:CM =1 Country Life Education Series IL FERTILITY AND RMANENT AGRICULTURE HOPKINS LIBRARY FACULTY OF FORESTRY UNIVERSITY OF TORONTO I COUNTRY LIFE EDUCATION SERIES Edited by Charles William Burkett, recently Director of Experiment Station, Kansas State Agricultural College j Editor of American Agriculturist TYPES AND BREEDS OF FARM ANIMALS By Charles S. Plumb, Ohio State University PRINCIPLES OF BREEDING By Eugene Davenport, University of Illinois FUNGOUS DISEASES OF PLANTS By Benjamin Minge Duggar, Cornell University SOIL FERTILITY AND PERMANENT AGRICULTURE By Cyril G. Hopkins, University of Illinois Other volumes in preparation SOIL FERTILITY AND PERMANENT AGRICULTURE BY CYRIL G. HOPKINS, PH.D. PROFESSOR OF AGRONOMY IN THE UNIVERSITY OF ILLINOIS, CHIEF IN AGRONOMY AND CHEMISTRY AND VICE DIRECTOR OF THE ILLINOIS AGRICULTURAL EXPERIMENT STATION V GINN AND COMPANY BOSTON • NEW YORK • CHICAGO • LONDON ENTERED AT STATIONERS' HALL COPYRIGHT, igio BY CYRIL G. HOPKINS ALL RIGHTS RESERVED giO.2 6 633 Cftt gUfaenaum 3PreS6 GINN AND COMPANY • PRO- PRIETORS • BOSTON • U.S.A. DEDICATED TO THE ASSOCIATION OF AMERICAN AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS THE RIGHTFUL GUARDIANS OF AMERICAN SOILS TO USE THE LAND WITHOUT ABUSING IT - J. OTIS HUMPHREY PREFACE Liebig said, "Agriculture is, of all industrial pursuits, the rich- est in facts and the poorest in their comprehension." To a large degree this statement is still true, and the chief purpose of this volume is to bring together in convenient form the world's most essential facts gathered from the field and laboratory, and to develop from them some foundation principles of permanent agriculture ; for, as Liebig also truly said, " Facts are like grains of sand which are moved by the wind, but principles are these same grains cemented into rocks." While one dare not believe that error has been completely avoided, the facts presented have been checked with all reasonable care, and they may be accepted with the confidence that they are accurately reproduced from the original data. Unsolved problems still remain, and some conclusions which seem to be indicated by the data thus far reported may be modi- fied later when more complete information is afforded. The author will always receive with deep appreciation suggested ad-, ditions, modifications, or corrections. It is, perhaps, unnecessary to say to the reader that his general knowledge of farm practice is presupposed, and no attempt has been made to include herein a thousand details with which every man experienced in the art of agriculture already is familiar. For the sake of himself and children it must be said to the practical farmer that he should encourage the teaching of the science of agriculture in the school, even though he may know much more than the teacher concerning the art of agriculture. viii PREFACE To encourage the teacher, let me say that much of the science of agriculture can be successfully taught without a field or a gar- den, and even without complete knowledge of the art. Thus, you may teach why clover should be grown and when it contains the most nitrogen, but leave the farmer to determine for himself when to plow it under, if he is the better judge of seasonal conditions and of their probable influences upon his own soil and crop. CYRIL G. HOPKINS UNIVERSIT^ OF ILLINOIS URBANA CONTENTS PAGE INTRODUCTION xvii PART I. SCIENCE AND SOIL CHAPTER I. FOUNDATION FACTS AND PRINCIPLES i II. THE MORE IMPORTANT ELEMENTS AND COMPOUNDS ... 12 III. PLANT FOOD AND PLANT GROWTH 26 IV. THE EARTH'S CRUST 46 V. SOIL FORMATIONS AND CLASSIFICATIONS 54 VI. SOIL COMPOSITION 58 VII. AVAILABLE PLANT FOOD 107 VIII. SOIL SURVEYS BY THE UNITED STATES BUREAU OF SOILS . .114 IX. SOIL ANALYSES BY THE UNITED STATES BUREAU OF SOILS . 136 X. CROP REQUIREMENTS FOR NITROGEN, PHOSPHORUS, AND PO- TASSIUM • 153 XI. SOURCES OF PLANT FOOD 156 PART II. SYSTEMS OF PERMANENT AGRICULTURE XII. LIMESTONE 160 XIII. PHOSPHORUS 183 XIV. ORGANIC MATTER AND NITROGEN 194 XV. ROTATION SYSTEMS FOR GRAIN FARMING 226 XVI. LIVE-STOCK FARMING 231 — XVII. THE USE OF PHOSPHORUS IN DIFFERENT FORMS . . . 236 — . XVIII. THEORIES CONCERNING SOIL FERTILITY 300 PART III. SOIL INVESTIGATIONS BY CULTURE EXPERIMENTS XIX. THE ROTHAMSTED EXPERIMENTS 344 XX. PENNSYLVANIA FIELD EXPERIMENTS 420 XXI. OHIO FIELD EXPERIMENTS 441 XXII. ILLINOIS FIELD EXPERIMENTS 453 XXIII. FIELD EXPERIMENTS IN THE SOUTH, INCLUDING SOUTHERN ILLINOIS 476 CONTENTS CHAPTER PAGE XXIV. MINNESOTA SOIL INVESTIGATIONS ...... 499 XXV. CANADIAN FIELD EXPERIMENTS 505 XXVI. SHORT-TIME POT-CULTURE AND WATER-CULTURE EXPERIMENTS IN COMPARISON WITH FIELD RESULTS 513 PART IV. VARIOUS FERTILITY FACTORS XXVII. MANUFACTURED COMMERCIAL FERTILIZERS . . . . 517 XXVIII. CROP STIMULANTS AND PROTECTIVE AGENTS .... 533 XXIX. CRITICAL PERIODS IN PLANT LIFE 538 XXX. FARM MANURE 541 XXXI. LOSSES OF PLANT FOOD FROM PLANTS 549 XXXII. LOSSES OF PLANT FOOD FROM SOILS 556 XXXIII. FIXATION OF PLANT FOOD BY SOILS- 562 XXXIV. ANALYZING AND TESTING SOILS 565 XXXV. RELATION OF FERTILITY TO APPEARANCE OF SOILS OR CROPS . 572 XXXVI. FACTORS IN CROP PRODUCTION 5715 XXXVII. ESSENTIAL FACTORS OF SUCCESS IN FARMING .... 584 XXXVIII. THE VALUE OF LAND 586 XXXIX. Two PERIODS IN AGRICULTURAL HISTORY ... . 1590 APPENDIX SECTION I. THE PRODUCTION OF PHOSPHATE ROCK 595 II. MODEL FERTILIZER LAW 599 III. COMPOSITION OF ANIMAL AND PLANT PRODUCTS 602 IV. STATISTICS OF AGRICULTURAL PRODUCTS 605 V. METHODS OF SOIL ANALYSIS 626 VI. COMPOSITION OF SOME EUROPEAN SOILS 634 VII. AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS IN THE UNITED STATES AND CANADA 643 INDEX 647 LIST OF TABLES TABLE PAGE 1. Elements, Symbols, and Atomic Weights 10 2. The More Important Elements : Occurrence . ... 13 3. Composition of Silicates 48 4. Composition of Rock 49 5. Composition of Fresh Limestone and Residual Clay . . .51 6. Soils: General Groups 55 7. Recognized Soil Types 5^ 8. Relative " Supply and Demand " of Seven Elements ... 59 9. Composition of Productive and Nonproductive Soils ... 63 10. Composition of Adobe and Coral Limestone Soils . -65 1 1 . Composition of Loess Deposits 69 12. Composition of Ten Residual Soils .... -73 13. Mineral Plant Food in Wheat, Corn, Oats, and Clover . . 75 14. Composition of New York Soils ....... 75 15. Fertility in Illinois Soils: Surface .82 16. Fertility in Illinois Soils : Subsurface 84 17. Fertility in Illinois Soils: Subsoil ....... 86 18. Composition of Southern Indiana Surface Soils .... 88 18.1. Plant Food in Surface Soils of Iowa 91 19. Composition of Surface Soils of Tennessee 93 19.1. Composition of Georgia Soil 94 19.2. Average Composition of Some Texas Soils ..... 95 19.3. Composition of Some Louisiana Soils ...... 96 20. Composition of Some Michigan Soils . . . . . .98 20.1. Composition of Canadian Soils • 103 20. 2. Certain Plant-food Elements in Illinois Surface Soils . . . 105 21. Annually Available Fertility in Illinois Soils no 22. Composition of Various Extensive Soil Types of the United States 138 23. Fertility in Farm Produce 154 24. Fertility in Manure, Rough Feeds, and Fertilizers . . . . 157 25. Pennsylvania Experiments with Lime ...... 165 25.1. Maryland Experiments with Lime ...... 167 26. Experiments with Magnesium Carbonate ..... 171 27. Losses of Calcium Carbonate from Broadbalk Field . . .174 28. Losses of Calcium Carbonate from Hoos Field . . . 175 29. Digestibility of Common Food Stuffs *99 xii LIST OF TABLES TABLE PAGE 30. Plant Food recovered from Food Consumed (Illinois) . . . 201 31. Plant Food recovered from Food Consumed (Pennsylvania) . . 202 32. Plant Food recovered from Six Months' Feeding (Ohio) . . 204 33. Fixation of Nitrogen by Alfalfa . 214 34. Nitrogen in Sweet Clover 220 35. Composition of Crimson Clover . ... . . . .221 36. Composition of Legumes and Other Plants ..... 222 37. 38, 39, and 39^. Comparison of Raw Phosphate and Acid Phosphate 247-253 40. Ohio Experiments with Manure, Phosphate, Kainit, Gypsum, and Complete Fertilizers ......... 256 41. Balance Sheet for Nitrogen and Phosphorus in Manure-phosphate Experiments .......... 257 42. Maryland Experiments with Different Phosphates .... 262 43. Pennsylvania Experiments with Different Phosphates . . . 264 44. 45, 44^, and 451:. Rhode Island Experiments with Nine Phosphates 268-274 46 and 47. Maine Experiments with Different Phosphates . . 276, 277 48 and 49. Massachusetts Experiments with Different Phosphates 279, 282 50. Illinois Experiments with Raw Rock Phosphate .... 285 51. Steamed Bone Meal and Raw Rock Phosphate .... 287 52-58. Rotation Crops on Agdell Field 346-352 59. Summary of Crop Yields and Values, Agdell Field .... 360 60 and 61. Wheat Yields, Broadbalk Field, Averages . . . 364,365 62. Wheat Yields at Rothamsted, Comparisons .... 372, 373 63: Wheat Yields, Broadbalk Field, Nitrogen Increments . . . 374 64. Wheat Yields at Rothamsted, Summaries 375 65. Rothamsted Records, Rainfall and Drainage 377 66 and 67. Barley Yields on Hoos Field 380, 381 68. Potato Yields on Hoos Field 386, 387 69. Residual Effect of Fertilizers on Hoos Field ..... 390 70. Hay Yields on The Park at Rothamsted . . . . . 393 71. Root Crops on Barn Field ....... 399, 400 72. Rothamsted Fields abandoned to Nature 404 73. Plant Food in Soil of Broadbalk Plots 411 74. Composition of Drainage Waters from Broadbalk Field . . . 413 75. Nitrogen in Soil of Agdell Plots 416 76. Compositio.n of Crops grown on Agdell Field 417 77. Composition of Hay from The Park, Rothamsted .... 418 78. Pennsylvania Crop Yields in Field Experiments .... 423 79 and 80. Pennsylvania Experiments by Twelve-year Periods . 428, 429 81. Pennsylvania Experiments : Financial Summary . . . -431 8iP. Pennsylvania Experiments : Twenty-five-year Average . . . 434 LIST OF TABLES xiii 82. Ohio Experiments : Five-year Rotation . . 442 83. Ohio Experiments : Potatoes, Wheat, Clover .... 448 84. Experiments at Strongsville, Ohio 452 85. Illinois Experiment Plots, Urbana 457 86. Comparable Corn Yields, Illinois Experiments .... 459 87. Crop Yields on Sibley Field, Illinois ....... 462 88. Crop Yields on Bloomington Field, Illinois . . . . 464 89. Crop Yields on Antioch Field, Illinois 467 90. Crop Yields on Sand Land, Illinois 468 91. Corn Yields on Deep Peat Soil, Illinois 471 92 and 93. Corn Yields on Peaty Alkali Soil, Illinois . . 473, 475 94. Crop Yields in Southern Illinois, Odin Field .... 478 95. Crop Yields in Southern Illinois, Du Bois Field . . . .481 96 and 97. Crop Yields on Worn Hill Land, Vienna, Illinois . 483, 485 98 and 98.1. Pot-culture Experiments with Worn Hill Soil . . 486,487 99. Southern Iowa Field Experiments 488 100. Georgia Fertilizer Experiments with Corn ..... 490 101. Rainfall Records at Experiment, Georgia ..... 49* 1 02 and 103. Georgia Fertilizer Experiments with Cotton . . 492, 493 104. Alabama Field Experiments with Cotton 495 104.1. Louisiana Field Experiments 496 105. Minnesota Soil Investigations 499 1 06 and 107. Canadian Field Experiments ..... 508,511 108 and 109. Comparison of Pot Cultures and Field Experiments . 513, 514 no. Composition of Farm Manures . . . . . • -543 in. Composition of Pulverized Dried Manures 545 112. Composition of Manure before and after Exposure . . . 547 113. Composition of Bean Crop at Different Periods of Growth . . 550 114. Composition of Barley at Different Periods of Growth . . . 552 115. Plant Food removed from Plants by Leaching .... 555 116. Nitrogen in Rothamsted Drainage Waters 557 117. Soluble Nitrogen in Cropped Soils, Rothamsted .... 55^ 118. Ammonia Fixation and Nitrification 563 119. Effect of Soil Preparation, Cultivation, Irrigation, and Fertilization on the Yield of Corn 578 120. Value of Land, measured by Crop Yields ..... 5^7 121. Composition of Animal and Plant Products 602 ILLUSTRATIONS PAGE GLACIAL MAP OF NORTH AMERICA 68 GENERAL SURVEY SOIL MAP OF ILLINOIS 76 A MAN OF SCIENCE: EUGENE WOLDEMAR HILGARD 102 MAP OF UNITED STATES SOIL PROVINCES 116 NITROGEN FIXATION BY CLOVER 218 TOPOGRAPHIC MAP OF OHIO EXPERIMENT FIELD 252 DIRECTOR CHARLES E. THORNE . . . 254 SIR JOHN BENNET LAWES 342 SIR JOSEPH HENRY GILBERT 344 TURNIPS ON AGDELL FIELD, 1908 362 DIRECTOR A. D. HALL 408 DIRECTOR EDWARD B. VOORHEES > 516 RAINFALL CHART OF NORTH PLATTE, NEBRASKA 580 RAINFALL MAP OF THE UNITED STATES 582 INTRODUCTION IT is the first business of every farmer to reduce the fertility of the soil, by removing the largest crops of which the soil is capable; but ultimate failure results for the landowner unless provision is made for restoring and maintaining productiveness. Every landowner should adopt for his land a system of farming that is permanent, — a system under which the land becomes better rather than poorer. If the independent farmer is to adopt and maintain permanent systems of profitable agriculture, he cannot accept " parrot " instruction ; he must know the why and wherefore, the reason for doing things, and the ultimate effect of his agricultural practice upon the productive power of the land. Every farm is an inde- pendent enterprise in which the farmer himself is the superin- tendent and general manager, and he must be able to direct the business, even though he may be the only man to execute his own plans. The agriculture of a state cannot be managed from a central office. The landowner must think for the land. The author is familiar with the often expressed idea that what the farmer wants is a simple statement of facts, but he is even more familiar with the absolute truth that what the farmer de- mands is the most positive proof of the correctness of such state- ment before he is willing to make any change from a practice based upon long experience. In the preparation of this book free use has been made of such technical terms as are necessary to the discussion of fundamental principles with scientific correctness. No apology is offered for this. Farmers and agricultural students have at least as good intellects as other classes of people; and if when they leave the farm they can learn to understand and manage successfully such lines of business as banking, contracting, building, operating railroads, factories, and other commercial establishments, — xviii INTRODUCTION which they are doing everywhere, — they can also understand their own business, if they will, when they remain on the farm or in control of land. Technical books are to be studied; they are not written for en- tertainment. They furnish definite facts, accurate data, and necessary information, relating to underlying principles upon which permanent successful practice must be based. The most important material problem of the United States is to maintain the fertility of the soil, and no extensive agricultural country has ever solved this problem. The frequent periods of famine and starvation in the great agricultural countries of China, India, and Russia, and the depleted lands and abandoned farms of our own eastern United States are facts that serve as a constant proof that the common practice of agriculture reduces the produc- tive power of land. The rule is almost universal that old land is less productive than new land. This simple and well-recognized fact points inevitably toward future poverty, not only for the individual or the family, but likewise for the commonwealth and for the nation. We may ignore this if we choose in America for a few more years, but with the decreasing productive power of our lands and with a rapidly increasing population the truth must strike us in the face in the near future. We cannot afford to let ignorance, prejudice, or bigotry blind us in this matter, neither in ourselves nor in others. Even the confi- dent assurance, by those who live in continued plenty, that the people of earth are not destined to suffer hunger, does not remove the positive fact that thousands, and sometimes millions, of people actually die of starvation within a single year in some of the old agricultural countries. An early recognition of these world-wide conditions and tend- encies is of paramount importance to the people controlling the more productive lands of the United States, not only for their own sake, but also for the sake of others who are dependent upon those lands for their present and future support, whether engaged directly in agricultural pursuits or in other industrial or professional lines, which cannot exist and prosper without agriculture. If the art of agriculture has ruined land, the science of agricul- INTRODUCTION xix ture must restore it; and the restoration must begin while some farmers are still prosperous, for poverty-stricken people are at once helpless and soon ignorant. Outside help will always be required to redeem impoverished soils, for poverty makes no in- vestments, and some initial investment is always required for soil improvement. It is the purpose of this book to teach the science of soil fertility and permanent agriculture, chiefly by reporting facts rather than by offering theories; and any one of common sense who reads the English language, and who can understand the common school arithmetic, can understand this book if he will study it. (The fact may well be recognized that some who have ample time for study, though physically industrious, are mentally lazy.1) The author suggests, however, that the busy farmer, who wishes to familiarize himself as quickly as possible with the most essential practical facts pertaining to the economical and perma- nent improvement of common or normal soils, and who is willing to pass over temporarily the discussion of foundation principles, may well begin the study of this book with " Systems of Perma- nent Agriculture," Part II, after first making the following facts a part of his ever ready knowledge: (1) Phosphorus and decaying organic matter are the two sub- stances which constitute the key to profitable systems of permanent agriculture on most of the normal soils of America ; although, when soils become sour, or acid, ground natural limestone should also be regularly applied, at the rate of about two tons per acre every four to six years. (2) There are six essential positive factors in crop production: the seed, a home for the plant, the food of which the plant is made (and this factor is just as important for plants as it is for animals), moisture, heat, and light. Of these six factors, the least appre- ciated and the most neglected is that of plant food, and yet this is a factor which the farmer can very largely control, whereas the others (except the seed) are largely beyond his control. (An important negative factor is protection from weeds, insects, and disease.) 