Cities over 10,000 .... .... 24.6 25.6 29.8 Stockholm 5.64 10.09 i9-47 18.30 21.13 36-74 __ Other cities 28.5 28.1 25.8 Cities without Stockholm 10.56 16.39 22.35 22.32 25.50 22.43 ' Bleicher, p. 239. It is worth noting that in Frankfort, a city of fairs, there was usually an excess of births in the citizen class : Aggregate excess of births (+) or deaths ( — ) : Frankfort. Leipzig. Entire pop. Resident class. 165I-1700 —1,056 +3.521 —55638 1701-1750 —6,059 +1,259 —9.310 1751-1800 —11,975 —2,224 —19.687 1801-1840 — 691 +1,513 — 2,452 1841-1890 -f 29,266 +43.153 ' Supplement to the census of 1890, Befolkningsstatistik, new series, xxxii. No. i : Bihang till Statistika, Centralbyraus Befolkningsstaiistik for Ar i8go, Folks- mdngdens Fordndringer Sverige Aren i88i-go, yenie Ofversigter for Aren i8i6-go (Stockholm, 1892), pp. ii-vii, etc. 238 THE GROWTH OF CITIES The percentages of total increase show how the urban population has been growing ever faster and faster, while the rural growth falls off. The urban population is, more- over, tending to be absorbed in Stockholm. But the per- centages of gain or loss through migration do not indi- cate that the more rapid growth of Swedish cities is due to immigration. Emigration from the rural districts has indeed increased, but much of it has been directed to foreign countries ; the net immigration to the cities has diminished since 1851-60. Even in Stockholm the immigration for 1881-90 barely exceeds the percentage of 1851-60. The real explanation of city growth is therefore to be found in the percentages of natural increase. In the first period, 1816-40, the per mille of loss {i. e., excess of deaths over births) was 3.49; in the succeeding decade this was turned into a positive gain of 0.56 per 1,000, which has steadily in- creased in the following periods until it reached 11.33 per mille in 1881-90, or almost as much as the rural natural in- crease. The tendency is even more marked in Stockholm, where a deficit of 12.06 per mille has finally been turned into an excess of 9.79. Further analysis shows that the birth-rate is higher in the urban than in the rural communities; although the differ- ence is not so great as in the case of the death-rates. But the difference between the urban and rural death-rates is now only 3.4 per mille, whereas in the period 1816-40 it averaged 12.2, and in the case of Stockholm nearly reached 23 per 1,000. Stockholm's death-rate of 45.1 per mille in 1816-40 was not an exceptional rate for cities in that period ; and great alteration for the better is typical of modern cities. This is the real explanation of the manner of city growth. The point of self-maintenance, which was reached in Paris before the close of the eighteenth century, in London in 1800, in the German cities in the first half of the present INTERNAL MIGRA TION 239 century, in Stockholm after i860, has not yet been univer- sally attained even in civilized Europe. In 1877 Dr. Dunant presented a paper' to the International Congress of Medical Science at Geneva in which he showed that of 30 great cities of Europe, 23 owed more than one-half their growth to immigration, and that seven ^ of these without it would have decreased. They were mostly Italian cities, but more recent statistics^ show that six of twelve great cities of France are also subject to an excess of deaths. On the other hand, a considerable number of English cities are losing more by emigration than they gain by immigration ; although as will appear hereafter the emigration is in large part directed toward the suburbs. The following list of the larger European cities {i. e. those having a mean population of at least 200,000 in 1 880-1 890 or 1881-1891) shows the proportion which immigration (or emigration) bears to the total increase per 1,000:* Immigration. Marseilles I>i83 Amsterdam 502 Lyons 1,115 Copenhagen 492 Bordeaux 1,060 Brussels 489 Rome 893 Vienna 410 Turin 881 Leeds 300 Buda-Pest 867 Birmingham 257 Milan 830 Naples 257 Munich 822 Edinburgh 94 Stockholm 734 Palermo 30 Paris 723 Manchester 717 Emigration. Breslau 715 Sheffield 4 Prague 700 Dublin 1 20 Berlin 697 Bristol 785 Hamburg 690 London 1,289 Belfast 654 Liverpool (decrease) 2,481 Dresden 588 ^ Injluence de r £tuigraiion de la Population des Campagnes dans les Villes ; published also in Annates de Dtmograpkie Internationale, vol. i. The important results are restated by Levasseur (ii, 386), and illustrated with a diagram. ^ Milan, St. Petersburg, Venice, Odessa, Prague, Rome, Naples. * Cf. Statistisches yahrbuch der Stadt Berlin for 1892, xix, 94-5 : " Movement of population in 88 European cities for ten years." * Ibid. The Italian figures are of slight value, because their population is esti- mated, no census having been taken since 1881. 240 THE GROWTH OF CITIES According to these figures, immigration would seem to play the largest role in the growth of the French and Italian cities, then in the German, the Scandinavian, and finally the English cities. The explanation lies in the fact that the English cities have a lower death-rate than the Italian and a higher birth-rate than the French ; in either case they have a larger excess of births over deaths, and derive a corre- spondingly smaller fraction of their growth from immigra- tion. The following table covers several decades and takes into account annexations of territory as well as natural in- crease and net immigration : ^ Table CXVII. Percentage of total increase due to Excess of Net immi- Incorporation births. gration. of suburbs. London, 1852-91 . . . . 84.03 15.97 Copenhagen, 1801-90... • 42.98 57-02 Cologne, 1821-90... 34.10 28.48 3742 Berlin, 1801-90... 26.72 Vienna, 1801-90.. . 20.52 32.48 47.00 Leipzig, 1801-90... 16.42 39.70 43-88 Paris, 1821-90.. . 15,23 64.21 20.56 Breslau, 1821-90... 15.16 79.21 5-63 Munich, 1811-90... 10.59 72.26 17-15 St. Petersburg, , 1801-90... . —26.81 (deficit.) The fact that London's percentage of growth due to natural increase is so large arises partly from the fact that no changes of area are made, and partly from the fact that the data are comparatively recent; but after all, the real reason is London's precedence in the making of sanitary im- provements. From the data concerning 88 European cities which Pro- fessor Boeckh publishes in the Berlin municipal Jahrbuch, he draws the conclusion that migration is a wider movement ^ Sedlaczek, " Die Bevolkerungszunahme der Grossstadte im XIX Jahrhundert und deren Ursachen," in Proceedings of Eighth International Congress of Hygiene and Demography at Budapest (1894), vii, 380. INTERNAL MIGRA TION 241 than the natural movement of births and deaths ; for in the former case the percentages range between +28.6 and — 16, and in the latter case between +19.1 and — 4.6/ As Table CXVII demonstrates, Professor Boeckh is entirely safe in concluding that fully one-half of the increase of population in the large cities is to be attributed to migration.^ In the United States, accurate vital statistics are too scanty to permit wide generalizations. Moreover, frequent suburban annexations, which are not always mentioned in the census reports, complicate matters. But if Boston be regarded as a typical American city, investigation shows that about one-half of the increase in population between 1865 and 1890 was due to immigration, three- tenths to annexations and two-tenths to natural growth, which is thus seen to play a subordinate part in the increase of Boston's population.* Statistics are also available which allow a comparison to be made of the natural increase in urban and rural communities in Massachusetts ; the urban population is here represented by the 28 incorporated cities, and the following data are averages for the five census years 1870, 1875, 1880, 1885, and 1890:3 Birth-rate. Cities 28.4 Rural remainder 22.0 '^ Op. cit., vii, 392 and 384: " Der Antheil der ortlichen Bewegung an der Zunahme der Bevolkerung der Grossstadte." ^ Cf. Mass, Census of i8gS, i, 220, and Forty-ninth Registration Report (1890), p. 156: Population of Boston in 1865 192,318 " " Roxbury, Dorchester, Charlestown, West Roxbury, Brighton in 1865 76,288 Births minus deaths (245,958—194,890) in 1865-90 51,068 (Add for net immigration in 1865-90 128,803) Population of Boston in 1890 448,477 ' Registration Report as above, p. 372. Rate ol Death-rate. natural increase. 21.4 7.0 17-5 4-5 242 THE GROWTH OF CITIES It is significant that the cities, instead of being the " de- stroyers of mankind," now produce a larger surplus of births than do the country districts. This result is due to a high city birth rate, as the urban death rate in Massachusetts still remains considerably above the rural death rate. Massa- chusetts may be taken as a typical state of the manufacturing East; in fact, entire New England is even more favorable than Massachusetts to the urban rate of natural increase."^ Re- garding the role of immigration in the growth of Massachu- setts cities, it may be noted that the increase of population in the ten years 1885-95 on the territory of the present 32 incorporated cities was 38.05 per cent., or about 38 per 1,000 for one year;^ but as was just indicated, the natural increase in the Massachusetts cities is only about 7 per 1,000 annually. The value of these Massachusetts statistics is considera- ble, especially as they have been confirmed by computations showing the refined birth rates; 3 for Massachusetts is the ^ The Sumt/iary of Vital Statistics of the Aew England States, iox 1892 (p. 56), gives the following figures : Population, 1892. Birth-rate. Death-rate. Urban group (towns of io,ooo-f ) • •• 2,441,418 est. 29.68 21.01 Rural group (towns of I o,coo — ) ... 2.444,987 est. 2C.oo 18.72 The natural increase is thus 8.67 and 1.28 per wzV/? for the urban and rural populations respectively. ■'' Mass. Census of i8gS, i, 49. •^See Dr. Crum's article in the Quar. Jour, of Econ., xii, 259, and one of his tables reprinted below (No. CXLV.) . While his figures are based upon density of population, it will be shown that the grouping by density corresponds closely to grouping according to the populousness of cities. With Dr. Crum's figures may be compared the birth and death rates and rate of natural increase in Massa- chusetts towns in 1890 (^Registration Report, as above, p. 374) : Number of Population of Birth Death Rate of towns. the groups. rate. rate. natural increase. 95 Under 1,000 15.5 17.5 — 2.0 84 1,000- 2,000 16.1 18.1 — 2.0 48 2,000- 3,000 19.0 18.7 0.3 22 3,000-4,000 21.6 17.I 4.5 30 4,000- 5,000 21.6 18.0 3.6 35 5,000-10,000 24.7 17.1 7.6 17 10,000-20,000 27.4 17.9 9.5 r. [from 25.7 17.0 5.8 20 Over 20,000 <^ . J'' ', ,^ „ ' (,to 34.4 22.5 12.2 INTERNAL MIGRA TION 243 leading industrial commonwealth of this country, and the type of that organization which seems destined to prevail more and more with the passage of years. It denotes the fact that the cities are no longer mere consumers of popula- tion produced in the rural districts, but will contribute their full share to the increase of population. In Europe the urban population as a whole still has a smaller natural increase than the rural population, but the difiference is not so great as in the case of the great cities. The Swedish statistics given in Table CXVI may be again referred to. It may be concluded that while the urban pop- ulation has a lower death rate in Europe than in America, its birth rate is in comparison still lower. In England the nat- ural increase in town and country is almost precisely the same;^ hence the more rapid growth of the towns is due to migration. The relation which this immigration has borne to the natural increase may be seen in the following figures,'' based on the assumption of a uniform natural increase throughout all categories of population, and showing the * Cf. the following figures given by Charles Booth (" On Occupations of the Peo- ple, 1801-81," in Jour, of Stat. Soc, 1886, p. 329) : Birth-rate. Death-rate. Nat. increase. London and 19 chief towns 37-21 Fifty large towns 36.79 Small towns 37-67 Urban population 37-12 22.09 14-03 Rural population 33-13 19.00 I4-I3 Cf. also Longstaff, Studies in Statistics, p. 25, whence the natural increase for 1871-81 per 1,000 population in 1871 may be deduced as follows: London 13.9 19 large cities 14.2 56 other cities 16.8 Total 76 cities 15.0 England and Wales 15.0 ^Sir Rawson W. Rawson, in your, of Stat. Soc, 1880, p. 501. Dr. Longstaff, op. cit., p. 24, prints an interesting table which shows the daily increase, daily 244 THE GROWTH OF CITIES percentage of gain or loss by migration in each decade, as well as the proportion which the migration bears to the total increase (or decrease) : Table CXVIII. Gain or loss (— -) by migration Ratio of net immigration to total increase. Other large Small Rural Other large Decade. London. towns. towns. districts. London. towns. 180I-H ... 4,46 6.00 — 1.20 — 2.19 23.8 29.7 181I-2I . . . 3.02 8.61 +0.94 —3-32 14.6 32.3 I 82 1-3 I ... 4.21 14.23 — C.83 —5-29 21.0 47-4 183I-4I ••• 3-25 12.00 — 2.02 —4-79 18.3 45-3 I 841-5 I ... 8.5s 12.12 —2.18 —6.80 40.2 48.8 185 1-6 I ... 6.75 7-77 —4.64 —4.64 36.2 39-4 1861-71 ... 2.78 7.72 — 2. 1 1 -4-83 17.4 36.8 These figures apparently indicate that the influx into London reached its height about 1850, as indeed did the migratory movement cityward in general. Since then London's gain by migration has diminished, and in 188 1-9 1 turned into an actual loss, its natural increase being 12.71 per 1,000 and the actual total increase only 9.86. It is only fair to remark, however, that part of the emigration from London is simply into the suburbs. In Germany, too, migration cityward can hardly be said destination, and daily migrants per i,coo, in several categories of population in England and Wales : Population Daily Daily Daily in 1 87 1. increase. destination, immigrants. London 143 133 165 +32 19 large cities 142 135 156 -|-2i 56 other cities 127 140 199 +59 Remainder of country 588 592 437 — 155 Foreign countries ••• 43 ' +43 I, coo I, ceo 1,000 INTERNAL MIGRATION 245 to be increasing, while the urban rate of natural increase on the whole approaches that for the entire country:^ Table CXIX. Twenty-five great cities of Germany. Prussia. Natural Net immi- Total Natural increase. gration. increase. increase. Jt86i-64 8,3 27.4 35.7 14.3 1864-67 4.3 1 7.7 22.0 10.9 1867-71 6.1 22.1 28.2 9.5 1871-75 10.4 21.7 32.1 12.3 1875-80 12.6 12.7 25.3 13.8 1880-85 9.9 14.3 24.2 12.0 In France the cities, considered in the aggregate, have virtually ceased to grow of themselves, but rely upon im- migrants from the rural districts for recruiting their popula- tion, which, as we saw in the first chapter, is by no means at a stand-still. The following data relate to the urban popu- lation of France:'' Table CXX. Aggregate numbers. Percentages. Total Excess of Net immi- Natural Immigra- increase. births. gration. increase. tion. 1861-65 805,582 141,350 664,232 17. 83. 1872-76 742,497 117,667 624,830 16. 84. 1876-81 1,119,146 38,480 1,080,666 3. 97. 1881-86 669,966 43,665 626,301 6, 94. 1886-91 544,784 — 1,129* 54S>9i3 — 0'2* 100.2 1872-91 3.076,393 198,812 2,877,710 6. 94. It thus appears that in France the percentage of the growth of cities due to their natural increase has diminished since 1861 until in the last period there was an actual defi- ciency. Migration into the cities, although not increasing in absolute numbers, has assumed relatively greater prom- ^ Allg, Stat. Archiv.y i, 167. '^ Resultats siatistiques dii denombretnent de i8gi, p. 72; and other census reports. •' Excess of deaths. * Decrease. 246 THE GROWTH OF CITIES inence. The statistics of 1876-81 may be further analyzed to show the movement of population in cities of different size. Without giving the actual numbers, the percentages may be reported as follows:^ Natural increase. Immigration. Paris 8.3 91.7 47 cities (pop. of 30,000-}-) 2.5 97.5 All cities except Paris 1.8 98.2 Total urban i 3.4 96.6 From this it appears that the deficit is due not so much to Paris and the larger cities, as to the smaller cities. But such statistics are not absolutely conclusive. In Austria the natural movement of population in the cities contributes more largely to their growth than in France. Thus, of the total gain, in 1880-90, of 193,341 in the eight larger cities (50,000-1-), 79,395 or 41. i per cent, came from the excess of births over deaths.^ In Hungary the natural increase constituted 32 per cent, of the total increase, 1880-90, both in the seven large cities (50,000+) and in the 25 free cfties.3 The results thus show the greatest diversity. In France the cities do not sustain themselves, nor do many of the Italian cities. In Germany, Sweden, Austria, Hungary, etc., the cities furnish from one-fourth to one- half of their in- crease according to size. But in Great Britain immigration has so diminished that even the largest cities provide three- fourths and often more of their increase. In the United States, where the cities now show a larger natural increase than do the rural districts, there is still a vast immigration, four or five times as large as the natural increase. ^ M. Loua, in yotirnil de la Societe Statistique de Paris (March, 1885) xxvi, 124. ^ Calculated from data in Stat. Monat., xviii, 234-46. ^ Calculated from data in the Hungarian census of 1890, pp. 55-57*. INTERNAL MIGRATION 247 The method here used is liable to grave inaccuracies, since it assumes that the original population of a city is not going and coming, and that a resident population is alone respon- sible for the births and deaths. Now between any two points of time for which the natural increase is computed, the deaths registered may be affected by the emigration or immigration of particular classes, say of old people. If old people come to the city to spend their last days, it is evi- dent that the number of deaths will be artificially increased, and vice versa when old people emigrate from the city. In the latter case we should have too few deaths, which would make the natural increase larger than it actually is and the immigration correspondingly smaller. Similarly, the num- ber of births may be affected in two ways : ( i ) it may be increased by the registration of children born in the city and removed before the census has been taken, or (2) by the birth of children to women who have moved to the city within the period under consideration. In either case the result would be the same, the city would appear to grow by reason of surplus of births, whereas it might be supported entirely by immigration.^ All these objections, however, are of theoretical rather than practical importance in any investigation of tendencies ; they are not of sufficient force in reality to invalidate conclusions based on comprehensive data such as those presented in the present chapter. Another method of approaching the question of city- growth is by means of statistics of birth-place. If, for example, 38.5 per cent, of the population of Boston were born there and 62.9 per cent, of London's population were native Londoners, it may be inferred that, other things being equal, Boston is receiving a proportionately larger ^ This is substantially the argument of Ballod, Die Lebensfdhigkeii dcr stddt- ischen tind Idndlichen Bevolkening (1897). 248 THE GROWTH OF CITIES immigration than London.' Hence for the study of internal migration the statistics of birth-place have a real value and will repay careful analysis.^ The vast dimensions of internal migration are often lost sight of in the contemplation of the horde of emigrants who go to foreign countries, though we in the United States, who have seen these shiftings of the native American population, are not so likely to fall into mistaken comparisons as are European students, whose attention is attracted to the loss of millions of their fellow citizens by trans-Atlantic migra- tion. In England, in 1891, 25 per cent, of the native born inhabitants were no longer residing in their native county, amounting in round numbers to 7,000,000 souls ; whereas the number of Englishmen residing in the United States in 1 890 was only 900,000. The following table, based chiefly on the official sources, ^ The assumption here made of " other things being equal " should be carefully noted. Circumstances may be conceived in which the conclusion would not fol- low. Suppose, for instance, that cities A and B are of the same size {e. g., i,ooo,- 000 inhabitants), and have the same annual increase, say 50,000, of which 25,000 represents the excess of immigration over emigration, and 25,000 the surplus of births over deaths. Now it might occur that city A had no emigration at all, in which case 25,000 would represent the pure immigration, and the immigrants would, in the first year, constitute 2.5 per cent, of the population of city A. But city B, we will say, sends out 100,000 of its population to other communities, and receives an inflow of 125,000, the net immigration being as in town A, 25,000. But in city B 12.5 per cent, of the population would now be outsiders. Hence it might be concluded that London's large percentage of native Londoners is due not so much to a higher birth-rate and lower death-rate than Boston's, as to less emigration. Nor is it possible to ascertain the amount of emigration from the city. Though the English census may return the total number of the natives of London within the British Empire, it does not give the number who are living in France or America. ^ In Germany they have a system of police registration of arrivals and depart- ures, and these Anmeldungen, as the notices are called, are sometimes used to study migration. But it is obvious that these data can be of little value, inas- much as they refer to cases of migration and not to the number of migrants. Cf. Bruckner's article in Allg, Stat. Archiv, vol. i. INTERNAL MIGRATION 249 exhibits the range of migration in several of the leading countries:' Table CXXI Percentage of the total populat on bom in Township Elsewhere County Elsewhere Foreign where in same where in same coun- Country. Date, enumerated. county.^ enumerated. country. tries.* Massachusetts, 1885... . 36.1 57-5 42.5 United States, 1890 66.86 18.37 14-77 Saxony,'' 1885 • 50-07 19.03 69.1 20.82 10.08 Prussia, 1890 • 53-9 15-8 69.7 27.1 3-2 Eng. and Wales 1 891 71.6 24-5 3-9 Denmark, 1890 77.88 18.62 3-5 France, 189I • 56.3 25.0 81.3 16,4 2-3 Switzerland, 1888 • 56.4 25-7 82.1 "-5 6.4 Austria, 1890 • 65.2 15.0 80.2 18.1 1-7 Belgium, 1890 • 65.2 2.8 Netherlands, 1889 • 65.4 21.6 87.0 II. 2 1.8 Norway, 1875 • 73.05 14-15 87.2 10.7 2.1 Hungary, 1890. • 73.6 15.6 89.2 9-3 1-5 Sweden, 1880 • 79-9 8.6 88.5 II. I 0.4 The table represents only the internal migration, not the migratory tendencies of the different peoples. The Swedes, from their position in the table, might be called a non- ' Cf. Wirminghaus, " Stadt und Land," in Conrad's yahrbucher filr Naiional- oekonomie und Statistik, Ixiv, 1 61, and Ravenstein, "Laws of Migration," in your, of Stat. Soc. (1889), lii, 241. 'In Saxony the birth place is given by towns or villages {Orte), since the township QGemeinde) is virtually identical with the town. ^ The " county " or district stands for the following political divisions : Saxony, Amtshauptmannschaft -, United States, State or Territory; Prussia, Kreis; Den- mark, Overovrighedskredsene; France, Department; Switzerland, Canton; Austria, Bezirk; Netherland, the Province; Hungary, Comitat; Sweden and Norway, Lan. *The term "foreign countries" includes all States and federal commonwealths outside the State specified. Thus in the case of Massachusetts, it includes all other American States and Territories; similarly in the case of Saxony and Prussia, the other members of the German Empire. Of the 42.5 per cent, attributed to foreign countries in the case of Massachusetts, 15.3 per cent, were native Americans, and 27.2 foreigners in the usual sense of the word. 250 THE GROWTH OF CITIES migratory people, while we know that proportionate to their numbers they send more migrants to the United States than any other country in Europe exeept Ireland.^ There are some other considerations that prevent these figures being accepted as entirely trustworthy indications of the relative strength of migration. The size of the township often seri- ously affects the statistics of birth place ; if the towns are ter- ritorially large, a short-distance migration might effect no change of residence, whereas, in case of townships of smaller area, even a half mile journey might make the migrant a res- ident of some other town. A similar result would follow if the townships of one country were perfect squares or circles, while in another country they were very irregular in forma- tion ; or if some townships were bounded by natural barri- ers of mountains or water, rendering them more or less iso- lated. While these facts must make us realize the short- comings of such statistics as are given in Table CXXI, they do not invalidate the table. While the fact that the Prussian township is very small in area compared with the others may partly account for its large inter-town migration ; and while in Saxony the primary unit is not the township but the vil- lage or compact dwelling centre, the fact remains that in Massachusetts, which has the largest inter-town migration of all, the township is several times larger than any of the European townships.^ This is true even if we exclude the for- eign-born element entirely ; for the percentage of native Mas- sachusetts people born in the township where enumerated is only 49.4. This brings out the fact of the superior mobility of Americans, which has long been familiar to us in a general way. Indeed, it appears from the table that Americans are more accustomed to migrate from State to State than are Europeans from county to county. The English are appar- ently the most mobile people of Europe, as regards internal ^ Supra, p. 152. ■* Supra, p. 142. INTERNAL MIGRA TION 251 migration at least; and yet the percentage of native English- men living outside their county of birth was in 1871 almost exactly equal to the percentage of native Americans living outside the State in which they were born — the percentages being 25.96 and 26.2 (1870) respectively. Within the last quarter of a century, however, internal migration in the United States has declined. Professor Will- cox has shown that interstate migration in the United States reached its maximum in 1860-70; in only three States, in- deed, (Maine, Indiana, lUinois,) has the maximum migra- tion occurred in a later period. The data for intrastate mi- gration are less conclusive, but in the three States which afford such data the indications are all in the same direction.^ On the other hand conclusive evidence exists that internal migration is on the increase in Europe, though not neces- sarily the migration cityward. As the evidence in favor of this statement has never been presented in comprehensive form, to the writer's knowledge, the following percentages drawn from the official sources may be of interest; in ^ Professor Willcox has computed the number of native New Yorkers hving in the United States in three State census years with the following result : Resident in 1855. 1865. 187^. County of birth 56.0 55.3 57.S Some other county of New York.. 19.9 17.8 16.0 Some other State 24.2 26.5 26.2 Thus the percentage of New Yorkers residing outside the county of birth was smaller in 1875 than in previous years. No later data are available in New York, the State census of 1892 having been a mere enumeration, but in Massachusetts there was a continued decline after 1875 : Natives of Massachusetts resident in 1875. 1885. Town of birth 48.61 51.00 Some other town of Massachusetts 30.62 29.46 Some other State 20.77 19.54 In Rhode Island, also, the percentage of natives resident in town of birth in- creased more than the other categories. — "The Decrease of Interstate Migration," in Pol. Science Qnnr., x, 603-614. 252 THE GROWTH OF CITIES several of the countries the statistics of birth-place refer to a larger civil division than the township, such as the county or Department : Table CXXII. Percentages of the population residing in the native Department County in England Lan in in France. i860. 1866. 1870. 1871. 1876. 1880. 1881. 1886. 1890. 1891. 857 84.0 83.2 and Wales.' 74.04 75-19 74.86 Table CXXIII. Sweden. 92.8 90.8 District in Denmark.' 83.15 Percentages of the population born in the town where enumerated: Prussia. Austria.- Switzerland.^ Netherlands. Belgium. 1846 1849 1850 1856 1859 i860 1866 1869 i?7o 1871 1879 1880 1885 1888 1889 1890 56.8 57-6 54.3 534 78.7 69.7 63-9 64.0 58.7 54.0 48.7 45-9 69.09 68.90 68.29 67.22 65.4 70.2 69.1 69.4 67.2 65- In Ireland internal migration has on the whole increased, ^ The percentages for England and Denmark are of the native, not of the total population. ''■ In Austria and Sw^itzerland the figures express the percentage of population residing in the town of legal settlement. INTERNAL MIGRA TION 253 although rather slowly. The following statistics of selected counties are given because they cover a longer period than do any of the more general British statistics : Percentage of population born and still residing, at census, in Dublin county. Leitrim county. Belfast county. 1841 73.8 95.6 84.9 1851 69.2 94.2 72.6 1861 68.9 92.7 75.2 1871 64.8 93.4 76.5 1881... 62.6 93.5 77.3 189I 62.7 92.8 76.9 Of the larger countries named in the above tables, Prussia and Austria represent one extreme and England and France the other. In Austria internal migration has had a remark- able increase in the last twenty years. In Prussia the in- crease has been large since 1880; in the preceding decade mobility apparently decreased, but this was no doubt due in some measure to the displacement of population in 1871, the year of the war with France. Now Schumann has shown that interstate migration in Germany is chiefly in the direc- tion of the centres of industry and commerce, ' — a fact which gives us the explanation of increasing mobility of the Ger- mans, for no country in Europe has progressed so rapidly in manufacturing industry in the last quarter century as has Germany. In Austria, too, as Rauchberg has conclusively shown,"* the current of migration is toward the cities, and the reason that it has increased so much is the transformation of Austria from an agricultural to a manufacturing and com- mercial country, which is now taking place. In England and France, on the other hand, this transformation took place long years since, and for that reason we see no marked tendency toward the increase of migration at pres- ent. Switzerland apparently has an increasingly mobile ^ " Die inneren Wanderungen in Deutschland," 'va.Allg. Stat. Archiv., i, 518-9. ' In Statistische Monatschrifi, xviii, 230 and 562. 254 THE GROWTH OF CITIES population ; but the figures given, which refer to the place of settlement, are dependent upon the laws of settlement and cannot therefore be accepted without reserve/ Owing to their peculiar situation and composition, the Swiss cantons are particularly liable to a large foreign immigration,^ and do actually receive more immigrants than any other country in Europe.3 Such immigration of course affects the figures of Swiss mobility. In Scandinavia, Holland and Belgium the internal migration has increased slightly more than it has in France. On the whole, it may be said that in the last quarter of a century the tendency to migrate has in- creased in a direct ratio with the distance east from the settled manufacturing and commercial countries of Western Europe. Great mobility in the latter countries was attained some decades ago, and now the other nations are rapidly ap- proaching thereto. As we have seen in Table CXXI, Eng- land, Germany and France now lead the European nations as regards mobility, but the other countries of Western Europe are not far behind, while Northern and Eastern Europe are still bound to the old order of things with settled populations. It may be conjectured that at the present time the most rapid increase in migration would be found in Russia, which is now going over from the village economy to the national ; but we are without statistics to support the supposition. But while there are considerable differences among the countries as regards the inter-town migration, there is much less difference in the inter-county migration. In France, Switzerland, Netherlands, Norway and Hungary more than one-half of the migration is in fact within the department or ^Thus the percentage of native town inhabitants was 63.8 in i860 and 56.4 in 1888, showing a smaller increase in migration. " Switzerland is bounded by France, Germany, Austria and Italy, and her peo- ple are allied in race and language to all four nationalities. » See Table CXXI. INTERNAL MIGRATION 255 province. Even in countries with a large migration over county lines, it is to be presumed that short-distance migra- tion prevails. Thus, among the countries with the smallest percentage of the native county element is England, and even in England the larger portion of the internal migration is confined to neighboring counties.^ Since migration is as active in England as in any settled country, we may regard as well established the proposition that internal migration is predominantly of the short- distance character."^ What, now, is the cause of this short-distance migration? And how is it connected with the growth of cities and the concentration of population? Both of these questions can be answered with a few statistics. Let us first look at the constitution of the native population of the German Empire in December, 1890, according to place of birth. The free city of Hamburg contained 294,174 Germans born in other States of the Empire, while only 47,674 Hamburgers were found residing outside their native city. Hamburg had therefore gained by the migration 246,500 or 686.5 P^^ cent, of its native population. This was the largest gain made by any province of the Empire. The free city of Bremen was second with 400.7 per cent., then followed Brandenburg and Berlin 21 1.3 per cent., Alsace-Lorraine 93.7 per cent.. Saxony 59.7 per cent, Westphalia 33.9 per cent., Rhine provinces 30.8 per cent., Schleswig-Holstein and Liibeck 'Of 1,000 migrants of English birth enumerated in the United Kingdom in 1 881, 524 were found in border counties, 451 elsewhere in England and Wales, and 25 in Scotland and Ireland. Ravenstein, " The Laws of Migration," in Jou?-. of Stat. Soc, 1885, p. 182. ^Dr. Schumann selected at random six rural townships in Oldenburg, and -found that over four-fifths of the migrants moved no further than two (German) miles. The percentage of immigrants whose birth-place was within two miles of their town of residence was 95.6, 60.1, 83.5, 78.2, 88.1, 80.2. Similarly, of all the natives who had changed their place of residence without going outside the grand duchy, 83, 90, 86, 84, 68, 70 per cent, had not moved farther than two miles. — .-illg. Slat. Arckiv, i, 509. 256 THE GROWTH OF CITIES 15.4 per cent, and Baden 8.8 per cent. All other divisions of the Empire lost in the exchange of population.^ Now manufacturing industry in Germany is concentrated in the Rhine-Westphalian provinces, the kingdom of Saxony, the city of Berlin, and the acquired French territory of Alsace- Lorraine. It will be observed therefore that the districts which have gained through migration are distinctively the commercial and industrial centers of the Empire, with the single exception of Baden ; Brandenburg and Schleswig- Holstein being in close dependence on Berlin and Hamburg- Altona respectively. But one cause can explain internal migration in Germany, namely, the growth of the centres of commerce and industry, or in other words, of the great cities. The figures just given cover migration for an entire gen- eration ; in the more recent years migration has concen- trated itself upon still fewer districts. This may be ascer- tained by means of the births and deaths, as previously noticed. Thus, in the five-year period 1885-90, all of the provinces of Germany had an excess of births over deaths, but most of them lost a part of this excess through migra- tion, their total increase being less than their natural in- crease. The results of the movement are shown in Table CXXIV.« Excess of Increase in births. population. Gain or loss by migration. Ratio to exctss Total. of births. Group I ( East Prussia) " II (West Prussia and middle German states) .... Group III (South Ger. states) . " IV (Indust. centres) . . 851,770 611,538 500,787 937,688 212,666 531,089 347,520 1,480,191 —639,104 — 80,449 — 153,267 + 542,503 —75-04 -13-15 —30.61 + 57-86 German Empire 2,901,783 2,571,446 —330,317 —11.38 Group IV comprehends Berlin with a percentage of 239.52, the district of Potsdam which encloses Berlin 140.54 per cent., the Hanseatic towns (Hamburg, Bremen, and ^ Vierteljahrshefie zur Statistik des Deutschen Reiches, 1893, Heft ii, p. 4. ■'Cf. Schriften des Vereins fiir Socialpolitik, Ivi, 6. INTERNAL MIGRATION 257 Liibeck) 214.82 per cent, the kingdom of Saxony 33.28 per cent., the Rhine provinces 14.65 and Westphalia 20.75 P^'^ cent. Nothing could more clearly and emphatically show the relation between migration and the centres of commerce and industry than these figures from Germany. If it were needful, similar figures could be adduced for the other in dustrial countries.' Since, then, the current of migration is toward the cities and yet the bulk of migration is for short distances only, we can see the manner of the movement ; it is a migration by stages having for its object the satisfaction of the de- mands for more labor in the cities. These demands are not met by the direct migration of superfluous labor from the fields to the cities, but by the flocking in of the inhabitants immediately surrounding the town ; the gaps thus left in the rural populations are filled up by immigrants from more remote districts until the attractive force of a rapidly grow- ing city makes its influence felt, step by step, to the most remote corner of the country.^^ Thus, an analysis of London's provincial element shows that the emigration from the provinces decreases with the distance from London. In an essay on " Influx of Popula- tion," in Charles Booth's Life and Labor of the People (vol. iii, chaps. 2 and 3), Mr. H. Llewellyn Smith presents the fol- lowing comparison for six concentric rings : 3 No. persons per i,ooo Average distance of pop. of each ring, Density of pop. per Ring. from London in miles, living in London, 1881. 1,000 acres. I 23.8 166.0 8CX) 2 52.5 I2I.4 488 3 90.9 61.2 540 4 126.0 32.0 516 5 175-7 16,2 800 6 236.9 24.9 406 ' Ravenstein {Jour, of Stat. Soc, 1885, p. 185, ff) established similar conclu- sions for England. The " counties of dispersion " are entirely agricultural, while the " counties of absorption " embrace the centres of manufacturing and commerce. ^ Ravenstein, ii>t(^., 199. ^ The average distance of a ring of counties is taken to be the result of multi- 258 THE GROWTH OF CITIES This table clearly establishes the fact that London's pro- vincial population was contributed by the outside counties in proportion to their distance from London. It fully con- firms the opinion of Mr, Ravenstein that internal migration is of short-journey type. Distance is the controlling factor. The only exception to Ravenstein's rule is in the sixth ring, which includes the counties of Durham and Northumber- land, Cornwall, Pembroke and Cardigan, and four others. It is to be noted that these are maritime counties with direct communication to London. It represents the current of migration, consisting of those who seek a distinct economic advantage ; the bulk of migration, however, is a " drift" to- ward the great centres by successive stages. If emigrants do thus move by stages, settling at intermediate points for considerable periods in the interval, then we should expect to find the average ages of those coming from great distances to be greater than those whose birth-places are nearer the city. Mr. Smith has constructed the following table from the census returns of 1 881, the last year for which figures were available : Percentage of total migrants who were Rings (as before). Under 20 years of age. Over 20 years. I 22.4 77.6 2 18.I 81.9 3 16.8 83.2 4 15-4 84.6 5 I9-I 80.9 6 15.9 84.1 In general, the proportion of adults increases with the dis- tance from London until we reach the fifth ring, which in- cludes the manufacturing districts of the North — Yorkshire, Lancashire, etc. Still, too much weight should not be attributed to these figures, for it may be that the short-dis- plying the population of each county by the distance of its centre from London, adding the products and dividing by total population of the ring. INTERNAL MIGRA TION 259 tance migration includes more families with children (which would reduce the average age) while the long-distance mi- gration might be equally direct, although consisting more of single young men between the age of 20 and 30. Again, with short-distance migration as the prevalent form, we should expect to find the native county element stronger in the great city than in the rural districts around it, while the border county element would be less strong in the city, having been, so to speak, deposited in the intervening rural parts of the county by the current toward the city. Raven- stein declares that this is the general rule in Great Britain. Forty-five out of 67 cities which he investigated recruit their population in the main from their own county, or in case of border towns, from two contiguous counties ; they contained a smaller percentage of migrants from outside their own county than did the intervening rural parts. ^ It is true, moreover, that with the development of railway communication the volume of direct and long-distance migra- tion to the city has increased ; men out of work will often migrate hundreds of miles in order to find work in their own trade, instead of remaining at or near home and changing their occupation. This exception to the rule that internal migration is chiefly a short-distance movement brings us to the second law of migration, namely, that the distance travelled by migrants varies in the same ratio as the magnitude of the city which is their destination. The larger the town, the wider its circle of influence in attracting immigrants; the small city acts as a magnet for the neighboring counties, a large city (ioo,ooo-t-) attracts strangers from other states or provinces, but only the great capitals exercise an international influence on ' But in cases where a large city is situated in a relatively small county, or where the city is growing much more rapidly than the rest of the county, it must needs have a larger feeder, and hence draws immigrants from beyond the county. — Ravenstein, loc. cit., pp. 200-218. 26o THE GROWTH OF CITIES migration. This fact is clearly established in the following statistics from the Austrian census of 1890, which show the birth-place of every 1,000 persons enumerated in the class of towns specified in the first column ; Table CXXV. ^ Another Another dis- Another pro- Dwelling-places. Town of town of trict of same vince of the Foreign {Ortscka/ten.) residence, same district.' province. kingdom. country. Total. Under 500 657 215 100 22 6 1,000 500-2,000 735 149 85 23 8 1,000 2,000-5,000. 699 132 119 36 14 1,000 5,000-10,000.... 556 142 210 67 25 i,oco 10,000-20,000... 464 122 310 77 27 1,000 Over 20,000 ... 431 13 253 231 72 1,000 Austria 652 150 128 53 17 1,000 The table also illustrates a third law of migration, to which attention should be called, namely, that the percentage of immigrants increases in the same ratio as the magnitude of cities, but in inverse ratio with the magnitude of rural com- munities.3 Thus, in Austria, 27 per cent, of the population of villages of 500-2,000 are outsiders and 73 per cent, are natives of their town of residence ; but in the cities of 20,000+ the respective percentages are 57 and 43. It seems somewhat singular that the native town element, which is weakest in the large cities and steadily increases in strength as the size of the town diminishes, should fall ofif so much in the very smallest towns, namely, those containing fewer than 500 inhabitants. Yet this is the fact to the best ' Rauchberg, in S(ai. Monaisckriji, xix, 129. * The Bezirk or district, of which there are 359 in Austria, contains 323 square miles, and is thus less than half the size of a New York county (average 750 square miles). •'' Georg von Mayr, who as chief of the Bavarian Bureau of Statistics, made the first thorough investigation in the field of internal migration in 1 871, is the author of this law. Cf. Heft XXXII dtr Beiirdge zur Statistik des Koiiigreichs Bayerns (Miinchen, 1876) : Die bayrische Bevolkerung nacJi dtr Gehiirtigkeit. INTERNAL MIGRATION 26 1 of the writer's knowledge in all the investigations thus far made — the Bavarian census of 1871, the 1880 census of the duchy of Oldenburg, 'and the Austrian census of 1890. One explanation of the fact may be that men living in the least populous places mustoftener seek their wives in a neigh- boring community, the range of choice in their own village being so small. The Austrian statistics, however, show that this happens more frequently in the villages of from 500 to 1 ,000 inhabitants, where the native town element is strongest of all;^ thus the number of women to 1,000 men was Among those born In another town of Places. In Austria. In town of residence. same district. Under 500 pop 1.047 9^5 i>297 500-1,000 " 1,049 1,001 1,373 From which it follows that the large percentage of im- migrants in the smallest places is not a result of marriages contracted by male residents with women of a neighboring place. The more probable explanation is that the least populous communities are unable, with their own members, to carry out the division of labor to a sufficient extent, and are there- fore obliged to recruit their force from neighboring places. But the conditions in Austria and Germany are peculiar to a fast disappearing civilization, in that the places with less ' The percentages of the native town element were as follows in the rural communes : Under 500 45-02 500-1,000 62.65 1,000-1,500 58.30 1,500-2,000 59-57 2,000-3,000 71-31 3,000-4,000 77-i8 4,000-5,000 79-42 5,coo-f- • 70.60 — Statistische Nachrichten ilber das Grossherzogtum Oldenburg^ Heft xix, p. 61. * Rauchberg, as above, xix, 130. 262 THE GROWTH OF CITIES than 500 population are usually manors. It is unfortunate that no data of this sort exist for countries without the manorial system. Von Mayr's law as stated above has never been success- fully disputed ^ and is confirmed by nearly all the statistics. Thus in Saxony the percentage constituted by the native element is as follows : Rural townships: Under 2,000 52.4 Over 2,000 49.7 Urban townships 48.0 Dresden 39.4 Leipzig 36.1 For other countries the facts may be summarized thus : Table CXXVI. Percentage of population born in town of residence: Rural. Urban. Capital city. 1. Prussia, 1890 53.9 43.8 2. Sweden, 1880 92.0 67.8 41.3 3. Denmark, 1890 86.77 75'i6 55-92 4. Netherlands, 1889 68.34 64.54 S.Belgium, 1890 70.4 59.4 57.0 6. Switzerland, 1888 60.7 34.7 24.3 7. France, 1891 ca 70.0 .... 32.4 ' The rule that the native town element decreases in cities in inverse ratio to their population has been denied by Hansen in his noted work. Die drei Bevolk- erungstufen, in which he prints a list of the 34 Bavarian cities (those with municipal institutions) arranged in order of size, and then points out irregularities. But it has never been claimed for the law that it applies to individual cities, but only to classes, and if the 34 cities mentioned be further divided into classes, the rule will be found to hold in most cases : Per cent of native city element in total population. 4 cities over 30,000 pop 43 3 " 20-30,000 " 53 II " 10-20,000 " 46 16 " under 10,000 " 54 Where the number of cases is so small, the liability to fluctuations is always present. lATERNAL MIGRATION 26$ EXPLANATIOMS. 1. The rural percentage is for all Prussia; the urban for cities of 20,000 + . 2. The unit as regards birth-place is here not the town, but the " Lan " or province. 3. The unit is the district, or province. 4. The urban percentage is for cities of 20,000+ ; the rural for the remainder oithe country. 5. Urban= communes of 5,000 + . The percentage in the last column refers to cities of 100,000 4- . 6. Urban — communes of 10,000 + . 7. In France we have the percentages of inhabitants who were bom in the commune where they were enumerated. Takiug the four departments that represent the largest and smallest per- centages of urban population, we have the following figures for 1891: Seine (99.13 per cent, urban) 32.4 Bouches de Rhone (83.5 per cent, urban) 54.7 Savoie Haute (91.7 per cent, rural^ 73. Cotes du Nord (90.0 per cent, rural) 69. These are the two extremes. There is not such a vast difference between the departements of Bouches de Rhone (containing Marseilles, the second largest city of France) and Cotes du Nord, (a rural di-partemenf) as regards the proportion of immigrants. England's statistics do not admit of ready comparison be- tween the urban and rural population in respect of immigra- tion. Certain typical counties may be selected, however; thus, among the following counties the percentage of popu- lation born in the same county was : Urban counties. Rural counties. London 65.5 Rutland 63.8 Middlesex 32.9 Cardigan 87.6 Lancashire 75.9 Suffolk 82,3 Hereford 70.0 It is certainly a striking fact that Lancashire, the typical manufacturing county of England, contains fewer residents of outside birth than do the rural counties of Hereford and Rutland. Even London has fewer immigrants than Rutland- shire. Middlesex, to be sure, has a large proportion of out- siders, but Middlesex should really be counted as part of one large metropolitan county including all London. Too much weight, however, should not be placed on these statis- tics, which are fragmentary at best and are in opposition to all others that we have on the subject. The three laws of migration now laid down hold good in the United States, so far as can be judged from imperfect statistics ; but the role of the foreigners is here a more im- portant one : 264 THE GROWTH OF CITIES Table CXXVII. Of each 1,000 INHABITANTS THERE WERE BORN IN The same province Another province, A foreign City named. or state. state or county. country. London, (1881) .... 629 732' 204' 63 Paris,'^ (1891) ... 324 358 •-' 541 61 Berlin, (1890) ... 407 589 394 17 Vienna, (1890) •••• 349 470 420 no Glasgow, (1881) ••• 513 614' 214 172 Boston, (1885) ... 385 492 167 341 « (1890) 505 142 353 Amsterdam (I 889) ... 683 769 203 28 New York, (1890) . 489 89 422 Chicago, (1890) 397 193 410 Among the seven cities in the foregoing table there ap- pears a wide variation in the percentage of the native town element, which constitutes from one-third to two-thirds of the population. Very nearly as wide a range, however, may be found among the cities of a single country ; among the 26 great cities of Germany, for example, it varies from 36 (Munich) to 62.4 per. cent, (Aachen), the average being 43.7 per cent. This percentage depends partly upon the individual city's industrial character ; partly upon its area, and in a larger degree, perhaps, on the composition of the population of the surrounding country. Thus the number of foreigners in the general population will afifect the city's composition, as it notably does in the United States. In Boston, the only large American city for which the neces- sary data were fully obtainable, the percentage of foreign born was 34.1, which is close to the average (31.8) of the 28 American cities of 100,000 and upwards. None of the European cities, of course, approaches this figure. ' For " province " read " county." " Foreign country " includes also Scotland and Ireland. 2 Department of the Seine, nearly identical with Paris; the first column gives the percentage (32.4) of inhabitants of the department who were born in the commune of residence. INTERNAL MIGRATION 265 Even Glasgow's 17.2 per cent, consists almost entirely of English and Irish born (3.1 and 13.1 per cent, respectively). Vienna has the next largest percentage of foreigners, but these are in large part (77 of the no in a thousand) natives of the confederated country, Hungary. But while foreign immigration contributes large numbers to the population of American cities, it is not true, as Mr. Ravenstein, for example, seems inclined to think, that " the migratory current from the country to the city is scarcely perceptible in the United States and other newly settled countries."^ We have seen in a former chapter how villages have decayed and cities prospered even in the West, and the strong tendency of young men to abandon the farm in order to seek their fortunes in the city is a matter of familiar observation. Th« only statistics that we have on this point, however, are those of Massachusetts. They show that 1 1 per cent, of Boston's population in 1885 had been born in other Massachusetts towns. Now it is possible, but hardly probable, that all these immigrants came from cities and none from the rural towns ; but one thing is clear at least : Boston's growth is almost as much due to immigration of native Americans as to her own natural increase or to foreign immigration. Of Boston's total population in 1885, 38.47 per cent, were born in the city itself and 27.39 per cent, in the United States outside of Boston.^ Now London con- tains only 30.7 per cent, of Englishmen born outside of London, Glasgow 31.5 per cent, of Scotchmen born outside of Glasgow, and Amsterdam 28.9 per cent, of Dutchmen born outside of the city itself. The French and German cities, it is true, have larger percentages, but in view of the comparison between Boston on the one side and London, 1 y. of St. Soc, 1889, p. 288. ■^The remainder, born abroad, is 34.14 per cent. Mass. Cens. of 188$, vol. i, pt. i, p. Ixviii. 266 I^HE GROWTH OF CITIES Glasgow and Amsterdam on the other, it can hardly be said that "the migratory current of the rural population toward the cities is hardly perceptible in the United States." ' Except for the larger immigration from abroad, the mi- gratory movement in the United States follows about the same direction that it does in Europe. In Massachusetts, for example, we have the following distribution of the popu- lation according to birth-place : Per cent, of total pop. Born in 1885.