m .Q-2 Issued March 21, 1908. U. S. DEPARTMENT OF AGRICULTURE, BUREAU OF SOILS— BULLETIN No. 48. MIL.TO>f WHITNEY, Chief. FERTILITY OF SOILS AS AFFECTED BY MANURES. BY FRANK D. GARDNER. WASHINGTON: GOVERNMENT PRINTING OF*^ICE. 19 8. Glass. ^Il5l Rnnk Q P 2. i Issued March 21, 190S. U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF SOILS— BULLETIN No. 48. MILTON WHITNEY, Chief. FERTILITY OF SOILS AS AFFECTED BY MANURES. BY FRANK D. GARDNER. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1908. CL-r-^-"-^ 7^ V APR 1 19C8 A Ota ^ x: LETTER OF TRANSMITTAL U. S. Department of Agriculture, Bureau of Soils, Washington, D. C, June 21^, 1907. Sir: I have the honor to transmit a manuscript entitled FertiHty of Soils as Affected by Manures, by Frank D. Gardner, in charge of Soil Management. It embodies the results of a uniform scheme of manurial treatments, as measured by the resulting increase in plant growth on soils collected from 220 fields in 23 States, located in the eastern half of the United States. The increasing use of manures, and especially of commercial fertilizers, by the farmers of the United States is sufficient justifica- tion for careful investigations as to the relative efficiency of various forms of manures on different soils. You will find here a vast amount of data, treated in a manner commensurate with the importance of the subject. It should be of great value to all who are interested in the use of manures. The manuscript has been gone over carefully with Assistant Secre- tary Hays, who concurs in my recommendation for its publication as Bulletin No. 48 of the Biu-eau of Soils. Respectfully, Milton Whitney, Chief of Bureau. Hon. James Wilson, Secretary of Agriculture. CONTENTS. Page. Introduction 7 Object of the investigations 9 The soils 10 The method , 10 Fertilizers used 13 Calculation of results 13 Arrangement and tabulation of results 15 Comparative efficiency of salts used separately and in combination 24 Nitrate of soda alone and with other salts 31 Sulphate of potash alone and with other salts 32 Acid phosphate alone and with other salts 32 Lime alone and with fertilizer salts 33 Relative efficiency of organic and chemical manures 37 Relative efficiency of all salts and combinations, when soils are grouped accord- ing to efficiency of PKN 38 Relative efficiency of salts when used alone and in combination .- 39 Relation of fertilizer requirements to character of soils 44 Relative efficiency of fertilizers by locality 46 Comparative fertility of soils 47 Relative response to fertilizers and expenditures for fertilizers 48 Average efficiency of fertilizer salts, by locality 50 Comparative efficiency of fertilizers, by soil series 51 Efficiency of fertilizer as related to soil texture 53 Natural fertility as related to responsiveness to fertilizers 53 Fertilizers and economy of water in plant growth 54 Summary 56 ILLUSTRATIONS. Fig. 1. Percentage increase in growth of plants, attributable to various ferti- lizer treatments of five principal soil types of the Waycross area, Georgia = = = .. 44 2. Percentage increase in growth of plants, attributable to various ferti- lizer treatments for all soils of Waycross area, Georgia, and Escambia County, Fla 45 3. Percentage gain in plant growth attributable to each of the salts P, K, X, and L when used alone 59 4. Relative efficiency of fertilizer ingredients by soil series, when the ingre- dients are used alone 52 5. Percentage increase in growth of plants attributable to fertilizer and accompanying increase per unit of water transpired. Average of 20 soils 55 FERTILITY OF SOILS AS AFFECTED BY MANURES." INTRODUCTION. The use of manures as a means of increasing the growth of crops dates back to ancient times. At the time of the discovery of America, the North American Indians used dead fish as a manure for corn. The Peruvians have used guano since the beginning of their recorded history, and the Chinese have long recognized the vahie of all kinds of excrements as fertilizers. In the early part of the last century De Saussure established the existence of the mineral constituents of the soil in plants, and while he believed that they were essential to the life of plants, his contemporaries regarded them as nonessential, or at the best useful only as a kind of stimulant. Justus Von Liebig, as a result of his investigations, published in 1843 a new edition of his notable work, "Chemistry in its Application to Agriculture and Physiology," in which it may be said he laid the foundation of the celebrated "mineral theory" of agriculture. Prior to the time of liebig the use of manures and fertilizers was purely an art, the beneficial results of which had not been satisfac- torily explained. Liebig's mineral plant food theory, apparently so ample, and in time widely accepted, attributed the beneficial effect of fertilizers solely to the plant food constituents which were supplied to the growing plants, and thereby laid the foundation for the compounding of fertilizers to meet the needs of soils and crops. There soon sprang into existence a soil chemistr^^ which had for its object an investigation of the stores of plant food in soils with a view to ascertaining the agricultural value of lands, as well as the character and amount of fertilizer, if any, that would give the largest net crop income per acre. It was soon learned that the soil con- stituents existed in various forms or compounds not all of which were available as plant food, and this led to a study of inexpensive methods of determining the available constituents, a study which has formed no small part of soil chemistry even down to the present a A comparative study of the fertility of the soils of the central and eastern United States, as influenced by manures and fertilizers and measured by the paraffin-pot method. 28220— Bull. 48—08 2 7 8 FERTILITY OF SOILS AS AFFECTED BY MANURES. time, with little indication that a method will ever be devised that will prove universally satisfactory for all soils and crops. So far as can be gleaned from the writings of soil investigators, only within recent years have any considerable number of them come to realize that there is no definite relation between the available plant food constituents of soils, as determined by mineral analysis, and their crop-yielding capacities. Neither had they arrived at that further realization that the crop-yielding capacity of a soil is dependent upon a complexity of factors, any one of which may be dommant, but all of which are concerned in large or small degree. Arable soils contain the organic remains of previous crops together with excretions produced during their growth and are also filled with living forms. Their bacterial flora is diversified and the activities and processes which take place within the soil, physical, chemical, and biological, are exceedingly complex, and at present but imper- fectly understood. It is the sum total of these activities and their products that determines largely the fitness of a soil for plants, rather than the variation in the character and composition of its mineral matter. \Vhile manure and fertilizers may often increase crop yields as a result of a direct supply of plant food or as a stimulant to the plants, there is now abundant evidence that their effect is rather the result of a direct action upon the soil, thereby changmg its relation to plants. Complex and imperfectly under- stood as is this action, it is a more satisfactory explanation of the benefits derived from fertilizers than the idea that the inconsequential amounts of nitrogen, potash, and phosphorus applied, as compared with the larger stores of those elements already in the soil, should be directly responsible for such marked increase in growth as fre- quently follows the application of fertilizers. The prime object of manuring the soil, whether with stable manure, green manure, or commercial fertilizers, is to increase its crop- jaelding capacity, and in order to justify the practice the resulting increase in product must be more than enough to offset the cost of the fertilizers applied. That is to say, the beneficial effect on the present and succeeding crops must be sufficient to compensate for the cost of the fertilizers and give a profit on the capital so invested. The first noteworthy use of commercial fertilizers in the United States was in 1848, during which year there was imported 1,000 tons of guano. This was followed the succeeding year by twent}^ times that quantity, after which date the importation steadily increased until 1880, when it reached its maximum and began to fall off because of a failing supply. Other materials, notably sodium nitrate from Chile and the potash salts from Germany, have taken the place of the failing supply of guano, and these, together with the development of our phosphate mines, the use of cotton-seed meal, and the utilization OBJECT OF THE INVESTIGATION. 9 of slaughterhouse by-products, have met the continually increasing demand for commercial fertilizers by our farmers. According to our census reports, the expenditure for fertilizers in the United States in 1880 was $28,500,000. Ten years later it was $38,500,000, and in 1900 it reached the significant sum of $54,750,000. There seems little doubt that this rate of increase in the use of fertilizers will continue for some time to come, and the subject is one of sufficient national importance to justify careful investigations. OBJECT OF THE INVESTIGATION. The object of this investigation is not for the purpose of explaining how fertilizers act nor of studying the relation of the composition of the soil to the beneficial effect of the fertilizer when applied to it, but rather to compare the effects of several liigh-grade standard fertilizer ingredients, lime, stable manure, and cowpea vines when applied under like conditions to a large number of soils, collected from widely separated areas and representing a wide range in soil texture, type, origin, and crop adaptation. By making a comparative test of a large number of soils it was thought that there might be established a relation between the manurial requirements and the origin, formation, type, or crop adaptation of the soil. While there has been recorded in agricultural literature, in the aggregate, a large number of field tests of fertilizers on a great variety oi soils, representing nearly every State in the eastern half of the United States, it is impossible to make a satisfactory comparison of the results obtained, because of the innumerable details in which the conditions of the tests do not agree. The variation in the amount, class, kind, and composition of the fertilizer used, the time and manner of its application, the test crop used, and the weather condi- tions which prevailed form such a wide range of possible combinations that it is rare to find two tests that are strictly comparable. In the present investigation the same variety of wheat was used throughout as the test crop, and all of the conditions for growth, excepting the manurial treatments to be tested, were maintained as nearly uniform as possible. With the same fertilizer ingredients, in like form and used in the same combinations, for a crop common to all, the results obtained from this large number of soils are strictly comparable, and far exceed in number any similar tests that can be brought together on such a basis. Mention should also be made of the fact that all tests reported in the following tables under areal surveys — and they constitute fully four-fifths of them — were made for the further purpose of gaining practical knowledge concerning the manurial requirements of the principal soil types as established by the soil survey parties and with 10 FERTILITY OF SOILS AS AFFECTED BY MANURES. the ultimate object of determining what manures or fertihzers would give best results on the soils. The results of all such tests are reported in general terms in the soil survey reports for the respective areas in which they occur. THE SOILS. The soils tested in this investigation, 220 in number, represent 90 types, half that number of soil series, and many geological forma- tions. The samples were taken from twenty-three States, extending from the Mississippi River to the Atlantic seaboard and from Rhode Island to Texas. With the exception of a limited number of mis- cellaneous samples, the greater number were collected from areas that were in process of being surveyed by the Bureau of Soils dur- ing the summer of 1905, and the following winter, and are representa- tive of the soils of those areas rather than the vast expanse of country included in the twenty-three States al)ove mentioned. Each sample, being a composite, made up of fifteen or twenty small samples, taken from different parts of the same field, is representative of the field from which it is taken as well as the type which it represents. The sam- ples were placed in strong grain bags and shipped to the Bureau's headquarters, where they were at once transferred to covered gal- vanized iron cans for storage. In this way the original condition and original moisture content were, as far as practicable, maintained until they could be tested. This is important in relation to the test, for it has been found by experience that as a result of long storage and resulting air-dry condition soils become more productive and are usually less responsive to fertilizers than when such a change is prevented. THE METHOD. The paraffin-pot method, described in Circular No. 18 of this Bureau and in the appendix to Farmers' Bulletin No. 257 of Iho United States Department of Agriculture, was employed in these tests, and while it is not designed to supersede field tests, results obtained with it, when compared with results obtained on the same soils at the agricultural experiment stations in Rhode Island,*^ New York, Ohio,'* Iowa, Missouri, and North Carolina, show that it will indicate the manurial requirements of soils for general farm crops with a fair degree of accuracy. It is also admirably adapted to an investigation of the character described in the following pages. The soil to be tested is thoroughly pulverized by crushing all lumps, and if containing stone or gravel this is removed by sifting. The sample is then thoroughly mixed and equal amounts weighed into a See Bui. 109 of the Rhode Island Expt. Sta. b See Buls. Nos. 167 and 168 of the Ohio Agricultural Expt. Sta. THE METHOD. 