1 " Many poor farmers have a lazy faith in the Lord ; they think or hope that He will somehow make up for whatever they fail to do." — HOARD. xx INTRODUCTION (3) Of the ten different chemical elements absolutely required for the growth of every agricultural plant, three come directly from air and water in practically unlimited amounts, and these three (carbon and oxygen from air and hydrogen from water) constitute about 95 per cent of the common mature crops. Nevertheless, each one of the seven elements obtained from the soil, though aggre- gating only 5 per cent, is absolutely necessary to the life and full development of the plant. Indeed, if any one of these elements be entirely lacking, the soil would be infertile and barren. So important are these plant-food elements, that soils are found so deficient in some essential plant food that the addition of a single element will more than double the crop yield. (4) The five elements, potassium, magnesium, calcium, iron, and sulfur, are contained in most normal soils in such large amounts, compared to the requirements of crops, that the supply rarely becomes depleted. Thus, in most cases, the problem is narrowed to the two elements, nitrogen and phosphorus, although, for various reasons, potassium also has come to have a recognized money value in commercial fertilizers. (5) Nitrogen is contained in the air in inexhaustible amount, but the legumes (clover, alfalfa, peas, beans, etc.) are the only agricultural plants which have power to utilize the free nitrogen of the air. Nitrogen in limited amount is contained in the soil in the organic matter, the principal material which gives a good soil its dark color. If the supply of organic matter is maintained, by plowing under farm manure, clover, cowpeas, or other green manures, then the supply of nitrogen will also be maintained. (6) The plowed soil of an acre (2 million pounds, for a depth of 6 § inches) of rich, well-balanced normal land in the Corn Belt contains about 8000 pounds of nitrogen, 2000 pounds of phosphorus, 35,000 pounds of potassium, and 15 tons of calcium carbonate (limestone) . (7) The surface soils of the United States vary in composition: (a) in nitrogen content, from 1000 pounds to 35,000 pounds; (b) in phosphorus content, from 160 pounds to 15,000 pounds; (c) in potassium content from 3000 pounds to 60,000 pounds, per acre; and many soils not only contain no lime, but are markedly acid and thus require heavy applications of lime, while some pro- INTRODUCTION xxi ductive soils contain as much as 20 per cent of calcium carbonate, corresponding to 200 tons of limestone per acre. (8) A loo-bushel crop of corn takes from the soil about 100 pounds of nitrogen, 17 pounds of phosphorus, and 19 pounds of potassium, in the grain, and about 48, 6, and 52 pounds of these respective elements in the stalks or stover. (9) One ton of average fresh farm manure contains about 10 pounds of nitrogen, 2 pounds of phosphorus, and 8 pounds of potassium; and 100 pounds of the most common " complete " commercial fertilizer contains about 2 pounds of nitrogen, 4 pounds of phosphorus, and 2 pounds of potassium. (10) One ton of clover hay contains about 40 pounds of nitro- gen, 5 pounds of phosphorus, and 30 pounds of potassium. When grown on soil of fair productive capacity, the roots and stubble of the clover plant contain no more nitrogen than the soil has furnished to the plant; but for each ton of clover plowed under, the soil is enriched by about 40 pounds of nitrogen. (n) Roughly estimated, the plant food liberated from an aver- age soil during an average season with average farming is equiva- lent to about 2 per cent of the nitrogen, i per cent of the phos- phorus, and ^ of i per cent of the potassium, contained in the surface stratum (about 6f acre inches, or 2 million pounds of average soil). (12) As an average in live-stock farming, the animals retain about one fourth of the nitrogen and phosphorus and destroy two thirds of the organic matter of the food consumed, and large loss is likely to occur in the manure produced, especially in nitrogen and or- ganic matter, a loss of one half of these constituents being easily possible during three or four months, in part from fermentation, which may occur even under cover, and in part from leaching where the manure is exposed to the weather or where too little absorbent bedding is used. (13) It is less difficult to maintain or increase the organic matter of the soil by means of legume crops and crop residues in a good rotation for grain farming than in any system of live-stock farming which does not include the purchase of feed. (14) Some satisfactory rotation plans for grain farmers are wheat, corn, oats, and clover ; or wheat, corn, and cowpeas ; or xxii INTRODUCTION cotton, corn, and oats and cowpeas. The first of these is a four- year rotation which should include a catch crop of clover seeded the first year and plowed under for corn as late as practicable in the spring of the second year. The other two are three-year rotations, and they should also include legume catch crops wherever practicable. In each rotation for grain farming, all products are to be returned to the soil excepting the grain, or seed, and the cotton lint. Either the whole cotton seed or the hulls and meal should also be returned for fertilizer. (15) In live-stock farming the feeding should be done on the fields so far as practicable, and manure produced in the barn should be hauled and spread in the fresh condition so far as possible. Sufficient bedding should be used to absorb all of the liquid excre- ment, which is as valuable, ton for ton, as the solid excrement. (16) To insure the maintenance of the phosphorus content of the soil where large crops are produced, about 20 pounds of phos- phorus per acre for each year in the rotation should be applied in grain farming and about 10 pounds per acre in live-stock farming (aside from that returned in the manure). To enrich the soil in phosphorus, heavier applications should be made for a time. (17) The average investment required for 25 pounds of phos- phorus is about 75 cents in 200 pounds of fine-ground natural rock phosphate of good grade, about $2.50 in 200 pounds of good steamed bone meal, about $3.00 in 400 pounds of good acid phos- phate, about $6.00 in 600 pounds of the average " complete " commercial fertilizer, and about $80 in manure made from corn costing 40 cents a bushel. The natural phosphate, if ground to pass through a sieve with 10,000 meshes to the square inch, gives satisfactory results when applied in liberal amounts (as 1000 pounds per acre every three or four years), if used in connection with decaying organic matter in sufficient amount to maintain the nitrogen. (18) Potassium salts are used with very great profit on soils positively deficient in that element, as on most well-drained ex- tensive peaty swamp lands; and soluble salts, such as kainit, may produce some profit for a time if used in connection with phosphorus on soils deficient in decaying organic matter, even where the total supply of potassium in the soil is very large. INTRODUCTION xxiii (19) Commercial nitrogen can usually be used with profit in market gardening, in cotton growing, and sometimes in the pro- duction of timothy hay near large cities; or, as a rule, wherevei the gross returns from an acre of produce exceeds $50 or $75. (20) As a rule, commercial nitrogen cannot be used with profit for the production of the staple grain crops, such as corn and wheat, although under some conditions small applications of nitrogen alone or with other elements, as in the ordinary so-called " com- plete " fertilizer, may stimulate the plants sufficiently to enable them to draw more heavily upon the soil, and thus return apparent temporary profit in a system of ultimate land ruin. And other seed fell on good ground, and sprang up, and bare fruit an hundred fold. — JESUS. I applied mine heart to know, and to search, and to seek out wisdom, and the reason of things. — SOLOMON. Every man shall receive his own reward according to his own labor ; for we are laborers together with God. — PAUL. SOIL FERTILITY AND PERMANENT AGRICULTURE PART I SCIENCE AND SOIL CHAPTER I FOUNDATION FACTS AND PRINCIPLES Science. Science means knowledge, nothing more and nothing less. It does not mean theory unsupported by fact. To plow the land and plant the seed and cultivate the crop is art, or prac- tice. To know what the soil and air contain and what the crop requires is science. If 10 cents are taken from 70 cents, only 60 cents remain. This is science, knowledge, fact, and not mere opinion. In the study of soil fertility we must make large use of two well-established exact sciences, mathematics and chemistry. Several other sciences furnish much exact data, but in some branches the data thus far secured are not sufficient to fully reveal the controlling facts and principles. Chemistry. Chemistry is the science which deals with the com- position of matter. All material things are composed of about eighty primary substances, called elements, which may exist sepa- rately or in various combinations, called compounds. About forty of the elements are more or less common, the others being rare elements. Air and soil and plants and animals contain less than twenty elements that are of interest to agriculture; while only ten different elements are known to be essential for the making of agricultural plants. (One other element, chlorin, may be essen- tial, but, if so, only in minute quantity.) 2 SCIENCE AND SOIL Chemical elements. An element is a substance which cannot be divided into two or more different substances. Sulfur (S) is a solid, nonmetallic element, easily melted to the liquid form. A piece of sulfur may be divided into two parts, but each part is sulfur, and if nothing else is added to sulfur, nothing but sulfur can be obtained from it. Carbon (C) is the principal element con- tained in coal. Iron is a well-known metallic element. Oxygen (O) is an element contained in the air in the gas form. Chemical compounds. A compound is a substance which con- tains two or more different elements and which possesses some properties or characteristics not possessed by either element alone. Thus, if carbon and sulfur be mixed together at the ordi- nary temperature, the product is only a mixture in which each element retains its own properties; but, at a higher temperature and under proper conditions, one combining weight of black carbon will unite with" two combining weights of yellow sulfur and form the compound called carbon disulfid (CS2), which is neither black nor yellow nor solid, but a colorless liquid somewhat resembling water, but which, when pure, contains absolutely noth- ing but the two elements, carbon and sulfur. Carbon, in charcoal for example, may be eaten in considerable quantity without harm, and sulfur is not dangerous in large doses; but the compound, carbon disulfid, is a deadly poison, and is fre- quently used as an insecticide and for the extermination of gophers and other burrowing animals. Thus the properties of the com- pound may differ in many respects from those of either element contained in it. On the other hand, when carbon is burned, by uniting with the oxygen of the air, the compound, carbon dioxid (CO2) , is formed, and when sulfur is likewise burned, the compound, sulfur dioxid (SO2), is formed; while if carbon disulfid is burned, by uniting with the oxygen of the air, the products of combustion are exactly the same as though the carbon and sulfur were burned separately with oxygen, carbon dioxid and sulfur dioxid being formed. Sodium (Na, natrium in Latin) is a soft metallic element which takes fire when thrown into water, and the element chlorin (Cl) is a greenish colored poisoned gas, but when united these two elements form the compound called sodium chlorid (NaCl) , salt. FOUNDATION FACTS AND PRINCIPLES 3 Chemical action. Chemical reaction is the union of two or more elements into a compound, or the separation of a compound into its elements, or the formation of new compounds from other com- pounds. In the most common chemical reactions heat is evolved. Place some coal in the stove, raise the temperature to the kin- dling point, and 32 pounds of the element oxygen entering the vent of the stove in gas form will unite with 1 2 pounds of the element car- bon in the coal and 44 pounds of the compound carbon dioxid (CO2) will form and pass off as a gas through the chimney. After this chemical reaction is completed, -the stove is found to contain only a few ounces of ashes, which represent the impurities in the coal. From this compound, carbon dioxid (CO2), which is always pres- ent in the air in small amount, all agricultural plants obtain their supply of carbon and oxygen, which together constitute about 90 per cent of the total dry matter contained in plants. Combining weights. Combining weights of elements are the relative proportions in which those elements combine to form compounds. The combining weight of the element hydrogen is smaller than that of any other element, and for this reason all other combining weights are referred to that of hydrogen as the stand- ard, or unit, of weight. The combining weight of hydrogen is i. One part of hydrogen will unite with 35.5 parts of chlorin to form the compound hydrogen chlorid (HC1), which is also properly called hydrochloric acid, and sometimes incorrectly called " mu- riatic " acid. Thus, the combining weight of chlorin is 35.5. We may take i pound of hydrogen and let it unite with 35.5 pounds of chlorin to form 36.5 pounds of the compound HC1; or we may use i ounce of hydrogen and 35.5 ounces of chlorin, or i gram of hydrogen and 35.5 grams of chlorin, or i milligram of hydrogen and 35.5 milligrams of chlorin. All that is necessary is, that we maintain these proportions. This is one of the absolute mathematical laws of chemistry and is fundamental to the prin- ciples of soil fertility and plant growth. If we try to combine 3 parts of hydrogen with 35.5 parts of chlorin, 36.5 parts of the com- pound HC1 would be formed and 2 parts of hydrogen would be left in its original form. Atoms. An atom is the smallest particle of an element. It is not known how small the atom is, but it is known that the weight of 4 SCIENCE AND SOIL an atom of carbon is 12 times, of oxygen is 16 times, of sulfur is 32 times, and of chlorin is 35.5 times, as great as the weight of an atom of hydrogen. Thus the atomic weights of all other elements are referred to the weight of the hydrogen atom as the chemical unit. Molecules. A molecule is the smallest enduring particle of an element or compound. The atom, if set free, instantly unites with another atom (either of the same element or of a different element) to form a molecule, and the molecule may endure permanently. One atom of hydrogen and one atom of chlorin unite to form one molecule of hydrochloric acid, HC1. The molecular weight of this compound 1536.5, which is the sum of the atomic weights, the weight of the hydrogen atom always being i. It is true that this is an arbitrary standard, but so is every common standard of weight or measure, such as the ounce or the inch or the dollar. The inch is an arbitrary standard of length to which we may refer other lengths or distances, and likewise the weight of the hydrogen atom is an arbitrary standard to which we may with equal accuracy refer the weights of the atoms of all other elements, and also the weights of all molecules of either compounds or elements. Atomic bonds. Atomic bonds are the links of union between atoms. This bond between atoms may be likened to the hand clasp between persons, except that under normal conditions the hand of one atom always grasps the hand of another atom. If the bond is broken, the freed hands immediately grasp other hands, breaking the bonds between other atoms if necessary to secure union. At the instant a bond is broken, when free hands exist, the atoms are called nascent, and in that condition they have unusual power to attack the molecules of other elements or compounds. Free hydrogen means hydrogen not combined with some other element. Thus we have nascent hydrogen (H), ordinary free hydrogen (H2), and- combined hydrogen, as in water (H2O). Of course, nascent hydrogen is also free hydrogen, but in an extraordinary form; namely, as a free atom, which as such can exist but an instant until it unites with another atom of hydrogen (or of some other element) to form a molecule. Valence. Valence refers to the number of bonds, or hands, pos- sessed by an atom. The hydrogen atom has but one hand (H — ) , while the oxygen atom has two hands ( — O — ), and the carbon FOUNDATION FACTS AND PRINCIPLES 5 atom has four hands (=C = ). In hydrogen molecules the atoms are always in pairs (H — H or H2) . Thus the weight of the hydro- gen molecule is two, because it contains two of the unit atoms. The atom of oxygen weighs sixteen times as much as the hydrogen atom, consequently 16 is the atomic weight, or the combining weight, of the element oxygen. The molecule of ordinary oxygen contains two atoms (O=O or O2), but there is a form of oxygen, /°\ called ozone, which contains three atoms in the molecule (O O or O3). The molecular weight of ordinary oxygen is 32, while the molecule of ozone weighs 48 times as much as one hydrogen atom. One oxygen atom has power to hold two hydrogen atoms (H — O — H or H2O). This is a compound which might be called dihydrogen oxid, but which is commonly called water. The molecular weight of water is 18, the sum of the atomic weights. Separately, hydrogen and oxygen are both gases under ordinary conditions, but the compound H2O possesses different properties, being a liquid at ordinary temperatures, although water becomes a gas at high temperature and a solid at low temperature. While it is common knowledge that this compound exists in three forms, solid, liquid, and vapor, and that it is easily changed from solid ice to liquid water and from liquid to vapor (steam), it is not so generally known that most elements and most com- pounds may exist in each of these three forms under proper condi- tions of temperature and pressure. One atom of carbon may combine with four atoms of hydrogen, ,Hk /H v forming the gas compound ( ;(Y or CH4 ) called methane or VH/ \H marsh gas, a constituent of illuminating gas, and sometimes formed in stagnant marshes. This hydrocarbon has a molecular weight of 1 6 and is the lowest in a very large series of compounds contain- ing only hydrogen and carbon. One atom of carbon with its four bonds may unite with two atoms of oxygen, forming carbon dioxid (O = C = O or CO2) , a chemical reaction which occurs in the com- bustion of coal or other substances containing carbon, as already explained. Monovalent atoms have one bond, or one hand (mono means 6 SCIENCE AND SOIL one, as in monotone); bivalent atoms have two bonds (bi or di means two) ; trivalent atoms have three bonds; tetravalent, four bonds; pentavalent, five bonds; hexavalent, six bonds; hepta- valent atoms have seven bonds; and octovalent atoms have eight bonds, with power to hold four of the bivalent atoms of oxygen. There are a few cases in which the atom does not make common use of all the bonds it possesses. Thus the nitrogen atom has five bonds, or hands, but in some compounds only three bonds are used to hold other atoms. It might be conceived in this case that the other two hands are clasped together, and this conception might even be extended to cover a molecule composed of a single biva- lent atom (such as mercury and, possibly, argon). One atom of nitrogen and three atoms of hydrogen form the compound called ammonia (NH8). This compound is frequently sold in fertilizers, but the hydrogen has no money value because water (H2O) con- tains hydrogen. The molecular weight of ammonia is 17, of which the nitrogen atom is 14 and the hydrogen atoms are 3. If a fer- tilizer is guaranteed to contain 17 per cent of ammonia, it should contain 14 per cent of the element nitrogen; while 8| per cent of ammonia is equivalent to only 7 per cent of nitrogen. Ammonia itself contains \$, or 82 per cent, of the element nitrogen. In the compound called ammonium chlorid (NH4C1), the atom of nitrogen is pentavalent; that is, it has and uses five bonds: H H ^N^-Cl. The molecular weight of this compound is 53.5 W ^H (14+4 + 35.5), and it contains 14&, or 26 per cent, of nitrogen. 