^ Town where enumerated 36.1 Some other town in Mass 20.8 Mass. — not specified 6 Massachusetts 57.5 Other New England States i x.2 " North Atlantic States 2.7 South Atlantic States 55 North Central States 55 South " " I Western States and Territorities i United States — not specified 2 United States 72.9 Foreign countries 27.1 100. ' The following table of comparisons may be found useful for reference ; Table CXXVIII. Percentage of population born in City of Territory immedi- Native coun- Foreign Date, residence, ately surrounding, try elsewhere, country. 1. German cities (26) of 100,000+ 1890 43.7 30.7 23.5 2.1 2. Austrian cities 132) of 20,000 -^ . 1890 43.1 26.6 23.1 7.2 3. Scotch cities (7') 1881 52.4 24.1 10.3 13.2 4. American cities (28) of ' v ' 100,000+ 1890 52.4 15.8 31.8 5. Boston i88s 38.5 12.9 14.5 34.1 Authorities. — The official German and Austrian censuses. The percentages for American cities are based on nth Cen., Pop., i, p. cxxvi. For Scotland, Ravenstein, op. cit., p. 195. The " immediately surrounding territory " comprises the Gebietstheilen (provinces, etc.) in Ger- many; the Land, or province in Austria; the native county and border counties in Scotland; State or commonwealth in America. England and Wales are reckoned as " foreign countries " to Scotland. '^ Based on the data given in Mass. Census of 188 j, vol. i, part i, p. Ixx. And if strength of migration be portrayed on county maps of any commonwealth by INTERNAL MIGRATION 267 It is unnecessary to go further in testing the hypotheses of von Mayr and Ravenstein. Detailed studies in several countries have been made, among which those of Dr. Rauch- berg, of the Austrian statistical bureau, are pre-eminent; these are illustrated with maps which convey to the eye the meaning of internal migration under the attraction of centres of population. They confirm the conclusions stated in this paper.^ They would also seem to meet the assertion some- times made regarding the mode of internal migration, namely, that it proceeds staffelweise, from farm to village, from vil- lage to town, from town to city, from city to metropolis. If, for example, the migration into a great city proceeds in the main from the immediately surrounding territory, there can be scant opportunity for smaller cities to act as feeders. Berlin, in 1890, cotained 936,143 persons born outside the city, or 59.3 per cent, of its total population. Brandenburg, the province in whose centre Berlin is situated, contributed 287,540 of the immigrants, and the other 25 great cities (100,000+) of Germany contributed 53,856, or one-fifth as many. Nevertheless, the population of Brandenburg in 1890 was half a million less than that of the 25 great cities (4,600,000 and 4,120,577 respectively). Statistics prove that great cities receive a small part of their immigration from other large cities, although a relatively large percentage of emigrants from these cities go to other cities to live. In the means of color or shading, it will usually be found that those counties which con- tain the largest percentage of natives of adjoining commonwealths are the border counties. Cf. maps showing interstate migration, in Stat. Atlas of (J. S. * " Die Gebiirtigkeitsverhaltnisse der Bevolkerung Oesterreichs nach den Ergeb- nissen der Volkszahlung von 31 December, 1890," in Stalls. Monat., xviii (1892), 517-574, especially p. 556. Also " Der Zug nach der Stadt," ibid., xix, 125-171. The student who is interested in the subject will find excellent studies of the in- ternal migration in Germany by Briickner (" Die Entwicklung der grossstadt- ischen Bevolkerung Deutschlands ") and Schumann (" Die inneren Wander- ungen ") in the first volume of the /illgemeines iitatisches Archiv. 268 THE GROWTH OF CITIES following table of the German cities, (a) denotes the per- centage of persons born in other great cities among the immigrants of the specified city, (b) the percentage of those born in a great city and emigrating who go to another of the 26 cities : a. b. a. b. Berlin 5.8 13.3 Nuremberg 2.4 22.9 Hamburg 10.9 36.7 Stuttgart 2.4 13,3 Leipzig 6.5 25.9 Chemnitz 3.9 27.1 Munich 2.4 9.5 Elberfeld 14.6 41.3 Breslau 2.2 27.8 Bremen 4.9 22.8 Cologne 6.9 25.5 Strassburg 3.7 9.2 Dresden 6.1 20.0 Danzig 3.9 34.1 Magdeburg 5.0 34.7 Barmen 11.8 44.0 Frankfort 3.9 21.3 Stettin 3.9 41.7 Hanover 5.6 27.9 Crefeld 5.5 27.0 Konigsberg 2.6 35.6 Aachen 5.6 28.1 Diisseldorf 9.3 31.9 Halle 6.2 30.5 Altona 14.6 67.5 Brunswick 5.0 32.7 Average (a) 5.9; (b) 26.6 The cities arej arranged in order of size, and it will be observed at once|that the percentages do not descend in the same order. Berlin, Munich, Breslau are all below the average. The cities that receive the largest proportion of their immigrants from other cities, are Hamburg, Altona, Elberfeld, Barmen ; and a glance at the map shows the reason why. Hamburg and Altona are almost parts of one city and Elberfeld-Barmen also constitute a pair of " twin cities." Hence migration between them must be large. On the other hand, the cities with the smallest percentages, Munich, Breslau, Frankfort, Konigsberg, Nuremberg, Stutt- gart, Strassburg — are all isolated cities ; size has therefore little if any influencej^on the origin of immigration. Loca- tion is the prime factor, because internal migration is for short distances. Nor if one looks at?migration from the opposite point of INTERNAL MIGRATION 269 view, does it appear that the largest city communicates primarily with other cities, and only secondarily with the rural districts or villages. While 67.5 out of every lOQ persons born in Altona and no longer resident there, are in the other great cities of Germany, only 13.3 per cent, of emigrating Berliners are residents of other great cities. Practically the same rule holds as before ; namely, it is distance rather than size of cities which determines the amount of immigration or emigration in connection with other places. Migration being predominantly short-dis- tance, the character of immigration to any city will partake of the local surroundings. And as in most cases the sur- rounding country is filled with hamlets and villages, the im- migrants will be of the rural class ; but as the distance widens the urban communities will contribute more and more. This appears in the military rolls of Frankfort^ 1890-2, containing the names of 2,293 young men not born in the city but living there and becoming subject to military duty in the years mentioned. The province or state of birth was given as follows:^ Rural communes. Towns. Prussia : Hessen-Nassau 273 160 Rhine province .... 50 69 Rest of Prussia 134 188 Hessen 238 127 Bavaria 259 132 Baden 126 84 Wurtemburg 134 75 Saxony 30 48 Other German states 60 86 Total 1,304 969 All Prussia, except the province of Hessen-Nassau wherein Frankfort is situated, sends a majority of this par- ' Bleicher, p. 37; the distinction between town and country is the legal one. 270 THE GROWTH OF CITIES ticular class of emigrants from the cities. From the four states of South Germany (being relatively near to Frank- fort) the migration was primarily rural, but from Saxony and the rest of Germany it was predominantly urban. The classification of the 969 city men follows : Towns under 5,000 inhabitants 369 " from 5,000^20,000 inhabitants 227 " " 20,000-100,000 inhabitants 233 " 100,000+ 140 969 It cannot be said, however, that statistics of birth-place entirely disprove the hypothesis of migration by stages through village, town, city and metropolis, inasmuch as a man's previous place of residence does not always coincide with his birth-place. The migrant to London, who is credited to the rural county in which he was born, may have passed his earlier years in a neighboring city. The German municipal statisticians are now devoting some attention to the ascertainment of the immigrant's last place of residence as well as his birth-place, and these statistics, imperfect as yet, show that the cities figure more largely as feeders to other great cities than appeared in the statistics of birth- place. But as the percentages afifected are those born in distant parts (who reach the city of their destination after residence in intermediate cities), the conclusions already stated need not be greatly modified. Dr. Bleicher, of the Frankfort municipal statistical bureau, has published a study of the migration for the year 1891. After eliminating trav- ellers, visitors, and other persons whose sojourn was tempo- rary, he had remaining 23,254 male, and 16,166 female immi- grants, 11,440, or 70.8 per cent, of the latter being domestic servants. Of the men, 8,656, or 37.2 per cent., were born in the town from which they moved to Frankfort ; the other two- INTERNAL MIGRATION 271 thirds had already left their place of birth. Of the women, excluding the servants, 43.9 per cent, came directly from the birth-place; or 58 per cent, if the servants be included. From these figures it is to be inferred that men are accus- tomed to migrate more frequently than women, although a larger number of women than men, as previously noted, par- ticipate in the migratory movement across township lines. A detailed analysis of the male immigrants to Frankfort in 1 89 1 is presented in Table CXXIX.i Male Immigrants from Women. Rural Neigh- German Other Foreign From Born in county of boring cities of places in coun- All all Frankfort, cities. ioo,ooo-i-. Germany, tries. places, places. Last place of residence. ... 9 City of Frankfort 6 Rural county of Frankfort. 2 A neighboring city 6 A German city of 100,000 + . 2 Some other German town. 72 A foreign country i 1 14.7 2 4-3 4 0-7 4 4-0 4-3 67-3 4-7 German Other Foreign cities of places in coun- 100,000 -•-. Germany. tries. 18.5 54-4 33-9 4.8 2.7 8.9 0.5 0.4 0.4 2.3 1-9 2.8 7-4 1.9 4.9 S7-0 37-2 20.0 9-5 1.5 29.0 37-2 43-9 4-1 S-7 0.6 0.7 2.6 3.6 3 4 3-S 470 34-9 S.o 7.7 99.9 lOO.O It appears from this table that the direct migration from birth-place to the great city is dependent upon distance. Of the immigrants from German cities of 100,000+ over eighteen per cent, were natives of the city from which they removed to Frankfort, showing that there is considerable mobility in great city populations ; seven per cent, of these immgrants were natives of another great city, and 57 per cent, natives of a smaller city or village of Germany. It will be noted that the immigrants include a large number of born Frank- forters returning to their native city; that, in fact, Frankfort contributed a larger percentage to the immigration than any other city. The role played by people thus returning to their native town or former place of residence, is a considerable one. Dr. ^ Bleicher, Heft II (1893), p. 46. 2/2 THE GROWTH OF CITIES Bleicher found that of 4,541 emigrants from Frankfort in 1891, 2,082, or 45.85 per cent, returned to the place from which they had come to Frankfort.^ Nor was this return current directed entirely to the small towns in the vicinity of Frankfort, as these figures will demonstrate:^ Emigrants to former place Original residence. Immigrants. of residence. Places within 12 kilometers 144 43=29.9 per cent. Cities of neighborhood 802 259=32.3 Great cities 1,002 411=^41.0 Other places in Germany 2,183 i»i 83=54.2 Foreign countries 410 i86=-45.4 Total 4,541 2,082=45.85 The emigration to the great cities and to foreign countries consisted of those engaged in business and trade, while that to smaller places in Germany was made up chiefly of day laborers. As between rural and urban communes in the vicinity of a large city, the latter will contribute more than proportion- ately to its immigration. Thus in the little German city of Oldenburg (pop. 20,575) ^^ the enterprising duchy of the same name, it was found in 1880 that among 8,541 immi- grants, 1,975 o'f 23 per cent, had been born in urban com- munities (places of 2,000+) and "jj per cent, in rural communities. But outside the capital the population is almost entirely rural, and it results that while the rural com- munes contributed 3.1 per cent, of their population to the capital, the towns contributed 8.3 per cent, of theirs.^ There are other instances. Switzerland has not a very large urban population, and yet the city Basel in 1888 contained a body of persons, amounting to 14.5 per cent, of its population, who were born in other cities, 47.3 having been born in ' Ibid., 53. = Ibid., 55. ' Statistische Nachrickten Oldenburgs, etc., xix, 212. INTERNAL MIGRA TION 373 rural districts and 38.2 in Basel itself.' In Bavaria again the free {unmittelbar) cities, excluding Munich, contained 12.6 per cent, of the population in 1890; but they con- tributed 17.83 per cent, of Munich's Bavarian immigrants. In Saxony, where the urban population, excluding Leipzig, falls somewhat below the rural population, it nevertheless contributed to Leipzig's population of 1885, 32.40 per cent, while the rural communes contributed 31.67 and Leipzig itself 35.57. The mobility of the great city populations may, however, be inferior to that of rural communities. Professor Karl Biicher, having in mind the factory operatives and sweat- shop victims of slum populations, affirms that poverty, as well as persistent clinging to familiar associations, prevents city populations from migrating as freely as country people. In support of his opinion he offers the following : ^ Table CXXX. Percentage of (a) new citizens coming from Towns. Villages and hamlets. Cologne, 1356-1479 37.4 62.6 Frankfort, 131 1-1400 28.2 71,8 " 1401-1500 43.9 56.1 (b) Journeymen book-binders: Frankfort, 1712-50 97.5 2.5 1751-1800 94.3 5.7 " 1801-35 89.2 10.8 " 1835-50 86.0 14.0 1851-67 81.2 18.8 Such figures are inconclusive. It may well be that the increasing proportion of rural bookbinders in Frankfort is ' Biicher, Die Bevolkerung des Cantons Basel- Stadt am i Dez., 1888. ' Etitstehung der Volkswt., 62 ff. Llewellyn Smith shows that the twelve lead- ing manufacturing counties of England had contributed only 2.4 per l,GOO of their population to the population of East London and Hackney in 1881, while the twelve leading agricultural counties had contributed 16 per mille. (Booth, op. cxt,, iii, 71. 2/4 THE GROWTH OF CITIES due to the extension of the printing trade to small towns. Formerly, this trade like all others not immediately con- nected with agriculture, was the virtual monopoly of the cities. Nevertheless, there is other evidence that the population of a metropolitan city is less migratory than the average. In 1 88 1 of all the persons born in London and still living in England and Wales, 80.4 per cent, were residents of London, while in the general population of England and Wales only 75.23 per cent, were residents of the county where born.^ With regard to Vienna, 84.7 per cent, of the native Viennese counted in Austria in 1890 were resident in Vienna; in the entire population only 66.3 per cent, were residing in the town where born : ^ that is, 153 out of 1000 born Viennese remove to other parts of Austria, and 337 of 1000 people on the average have left their town of birth. Emigration from the great city follows the general laws already formulated. It is overwhelmingly short-distance migration, since so much of it is directed into the suburbs. 3 In the distribution of the natives of London in other parts of England and Wales in 1881, the contiguous counties (Middlesex, Surrey, Kent, Essex, Hertford) forming an extra-metropolitan group received 53.73 per cent, of the migrants. Wales received only 1.36 per cent., but the Northern groups (manufacturing counties) received a slightly larger percentage than the midland group,'* although situated at a greater distance. A better idea of London emigration will perhaps be gained from the following table, iRavenstein, in Jour, of St. Soc, xlviii (1885), 195, 171. *Rauchberg, St. Mctt., xix., 152; xviii, 534. * Economically this should not be regarded as migration since it does not usually involve a change in the place of business. *Raven8tein, jf. of St. Soc, xlviii, 206-7. INTERNAL MIGRATION 2/5 in which the counties of England and Wales are arranged in concentric rings about London:^ Natives of London living in each ring of counties, per i,ooo of Average distance from the population of the Ring. London in miles. ringin 1881. I 23.8 142.3 2 52.5 42.5 3 90.9 17-7 4 1 26.0 9.8 5 175-7 8.5 6 236.9 6.5 Comparing this table with the one showing the influx into London from the same rings, it will be found that the first ring is more conspicuous here than in the inflowing move- ment, while the outside rings participate more largely in the inflow. This is what would naturally be expected ; the great city's attraction extends to the remotest boundary, and the in- flowing movement creates a counter current of emigration, which is, however, much less intense. On the other hand, the suburbs receive more than they give. Remarking that the reg- istration county of London includes parts of the historical counties of Middlesex, Surrey and Kent, we may note that in 1 88 1 there were jGyyyi natives of extra-metropolitan Middle- sex in metropolitan Middlesex, and 80,271 natives of the city part of the county in the extra-city part, — again of 3,500 for the extra-metropolitan portion. These large figures repre- sent migration in no real sense ; they merely express part of the movement across an imaginary boundary line between registration London and Greater London. But whither go the real emigrants and why do they leave London ? Mr. Smith has followed up the native Londoners outside of London and the contiguous counties of Essex, Middlesex, Sussex and Surrey, and found 64,918 residing in ^ H. L. Smith, in Booth, op. cit., \\\, 67, 126. 2/6 THE GROWTH OF CITIES cities of 100,000 or more, this being equal to 17.4 per 1000 of their population, and representing their power of attrac- tion upon London ; 24,079 in cities of 50-100,000, or 14 per mille of their population ; and 196,415 in all places under 50,000, or 1.29 per mille of their population.' Outside the suburban movement, therefore, it is the great cities that draw most heavily upon London. The movement toward the smaller places is partly non-economic, /. e., non-workers seek- ing retirement in a small town amid inexpensive surround- ings ; the larger part of this outflow however is not repre- sented in the figures given, for it consists of those returning to the county of birth. The outflow to the larger cities is without doubt on business lines and consists of artisans and business men of various kinds.* The nature of the migratory movement having been con- sidered at some length, it is now time to investigate the character of the migrants, and especially of the migrants to the cities. We have seen something of their character in studying the sources of the migration, but we may gain further light by considering the sex, age and social rank of the im- migrants. As to sex, it may be confidently declared that woman is a greater migrant than man, — only she travels shorter distances. In considering the sex of city populations we shall see that the excess of women among the immigrants is one of the causes of the general surplus of women in cities, which ex- ceeds that of the rural districts. ^ Smith, in Booth, op. cit., iii, 126-7. '^ In Germany, also, the larger portion of the emigration from the great cities is either to the surrounding province or to other great cities; in 1890, of the native* of the 26 great cities who vi^ere living in Germany outside the city of birth, 56.1 per cent, resided in the surrounding province, and 26.5 per cent, in the other great cities. (Calculated from statistics given in Statistik cles Deuiscken Retches, N. F., vol. 68, p. 71* ff.) Cf. Table CXXIX, supra, and Bleicher, ii, 52. INTERNAL MIGRATION 277 Following was the number of women (all the females) to 100 men (all the males) in 1881,' among natives of England and Wales. Scotland. Ireland. Residing in the county of birth 104 108 104 " in another county of same country ... . 112 114 116 " in another part of the United Kingdom. 81 91 92 The most marked predominance of the female sex is not among the " stay-at-homes," /. e., among those residing in the county of birth, but among those who have moved to some other county. On the other hand, however, the women predominate only among the short-distance migrants, for when it comes to emigrants to another of the three kingdoms, their proportion falls immensely. Other countries present similar statistics.'' So do cities. Thus of all the persons born in London and living anywhere in the United Kingdom in 1 88 1, there were 112 females to 100 males; but among those still residing in London, only 109.3 1 J. of St. Soc, 1885, p. 197. *Some German statistics of 1S90 present the facts clearly, the following figurcB being the number of females to i,ooo males: Born. Prussia.^ Austria. ° In township of residence l,oi8 i,oii Elsewhere in same county * i>254 1,31 1 " " the same province 1,064 1.023 " " " " country 820 889 In foreign countries 874 950 Total population 1,038 i>044 ' Preussische Statistik, Heft cxxi. ' Rauchberg, Stat. Mon., xix, 130. ^ Kreis in Prussia, Bezirk in Austria. Taking the average of towns in these countries, it appears that in the native town element the women are not so strong as they are in the general population; but the female newcomers are chiefly from the same county, their proportion among other immigrants falling below the general average. The farther the dis- tance, the smaller the proportion of women. This rule apparently breaks down in the case of foreign countries; but as a matter of fact, international migra- tion like that between Austria and Hungary is really short distance. Germany with its long boundaries is in about the same condition. *Ravenstein, J. of St. Soc, 48: 195. 278 THE GROWTH OF CITIES The statistics show that while women move from town to town within the same county or province more readily than the men do, they are more loath to move to a distance. The reason is probably to be found in the marriages which take women into a neighboring town, as well as the demand for domestic servants,' while men go longer distances in search of the best labor market. The relative proportion in which the two sexes participate in the migratory movement appears with considerable distinctness in the following Aus- trian statistics : ' Table CXXXI. Females to i,ooo males among those born Elsewhere Elsewhere Another Foreign In town of in same in same province of coun- Ratio in the Places with a pop. of residence. Bezirk. province. Austria. tries. whole pop. Less than 500 985 i.^gy 1,038 861 920 1,047 500-2,000 1,001 1,373 1,047 882 945 1,049 2,000-5,000 1,027 1,27s 1,012 854 995 1,347 5,000-10,000 1,079 ^>^T^ 949 729 84s 1,029 10,000-20,000 1,067 1,200 955 655 914 1,004 ao,ooo+ 1,100 1,099 1,039 95^ 968 1,039 Austria 1,011 1,311 1,023 889 950 1,044 Vienna 1,095 •••• ••-• ■■■■ ••■• 1,061 Graz 1,1:36 •••• •••• ■ — •••• 1,086 It will be observed that the number of women to looomen is the largest in the smaller places and steadily decreases until the class of cities having 20,000 inhabitants is reached, when it once more increases. In the very largest cities it exceeds any of the ratios, being 1,061 in Vienna, 1,086 in ^ Thus, the distribution of the female immigrants in Frankfort (Cf. Bleicher, p. 15), as compared with the native Frankfort women was as follows: Bom Foreign bom to In Frankfort. Outside. loo natives. Children under 15 18,876 4.237 22.4 Dependents (members of family, etc.) . 13,143 28,620 217.8 Domestic servants 383 16,242 4,240,7 Other occupations 3,881 9,21 1 237.3 Total 36,283 58,310 160.7 * Rauchberg, Stat. Mon., xix, 130. INTERNAL MIGRA TION 279 Graz and Triest,' etc, ; here is where the influence of large numbers of domestic servants makes itself felt. Considering the '* stay-at-homes " alone, it will be noticed that women predominate the more conspicuously, the larger the town ; that is, the males migrate in relatively greater numbers from large cities than from small towns. But among the migrants the predominance of the females is especially marked, not in the cities, but in the smaller places. Both facts are explained by the fact already emphasized that the large places alone draw migrants from a distance ; hence they attract a majority of men, since, as we just noted, women are short-distance migrants. Finally, the table shows that in the larger cities the female element predominates less strongly among the immigrants than among the natives, the contrary being the case in the smaller places. Thus immigration strengthens the native female element in small places, but weakens it in towns of 5,000 and upwards. The distribution of the sexes in the United States has been thoroughly treated by Professor Willcox ^ who calls attention to two tendencies : while, in all the elements of the pop- ulation, the females show a tendency toward concentration in the cities, the tendency is more marked among the negroes of the South and the immigrants of the North than among the native whites. In New York State, 50.46 per cent of the foreign born living in 118 cities and towns are females, and only 45 per cent in the rural districts, — a difference of 5'46- But among the native Americans the percentages are 51-32 and 49.74 respectively, or a difference of only 1.58. In Georgia 54.84 per cent of the negroes in the cities are females and 49.76 per cent in the rural districts, — a difference of 5.08. But among the whites the percentages are 50.33 and 49.88, or a difference of only .45. Professor Willcox does not think ' Rauchberg, ibid., xix, 153. '^ Am. yottr. of Sociology, i, 725. 28o '^HE GROWTH OF CITIES that this tendency toward dissociation and concentration among the female negroes and foreigners can be wholly ac- counted for by migration, but is in part due to the higher infant mortality among these classes in the cities/ Age. — It is a matter of common observation that the migrants to the cities are chiefly young people ; and in this case the results of common observation are confirmed by statistics. A good idea of the age classification of migrants to the cities is given in the following table in which the native and immigrant elements of the city of Frankfort are compared with the average of Germany'' (1890 in each case) : Table CXXXII. Frankfort. Germany. Born in Frankfort. Born elsewhere. Age. Male. Female. Male. Female. 0-5 227 204 17 14 13O.I 6-10 182 163 26 23 1 1 1.9 II-15 169 154 41 36 109.5 16-20 Ill 112 no 117 93.2 21-30 105 130 299 316 161.9 31-40 67 75 218 209 127.6 41-50 57 62 158 143 103.8 51-60 41 48 82 81 78.3 61-70 27 33 33 40 52-0 71-80 12 15 12 17 23.6 81+ 2 4 1.6 3.2 4.2 Unknown .... 2.4 .8 1,000 1,000 1,000 1,000 1,000 Only 7 or 8 per cent of the migrants to a typical city are under the age of 15, while in a normal population the pro- portion of children of that age is about 33 per cent. And while normally about one-half of any population will be at the age of 16-50, the age of activity, the proportion among the immigrants to a city is fully three-fourths. Over one- ' Cf. infra, ch. v, Sec. I. "Bleicher, p. 6; Stat, des Deutschen Reiches, N. F. xliv, 24* ff. INTERNAL MIGRA TION 2 8 1 half the immigrants are of the age 20—/J.0, — in many cases 20— JO. Above the age of fifty, the percentages do not show any considerable variation. ^ The effect of the migration of persons in the active period of life to the cities, is to wrest away from the city-born the real work of the city. When it is said that 58 per cent of Berlin's population was born outside of Berlin, one is hardly prepared for the information that about 80 per cent of its male workers of the age of 30-60 are outsiders. But the following table shows the number of immigrants per 1000 of each age group in Berlin in 1885 :^ Male. Female. 0-15 166 163 Over 15-20 495 512 " 20-25 765 697 " 25-30 ..771 763 " 30-35 791 779 " 35-40 816 789 " 40-45 811 785 " 45-60 794 764 " 60 767 764 All ages 580 573 Of children under 15, the immigrants contribute only 17 ' The foregoing statistics do not give the age at the time of Jtiigraiion. Such statistics would show that at least four-fifths of the migrants are young men and women. For example, of 295 migrants from English villages to London and other cities, 16 were under the age of 15; 235 (80 per cent.) were between 15 and 25, 27 were between 25 and 30, and only 17 were over 30. (Booth, Life and Labor of the People^ iii, 1 39.) Bruckner, in his careful study of the German cities {Allg. Stat. Ar., i, 650), has shown that the age-grouping depends chiefly upon the volume of immigration, and its duration. The greater the immigration, the larger the percentage of the higher age classes, and the smaller the percentage of children. The only excep- tion occurs in the case of cities that have a strong current of emigration to the suburban districts, which embraces the young married people, and thus reduces the percentage in the higher age groups. Leipzig was, until the incorporation of the suburbs, a typical city of this kind. * Allg. Stat. Archiv, i, 634. 282 THE GROWTH OF CITIES per cent in Berlin, but of adults at the age of 30-60, the im- migrants form about 80 per cent of the entire population at those ages. Such a vast number of outsiders in the import- ant ages of life must necessarily have a deep influence on city life. ^ The length of the residence of immigrants to the city is alsa of interest. The Berlin statistics may be considered fairly typical, although there is always variation between different times and places : ^ 1885. Male. Female. Total. Berlin born 420 427 424 Residence o- 1 year 94 67 80 " over 1-2 years... 49 39 44 " 2-5 " • • • 94 92 93 " " 5-10 "... 88 104 96 " " 10-15 " ••• 96 ^oo 99 "15 "... 159 170 165 1000 1000 1000 The immigrants who had been in Berlin less than 5 years in 1885 constituted one-fifth of the entire population, over- balancing those who had resided in the city more than 1 5 years. The proportion that had lived in the city over 5 years and less than 16 was also about one-fifth. In these data, the age of persons has not been considered ; but it has an important influence. Thus, most of the child- ren are to be found in the class of Berlin born, the recent emigrants are mainly young adults, and the immigrants of 10 1 In this respect Berlin is not an exceptional city. In the same year (1885) the percentage of immigrants in the adult population (20 years -f-) was as follows: Male. Female. I.Hamburg 68.3 61,2 2. Berlin 78.7 76.0 3. Frankfort - 79.4 76.6 4. Breslau 80.1 76.0 5. Leipzig 84.6 78.4 '■ Briickner, ibid., i, 632. INTERNAL MIGRATION 283 years' residence, middle-aged and elderly people. But the following table shows the Length of residence in Berlin of 1,000 adult immigrants (over 25 years) in 1885:' Male. Female. o-i year 71 51 Over 1-5 years 130 116 " 5-10 " 123 135 " 10-15 " 164 163 " 15 " 302 308 Berlin-born 208 227 998 1000 It is somewhat reassuring to find that fifty per cent of the city's adult population consists of native Berliners at least 25 years old and immigrants who have resided in the city at least fifteen years. But the other half of the adult population are such comparative strangers, that one can understand why municipal government resting upon manhood suffrage pre- sents so many difficulties in these latter days. To recapitulate : the manner in which the modern growth of cities has taken place is rather a larger natural increase in the city populations themselves (lower death rate!) than an increase in immigration from the rural districts ; the current of migration cityward has been observed for several centuries, but it is only in the nineteenth century that any considerable number of cities have had a regular surplus of births over deaths. Migration is predominantly a short-distance movement, but the centres of attraction are the great cities, toward which currents of migration set in from the remotest counties. The larger the city, the greater its power of attraction {i. e., the larger its proportion of outsiders, and the more distant the counties or districts which contribute to it). The mobility of great-city populations is below the aver- age mobility. ' Briickner, ibu^., \, 639. 284 THE GROWTH OF CITIES Women are greater migrants than men, but move only short distances ; marriage and domestic service being the levers of their action. Most migrants are young people, so that about 80 per cent, of the adult population of great cities is of outside birth. Two-thirds of the immigrants have lived in the great city less than 15 years. CHAPTER V. THE STRUCTURE OF CITY POPULATIONS. Before any estimate can be made of the influence which the process of concentration of population exerts upon the industrial and social life of the nation, it will be necessary to study closely the structure and composition of the city pop- ulations themselves, and note the characteristic traits which distinguish them from the rest of the people. For after all that is said in derogation of the effects of mere association upon people's lives, it remains true that differences in the physical composition of any population do really explain many of its peculiarities. For the purpose of this analysis it will be best to study a few great cities, for they are the type of urban population that all masses of people dwelling in compact centres will tend to resemble as they approximate the great cities in populousness. r. SEX. The simplest distinction made between individuals is that of sex ; it is important, because all social life is affected by the proportions of the sexes. Wherever there exists a con- siderable predominance of one sex over the other, in point of numbers, there is less prospect of a well ordered social life. The determination of the numbers of the sexes is there- fore preliminary to discussions of marriage ties and home life under the influence of agglomerated populations. The important fact here to be recorded is that the cities contain a larger proportion of women than does the rest of 285 286 ^'^-^ GROWTH OF CITIES the country ; and as women outnumber the men throughout Europe and the Eastern States of America, it follows that the preponderance of women is accentuated in the cities. As examples will serve the following statistics indicating the percentage of women in representative American cities and the commonwealths to which they belong: New York city 50.66 State 50.37 Brooklyn 51-12 " 50»37 Philadelphia 51.18 " 49-29 St. Louis 49.5 1 " 48.30 Boston 5 1 -44 " 5i'42 Indeed, of the fifteen leading cities of the United States, all but three, (Chicago, Bufifalo and Pittsburg), contained a larger proportion of women in 1890 than did the States in which they are situated. In European countries, where the population is usually more homogeneous than it is in the United States, with its industrial East and its newly-settled, masculine West, the tendency of cities to produce a surplus of women is more conspicuous. The German figures, for instance, show that to every 1,000 males, there are the following number of females : Germany (1890) 1,040 Small cities (5,000-20,000) 994 Middle-sized cities (20,000-100,000) 1,004 Great cities (ioo,ooo-|-) i>057 Here one observes a regular increase in the proportion of women to men, as one ascends from the smaller to the larger cities.'' Whether the same rule would hold in the case of villages does not appear; the Austrian statistics (Table CXXXI, last column) would indicate the reverse. But it is ' The garrisons, it appears, form a larger percentage in the populations of the middle-sized and small cities than in those of the great cities; but the difference is not great. — Kuczynski, i8. THE STRUCTURE OF CITY POPULATIONS 287 capable of statistical demonstration that in the majority of the older countries, the cities contain relatively more women than do the smaller places. The following statistics, drawn from the most recent censuses available, show that the only exceptions are Bavaria (due to garrisons in the small towns — Kuczynski, p. 27), European Russia, Servia and Aus- tralia : Females to t,ooo males. England and Wales, 1891 1,064 1. London 1,116 2. Urban sanitary districts 1,090 3. Rural " " 1,010 Scotland (1881) 1,076 1. Rural districts 1,031 2. Villages i>043 3. Towns 1,102 Sweden (1890) 1,065 1. Stockholm 1,204 2. Urban 1,191 3. Rural 1)038 Denmark (1890) 1,051 1. Copenhagen I>I79 2. Urban "^t^ZZ 3. Rural 1,01 1 Hungary (1890) 1,044 1. Buda-Pest 1,066 2. 27 other cities 1,084 3. Rural remainder 1,027 Netherlands (1889) 1,024 Urban (cities of 20,000-f) 1,123 Rural 982 Spain (1887) 1.039 Urban (^ca pi tales and communes of 20,000-f) 1,082 Rural remainder 1.029 Belgium (1890) 1.005 Urban 1.049 Rural 966 Servia (1890) 94S UrbaTi 777 Rural 977 Trance (189 I ) 1.014 Paris (1886) 1,045 288 "^^^ GROWTH OF CITIES Females to i,ooo males. Russia (1897) 999-7 European Kussia 1,028 St. Petersburg 826 Moscow 763 Poland 986 Warsaw 1,064 Percentage of males. New South Wales (1891) 53-89 Rural 58.07 Urban 51-67 Country towns 52.91 Metropolis 50-55 Professor Willcox was the first to give the data for the United States : ^ Males. Females. Per cent, of females. Towns (2,500+ inhabitants) .. 11,358,986 11,373,439 50.03=1001 Rural districts 20,708,894 19,180,931 48.08= 926 32,067,880 30.554,370 48-79= 953 The causes of the pecuHar distribution of the sexes are imperfectly understood. It is not an entirely new phenome- non, for many cities have long contained more women than men among their inhabitants. It existed in some German cities as far back as the fourteenth century.'' In the Nether- lands there was a large surplus of women in the cities sixty ^ The difference is more marked if we exclude from the urban population the smaller towns. Thus, in New York State all but three of the cities of 25,000 and upwards have a preponderance of women, the average for this whole class of cities being 50.88 per cent., and only 49.80 for the remainder of the common- wealth. In Massachusetts the percentage of women in towns of 25,0004- is 52.70, in towns of less than 1,000 population, 49.63. Finally, in the aggregate popula- tion of the 28 American cities of 100,000+ the ratio is 999 as against 953 in the total population. — Cf. " Distribution of the Sexes in the United States," Amer. Jour, of Sociology (May, 1896), i, 732. ^ Bucher, Die Bevolk^rung von Frankfort i59 i»i48 Rural districts 1,005 1,002 990 ' Levasseur, ii, 392. 'Cf. Mass. Census of 188^, vol. i, pt. i, p. 550. 2go 2"v7£ GROWTH OF CITIES city-born rather than the newcomers. This is notably true of the great German cities, as the following figures for 1890 demonstrate :^ Native city element. Immigrants. Total pop. Females 1,457.433 1.785.906 3,243,339 Males 1,303,927 1,756,175 3,060,102 Excess of females 153,506 29,731 183,327 Ratio, females to males. I,ll8 1,017 1,060 Professor Karl Biicher, thus observing that the Frauenuber- schuss of the cities was not the result of an immigration of females in excess of males, advanced a biological theory in explanation, saying that a city population has of itself a ten- dency to produce an excess of women above the general average." But he apparently overlooked the influence of emi- gration from the cities themselves. Unfortunately, no data are at hand for the ascertainment of the relative numbers of men and women born in the great cities of Germany who are no longer resident in the city of birth. But it is possible to compare the cities which are subject to the greatest amount of emigration with those subject to the smallest emigration and see what effect, if any, is exerted upon the proportion of women in the native population still remaining in the cities. The cities in which the percentage of those born in the specified city but now resident elsewhere in Germany was highest and lowest respectively, at the census of 1890, are given below, the second column denoting the number of females to 1,000 males in the native element remaining in the city: 3 ^Bleicher, p. 12. ' Cf. Die Bevolkerung des Kantons Basel-Siadt am i Dez. 1888, p. 19, and " Die Vertheilung der beiden Geschlechter auf der Erde," in Allg. Stat. JrcAw,ii, 390. ' Statistic des Deutschen Reiches, N. F., vol. 68, p. 71*, and Bleicher, i, pt. 2, p. 12. THE STRUCTURE OF CITY POPULATIONS 20 1 Stettin 38,78 1,182+ Konigsberg 32.77 1,238+ Hanover 32.72 1,095 — Dresden 32.55 I,I44+ Danzig 31.97 1,262+ HaUe 31.55 1,118 Hamburg 13.92 I>130+ Cologne 17.90 i)093 — Crefeld 18.23 i>075 — Frankfort 18.47 1,109 — Berlin 18.70 1,098 — Aachen 18.83 1,076 — Average of the 26 cities 22.31 1,118 Now it appears that in the six cities having the largest emigration (within Germany) the excess of females is usually- above the average (as denoted by the + sign), while in all but one of the six cities with the least emigration, the excess of females is below the average. This indicates that more males than females emigrate from the cities, producing a surplus of females which roughly corresponds to the amount of emigration. More decisive are the statistics of Austrian cities. Taking all the people born in Vienna and enumerated within the Austrian empire in 1890, there were 1,082 women to 1,000 men ; but in that portion of the native Viennese still resident in Vienna, the proportion of women was larger, namely i>095) — thus showing that more men than women had migrated. And this leaves out of account the emigration to foreign countries ^ which drafted even more males from Vienna than did the local, or internal, migration, by virtue of the law that men are the long-distance migrants.'' ^ Sixty per cent, of the emigrants to the United States are males. * Cf. the following figures regarding the native Viennese : Resident in Vienna 1,095 women to 1,000 men. " " neighboring districts ^,057 " " " " " " other districts of the same province 1 ,050 " " " " Resident in another province 972 " " " " — Rauchberg, Stat. Man., 19: 153. 292 THE GROWTH OF CITIES But it remains true that taking in the city population as a whole and including that portion which had emigrated as well as the portion which was at home, the proportion of females is above the average. Thus, the number of women to i,ooo men was found to be as follows in the latest census of Austria:^ Austria, average i>044 Towns of 500-2,000 (the maximum outside of the great cities). 1,049 All citizens of Austria born in Graz 1,066 " " " " « " Triest 1,040 " " " " " "Prague 1,130 " " « « " " Brunn 1,135 « " " " " " Lemberg 1,095 " " " " " " Cracow 1,190 Now unless it can be shown that the small towns lose fewer males by international emigration than the large cities do, it may be regarded as an established fact that the cities produce of themselves a larger preponderance of women than do the rural districts. This seems the more probable when it is considered that already among young children, among whom migration would affect both sexes alike, the ratio of females is higher in the city than in the country. In the entire United States^ there are 960 girls to 1000 boys under the age of one year ; in the 28 great cities there are 976, Again there are, in the United States, 965 girls to 1000 boys under the age of five years; in the great cities 982. In Maryland, the commonwealth surrounding Washington on three sides, there are only 981 girls to 1000 boys under the age of 15, and in the United States there are only 970 ; but in Washington there are 1,015, ^^ excess which can scarcely be explained by the phenomenon of migration. The difference is even 'Rauchberg, Stat. Mon., 19: 153. 'Cf. nth Census, Vital Statistics 0/ Cities, p. 13, and Report on Pop. THE STRUCTURE OF CITY POPULATIONS 293 more marked in the case of the colored children, for while Maryland has 994 girls to 1000 boys, Washington has 1,061. In Germany, too, the cities have relatively more girls than boys as compared with the country at large. In 1890 the number of females under the age of 15, to 1000 males of the same age was 986 in cities having 5,000-20,000 inhabitants, 989 in cities of 20,000-100,000 population and 1,005 i"^ cities of 100,000 or over. As a further piece of evidence it is to be noted that the immigrants to the cities form a larger per- centage of the males than of the females under the age of 1 5 ; thus, in 1885 the immigrants constituted the following per- centages : ^ Boys. Girls. Breslau 19.1 18.0 Leipzig 25.9 25.1 Berlin 16.6 16.3 Frankfort 19.3 19.2 Hamburg 13.9 14.2 In Hamburg alone do the immigrants add a larger per- centage of girls than of boys to the population under 1 5 years.'' These facts would indicate that some other force than migration has been at work to cause the large excess of women in cities. But how do cities of themselves produce the Fraueni'iberschuss ? In the first place, it is to be noted that the proportion of ^ Briickner, Allg. Stat. Archiv, i, 634. ^ In Frankfort it is only in the age period of 16-20 that the females begin to predominate among the immigrants; in 1890 the number of females to 1,000 males in each age-group was : City-born. Immigrants. Total. 0-5 997 918 988 6-10 993 976 990 11-15 1,004 955 994 16-20 1,122 1,178 1,157 — (Bleicher, pp. 5, 6.) 294 THE GROWTH OF CITIES female births is larger in the city than in the country. This fact has long ago received statistical proof.^ A good illus- tration is to be found in France. According to Levasseur,"^ the average number of male births to lOO female births dur- ing the period 1801-65 was 103 in the department of the Seine (which is nearly coincident with the city of Paris), 104.3 iri the remainder of the urban population and 105.3 in the rural population. The difference is not due to the in- fluence of illegitimate births, in which there are fewer boys than girls and which are especially prevalent in the cities ; for, restricting ourselves entirely to the legitimate births .he figures would be 103.6, 104.6, 105.7 ii^ the order given above. In the city of Frankfort in Prussia the ratio has been remarkably steady for nearly two centuries, having been in the period 1701-50, 103.3; in 1751-1800, 103.4; in 1801-50, 103.3; in 1851-90, 103.7; the average in 1635- 1890 being 103.8.3 In Prussia itself the ratio has been higher, having fluctuated between 105 and 106 in 1874-91. -^ The causes of sex are still too imperfectly understood to permit an assignment of reasons for the great excess of male births in the country, but the most likely reason seems to be the inbreeding of a few family stocks, while in the city there is vastly more crossing. The fact ' The ratios for the middle of the century are given by Wappaeus and Legoyt, the latter's being as follows (p. 69) : Births. Boys to 100 girls. City. Country. France 105.06 106.75 Prussia 105.31 105.95 Belgium 104.51 105.57 Holland 107.73 106.72 Denmark 105.73 106.19 Sweden 104.62 105.06 ' La Population Frangahe, ii, 20. ' Bleicher, p. 343. * Conrad's Hdwbh., Sup. i., 215. THE STRUCTURE OF CITY POPULATIONS 295 that Jewish famiHes have a notably large excess of boy- births argues in favor of this hypothesis.^ Not only are relatively fewer boys than girls born in the city, as compared with the country, but more male children die in the earlier months of life. The science of demography recognizes the fact that infant mortality bears with more severity on boys than on girls, and as infant mortality is in most countries higher in the city than in the country,^ it fol- ^ Mayr, Bevolkerungsstatisiik, 188, where a bibliography of the subject of sex at birth will be found; to it should be added the Theory of Sex Development, by Dr. Leopold Schenk, as well as the Evolution of Sex, by Geddes and Thompson (in Contemporary Science series), and Bertillon's monographs on Nataliti. On the statistics, see especially Boeckh, in Bulletin of the Intern. Institute of Statis- tics, vol, V. * In the urban population of Massachusetts, 1881-90, the infant mortality was 175, in the rural population, 129.5. {,Rep- of State Board of Health for i8g6, p. 753.) Some recent Prussian statistics may be cited in illustration of both the propositions in the text (Bleicher, p. 268) : Table CXXXIII. Number of deaths of children under one year to each i,ck)o living births (i8go-i) IN each category. Legiti- Illegiti- Differ- Male. Female. Total. mate. mate. ence a. b. c. d. e. a-b. Prussia . 220.9 188.7 205.2 192.7 357-3 32.2 Rural communes . . . . 210.8 180.0 195.8 185.7 333-1 30.8 Urban . 238.0 203.4 221. 1 204.8 389.0 34-6 Cities under 20,000. • • 225.3 190.7 208.5 194.4 378.3 34.6 Cities 20-100,000.. • 230.5 1978 214.5 199.8 389.9 32-7 " 100,000-f . . • • 259.3 223.2 241.7 221.9 397-4 36.1 Berlin . 269.2 230.8 250.5 227.6 412.7 38.4 Dr. F. S. Crum has recently shown {Quarterly Journal of Economics, 1897, xi, 259) that in Massachusetts infant mortality increases in direct ratio with density of population, or, in other words, with the numerical size of cities. The subject of infant mortality is discussed in all hand-books of vital statistics or demography, — Newsholme, Farr, Bertillon, etc. The le.-iding American authority is Dr. John S. Billings doth Census, vol. xi, " Relation of Age to Deaths "). An excellent article on infant mortality is con- tributed by Dr. T. B. Curtis to Buck's Hygiene and Public Health, vol. ii, pp. 269-301. A comprehensive sociological study has been made by Seutemann, Kindersterblichkeit sozialer Bevolkerungsgruppen, which appears as vol. v of F. J. Neumann's Beiirage zur Geschichte der Bev'dlkerung in Deutschland. Cf. also Silbergleit, " Ueber den gegenwartigen Stand der Kindersterblichkeit, ihre Erscheinungen und ihre Entwicklung in Europaischen Grossstadten,' Hygienische Rundschau, 1895. 296 THE GROWTH OF CITIES lows that the ranks of the males in the city are depleted by this natural cause. The effect of infant mortality upon the numerical relation of the sexes is conspicuous in the case of the negro race. In New York city, for example, the number of deaths of children under the age of five years to 1000 living of the same age was 231.09 for the colored and 115.65 for the whites ; ' in the same city the ratio of females in the colored population was 1,029 ^s compared with 1,011 in the native population of native parentage. Professor Willcox has observed that the six common- wealths in which the cities (towns of 2,500+ inhabitants) have a percentage of females above 52 are States with a large negro element (Mississippi, South Carolina, Louisiana, Georgia, District of Columbia, Maryland, North Carolina).'' In the cities of these States the whites do not have a very large excess of females; it is rather among the negroes, thus: Percentage of females in total population of the cities; Whites. Colored. Mississippi S^-^S 56.00 North Carolina 51.12 53-33 District of Columbia S^-OS 55-34 Georgia 50.33 54.84 But it is not alone the high rate of infant mortality which depletes the ranks of males in the city; there are other dangers to the health of males which show their effects oftener in the city than in the country, and thus tend to ac- centuate the urban excess of females. City occupations are oftener subject to fatal accidents; and, in fact, violent deaths ^ nth Cen., Vital Statistics of Cities, p. 44. * American yournal of Sociology, vol. i. THE STRUCTURE OF CITY POPULATIONS 297 of all kinds are more frequent in cities, as is expressed in the following Scotch statistics : ^ Proportion to Proportion of entire pop., 1871. violent deaths, 1878. Principal cities (25,000-f- pop.") 31.80 39-98 Cities of 10,000-25,000 9.95 28.15 Towns of 2,000-10,000 23.10 1 1.83 Rural pop. (mainland) 31.22 11.94 " " (islands) 3.93 8.10 Violent deaths, of course, affect principally men rather than women. On this point the statistics of the little duchy of Oldenburg once more afford invaluable evidence. On the average population of 1871-85, for each 10,000 living persons, there were the following number of deaths : ^ Table CXXXIV. Urban (towns of 2,000 + ). Rural. Age-groups. Male. Female. Difference. Male. Female. Difference. 0-5 654 573 81 534 471 63 5-10 97 87 10 87 86 I 10-15 51 62 — II 50 56 — 6 15-20 77 73 4 65 58 7 20-30 125 96 29 95 84 II 30-40 158 138 20 103 126 — 23 40-50 230 156 74 156 139 17 50-60 323 232 91 243 215 28 60-70 567 453 114 479 490 —II 70+ 1.213 1,166 47 1,253 1,252 I All ages 244 222 22 225 209 16 In the entire town population there is thus found a death rate of 24.4 for males and 22.2 for females, and in the country the difference is somewhat less, the rates being 22.5 and 20.9 respectively. In only two periods does the city, ^ Walford, " On the Number of Deaths from Accident, Negligence, Violence and Misadventure," y. of St. Soc, Sept., 1881, p. 28. * Statistische Nachrichten iiber das Gi'ossherzogthum Oldenburg, Heft xxii, p. 1 14. 