/^ V: 
 
 ■:\>^ 
 
 ^^-t^. 
 
 
 % 
 
 ^2' 
 
 ^^^ 
 
 .*-..^ 
 
 
 ^ 
 
 
 
 
 
 
 '^ I" ~ 
 
 ^ 
 
 /^..^^H 
 
 '***-.;: 
 
 
 ^'.v^A 
 
 ,.^ 
 
 'V^i 
 
 -r' 
 
 
 -A 
 
 
 
 
 
 ^•^ 
 
 ■41 
 i 
 
 MM-S'f'tirVv,?. 
 
 f 
 
 , ^ 
 
 •.V- 
 
 
 
 
 
 ^■■- 
 
 
 ■■■'Ipl^f-' 
 
 
 
 /iiff' "■■■■■ 
 
 
 
 
 ,„,::'. 1 
 
 ^i- 
 
 ■>'•'- . Uv-ZV -■■■■•■"■■ — ^' 
 
 ^:^^X 
 
 ^^ 
 
 
 •Ssisi, 
 
 V 
 
 ,2fii*^tSv^':!^S 
 
 ■ssS 
 
 <»— "^ 
 
 '"■^0S1, '■'& 
 
 
 " 
 
 A.. 
 
 ■■?■'■■■;- 
 
 
HA 37.U5''l90r""'"'"* '■"'""* 
 *KiiiS,.Sf.';!S,I?.'''"S from the first c 
 
 3 1924 013 916 030 
 
DEPARTMENT OF COMMERCE 
 AND LABOR 
 
 AMERICAN 
 ICENSUS TAKING 
 
 '•*-j 
 
 [; BUREAU OF THE CENSUS 
 
Cornell University 
 Library 
 
 The original of tiiis book is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924013916030 
 
o 
 
 cq 
 ■s> 
 
 7, 
 W 
 u 
 
 w 
 
 M 
 
^mternmrttt ^Pnnting ®ffi« 
 
BUREAU OF THE CENSUS 
 S. N. D. NORTH, DIRECTOR 
 
 AMERICAN 
 CENSUS TAKING 
 
 FROM THE FIRST CENSUS 
 OF THE UNITED STATES 
 
 P 
 
 ^^ 
 
 REPRINTED {rom THE CENTURY <<: 
 
 MAGAZINE for APRIL, 1908* ' ' \ 
 
 C-, '"•t'f./t [C 
 
 \::% 
 
 ■ ^^^^vzai 
 
The developments of statistics are causing history to be re- 
 written. Till recently, the historian studied nations in the aggre- 
 gate, and gave us only the story of princes, dynasties, sieges, and 
 battles. Of the people themselves — the great social body, with life, 
 growth, forces, elements, and laws of its own — he told us nothing. 
 Now, statistical inquiry leads him into hovels, homes, workshops, 
 mines, fields, prisons, hospitals, and all other places where human 
 nature displays its weakness and its strength. In these explora- 
 tions he discovers the seeds of national growth and decay, and thtis 
 becomes the prophet of his generation. 
 
 The chief instrument of American statistics is the census, 
 which should accomplish a twofold object. It should serve the 
 country, by making a full and accurate exhibit of the eletnents of 
 national life and strength; and it should ser-oe the science of sta- 
 tistics by so exhibiting general results that they may be compared 
 with similar data obtained by other nations. The census is indis- 
 pensable to modem statesmanship.— f AMES A. GARFIELD. 
 
 (4) 
 
 b'~^7^1 
 
DEPARTMENT OF COMMERCE AND LABOR 
 
 SECRETARY 
 
 HON. GEORGE B. CORTELYOU 
 
 BUREAU OF THE CENSUS 
 
 DIRECTOR 
 S. N. D. NORTH 
 
 CHIEF CLERK. 
 
 EDWARD MCCAULEY 
 
 DISBURSING AND APPOINTMENT CLERK 
 
 JOHN W. LANGLEY 
 
 CHIEF STATISTICIANS 
 
 WILLIAM C. HUNT, population 
 WILLIAM M. STEUART, manufactures 
 LE GRAND POWERS, agriculture 
 WILLIAM A. KING, vital statistics 
 
 CHIEFS OF DIVISIONS 
 
 CHARLES S. SLOANE, geographer 
 
 WILLIAM H. JARVIS, population 
 
 JOSEPH D. LEWIS, manufactures 
 
 HART MOMSEN, agriculture 
 
 R. C. LAPPIN, VITAL statistics 
 
 JOSEPH A. HILL, revision and results 
 
 WILLIAM S. ROSSITER, printing and mailing 
 
 (5) 
 
DIRECTORS OF THE FEDERAL CENSUSES 
 
 CENSUS OF 1840. 
 WILLIAM A. WEAVER, superintending clerk (1840-1842). 
 
 CENSUS OF 1850. 
 
 JOSEPH C. G. KENNEDY, superintendent (1850-1853). 
 
 JAMES B. DE BOW, superintendent (1853-1854). > 
 
 CENSUS OF I860. 
 JOSEPH C. G. KENNEDY, superintendent. 
 
 CENSUS OF 1870. 
 FRANCIS A. WALKER, superintendent. 
 
 CENSUS OF 1880. 
 FRANCIS A. WALKER, superintendent (1879-1881). 
 CHARLES W. SEATON, superintendent (1881-1885). 
 
 CENSUS OF 1890. 
 ROBERT P. PORTER, superintendent (1889-1893). 
 
 CARROLL D. WRIGHT (commissioner of labor, in charge 
 1893-1897). 
 
 CENSUS OF 1900. 
 WILLIAM R. MERRIAM, director (1899-1903). 
 
 S. N. D. NORTH, director (1903- ). 
 
 I Mr. De Bow served again in 1855, when the census work was resumed 
 temporarily, and later Mr. Jo-<5eph C. G. Kennedy was reappointed as super- 
 intending clerk, and served from 1858 until he was appointed Superintendent 
 of the Eighth Census. 
 
 (6) 
 
AMERICAN CENSUS TARING' 
 
 THE Constitution of the United States requires an 
 enumeration once in ten years as a basis for the 
 apportionment of Representatives in Congress. The 
 modern American census, which collects statistics 
 relating to everj- important feature of national 
 development, is the outgrowth of that requirement. 
 Beginning with six simple questions relating to pop- 
 ulation, the amount and scope of the information 
 secured were steadily increased until it became too 
 extensive to be tabulated by hand within a reasonable period. 
 
 The first Census Act was passed at the second session of the First 
 Congress. In accordance with the practice of that period, the 
 task of securing the first enumeration of inhabitants was placed 
 upon the President, whose duties included active supervision of 
 all the routine affairs of Government. In the days of the early 
 Presidents even the issuing of a patent, afterwards a mere incident ■ 
 in bureau routine, was a matter for Presidential consideration, 
 requiring a parchment from the State Department, signed by the 
 President, the Secretary of State, and the Attorney-General. 
 Indeed, all duties which did not clearly devolve upon some 
 department or ofiicial of the youthful Republic fell to the Presi- 
 dent himself. 
 
 The census law was signed by President Washington, March i , 
 1790. By it the marshals of the several judicial districts were 
 required, with the aid of assistants to be appointed by themselves, 
 to ascertain the number of inhabitants within their respective 
 districts, omitting Indians not taxed, and distinguishing free 
 persons (including those bound to service for a term of years), 
 the sex and color of free persons, and the number of free males 
 sixteen years of age and over. 
 