53-5 Phosphorus is another element which sometimes uses only three /H bonds, as in hydrogen phosphid, P^-H, and in phosphorus trichlorid, , Pe-Cl, and sometimes five bonds, as in phosphorus pentachlorid, XC1 CL ,C1 yP^-Cl. Thus, the hydrogen phosphid contains f|, or 91 per CK XC1 FOUNDATION FACTS AND PRINCIPLES 7 cent, of phosphorus (the atomic weight of phosphorus being 31), while the phosphorus pentachlorid contains less than 15 per cent of phosphorus. Nitrogen and phosphorus are in some resp'ects very much alike, and in other respects they are very unlike. They are the two most precious elements of plant food, and they deserve from the author and from the reader all of the consideration they are to receive in this book. The gas law. This law is that equal volumes of gases under like conditions of temperature and pressure contain the same number of molecules. In other words, in the gas form, every molecule occupies, or controls, the same amount of space. Thus, the hydro- gen molecule, with a weight of 2, occupies as much space as the oxygen molecule, which weighs 32, or the molecule of carbon dioxid, weighing 44, or of sulfur dioxid with a molecular weight of 64. (The atomic weight of sulfur is 32.) If a 6-gallon bottle holds 2 grams of hydrogen (H2), it will hold 32 grams of oxygen (O2), 28 grams of nitrogen (N2), 16 grams of methane (CH4), 17 grams of ammonia gas (NH3), 44 grams of carbon dioxid (CO2), 64 grams of sulfur dioxid (SO2),and a gram- molecule (the molecular weight in grams) of any other gas. This law does not apply to liquids or solids, but only to gases. Chemical symbols. A symbol is used to represent one atom of an element. H stands not only for the element hydrogen, but also for one atom of hydrogen with a combining weight of i. Likewise S stands for sulfur and for one atom of sulfur, and represents a weight of 32. Chemical formulas. A formula is used to represent a molecule and shows the kind and the number of atoms contained in the molecule. The formula H2O represents one molecule of water, containing two atoms of hydrogen, each having a combining weight of i, and one atom of oxygen, with a weight of 16. Thus, the molecular weight of water is 18. The formula Ca3(PO4)2 (read: Ca three, PO four, twice) represents one molecule of tricalcium phosphate, the valuable phosphorus compound contained in bones and in natural phosphate rock. The metallic element calcium (Ca) is also contained in limestone, which is calcium car- bonate (CaCO3), and in quicklime, or burned lime, which is calcium 8 SCIENCE AND SOIL oxid (CaO). The atomic weight of calcium is 40. The subscript figures used in a chemical formula always refer to the preceding element, or to the inclosed group of elements if parentheses are used. In the calcium phosphate, Ca3(PO4)2, the subscript 3 shows that three atoms of calcium are contained in the molecule. The parentheses are used to inclose a group of atoms (one atom of phosphorus and four atoms of oxygen) which occurs in many other compounds, as in H3PO4 (phosphoric acid), FePO4 (iron phosphate) , etc. (Fe is from ferrum, the Latin word for iron, and I is the symbol used for the element iodin.) The subscript 2 follow- ing the parenthesis in Ca3(PO4)2 means that the molecule contains the PO4 group twice, and for this reason Ca3(PO4)2 is a better for- mula than Ca3P2O8, which may also be correctly used. A mole- cule of tricalcium phosphate contains three atoms of calcium (3 x 40 = 120), two atoms of phosphorus (2 x 31 = 62), and eight atoms of oxygen (2 x 4 x 16 = 128), and the molecular weight is 310 (120 + 62 + 128), of which the phosphorus represents only 62. Thus, tricalcium phosphate contains -/^$, or 20 per cent, of the element phosphorus. In other words, one fifth of pure tricalcium phosphate is phosphorus, while the remaining four fifths consist of calcium and oxygen in nearly equal parts. The law of constant proportions. This law is that in every chemi- cal combination the constituents occur in definite and constant proportions by weight. The percentage of phosphorus in pure tricalcium phosphate is absolutely constant. It matters not whether the compound is made in the United States, in Germany, or in Japan, nor whether it is obtained from bones or from phos- phate rock, the percentage of phosphorus it contains is always the same, if the compound is pure. This percentage is exactly 20, according to the most accurate accepted determinations. The atomic weights are determined by several different methods, but, even with the finest and most accurate balances and other instru- ments and means, absolute exactness may not be achieved, be- cause of the human error. No man can measure a mile with absolute exactness, because two different measurements made by one man may vary by an inch, a tenth of an inch, or, at least, by a hundredth or a thousandth of an inch. According to the chemical law, the proportion of the different FOUNDATION FACTS AND PRINCIPLES 9 elements in any pure compound is absolutely definite and constant, in strict accordance with the chemical law, and the proportion can be determined with ten times the degree of accuracy required for all practical purposes ; nevertheless, the determination may not be absolutely exact. While all atomic weights are essentially referred to hydrogen as unity, the mathematical basis is exactly 16 for the atomic weight of oxygen, because the element oxygen constitutes in quantity one half of the earth's crust (including the air, the ocean, and the solid crust to a depth of ten miles), and forms compounds with nearly all other elements, thus affording closer comparisons than hydrogen. The known chemical elements. In the accompanying table is the complete list of 80 known elements. For convenience the more common elements, that every one should know, are given in one group, and the rare elements, that few people have ever seen, are grouped by themselves. The symbols used for the chemical elements are essentially the same in all languages. In a few cases where the modern name varies in different languages, the nations have agreed upon a symbol derived from the Latin name of the element; as, for example, Fe for iron (ferrum in Latin), K for potassium (kalium), and Na for sodium (natrium). Some of the atomic weights of the rare elements have not yet been determined with a sufficient degree of accuracy to justify assigning a more specific value than the nearest whole number. Further investigation must determine whether such whole numbers are correct. It is a noteworthy fact that the chemists of the world are agreed that, of the forty or more common elements, sixteen have atomic weights that differ from whole numbers by less than .05, and twelve others differ only by about .1, thus showing twenty-eight of the best-established atomic weights apparently grouped with relation to the unit, with only twelve scattering ; and some of these (as nickel) are doubtful, while others are high atomic weights with consequent possibilities of error, the accepted atomic weights of gold and platinum both having been changed by .5 within recent years. These data and the recognized periodic law, that the properties of the elements are periodic functions of their 10 SCIENCE AND SOIL TABLE i. ELEMENTS, SYMBOLS, AND INTERNATIONAL ATOMIC WEIGHTS FOR 1909. (Decimals, including ciphers, indicate supposed accuracy.) THE MORE COMMON ELEMENTS THE RARER ELEMENTS Name Symbol Atomic Weight Name Symbol Atomic Weight Aluminum .... Antimony (Stibium) . Argon Al Sb A As Ba Bi B Br Cd Ca C Cl Cr Co Cu F Au H I Fe Pb Li Mg Mn Hg Mo Ni N O P Pt K Si Ag Na Sr S Sn Ti Zn 27.1 120.2 39-9 7S-o 137-4 208.0 II. 0 79-9 112.4 40.1 12.0 35-5 52.1 59-Q 63.6 19.0 197.2 1.008 126.9 55-9 207.1 7.0 24-3 54-9 2OO.O 96.0 58.7 I4.O IO.OOO 3I.O I9S.O 39-i 28.3 107.9 23.0 87.6 32.1 119.0 48.1 65-7 Caesium .... Cerium Cs Ce Cb Dy Er Eu Gd Ga Ge Gl He In Ir Kr La Nd Ne Os Pd Pr Ra Rh Rb Ru Sa Sc Se Ta Te Tb Tl Th Tm W U V Xe Yb Y Zr I32-8 140.3 93-5 162.5 167.4 152 157-3 69.9 72-5 9.I 4.0 114.8 I93-I 8l.8 139.0 144-3 2O 190.9 106.7 140.6 226.4 102-9 85-5 IOI-7 150.4 '44.1 79-2 181 127-5 159-2 204.0 232.4 168.5 184.0 238-5 51-2 128 172.0 89.0 90.6 Columbium Dysprosium . . . Erbium Arsenic Barium Bismuth Europium .... Gadolinium . . . Gallium .... Germanium Glucinum .... Helium '. Boron Bromin Cadmium .... Calcium Carbon Chlorin Indium Chromium .... Cobalt Iridium .... Krypton .... Lanthanum . . . Neodymium . . . Neon Copper (Cuprum) Fluorin Gold (Aurum) . . . Hydrogen .... lodin Osmium .... Palladium .... Praseodymium . . Radium .... Rhodium .... Rubidium .... Ruthenium . . . Samarium .... Scandium .... Selenium .... Tantalum .... Tellurium .... Terbium .... Thallium .... Thorium .... Thulium .... Tungsten (Wolfram) Uranium .... Vanadium .... Xenon Iron (Ferrum) . . . Lead (Plumbum) . . Lithium .... Magnesium . . . Manganese . . . Mercury (Hydrar- gyrum) .... Molybdenum . . . Nickel Nitrogen .... Oxvffen . Phosphorus . . . Platinum .... Potassium (Kalium) . Silicon Silver (Argentum) Sodium (Natrium) . Strontium .... Sulfur . . Tin (Stannum) .. . Titanium .... Zinc Ytterbium .... Yttrium .... Zirconium .... FOUNDATION FACTS AND PRINCIPLES n atomic weights,1 together with the recently discovered radium and radio activity, and the evidences 2 of accomplished transformation of one element into another, strongly indicate a common origin for different elements, and lend to the subject a present-day in- terest as intense as ever moved the alchemist to try to turn the baser metals into gold. 1 It is worth while to note some relations that exist between the monovalent elements fluorin, chlorin, bromin, iodin; between the bivalent elements oxygen, sulfur, selenium, molybdenum; between the trivalent (or pentavalent) nitrogen, phosphorus, arsenic, antimony; and also between the tetravalent carbon, silicon, titanium, and germanium: Fluorin Chlorin Bromin Iodin F = i9 01 = 35.5 Br = 8o 1 = 127 HF HC1 HBr HI Oxygen Sulfur Selenium Molybdenum O = 16 8 = 32.1 Se = 79.2 Mo = 96 H2O H2S H2Se H2SeO4 H2MoO4 Nitrogen Phosphorus Arsenic Antimony N = 14 P = 31 As = 75 Sb = 120.2 NH3 PH3 AsH3 SbH3. N2O5 P2O5 As2O6 Sb2O6 Carbon Silicon Titanium Germanium C = 12 Si = 28.4 Ti = 48 Ge = 72.5 CO2 SiO2 TiO2 GeO2 Aside from the similarity of valence and other properties and of compounds formed, there is interest in the relation of atomic weights, especially in the second and fourth groups, and, even in the fact that the atomic weight of antimony is so nearly the sum of the other three in the group. 2 In 1907 Ramsay and Cameron, of England, reported that they had reduced copper, in the presence of radium emanation, into other elements of the same series: potassium, sodium, lithium. (See Nature, July, 1907, and Journal of the Chemical Society, September, 1907.) The correctness of Ramsay and Cameron's experiments has been called in question by Mme. Curie and Mile. Gleditsch; Comptes rendes, 147, 345 (1908); Science, December 4, 1908. CHAPTER II THE MORE COMMON ELEMENTS AND COMPOUNDS Important elements. Fifteen elements are of special interest and importance in the study of soil fertility, because they are commonly found in plants and animals and because they so largely constitute the soil and air and ocean and the common things of earth. Of these fifteen elements, ten are known to be essential to plant growth; eight of them constitute 98 per cent of the solid crust of the earth; four of them constitute 99.6 per cent of the ocean, about 96.4 per cent being water (H2O) and 3.2 per cent common salt (NaCl); and two of them (nitrogen and oxygen) constitute 98.5 per cent by weight of the dry atmosphere, about 1.5 per cent of the air consisting of the recently discovered element, argon. The ten essential elements of plant food may be grouped as follows : C, H, O, obtained by plants from air and water. P, K, N, sometimes deficient in soils, and of money value as plant food. S, Ca, Fe, Mg, required in small amounts and not likely to be deficient in soils. The five other elements commonly present in plants are silicon, aluminum, sodium, chlorin, and manganese. The reader is earnestly advised to learn by groups 1 the name and atomic weight and the valence of each of these important elements, and the following table is constructed for this purpose. Aside from the name, symbol, atomic weight, and valence, Table 2 furnishes some extremely valuable and useful information concerning the occurrence and relative abundance of the elements which essentially constitute the crust of the earth, the ocean, the air, and the agricultural plants and animals. These data are based 1 The author consents to the students' memory key: "C. HOPK'NS' CaFe,-Mg, " if Mg stands for "Mighty good " and the omission of I, for modesty. upon the recent computations of Professor F. W. Clarke of the United States Geological Survey for the average composition of the ocean and of the earth's crust * to a depth of ten miles, essen- tially upon Sir William Ramsay's recent estimate for the average composition of air, and upon the author's compilations and com- putations for the average composition of shelled corn (maize) . TABLE 2. THE MORE IMPORTANT ELEMENTS NAME SYMBOL ATOMIC WEfGHT (0=i6) VALENCE, OR NUMBER OF BONDS OCCURRENCE In Earth's Crust (Per Cent) In the Ocean (Per Cent) In the Air (Per Cent) In the Corn Ker- nel (Per Cent) Oxygen . . Carbon . . Hydrogen o c II 16 12 I 2 4 I 47.07 .20 .22 85-79 23.00 .OI 46.000 45.000 6.400 10.67 Nitrogen Phosphorus Potassium . N P K 14 31 39 3 or 5 3 or 5 i trace .11 2.46 75-50 1.760 .300 •340 .04 Magnesium Calcium . . Iron . . . Sulfur . . Mg Ca Fe S 24-3 40 56 32 2 2 2, 3, 6, or 7 2, 4, or 6 2.40 3-44 4-43 .11 .14 •05 •125 .022 .008 .004 .09 Silicon . . Aluminum . Sodium . . Chlorin . . Manganese Si Al Na Cl Mn 28.3 27 23 35-5 55 4 3 i i, 3> 5, or 7 2, 3, 6, or 7 28.06 7.90 2-43 .07 .07 .014 I.I4 2.O7 .013 .013 Titanium Argon . . Ti A 4? 40 4 (?) .40 1.48 'Totals 99.36 2 99-99 99-99 3 99-99 1 United States Geological Survey Bulletin 330 (1908). The data for phos- phorus and potassium include analyses of 1671 and 2110 different samples, re- spectively, of representative rocks, some of which were kindly furnished to the author by Doctor Clarke since Bulletin 330 was published. 2 About sixty other elements (most of them very rare) must account for this deficiency. 3 Constant traces of helium, neon, krypton, and xenon are also found in the air, of which they may constitute five parts per million. Varying amounts of moisture, compounds of nitrogen, sulfur, chlorin, and more or less dust, also exist in the air. 14 SCIENCE AND SOIL It may well be stated here that plants secure their supply of both carbon and oxygen from the carbon dioxid of the air. The .01 per cent of carbon (C = 12) shown in the table is equivalent to nearly .04 per cent of carbon dioxid (CO2 = 44). The hydrogen of plants is taken from the water absorbed by the roots. The corn plant secures its supply of nitrogen from the " trace " contained in the earth's crust, which, however, amounts to about .25 per cent, in the tilled stratum of a good soil. Under proper conditions legume plants secure more or less of their nitrogen from the air. The remaining six essential elements are secured only from the soil by all plants. Of the atmosphere, ocean, and solid crust (ten miles deep), the solid crust constitutes about 93 per cent of the whole ; while the entire atmosphere amounts to only .03 per cent. These addi- tional facts make possible a mathematical comparison between the supply and crop requirements of carbon and oxygen (in CO2) and nitrogen in the air, and emphasize the importance of the carbon cycle and of the circulation of some other elements, all of which is more fully discussed and explained in the following pages. A ready working knowledge, sufficient for everyday use, lies at the basis of success in every industry and profession. It is worth while to have in mind a few fundamental facts relating to the seven- teen elements named in Table 2, which constitute more than 99 per cent of earth, sea, and air, and of all plants and animals. Nothing can be made of nothing. Compounds consist of two or more elements, and the molecule of a compound must contain two or more atoms. If one knows the valence of the elements, he is then in control of much information of very great value in relation to compounds. Valence is the key to the understanding of compounds and chemical reactions. Table 2 gives this information for the very important elements. Three of these elements — hydrogen, potassium, and sodium — have only one bond, or hand, for each atom (H — , Na — , and K — ); while chlorin (Cl) may use i, 3, 5, or 7 bonds. Three other elements have only two bonds for each atom (O = , Mg = , and Ca=), these elements being strictly bivalent. Sulfur sometimes uses only two bonds (in H2S and CS2), but may use four or six. THE MORE COMMON ELEMENTS AND COMPOUNDS 15 Iron and manganese are alike peculiar, with a valence of 2, 3, 6, or 7. Aluminum is a trivalent element with three-handed atoms, while nitrogen and phosphorus may use either three or five bonds. This leaves only the strictly tetravalent family, carbon (= C = ), sil icon ( = Si = ) , and titanium ( =Ti = ) .