298 THE GRO WTH OF CITIES as compared with the country, show more favor to man than to woman; during the ages of 10-15 ^"^^ 15-20, city females die more rapidly than the men, as compared with the rural rates. In the ages 5-10 and 20-30, there is also little differ ence between city and country, as regards the comparative death rates of males and females. But that infant mortality is higher in the city than in the country and is more severe on boys than on girls, appears once more in the figures for the age-period 0-5. It is after the age of 30, however, that city life appears to cause the greatest mortality of men as compared with women. In the years 60-70, the rural death-rate is higher for women than for men, while the city death-rate for men is far above that for women. Why do so many more men, relatively, die in the city than in the country? All things considered, it must be ex- plained on the ground that city occupations are more dangerous to the health of males than are country occupa- tions, while to females over the age of 5 years there is not a great difference between city and country. In other words, it is not the city air and conditions of life, so much as the peculiarity of city occupations, that cause higher mortality in the city than outside. The number of deaths among adult males has apparently been increasing in the city faster than in the other classes of population. Thus in Frankfort during four decades the number of female deaths to 100 males dying was as follows : ^ Married Total. women. Widows. Girls. i8ii-20 373 51 46 276 1821-30 326 47 47 232 1831-40 231 37 36 158 1841-50 189 33 32 124 ^ Bleicher, 244. Cf. with the foregoing conclusions, the conclusions reached by Dr. Kuczynski (p. 231) after a comprehensive statistical investigation: "While it is true that on the whole the mortality of the male population in the great cities of Prussia is somewhat more unfavorable than in the other parts of the Prussian monarchy, the same cannot be affirmed of the female population." THE STRUCTUFE OF CITY POPULATIONS 299 Since, then, the cities have a high mortality among males, both infant and adult, and a relatively low percentage of male births, it follows that the cities would have an excess of females, even though they were entirely isolated and cut off from immigration or emigration. In the face of these facts, advanced students have abandoned the old belief that nature tended to equalize the number of the two sexes and that in- equalities must be explained by migration, war, pestilence, etc.,^ and Prof. Biicher has laid down the rule that the relative numbers of the two sexes are determined by the relation be- tween the rates of natural increase of the two sexes and cannot stand permanently above or below this ratio, provided there be no migration. There will be an equilibrium, of course, when females become so numerous that even with a lower death-rate, the number of deaths among them would exceed the number of births. To summarize : The excess of females in any population is usually ascribed, first, to the heavier mortality of male than of female infants, which within the first year usually efifaces the superiority of male births. Then comes the great mortality of adult males due to the dangers of their occupa- tions, as well as to vice, crime and excesses of various kinds which shorten life." Now all these forces are accentuated in the cities, producing a greater excess of females there than elsewhere, even without the influence of emigration, which in- creases the surplus of women in cities. In the cities also, the ' Hence it was easy to explain the preponderance of females in Europe and of males in America as being the result of international migration. This is indeed the theory of the nth Census (vide Rep. on Pop., vol. i, p. Ixxi). Unfortunately for the theory, however, there are some European countries, notably Italy, that are subject to heavy emigration, and yet have an excess of males, while in the United States the " native whites of native parents," a class obviously not affected by migration, also have a large excess of males (966 females to 1,000 males) in- stead of an equal number or slight excess of females, as the theory demands. ' Mayo- Smith, Statistics and Sociology, 41. 300 THE GROWTH OF CITIES superiority of male births over female births is smaller than in the country. II. AGE. The age classification of any population is a record of biological facts second only to the distinction of sex. Not to mention the economic significance of the predominance of the productive classes, or of children and aged persons, one must recognize the influence of the age-distribution upon percentages that have wide-reaching social significance, such as the birth-rate or death-rate. As the standard text-books have pointed out, the curve of ages approximates to the form of a pyramid, the newly-born forming the base and the very aged the apex. The more rapid the increase, the broader the base. With a stationary population, on the other hand, the base narrows and the upper part of the curve, representing the middle and advanced ages, bulges out, so that the age curve becomes bell-shaped. Finally, when the normal age-classification is disturbed by migration, there result two typical forms of the curve: (a) in the case of immigration, the curve may be likened to a top, on account of the expansion of the middle age-periods; (b) in the case of emigration, the curve sinks in the middle and the figure becomes spindle-shaped. All but one of the forms are found in the United States, that of the stationary population, which France tends to approach.^ Now the age-curve of city populations inclines to the top- shape, apparently as a result of immigration. Compare, for example, the distribution by age-periods of the population of 1 Levasseur (vol. ii, ch. 15) has a model chapter on age-groups, copiously illus- trated with diagrams. Compare the summary in v. Mayr, Bevolkerungsstatistik, § 30; Mayo-Smith, Statistics and Sociology, ch. iv; and diagrams in Statistical Atlas of the United States. THE STRUCTURE OF CITY POPULATIONS 301 the United States and that of its 28 great cities (those of 100,000+ inhabitants) : ^ U. S. Citiei. Under 15 355.3 299-46 15-65 603.5 668.17 65+ 38.6 29.43 Unknown 2.6 2.94 1,000. 1,000. Here, then, is a noticeable difference in age-grouping; the city population is markedly strong in people of the middle, active, productive ages, and has relatively few children and aged people. Similar contrasts are found in European countries. The Hungarian census for 1890 (p. 143*), for instance, compares the total population in the three classes, taking one class as a standard at 100: 37 free Remaining Ages. Budapest. cities. population. 0-15 70 88 100 15-40 137 no 100 40-60 95 ICO 100 6o-|- 78 III 100 The German statistics too, are, very informing : Table CXXXV. Age distribution in cities of 100,000 + . 20,000-100,000. 5,000-20,000. Germany. Under 15 292 321 345 351 15-40 474 450 417 387 40-60 177 169 170 182 604- 57 60 68 80 1,000 1,000 1,000 1,000 Females TO 1,000 males: Under 15 1,005 989 986 995 15-40 1,010 916 909 1,027 40-60 I«I36 I>i42 1,109 1,094 60+ 1,616 1,518 1,364 1,196 All ages 1,057 1,004 994 1,040 * Calculated from the nth Census Reports on Pop. and Vital Statistics of Cities, p. 16. 362 THE GROWTH OF CITIES Starting with the percentages for all Germany, there is a steady decrease in the proportion of children and aged people, as one progresses through the small cities to the great cities, and on the other hand a steady increase in the number of persons at the active age of 15-40, while the pro- portion at the age 40-60 remains nearly constant. ^ The cause, as stated above in Chapter IV (p. 280) is in the main the migratory movement. The percentages there given to show the age-grouping among the native Frankforters and the newcomers would be more fully realized in a diagram," wherein the curves should show the ages of native residents of Frankfort, and of immigrants. The former is bell-shaped, the latter top-shaped, as is the curve for the entire popula- tion, which is typical of all city populations. (Cf. the dia- gram representing the ages of men and women in Stadt and Land in Switzerland in the Census of 1888, Schweizerische Statistik., Bd. 88.) This difference of city and country as regards age group- ing was remarked as early as 1872 by Georg von Mayr in taking the Bavarian census. And recent investigations show ^The Austrian statistics, even minuter, confirm the foregoing figures and deductions : Table CXXXVI. Classes of towns: 500- a.ooo- 5,000- 10,000- Under 500. 2,000. 5,000. 10,000. 30,000. 20,000 + . Austria. 10 years and under. 268 278 269 243 226 193 260 11-20 194 195 197 199 202 193 195 21-30 144 150 159 183 201 214 160 31-40 124 125 130 132 134 151 129 41-50 107 106 104 103 100 113 107 51-60 84 79 76 73 72 73 79 61-70 54 47 45 45 44 44 48 70-}- 25 20 20 22 21 19 22 — Rauchberg, Stat. Monats., xix, 133. For the elaborate statistics of France, see Risultats statistiques du dinotnbrement de i8gi, p. 222. ^ The usual form of diagram consists of two curves, one for each sex, thus mak- ing the figures already discussed. THE STRUCTURE OF CITY POPULATIONS 303 that a difference has existed throughout the century in Germany. As preceding tables have indicated, the differ- ence at present is about five per cent, in favor of the cities for the middle ages and of the country for childhood. In Prussia the same differences existed in 1816 and 1858: ^ Males. 0-14. 14-60. 60+. Total. g^ f Urban 32.10 61.09 6.81 100. /Urban 31.92 63.03 5.03 99.98 ^ Rural 37.70 55.57 6.73 100. 1858. ' •-Rural 36.36 57.93 5,68 99.97 In Leipzig, the age-grouping in 1792-94 was about the same as it is now, and Dr. Kuczynski, to whose labors these results are due, concludes that at least in Berlin, Leipzig and Frankfort, the strong representation of the middle age classes and the small percentage of children are not phe- nomena of the most recent times, and that in fact the differ- entiation is not even a product of the present century. =* Hence the conclusion that migration cityward began earlier than 1800. Table CXXXV also illustrates the fact mentioned in the preceding section, that the preponderance of women in the great cities begins at an age when only biological forces could be at work, although its rapid increase in the later years shows that the high death-rate of males in middle life, and their emigration from the city, are also to be reckoned with. The preponderance of aged persons in the village and country districts, as contrasted with the city, is not to be explained by the existence of a " return current." In that case the excess of aged women would be greater in the country than in the city. 3 As the resnlt of the presence of a relatively large number ^ Kuczynski, 262. * //5tV/., 270. ' See further on the subject of a " return current of migration," Ogle, " Alleged Depopulation," etc., in J. of St, Soc, 1889, 216. 304 THE GROWTH OF CITIES of persons in the active period of life in urban populations, one would expect city life to be easier and more animated, the productive classes being large and having a smaller burden to bear in the support of the unproductive class. One would also expect to find more energy and enterprise in cities, more radicalism, less conservatism, more vice, crime and impulsiveness generally. Birth-rates should be high in cities and death-rates low, on account of age grouping. III. RACE AND NATIONALITY. A third natural difference among men is that of race, which may or may not be accompanied with a differ- ence of nationality. The Italian and the German are racially differentiated, but men of both races are citizens of the Swiss republic. It will therefore be convenient to consider race and nationality together, although nationality is a political rather than a natural distinction. In the chapter on migration it was shown that foreigners are found in largest numbers in cities. Table CXXV, indeed, shows that the larger the city, the greater the percentage of foreigners in its population. But while the foreign ele- ment is strong in European capitals compared with pro- vincial cities, its numerical strength there is nothing com- pared with its position in American cities. The difference was brought out in Table CXXVII. The composition of the city populations of the United States is therefore of especial interest on account of their large contingent of European- born. Taking the 28 cities of 100,000 population and up- wards in 1890, we find the following elements of population: Bom in State where resident 5,082,637 52.4 Some other American State '■>530>67S I5>8 Foreign countries 3,084,648 3 1 .8 9,697,960 100.00 THE STRUCTURE OF CITY POPULATIONS 305 Somewhat over one-half the population of our cities was born either in the city itself or in the commonwealth to which it belongs, while a little less than one-third was born in foreign countries. In the entire country only 14.77 P^^^ cent, of the population is foreign-born. The cities therefore con- tain more than their due proportion of foreigners. Ex- pressed in another form, the 28 great cities, while constitut- ing in 1890 15.5 per cent, of the entire population of the United States, contained 12.4 per cent, of all the American- born, and 33.4 per cent, of the foreign-born in the United States. There is therefore a decided tendency on the part of the foreigners to settle in our largest cities. The ques- tion is. Is this tendency recent and increasing, or is it a natural and permanent incident of the process of distributing the newcomers? As to the process of distribution, it may be remarked at the outset that it is not entirely the task of the seaports ; Chicago has had at nearly every census a larger percentage of foreigners than New York or San Francisco. Table CXXXVII shows the proportion of the foreign-born in the population of the ten leading cities since i860, com- pared with the proportion of the foreigners in the population of the commonwealths and the entire country. From this table may be deduced the rule that in the United States and in most of the commonwealths, the percentage of foreigners has uniformly and almost steadily increased since 1850, while on the other hand, it has decreased in the cities. The conclusion is the opposite of a very general belief,^ which is probably founded on a comparison of the census figures of 1890 and 1880. The percentage of foreign-born in 1880 was almost ^ " The tendency is to increased concentration of the foreign-born in large cities owing to the increased immigration of Latins and Slavs." Mayo-Smith, op. cit., p. 302. Cf. the arguments advanced in behalf of the Lodge bill restrict- ing inmiigration. 3o6 THE GROWTH OF CITIES uniformly lower, both in the city and in country, than at the preceding censuses. This was a result of the relatively small immigration in 1870-80 (2,800,000 as against 2,300,000 and Table CXXXVII. Percentage of foreign-born in the total population of States and cities named ; 1850. i860. 1870. 1880. 1890. United States 9.68 13.16 14.44 "S-SZ I4-77+ North Atlantic States (9)... 15.37 19.10 20.49 I940 22.34+ Massachusetts 16.49 21.13 24.24 26,78 30.77+ I.Boston 35-874- 31-53 31-64 35-27 New York 21.18 25.80 25.97 23.83 26.19+ 2. New York city 47-i6+ 44-49 39-68 42.23+ 3. Brooklyn 36.54+ 31-36 32.46 Pennsylvania 13.12 14.81 15.48 13.73 16.08+ 4. Philadelphia 29.96+ 27.24 24.12 25.74 South Atlantic States (9)... 2.24 3.03+ 2.85 2.29 2.35 Maryland 8.78 11.28+ 10.68 8.86 9.05 5. Baltimore 24.71+ 21.09 16.89 15.88 District of Columbia ... . 9.51 16.63+ 12.34 9.64 8.15 North Central States (12)... 12.04 "6.97 17.97 ^6.80 18.16+ Ohio 11.02 14.03+ 13.98 12.35 ^2.51 6. Cincinnati 45.09+ 36.81 28.09 24.05 Illinois 13.14 18.96 20.28 18.96 22.01 + 7. Chicago 49.99+ 45.01 40.71 40.98 Missouri 11.23 13.58+ 12.91 9.76 8.77 8. St. Louis 50-53+ 36.1 1 29.96 25.43 South Central States (8)... 3.18 3-99+ 3.62 3.08 2.93 Louisiana 13.18+ 11.44 8.51 5.76 4.45 9. New Orleans 38.31+ 25.32 19.05 14.20 Western States and Ter- ritories (11) 15.11 28.92 31.64+ 28.29 25.46 California 23.55 38.56+ 37-95 33-87 30.32 10. San Francisco .... 49.32+ 44.56 42.11 The sign (+) designates maxima. In most of the cities, the maximum occurs in i860, while in the United States (and in many of the commonwealths, including most of the Northern States) the maximum was at the latest census, 1890. In connection with this table, the author desires to acknowledge his indebtedness to Dr. Dclos S. Wilcox. 2,600,000 in the decades 1860-70, 1850-60.) Unless the im- migration constantly increases, the percentage of foreigners in the total population will diminish, as the children of the immi- grants go to swell the number of natives. With the exception THE STRUCTURE OF CITY POPULATIONS 307 of 1880, therefore, the percentage of foreigners has constantly increased in the general population and as constantly decreased in the largest cities, signifying a slow but certain process of equalization. The cities, then, act as centres of dispersion for the immigrants, and the fact that in New York and most of the other great cities the percentage of foreigners was smaller in 1890 than in 1870 shows how rapidly dis- persion takes place. The number of immigrants in 1880-90 was unprecedently large (5,200,000, or double the average of the three preceding decades), and without rapid distribu- tion would have greatly increased the percentage of the foreigners in 1890 as compared with previous census years. Even as compared with 1880, the percentage of foreigners had increased less rapidly in the larger cities than in the rest of the country. Computations by Professor Willcox show that in the fifty most populous cities of the country in 1880, 29.9 per cent, of the population was of foreign birth, while in 1890 the percentage in the same cities had risen only to 30.8. Whereas the percentage of persons of foreign birth in the rest of the country (including substantially all places of less than 56,000 inhabitants in 1890) had increased from lo.o per cent to 11.3 per cent.^ As to the inclination of the various nationalities to dwell in the large cities, the eleventh census gives us the following proportions of their total number who in 1 890 were dwelling in the 124 cities of at least 25,000 inhabitants:' ^ The Federal Census, Publications of the American Economic Association, New Series, No. 2, p. 24. * Rept. on Pop., vol. i, p. cl. 3o8 THE GROWTH OF CITIES Total number in Per cent, thereof in prin- United States. cipal cities (25,000+). Italians 182,580 58.79 Russians 182,644 S7'90 Poles 147)440 57-1 1 Irish 1,871,509 55.97 Austrians 123,271 48.33 Bohemians 118,106 48.32 Germans 2,784,894 47-71 French 113,174 45-69 Hungarians 62,435 44-78 Scotch 242,231 41-25 English 1 909,092 40.70 Chinese 106,688 40.19 Dutch 81,828 33.54 Swedes 478,041 31.81 Canadians ^ 980,938 31-36 Swiss 104,069 31.15 Welsh 100,079 25.80 Danes 132,543 23.24 Norwegians 322,665 20.78 Mexicans 77,^53 7-97 Other nationalities 127,467 39.22 Total 9,249,547 44-13 From this table it would appear that the least desirable class of immigrants — those from Southern Europe — are most prone to remain in the great cities. The fact is, how- ever, that much, if not most of this immigration is very recent and there has hardly been time for the new arrivals to disperse. Comparing [890 with 1880 it will be found that the percentage of those remaining in cities declined among the Italians, Poles and Hungarians, while only among the Russians did the increase considerably exceed the increase for the Germans and Irish. For the following data the author is indebted to Professor Willcox of Cornell : ' Includes Great Britain not specified. ' Includes Newfoundland. THE STRUCTURE OF CITY POPULATIONS 309 Percentage in cities ' of total in U. S. (Increase or decrease between 1880 and 1890 is denoted by ~r or — .) 1880. 1890. j- Russians 24.2 53.8 — Italians 61.3 53.7 —Poles 53.5 51.7 -|- Bohemians 39,9 44.4 +Austrians 35.0 43,8 — Hungarians 58.6 39^, -|- All six 44.0 49.4 — Other foreign-born 34.6 32.9 +Native 12.6 14.5 +Total 15.5 17.9 -j-Germans 39.4 42.0 -|- Irish 45.8 48.5 +English 30.7 33.0 + Canadians 21.9 224 +Swedes 17.0 25.1 While the immigrants from Southern Europe have been recent arrivals and have not widely dispersed, the Canadians, who have been coming across the boundary line for a long time without having to pass through a great city on their entrance, were widely scattered in 1890 and nearly as much so in 1880. The Mexicans, of all nationalities with a repre- sentation of over 50,000 in the United States, are the least concentrated in large cities ; a fact which without doubt re- sults from their manner of entrance and the size of the centres of population nearest their own country. With the decline of railway building and the complete occupation of public lands it may be expected that immigrants in the future will disperse less readily than in the past. As they belong almost entirely to the lower ranks of laborers they will be able to find employment only in the cities, whose large public works and manufactures demand muscular, un- skilled labor. ' The cities referred to are the 50 principal cities of 1880. 310 THE GROWTH OF CITIES It has been noted that while the foreign-born element constitutes 14.77 P^*" cent, of the population of the United States, it forms 31.8 per cent, of the population of the great cities (100,000+ pop.). In the case of the negroes, the per- centages are at present reversed, but since their emancipa- tion they have been migrating to the cities to a considerable extent. Hoffman gives the following comparative rates of increase, 1860-90, in the Southern States and their sixteen principal cities (14 in i860) for the whites and the negroes : ^ White. Colored. Cities 94.1 1 242.60 Remainder of popiilation 45-52 A^-^Z This indicates a movement of the colored people city- ward, causing a smaller rate of increase in the rural parts of the country and a much higher rate in the cities than ob- tained among the whites. The colored constituted the fol- lowing percentages of the entire population of the i860. 1890. Ten States 36.CO 35-96 Cities specified 18.85 29.08 "In i860, 11.67 P^"* cent, of the white population lived in the large cities, increasing during 30 years to only 14.89 per cent. ; in contrast with an increase from 4.82 per cent, of colored urban population in i860 to 10.93 in 1890." "" If pres- ent tendencies continue, the negroes will be more inclined to city life than the whites of the same State. Hoffman, in- deed asserts that "during the last decade this migratory tendency of the colored population has been more pro- nounced than ever." * But a critic in the New York Even- ^ Race Traits and Tendencies of the American Negro {Pubs, of Amer. Ecoti. Assn., vol. xi, pp. 9, 10). ^ Ibid., II. ^ Ibid., 12. THE STRUCTURE OF CITY POPULATIONS 311 ing Post has pointed out that the changes noted by Mr. Hoffman as occuring in the period 1860-90 were for the most part effected in the era of emancipation (1860-70), when the negroes naturally felt themselves under a strong impulse to see something more of the world than could be viewed from the plantation. The percentage of negroes in the entire population of the selected cities increased as fol- lows: i860 18.85 1870 27.74 1880 28.55 1890 29.08 Mr. Hoffman also errs in his statistical tables designed to show that the negroes are increasing more rapidly than the whites in the large cities of the country. He includes in one table cities that contained more than 20,000 negroes in 1890, and in a second table those that contained between 10,000 and 20,000 negroes; but the sum total of all the cities in both classes contradicts his inference that the negroes are increasing in these cities more rapidly than the whites : ^ White population. Colored population. 1880. 1890. Increase ^. 1880. 1890. Increase ^. Group I .. 3,117,174 3.965.7" 27.22 376,316 498,104 32,36 Group II.. 1,407,834 2,386,493 69.51 74,875 129,849 73.42 Total 4,525,008 6,352,204 40.00 451. 191 627,953 39.00 Evidently Mr. Hoffman's methods are not beyond criti- cism. As a matter of fact, the percentage of negroes in- creased in very few of the great cities. Taking Mr. Hoff- man's figures for a few of our largest cities that contained more than io,ooo persons of African descent in 1890, we find ^ Race Traits and Tendencies of the American Negro {Pubs, of Anier. Econ. Assn., vol. xi, pp. 12, 13.) 312 THE GROWTH OF CITIES that they constituted the following percentages in the popu- lation of 1880 and 1890: 1S80. 1890. New York 1.66 1.60 Chicago 1.30 1.31 Philadelphia 3.9 3.9 Brooklyn 1.45 I.29 St. Louis 6.8 6.3 Baltimore 19.3 18.3 Cincinnati 3.3 4.1 New Orleans 36,4 36.2 Washington 48.5 49.0 While there undoubtedly exists an important movement of the negroes to the North, directed in the main toward the cities, ^ negro mortality is here so high on account of climatic and other conditions, that in most of the great cities it prevents the negro race gaining upon the whites, or even holding its own. But in the smaller cities of the South it is otherwise, and trustworthy statistics clearly show that the movement cityward in the Southern States affects the negro more than it does the white population. Thus of the total white and negro populations in the States specified, the following percentages dwelt in towns of 4,000 and up- wards in 1880 and 1890:^ Table CXXXVIII. Whites. Negroes. 1880. 1890. Difference. 1880. 1890. Difference. Alabama 4.44 7.79 3.35 4-82 8.68 3.86 Arkansas 1.29 4.63 2.34 2.14 7.25 5. 11 Florida 10.02 14.39 4.37 7.96 14.89 6.93 Georgia 7.68 13.74 6.06 7.75 14.59 6.84 Kentucky 36.16 33.35 2.81 ' 13.79 13.96 .17 Mississippi 2.87 4.38 1.51 2.20 3.68 1.48 In five of the six commonwealths here represented, the 1 Cf. F. J. Brown, The Northward Movement of the Colored Population, 2 Lectures of Professor W. F. Willcox on Social Statistics. ' Decrease. THE STRUCTURE OF CITY POPULATIONS 313 urban percentage increased more among the negroes than among the whites in the decade 1880-90; and Maryland is the only other Southern State that stands with Mississippi as an exception in this respect. The consequences of this tendency of the negro race to move into the towns and cities are, in the opinion of the present writer, likely to be wholesome rather than the oppo- site, at least so long as the movement is toward the towns and smaller cities in the South, rather than toward the great cities of the North, where the conditions of life in the negro quarters are pitiable in the extreme. ^ To learn industry, thrift and self-reliance, the negro needs to be removed from the isolation, ignorance and shiftlessness of the farm or plan- tation and brought under white superintendence in the town. If the present experiments in the direction of employing negro labor in Southern cotton factories result successfully, there promises to be a period of real advance in the econ- omic efficiency of the negro, and thus an eventual solution of the negro problem. Mr. Hoffman pictures in lurid light the effects of the ex- cessive mortality of the negro race (especially in the cities) ; its increase since emancipation ; the impairment of the con- stitutional vigor of the race, as seen in the abnormal preval- ence of consumption and other pulmonary troubles, and in the excessive infant mortality ; the enormous economic waste involved in the death of many of the young and the survival of comparatively few to the productive ages ; the hindrance that such a population offers to the development of our cities, etc.'' It may be doubted, however, whether the negro race has not a considerable power of resistance to these forces of deterioration that Mr. Hoffman predicts will wipe ^ See, for example, the letter of Mr. Paul Lawrence Dunbar to the New York Sun, Sept. 4, 1897, ^i^d foJ^ details, Hoffman, op. cit. passim, but especially ch. ii. - Cf. the summary, op. cit,, pp. 145-8. 314 THE GROWTH OF CITIES it out of existence after the manner of the Maories and Sandwich Islanders. For example, there is some evidence, not presented by Mr. Hoffman, to the effect that negro mor- tality is now diminishing rather than increasing in the Southern cities.^ Whether or not an inroad has as yet been made upon the sexual immorality that is the bottom cause of race degeneration, there is this consolation : A band of intelligent, highly educated, self-sacrificing negroes has fully possessed itself of the perilous situation of the race and is struggling manfully and hopefully toward the true goal. So long as the race can produce leaders like Booker Washing- ton and Paul Lawrence Dunbar, there is no cause for despair. Whatever tendency can be shown to make for the im- provement of eight million Americans cannot fail to be of vital import to the nation as a whole. In the author's opinion, the nation can view with equanimity the movement of the negroes from the farm to the smaller cities. This is certainly preferable to a concentration of the race on farm- ing lands in a " black belt." IV. OCCUPATION. The connection between occupation and place of habita- tion is as close as possible and requires few words of expla- nation in this place. The entire essay, as a matter of fact, is occupied with the theme in its broadest aspects, — how in- dustrial organization conditions the dwelling-places of individuals. ' The Bulletin of the U. S. Dept. of Labor., May, 1897, presents the following statistics compiled from the official health reports of the several cities (pp. 270- 283): Deaths pbr 1,000 of negro population: Atlanta. Memphis. Richmond. Baltimore. Charleston. 1882-85 37-96 43.01 40.34* 1880-84 36.15 44.08'^ 1886-90.... 33.41 29.35 38.83 1885-89 30.52 46.74 1891-95 32.76 21. II 34.91 1890-94 31.47 41.43 •^ For 1881-85. ^ For 1881-84. THE STRUCTURE OF CITY POPULATIONS 315 The German census of occupations in 1882 affords the fol- lowing comparison : Table CXXXIX. Percentage of inhabitants belonging to specified occupation in Cities Cities Towns Cities 20,000- 5,000- 2,000- Rural All 100,000 + . 100,000. 20,000. S,ooo. parts. Germany. Agriculture i .4 3.4 9.9 26.3 64.5 42.5 Mfs. and mining 47.3 52.8 53.6 49.0 24.4 35,5 Trade and commerce . . 26.6 19.5 15.6 11.6 4,9 lo.o Casual day labor 5.0 4.5 4.3 2.9 0.7 2.1 Liberal professions, etc. 10.7 11.2 9.1 4.9 2.3 4.9 Free incomes; unknown. 8.9 8.6 7.6 5.4 3,7 5.0 99.9 loo.o loo.i loo.i 100.5 loo.o Agricultural pursuits naturally figure poorly in the great cities, and gain ground as one descends to the village and hamlet. Industry, it is worth noting, makes a smaller per- centage in the great cities of Germany than in the smaller towns. While this is doubtless due in part to suburban de- velopment, and while the inclusion of mining in the rubric must also favor the smaller towns, it confirms the conclusion reached earlier in the essay that it is not so much manufac- turing industry as commerce that builds up great cities. Casual labor is also affected by the degree of agglomera- tion ; the statistics show that the process of concentration of population increases this unfortunate class. On the other hand, it also increases the strength of the liberal professions, which would usually be counted a blessing unless these ranks absorbed too many men in the effort to cure abuses in law, hygiene, medicine, religion and education. The sixth class is made up largely of people without occupation, in- mates of institutions, and people living on their incomes. This class, too, is attracted to the cities. It will be of interest to see what percentage of the popu- lation is engaged in gainful occupation in the various cate- gories of towns, although, to be sure, such percentages 3i6 THE GROWTH OF CITIES depend in a measure on the proportion of children. This proportion being smallest in the great cities, one would ex- pect to find there the largest working population. Such was the case according to the German occupation statistics of 1882: Percentage of adult Percentage of inhab- women itants engaged. occupied, in domestic service. Cities loOjOOo-f- 40.3 24.1 14.4 " 20,ocx)- 1 00,000 38.1 21. 1 12.6 " 5,000-20,000 37.4 20.9 10.7 Towns 2,000-5,000 37.3 23.9 8.2 Rural parts 39.7 31.2 6.2 The one exception to the rule stated above — that the per- centage of workers in the population increases with the size of the town — exists in the case of the rural districts and is due to the common practice of women working in the fields, as appears in the second column of the table. Few women act as servants in the country, but the number steadily in- creases as one progresses towards the great cities. A third question arises in this connection : What efifect does agglomeration have upon the industrial rank of the in- dividual? The following Austrian statistics would seem to indicate that agglomeration favors the upper ranks — em- ployers, independent workers, and the higher employees — at the expense of the lower — the artisans and day laborers :^ Percentage of persons occupied in Other places Places of Great cities. ' of 2,000+. 2,000 — . Independent^ 33.9 30.3 27.1 Salaried employees 15.8 8.6 1.3 Artisans 47.1 52.7 63.0 Day laborers 3.2 8.4 8.6 lOO.O loo.o loo.o 1 Rauchberg, St. Mon., xx, 391. ^ The seven principal cities; all but two, Briinn and Cracow, having more than 100,000 inhabitants. •'The German terms are Selbststandige, Angestellte, Arbeiter, Taglohner. " Angestellte " includes the superintendents, foremen, clerks, etc. — the higher personnel. 7 HE STRUCTURE OF CITY POPULATIONS 317 But this apparent advantage of the cities partly disappears on analysis into groups of industries, which shows that the high percentage of employers in cities is due to small under- takers in trade and commerce. In manufacturing industry, wherein the labor problem is more acute, the percentage of entrepreneurs is smallest in the cities, and of workmen larg- est, while the cities' advantage in case of the higher em- ployees is minimized : ^ Table CXL, Trade and Professions and Agriculture. Industry. Commerce. Civil Service. Large Ur- Large Ur- Largi e Ur- Large Ur- cities. ban. Rural. cities, ban. Rural, cities. ban. Rural. cities. ban. Rural. Independen t. 275 262 233 187 211 213 409 380 329 707 715 871 Employees. .29 42 25 16 7 2ZO 106 65 211 207 104 Artisans • • . 629 6i6 670 764 739 741 290 280 317 78 76 53 Laborers. . . ,. 67 118 95 24 34 39 8i 234 289 422 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 Thus it turns out that the growth of cities, like the growth of manufacturing, upon which it rests, favors the develop- ment of a body of artisans and factory workmen, as against the undertaker and employer. That the class of day labor- ers is relatively small in the cities is reason for rejoicing, though much may here depend on the methods of classifica- tion adopted by the census authorities. * Ibid,, XX, 394. CHAPTER VI. THE NATURAL MOVEMENT OF POPULATION IN CITY AND IN COUNTRY. In the preceding paragraphs, an analysis of city popula- tions at rest has been given — the static aspect ; it is now proposed to analyze city populations in motion — the dynamic aspect ; and the three great subjects to be discussed are marriages, births, and deaths. The following typical figures express the ratios to i,ooo of population, in Massa- chusetts, for the two years 1894 and 1895 • ' Table CXLI. Persons marrying. Births. Deaths. Boston 23.10 3I.24 23.23 Cities 100,000-50,000 18.89 29.72 19-49 " 50,000-25,000 18.08 • 29.00 18.0.3 " 25,000-10,000 15-92 27.57 16.68 Total urban 19.47 29.67 19.85 " rural 13,77 21.76 17.38 " State 17.68 27.19 19.07 It is clear that such a regular progression must have ac- countable causes. Let us analyze them. 1 28 th Annual Reft, of State Board of Health of Mass. for 1896, p. 826. The Summary of the Vital Statistics of the New England States for 1892 (p, 56) shows the relation of the urban and rural groups by comparing their rates with the New England rate as a standard of 1,000: Marriages. Births. Deaths. Urban pop 1,114 1,195 ^»°58 Rural " 886 805 943 (318) ]>ja tural mo vement of popula tion 319 I. MARRIAGES. The law of population is the most difficult subject fn social statistics, and statisticians are not even yet agreed upon a law or theory of population that has universal ap- plication. We may begin our researches in this field with the statistics of marriage, which naturally lead up to those of births. In Massachusetts, the number of persons marrying to 1,000 of the total population is smallest in rural districts, and, increasing with the magnitude of the dwelling-centre, attains its maximum in Boston. Other countries, too, have their largest marriage-rates in the urban populations. ^ But does this crude rate indicate a stronger tendency toward marriage in the city? Considering that the cities contain a relatively larger number of persons of marriageable age, they naturally ought to have more marriages ; ' and when the marriages are compared with the adult and unmarried rather than with the total population, the superiority of the city tends to disappear. 3 Only in appearance, concludes Levas- seur, is nuptialite greater in the city than in the country. * Thus, the Swedish statistics show an apparently higher mar- 1 For England, cf. J. of St. Society (1890), 53: 267. For France, Levasseur (ii, 77) gives the ratios of marriages, which are here doubled to show the ratios of persons married : i860. 1885. Department of Seine 19.8 16.6 Urban population 16.4 14.8 Rural " 15.4 14.6 France 15.8 14.8 " Mayo-Smith, Statistics and Sociology, 98. ' Persons marrying (1886) to 1,000 of the adult unmarried population of each sex (Levasseur, ii, 397) : Men over Women orer 20 years. 15 years. France 61.7 66.0 Paris 59.4 58.6 * Op. cit.y ii, 86. 320 THE GROWTH OF CITIES riage rate for Stockholm and the urban population through- out the century : ^ 1816-40. 1841-50. 1851-60. 1861-70. 1871-80. 1881-90. Rural 15-70 I4'56 15.CO 12.88 13.20 11.94 Urban 15-72 14-44 16.98 14.64 16.20 iS-34 Stockholm 16.34 14.62 18.44 15-48 18.14 17-72 But the city rate is increased by the relatively small num- ber of children in urban populations; hence when the number of marriages in 1881-90 is compared with the adult population, the ratios are transformed : Sweden. Stockholm. Marriages per 1,000 men of 20-50 years 68.9 59.8 " " " women of 20-45 years .. • 75-^ 57-° In England and America and perhaps other countries,^ however, the marriage rate remains higher in the cities even when based on the adult population ; ^ and Dr. Ogle put forth the hypothesis that the marriage-rate varies with the amount of employment among young women. He noticed that the marriage-rate among single men between the age of 20 and 45 was highest in the rural county of Bedford, which is particularly distinguished for its straw-plaiting and lace industries, occupying many young women. His hypothesis that " men are more ready to marry girls or young women who are themselves earning money " is substantiated by the statistics showing the proportion of women 15-25 years old ^ Swedish Census of 1890, Bihang, p. vii. (For title, see Table CXVI.) ''■ See Dr. F. S. Crum's essay, " The Marriage Rate in Massachusetts," Publica- tions of the American Statistical Association (Dec, 1895), ^^> 33^- ^^ ^^ ^°'' lowing, " adult " means a person over 15 years of age : Persons marrying, 1885. The 28 cities. Rural remainder. Per 1,000 of total population 19.2 15.0 " " " adult " 27.0 20.8 " " " " marriageable population iio.i 97-^4 NATURAL MOVEMENT OF POPULATION 321 industrially occupied. ^ And as a general rule this propor- tion is largest in the cities, with their factory operatives. This hypothesis also seems to afford an explanation of the high (refined) marriage-rate in the industrial towns of Massachusetts, where conditions are nearly similar to those in England. The cities with the highest marriage rates in the years 1 894 and 1 895 were : = Chicopee 27.3 Boston 23.1 Lawrence ^ 22.1 New Bedford 21.9 Lowell 21.4 Chelsea 20.6 Lynn 20.4 Fall River 20.2 Fitchburg 19.7 Now it will be seen at a glance that this list includes the principal textile cities of Massachusetts, /. e., those cities de- voted to an industry that employs a large percentage of female labor. And in the list of textile cities appearing in the State census of 1895, there are only three other cities mentioned — North Adams, Pittsfield and Taunton, all of which have other and probably more important manufactur- ing interests. Boston, although not a textile city, offers a vast amount of employment to women ; Lynn is devoted to the boot and shoe industry ; but Chelsea, the only remain- ing city unaccounted for, is a Boston suburb, and no appar- ent reason exists for its abnormally high marriage-rate. At the opposite extreme are the cities and towns of Woburn, whose marriage-rate was 13.9; Medford, 13.2; Marlborough 13. 1 ; Quincy, 12.8 and Peabody 12.8. Three of these 1 " On Marriage-Rates and Marriage- Ages," J. of St, Soc. (1890), 53 : 267. "^ 28ih Rep. of Mass. State Board of Health, 826, 222 THE GROWTH OF CITIES towns are devoted to the leather and boot and shoe indus- tries, while Medford and Quincy are suburbs of Boston. The comparative statistics now presented show wide dif- ferences in the marriage-rate among countries and cities. As a rule, the tendency toward marriage appears stronger in the rural districts than in the cities ; but industrial cities, and especially cities devoted to industries that employ female labor, are exceptions, marriage being there more frequent and at an earlier age than elsewhere. ^ When, however, we come to the question of family life, there is no doubt that the cities make the more unfavorable showing. From the relatively small number of children, or non- marriageable persons in the urban population, one would expect that its percentage of married persons would be con- siderably larger than the average. On the contrary, the sta- tistics show the cities have abnormally large percentages of the unmarried. This fact is illustrated in some very old (1830) Swedish statistics :=" Urban. Rural. Children 275 360 Single 367 223 Married 267 355 Widowed 91 62 1,000 1,000 It is thus found that the bachelors and spinsters as well as the widowed predominate in the cities, while the country has ^On the other hand, Parisians marry later than do other Frenchman. In 1885, the average age at marriage (Levasseur, ii, 81) was: Men. Women. Department of Seine. . 31 years, 9 months. 27 years, 5 months. Urban population .... 29 " 7 " 25 " 8 " Rural population 29 " 3 " 24 " 8 " *Legoyt, p. 38, where other data are presented similar to these. NA TURAL MO VEMENT OF POPULA TION 323 the larger percentage of children and the married. This is the relation which exists in the United States to-day; thus, the conjugal condition of the adult population (20 years and over) in 1890 was — ^ Males. Females. , ^ , , , The 28 great United The 28 great United cities. States. cities. States. Single 362 309.5 263 199 Married 590 638. 588 664 Widowed 42 46.5 145 132 Divorced 2 3 3 4 Unknown 4 3 I i 1,000 1,000 I, coo 1,000 The exclusion of children is necessary to statistical ac- curacy, as they are incapable of marriage. But it some- times happens that the cities, notwithstanding the large pro- portion of adults residing in them, still have fewer married people than do the rural districts. Such is the case in Austria:' Places of Single Less than 500 612 500-2,000 592 2,000-5,000 609 5,000-10,000 627 10,000-20,000 632 20,000 \- 629 Vienna 620 All places 608 Table CXLII. Widowed, divorced le. Married. or separated. Total. ' 335 53 1,000 ! 353 55 1,000 > 335 56 1,000 314 59 1,000 ! 309 59 1,000 • 305 66 1,000 311 69 1,000 336 56 1,000 In Europe generally, with the exception of England, ^ nth Cen., Pop., i, p. clxxxvi. * Rauchberg, St. Mon., xix, 135. 324 THE GROWTH OF CITIES the percentage of married people in the cities is below the average : ^ ENGLAND. Number of persons married in 1,000 of each sex between the ages of 15 and 45 years. (Census of 1891.): Male. Female. London 1 u ..• f 464 ^Sfi T 1- V urban counties ■{ cZ. \.. Lancashire J 1 469 475 Rutlandshire "I , .• f 388 427 ■tr r J u- y rural counties \^„ ' Herefordshire / 1 402 422 In the main, the statistics of conjugal condition agree with the statistics of annual marriages ; in both cases Eng- land's urban communities show a stronger tendency toward marriage and home life than do her rural communities, while on the continent the reverse is usually true. But in the United States, the singular result is reached of a rela- tively high urban marriage-rate and at the same time a relatively small proportion of married people in the cities. Inasmuch as the marriage-rate has been high in Massa- chusetts cities during a long period of time, one would naturally expect that the proportion of married people as revealed in the census would be large; but just as the 28 great cities of the country have fewer married people than the United States as a whole, so has Boston fewer than Massachusetts. ' ^ For France (Levasseur, ii, 390) : Population between ages of 20 and 60 years. Male. Female. Paris. France. Paris. France. Single 385 348 314 270 Married 570 609.5 5^6 640 Widowed 43 42 117 89.5 Divorced 2 .5 3-5 1,000 1,000.0 1,000 i,ooo.c ^Of the population 20 years old and over in 1890, the following percentages were married (^iitk Cen., Fop., i, 851, 888) : Males. Females. Massachusetts 61. i 55.8 Boston 55.3 50.5 Of the adult male population, 32.8 per cent, were single in Massachusetts, 38. i per cent, in Boston. NATURAL MOVEMENT OF POPULATION 325 A number of explanations may be offered for such an ap- parent contradiction. For one thing, rural emigration takes away most of the bachelors and maids, leaving in the country a population with a large proportion of married people ; and at the same time that marriages are compara- tively infrequent, social circumstances may be such as to impel rural couples to go to the cities for the performance of the marriage ceremony. Moreover, in many German cities it is found that city young people often remove to a suburb shortly after marriage in order to begin house-keep- ing in a cottage of their own ; ^ the marriage is thus credited to the city, while the census counts the married couple in the suburb. The most probable explanation, however, is that city marriages take place at an earlier age than country marriages, where the city marriage-rate is the higher of the two, and that they are dissolved sooner by the relatively high mortality to which males are subject in the city. This would account for the larger number of widows in urban populations. Divorce is also more frequent in the city. By the re-marriage of widowed and divorced persons, the city marriage-rate is raised, without any real addition to the number of married people, as compared with the rural com- munity where the first marriage would have continued longer. If the city of Copenhagen, for example, be compared with a Danish rural district (Fiinen), it will be found that Copen- hagen has the higher marriage-rate. But it is due simply to early marriages, for by the time the age of 45 years is reached, Copenhagen has relatively fewer married people than has Fiinen. The rural population marries later, but of all the males who reach the age of 45, and of all the females who reach the age of 35, more are married in the rural than in the urban population.^ Now, whether people marry early ' Briickner, Allg. Stat. Archiv, i, 662. ^ Rubin and Westergaard, Statistik der Ehen. 226 ^-^^ GROWTH OF CITIES or marry late depends chiefly on their foresight and self- control ; as a rule it is the higher social classes, the wealthier or at least the propertied classes, which possess these quali- ties. And in proportion as they predominate in a city, will the city's marriage-rate be low and its birth-rate also. With this brief indication of the theory of population, it may be reserved for more careful discussion in connection with birth- rates and the fecundity of marriage. The explanation advanced in the eleventh census for the relatively small number of married people in the urban pop- ulation is its peculiar composition as regards the native and foreign elements.'' Now as a matter of fact, the largest sin- gle element in the population of the 28 great cities consid- ered, is the foreign element, which contains a larger propor- tion of married people than any other element, and so far as race enters into the question, it is the native white element in which the greatest difTerence between city and country is found. A similar difference, explained very likely by the number of domestic servants in the city, is found among the negroes, who, however, constitute a much smaller element of the aggregate population of the 28 cities. On the whole, it appears to be the postponement of marriage on the part of native Americans that tends to reduce the proportion of the married in the city below that in the country. In Austria such a postponement of marriage in the cities is shown by an admirable compilation of statistics, which can be only partially reproduced here : * Pop., \, p. clxxxvi. The facts : Twenty-eight Great Cities. United States. Males. Per cent, married. Per cent, married. M. F. M. F. Foreign white i.357»779 67.3 62.7 65.9 68.1 Native white, native parents . . 780,947 57.1 58.0 66,1 67.9 Native white, foreign parents . . 656,053 45.6 54,0 48.6 58.8 Negro 128,145 59.5 51.9 69.0 65.0 Aggregate 2,952,238 59.0 58.8 63.8 66,3 NA TURAL MO VEMENT OF POPULA TION 327 Table CXLIII.i Number of women married among each 1,000 females of the age Places with a pop. of ii-ao. 21-30. 31-40. 41-50. 51-60. 61-70. 70+. All ages. Less than 500 .... 28 482 757 738 611 425 221 330 500-2,cxx) 47 566 798 748 598 405 207 347 2,000-5,000 45 532 744 694 552 378 197 328 5-10,000 32 465 696 655 531 353 176 311 10-20,000 23 431 686 643 513 333 164 305 20,000 -J- 19 349 621 618 490 309 139 298 Austria 36 495 745 714 578 392 199 331 Number of bachelors among each iooo males of the age Places with a population of 0-10. 11-20. 21-30. 31-40. 41-50. 51-60. 61-70. 70+. All ages. Less than 500 . 1,000 1,000 677 238 149 129 128 121 627 500-2,000 1,000 999 602 164 101 89 86 91 611 2,000-5,000... 1,000 1,000 645 207 140 123 113 loi 631 5,000-10,000 .. 1,000 1,000 734 257 172 135 127 108 656 10,000-20,000. 1,000 1,000 755 255 158 133 123 116 661 20,000+ 1,000 1,000 799 310 173 134 117 113 657 Austria 1,000 1,000 677 219 135 114 no 107 628 It thus appears that among the women, the married in every age group become relatively fewer as the size of the dwelling-place increases ; and the corresponding table, not here reproduced, shows the same thing for men, although in the advanced ages the differences between the various cate- gories of towns are much less marked than they are in the lower age classes. The second table shows that in the higher age classes the cities have no more unmarried men than the small towns, a result not irreconcilable with the foregoing statement, as the widowed and divorced, not the married, absorb the difference. Rauchberg's conclusion for Austria is of wide application : " The marriage-rate, the age at mar- riage, the preponderance of unmarried among the migrants and the direction of migration (toward the cities) work to- gether to raise the participation in married life, especially in the younger years, to a higher level in the villages and coun- try places than in the middle-sized and large cities." ' ^ St. Mon., xix, 136. ' Ibid., 137. 328 THE GROWTH OF CITIES The influence of migration upon marriage-rates is readily- seen. Rural emigration, being predominantly of young unmarried persons, leaves a large married population in the country. But it is doubtful if it materially diminishes the percentage of married in the city. Frankfort statistics (1890) afford the following example:^ Female population over 20 years of age Born Boni Frankforters. outside. Single 35.2 40.9 Married 47.5 47.0 Widowed and divorced 17.3 1 2. 1 ico.o 1 00.0 The young people who go from the village or farm to the city cannot, to be sure, marry at once, for they are obliged to obtain a position in the ranks of industry that will enable them to earn a living and support a family ; but this requires only a few years. Hence, as years pass, the percentage of the unmarried in the city steadily diminishes and approaches the rural percentage, as already noted in Table CXLIII. That the two percentages do not coincide seems to be due, not to the immigrants to the city, but to the native citizens themselves, who are less inclined to marry than the immi- grants, as appears in the following statistics from the Berlin census of 1885 : Per cent, married of all over "^ Male. Female. 20 yrs. 30 yrs. 20 yrs. 30 yrs. Berlin-born 55.6 77.6 49.6 57.8 Born elsewhere 58.8 79,5 53.8 62.6 Immigrants within preceding 5 years 30.7 66.9 37.0 51.9 Other immigrants 74.6 82.2 60.5 64.7 Comparing the city-born with those born outside, it will be seen that the proportion of married people among the 1 Bleicher, p. 7. » Allg. Si. Ar., i, 640. NATURAL MOVEMENT OF POPULATION 329 latter is 3 or 4 per cent, higher than among the former, and this notwithstanding the large number of immigrants of youthful age, among whom the percentage of the married is necessarily low. As a result the proportion of the married among the older immigrants is extremely high (74.6 and 60.5) as compared with the Berlin-born. Nor is the difference merely the result of difiference in age classification, for closer analysis shows that only in the lower age groups (below the age of 30) is the percentage of the married greater among the Berlin-born than among the immigrants.^ The latter, moreover, have a larger percen- tage of second marriages than the natives. The strong ten- dency of immigrants toward marriage is, perhaps, to be attributed to lack of prudence and foresight, as compared with the better-fed city people. Mention has just been made of the heavy percentages of city marriages dissolved by divorce. This does not contra- dict the statement in the Eleventh Census that the " propor- tion of divorced persons in the cities is less than that in the country at large ;"== for many persons who are divorced re- marry and are then no longer counted among the divorced. The evidence is conclusive that in Europe the number of divorces is three or four times as great in the cities as in the rural parts. 3 In France the peasantry, constituting one-half ' Bruckner, Ibid., i, 641, note i. ' Rep. on Pop., i, p. clxxxvii. As a matter of fact, the census enumeration of divorced persons is acknowledged to be extremely defective. *Cf. Korosi's Statisiiques des grandes Villes, and J. Eertillon's Ehide demogra- phique du divorce et de la separation de corps (p. 55), from which the follo^ving are taken : Divorces per 100 marriages in Brussels 1871-73 12.4 Belgium 3.5 La Hague.. 1865-74 11. i Netherlands 4.6 Munich ... 1868-74 15.3 Bavaria 5.0 Stockholm. 1864-73 28.1 Sweden 6.4 Copenhagen. 1 87C-74 29.2 Denmark 12.6 330 THE GROWTH OF CITIES the population, contributes only 7 per cent, of the divorces. The number of divorces (per 100,000 inhabitants) in France was as follows in 1885 : Department of the Seine (Paris), 47 ; urban population, 19; rural population, 3.5/ In the United States the evidence against the cities is less conclusive, but it justifies Hon. Carroll D. Wright in stating " that not only is the increase of divorce proportionately greater in the cities than in the country, but the ratio of married couples to each divorce is also greater."