11 granite-ware pans, where they receive their respective apphcations of manure or fertiHzers and are made up to their optimum water content with distilled water. After remaining in the pans for about a week, being occasionally wet with distilled water, and frequently stirred in order to secure a thorough incorporation of the fertilizer and a good soil condition, it is ready to pot and plant. Five small wire-gauze pots are used for each treatment. About 350 grams of soil is then placed in each pot, uniformly packed, and planted with six selected germinated kernels of wheat, after which the pots are dipped in melted paraffin, which not only forms an intimate contact with the soil but makes the pot water tight. The soil is then cov- ered with a thin layer of washed quartz sand and the pot and con- tents weighed and weight recorded. The pots are next placed in trays and given a favorable exposure in the greenhouse for three or four days or until the plants attain a height of about 1^ inches, at which time the pots are sealed. The sealing consists of covering the tops with paraffined paper disks in which are slits through which the plants grow. The disks are sealed to the sides of the pots by rheans of melted paraffin. The loss of water by direct evaporation from the soil is in this way reduced to a negligible quantity. During the growing of the plants, which usually continues from eighteen to twenty-one days from the date of sealing, the pots are weighed at intervals of two or three days and watered with distilled water in order to retain a favorable moisture content for plant growth. By this method the loss of water or the amount transpired by the plants may be ascertained periodically, and at the end of the experiment the total amount of water given off through the plants of each pot obtained for comparison with the growth and green weight of the plants, which is ascertained by cutting and weighing the plants at the time the experiment is concluded. All conditions of the experiment are so carefully controlled that the average result of five pots rarely differs more than 5 per cent from the average result of any other five pots that have been treated throughout in precisely the same manner (see Table I). Differences which occur beyond this amount may therefore safely be attributed to the different manurial treatments which have been given. Tliis method has several advantages over the growing of plants in open and porous pots. The method of coating the soil with paraffin prevents any accumulation of roots between the soil and the receptacle, a trouble which is common in pot experiments. The complete sealing up of the soil also enables the experimenter to determine the amount of water which the plant has actually used and transpired in its process of growth, and this, together with the small size of the pots, enables the moisture content and its fluctuations to be carefully controlled. 12 FERTILITY OP SOILS AS AFFECTED BY MANURES. On good soils, or as a result of favorable treatment, plants are produced in the little pots in twentj^-five daj^s or less time, the green weight of wliich sometimes equals or exceeds 1 per cent of the weight of the soil in wliich they grew. Such plants are approximately 85 per cent water; they have transpired about 100 units of water for each unit of green matter produced, and their dry matter is relatively richer in mineral constituents and nitrogen than that of mature plants. On this basis, an acre foot of soil, weighing 3,500,000 pounds, would produce 35,000 pounds of green matter, requiring for its pro- duction the equivalent of 17^ inches of rainfall. This green matter would be equivalent to 5,900 pounds of air-dry material, or about 50 bushels of wheat and H tons of straw to an acre. While the removal of green matter equal to 1 per cent of the soil in the little pots is somewhat above the average, it serves as an illustration of the heavy draft made upon the soil in a very short time, a draft, as regards moisture and mineral constituents, although part of the latter comes from the seed, greater than that which takes place under field conditions by the removal of a large mature crop, assuming that the removal takes place to a depth not greater than 1 foot, which for wheat and similar crops would be approximately correct, and providing also the movement of plant food by capillarity from below 1 foot be ignored. Table I. — Actual transpiration in grams for each of twenty pots, on three soils, under uniform, treatment, and the percentage variation from the average for each pot, also average percentage variation of each group of five pots. Leonardtown loam . Corn. Cecil clay— poor. Leonardtown loam. Transpi- ration. Variation. Transpi- ration. Variation. Transpi- ration. Variation. For eacti For each For 5 For each For each For 5 For each For each For 5 pot. pot. pots. pot. pot. pots. pot. pot. pots. Grams. Per cent. Per cent. Grams. Per cent. Per cent. Grams. Per cent. Per cent. 69.9 - 5.5 39.3 +10.1 87.5 + 0.4 77.6 + 4.9 25.7 -28.0 90.5 + 3.9 58.7 -20.1 ■ -0.9 37.7 + 5.6 ■ -2.8 89.1 + 2.2 ■ +3.3 83.5 + 12.8 34.8 - 2.5 98.4 + 12.9 76.5 + 3.3 35.9 + 0.6 84.8 - 2.6 59.1 -20.2 ■35.8 + 0.3 89.9 + 3.2 72.1 - 2.6 37.3 + 4.4 99.2 + 13.9 79.2 + 7.0 -3.0 38.6 + 8.1 ■ +4.1 87.7 + 0.7 ■ +1.6 73.9 - 0.2 35.4 - 0.8 80.5 - 7.6 74.1 + 0.8 38.9 + 8.9 85.2 - 2.2 82.3 + 11.1 38.8 + 8.6 79.8 - 8.4 1 66.8 - 9.8 33. 6 - 5.9 79.5 -8.7 70.6 - 4.6 • +2.2 37.8 + 5.9 • +0.8 86.1 - 1.1 • -2.7 68.9 - 6.9 38.1 + 6.7 88.8 + 1.9 89.9 +21.4 31.7 -11.2 89.5 + 2.7 84.3 + 13.8 34.7 - 2.8 82.4 - 6.4 79.0 + 6.7 32.3 - 9.5 83.0 - 4.7 76.4 + 3.2 • +2.0 39.8 + 11.2 ■ -2.5 94.3 + 8.2 -2.1 68.3 - 7.7 36.9 + 3.4 87.4 + 0.3 69. 7 - 5.8 31.2 -12.6 79.3 - 8.9 FERTILIZERS USED. 13 FERTILIZERS USED. In this investigation the tests were confined to the use of high-grade, standard fertilizing materials, consistmg of nitrate of soda, sulphate of potash, and acid phosphate, together with air-slaked lime, well- decomposed stable manure, and green cowpea vines. These ingredi- ents were applied to the soil separately and in various combinations as follows: Treatment and rate per acre. 1. Untreated. 2. Manure, 10 tons. 3. Lime, 1 ton. 4. Nitrate of soda, 200 pounds. 5. Sulphate of potash, 200 pounds. 6. Acid phosphate, 200 pounds. 7. Nitrate of soda and sulphate of potash, 200 pounds each. 8. Nitrate of soda and acid phosphate, 200 pounds each. 9. Sulphate of potash and acid phosphate, 200 pounds eacli. 10. Nitrate of soda, sulphate of potash, and acid phosphate, 200 pounds each. 11. Same, plus lime, 2,000 pounds. 12. Cowpea vines 5 tons, lime 2,000 pounds. The rate of application is based on the weight of an acre of soil to the depth of 7 inches, which is approximately 2,000,000 pounds. The fer- tilizers were applied in solution, and the lime, manure, and cowpea vines in bulk, each being finely ground. CALCULATION OF RESULTS. While tliis report combines the results of more than 13,000 pots, individual record of which has been made, the results were reported on forms, of which the following is a sample : [Can 101.] Actual transpiration and green weight and relative transpiration and green iveight of SO wheat plants on the basis of 100 for the untreated soil. Soil: Norfolk Sandy Loam. Locality: Waycross, Ga. Planted: Aprils. Sealed AprillO. Disctd.: April 28. Basket Nos.: 2066-2125. Num- ber. Treatment and parts per million.a Transpira- » tion, 18 days Green weight. Relative varia- tion by — Transpi- Green ration, weight. Untreated Manure 10,000 Cowpeas 5,000, lime 1,000 Nitrate of soda, potassium sulphate, and acid phos- phate 100 each Same + lime 1,000 Nitrate of soda and potassium sulphate 100 each. . Nitrate of soda and acid phosphate 100 each Nitrate of soda 100 Potassium sulphate 100 Potassium sulphate and acid phosphate 100 each . . Lime 1,000 Acid phosphate 100 757.6 1295.2 1135.7 1160.3 1090. 1 1047. 2 997.6 1053. 925.9 833.6 803.0 651.9 7.0 14.2 13.6 11.5 11.0 10.8 9.6 9.5 8.0 8.0 7.0 6.5 100 171 150 153 144 139 132 139 122 110 106 86 100 203 194 164 157 154 137 136 114 114 100 93 a Parts per million multiplied by 2 equals the pound rate per acre on the basis of 2,000,000 pounds as the weight of an acre of soil 7 inches deep. 14 FEKTILITY OF SOILS AS AFFECTED BY MAISTUKES. On tliis form are given the total transpiration and total weight of green plants for each treatment, consisting of five pots. The trans- piration and the green weight of plants for each treatment are entered in the third and fourth columns, respectively, wliile the relative transpiration and weight, based on 100 for the untreated soil, are given in the fifth and sixth columns and are merely com- puted from the actual transpiration and weight in order to convert the comparison to a percentage basis and make it more simple. The variation by transpiration, wliile a good indication of the relative growth and effect of the treatments, frequently gives a range of lesser magnitude than the actual growth of plants. Plants that have made a marked increase in growth as a result of soil treatment usually contain a higher percentage of water than untreated ones and therefore show a slightly lesser variation by dry, or water-free, weight than by green weight. See Table II, wliicli follows. The increase in growth is also accompanied by an improvement in con- dition which can only be measured by appearance, but wliich will usually prove an advantage to the better plants if grown to maturity. The results given in this bulletin are all based on the actual green weight of plants, the weighings being made immediately upon cutting the plants and under uniform conditions. Table II. — Percentage gain or loss attributable to various fertilizer treatments on four soils. [P=acid phosphate. K=sulphate of potash. N=nitrate of soda. L=lime. M=stable manure.] A.— BASED ON GREEN WEIGHT OF PLANTS GROWN ON UNTREATED SOILS. Green weight of plants grown on un- treated soils. Percentage gain or loss attributable to — Soil. P. K. N. PK. PN. KN. PKN. L. PKNL. M. Grams. 8.8 8.8 8.3 6.5 - 1 16 - 7 - 2 1 7 15 - 4 47 8 9 - 6 41 14 7 47 12 - 4 5 46 22 5 6 10 1 9 Cecil clay, poor -14 17 Leonardtown loam, poor.. 24 7.6 2 5 13 11 20 17 5 9 B.— BASED ON DRY WEIGHT OF PLANTS GROWN ON .UNTREATED SOILS. Dry weight of plants grown on un- treated soils. Percentage gain or loss attributable to— Soil. P. K. N. PK. PN. KN. PKN. L. PKNL. M. Cecil clay, good Grams. L60 1.55 L20 L22 - 1 - 3 13 2 -11 - 4 4 18 - 9 3? 5 - 1 - 8 25 3 8 47 9 13 5 36 24 -5 -1 8 6 -11 -17 Leonardtown loam, good. Leonardtown loam, poor.. 18 1.39 3 2 8 4 17 19 2 ARRANGEMENT AND TABULATION OF RESULTS. 15 ARRANGEMENT AND TABULATION OF RESULTS. Table III, which follows, gives the actual green weight of plants grown on the untreated soils and the percentage increase or tlecrease in growth, the latter indicated by the minus ( — ) sign, resulting from th^ various fertilizer ingredients and combinations. The data arc arranged by States, areas, and types and averages are given for areas and for types within areas. The sjniibols at the head of the columns are used for the sake of brevity. P stands for acid phosphate ; K for sulphate of potash; N for nitrate of soda; L for air-slaked lime; M for well-rotted stable manure, and Cv for green cowpea vines. The ingre- dients and rate of application are given in the preceding form. All data resulting from the tests are embodied in Table III. Marked variations from the general trend are exceptional and most frequently occur with manure or cowpea vines. These substances have occa- sionally given negative results apparently as a result of decomposition products from which the soil did not have sufficient time to recover prior to the planting of the seed. The blanks in the table indicate no tests for the ingretlients or combinations heading the columns in which they occur. Practically all such omissions occurred in miscellaneous samples which were tested by the Bureau parties at several of the State experiment stations. Table III gives the results for 220 soils and from these results many tabulations w^ere made, parts of which are given on subsequent pages. The arrangement and segregation of data in subsequent tables are for the purpose of a further compara- tive study of the facts as well as to enable the reader to verify state- ments that may follow in the text, although a number of facts are given from tabulations other than those wdiich follow. Table III. — Percentage increase in growth attributable to the various fertilizer appli- cations. Weight of plants grown on un- treated soil. Gain or loss attributable to— Soil type and locality. P. K. N. PK. PN. KN. PKN. L. PKNL. M. CvL. WISCONSIN. Portage County: Miami stony sand Miami sand Grams. 9.9 6.0 9.0 11.0 8.4 10.0 8.7 P.ct. 6 7 4 - 1 7 -11 2 P.ct. 13 15 8 - 2 9 - 9 -16 P.ct. 11 33 2 14 31 - 6 - 2 P.ct. - 3 5 - 1 2 13 - 1 5 P.ct. •11 24 10 3 23 -10 - 9 P.ct. 21 29 14 16 32 - 3 10 P.ct. 22 39 3 4 35 3 1 P.ct. - 3 7 4 2 11 5 9 P.ct. 19 43 21 3 8 10 -14 P.ct. P.ct. 47 8 80 4 Miami sandy loam Marshall gravelly loam Marshall sand 29 36 34 21 42 12 14 3 Portage silt loam Portage silt loam 10 -14 Average for area 9.0 2 3 12 3 7 17 15 5 13 41 4 MISSOURI. Crawford County: Clarksville silt loam. . . Wabash silt loam Clarksville stony loam. 6.6 7.6 14.6 26 -11 4 15 1 3 17 - 3 - 2 27 - 2 - 9 25 3 2 15 - 5 11 7 - 1 3 14 - 3 1 2 - 2 43 23 -10 32 24 13 Average for area 9.6 1 6 6 4 5 10 7 3 4 19 23 28220— Bull. 48—08- 16 FEKTILITY OF SOILS AS AFFECTED BY MANURES. Table III. — Percentage increase in growth attributable to the various fertilizer appli- cations — Continued . Weight plants grown on un- treated soil. Gain or loss attributable to— Soil type and locality. P. K. N. PK. PN. KN. PKN. L. PKNL. M. CvL. MISSOURI— continued. Scotland County: Shelby silt loam Miscellaneous: Loess, McBain County Grams. 9.6 7.8 7.0 5.9 P.ct. 7 8 P.ct. 24 22 P.ct. 14 28 P.ct. 22 26 25 29 P.ct. 17 26 50 54 P.ct. 32 47 60 44 P.ct. 28 50 68 61 P.ct. 11 19 25 7 P.ct. 28 P.ct. 69 23 57 34 P.ct. 30 35 Clay loam, Salem 50 66 35 Silt loam, Columbia. . . 