* When an iron atom uses three hands, it is called -ic iron, but if it uses only two hands it is called -ous iron. Thus we have the ferrous chlorid (FeCl2) and ferric chlorid (FeCl3) ; also ferrous oxid (FeO) and ferric oxid (Fe2O3). Likewise, when phosphorus uses only three bonds, it is called -ous phosphorus, but with the five bonds in use it is -ic phosphorus, as in phosphorous chlorid (PC13) and phosphoric chlorid (PC15), which are also called phosphorus tri- chlorid and phosphorus pentachlorid, the endings, -ous and -ic, being unnecessary when the valence is specified in the number of chlorin atoms held. Matter may exist in three distinctly different forms or classes which might be called " wonary " (single), "fo'nary" (double), and " trinary," or ternary (triple). First, matter may exist in the form of free or uncombined ele- ments; as solid metallic iron, aluminum, magnesium, calcium, sodium, or potassium; as solid nonmetallic carbon, phosphorus, sulfur, or silicon; as liquid mercury or bromin; or as gaseous oxygen, nitrogen, hydrogen, or chlorin. This might be termed the " wonary " form, all atoms in the molecule being of the one element. Second, matter may exist in binary compounds ; that is, with 1 Argon is of interest chiefly because it is so very common and yet so recently discovered. Argon is everywhere present in the air and we respire more than an ounce a day of that element. It is an invisible gas, and because it is mixed with so much nitrogen (which it resembles somewhat) and oxygen, and cannot be seen, it would be less easily discovered than many other elements; but the chief rlimculty in detecting it by chemical methods was due to its chemical inaction. Because of this inaction, it has been named argon, which means without action. When the ele- ment was discovered and very thoroughly investigated, the discoverers (Rayleigh and Ramsay, in 1894) concluded that the argon atom has no valence, — no hand with which to clasp the hand of another atom, either of argon or of any other element. In other words, they discovered that argon forms no compounds, and that the molecule of argon is monatomic (man or mono means one, as in monotone = one tone). Later investigations indicate, however, that argon has some weak affinities. 1 6 SCIENCE AND SOIL two elements represented in the molecule. In the name of such a compound both elements are expressed, and sometimes the name also includes the number of atoms of each of the elements in the molecule, as phosphorus pentachlorid (PC16). As a rule, the name of one element is modified slightly so as to end in -id, a termination that means that the compound is binary, containing but two elements. (Any exception to this rule will be self-explanatory.) Thus sodium chlorid must contain only two elements, sodium and chlorin, because of the names and the ending -id] and, since sodium must be monovalent and chlorin may be, the formula for the molecule is probably Na — Cl, which is correct for com- mon salt. As a matter of fact, chlorin is always monovalent in binary compounds with metals. Calcium oxid (quicklime) must be a binary compound of the two strictly bivalent elements, calcium and oxygen, and the molecular formula may be Ca=O, which is also correct. Calcium chlorid should be Ca=Cl2; and magnesium chlorid, Mg=Cl2; and potassium oxid, K2=O; hydrogen sulfid, H2 = S; sulfur dioxid, SO2 (O = S=O); and sulfur trioxid, SO3 (in which the sulfur atom must use six bonds), all of which are correct. More complex molecules, which, however, are easily understood, are aluminum oxid (A12O3) and phosphorus pentoxid (P2O6) . The aluminum atom has three bonds (or hands) , and the phosphorus uses five bonds in this oxid, while the oxygen atom is always bivalent, having but two hands. There are six bonds of union in aluminum oxid and ten bonds in phosphorus pentoxid, p=o thus: A1;>O, /O. Thus, if one knows the name, the atomic A\0 P=0 %0 weight, and the valence of the fifteen most important elements, he has the key to the formula and percentage composition of their binary compounds. Third, matter may exist in ternary (triple) compounds, with three elements represented in the molecule. Most ternary com- pounds contain oxygen, and, in naming such compounds, the most common rule is to express the names of only two of the elements THE MORE COMMON ELEMENTS AND COMPOUNDS 17 and then to employ a short ending to designate the oxygen. Thus, the ending -ate commonly means oxygen. Ternary com- pounds may best be studied in groups, in which the relation of oxygen to one of the other elements is constant. Thus, the car- bonates constitute a large class of compounds in which the group, or radicle (=CO^), is always present. The structural formula for ~°\ this radicle is as follows: yC = O. From this it will be seen -0/ that the carbonate radicle has two free hands, or bonds, capable of holding two monovalent atome or one bivalent atom. Now, the fact is, that almost any metallic element can join hands with this radicle. Thus we have calcium carbonate (CaCO3), magnesium carbonate (MgCO3) , ferrous carbonate (FeCO3) , sodium carbonate etc. The nitrate radicle is — NO3, as in sodium nitrate ( The chlorate radicle is — C1O3, as in potassium chlorate (KC1O3). The silicate radicle is =SiO3 (like the carbonate radicle =CO3). The sulfate radicle is =SO4, as in calcium sulfate (CaSO4). The phosphate radicle is =PO4, as in ferric phosphate (FePO4). If we can remember these six radicles, we have the key to the constitution and composition of a large number of ternary com- pounds, some of which are of the greatest importance in soil fer- tility; as, for example, limestone, which is calcium carbonate (CaCO3) ; land plaster, which is calcium sulfate (CaSO4) ; and the important compound in phosphate rock and in bones, called " bone phosphate," which is calcium phosphate, Ca3(PO4)2, also properly called tricalcium phosphate. When the element chlorin, or the element sulfur, or any of these radicles join hands with metallic elements, the resulting compound is called a salt; as NaCl (common salt),Na2SO4 (Glauber's salt), MgSO4 (Epsom salt) ; and even limestone (CaCO3) may properly be called a salt of calcium, and ferrous sulfate (copperas, which, however, contains no copper) is a salt of iron (FeSO^). Oxids and hydroxids. As already explained, matter may exist in the form of free elements, as nitrogen (N2), sulfur (S2), or phos- phorus (P4). 1 8 SCIENCE AND SOIL Oxygen and hydrogen are not only very abundant elements (water, the " universal solvent," being H2O), but they are also very active, chemically, and one or both have some part in nearly all important groups of compounds. Oxids are binary compounds of oxygen with other elements, and they constitute a very large class, because almost all other elements form compounds with oxygen. Important examples are common quartz sand, which is silicon dioxid (SiO2) ; water itself, which is hydrogen oxid (H O) ; and carbon dioxid (C02). Hydroxids are compounds which always contain the radicle — OH, which is known as hydroxyl, or the hydroxid group, or the hydroxid radicle. This group, consisting of one atom of oxygen holding one atom of hydrogen with one hand and with the other hand free, or clasping some other atom, is the most important group of atoms in all chemistry. As a group it is monovalent, having one free bond, and it unites with almost all elements. Thus, we may consider: Hydrogen hydroxid, HOH, or water. Potassium hydroxid, KOH. Calcium hydroxid, Ca(OH)2. Iron hydroxid, Fe(OH)3. Silicon hydroxid, Si(OH4). Phosphorus hydroxid, P(OH)5. Sulfur hydroxid, S(OH)6. Ethyl hydroxid, C2H5OH, or alcohol. In these compounds, the valence of the different elements varies from i to 6, and a corresponding number of hydroxyl groups ( — OH) may be held. While these various hydroxids, as KOH, are not strictly binary compounds, the unbroken — OH group acts much like an element and the ending -id is used for these com- pounds. This cannot be misunderstood, because potassium hy- droxid (for example) plainly indicates the three elements, potas- sium, hydrogen, and oxygen. NOTE. The meanings of a few word endings and prefixes are noted here for reference : THE MORE COMMON ELEMENTS AND COMPOUNDS 19 The ending -ic means common, as in ferric chlorid (FeCl3), and sulfuric acid (H2SO4). The ending -ous means less, as less chlorin, in ferrous chlorid (FeCta), and less oxygen, in sulfurous acid (H2SO3). The ending -ate means common; usually suggests oxygen, as in sodium nitrate (NaNO3). The ending -lie means less, as in sodium nitrite (NaNO2).~ The prefix hypo- means still less, as in hyposulfurous acid (H2SO2), or in sodium hyposulfite (Na2SO2). The prefix per- means more, as in persulfuric acid (H2S2O8). The prefix hydro- means hydrogen and no oxygen, as in hydrochloric acid (HC1). The ending -ic in the names of acids without the prefix hydro- suggests oxygen, as in sulfuric acid (H2SO4). The ending -id is used for binary compounds. Hydro-ic acids yield -id salts; other -ic acids yield -ate salts; and -ous acids yield -ite salts. Hydroxid oxids. We have, not only oxids and hydroxids, but also compounds that are part oxid and part hydroxid, as shown by the structure formulas given in the first column in the following classified list: C1OH . . . Hypochlorous acid, HC1O. OC1OH . . Chlorous acid, HC1O2. O2C1OH . . Chloric acid, HC1O3. OgClOH . . Perchloric acid, HC1O4. S(OH)2 . . Hyposulfurous acid, H2SO2. OS (OH) 2 . . Sulfurous acid, H2SO3. O2S(OH)2 . Sulfuric acid, H2SO4. ONOH . . Nitrous acid, HNO2. O2NOH . . Nitric acid, HNO3. Here we have an atom of chlorin, sulfur, or nitrogen, one or more oxygen atoms, and also one or more hydroxyl groups, in the same molecule, and the increasing valence of certain elements is illustrated, — that of chlorin from i (in HC1) to 3, 5, and 7, in the compounds shown; that of sulfur from 2 (in H2S and CS2) to 4 and 6, in the compounds here shown; and nitrogen with three and with five bonds. The formulas, showing both oxid and hydroxid characters, can 20 SCIENCE AND SOIL all be derived from the corresponding hydroxids by subtracting water (H2O), thus: C1(OH)8 yields OC1OH and H2O. C1(OH)6 yields O2C1OH and 2 H2O. C1(OH)7 yields O3C1OH and 3 H2O. S(OH)4 yields OS(OH)2 and H2O. S(OH)6 yields O2S(OH)2 and 2 H2O. (Sulfur hexahydroxid) (Sulfuric acid) (Water) / . , , OV:/OH HO>S04 = CaS04 + 2 H2O. X|OH Hi' The dotted line shows how the two compounds are broken. The two hydrogen atoms are broken off from the sulfuric acid and the two hydroxyl groups are broken off from the calcium hydroxid, and each hydrogen atom (H — ) immediately joins hands with a hydroxyl group ( — OH) and thus are formed two molecules of water, 2(H — O — H). The bivalent calcium atom (Ca=) also grasps the two free hands of the radicle =SO4, and forms the compound CaSO4, which is calcium sulfate. Thus, from a base (alkali) and an acid, we have formed water and a neutral salt, neither acid nor alkaline. This is a typical reaction between a hydroxid acid and a hydroxid base, resulting in the formation of a salt. Many other salts may be formed by similar reactions between basic hydroxids, oxids, or free metals, on the one hand, and different kinds of acids, on the other hand, which may contain oxygen in the form of hydroxid only, or in both hydroxid and oxid form, or the acid may contain no oxygen, but only hydrogen joined directly to some nonmetallic element, as in hydrochloric acid (HC1) or hydrosulfuric acid (H2S). Salts are the most abundant and important compounds. The earth's crust consists largely of insoluble mineral salts, called sili- cates. These are, as a rule, very complex salts, one of the simplest being common felspar, potassium aluminum silicate, KAlSi3O8. This double salt is the principal source of potassium, one of the essential elements of plant food. Common salt is sodium chlorid, NaCl. Epsom salt is magnesium sulfate, MgSO4. Glauber salt is sodium sulfate, Na2SO4. Even limestone is a kind of salt, calcium carbonate, CaCO . For convenience, the formulas of salts containing oxygen are 22 SCIENCE AND SOIL written with the oxygen in one group, as Na^SO^ but it may well be remembered that oxygen is regularly the linking element be- tween the metal and the nonmetal, and the structure of the mole- cule is better shown if written thus: A few more equations for typical reactions will show the various ways by which salts may be formed. Other similar acids and bases give similar products : Ca(OH)2 + H2SO4 = CaSO4 + 2 H2O. KOH + HNO3 = KNO3 + H2O. These equations represent, first, the reaction between the cal- cium hydroxid and sulfuric acid with the formation of calcium sulfate (gypsum, or land plaster), and, second, the reaction be- tween potassium hydroxid and nitric acid with the formation of potassium nitrate (saltpeter). In place of a hydroxid base, we may use an oxid of some metal, thus: CaO + HjSO* = CaSO4 + H2O. CaO + 2 HNO3 = Ca(NO3)2 + H2O. K2O + H2SO4 = K2SO4 + H2O. The salts formed are calcium sulfate, calcium nitrate, and potas- sium sulfate, and in each reaction only one molecule of water is formed, when the oxid base is used, in place of two molecules of water from the use of a bivalent hydroxid base. The base may be not only in the form of an oxid or hydroxid of the metal, but we may use the free metal itself as a base, thus: Fe + H2S04 = FeS04 + H9. Mg + 2 HN03 = Mg(N03)2 + H2. In these reactions the hydrogen atoms are displaced by the metal and liberated as a gas. While most acids contain oxygen either as a hydroxid (C1OH) or a hydroxid oxid (OC1OH orO2ClOH), a few acids, with the prefix hydro-, contain no oxygen; as, for example, hydrochloric acid THE MORE COMMON ELEMENTS AND COMPOUNDS 23 (HC1); but these acids also react with bases (metals, metallic oxids, or metallic hydroxids), as shown above for the oxygen acids, thus: Fe + 2 HC1 = FeCl2 + H2. CaO + 2 HC1 = CaCl2 + H2O. NaOH + HC1 = NaCl2 + H2O. In these reactions the salts formed are chlorids. The last equa- tion shows the reaction between sodium hydroxid, which is ordi- nary concentrated lye, a poisonous substance, and hydrochloric acid, which is also a poisonous substance; and the salt formed is sodium chlorid (NaCl), the common table salt, and the most abundant salt of the ocean. Hydrosulfuric acid (H2S) also forms many salts, called sulfids, as iron sulfid (FeS), calcium sulfid (CaS), etc. In all of these reactions the number of pounds of any base re- quired to react with a given amount of any acid is easily and accu- rately computed, and the amount of the salt to be formed and of the water or hydrogen to be liberated can also be told in advance, if one knows the atomic weights and the equation for the reaction. Thus, 40 pounds of sodium hydroxid will react with 36.5 pounds of hydrochloric acid, and form 58.5 pounds of common salt and 1 8 pounds of water. It will be seen that 76.5 pounds of materials are used, and that 76.5 pounds of products are formed. In like manner, all such equations must balance. In other words, the num- ber and kind of atoms and the total quantity of materials put into the reaction on one side of the equation must be exactly the same as appear in the products on the other side of the equation. Acid salts are salts which still contain some of the hydrogen of the acid from which they were made. Thus, if we supply only one half as much potassium as would be required to react with a cer- tain amount of sulfuric acid, if an acid salt is possible, it will be formed : KOH + H2SO4 = KHSO4 + H2O. The sulfuric acid molecule has two hydrogen atoms, and, to make neutral potassium sulfate, both hydrogen atoms must be replaced with potassium atoms; but in the equation here shown, only one potassium atom is provided. Consequently, only one 24 SCIENCE AND SOIL hydrogen atom is replaced by potassium, and the compound formed is acid potassium sulfate (HKSO4 or KHSO4) . The charac- ter of this compound is one half acid and one half salt. If applied to the soil, it would tend to make the soil sour, or acid, which is also true of the common acid phosphate of the fertilizer trade, which is further discussed under phosphorus. Common acids are so few in number that it is well worth while to memorize the formulas. The following are important in the study of soil fertility : HC1 . . . Hydrochloric acid (no oxygen)-. HNO3 . . Nitric acid. HNO2 . . Nitrous acid (less oxygen) . H2SO4 . . Sulfuric acid. H8PO4 . . Phosphoric acid. H2CO3 . . Carbonic acid. HoSiOo . . Silicic acid. a O H2S . . . Hydrosulf uric acid (no oxygen) . The salts made from these acids, and their derivatives, by reac- tion with the bases represented by the six metals (iron, aluminum, calcium, magnesium, potassium, and sodium) constitute, in large part, the solid crust of the earth and the salts of the sea. One non- metallic oxid (SiO2) and two oxids of metallic elements (Fe2O3 and A12O3) are also found native in very considerable amounts. Some acids are strong and some are weak. The weakest acid is carbonic (H2CO3), but silicic (H2SiO3) and hydrosulfuric (H2S) are also weak acids; while hydrochloric (HC1), nitric (HNO3), sulfuric (H2SO4) , and phosphoric (H3PO4) are all very strong acids. A strong acid may take a base away from a weak acid, thus: CaCO3 + H2SO4 = CaSO4 + H2CO3. Here we have the strong sulfuric acid reacting with calcium carbonate (limestone) to form a neutral salt, calcium sulfate, and the weak carbonic acid. The carbonic acid is so weak that it may break in two, forming water and carbon dioxid: HoCOq = U0O -+- CO0. 