^ Any marked divergence from this rule may partly be due, as Professor Willcox suggests,^ to the presence of a foreign-born Catholic population in the cities, and partly to the more approximate equality of the rural and urban populations in this country as regards the reception of new ideas than is the case in Europe with its rural population composed of peasants. II. FECUNDITY. The fruitfulness of marriage is commonly expressed in the ratio of annual births to each 1,000 of total population, — the birth-rate. In former times the cities had a considerably lower birth-rate than the country, which was explained by the statisticians as the result of late marriages and limitations on the size of the family. " More wants," says Siissmilch in I76i,"and increased splendor, with higher prices for the necessaries of life, keep men from marrying in the cities." * ^Levasseur, ii, 91. ' Report, as Commissioner of Labor, on Marriage and Divorce in the United States, 1867-86, p. 162. Cf. tables on pp. 159-161. * The Divorce Problem, Columbia Studies, i, 33. Professor Willcox also notes that the divorce rate in the District of Columbia (an urban population) is nearly three times that of the adjacent commonwealths — a ratio which approaches European conditions. (^Ibid., 39.) ^ Die gottliche Ordnung (2d ed.), i, 257. Siissmilch's statistics of births in city and country are not very satisfactory; perhaps the rates 42 (i birth to 24 living) and 36 (i: 28) for the rural and urban populations, respectively, would express his opinion, if reduced to figures. (Cf. i, 225.) He estimates the num- NATURAL MOVEMENT OF POPULATION 331 In another epoch-making work on population statistics^ a century later, a countryman of Siissmilch's could present statistics showing that, in most European countries, the ur- ban birth-rate exceeded the rural. Only Sweden and Prus- sia were exceptions, and as indicated in Table CXVI, Sweden ceased to be an exception at the very time Wappaus was writing. At the present day the crude birth-rate is almost universally higher in city than in country,'' and, indeed, as shown in the Massachusetts statistics given in the introduc- tory treatment of the movement of population, the birth-rate increases with the populousness of cities. The most obvious explanation of a high birth-rate would be a large proportion of women in the child-bearing period. The cities have a larger percentage of such persons ; hence for this reason, and not because of greater fecundity of city women, do the cities often have a high birth-rate. The following table shows, in the ideal or normal rate, that the fecundity of women should be greatest in the cities on ac- ber of children to a marriage at 3.5 in the city and 4in the whole country (i, 175). The birth rate in large cities fluctuated enormously in previous centuries because of frequent epidemics. ^ Die allgemeine Bevolkerungsstaiistikhy yE. Wappaus (Leipzig, 1861), ii,48i. ^ In Prussia there is still some difference between city and country : * Urban. Rural. 1849-55 37-98 40.6 1856-61 38.1 40.9 1862-67 39'0 41.2 1867-86 39.3 40.4 25 great cities Prus.sia. Germany. of Germany. 1861-64 40-8 •••• 36.9 1864-67 40.4 .... 38.7 1867-71 38.5 .... 38.3 1871-75 414 414 41-9 1875-80 40.9 40.8 42.1 18SC-85 38.9 38.5 37.4 * Bruckner, Allg. Siat. Arckiv, i, 160-1. 332 THE GROWTH OF CITIES count of the favorable age- distribution; following the crude, or uncorrected, birth-rate, is the true rate, — for both legiti- mate and illegitimate births (1890-91) :^ Table CXLIV. To 100 women aged 16-50. Births to 1,000 inhabitants. AH Married: Single, widowed, 4 * 1 women: legiti- etc.: illegiti- Normal. Actual. total births. mate births. mate births. 16 great cities 42.4 35.9 12. i 22.8 2.6 Cities 20-100,000 39.4 37.9 14.5 26.7 2.2 Cities under 20,000 .. . 37.3 35.6 14.4 26.3 2.2 All cities 38.9 36.3 13.5 25.1 2.4 Rural 38.2 40.0 16.8 28.9 2.5 Prussia 38.5 38.5 15.4 27.4 2.5 These figures plainly show that fecundity is greatest among the women of the rural districts of Prussia, and stead- ily decreases as the dwelling-centre becomes more popu- lous, — except that the middle-sized cities show no decrease over the smaller cities. Illegitimate births, it may be noticed incidentally, are relatively more numerous in the country than in the city, although the larger proportion of maidens and widows in the city would lead us to expect the contrary. In the great cities, indeed, the percentage of ille- gitimates increases, but here there is the additional factor of maternity hospitals and clinics which receive a great many women from outside, — the majority of such women, it is un- necessary to add, being unmarried. The Prussian statistics hardly justify us in making the ^ The computations for the ideal or normal rate were made by Dr. Bleicher of Frankfort, and will be found in his paper on "Die Eigenthiimlichkeiten der stadtischen Natalitats und Mortalitatsverhaltnisse," in the report of the proceed- ing of the 8th International Congress of Hygiene and Demography at Budapest (1894), vii, 468. The other statistics appear in his tables, and are also to be found in Bleicher, 267. NATURAL MOVEMENT OF POPULATION 333 generalization that city marriages are less fruitful than country marriages. Indeed, the opposite is true in several countries, if the great cities be excepted. Thus, in Denmark the average annual number of births (1880-9) to 1,000 women, aged 16-50, married and unmarried, respectively, was:^ Legitimate. Illegitimate. Copenhagen 234.1 48.5 Other towns 259.6 23.1 Rural communes 245.9 24.3 Denmark 246.3 28,6 Additional examples from European statistics might be given,^ but the most pertinent one is found at home. Dr. F. S. Crum has shown that in Massachusetts the fecundity of marriage increases with the density of the population, and reaches its maximum in the largest city: 3 ' Rubin, " Population, Natalite et Mortalite du Royaume de Danemark," Int. Cong, of Hygiene, 1894, Proceedings, vii, 489. ^ In Saxony the number of births to 100 women of child-bearing age (1879-83) in the five government districts is as follows {^Zeitschrift des K. Sachsischen Stat. Bureaus, 1885) : Legitimate. Illegitimate. (Rural 28.0 5.9 Urban 28.2 4.2 Dresden city 21.9 4.6 (Rural 30 4 6.9 Urban 31.9 8.2 Leipzig city 28.3 3.6 Zittau {^^"^^1 ^9.6 7.3 *- Urban 20.8 5.0 Bautzen j ^'^'^^^ ^^-S 6.7 I Urban 23.9 3.4 {Rural 32.8 9.4 Urban 26.3 8.5 Chemnitz city 26.0 4.8 '"The Birth-Rate in Massachusetts," Quar. Jour, of Economics, xi, 259. 334 THE GROWTH OF CITIES Table CXLV. Deaths of chil- No. of surviv- Percentage of dren aged o-i year, ors at one increase (or Births per i,ooo married Per i,ooo Per i,ooo year per loo decrease — ) Pop. per No. of women. total living married in pop., sq. mile. towns. Total. Native. Foreign, deaths. births. women. 1875-85. 0-24 39 59-7 57-1 64-5 97-4 106.6 5.34 —13.9 25-49 64 62.9 58.2 89.1 103.7 121. 5 5.53 —13-3 50-99 103 74.1 61.5 117. 9 I40-3 126.6 6.45 +2.2 100-199 44 83.0 67.2 iig.o 148.9 122.5 7-28 +6.3 300-499 53 106.3 87-6 133-2 193-1 135-3 9-19 -fiS-9 500-999 22 106.0 84.1 146.3 194-3 136.3 9.13 +37-0 J, 000-4,999 ... . 19 120.0 86.0 153.0 249.7 175-3 9-91 +29.5 5,000 + 4 124.0 gg.o 139.0 229.9 185.8 10.10 +17.0 In the first place, it should be observed that the corres- pondence between density and populousness is close. The four cities in the densest group are Boston, and its three principal suburbs, Cambridge, Somerville, Chelsea; and all the thirteen cities which, in 1885, contained 25,000+ inhabi- tants are found in the last two groups of the table. Roughly speaking, the last group signifies Boston, the second densest group, the industrial cities. In the second place, the table shows that the high birth- rate of the large cities is not solely due to difference in race, as has often been asserted ; for native women, as well as foreign, are more productive in the cities, the only irregu- larity in the progression being in the group of towns with a density of 200-500. In the third place, it is to be remarked that the high birth- rate of Massachusetts cities is not accounted for, as is some- times urged,' by the fact that " in the cities there is a larger proportion of population between the ages of fourteen and fifty." Calculations by the present writer show that whereas the number of births to 1,000 women aged 18-45 years was about no in Massachusetts in 1894-5, it was about 125 in Boston.'' True, the Boston women marry younger than do ^Annals of American Acad,, v, 87. ' It might be objected that the percentages given above do not distinguish the native and foreign mothers, and that the predominance of the latter gives Boston NATURAL MOVEMENT OF POPULATION 335 the women of Massachusetts generally, and Boston conse- quently has a larger proportion of its married women in the age-group 20-30 years, the most fertile period of woman's life, than Massachusetts has ; the percentages being 27 and 24.1 respectively.^ But such a slight difference cannot alone explain the relatively high birth-rate of Boston. The evidence as to relative fecundity of city and country women is therefore conflicting. Heretofore the European statisticians have explained the relatively high birth-rate (crude) in cities as a result either of favorable age distribu- tion or of more illegitimacy. Wappaus, for example, showed that the number of children to a marriage was almost invari- ably higher in country than in the city; but he found a larger number of illegitimate children in the cities, which fact, in connection with the more frequent marriages in the city, gave the latter a higher birth-rate.^ At the present its high rate. The probabilities are against this; in Boston 70.5 per cent, of the ioreign and 73.7 per cent, of the native married women are under 45 ; in the rest of Massachusetts the percentages are 66.0 and 64.6 respectively {^iith Cens., Pop., i, 851, 888). 'Calculated from utk Cens., Pop., i, pp. 851, 888. In Prussia the respective percentages for cities of 20,000+ and rural districts are 23.82 and 20.35 ! '" Bavariat for town and country, 21.94 'ind 15.49 (Cf. Zeitschrift des K'dnigl. Bayerischen Statis. Bureaus, 1892, p. 309. - By deducting the number of deaths of infants under 5 years, Wappaus was able to show a still larger fruitfulness of marriage among countrymen. A portion of his table is reproduced herewith (Cf. AllgemeineBevolkerungsstatistik, 11,483-4) : Table CXLVI. Children to a Same at end of Percentage of illegitimate marriage. fifth year. births in total. City. Country. City. Country. City. Country. France 3.16 3.28 2.03 2.34 15.13 4.24 Netherlands 3.91 4.32 2.49 3.07 7.71 2.84 Belgium 3.80 4.17 ... ... 14-49 5-88 Sweden 2.99 4.19 1.83 3.16 27.44 7.50 Denmark 3.04 3.34 2.14 2.58 16.05 10.06 Saxony 4.60 4.13 2.77 2.64 15-39 14.64 Prussia 4.00 4.44 2.56 3.13 9.80 6.60 (Saxony 4.30 4.22 2.59 2.81 15-34 "'-58) Saxony is the only exception to the general rule, and Wappaus accounts for it 336 THE GROWTH OF CITIES day it is not so generally true that the larger families are found in the rural districts. In France, indeed, the proposi- tion remains true, as the following comparison (from Levas- seur ii, 398), indicates: Families having (1886) Dcpt. of Seine. France. child 323 2CO 1 child 276 244 2 children 201 218 3 " 105 145 4 " 53 90 5 or more children 12 103 1,000 1,000 But in the United States, on the other hand, the average number of persons to a family (4.93) is smaller than it is in the 28 great cities of 1890 (4.99). The difference is really greater than appears, because the United States average is raised by the large families of negroes in the South, which contains only four of the 28 great cities.^ Comparing the ten leading cities with the commonwealths to which they belong, it will be found that the difference is nearly always in favor of the city : =" by the fact that, in this industrial kingdom, manufacturing industries had spread from the towns into the open country and villages. If the population be divided into industrial and agricultural groups, instead of city and country, the result will agree with the other countries. (See the last line of the table.) ' Cf . the following : North Atlantic States 4.69 North Central " 4.86 Western States 4.88 South Atlantic States 5.25 South Central " 5.30 "^ nth Cens., Pop., vol. i, p. cucff. The census interpretation of the decline in the average size of a family is open to criticism. NA TURAL MO VEMENT OF FOPULA TION -^^-i^'j City. State. New York 4.84 4.59 Chicago 4.99 4,92 Philadelphia 5,10 4.95 Brooklyn 4.72 4.59 St. Louis 4.92 5.07 Boston 5.00 4.67 Baltimore 5.01 5.16 San Francisco 5.69 4.92 Cincinnati 4.67 4.68 Cleveland 4,93 4.68 The statistics of families, however, are not absolutely trust- worthy information regarding the fruitfulness of marriage. The census definition of family is necessarily loose, and in- cludes the inmates of an hotel, an asylum, etc. ; and many other factors have to be considered. A possible explanation is the presence of so many foreigners in the great cities, amounting to 32 per cent, of their total population as against 1 5 per cent, for the United States. That the foreigners have larger families than the Americans is well-known, and is demonstrated anew by the Massachusetts statistics just re- ferred to. But these latter statistics show that the native, as well as the foreign married women, have more children in the cities than in the country. In fact, there seems to be no direct connection between agglomeration and fecundity. In Berlin, Leipzig and Mun- ich, three of the four leading cities of Germany, the refined birth-rate is below the average of the twenty-six great cities, while in the smallest six cities of this class the birth-rate is above the average. And among the cities of equal size, even in the same country, notable differences in the birth-rate are familiar; factory towns generally having a high birth- rate without any advantage in the age distribution. The conditions affecting the fruitfulness of marriage are so nume- rous and complicated that statisticians and social philoso- phers are still in dispute as to their relative influence. The 338 THE GROWTH OF CITIES theory which at present commands the largest body of ad- herents is the one set forth in Professor Fetter's Versuch zu einer Bevolkerungslehre, which, it is hoped, may be put by the author into more easily accessible form for American readers. Substantially the same theory, however, is pro- pounded in Professor Hadley's new work. According to this theory, the birth-rate (as the expression of fecundity) is not dependent on the peculiarities of population groups, nor even the general economic condition of the population ; but both the rate of increase and the economic condition depend on a third factor, — the economic foresight and prudence of the individual. " High comfort and low birth-rate are com- monly associated, because comfort is made to depend upon prudence. Let the comfort be made independent of pru- dence, as in the case of the pauper or criminal, and the birth- rate tends to increase rather than diminish . . , It is not that social ambition in zV.?^//" constitutes a greater preventive check to population than the need of subsistence ; but that the need of subsistence is felt by all men alike, emotional as well as intellectual, while social ambition stamps the man or the race that possesses it as having reached the level of in- tellectual morality. Ethical selection can therefore operate on the latter class as it does not on the former. The intel- lectual man has possibilities of self-restraint which the emo- tional man has not."^ One of the statistical proofs that the birth-rate diminishes with each advance in civilization was furnished by Miss J. L. Brownell in a study of the American statistics in the Tenth Census. Miss Brownell took as statistical indices of civiliza- tion the percentage of all deaths (from known causes) that were due to nervous diseases, the density of population, the intensity of the cultivation of the soil as indicated by the ^ Hadley, Economics, pp. 48-9. NA TURAL MO VEMENT OF POPULA TION 339 value of agricultural products per acre, and the value of manufactured products per capita. In comparing the forty- seven states and territories, Miss Brownell found that a low birth-rate generally accompanied high percentages of the other factors, and vice versa. Her table ^ may be summarized as follows : Value of agri- Value of mf. Deaths from ner- cultural pro- products per vous diseases. Density. ducts per acre. capita. Coherences with birth-rate 8 8 16 7 Opposition to birth-rate 39 39 31 40 Total States and Territories 47 47 47 47 In order to ascertain the relation of the birth-rate to agglomeration of population, Miss Brownell's comparison has been extended by the writer. In the following list the plus or minus mark signifies that the rate or percentage for the State specified is above or below the average for the United States; the asterisk simply denotes coherences:" Table CXLVII. Density of Percentages of Birth-rate. pop- pop- urban. 1 . Alabama -\- — — 2. Arizona — — * — * 3. Arkansas -|- — — 4. California — — * -j- 5. Colorado — — * -|- 6. Connecticut — -|- -|- 7. Dakota -f — — 8. Delaware — + -|- 9. Dist. of Col — -f- 4- 10. Florida -|- — — 11. Georgia 4- — — 1 2. Idaho + — — 13. Illinois -)- -f -j- ^ Annals Amer. Acad., v, 74. 'The percentage of urban population for the United States in 1880 was 21.47, the line being drawn at towns of io,ooo-f . The authority is iiih Cen., Stat, of Cities, p. I. 340 THE GROWTH OF CITIES Birth-rate. 14. Indiana — 15. Iowa + 1 6. Kansas + 1 7. Kentucky + iS. Louisiana -\- 19. Maine — 20. Maryland — 21. Massachusetts — 22. Michigan — 23. Minnesota \- 24. Mississippi + 25. Missouri + 26. Montana ■\- 27. Nebraska + 28. Nevada — 29. New Hampshire — 30. New Jersey — 31. New Mexico + 32. New York — 33. North Carolina + 34. Ohio — 35. Oregon -)- 36. Pennsylvania — 37. Rhode Island — 38. South Carohna + 39. Tennessee + 40. Texas + 41. Utah + 42. Vermont — 43. Virginia -j- 44. Washington + 45. West Virginia -j- 46. Wisconsin -\- 47. Wyoming -}- Density of Percentages of pop. pop. urban. + ♦ +* +■• + + + + + —■ + —■ -1- + + + + + + + + + +* • — +* + __ +* — In 39 out of 47 cases, a high birth-rate is opposed to the concentration of population ; where a large proportion of the population is in cities, the birth-rate is low, and vice versa. The States in which there are coherences are Arizona, Indi- ana, Maine, Michigan, Nevada, New Hampshire, and Ver- NA TURAL MO VEMENT OF POPULA TION 341 mont, with both rates below the average of the United States ; and Louisiana, with both rates above the average. The theory of population herein adopted is, in brief, that a declining birth-rate accompanies an advancing civilization.^ In so far as cities represent the highest culture and comfort of a country, just so far will they have a low birth-rate and families below the average size. Now, in the past, the classes devoted to manufactures and to commerce have had radically different standards of living. Under the factory system a man marries early because, with female and child-labor in demand, his family soon becomes a help rather than a bur- den. We have but lately seen that in both England and the United States (Massachusetts), marriage is most frequent in towns where women can find employment. Dr. Ernst Engel was probably the first statistician to ad- vance statistical data in favor of the proposition that it is chiefly the occupation rather than the mere association of people in large or small dwelling centres which causes the difference in fertility of city and country women.^ In a thorough analysis of births in various groups of the popula- tion of Saxony, for the decade 1840-49, he demonstrated that the birth-rate in towns 91 to 100 per cent, of whose ' Additional statistical proofs of this theory consist in compilations showing that the birth-rate in large cities diminishes as one goes from the poor to the rich quarters, and that the age at marriage increases, and size of family diminishes, as one passes from the classes low in the social scale to the responsible mercantile and professional classes. Cf. Fetter, op. cit.; Levasseur, ii, 398, and iii, 218; Charles Booth on the birth-rate in London districts, your, of St. Soc, 1893; crit- icism by R H. Hooker, ibid., Ix, 753; Brownell, " The Significance of a Decreas- ing Birth-Rate," in Annals of the Amer. Acad, of Pol. and Soc. Sc, v (with references); Billings, "The Diminishing Birth-Rate in the United States," Forum, June, 1893; Edson, "American Life and Physical Deterioration," in North Amer. Review, Oct., 1893. A. very good summary is given by Prof. Mar- shall in the third edition of his Principles of Economics, p. 263 ff. And see especially Tallqvist, La Tendance a une moindre Feconditi des Marriages, and Rubin- Westergaard, Statistik der Ehen. ^ Cf. Wappaus, op. cit., ii, 481, and Table CXLVI, supra. 342 THE GROWTH OF CITIES population was devoted to agricultural industry was 30 per 1000, and in towns where 91 to 100 per cent, of the popula- tioo was engaged in manufacturing and commerce, 48.4 per 1000.^ English statisticians, moreover, long since pointed out the high birth-rate peculiar to mining and industrial districts.^ In Germany, the cities which indisputably have the largest proportion of births to child-bearing women, are the purely industrial cities of the Rhine-Westphalian dis- trict.3 On the other hand, the commercial cities, with their greater wealth, comfort and culture, have the lowest birth- rate. What effect has migration cityward exercised on the urban ' Statis, Mitteilungen aus d, Konigr. Sachsen, herausgeg. vom Statis. Bureau des Minist. des Innern, Bewegung der Bevolkerung, etc., in d. yahren, 1834-jo, Dresden, 1852, Introd., pp. 20, 56. (Also privately published under the XK'iS.Q Die Bewegung der Bevolkerung im Konigr, Sachsen, etc.: ein Beitrag zur Pkysio- logie der Bevolkerung, vom E. Engel, Dresden, 1854.) .... Taking the entire population, the standard birth-rate of Saxony was 41.0; of the towns which were predominantly agricultural, 38.8; of the towns predominantly industrial and com- mercial, 42.2. ^Cf. Newsholme, Vital Statistics, 57. ' Bruckner, Allg. Stat. Archiv, i, 1 62. Bruckner's comparisons are based on the crude birth-rate, but inasmuch as the Rhine cities have a small stream of im- migration, it is fair to infer that the proportion of adult women is not unduly large, as indeed proves to be the case. Bruckner's study covers, for most of the German Grossstddte, the period 1861-85, and his grouping of the cities according to birth-rate is copied below. Those familiar with German industrial conditions will recognize the identity of the first group with the factory industries, that of the third with commerce : High birth-rate. Medium birth-rate. Lov/ birth-rate. Chemnitz. Danzig. Strassburg. Barmen. Breslau. Dresden. Elberfeld. Cologne. Hanover. Krefeld. Halle. Konigsberg. Altona, Dusseldorf. Bremen. Aachen. Stuttgart. Stettin. Magdeburg. Nuremberg. Leipzig. Berlin. Hamburg. Frankfort. Munich. NATURAL MOVEMENT OF POPULATION 343 birth-rate? In many countries the greater fecundity of women in industrial centres has been explained by the fact that, one or two generations back, the workmen came from rural districts, and naturally from the largest families, not all of whom could be provided for out of the family estate. In other words, the urban immigrants were children of produc- tive women, and fecundity is an hereditary quality.' Capt. John Graunt's authority was cited in support of the theory ; but when he speaks of "breeders " migrating to London he seems to be referring rather to women of marriageable age. In the United States, especially in New England, the high birth-rate of industrial centres is largely due to the fertility of the women of French Canadian and Irish stock ; but aside from the element of race, the determining influence on the birth-rate is social and psychological, rather than physiologi- cal. Whatever their capabilities, the rural emigrants have small families when they have once attained a certain stand- ard of life adherent to the higher social classes. III. DEATH RATES. The discussion of mortality is not so beset with statistical difficulties as are the subjects of births and marriages ; for it is almost everywhere true that people die more rapidly in cities than in rural districts. The statistics of deaths already incidentally introduced do not, therefore, require augmenta- tion, for they are typical of all countries. To render clearer the relations of population-centres to mortality, the following ratios have been computed from The Vital Statis- tics of New England ( 1892, p. 57), which are regarded as equal to the European statistics in accuracy : ' This is urged with force by Ziegler, Die Naturwhsenschaft und die socialde- mokratische Theorie, 147. 344 ^^^ GROWTH OF CITIES Ratios to the New England rate taken as ioo: Cities. I00,O0O-(- Il6 50,000-100,000 1 10 25,000-50,000 105 10,000-25,000 95 Total urban 106 " rural 94 The death-rate is the lowest in the rural parts and steadily increases with the size of the city. Further, it is to be ob- served that these crude rates are too favorable to the cities in that they do not take into account the larger proportionate number of people of healthy ages residing in the cities, as set forth in a preceding paragraph on age classification. The following is a comparison between the actual death- rates with the rates that would prevail in Prussia, if all groups of population had the same mortality at the same ages : Table CXLVIII. Deaths per 1,000 population. In 1890-91.' With uniform mortality Prussia 23.5 23.5 Rural 23.4 26.7 Urban 23.6 24.4 Cities under 20,000 24.2 25.9 " of 20,000-100,000 23.5 23.5 " " 100,000+ 22.8 22.6 That is to say, the age constitution of the rural popula- tion is such that if the same mortality prevailed at each age as prevails in the city, its death-rate would be much higher (26.7) than the urban rate (24.4) ; whereas, actually, the rural rate is the lower of the two (23.4 as against 23.6). The smaller cities make a good showing, but the large cities (ioo,ooo-f ) have a higher death-rate than they should have. These statistics controvert the long-current dictum that mortality always increases in the same ratio with density of ' Bleicher, 268-9. NATURAL MOVEMENT OF POPULATION 345 population, and the still more refined rule of Dr. Farr's that the mortality of a district is approximately equal to the sixth root of its density/ But the only satisfactory method of contrasting urban and rural mortality is to compare the mor- tality at different ages. In the United States such a com- parison results as follows:^ ' This hypothesis never gained general acknowledgment, and in recent reports of the Registrar-General of England it is abandoned. But there still exists in the United States and France, as well as in England, a close connection between mortality and density. Thus the Supplemettt to the ^^th Annual Report of the Registrar- General (Part i, p. xlvii), divides the population into 15 groups, in- creasing in density from 138 to 19,584 persons to the square mile, and in mor- tality from 14.75 '^^ 3°'70 (crude death-rates; the corrected rates being respec- tively 12,7 and 33.0) The American Statistics (^iith Cens., Soc. Stat, of Cities, p. 7) : No. of Pop. per Persons to each Death- Cities of cities. acre. dwelling. rate. 10-15,000 41 2.43 5.45 17.86 15-25,000 39 2.79 5.85 19.45 25-50,000 40 4.67 6.06 21.81 50-100,000 25 9.04 6.28 22.43 ioo,ooo-|- 28 I5.I5 7'64 23.28 Total 173 8.73 7.05 22.62 The French Statistics {^Statistique sanitaire des villes de France, 1886-90) : Cities. Deaths per 1,000. Under 5,000 20.91 5,000-10,000 21.58 10,000-20,000 25.80 20,000-100,000 25.75 100,000-400,000 26.65 Paris 23.69 France 22.21 * Exclusive of still-births. Source, nth Cens., Rep. on Vital and Social Sta- tistics, 'Pt.i,^!'^. 17-19. The registration States are New Hampshire, Vermont, Massachusetts, Rhode Island, Connecticut, New York, New Jersey, Delaware and District of Columbia. Cities of 5,0004- population in these nine districts are classed as urban, the remainder of the population as rural. The statistics of the great cities (28 in number) are less trustworthy, as their registration systems are still often defective. The metropolitan district includes the counties of New York, Kings, Queens, Richmond, Westchester (in New York), Hudson and Essex (in New jersey), and the cities of Paterson and Passaic (^op. cit., appendix). 346 ^^^ GROWTH OF CITIES Table CXLIX. Dfaths per i,ooo population at specified ages: Registration States. Cities of Metropolitan Rural. Urban. ioo,ooO"r. District (N. Y.). All ages 15-34 22.15 21.62 24.61 Under i year 121.21 243.32 236.81 264.35 Under 5 years 37-12 80.40 78.00 89.25 5-15 years 4.03 6.21 5.96 6.16 15-45 years 6.89 10.80 10.71 12.07 45-65 5 ears ^S-^9 26.27 26.62 3"-52 65 years+ 67.83 88.60 89.76 96.62 Unknown 54-98 20.65 20.76 '^A-l- In the United States, therefore, the mortality is heavier in the city than in the country in every period of life; and as a rule it increases in severity in the same ratio as the magni- tude of the city. Only in the age-period 5-15 years is the mortality less in the metropolitan district than in the regis- tration cities, and nowhere does it approach the more favor- able rate of the rural districts. What significance has the heavy urban mortality to the average citizen? It means that whereas the average person born in Massachusetts may expect to live 41.49 years, the average person born in Boston may expect to live only 34.89 years ;^ that while 426 out of 1000 men born in Prussia sur- vive to the age of 50 years, only 318 out of 1000 native Ber- liners reach the same age '^^ that while the mean age at death is 42 years 2 months in France, it is but 28 years and 19 days in Paris; 3 that while the average duration of life in the rural population of the Netherlands is 38.12 years, in the urban population it is only 30,31 years.'* The contrast be- 1 op. ciL, 484-5. ■^ Levasseur, ii, 312. •■^Turquan, "La vie moyenne en France," Revue Scientifique, 24 Dec., 1892, pp. 812, 817. Cf. Lagneau, Essai de statistique anthropologique sur la popula- tion parisienne. * Bevolkingsiafeln, etc., voorhetKonigr. der Nederlanden, Staat A, pp. 388, 391 . NATURAL MOVEMENT OF POPULATION 347 tween what is and what might be the healthfulness of city people, is strikingly shown by recent computations of the Registrar-General of England, who has grouped together the districts with low mortality (14-15 deaths per 1000 popula- tion) under the term of " selected healthy districts." They embrace about one-sixth of the population of England and Wales, and are chiefly rural, Manchester township, on the other hand, is taken as a fair representative of the urban and industrial population. On the basis of mortality in 1881-90, it is figured that the expectation of life at birth is as follows : ^ Selected healthy districts 51-48 years. Manchester township 28.78 " England and Wales 43.66 " That is to say, a person born in one of the " selected healthy districts" of England may expect to live, on the average, nearly twice as many years as a person born in urban Manchester ! To show the social waste involved in such heavy mortality, it is sufficient to point to the fact that 100,000 males born in Manchester would be reduced to 62,326, and 100,000 females to 66,325, in five years; while in the healthy districts it would take fifty and forty-eight years, respectively, to bring about the same reduction." Clearly, the concentration of population produces an enor- mous drain on the vitality of a people. * Supplement to t/ie ^jt/i Annual Report, Part ii, p. cvii. - Ibid. Comparisons of the duration of life and expectation of life, based on life tables of urban and rural populations, are now becoming abundant. They may be said to have begun with the English Friendly Societies, which furnished statistics for rural, town and city districts. Cf. Radcliffe, Observations on the Rate of Mortality a7td Sickness existing among Friendly Societies (Manchester, 1850, Colchester, 1862, Sup. Rep., 1872); F. G. P. Neison, Contributions to Vital Sta- tistics \{\%']^'). On the subject of longevity, compare Levasseur (vol. ii), von Mayr, Bevolker ungsstatistik, and Mayo-Smith, Statistics and Sociology. 348 THE GROWTH OF CITIES But it does not follow that the case is hopeless for the cities and hence for society in general, which is destined to see an increasing number of its members concentrated in cities. There is no inherent and eternal reason why men should die faster in large communities than in small hamlets, provided they are not too ignorant, too stupid, or too selfishly individualistic to cooperate in the securing of common bene- fits. In some degree, doubtless, the mortality of city adults must exceed that of rural adults, on account of the danger- ous nature of city occupations ; much of this occupational mortality is irremediable, but it should no more be charged up against the city than the mortality in railway accidents should be charged against the country. In each case the mortality is the price paid for progress ; we might secure relief by abandoning both railways and machinery, and re- turning to the economic system of previous centuries. But leaving aside accidental causes, it may be afifirmed that the excessive urban mortality is due to lack of pure air, water and sunlight, together with uncleanly habits of life in- duced thereby.' Part cause, part effect, poverty often ac- companies uncleanliness : poverty, overcrowding, high rate of mortality, are usually found together in city tenements." Even though indigence be not carried to the point of starva- tion, it has a decided effect on the production of efifluvial poisons as well as on the tendency to disease of every kind. ' As Dr. Farr once said {Sup. to 2j(h Annual Rep. of Reg.- Gen., p. xxxiii-v), *' there can be no doubt that mere proximity of the dwellings of people does not necessarily involve a high rate of mortality. When any zymotic matter, such as varioline, scarlatinine or typhine finds its way into a village or street, it is more likely to pass from house to house than it is when the people are brought less fre- quently into contact. The exhalations in the air are thicker. But if an adequate water supply, and sufficient arrangements for drainage and cleansing are secured, as they can be by combinations in towns, the evils which now make dense dis- tricts so fatal might be mitigated. Indeed, some of the dense districts of cities are at the present day comparatively salubrious." '■'Cf. Report of New York Tenement House Com., 1894, pp. 433-4- NATURAL MOVEMENT OF POPULATION 349 Thus a report of the medical officer of health for the London County Council shows that mortality (a) increases in the same ratio as the proportion of population living more than two in a room in tenements of less than five rooms (b) — a. b. 17.51 under 1 5 per cent, 19.51 15-20 " 20.27 20-25 " 21.76 25-30 23-96 30-35 25-07 35-40 Still more striking was the result of the Berlin inquiry of 1885: a. 73,000 persons living in families in tenement of i room. b. 382,000 " " " " " " " 2 rooms. c. 432,000 «' " " " " " " 3 " d. 398,000 " " " " " " " 4 " Class (a) suppHed nearly one-half of the total deaths, although it constituted only six per cent, of the population. While government cannot cure poverty, it can remedy some of the results of poverty, and that is what the government of Berlin has done in the enactment of wise building laws, which have abolished the insanitary conditions that led to such frightful mortality.^ That cleanliness and healthfulness may co-exist with indi- gence is shown by the example of the tenth ward in New York city. For it is not only by far the most densely popu- ^ The death-rate in the first ward of New York in tenements where there were front and rear houses on the same lot was 61.97, while it was 29.03 in houses, of the same ward, standing singly on a lot. (Report of Ten. House Com. of 18^4, p. ^■^.') In an English city the death-rate was 37.3 in districts which contained 50 per cent, of the back-to back houses, as against 26.1 in districts containing no such houses. In Glasgow the death-rate was 27.74 for families living in one and two rooms; 19.45 ^^^ those in three and four rooms, and 11.23 ^^r those in five or more rooms. (Newsholme, Vital Statistia, 140, 155.) 350 THE GROWTH OF CITIES lated ward in New York both as regards number of inhabi- tants to the acre and number of tenants to the house,^ but it also contained a large number of the rear tenements ^ so scathingly denounced as death-traps by the Tenement House Committee of 1894. Yet, notwithstanding these conditions, the tenth ward had the extremely low death-rate of 17.143 and was surpassed in healthfulness by only two wards (one a business and the other a suburban district) out of the entire twenty-four wards in the city.'^ This favorable death-rate was not the result of superior economic conditions ; on the con- trary, the population of this ward consists almost entirely of Russian and Polish Jews, who are among the poorest classes of the city. Nor was it the result of a favorable age consti- tution, for the tenth ward swarms with children. We may, indeed, exclude the adults entirely, and we shall still find that the death-rate in the tenth ward is more favorable than that in all but three other wards of the city : the number of deaths of children under five years of age to 1,000 living of that age in the tenth ward was 58.32, a remarkable showing when the rates in the other wards, running as high as 183, are taken into consideration.^ What, then, is the explanation of the handsome record made by the tenth ward in the face of its unfavorable conditions? There is but one answer: its peo- ple are careful in the observance of sanitary laws. Being ^ The density per acre was, by the census of 1890, 543; that of New York city, 59. (^Social Stat, of Cities, 11.) The density of the tenement house population of New York was 103, that of the tenth ward, 622 (^Report of the Tenement House Committee 0/18(^4, pp. 23-4). The average number of tenants to a house was 57.2 in the tenth ward, 34 in the entire city {Ibid., p. 25). ' Cf. Table A, p. 274 of the Report. ^ The statistics all refer to the year 1893. * Ibid., p. 25. ''Ibid., Table B, p. 278. The average rate for the whole city was 76.6. The rate for London in 1891 was 66.4, in England, 59, and in the rural county of Hereford, 39.2. (^Registrar-General's Report, p. xliii.) NA TURAL MO VEMENT OF POPULA TION 3 5 i Hebrews, they observe the strict Mosaic laws regarding cleanhness, the cooking of food, habits of eating, drinking, etc. This illustration is sufficient to show that the most difficult problem connected with the health of cities is not incapable of solution. Personal cleanliness in the home has become more and more a subject of legislation in the way of strin- gent building laws'" and proceedings for the condemnation of old tenements. All the great cities now have laws regulat- ing the proportion of the building-lot that must remain un- built {i. e. the amount of court space), the height of buildings, their construction, the size of rooms, height of ceilings, sani- tary appliances and methods of artificial lighting. Where such regulations cannot be enforced on account of the age and situation of the buildings, the buildings have been con- demned and demolished by public authority, as instance the British Housing of the Working-Classes Act of 1890 (53 and 54 Victoria, ch. 70, sec. 21) which provided for the destruc- tion of the fearfully unhealthy "back-to-back" houses, and the New York Law of 1895 (chapter 567, sec. 7) which provided for the condemnation of the equally insanitary rear tenements and other disease-breeding structures. The length to which sanitary legislation may go in order to pre- vent the city from becoming a "pest-house" is to be ob- served in the measures taken against overcrowding. Thus, the New York Law of 1895 (ch. 567, sec. 10) makes it manda- tory upon the Board of Health of New York city to see that at least 400 cubic feet of air shall be afforded to each adult and 200 to each child occupying a room in tenement houses with insufficient ventilation. The recent movement toward ' According to the New York Tenement House Committee of 1894 (^Report, p. 62), the New York building laws, as regards tenement-houses and dwellings for the poor, are " superior to any that prevail elsewhere." But they are not strictly enforced. 352 THE GROWTH OF CITIES the building of model tenement houses, which has assumed large proportions in New York city, is another step in the direction of preventing overcrowding. The possibilities of progress toward healthfulness which are afiforded by such im- provements in the housing of the people, are demonstrated by the mortality statistics of the Peabody tenements in Lon- don. In these buildings population is very dense, — 75 1 to the acre as compared with 58 for all London, — and the age distribution is less favorable to low mortality than is that of the city at large. But the death-rate in the Peabody build- ings (1886-89) was 18.6 per 1000, as compared with 19.6 for London, 25.6 for the central districts, and 26.9 for the eastern districts of London.^ Even more encouraging is the reduction in infant mortality (deaths of infants under one year to 1,000 births) : 1887. 1889. 1891. 1893. Peabody buildings 141 127 134 126 London 188 141 154 164 Central districts 175 151 177 181 Eastern districts 172 146 161 175 While in other groups of population the infant mortality increased from 1889 on, it slightly diminished in the model tenements. The housing of the working classes is therefore one of the most urgent duties of cities to themselves. It has often been objected to the policy of demolition and reconstruction that new slums are created as fast as the old ones are abolished. The Royal Commission on the Housing of the Working Classes in 1884 reported that "where the demolitions are so extensive that the people have to depart, then new slums are created." Since workmen must and will live near their work, the immediate efTect of any consider- able demolition of their dwellings is the overcrowding of the 'Newsholme, "Tlie Vital Statistics of Artisan Block-Dwellings," in Proceedings of the 8th Inter. Cong, of Hygiene, Budapest, 1S94, vii, 430. NA TURAL MO VEMENT OF POPULA TION 353 neighboring houses ; and it is only by the enforcement of stringent building and sanitary laws that the city can prevent the formation of new slums. The demolition of old build- ings needs to be done with judgment, since it is often bet- ter policy to wait for the pressure of high rents to compel the factory to remove to the suburbs, taking the workmen with it. To the erection of model tenements in the suburbs no posssible objection can be offered, and this is unquestion- ably the direction private philanthropy should take.^ There are other ways in which government may provide material remedies for personal and family uncleanliness en- gendered by the conditions of life in tenements ; such as, for example, public baths, or still better, cheap water rates that will encourage house-owners to put bath-rooms in the tene- ments. Public baths is the European policy,'' but the Amer- ican policy tends rather toward cheap water rates, as is notably the case in Buffalo. As for the rest, education must be trusted to teach the city-poor proper sanitary habits. Education is no doubt a process both long and toilsome ; but it is withal a hopeful process and forms the basis of modern democracy. Outside of private dwellings of the people, the scope of private and municipal activity is less restricted. In the matter of securing light and air, it is now admitted that the provision of numerous small parks or recreation piers as ' The question of the housing of the working classes has given rise to an abun- dant literature, references to which may be found in the encyclopedias of political science {e. g., Art. " Arbeiterwohnungen," in Sch5nberg's Handbuck der Pol. Economie ; Art. " Wohnungsfrage," in Conrad's Hdwbh.') An excellent sum- mary of the legislation together with a clear statement of the problem, is given in Bowmaker's little book, The Housing of the Working Classes, which also con- tains a bibliography; see also The Housing of the Working People, by Dr. E. R. L. Gould (8th Special Report of the U. S. Commissioner of Labor). ■' Twenty-five parishes in London maintain 31 bathing establishments. Cf. Hartwell, in Bui. of the Dept. of Labor, July, 1897 : " Public Baths in Europe." 354 THE GROWTH OF CITIES breathing spaces is not a luxury, but a matter of absolute necessity to the city. Of equal or greater importance is the provision of children's play-grounds on an adequate scale. People having once realized that city life is to be the per- manent lot of a majority of the inhabitants of civilized coun- tries, it becomes the undoubted policy of their governments to make the city healthful. In line with this policy there has grown up a vast administrative system, unknown to our rural ancestors, which is charged with the inspection of workshops and the conditions under which clothing is man- ufactured ; with the inspection of food and the prevention of adulteration, etc. Perhaps the most important single factor in the reduction of the death-rate in New York city has been the reduction of infant mortality brought about by a rigid inspection of the milk supply and the encouragement of the use of sterilized milk."^ As will shortly appear, the widest divergence of rural and urban mortality occurs in infancy; hence all measures designed to affect infant mortality favor- ably are to be energetically promoted. The medical inspec- tion of children in the pubhc schools and the care taken to isolate contagious diseases are also factors in the reduction of the city death-rate.^ Finally, the modern Babylons are endeavoring to secure health for their residents by improv- ing methods of communication between their business cen- tres and residential outskirts. The London County Council, as Dr. Albert Shaw has observed, is giving more attention, if possible, to efforts to induce the railroads to improve their suburban morning and evening train service for workingmen, and to develop all parts of London's underground, surface and suburban transit systems, than it is to dealing directly ' Reports of the Board of Health of New York City, passim. ^ The decrease in the New York death-rate from contagious diseases for the first half of the year was 3.17 per cent, in 1894, 2.23 in 1895, 2.07 in 1896, and 1.49 in 1897. NATURAL MOVEMENT OF POPULATION 355 with the housing question/ This subject will engage discus- sion in a subsequent chapter. Considering the effort devoted to improving the healthful- ness of cities, one would naturally expect a diminution of their death-rates in recent years, and of course such a dimin- ution is familiar to all. In a preceding chapter a table was given showing the progress made in the capital and other cities of Sweden. In England the gradual approximation of the urban to the rural death-rate is shown in the following percentages from the Registrar-General's Reports of 1891 (p. Ivii) and 1893 : Urban Rural Deaths in town to sanitary districts. loo in country. 1851-60 24.7 19.9 124 1S61-70 24.8 19.7 126 1871-80 23.1 19.0 122 1S81-9O 20.3 17.3 117 1893 20.2 17.4 116 Like the Swedish rate, these are the crude death-rates, in the calculation of which age and sex are neglected ; but it is legitimate to use crude rates in comparison as regards time, since in any given place the age-distribution is not subject to great change. Compare these rates with those given by Sussmilch^ more than a century ago, rates which continued until the middle of the present century : Rural 25 Small towns 31 Large cities 36 Capital cities 40-|- Within a period of twenty-five years, London reduced its death-rate from 50 to 25, thereby increasing the average 'Testimony before the N. Y. Tenement House Com. of 1894, in Rtport,"^. 373. '^ Die g'dtil. Ordnung, 1761, i, 79-91. 356 THE GROWTH OF CITIES length of life from 25 to 37 years. New York city has also made some progress in the last 40 years : ^ Deaths per i,ooo. 1856-65 32.19 1866-75 29.77 1876-85 26.32 1886-95 25.18 1896 21.52 In the introduction of enlightened sanitary methods we should expect the largest cities to lead the way. How effect- ually they have met their obligations may be ascertained by a review of the mortality statistics of the Germanic countries. Within a century Vienna has reduced her death-rate from 60 in the 1000 to 23 (1886-90); the urban rate (50-53 cities), following Vienna's, was 24, while for the entire state the rate was 29 ."" Allowing for differences in age distribu- tion, it remains true that the cities of Austria are more healthful than the rural districts. This is indicated by the relatively low infant mortality in Vienna and the other cities ; it being 208 per 1000 hving births in Vienna, 227 in 53 cities and from 243 to 260 in the entire state.3 Still more instructive are the Prussian statistics. The fol- lowing table covers the years 1890-91, but for the sake of comparison the mortality of each group of towns for 1880-81 ' Report of the Board of Health for year ending Dec. ji, i8g6, p. 14. " Von Juraschek, " Die Sterblichkeit in den Oesterreichischen Stadten," Pro- ceedings of the Eighth Inter. Cong, of Hygiene, Budapest, 1894, vii, 491. ' Loc. cit., p. 502. The urban mortality in Austria can hardly be called favor- able; it is rather that the rural rate is so unfavorable. Still it demonstrates the fact that the cities in Austria can no longer be called " destroyers " of population, but are rather showing the rural districts the path toward healthfulness. In Hun- gary a similar condition exists, so far as we can judge from imperfect data. Dr. Thuroczy calculated the death-rates in 3,284 towns of various sizes, and found the highest rate (47) in the smallest places, and a steady decrease as the towns be- came larger, until a population of 40,000 was reached (27.9). After that, the rate increased slightly. ((?/. cit., vii, 222.) NA TURAL MO VEMENT OF POPULA TION 357 has been added and for the 16 great cities the mortality in each age period :^ Table CL. Deaths per iooo in each age group. 16 largest cities. Cities 20,000- Cities under Urban Rural 1880-1. 1890-1. 100,000. 20,000. communes, communes. Prussia. °"^5 53-5 41-6 37.6 33.7 37.2 32.3 34.0 ^S-20 4.9 3.9 4.5 4.6 4.4 4.5 4.5 20-30 7-7 5-8 6.S 7.0 6.4 6.2 6.3 30-40 12-3 9-8 10.8 10.1 10.2 8.1 9.0 40-50 17.2 15.2 16.2 14.5 15.2 11.4 12.9 50-60 26.3 24.4 26.3 23.7 24.5 20.8 22.1 60-70 46.9 46.8 50.2 48.7 48.S 46.7 47.3 70-80 99.4 99.1 107.3 103.7 103.3 105.7 104.9 80 -r ^. ... 210.5 221.8 229.3 227.1 226.2 230.9 229.2 All ages, 1890-91. 22.8 23.5 24.2 23.6 23.4 23.5 " 1880-81. 27.5 ... 26.1 25.8 26.4 24.6 25.2 Observe, first, that the crude death-rate has decreased in the rural population only from 24.6 in 1880 to 23.4 in 1890, while the urban rate diminished in the same period from 26.4 to 23.6, and the rate of cities with population of 100,000 and upwards from 27.5 to 22.8. Obviously, it is the class of great cities that made the best showing. The inference is confirmed by an examination of the death-rate in the various age- groups, from which it appears that the great cities of Prussia have a lower death-rate than the smaller cities at every age-period except 0-15 and 40-60 years ; and lower than the rural population itself at the ages of 15-30 and 70+. In an extended comparison of mortality by age-classes in the several Grossstadte of Prussia and the provinces in which they are situated, it has been pointed out that among males at least one-half of the cities have as low a rate of mortality as the provinces, at the ages of 4 and 5 years, and nearly all in the period of 20-25 years; but in the later ages the cities are much more unhealthful. On the other hand, the female mortality is less in the majority of the cities ^ Bleicher, 268-9. 2^8 ^^^ GROWTH OF CITIES at almost every age after the first three years of childhood. On the whole, the cities make the worst showing in the age- period 1-3 years, and the best in years 5-25.^ Somewhat similar results are yielded by the statistics of other countries. Thus, the only age-period in which the male mortality is less in the urban than in the rural popula- tion of Denmark is 10-15," while Copenhagen stands above the urban population for the age-classes 5-25. As regards the mortality of females, on the other hand, the urban popu- lation leads the rural in the years 10-25, ^"<^ Copenhagen has the advantage of both from 10 to 35 years, as well as in old age. In Bavaria, again, the cities make the better show- ing during the age-periods 10-15 ^"^^ 21-30 among males, and 21-40 among females.3 In England, the cities have the advantage of a lower rate of mortality than the rural districts during the ages 15-35 for females alone. Taking the national death-rate at each age as 100, the death-rates of selected locaUties will be expressed in the following figures :■♦ Table CLI. I. Ratios of male death-rates. 0-. 5-. 15-. 25- 35-- 4S-- S5-- ^S-- London ill 103 92 106 113 115 113 106 Pleasure places 77 75 90 105 105 Seaside places 78 78 108 ll i 98 Manchester 139 141 124 136 152 Dockyard towns 104 104 iii iii 109 Staffordshire potteries. 134 loi 98 lOi 125 Lancashire mfs 123 116 100 104 109 North collier districts. 109 128 118 89 87 Rural districts (450) . 74 83 91 88 79 ^Kuczynski, 231. "^ And the oldest ages. See Rubin, vol. vii, p. 490, of the Proceedings of the Eighth Inter. Cong, of Hygiene and Demography, Budapest, 1894. ' Kuczynski, 214 ff. * Jour, of Stat. Soc. (1897), 60 : 65. 93 91 98 90 82 85 157 153 144 109 96 95 148 144 132 120 130 130 91 lOI 106 76 78 86 NATURAL MOVEMENT OF POPULATION 359 2. Ratios of female death-rates. t>-. S-. IS-. 2S-. 3S-. 45-. SS-. 6s-. London 113 100 78 89 102 108 104 loi Pleasure places 74 80 75 78 84 87 87 91 Seaside places 76 83 83 84 82 82 79 82 Manchester 141 137 iii 126 143 158 159 145 Dockyard towns 105 106 ico 99 100 102 93 97 Staffordshire potteries. 131 94 iii 119 118 118 128 129 Lancashire mf. dist . . 121 109 in 112 no 122 137 133 North collier districts. 112 126 124 118 no 106 107 in Rural districts (450) . 