 The object of the inquiry last mentioned was, undoubtedly, to 
 obtain definite knowledge as to the military and industrial 
 strength of the country. This fact possesses special interest 
 because the Constitution directs merely an enumeration of inhabi- 
 tants, and it appears, therefore, that the demand for information 
 
 Origin. 
 
 The First 
 Census law. 
 
 Early 
 inquiries. 
 
 • This paper, by Hon. "William R. Merriam, formerly Director of the Census, 
 appeared in the Century magazine for April, 1903. A few paragraphs as it 
 was originally printed have been omitted from this reprint; the sections 
 describing automatic electrical machinery were not part of the original paper. 
 
 (7) 
 
8 
 
 Method of 
 procedure at 
 the First 
 Census. 
 
 Problems 
 encountered 
 at the First 
 Census. 
 
 more extensive than that previously required, which has been so 
 marked a characteristic of census legislation, began -with the first 
 Congress that dealt with the subject. 
 
 The method followed by the President in putting into operation 
 the First Census law, although the object of extended investiga- 
 tion, is not definitely known. It is generally supposed, and 
 occasionally stated as though beyond challenge, that the Presi- 
 dent or the Secretary of State dispatched copies of the law, and 
 perhaps of instructions also, to the marshals. There is, however, 
 some ground for disputing this conclusion. 
 
 At least one of the reports in the census volume for 1790 was 
 furnished by a governor. This, together with the fact that there 
 is no record of correspondence on the subject of the census with 
 the marshals but that there is a record of such correspondence 
 with the governors, makes very strong the inference that the mar- 
 shals received through the governors of the states their instruc- 
 tions relating to the First Census of the Republic. This inference 
 is strengthened by the fact that in 1790 the state of Massachusetts 
 furnished printed blanks, which it would have been unlikely to 
 do had the state officials been in no way connected with the 
 enumeration, and also by the fact that the law relating to the Sec- 
 ond Census specifically charged the Secretary of State to superin- 
 tend the enumeration and to communicate directly with the 
 marshals. 
 
 It will be interesting to consider briefly the difficulties which 
 confronted President Washington, the first Superintendent of 
 Census. In March, 1790, the Union consisted of twelve states — 
 Rhode Island, the last of the original thirteen to enter the Union, 
 being admitted May 29, and Vermont, the first addition, in the 
 following year, before the results of the First Census were an- 
 nounced. Maine was a part of Massachusetts, Kentucky was a 
 part of Virginia, and the present states of Alabama and Missis- 
 sippi were parts of Georgia. The present states of Ohio, Indiana, 
 Illinois, Michigan, and Wisconsin, with part of Minnesota, were 
 known'as the Northwest Territory, and the present state of Ten- 
 nessee, then a part of North Carolina, was soon to be organized 
 as the Southwest Territory. 
 
 The gross area of the United States was 827,844 square miles, 
 but the settled area was only 239,935 square miles, constituting 
 about 29 per cent of the total. Western New York was a wilder- 
 ness, Elmira and Binghamton being but detached hamlets. With 
 the exception of a portion of Kentucky, the territory west of the 
 Allegheny mountains was unsettled and scarcely penetrated. De- 
 
troit and Vincennes were so small and isolated as to merit no consid- 
 eration, and they were not included in the report of the First Census. 
 Philadelphia was the capital of the United States. Washington 
 was a mere Government project, not even named, but known as 
 the Federal City. Indeed, by the spring of 1793 only one wall 
 of the White House had been constructed, and the site for the 
 Capitol had been merely surveyed. 
 
 75 
 
 70 
 
 65 
 60 
 55 
 50 
 *5 
 40 
 35 
 30 
 25 
 20 
 15 
 <0 
 5 
 
 90 (E 
 
 00 18 
 
 18 
 
 ZO '& 
 
 30 I& 
 
 40 IS 
 
 50 le 
 
 00 le 
 
 If 
 
 eo 18 
 
 730 / 
 
 19 
 
 iSOO 
 00 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 , 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 A 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 y 
 
 / 
 
 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 DIAGRAM SHOWING THB TOTAI< POPUI<ATION OF THB UNITEB 
 STATES AT THE SEVERAI, CENSUS PERIODS FROM 1790 TO 1900. 
 
 The United States was bounded on the west by the Mississippi 
 river, beyond which stretched that vast and unexplored wilder- 
 ness belonging to the Spanish King which was afterwards ceded to 
 the United States as the Louisiana Purchase, and which, now com- 
 prises the great and populous states of Louisiana, Arkansas, Indian 
 Territory, Oklahoma, Missouri, Kansas, Iowa, Nebraska, South 
 Dakota, North Dakota, and Montana, and most of Colorado, 
 Wyoming, and Minnesota. The Louisiana Purchase was not 
 completed for more than a decade after the First Census was taken. 
 
10 
 
 and yet the centennial of the participation of the states of that 
 vast area, a dominion in itself, in the privileges of the Union, is 
 being celebrated in a city of more than 600,000 inhabitants, which 
 at the time of the purchase was not even atrader's outpost. On the 
 south was another Spanish colony known as the Floridas. Texas, 
 then part of the colony of Mexico, also belonged to Spain, as did, 
 indeed, those far-away southwestern regions now divided into 
 California, Utah, Arizona, and New Mexico, which, while pene- 
 trated here and there by venturesome explorers and missionaries, 
 were, for the most part, an undiscovered wilderness. 
 
 Though the area covered by the enumeration in 1790 seems 
 very small when compared with the present area of the United 
 States, the difficulties presented were vastly greater than in 1900. 
 In many localities there were no roads, and where these did exist 
 they were poor and frequently impassable; bridges were almost 
 unknown. Transportation was entirely by horseback, stage, or 
 private coach. A journey as long as that from New York to 
 Washington was a serious undertaking, requiring eight days 
 under the most favorable conditions. According to the unpub- 
 lished diary of a citizen of Connecticut, who in the spring of 1793 
 traveled on horseback from Georgia to his home near New Haven, 
 the trip consumed thirty-nine days, and required an expenditure 
 of $72 for subsistence alone. Moreover, this amount was paid in 
 different kinds of currency, each of which was accepted only in 
 certain localities. 
 
 Mails were transported in very irregular fashion, and correspond- 
 ence was expensive and uncertain. New York city, which in 1900 
 ranks as the second city in the world in population, with 3,437,202 
 inhabitants, in 1790 possessed a popiilation of only 33,131 
 although it was the largest city in the United States; Philadelphia 
 was second, with 28,522; and Boston third, with 18,320. 
 
 There were, moreover, other difficulties which were of serious 
 importance in 1790, but which long ago ceased to be problems in 
 census taking. The boundaries of towns and other minor divi- 
 sions, and even those of counties, were in many cases unknown 
 or not defined at all. The hitherto semi-independent states had 
 been under the control of the Federal Government for so short a 
 time that the different sections had not yet been welded into a 
 harmonious nationality in which the Federal authority should be 
 unquestioned and instructions promptly and fully obeyed. The 
 inhabitants, having no experience with census taking, imagined 
 that some scheme for increasing taxation was involved, and thus 
 to the other difficulties of the assistant marshal was added the 
 
11 
 
 caution of the citizen lest he reveal too much of his own affairs. 
 Moreover, there was opposition to enumeration on religious 
 grounds, for the Old Testament recorded a most unpleasant 
 account of a venturesome king who brought down the wrath of 
 Heaven bv taking a census of the children of Israel— regarded by 
 many as a warning of possible disaster to the Republic. 
 