50 INDIANA. Newton County: Marshall fine sandy 6.3 7.6 12.6 5 17 -21 12 22 - 2 9 17 -14 20 - 8 5 14 -23 12 21 13 - 4 16 -17 -12 22 -12 - 1 21 - 7 60 77 32 — 2 34 Clyde fine sand - 4 Average for area 8.8 11 4 4 - 1 15 - 2 - 1 4 56 9 Tippecanoe County: Marshall silt loam 7.2 7.5 4 - 3 30 1 2 21 6 7 28 10 - 6 26 58 3 18 46 12 6 58 - 7 - 3 17 6 7 35 - 2 42 71 4 — 2 Miami silt loam Average for area 5.6 30 6.8 10 8 14 10 31 22 25 2 16 37 11 OHIO. Westerville area: Miami black clay loam. Miami clay loam 8.3 6.1 8.6 -11 11 - 8 6 5 , 3 - 2 15 - 2 9 3 - 3 13 27 1 7 11 8 8 13 - 5 1 7 2 11 11 2 22 15 48 5 45 15 Average for area 7.8 - 3 5 4 3 14 9 5 3 8 28 22 Miscellaneous: Volusia silt loam. 5.8 7.1 4 - 4 14 6 - 9 47 24 — 6 46 30 — fi 62 27 - 6 21 9 30 76 4/ 52 10 9 Miami clay loam, Ger- man town 22 Miami clay loam. Strongs ville - 5 12 NEW YOEK. Tompkins County: Dunkirk clay loam Dunkirk clay loam Average for type 9.0 4.5 -IS 9 4 13 55 - 1 - 6 6 27 23 47 11 42 -14 29 12 68 33 - 5 54 6.8 - 9 7 34 - 3 17 35 27 8 51 25 Dunkirk loam 8.4 9.6 - 4 4 2 1 15 1 - 1 - 4 16 - 8 20 12 11 - 7 11 7 46 .36 Miami stony loam 26 Average for area 7.9 - 5 4 21 - 3 10 ! 24 1 19 2 10 46 19 Bingham ton area: Dunkirk gravelly sandv loam 6.6 10.6 10.6 4 9 - 1 1 9 11 11 1 5 4 - 8 16 20 12 16 9 22 1 7 26 26 - 3 10 27 -11 15 29 20 31 69 Dunkirk gravelly loam Wabash loam - 9 10 Average for area 9.3 4 7 6 4 11 16 11 11 10 27 23 Other localities in State: Volusia silt loam Volusia silt loam Volusia silt loam Dunkirk clay loam Dunkirk clay loam 5.4 4.8 6.6 1 8.9 10.2 2 - 2 4 16 11 15 15 22 29 11 13 34 10 20 18 11 15 11 31 - 5 8 21 4 10 - 6 38 53 19 'I 20 35 23 28 5 9 15 8 19 8 32 62 15 22 51 81 77 17 17 29 ■■"i3 28 21 ARRANGEMENT AND TABULATION OF RESULTS, 17 Table III. — Percentage increase in growth attributable to the various fertilizer appli- cations — Continued. Weight of plants grown on un- treated soil. Gain or loss attributable to — Soil typo and locality. P. K. N. PK. PN. KN. PKN. L. PKNL. M. CvL. PENNSYLVANIA. Montgomery County: Hagerstown loam Penn silt loam Lansdale silt loam Chester loam. . . . Grams. 10.3 12.3 9.9 7.8 P.ct. 13 8 P.ct. - 2 12 23 9 P.ct. 1 8 1 4 P.ct. 11 11 27 21 P.ct. 13 11 - 7 18 P.ct. 6 24 31 15 P. ft. 9 13 7 27 P.ct. 6 - 8 19 4 P.ct. 23 4 17 33 P.ct. 20 24 9 13 P. cl. 29 36 30 8 Average for area 10.1 5 10 4 18 9 19 14 1 5 19 17 24 RHODE ISLAND. The State: Miami silt loam, King- ston 3.6 - 8 -19 -14 28 22 22 31 133 111 39 Miami stony loam, Middlcton 7.8 5.8 6.7 0.1 15.4 5.6 18 24 7 10 5 4 9 60 5 10 21 41 94 4'! 17 8 35 i Miami stony loam, Jamestown 1 Miami stony loam, Curtis Corners 1 Miami stony loam, Woonsoeket Miami stony loam. Little Compton Miami stony loam, Ashaway 1 Average for typo 8.4 1 11 18 40 t Gloucester stony loam, East (ireonwich 8.7 8.5 " 15 4 3 2 -22 2 9 25 8 37 21 14 30 Gloucester stony loam, Chepachet Gloucester stony loam, Tarkiln 11.4 8.5 Gloucester stony loam, Foster Gloucester stony loam, Greene 7.2' ! Average for type . . 8.9 1 9 20 Alton stony loam, Hope ValleV 5.5 4.7 1 1 34 51 38 - 2 54 51 1 Alton stony loam, Wickford i Average for type . . 5.1 43 18 53 1 ' Warwick sandy loam, Alton 7.1 7. 2 1 41 23 - 8 28 42 42 Warwick sandy loam, Harrington 1 1" ■ Average for type.. 7.2 1 I 1 32 10 42 1 1 Average for area . . 7.9 1 15 14 35 1 1 , VIRGINIA. Louisa County: Cecil sandy loam Cecil loam 7.2 4.6 24 5 36 47 13 75 31 26 13 75 27 100 30 102 5 71 39 153 25 63 57 158 Average for area .5.9 30 1 26 44 1 28 44 64 66 38 96 44 1 108 Hanover Couhty : Wickham sandy loam. Norfolk sandy loam . . . 7.5 7.5 -3 5 3 : 11 11 i -1 10 20 10 17 10 35 36 30 9 - 1 35 47 31 44 -12 - 8 Average for area .... 7.5 I 8 16 8 14 26 33 4 41 3S 10 18 FERTILITY OF SOILS AS AFFECTED BY MANURES, Table III. — Percentage increase in growth attributable to the various fertilizer appli- cations — Continued. Weight plants grown on un- treated soil. Gain or loss attributable to— Soil type and locality. P. K. N.' PK. PN. KN. PKN. L. PKNL. M. CvL. VIRGINIA— continued. Appomattox County: Gram.i. 12.9 11.7 13.2 P.ct. P.ct. P.ct. 6 - 3 4 P.ct. -13 - 2 3 P.ct. 4 P.ct. 12 P.ct. 2 9 6 P.ct. - 1 - 6 5 P. ct. 2 7 - 4 P.ct. - 3 _ 2 2 P. ct. 24 Cecil clay, Pocahontas field Cecil clay, Mitchell field — 3 4 7 6 NORTH CAROLINA. New Hanover County: ■ 4.1 6.4 7.3 3.0 37 3 3 13 41 9 3 30 122 37 22 77 61 20 3 40 124 45 20 87 131 51 33 117 151 63 29 110 66 16 8 70 l.-)4 70 31 173 241 94 57 130 144 -lorfolk fine sand Norfolk fine sandy loam . 37 34 Portsmouth fine sand . 230 Average for area 5.2 14 21 65 31 69 83 88 40 107 131 111 Miscellaneous: Cecil clay (good), Statesville. 8.8 8.8 4.5 . 4.1 .'•,.2 4.8 3.0 12.4 4.5 4.8 4.5 4.6 4.7 - 1 m 20 44 25 - 6 11 4 42 2 - 4 9 - 2 1 34 31 21 42 22 - 2 15 44 26 19 11 - 4 42 30 16 63 42 29 18 20 42 14 32 28 99 29 19 33 29 -14 16 31 35 16 19 20 9 - 6 97 23 9 52 51 24 26 20 43 23 19 17 14 7 35 39 28 71 59 40 11 51 38 62 21 22 - 4 5 105 49 34 56 63 19 25 24 56 62 42 30 5 6 45 52 - 1 1 10 31 43 20 19 4 11 9 -14 90 53 33 66 38 8 61 01 41 46 28 Cecil clay (poor) , Statesville 3 Cecil clay, Iredell test farm. 105 103 28 70 53 26 22 46 57 51 16 Cecil sandy loam, Raleigh Iredell clay loam, Statesville 54 Norfolk fine sandy loam, Edgecombe . . . Norfolk fine sandy loam, TarV)oro Porters clay, Asheville. Porters clay, Biltmore. Porters sandy loam, Blantyre 78 15 24 31 Portsmouth s a n d, Pinehurst 45 Portsmouth silt loam, Chowan County Silt loam, U li i o n County 74 Tobacco soil. Wake County SOUTH CAROLINA. York County:- Cecil sand 3.9 7.0 7.7 -15 40 - 8 3 6 26 28 19 28 - 9 1 8 6 20 40 16 54 34 18 S 33 3 71 41 17 90 76 51 00 Cecil sandy loam Iredell clay loam 58 19 Average for area 6.2 8 24 7 5 25 35 15 43 72 46 Cherokee County : Cecil clay 7.4 6.6 5.9 6.9 9.1 - 4 2 - 3 -12 - 3 - 6 1 7 28 7 36 52 46 35 32 - 2 9 - 1 9 1 26 47 46 33 27 39 52 46 60 46 35 48 52 58 27 4 39 29 4 14 44 56 68 42 42 81 88 71 36 12 Cecil silt loam . . . 74 Cecil fine sandy loam. . Cecil sandy loam Iredell clay loam 123 5 7 Average for area 7.2 - 4 7 40 3 36 49 44 18 55 64 44 Lancaster County: Cecil silt loam 7.3 11.8 5.3 75 15 13 15 5 9 63 16 35 61 14 26 67 26 30 8 19 40 73 28 105 10 25 44 111 Miscellaneous: Orangeburg sandy loam, St. Matthews. . Portsmouth sandy loam, Darlington -2 17 138 ARRANGEMENT AND TABULATION OF RESULTS. 19 Table III. — Percentage increase in growth attributable to the various fertilizer appli- cations — Continued . Weight of plants grown on un- treated soil. Gain or loss attributable to— Soil type and locality. P. K. N. PK PN. KN PKN. 1 L. PKNL. M. CvL. TENNESSEE. Henderson County: Lexington silt loam... KENTUrKY. McCracken County : Memphis silt loam Grams. 7.5 5.0 P.ct 1 - 3 P.ct 1 3 P.ct 36 39 P.ct 5 8 P.ct 16 25 P.ct 44 42 P. ct. 36 42 P.ct 5 18 P.ct. 57 48 P.ct 72 70 P.ct. 17 j - 2 r,E(>RGi.\. Wayeross area: Norfolk sand, 12 miles west Wayeross Norfolk sand, 4J miles SE. Wayeross Norfolk sa'nd, 3J miles NW. Waltertown ... 5. (i 6.1 - 9 - 4 •- 2 24 30 - 5 45 61 36 27 18 5 49' a5 45 . .50 22 j 41 72 57 24 ! 64 20 34 109 121 55 124 114 77 118 132 65 Average for ty^e 5.7 - 5 16 47 17 39 1 52 51 39 95 105 105 Norfolk fine sand, 3 miles W. Wayeross. . Norfolk fine sand, S miles SE. Wayeross . Norfolk fine sand, 2 miles W. Wayeross.. (;. 1 (i.O 7.4 7 5 - 9 15 8 - 8 55 53 - 3 23 - 2 - 6 49 18 -12 69 38 74 34 50 1 27 14 6 84 32 25 102 113 28 107 107 36 Average for type . . G.5 1 5 35 5 18 36 46 22 47 81 83 Norfolk Sandy loam, 2i miles N. way- cross /.O (i.5 5.0 - 7 11 14 15 16 36 54 20 14 15 3 37 77 15 54 102 42 64 85 25 35 -9 57 100 43 103 109 54 94 109 42 Norfolk sandy loam, 3 miles N W . Vv ay- cross Norfolk sandy loam, 2J miles SW. Elsie Average for type.. G.2 1 15 37 11 43 66 58 9 67 89 82 Norfolk fine sandy loam, b\ miles W. Wayeross 7.5 7.9 3.9 17 1 11 28 5 29 60 29 74 23 6 35 63 21 51 83 38 79 73 24 80 40 22 64 89 35 105 113 96 150 113 67 176 Norfolk fine sandy loam, 3 miles E. Wayeross Norfolk fine sandv loam, 4 miles N. Manor Average for type.. 6.4 10 21 54 21 45 67 61 42 76 120 119 Portsmouth fine sand. 5i miles SE. Way- cross 5.2 5.2 5.5 6 1 2 21 15 18 73 50 33 31 26 24 92 75 33 104 80 29 77 90 71 29 92 20 117 153 75 150 219 64 148 175 109 Portsmouth iinesand, 1 mile S. Needham Portsmouth fine sand, 1 mile S. Glemnore . . . Average for type. . 5.3 3 18 52 27 67 71 79 47 115 144 144 Average for area... 6.0 2 15 45 16 42 59 59 32 80 108 107 FLORIDA. ; Kscambia County: Portsmouth sand 8.3 2 5 12 5 18 11 11 48 48 47 : 35 Norfolk sand 5.4 5.9 7.2 16 5 4 18 32 19 13 45 33 11 15 9 11 20 16 13 84 41 11 46 22 46 24 29 74 55 23 83 94 66 107 97 65 Norfolk sand Norfolk sand Average for type... 6.2 8 23 30 12 16 46 26 33 51 81 90 20 FERTILITY OF SOILS AS AFFECTED BY MANURES. Table III. — Pcrcenkuje inarase in growth attributable to the various fertilizer appli- cations — Continued . Weight plants grown on un- treated soil. Gain or loss attributable to— Soil type and locality. P. K. N. PK. PN. KN. PKN. L. PKNL. M. CvL. FLORIDA — continued. Escambia County— Cont'd. Norfolk fine sandy Grams. 8.4 5.5 P.ct. - 7 10 P.ct. 14 12 36 P.ct. 15 -5 27 P.ct. 8 32 33 P.ct. -12 39 48 P.ct. -4 30 12 P. ct. 11 32 48 P.ct. 35 45 136 P.ct. 14 82 130 P.ct. 32 32 90 P.ct. 51 Norfolk fine sandy 85 Norfolk fine sandy 3.3 24 136 Average for type . . 5.7 1 9 21 12 24 26 13 30 72 75 51 91 Average for area . . 6.3 8 19 20 16 20 27 26 52 61 63 82 ALABAMA. Lee County: 8.6 5.6 1 -11 9 - 7 14 29 - 2 -11 8 13 31 11 15 43 10 13 28 52 "4.3' 58 Average for type. 7.1 - 5 1 21 - 7 10 21 29 12 40 ! 6.8 4.7 9 - 1 10 40 - 1 30 -I 63 21 77 40 62 35 57 50 103 135 107 1 160 Norfolk sand . 108 j 143 -Vverage for type . 5.8 4 4 35 42 58 48 53 119 108 i 151 Norfolk sandy loam Norfolk sandy loam 6.1 7.7 -14 12 4 19 42 55 15 20 37 50 43 1 46 53 j 63 11 19 64 87 42 ! 104 61 j 164 Average for type . 6.9 - 1 12 48 18 43 48 54 15 75 52 134 Norfolk coarse sand Norfolk coarse sand 5.3 6.4 - 7 -12 15 48 20 -i 34 3 49 13 50 25 5 22 70 23 69 33 101 41 Average for type . 5.8 - 9 7 34 3 18 31 i 37 13 46 46 1 71 Average for area . , 6.4 - 3 6 35 3 28 40 i 42 23 70 65 1 110 Miscellaneous: Clarksville clay loam, 11.5 6 25 17 42 4 25 -4 17 6 37 5 19 - 4 13 40 45 9 59 3 19 - 1 10 -15 -13 Clarksville silt loam. !1. 5 17 ' 20 19 19 34 ' 67 2 12 28 j 69 4 16 20 21 12 1 15 - 3 i 30 Orangeburg clay, Kings farm Orangebmg clay. Ford's farm "... Orangeburg sandy loam, Rawl'sfarm... Orangeburg sandy loam, Marion 10.3 9.1 S. 1 4.8 3 22 6 14 6 24 - 5 7- 35 32 55 49 17 ' 17 69 1 37 26 35 20 60 MISSISSIPPI. Pontotoc County: 6.4 5.4 6.6 7.6 8 41 - 9 17 8 44 3 - 9: 28 52 18 28 13 6 32 2 6 22 39 14 10 23 14 44 27 21 25 28 50 45 26 14 6 63 6 9 31 9 56 1 37 47 1 18 39 25 18 1 29 48 Lufkin silt loam. 78 32 Orangeljurg sandy loam 24 8. ! — 4 !- 6 39 Average for area 6.8 10 9 28 9 22 26 32 16 38 24 44 Montgomery County: Orange iJin-g f" i n e 7.5 24 8.5 7 20 17 43 29 13 27 53 17 53 48 24 24 64 60 60 26 ' 22 16 30 48 ARRANGEMENT AND TABULATION OF RESULTS. 21 Table III. — Percentage increase in growth attributable to the various fertilizer applica- tions — Continued . Weight plants grown on un- treated soil. Gain or loss attributable to — Soil type and locality. P. K. N. PK. PN. KN PKN. i;. PKNL. M. CvL. MISSISSIPPI— continued. Montgomery Co.— Cont'd. Memphis silt loam Memphis silt loam Memphis silt loam Memphis silt loam Memphis silt loam Memphis silt loam Orams. 10. 6 9.5 8.6 10.2 11.2 7.0 P.ct. 1 -13 ' -10 -13 - 3 -10 7 P.ct - 3 - 9 6 - 2 - 7 9 P.ct - 2 12 -15 -11 14 P.ct. -16 - 8 3 - 6 -10 4 P.ct 10 5 -13 14 P.ct 7 8 14 9 5 12 P. ct. 1 13 12 8 14 p.ct -1 -5 6 1 13 ; p. ct. ■ 16 1 3 20 -3 17 P.ct. -4 -5 6 18 5 36 P.ct. 6 11 34 28 9 40 Average for type 9.5 - 7 - 2 - 2 - 6 11 7 3 11 9 21 Average for area . . . 9.3 - 1 3 8 9 18 14 8 18 18 30 LOUISIANA. Caddo Parish: Caddo fine sandy loam. Cadrto fine sandy loam. Caddo fine sandy loam. Caddo fine sandy loam. 4.4 5.0 8.8 6.0 18 -10 - 7 6 18 4 5 17 86 33 7 26 25 - 4 - 8 - 4 16 2 15 20 84 62 2 12 36 4 9 8 18 8 8 19 52 72 14 16 20 44 6 79 186 148 41 23 Average for type... 6.0 2 11 38 2 13 40 14 13 39 37 100 Miller fine sand Miller fine sand 5.8 7.3 12 21 12 26 38 10 3 21 24 21 38 17 19 1 22 14 34 16 81 43 48 36 Average for type 6.6 17 19 24 12 31 1 19 10 18 25 62 42 Norlollc fine sand Norfollv fine sand Norfolk fine sand 3.7 5.4 5.3 - 3 21 15 51 19 21 62 41 32 65 5 36 54 65 61 62 32 61 51 56 59 41 39 9 113 80 74 146 92 28 195 98 80 Average for type 4.8 11 30 45 35 j 49 63 55 30 89 89 124 Norfolk fine sandy loam 8.7 9.6 5.6 7.9 4 9 - 8 9 - 3 20 1 - 5 7 13 35 -10 9 -19 19 3 17 8 76 16 lb „ 21 13 44 AA 9 -1 33 35 t 1 30 9 22 54 18 25 1 52 47 Norfolk fine sandy loam Norfolk fine sandy loam 28 2 54 56 Norfolk fine sandy loam Average for type 8.0 - 1 8 11 3 29 31 1 19 9 40 26 31 Orangel)urg fine sand . . Orangeliurg fine sand. 7.4 5.0 8 -22 10 -10 21 -10 8 -10 15 20 28 14 21 28 20 12 28 50 15 114 28 124 Average for type 6.2 - 7 6 - 1 18 21 25 16 39 115 76 Orangeburg fine sandy loam 6.5 5.0 3 3 6 ft 10 7 20 30 15 32 7 15 44 20 26 34 96 Orangeburg fine sandy loam 10 22 Average for type 5.7 2| 5 5 11 15 19 24 4 30 27 61 Average for area 6.3 3 1 13 24 1 10 26 35 24 15 1 45 54 j 74 ARKANSAS. Prairie County: Calhoun clay 8 3 12 12 14 12 16 16 26 IS 27 4 13 Crowley silt loam Crowley silt loam Crowley silt loam Crowley silt loam 7.1 10.1 7.0 8.1 - 8 1 - 1 3 10 9 14 9 43 2 9 i -6-5 - 6 22 - 3 j 7 -2 -2 26 13 7 3 27 36 2 43 46 60 5 71 33 19 7 64 18 38 9 80 42 Average for type 8.1 7 14 - 1 6 9 9 32 42 27 1 42 22 FERTILITY OF SOILS AS AFFECTED BY MANURES. Table III.— Percenia^e increase in growth attributable in the 'various fertilizer appli- cations — Continued . Weight plants grown on un- treated soil. Gain or I ass attributable to — Soil type and locality. P. K. N. PK. PN. KN. PKN. L. PKNL. M. CvL. ARKANSAS— continued. Prairie County— Cont'd. Grams. 8.8 10.8 11.1 P.ct. - 9 4 5 P.ct. 8 - 3 13 P.ct. 1 2 - 3 P.ct. 2 - 7 4 P.ct. U 2 3 P.ct. 6 6 9 P.ct. I,! P.ct. 6 '2 - 7 P.ct. - 7 4 - 1 P.ct. 6 4 5 P.ct. 22 Morse clay 9 4 Average for type 10.2 6 5 7 - 5 - 1 5 12 Acadia silt loam Acadia silt loam Acadia silt loam Acadia silt loam 11.2 7.3 7.4 9.0 4 7 20 11 29 10 29 6 11 12 31 11 25 1 18 31 24 10 22 26 27 15 41 18 12 14 23 -11 33 37 48 41 46 7 31 32 27 16 38 21 Average for type 8.7 6 11 19 16 19 21 25 10 40 29 26 Waverly silt loam Waverly silt loam .... Waverly silt loam 6.5 7.8 5.0 15 - 7 28 20 9 20 27 - 7 18 27 - 3 23 15 - 4 22 31 4 29 38 - 3 28 61 37 104 55 40 150 38 18 10 54 41 132 Average for type . . . 6.4 12 16 13 16 11 21 21 67 82 22 76 Average for area . . 8.3 5 10 12 8 11. 15 14 26 40 21 36 TEXAS. San Marcos area: Blanco loam ..... 8.7' 17 21 19. 21 23 10 33 20 21 28 Crawford silt clay . . . Crawford silt clay. . . Crawford silt clay. . . 9.2 9.2 8.8 20 - 7 13 21 2 25 31 19 30 20 - 1 18 40 8 29 29 19 30 22 23 42 20 4 24 42 31 46 40 14 56 52 34 11 Average for type . . . 9.1 9 16 27 12 26 26 29 16 40 37 32 Houston black clay. . . Houston black clay ... Houston black clay. . Houston black clay. . 7.9 8.9 7..0 8.1 8 20 - 4 8 4 21 - 5 21 15 31 21 8 17 25 5 23 4 33 12 9 9 35 21 30 17 35 19 15 5 17 6 21 23 42 26 30 8 10 25 42 23 38 32 Average for type . . . 