202 & THE MORE COMMON ELEMENTS AND COMPOUNDS 25 Thus we may apply limestone (CaCOs) to an acid soil (containing organic acids, silicic acid, or acid silicates) , and the soil acids will take the base (calcium), and the liberated carbonic acid will break in two, the gas carbon dioxid (CO2) passing out of the soil into the air and thus leaving the soil with no acid in it. CHAPTER III PLANT FOOD AND PLANT GROWTH Oxygen. Oxygen is the most abundant element. It constitutes about one half the sum of all known matter. It forms chemical compounds with nearly all other elements,1 and is, in consequence, termed a chemically active element. In the free state (O2), it is a gas. In this form it constitutes about 23 per cent of the air. It is thus present everywhere and ready to form compounds with other elements, or to attack other compounds under favorable conditions. The compounds formed with oxygen may at ordinary temperatures be gases, such as carbon dioxid (CO2), liquids, such as water (H2O), solids, such as iron oxid (Fe2O3). All ordinary combustion consists of chemical reaction with oxygen, and the principal products formed are carbon dioxid and water. Water is eight ninths oxygen, and carbon dioxid is eight elevenths oxygen, as any one can determine for himself if he knows the atomic weights given in Table 2. The grain of corn is nearly one half oxygen (46 per cent). Carbon. This is a very common element, but not very abundant as compared with nine other elements. Even titanium, a tetrava- lent element belonging to the same periodic group as carbon and silicon, is one half more abundant than carbon in the earth's crust. But titanium has no agricultural value, while carbon is one of the most important elements in the structure of plants and ani- mals. About 45 per cent of the corn kernel is carbon. Carbon in the free state is the principal element in coal and char- coal. Soft coal (bituminous) contains about 90 per cent of carbon, and hard coal (anthracite) contains about 97 per cent of carbon. Graphite, the " lead " used in lead pencils, is not lead, but car- 1 No oxygen compounds are known with fluorin, argon, or helium. 26 PLANT FOOD AND PLANT GROWTH t 27 bon. It differs in some manner from the carbon in coal or char- coal, probably because of a different number, or a different arrange- ment, of the atoms in the molecule. The diamond is very pure carbon in crystallized form. Both the diamond and ordinary carbon may be converted into graphite, and small diamonds have been made artificially from ordinary carbon. Free carbon in any form burns with oxygen to form carbon dioxid (CO2), but with graphite and the diamond the reaction occurs only at higher temperatures than are necessary with ordinary carbon. Carbon in the free state has never been liquefied, but it has been Volatilized at the temperature of 3500° Centigrade (6332° Fahrenheit). On the other hand, many of the compounds with carbon are liquids or gases at ordinary temperatures. Examples of liquid compounds are carbon disulfid (CS2), benzene (C6H6), and many other hydrocarbons contained in petroleum; while carbon dioxid (CO2), methane (CH^), and acetylene (C2H,j) are well-known gases containing carbon. In the solid form, carbon occurs not only in the free state, but also in compounds, of which two very important groups are the carbonates and the carbohydrates. Marble, limestone, chalk, and marl are different forms of calcium carbonate (CaCO8), one of the few compounds with a molecular weight of 100, and thus containing 12 per cent of carbon, 40 per cent of calcium, and 48 per cent of oxygen. Of these four materials marble is nearly pure calcium carbonate, while the others may be nearly pure or very impure. When calcium carbonate is heated, two bonds are broken and then joined in another way, so that the one compound (CaCO8) is broken into two (CaO) and (CO2), thus: = 0 becomes Ca^O, or CaCO3 = CaO + CO2, The calcium oxid (CaO) is burned lime, or quicklime, which remains in the kiln; while carbon dioxid (CO2) is a gas which passes off into the air. It is easy to see that only 56 pounds of quicklime could be made from 100 pounds of pure limestone. 28 . SCIENCE AND SOIL Limestone is a constituent of good soils, and if limestone is not present in a soil, then it should be applied; for, as already explained, carbonic acid is the weakest of all acids, and, consequently, if the soil contains limestone, it cannot be an acid soil, because the soil acids will take the base away from calcium carbonate (and also from any other carbonate), and the liberated carbonic acid is broken up into water and carbon dioxid. Magnesium carbonate and iron carbonate are found in rock deposits; while potassium carbonate (K2COg) is the lye (alkali) obtained from wood ashes, and sodium carbonate (Na2CO3) is the most harmful alkali in alkali soils. The carbonic acid is so weak that the carbonates of the strongest bases (potassium and sodium) are almost as basic (alkaline) as the hydroxids of the same ele- ments (KOH and NaOH). The term hydrate used in the name of a chemical compound means that it contains water combined with some other constitu- ent, or that hydrogen and oxygen are present in the same propor- tion as in water (H2O). Thus, carbohydrates contain carbon and water. This very important group of carbon compounds, includ- ing sugar, starch, and cellulose (wood fiber) will be explained after hydrogen has been discussed. Hydrogen. Hydrogen is the third most important element in plants, constituting about 6.4 per cent of the corn kernel. Water is the only abundant source of hydrogen, although the element is found in the earth's crust in appreciable amount, chiefly in hy- drated mineral compounds containing, as the name indicates, water in combination with salts. In some cases the combined water corresponds to the amount that might be held in hydroxid form; as, for example, in the abundant mineral called gypsum, or land plaster, which is calcium sulfate crystallized with two molecules of water, CaSO4 : 2 H2O, or CaO2S(OH)4. Hydrogen in the free state (H2) is a gas. From Table i it will be seen that the molecule of hydrogen (H2) is lighter than the atom of any other element; and, according to the gas law, a given volume will be filled by the same number of molecules of hydrogen as of any other gas. Consequently, hydrogen gas is the lightest of all known gases, so that a balloon filled with hydrogen easily floats in the atmosphere of nitrogen (N2) and oxygen (O2) , one of which PLANT FOOD AND PLANT GROWTH 29 is fourteen times, and the other sixteen times, heavier than hy- drogen. At the ordinary temperature, hydrogen and oxygen gases can be mixed together and remain a mixture; but, if heated or ignited, the bonds which hold the atoms in molecules (H2 and O2) are broken, and the mixture explodes with terrific force and loud re- port. The only product of the explosion, or at least of the reac- tion, is water, H2O; and, if either gas was present in the mixture in excess of these proportions, the excess remains unchanged. The corn kernel contains 6.4 per cent of hydrogen, or about 97^ per cent of the three elements, oxygen, carbon, and hydrogen. Life. The fixation of carbon is the most important process in the growth of plants. By the term fixation is meant the changing of a gas or soluble substance to a solid or insoluble form by means that involve chemical reaction. (The fixation of atmospheric nitrogen, the fixation of soluble phosphorus in soils, and the fixation of po- tassium and other bases will be explained in the following pages.) The process known as the fixation of carbon is the more impor- tant, because it involves, not only the fixation of carbon itself, but likewise the fixation of both oxygen and hydrogen. Its im- portance is better appreciated by recalling that these three ele- ments compose about 95 per cent of the entire weight of most agricultural plants or crops. Both the carbon and oxygen utilized in plant growth are derived from the carbon dioxid contained in the air. It is truly remarkable that 90 per cent (90 pounds in 100) of our common crops must be secured from .04 per cent (4 parts in 10,000) of the air. The fixation of carbon, oxygen, and hydrogen takes place in the green parts of plants. The carbon dioxid enters through the breath- ing pores1 on the under side of the leaf; and the water, composed 1 A breathing pore consists of two guard cells with a slit between them, passing through the outer coat of the leaf. These slits or openings are greatly influenced by the moisture and temperature of the air. In the absence of light they remain closed. When the breathing pores are open, the outside air has free access to the intercellular spaces and passages within the leaf. The number of breathing pores varies with different plants. About 17,000 per square inch have been found on oat leaves, 102,000 on corn leaves, and 216,000 per square inch on the leaves of red clover. They are found chiefly on the under side of leaves and on green stems, but sometimes in small numbers on the upper leaf surface and even on underground stems. 30 SCIENCE AND SOIL of hydrogen and oxygen, enters the leaf through the stem, having passed from the soil into the plant through the roots. As the carbon dioxid and water come together within the leaf, a chemical reaction occurs which may be illustrated by the following equation: HJ_ fV_i_'o r* C} TT r'o _i_ r\ a — \J T \J — \_y=vj = jnoV^V-' T v»« The dotted line shows how the bonds are broken. The two atoms of oxygen that are set free immediately join hands to form a mole- cule of oxygen (O2), which passes from the leaf into the outer air. The two hydrogen atoms (H2 = ) attach themselves to the group, =CO, forming the compound, H2CO, which may also be written CH2O, to show its hydrate character, and which might be called monose, but is commonly known as formic aldehyde. This reac- tion occurs only in the light and only in the presence of active living chlorophyll (the green coloring matter of leaves). In other words, this compound is formed under the influence of life, and by it we enter a new field known as organic chemistry. Organic matter. Organic matter consists of compounds formed by life processes, — compounds that are, or have been, living matter; whereas, inorganic matter consists of rocks, minerals, and metals, of salts, liquids, and gases, whose origin has no necessary connection with any living substance. Organic matter consists of carbon compounds,1 such as hydrocarbons (containing only hydrogen and carbon), fats (containing much carbon and hydrogen with little oxygen), carbohydrates (containing carbon with hy- drogen and oxygen in proportion to form water — as the name in- dicates), and proteids, which contain not only carbon, hydrogen, and oxygen, but also nitrogen, and sometimes phosphorus and sulfur. These are the great groups of organic compounds compos- ing plants and animals. The carbohydrates are the most abundant in plants, while the proteids, although a necessary part of plants, are the most abundant compounds in animals. Aldehydes and carbohydrates. Formic aldehyde is only one of many aldehydes, which constitute a large class or series of organic compounds. The aldehydes are extremely active substances and 1 Many of the simpler carbon compounds can be made artificially. PLANT FOOD AND PLANT GROWTH 31 have power to attack and decompose other substance. Formic aldehyde is often used as a disinfectant, and a 40 per cent solution, known as " formalin," is employed (at the rate of one pound of formalin in 50 gallons of water) to destroy smut in seed oats, for example. A remarkable property of the aldehydes is the power of condensa- tion, by which two or more molecules are condensed into one. Thus, two molecules of formic aldehyde, or monose, 2 CH2O, may become one molecule of diose, C2H4O2; while three molecules may form one of triose, C3H6O3; and four may form tetrose, C4H8O4; and five, pentose, C5H10O5; etc. The condensation process is so rapid that formic aldehyde itself is found in plants only in very small amount, while the condensa- tion products constitute commonly 80 to 90 per cent of the entire plant. The ending -ose means sugar, and the prefix mon-, di-, tri-, etc., designate the number of carbon atoms in the molecule. The following may illustrate this series of carbohydrates: l CH2O, monose (formic aldehyde). C2H4O2, diose (unknown). C3H6O3, triose (glycerose). C4H8O4, tetrose (erythrose). C5H10O5, pentose (xylose). C6H12O6, hexose (glucose). C12H24O12, lactose (milk sugar). C12H22On, sucrose (common sugar). Here we may see the possible development of the well-known glucose, milk sugar, and common sugar (obtained from sugar cane and sugar beets), as condensation products from monose, or formic aldehyde, formed in the living plant from carbon dioxid and water. 1 It cannot be considered as absolutely proven that formic aldehyde is always the first product of this fixation process; and, if it is, it seems that the first condensa- tion product results from the union of three molecules, because the compound that might be called diose is not found in plants and is not known to exist. The known facts are that carbon dioxid is condensed in the leaves of plants and that oxygen is given off in the proportions required for this reaction (aside from the oxygen normally exhaled), also that formic aldehyde is found in plant leaves, that aldehydes have the power of condensation, and that multiples of the formic aldehyde molecule are actually present in plants (as hexose) or represented (as in starch, cellulose, pentosans, etc.). 32 SCIENCE AND SOIL It may be noted that cane sugar differs from milk sugar by one molecule of water (H2O). The sugars are a very important group of compounds, but perhaps the starches are a still more important group. Starch (C6H10O6) appears todifferfrom glucose (C6H12O6) by one molecule of water (H2O) , but it is known that the starch mole- cule is not simply C6H10O5, but some multiple of this formula, which is best written (Cgll^C^)*, in which x stands for the number by which this formula should be multiplied, for as yet x is unknown, although the proportion or percentage of each element in starch is known. The formula for cellulose (plant fiber) must also be recorded as (CgHjoOg)^, but this x may be a different number than the x of the starch molecule. Growth. Rapidity of growth is related to leaf surface. Sugar, starch, and fiber constitute the great carbohydrate group of plant structure, and their formation is dependent primarily upon the fixation of carbon, with oxygen and hydrogen, in the leaf; and, with all necessary things provided in proportionate amounts, this process goes on in direct proportion to leaf surface. In other words, under perfect conditions, a leaf four inches long will grow four times as much during the day as a leaf only one inch long; and, with sufficient moisture and with plant food provided in abundance, a pasture with the grass kept six inches long will furnish twice as much feed as one with the grass kept down to three inches. If the foundation principles and the controlling factors in plant growth can be known, then the ideal conditions for crop production may be provided much more nearly than is common. The ideal condition is to provide all controllable factors in such abundance or perfection that the crop yields will be limited only by the sun- shine and rainfall. With all other limiting factors removed, the average yield of corn in the corn belt would undoubtedly exceed 100 bushels per acre. (See the records of actual yields, in the follow- ing pages.) Carbon cycle. The carbon cycle includes both the fixation and the liberation of carbon. Animals feed upon plants and plant prod- ucts rich in carbon compounds, which in part are digested and car- ried into the blood to meet the oxygen inhaled through the lungs. The carbon is burned, or oxidized, to carbon dioxid, furnishing PLANT FOOD AND PLANT GROWTH 33 to the animal the energy or heat equivalent to that of ordinary combustion in the furnace, of the same materials; and the carbon dioxid is then thrown off through the lungs into the air, again to become the source of carbon and oxygen for plants. Thus, the fixation of carbon by the plants on the one side, and, on the other, all forms of combustion, including the visible flame, the consump- tion and oxidation of food by animals, or the oxidation of organic matter in the soil, completes the endless carbon cycle. But for this carbon cycle, plant growth and crop production would soon cease. A simple computation reveals facts not com- monly appreciated : A column of air one inch square and the height of the atmosphere weighs 15 pounds, which is equivalent to 2160 pounds per square foot, or less than 95 million pounds per acre. In ten thousand pounds of average air there are less than four pounds of carbon dioxid (CO2) or about one pound of carbon. Consequently, there is less than 10,000 pounds of carbon in the air above one acre of land. In 100 bushels of corn (5600 pounds), there are 2500 pounds of carbon. (See Table 2 , or compute from the per cent of carbon in starch and fiber, C6H10O5.) Thus, the total supply of carbon over an acre of land is only equal to the needs of four such corn crops as are commonly produced on the best-treated corn-belt land in the best seasons, the grain only being considered, or to only two crops, considering both grain and stalks. If, however, only one fourth of the earth's surface is land, if only one fourth of the land is cropped, and if only one fourth of 100 bushels is the average crop, then the supply of carbon is sufficient, not for two years only, but for 128 years, which, however, still emphasizes the fact that the carbon cycle makes possible the continuation of plant life on the earth. A maintenance ration for animals is a supply of food sufficient only to support the animal body in health, to provide food materials for repairing the daily waste, and to furnish energy sufficient to keep the body warm and to maintain the circulation of the blood and other necessary activities. Plants also have some vital pro- cesses to provide for, and, to a limited extent, plants are consumers of energy day and night. Food materials are stored by the plant, chiefly to be utilized in subsequent plant development. Thus 34 SCIENCE AND SOIL sugars are converted into starch and stored away in roots, tubers, or seeds, to supply the future needs of the same plant or of new plants. At the proper time the plant reconverts the insoluble starch into soluble sugar 1 and carries it through the circulation to the point of consumption as food by the plant, either for energy, repair, or growth. Food materials are thus consumed or oxidized within the plant, and carbon dioxid is constantly given off from all its living parts, including the roots. During the day the fixation of carbon is commonly so great as to completely mask the liberation of carbon dioxid in the green parts of the plant. Vegetable fats. Before leaving the subject of the fixation of carbon, oxygen, and hydrogen, some further mention should be made of the group of compounds called fats, whose importance is exceeded only by that of carbohydrates and protein. The vegetable fats and oils show distinct relationship to a con- densation process similar to the formation of sugars and other carbohydrates from formic aldehyde, the photosynthetic product of the reaction between carbon dioxid and water in the leaves of plants. The following series of compounds will show this relation- ship: . HYDROCARBONS FATTY ACIDS HCH3 — Methane ........ HCOOH — Formic acid. CH3CH3— Ethane ........ CH3COOH— Acetic acid. C2H5CH3 — Propane ....... C2H5COOH — Propionic acid. C3H7CH3 — Butane ........ C3H7COOH — Butyric acid. C4H9CH3 — Pentane ........ C4H9COOH — Valeric acid. C5HUCH3 — Hexane ....... CSHUCOOH — Hexoic acid. C6H13CH3 — Heptane ....... C6H13COOH — Heptoic acid. C7H15CH3— Octane ........ C7H15COOH — Octoic acid. CUH23CH3 — Dodecane ....... CUH23COOH — Laurie acid. C15H31CH3— Hecdecane ...... C15H31COOH — Palmitic acid. C7H35COOH — Stearic acid. Unsaturated C17H33COOH — Oleic acid. C17H3COOH — Linolic acid. 1731 — Linolenic acid. The hydrocarbons, which constitute the simplest series of carbon compounds, are shown for direct comparison with the fatty acid 1 Glucose sugars and sirups are manufactured in large quantities by use of strong acids for converting the starch into glucose. PLANT FOOD AND PLANT GROWTH 35 series. All of the compounds here illustrated are known. The two series differ only by the — COOH group in the fatty acid series in place of the — CH3 group in the hydrocarbon series. The — CH3 group is methane from which one hydrogen atom is removed, leaving the radicle — CH3, which is called methyl (one of the alkyl radicles), and acts as a monovalent radicle, replacing one hydrogen atom. It finds a place in many organic compounds, as in ethane (CH3— CH3), butane (CH3— CH2— CH2— CH3), etc. The group, — COOH, is called carboxyl, or the acid group. It may be represented: — C— O— H II O This group also has one free bond and acts as a monovalent radicle. Whenever this group is contained in an organic compound, the compound is an acid. The hydrogen in the hydroxyl part of this group may be replaced by metals, thus forming salts. If the free hand in this carboxyl group is grasped by a hydrogen atom, the compound formed is formic acid, but if methyl ( — CH3) joins hands with carboxyl ( — COOH), the compound formed is acetic acid (CH3COOH), the acid which gives to vinegar its sour taste. When lead (Pb) is used as a base to form a salt with acetic acid by re- placing the acid hydrogen of the hydroxyl group, the sourness is destroyed and the salt is known as sugar of lead, or lead acetate, (CH3COO)2Pb. When the hydroxyl group joins alkyl radicles ( — CH3, — C2H5, etc.), alcohols are formed, as methyl alcohol (CH3OH), called wood alcohol, and ethyl alcohol (C2H5OH), which is common alcohol. Common glycerin, which is also called glycerol (because it is an alcohol), is an organic compound consisting of a trivalent radicle, called glyceryl, united with three hydroxyl groups, C3H5(OH)3. Common animal fats consist chiefly of palmitic, stearic, and oleic acids combined with this radicle, =C3H5, and the fats themselves are called palmitin, stearin, and olein. The harder fats, like tallow, contain more stearin (C17H35COO)3C3H5, while the softer fats, like lard and butter, contain considerable olein, which differs from 36 SCIENCE AND SOIL stearin by having two less hydrogen atoms in each acid radicle. By itself, olein is a liquid or oil. The oil of corn contains about 4 per cent of stearin, 45 per cent of olein, and 48 per cent of linolin, which differs from olein by two hydrogen atoms, and from stearin by four hydrogen atoms, in each acid radicle. When these fats and oils are heated with a strong base (alkali) such as potassium hydroxid, three potassium atoms displace the glyceryl radicle (=C3H5) and form potassium stearate (C17H35COOK), potassium oleate (C17H33COOK), etc.; while the three hydroxyl groups unite with glyceryl to form glycerin, C3H6(OH)3. The salts formed by potassium or sodium with these fatty acids are what we call soap, the potassium compounds being soft soap, and the sodium, hard soap. While the fixation of carbon, oxygen, and hydrogen, resulting ultimately in the formation of carbohydrates and fats, is properly considered the most important process in plant growth, we may well remember that no fixation and no growth occur in the absence of the other seven essential elements of plant food. Indeed, from the standpoint of possible control of crop production, another tripod is more important than these three; namely, nitrogen, phosphorus, and limestone. Nitrogen. This element has received more consideration as plant food than any other essential element. In the free state (N2) it is a gas, and in this form it constitutes three fourths of the air. The total supply of nitrogen over each acre of the earth's surface, if available, would meet the needs of a hundred-bushel crop of corn every year for 500,000 years; whereas the supply of carbon is sufficient for such crops for only two years. Nevertheless, carbon has no commercial value as plant food, while nitrogen in available form is worth 15 to 20 cents a pound in the markets. These facts only emphasize the need of science in agriculture. Nitrogen is not contained in the mineral matter of the earth, but it is a constituent of common organic matter. It is an essential part of the structure of every plant and animal, and is present in all crops and crop residues and, consequently, in the organic matter, vegetable matter, or humus, of the top soil; and it is from the decomposition products of this organic matter that nitrogen is PLANT FOOD AND PLANT GROWTH 37 furnished to most growing crops, by a process (nitrification) that is more fully explained in the following pages. Protein. Protein is the general name for organic nitrogen com- pounds, including the proteids, or final products, and the amids, or intermediate products. The amids and proteids of the protein group might be compared with the sugars and starches (and fibers) of the carbohydrate group in which the sugars are the intermediate form and the starches (and fibers) the more permanent form. Protein always contains nitrogen in addition to oxygen, carbon, and hydrogen. The chemical reactions involved in the formation of proteids are not yet well understood, although many of the intermediate products (amids) are well known, and some can be made arti- ficially. The amids are especially abundant in young or immature plants, and they are also liberated as intermediate decomposition products. Thus, carbamid, O=C = (NH2)2, which is also called urea, is a common nitrogen compound in urine, the medium by which most of the nitrogen waste is thrown off from the animal body. This compound might be considered as formic aldehyde, or monose (O=C = H2), in which the two hydrogen atoms are re- placed by two amido groups, and the amido group ( — NH2) may be considered as ammonia in which only two monovalent hydrogen atoms are joined to the trivalent nitrogen atom, thus leaving one free hand by which this group may be attached to other groups or atoms in the building of molecules. The hydroxyl group ( — OH) and water (OH2),in relation to oxygen, correspond to the amido group ( — NH2) and ammonia (NH3), in relation to nitrogen, and also to the methyl group ( — CH3) and methane (CH4), in relation to carbon. The amido group ( — NH2) acts as a monovalent radicle, and by replacing hydrogen atoms in various compounds forms new compounds called amids, or amido compounds, and these by con- densation or combination with other groups may form the final nitrogenous organic compounds called proteids, which constitute chiefly the flesh (not fat) and vital organs of animals, and the pro- tein of mature plants. It has been suggested that amido formic aldehyde, H2NCHO, or amido acetic aldehyde, CH2(NH2)CHO, or aspartic aldehyde (see aspartic acid and asparagin in the following list) may furnish the 38 SCIENCE AND SOIL initial molecules whose condensation produces proteids, but this is largely speculative. The following list illustrates some instructive relationships of important and well-known compounds: HCOOH, or H— C— OH .... Formic acid. O CH2COOH, or H3=C— C— OH . . Acetic acid (the acid in vinegar). O CH2(NH2)COOH Amido acetic acid. COOH COOH, or (COOH)2 Oxalic acid. CH2COOH CH2COOH Succinic acid. CH2COOH I ' CH(NH2)COOH Amido succinic acid (the aspartic acid in pumpkin seed, beets, etc.). CH2CONH2 CH(NH2)COOH Amido succinamic acid (the asparagin found in asparagus, in beans and peas, and in many seeds when ger- H minating). H— C C— H C6H6, or || ... Benzene. H— C. ,C— H C6H5 — OH Hydroxy benzene, or phenol (carbolic acid). C6HS — NH2 Amido benzene, or anilin. C6H3(OII)3 Trihydroxy benzene, or pyrogallol (pyrogallic acid). C6H3(NH2)3 Triamido benzene. C10H14N2 Dipyridyl hexahydrid, or nicotin (the alkaloid of tobacco). C17H21NO4 Morphin (the alkaloid of opium, from the poppy). C21H22N2O2 Strychnin (the alkaloid of nux vom- ica). £2*0^392^07583 • : Albumen, or the white of egg. (For- mula suggested by Schutzenberger ) PLANT FOOD AND PLANT GROWTH 39 Zein, the most abundant proteid in corn (Zea mays], has the following composition: Carbon .... 55.15 per cent. Hydrogen ... 7.24 per cent. Oxygen .... 20.77 Per cent- Nitrogen . . . 16.22 per cent. Sulfur .... .62 per cent. According to this analysis, the molecule of zein might be repre- sented by the following formula: Ordinary corn contains n per cent of protein, of which about one half consists of the proteid zein. This nitrogenous substance has been separated, purified, and investigated with very great care, especially by Chittenden andOsborne (American Chemical Journal (1891), 73, 453, 529; (1892), 14, 20). The percentage composition represents the average of several closely agreeing analyses of what was believed to be very pure zein. Based on the percentage of sulfur, the molecular weight cannot be less than about 5000, and the formula given above or some multiple of it must be approxi- mately correct. Certainly the proteid molecule is exceedingly complex, and the number of different proteids is very large. They all contain nitrogen, usually about 16 per cent, and some of them contain also sulfur and phosphorus. Sulfur. This is an essential element for all plants, but the amount required for normal growth and full development is relatively very small, even when compared with the small percentage present in the earth's crust. Most proteids (as zein, for example) contain sulfur, but the percentage is usually very low. It is present, however, in organic combination, and does not give the ordi- nary tests for sulfates, the form in which it is usually taken from the soil. Many of the simpler organic compounds of sulfur are well known, and some can be made artificially. The oil of onions and garlic, which gives to those plants their peculiar odor and taste, consists 40 SCIENCE AND SOIL chiefly of allyl * sulfid (C3H5)2S; and mustard oil is also composed of organic sulfur compounds. Phosphorus. Phosphorus is the Greek word for the morning star, and signifies light. The element phosphorus is closely asso- ciated with the beginning of all forms of mortal life. The nucleus of every living cell in plants and animals is rich in phosphorus. Nuclein, the phosphorized nitrogenous constituent of the cell- nucleus, contains as high as 10 per cent of the element phosphorus, although it may contain no sulfur. The following formula has been suggested byMiescher for nuclein derived from animal cells: C29H49022N9P3. Lecithin, C44H90O9NP, is a well-known organic phosphorus com- pound, which it is thought may have some controlling influence in the formation of fats and oils. Ordinary corn contains about 5 per cent of oil, of which 1.5 per cent consists of lecithin; that is, i^ pounds of lecithin are found in 100 pounds of the oil. It should be remembered that, while sulfur is contained in many proteids, phosphorus is present in every cell of every plant. The grain or seed of plants contains, as a rule, more than fifty times as much phosphorus as sulfur. The phosphorus of the corn kernel is found largely in the germ. In 1000 pounds of corn there are about 100 pounds of germs containing more than two pounds of the ele- ment phosphorus. About 95 per cent of the ash obtained from the burning of corn consists of the phosphates of potassium and magnesium. Hay, straw, and other coarse products usually contain more sulfur than the grain, but these coarser parts com- monly remain on the farm, while the grain is more likely to be sold. 1 The monovalent allyl group ( — CsHs) differs from the trivalent glyceryl (=CsH6) only in having a double bond, as shown in allyl alcohol and glycerin (which might also be called glyceryl alcohol) : Allyl alcohol Glycerin CH2 CH2OH II I CH CHOH I I CH2OH CH2OH PLANT FOOD AND PLANT GROWTH 41 The proteid of milk, called casein, has the following composition : Carbon .... 53.30 per cent. Hydrogen . . . 7.07 per cent. Oxygen . . . . 22.03 Per cent. Nitrogen . . . 15.91 per cent. Sulfur 82 per cent. Phosphorus . . .87 per cent. A ton of wheat bran contains about 24 pounds of phosphorus, or 1.22 per cent. About 86 per cent of the total phosphorus in bran is soluble in water, and, according to Patten and Hart (New York State Agricultural Experiment Station Bulletin 250), this water-soluble phosphorus is contained in the salt of an organic acid, which is probably identical with a compound investigated by Posternak, and called by him anhydro-oxymethylene diphos- phoric acid, the formula being C2H8P2O9. As determined by Patten and Hart, the complex salt of this acid which constitutes the prin- cipal phosphorus compound in wheat bran, has the following composition, as found by the ultimate analysis of the isolated compound : Calcium 1.13 per cent. Potassium 2.60 per cent. Magnesium 5.80 per cent. Carbon 17.30 per cent. Hydrogen 3.63 per cent. Phosphorus 16.38 per cent. Oxygen (by difference) . . . 53.16 per cent. The free acid was found to contain 10.63 per cent of carbon, 3.38 per cent of hydrogen, and 25.98 per cent of phosphorus, which corresponds fairly closely, especially in phosphorus, with the theoretical percentages, which any one can compute for the for- mula C2H8P2Oa. m 9 m 9 The mineral part of animal bone consists largely of tricalcium phosphate, Ca3(PO4)2, which, when pure, contains 20 per cent of the element phosphorus, as can be easily computed by any one who knows the atomic weights. In 100 pounds of raw bone are about 42 SCIENCE AND SOIL 10 pounds of phosphorus. The phosphorus required by animals must first be supplied in the plants that serve as animal foods. The percentage of phosphorus in the earth's crust is small when compared with the requirements of plants, especially when we also consider that the phosphorus accumulates in the more concen- trated and more salable products, as in the seed or grain, and also in the flesh, bone, and milk, of animals. Phosphorus is usually taken up by plants in the form of phos- phates, but within the plant it enters into organic combination as shown above. The six elements thus far discussed in some detail — carbon, oxygen, hydrogen, nitrogen, phosphorus, and sulfur — are all non- metallic. Three of them — oxygen, hydrogen, and nitrogen — are gases in the free state, and the other three — carbon, phosphorus, and sulfur — are nonmetallic solids. Four other elements are also absolutely essential to the growth of all agricultural plants. Potassium and magnesium. These are metallic elements which have very important functions in plant growth and which are re- quired in considerable amounts. Both are stored in the seed in relative abundance, and are found in the ash of grains in the form of phosphates, although still larger amounts of potassium are stored in the coarser parts of plants (as in straw, cornstalks, etc.) . It is not known that potassium and magnesium are essential constituents of protoplasm, but, like nitrogen and phosphorus, they are found in largest proportions in the embryo tissues. It is suggested that one of their essential functions may be as carriers of nitrogen and phosphorus in the form of definite salts (as nitrates and phosphates) capable of reaction with certain products result- ing from the fixation of carbon, oxygen, and hydrogen. Certainly it is not sufficient that phosphorus, for example, shall merely be carried in the form of some soluble phosphate into the laboratory (the leaf) of the plant, but the compound must be such that the metallic base will release the phosphorus at the proper time, in order that it may enter the organic combination and thus become a part of the living organism. It is known that organic acids are developed by the plant with which potassium and other bases carried into the plant in the form of nitrates, phosphates, etc., may unite, and do unite, at some time PLANT FOOD AND PLANT GROWTH 43 during the life of the plant; for some of the potassium that enters the plant roots as nitrates, phosphate, or sulfate is afterward found in the plant in organic salts, as tartrate (in grapes) , oxalate (in sorrel), etc. There appears to be little or no evidence that any living organic compounds of potassium or magnesium exist. It is the common belief that potassium has large influence over the formation of carbohydrates; but the information is not suffi- cient to determine whether this influence is direct or very indirect, as