 1790 jBoo raio If 
 
 !0 1830 18 
 
 afihelM 
 to M. 
 
 so 18 
 
 SO 1870 1880 (ff 
 
 30 
 00 
 
 12 
 
 ■0 
 
 9 
 8 
 
 7 
 
 6 
 
 b 
 
 A 
 3 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 s 
 
 / 
 
 
 
 
 
 
 
 
 / 
 
 \ 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 DIAGRAM SHOWING THE ABSOI<UTE INCREASE IN POPULATION 
 OF THE UNITED STATES FOR EACH CENSUS PERIOD FROM 1790 
 TO 1900. 
 
 At the First Census nine months were allowed in which to carry 
 out the provisions of the law. There were 17 marshals. The 
 records showing the number of assistant marshals (now called 
 enumerators) employed in 1790, 1800, and 1810 were destroyed by 
 fire, but the number employed in 1790 has been estimated at 650. 
 
 The returns of the marshals were sent to the President, and by 
 him turned over to the Secretary of State. Little or no tabulation 
 was required, and the report of the First Census, as well as the 
 reports of the Second, Third, and Fourth, was produced without 
 
 Time 
 
 consumed in 
 enumeration. 
 
12 
 
 Cost of the 
 First Census. 
 Comparison 
 of cost of the 
 First and 
 Twelfth 
 censuses. 
 
 the employment of any clerical force, the returns being trans- 
 mitted directly to the printer. The population of the United 
 States in 1790 was 3,929,326, and the entire cost of the census was 
 
 144,377. 
 
 The difference between the cost of securing the returns from 
 the six simple questions asked in 1790, and that of the extended 
 inquiry made a century later, is illustrated by the per capita cost, 
 which in 1790 was 1.13 cents and in 1900, 15.5 cents. In 1790 
 Virginia was the most populous state in the Union, having 
 
 DIAGRAM SHOWING PBR CENT OF INCREASE IN THE POPUl,A- 
 TION OF THE UNITED STATES FOR EACH CENSUS PERIOD 
 FROM 1790 TO 1900. 
 
 747,610 inhabitants. The records of the Treasury Department 
 show that at the First Census the cost of making the enumeration 
 in that state was $7,553. 90. Moreover, at that enumeration the 
 underpaid assistant marshals supplied their own blanks, an item 
 which was of considerable importance in the days when all paper 
 was made laboriously by hand. In 1900 the population of 
 Maine — about 700,000 — most nearly approximated that of Vir- 
 ginia in 1790. At the Twelfth Census the cost of actual enumera- 
 tion in Maine, including the pay of supervisors, was $34,560.90, 
 or more than three-fourths of the amount expended for the 
 
13 
 
 enumeration of the entire I'nited States in 1790, though the pay 
 of an enumerator in 1900 did not exceed the wages of an intelli- 
 gent day laborer. 
 
 Table i shows ver}' clearly the slow, steady expansion of census Expansion 
 inquiry up to 1850. Table 11 shows the variations in the time of census 
 allowed and consumed for enumeration, as well as other interest- ^^^^^O'* 
 ing details, to which no more detailed reference need be made. 
 
 Table I.— THK KXPANSION OF CKNSUS INQUIRY FROM 1790 TO 1850. 
 
 Head of family 
 
 Free white males over 16 
 
 Free white m.ales under 16 
 
 Free w^hite females 
 
 All other free persons 
 
 Slaves 
 
 Name of county, town, or city 
 
 Extension of age inquiry 
 
 Manufactures 
 
 Colored, by sex and age 
 
 ForeigTiers not naturalized 
 
 Occupation by three classes 
 
 Further extension of age inquiry 
 
 White deaf, dumb, and blind 
 
 Colored deaf, dumb, and blind 
 
 Printed schedules 
 
 Extension of occupation inquiry 
 
 "White and colored insane and idiots — 
 
 Education 
 
 Ag'riculture 
 
 Fish and fisheries 
 
 Mines and mining 
 
 Commerce 
 
 Revolutionary pensioners, with age 
 
 Dwellings and families numbered 
 
 Name of every free person 
 
 Value of real estate owned 
 
 Nativity 
 
 Married within the year 
 
 Paupers and convicts 
 
 Separate schedule for slaves 
 
 Mortality 
 
 Social statistics _ — 
 
 Inquiries concerning population, agricul- 
 ture, and manufactures systematized . 
 
 1820. 
 
 1830. 
 
 1840. 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 * 
 
 
 * 
 
 * 
 
 
 * 
 
 * 
 
 
 * 
 
 * 
 
 
 * 
 
 * 
 * 
 
 * 
 * 
 * 
 * 
 * 
 * 
 
 1850. 
 
 A summary of the results of the First Census, not including 
 the returns for South Carolina, was transmitted to Congress by 
 President Washington, October 27, 1791. The legal period for 
 enumeration, nine months, had been extended, the longest time 
 consumed being eighteen months in South Carolina. The report 
 of October 27 was printed in full, and published in what is now 
 a very rare little book of 56 pages; afterwards the report for 
 
fl 
 
 
 % 
 
 
 
 
 H 
 
 
 < 
 
 
 i-r 
 
 
 t3 
 
 
 m 
 
 
 •< 
 
 
 tn 
 
 
 Q 
 
 h 
 
 'f. 
 
 n 
 
 < O 
 
 f1 
 
 Q 
 
 ;? » 
 
 H 
 
 <! 
 
 iT 
 
 Q 
 W 
 
 
 
 
 
 (4 
 
 O 
 
 P 
 
 
 rn 
 
 a H 
 
 W »1 
 
 p( 
 
 
 
 
 
 & w 
 
 
 
 
 w 
 
 <! 
 
 Pi 
 
 S 
 
 M 
 
 o 
 
 
 
 U5 
 
 w 
 
 < 
 
 f1 
 
 of 
 
 
 B< 
 
 h 
 
 W 
 
 o 
 
 i4 
 
 W 
 
 n 
 
 « 
 
 !5 
 
 S 
 
 < 
 
 tJ 
 
 m 
 
 k; 
 
 111 
 
 W 
 
 o 
 
 H 
 
 H 
 
 
 < 
 
 is 
 
 rt 
 
 
 W 
 
 ffi 
 
 S^ 
 
 n 
 
 « 
 
 f5 
 
 w 
 
 5-1 
 W 
 
 B. 
 
 B( 
 
 O 
 
 rn 
 
 Oi 
 
 P 
 
 M 
 
 02 
 
 m 
 
 ^ 
 
 p> 
 
 M 
 
 t> 
 
 u 
 
 ^ 
 
 
 < 
 
 ELAPSED TIME 
 
 FROM CENSUS 
 
 DATE TO DATE 
 
 OF PUBLICATION 
 
 OF POPULATION 
 
 VOLUME. 
 
 § 
 
 oovD tnf>.o o^a\cf\ 
 
 ■^ 
 
 O O 
 
 
 
 i 
 
 It 
 > 
 
 KHMMl-iMMMtO 
 
 " 
 
 N rt- 
 
 " 
 
 
 1- . 
 
 is 
 
 It 
 
 N Tl-vq ^O -*0O rO\0 
 M M N N N ■^\0 VD 
 
 oooooooo 
 
 ■8 
 
 1^ 
 
 MCO 
 
 1 
 
 
 o o o o o o o o 
 
 O 
 
 O o 
 
 o 
 
 
 1 
 
 03 -rt t^ o\ m trun o\ 
 
 R 
 
 ^s 
 
 CT\ 
 
 
 COM -d-iO10tO(0t0 
 
 ? 
 