8.0 6 7 22 14 18 19 25 11 28 18 34 Average for area 8.5 8 12 23 14 22 20 28 14 32 26 34 Rusk County: Caddo fine sandy loam Caddo fine sandy loam 8.2 8.1 6 12 2 11 53 40 5 15 51 43 53 35 47 34 7 17 47 36 61 62 57 40 Average for type . . . 8.2 9 7 47 10 47 44 40 12 42 62 49 Norfolk fine sand Norfolk fine sand Norfolk line sand 8.5 7.8 8.6 14 - 4 5 16 1 41 21 19 16 6 - 3 29 17 21 54 28 36 49 28 36 21 3 10 55 60 26 86 103 61 48 69 37 Average for type 8.3 5 10 27 , 6 22 39 38 11 47 83 ^ 51 Norfolk fine sandy loam . . 7.5 6.1 20 31 28 31 38 48 41 48 42 52 62 56 65 48 21 40 64 41 105 48 62 Norfolk fine sandy loam 80 Average for type fi.8 26 30 43 45 47 59 57 31 53 76 71 Orangeburg fine sandy loam 9.2 9.2 7.5 7.0 -12 - 7 - 3 - 6 2 20 14 34 31 9 - 9 - 1 24 - 3 - 6 26 27 3 14 26 39 3 36 26 40 9 13 12 8 -11 39 37 60 26 68 37 80 29 53 Orangeburg fine sandy 66 Orangeburg fine sand . . Susquehanna fine sandy loam 68 36 Average for area 8.0 6 11 32 13 28 37 38 13 45 67 54 AREANGEMENT- AND TABULATION OF RESULTS. 23 Tablk III. — Percentage increase in (jroivth attributable to the various fertilizer appli- cations — Continued. Soil typo and loc-tility. TEXAS — continued. Miscellaneous: Galveston clay, Galveston Houston black clay loam, San Antonio.. Orangeburg clay, Maroney's farm Orangel>urg fine sandy loam, Crockett Orangeburg fine sandy loam, Palestine Orangel.)urgfine sandy loam, Nagadoches . . . Orangeburg fme sandy loam, Palestine Miscellaneous: Wisconsin drift (good) Wisconsin drift (poorj lowan drift Missouri loess South lowan loess Kansan till ILLINOIS. Marion silt loam. Red silt loam NEW JERSEY. Miscellaneous: Norfolk sandy loam. Woodbine Norfolk fine sandy loam, Woodbine CONNECTICUT. Connecticut Valley area: Hartford sandy loam. . MARYLAND. Miscellaneous: Leonardtown loam (good) Leonardtown loam (poor) Muck, peat, and swamp soils: Illinois Iowa Wisconsin, Portage County Indiana, Newton County . Virginia, Portsmouth . Average for type Weight of plants grown on un- treated soil. Gain or loss attributable to- Grums. 11.2 .'■).8 8.6 8.3 y.s s.s 11.4 9.6 7. 7 9.0 11.6 .5.4 5.9 6.0 6.5 8.3 6.5 9.4 15.0 4.4 3.2 P. ct. -14 S - 3 P.ct. -12 -25 53 PN. KN. P.c/. P.ct. P.ct 3 -23 - 9 14 - 3 17 30 - 7 31 I 10 11 -13 30 25 - 3 19 -11 - 4 17 14 - 4 P.ct 4 35 7 67 31 43 14 39 PKN. L. P.cl. P.ct -20 '- 2 - 8 :-i3 22 16 46 i 10 22 I 1 100 69 7 -14 23 P. ct. - 8 - 4 38 29 M. P.ct. 4 19 29 19 -14 56 37 CvL. P.ct. 31 — 7 5 58 40 15 43 47 [ 86 '...... 69 I 5 -17 47 24 -12 45 58 39 28220— BulL 48—08- 24 FERTILITY OF SOILS AS AFFECTED BY MANURES. COMPARATIVE EFFICIENCY OF SALTS USED SEPARATELY AND IN COMBINATION. Table IV, which follows, is primarily for the purpose of showing the wide variation in the aggregate effect of the three fertilizer in- gredients when used separately as compared with their efficiency when used together. This table gives the percentage increase or decrease in growth (the latter indicated by the minus sign) produced by acid phosphate, sulphate of potash, and nitrate of soda used separately; the aggregate increase attributable to these three ingredients; the increase when the three ingredients were combined and the difference between the aggregate and that observed when the same ingredients were used in combination. All soils having complete data for these observations are included, the total number being 190, and the data are arranged in a descending series, beginning with the greatest difference in favor of the sum of the ingredients, until the difference becomes zero, after which the series is ascending and the differences are in favor of the combination. A wide varia- tion is shown in the last column, the difference ranging from a max- imum of 165 per cent in favor of the sum of the ingredients to zero and then increasing to a maximum of 72 per cent at the other extreme in favor of the combination. Of the total number of comparisons (190) there are 71 soils in wliich the efficiency of the combination exceeded the aggregate of the separate ingredients, 117 soils in which the aggregate of the individual effects exceeds that of the combination, and 2 soils in which the aggregate effect of the single ingredients exactly equals the increase in growth when those ingredients were used in com- bination. The number of soils that show very wide differences in respect to the aggregate efficiency of the ingredients as compared with the effect when the same ingredients are combined is small, the number increasing greatly as the lesser differences are approached. For example, 60 soils show differences of 9 per cent or less, while for 115 soils the difference is within the limit of 19 per cent. Within this limit of 19 per cent there are 51 soils in which the difl'erence is in favor of the combination and 62 in which it is in favor of the aggregate increase for the same ingredients used separately. (See Table V following.) As a result of averages within these limits, acid phosphate gives an increase of 0.64 per cent, sulphate of potash 6.68 per cent, and nitrate of soda 19 per cent, or a total of 26.82 per cent, as compared witli 24.5 per cent as a mean effect of the same ingre- dients used in combination on the same soils. Within these limits, which includes 60 per cent of the soils, the average efficiency of the ingredients as measured by the growth of the plants is nearly the same whether used separately or in combination. It therefore ap- COMPAKATIVE EFFICIENCY OF SALTS. 25 pears that each ingredient has performed sj)ecial functions wliich differ from and are independent of those performed by the others, whether the ingredients are appHed separately or in combination. Of the soils in which the effect of the ingredients in combination differs from the aggregate effect of the same ingredients used sepa- rately by 20 per cent or more, 55 are in favor of the ingredients used separately and 20 in favor of their combination. Within the former group of soils there are 15 instances in wliich the aggregate increase of the ingredients when used separately is more than three times as great as when the same ingredients were used in combination, and 12 instances where the aggregate is less than three times but more than two times as great as when used in combination. Within this group, therefore, the function of the several ingredients is to a considerable degree identical, at least so far as they effect an increase in the growth of plants, and one may be substituted for another to a considerable extent. Of the soils in which the efficiency of the combination exceeds that of the same ingredients used separately by 20 per cent or more there are only 20. / In tliis group the function of each ingredient is not only different from that of the others, but each is depentlent upon the presence of others in order that it may have the greatest effect. There is no apparent relation either by soil type or locality with this grouping of the soils; neither do the soils of one group appear to be particularly more responsive to the complete fertilizer than those of another. Of 190 observations on each, the individual ingredients give negative results as follows: Acid prosphate 66 times, sulphate of potash 31 times, and nitrate of soda 22 times. The combination of all three ingredients gives a negative result only 9 times, wliile the aggregate effect of the individual ingredients becomes negative 21 times. Table V shows the average increase produced by each ingredient, the aggregate increase, and the actual increase when the ingredients were combined for soils in which the aggregate increase from indi- vidual ingredients differs witliin certain limits from that obtained when the same ingredients were combined. There are ten soils in which each of the three ingredients when used separately gave a greater increase in growth than when all were com- bined and ten others in which two of the three each equaled or exceeded the gain produced by the three combined. (See Tables VI and VII.) These soils are not confined to particular localities or types, neither are they associated with high or low degree of natural fertility, as may be seen by the wide range in weight of plants on the untreated soils. 26 FERTILITY OF SOILS AS AFFECTED BY MA] Table IV. — Percentage increase in growth attributable to 1\ K, ai the same, as compared with the observed increase when a mixture was applied. [Results arranged in ilcsccniiiiig series, beginning with the grciilcst Hetiml sum of the ingredients.] 1 Weight Increase { ±1 EC i EU a \ iVi AJN U KKS. ibiitable to 1\ K, and N, and the sum of rase when a mixture of the three fertilizers dillerencc In fiivoruf the Stale. Virginia North Carolina. Mississippi. Louisiana.. Louisiana.. - North Carolina. Virginia Louisiana North Carolina Missouri Iowa Texas North Carolina. North Carolina Locality. Portsmouth . Edgecombe.. Pontotoc County, Caddo Parish Rusk County. Caddo Parish Jredcll test farm Louisa County . . Caddo Parish".... Pinehurst Crawford County Soil. Ames San Marcos New II a n o ve r County. Blantyre Illinois Manito Louisiana Caddo Parish . Louisiana Caddo Parish . Wisconsin.. Indiana. . . Mississippi. Florida . . . . Arkansas . Texas South Carolina. New York Florida Florida New York Florida Georgia North Carolina. Mississippi Arkansas . Indiana. . Georgia . Texas . . . Newton County . . Montgomery County. EscambiaCounty. Prairie County . . . San Marcos area . . York County Peat Norfolk fine sandy loam. Lufkin silt loam.. . Caddo fine sandy loam. Norfolk fine sandy loam. Norfolk fine sand . . Cecil clay Cecil loam Miller fine sand Portsmouth sand.. Clarksville slit loam. Peat Crawford silt clay. . Norfolk sand ...... Geneseo EscamliiaCounty. EscaniliiaCdiint y. Genesee County... EscamltiaCouiity. Waycross area . . . Raleigh Montgomery County. Prairie County . . . Newton County . . Waycross area ... Rusk County Georgia North Carolina. Waycross area . Tarboro Texas \ San Marcos area. Arkansas Prairie County... Arkansas Prairie County... Louisiana Caddo Parish Arkansas Prairie County... Iowa Ames Porter's sand y loam. Muck Miller fine sand . . . . Caddo fine sandy loam. Muck Marshall loam Orangeburg fine sandy loam. Norfolk fine sandy loam. Vv'averly silt loam. Houston black clay. Cecil sandy loam . . Dunkirk clay loam Norfolk sand Norfolk sand Dunkirk clay loam Norfolk sand Norfolk fine sandy loam. Cecil sandy loam.. Lintonia loam Weight of plants on un- treated soil. Arkansas \ Prairie County . . Texas San Marcos Maryland Leonardtown Georgia Waycross area Alabama -Lee County Louisiana Caddo Parish.. Acadia silt loam. . Marshall fi n e sandy loam. Norfolk sand Caddo fine sandy loam. Norfolk fine sandy loam. Norfolk fine sandy loam. Crawford silt clay. . Crowley silt loam.. Morse clay Norfolk fine sand.. Crowley silt loam.. Wisconsin drift (poor). Waverly silt loam. Blanco loam Leonardtown loam (good). Portsmouth fine sand. Norfolk sandy loam. Caddo fine sandy loam. Grams. 3.2 4.8 5.4 4,4 0.1 Increase at- ijj. ' tributable to— Sum please j j... p«^ attrib- Y'l P.Ct. P.Ct. P.Ct. 53 112 '44 42 i 63 31 3.7 i- 3 4.5 86 4.6 36 7.3 21 4.8 I 42 6. 6 26 1.5.0 9.2 4.1 4.5 9.4 5.8 6.0 7.2 7.6 7.5 5.0 7.0 7.0 8.9 5.9 5.4 10.2 7.2 7.5 4.1 8.5 7.4 6.3 5.6 8.1 3.9 5.3 7.0 11.1 5.4 8.1 7.7 0.5 8.7 8.3 5.2 7.7 5.0 ^- N, P.Ct. 234 149 137 122 110 162 158 57 110 58 71 72 200 "to'' !«"««• 'KN. I . 53 15 81 25 106 62 57 70 ino 86 27 P.Ct. 51 105 102 1 56 ^1 20 i 22 I 151 i P. ct. I 165 i 93 69 25 62 19 49 8 88 47 56 16 87 48 •87 48 60 28 72 35 71 34 65 28 82 46 47 11 40 5 56 22 105 73 41 - 4 COMPARATIVE EFFICIENCY OF SALTS. 27 Tablk IV. -Fercenkujt increase in growl h attributable to the savie, etc. — Continued. /'. A', and N, and the. sum of State. Texas , Alabama. Texas , Indiana.... Louisiana. North Carolina North Carolina Louisiana Ohio Mississippi... New Jersey. Missouri Pennsylvania... Georgia Wisconsin.. Mississippi. Mississippi. Pennsylvania. New Yorlc Louisiana Texas Texas New York. Iowa New York. Wisconsin., Virginia Arkansas.., New York. Georgia. Florida.. New York. Louisiana.. Wisconsin . Georgia North Carolina. North Carolina. Texas New York Louisiana South Carolina. Alabama Ohio , Illinois.. Wisconsin Florida South Carolina. Texas Alabama South Carolina. Wisconsin Alabama. Arkansas. Locality. Rusk County. Ford's farm... Rusk County. Tippecanoe County. Caddo Parish Biltmoro fann. Wake County.. Caddo Parish.. Soil. Westerville area. . Pontotoc County. Woodliine. Scotland County. Montgomery County. Waycross Portage County... Pontotoc County Montgomery County. M o n t g o m e r y County. B i n g h a m t o u area. Caddo Parish Rusk County. Nagadoches... Portage County.. Louisa County".. . Prairie County Binghamton areji. Waycross area Escambia County Tompkins County Caddo Parish Portage County.. Waycross area N c \v Hanover County. Statesville Rusk Countv Caddo Parish Cherokee County. Lee County Strongsville Pulaski Portage County . . Eseambi^ County. Cherokee County . San Marcos .". . Kin.g's farm York County Portage County . . Lee County , Prairie County ... Norfolk fine sand. Orangeburg clay... Norfolk fine sandy loam. Miami silt loam Norfolk fine sandy loam. Porters clay Toliacco soil Orangeburg fine sand. Miami clay loam.. . Orangeburg sandy loam. Norfolk sandy loam. Shelby silt loam Lansdale silt loam. Norfolk fine sand. Miami sand Lufkin clay Memphis silt loam. Penn silt loam Dunkirk gravelly sandy loam. Norfolk fine sandy loam. Caddo fine sandy loam. Orangeburg fine sandy loam. Volusia silt loam. . lowan drift Volusia silt loam.. Marshall sand Cecil sandy loam. . . Calhoun clay Dunkirk gravelly loam. Norfolk fine sandy loam. Norfolk fine sandy loam. Dunkirk clay loam. Norfolk fine sandy loam. Miami sandy loam Norfolk sandy loam. Portsmouth fine sand. Iredell clay loam. . . Orangeburg fine sandy loam. Volusia silt loam.. Norfolk fine sand, . Iredell clay loam.. Cecil sandy loam... Miami clay loam... Red silt loam Miami stony sand . Portsmouth sand.. Cecil silt loam Houston blackclay Orangeburg claj^ . . Iredell clay loam... Marshall gravelly loam. Norfolk coarse sand. Acadia silt loam... Weight of plants on un- treated soil. Grams. 