in maintaining the general health of the plant by having some absolutely necessary part in reactions involving the transference of nitrogen or phosphorus from inorganic compounds to the living organic combination. The potassium contained in plants is in large part very easily removed by leaching with water, and hence peaty swamp soils consisting largely of organic matter are frequently very deficient in potassium. While potassium and magnesium are required by plants in very considerable amounts, as stated, and as shown in Table 2, yet, when measured by the average composition of the earth's crust and by average crop requirements, the supply of these two elements is very great. Calcium and iron. These elements are absolutely essential to the normal growth and development of all agricultural plants, but for the grain crops the amounts positively necessary are so extremely small and the quantities present in the earth's crust are so extremely large that it is rarely that either calcium or iron is furnished to such plants in amounts insufficient to perform their essential functions, except, of course, when they are artificially withheld, as in investigational work. Legume plants are a very marked exception, however, so far as calcium is concerned. Iron evidently has some important connection, direct or indi- rect, with the formation of chlorophyll (the green coloring matter of leaves) ; for, if iron is withheld from the plant, the leaves do not become green, and if later iron is supplied, the chlorophyll soon begins to develop. On the other hand, analysis has shown that the chlorophyll itself does not contain iron, and the somewhat common assumption that the green color of plants is due to the presence of iron compounds of that color is incorrect. The iron held in nuclein compounds is not dissolved out by dilute 44 SCIENCE AND SOIL hydrochloric acid, — a fact which indicates that the iron is a con- stituent of the living matter of plants. Animals also have a small but absolute requirement for iron. Aside from oxygen, iron is the most abundant essential plant- food element, constituting about 5^ per cent of the solid crust of the earth, although the amount required by plants is very insig- nificant. Thus, the earth contains more than 50 times as much iron as phosphorus, while the corn kernel contains nearly 40 times as much phosphorus as iron, so that the supply of phosphorus would be depleted as much by the removal of 100 crops as the supply of iron would be by 200,000 crops. While a very small supply of calcium is of vital importance, considerable amounts of that element are commonly taken up and deposited in the coarser parts of plants, as in straw, cornstalks, and hay, and large supplies of calcium are required for legumes, especially for clover and alfalfa. This larger use of calcium appears to be due, especially in grain crops, to its power as a base to unite with organic acids that might otherwise injure the plant; and the salts formed are commonly deposited, not in the seed or with stored food materials, but in the older tissues as inert matter. The common use of certain calcium compounds, such as burned lime and ground limestone, for correcting soil acidity should not be confused with the essential need of the element calcium as plant food. Even strongly acid soils often contain abundance of the ele- ment calcium for plant food, not in the form of carbonates, but in silicates, which, however, have no power to correct soil acidity. Aluminum, silicon, sodium, chlorin, and manganese. These elements are not known to be essential to plant growth, but they are commonly found in plants, although the amount of manganese is very small and that of aluminum still smaller. The opinion that silicon was essential and gave stiffness to the straw of cereals is not correct, and the report that manganese exerts a marked stimulating action on plant growth has not been verified upon more thorough investigation. Sodium is now known to be a nonessential, but there is still a possible question regarding chlorin. According to Pfeffer, " it remains for precise researches to determine whether a minimal amount is essential, or whether chlorin simply favors growth under special cultural conditions." PLANT FOOD AND PLANT GROWTH 45 Common functions. Some common functions may be performed by several elements. Thus, if there is need for a base to correct an excess of acid that has developed in the plant, sodium may serve as well as potassium, although with enough potassium provided, no sodium is needed. As already stated, the largest use of calcium appears to be in this line, in which, perhaps, manganese, magne- sium, or iron might serve equally well if they were present in the plant in sufficient amount. Likewise, in solvent compounds, chlorin may serve as well as nitrogen or phosphorus, but cannot take their place in living tissue. We shall also consider in the following pages the value to plants of certain materials when applied to certain soils, which serve not as plant food, but rather as soil stimulants, having power to liberate from the soil some essential plant-food element more rapidly than it would otherwise become available — an action that may result in temporary profit and ultimate land ruin. Caustic lime, salt, gypsum (land-plaster), and, under certain conditions, commercial fertilizers, and even farm manure, clover, and green manures, may act in part, at least, as soil stimulants; and, to guard against such injurious action, practice must be controlled by science (knowl- edge). CHAPTER IV THE EARTH'S CRUST NEARLY 98 per cent of the solid crust of the earth consists of silicates of the six metals, aluminum, iron, calcium, potassium, sodium, and magnesium (in this order of relative abundance); and the remainder is largely composed of the closely related titan- ates. Silicon. Silicon in the mineral matter constituting the earth's crust corresponds to carbon in the organic matter of the vegetable and animal kingdoms. In all of the great groups of organic com- pounds the molecule is built up by the linking power of the four- handed carbon atom, as, for example, in the hydrocarbon, hexane (C6H14): H H H H H H I I I I I I H— C— C— C— C— C— C— H I I I I I I H H H H H H Silicon is the second member of the carbon group * in the periodic system, as shown on page n, and its linking power is also very great, although alternating with oxygen and metals and restricted mainly to silicates. Thus, instead of the almost unlimited number of hydrocarbons, carbohydrates, and other numerous compounds of carbon, hydrogen, and oxygen (alcohols, fats, organic acids, etc.), there are but four such silicon compounds known: SiH4, SiO2, OSi(OH)2 or H2SiO3, and Si(OH)4 or H4SiO4, which differs from silicon dioxid by two molecules of water. 1 A most interesting compound is SiC, silicon carbid, so-called carborundum, formed by the union of the two tetravalent elements and, next to the diamond, one of the hardest known substances. 46 THE EARTH'S CRUST 47 The numerous natural polysilicates (poly- means many) compos- ing granite and most other rocks of the earth's crust are salts of poly silicic acids, although the acids themselves are not known to exist free from the basic elements or radicles. The following may illustrate a few of the possible combinations, the last three being known only in salts in which bases appear in place of the acid hydrogen : Silicon dioxid, SiO2 . . . . O = Si=O. Metasilicic acid, H2SiO3 . . O = Si = (OH)2. Orthosilicic acid, H4SiO4 . . (HO)2 = Si = (OH)2. Disilicic acid, H2Si2O5 . . . HO-Si=O3=Si-OH. Polysilicic acid, H4Si3O8 . . (HO)2 = Si=O2 = Si = (OH)2. = Si"°2 = Si ^°2 = Si =°- Si = (OH)- Among the most common mineral compounds found in granite is ordinary felspar, or orthoclase, or potassium aluminum poly- silicate, KAlSi3O8, or (KAlSi3O8)2, whose structural formula may be represented thus: /o o o o K/ Si Si Si O O Si Si This is sufficient to illustrate what is meant by polysilicates. Other silicates differ from the common felspar by the substitution of other elements for potassium or aluminum or both, and also by different proportions of the various constituents, as: Orthoclase (potassium felspar) .... KAlSi3O8. Albite (sodium felspar) ....... NaAlSi3O8. Anorthite (calcium felspar) ...... CaAl2Si2O8. Crysolite (magnesium iron silicate) . . . MgFeSiO4. In some cases hydroxyl groups are included, and when such com- pounds are heated, two hydroxyl groups are broken, leaving one oxygen atom in their place, thus yielding water and anhydrous 48 SCIENCE AND SOIL silicate. Silicates from which water can be separated are called hydrated silicates or, sometimes, acid silicates: Steatite (soapstone) Mg6Si4O13(OH)2. Kaolin (clay) Al2Si2O5(OH)4. Commonly, the silicates of the earth's crust are more or less mixed, so that samples of pure compounds are rarely, if ever, found in native state. Following are the results of analysis of specimens of orthoclase, kaolin, and steatite as found in nature : TABLE 3. PERCENTAGE COMPOSITION OF SILICATES CONSTITUENTS ORTHOCLASE (POTASSIUM FELSPAR) KAOLIN (CLAY) STEATITE (SOAPSTONE) Potassium 11.64 .04 .24 trace 30.89 9.84 •93 •34 .07 .11 •55 22.87 iQ-59 •°3 12.83 43.61 .37 15-43 2.32 9-15 19.91 3.22 .12 8-45 4I-I3 Magnesium Calcium Iron Silicon Aluminum Sodium \Vater of hydration Oxygen, etc 46.42 While the samples of orthoclase and kaolin were fairly pure, the steatite contained other metals aggregating almost as much as the magnesium. Granite and gneiss. These are among the most common rocks, the former being of igneous or eruptive origin, while gneiss is essentially the same material in sedimentary stratified form. In other words, when granite has been disintegrated by the action of heat and cold, rain and frost, has been transported by wind and flowing water, has been redeposited in strata over river bottoms, or ocean beds, and has become reformed into compact masses by the cementing action of acids, alkalies, or salts, it is then called gneiss. Gneiss is one of the oldest stratified rocks, and was formed chiefly previous to the beginning of plant or animal life on the earth. Granite and gneiss consist principally of the four mineral groups, felspar, hornblende, mica, and quartz. Of these the felspar group THE EARTH'S CRUST 49 has already been discussed. The hornblendes (or amphiboles) in- clude certain white or light-colored silicates of calcium and mag- nesium, often with fibrous structure, of which common asbestos is a good example; also silicates of aluminum, magnesium, and iron, of darker colors, green or black. The micas include light-colored or transparent potassium aluminum silicates and black silicates of aluminum, magnesium, and iron. While the hornblendes are often fibrous, the micas, as a rule, are easily split into the well-known mica sheets. Quartz. Quartz, when pure, is crystallized silicon dioxid (SiO2), but it is often colored by small amounts of metallic compounds. Aside from being a common constituent of granite and gneiss and of many other less abundant silicate rocks, quartz is often found in rock masses or seams in a nearly pure state. Quartz sand is not uncommon, but the opinion that sand and quartz are synonymous terms is very incorrect, for sand usually includes very considerable amounts of granite or gneiss and other mineral particles. The following statement shows the composition of common samples of original granite, fresh gneiss, and decomposed gneiss; also the percentage of each constituent saved from the fresh gneiss and found in the decomposed gneiss, as computed by Merrill,1 assuming no loss of aluminum, which indicates a total loss of 44.67 per cent of the original rock. They serve only as illustrations, and other samples may vary greatly from these. TABLE 4. PERCENTAGE COMPOSITION OF ROCK CONSTITUENTS FRESH GRANITE FRESH GNEISS DECOMPOSED GNEISS PERCENTAGE SAVED Phosphorus .04 3-3° .29 1.32 3-34 32-77 7.64 2.90 .89 47.60 .11 3-54 .64 3-17 6-34 28.49 8.94 2.09 .62 46.06 .21 .92 .24 trace 8.52 21.27 14.05 .16 13-75 40.88 100.00 16.48 25-30 o.oo 85.65 47-55 IOO.OO 4-97 Potassium Magnesium Calcium Iron Silicon Aluminum Sodium Water of hydration . . . Oxygen etc " Rocks, Rock Weathering, and Soils," 1897, p. 215. 50 SCIENCE AND SOIL Gneiss may contain constituents not always present in granite, because of admixture of other materials in transportation; also certain constituents are likely to be lost to some extent in the origi- nal disintegration and transportation, and to a great extent in the subsequent more complete decomposition, so that often certain constituents may show higher percentages in the final residue. Zeolites. Zeolites are formed from partially decomposed min- erals, like granite and gneiss. They are hydrated double silicates of aluminum with calcium or sodium, and may contain other bases, especially potassium. They are credited with important functions in soils to which further reference will be made. Shale, kaolin, and clay. These materials consist chiefly of hydrated aluminum silicate related to the mineral kaolinite, Al2Si2O5(OH)4, and representing in part the final residue from the decom- position of felspar, hornblendes, micas, etc., from granite, gneiss, and other silicate rocks. They may be grouped under the general term argillites (from argil, meaning potter's clay) . Slate is the well- known roofing material. Shale is the term applied to the more thinly stratified formations which disintegrate more or less readily when exposed to the weather. Kaolin is common fire clay. Ordinary brick clay belongs in the same group, and, in fact, shale itself is often ground and used for making brick or tile. Aluminum silicate is the final residue from the disintegration of many different rocks, and consequently is itself one of the most permanent substances. The oldest records of man have been pre- served in burnt clay, both in tablets and in pottery. Carbonates. The carbonates include a very important group of rocks, although they constitute a small portion of the earth's crust when compared with the silicates. Of the carbonates, the common limestone, calcium carbonate, CaCO3, is by far the most abundant. It is frequently quite impure. Marble is calcium car- bonate, mottled or colored with impurities and of sufficiently close texture to admit of polishing. The mineral calcite is very pure crystallized calcium carbonate, CaCO3. Magnesian limestone (dolo- mite) is a double carbonate of calcium and magnesium, CaMg(CO3)2, but this compound is frequently mixed with calcium carbonate, CaCO3, so that varying percentages of calcium and magnesium are found in dolomitic limestone. THE EARTH'S CRUST Most limestone deposits are marine formations, and frequently consist largely of shells, but this is not always the case. Small amounts of calcium carbonate are found in many other stratified rocks. Impure limestones containing silicate minerals may lose, by weathering and leaching, practically all of the calcium carbonate or magnesium carbonate which they originally contained and leave a residue free from carbonates, as shown by the following analyses: TABLE 5. COMPOSITION OF FRESH LIMESTONE AND ITS RESIDUAL CLAY CONSTITUENTS FRESH LIMESTONE RESIDUAL CLAY PERCENTAGE SAVED OF EACH CONSTITUENT Phosphorus i-33% .29 .18 3x-99 1.64 1.94 2.22 .12 3-36 9-3° 2.26 45-37 1.11% .80 .16 2-79 i-39 15.82 16.03 •45 ii. 61 none 10.76 39.08 10.24% 33-63 IO.62 1.07 10.44 IOO.OO 88.65 46.74 42.41 none Potassium Magnesium Calcium Iron . Silicon Aluminum Sodium Manganese Carbonate carbon Water of hydration Oxygen etc Penrose 1 presents convincing evidence that this peculiar man- ganese clay was derived from impure limestone, and Merrill ("Rocks, Rock Weathering, and Soils," 1897, pp. 232, 233) computes that more than 93.6 per cent of the original rock was lost during the processes of decomposition, weathering, and leaching, assuming no loss of silicon. While these analyses represent very satisfactorily the general results of rock weathering, it must not be assumed that the representation is exact in all details. In minor constitu- ents the sample of rock taken for analysis may vary greatly from the particular rock stratum of which the sample of clay was the residue, and the loss by leaching of the same constituent may also vary greatly in different rocks. Thus, in the decomposition of the 1 Arkansas Geological Survey, Annual Report for 1890, p. 179. 