72 87 105 96 86 80 82 88 These statistics will throw some light on the causes of the high rate of mortality in towns, — whether town-made, due to occupation, or brought about by the migratory movement. First, it is to be noted that the rural death-rate is below the average for England at every age for male and every age except 15-25 for female. But the urban rates (London, Manchester, Stafifordshire and Lancashire) are almost invar- iably higher than the national rates, although in London the rates of female mortality are very little above the national average except during the first five years of age. The pres- ence of so many domestic servants is sufficient to explain this, since the male mortality- rate is not so favorable in Lon- don, nor the female in manufacturing and mining districts. Generally speaking, the urban rate is most unfavorable in infancy, Manchester being an exception. But Manchester's excessively high rate of adult mortality is not representative even of the manufacturing cities of Northern England. There are many factors other than conditions of residence which cause the differences in mortality heretofore noted ; such, for example, as the presence of many female servants in London. Hospitals and similar institutions also disturb the rates. Occupation of course exercises a very consider- able influence, which may be seen in the high death-rates of Manchester and the Staffordshire potteries. According to Dr. Ogle's investigation, the lowest mortality was among the 36o THE GROWTH OF CITIES clergymen, and the highest among hotel servants, while coal miners had a lower mortality than commercial clerks, and the street hawkers, who lead a healthy outdoor life, had about the same mortality as the men engaged in the lung-destroying pottery manufacture.^ It is obvious that statistics of occupa- tion and mortality will aid us little. Dr. Arlidge, who has written the most authoritative book on the subject, virtually confesses that it is impossible to determine how far diseases are trade-made, and how far town-made.^ Some influence on urban mortality is doubtless exercised by those trades in which men are subject to injuries arising from dust fumes, or from contact with poisonous substances, or from handling heavy tools and machinery. It is doubtful if indoor or sedentary labor, of itself, produces a high death-rate ; social position and economic condition, on the other hand, seem to have more influence. That the high urban death-rate is not primarily due to the nature of the city trades may be inferred from the fact that the widest difference between urban and rural mortality exists among children under the age of five years. In other words, the high urban death-rate is primar- ily a result of high infant mortality. In the average com- ^ Taking loo as the standard, the comparative mortality in certain trades was as follows : Clergymen loo Printers 193 Farmers 114 Glassworkers 214 Paper-makers 1 29 Cutlers, scissors makers .... 229 Lawyers 152 File-makers 300 Coal miners 160 Costermongers, street sellers. 308 Bakers 172 Earthenware makers 314 Commercial clerks 1 79 Hotel servants 397 Railway and road laborers ... 1 85 Additional figures are given and discussed at some length by Professor Mayo- Smith in Statistics and Sociology, pp. 165-7. ' Hygiene, Diseases and Mortality of Occupations, p. 33. NATURAL MOVEMENT OF POPULATION 361 munity about one-fifth of all deaths are children under one year of age.^ Not only in England, but in Prussia (r/. Table CXXXIII, ante), France and the United States (Massachusetts, cf. Table CXLV), do the cities show a higher rate of infant mor- tality than the country. But the situation is not irremedi- able. Not all the evils are of a permanent nature; for example, the employment of married women in factories.' In addition to the lack of proper sanitation already de- scribed, the essential reason of excessive infant mortality in cities is poor nourishment. This was proved beyond ques- tion by the classical investigation of Boeckh in Berlin. A supply of pure milk, fed through non-rubber tubes, is abso- lutely necessary where wet-nurses are lacking; even with proper precautions the infant death-rate always increases in the summer-time when the mothers Work out, and it is so much more difficult to obtain sterilized milk. That the cities ' Levasseur ii, 164. The fact will be made plainer than it has yet been made by the following figures showing how many persons out of 1,000 at the beginning of a quinquennial period will on the average survive to its end (1881-90) : * England Manchester Healthy and Wales. township. districts. 0-5 751 623 827 5- 976 , 952 982 10- 990 970 989 15- 981 967 983 20- 974 951 977 25- 965 928 971 30- 956 903 966 35- 946 880 961 40- 933 854 954 After the age of 5 years is reached, Manchester remains for a time subject to only a slightly greater mortality than the healthy districts, although the difference increases again after the age of 35. * Demonstrated by Miss Collet, " The Extent and Effects of Industrial Employ- ment of Women," Jour, of Stat. Soc. (June, 1898), 61 : 219-60. ■ Sup. to jjth Annual Pep. of Registrar- General, Part ii, p. cxi. 362 THE GROWTH OF CITIES will ultimately learn how to deal with this problem, is indi- cated by the successful efforts of several European countries. As has already been pointed out, the Austrian cities have a lower rate of infant mortality than the rural districts ; and the following statistics show that Austria does not stand alone :^ Deaths of infants under one year to i,ooo living births: Countries. Cities. Germany, 200 193 and 206 cities in 1889-92 237.0 Belgium, 166 70 cities in 1889-92 187.5 France, 166 Cities of 25,0004- in 1891-92 i70-7 England,- 145 28 and 33 cities (80,000+) in 1891-92. 163.4 Austria, 254 57 cities (i2,0GO-|-) in 1889-92 238.6 Switzerland, 194 15 of the large cities in 1890-92 157-5 Netherlands, 203 12 large cities in 1891-92 I95'0 In Bavaria, as was recently pointed out by Kuczynski,* the cities have been able to present a lower rate than the rural districts since 1882, with the exception of the years 1886 and 1893. The most favorable showing was in the provinces containing the largest cities, Munich being especially distin- guished for a low rate. In Bavaria, too, the cities have a lower rate of infant mortality among the illegitimate children, than have the rural districts. Elsewhere, the fact that the city illegitimate children form a larger proportion of the total number of children than is the case in the rural districts,* often accounts for the more unfavorable general rate of infant mortality ; since illegitimate children are everywhere subject to a heavy mortality. In Saxony the urban rate of infant mortality is elevated by a high rate in industrial Chemnitz ; nevertheless, the large ' Silbergleit, " Kindersterblichkeit in Europ. Grossstadten," Proceedings ofBuda~ pest Cong, of Demography, vii, 445. *In 1881-90 the rural rate was 128 as compared with an urban (78 cities) rate of 160. — Suppl. to Jjih Rep. of Registrar- General, Pt. i, p. Ixviii. •'P. 199. * Not necessarily a larger number of illegitimates in proportion to the number of unmarried women. NATURAL MOVEMENT OF POPULATION 363 cities make a more favorable showing than do the small cities.^ The reduction of the rates in Prussian cities in the last decade is noticeable :^ 1880-1. I 890-1. Prussia 208.3 205.2 Rural ' I95-0 195-8 Urban 233.2 221. i In 16 largest cities 267.0 241.7 In all cities 20,000-100,000 224.4 214.5 " " " 20,000 216.3 208.5 The infant mortality of illegitimate children increased slightly in the rural communes, while it diminished in all the cities, most of all in the great cities. In discussing remedies for a high rate of infant mortality, it is well to bear in mind that the real excess in cities does not occur in the first week after birth or even in the first month. It really begins with the second month and reaches its maximum in the sixth month.3 In Prussia, indeed, an in- ' Zeitschrift des kdnigl. Sachs. Statis. Bureaus, xxxiv, 16; xl, 4, 1 1, 12 : Table CLII. Average annual number of deaths of children under one year to 100 living births: Cities of 1881-85. 1886-90. 100,000-f inhabitants 25.9 25.7 20,000-100,000 31. 1 28.8 10,000-20,000 30.5 31.2 5,000-10,000 30.1 29.0 3,000-5,000 29.6 30.7 2,000-3,000 28.6 28.6 1,000-2,000 \ 2^^ 24.8 Under 1,000 > 26.I All cities 28.55 28.05 ^Bleicher, 267. * English experience. The Registrar-General (^Sup. ^jth Rep., Pt. ii, p. cix) makes a comparison between three cities (Preston, Leicester and Blackburn) which have a high rate of infant mortality, and three rural counties (Hereford- 3^4 THE GROWTH OF CITIES fant born in the city has a better chance of living, during the first 15 or 30 days, than has a country child ; of 100 children dying in first year after birth the following number died in the first 1 5 days : ^ Male. Female. Prussia 21.6 19.6 Rural 23.5 21.4 Urban 18.6 16.7 Cities under 20,000 ; 20.1 18.0 " 20,000-100,000 18.3 16.6 " 100,000 \- 1 6.4 14.8 The geographical distribution of infant mortality in Prussia hardly confirms the current opinion that a high birth-rate must be accompained by a high rate of infant mortality ; the Rhine cities have the former without the latter. On the other hand, in Massachusetts, as we saw in Table CXLV, the rate of infant mortality moves hand in hand with the birth- rate. But the infant mortality is probaby as much cause as shire, Wiltshire, Dorsetshire) which have a low rate. Taking the latter's rates as 100, the following figures express the urban rates : Age in days. Ratio. Age in months. Ratio. o 120 o 127 I 164 I 221 2 123 2.... 301 3 102 3 308 4 95 4 303 5 109 5 373 6 136 6 337 Weeks. 7 279 o 123 8 325 1 164 9 292 183 10 278 197 " 27s p. ex. — "In the first week of life the town rate exceeds the rural rate by 23^ ; 2dwk., 64%; 3dwk., 83%; 4thwk., 97^, First month, 27% ; 2d mo., 121%; 6th mo., 273% (max.). 1 Bleicher, Proceedittgs of Budapest Congress, vii, 477. NA T URAL MO VEMENT OF POPULA TION 365 efifect ; by shortening the period of suckling and diminishing the intervals of child-bearing, it leaves a vacancy to be filled by another birth/ It has been aflfirmed that the decrease in the death-rate, so far as it comes from a reduction of infant mortality, is no economic benefit because it merely preserves unsound bodies and minds. The fallacy of such reasoning has frequently been pointed out^ and need not be repeated here. The smaller the infant mortality, the heavier must be the mortality in the latter years ; for men must die sooner or later. The crucial point is whether the lives saved will be extended to old age or will be lost in earlier adult years as the result of weak constitutions. Dr. Ogle maintained the latter, in which event the extension of life could hardly be regarded as an unmitigated blessing. But his statistics 3 are now superseded ; in the last decade, 1881-90, the English death-rate decreased for both sexes at every age-period except 65-75, clearly in- dicating a greater length of life. And the Prussian statis- tics already presented point to the same conclusion : mor- tality diminished at every age except among those over 80 years old. One question still remains, — ^What effect, if any, have migratory movements between city and country upon the death-rates of each? We found that as a rule the urban rates were most favorable at the age 15-35 and the rural rates then most unfavorable. Many people regard this simply as a result of migration. Mr. Thomas A. Welton in an article on "Local Death-Rates in England" (from which the statistics of Table CLI are derived) assumed at the outset that if the death-rates in the age-periods 5—45, and especially 15-35, ^ Newsholme, Vital Statistics, 57. * Mayo-Smith, Statistics and Sociology, 179-80. Cf. Bleicher, 274. ' According to Dr. Ogle, the death-rate for males over 35 and for females over 45 increased between 1838-54 and 1871-81. (Cf. Mayo-Smith, op. «V,, p. 178.) 366 THE GROWTH OF CITIES were depressed in London and other places attracting workers, and simultaneously raised above the national pro- portion in the districts supplying the bulk of such workers, his case would be made out. It was easy to prove the facts, but the assumption cannot be granted. It is quite as likely that the low urban rate at the age-period 15-35 is due to superior nourishment as to the immigration of healthy per- sons from the country. Ammon has compiled statistics tend- ing to show that city workers are better fed than country workers.^ At any rate it is a reasonable hypothesis. In Frankfort, Germany, an attempt was made (1890-1) to compare the mortality of citizens born in the country and those born in the city itself, with the following result : " Table CLIII. Death-rate of Frankforters born In the city. Outside. 0-5 72-3 68.5 5-10 10.2 1 1.8 10-15 3.8 5-2 15-20 4-3 3-5 20-30 6.5 5.7 30-40 8.4 9.8 40-50 13-5 16.3 50-60 22.4 27.0 60-70 43.8 53.2 70-80 109.5 ^05-3 8o-|- 224.0 251.8 All ages 25 o 14.5 Excluding children under 5 years.. . . 13.0 13.7 Clearly, the influx of countrymen does not reduce the city death-rate among adults; for the outsiders have a heavier mortality at every age except 0-5, 15-30, and 70-80. '^ Die Gesdhchaftsordnung und ihre natiirlichen Grundlagen, p. 117; Die naturliche Auslese beim Menscken, 1 23, 1 70. ' Bleicher, II Heft, p. 24. NA TUBAL MO VEMENT OF POPULA TION oQy Whether they are stronger than the city-born or not, cannot be easily determined ; but they certainly endure less per- tinaciously the heavy draughts on vitality made by city-life, than do the city-born. What John Graunt wrote 200 years ago seems true to-day : " As for unhealthiness, it may well be supposed that although seasoned bodies may and do live near as long in London as elsewhere, yet newcomers and children do not." ^ To recapitulate : The tendency to marry and the fruitful- ness of marriage are but slightly affected by the concentra- tion of population. The theory of population now accepted makes economic and social position the determining factors, rather than the degree of density of population. Hence marriage and birth rates dififer in cities of the same magnitude according to the prevailing industry and occupation. Death-rates, however, vary with the degree of agglomera- tion of population. But there is no inherent reason for the relatively high urban mortality except man's neglect and indifiference. Recent tendencies show that the great cities are leading the way in making sanitary improvements, and in several countries, of which Bavaria is an excellent ex- ample, the large cities now make a more favorable showing as to mortality than do the other communities. This holds true even of infant mortality, which is one of the most decisive indices of a locality's healthfulness, ^ Bills of Mortality, p. 90. CHAPTER VII THE PHYSICAL AND MORAL HEALTH OF CITY AND COUNTRY The field opened up in the study of death-rates is the most important and one of the most interesting yet encountered. That the townsman on the average is shorter-lived than the countryman is incontrovertibly established ; and it is com- monly believed that the city man is also less healthy, vigorous and capable, both physically and mentally, than the country- man. In short, cities are the site, and city life the cause, of the deterioration of the race.' The severest indictment is drawn by Nordau, the apostle of Degeneration : " The inhabitant of a large town, even the richest who is surrounded with the greatest luxury, is continually exposed to the unfavorable in- fluences which diminish his vital powers far more than what is inevitable. He breathes an atmosphere charged with or- ganic detritus ; he eats stale, contaminated, adulterated food ; he feels himself in a state of constant nervous excitement^ and one can compare him without exaggeration to the in- habitant of a marshy district. The children of large towns who are not carried ofif at an early age suffer from the pecul- iar arrested development which Morel has ascertained in the population of fever districts. They develop more or less normally until they are 14 or 15 years of age, are up to that ' Such is the concurrence of opinion among all the writers of former genera- tions. Says Rousseau {^Entile, 1819, vi, 6i) : " Les villes sont le gouffre de I'es- p6ce humaine. Au bout de quelques generations, les races perissent ou degener- ent," Similarly, Henry George {Social Problems, 317) : "This life of the great cities is not the natural life of man. He must under such conditions deteriorate physically, mentally, morally." (368) PHYSICAL AND MORAL CONDITIONS 369 time alert, sometimes brilliantly endowed, and give the high- est promise. Then suddenly there is a standstill. The mind loses its facility of comprehension and the boy, who only yesterday was a model scholar, becomes an obtuse, clumsy dunce, who can only be steered with the greatest difficulty through his examinations. With these mental changes, bodily modifications go hand in hand."^ And it is in the re- markable growth of great cities that Nordau finds the explana- tion of the equally striking increase in the number of degen- erates in the last half century.^ Nordau, however, is an extremist, whose opinion many people regard as too pessimistic, not to say fanciful. Let us therefore quote the testimony of a sane, conservative English physician, — Dr. G. B. Longstafif, one of the best- known statisticians in England : " That the town life is not as healthy as the country is a proposition that cannot be contradicted. . . . The narrow chest, the pale face, the weak eyes, the bad teeth, of the town-bred child are but too often apparent. It is easy to take an exaggerated view either way, but the broad facts are evident enough ; long life in towns is accompanied by more or less degeneration of race. The great military powers of the continent know this well enough, and it may be surmised that with them agricultural protection is but a device to keep up the supply of country- bred recruits. "3 Finally, the theory of city degeneracy is met with in the proverb that one cannot find a London cockney whose father was born in the city, and in the oft-quoted assertion that no business houses can be found in the city whose members have resided in the same city more than one or two '^Degeneration, Trans, from 2d Ger. Ed. (N. Y., 1895), p. 35. 2/<5iV.,p.36. ' your, of Stat. Society, 1893, p. 416. 370 THE GROWTH OF CITIES generations.^ The belief that city families die out early is widespread, and is expressed in nearly all literatures. Its most impressive statement has been formulated by Dr. Georg Hansen in his oft-abused and oft-praised work, Die Drei Bevolkerungsstufen,^ which is essentially an argument for the preservation of a peasantry or agricultural class, not only as a military measure, but as the fundamental condition of national vigor and well-being. Hansen's argument for the superiority of country-bred people embraces a consider- able number of propositions that require critical examina- tion, the principal ones being the following : (i) The city-born reside in the poorest quarters of the city; the country-born in the wealthiest (p. 147). (2) The city-born predominate in the lowest occupations and the lowest social classes (p. 150). (3) The city-born contribute an unduly large proportion to the class of degenerates (criminals, lunatics, suicides, etc., pages 196-202). (4) The cities have a low rate of natural increase, often indeed a deficit of births (p. 28). (5) The city population always consists of at least as many country-born as city-born (p. 27). (6) The typical city class, the middle class or bourgeoisie , is incapable of self-perpetuation (p. 27). The fact that the poorest districts of great cities often ^ Thus Cantlie, author of Degeneration among Londoners, " after prolonged and careful search could not find a single person whose ancestors, from their grandparents downwards, had been born and bred in London." — Strahan, Mar- riage and Disease, p. 31. Dr. Pratt, in a paper before the American Social Science Association in 1887, accepted the statement as authoritative. Cf. also Booth, Life and Labor of the People, iii, 65 : "There is a strong conviction in the minds of many, incapable however of strict verification, that Londoners tend to die out after the second or, at least, the third generation." ^ Ein Versuch, die Ursachen fur das Bliihen und Altern der Volker nachzu- weisen, Miinchen, 1889. PHYSICAL AND MORAL CONDITIONS 371 contain the largest percentage of city-born has led other writers than Hansen into serious error. From this error even the great work of Charles Booth (^Life and Labor of the People of London) is not free; thus, his investigators found that while 34.3 per cent, of the population of all London in 1881 were countrymen {i. e., born in the United Kingdom outside of London), only 24.2 per cent, of East London's population were countrymen. For the other dis- tricts the percentages were as follows : North, 44.4, West, 37.3, South, 34.1, Central, 30.4. North and West London, it is needless to say, are the wealthy residential quarters, while East London is the home of the " submerged tenth." Bethnal Green in East London is one of the poorest quarters in the city, and it contained 12.5 per cent, of countrymen; while London City, the heart of the metropolitan business and commercial interests, contained 39.5 per cent.^ But it is a mistake to infer off-hand, as do Mr. Booth's investigator and Hansen, that the countrymen constitute the wealthy class and the city-born the proletariat. In the first place, a large portion of the countrymen resid- ing in the wealthy districts are servants, janitors, etc., a class which, as will shortly appear, is recruited almost entirely from the country. Another large portion are clerks and other subordinate employees. But even were the rich them- selves largely of outside birth, it remains to be shown that they are rural rather than urban-born. Take the newcomers in Friedrichstadt, the business centre of Berlin, for instance ; they consist not only of provincials but also of foreigners, and the latter are more likely than not to be city-born, for as a previous chapter showed, long-distance migrants are more likely to be urban than rural. And in thickly settled coun- tries, a large part of the short-distance migration originates 1 Booth, op. cit., hi, 121-3. 372 THE GROWTH OF CITIES from the smaller cities and towns. The Leipzig census of 1885, for example, (Theil, II. p. 7, ff.) showed that fully one half (50.6 per cent.) of the immigrants had been born in places of 2,000 + . But the main reason, after all, why the city-born predom- inate in the poorest quarters of Berlin, London, New York, etc., is that the poorer classes of immigrants have so many children, who, of course, are classed as natives of the city. The slums of these cities were originally created by the flocking in of the most degraded peasant classes, who have married and propagated their kind until they now figure in the statistics as natives of the city, the product of urban con- ditions, the urban proletariat. The fact is amply demon- strated by the Vienna statistics, which show that the central and wealthy districts contain a larger percentage of immi- grants than the outer districts when the children of the im- migrants are credited to Vienna ; but when the children are credited to the father's birthplace {i. e., place of settlement),, it appears that the immigrants predominate in the outer and poorer districts of the city.' In Frankfort, again, the heavi- est immigration is not into the wealthier, but into the poorer districts.' ^ Rauchberg, " Der Zug nach der Stadt," in Stat. Mon., xix, 162. The per centage of immigrants was as follows according to place of — Birth. Settlement. Central districts 57-35 63.23 Outer districts 50-53 67.10 It is not always true, as Hansen assumes, that the centre of a city is its richest and most prosperous part; in fact, the reverse holds true in most American cities,, it being the well-to-do classes who can afford to live in the suburbs at a distance from their shops and ofi&ces. Even in some of the continental cities, where the storekeeper usually resides over his store, and the laborer near the factory, this relation is beginning to disappear. In Frankfort, for example, the inner city, where the business is transacted, receives the largest immigration, but this does not prove the truth of Hansen's theory that the immigrants step at once into the most important positions, for, as a matter of fact, the leading business men (judged by their wealth) dwell in the outer districts. The following table show- PHYSICAL AND MORAL CONDITIONS 373 Even more conclusive evidence that the wealthy classes are not to be identified with the country-born is furnished by the statistics of occupation and social rank, contrary to the interpretation imposed upon them by Hansen. The Berlin census of 1885 shows that the percentage of native Berliners was largest in manufacturing and trade, and smallest in personal service. But the class of casual laborers also contained a large proportion of the city-born, especially among females. The summary is as follows : Table CLIV, Showing the representation of native Berliners in the city's industries, 1885. Born in Berlin. Per cent. Per cent, of Per cent, of of all Total, census total. Males, all males. Females, females. Population, 1885- 557,226 42.4 265,184 42.0 292,042 42.8 Dependents 391,819 60.6 157,781 82.5 234,038 51.4 In all occupations 165,407 24.74 107,403 24.31 58,004 25.57 (1-24) Industries 112,317 29.53 80,321 27.59 31,996 35-87 (25) Personal services... 7,683 8.78 1,653 10.68 6,030 8.72 (26) Casual laborers 20,428 26.51 11,791 22.37 8,637 35.5 (27-38) Liberal profes- sions a .... 10,993 16.01 8,809 14.33 2,184 30'75 (39) Free income 11,898 26.5 4,696 21.8 7,202 30.7 (40) Not given 2,088 20.4 133 18.5 1,955 ^°-^ a Professions, without mil- itary 10,011 20,9 7,827 19.3 2,184 30.75 ing the distribution of the male immigrants to the city in a single year (1891) .will make this clear (Bleicher, II, p. 37, and I, pt. 2, table 35) : Am't of income tax Immigration. Population. per taxpayer. 1891. iSqo. Marks. Old city 25.18 15.24 8.8 New city 25.92 19.69 12.1 Southwest 7.39 4.22 13. 1 West end 2.17 3.56 11. 2 Northwest 3.59 6.78 16.4 North end 6.94 11. 14 16.8 Northeast 8.55 12.19 12.8 East end '. 5.96 7.62 13.5 Bornheim 2.90 5.90 8.4 Sachsenhausen-Inner 5.12 6.19 9.2 " " -Outer 6.8 7.47 14.3 1 00.0 ico.o 12.0 The most wealth is in the Northwest district and North End, where immigra- tion is proportionately the lowest. 374 '^HE GROWTH OF CITIES Explanations. The source of all the Berlin statistics is the census of 1885 : Die Bevolkerungs-, Geiverbe-, und Wohmmgsanfnahtne vom 1 Dez. 188$ in. der Stadt Berlin, itn Auftrage, etc., herausgegeben von Richard Boeckh, Berlin, iSgo,^. It is impracticable to refer to volume and page, as com- putations have been made from many different places The numbers in parenthesis refer to the official classification of occupations. The term " casual laborers " is a translation of " Arbeiter ohne nahere Angabe;" "free-income" of " Ohne Beruf," most of this class bemg rentiers ssvA pensioTidre or students (table CLV). The 40th group, " Ohne Berufangabe," con- tains about 1,000 persons in hospitals, prisons, etc. The other German expressions translated arc as follows: Entrepreneurs or undertakers, " Selbststandige ; " emploj'ees, " Abhangige " Each of the 40 classes of occupations contains several minor groups which ought to be mentioned. The entire classiiication is admirable. Table CLV. Occupations of Berliners, 1885. Males. Females. Percentage thereof Percentage thereof I. Industries (1-24) : Total. Berlin-born. Total. Berlin-born. 1. Trade (19) 60,494 28.34-]-' i3>53i 29-SS— ' 2. Clothing, etc. (15) 36)516 15.83 — 63,005 3S.88-f 3. Metal working (7) • 32,122 yj.di^ 505 52.87 4- 4. Wood working (13) 3i)973 32-i5-f 664 48.78-I- 5. Building trades 1 16) 27,978 16.53 — 25 44.00-I- 6. Food supplies (14) 18,562 18.92 — 1,128 32.27 + 7. Transportation (22) 15.456 15-64 — 210 29.52-I- 8. Paper and leather (12) 13.324 41.03-r 1.703 61.59 + 9. Machinery, tools, etc. (8) 12,050 36.93+ 162 45.67 + 10. Hotels and restaurants (23I 11,586 13.38 — 1.477 i4'i5 — II. Printing (17) 9,178 5S-I5+ 45i 59-42+ 12. Textile (11) 7.405 42.63 I- 3.739 50.26 + 13-15. Clay and stone, chemical, heat and light (6, 9, 10) 6,753 3I-9+ 162 37-6 + 16-20. Agr., gardening, fishing, mining (1-5) 2,985 18.00 — ,208 30.0+ 21. Art industries (18) 2,715 60.11+ 64 39.06+ 22. Insurance (20) 1.444 23.62 — i o. — 23. Peddling (21) 102 16.68 — 62 12.90 — 24. Amusements (24) 513 3I.3S+ 108 22.22 — Total (1-24) 291,156 27.59 89,205 35.87 11. Liberal professions (27-38) total.... 61,516 14-33 — 7.^°'' 3°-75 + 1. Army and navy (37) 20,607 4-79 — ° °- — 2. Public adminis. (34, 35, 38) 13.124 14.00 — 78 19.2 — 3. Rys., telegraphs, post (27, 28) ii,7^9 15.5 — 103 22.3 — 4. Teachers (30) 3,802 17.26 — 3.648 39.09+ 5. Arts (31) 3,628 41-46+ 1.400 33-85 + 6. Legal prof. (36).'. 3,069 23.50 — o o. — 7. Literary prof. (32) 3,003 27.93+ 78 25.64-1- 8. Medical prof. (29^ 2,035 18.08 — 1.790 12.96 — 9. Clerical prof. (33) 529 17.20 — 4 o. — III. Rentiers (39) 10,938 24.0 — 14,280 32.8 + Students (39) 8,292 17.0 — 1.117 24.08 — * The plus and minus signs in these columns indicate that the percentage is above or below the average in all occupations. PHYSICAL AND MORAL CONDITIONS 375 Table CLV presents in a clearer light the facts we are seeking, A few words will suffice to state the position of the 226,885 women engaged in gainful occupations in Berlin in 1885. As shown by the table 89,205 are found in the general industries, and the table also shows that the in- dustries that absorb nine-tenths of these are the clothing- manufacture, trade (store-keeping or clerking), and the textile industries. Nearly as many (73,335) are found in personal service, and the remainder are chiefly classed as lab- orers (24,336) or as living on free income, the latter including rentiers and students. A few are also in the literary pro- fessions. The percentage of Berlin-born women in these eight groups is as follows : Servants 8.72 Students 24.08 Trade 29.55 Liberal professions 30-75 Rentiers 32.8 Laborers 35.5 Clothing manufactures 35-88 Textile industries 50.26 The largest percentage of native Berlin women is found in the lowest skilled, worst-paid industries. Even the domestic servants, who are chiefly immigrants, enjoy an economic condition superior to that of the poor sewing-women engaged in the ready-made clothing business ; nor are they much poorer than the factory operatives in the textile industry. In trade, moveover, the native women are more often clerks than employers, their percentage in the two ranks being respectively 39.4 and 22.7.^ Similarly, in the clothing manufacture: undertakers, 11,507 out of 36,682 or 31.4 per cent.; employees 11,783 out of 28,323 or 41.6 per cent. ^ Of the 7,948 female undertakers, 6,146 are immigrants and 1,802 Berlin-bornj of the 5,583 employees, 3,386 are immigrants and 2,197 Berlin-born. 376 "^H^ GROWTH OF CITIES Indeed, the social rank of the native Berlin women in the whole group of industries 1-24 is inferior to that of the immi- grants (Table CLIV). Of the 89,205 women in these occu- pations nearly one-half are employees, and of these the Berlin-born constitute 42.5 per cent., although they form but 30 per cent, of the entrepreneurs and only 25.6 percent, of all women actively engaged in work. Among the em- ployees, again, there is little encouragement for the city woman; 5,940 of the employees are classed as salaried officers or superintendents, but as 4,726 of them are in mercantile businesses, it is plain that most of them are noth- ing but clerks. Nearly 25,000 are ordinary laborers, of whom the city-born constitute 35.5 per cent. — an ominous sign of a city proletariat. Let us now confine our attention to the male population. If we add to Table CLV the domestic servants and casual laborers, and then arrange the percentages according to maxima and minima|of the Berlin-born, we shall have these two groups : Table CLVI. Above the average (24.31). Below the average. Art industries 60.1 1 Servants 10.68 Printing S^-^S Hotel and restaurant 13-38 Textile industries 42.63 Transportation 15.64 Arts 41.46 Clothing manufactures 15-83 Paper and leather 4I-03 Building trades 16.53 Metal working 37.62 Peddling 16.68 Machinery, etc • 36.93 Students 17.01 Woodworking 32.15 Teachers 17.26 Chemical industries, etc . . • 31.9 Agr., gardening, etc 18.0 Amusements 31.35 Food supplies 18.92 Trade 28.34 Lib. professions (excluding Literary professions 27.93 military) 19.3 Casual laborers 22.37 Legal professions 23.50 Insurance 23.62 Rentiers and pensionars. •• 24.00 PHYSICAL AND MORAL CONDITIONS 377 The foregoing table would seem to be a sufficient answer to the pessimism of those who regard the London " sub- merged tenth " as a type of the city-dweller. The industries in which more than the average proportion of Berlin-born men are occupied are, almost without exception, the skilled trades ; in fact, the very highest percentage is found in the art industry. On the other hand, the industries filled with immigrants are the low-skilled ones, or trades requiring muscular strength more than mental ability. To this general rule there is an exception in favor of the liberal professions, which in most cases contain an unusually large percentage of immigrants. This is to be expected ; the cities are the centres of art and culture, and not only attract to themselves from the most distant quarters of the land students in search of artistic and professional training, but retain the best of them after the schooling has been com- pleted. Neither table makes the distinction between entrepreneur and employee for the men engaged in industry. The native Berliners apparently make a better showing among the em- ployees than among independent workers. It should be remarked, however, that most of the latter, although entre- preneurs, are not employers, and under modern conditions may not be so well situated as employees — especially the employees of the higher class (designated as angestellte Beamten, or salaried officers, in the tables). The classifica- tion of entrepreneurs (males), according to number of men employed, results as follows : Per cent, thereof. Total. Berlin-born. With no employees 56,046 21.2 1-5 " 22,424 23.9 " more than 5 employees 6,021 31.2 Total 84,491 22.7 This shows that in the higher ranks of entrepreneurs the 378 THE GROWTH OF CI TIES native Berliners are stronger than they are among the em- ployees of all classes, except apprentices. At the same time, it must be noted that this superior showing of the city- born among large employers is confined principally to store- keeping, inn-keeping and the clothing manufacture ; it exists in a small degree, if at all, in the skilled trades. It can hardly be said that the Berlin statistics favor Han- sen's contention, and even his own manipulations of them have failed to put his cause in a good light. The recent' Austrian statistics ^ are even less useful to Hansen's theory : Table CLVII. Birth-place. Vienna. Elsewhere Foreign Percentage oi Both in Austria, countries. females in Males. Females. sexes. Total, the group. Undertakers 27.5 42.3 34.4 49.2 16.4 100 46.8 Salaried employees. 35.7 44 5 36.7 47.8 15.5 lOO 8.5 Artisans 30.4 44.4 34.4 57.4 8.2 100 28.5 Unskilled laborers . . 23.6 13.4 21. i 66.4 12.5 100 24.4 Servants 16.8 12.2 12.4 73.3 14.3 100 94.3 All occupations 30.2 42.7 34.2 54.1 I1.7 100 31.4 Dependents 83.8 54.1 63.3 30.1 6.6 100 68.9 Total population .. . 44.0 45.4 44.7 45.2 lo.i 100 51.5 The term "undertakers" here designates all persons carry- ing on independent enterprises, whether they employ other labor or not. The class of higher employees (Angestellte) includes approximately all persons on a salary, i. e., not only public officials, but also the superintending /^r5^;z«id., 80.) ^Ibid.,ZT,. "/(JtV., 194. * Dr. Kuczynski's discussion of this point (pp. 71-8) is not restricted to Han- sen, but includes also Ammon and others. He mentions the fact that Berlin in 1840 contained 165,722 native Berliners; 50 years later (1890) there were 39,782 survivors. Naturally some of them had descendants, even if Hansen re- fused to admit it. Paris, on the other hand, is not self-sustaining, and Lagneau reckons that if left to itself, it would diminish 50 per cent, each generation, and in 18 generations or 5^ centuries would be effaced. (Essai de statistique anthro- pologique sur la population parisienne.) 386 THE GROWTH OF CITIES Of rather more interest is Hansen's final point about the dying out of the real city-makers, — the intellectual, mercan- tile, employing class.' In order to clear the ground for his case, he first attempts to prove that the only cities which provide their own natural increase are the factory cities, which require only " hands," and are supplied with the progeny of the proletariat; while commercial cities, where the intel- lectual ability identified with the middle class is concentrated, are dependent for their growth upon immigration.' A ker- nel of truth may doubtless be found in this proposition, as reference to Chapter VI on the natural movement of popula- tion will show. But this is the only statistical proof Hansen advances in favor of his theory that the intellectual workers are incapable of self-continuation. For the rest, he quotes proverbs about wealth not remaining in a family for more than three genera- tions, about the disappearance of great mercantile families,^ etc., and instances the rise of new men into the ranks of the " captains of industry," etc. The reasons he gives for the dying out of the merchant princes and directors of industry are failure of intellect, causing them to sink into the prole- tariat, and late and infrequent marriages and consequent small families. It is not a necessary consequence of the latter, however, that fewer children should be reared by the prosperous classes than are reared by the poorer classes with their larger families. The Massachusetts statistics already presented do indeed indicate that the diminution in infant mortality in small and well-nurtured families is too slight to ^ Drei Bevolkerungsstufen, Bk. iii, ch. 4, 5, esp. pages 174- 180. * Op. ciL, 27, 39, 208, 209. ' It may not be out of place here to observe that the export hand-book of Hamburg, the great German commercial centre, contains a list of 62 firms who were in business in Hamburg in the eighteenth century. Details of their careers, with old documents, etc., will be found in Hamburgs Handel und Verkehr : IlluUrirtes Export Handbuch der Borsen-Halle, iSgj-gg, i, 438. /' / PHYSICAL AND MORAL CONDITIONS 387 counterbalance the diminution in fecundity; as Dr. Crum remarks, " foresight and prudence seem to exercise a more powerful influence in restricting fecundity than in reducing infant mortality." But the careful study of marriage statis- tics according to social groups of the population which was made in Copenhagen by Rubin and Westergaard, leads to the conclusion that the upper classes, with a low birth-rate, often bring up as many children as do the lower classes, with a high birth-rate. Comparing the rates of the several classes or social groups with that of the fifth class (factory opera- tives, day laborers, sailors, etc.) as a standard (100), the following relations are found to exist: ' Births. Survirals. Class I 97 109 " 2 94 97 " 3 84 90 '• 4 90 94 " 5 100 100 The subject of the preservation of an intellectual aristoc- racy which shall add to human knowledge and lead the human race in its progress, has been ably discussed,^ and it must be admitted that the trend of opinio|i is against the conclusion reached by Rubin and Westergaard in their authoritative investigation of the marriage statistics of a single city.3 Sir Francis Galton is inclined to the opinion ' Statistik der Ehen, p. 1 22. The first class consists of employers and profes- sional men; the second of independent artisans, small tradesmen, superintend- ents, etc.; the third of teachers, public officials, etc.; the fourth of clerks, servants, etc, ''■ See the admirable discussion by Professor Marshall, Principles of Economics^ 3d ed., pp. 283-5. ' But Dr. Engel, who investigated the marriage statistics of Saxony, reached a similar conclusion. He says : " We derive from this investigation the conviction that while indeed more children are born from a marriage in the industrial pop- ulation than from one in the agricultural population, nevertheless the children of the latter have a greater vitality." (Cf. Wappaus, Allg. Bevolkerungsstaiistik, ii, 487.) 388 THE GROWTH OF CITIES that while the "upper" classes are capable of producing large families, they either consciously refuse to do so, or else marry heiresses who are of course hereditarily unproductive. The Spencerian school takes the ground that the higher the development of the individual, the smaller his capacity for reproduction, etc. The salvation of society therefore depends upon a mobility sufificient to permit or even encourage the rise of individuals from the lower to the upper social ranks. The process of recruiting the real aristocracy of ability and character must be unimpeded. And it is the concentration of population in cities which best promotes the process of bringing capable men to the front. Here is the one kernel of truth in Hansen's work, — the cities are the instruments of natural selection. As such they may be destroyers of human vigor, but not in the sense understood by Hansen. It is rather a social service that they perform in weeding out the incapable and inefficient, while advancing the more capable members of society. Let us briefly consider Hansen's suggestive theory of the Bevolkerungs Strom, which may help us in reaching a right conclusion concerning the process of natural selection. Society, according to the Hansen theory, consists of the three classes, — (i) land-owners; (2) intellectual workers or the middle class, including artisans, merchants, and profes- sional men; (3) the unskilled laborers and factory opera- tives, or the city proletariat. The land- owning, agricultural class is the great reservoir of vigor and life in any nation ; but it cannot hold itself together, for the reason that men multiply more rapidly than land can be made. The land- owning class is therefore continually throwing ofif a portion of its recruits and these form the current of migration to the cities. There they enter the middle class and struggle up- ward toward leadership ; but no family can long sustain the rigor of city competition, and eventually deteriorates whether it has attained the highest position or not. Hence there is op- PHYSICAL AND MORAL CONDITIONS 389 posed to the upward current a downward current of degener- ates. The upward current is composed almost wholly of the country-bred (offspring of the land-owners), but the down- ward current is composed of the city-bred. The reservoir which absorbs the survivors in the downward current is the class of unskilled laborers or the city proletariat, who multi- ply abundantly but never rise in their position. The city thus becomes an instrument of social degeneration. It takes the crude vigor and vitality of the agricultural population, develops and appropriates to itself their highest intellectual abilities, and then casts them aside into the ever-increasing number of non-efificients. It is obvious that if the cities keep on growing at their nineteenth-century rate, they will dry up the reservoirs of strength in the population and leave in their place an immense proletariat, practically good for nothing. Now is it true that cities would stagnate and decay, if the stream of migration were stopped? Are they incapable of producing the intellect and energy requisite for progress? Is it " the result of the conditions of liie in great towns that muscular strength and energy get gradually used up ; that the second generation of city men is of lower physique and has less power of persistent work than the first, and the third generation (where it exists), is lower than the second ?"^ All of our investigations in the course of the present chapter point to the conclusion that the townsman is on the average a more efficient industrial unit than the rural immi- grant. The city proletariat, contrary to Hansen's theory, appears to be recruited from the country-born rather than from the real city- dwellers. In fact, the countryman coming to the city begins a slow ascent, rather than a descent ; his children, instead of being men of " lower physique with less power of persistent work" advance to a higher rank on the industrial and social ladder, while the third generation, in- ■ H. L. Smith, in Booth, op. cil., iii, no. 390 THE GROWTH OF CITIES Middle " Stud- classes. ierten." Total. H 4 100 49 lO 100 35 25 100 Stead of dying out, is still more capable and efficient. This is also the opinion of Ammon, Hansen's most famous dis- ciple, whose researches in Carlsruhe may be summarized thus (without guaranteeing their value for generalizations) : ^ Lower classes. Immigrants S2 Their sons 41 " grandsons 40 As a class, the country immigrants do not at once assume the higher positions in the economic organism, but enter the unskilled occupations where muscular strength and vigor are in demand. Among the rural immigrants there are indeed some few skilled artisans, but there are very few cases of country laborers becoming artisans in the cities.' The im- ' G esdhchaftsordnuiig, 145. - H. L. Smith who investigated the cases of 500 village emigrants (the great majority of whom moved to London), found only six such cases. His table show- ing the occupations of the 500 migrants before and after migration, confirms the conclusions in the text, which are based chiefly on the more precise data of Con- tinental cities. The table may be condensed as follows : Percentage of migrants in the occupations named: Before After migration. A. Outdoor labor. Laborer 64.0 16.9 Other 5.5 25.2 B. Service — personal and domestic 5.8 15.8 C. Public service — soldier, policeman, etc 14.5 D. Building trades 8.3 6.4 E. Other industries ' 7.6 9.1 F. Retail dealers and innkeepers 7.8 9.7 G. Miscellaneous: Clerk 7 1.5 Teacher, etc . .3 .9 100. 100. — Booth, Life and Labour of the People, iii, 140. " In the original table this group as designated is " other skilled occupations," but, as Mr. Smith observed, the soap, chemical, gas-works, etc., here included probably employ many ordinary laborers. PHYSICAL AND MORAL CONDITIONS 391 migrants at first take up with such menial occupations as domestic and personal service, work in hotels and restaur- ants, postmen, cab-drivers and truckmen, and, in some cases, with the building trades. It is only gradually that they work their way into the skilled industries, in which the city-born have a far larger representation. An illustration of this fact is given in the following figures from the Berlin census of 1885, showing the distribution, for both sexes, of the erwerhs- thdtigen native Berliners, immigrants, and immigrants within the five years just preceding the census : ^ Table CLIX. Berlin-born. Total Immigrants. Immigrants in 1881-5. Male. Female. Total. M. F. Total. M. F. Total. Industry, trade, trans- portation 748 551 679 631 339 533 580 228 453 Common laborers ... no 149 123 122 93 113 94 59 81 Domestic servants .. . 4 89 35 13 375 135 21 593 229 Other servants 11 15 12 24 23 24 26 12 21 Military 9 ... 6 58 ... 39 135 ... 86 Public service 73 38 61 99 29 75 78 27 59 No profession 45 158 85 52 141 82 67 81 72 1,000 1,000 1,000 1,000 i,coo 1,000 1,000 1,000 1,000 And just as the newcomers work their way up from the unskilled into the skilled industries, so do they rise from the ranks of employees into those of undertakers and employers. The immigrants who had been in Berlin five years or less in 1885 formed 29.3 per cent, of the working population, but only 12 per cent, of the entrepreneurs, and only 5.5 per cent, of the entrepreneurs who employed more than five workmen. On the other hand the older immigrants, who had been in Berlin over 15 years, constituted 22.7 per cent, of those in gainful pursuits, 40.9 per cent, of the entrepreneurs, and 45 per cent, of the large employers : ^ Briickner, Allg. St. Archiv, i, 645. 392 THE GROWTH OF CITIES Table CLX. Popu- Entrepreneurs (males), lation All occupations. with number of employees. Residence in Berlin. 1885. Females. Males. o. 1-5. Over 5. Total. Less than 5 years 21.6 32.6 29.3 13.0 g.g 5.5 11.7 S-15 years 19.5 22.1 23.7 25.4 24.8 18.7 24.7 Over 15 years 16.5 19.7 22.7 40.4 41.4 44.6 40.9 Total immigrants 57.6 74.4 75.7 78.8 76.1 68.8 77.3 " Berlin-born 42.4 25.6 24.3 22.7 23.9 31.3 21.2 Total 100. o too.o loo.o 100. o loo.o loo.o 100. o Notwithstanding such facts, it is commonly held that city- life produces dwarfed, stunted men and degenerates ; for- tunately, statistics of physical infirmities exist which dispel such fears about the effects of city life. It is now generally recognized that a connection exists between congenital blind- ness, congenital deaf-mutism and congenital imbecility or feeble-mindedness, i. e., they are all results of impaired con- stitutional vigor. Now recent statistics show that these in- firmities are rather more prevalent in rural districts and small towns than in the cities, while insanity, which is rather a nervous than a bodily failing, prevails chiefly in the cities. From the exceedingly valuable report by Dr. John S. Billings on the Insane, Feeble-minded, Deaf and Dumb, and Blind, in the United States at the Eleventh Census, the following figures are derived, showing the ratio of the specified classes to 100,000 of the population: United Cities of States. 50,000 J . Authority. Insane 170.0 242.9 Tables 151, 153. Feeble-minded 152.7 74.3 " 172,174. Deaf mutes 64.8 48.7 " 189,191. Blind 80.8 53.5 " 223,225. These results are confirmed by European statistics.^ And lest it be inferred that the difficulty of apportionating inmates 'Levasseur, La population fianfaise, i, 345; Rauchberg, Die Bevolkerung Oesterreichs, 232 ff. PHYSICAL AND MORAL CONDITIONS 393 of institutions to their real home, city or country, impairs the value of the statistics, it may be well to refer to the classical work of Dr. Mayr, in which these classes in Bavaria were distributed according to birth-places.' For the last three classes, Dr. Mayr found geographical differences, but for the insane the local variation was almost entirely due to the size of the town. His results may be summarized thus : Rural districts. Cities. Munich. Insane 88.1 185.4 221. Feeble-minded 153-3 '^36-5 130. Deaf and dumb 91. i 73.3 59. Blind 78.6 1 19.8 117. It is clear then that while city life produces, or at least maintains fewer of the severer physical infirmities, like blind- ness, deaf-mutism and idiocy, than does the country, it does favor the increase of insanity. The average height and girth of chest are significant cri- teria of physical vigor ; and of the two, the latter is the more important, since it is indisputable that the strongest indi- viduals and races are those that have the greatest chest capacity and lung power. As regards stature, the preponderance of opinion in the past has been that city life exerts a depressing effect upon the individual.^" The city of Hamburg is below the average for Germany, Geneva below the average for Switzerland, and Madrid has almost the shortest male population in all Spain. Ammon, the Carlsruhe anthropologist, holds the contrary ^Die Verbreitung der Blindheit, der Taubstummheit, des Blddsinns und des Irrsinns in Bay em. XXXV Heft der Beitrage zur Statistik des Konigr. Bayerns, iSyy, pp. 71-2, 304. Additional references are given in Mayr's Bevolkerungsstatistik, § 33. ''■ This is the conclusion of Professor Ripley, who surveys the evidence in an essay (" Racial Geography of Europe, xii, Urban Problems ") in the Popular Science Monthly, March, 1898. Most of the statements made above depend on the authorities cited by Dr. Ripley {loc, cit., 52 : 602). 394 THE GROWTH OF CITIES opinion;^ but his statistical methods are open to grave sus- picion,^ and his deductions from observations in Baden are not necessarily true of large populations. Dr. Beddoe, the leading British authority, declares that " it can be taken as proved that the stature of men in the large towns of Britain is lowered considerably below the standard of the nation, and as probable that such degeneration is hereditary and progressive." 3 Ammon finds, moreover, that the townsman, who works in closed rooms and makes little muscular exertion, is consid- erably inferior in chest capacity to the countryman.* And the difTerence is not due to the migration cityward of countrymen below the general standard of the nation, for the countrymen residing in cities have a larger girth of chest than the city-born. From Cato's time down, statesmen have declared that the bravest men and most daring soldiers have come from the land.5 In 1856, Dr. Engel, the statistician, concluded that of 100 candidates examined from the country districts, 26.6 "^ Die Gesellsckaftsordnung (1895), P* ^^7* Ammon affirms that townsmen grow taller and mature earlier than countrj'men as a result of superior nourish- ment (Z>»> natiirliche Auslese, 123). It is undoubtedly true that a connection exists between stature and economic position ; thus, it is found that the height of persons in Paris uniformly increases as one passes from the poorer to the wealth- ier wards. In Menilmontant (ward xx), where 80 per cent, of the funerals are at public expense, the average height was 1.637 metres; in the Opera (ix), with only 27 per cent, of public funerals, the average was 1.660 m. — Manouvrier, " Sur la Taille des Parisiens," Bulletin de la Soc. d^Antkrop. 3d ser., ix, 168. '' Cf . Kuczynski, 1 24-9. ' On the Bulk and Stature of Man in Great Britain (1867), p. 180. * Die natiirliche Auslese, 170. ' M. Porcius Cato, de Rustica, c. i : Foriissimi viri et milites strenuissimi tx agricolis gignuntur, minimeque male cogitantes. Plinius, Hist. Nat., Lib. 18, c. 5. Mhnoires de Maximilien de Bethune, Due de Sully, etc. (Lond., 1747), T. ii, p. 289. Engel, " Die physische Beschaffenheit der militS.r pflichtigen Bevolkerung im PHYSICAL AND MORAL CONDITIONS 395 were declared fit for service, as against 19.7 from 100 urban candidates. Now it is to be observed that the believers in town degen- eracy base their arguments on antiquated statistics. There can be no doubt that down to very recent times the health and vigor of urbanites compared unfavorably with that of men who worked in the open air, just as their death-rates did.' But in the last quarter century the evidence in both cases has changed. In 1874 a French authority '^ declared that fitness for army service depends less on density of popu- lation than on wealth, climate, daily life.