8.5 9.1 7.5 5.6 12.4 4.7 7.4 6.1 7.6 9.6 9.9 6.0 6.0 6.4 7.0 12.3 6.6 9.6 8.2 6.6 9.0 4.8 8.4 7.2 8.3 10.6 7.9 8.4 9.0 8.7 9.0 6.0 3.0 5.2 9.2 5.4 5.3 9.1 ae 7.0 9.9 8.3 6.6 7.9 10.3 7.7 11.0 11.2 Increase at- tributable to— P. K. N. P.ct.'p.ct. P.ct. 14 16 41 22 24 42 20 28 38 30 21 28 - 3 ' 20 35 11 15 18 9 1 11 28 8 10 21 11 5 15 17-2 28 12 5 20 7 24 14 23 1 5 8 53 7 15 33 8 8 28 / 9 14 8 12 8 4 1 11 - 8 9 13 6 2 53 1 (■) 31 4 22 10 16 25 28 - 2 15 34 7 9 31 24 5 13 12 12 14 9 9 1 1 5 29 - 7 14 15 9 13 4 9 7 4 8 2 16 20 13 30 77 7 21 16 2 34 9 15 13 ir, 21 32 - 3 7 32 1 9 14 - 4 - 5 12 3 8 19 13 11 2 5 12 2 1 52 8 8 21 3 6 17 6 19 - 1 - 2 14 - 7 15 48 4 29 Sum of P,K, and N. P.ct. 71 In- crease attrib- utable to PKN. P.ct. 49 67 65 58 33 25 30 21 13 26 20 28 50 39 28 14 13 1 - 1 47 24 23 57 35 35 30 26 7 24 U 11 9 3 25 110 34 20 Dif- fer- ence. P.ct. 22 21 21 21 19 19 18 18 18 17 20 10 59 9 27 9 l.'-i 9 - (i 9 22 8 22 8 11 8 48 7 30 7 19 7 18 7 4 7 50 6 27 6 28 FERTILITY OF SOILS AS AFFECTED BY MANURES. Table IV. — Percentage increase in growth attribntable to F, K, and N, and the sum of the same, etc. — Continued. State. Locality. Soil. Weight of plants on un- treated soil. Increase at- gu^i tributable to— | of P,K, and P. K. N. N. In- crease attrib- utable to PKN. Arkansas Georgia Texas Prairie County ... Waycross area. .. Rusk County Hanover County.. Waycross area Prairie County . . . Montgomery County. Henderson County. Asheville Virginia Georgia Arkansas Pennsylvania... Tennessee North Carolina Texas Missouri Texas Crawford County. New York Indiana Arkansas Arkansas Indiana .\rkansas North Carolina. New York Ohio Tompkins County Tippecanoe County. Prairie County . . . Prairie County . . . T ippecan"oe County. Prairie County. . . New Hanover County. Tompkins County Arkansas South Carolina. North Carolina. Texas . Prairie County . Cherokee County.. States ville Galveston Texas Rusk Countv Texas Kentucky Alabama Rusk County McCracken County. Lee County Louisiana Texas... North Carolina. New York .... Georgia Arkansas Maryland Texas Alabama Texas. . Caddo Parish San Antonio States ville Tompkins County. Waycross Prairie County . . Leonardtown! Rusk County Rawles's farm San Marcos area Alabama South Carolina. Mississippi Louisiana Texas. . Lee County Cherokee County . Montgomery County. Caddo Parish San Marcos area. . South Carolina. Virginia Mississippi New York Missouri Georgia South Carolina. Texas Cherokee County. Pocahontas field . . Pontotoc County. Binghamtonare i. Crawford County. St. Matthews Palestine Morse clay Norfolk sand Orangeburg fine sand. Norfolk sandy loam. Norfolk fine sand. . Acadia silt loam.. . Hagerstown loam. Lexington silt loam. Porters clay Houston black clay Clarksville stony loam. Orangeburg clay . . Dunkirk loam Marshall loam Morse clay Crowley silt loam . . Marshall silt loam. Acadia silt loam.. . Norfolk fine sandy loam. Dunkirk clay loam Miami loam Waverly silt loam. Cecil fine sandy loam. Cecil clay (good) . . Galveston clay Susquehanna fine sandy loam. Norfolk fine sand. . Memphis silt loam. Norfolk sand Kansan till Caddo fine sandy loam. Houston black clay loam. Cecil clay (poor). .. Miami stony loam . Norfolk sandy loam. Crowley silt loam . . Leonard town loam (poor). Norfolk fine sand . . Orangeburg sandy loam. Houston black clay Norfolk sand Cecil sandy loam. . . Memphis silt loam. Orangeburg fine sandy loam. Crawford silt clay. Cecil clay Cecil clay Monroe silt loam. . . Wabash loam Wabash silt loam. . Norfolk sand Orangeljurg sandy loam. Orangeburg fine sandy loam. Grams. P.ct. 10.8 4 6.1-2 7.5-7 7.5 6.1 7.3 10.3 5.8 8.9 14.0 8.6 8.4 7.5 10.1 7.2 9.0 7.3 4.5 8.6 7.8 5.9 8.8 11.2 7.0 6.8 5.9 8.8 9.6 7.1 6.5 8.6 8.1 8.1 4.7 6.9 8.6 6.5 9.2 7.4 11.7 8.0 10.6 7.6 5.5 11.8 11.4 P.ct - 3 - 5 20 11 15 7 -2 P.ct. I P.ct. 0-2 -14 1-12 -3 - 4 ! 8 -3 3 - 8 - 7 15 5 6 6 !- 5 - 4 ;- 5 -11-1 -12 28 -13 ' 6 41 - 7 - 5 50 - 6 -23 6 25 39 59 70 5 14 26 - 2 ' 3 15 -13 60 13 P.ct. - 3 24 40 30 74 15 36 21 19 19 17 5 3 15 14 13 12 6 6 2 3 11 12 17 18 28 29 - 3 52 - 4 -20 COMPAKATIVE EFFICIENCY OF SALTS. 29 Table IV. — Pcrcentaye increase in growth attributable to P, K, and N, and the sum of the same, etc. — Continued. Pennsylvania. . North Carolina. Alabama North Carolina. Ohio Florida Illinois Louisiana. Montgomery County. Union County. . . Lee County New Hanover County. Westerville area. . Escambia County Caddo Parish. Wisconsin. Alabama. . Portage County. 1,( e County Iowa Georgia Mississippi. Indiana Mississippi. New Jersey Georgia Virginia Alabama Georgia j... North Carolina. Wisconsin.. Mississippi. Waycross Montgomery County. Newton County . . Mont gome ry County. Woodbine Alabama. . . Mississippi. Georgia Mississippi. Texas Louisiana. Waycross Hanover County . . Marion Waycross Chowan County.. Portage County. . M o n t go me r y County. Lee County Pontotoc County. Waycross Montgomery County. Rusk County Chester loam. Caddo Parish. Indiana South Carolina. Louisiana Newton County York County. .. Caddo Parish. .. Rhode Island.. Kingston. Silt loam Norfolk sandy loam. Norfolk fine sand . . Miami black clay loam. Norfolk fine sandy loam. Marion silt loam. . . Orangeburg fine sandy loam. Portage silt loam. . Norfolk c Q a r s e sand. Wisconsin drift (good ) . Portsmouth fine sand. Memphis silt loam. Clyde fine sand Memphis silt loam. Norfolk fine sandy loam. Norfolk sandy !oani. Wickham sandy loam. Orangeburg sandy loam. Portsmouth fine sand. Portsmouth silt loam. Portage silt loam. . Memphis silt loam. Cecil sandy loam. . . Orangeburg clay. . . Norfolk fine sand. . Memphis silt loam. Orangeburg fine sandy loam. Norfolk fine sandy loam. Muck Cecil sand Orangeburg fine sand. Miami silt loam Weight Increase at- of . tributable to— plants I on an- ; treated I soil. Grams. 7.8 4.6 6.1 6.4 8.3 5.5 7.7 5.0 8.7 6.4 9.6 5.5 10.6 12.6 9.5 6.5 7.0 7.5 4.8 5.2 4.5 10.0 10.2 5.6 6.6 7.4 11.2 9.2 7.9 4.4 3.9 5.0 P.ct. - 4 -14 3 -11 10 -13 -16 15 18 - 3 - 2 - 9 1 14 5 15 19 - 9 - 2 -11 - 7 - 9 3 1-9-8 -10 - 7 -12 - 6 '. 1-5 -25 Gl -15 - 8 -22 -10 10 2 -13 -21 -16 - 3 14 1 -11 - 3 K. 3.6 - 8 -19 N. P.ct. 4 32 42 37 - 2 - 5 - 3 10 - 2 20 30 33 - 2 -14 9 36 11 25 50 14 - 6 -15 29 18 - 3 -11 14 -10 14 26 -10 S;^ crea'se ■p% 'attrib- and' "table N- I PKN. P.ct. 13 28 32 49 - 7 17 -24 16 -16 8 35 53 -18 -37 -25 11 43 13 46 66 35 -26 -20 11 12 -20 -28 - 4 -14 50 3 -42 P.ct. 27 32 1 25 52 71 1 -17 - 5 32 64 36 69 90 62 3 12 43 45 14 8 36 35 100 54 28 Dif- fer- ence. P.ct. 14 14 14 14 15 15 16 16 17 17 17 18 19 20 20 21 21 23 23 24 27 29 32 32 33 34 36 40 49 50 51 70 80 FERTTLTTY OF SOILS AS AFFECTED BY MANURES. Table V. — Average percentage increase in growth attributable to P, K, and N and the aggregate increase of the same salts used individually as compared with the actual increase when the same salts were used together, averages being for the soils in ivhich the difference between the aggregate and actual increase is within limits indicated in the first column. Range of differ- ence. Number of soils. Green weight of plants. Increase attributable to— Sum of P, K, and N. Increase attribu- table to mixture of PKN. Differ- ence. P. K. N. Per cent. 60+ 50-59 40-49 30-39 20-29 10-19 0-9 9-0 19-10 29-20 39-30 49-40 50+ 5 8 7 16 19 35 28 .32 20 9 5 2 4 Grams. 4.8 7.0 6.4 6.7 7.3 7.6 8.3 7.6 7.9 7.5 8.2 8.6 4.2 Per cent. 37 31 19 16 13 6 1 :! - 8 - 6 -18 Per cent. 49 31 30 22 19 12 5 4 4 4 -5 -5 6 Per cent. 64 38 35 31 38 22 22 17 14 14 4 2 4 Per cent. 150 100 84 69 70 40 28 19 13 14 -9 -9 -8 Per cent. 52 46 41 31 46 20 23 23 28 37 24 36 53 Per cent. 98 54 43 35 24 14 5 - 4 -15 -23 ■ -33 -45 -61 Table VI. — Ten. soils in which percentage increase attributable to each ingredient is- greater than that produced when all are combined. State. Locality. Soil type. Weight of plants grown on un- treated soil. Increase attributable to— Sum of PKN. In- crease attrib- utable to mix- ture of PKN. Dif- fer- ence. P. K. N. Louisiana Caddo Parish MiUer fine sand . . . Peat Grams. 7.3 15.0 6.6 7.6 5.4 10.2 6.3 11.1 8.1 6.6 P.ct. 21 24 26 17 16 11 5 5 9 4 P.ct. 26 26 15 22 18 11 12 13 P.ct. 10 21 17 17 13 18 9 -.3 2 11 P.ct. 57 71 58 56 47 40 26 15 25 16 P.ct. 1 20 P.ct. 56 51 Missouri Indiana Florida Crawford County. Newton County . . Escambia County. Geneseo Newton County . . Prairie County ... Prairie County . . . Binghamton area. Clarksville silt loam. Marshall loam Norfolk sand Dunkirk clay loam. Marshall fine sandy loam. Morse clay Crowley silt loam. Dunkirk gravelly sandy loam. 7 1 51 16 i 40 11 36 New York Indiana Arkansas Arkansas New York 5 - 4 -11 1 35 30 26 • 25 15 Average 8.4 14 16 11 41 5 36 NITRATE OF SODA ALONE AND WITH OTHER SALTS, 31 Table VII. Tni sails in luhich two of the three ingredients each produced a greater increase than when all three were combined. NITRATE OF SODA ALONE AND WITH OTHER SALTS. A study of Table IV shows that in 66 per cent of the instances nitrate of soda when used ah)ne has produced an increase of growth which equals or exceeds that produced by either sulphate of potash or acid phosphate. A tabulation of the data for nitrate of soda and of combinations v^ which it enters shows that the nitrate of soda has a marked bene- ficial effect on the majority of soils which on an average is not increased by the addition of acid phosphate. A comparison of the nitrate with the nitrate-phosphate column shows a difference in favor of the former in slightly more than half of the comparisons. The average of these two columns shows a difference of less than one- tenth of 1 per cent. Nitrate of soda when supplemented with sulphate of potash gives an increase in growth which in 74 per cent of the instances equals or exceeds that produced by nitrate of soda alone, the average result being 8.5 per cent in favor of the combination. When this nitrogen- potash combination is still further supplemented by acid phosphate there is on an average no additional increase in growth. In fact tlie instances in which the nitrogen-potash combination equals or exceeds that of the complete fertilizer are slightly in the majority. The tabulation above mentioned is not published, but the same facts may be gathered by careful study of those columns in Table III, which give the results for nitrogen and all combinations into which 'it enters. 32 FEETILITY OF SOILS AS AFFECTED BY MANURES. SULPHATE OF POTASH ALONE AND WITH OTHER SALTS. Sulphate of potash when supplemented by acid phosphate pro- duced an increase in growth, which in 50 per cent of the soils exceeded that produced by potash alone, the average increase being one- tenth of 1 per cent in favor of the combination. When the potash salt is combined with nitrate of soda there is a marked increase in growth over that produced by potash alone, which, on the average, is not further increased by the addition of acid phosphate. In 86 per cent of the soils the effect of potash combined with nitrogen has equaled or exceeded the effect of potash alone. These facts are brought out by a study of the columns in Table III, in which potash occurs, omitting all soils for which the data for this salt and all its combinations f),re incomplete. ACID PHOSPHATE ALONE AND WITH OTHER SALTS. Referring to Table IV, the striking fact is the large number of instances in which acid phosphate gives a small or negative effect. The efficiency is quite regularly increased (see all columns of Table III in which ])hosphate enters) as each fertilizer salt is added, the exceptions to this being mostly for those soils in which the efficiency of the complete fertilizer as a whole becomes quite small or is negative. In the life and economy of the plant, phosphates are generally recognized to be instrumental chiefly in the production of seed or fruit, and have a tendency to hasten maturity, and may influence the color and quahty of the fruit. It is also known that plants do not absorb extraneous phosphates during the very early stages of growth. From the standpoint of the requirements of the plant, no response would be expected from the phosphates, and it is, there- fore, not strange that in many instances a small or negative effect has resulted from their application. It should be borne in mind, however, that this is not a study of the plant's requirements, but a study of the fertihty of the soil as affected by fertiUzers and that the plants have been used merely as the indicator of the degree to which fertihty has been affected by the various applications. It is now contended that aside from the direct action that various salts have on the growth and economy of the plant, they may also act directly on the soil, thereby increasing its fertility in a way not fully under- stood. An inspection of the tables will show that while small and negative effects have often resulted from the phosphate, there are 55 soils in which the increase in growth attributable to this salt when used alone has ranged from 10 to 50 per cent and in two instances even^ more. LIME ALONE AND WITH FERTILIZER S\LTS. 33 LIME ALONE AND WITH FERTILIZER SALTS. Table VIII shows the increase in growth attributable to lime, to a complete fertilizer with and mthout lime, to manure, and to cowpea vines with lime. The soils are arranged in a descending series as determined by the efficiency of the complete fertilizer without lime. Lime alone shows quite a marked effect in the majority of the soils, and while its efficiency corresponds in a general way with that of a complete fertilizer there are numerous exceptions to this. The hme may have little or no effect where the fertilizer is markedly beneficial. For example, in the upper portion of the table where the figures for the complete fertilizer range from 50 to 75 per cent increase in growth there are 8 soils on which lime gives less than 10 per cent increase, while in the lower portion of the table where the complete fertilizer does not exceed 1 1 per cent there are 7 instances where the increase in growth attributable to lime exceeds 30 per cent. As a rule, the efficiency of the complete fertilizer has been appreciably increased by the addition of lime. There seems to be little evidence in support of the contention sometimes made that the benefits due to lime are attributable to its favorable action on nitrification. A study of Table IX, which summarizes the results of Table VIII and all data for the same soils in Table III — i. e., gives the mean per- centage increase in growth attributable to the several fertilizer salts and combinations for groups of soils in which the efficiency of a complete fertilizer falls within certain limits as given in the second column — shows that as the efficiency of a complete fertilizer declines there is a corresponding decline in the effect of practically all ingre- dients and combinations. This is equally as true for manure or for potash and phosphate as it is for nitrate of soda and lime, so that the parallel decline in the lime and nitrate columns in the table loses significance. Of the soils in Table VIII, where the efficiency of a com- plete fertilizer is not less than 25 nor more than 75 per cent, there are 17 in which the efficiency of lime exceeds 35 per cent and 31 in which it is 10 per cent or less. As a result of aVerages of these two groups of soils, we get the following results: Seventeen soils (lime giving increase of 35 per cent or more) average efficiency, lime 63, N 35, NKP 49; 31 soils (lime giving increase of 10 per cent or less) lime 4, N 24, NKP 40. The effect 6i nitrogen is obtained from Table III. The difference in the efficiency of lime is here very marked, drop- ping from an average of 63 per cent for the 17 soils to only 4 per cent for the 31 soils. The relative decline in the effect of nitrate of soda and the complete fertilizer, however, is small, being from 35 to 24 for the former and 49 to 40 for the latter. 