 00 t^ 
 \0 1- 
 
 CO 
 
 
 3«^lft^g^ 
 
 to 
 
 iOm" 
 
 1 
 
 
 3 
 110 
 
 Tf fO M to O fO^O 1-1 
 
 
 S« 
 
 1 
 
 
 OS ro c?\£'cO O Hi 5 
 
 % 
 
 
 ? 
 
 
 tOlOt^cf.« Kro- 
 
 M H N to 
 
 % 
 
 M 
 
 «. 
 
 
 Total 
 number 
 of pages 
 in pub- 
 lished 
 reports. 
 
 --3I&HI 
 
 
 P. 
 
 lOO' 
 
 8 
 
 CTN 
 
 o" 
 
 
 Great- 
 est 
 num- 
 ber of 
 clerks 
 
 em- 
 ployed. 
 
 O O Ow 
 
 00 
 
 i-Tfo 
 
 
 
 Time ac- 
 tually con- 
 sumed in 
 complete 
 enumera- 
 tion 
 (months). 
 
 O lOO O O lO 
 
 00 <6 6 tri'i- 6 
 
 O 
 
 o o 
 
 o 
 
 
 Legal 
 period for 
 enumera- 
 tion, ex- 
 cluding 
 exten- 
 sions 
 (months). 
 
 OiOMDMDVD lO-^in 
 
 
 
 
 
 Number of 
 assistant 
 marshals 
 or enu- 
 merators. 
 
 mo oco M\o toMio tooo o Tt t~- 
 
 VO ON M M UD M W ^^ lO COQO >0 00 
 
 
 '-'-' wMMCNrOtf 
 
 vo 
 
 
 
 Num- 
 ber of 
 mar- 
 shals or 
 super- 
 visors, 
 
 M M CO to to -a- -^VO 
 
 K 
 
 
 1 
 
 
 si 
 < 
 a 
 >• 
 m 
 P 
 m 
 
 W 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 OOOOOOOO 
 Kco (M (S O?00 Oo'ot 
 
 
 II- 
 
 i, 
 
 
 O cd 
 
 
 iH <U 
 
 a PI 
 
 as 
 
 • rH O 
 
■Si 
 
 M 
 o 
 
 H 
 
 M 
 
 w 
 w 
 
 tl. 
 o 
 
 w 
 
 P 
 
 a 
 n 
 H 
 o 
 «J 
 
 i4 
 «! 
 
 M 
 
 (U 
 
 w 
 w 
 
 o 
 w 
 
 o 
 
 tn 
 
16 
 
 Early census 
 schedules. 
 
 Marginal 
 comment by 
 enumerators. 
 
 South Carolina, which was printed later, was "tipped in." To 
 contain the results of the Twelfth Census, ten large quarto 
 volumes, comprising in all 10,400 pages, are required. No 
 other illustration of the expansion of census inquiry can be more 
 striking than this. 
 
 The original schedules of the First Census are now contained 
 in 26 bound volumes, preserved in the Census Ofiice. For the 
 most part the headings of the schedules were written in by 
 hand. Indeed, up to and including 1820, the assistant marshals 
 generally used such paper as thej' happened to have, ruling it, 
 writing in the headings, and binding the sheets together them- 
 selves. In some cases merchants' account paper was used, and 
 now and then the schedules were bound inside of a newspaper. 
 
 In consequence of the informal method resulting from requir- 
 ing marshals to supply their own blanks, the 193 volumes con- 
 taining the schedules from 1790 to 1820, inclusive, var}' in size 
 from about 7 inches long, 3 inches wide, and % inch thick, to 
 21 inches long, 14 inches wide, and 6 inches thick. Some of the 
 sheets in these volumes are only 4 inches long, but a few are as 
 much as 3 feet in length, necessitating several folds. In some 
 cases leaves burned at the edges have been covered with trans- 
 parent silk to make them easy to turn. 
 
 In the early days of census taking the hard-worked and under- 
 paid enumerator appears to have felt free to express his feelings 
 occasionally on the margins of the schedules which the Federal 
 Government expected him to supplj'. At the distance of a cen- 
 tury, some of these faded documents are very interesting. The 
 returns for Delaware for 1800 are typical. On the left-hand mar- 
 gins of the sheets are the names of the places visited. The names 
 of the heads of families are given in the second column, and other 
 information, by numbers, in columns to the right. The following 
 entries illustrate the problem of uncertain geographic boundaries, 
 so serious in the early censuses, and still somewhat in evidence 
 in 1900: 
 
 Kent County. 
 
 Duck Creek Hundred. 
 Forrest and Grog Town. 
 
 Duck Creek Neck, embracing pcill's Blacksmith Shop. 
 Those persons between ye road leading from Worrell's Bridge to Old 
 Duck Creek, from thence to Grog Town and from thence down the 
 stream to the beginning. 
 The alley. 
 
 Near Jamestown (a village of no acct.). 
 Duck Creek Hundred Forrest, near Lewis' X Roads. 
 Binding on the County Road below Frederica. 
 Bottom of ye Neck. 
 
17 
 
 Kent County — Continued. 
 
 Between the Neck and Milford. 
 
 New Whark — so called — on Mispillion Creek. 
 
 A small Village called CuUen Town. 
 
 Mashahope Broad Arm — so called. 
 
 Near ye Maryland line, also a small village called Whitelesburgh, a place 
 
 of no note. 
 Canterbury- commonly called Irish Hill. 
 Binding on the Forrest of Murderkill Hundred. 
 Binding on Murderkill Neck in ye neighborhood of Smack's Hill. 
 Woodley Town, a small village containing g dwellings & 43 inhabitants, 
 
 bdng'on a small prong of St. Jones's Creek, about g miles from Dover. 
 Binding on the Forrest and Stone I,ine near Cow Marsh. 
 Georgetown, a small village in the Forrest of Cow Marsh. 
 Tappahannah, it contains 8 houses and 36 inhabitants. 
 
 Miscellaneous information, such as the names of persons tem- 
 porarily residing in the district at the time the census was taken, 
 was occasionally written on the schedules. In 1800 some of the 
 assistant marshals in Massachusetts added on the right-hand 
 margin of their schedules statements concerning the number of 
 "people of color housekeeping by them.selves " ; others estimated 
 the value of houses occupied by persons enumerated, and one 
 closed his report by giving some figures to show the total number 
 of "soles" in the three towns enumerated by him. 
 
 For all the early censuses the six questions asked at the First Census 
 Census formed the basis of the inquiry. The Second Census inquiry wnder 
 extended the age questions; the Third attempted a record of manu- '^°^^[/lg„^ 
 facturing establishments; the Fourth enumerated colored persons, 
 by sex and age, and foreigners not naturalized, gave statistics of 
 occupations, and again attempted a record, slightly more elabo- 
 rate, of manufacturing interests. The Fifth Census is memorable 
 for the adoption of printed schedules. That census, and the Sixth 
 as well, further broadened the range of census inquiry. The 
 Seventh Census— authorized by the passage of a law drawn with 
 great deliberation— marked the beginning of scientific census 
 inquiry in the United States. In consequence the census of 1850 
 is known as the first "modern census" of the United States. 
 
 By 1870 census inquiry had become so extended, and the increase Mechanical 
 in population and material development was so enormous, that assistance 
 tabulation by hand was necessarily inaccurate and extremely ^" '^7°- 
 expensive. It was clear, moreover, that a point must be reached, 
 before many more decades had passed, where complete tabulation 
 within the census period would be actually impossible. General 
 Walker, Superintendent of the Ninth Census, casting about for 
 mechanical assistance, found some relief in the Seaton machine, 
 a simple affair composed of parallel rollers by which blanks were 
 unwound from a roll, and six columns— one from each of six 
 different blanks or tables— were brought into close proximity, 
 23018 — 04 -^ 
 
18 
 
 thus effecting a great saving in the time of the clerk. At the 
 Tenth Census the Seaton machine was extensively used and thor- 
 oughly tested. Like hand tabulation, however, this device became 
 inadequate, as the mass of detail rapidly increased. 
 