^ Health and vigor may always be preserved if men in cities will make proper provision for open-air exercise, cleanliness and a pure food supply. Professor Marshall, who is not afraid of looking the Konigr. Sachsen," in Zeit. des Sta{. Bureaus des K. Sachs. Ministeriums des /nnern, 1856, Nr. 4-7 (esp. pp. iii, 112). E. Helwing: " Ueber die Abnahme der Kriegsiiichtigkeit der angehobenen Mannschaften, namenilich in der Alark Branden burg, l&erlin, i860. ^ Exceptions, however, can be found in former centuries, A recent writer has called attention to the facts that the city-bred infantry of the Flemish towns was more than a match for the best troops French chivalry could bring against them in 1302, and that the citizen soldiers of southeastern and eastern England in the War of the Roses gave the victory to the Yorkists over the masses of peasants and huntsmen of the North and West. — Contemporary Review (Oct., 1891), 60: 554- - Art. Recruitment, Dictionnaire encyclopedique des sciences medicates, 3d series, vol. ii. ' Taking the three densest departments of France and comparing them with the three least dense, it was found that the number of men who had to be examined to secure 1,000 soldiers was (^Ibid., 644, 646) : Pop. per sq. Number kilometer. examined. Seine 64.5 i>790 (Paris) Nord 3.5 1,815 (Lille, Roubaix, etc.) Rhone 3.5 1,776 (Lyons) Hautes Alpes 5 2,580 Lazere 3 2,050 Basses- Alpes 3 2,190 396 THE GROWTH OF CITIES evils of town life in the face, is right when he says that " it is not to be concluded that the race is degenerating physically, nor even that its nervous strength is, on the whole, decaying. On the contrary, the opposite is plainly true of those boys and girls who are able to enter fully into modern outdoor amusements, who frequently spend holidays in the country, and whose food, clothing and medical care are abundant and governed by the best knowledge,"^ But after all, progress depends less on purely physical strength than on moral resolution or nervous strength. In the words of the writer just quoted, "the power of sustain- ing great muscular exertion depends on force of will and strength of character as well as on constitutional strength. This energy (strength of man, not of body) is moral rather than physical; but yet it depends on the physical conditions of the nervous strength. This strength of the man himself, this resolution, energy, and self-mastery, or in short this 'vigor' is the source of all progress: it shows itself in great deeds, in great thoughts, and in the capacity for true relig- ious feeling.'"* Now, it is precisely the high nervous organization of city- bred soldiers that has enabled them to last through long campaigns as well as or better than countrymen with their rude physical health. It made the students of Berlin Uni- versity able to bear fatigue better than the average soldier in the war of 1 870-1, and rendered the New England store- clerks equal to all the strain of Sherman's march to the sea in 1864. When to nervous strength is joined the muscular development to be found among the athletic middle-class youth in American and English suburban towns, one is jus- tified in haiUng them as world-conquerors. Seven years ago an English writer, referring to the young men of Wimbledom and Battersea, near London, had the foolhardiness to eulo- ^ Principles of Economics, p. 281. ''■ Ibid., 275. PHYSICAL AND MORAL CONDITIONS 397 gize them as future victors of Marathons ;^ but yesterday, and he saw his prediction fulfilled as American city lads marched to victory on Cuban soil, side by side with the rough cowboys of the western plains. Having dealt with the subject of physical health and vigor, it now remains to consider the influence of city life upon in- telligence and morals. Education. — As regards education, it must be obvious that the agglomeration of population is more favorable than its dispersion can be. In fact, one would naturally turn to the cities and towns for the best schools, since they alone can afiford to provide the expensive advantages incident to the grading of pupils and the division of labor educationally. It is not surprising, therefore, that the urban schools of the United States have 190 class days per annum, and the rural schools only 115; and that the attendance in the city is 70 per cent, of the enrolment, while in the country it is 62 per cent. Moreover, the statistics of illiteracy in the United States are favorable to the cities, notwithstanding the recep- tion by the cities of the bulk of illiterate foreigners. The following comparison embraces all of the 28 large cities (100,000+) of the United States and the commonwealths in which they are situated (Washington, which coincides with the District of Columbia, being ranged under Maryland) : Percentage of illiterates in the population, io years of age and over, 1890:' New York 5.53 Pennsylvania 6.78 New York 7.69 Philadelphia 4.97 Brooklyn 3.25 Pittsburg 6.93 Buffalo 5.38 Allegheny 3.77 Rochester 3.56 Missouri 9.09 Illinois 5.25 St. Louis 5.89 Chicago 4.63 Kansas City 5.78 ' Low, " The Rise of the Suburbs," in Contemp. Review, 60 : 555. * iitk Cens., Cotnp. iii, 301, 317. 398 'J'^^ GROWTH OF CITIES Massachusetts 6.22 New Jersey 6.50 Boston 5.69 Newark 4.81 Maryland ^S-TO Jersey City 5.91 Baltimore 9.80 Minnesota 6.03 Washington 13.20 Minneapolis 2.39 California 7.67 St. Paul 4.54 San Francisco 5.35 Kentucky 21 .65 Ohio 5.24 - Louisville 10.69 Cincinnati 4,27 Nebraska 3. 1 1 Cleveland 6.49 Omaha 2.86 Louisiana 45»83 Rhode Island 9.76 New Orleans ^S-TO Providence 7.73 Wisconsin 6.73 Colorado 5.24 Milwaukee 5.34 Denver 2.83 Michigan 5.92 Indiana 6.32 Detroit 6.66 Indianapolis 6.12 With very few exceptions (New York City, Pittsburg, Cleveland, Detroit), the cities have a better educated popu- lation than the rest of the State in which they are situated. The difference in favor of the cities is in many instances very marked, although in the case of Baltimore, New Orleans and Louisville, it is explicable by the different proportions of negroes in the population. There can be no doubt about the superiority of the city schools, both primary aud secondary. Educators in fact now recognize the inferiority of rural schools as one of their most pressing problems, and the National Educational Association is even now discussing the ample report on rural schools presented at its 1897 meeting by its Committee of Twelve. But the education of the schools forms only a part of a man's education. Their discipline must be supplemented by outside reading and experience ; alone it too often promotes superficiality. And this is the pecuHar danger of urban habits of life. The city boy is taught to read, but not to think ; the result is seen in the immense constituency of " yellow journalism." Country newspapers are trivial enough, but they do not descend to the depths of moral PHYSICAL AND MORAL CONDITIONS 399 degradation of sensational metropolitan journals, manufac- tured for city readers. Town education has been so well described by Mr. Hob- son ' that it would be a loss not to quote his words : " That town life, as distinguished from town work, is educative of certain intel- lectual and moral qualities, is evident. Setting aside that picked intelligence which flows to the town to compete successfully for intellectual employment, there can be no question but that the townsman has a larger superficial knowl- edge of the world and human nature. He is shrewd, alert, versatile, quicker and more resourceful than the countryman. In thought, speech, action, this superior- ity shows itself. The townsman has a more developed consciousness, his intelli- gence is constantly stimulated in a thousand ways by larger and more varied society, and by a more diversified and complex economic environment. While there is reason to believe that town work is on the average less educative than country work, town life more than turns the scale. The social intercourse of the club, the trade society, the church, the home, the public-house, the music-hall, the street, supply innumerable educative influences, to say nothing of the ampler op- portunities of consciously organized intellectual education which are available in large towns. If, however, we examine a little deeper the character of town edu- cation and intelligence, certain tolerably definite limitations show themselves. School instruction, slightly more advanced than in the country, is commonly utilized to sharpen industrial competition and to feed that sensational interest in sport and crime which absorbs the attention of the masses in their non-working hours; it seldom forms the foundation of an intellectual life in which knowledge and taste are reckoned m themselves desirable. The power to read and write is employed by the great majority of all classes in ways which evoke a minimum of thought and wholesome feeling. Social, political and religious prejudices are made to do the work which should be done by careful thought and scientific in- vestigation. Scattered and unrelated fragments of half-baked information form a stock of 'knowledge' with which the townsman's glib tongue enables him to pre- sent a showy intellectual shop-front. Business smartness pays better in the town, and the low intellectual qualities which are contained in it are educated by town life. The knov.'ledge of human nature thus evoked is in no sense science; it is a mere rule-of-thumb affair, a thin mechanical empiricism. The capable business man who is said to understand the ' world ' and his fellow-men, has commonly no knowledge of human nature in the larger sense, but merely knows from obser- vation how the average man of a certain limited class is likely to act within a narrow prescribed sphere of self-seeking. Town life, then, strongly favors the education of certain shallow forms of intelligence." Religion and Morality. — According to the special report of ^The Evolution of Modern Capitalism, pp. 338-9. 400 THE GROWTH OF CITIES Dr. H. K. Carroll on Statistics of Churches and Religious Denominations at the Eleventh Census,^ the cities contain a larger proportion of church members, or communicants, than do the smaller places ; but on the other hand, the cities have by far fewer church buildings, as will be observed in the following statistics : Communicants (percentage One church edifice of total population!. to a pop. of United States 32.85 440 All cities of 25,0004- (124) 37-90 i»439 Cities of 25,000-100,000 (96) 39-IO i>o52 " " 100,000-500,000 (24) 3S.90 1,468 " " 500,000-f- (4) 35-6o 2,147 Thus the population to a church building steadily increases with the size of the city, and it is a question whether the seating capacity of the buildings increases in the same ratio. It is also noticeable that while the cities have a larger pro- portion of communicants than the entire United States, yet this proportion steadily diminishes as we pass from the class of smaller cities to that of the larger. To put the facts in another light they may be thus summarized : Ratio to the U. S. Population of 124 cities of 25,000-!- 22.5 Communicants in 124 cities of 25,000-!- 25.72 Number of church edifices in 124 cities of 25,ooo-(- 6.82 Value of church property in 124 cities of 25,000-f- 46.13 Of the leading denominations (numerically), the Roman Catholics are strongest in American cities, although the Hebrews tend to concentrate in cities even more than the Catholics. Of the total strength of the several denomina- tions in the United States the following percentages were in the cities of 25,0004- in 1890:^ ' Pp. xxvi and xxvii. ' Op, cit., p. xxvii. PHYSICAL AND MORAL CONDITIONS 40 1 Jewish (orthodox) 9^•^^ " (reformed) 84,57 German Evangelical Protestant 77-97 Roman Catholic 48.26 Unitarians 48.08 Protestant Episcopal 48.03 German Evangelical Synod 38.56 Reformed in America 30.85 Presbyterian (North) 29.85 Lutheran Synod. Conference 29.77 Lutheran General Council 25.75 Congregational 25.57 Regular Baptist (North) 25.06 And since only 22 per cent, of the total population of the country is resident in these cities, the denominations above mentioned are disproportionately strong in the cities. The only statistical measure of morality, as distinguished from religion, is negative, being in fact a measure of immor- ality, — the amount of vice and crime recorded by the police authorities. But first, let us consider the phenomenon of suicide. As is generally known, this phenomenon is more frequent in the city than in the country/ In the United States, as far as we can judge from the imperfect returns of vital statistics in the Eleventh Census, the disparity is not so great as it is elsewhere ; the rate per i ,000,000 population being 92.9 in the cities of registration States, and 80 in the rural parts, while in the outside cities it was 126.5.=' The following table is from Morselli, Levasseur, von Mayr, etc. : Suicides per 1,000,000 population. Urban. Rural. Capitals. France (1883) 263 172 Dept. of Seine 472 Prussia (1869-72) 162 97 Berlin '... 191 Italy (1877) 66 29 Rome, iii; Milan 199 Norvi^ay (1866-9) 92 72 Sweden (187 1-5) 167 67 Stockholm 440 Denmark (1869-75) 283 257 Copenhagen 350 Belgium (1858-60) 64 34 Bavaria (1876) 118 104 Munich (1860-69) 19° Saxony (1859-63) 317 219 ^ Morselli, Suicide, pp. i6i-l86, esp. 172; Levasseur, La. pop. fran., ii, 133. ' Vital and Social Statistics, Part i, pages 463-4. 402 THE GROWTH OF CITIES Without exception, the suicide-rate is higher in the urban than in the rural communities, and highest of all in the great cities. Botli von Oettingen and Wagner have called attention to the excessive rate in the capitals:^ St. Petersburg i8o Russia 28 Vienna 290 Austria 130 Leipzig 450 Saxony 394 London 85 England 69 New York 121-181 United States 32 (?) In fact, there seems to be a regular progression in the sui- cide-rate from small centres to large centres, as may be seen in the following Prussian statistics:* Suicides to i,ooo,ocx3 inhabitants. Prussia, 188X-90 202 " cities of 1 5,000-j- 256 " " "20-100,000(1892) 247 " " " ioo,ooo-|- " 308 " Berlin 329 The fact that cities contribute more heavily to the num- ber of suicides than do rural communities has been estab- lished. But it is not to be inferred that those countries which have the largest urban popalations also have the largest relative number of suicides ; in fact Morselli has shown by comprehensive statistics 3 that many of the coun- tries in which population is most concentrated {^e.g. England, Holland, Belgium), have a low suicide-rate. Race is the most important single factor in the production of suicides ; hence in the Netherlands the highest suicide-rate is not in the * Von Oettingen, Z>/(f Moralstatistik (1882), p. 765; Nagle, Suicide in New York City ; Wagner, Die Geseizmassigkeit in den scheinbar ■willki'irlichen mejisch- lichen Handlungen vom Standpunct der Statistik, 2 Theil, i, Vergleichende Selbstmordstaiistik Europas, etc., Hamburg, 1864. 'Von Mayr, Art. " Selbstmordstatistik " in Conrad's Hdwbh., ist Sup., p. 700. ' Op. cit., 170, 157. PHYSICAL AND MORAL CONDITIONS 403 south-western provinces which contain the large cities, but in the north-east, where the Germanic element is strong.^ The analysis of suicides by occupation shows that the phe- nomenon is also connected with the predominant professions. Thus, in France the number of suicides per 1,000,000 of pop- ulation, is 120 in agriculture, 130 in commerce, 190 in in- dustry, 290 among domestic servants, 550 in the liberal professions and 2,350 among those of no profession or of unknown profession.' Suicide is really one of the penalties paid for progress. It is one of the processes of natural selection, resulting from failures in the " struggle for exist- ence " and is therefore most prevalent where the competi- tive struggle is keenest. As cities are the centres of the severest competition, they naturally have the largest number of suicides. It is to be observed, however, that suicide is not increasing in the large cities, or at least is increasing less rapidly than in the smaller places. Morselli reviews the evidence at length and regards it as establishing the " fact that the tide of suicides rises in all countries, and especially in the prov- inces, whilst it remains stationary or decreases in the great and most civilized capitals of Europe."^ Criminal statistics undoubtedly put the cities in a bad light. In England, for example, the cities have double or even quadruple the amount of crime that the rural commu- nities have, as will be seen in the following table giving for the year 1894 the number of offenses per 100,000 popu- lation :4 ' op. Cit.,^-]. - Levasseur, ii, 127. •' Op. cit., 183. According to the latest English yttdicial Statistics (for year 1894, published in the 1896 Parliamentary papers), the county of London ranked tenth among English counties, with a suicide-rate of ico.6 (p. 47). * Jttdicial Statistics, in Pari. Papers, 1896, vol. xciv, p. 24. 404 ^^-^ GROWTH OF CITIES Table CLXI Indictable crimes. Offenses against Crimes; all persons. Offenses tried summarily. indictable Offenses Crimes Crime , ' , offenses against of vio- against Drunk- Vagrancy reported, property, lence. morals. Assaults. enness. Acts. Metropolis 416.7 386.2 10.6 5.9 390.1 637.4 148.8 Mining counties .... 234.3 214.3 83 8.1 286.8 1,136.7 280.3 Mf. towns 351-8 332.4 6.6 4.4 272.6 470.1 244.9 Seaports 643.6 597-9 22.5 8.4 426.0 1,260.8 368.3 Pleasure towns 265.7 250.4 4.3 4.1 180.5 289.3 82.9 Agr. counties: East " 128.2 iig.i 3.7 3.6 120.3 109.9 55-4 S.-W. " 182.9 163.5 5.2 8.1 150.1 209.4 155.7 Home counties 202.1 185.9 4-2 6.5 146.7 245.0 52.2 Eng. and Wales 296.7 275.9 7-2 6.1 252.2 616.3 191.1 The vast majority of crimes are against property, having numbered 53,621 in a total of 56,281, and these are largely larcenies. As regards the graver offenses against the person, London occupies a middle rank (murder) and a low rank in sexual crimes (rape). But it does not appear that crime is increasing dispropor- tionately in the cities. The French statistics, for example, separate criminals according to their domicile or legal resi- dence. In 1 841-5, 38 per cent, of those charged with offenses were domiciled in the urban communes (2,000+ pop.), which then contained about 23 per cent, of the population. In 1866-70 the respective percentages were 44 and 31; in 1 88 1 -5, 46 and 35.^ Crime has therefore increased less rapidly than population in the towns.'' Our statistics of vice are mainly restricted to the subject of illegitimacy. We have indeed just seen in Table CLXI that violations of the person are less frequent in the English ' Levasseur, op. cit., ii, 455. '■'Porter {Progress of the AW?(7«, p. 646) figured that between 1805 and 1841 crime increased by 1,277 P^"^ cent, in 20 agricultural counties of England and 1,252 in 20 industrial counties. But Leone Levi denies the validity of Porter's pro- ceeding, {your, of St. Soc, 1880.) PHYSICAL AND MORAL CONDITIONS 405 cities than in the rural counties. Illegitimacy, on the other hand, appears to center in the cities. It is nearly a half- century since Wappaus showed that illegitimacy was about twice as great in the cities as in the rural parts of Europe; the average percentage of illegitimate births in all births being 14.7 in urban populations and ^.6 in the rural.^ The difference between city and country is very marked in France, and illegitimacy culminates in Paris, where between one-fourth and one-third of all the births are illegitimate." In 1879-83 there were the following numbers of illegitimates per 100 births;* Department of the Seine (Paris) 24.1 Urban population lo.i Rural " 4.2 France 7.4 Judging from the foregoing statistics, the cities must be hot-beds of immorality. But their case is not so bad as it seems. It must be re membered that maternity hospitals are always located in cities, and many of the women who enter these are countrywomen who come to the city to conceal their shame. Levasseur is the authority for the statement that over one-fourth (4,405) of the illegitimate births in 1884 (16,137, or "^^'7 percent, of all births) were such cases. He says that nearly one-half were the fruits of liaisons, which in Paris are regarded as a form of marriage, leaving something over one-fourth the alleged number as the real product of Parisian immorality. With these deductions, the Paris rate '5«/r«, Table CXLVI. - Even Paris, however, ranks below some other European cities. In Munich, Vienna and Prague one-half the births were illegitimate a few years since, and Rome had a percentage of 44.5; Stockholm, 40; Moscow, 38.1; Budapest, 30.5; Paris at that time 28.6, and London 3.9. — Levasseur, ii, 4CX)-i. ' Levasseur, ii, 34. 4o6 I^HE GROWTH OF CITIES would be somewhat less than twice that of the rural popula- tions. But there is still another factor in the problem, namely, the larger proportion of young unmarried women in the cities. This factor, taken into consideration, will account for much of the city illegitimacy ; in Germany, as we have seen, (Table CXLIV) the number of illegitimate births to i,ooo unmarried women of child-bearing age is actually less in the cities than in the country, and the same fact has been observed in Scotland.' On the whole, it is to be doubted if the cities are much worse than the rural districts as regards illegitimacy; the question cannot be determined definitely until other countries furnish the refined rate. In this coun- try, unfortunately, no distinction between rural and urban populations has been made in the matter of illegitimacy. Infanticide, as the European criminal statistics have shown, is more prevalent in the country than in the city, while abor- tion seems to be less prevalent there. Prostitution, regarded as a profession, is certainly a city institution, but many social workers doubt whether the sexual morality of the country is on a higher plane, from their knowledge of the large proportion of prostitutes who were first corrupted in country homes. The morals of " wicked Paris " have frequently been impeached, but sociologists who know the facts declare that a very large part of the Parisian vice is supported by travellers and foreign sojourners. If such is the case, it is wrong to regard the entire Parisian population as immoral. Similarly with other cities: they have a great deal of vice, to be sure, but it is the property of a distinct class of the population. In the United States, the number of drinking saloons af- fords a rough index of a town's morality, and the New York State figures for 1897 show that while the cities on the whole ' Levasseur, ii, 206. PHYSICAL AND MORAL CONDITIONS 407 have a larger relative number of saloons than the rural parts, the largest cities do not take the lowest rank:' No. of saloons per 1,000 population. New York State 3.6 Rural 2.8 Urban > 4.5 Cities of first class ■■* 4,3 " " second class '^ 4.0 " " third class * 4.5 The amount of viciousness and criminality in cities is probably exaggerated in popular estimation from the fact that the cities have long been under the blaze of an Argus- eyed press, so that the worst is known about them. They have hitherto overshadowed the evils in the moral life of vil- lages, but several recent rural crimes of unwonted atrocity have awakened fn the nation a truer realization of the actual facts.s Many sociologists have also realized that the rural centre is not so "idyllic" as has been imagined.^ ^ Second Annual Report of the State Commissioner of Excise, for year ending Sept. 30, 1897, p. ^°5' * New York, Brooklyn, Buffalo. ' Rochester, Albany, Syracuse, Troy. * Thirty-four cities with population ranging (1892) between 8,000 and 50,000. * Reference is here made particularly to the assault, robbery and murder of Mrs. McCloud of Shelburne Falls, Mass. Note the following specimen of newspaper comment on the crime (Hartford Times') : " There is a lesson for our New Eng- land communities in the career of Jack O'Neil, the Shelburne Falls hoodlum and ne'er-do-well. O'Neil was what the specialists would describe as a true degener- ate. He was an idle, worthless, drunken, penniless fellow, hanging around the en- trances to the village dramshops (of which Shelburne Falls plainly has too many) , sponging his food and lodging out of his mother, a hard-working washerwoman, and as sure to develop into a criminal as darkness is to succeed daylight The evidence at the trial in Greenfield showed that O'Neil was only one of a gang of youthful * bums ' and hoodlums who are tolerated in Shelburne Falls, and whose ill-gotten gains furnish considerable support for a lot of cheap and nasty dram- shops which disgrace the place. How many other New England villages present the same conditions? A good many to our certain knowledge." ® Vide Prof. Blackmar's studies of the " Smoky Pilgrims " in the American yournal of Sociology, January, 1897. 408 THE GROWTH OF CITIES But if there is actually a larger criminal and vicious class in the cities, as would be a priori expected from the fact that the cities are the foci of attraction, it does not follow that the danger of contamination is greater. The fact is that in the city the crime is localized; it is confined to particular classes and the remaining social classes are so much the cleaner. There are perhaps relatively more offenses com- mitted in the city than in the village,' but not so many more offenders. And most people will admit that there is con- siderable difference between a society where the same man comes before magistrate six times, and another society where six men come before the judge once. The cities, moreover, have the benefit of an educated pub- lic opinion on moral questions which is often effective to suppress the beginnings of vice. The power of social opin- ion, supported by legislation, has been abundantly demon- strated in the transformation of factory labor. There was a time when factories were actual " men-consumers," produc- ing a morally and physically dwarfed and stunted race. That time is forever past in America and England, while to-day the worst conditions are found in the home ("sweating") industries. The same strong social opinion that wiped out factory abuses by the Factory Acts must now be concentra- ted on the evils of city life. Finally, the fact must not be overlooked that the city af- fords more opportunities for the exhibition of virtues as well as of vices, and " if our annals of virtue were kept as care- fully as our annals of vice, we might find that town life stood higher in the one than in the other." Every day the city witnesses the performance not merely of acts of generosity and self-denial, but of heroic self-sacrifice. Over against the 1 In using statistics of offenses to compare the moral conditions of different places, care must be taken to exclude such offenses as consist merely in a viola- tion of a local ordinance, e.g., neglect to clean a sidewalk of snow. PHYSICAL AND MORAL CONDITIONS 409 professional criminal is to be put the policeman ; against the roue, the fireman who uncomplainingly faces danger and death day after day. The records of city charitable soci- eties would reveal innumerable deeds of kindness, but would still leave unrevealed the thousand and one generous acts of service performed by the poor themselves for the relief of the unfortunate in their midst. CHAPTER VIII GENERAL EFFECTS OF THE CONCENTRATION OF POPULATION I. ECONOMIC EFFECTS Before the effects of the concentration of population in cities can be treated in their broad and general aspect, with reference to the nation or social body as a whole, it will be necessary to compare the economic condition of city-dwellers with that of rural workers. This can be done but imper- fectly by means of statistics of wages, cost of living, etc., be- cause averages, in the case of extremes, have little significance. An average height of 5 feet 6 inches would be the numerical mean between 7 feet and 4 feet, but would not be a true average since it approaches neither of the two men compared. So with regard to wealth, it is well known that the wealthy men of this country dwell in the great cities, and that the most degrading poverty is found in the cities. It is almost incredible that men in the country should suffer such depri- vation and come so near starvation through lack of employ- ment, as do masses of the urban population, — at least not in this country, where famines are unknown. In Russia or India, when the crops fail, a farm laborer may be reduced to the direst straits ; but wherever the modern railway has pene- trated, agriculturists as well as other classes are relieved from fear of starvation. Aside from famines, moreover, it is doubt- less true that the economic condition of the Polish Jews, Bohemians, etc., now living in New York tenements was con- siderably worse when they were living on their farms in C410) GENERAL EFFECTS OF CONCENTRATION 411 Europe and had the scantiest clothing, the most wretched shelter and most miserable surroundings. Bad as are the homes in crowded city tenements, they are an improvement. Still, one would say, comparing city and country as we know them in the United States, that the most hopeless poverty, as well as the most splendid wealth, are found in the cities. The Prussian income tax returns show that the income tax, which is levied on all single males or heads of families en- joying an annual income of 900 marks ($225) or more, is paid by a larger percentage of people in the city than in the country, — 38.4 per cent, as against 24.4 per cent.^ Among the taxpayers, the cities have a larger percentage among the higher incomes :' Table CLXII. Annual income Distribution of each i,ooo taxpayers in in marks. Cities. Rural districts. 900-3,000 429.9 679.5 3,000-6,000 156.4 124.4 6,000-9,500 85.5 44.7 9.500-30.500 1544 65.0 30,500-100,000 95.1 46.2 ioo,ooo-f 78.7 40.2 Total 1 ,000 1,000 Another method of measurement is to compare the wages of unskilled labor in the city and country. Such compari- sons have frequently been made, and show that the wages of this unspecialized labor are invariably higher in the cities.^ ' Bleicher, 146. "■' See " incoraes for city and country " in Alittheilungen aus der Verwaltung der directen Sleuern iin preus. Staat, " Statistik der Einkommensteuerveranlagung," 1892-3, Berlin, 1892, pages 308, 311. •^ The compulsory sick insurance system of the German empire grants an allow- ance corresponding to the local daily wage of unskilled laborers (" ortsiiblicher Tagelohn gewohnlicher Tagesarbeiter ") . The local rate was duly ascertained in every town in the empire, allowance being made in money for wages paid in "truck." From the first official revision (1S84) the following comparative sta- 412 THE GROWTH OF CITIES Similarly, comparisons have been made of wages in the various skilled trades, one with another. Here, especially, do city wage-earners maintain a superior rate, partly in con- sequence of the strength of city trade unions. What is the comparative cost of living in city and country? Does it tend to counterbalance the advantage of the city in higher wages? First, as to rent and personal services of all kinds, no doubt exists as to the disadvantages of cities. Villagers also secure lower prices in buying vegetables and such provisions as are brought to the market from the sur- rounding agricultural districts. But here their advantage over the city dwellers ends. The townsman buys his bread and meat and various other staple food-products fully as cheaply as does the villager, and far more cheaply than does the suburbanite. And in all other purchases, the townsman has an immeasurable advantage. Clothing, furniture, books and comforts and luxuries of every kind are offered in a var- iety and at prices in the cities that are not approached in the village store. In general, it may be said that the consumers' rent is much larger in the cities than in smaller dwelling-cen- tres. By " consumers' rent " economists mean the surplus of enjoyment that a man derives from purchasing an article at a price lower than the price he would be willing to pay in barter. The man with an income of $50,000 would no doubt tistics were compiled (Hirschberg, " Ergebnisse der fiir die Arljeiter-Kranken- versicherung vorgenommene Lohnstatistik in Preussen und den freien Stadten," yahrbiicher filr National-oekonomie und Statisiik (1S85), xhv, 265 : Daily wages (in marks) of adults. Males. Females. In cities of 100,000-f- 2.16 1.44 " " " 50,000-100,000 2.06 1.27 " " "20,000-50,000 1.77 1. 14 " " " less than 20,000 1.44 .94 Average i .46 .95 Additional statistics are given in the article " Arbeitslohn, Statistik " in Con- rad's Hdwbh. GENERAL EFFECTS OF CONCENTRATION 41 ^ be willing to give 50 cents or more for a loaf of bread; but in the market, the price has been fixed by the marginal buyer and seller at eight cents, and the rich man, as well as the poor man, can buy bread at that price. The consumers' rent of the rich man would in this case be designated by forty- two cents. It is the fierce competition of the great stores in the cities that lowers prices and secures to the city-dweller a large consumers' rent. The existence of high rents in the city may more than counterbalance the advantages of lower prices in staple articles, thus making the laborer's cost of living higher, and his actual wages lower than in the country. In the average case, a workingman expends from 12 to 15 per cent, of his income for lodgings.^ Hence when rents run at a figure that averages between 25 and 33 per cent, of the normal wages of an unskilled laborer, as they did at one time in Munich, Dresden, etc., or even at more than 33 per cent., as they are said to have done in Frankfort, Breslau and Danzig, Stettin and other German cities,^ there is bound to be much misery and suffering. Either the expenditure for rent will encroach upon the other necessary expenses, or else several families must live together in narrow quarters.^ There is reason to ^Engel's Law, formulated in 1857. Engel classified the expenditure of three groups of workmen in Saxony: (i) those with an income of ;^2oo-300; (2) in- termediate class, income of ^450-600; (3) well-to-do-class, $750-1,000. He found that each class expended 12 per cent, of its income for lodging. (Cf^ Marshall's Principles of Economics, p. 191.) l^he United States Bureau of Labor^s Report on Cost of Production, 1891, classified the expenditures in normal families, and found that the proportion of income expended in rent closely ap- proached 15 per cent, in ten out of the twelve classes. ^ Cf Schfinberg's Handbuch der Politischen Oekonomit, 3d ed., ii, 736. * The direct relation of income to overcrowding is shown by an investigation of 600 families in the New York tenement house district : No. of persons Weekly No. of persons to a family. wage. to a room. Families living in one room 3.3 $8.50 3.3 " " " 2 rooms 4.6 10.90 2.3 " " " 3 " 5.4 12.00 r.8 " " 4 " 54 16.50 1.35 — Report of N. V. Tenement House Com. of i8g4,\)-p.^2iZ-A- 414 THE GROWTH OF CITIES fear that too often both alternatives have had to be accepted. Add to this overcrowding, the negligence of the public authorities in regard to the construction of buildings and their sanitary condition, and you have the city slums, with their sickening odor of disease, vice and crime. Whose is the fault of the slums? and what are the reme- dies? are questions that do not demand extended discus- sion here. Society and the state are to blame for the worst features of the slums and in most civilized countries are now applying the remedies. Minute building regulations and careful supervision will prevent the erection of future death- traps ; strict inspection will prevent the use of a building, or part of a building, for a purpose other than the one for which it was built ; the requirement of a definite amount of air space to each occupant of a room will prevent some of the worst evils of overcrowding ; plenty of water, good paving, drainage etc., will render the sanitary condition good. The existing "death-traps" must be condemned and torn down. These measures, or most of them, have been taken in New York city, whose tenement house laws are probably as good as any. Additional legislation has been proposed, such as limitation on the right to migrate to a city whose housing accommodations are already insufifiicient, the prohibition of rent-taking above a legal maximum (like usury), and re- strictions upon the right of contract and the right to seize household goods in default of rent. These proposals are of German origin and have found little support in other coun- tries. To effect a reduction in house rents, however, is a less simple matter. In some places, notably in Germany, the overcrowding may have been due to a dislike on the part of the landlords to building and managing laborers' dwelHngs, which are subject to the annoyance of petty accounts, fre- quent changes of habitation and numerous losses. But GENERAL EFFECTS OF CONCENTRATION 415 where the workingmen are able to pay a reasonable rent, as they usually are in this country, the supply of laborers' dwellings will usually meet the demand unless there be mon- opoly sites.' Where the building space in the vicinity of a group of factories and other industries is limited, the law may fiot be able to prevent overcrowding without inflicting hardship upon the workingmen. Thus, in Glasgow after an era of reform in which strictly sanitary buildings had been constructed by the city on the site of old tenements, there were too few dwellings for the accommodation of those who wished to remain in the centre of the city, and overcrowding was almost an immediate result. In such a case, the only successful way out is an improved transit service which will permit more of the workingmen to reside in the suburbs. The London County Council has come to recognize this as the only really effectual remedy against overcrowding, and while it has secured Parliamentary Acts which will forbid the future erection of insanitary tenements, and has also devoted some attention to the condemnation and destruction of exist- ing " death traps," the Council's principal aim is the devel- opment of a rapid transit system between London and the suburbs. " Cheap trains for workingmen " is a rallying cry which has caused Parliament to abate the passenger-taxes in favor of railways that afford facilities for suburban, travel. In New York city, whose island situation prevents the popu- lation settling round the business centre in circular fashion, rapid transit is the only hope of keeping rents down. It is hardly worth while to set forth the statistics regarding overcrowding. The elaborate statistics of the Eleventh Census showing the average number of persons '^ to a dwell- ' There was no known scarcity of laborers' dwellings in Chicago, and no over- crowding, until the arrival of poverty-stricken immigrants from Southern Europe. ^ The average number of persons to a dwelling in the five principal American 4i6 THE GROWTH OF CITIES ing possess little value, as everything depends on the man- ner of construction of dwellings and on the relative number of lodgers (young unmarried men) in the town. Nor can the European comparison of overcrowding in various cities be regarded as entirely trustworthy. An instance in point is a recent essay on overcrowding by one of the most emi- nent statisticians in Europe.^ The English census of t 89 1 , however, affords a valuable comparison between urban and rural overcrowding. Re- garding as overcrowded all the " ordinary tenements that had more than two occupants to a room, bedrooms and sitting rooms included," the census statisticians found the cities in 1890, compared with the commonwealths in which they are situated, was : City. State. New York 18.52 6.70 Chicago 8.60 5.71 Philadelphia 5.60 5.26 Brooklyn 9,80 6.70 St. Louis 7.41 5.52 The statistics of European cities may be found in the manuals of population statistics. ^ J. Bertillon, Essai de statistique cotnparie du surpeuplement des habitations a Paris et dans les grandes villes Europeennes. Paris, 1895. '^^^ table referred to in the text gives the percentage of inhabitants living in overcrowded dwellings («. e., those logements in which the number of persons exceeds double the number of rooms — pieces^ : Paris 14 London 20 Berlin 28 Vienna 28 Moscow 31 St, Petersburg 46 Budapest 74 The Budapest statisticians said that Bertillon counted only the chambers among the pieces in Budapest, whereas he ought to have included kitchens and other rooms as he had done in the other cities. — -Cf. Eighth Inter. Cong, of Hygiene and Demography at Budapest, 18^4, vii, 425, where a r6sum6 of Bertillon's Essai is given. GENERAL EFFECTS OF CONCENTRATION 417 following percentages of the population to be living in over- crowded tenements : Urban Rural Size of tenements. sanitary districts. sanitary districts. 1 room 1. 61 0.25 2 rooms 4.42 2.48 3 " 346 2.83 4 " 2.82 2.90 Total 12.31 8.46 But with the recent improvements in city tenements, it is reasonably open to doubt if they are not cleaner and healthier habitations than thousands of dwellings in the country. We hear the most about city-tenements ; on them is focussed all the light of public attention. But no com- mittees have investigated rural dwellings, nor does the met- ropolitan press spy them out. From personal observation, however, the writer believes that for every ill-kept city tene- ment, there is at least one rural shanty in as bad or worse condition. The factory having been purified by the pressure of public opinion and legislation, the city-tenement house is now yielding to similar pressure. That the purely economic effects of the concentration of population are beneficial to society as a whole clearly follows from the fact that the movement itself is mainly in obedience to economic causes. Did it not result in the production of greater wealth, it would soon cease. Production increases with increasing density, and more particularly with increasing concentration, because there is opportunity for greater spec- ialization ; every man is placed where his strength and skill are exerted to the best advantage. Ambition has a wider field, and pre-eminent talent is more frequently brought to light. The statesman, from his acquaintance with the tax revenue returns, knows that cities, especially commercial 4i8 THE GROWTH OF CITIES cities, are the seats of wealth.^ And the economists have shown that the urban wealth redounds to the advantage of the rural districts as well, for, as Adam Smith long ago pointed out, the cities afford a convenient and profitable market for the rude products of the country. The volumi- nous reports of the recent British Parliamentary commissions merely serve to emphasize the statement, for they show that the only agricultural districts in England that have been able to endure the long agricultural depression without showing signs of distress, are those around London and the industrial towns. City markets stimulate intensive cultivation and im- provements in agricultural methods; while commerce and manufactures introduce order and good government, and guarantee the liberty and security of individuals in both city and country.^ Larger production, other things being equal, is identical with greater individual wealth, for distribution is merely the process of assigning to each worker the value of his pro- duct. If the average individual's share is expressed by the quotient of total production divided by total population, it must naturally increase when the dividend increases faster than the divisor, as it does in the case of agglomerations. 'James Lowe, author of the famous Present State of England (1821), con- stantly refers to the poverty of rural places and their small share of the taxes. The rich and powerful countries are those in which the concentration of population has gone farthest, e. g., the Dutch provinces. England is in proportion to popu- lation a richer and happier country than France; " in the size of her towns, this great kingdom, so long the dread of our forefathers and of Europe, has in the last present age been altogether surpassed by England and Scotland; for although our island boast only half her population, the distribution of it is made in a manner far more conducive to efficiency in a commercial and financial sense " (p. 217). '^ Wealth of Nations, Bk. iii, ch. 4. Smith also notes that the wealth acquired in cities is often employed in purchasing lands, and thus devoted to agriculture. But this ambition to enter the landed aristocracy was peculiar to England and some other European countries, and is now fading away. GENERAL EFFECTS OF CONCENTRATION 419 But normal distribution presupposes perfect competition, which in turn depends a great deal on the legal norms. When, for instance, laborers are forbidden by law to unite in associations for their own protection, the competition among employers for their services is imperfect. Given a tacit agreement among them not to pay more than a certain wage, and the weakening of a single workingman will establish an abnormally low wage for the entire body of laborers. States- men have learned that laissez-faire is the last thing to secure the competition on which all economic laws are based. On the other hand, there are many laws which interfere with this perfect competition. Among these are restraints upon freedom of movement,' which is one of the greatest factors in bringing about competition and justice in distribution. One of the conditions of the high wages prevailing in the United States is, without doubt, the unhampered ability of a laborer to migrate wherever his best interests lead him.^ Now with just public laws supplementing economic law in the regulation of the process of distribution, the concentra- tion of population is particularly favorable to the working- men. It gives every man the chance to show "what is in him." Moreover, and here is the strongest point, a dense population is the most favorable to strong organization. The trade union movement, which has been a conspicuous force in improving the condition of English workingmen in the 19th century, (not so much, perhaps, from the economic or materialistic standpoint, as from the moral, intellectual and educative standpoint,) would have been impossible without the association of large numbers in the cities. The trade union is in fact the only hope of those who have seen materialism prevail over spiritualism ever since the disrup- ' But there is considerable justification for such an exception to this rule as the United States contract labor immigration law. » Cf. Walker, Wa^es Question, pp. 178-88. 420 THE GROWTH OF CITIES tion of the familiar and friendly relations of master and em- ployee by corporations and the system of centralized indus- try. And the trade union is peculiarly a city institution. The movement towards the cities is therefore justified from the economic point of view, provided it does not go too far. But what is the limit? In the first place, it must be recognized that such a movement may continue after the forces that generated it have ceased to act. The reputation that cities enjoy for the payment of high wages may attract laborers after the adjustment between urban and rural rates of wages has been effected. The information on which men act may be misleading. Or men may over-estimate their prospects of success, just as they do when they flock to the gold fields. As a matter of fact, the average income in a mining camp is almost always small ; it is even affirmed that less wealth has been taken out of the Klondike than men took into it. But men went there on the expectation of ac- quiring wealth. Moreover, it is possible that a goodly por- tion of the migrants to cities may not act on economic mo- tives at all, but rush to the cities out of general discontent and a desire for change, rather than from any real inferiority in their economic rewards. The result would be superfluity of labor in the city and scarcity in the country. It is hard to say whether this condition actually exists. We hear a great deal at times about the masses of the un- employed in our cities ; but this is occasional and comes with commercial crises, which seem, unfortunately, to be a necessary part of the modern industrial system with its sep- aration of producer and consumer, its rapid dynamic move- ment and its substructure of credit. In a time of ordinary business prosperity there are not many capable men perma- nently out of work in the cities. The complaints of pauper- ism result from the influx of tramps and country good-for- nothings, who are attracted by reports of extensive city GENERAL EFFECTS OF CONCENTRATION 421 charities. As a matter of fact, the number of really efficient workers who cannot find remunerative employment in our great cities is small ; the testimony of those who have been connected with charity organization societies or with private employment bureaus will, it is believed, entirely substantiate this opinion.' Nevertheless there is a widespread opinion in many coun- tries that the movement has proceeded too far ; that so long as the city received only the best blood of the rural districts, the cream of rural youth, it was a healthful tendency. Now, however, it is thought that the migratory current sweeps along vast numbers who are not adapted to town life.'' This lower grade of labor (represented in American cities by the foreign-born) is said to imperil the standard of life of city laborers, undermining their forces in the battle with capital. But there is another side to the picture, and before govern- ment or society attempts to put up the bars to the migration ' Messrs. R. Fulton Cutting and Walter L. Suydam, in reporting to the New York Association for Improving the Condition of the Poor on the subject of " Agricultural Depression in New York State " (^Lea/let A'o. I ), came to the con- clusion that " some attempt should at least be made to check that tide of migra- tion to the city that threatens to make the condition of multitudes there quite in- tolerable It is quite true that as far as regards the difficulty of obtaining employment in New York City for able-bodied men, we are far better situated than is London or many continental cities; yet we have a vast number of the very poor against whom the door of material improvement is well-nigh closed. Unskilled labor is always travelling too near the dead-line of dependence. The intermission of employment and occasional enforced idleness from sickness, coupled with the terrible rent-charge in New York city, makes it well-nigh im- possible for the common laborer to save anything from his earnings " (p. 17). But this was written at a time of business depression (1896). - Fide Graham's Rural Exodus, p. 2, and passim. Kingsbury (" The Ten- dency of Men to Live in Cities," an address at the i S95 meeting of the American Social Science Association, published in the Journal of the Association) observes that the newspaper advertisements for boys as clerks no longer read " one from the country preferred" as they did fifty or sixty years ago; which may signify either deterioration on the part of the rural migrants or improvement in city youths. 422 THE GROWTH OF CITIES from country to city, it should consider the advantages which cities possess for assimilating elements that could not be utilized elsewhere, and of educating or taking care of the more helpless persons. On this point the conclusion reached by Dr. Devineinhis discussion of the "Shiftless and Floating City Population " must be accepted by every one who really stops to consider the matter:^ "Taking into account the national interest as a whole, the city is a better and less dangerous and less expensive place for the vagrant than the country. His migration to the city should be welcomed rather than discouraged. If he is in the city we shall be more conscious of his existence, but for that very reason we shall be better able to deal with him. There is greater tax- able wealth and therefore greater resources for charitable relief and for correctional discipline. The whole of the re- pressive and remedial work can be done more efficiently and with better opportunities to watch the results than in the country." On the other side, it is possible that a scarcity of labor may exist on the farm. Professor Sering affirms that in the agricultural fields of East Prussia, which are being depopulated by emigration, harvests have rotted for lack of labor.^ And the extensive investigations of the German Socio-political Association show that the wages of farm hands have risen to a height that makes it impossible for many farmers to em- ploy the labor they need.3 It is complained that there is also a scarcity of farm servants in England,* but such is not the finding of the Royal Commission on Agriculture. In an ^ Annals of the Am. Acad. (Sept., 1897), x, 159. ■^ Die innere Colonisation im ostlichen Deutschland (vol. Ivi of the publications of the Verein fur Socialpolitik) . ^ Die Verhaltnisse der Landarbetter in Deutschland, Schriften des Vereins filr Socialpolitik, vols, liii-lviii. * Graham, Rural Exodus, p. 20. GENERAL EFFECTS OF CONCENTRATION 423 investigation into the causes of agricultural depression in New York State it was found that the scarcity of good farm labor played a part ; ten per cent, of the replies to the ques- tion, " What is the cause of the tendency among farmers and their families to leave their farms and live in towns and cities,' assigned as a cause " the difificulty of obtaining good help in the house and on the farm."' Now it may well be that urban expansion has at times out- run the growth of the contributory territory, so that the cities have become swollen with a surplus population without employment. This condition has been sorely felt in Austra- lia, where vast government works have been completed and the laborers, temporarily thrown out of employment, have remained in the cities. But as to the alleged scarcity of labor in the agricultural districts being due to an excessive rush to the cities, it is sufficient to observe that the very provinces in Germany that make the loudest complaints have sent the largest number of emigrants across the ocean. Clearly, the migratory movement is not called into being by the cities alone ; there must be dissatisfaction at home to cause such an outpouring across the sea as well as toward the city. The spirit of adventure, or the pressure of subsist- ence, or some other impelling motive, induced this great movement ; not the mere attraction of the city on well-paid, comfortably-housed agricultural laborers. In the United States there has always been a relative scarcity of good farm labor, from the fact that most men de- sired to own their farms and were usually able to do so on account of our vast domain. To-day, the depopulation of the rural districts in the East is caused not less by the migra- tion westward than by the movement toward the cities. As to England, evidence was presented in the second 1 A. I. C. P. Lea/let, cited above, p. 8. 