34 FERTILITY OF SOILS AS AFFECTED BY MANURES. Table VIII. — Weight of plants on the untreated soil and the percentage incrcme ingrotvth produced by lime, a complete fertilizer with lime, a complete fertilizer, mavvre, and cow- pea vines with lime. Looality. Soil. Weight of plants, un- treated soil. Increase attributable to — NPKL. NPK. M. CvL. North Carolina. North Carolina. North Carolina. Virginia Indiana , . . . Georgia., Georgia. Georgia. Iowa New Hanover County. New Hanover County. Iredell test farm... Louisa County Newton County. . Waycross area.. . Waycross area Waycross area... Ames Georgia , Waycross area... Georgia. Georgia. Iowa Georgia. Georgia. Waycross area. Waycross area. Leon Waycross area. Waycross area. Virginia. . Alabama., Texas Georgia North Carolina. . North Carolina. Alabama North Carolina. . Marion . Portsmouth.. Marion Rusk County. Waycross area New Hanover County.. Tarboro , Lee County Chowan County. Alabama Lee County. Louisiana., Indiana. . . South Carolina. Georgia Iowa Nortli Carolina. . North Carohna. .| Louisiana South Carolina...' South Carolina... Caddo Parish Tippecanoe County. Cherokee County. Waycross area..". Blackhawk County. Edgecombe Pinehurst Caddo Parish York County. — Cherokee County. Iowa Ames. Louisiana... Alabama... Mississippi. Georgia North Carolina. Texas South Carolina.. Florida Mississippi. Texas Texas Wisconsin. Caddo Parish. . Lee County Pontotoc area . Waycross area. Raleigh Rusk County Cherokee County.. Escambia County Montgomery County. Rusk County Norfolk sand. Portsmouth fine sand. Cecil clay , Cecil loam Peat , Rusk County Portage County . . Portsmouth fine sand. Norfolk fine sandy loam. Norfolk sandy loam Wisconsin drift (poor). Portsmouth fine sand. Norfolk fine sand Norfolk fine sandy loam. Kansan till Norfolk sand Portsmouth fine sand. Orangeburg sandy loam. Peat '. Orangeburg clay . . . Norfolk fine sandy loam. Norfolk sandy loam. Norfolk fine sand... . Norfolk fine sandy loam. Norfolk sandy loam. Portsmouth silt loam. Norfolk sand Norfolk fine sand Miami silt loam.. Cecil sandy loam. Norfolk sand lowan drift Norfolk fine sandy loam. Portsmouth sand. . Norfolk fine sand . . . Cecil sand Cecil fine sandy loam. Wisconsin drift (good). Norfolk fine sand . . Norfolk coarse sand, lyufkin silt loam Norfolk fine sand . . Cecil sandy loam Ndrf oik fine sand . . Cecil silt loam Norfolk fine sandy loam. Orangeburg fine sandy loam. Norfolk fine sandy loam. Caddo fine sandy loam. Muck Grams. 4.1 3.0 4.5 46 4.4 5.2 3.9 6.5 7.7 G. L 7.5 5.9 5.5 5.5 3.2 9.1 7.5 7.0 6.4 3.0 7.7 4.5 5.3 5.6 6.9 5.6 9.0 4.8 5.4 3.9 5.9 3.7 5.3 5.4 6.0 4.1 8.5 6.6 3.3 7.5 6.1 8.2 7.2 P.ct. 66 70 45 71 14 92 64 35 17 29 34 40 6 64 20 26 39 8 29 - 5 41 5 63 27 52 21 39 136 24 40 7 23 P.ct. 154 173 105 153 70 153 105 100 84 89 42 109 75 69 97 55 64 57 70 70 87 57 103 74 35 65 121 113 70 56 32 103 55 56 130 64 41 47 41 P.ct. 151 110 105 102 100 90 86 85 82 P. ct. P. ct. 254 { 144 63 219 150 109 86 150 102 113 24 124 64 37 47 49 105 103 94 38 61 46 71 114 146 59 37 113 53 86 81 90 60 48 61 51 LIME ALONE AND WITH FERTILIZER SALTS, 35 Table VIII. — Wright of plants on the untreated soil and the percentaije increase in groivth produced by lime, etc. — Continued. State. Locality. Florida .Mabama. . Mississippi .Mabama. . . New York. North Carolina. Kentucky ...... Texas Arkansas. Texas Wisconsin. .\rkansas.. Virginia. .. Tennessee. Louisiana. Texas. Texas. Escambia County Lee County Pontotoc Coimty. L<^e County Tompkins Coun- ty. Union County. .. . MeCracken Coun- ty. San Marcos ar<'a. . Prairie County .. . Rusk County Portage County Prairie County .. Louisa County... Henderson Coun- ty. Caddo Parish Norfolk sand Norfolk sandy loam. Orangel)urg clay .. Cecil .sandy loam . . Dunkirk clay loam. Silt loam Memphis silt loam. Rusk County . Rusk County. Wisconsin i Portage County New York . . South Carolina .: ChercJ^ee County. Alabama j Lee County ..'..,. . Louisiana ' Caddo Parish Texas North Carolina . South Carolina. Texas Louisiana . Texas. . Florida. Louisiana. New Jersey Rhode Island... Virginia Virginia North Carolina. Texas North Carolina . Missouri.. . New York. Mississippi Louisiana . .\rkansas Texas Pennsylvania... South Carolina. .\rkansas Arkansas South Carolina . New York. Mississippi. Arkansas.. Texas North Carolina.. San Marcos area. Statesville York County Rusk County Caddo Parish. San Marcos area. . Escambia Coun- ty. Caddo Parish Woodbine, Kingston Louisa County Hanover County.. Wake County San Marcos area. . New Hanover County. Scotland County.. Geneseo Pontotoc area . . . . Caddo Parish Prairie County .. Rusk County Montgomery County. Cherokee County Prairie County . . Prairie County . . St. Matthews Bingham ton County. Pontotoc County. Prairie County . . . Rusk County Biltmore. Crawford silt clay . . Acadia silt loam" . . Orangeburg fine sand. Miami sand Waverly silt loam. Wickham sandy loam. Lexington silt loam. Caddo fine sandy loam. Norfolk fine sand., Orangeburg fine sandy loam. Marshall sand Volusia silt loam. . . Cecil clay. Norfolk sand Norfolk fine sandy loam. Houston black clay. Iredell clay loam Cecil sandy loam Caddo flric sandy loam. Norfolk fine sandy loam. lilanco loam Norfolk fine sandy loam. Orangelnirg fine sandy loam. Norfolk fine sandy loam. Miami silt loam Cecil sandy loam... Norfolk sandy loam Tobacco soil Houston black clay. Norfolk fine sandy loam. Shelby silt loam Dunkirk clay loam . Lufkin clay Orangeburg fine sand. Waverly silt loam. . Norfolk fine sand. . . Chester loam I redeU clay loam Crowley silt loam..-. Acadia silt loam Orangeburg sandy loam. Wabash loam Orangeburg sandy loam. Calhoun clay Orangeburg fine sandy loam. Porters clay Weight Of plants, un- treated soil. Grams. 5.9 ti. I 4.0 5.0 Increase attributable to- 8.8 7.4 7. .5 6.0 6.5 7.5 8. (i 9.2 8.4 4.8 7.4 4.7 7.0 7.0 5.2 7.0 8.1 3.() 7.2 7.5 4.7 7.9 9.6 8.9 6.4 5.0 5.0 7.8 7.8 9.1 7.0 11.2 11.8 10.6 7.6 8.3 9.2 12.4 I L. NPKL. NPK. M. CvL. P.ci. P.ct. P.ct. P.ct. P.ct. 24 55 46 94 97 11 64 46 42 104 - ;i 47 45 IS 32 18 52 43 43 ■ 29 42 42 33 28 54 4 51 18 48 42 70 2 24 46 42 56 11 23 48 41 31 38 8 60 40 80 68 7 43 39 SO - 4 61 .55 38 38 54 9 35 36 31 -12 5 57 36 72 17 18 52 36 20 186 10 26 36 61 37 13 39 36 68 53 11 8 35 34 3 15 62 35 81 4 44 35 42 12 50 135 35 108 143 2 56 3!> 18 47 17 42 35 10 38 - 1 28 34 33 33 41 34 76 58 17 36 34 62 40 28 54 33 54 . 52 20 21 33 28 37 45 82 32 32 85 44 32 34 96 7 31 32 28 42 133 111 31 39 5 39 30 25 57 - 1 47 30 44 - 8 11 16 21 30 30 15 30 " 42 8 31 29 57 34 11 28 28 69 30 19 22 28 17 28 6 31 28 9 48 12 15 28 114 124 104 150 28 10 132 3 60 28 103 69 4 33 27 13 8 14 42 27 36 7 43 71 27 64 80 12 33 27 46 27 19 28 26 25 17 10 15 26 31 10 6 39 26 25 24 18 27 26 4 13 12 37 26 37 66 31 26 25 61 24 36 FERTILITY OF SOILS AS AFFECTED BY MANURES. Table VIII. — Weight of plants on the untreated soil and the percentage increase in groivth produced by lime, etc. — ^Continued. State. Locality. Georgia Alabama Illinois North Carolina. Georgia Georgia Texas. New York. Texas Wisconsin. Florida Mississippi. Texas Illinois Louisiana. New York.. New Jersey. Iowa North Carohna. Alabama Louisiana Texas South Carolina. Arkansas Texas Indiana Alabama Louisiana Arkansas.. Georgia Mississippi. Texas Texas Mississippi. Ohio Pennsylvania. . . Mississippi Indiana New York . Alabama. . Mississippi. New York. New York. Florida Florida Florida Texas. Pennsylvania. Louisiana Louisiana Texas Ohio Mississippi. . . Wayeross area . . Lee County Manito Blantyre Wayeross area . . Wayeross area . . Nacogdoches San Marcos area. . Portage County .. Escamliia County Montgomery County. San Marcos area.. Pulaski County. . . Caddo Parish Ithaca Woodbine. Ames , Asheville Perry County . . . Caddo Parish San Marcos area. York County Prairie County . . San Marcos area. Newton County . Lee County Caddo Parish Prairie County. .. Wayeross area . . . Pontotoc County. Nacogdoches Palestine Montgomery County. Westville area Montgomery County. Montgomery County. Tip pecanoe County. Tompkins County Perry County Montgomery County. Tompkins Tompkins Escaml)ia Escaml)ia Escambia County County County County County San Antonio . Montgomery County. Caddo Parish Caddo Parish Rusk County Westerville area . . Montgomery County. Norfolk sandy loam Norfolk coarse sand Muck Porters sandy loam. Norfolk sand Norfolk fine sandy loam. Orangeburg fine sandy loam. Volusia silt loam. . . Crawford silt clay .. Miami stony sand. . Norfolk sand Lintonia loam Crawford silt clay .. Red silt loam Orangeburg fine sand. Volusia silt loam. . . Norfolk sandy loam Muck Porters clay Orangeburg clay . . . Miller fine sand Houston lilack clay. Iredell clay loam . . . Acadia silt loam Houston l)lackclay. Marshall loam Cecil sandy loam . . . Orangeburg fine sandy loam. Acadia silt loam Norfolk fine sand. . . Monroe silt loam . . . Orangeburg clay . . . Orangeburg fine sandy loam. Memphis silt loam.. Miami clay loam Penn silt loam Memphis silt loam. . Marshall silt loam . . Dunkirk loam Orangeburg sandy loam. Memphis silt loam . . Weight of plants, un- treated soil. Dunkirk clay loam.. Miami stony loam.. Portsmouth sand . . Norfolk sand Norfolk fine sandy loam. Houston black clay loam. Hagerstown loam . . Caddo fine sandy loam. Norfolk fine sandy loam. Susquehanna fine sandy loam. Miami " l)lack clay loam. Memphis silt loam. . Grams. 5.0 6.4 9.4 4.5 6.1 7.9 6.6 9.2 9.9 7.2 8.5 9.2 7.0 7.4 5.4 6.0 15.0 5.8 10.3 5.8 8.1 7.7 9.0 8.9 7.6 8.6 6.5 7.3 7.4 8.0 6.1 12.3 8.4 8.1 9.0 9.6 8.3 5.4 8.4 8.7 7.0 8.3 11.2 Increase attributable to — L. NPKL. P. ct. P.ct. - 9 43 22 23 / 33 43 22 34 55 22 35 16 24 8 15 4 31 - 3 19 29 23 24 16 20 42 16 38 20 28 9 32 2 17 -14 10 53 16 35 22 34 6 26 3 17 -11 41 5 23 22 21 10 28 7 -15 14 37 6 25 9 18 4 5 -15 9 13 17 7 11 - 8 4 6 9 - 7 6 - 7 11 21 17 1 20 -14 12 7 48 48 46 74 35 14 11 19 6 23 8 14 30 -U 26 1 11 -3 NPK. M. CvL. P.ct. P.ct. P.ct. 25 54 42 25 33 41 25 45 51 24 61 31 24 77 65 24 96 67 24 56 15 23 77 13 23 14 34 22 47 8 22 66 65 22 30 48 22 40 52 22 ■ 26 45 21 15 28 20 51 29 20 25 14 20 64 19 9 16 19 32 26 19 81 48 19 25 32 18 51 19 18 32 21 17 8 23 16 77 34 15 58 15 20 26 15 7 16 14 28 36 14 29 39 14 29 5 14 37 43 14 36 40 13 15 45 13 24 36 13 6 34 12 -2 4 12 46 12 17 20 12 18 28 11 68 - 5 11 36 26 11 47 35 11 83 107 11 32 51 10 19 - 7 9 20 29 9 6 41 9 9 25 9 29 36 8 22 5 8 5 9 EFFICIENCY OF ORGANIC AND CHEMICAL MANURES. 37 Table VIII. — Weight of plants on the untreated soil and the percentage increase in growth produced by lime, etc. — Continued. Locality. Soil. Weight plants, Increase attributable to - State. 1 treated L. NPKL. NPK. M. CvL. soil. Grams. P.ct. P.ct. P.ct. P.ct. P.ct. Louisiana Caddo Parish Caddo fine saiidy loam. 6.0 19 16 8 79 23 Crawford County. Montgomery Clarksville siP. loam Lansdale silt loam.. 6.6 9 9 14 . 19 1 17 7 43 9 32 Pennsylvania 30 County. New York Bingham ton area. Dunkirk grayelly loam. 10.6 - 3 -11 7 20 - 9 Arkansas Prairie County... Crowley silt loam. . . 7.1 36' 60 7 19 38 Indiana Tippecanoe County. Marshall loam 7.5 - 3 7 6 42 - 2 Dunkirk cla / loam. . Marshall gravelly lo-m. Caddo fine sandy 10.2 11.0 8 2 3 5 4 17 36 21 Wisconsin Portage County.. 14 Louisiana Caddo Parish 5.0 8 72 4 44 148 - loam. Wisconsin Portage County.. Miami sandr loam. . 9.0 4 21 3 29 12 Wisconsin Portage County. . Portage silt loam... 10.0 5 10 3 21 10 Missouri Crawlord County. Clarksville stony loam. 14.6 - 3 - 2 3 -10 13 Arkansas Prairie County. . . Crowley silt loam.. . 10.1 2 5 3 7 9 Wisconsin Portage County. . Portage silt loam... 8.7 9 -14 1 42 -14 New York Binghamton area. DunlH -1 to - 20 }-' 2 -7 2 7 4 21 11 9.3 8.7 11.2 RELATIVE EFFICIENCY OF SALTS WHEN USED ALONE AND IN COMBINATION. In these tests as shown in Table III, beginning on page 15, each of the three salts, acid phosphate, sulphate of potash, and nitrate of soda, has been used separately and in three combinations, thus giving four observations relative to the increase in growth pro- duced by each. For lime there are two observations, while for manure and cowpea vines and lime there is only one observation each. While it is disputed ground as to whether the efficiency of a fertilizer salt should be measured by using it alone or by combining it with all other elements in which the soil may be deficient, many experimenters have adopted the combination as the most reliable means. It seems probable, however, that theoretical considerations, rather than observed results, have led to the adoption of such a system, the idea being that if the soil is deficient in two or more 40 FERTILITY OF SOILS AS AFFECTED BY MANURES. ingredients the full benefits of any one of them can not be obtained in the absence of a sufficiency of the others. While such reasoning seems logical it is not always supported by facts, as may be seen by a study of Table IV. Here there are only 20 soils out of 190 in which this seems to be noticeably true, while there is a much larger number, viz, 50 soils, in which the efficiency of the ingredients used separately is much more marked than when used in combination, and this in spite of the fact that all three of them produce quite marked results. Indeed, when we calculate the average efficiency of the ingredients for all soils we find that each is slightly more efficient when used alone than when in combination with one or more other ingredients. As an average of many field tests on various crops and in numerous localities the conclusion is that the efficiency of a fertilizer ingredient is about equally as marked whether used alone or in combination with one or more other ingredients. On the less productive soils the tendency is toward a better effect from the ingredient when in com- bination with other ingredients which are also beneficial. On better soils the tendency is in the other direction." For many individual soils this tendency is as marked under field conditions as it is in the pots. Numerous bulletins of the Alabama agricultural experiment station report the results of fertilizer tests tliat were conducted at intervals during the years 1889 to 1900. Of such tests 156 were cooperative experiments with cotton and were carried on in 53 of the 67 counties of the State. The fertilizer materials used per acre consisted of nitro- gen derived from 90 pounds of sulphate of ammonia or 96 pounds of nitrate of soda or 200 pounds of cotton-seed meal, potash derived from 150 or 200 pounds of kainit or 64 pounds of muriate of potash, and phosphorus derived from 195 or 200 pounds of dissolved bone- black or 240 pounds of acid phosphate. The three classes of fertilizer were used separately and the three were also combined, each ingre- dient entering into the combination in the same amount as when used alone. The results are reported in pounds of seed cotton per acre over and above that produced when no fertilizer was used, and the data for each of the three fertilizers when used alone and in com- bination are complete for 134 tests. The aggregate increase attrib- uted to the three fertilizers when used separately was 443.2 pounds of seed cotton per acre, as compared with an average of 444.7 pounds increase per acre when the same ingredients were combined. Table X shows the percentage increase in growth attributable to each ingredient when used separately, as compared with the calcu- " See tabulated results on corn in ' Results in Field Experiments with Various Fertilizers," by Prof. W. O. Atwater, Ph. D., U. S. Dept. Agr., 1883. EFFICIENCY OF SALTS ALONE AND IN COMBINATION. 