 THB SEATON DEVICE FOR TABUI<ATION, MODEI, OF 1880. 
 
 This machine (the invention of C. W. Seaton) was first utilized in 1872 and 
 extensively used in tabulating the census of 1880. It was constructed of 
 ■wood, and contained a roll of paper which unwound from one of the large 
 rollers near the bottom, back and forth over two series of small rollers, one 
 series located at the top of the machine and the other near the bottom, and 
 finally wound on the second large roller at the bottom. The object of the 
 machine was to condense a lengthy tally sheet, so as to present a small sur- 
 face, upon which could be indicated rtems relating to seven different classes 
 of facts. By a partial turn of the winding roller the blanks advanced and 
 exposed on each small roller at the top another series of columns. 
 
 Change 
 of Census 
 fnethods in 
 1880. 
 
 The census of 1880 is notable for the radical change effected in 
 the method of enumeration. From the First to the Ninth Cen- 
 sus, inclusive, this task had been performed by the marshals, 
 though it was early evident that such an arrangement was highly 
 unsatisfactory. By the law providing for the Tenth Census there 
 was created a new body of census officials, known as supervisors, 
 of whom there were to be one or more for each state. The super- 
 visor's district was divided into enumeration districts, each of 
 which was assigned to an enumerator. The change to supervisors 
 
#»- 
 
 w 
 
 o 
 
 2; 
 p> 
 di 
 
 a 
 
 CrS 
 <! 
 O 
 
 n 
 >< 
 » 
 
20 
 
 and enumerators was so beneficial that this provision of the law 
 of iS8o has become an established census practice. 
 Electric As the Eleventh Census approached, it was clear that the scope 
 
 tabulation of inquiry m.ust be curtailed, or some form of very rapid mechan- 
 " ical tabulation must be supplied. This requirement was effect- 
 
 ively met; the census of 1S90 is notable for the introduction of 
 electric tabulation, the principle of which is apparently capable 
 of responding to all the demands for many 3'ears to come. 
 
 By this system, invented by Dr. Herman Hollerith, of Wash- 
 ington, D. C, the facts entered upon the various schedules are 
 copied upon cards by punching holes. The position of each hole 
 indicates its significance; no writing is required. 
 
 The cards, when punched, are tabulated by an ingenious 
 machine provided with a pin box which contains a needle, set on 
 a fine spiral spring, for each possible hole in a card. The machine 
 used in 1890 and 1900 was operated by hand; the pin box was 
 brought down over each card in turn; those needles which met 
 an unpunched surface were repressed, while those which passed 
 through a hole made an electric contact below, and by an arrange- 
 ment of relays, permitting any desired combination, caused one 
 or more counters, or dials, to register. At the conclusion of each 
 ' ' run ' ' the counters were read and the results recorded. 
 
 "With the aid of one of these machines an experienced and 
 
 capable clerk could tabulate on an average from 8,000 to 10,000 
 
 cards in a working day of six and a half hours. The importance 
 
 of this system laj- in its ability to count combined facts. 
 
 The Twelfth The law providing for the Twelfth Census was a notable one, 
 
 Census. drawn with great care. It recognized the fact that the question 
 
 of tabulation, which for several decades had been the leading 
 
 problem in census methods, had now given place to the general 
 
 question of the organization and equipment of a vast establishment 
 
 in a limited time. The maximum force required had grown from 
 
 about 650 employees in 1790 to 59,373 in 1900, more than the 
 
 entire population in 1790 of New York and Boston combined. 
 
 Perfected At the Twelfth Census, electric tabulating machines of the same 
 
 electric _ model as those of 1890 were extensively used, and special tabu- 
 
 ~ " " lating machines with adding machine attachments were employed 
 
 in preparing the statistics of agriculture, in connection with which 
 
 automatic electric sorters were also used to great advantage. 
 
 Toward the close of the Twelfth Census work the tabulating 
 machuie was perfected by the addition of automatic feeders, and 
 an automatic sorting machine was also invented. These were 
 used upon the final work of the Twelfth Census, and adopted for 
 
 tabulation. 
 

 
 27 87 
 
 
 
 
 \ 
 
 \ 
 
 §5 IS' 
 
 1^ 
 
 fe;H 
 
 So. 
 
 &^|g |s :| 
 
 Is 
 
 1 
 
 Is 
 
 
 
 !z; 
 
 ^^13 1§ 
 
 is 
 
 M 
 
 o 
 fa 
 
 N 
 
 IS l§ |5 
 
 • 
 
 & 
 
 ^ 
 
 S 
 
 
 sg 
 
 §5 33 £53 
 
 t3 
 
 M 
 
 g 
 
 S 
 
 fi 
 
 s, 
 
 
 
 ^w ^*-' o« 
 
 ;^Ui Ou 5° 
 
 Uu <^FH p^n 
 
 g 
 
 s 
 
 
 
 g 
 
 o 
 
 • 
 
 EQ 
 
 d 
 
 ^ 
 
 & 
 
 4> 
 l-t 
 
 !§ |5 
 
 a:5 
 
 SI 
 
 35 §2 
 
 
 
 5 
 
 1 
 
 l4 
 fa 
 
 o 
 
 fa 
 
 o 
 
 
 » 
 
 •i 
 
 C9 
 
 °* S S 
 
 <0 ^ lO 
 
 o 
 
 OI 
 
 a 
 
 t3 
 
 CO 
 
 o 
 
 ^ 
 
 kn 
 
 » 
 
 • 
 
 S 3 
 
 O 1-1 
 
 CO 
 
 3 S S 
 
 " s s 
 
 DS ^ us 
 
 • 
 
 fa 
 
 H 
 
 g 
 
 o 
 
 r- 
 
 eo 
 
 GO 
 
 •tft 
 
 • 
 
 o 
 I-l 
 
 CO 
 
 
 g ». 
 
 i 
 
 • % « 
 
 • -» 
 
 wt- 
 
 toco ^o» 
 
 B 
 
 • 
 
 rH 
 
 »I 
 
 CO 
 
 o S 
 
 8 
 
 g ^ g 
 
 S S g 
 
 itfi 
 
 lO 
 
 ts 
 
 t- 
 
 CO 
 
 o 
 
 
 ,a 
 
 " 
 
 r3 
 
 « 
 
 
 
 -- 
 
 ■* 
 
 r- 
 
 o 
 
 CI 
 
 t-l 
 
 • ^ 
 
 s s 
 
 o 
 
 CO 
 
 
 g 
 
 M 
 
 o 
 
 • » 
 
 g 
 
 
 s 
 
 
 s 
 
 C9 
 
 + 
 
 00 
 
 1^ « 
 
 <* 
 
 CO 
 
 ^ CO 
 
 
 CO 
 
 ■«*• 
 
 • 
 
 ■* 
 
 00 
 
 CO r- 
 
 M 
 
 r- 
 
 CO t- 
 
 CO 
 
 1- 
 
 eo 
 
 t- 
 
 CO 
 
 I- 
 
 • ° 
 
 (M 
 
 o 
 
 ©a a 
 
 ca 
 
 o 
 
 N 
 
 eo 
 
 • 
 
 CO 
 
 -. .» 
 