424 THE GROWTH OF CITIES chapter of this essay ^ to the effect that no real depopulation of the rural districts had taken place. The conclusions of Drs. Ogle and Longstafif are confirmed in the Report of the Royal Commission on Labor, which states positively that rural emigration proceeds without reference to the local rate of wages, whether labor is relatively scarce and wages high or the opposite, =" It is worse than useless to attempt to stem the current of emigration from distressed farming districts, since migration is the one efficient remedy for such distress. And it should be thoroughly understood that a large migration from the country to the city is a perfectly natural phenomenon. Sir James Steuart has analyzed the movement in his excellent treatment of the distribution of population. "What occa- sion," he inquires, "has the country for supernumerary hands? If it has enough for the supply of its own wants and of the demands of the cities, has it not enough? Had it more, the supernumeraries would either consume without working, or if added to the class of laborers instead of being added to the number of free hands, would overturn the balance between the two classes ; grain would become too plentiful and that would cost a general discouragement of agriculture, whereas by going to the cities they acquire money and therewith purchase the grain they would have consumed had they remained in the country ; and this money which their additional labor in cities will force into circulation would otherwise have remained locked up, or at least would not have gone into the country but in conse- quence of the desertion of the supernumeraries. The proper and only right encouragement for agriculture is a moderate and gradual increase of demand for the produc- tions of the earth, . . . and this demand must come from ^ Supra, p. 45. ^ Fifth Report (W. C. Little) in Pari. Documents, 1894, xxxv, no. GENERAL EFFECTS OF CONCENTRATION 425 cities, for the husbandmen never have occasion to demand ; it is they who offer for sale." ^ II. POLITICAL EFFECTS. The changes in the distribution of population which have been considered in the present paper have necessarily- effected changes in State and national politics and national power. In the first place, it will be observed that the causes of concentration are forces which augment national wealth. Compare the distribution of wealth in the United States at the present time and in 1787. In the Constitutional Con- vention it was held ^ that the distribution of wealth was so even throughout the country, that any system of taxation might safely be based on numbers. How diiferent in 1898 ! The rural population is not less wealthy now than it was a century since, but the urban population has amassed incal- culably greater wealth. Had England remained an agricultural country without commerce and cities, she would not now be the powerful and wealthy state that she is ; and in saying this, it is not for- gotten that national power depends not so much upon wealth as upon manhood. But the controversy as to the relative fighting capacity of the townsman and the countryman is after all an idle one. Man for man, it is possible that the agriculturists might be able to overpower the industrials, though even this is very doubtful, as we saw in the last chap- ter. But on a given area industry will support so many more men than agriculture will, that the former would easily triumph in an armed conflict. A German student has recently shown that the agricultural counties of England ' Sir James D. Steuart : An Inquiry into the Principles of Political Economy being an Essay on the Science of Domestic Policy in Free Nations ; in Works, i, 70. London, 1805. - I'.liot^s Debates, v, 297 ff. 426 THE GROWTH OF CITIES merely doubled their population in the period 1 801-91, and even this increase depended in part on the presence of wealthy consumers in the cities/ Hence had England re- mained an agricultural country she would now have a popu- lation of about 16 million, instead of 28 million. The terri- tory of England could not afford support to 28 million peo- ple producing their own food supply. The policy of Ger- man statesmen who would keep Germany an agricultural country for the sake of her army is therefore seen to be mistaken. Of even more fundamental importance than national power is national stability. Anything affecting the constitution of the electorate must be of great interest; anything that di- minishes the elector's love of country or interest in its pres- ervation must excite distrust. Now land ownership has in the past been recognized as the most important conservative force in politics, and the statesman's ideal — even in England, the land of great landlords — has been a country of small landowners like France. But with increasing concentration of population goes an increase of tenancy, both as regards land and dwelling houses. The American statistics concern- ing the private ownership of dwelling houses show the fol- lowing proportions of tenants in 1890:* Per cent. Farms 34.o8 Rural population 56.22 Cities of 8,000-100,000 64.04 " " 100,0004- 77-^7 United States 52.20 ^ J. Golcistein, Berufsgliederung urid Reichtum (Stuttgart, 1897), P- ^^' '^ ^" interesting monograph on the subject. It contains a full bibliography. On the opposite side of the question, see von Bindewald, " Eine Untersuchung iiber den Unterschied der Militartauglichkeit landlicher und stadtischer Bevolkerung," in Jahrb. fiir N.- O. und Statistik, Ixx, 649 seq. ^ Abstract of the uth Cens., 223. Holmes, " Tenancy in the United States," Quar. your, of Econ., x, 37, GENERAL EFFECTS OF CONCENTRATION 427 Several cities exceed the average of 77.17 percent, for the great cities ; thus New York's percentage was 93.67 (Greater New York about 85), Brooklyn, 81.44; Jersey City, 81.20; Boston, 81.57, ^tc- Rochester (56.02) and Milwaukee (57.87) make the most favorable showing. It may be worth while to note that the percentage of tenancy in Berlin (96.65) exceeds that of any American city. Thus statistics show that the ownership of the home be- comes less common in the degree that we leave the farm and village and proceed up (or down) the scale to the great cities. We do not wish to minimize the importance of this fact from the political point of view, and yet we must remember that there are other forms of property than real estate. If a man who rents his home has a good bank account, he is not likely to vote for the overthrow of government.^ The extensive distribution of government bonds among French citizens is felt to be an influence favorable to conser- vatism scarcely inferior to the ownership of the soil of France by millions of peasant proprietors. And the American election of 1896 proved that the land-owner is not necessarily a conservative, nor the city man necessarily a radical. For whatever our opinion as to the merits of the controversy, it must be conceded that the cause supported by the farmers involved radical changes, while the city populations voted to conserve the status quo. Nevertheless, the danger of class antagonism is particu- larly grave in the cities. Dives and Lazarus become figures too familiar to let us rest in peace. The chasm created by the industrial system yawns widest in the cities ; and the means of bridging it will require careful consideration. In internal politics the changes causing or accompanying ^ And it may be observed that England's government has been more stable than any continental government, notwithstanding the concentration of its landed property in a few hands. 428 THE GROWTH OF CITIES the concentration of population have brought many new problems to the front, of which the question of taxation is a good specimen. As previously noted, wealth followed pop- ulation so evenly in the United States of 1790 that even a per capita tax would have been equitable, while the general property tax generally adopted gave perfect satisfaction. But the general property tax has utterly broken down in the last few years on account of the growth of other forms of wealth than immovable property.^ And so it is generally ; the complexity of modern city civilization demands new laws and new policies. And the most conspicuous problem is the difficulty of governing the cities themselves. . Aristotle saw the difficulty of governing a vast agglomera- tion of people, and limited the population of his ideal city- state to 10,000.^ There is, indeed, a vast difference between the government of a city and that of a village. A dense population engenders problems that are never thought of in a village. Run the eye over a directory of the public offi- cials of a great city and observe how few of them are known to village governments. Building departments, paving, fire, health, park, public improvement, library, public building boards ; city auditor, corporation counsel, city architect, surveyor, superintendent of markets, sewers, bridges, print- ing; inspectors of milk, provisions, lime, petroleum; a sealer of weights and measures, etc., etc. It would take pages merely to enumerate the officials of New York city, the majority of whom perform functions to which no parallel is found in the village. The complexity of a city government, the multifariousness of its duties, make it the most difficult kind of government to watch. Even the national government does not under- take to regulate so many details, and the general supervision ' Seligman, Essays in Taxation, chs. i and ii. ^ De Republica, 1. 7. GEN{ERAL E'FFECTS OF CONCENTRA TION ^-jq A to which it is\ mainly limited can be more easily watched. The city is in riaany respects a great business corporation ; it calls for a careful, systematic, business-like administration. Now, administration is a branch of politics in which Ameri- cans have hitherto shown more awkwardness than is pleasant to think of. Cc uption if not inefficiency has been the characteristic mj.^_. of public administration, especially in the cities. And the reason why Americans have submitted to such service, has been its small sphere, as compared with the sphere of local {i. e. rural) government, which has been the strength of our democratic institutions. As Mr. Bryce has said : " Americans constantly reply to the criticisms which Europeans pass on the faults of their State legisla- tures and the shortcomings of Congress by pointing to the healthful efificiency of their rural administration, which enables them to bear with composure the defects of the higher organs of government." ' Now it is obvious that with the rapid moveTient of population from the rural dis- tricts to the cities, the sphere of local rural government (typified in the towntmeeting, the glory of New England and New England's sons in every part of the Union) has been continually narrowed. With only 25 or 35 per cent, of our people residing in *:he rural districts (Massachusetts), the "healthy efficiency of rural administration" signifies but little. The difficulties of city government (" the one conspicuous failure in American politics ") are enhanced by the large floating population which is a necessary accompaniment of a great migratory movement toward the cities. The thou- sands of new residents are strangers to the city's history and traditions, h?.ve no local attachments, and do not readily acquire any civc pride. The vast majority are non-tax- payers, and fee) little concern in the city's government. ^\imerican Commonwealth, 2d ed,, 1,591. 4^0 ^^^ GROWTH OF CITIES The ignorance of local history and geogn\phy is in fact almost appalling, and our munic^alities have but just begun to provide instruction on these subjects, which may in time awaken civic pride.^ On top of these obstacles to good government comes the problem of assimilating the foreign natio lities. When the foreign immigrants settle in isolated " qu^^**:;rs," which is the natural tendency, much effort is required to raise them to the city's standard. This is perhaps thr one danger of the " movement toward the cities " so far as the United States is concerned : the influx of a shiftless ar d degraded popula- tion from foreign lands, which cannot b-: readily distributed throughout the country. Inasmuch as this is not a treatise on municipal govern- ment, no discussion of proposed reforms is in place. But one thing is to be strenuously insisted upon, and that is the right of the cities to self-government. The strength of our political institutions has always been in local government, and the only hope for our cities is freedom to work out for themselves a plan of government which shall take the. place of that rural local administration that has been our boast in times past. "The problem of modern jtimes is how to make life possible in large cities devoted to industrial activities, and this is a problem which cannot be dealt with except by the cities themselves." The only hope for the cities is to educate the mass of the propertyless, and this will never be accomplished until the liberal and genero^is minds of the city have the assurance that their work is not liable to be undone at any moment by the State legislature. ' New York and other large cities have popular lecture .ourses in which these themes are sometimes handled. Many of the cities also have societies of large membership devoted to the investigation of local history aid propagation of re- sults. The city of Brookline, Mass., has recently prepared a text-book for its school children, which is a local geography, a botany, a gk>logy, a history, and a treatise on civil government as related to that town. GENERAL EFFECTS OF CONCENTRATION 431 Amidst the discouragements incited by a contemplation of the failure of our city governments to achieve anything like the success of American rural local government, we may derive some small consolation from the fact that things are not now so bad as they used to be. Let us read De Tocqueville's description of our cities in the thirties, and take fresh courage to renew the struggle for municipal reform. Says the illustrious author of Democracy in America (Reeve's trans. 2d Am. ed., p. 270) : "The United Stales have no metropolis; but they already contain several very large cities. Philadelphia reckoned 161,000 inhabitants, and New York 202,000 in 1830. The lower orders which inhabit these cities constitute a rabble even more formidable than the populace of European towns. They consist of freed Blacks in the first place, who are condemned by the laws and by public opinion to an hereditary state of misery and degradation. They also contain a multitude of Europeans who have been driven to the shores of the New World by their mis- fortunes or their misconduct, and these men inoculate the United States with all their vices, without bringing with them any of those interests which counteract their baneful influence. As inhabitants of a country where they have no civil rights, they are ready to turn all the passions which agitate the community to their own advantage; thus, within the last few months serious riots have broken out in Philadelphia and New York. Disturbances of this kind are unknown in the rest of the country, which is nowise alarmed by them because the population of the cities has hitherto exercised neither power nor influence over the rural districts. " Nevertheless, I look upon the size of certain American cities, and especially on the nature of their population, as a real danger which threatens the future security of the democratic republic of the New World; and I venture to predict that they will perish from this circumstance, unless the Government succeeds in creating an armed force, which, whilst it remains under the control of the major- ity of the nation, will be independent of the town population and able to repress its excesses." III. SOCIAL EFFECTS Having considered the economic and political effects of the concentration of population, we may now conclude with an estimation of the social effects upon urban and rural com- munities, and a general summary from the point of view of society as a whole. 432 THE GROWTH OF CITIES As to the cities themselves, we have just noted how good government and even social solidarity are threatened by- class antagonisms. The actual cause of such social antipa- thies will be found in an exaggerated individualism, which has been developed by an era of industrialism, out of mediaeval militarism. The new industrial forces which transformed the solidified Age of Authority into a liquefied Age of Freedom, have naturally been more predominant in the cities than elsewhere, for the close contact of man with man in a dense population removes prejudices and engenders liberalism. The cities have always been the cradles of lib- erty, just as they are to-day the centres of radicalism. Every man of the world knows that isolation and solitude are found in a much higher degree in the crowded city than in a country village, where one individual's concerns are the concern of all. The cities, then, are favorable to free thought and the sense of individual responsibility.^ But it is a question whether the loosening of the ties of individual responsibility has not gone too far. " The great danger to morality and good government," says Roscher, "is that the individual is lost in the multitude of atoms, — a con- dition that may abolish the sense of duty and make the great city as insecure as the opposite extreme, the wilderness."'' Now this extreme individualism of the cities is merely one manifestation of the — shall we say excessive — fluidity of modern society, and its cause is chiefly industrial. Cities vary in their lack of social feeling {i. 633,332 1895 " " " " «' " est. 6,048,555 A fair estimate of the present population of the metropol- itan police district, which includes every parish within 12 miles of Charing Cross, would be six and one-half million souls. And the New World agglomerations are following close after London. A careful estimate by Dr. Roger Tracy, reg- istrar of vital statistics in New York city, gave the present New York on January i, 1898, a population of 3,388,771 ; to which should be added Hudson county, Newark and Elizabeth, New Jersey, making a grand total of 4,029,517, "Greater Paris" had over 2,700,000 already in 1891 and now has at least 3,000,000; while "Greater Berlin" on January I, 1896, was credited with a population of 2,666,000. Even Boston, which ranked sixth among American cities in 1890, with a population of 448,477, was credited with 1,004,424 ' Ibid., p. 23. 454 THE GROWTH OF CITIES at the State census of 1895 in the territory within a radius of 12 miles of the State House.' The advantages and disadvantages of such concentration of population have been discussed at length in a previous chapter. The discussion of remedies began at least twenty- centuries ago, and will perhaps continue twenty centuries hence. Plutarch's warning against the overgrowth of the great cities^ and Cicero's constant effort 3 to turn back the current of emigration from the country alike came to nought. Justinian tried to stop the current by legal meas- ures/ and mediaeval statesmen and monarchs followed a sim- ilar course. The extension of Paris beyond certain limits was prohibited by law in 1549, 1554, 1560, 1563, 1564 and 1672.5 In the time of the later Tudors and Stuarts, procla- mation after proclamation was issued forbidding the erection of new houses in London and enjoining the country people to return to their homes.^ There were many good reasons for such action — the difficulties of municipal government, the fear of local pressure on Parliament, the difficulty of pro- viding an adequate food-supply ^ and water supply, the danger of fires (the Great Fire of 1665 !), and especially the danger- of plagues and epidemics arising from insanitary conditions. The evils enumerated in the Act of 1593^ are almost identical with those depicted in the recent report of the New York tenement house commission. " For the reforming of the ^ op. cit., pt. i, p. 47. ^ PrcEcepta Politica, ' Ad. Ait., i, 19. * Pohlmann, Uebervolkerung der antiken Grossstddte, 169. ^ Roscher, System der Volkswirtkschafi, iii, 39. ''Cunningham, Growth of English Industry and Commerce, ii, 172. ' One of James T's proclamations was issued " on account of the present scarcity and dearth and of the high prices of corn and grain." * 35 Eliz., c. 6. TENDENCIES AND REMEDIES 455 great mischiefs and inconveniences that daily grow and in- crease by reason of the pestering of houses with diverse families, harboring of inmates and converting of great houses into several tenements or dwellings, and erecting of new buildings within the cities of London and Westminster and other places near thereunto, whereby great infection of sick- ness and dearths of victuals and fuel hath grown and in- creased," it ordained that no new buildings should be erected (except for inhabitants of the "better sort,") and that houses should not be broken up into tenements, etc. While legislative prohibitions of city growth are now a thing of the past, it has been seriously proposed in Germany^ to check the overflow of rural laborers into the great cities by means of settlement fees. Any such proposal to limit individual freedom of movement would not be entertained in Anglo-Saxon communities. Not less hopeless are the schemes promoted by agricultur- ists to make farming more attractive and more remunerative with the help of scientific cultivation, allotments for laborers, and various legislative measures of relief.^ In themselves such measures are often praiseworthy, as is every plan of improving the moral and material condition of farmers.3 But it is idle to hope that the adoption of any of these plans will stop the drift of farmers' sons cityward. The production of the world's food-supply calls for a definite amount of labor ; any project for increasing the per capita product of agricul- tural labor simply releases a certain amount of such labor for other occupations. From the same point of view are to be judged schemes of colonization of the city poor on farm- ' By Roscher and others. Cf. Art. " Wohnungsfrage " in Conrad's Hdwbh. 6 : 751, and 7: 21. - Cf. Leaflet No. i of the New York Association for the Improvement of the Condition of the Poor, on " Agricultural Depression." *Cf. Emerick, " An Analysis of Agricultural Discontent," in Pol. Sc, Quarterly, vol. xi. 4^5 THE GROWTH OF CITIES ing lands. Aside from the difficulty of getting the tenement classes permanently away from the city/ and the question- ability of trying to make farmers out of men who fail in other trades, the world already has all the farmers it needs. Col- onization of the city poor may, indeed, be the salvation of individuals ; but it simply necessitates the transfer of other people in corresponding numbers from the country to the city.'' Another scheme of stopping migration cityward is to make village life more attractive. There is certainly opportunity for work in this direction. But so long as the present indus- trial organization endures, no amount of village improvement will keep ambitious youths at home, for the reason that all the opportunities for rising in the world are in the cities. If domestic industries could be re-established, villages would soon pick up. Efiforts are making on the part of philan- thropists to put out portions of their work in the country, but the success attained has been exceedingly limited, for employers have to contend not only with the irregularity of country labor, but with the hostility of city trade unionists, who resent the policy. In analyzing the conditions of pro- duction, we found little encouragement for the hope that improvements in electric motors would bring about decen- tralization of industry. It is a mistake to regard this as " at present the most hopeful method of withdrawing the pressure from our large industrial centres." 3 Still another remedy proposed is administrative decentral- ization,* the building up of rural self-government and the ^ Supra, p. 221. * Altogether different is the plan of sending city children into the country for part of the year. Such schemes as the George Junior Republic have as their ob- ject the training of citizens, not simply the making of farmers. " McKenzie, Introduction to Social Philosophy, io8. •A popular cry in England. Cf. Stephens, Rural Administration, 1896. TENDENCIES AND REMEDIES 457 removal of garrisons and government offices to villages or small towns. But strategic reasons compel the concentration of military forces in cities, and the tendency toward consoli- dation is antagonistic to the dispersion of government offices. On the other hand, it cannot be doubted that the conferring of more responsibility upon village officials than they enjoy at present on the continent of Europe, would afford a field for the ambition of men who now have to re- move from the village if they desire to enter politics. Perhaps some of the many gratuities (such as hospitals and medical service) in cities might be restricted with ad- vantage. There are those who advocate an abrupt discontin- uance of all public improvements in the city, lest they attract more migrants from the country; such persons would prefer to have the city remain a mud-hole. They are to be found in the class that can own country homes and thus escape city dangers. Is there, then, " no remedy until the accumulated miseries of overgrown cities drive the people back to the country?" One remedy is to admit the harmful tendencies of city life, to fight city degeneration on its own ground, and free city life from as many ills as possible. This work is now pro- ceeding on a vast scale, and in a vast number of ways. Pri- vate philanthropy and public supervision go hand in hand. Not only complete drainage, paving, water-supply, inspection of food, etc., are required from the municipality, but also small parks, playgrounds, public baths and laundries, and a variety of other institutions. A vast deal has been accom- plished in this line, and the work is only begun. Much may be expected from the progress of invention and discovery and the growth of capital. Prof. Marshall indicates how certain improvements (some of which have already been made in American cities) would " enable a large part of the popula- 458 THE GROWTH OF CITIES tion to live in towns and yet be free from many of the pres- ent evils of town life. The first step is to make under all the streets large tunnels, in which many pipes and wires can be laid side by side, and repaired when they get out of order, without any interruption of the general traffic and without great expense. Motive power, and possibly even heat, might then be generated at great distances from the towns (in some cases in coal mines) and laid on whenever wanted. Soft water and spring water, and perhaps even sea water and ozonized air, might be laid on in separate pipes to nearly every house; while steam-pipes might be used for giving warmth in winter, and compressed air for lowering the heat of summer ; or the heat might be supplied by gas of great heating power laid on in special pipes, while light was derived from gas especially suited for the purpose, or from electricity; and every house might be in electric communi- cation with the rest of the town. All unwholesome vapors, including those given off by any domestic fires which were still used, might be carried away by strong draughts through long conduits, to be purified by passing through large furnaces and thence away through huge chimneys into the higher air."^ But while much is to be expected in this direction in the near future, the most encouraging feature of the whole situ- ation is the tendency, heretofore alluded to in the present essay, toward the development of suburban towns.'' The ^Principles of Economics, 3d ed., p. 305, note. ^ In Vienna the suburbs in close connection with the city itself have long had a rapid growth. After 1870, with a population equal to about one-third of the city, they had larger increments of increase than the city itself. In 1891 they were in- corporated into the city having a population of 464,110 as compared with 798,719 for the old city (figures for 1890; cf. Rauchberg, in St. Man., xix, 140 ff). Sim- ilar statistics might be given for the Saxon cities : Dresden, for instance, in TENDENCIES AND REMEDIES 459 significance of this tendency is that it denotes, not a cessation in the movement toward concentration, but a diminution in the intensity of concentration. Such a new distribution of population combines at once the open air and spaciousness of the country with the sanitary improvements, comforts and associated life of the city. The question is, however, whether the marked growth of the suburbs is because the cities are already too full to hold more, or because the populations of the congested districts are overflowing into the suburbs and thereby leaving a more tolerable condition for those behind. Hon. C. D. Wright, Superintendent of the eleventh census, takes the latter view and presents the following statistics in substantiation thereof :^ Increase or decrease per cent, of population, 1870-90. New York.^ Philadelphia. » Boston.* Congested wards 9.38 — 6.56 16. Remaining " 131-56 168.91 156. Entire city 60.81 55.33 76. The only one of the three cities in which the crowded dis- tricts actually lost population in 1870-90 was Philadelphia. creased by 56 per cent. 1871-90, while 28 of her suburbs increased by 233 per cent. •? Increase, or decrease. Population, per cent., 1870-90. 1890. 28 towns under 2,000 — .3 39j09I 93 towns 2,000-10,000 -i-26.4 454,910 22 towns over io,oco •\-^l-'^ 987>46o 150 suburbs of the 22 cities +146.3 462,575 1" Urban Population," in Popular Science Monthly, xl (Feb., 1892), p. 463. 2 The congested wards are 1-17, with the exception of 12, which is in the upper part of the city; this includes that part of the city below Fourteenth street and one ward (16) above. 'Congested wards, 2-20, except 15. * Here the " congested wards " designate Boston proper, and the " remaining wards " the annexations. ^ Cf. Lommatzch, Die Bewegung der Bev'olkerung Sachsens, pp. 24-8. For growth of the cities and suburbs, 1834-75, vide Zeitsckrift des Kngl. Siichs. Stat. Bureaus, 1876, p. 302. 46o THE GROWTH OF CITIES Table CLXIV. Population per acre in New York City wards.' Ward. i860. 1880. 1890. I -f-Ji7-S 116.5 72.2 2 +30-9 Jf9-8 ".4 3 39-5 37-7 +39-6 4^ 264.9 252.9 214.5 5 •• +132.9 94.3 73-7 6 +310.4 233.6 268.8 7 201.9 252.9 -1-289.7 8 +215.3 196.0 170.6 9 137-8 +169.5 169.0 10 272.7 432.3 +523-6 " 303-9 350-9 +384-3 12 5.5 14.8 +44.5 13 307-6 353-2 +428.8 14 292.5 +314-3 292.6 15 139-3 +161.0 128.3 16 129.4 +149.5 140.8 17 220.4 +316.7 3"-6 18 127.7 +148.0 140.6 19 22.1 106.8 +158.5 20 152.0 +193-7 189.9 21 119-5 +161.8 153.3 22 40.3 72.9 -|- 100.6 23 6.6 +12.6 24 1.6 +2.5 Table CLXIV affords means of analyzing New York's growth more carefully ; it shows the average number of in- habitants per acre in each ward at the censuses of i860, 1880 and 1890, the years in which the greatest density (/. ro ^ r^ o VO vO in ro o VO VO r^ N cn VO i-i ro HH Tt- t^ H^ HH !>. ^ t^ N oo Ov O o 00 vq_ VO ro ro r^ l^ m ro m o cT HH N 00 ro 00 CO ri vd CO •o VO 5 t^ VO •* in •^ vO Tj- m C) 00 Ov M ►H M ■* ro t-~ M M t^ VO ro Tt- VO Ti- Ov l^ CO •^ ro N t^ t>» r£> HI O o ro O fO ro N '^J- n- ro t^ ■. VO o ro o ro Tj- M N VO o O; 00^ 1^ o ts. ro lO ri ro cfv O 00 in 00 00 M Ov > , — .. O cq m r~. t^ 00 m 00 ro ro m ^ ro in ov od in VO qv d VO in l-^ OT i^ d" c^ N 00 6 OS H-t '^ r^ x> ro 00 CVi m N m Ov vO O VO 00 O m M r) ■+ o t>. t^ VO •^ o ■vj !S ^ ^-N 5 »> U) o s ^ ^ a B m HI terril e wall 1) 13 1) _o 1) .2 X U5 5i in 73 "o a. O S-i O T3 O -g < s 5 ^ CO "c TENDENCIES AND REMEDIES ^^y (within a radius of one kilometre of the city hall) was 32,589, while in the second ring (radius, 1-2 kilometres) it was 54,- 024.' Similarly in Vienna, the respective figures of density in two concentric districts were 25,154 and 38,894.^ A strik- ing example of the tendency is revealed in the statistics of Hamburg: 3 Population, in thousands. 1867. 1880. 1890. Inner city 157 171 161 Remainder 64 n6 158 15 suburbs 45 120 245 The process thus sketched for New York, London, etc., is known as " city-building." The original settlement becomes the business centre and for some time continues to grow rapidly. But if the city prospers, the time will come when this old centre is more and more needed for strictly business purposes; houses disappear before the march of office-build- ings, government buildings, banks, etc., until the only resi- dents left are the janitors and portiers, the keepers of the great buildings.'^ With continued growth, the business cen- tre extends itself and steadily pushes the dwellings toward the circumference, until at length the municipal limits are reached and passed. American cities are not so compactly built as European cities. On the continent especially, where it is still the prac- tice to live in rooms connected with one's store or workshop, ^ Die Bevolkerungs- tmd Woknujtgsaufnahme votn i Dez. i8go in der Stadt Berlin, i Heft., Berlin, 1893, p. xiv, ff. *See the article of E. Hasse, entitled " Die Intensitat grossstadtischer Men- schenanhaufungen, Allg. Stat. Archiv, ii, 615 ff. It is only the very largest cities, as Hasse's investigation shows, which have reached this condition. In Paris several of the central arrondissements have been losing population. (Cf. Meuriot, op. cit., ch. xii, where the subject is more extensively treated.) '^ Statistik des Hamburgischen Staates, Heft xvi, Census of 1890. * In 1 85 1 there were 14,580 inhabited dwellings in London City; in 1881,6,493. 468 THE GROWTH OF CITIES the density of population is remarkable in comparison with American cities, thus:^ No. of cities Area in Population considered. acres. Total. Per acre. United States 28 638,235 9,670,000 15.2 England 22 231,150 8,840,000 38.3 Germany 15 193,290 5,000,000 25.9 Almost as many English urbanites dwell on 230,000 acres as Americans on 638,000 acres. The German percentage is somewhat more favorable, until we restrict the comparison to the building area. Then the population per acre in fifteen American cities is 22 as compared with 157.6 in thirteen German cities.^ Paris is one of the most densely populated cities in the world. Upon an area slightly smaller than Kansas City's (20,774 acres), Paris concentrates about two and one-half million persons as contrasted with the latter's 133,000. The following comparative table of individual cities will further illustrate the point here insisted upon:^ Table CLXIX. Acreage. London 74,692 Present New York 230,000 Paris 19,295 Berlin 15,661 Chicago 102,765 Philadelphia 82,807 Brooklyn 18,084 Liverpool 5.210 Manchester 12,788 Hamburg 18,544 St. Louis 39>276 Boston 2/1,231 Baltimore 18,867 Birmingham : 8,400 5 i.i ^ iitk Cen., Social Statistics of Cities,^. 14. Population is given in round numbers. * Ibid. ^ Ibid., 13. Names of American cities are indented, and their density put in a separate column. Pop. per acre. 564 — 13.0 126.9 100.8 10.7 12.6 — 44.6 99.4 39-5 30.7 "•5 18.5 23.0 TENDENCIES AND REMEDIES 469 These figures are to be used cautiously, as they depend somewhat upon the amount of suburban territory recently annexed:^ but on the whole, they demonstrate that popula- tion is spread out over a larger territory in American than in European cities. It has sometimes been urged that this is largely a result of the development of the electric street railway in America, but the causal connection is not appa- rent. The first street railway using electric propulsion was opened in 1886, and the number of miles in operation at the time of the latest census was not only small in the aggregate, but was restricted for the most part to smaller cities than those at present under consideration. It should rather be said that the American penchant for dwelling in cottage homes instead of business blocks after the fashion of Europe is the cause, and the trolley car the effect. Philadelphia was the " city of homes " long before rapid transit. Philadelphia in 1880 led all other American cities in length of horse-car lines, but the horse-car is too slow to carry the majority of workingmen to and from their work each day. Hence the comparative figures of mileage and number of rides per in- habitant of American and European cities are indications of low or high density of population, which may be regarded as the cause of street railway building.^ ^ Thus, in the old New York there were 58.7 inhabitants to the acre in 1890-1, while Paris had 126.9 ^nd Berhn 100.8; and yet it is altogether probable that New York suffers more from dense crowding than do the two European capitals, for the vast majority of New Yorkers live below the Harlem, where the density in 1894 was 143.2 (iV. Y. Tenement Hotise Co7nmittee Report, p. 256). ^ The nth Cen. Rep. on Transportation by Land (p. 685) contain scompara- tive figures of the average number of rides annually per inhabitant : New York 297 Buffalo 65 Kansas City 286 London 74 San Francisco 270 Liverpool 51 Boston, Lynn and Cambridge. 225 Glasgow 61 Brooklyn 183 Berlin 87 Chicago 164 Hamburg 78 Philadelphia 158 Vienna 43 St. Louis 150 Budapest 37 Berlin, with the best street railway system in Europe, would rank twenty- second among the 28 American cities of 100,000-)-. Ayo THE GROWTH OF CITIES A better index of suburban travel is the number of com- muters carried by the steam railways and their percentage in the total number of passengers. Such statistics are furnished by the Eleventh Census {Social Statistics of Cities, pp. 49- 50) and it appears that, on the whole, suburban traffic in- creases in the same ratio with the magnitude of cities : Table CLXX. No. of Ratio of commuters Cities. cities. Commuters. to all passengers. 10,000-25,000 75 4,765,884 28.6 25,000-50,000 29 6,667,220 29.3 50,000-100,000 23 3.956,938 SI'I 100,000+ 24 79,945,182 52.1 Total 151 95»335>224 464 In the matter of street railway travel, Chicago and Phila- delphia ranked far below New York, Evidently this is con- nected with the fact that they have a larger traffic on the regular railway lines. But the palm for suburban travel be- longs to Boston, which had almost as many commuters as New York and Chicago put together : ' Commuters. Ratio to all passengers. Boston 24,587,000 62.9 Chicago 16,903,000 85.9 Philadelphia 10,714,000 70.7 New York 8,643,000 26.9 Cincinnati 3,697,000 86.9 Pittsburg 2,698,000 48.8 San Francisco 2,367,000 37.3 St, Louis 2,164,000 75.8 It is clear that we are now in sight of a solution of the problem of concentration of population. The trolley car and the bicycle may serve the purpose in middle-sized cities or even in the less populous cities of the first class. But when the city attains a population of a quarter of a miUion, more ^Ibid., 50, TENDENCIES AND REMEDIES 471 rapid transit than the electric surface railway can furnish is imperatively demanded. A surface railway cannot well run cars at a speed of more than nine miles an hour, and the legal Hmit in New York State is ten miles an hour. But, as few workingmen can afiford to spend more than half an hour in going to their work, they would then be compelled to dwell within three or four miles of the factory and could not have homes in the open country or suburbs, which are at least seven miles beyond the centre of the large city. Even the elevated system would not serve the purpose with its regular trains, which cannot be run at a speed in excess of 12 miles an hour. On the other hand, the regular railway lines with fast suburban trains are too few in number to serve a large territory. The sole remedy is the multiplication of steam railroads or the building of elevated and underground four-track systems, thus providing for express trains with a speed of at least 25 miles an hour. Then the workingman could establish his dwelling in the suburbs anywhere within a radius of ten miles of the centre of the city. Moreover, by virtue of the geometrical proposition that the areas of two circles are to each other as the squares of their radii, you will quadruple the area for residences every time you double the distance travelled. If in the first case (3 mile radius), the land within the circle afifords room for 20,000 dwellings, when you double the speed (ordinary elevated system) you will have an area large enough for 80,000 dwellings. Double it again (the underground railway supposition) and you will have ground for 320,000 houses. The transcendent importance of rapid transit as a remedy for overcrowding has been recognized most adequately in Belgium, where the railways are principally state-owned. The government there has not only provided an adequate train service for workingmen residing in suburban towns, but has established the rates of fare on a cheap basis that permits 472 THE GROWTH OF CITIES the train service to be used daily by workingmen. The ser- vice to and from work costs 2i cents a week to those travel- ing three miles or less, and gradually increases up to 57 cents for 42 miles. By the Cheap Train Acts of 1883, the English Parliament subsidized the railways entering London with about $2,000,- 000 a year, in the shape of remission of taxes, for the provis- ion of workmen's trains. A season ticket for one year (600 journeys for a distance of 22 miles) can be obtained for about four cents per journey. New York still lags behind, but Chicago and more especially Boston (where the legislature has aided the public) are developing a first-rate system of suburban communication. It cannot be doubted that the extremely satisfactory housing of Bostonians as compared with European urbanites,' is due not less to the fostering of suburban travel by the steam railways than to the development of the trolley system. Another striking example is that of Sydney, New South Wales:'' Population of . The city. The suburbs. 184I 29,973 1851 44,240 9,684 1861 56,840 38,949 1871 74,566 63,210 1881 100,152 124,787 1891 107,652 275,631 In the city itself growth has almost ceased, while the suburbs more than doubled their population in the last de- cade. But the city cannot be called overcrowded, for in ' Robert P. Porter in the Report of the Mass. Special Commission 07t the Rela- tions between Cities and Street Railway Companies (1898, p. 218) calls attention to the fact that Boston has only 1,053 families dwelling in a single room as com- pared with 120,000 in Glasgow. The percentages of all families are respectively ly^ and 33. ^ Census of N. S. W., i8gi. Statistician's Report, p. 120, TENDENCIES AND REMEDIES 473 1 89 1 it had only 37.4 inhabitants to the acre. This is a very- low density ; in New York city there are only three wards in which the population is less dense — the Second, which is al- and most wholly given over to business, and the Twenty-third Twenty-fourth, north of the Harlem river. (Table CLXIV). Attention has often been called to another encouraging tendency favoring suburban growth, namely, the transference of manufacturing industries to the suburbs. The local ad- vantages of a suburban town have been pointed out ; they include not only a great saving in rent and insurance, but economy in the handling and storing of goods. All carting is avoided by having a switch run directly into the factory ; saving to machinery is effected by placing it all on solid foundations on the first floor ; and plenty of space is at hand for the storing of fuel and materials, so that these may be bought when the market offers the most favorable terms. Finally, the suburban employer is likely to secure a high grade of employees. On the one hand, he is not antagon- ized by the trade unions, who can treat with him as effectively as if he were in the city itself ; on the other hand, his large workshops, and the prospect of a cottage and garden, and open air Hfe, attract operatives of the best class. Statistical data regarding the location of factories in suburbs are not available, but the strong tendency in that direction is familiar to all Americans. A similar tendency is noticeable in Eu- rope, and it has been remarked that although Manchester, Leeds and Lyons are still the chief centres of the trade in cotton, woollen and silk goods, they no longer produce any great part of these stuffs. Manchester is surrounded with scores of industrial cities and villages. Suburban growth as a result of this tendency cannot be forced ; it must wait upon economic forces. But the growth of purely residential suburbs can be influenced a great deal by public policy. In the past it was chiefly the middle 474 THE GEO WTH OF CITIES classes who could afford to dwell in the suburbs. But if society wishes to minimize the evils of concentration of pop- ulation, it must abandon the hope of accomplishing great things by such palliatives as model tenements, (which, if lo- cated in the city, often serve merely to prevent factories from moving to the suburbs), building laws, inspection of build- ings, and the various other ameliorations already discussed. Four goals are of fundamental importance: (i) a shorter working day, which will permit the workingman to live at a distance from the factory; (2) associations for promoting the ownership of suburban homes by workingmen ; (3) cheap transit ; (4) rapid transit. The importance of the two latter policies has been urged in so eloquent words by Dr. Cooley that they deserve quotation: " Humanity demands that men should have sunlight, fresh air, the sight of grass and trees. It demands these things for the man himself, and it demands them still more urgently for his wife and children. No child has a fair chance in the world who is condemned to grow up in the dirt and confinement, the dreariness, ugliness and vice of the poorer quarters of a great city. It is impossible to think with patience of any future condition of things in which such a childhood shall fall to the lot of any large part of the human race. Whatever struggles manhood must endure, childhood should have room and opportunity for healthy moral and physical growth. Fair play and the welfare of the human race alike demand it. There is, then, a permanent conflict between the needs of industry and the needs of humanity. Industry says men must aggregate. Humanity says they must not, or if they must, let it be only during working hours and let the necessity not ex- tend to their wives and children. // is the office of the city railways to reconcile these conflicting requirements." The extent to which this function may be fulfilled is indi- cated by the progress already made in Boston, Sydney, etc. Even in the European city of Frankfort it was found in 1893 that about 60 per cent, of the population doing business there lived outside.^ The electric trolley car is helping in the transformation, and its influence will undoubtedly be ap- parent in the Twelfth Census. ^ Bleicher, in Proceedings of Budapest Cong., vii, 466. TENDENCIES AND REMEDIES ^y^ The " rise of the suburbs " it is which furnishes the solid basis of a hope that the evils of city life, so far as they result from overcrowding, may be in large part removed. If con- centration of population seems destined to continue, it will be a modified concentration which ofifers the advantages of both city and country life. It will realize the wish and the prediction of Kingsley {^Miscellanies: "Great Cities"), — "a complete interpenetration of city and country, a complete fusion of their different modes of life and a combination of the advantages of both, such as no country in the world has ever seen." BIBLIOGRAPHICAL NOTE. The first scientific monograph devoted to the subject of urban and rural popu- lations and the migration from the latter to the former was written by M. A. Legoyt, the distinguished chief of the French bureau of statistics : Du Progres des Agglomerations Urbaines et de I'Emigration Rurale,^ Paris, 1870, pp. 2S0. It was an admirable piece of work for the period in which it was written, but is now ren- dered obsolete by the publication of new material. The second monograph worthy of mention was by the Swede, J. Gamborg : Om Byerne og Landt, i deres indbyrdes forhold med hensyn til Befolkring og Produktion, Christiana, 1877. Nothing more in the shape of scientific treatises appeared, to the author's knowl- edge, until very recently, when several writers on the continent of Europe have put out monographs in rapid succession: (i) M. Heins, La Belgique et ses grandes Villes au XIX^ Siecle, La Population, Ghent, 1897; (2) R- Kuczynski, Der Zug nach der Stadt; Statistische Studien iiber Vorgange der Bevolkerungs- bewegung im Deutschen Reiche,^ Miinchen, 1897; (3) ^^^^ Meuriot, Des Ag- glomerations urbaines dans I'Europe contemporaine ; Essaisur les Causes, lesCon- ditiones, les Consequences de leur Developpment, Paris, 1897. The last is a Doctor's thesis and is the most thorough study made since Legoyt, but is greatly inferior to the latter in originality and freshness, as well as arrangement. Noth- ing at once systematic and scientific has been published in English, altho the city problem has been attacked by Robt. Vaughan, The Age of Great Cities, 2d edition, London, 1843; Fothergill, The Town Dweller; and Strong, The Twent- ieth Century City, New York, 1898. In periodical literature and encyclopedias there is a valuable body of writings on this subject. Dr. G. B. Longstaflf has a short but well-written chapter in his Studies in Statistics, and has contributed noteworthy articles on Rural Depopula- tion to Palgrave's Dictionary of Political Economy, and the Journal of the London Statistical Society, 56: 380 (Sept., 1893). The chapter on Les Populations Urbaines in Levasseur's La Population Frangaise (vol. ii, Paris, 1891) can not be neglected, nor can the pertinent chapters in Rauchberg's Die Bevolkerung Oester- reichs. One of the most important resumes of the recent results is given by A. Wirminghaus, Stadt und Land, unter dem Einfluss der Binnenwanderungen, in Jahrbiicher fiir Nation aloekonomie und Statistik, Ixiv, pp. i, 161. "Writers who have touched the subject incidentally but luminously are Bucher, essay on Die inneren Wanderungen in his EntstehungderVolkswirthschaft; Hobson, Evolution of Modern Capitalism; Pearson, National Life and Character. Finally, it remains * Cited in text as Legoyt. * Cited in text as Kuczynski. (476) BIBLIOGRAPHICAL NOTE 477 to mention some of the fuller text-books, such as Roscher's System der National- okonomie, vol. Ill; Wagner's System, etc., vol. I (Grundlagen der Volkswirth- schaft), etc. Reference should also be made to Schonberg's Handbuch der Pol- itischen Oekonomie, Art. Bevolkerungslehre, and Conrad's Handworterbuch der Staatswissenschaften, Art. Bevolkerungswesen, etc. An extended discussion of nimierous phases of the subject was held at the Eighth International Congress of Hygiene and Demography at Budapest, 1894, a full report of which is published in the Proceedings (Vol. VII). Most of the works above-mentioned contain numerous references to authorities or sources of information, but the best bibliographies are to be found in the fol- lowing : For Chapter II. A recent account of the organization of government statistical offices and their publications on population statistics will be found in Meuriot, op. cit. Additional references to the German statistics are to be found in the article of Wirminghaus, above cited, and, especially for the municipal statistics, in the article of Bruckner, Allgemeines Statistisches Archiv, i, 135. For the United States, see E. C. Lunt, Key to the Publications of the United States Census, in publications of the Amer- ican Statistical Association, i, 105. The most important compendium of recent statistics of the population of towns and cities is the Ortsstatistik of Alexander Supan, constituting the ninth number of Wagner-Supan's Bevolkerung der Erde ^ and Erganzungsheft No. 107 of Peter- mann's Mitteilungen, Gotha, 1893. Nearly every country in the world is em- braced in this compendium, and their statistical publications containing data re- specting the population of towns are, of course, cited in full. Statistical data for the first half of the century have frequently to be sought in the handbooks of Staatenkunde, of which the more important are : Hassel, Statistische Uebersichtstabellen der sammtlichen Europaischen Staaten, Gottingen, 1809. (Cited as Hassel, 1809, in text.) Hassel, Statischer Umriss der sammtlichen Europaischen und der vomehmsten aussereuropaischen Staaten in ihrer Entwicklung, Grosse, Volksmenge, Finanz- und Militarverfassung tabellarisch dargestellt. 3 Hefte. Weimar, 1823-4. (Cited in text as Hassel, 1822.) J. E. Worcester, Geographical Dictionary or Universal Gazetter, 2 vols., An- dover, 181 7. (Cited as Worcester in text.) Malchus, Statistik und Staatenkunde von Europa. Stuttgart and Tiibingen, 1826. (Contains ample bibliography, pp. 2C-39.) Bernoulli, Handbuch der Populationsstatistik. Ulm, 1841. Harper's Statistical Gazetteer of the World. New York, 1855. (Cited as Plarper.) Wappaus, AUgemeine Bevolkerungsstatistik. Leipzig, 1861. (Especially im- portant for the statistics of 1845-55.) ' Cited in text as Supan. 478 THE GROWTH OF CITIES Kolb, Handbuch der vergleichenden Statistik. 2d edition, i860; 8th edition, 1879. (Cited as Kolb, i860, and Kolb, 1879, respectively.) Almanach de Gotha. Statesman's Year Book, since 1864. Brachelli, Die Staaten Europas. 