41 lated increase where it occurs in the several combinations. The results are for each of the four ingredients, lime, nitrate, potash, and phosphate, and are the averages for each of the 25 areas given in Table III. As an illustration of how the results are calculated, take the average results of the five soils from Cherokee County, S. C. Nitrate alone gave an increase of 40 per cent. Nitrate and phosphate pro- duced an increase of 36 per cent, or 40 per cent more than phosphate alone, while nitrate and potash produced 49 per cent, or 42 per cent more than potash alone. The average effect for nitrogen in these two combinations is therefore 41 per cent and is entered in the second column under nitrogen. As NP = 36 and NK = 49, by addition (2N)PK = 8o. However, PK = 3, so by difference we have 2N = 82 or N = 41, which is entered in the third column imder nitrogen. NPK = 44 and since PK = 3, N in the NPK combination produced an increase of 41 per cent. The four values each for potash and phos- phate and the two values for lime are obtained in the same manner. For this area the efhciency of the several ingredients is very uniform, the range of variation for nitrate, potash, and phosphate being only 1 per cent, while for lime the difference between the two observations is 7 per cent. Since the results in Table X are based upon the average of all soils from each area the variation in the efficiency of the ingre- dients as they occur in the various combinations is not marked. By making the same calculations for each of the soils, however, many of them show a marked variation in the efficiency of the ingredients as they occur in the several combinations, a difference far too great to be attributable to error of observation, which by actual test has been found not to exceed plus or minus 5 per cent. Where the variation in the efficiency of a fertilizer ingredient, as it occurs alone and in several combinations, exceeds the limit of error in observations as above stated, such portion of the variation as exceeds that limit must be attributable to its association witli the other ingredients. Just as in a solution of several salts the ])resence of one may increase or decrease the solubility of others, so in the soil the addition of one ingredient may increase or decrease the effectiveness of another. It has already been shown that for quite a number of soils the effect of three ingredients combined is no better than for one of them and that one ingredient may be substituted for another witli equally good effect. See Table VI, page 30. On the other hand, there are many soils in which the effect attributable to a combination of tliree ferti- lizer ingredients is two or three times greater than the aggregate effect of the same ingredients when used separately. See latter portion of Table IV, page 26. The variation in the efficiency of a fertilizer ingredient as used sep- arately and in several combinations bears no consistent relation to the 42 FERTILITY OF SOILS AS AFFECTED BY MANURES, efficiency of the ingredients with which it is associated or to the effi- ciency of the combination as a whole. If nitrate of soda produces an increase in growth which equals or exceeds that produced when it is associated with potash and phosphate, as is frequently shown in Table IV, pa;ge 26, we would theoretically expect little or no effect from either potash or phosphate when used alone on the same soils. Con- trary to this theory, however, we find that, with an occasional excep- tion, both potash and phosphate are markedly efficient on these soils. On the other hand, if nitrate of soda produces little or no increase as compared with a fair to good increase obtained when it is combined with potash and phosphate,"^s is frequently shown in the latter por- tion of Table IV, we should expect to obtain fair to»good results from potash or phosphate when used alone on the same soils. But what do we find? Usually little or no effect from. potash, and a negative effect from the phosphate. Table V, page 30, which is a condensation of the results in Table IV, shows up this relationship in a striking man- ner. In that table is given the average percentage increase in growth attributable to each of three ingredients and their combination for groups of soils in which the aggregate effect of the ingredients dif- fers within certain limits from that obtained when they are used in combination. It will be noticed that there is a more or less regular and parallel decline in the efficiency of each of the ingredients when used separately, from a maximum of 37, 49, and 64 per cent for phosphate, potash, and nitrate to a minimum of — 18, — 5, and 2 for the same ingredients, respectively. Notwithstanding this marked de- cline in the efficiency of the ingredients when used separately, the efficiency for the three ingredients combined is essentially as marked at the end of the series as at the beginning, the average effect for the first and last groups being 52 and 53 per cent, respectively. The aggregate effect of the three ingredients used separately is 150 and — 8 per cent for the same groups. (See Table V, p. 30.) EFFICIENCY OF SALTS ALONE AND IN COMBINATION, 43 Table X. Percentage increase in growth attributable to each fertUizcr salt and lime, when used aloiu and when used in'various combinations, as calculated by the difference method. Average, by areas. State. Locality. Indiana Newton County. . . Missouri Crawford County . . Ohio Westerville area. . . New York I Tompkins County . Wisconsin. New York Mississippi Pennsylvania . . . Indiana Missouri Virginia Texas Arkansas Mississippi South Carolina. , Louisiana Texas Tennessee Kentucky South Carolina. , Alabama Florida Georgia Virginia North Carolina . Portage County. Binghamton area Montgomery County. . Montgomery County. . Tippecanoe Coimty... . Scotland County HanoverCounty San Marcos area Prairie County Pontotoc County York County Caddo Parish Rusk County Henderson County McCracken County Cherokee County Lcc County Escambia County Waycross area Louisa County New Hanover County. State. Locality. Indiana Newton County . . . Missouri ' Crawford County . . Ohio I Westerville area.. . New York Tompkins County . Wisconsin 1 Portage County. New York Mississippi Pennsylvania. . . Indiana Missouri Virginia Texas Arkansas Mississippi South Carolina. Louisiana Texas Tennessee Kentucky South Carolina. Alabama Florida Georgia Virginia North Carolina . Binghamton area Montgomery County. . do Tippecanoe County Scotland County Hanover County San Marcos area Prairie County Pontotoc County York County Caddo Parish Rusk County Henderson County McCracken County Cherokee County." Lee County Escambia County Waycross area Louisa County New Hanover County. Increase with lime. P. cl. 6 - 3 3 - 9 _ 2 - 1 4 - 9 8 4 26 6 8 21 7 21 6 11 28 35 21- 30 19 Increase with phosphoric acid. P. rl. % 1 iz; !5| wl + M 1 W Ph PL, + Ph . ci. -6 2J 4 -9 -2J 1 -1 «i 94 I I -u -3 -ti -1 j -8 I -4i' -4 lu^ -l' 1 7 P. ct. - 6 4 4 - 8 - 3i - 4 4 9 3 2 P. ct. -17 - 4 - 4 - 5 _ 2 - 5 - 4 - 5 3 - 4 7 8 - 1 6 10 1 - 8 - 5 2 - 1 Increase with potash. P.ct. 7i 1 5i 2i 3 .5 14 4 KiJ 9 li 3 P.ct. 10 1 -1 () a 5"; 5 14 IS 10 (i ' Increase with nitrogen. 6 13 9 li 11 t) 1(5 7 13 8 8 5i 7* 12 14 Ki 9 24 17^ 22 P.ct. - 1 - 7 - 9 9 8 5 ,5 - () U 19 5i 10 30 I - 2 i 10 20 7 8 14 6 17 22 19 P.ct. 4 4 4 21 12 6 8 4 14 14 10 23 12 28 24 24 32 36 39 40 35 20 45 44 65 P.ct. 1§ 2i lOi 17i 9| 8 12J 6J 17i 9 16 11 . 54 14i 21 22J 24 29 33i 41' 32i 10' 42 26 58i P.ct. 5 6 10 18 lOi 12 13 5 21 13 16 H 9 20 12 25 26 28 30 41 32 15 42 P. ct. - 6 _ 2 2 22 12 7 14 — 5 15 6 25 14 6 23 28 14 25 31 34 41 39 10 43 38 57 44 FERTILITY OF SOILS AS AFFECTED BY MANURES. RELATION OF FERTILIZER REQUIREMENTS TO CHARACTER OF SOILS. That soils vary greatly in their crop-producing capacity and in the degree to wliich they respond to fertilizers under field conditions has long been known as a result of experience and also as a result of care- fully recorded field tests. This marked difference in the degree to which various soils respond to fertilizer treatments is the most strik- P K N L PK PN KN PKN PKNL M CVL 1 50 140 130 120 it ^ 110 o lOO It 90 - 80 Uj w 70 ^ 60 ^ 50 S 30 S; (0 10 — 1 i i 1 1 i 1 1 1 1 1 1 1 M X > * X /^v*^ ' /// ,' J/ \\W/ PORTSMOUTH FIN E SAND »*>.xxxxx NORFOLK FtNE SAN Or LOAM i ■ ■ ■ NORFOLK SAND NORFOLK FINE SAN D NORFOLK SANDY LOAM Fig. 1. — Percentage increase in growth of plants attributable to various fertilizer treatments of five principal soil types of the Waycross area, Georgia. ing fieature manifested in the results recorded in the preceding tables. It has also been long contended that there is a relation between the origin and the character of the soil and the character of the fertilizer that would be required to produce good results. It is doubtful, how- ever, if such a claim can be substantiated by observed facts over a large territory. It certainly is not borne out by the results obtained on the 220 soils here reported, except in case of the muck soils which uniformly respond to potash. In fact, the character of the fertilizer KELATION OF FERTILIZEE REQUIREMENT TO SOILS. 45 required for the same types or series of soils, as found in widely sep- arated areas or localities, and as shown by these results, varies more than that required for different types and series when they occur in the same area or locality. In illustration of this point, five soils from Waycross, Ga., tested in triplicate, show a marked uniformity in the character of their manurial requirements, regardless of the fact that they represent five types and two soil series. The same soils in Escambia County, Fla., show manurial requirements of a very differ- ent character. (See Table III, p. 15, and figs. 1 and 2.) In the Waycross area nitrate of soda is much more effective than sulphate of potash, and while lime is decidedly beneficial it is not P K N L PK PN KN PKN PKNL M CVL £:sc/iAtB//i COUNTY riORj DA Fig. 2.— Percentage increase in growth of plants, attributable to various fertilizer treatments for soil of the Waycross area, Georgia, and Escambia County, Fla. equal to nitrate of soda. In Escambia County sulphate of potash is nearly as efficient as nitrate of soda, and lime is about twice as effect- ive as the three fertihzer ingredients combined. These differences are based on average results for the two areas, and notwithstanding that the soils are of the same type, series, and formation, they show as marked differences in the character of fertihzer required as will be found between any of the areas, even when the soil types are entirely different. A Cecil sandy loam from Raleigh, N. C, was markedly improved by each of the three fertilizer salts, and also by lime, the increased growth of plants obtained when all of these ingredients were com- 46 FERTILITY OF SOILS AS AFFECTED BY MANURES. bined being 103 per cent. The same soil type from Lee County, Ala., as a result of tests on two samples, gave no response to either potash or phosphate, but produced a fair increase in growth as a result of nitrate of soda and also for lime, the maximum increase being 40 per cent where all three fertilizer salts were used in conjunction with lime. (See Table III, p. 15.) RELATIVE EFFICIENCY OF FERTILIZERS BY LOCALITY. In comparing the relative efficiency of the fertilizer ingredients, it is found that in the vast majority of instances nitrate of soda ranks first and lime second. This is especially true in the southern areas, where fertilizers are most extensively used. In the Central and North- ern States, where the efficiency is less marked, there are quite a num- ber of instances in which potash or phosphate outranks the lime or nitrate, although on an average the nitrate holds first rank and lime second. Sulphate of potash ranks next to lime, while acid phosphate having the lowest value, has in numerous instances shown a slightly negative effect. With soils from the States of Wisconsin, Iowa, Missouri, Illinois, Indiana, Ohio, New York, and Pennsylvania the response to commer- cial fertilizers has been moderate, slight, or in many instances almost imperceptible. On those soils, also, organic manures, in the form of cowpea vines or stable manure, have been very much more effeotive than the chemical ones, the mean relative increase attributable to manure being nearly three times that for a complete fertilizer with lime, i. e., barnyard manure gave an average increase of 38 per cent, while the complete fertilizer with lime gave an average increase of only 13 iper cent. (See Table XI, from which the averages were com- puted.) In these States lime alone seldom showed a marked effect. While the tests for these States are insufficient in nund)er to represent all soil types and conditions, yet the results are in conformity with the general practices of the vast majority of farmers of these sections, who as a ride do not purchase commercial fertilizers for use in the growing of general farm crops, but who for the most /part recognize the high value of barnyard manure and generally utilize all that is pro- duced on the farm. In the Atlantic and Gulf Coast States, including Kentucky, Tennes- see, and Arkansas, for which a larger number of observations have been made and where most of the important soil types have been tested, the soils, with few exceptions, respond in a marked degree to commercial fertilizers and lime, and while cowpea vines and stable manure are on an average somewhat superior to commercial fertilizers, the difi'erences between the relative value of these materials is small, being 54.7 and 58.6 per cent increase in growth for commercial fertil- izers and manure, respectively, as compared with 13 and 38 per cent for the same materials in the former group of States. (See Table XI, rOMPARATTVE FERTILITY OF SOILS. 