 • 
 
 lO 
 
 • -o 
 
 t-t 
 
 in 
 
 iH 
 
 U3 
 
 1-H 
 
 w 
 
 
 
 COPYRIGHT, 1B99, BY H.H 
 
 □LLER 
 
 TH. 
 
 
 
 
 u , -r O e ft '1' 
 •r* bo H ^'t d 
 
 •ScdKW>.a 
 
 H ° b W 
 
 StJ ." »i O J J 
 
 fl "= ^ 
 
 ^^ , r Si ■,■2'-' a^ 
 
 O ..Oril-'iur': 
 
 iftA 
 
 U] 
 
 1 1'iiS" 
 
 -lU'C-lH 
 
 OJ d t 
 .-< [fl It oj r? 
 
 ?ai IBIS'S 
 
 uj=ii ■« B " ° 
 u g-it: S s *" ■!» 
 
 (20 
 
22 
 
 routine tabulation and the various requirements of the permanent 
 office, including the tabulation of the Philippine census. 
 
 The principle upon which the automatic tabulator and the auto- 
 matic sorter are operated is the same as that of the hand tabulating 
 machine already referred to. The automatic portion of these 
 machines is about 18 inches square. Punched cards are placed in 
 the card receiver in lots of 400 or 500, and are pressed forward by 
 a pressure plate against a knife edge, which pushes downward 
 one card at a time. As each descends it pauses before the pin 
 box, which in this :nodel is placed vertically instead of horizon- 
 tally. As in the case of the hand machine, where the pins encoun- 
 ter the card, they are pressed back into the pin box, but where 
 there are holes the pins pass through and establish electric 
 connection. 
 
 In the electric tabulator, after contact has been made, the card 
 is fed downward, and if it has been tabulated a deflector operates 
 a deflector magnet, to allow the card to pass into a receptacle for 
 tabulated cards; but if it has not been tabulated the card falls into 
 what is known as the reject box, for further examination. 
 
 The adjustment of the automatic tabulating machine is such 
 that if a card is defective or out of position, or if the last card has 
 been fed through, the machine will stop , automatically. This 
 machine has a record of tabulating 415 cards a minute, but the 
 average speed in the work of the Census Office is about 350 cards 
 a minute. This is an average of 80,000 or 90,000 cards in a work- 
 ing day of seven hours, including stoppages for reading of dials, etc. 
 Speed is easily regulated by a rheostat. A small stand is sufficient 
 to hold the automatic tabulator. It is connected with the case of 
 dials upon an adjoining table by wires transmitting the electric 
 current. The dials may number from i to 70, and thus inay record 
 simultaneously from i to 70 primary facts or combinations pro- 
 ducing an immense variety of facts. 
 
 The automatic sorter stands upon a frame about 3 feet square 
 and 5 feet high, directly over 10 boxes arranged one above 
 another. As in the tabulating machine, the cards are fed down- 
 ward one by one, before the pin box. The circuit established 
 through holes in the cards by means of electric magnets controls 
 10 chutes of varying lengths which lead to the 10 boxes. As each 
 card is pushed downward it enters the chute which has been 
 determined by the electro-magnet as the proper one to receive it, 
 and is assisted by fingers into its proper receptacle. In this way 
 cards which have been placed in the feeder pass through the 
 
w 
 g 
 
 3 
 
 o 
 < 
 
 o 
 g 
 
 <! 
 
 ►J 
 O 
 
 <) 
 
 o 
 
 <! 
 W 
 W 
 
24 
 
 other 
 mechanical 
 devices 
 tdilized. 
 
 Magnitude 
 of the Twelfth 
 Census and of 
 its results. 
 
 machine and are deposited in one of the lo boxes, according to 
 the combinations which have been worked out previously by the 
 adjustment of the pins in the pin box. Thus, for example, if it 
 were desired to sort a large number of cards according to single 
 years of age, it would be necessary first to pass the cards through 
 the machine in order to separate them by the tens column; they 
 would then be passed through again, to be sorted by the units 
 column, and the passage of the cards twice through the machine 
 would bring together all cards for each separate year of age. The 
 separation of cards into not more than lo classes requires but one 
 run through this machine; any number of classes exceeding lo 
 requires two or more runs. If a card is imperfect, or is improp- 
 erly presented to the pin box, the machine does not advance it, 
 but passes it into what is called the reject box. The automatic 
 sorting machine can be operated at a speed of from 75,000 to 
 175,000 cards per day; the record of actual operation of this 
 machine in the Census Office for a seven-hour day was the sort- 
 ing of 162,000 cards into 10 classifications — the equivalent of 
 1,620,000 individual classifications. Dials upon the machine 
 record the number of cards which pass through it, and it is thus 
 possible to read at any moment the product up to that time, and 
 to keep a record of the total and average product per day. 
 
 Electric tabulation has proved a boon to the Census Office. It 
 need only be said that if, at the Twelfth Census, the three tallies of 
 age and sex, nativity, and occupation had been made by hand, it 
 has been estimated that the tabulation of the statistics for these 
 three subjects alone would have required the time of a hundred 
 clerks for 7 years, 11 months, and 5 days. 
 
 In the Twelfth Census, machines were first employed, whenever 
 possible, for the tabulation of manufactures schedules. These 
 machines were large typewriters, operated for the most part by 
 men. Mechanical calculating contrivances also were employed, 
 greatly facilitating the handling of the tabulated returns. Huge 
 columns of figures were totaled, long percentages computed, and 
 even multiplications were performed with amazing rapidity and 
 with perfect accuracy. In calculating the percentages required 
 in the published tables of the Twelfth Census, which numbered 
 in the aggregate more than one million, the slide rule was found 
 to be a mechanical device of very great assistance. 
 
 It is difficult for one not connected with such work to understand 
 how vast an undertaking is a modem American census. Perhaps 
 the following figures may suggest .some idea of the extent of detail 
 handled by the Census Office at the census of 1900: 
 
25 
 
 EMPLOYEES OF THE TWELFTH CENSUS. 
 
 Supervisors 300 
 
 Special agents in the field 2, 648 
 
 Enumerators 52, 871 
 
 Clerical force (maximum) 3, 554 
 
 Total 59> 373 
 
 SCHEDULES RECEIVED. 
 
 Population 900, 000 
 
 Agriculture 5, 738, 524 
 
 Manufactures 608, 401 
 
 Vital statistics 50, 000 
 
 Total 7. 296. 925 
 
 CARDS PUNCHED. 
 
 Population: 
 
 Individual 7S, 3°3i 387 
 
 Family 16, 239, 797 
 
 Agriculture : 
 
 Farm 5. 739. ^57 
 
 Crop 116,571,239 
 
 Vital statistics i, 039, 094 
 
 Total 215. 893, 174 
 
 NUMBER OF TIMES CARDS PASSED THROUGH TABULATING MACHINES. 
 
 Population 375. 4io, 161 
 
 Agriculture 237. 929. 2S9 
 
 Vital statistics 6, 234. 564- 
 
 Total 619, 574, 014 
 
 It may be that these figures, which possess such weighty sig- 
 nificance in the Census Office, will not impress the reader so 
 deeply as some of the statistical facts which the Twelfth Census 
 recorded. Since 1790 the area of the United States has increased 
 from 827,844 to 3,622,933 square miles; the number of counties 
 has increased from 307 to 2,867; and the total population has 
 increased from 3,929,214 to 76,303,387, or nineteenfold. There 
 are four states each possessing, in 1900, a population greater 
 than that of the entire nation in 1790, at which time two of the 
 four were an untrodden wilderness. The number of cities with a 
 population of 8,000 or over has increased from 6 to 546, and the 
 number with a population of 25,000 or over from 2 to 161. There 
 are now 38 cities having a population exceeding 100,000, and 3 
 of these have over 1,000,000 each. In igoo the record of capital, 
 wages, and value of products of manufactures rose to figures 
 almost beyond comprehension. The capital invested was f 9,846,- 
 628,564; the salaries and wages paid amounted to 12,735,430,848; 
 and the value of products was $13,039,279,566. In agriculture the 
 figures are almost equally impressive. The total value of farms 
 
task. 
 