4th ed., 1884. The books above mentioned, together with the citation of authorities in the statistical tables of Chapter II., comprise the principal sources of information. Chapters IV to IX. The authorities used in Chapter III are enumerated under the several sections of that chapter, and through the remaining chapters of the essay in foot-note ref- erences. It is not, therefore, necessary to repeat titles here, especially as classi- fied bibliographies are accessible in the encyclopedias or handbooks of economics and social sciences, notably those of Conrad and Schonberg and Wagner. Particularly valuable are the bibliographies appended to the separate sections of Mayr's Bevolkerungsstatistik, 1897 (especially Sees. 26, 27, 37, 39, 81), and the classified bibliography of 100 pages in Bevolkerungslehre und Bev5lkerungspolitik by A. von Fircks (Leipzig, 1898). As to the municipal conditions and the problems of city life, see Robt. C. Brooks's Bibliography of Municipal Administration and City Conditions (Muni- cipal Affairs, March, 1897). INDEX OF AUTHORS Alexander, E. P. 204 Allen, W. F. 17 Ammon, O. 366, 385, 390, 394, 441 Andrews, E. B. 227 Aristotle 428, 452 Arlidge 360 Asboth, von 121 Aschrott, P. F. 214 Ashley, W. J. 169, 177, 178, 193 Atkinson, Edward 184 Bagehot, Walter 185 Balbi 120 Ballod, C. 247 Beaumaire, Vicomte de 211 Becher, S. 95 Beddoe, John 394 Beloch 163 Bernouilli 477 Bertillon 295, 329, 416 Billings, J. S. 32, 295, 341, 392 Bindewald, von 426 Blackmar, F. W. 407 Bleicher, H. 269, 270, 271, 272, 280, 290, 293, 294, 295, 298, 332, 344, 357, 363, 364, 365, 373. 381,383,411,474 Block, M. 78, 79, 120 Boeckh, R. 240, 241, 295, 361, 374 Botzow, C. 172 Booth, Charles 243, 257, 281, 341, 371, 382, 390 Bowmaker 353 Boyd, Carl 34 Brachelli, H, 478 Brooks, R. C. 478 Brown, F. J. 312 Brownell, Miss J. L. 338-41 Bruckner, N. 85, 86, 248, 267, 281, 283, 293, 325, 329, 331, 342, 391, Bryce, James 223, 429 Buchenberger, A. 211 Bucher, Karl 169, 175, 180, 185, 231, 273, 288, 290 299, 476 Bulgarius, Thaddaeus 107 Buschen, A. von 107 Cannan, Edward 51 Cantlie 370 Carroll, H. K. 400 Cato 394 Chastellux, Comte de 75 Chisholm, J. C. 54 Cicero 454 Clark, John B. 228 Closson, C. C. 441 Coghlan, T. A. 138, 139, 140 Collet, Miss E. B. 361 Cooley, C. H. 169, 183, 198 Coulanges, de 169 Crawford, J. M. 105 Crooker, J. H. 220, 438 Crum, F. S. 242, 295, 320, 333, 387 Cunningham, Wm. 169, 171, 454 Curtis, T. B. 295 Cutting, R. F. 421 Delille, Leopold 231 Deparcieux 235-6 Devine, E. T. 422 Dixon, F. H. 200 Dunbar, P. L. 313 Durkheim, E. 185 Edson, C. 341 Emerick, C. F. 211, 455 Engel, E. 82, 83, 341, 342, 387, 394, 413 Epps, Wm. 217 Farr, Wm. 345, 348 Fetter, F. A. 338, 341 Fircks, A. von 478 Fiske, J. 14 Fletcher, H. F. 212, 220 Fothergill, J. 476 Foville, A. de 78, 79, 169 Fowler, W. W. 169 Franscini, S. 117 Galton, Sir Francis 387 Gamborg, J. 476 Gannett, Henry 13 Geddes, P. 295 George, Henry 216, 368, 439 Gibbins, H. de B. 165, 166, 179 (479) 278, 328, 366, 37°: 282, .477 222, 48o lADEX OF A UTHORS Giffen, Sir Rob't. 226 Gilman, N. P. 2 Gladden, Washington 437 Gohlert, J. V. 95 Goldstein, J. 426 Goltz, von der 216 Goodnow, F. J. 15 Gould, E, R. L. 353 Graham, P. A. 188, 210, 211, 213, 216, 421, 422 Graunt, Capt. John 232, 233, 234, 235, 367 Green, Mrs. J. R. 169 Hadley, A. T, 79, 184, 200, 338 ' Hackel, Ernst 185 Ilalley, Edmund 235 Hansen, Georg 262, 370, 371, 373, 378, 379, 381, 384, 385, 386, 388 Harper 127, 132, 134, 135, 136, 137, 477 Harrison, Frederic 443 Hartwell, E. L. 353 Hasse, E. 467 Hassel 95, 119, 120, 121, 122, 123, 128, 477 Hearn 169 Heins, M. 476 Held, Adolph 184 Hirschberg 412 Hobson, J. A. 52, 53, 56, 148, 181, 182, 184, 225, 229, 399, 433, 434, 476 Hoffman, F. L. 310, 311, 313, 314 Hoffman 84 Holmes 426 Hooker, R. H. 341 Hourwich, I. 170 Hume, D. 163, 452 Inama-Sternegg, von 383 James, E. J. 19 Jannasch, R. 82, 83, 107, 203 Juraschek, F. von 356 King, Gregory 44 Kingsbury 221, 421 Kingsley, Charles 475 Koppen, P. von 107 Kohl, J. C. 169 Kolb 119, 120, 121, 128, 134, 478 Korosi, J. 329 Kuczynski, R. 233, 234, 286, 287, 298, 303, 358, 384, 385. 394, 476 Kuhn, £.171 Lagneau, E. 346 Laing 7 Lapouge, G. 441, 442 Laspeyres, E. 197, 201 Laveleye, E. 216 Lavergne, L. de 167 Lecky, W. E. H. 222 Legoyt, M. A. 49, 211, 230, 294, 322, 476 Lehr, A. 383, Levasseur, Emile, 49, 70, 73, 76, 77, 184, 211, 239, 289, 294, 300, 319, 322, 324, 330, 341, 346, 347, 361, 392, 401, 403, 404, 405, 406, 446, 476 Levi, L. 404 Lichtenstein, J. M. 95 Little, W. C. 221 Lommatzch, G. 8, 93, 459 Longstaff, Geo. B. 45, 63, 65, 211, 217, 243, 369, 476 Losch 8, 185 Loua 246 Low, S. J. 397 Lowe, James 44, 418 Lunt, E. C. 477 Mackenzie, J. S. 2, 434, 436, 456 Maine, Sir Henry 169, 170 Malchus 477 Maltbie, M. R. 215, 219, 435 Manouvrier 394 Marshall, Alfred 184, 203, 207, 208, 341, 387, 395, 396, 413,44s- 457, 458 Marx, Karl 176, 184, 195, 196 Mataja, Victor 184 Mayo-Smith, Richmond 5, 299, 300, 305, 319, 360, 365 Mayr, Georg von 19, 260, 267, 295, 300, 302, 347, 393, 401, 402, 478 Meuriot, Paul 76, 449, 467, 476, 477 Mill, J. S. 184 Minutoli 120 Moore, H. E. 211 Morselli 401, 402, 403 Mulhall, M. G. 167 Neison, F. G. P. 347 Neumann, F. J. 295 Neumann-Spallart 97, 182 Newsholme, A. 342, 349, 352, 365 Nordau, M. 368-9 Oettingen, A. von 384, 402 Ogle, Dr. W. 45, 47, 188, 211, 303, 320, 359, 365, 444 Oldberg, E. von 107 Palgrave, R. I. 63, 211, 217, 476 Patten, S. N, 227, 436 Pearson, Charles 222, 439, 441, 476 Petty, Sir William 235, 452, 453 Philippovich, E. von 88 Pliny 394 Plutarch 454 INDEX OF AUTHORS 481 Pohlmann, R. 454 Porter 404 Porter, R. P. 472 Pratt 370 Preston-Thomas 214 Prothero, R. E. 164, 165, 166 Quesnay 230 Radcliffe 347 Ralph, ] ulian 220 Ratzel 2, 169, 183 Rauchberg, H. 94, 95, 96, 97, 98, 253, 260, 261, 267, 274, 277, 278, 279, 291, 292, 302, 316, 323, 327, 372, 378, 380, 392, 458, 476 Ravenstein, 255, 257, 258, 259, 265, 266, 267, 274, 277 Rawson, Sir Rawson W. 49, 243 Reden, Fr. Fred W. 83 Ripley, W. Z. 169, 393, 442, 443 Rogers, Thorold 44, 170 Roscher, Wm. 169, 197, 432, 454, 455, 477 Ross, Edward A. 197 Rousseau, J. J. 368 Rubin, M. 325, 333, 341, 358, 387 Rumelin, Gustav 15, 439 Salmon, Lucy 220 Sax, E. 169 Schaffle, Albert 185 Schenk, Leopold 295 SchmoUer, Gustav, 169, 170, 175, 176, 179, 180, 181, 184, 185, 187, 194 Schnitzler 107 Schonberg, G. 477 Schonhof, Jacob 184 Schulze-Gavernitz 184, 203, 209 Schumann, 253, 267 Schwabe, H. 82, 200, 384 Sedlaczek 240 Seeley, J. R. 180 Seligman, E. R. A. 428 Sering, Max 168, 216, 422 Seutemann, 295 Shaw, Albert 219, 437 Silbergleit 295, 362 Simmel, Georg 184 Sinzheimer, L. 185 Smith, Adam 176, 184, 418 Smith, H. Llewellyn 257, 258, 273, 275, 276, 382, 383. 389* 390 Sohnrey, H. 216 Spencer, Herbert 158, 169, 184, 185 Sprmger, J. 95 Stephens, H. C. 215, 456 Steuart, Sir James 169, 424, 425^ Stickney, A. B. 199 Stieda, W. 184 Storch, H. 107 Strahan 370 Strong, Josiah 477 Siissmilch, J. 7, 330, 355, 444 Supan, Alexander 7, 10, 11, 18, 58, 66, 115, 116, 120, 121, 122, 128, 129, 133. 135. 136, 137. 139, 477 Suydam, W. L. 421 Tallqvist 341 Thompson 295 Thuroczy 356 Tocqueville, A. de 431 Toynbee, Arnold 45, 166, 21 7 Tucker, Geo. iii, 40 Turquan 346 Vaughan, Robt. 476 Wagner, A. 402, 477 Walker, Francis A. 184, 419 Walford, C. 297 Wappaeus, J. E. 7, 294, 331, 335, 34 1, 387. 477 Weeden, W. B. 184 Wells, David A. 184 Welton, Thomas A. 365 Wernicke, J, 234 Westergaard, H. 325, 341, 387 Willcox, W. F. 4, 173, 213, 251, 279, 288, 296, 307, 308, 312, 330 Williams, R. Price 48, 49, 51, 52, 53, 464, 465 Wirminghaus, A. 211, 249, 476, 477 Wirth, M. 117 Worcester, J. E. 477 Wright, C. D. 184, 330, 459 Young, Arthur 45, 165 Ziegler 343 INDEX OF SUBJECTS Abyssinia, concentration of population 137 Accidents, fatal 296-7 Administration, centralization of, and city growth 215; public, in x\merica 428-9; decentralization of 456-7 Advertising, expenditures for 195 Age of migrants 280-2, 284; distribu- tion in urban and rural districts 300-4; distribution, curves of 300; at marriage 322, 325 Agglomeration in France and Italy 9; in final classification 16 Agricultural implements, American 168 Agriculture and the distribution of pop- ulation 160-9; the mother-industry 1 61-2; application of science to 162; machinery in 162; progress in, since 1750, 164-8; influence of transportation upon 197; depres- sion of 210; favors dispersion of population 223; in city and country 315, 317; and the birth rates 339- 42; stimulated by city markets 418 Air, compressed, for reducing tempera- ture 458 Alexandria, population 448 Algiers, concentration of population 137 Allotments, small, and emigration 216; as a remedy for concentration 455 American and European cities, compar- ative rates of growth 450 Americans, superior mobility 250; post- ponement of marriage 326; see also United States. Amusements in city and country 218 Antiquity, city walls in 5-6; great cities in 448-9 Antwerp, population 1802-1890, 116 Apprentices 438 Argentina, city growth 134, 135 Aristocracy, intellectual, continuation of 386-9, 44=; Arithmetic, Political 231 Armies, disbanding 217; country-bred recruits 369-70, 394-5; effect of city growth upon 425-6 Art in cities, 218, 224; city vs. country- born, in 374-6 Artisans in city and country 316-7; city vs. country -born 378-80; birth rates among 387; in intellectual aristoc- racy 388; from the country 390 Association of people, effects of, intel- lectual in cities 220 Athenian's city and country residence 175 Athens, modem 120; ancient 448 Athletics prosper in cities 445 Atmosphere in cities 348, 368 Australia, 1891, compared with United States 1790, I; mode of life of na- tives 3; city growth in 138-142; city growth and physical features 149; land systems 217; suburban growth 472 Austria, city growth 94-100, esp. 95; natural increase and immigration in cities 246; internal migration 249— 253 ; statistics of birth-place 260-1 ; sex in migration, 277-8, 291 ; ratio of the sexes 292; age-grouping 302; conjugal condition, 323, 327; mor- tality 356. See also Vienna. Averages, difficulty with 410; fallacy of 444 Ayuniamettto, average area of, in Spain 142 Babylon 6, 448 Bachelors, preponderance in cities, 322-7 Bagdad, population 122, 123, 449 Baker's industry centralized 196 Baltimore, population 21, 450 ; density of population, 468 Baptisms, ratio of, to burials 234 Baths, public 353 Bavaria, growth of cities 93, 94; mobil- ity of urban and rural populations (483) 484 INDEX OF SUBJECTS 262; natural increase in cities 385; insane, feeble-minded, blind, deaf and dumb 393 Belfast, growth 65, 66, 67 Belgium, city growth 115- 116; a manu- facturing country 18 1-2; internal migration 249-252 Belgrade, population 1800-90, 121 Berlin, density of population 4; popula- tion 1819-1890,84; relative growth 86,91; area 139; migration 233-4; excess of births over deaths 236; manner of growth 239, 240; age of immigrants 281 ; length of resi- dence of immigrants 282-3; mar- riage among natives and immigrants 328-9; overcrowding and mortality 349 ; occupation and social rank of immigrants 373-8, 391 ; prostitution among natives and immigrants 384; Greater, population 453, 466; growth of population by districts 465-6; density of population 468 Bevolkerungsstrom, Hansen's theory of, 388-9 Bills of mortality, in sixteenth century 231 Birmingham, immigration vs. natural increase 239; density of population 468 Birth-place, statistics of 247-276 Birth rates of urban and rural popula- tions 330-43; of social classes, Copenhagen 387 Births in urban and rural France 69; relation of to deaths 231-246; in London 232, 235 ; in German cities 234, 236, 237, 245; in Sweden, 237-8; in modern cities 239; in Massachusetts cities 241-2; in French cities 245; influenced by migration 235-6; sex at 294; ex- cess of, over deaths in Bavarian cities 385 Blindness in city and country 392-3 Bolivia, concentration of population, 136 Bombay, growth 126, 127, 450 Bonds, ownership of French government 427 Bornu (Soudan) concentration of popu- lation 137 Bosnia, cities 121 Boston, population in 1800,21; growth 37-39; incorporation of suburbs 241; manner of growth 241; char- acter of immigrants to 265; ratio of the sexes 286; excess of women 289; natural movement of popula- tion 318; marriage rates 321; con- jugal condition 324; Greater, pop- ulation of 453-4; congested wards 459; density of population 468; suburban travel 470, 472 Brachycephalic race 441-2 Brandenburg, density of population 4 Brazil, distribution of population 133-4 Brescia, township and community 17-18 Breslau, mortality 235; manner of growth 239 Brookline, local text-book prepared by 430 Brussels, population Ii6, 450 Bucharest, growth 122 Budapest, population loi, 450 Buenos Aires, growth 134-135, 450; natural advantages 149 Building laws in New York, London, etc. 351.414 Building trades, city vs. country-born in 374-81, 390 Bulgaria, cities 121 Burials, relation of, to christenings 232- 3-4 Business firms in cities 369-70, 387 Cahiers respecting scarcity of farm labor 231 Cairo 449 Calcutta, growth 126-128, 450 Canada, city growth 130-132 Canals, era of, in the United States 23- 25; inadequacy of 184 Cape Colony, concentration of popula- tion 137 Capital, beginnings of era of, 186; ad- vantages of concentration of 194-6; an aid to cities 457-8 Carthage, population 448, 452 Castles as nuclei of towns 174 Catholics and divorce in cities 330; strength of, in cities 400-1 Census, Austrian, of 1890, a model, 94 Centralization among cities 446-8; in administration 215, 456-7 Cereal production in the United States 26 Charity in cities 436 Charleston, population in 1800, 21 Charters of freedom bought by cities 178 C/tefs lieux, in France 72-74 Chest measurement in city and country 394 Chicago as a centre of newly-occupied territory 20; area of 139; vs. New INDEX OF SUBJECTS 485 York, future growth 174; and New York, freight rates between 204; ratio of the sexes 286; growth 450; density 468; suburban travel 470 Children in cities 368-9, 396, 474 Chile, city growth 135 China, estimated population 129; urban population 129; lack of trustworthy data 45 1 Christenings, relation of, to burials 232- 3-4 Christiania, population 1801-91, in Churches in city and country 400-1 Cities, power of attraction of great 259- 60; ancient, population of, 448-9; number in Europe in modern times 449-50; population of fifty princi- pal cities of the world 450; limits upon the growth 05451-3 Citizen, significance of the word 3, 6 "City-building," process of 467 City-growth vs. concentration of popu- lation 155-157 City life, advantages of 218-221 City-location, principle of 172 City-state of antiquity 6; Aristotle's ideal 428 Civilization, identified with city 6; and the birth-rate 338-41 Classes, antagonism of, in cities 427 Classification of dwelling centres, ancient 5-6; mediaeval 6-8; modern 8-16 Clerks, from the country 371, 390; birth- rates 387; advertisements for coun- try boys 421 Clothing, ready-made, and the factory system 195 Clothing trade, leading industry of New York City 206-7 Cockney, London 369 Cohesiveness lacking in city populations 432-3 Colbert 230 Colbertism 180 Colombia, concentration of population 136 Colonization of city poor 455-6 Commerce, expansion of, in England 55 ; place of, in industrial evolution 159, 169-184; between town and country 176; in middle ages 177; affected by Crusades 179; between towns 179; of Holland 181; bio- logical analogy to growth of 183; and distribution of population 223 ; in city and country 315, 317? ^^*i the birth rate 341-2; city vs. coun- try born in 373-81 Commons, enclosure of, in England 165 Commune, average area of, in France, Switzerland, Belgium 142 Communication, means of, as affecting the food-supply 3 Commuters, number of 470 Competition presupposed in normal dis- tribution 419; and wages 419 Concentration of population and growth of cities, distinction between 155-7 Condition, conjugal, in city and country 322-30 Congestion, relief of, in Philadelphia and Boston 459; New York city 460-2; London 462-5; Berlin 465-6; Vi- enna 467; Hamburg 467 Conservatism of rural populations 439- 40, 443 Constantinople, population of 120, 449- 50, 452 Consumption, local, and city industries 206-7; ^"^^ future city growth 227 Co-operation, political, required in cities 433-5 Copenhagen, growth 112-3,450; early marriages in 325; fecundity of wo- men 333; death-rates by ages 358; birth-rates in social classes 387 Costa Rica, concentration of population 136 Cottage industries 195; outlook for 196; in England and the United States 206 Cotton industry in England 52; France Country-born, residence quarters of, m cities 370-2; in city occupations 370, 373-82, 390; among degener- ates 370, 383-4, 392-4; in the up- ward current 389; intelligence of 397-9 ^ .^ Countryman contrasted with townsman 3 Cowboys as soldiers, 397 Crime, city vs. country-born in 384 ; in town and country 403-9 ; localized in cities 408; in cities, explanation of 443 Crisis of 1873, and urban growth 27; 1877, and urban growth 89 Crusades aid the towns 178; promote commerce 179 Cuba, concentration of population 136 Cuhur and city growth 441 Current, downward, of failures 443 486 INDEX OF SUBJECTS Damascus, population of 122, 123, 449 Danes, founders of towns in England 171 Day, need of a shorter working 474 Deafmutism in city and country 392-3 Death rates, limitations upon value of 359; in exceptional wards, 444 Deaths, in urban and rural France 69; relation of, to births 231-46; in London 232, 235, 236; German cities, 234, 236, 237, 245; Sweden 237-8; modern cities 239; Massa- chusetts cities 241-2; French cities 241;; influence of emigration upon 235-6; violent 296-7; in city and country 343-68 Deer forests 217 Degenerates, downward current of 389; rural 407 Degeneration, the city and 368-9, 388-9, 392-5 ; in cities and counter forces 457-8 Denmark, city growth in 112-114; in- ternal migration 249-253; fecun- dity of women 333; death-rates by ages 358 Density of population dependent upon natural conditions 2; value of the term 4-5; relation to concentra- tion 146; in principal countries 147; and the birth rate 338-40; death rate 344-6, 350; in New York wards, 460; central area of London, decrease of 463-5 ; Berlin 466-7; Vienna, 467; of American and European cities compared 467-8; and street railways 469; of Sydney 472 Depopulation, rural, in England 45; France 69; Germany 89; United States 210-2; causes of 213, 423 Deterioration of physique in cities 434; not a result of city life 444-5 Differentiation performed by cities 442 Diseases, nervous and the birth rate 338-40 Distance between houses in England 9 ; the chief factor in migration 255- 271 Distribution of population, influence of nature upon 2; social environment 3; statistical methods for determin- ing 4; distinction between and growth of cities 19 Distribution of wealth, process of 418-9 Division of labor, beginnings of 161, 170, 174, 175; in Greece, 175; Rome 175; importance of 193, 195; theory of 194; evolution of 197; and migration 261; promotes production 417 Divorce more frequent in cities 325, 329-30 Dclicocephalic race 441-2 Domestic system 186-7; dangers of 195; and electric motors 196; edu- cation under the 438; prospect for revival of 456 Dorfzxid township 10 Drunkards, methods of treatment of 436 Drunkenness in England 404 Dublin, growth of 65, 66, 67, 450 Economy, household or village 170, 185; town 170, 176, 186; national 170, 177-180; international 170, 181 Ecuador, concentration of populat'n 136 Education, advantages in cities 218; popular attacked 222; in town and country 397-9; transformation in methods of 437-8 Egbatana 448 Egypt, concentration of population 137 Election of 1896, vote of farmers in 427 Electric motors and cottage industries 196, 456; power and distribution of population 192; the factory sys- tem 196 Elizabeth, proclamation forbidding mi- gration to London 231 Emigration from Ireland 64, 152; Ger- many 89, 152; Norway and Sweden 152; England and Wales 152; great cities 274; effect of upon ratio of the sexes 290-1 ; rural, causes of 423; advantages of 424. See also Migration. Employees, in city and country 316-7 Employers, in city and country 316-7; city vs. country born 375-80, 391; birth-rate among 387; in intellect- ual aristocracy 388 Enclosure Acts 165, 217 Energy, physical and moral 396 England, definition of city 13; statistics of city growth 40-57, esp. 46; the typical industrial nation 41 ; classi- fication of urban centres 41-2; urban and rural sanitary districts 42; rural depopulation 45-7; in- dustrial history 51-6; immigrants to the United States from 152; and INDEX OF SUBJECTS 487 the food supply 164; agricultural improvements 164-7; agricultural population 166; end of feudalism 178; national unity 178; Black Plague 178-179; natural increase in cities 243-4; internal migration 249-53; mortality in town and country 345, 347, 355; deaths by sex and age, 358-9; crime 404; increase in wealth and power 425-6; rapid growth of the suburbs 446-7; density and acreage of cities 468. See also London. Englishmen, mobility of 250 Entrepreneur, appearance of the 178; city vs. country born 375-80, 391, See also Employer. Environment, physical and population 2, 5; in Uruguay 148; Argentina 149; India 149-150; Australia 149 Erie canal and New York city 25 Essex contains London suburbs 447 Evolution of industrial society 158-160; bibliography of 169; periods of 1 70; stages in 185 Excitement in cities 368 Factories, advantageous location of 197-209; movement of toward suburbs 8, 202, 203, 224, 228, 473 Factory system, in England 53; France 78; Germany 88; Russia 105; be- ginnings of 188, 192-3; ancient Egypt 193; advantages of 193-6/ and distribution of population 196; education under 438 Factory towns and the marriage rate 321; birth rate 337-8; 341-2 Fairs develop out of weekly town market 179 Families, city, extinction of 370, 386-9; 445 Family life in city and country 322-30; size of families 336-7; life in Paris 405 Famine in Austria 96; Hungary 102; and distribution of population 169 Farm labor, scarcity of 422-4; wages of 422 Farmers, isolation of 221 ; the radical party in 1896, 427 Farms, increase of in the United States 26; plans to make more attractive 455 Fashion dominates city dwellers 432 Fecundity of women 330-43 Fees for settlement 455 Fertilizers and the law of diminishing returns 225 Flanders, seat of the woollen trade 179 Florence, influence of 6 Food in cities 368 Fords, sites of cities 172 Forecast concerning city growth 225-9 Foreigners, female, in American cities 279-80; concentration of in cities 304-9 ; percentage of in cities 306 : conjugal condition 326; birth-rates among 334; size of family 337; tendency of to herd together 430 Forts as nuclei of towns 174 France, agglomerated population 9; definition of cities 14; urban growth 67-80; rural depopulation 68, 69; natural increase 1881-91, 69; in- ternal migration 1881-91, 69; in- dustrial development 76-80; agri- cultural progress 167; national unity 180; railway policy 200; natural increase vs. immigration in cities 245 ; internal migration 249- 252; sex at birth, 294; marriage- rate 319; conjugal condition 324; divorces 330; size of families 336; death-rates, 345; crime 404; ille- gitimacy 405, See also Paris. Frankfort, excess of births over deaths 237; military rolls 269; birth-place of immigrants 271; return of mi- grants to native places 272; urban and rural origin of immigrants 273; sex of immigrants 278; age of im- migrants 280; sex and age of immi- grants 293; sex at birth 294; deaths in the sexes 298; mortality among natives and immigrants 366; resi- dence of natives and immigrants 372-3; occupation of natives and immigrants 381 ; pauperism of na- tives and immigrants 383 Functions, municipal 353-5 Garrisons in cities 457 Geme'ente, average area of, in Holland 142 Gemeinde, average area of, in Hungary, Austria, Germany, Prussia 142 Genius, in citv and country 439, 442 Genoa, influence of 6 Germany, classification of dwelling-cen- tres 15; city growth 80-94; indus- trial development 87-9; immigrants to the United States 152; railways and cities 200; freedom of move- 488 INDEX OF SUBJECTS merit 214-5; natural increase vs. immigration in cities of 245 ; cause of migration 255-7; excess of wo- men in cities 286, 290; excess of girls in cities 293; age-grouping in cities 301 ; occupations in cities 315 Ghent, population of, 1802-1890, 116 Glasgow, growth 59, 60, 450 Government, officials in a city 428; com- plexity of municipal 428-9; corrup- tion in city, in U. S. 429; excellence of local, in U. S. 429; difficulties of city, in U. S. 429-30; city, in 1830 431 Grain-trade in the United States 168 Greece, city-growth 120; cities of ancient 6,448 Gregariousness 221 Guatemala, concentration of population 136 Guiana, British, concentration of popu- lation 136 Guild system, education under 438 Hamburg, continuance of mercantile firms 386; growth 450; density of population 468 Handicraft system of production 186; and decline of villages 196; educa- tion under 438 Handicraftsman, development of the mediaeval 176 Hanseatic towns, influence of 6 Health in city and country 368-409 Heat, future method of distributing 458 Heathen or countrymen 3 Hebrews, sanitary regulations and health of 350-1 Herzogowina, cities of 121 Holland, city growth 1 14-5 ; a commer- cial country 181-2; internal migra- tion 249-252 Home rule for cities, basis of 430 Homes, suburban, need of associations for building 474 Honduras, concentration of population 136 Hospital gratuities in cities 457 Household economy 170 Housekeeping in city and country 219 Housing of the working-classes 35 1-3 Humanitarianism, the new era of 434-5, 440-1 Hungary, definition of cities 8; city growth 100-4; manner of growth of cities 246; age-grouping 301 Idiocy in city and country 392-3 Illegitimacy and ratio of sexes at birth 294-5; in Prussian towns 332; in Saxon towns 333 ; in city and coun- try 335-6; and infant mortality 362-3; in Paris 405; in cities 405-6 Illinois, urban population of, in 1890, 28; percentage of urban population 31 Illiteracy in town and country 397-8 Immigrants to cities, residence of 370-2; occupation 373-81, 390; social rank 377-81 ; efficiency of 389-92. Im- migrants to the United States, see Foreigners. Immigration, ratio of, to natural growth of modern cities, 239-246; effects upon community life 443-4. See also Migration. Improvements, public, in cities 457 Income, in town and country 41 1; Engel's law of expenditure of 413 Increase, natural, vs. immigration 230- 246; in cities 283; (Bavaria) 385 India, British, city growth 123-128; in- dustrial organization 123; popula- tion of provinces 124; definition of town 124 Indiana, composition of urban popula- tion 30; percentage of urban popu- lation 32 Indians, conditions of residence of 3 Individualism 432, 434, 436 Industrialism 432-4 Industries, groups of 223; effect of various on distribution of population 223-5; of city and country 314-5; city vs. country-born in 374-81, 391 Infanticide 406 Infantile mortality. See Mortality. Infirmities, physical 392-3 Insanity in city and country 392-3, 443 Intelligence of city people 369; of city and country people 397-9 Invention, " heroic " theory of 184; and city life 457 Iowa, density and concentration of pop- ulation 4; percentage of urban population 32; railway discrimina- tions 199-200 Ireland, growth of cities 64-67; potato famine 64; emigrants to the United States 152; internal migration 252 Iron industry in England 54-56; Scot- land 61; France 78 Italy, agglomerated population 9; urban and rural population II ; city growth 1 1 7-9 INDEX OF SUBJECTS 489 Jamaica, concentration of population 136 James I., proclamation forbidding mi- gration into London 231 Janitors, from the country 371, 391 Japan, distribution of population 129- 130 Kansas, so-called cities 12; percentage of urban population 32 Kleinstddtisch, a term of reproach 3 Klondike, wealth taken in and out 420 Labor-supply in large cities and small towns 205-206; a factor in de- termining location of factories 207; of both sexes influences in location of factories 208; scarcity of in rural districts, an ancient complaint 230; casual, in city and country 315-7; irregularity of in country JLaissez faire, policy of 434-5 Lancashire, growth 53; compared with Glasgow 61 iand, unoccupied, on the globe 226; tenure, effect of different forms on city growth 215-7; owners of, as a distinct class 388,427; ownership of, a conservative force 426; and city growth 426 Xegislation affecting the distribution of population 214-8 Leipzig, excess of births over deaths 237, 239, 240 Levasseur's law of large city growth 49-50; in United States 34-36; England 48-5 1 ; Scotland 64; Ire- land 65; France 70-77; Germany 89; Austria 99-100; Hungary 103, 104; Russia, 109; Norway 112; Belgium 116; Australia 142; sus- tained 446-8 Liberalism engendered in cities 432, 439 Life, expectation of, in city and country 346-7 Limitations upon size of a city 452-3 Liquors 404, 407 Liverpool, explanation of decline of population 51; population 450 density 468 Living, cost of, in city and country, 412-3 Location of cities, principles of 172 Locks in America and Germany 443 liondon, population in 1801, 44; 1801- 1891, 46; proportion to population of England 1801-91, 47; propor- tionate growth 1 80 1-7 1, 49; 1881- 91, 50; commercial development 56; area 139; commercial centre of the world 181; centralization of baker's trade 196; future growth 229; royal proclamations designed to check its growth 231; immigra- tion to, in 1580 and 1650 compared with 1 87 1-8 1, 232; rate of growth in 17th century 232-3; excess of births over deaths 236, 243; manner of growth 239, 240, 243-4; origin of immigrants 257-8; mobility of natives 274; direction of emigra- tion 275-6; sex of emigrants 277; mortality and overcrowding 349; death-rates by sex and ages 358-9; distribution of countrymen 371; crime 404; rapid growth of suburbs 446-7; rank 449-50; limit to growth 452-3; population of met- ropolitan district 453; extension of prohibited by law 454; relief of congested districts 462-5; popula- tion by districts in 1 89 1, 463; City, population since 1801, 464; density 468 Louisiana, stationary urban percentages 20; percentage of urban population 32 Luxury in cities 368 Lyons, now a commercial centre 473 Machine industry, relation to concen- tration of population 148, 152, 158 Machinery and the factory system 195 Madras, growth 1 26, 1 28, 450 Manchester, population 450; density of population 468; removal of factories 473 Manila, population 1887, 128 Manner of city growth 283 Manufacturing industry, in the country 8; United States 25, 27; and con- centration of population 148, 224; consolidation of processes in 191; industries, influence of transporta- tion on 197-8; facilities of produc- tion in 197-208; in the United States, centralization of 201; de- centralizing tendencies 202-4; enterprises peculiar to cities 206; in city and country 315, 317; and the birth-rate 339-42 Marriage, influence on migration 278, 284; rates of in city and country r 490 INDEX OF SUBJECTS 3 1 9-30 ; among self-supporting women 320-1; age at 322; among city-born and immigrants 328-9 Massachusetts, urban population in 1880, p 27; 1890, 28; distribution of population 36-8; growth of cities 37; percentage of urban pop- ulation 31 ; birth places of inhabi- tants 266; natural movement of population 318-9; marriage rates 320-2; conjugal condition 324; fecundity of women 334; infant mortality 334 Materialism engendered in cities 433; and industrialism 434-5; giving way to humanitarianism 440 Mediaeval cities, influence of 6 Melbourne, population 138, 450; area 139 Memphis, ancient 6, 448 Mercantile system 1 78-181, 214 Methods of measuring association 4 Mexico, city growth 132, 133 Middlesex, rapid growth 447 Migrants, sex of 276-280, 284; age of 280-2, 284; length of residence of 282-3, 284 Migration cityward in sixteenth century 222; in middle ages 230 et seq ; in France 230-1; England 231 ; Ger- many 231; not new 283; a short- distance movement 283; effect on marriage 325, 328-9; birth-rate 342-3; death-rate 365-6; harmful tendencies of 382; function of in evolution 388 ; and municipal gov- ernment 429-30; effects of, on villages 437-9 Migration, freedom of, and city growth 214; internal and urban growth, ch. iv, 230-285; effect upon birth and death-rates 235-6, 239, 247; volume of 248 ; range of, in leading countries 249-253; increase of in Europe 251-254; laws of 255-263; chiefly for short distances 255, 268, 270; by stages 257, 258, 267, 270- I; induced by large cities 257; of rural and urban populations 260—3 Military considerations in city location 174; concentrated in cities 457 Milk, sterilized, and infant mortality 354, 361 Mining favors dispersion of population 223; and the birth-rate 342; camps, average income small in 420 Mir 170 Mississippi, density and concentration of population 4; stationary urban percentages 20; percentage of urban population 32 Mississippi Valley, settlement 25 Missouri, density and concentration 4; urban population 28-9, 30, 31 Mobility 249-25 1 ; of rural and urban populations 260-5, 272; great city populations 273-4, 283; in industry and society 388 Monasteries as nuclei of towns 174 Money, introduction of 186 Montivideo, natural advantages 148 Morality in city and country 399-409 Mortality, in city and country 343-68; cities, causes of, 348-50; preventive measures against 351-4; reduction of in cities 355-65; according to occupation 360; infantile, in city and country, Prussia 295-6, 363- 4; Mass. 334 Mortality infantile, in Peabody buildings 352; New York 354; Austria and Vienna 356; increases urban mor- tality 360; cause of 361 ; reduction of in European cities 362; Bavaria 362; remedies for 361; period of highest 363-4; relation to birth-rate 365; to death-rate 365 Moscow, factory industries 105; popu- lation 450 Motive power, future distribution of 458 Motives of migrants to the cities 213-222 Music in cities 218, 224 Natal, concentration of population 137 National economy 170 Nationalities, inclination for city life 308-9 Negroes, concentration of female, in cities 279-80; infantile mortality among 296; migration of to cities 310-4; mortality of 313-4; in factories 313; conjugal condition 326; in New York and Philadel- phia in 1830, 431 Neighborliness, lack of, in cities 432-6 Netherlands. See Holland. New England, manufactures 25; births and deaths 242, 318, 344 Newfoundland, concentration of popu- lation 136 New Hampshire, density of population 4; percentage of urban population 31 INDEX OF SUBJECTS 491 New Jersey, urban population 29, 30, 31 New South Wales. See Australia. Newspapers, city and country 398-9 New York, definition of cities 13; urban population 28, 30, 31 New York City, population 21, 450; and the Erie canal 25; area 139; vs. Chicago, future growth 174; and Chicago, freight rates between 204; leading industries 207; ratio of the sexes 286; metropolitan dis- trict, mortality 346; healthfulness of Tenth Ward 349-50 ; reduction of death-rate 356; growth of metropolitan 447, 450, 453; con- gested districts 459; decrease of density by wards 460-2; density 468; suburban travel 470 New Zealand. See Australia Nicaragua, concentration of population 136 Nineveh 6, 448 North Dakota, so-called "cities" 12; percentage of urban population 32 Norway, definition of cities 7; city growth 1 1 1-2; emigrants to the United States, 152 Nourishment of urban and rural dwell- ers 366 Nuptialite. See Marriage Occupation statistics 228 9; of city peo- ple 296 8; and country people 314-7; and the birth-rate 341-2; death-rate 359-60; city and country born 373-81, 391 ; and suicide 403 Ocean, as a food-producer 227 Octroi, levied by cities 7 Ohio, urban population 28, 31 Oldenburg, migration 261, 272; deaths by ages 297 Opinion, public, in cities 433 Opportunity, the characteristic of city life 443 Orange Free State, concentration of population 137 Organism, industrial society likened to an 159, 183 Overcrowding in principal cities 349, , 416; relation to income 413; statis- tics of 415-7 Pagan, signified countrymen 3, 440, 443 Paraguav, concentration of population 136' , . i Paris, percentage increase of population 1801 to 1881, 7c; population 73, I 74; relative growth 76; area 139; I the centre of a great plain 172; excess of births over deaths 236; manner of growth 239, 240, 246; surplus of women 289; sex at birth 294; illegitimacy 405; rank 449- 50; Gi eater, population 453; growth of, prohibited by law 454; density 468 Parish in Louisiana 16 Parks, provision of 353 Pauperism, city vs. country born in 383-4 Peasantry, fitness for army service 369— 70, 425; and cit}' slums 372, 383; the reservoir of \'igor 388, 394 Pennsylvania, urban population 28, 31 Persia, distribution of population 123 Peru, concentration of population 136 Philadelphia, population 21, 450; out- grown by New York 25; area 139; ratio of the sexes 286; congested wards 459; density 468; streetcar systems 469; suburban traffic 470 Philanthropy in cities 436-7 Philippine Islands, distribution of popu- lation 128 Philistine, peculiar to small cities 440 Physical environment, effect of, upon distribution of population 2, 5 Physiocrats 230 Piers, recreation 353 Plague, visitation of, in former centimes 233 Play-grounds, necessity of 354 Poland's factory industries 105 ; density of population 106 Policemen, city vs. country bom 382, 390 Politics, significance of the word 3, 6; and location of cities 172, 174 Population, natural movement of ch. vi, 318-67; law of 319, 326, 337-341; Spencer's theor}' of 388; argricul- tural, produces a surplus 424; agri- cultural, full of error 440 ; two modes of increase of 443-4 Portugal, city growth 120 Poverty and mortality 349-50 Power, military, increased by concentra- tion of population 425 Precinct in Texas 17 Primitive man, mode of residence 3 Proclamation, Fast Day, of Governor of New Hampshire 438; against the growth of T.ondon 454—5 "i Production on a large scale, advantages of 193-6 492 INDEX OF SUBJECTS Professional men, city born vs. country born 373-6; birth-rates among 387; in intellectual aristocracy 388 Professions in city and country 315 Progress, penalty for, paid by cities 443 Proletariat, the urban 372-83, 388-9 Property holders diminish in cities 426-7 Prostitution, city vs. country born in 384; in cities 406 Protectionism 181, 214, 369 Prussia, definition of cities 8; city growth 80-83, 92; internal migra- tion 249-253; sex at birth 294; infant mortality 295; age-grouping 303; birth-rates, 331-2; death rates 344; deaths by ages 356-7; incomes 41 1 ; wages 412 Queensland. See Australia Race, in city and country 304-14; and the birth-rate 334, 343; and suicide .4°3 Radicalism engendered in cities 432, 439 Railway discriminations 2,Z Railways and cities in the United States 25, 172; England 55-56; France 79-80; Germany 87; Austria 97- loo; as a cause of concentration of population 152; and village trade 1 89-90; relation to manufacturing sites 198-9; under the competitive regime 199; and cities in Germany 200-1 ; tending to favor the small factory town 203-4; a single uni- form rate on 204; effect of, upon migration 259; function of street 474 Rank, social and industrial 3x6-7 Recruits for the army, city vs. country born 382 Religion, place of, in industrial evolu- tion 159, 171; in city and country ! 399-401; state of, in villages 438 Remedies for concentration of popula- tion 454-75; the four commend- able 474 Rent, consumers, in cities 219, 412-3; ' house, problem of 413-5 Residence, length of, of outsiders in cities, 282-3, 284 Retail trade, evolution of 190 Returns, law of diminishing discussed 225-7; constant 228 Revolution, agrarian 164-167 Revolutionary movements in cities 7 Rides, number of, in street cars, in various cities 469 Rio de Janeiro, growth 134, 450 Riots, in Philadelphia and New York 431 Roman cities, influence of 6 Rome, food supply 163; population 448, 450 Roumania, city growth 122 Russia, city growth 105-109; manufac- tures 105 Russian Jews, health of 350 Saint Louis, population 450; density of population 468; suburban travel 470 Saloons in New York cities 406-7 Salvador, concentration of population San Francisco, population of in 1852, 22 Sanitary districts, urban and rural, in England 13 Sanitation and death-rates 233 "Saturation," point of 51 Savings, small among common laborers in cities 421 Saxony, city growth 92-3 ; fecundity of women 333; illegitimacy 333; in- fant mortaHty 362-3; suburban growth 458-9 Scattered populations, in primitive so- ciety 3; in France and Italy 9; in final classification 16 School-children, medical inspection of 354 Schools, in cities 218; rural districts 218, 222, 437; urban and rural 397. 438 Scotland, urban growth 57-64, esp. 58; creation of forest preserves 63 Sects, religious, in Americrn cities 400-1 Selection, cities as instruments of natural 441-5 Servants, domestic, m migration, 278-9, 284, 289; from the country 371; origin of city 374-6, 391; birth- rates 387 Servia, city growth 121 Settlement fees proposed to check mi- gration 455 Settlements, social 436 Sex of migrants 276-280, 284, 290; in cities 285-300; atbiith294; causes of 294-5; ratio of, effected by deaths 297-8 Sexes, distribution of in the United States 279-280 Shelburne Falls murder 407 Slums, cheap labor supplied by 206; INDEX OF SUBJECTS 493 both cause and effect 210; and demolition of tenements 352-3; created by the country-born 372; connection with high rents 414; remedies for 414 Socialism in cities 219, 435; (social legislation) spread of 435 Society, industrial, likened to an organ- ism 159 Sofia, population 121 Soil, feitility of, influences location of city 172 Soldiers from city and country 369-72, 383. 388, 394-7 Solitude in city and country 432 South, influence of city growth upon the 440 South Carolina, percentage of urban population 20, 32 Spain, city growth 119 Specialization, geographical, in middle ages 179 Spectroscope, similarity of city to 442 Speed of transit system 471 Spinsters, preponderance in cities 322-7 Standard of life in city and country 218 Stature in city and country 393-4 Steam, relation of to concentration of population 158, 173; the factory system 192-3 Steam-power, increase of, in France 78 Stock-breeding, Bakewell's contribu- tions to 165 Stockholm, relative growth 1 805-1 890, no; vital statistics of 181 6-1 890, 237; marriage rates 320 Stores, department 190, 195, 413 Strand, The, in London, population since 1801, 464 Strangers and city government 429-30 Straw-plaiting, a rural industry 320 Stream, downward, of failures 443 Street railways, electric, and density of American cities, 469 Students, birth-place of city 374-7; o* Berlin University in war 396 Suburban annexations, mode of treating 18-19; ratio to total increase 240; of Boston 241 Suburbs, growth of, in New York 36; Massachusetts 37-38; England 51, 57; Scotland 62; Ireland 65; France, 75, 76, 77; Germany 86; Austria 95; Denmark 112-113; Belgium 116; India 126; Australia 139, 141 ; movement of factories toward 202 ; in England and Ger- many 203, 224, 228, 473; move- ment to the, from London 276; young couples 325; model tene- ments in 353; wealth in Frankfort 373; athletic development in 396- 7; promotive of health fulness 445; grow fastest among Austrian cities 446, 458; English 446-7; Saxon 458-9; significance of the growth of 458-74; growth of Berlin 465-7; Hamburg 467; travel to and from 470-1; growth of Sydney 472; advantages of for manufacturing, 473; for residences, public policy toward 473-4; in Frankfort 474 Suicide 401-3, 443 Surrey contains London suburbs 447 Survival of the unfittest 444 Susa 448 " Sweat shop," attitude of society to- ward 196-7; and the labor market 206 Sweden, definition of cities 7; city growth 109-110; vital statistics of 1816-90, 237; internal migra- tion 249-253; marriage rates 320 Switzerland, city growth 117; Allmende 216; internal migration ■2\()-'Zt)T, Sydney, population 1891, 138; area 139; suburbs 139, 472; growth 141 Tasmania. See Australia. Taxation and cities 214, 425, 428 Teachers, city vs. country-born 374, 390; birth-rates 387 Tenancy increased by city growth 426 Tenement population, lack of mobility in 221, 456; houses, mortality in 348-52; legislative policy towards 351; model 352-3,415; Peabody, in London 352; problem of 414; overcrowded 416-17; evils in 1593, 454 Thebes 6, 448 Town, Is it urban or rural? 14; vs. town- ship 17; in American and English usage 17; origin of, from trade l6o, 171; in middle ages 170, 176; and country, differentiation of 1 76 Township vs. town 17; variations in size of 18; average size in principal countries 142 Trade and the food supply 3; internal, and city growth 214; in city and country 315, 317; city vs. country born in 374-81, 391 494 INDEX OF SUBJECTS Trades, place of the, in industrial evolu- tion 159 Trade Unions, strength of, in cities 205, 419; attitude of, toward domestic industries 456; suburban industries 473 Trains, workingmen's, in Belgium 472; England 472 Tramps, influx of, into the cities 420-1 ; taking care of, in cities 422 Transit, rapid, necessity of 354, 41 5, 470- 4; Belgian solution 47 1-2; English solution 472; need of cheap 474 Transport facilities and the food-supply 3 Transportation facilities and location of cities 1 72; and distribution of popu- lation 183; influence upon agricul- tural and mining populations 197; upon manufacturing populations 197-204; as a factor in production 204; city vs. country born in 374- 81,391 Transvaal, concentration of population 137 Travel, street railway 469 ; suburban, in American cities 470 Trolley car 469-74 Tunnels under all city streets 458 Turkey, city growth in European 120; Asiatic 122 Undertakers, see Entrepreneurs. Unemployment in cities 420-1 ; Austra- lian cities 423 United States in 1790, compared with Australia in 1891, i; density and concentration of population 4; dwelling centres in, law regarding 12; number of incorporated cities 13; statistics of urban growth 20- 40, esp. 22; census definition of urban population 21 ; leading events in industrial history 23-27; number of farms 26; production of cereals 26; growth of manufactures 27; distribution of its urban population 28; large cities and villages 29; percentage of urban population in the individual States and Territories 31-32; relative rates of increase of groups of cities 34-36; distribution of population i8co, 1850, 1890, 39- 40 ; future growth of cities 39; agri- cultural progress 167; the grain trade 168; factory system 1 87 ; de- cline of villages i88; effect of rail- way discriminations 199; centrali- zation of manufacturing 201-2; decentralizing tendencies in manu- factures 202-4; rural depopula- tion 211; birth and deatli rates in cities 241-3 ; statistics of birth-place 249; decline of inter-state migra- tion 251; character of the migra- tory movement 263-6; sex of mi- grants within the 279-80; surplus of women in cities 286, 288; girls in cities 292; infantile mortality among negroes 296; age-grouping 301 ; concentration of foreigners 304-9; natural movement of popu- lation 318; conjugal condition 323, 326; divorces 330; size of families 336-7; and the theory of popula- tion 338-41; death-rates 345-6; the insane, feeble-minded, blind, deaf and dumb 392; urban and rural education 397-8; churches 400-1; suicide 401; saloons in New York State 406-7; distribution of wealth 1787 and 1898, 425; tenancy 426; city-government 428-31; local (rural) government 429; separa- tion of social classes 434; suburban growth 447-8; density of principal cities 468; suburban travel 470 Urbane, significance of the word 3 Uruguay, concentration of population 136; natural configuration 148 Vagrants in cities 420-2 Vapors, how carried off in future cities Venezuela, concentration of population 136 Venice, influence 6, 177 Vermont 4, 32 Vice, in cities 404-7 Victoria, agrarian policy 217. See also Australia. Vienna, population 1786, 1890, 95; pro- portionate growth 96; and the railways 97; mobihty of natives 274; ratio of sexes 278, 291 ; re- duction of death rate 356; city and country born 372, 378-80; growth of suburbs 458 Vigor of city and country men 368-97 Village economy 170 Villages, decline of 188-189; advan- tages of, for manufacturing 204-9; lack of prizes 213; emigration from, produces stagnation 437-8; pro- posals to increase attractions of 456 INDEX OF SUBJECTS 495 Virtues in cities 408-9 Wages in city and country 411, 412; attract immigrants 420 Waiters, city vs. country born 374-Sl, 390 Wales. See England. War of I-812, and stagnation of com- mercial cities 23; Civil War and city growth 26; place of. in indus- trial evolution 159. See also Army, Military, Soldiers. Water communication vs. railways 173; rates for city 353; for city houses 458 Wealth, growth of, favors concentration of population 223-4; distribution of, in cities 372-3, 410-11; dis- tribution of, in 1787 and 1898, 425 Wen, similarity of great city to a 439 Women, greater migrants than men 276- 280, 284; are short-distance mi- grants 279, 2S4; excess of, in cities 285-9; excess of, in cities, causes of 289-300 ; employment of, and the marriage rate 320-1; married, in factories 361 ; city-born in Ber- lin industries 373-6; city-born in Vienna industries 378-80 Workers, percentage of, in city and country 315 Workingmen in cities 419-421; public policy toward 474 Widows, preponderance of, in cities 325 Yokohama, growth 130 STCDIES IN HISTORY, ECONOMICS AND PUBLIC LAW EDITED BY THE FACULTY OF POUTIGAL SCIENCE OF COLUMBIA UNIVERSITY. VOLUME 1, 1891-2. Second Edition, 1897. 396 pp. Price, $3.00 ; bound, $3.50, X. The Divorce Problem — A Study in Statistics. By Walter F. Willcox, Ph. D. Price, 75c. 2. The History of Tariff Administration in the United States, from Co- lonial Times to the McKinley Administrative Bill. By John Dean Goss, Ph. D. Price, $1.00. 3. History of Municipal Land Ownership on Manhattan Island. By George Ashton Black, Ph. D. Price, $2.00. 4. Financial History of Massachusetts. By Charles H.J. Douglas, Ph. D. {Not sold separately.') VOLUME II, 1892-93. 503 pp. Price, $3.00 ; bound, $3.50. 1. The Economics of the Russian Village. By Isaac A. Hourwich, Ph. D. {Out of print.) 2. Bankruptcy. A Study in Comparative Legislation, By Samuel W. Dunscomb, Jr., Ph. D. Price, $1.00. 3. Special Assessments : A Study in Municipal Finance. By Victor Rosewater, Ph. D. Second Edition, 1898. Price, $1.00. VOLUME III, 1893. 465 pp. Price, $3.00; bound, $3.50. 1. History of Elections in the American Colonies. By Cortlandt F. Bishop, Ph. D. Price, $1.50. Vol. Ill, no. 7, may also be obtaijied bound. Price, I2.00. 2. The Commercial Policy of England toward the American Colonies. B5' George L. Beer, A. M. Price, $1.50. VOLUME IV, 1893-94. 438 pp. Price, $3.00 ; bound, $3.50. 1. Financial History of Virginia. By William Z. Ripley, Ph.D. Price, $1.00. 2. The Inheritance Tax. By Max West, Ph. D. {Out of print .) 3. History of Taxation in Vermont. By Frederick A. Wood, Ph. D. Price, $1.00 VOLUME V, 1895-96. 498 pp. Price, $3.00 ; bound, $3.50. 1. Double Taxation in the United States. By Francis Walker, Ph, D. Price, $1.00. 2. The Separation of Governmental Powers. By William Bondy, LL. B., Ph. D. Price, $1.00. 3. Municipal Government in Michigan and Ohio. By Pelos F. Wilcox, Ph. D. Price $1.00. VOLUME VI, 1896. 601 pp. Price, $4.00 ; bound, $4,50. History of Proprietary Government in Pennsylvania. By William Robert Shepherd, Ph. D. Price, $4.00; bound, $4.50. VOLUME VII, 1896. 512 pp. Price, $3.00 ; bound, $3.50. 1. History of the Transition from Provincial to Commonwealth Govern- ment in Massachusetts. By Harry A. Cushing, Ph. D. Price, $2.00. 2. Speculation on the Stock and Produce Exchanges of the United States. By Henry Crosby Emery, Ph. D. Price, $1.50. VOLUME VIII, 1896-98. 551 pp. Price, $3.50; bound, $4.00. 1. The Struggle between President Johnson and Congress over Recon- struction. By Charles Ernest Chadsey, Ph. D. Price, $1.00. 2. Recent Centralizing Tendencies in State Educational Administration. By William Clarence Webster, Ph. D. Price, 75c. 3. The Abolition of Privateering and the Declaration of Paris. By Francis R. Stark, LL. B., Ph. D. Price, $1.00. 4. Public Administration in Massachusetts. The Relation of Central to Local Activity. By Robert Harvey Whitten, Ph. D. Price, $1.00. VOLUME XI, 1897-98. 617 pp. Price, $3.50 ; bound, $4.00. 1. English Local Government of To-day. A Study of the Relations of Cen- tral and Local Government. ByMilo Roy Maltbie, Ph. D. Price, $2.00. Vol. IX, no. J, may also be obtained botmd. Price, I2.50. 2. German Wage Theories. A History of their Development. By James W. Crook, Ph. D. Price, $1.00. 3. The Centralization of Administration in New York State. By John Archibald Fairlie, Ph.D. Price, $1.00, VOLUME X, 1898-99. 500 pp. Price, $3.00 ; bound, $3.50. I Sympathetic Strikes and Sympathetic Lockouts. By Fred S. Hall, Ph.D. Price, $1.00 2. Rhode Island and the Formation of the Union. By Frank Greene Bates, Ph.D. Price, $1.50. Vol. X, no. 2, may also be obtained bound. Price, I2.00. 3. Centralized Administration of Liquor Laws in the American Common- wealths By Clement Moore Lacey Sites, Ph.D. Price, $i.co. VOLUME XI, 1899. 495 pp. Price, $3.50; bound, $4.00. The Growth of Cities. By Adna Ferrin Weber, Ph.D VOLUME XII, 1899-1900. 586 pp. Price, $3.50; bound, $4.00. 1. History and Functions of Central Labor Unions. By William Maxwell Burke, Ph. D. Price, $1.00, 2. Colonial Immigration Laws. By Edward Emberson Proper, A.M. Price, 75c. 3. History of Military Pension Legislation in the United States. By William Henry Glasson, Ph.D. Price, $1.00. 4. History of the Theory of Sovereignty since Rousseau. By Charles E. Merriam, Jr., Ph.D Price, $1.50. VOLUME XIII, 1901. 570 pp. Price, $3.50 ; bound, $4.00. I. The Legal Property Relations of Married Parties. « By Isidor Loeb, Ph.D. Price, $1.50 3, Political Nativism in New York State. By Louis Dow Scisco, Ph. D. Price, $2.00. 3, The Reconstruction of Georgia. By Edwin C. Woolley, Ph.D. Price, $1.00. 1 VOLUME XIV, 1901-1902. 576 pp. Price, $3.50 ; bound. $4.00. 1. Loyalism in New York during the American Revolution. By Alexander Clarence Flick, Ph. D. Price, $2.00 2. The Economic Theory of Risk and Insurance. By Allan H. Willett, Ph. D. Price, $1.50. 3. The Eastern Question. A Study in Diplomacy. By Stephen P. H. Duggan, Ph. D. Price, $1.50. VOLUME XV, 1902. 427 pp. Price, $3.00; bound, $3.50. Crime in its Relations to Social Progress. By Arthur Cleveland Hall, Ph. D VOLUME XVI. 1902. 1. The Past and Present of Commerce in Japan. By Yetaro Kinosita, Ph.D. Price, $1.50. 2. The Employment of Women in the Clothing Trade. By Mabel Hurd Willett, Ph. D. Price, $1.50. The set of fifteen volumes (except that Vol. II can be supplied only in unbound nos. 2 and 3) is offered bound for $50. For further information apply to Prof. EDWIN R. A. SELIGMAN. Columbia University, or to THE MACMILLAN COMPANY, New York, London : P. S. KING & SON, Orchard House, Westminster. * rf> %4" ^ -n^. -c^* •%/" .x^''"- :-^:^ .x\~^' ^^i^/!?Z^ ,-0 r ..o>^.. "-<,/^-^^ xx^ ""'::^. ■;'^;^. .^^ -"-h. ^J~ V •^^ ' -v '■■■ V.^-^A"'^ -> ^"' ,-^ <^' ON'. ^V\^-- -^'^ -O ^0 o. .0' ''b, xv>' ■^>. - ■'^. 'a V Oo. '^., .0 \- ^'^ ^tK%^n^ <-■ ',r'. .^' ■■ ' -^^ a\ aN -^ -v ■ ^ * 1 ^^^ ■ q^ ^J> %/*::^;>' '\^^ ^ :".: 0- s^ ■■ ' '/. c- >' vx'^-