47 from which these averages were computed.) In other words, the commercial fertihzers and Hme are more than four times as effective on soils of the Altantic and Gulf Coast States as they are on those of the Xorth Central ^States, including New York and Pennsylvania, while manure, when compared on the same soils, is only a half better on the soils from the former States. In this respect the results are again in conformit}^ with the practices in the Atlantic and Gulf Coast States, where the bulk of the commercial fertilizers are used. COMPARATIVE FERTILITY OF SOILS. Another point brought out in the tables is the relative fertility or crop-3nelding capacity of the untreated soils. Under the prevailing condition of crops and climate in the field, marked differences in this respect exist, some of the better soils showing a crop-producing capacity four or five times as great as that of the poorer ones. In these tests, where all soils are put into excellent physical condition and where moisture and temperature are always favorable, these dif- ferences are less marked but nevertheless exist to a considerable degree, as may be seen by comparing the actual weight of plants grown on "the untreated soils. The average growth of plants on untreated soils from the North-Central States, including New York and Pennsylvania, is 28 per cent greater than the average of those from the Atlantic and Gulf Coast States. This difference in the initial crop-producing capacity of the soils from the two sections as indicated in the pots is sufficient to make the actual increase in growth from the stable manure nearly as great from one section as from the other. Table XI. — Increase in growth attributable to complete fertilizer with lime and manure; and proportionate cost of fertilizer to value of crops as obtained from Census, 1900. State. Missouri Indiana Ohio New York Wisconsin Indiana Mississippi Pennsylvania. New York Missouri Texas Mississippi Arkansas Locality. Crawford County Newton County Westerville area Binghamton ajea Portage County Tippecanoe County . . Montgomery County . do Tompkins County Scotland County San Marcos area Pontotoc County Prairie County Hanover County. . . York County Caddo Parish Rusk County McCracken County . Cherokee County . . . Henderson County . Escambia County . . Lee County. Virginia South Carolina . . Louisiana Texas Kentucky South Carolina . Tennessee Florida Alabama Georgia j Waycross area . Virginia ! Louisa County. North Carolina 1 New Hanover County . Weight of plants untreat- ed soil. Grams. 9.6 8.8 7.8 9.3 9.0 6.8 9.3 10.1 7.9 9.6 8.5 6.8 8.3 7.5 6.2 6.3 8.0 5.0 7.2 7.5 6.3 6.4 6.0 5.9 .-1. 2 Increase from fer- tilizer Increase from and lime. Per cent. Per cent. 19 4 56 8 28 10 27 13 41 16 37 18 18 19 17 10 46 28 69 32 26 38 24 40 21 41 38 43 ■ 72 45 54 45 67 48 70 55 64 57 72 61 63 70 65 80 108 96 44 107 131 Cost of fertilizer to value of crop. Per cent. 0.23 .03 .62 1.13 .32 .39 .49 3.28 1.98 .08 .04 .03 4.04 6.34 .21 .29 .22 5.54 .10 6.82 4.81 4.45 4.. 50 7.82 48 PERTTLTTY OF SOILS AS AFFECTED BY MANURES. RELATIVE RESPONSE TO FERTILIZERS AND EXPENDITURES FOR FER- TILIZERS. In Table XI, where the average percentage increase in growth attributable to the complete fertilizer and lime is given for each area and arranged in an ascending series, there is also given the average percentage increase attributable to stable manure for the same soils as well as the expenditure for commercial fertilizers for the same areas or counties. The expenditure for fertilizers in the area is expressed in percentage of the valuation of products other than those fed to live stock and is computed from statistics reported in the United States Census for the year 1900. This table, besides showing the relative response to the chemical fertilizer and stable manure, which was discussed in preceding pages, also brings out the relation between the response to a complete ferti- lizer and lime, as obtained by the paraffin pot test, and the expendi- ture for fertilizer in each county or area expressed in percentage of the valuation of crops produced. The table is separated into two portions, the first portion embracing 13 areas or counties in wliich the increase in growth attributed to a complete fertilizer and lime ranges from to 40 per cent, and the second portion embracing 12 areas or counties in which the increased growth for the same treatment ranges from 41 to 107 per cent. In the first portion of the table there are no localities where the valuation of commercial fertilizer used equals 4 per cent of the value of the products grown, while in the second portion there are only four locali- ties where it does not exceed such percentage. In the first portion of the table where the response to the fertilizer with lime does not exceed 40 per cent there are only three localities where the cost of fertilizer actually used in 1900 exceeded 1 per cent of the valuation of the crop grown. These exceptions are in New York and Penn- sylvania, in areas where special crops probably receive more attention than in the others. The one in Pennsylvania was immediately about Philadelphia. Their nearness to both the great produce and fertilizer markets is also a factor tending to the more extensive use of fertilizers. Of the four counties in the second portion of the table in wliich the cost of fertilizers used does not exceed 4 per cent of the value of crops grown, McCracken County, Ky., and Henderson County, Tenn., are each represented by only one sample. If better represented, they would probably fall in the first portion of the table, as indicated by the small amount of fertilizer used. The remaining two counties are Caddo Parish, La., and Rusk County, Tex., in each of which cases but little commercial fertilizer is used, though they are grouped in the second portion of the table as a result of their response to fertilizers in EESPONSE TO AND EXPENDITURES FOR FERTILIZERS. 49 the pot test. In these two counties the soils were well represented by the samples tested, and the results indicate that the moderate use of commercial fertilizer might prove profitable. There are other factors, Ohio, Westervil/e ^rea /no/, h/e avion Co Miss-, Moniqomery Co Ma, Crain/forc/ Co NY. Tomp/iins Co Wis , Portage Co Po , Moniaomery Co A/ y , Bin^howton Area Va , Honot^er Co /no/. Tippecanoe Co. Tenn , Henc/erson Co SC. YorA Co Ar/< . Proir/e Co L a. , Coc/c/o Poris/j Mo, Scot /one/ Co- Ten , Son Marcos /Irea Ky., Mc Croc /(en Co SC. C/ieroAee Co /l/o . L ee Co Tex, Rus/< Co. Miss. Pontot oc Co. Go. Woycross /Irea f/o. £^scam/oi o Co. Vo. Louiso Co A/ C. New //on over Co "=^&S °^ ^^m J O 5 10 P C n K ^^^S3 A/ L y////^//x Fig. 3.— Percentage gain in plant growth attributable to each of the salts P, K, N, and L when used alone. however, besides the comparative response to fertilizers which are concerned in their profitable use. Low value of land, long distance from markets and seat of supplies, including fertilizer, would all tend 50 FERTILITY OF SOILS AS AFFECTED BY MANURES. toward the unprofitable use of fertilizers. In Rusk County, Tex., and Caddo Parish, La., a few farmers have begun the use of commercial fertilizers and find them profitable. AVERAGE EFFICIENCY OF FERTILIZER SALTS, BY LOCALITY. Figure 3 shows graphically the percentage increase in growth attributable to each of the fertilizer salts and lime w^hen used sepa- rately, and is based on the average for all soils of each area as given in Table X on page 43. The areas are arranged in an ascending series according to the aggregate increase from all ingredients, which ranges from a minimum of only 9 per cent for the Westerville area, Ohio, to a 'maximum of 140 per cent for New Hanover County, N. C. The range in the efficiency of the individual salts is also marked, the increase attributable to nitrate of soda varying from 4 per cent in several of the areas in the upper part of the diagram to 65 per cent in New Han- over County, N. C, and that for lime varying from a slightly negative result for Newton County, Ind., to an increase of .52 per cent for Escambia County, Fla., with somewhat lesser variations for potash and phosphorus. In the majority of the areas the efficiency of the three fertilizer ingredients and lime assume the same order. In 16 of the 25 areas, nitrate of soda w^as the dominant salt, while lime was dominant in 3 and potash in 3 of the areas. In 14 areas out of 25 lime ranks second in efficiency. In the States of Indiana, Missouri, Pennsylvania, and Virginia potash seems to have been relatively more effective than elsewhere. In the Arkansas and Florida areas lime is largely responsible for their position in the series, and in both of them nitrate of soda holds a relativelv low rank. Table XII. — Average percentage increase in growth for all fertilizers and combinations, by soil series. No. of soils. Weight of plants grown on un- treated soil. Increase in growth attributable to— Soil series. Ph' P.ct. 19 16 12 4 9 8 62 i^i p^ Pi p.ct. 53 34 32 23 15 10 33 M p.ct. 63 47 46 23 18 20 49 S5 p- P.ct. 68 4.5 52 31 19 12 60 Ph s > Portsmouth Norfolk 45 12 18 8 5 4 Grams. 5.2 6.3 6.0 8.0 7.9 7.0 6.0 P.ct. 10 6 14 2 5 5 21 p.ct. 43 35 36 20 11 16 27 p.ct. 25 16 17 8 4 6 56 P.ct. 43 28 26 11 4 4 30 P.ct. 96 64 69 37 19 8 60 p.ct. 100 76 62 42 44 50 46 P.ct. 117 83 Cecil 65 44 Miami 17 8 Muck 39 EFFICIENCY OF FERTILIZERS, BY SOIL SERIES. 51 Table XIII. — Peirentage increase in growth attributable to each fertilizer salt and to lime when used alone, as compared with its calculated effect when used in various com- binations. Averages for each of six series and Muck. L. P. K. N. Soil scries. 4 1 ■ z Ui Pk ■3 1 !? + 1 w Pi N 14 1 Z + (N 1 Zl 1 1 !^ £■' 1 55 P- 1 + p- e-i P, 1 PL, i ■3 1 Ph + 1 W N Ph 1 Pi + (N Pi 1 W p^ Portsmouth.. Norfolk Cecil 43 28 26 11 4 4 30 28 19 17 6 -4 10 6 14 2 5 5 21 8 - i i 3i - i -4 Vi li H 4 -2^ 20 5 -2 t) 19 16 12 17i 11 6* 41 3 25 28i 17.J 14^ 15* 15 11 20 43 35 36. 20 U 16 27 43i 29J 26 20 9i 8i 45| 32J 30J 19 14* 12" 13 43 29 35 Orangeburg . . Miami Marshall Muck... 8| 4 1 ! 9 -8 '■ 8 11 62 4"! -8 3V 4 8'^ 2 36 , 27 23 15 6 4 COMPARATIVE EFFICIENCY OF FERTILIZERS, BY SOIL SERIES. Table XII gives the average percentage increase in growth attribu- table to each fertilizer salt and combination of salts, by soil series, using only those in which the treatments were uniform and complete. No series were used where less than 5 samples were available, except in the case of Muck, in which only 4 samples were used. Table XIII gives the percentage increase in growth attributable to each of the three fertilizer ingredients and lime, as compared with their effect when used in the combinations, the efficiency in the com- binations being calculated by the difference method previously explained. These two tables show that the relative efficiency of the three salts and lime is essentially the same for all the soil series except the Muck. The range of efficiencj^ of the complete fertilizer with lime is very marked, varying from a maximum of 96 per cent for the Portsmouth series to only 8 per cent for the Marshall series. It is noticeable that barnyard manure is relatively much more efficient for the Marshall and Miami series than for the others. Figure 4 shows graphically the average relative percentage effect- iveness of the three fertilizer ingredients and lime by series when used separately, as given in Table XIII. The variation in the aggregate effectiveness of the three salts and lime for the several series is nearly as great as when the soils were grouped by localities. It is evident, however, that the serial grouping is, to some extent, a locality group- ing also, the Marshall and Miami series, which are the least responsive to the commercial fertilizers, occurring only in those States in which fertilizers are but little used, wliile the Portsmouth and Norfolk series are confined wholly to the Atlantic seaboard, where fertilizers are most extensively used. While there is a marked difference in the aggregate effectiveness of the three salts and lime on different series, 52 FERTILITY OF SOILS AS AFFECTED BY MANURES. ranging from a minimum of 29 per cent for the Miami series to a maximum of 115 per cent for the Portsmouth series, broadly speaking the relation between the value of the individual ingredients in the several series is remarkably uniform, even more so than occurs when the soils are considered by areas. The Muck, represented by only 4 samples, is exceptional, and shows the characteristic importance of potash, that ingredient being on the average about as effective as the total of lime, nitrogen, and phosphate. Tliis fact is also in conformity with our field knowledge of Muck and Peat, which are greatly bene- fited by applications of potash salts. Portsmo uth Gee// A/orfo/k Oronoe >6 urq Mars/70 // M/o/v/ A/fuoA so//s \})))mM}))} . //y/^y/-////^///^/^yy//\ ^^v^^^ H^^^^^^^ Wmmwmimm O /O 20 H H H H K I— P I I K t.^\\\\\vH N L W////A Fig. 4. —Relative efficiency of fertilizer ingredients by soil series, when the ingredients are used alone. Table XIV. — Average percentage increase ingrowth attributable to each fertilizer salt and lime for soils of the Norfolk, Portsmouth. Orangeburg, and Cecil series when classified by texture. L. P. K. N. 12; ^- 14 Soil class . ^ M PL, •z i4 P^ 1 W fc + Ci 1 'Z + (N 1 (^ + 01 1 Pu ■3 1 c "3 1 Ui p- 'Z p- +