 26 
 
 in 1900 was J16, 674,690,247, and that of agricultural products in 
 1899, ^4,739,118, 752. To gather and collate such stupendous 
 figures, not only with accuracy, but so swiftly that the record of 
 population in 1900 appeared as quickly as did the little report of 
 the First Census, was a task of the first magnitude. 
 
 Such is the modern census of the United States. It is a decen- 
 nial snap shot of the nation for the benefit of all time. Patrick 
 Henry declared that there was but one lamp by which his feet 
 were gidded — experience. But so important has the study of 
 facts become that statistics presenting the facts analyzed and 
 classified is the lamp which guides the statesman and the student 
 of to-day. 
 The perma- By the act of March 6, 1902, the Census Office was made a per- 
 
 nent Census manent bureau of the Government. This act became effective 
 Office audits j^^^ ^^ j^^^, fey which date the main reports of the Twelfth 
 Census had been completed. The secondary inquiries specified 
 by Congress were then commenced. Among those now in pro- 
 gress are: 
 
 Mines and Quarries. 
 Street Railways. 
 
 Electric Light and Power Stations. 
 Telegraphs and Telephones. 
 Deaf, Dumb, and Blind. 
 Statistics of Births and Deaths. 
 Wealth, Debt, and Taxation. 
 Social Statistics of Cities. 
 Crime, Pauperism, and Benevolence. 
 To these should be added the collection and compilation of 
 annual statistics of cotton ginning, and the tabulation of annual 
 statistics of immigration. 
 
 B}' direction of the President the tabulation of the census of the 
 Philippine Islands was undertaken by the Census Office, begin- 
 ning in January, 1904. 
 
 Toward the close of 1904 preparations will begin for taking and 
 tabulating the census of manufactures, which Congress directed 
 to be taken in 1905. 
 
 Perhaps the most important and difficult task, however, at 
 present before the Census Office is to secure uniformity in the 
 extensive statistical work of the various state and municipal 
 bureaus, and to devise a system of cooperation with the General 
 Government. 
 
 Cooperation of the Census Office with state bureaus has lately 
 received a practical illustration in the act of Congress enabling 
 
THE AUTOMATIC SORTING MACHINE. 
 
28 
 
 the Census OflBce to unite with the Census Bureau of the state of 
 Michigan in taking a census of manufactures. As the constitu- 
 tion of Michigan requires a census in the year 1904, including 
 returns for manufactures, a Federal census of manufactures in 
 1905 in that state would be an unnecessary expense and duplica- 
 tion of labor; hence Congress has authorized the Director of the 
 Census to share with the state of Michigan the expense of secur- 
 ing the returns relating to manufactures. 
 Publications During the progress of the Twelfth Census 250 bulletins, aggre- 
 
 of the Census gating 4,046,250 copies, were published to present advance infor- 
 mation concerning population, mortality statistics, agriculture, 
 and manufactures. 
 
 The main reports of the Twelfth Census are comprised in 10 
 quarto volumes, aggregating 10,470 pages. 
 
 Since the establishment of the permanent Census Office, there 
 have been published 7 bulletins relating to inquiries in progress, 
 a complete statistical atlas of the United States, and two large 
 volumes, one containing statistics of employees and wages, and the 
 other presenting detailed statistics of occupations at the Twelfth 
 Census. 
 
 Office. 
 
 PLANS FOR FUTURE CENSUS WORK' 
 
 Advantages 
 resulting 
 from the 
 permanent 
 Bureau. 
 
 BY HON. GEORGE B. CORTELYOU 
 
 All the advantages anticipated from placing the Census Bureau 
 on a permanent basis, including more thorough preparation for 
 the Federal decennial censuses, their earlier compilation and 
 publication, their greater comparability with each other, their 
 greater accuracy and continuity are destined to be realized. But 
 more than this, the establishment of a permanent Office has opened 
 up unexpected opportunities for beginning much-needed statis- 
 tical reforms all along the line, which are certain to result not 
 only in the avoidance of duplication in Federal statistical work, 
 but also in bringing into closer harmony with Federal' statistics a 
 great body of statistical material collected in different states and 
 municipalities. 
 
 I From the first annual report of the Secretary of Commerce and I<abor. 
 
29 
 
 Much of this material relates to subjects closely akin to those 
 covered by the census reports, bvit it has heretofore been compiled 
 on lines so diversified and so dissimilar to those of the census that 
 it has been impossible to coordinate the results. Undoubtedly it 
 will take many years fully to accomplisTi coordination and collabo- 
 ration between the Federal census and the many state bureaus 
 engaged in similar statistical work; but the Director's report shows 
 that substantial progress has already been made in several direc- 
 tions; and I can not resist the conviction that as this movement 
 for the standardization of all classes of official statistics proceeds 
 and develops it will come to be recognized as marking one of the 
 greatest and most practical reforms in official statistics. 
 
 The piurely scientific work of the Census Bureau is closely related 
 at many points to the practical affairs of the nation. A striking 
 illustration of this fact is found in the current investigation of the 
 bureau concerning the receipts and expenditures of cities. The 
 schedule prepared for collecting data relating to this subject has 
 become the pivot around which is now centering a well-organized 
 movement for securing a uniform classification of municipal 
 accounts and a more intelligent presentation of them. A con- 
 ference for the critical stud}' and perfection of the schedule of 
 municipal receipts and expenditures, recently called by the 
 Director of the Census, brought together a large gathering of the 
 representatives of the oflBces charged with the administration of 
 finances in our greater cities, and of others interested in the more 
 general adoption of improved methods of municipal accounting. 
 
 Still another great advantage arising from the establishment of Proposed 
 the permanent Census Bureau, and from its transfer to a depart- transfer of 
 ment containing other bureaus engaged more or less in statistical ^„^u 
 work, has impressed itself upon my mind from the first. The 
 Census Bureau is a purely statistical office, employing a body of 
 experts whose main business it is to study statistics and statistical 
 methods, with a view to their improvement and perfection. To 
 this work they give their undivided attention, and it is reasonable 
 to believe that a steady improvement in the character of official 
 statistics will result from the concentration in such an office of as 
 much of the statistical work of the department, no matter what 
 its immediate character, as can be centered there without inter- 
 ference with the administrative duties of the other bureaus. 
 
 In accordance with these views, and acting under the authority 
 conferred by the organic act of the department, I have already 
 transferred from the Bureau of Labor to the Bureau of the Census 
 
30 
 
 the compilation of the annual statistics of cities of 30,000 popula- 
 tion and over, provided for by the act of Congress approved July 
 I, 1898, and from the Bureau of Immigration the compilation of 
 the statistics of immigration. 
 
 The transfer of the compilation of the immigration statistics to 
 the Census Bureau has long been advocated by statisticians, and 
 will bring immigration statistics into proper harmony with the 
 population statistics of the Federal census. It will enable the 
 Department to present them with the fuller detail and analysis 
 which have become imperative, in view of the rapid increase in 
 immigration and its changing character. 
 
 I desire here to call attention to the recommendation made by 
 the Commissioner-General of Immigration and the Director of the 
 Census that provision at once be made by law for securing the 
 proper statistics of foreign born emigrants from this country. No 
 satisfactory statistical statement of immigration and its permanent 
 effect upon the population of the country can be compiled unless 
 the statistics of immigration are accompanied by the correspond- 
 ing statistics of foreign born emigration. 
 
 It is my intention, from time to time, as opportunity presents 
 itself, to transfer other important' branches of statistical work to 
 the Bureau of the Census. As an illustration of how widely use- 
 ful such a recognized center of statistical work can become, and 
 how it may be possible and desirable for all of the departments of 
 the Government to utilize its services in their own statistical com- 
 pilations, I may refer to the recent request of the Civil Service 
 Commission, which I have approved, that the Bureau of the Cen- 
 sus shall compile the statistics of the classified service of the United 
 States, collected under the order of the President dated March 31, 
 1903. 
 A great Heretofore there has been no bureau of the Government to which 
 
 statistical sasih distinctive and exclusive functions could be assigned, and the 
 
 ora ory. quality and the value of our governmental statistics have suffered 
 correspondingly in comparison with those of other nations. But 
 with this conception of the true function and the proper develop- 
 ment of a permanent Census Bureau it will necessarily become, in 
 a comparatively short time and under proper direction and man- 
 agement, the great statistical laboratory of the United States Gov- 
 ernment, worthy to take rank with the best statistical ofiices 
 maintained by European governments. 
 
 The success of this general plan for the development of a Fed- 
 eral laboratory of statistics depends upon its organization upon a 
 
31 
 
 strictly nonpartisan basis, such as will command universal confi- 
 dence. So organized and carried on, the Bureau of the Census 
 ■will very quickly come to be recognized as belonging strictly in 
 the category of the scientific bureaus of the Government, and as 
 one of the most useful and important of them. 
 
SUMMARY or CENSUS PUBLICATIONS 
 
 Pages. Pages. 
 
 1790. I vol 56 . 1850. I vol 1 158 
 
 1800. I vol 74 i860. I vol 310 
 
 1810. I vol 180 1870. I vol 854 
 
 1820. I vol 160 1880. I vol 1054 
 
 1830. I vol 163 1890. 2 vols 2332 
 
 1840. I vol 480 igoo. 2 vols 2212 
 
 MANUFACTURES. 
 
 Pages. Pages. 
 
 1810. I vol 233 1870. I vol." 850 
 
 1820. T vol 128 1880. I vol 1248 
 
 1850. I vol 143 1890. 3 vols 2614 
 
 i860. I vol 963 igoo. 4 vols 4286 
 
 AGRICUI,TUR13. 
 Pages. I Pages. 
 
 i860. I vol 464 1890. I vol 950 
 
 1880. I vol 1 149 1900. 2 vols I93X 
 
 VITAI< STATISTICS. 
 
 Pages. Pages. 
 
 1850. I vol 304 1880. 2 vols 1794 
 
 i860. I vol.^ 651 1890. 4 vols 4368 
 
 1870. I vol 702 1900. 2 vols 2063 
 
 STATISTICAI, ATI^ASES. 
 
 1870. I volume. Diagrams, 39; maps, 66. 
 
 1880. I volume. 3 Diagrams, 79; maps, 13; diagram-maps, 249. 
 
 1890. I volume. Diagrams, 227; maps, 180. 
 
 1900. I volume. Diagrams, 133; maps, 74. 
 
 SPECIAIv REPORTS. 
 
 Sixth Census: 1840. 
 
 Pages. 
 Census of Pensioners for Revolutionary or Military Services . 195 
 
 I Includes industry and wealth, 
 z Includes miscellaneous statistics. 
 3 Published by Charles Scribner's Sons. 
 (32) 
 
33 
 
 Tenth Census: 1880. 
 
 Pages. 
 
 Cotton Production, 2 volumes 1772 
 
 Defective, Dependent, and Delinquent Classes of the Popula- 
 tion 638 
 
 Forests of North America, exclusive of Mexico 622 
 
 Mining Industries (exclusive of precious metals), with spe- 
 cial investigations into the iron resources of the Republic 
 
 and the Cretaceous coals of the Northwest 1064 
 
 Mining Laws and Regulations, United States, State, and 
 
 Territorial 716 
 
 Newspapers, etc., Alaska, Fur-Seal Island, and Shipbuilding 1126 
 
 Petroleum, Coke, and Building Stones 876 
 
 Power and Machinery 666 
 
 Social Statistics of Cities, 2 volumes 1774 
 
 Statistics and Technology of Precious Metals 558 
 
 Statistics of Wages in Manufacturing Industries 796 
 
 Transportation, Agencies of 870 
 
 Valuation, Taxation, and Public Indebtedness 920 
 
 Water Power, 2 volumes 1662 
 
 Eleventh Census: 1890. 
 
 Crime, Pauperism, and Benevolence, 2 volumes 1468 
 
 Education in the United States 149 
 
 Farms and Homes, Proprietorship and Indebtedness 65S 
 
 Indians in the United States, except Alaska 692 
 
 Insane, Feeble-minded, Deaf and Dumb, and Blind in the 
 
 United States 767 
 
 Insurance Business in the United States, 2 volumes 162S 
 
 Mineral Industries 874 
 
 Population and Resources of Alaska 294 
 
 Real Estate and Mortgages in the United States 956 
 
 Social Statistics of Cities of United States 146 
 
 Statistics of Churches 840 
 
 Transportation Business, 2 volumes 1422 
 
 Vital Statistics of Boston and Philadelphia (covering 6 years) 278 
 Vital Statistics of New York City and Brooklyn (covering 6 
 
 years) 53^ 
 
 Vital Statistics of the District of Columbia and Baltimore 
 
 City (covering 6 years) 250 
 
 Wealth, Debt, Taxation, 2 volumes 1564 
 
 23018 — 04 3 
 
34 
 
 Twelfth Census: 1900. 
 
 Pages. 
 
 Abstract of the Twelfth Census 408 
 
 Employees and Wages 1344 
 
 Special report on Occupations 1000 
 
 Electric Light and Power Stations (in press) 
 
 Mines and Quarries (in press) .... 
 
 Street Railways (in press) 
 
 Crime, Pauperism, and Benevolence (inquiry in progress) 
 
 Deaf, Dumb, and Blind (inquiry in progress) . • 
 
 Social Statistics of Cities (inquiry in progress) 
 
 Statistics of Births and Deaths (inquiry in progress) 
 
 Telegraphs and Telephones (inquiry in progress) 
 
 'S^ 
 
.f^K^' 
 
 
 
 
 '^ 
 
 V 
 
 
 l*?er' 
 
 "1^ 
 
 X 
 
 .-H^ A. 
 
 
 
 ^ 
 
 
 J:- 
 
 ;•i^S^, 
 
 
 
 
 
 ■ -■■ife'::"'''i'^'.-.'.-/'<jf- 
 
 'ii--*^— 
 
 
 
 ^k 
 
 
 
 
 
 
 ^'i 
 
 ■' -'/-v ^--^J ^r^- 
 
 
 
 i^' 
 
 -ciS ? 
 
 
 
 > 
 
 1 ■^*¥- n..^^^. 
 
 -v" 
 
 )> 
 
 
 ,>-^ 
 
 
 M .. ^* 
 
 •* ^ 
 
 
 
 
 -*-Vi;^ ^ftl 
 
 ■;^ 
 
 1^- 
 
 'A,^ 
 
 
 tS.*<fe 
 
 ,^._f7,-' 
 
 
 ■X'tV^'M/f^X 
 
 i»i.