52 SCIENCE AND SOIL more common rocks (see Table 4) the loss of potassium and sodium is very great, while most of the iron and phosphorus is likely to be found in the residue. Sulfates. Natural sulfates are confined chiefly to hydrated cal- cium sulfate, CaSO4(H2O)2 or CaO2S(OH)4, containing about 18.6 per cent of sulfur and more than 20 per cent of water of hydration. This is the mineral called gypsum. It occurs in numerous deposits, at various depths, and sometimes extends over hundreds of square miles, as in northern Ohio. Under the name of land-plaster this mineral has been used very extensively in places as a soil stimu- lant. Traces of calcium sulfate are found in most limestones and in some other rocks. Sulfids. The sulfids of iron are widely distributed in nature. Iron disulfid, FeS2, is commonly known as pyrite, also called " fool's gold," because of its glitter and yellow color. One form of iron disulfid decomposes quite readily when exposed to air and moisture, and yields ferrous sulfate, FeSO4, as one of the products. Phosphates. These occur in small amount in connection with many other rocks and minerals, principally in the form of calcium phosphate, Ca3(PO4)2, specifically called tricalcium phosphate, when mentioned in connection with the artificial dicalcium phos- phate, Ca2H2(PO4)2, or monocalcium phosphate, CaH4(PO4)2. Granite commonly contains a trace of phosphorus, and in gneiss about . i per cent of phosphorus is found as an average, correspond- ing to 10 pounds of calcium phosphate (2 pounds of phosphorus) in one ton of gneiss. Limestones also contain calcium phosphate, as a rule. While the amount is usually less than i per cent, some quite extensive deposits of phosphatic limestone exist which con- tain from 10 to 30 per cent of tricalcium phosphate. In places where such rocks have been long exposed near the surface, the cal- cium carbonate has been largely removed by leaching, so that the remaining porous rock may contain as high as 75 per cent of the phosphates of calcium, iron, and aluminum, in which the calcium compound, Ca3(PO4)2, greatly predominates (as in the Tennessee brown rock phosphate). There are also some natural deposits of compact calcium phos- phate rock, varying in purity from about 40 to 80 per cent (as the Tennessee blue rock phosphate). These deposits of phosphate THE EARTH'S CRUST 53 and of phosphatic limestone show evidence of living organisms having been connected with their origin, as in limestone shells and bony skeletons. Apatite is crystallized calcium phosphate, containing small amounts of calcium chlorid or calcium fluorid. This mineral is largely found in masses, but traces of it are found in nearly all other rocks, whether of igneous or aqueous formation. Oxids. Oxids of silicon, iron, and aluminum are more or less abundant and distributed almost universally, in quartz (SiO2) and quartz sand, in the iron ore, hematite (FegOg) , and in the aluminum ore, bauxite (A12O8) and (Fe2O3). Other deposits. Various other deposits found naturally in the earth, but constituting extremely small percentages of the earth's crust, include common rock salt (NaCl) ; potassium salts, as carnal- lite (KClMgCl2 6 H2O) ; and kainit (K2SO4MgSO4MgCl2 6 H2O), from which potassium chlorid (KC1) and potassium sulfate (ILjSO^) , respectively, are separated; saltpeter (KNO3), and Chile saltpeter or sodium nitrate (NaNO3) ; also the extensive deposits of anthra- cite and bituminous coal, the former consisting of nearly pure carbon, while the latter contains considerable amounts of hydro- carbon compounds in addition to the free carbon. CHAPTER V SOIL FORMATIONS AND CLASSIFICATIONS Residual soils. Residual soils are those that are formed in place from the disintegration of rocks. They consist of the least soluble decomposition products, which often constitute but a small pro- portion of the original rock. Thus a limestone containing 80 per cent of calcium carbonate and 20 per cent of impurities (as poly- silicates, etc.) may weather to a soil composed entirely of the im- purities from which the calcium carbonate has been completely removed by leaching, and the polysilicates may have partially broken down into acid silicates, zeolites, clay, oxids of silicon and iron, etc. Transported soils. These are also formed from disintegrated and partially decomposed rock, but instead of remaining in the place previously occupied by the rock, they have been transported, and often retransported, by various agencies (materials from many sources sometimes being mixed together), and finally deposited in the places which they now occupy. Wind, water, and glaciers are the chief carrying agencies. Glacial material (bowlder clay) is characterized by the presence of worn or rounded stones, varying in size from sand grains to bowlders, embedded in silty clay. While glacial drift covers exten- sive areas in northern United States, sometimes to a depth of 100 feet or more, the glacial material is covered in many areas by a deposit of loess,1 varying in depth from a few inches to several feet. Loess is characterized in part by the absence of pebbles. It consists largely of silt, with some very fine sand and but little clay. It has been transported by wind, as a rule, and in places is found in high elevations and even overlying residual soils, but in deep loess deposits, as in the bluffs along the Mississippi and other large streams, evidences are found of some transportation by water. 1 This word is taken directly from the German (like sauerkraut) and pronounced like less, with the lips protruded as in whistling. Similarly, the English word beef- steak has been adopted into the German language. 54 SOIL FORMATIONS AND CLASSIFICATIONS 55 Alluvial soils are the common formation in river valleys and other lowlands that receive deposits of material washed from the higher lands. Soil materials. Soil materials consist of stones, gravel, sand, silt, clay, and organic matter. The term day, as correctly used, is applied to the material that gives to certain soils their sticky, plastic property, including hydrated aluminum silicate and other plastic substances, in part reduced probably to the molecular state of division and without granular character, although most so-called " clay " contains more or less undecomposed, or but partially decomposed, mineral particles. Silt includes a grade of particles that are smaller than sand, impalpable in fact, but still granular as seen through the microscope, and not plastic when free from clay. Soil types. Soil types are based largely upon the relative propor- tion of these several soil materials, as may be noted by inspection or determined by mechanical analysis. The following general groups are recognized: TABLE 6. SOILS: GENERAL GROUPS NUMBER LIMITS GROUP NAMES DESCRIPTION o to 9 IO tO 12 13 to 14 Peat . . . Peaty loam Muck . . . With 25 to 75 per cent or more of organic matter. 10 to 25 per cent of organic matter with loam. 10 to 25 per cent of organic matter with much clay. 15 to 19 20 tO 24 Clay . . . Clay loam . . Plastic clay predominating. Much clay with loam. 25 to 49 5° to 59 Silt loam . . Loam . . . Much silt with loam. Sand, silt, clay, and organic matter with neither markedly predominating. 60 to 79 80 to 89 Sandy loam . Sand . . . Much sand with loam. Sand without much silt or clay. 90 to 94 95 to 97 r>R Gravelly loam Gravel . . . Stony loam Rock outcrop Gravel with loam. Gravel without much silt or clay. Stones with loam. Disintegrating rock. 95 99 SCIENCE AND SOIL TABLE 7. SOME RECOGNIZED SOIL TYPES No. NAME No. NAME I Deep peat. 32 Light gray silt loam on tight 2 Medium peat on clay. clay. 2.1 Medium peat on clayey sand. 32.1 White silt loam on tight clay. 2.2 Medium peat on sand. 33 Gray-red silt loam on tight 2-3 Medium peat on rock. clay. 3 Shallow peat on clay. 34 Yellow -gray silt loam. 3-1 Shallow peat on clayey sand. 34-i Yellow-gray silt loam on 3-2 Shallow peat on sand. tight clay. 3-3 Shallow peat on rock. 35 Yellow silt loam. 10 Peaty loam on clay. 3S-i Yellow silt loam on tight clay. 10. 1 Peaty loam on clayey sand. 35-2 Yellow silt loam on clay. 10.2 Peaty loam on sand. 35-3 Yellow silt loam on sand. 10.3 Peaty loam on rock. 35-4 Yellow silt loam on gravel. *3 Muck on clay. 35-5 Yellow silt loam on rock. I3-I Muck on clayey sand. So Black loam. 13.2 Muck on sand. 50.1 Black loam on clay. T3-3 Muck on rock. Si Brown loam. IS Drab clay. Si-i Brown loam on clay. i5-i Sandy drab clay. 51.2 Brown loam on silt. 15-2 Gravelly drab clay. Si-3 Brown loam on sand. 16 Gray clay. 5i-4 Brown loam on gravel. 20 Black clay loam. 5i-5 Brown loam on rock. 2O. I Sandy black clay loam. 52 Gray loam. 20. 2 Gravelly black clay loam. 53 Yellow loam. 21 Drab clay loam. 54 Mixed loam. 21. 1 Drab clay loam on sand. 60 Brown sandy loam. 22 Gray clay loam. 60. i Brown sandy loam on silt. 25 Black silt loam. 60.2 Brown sandy loam on sand. 25-1 Black silt loam on clay. 60.4 Brown sandy loam on gravel. 26 Brown silt loam. 60.5 Brown sandy loam on rock. 26.1 Brown silt loam on clay. 61 Mixed sandy loam. 26.2 Brown silt loam on sand. 62 Brown fine sandy loam. 26.3 Brown silt loam on till. 63 Light brown fine sandy loam. 26.4 Brown silt loam on gravel. 64 Yellow fine sandy loam. 26.5 Brown silt loam on rock. 65 Gray fine sandy loam. 2? Brown silt loam over gravel. 80 River sand. 28 Brown -gray silt loam on 81 Dune sand. tight clay. 82 Beach sand. 29 Drab silt loam. 90 Gravelly loam. 29.1 Drab silt loam on clay. 95 Gravel. 3° Gray silt loam on tight clay. 98 Stony loam. 31 Deep gray silt loam. 99 Rock outcrop. SOIL FORMATIONS AND CLASSIFICATIONS 57 The soil strata are commonly classed as top soil and subsoil, and in the name of a soil type the character of the top soil is indicated, and also that of the subsoil if it is peculiar or markedly different from the top soil. In the detail soil survey of Illinois conducted by the State Experiment Station, which now covers about thirty counties, or one third of the state, the preceding soil types have been recognized and mapped, and records are kept under the numbers and names given. The system of numbering (similar to the Dewey library system) is flexible, and permits additions of main types or related types (by decimals), and the name is designed to carry with it a definite suggestion of the character of the soil. CHAPTER VI SOIL COMPOSITION SOILS IN GENERAL ASIDE from the organic matter, any soil material (excepting quartz sand, but including granitic sand) will commonly contain all of the elements found in ordinary silicate rocks, but, of course, in very varying proportions. Soils contain large amounts of silicon and much aluminum and sodium, none of which are essential to plant growth, also very large amounts of oxygen, an element which as plant food is supplied in the carbon dioxid taken into the plant through the leaves. This means that about 85 per cent of the solid crust of the earth has no value as plant food. This includes not only silicon dioxid (as quartz sand), aluminum silicate (as pure clay), and aluminum sodium polysilicates, but also these elements when present in other combinations. The remaining abundant elements, iron, calcium, magnesium, and potassium, are all essential as plant food. Of these four, iron is the most abundant in the earth and the least abundant in plants, and, so far as the writer is aware, soil has never been known to be- come deficient in iron as measured by crop requirements. Calcium and magnesium are somewhat less abundant than iron, and are required by crops in very much larger amounts, and on some soils crop yields are appreciably increased by the application of one or both of those elements in suitable compounds, but in many or most such cases the increase in crop yields is not due to the direct effect of the calcium or magnesium as plant food, but rather to the indirect effect their compounds may produce in increasing the availability of other less abundant plant-food elements. In the average crust of the earth, potassium is slightly more abundant than magnesium but less abundant than iron or calcium. Of these four elements, potassium is required by plants in greatest 53 SOIL COMPOSITION 59 amount, but nevertheless the total supply of potassium in nearly all soils is exceedingly large compared with crop requirements; and, while it has a money value in commercial fertilizers and is quite extensively used, there is much evidence to show that on many soils the influence which it produces is due in part at least to in- direct effects, as in the liberation of other more deficient plant- food elements. Sulfur and phosphorus are not in the same class with the eight abundant elements composing the silicates; and between these two elements there are also marked differences, since sulfur is brought to the earth in rain in considerable amounts and is also about as abundant as phosphorus in the earth's crust, while crops require from three to ten times as much phosphorus as sulfur. If we disregard the three elements which agricultural plants ob- tain from the air and water (in CO2 and H2O), as being in large measure beyond our control, we may secure a clear conception of the relative abundance of the remaining essential plant-food ele- ments, based both upon the most original natural supplies and upon crop requirements, by a study of Table 8. TABLE 8. RELATIVE "SUPPLY AND DEMAND" OF SEVEN ELEMENTS ESSENTIAL PLANT-FOOD ELEMENTS POUNDS IN 2 MIL- LION OF THE AVERAGE CRUST OF THE EARTH POUNDS IN 100 BUSHELS OF CORN (Grain only) NUMBER OF YEARS' SUPPLY INDICATED (See Table 2) Phosphorus 22OO 17 I3O Potassium 4Q2OO 10 2600 Magnesium 48000 7 7OOO Calcium 68800 ii iCtJOOO Iron 88600 I 2OOQOO Sulfur 22OO I IOOOO Nitrogen in air 70 million Ib. IOO 7OOOOO over one acre Two million pounds correspond to the weight of the plowed soil of an acre of average land to a depth of -6| inches (counting 300,000 pounds per acre inch), so that the supply of the plant-food elements given in Table 8 is simply what would be contained in an acre of 60 SCIENCE AND SOIL plowed soil if it represented the average composition of the solid crust of the earth. Corn is the most important American crop, and the common farm practice is to retain on the farm the corn- stalks (stover), so that the plant food removed in the grain is of the greatest consideration. While there is probably no cultivated soil whose composition is exactly the same as the average of the earth's crust, and while 100 bushels of corn per acre is about four times the average yield for the United States, nevertheless the data given in Table 8 pre- sent the broadest possible conception of the great problem of soil fertility in relation to permanent agriculture; because all soils are made essentially from the earth's crust, and, if some are richer, others are certainly poorer, than this general average. Likewise the loo-bushel yield of corn is of immediate interest, for it has been produced, — and can be produced throughout the corn belt in normal seasons with good farming on the richest and best-treated soils; and the production of large yields is an essential considera- tion, both from the standpoint of profitable farming and for the future support of a rapidly increasing population. There are natural agencies which may operate under different conditions to enrich, deplete, or maintain the fertility of the soil. In the formation of residual soils from the leaching of disinte- grating and decomposing rock materials, as illustrated in Tables 4 and 5, the percentage of a given plant- food element may increase or decrease or remain constant, depending upon whether the compound in which that element occurs is proportionately less or more soluble than the bulk of the material. Thus in the decom- position of gneiss (Table 4), it is evident that the alkali bases, as potassium, magnesium, calcium, and sodium, were leached out much more rapidly than the iron, aluminum, silicon, and phos- phorus; and consequently the per cent of phosphorus doubled and the per cent of potassium and magnesium markedly decreased, while the calcium practically disappeared. On the other hand, in the formation of residual clay from limestone (Table 5), the per cent of phosphorus increased distinctly and the per cent of calcium very greatly, while most of the elements in the silicate minerals, including potassium and sodium, very markedly increased in per- centage. SOIL COMPOSITION 61 In level upland areas, such as the loess-covered prairies of the Central West, which neither receive deposits from overflow nor lose partially depleted soil by erosion (especially while covered by prairie grasses), the operation of the natural laws tends steadily toward soil depletion, with respect to the valuable mineral elements; and this law has been in operation since the glacial age, or since the loess was deposited, wherever the climatic conditions have been similar to those prevailing in historic time. Thus we find (as hereinafter shown) that the oldest glacial or loessial soils (as in the lower Illinoisan glaciation) are markedly poorer in total phos- phorus, potassium, magnesium, and calcium than are the simi- larly formed soils of more recent formation (as in the late Wiscon- sin glaciation). With some elements the difference is most marked in the surface soil, and with others in the subsoil. The accumulation of organic matter in the glacial or loessial soil begins sometime after its deposition and continues until a maximum is reached, after which the organic matter, as well as the valuable mineral elements, tends to decrease, the latter because of leaching, as from the beginning, and the former because the rate of decay finally exceeds the rate of growth or accumulation. Ultimately, under these natural processes, the level lands would become practically barren. All of the level upland soils of southern Illinois were far past the maximum in productive power when this country was first settled. Indeed, much of the land of central and northern Illinois was past the maximum and tending toward depletion. Probably the black clay loam soil of the flat prairie lands in the Wisconsin glaciation was almost at its maximum condition of productiveness when the White Man took possession, but even the soil of this topography (drab silt loam) was far past its prime in the lower Illinoisan glaciation. In some of the Southern states there are still to be found level upland virgin soils that are known, as a class, to be too unproduc- tive to justify cultivation. The author has collected representative samples of this class of virgin gray silt loam soils that were found upon analysis to contain less than 400 pounds of total phosphorus in 2 million pounds of surface soil, while the subsoil of adjoining moderately productive slopes contained 1500 pounds of phos- phorus. The carbonates of calcium and magnesium have entirely 62 SCIENCE AND SOIL disappeared from these level upland soils, and in their place marked acidity has developed. Under these conditions the growth of vege- tation and the fixation of nitrogen by legumes become very lim- ited, and level virgin soil, not subject to erosion, was found to contain less than one fifth as much as the average nitrogen content of the black clay loam of the late Wisconsin glaciation in northern Illinois. In the progress of geologic time, surface drainage courses are de- veloped, and all level uplands become eroded hills and valleys, thus exposing the lower subsoils with their larger supplies of unleached mineral plant food, more or less of which is spread out over the lower lying slopes or level bottom lands, which sometimes again become depleted, as broad terraces above the deepened channel. Thus, moderate soil erosion is not an unmixed evil; and, with no adequate return of mineral plant food, the bottom lands and the sloping hill lands are more permanently productive (with legumes made prominent in the crop rotations) than are the level upland soils. It is doubtful if there has ever been a land on the face of the earth, where the same soil particles have been turned with the plow year after year, that has remained productive for two centuries, with no return of mineral plant food. Even in populous China there are many level upland areas, sometimes of a hundred square miles in extent, where no one lives; and the restoration of these areas has been called the " Problem of China." " In nature all things are in equilibrium " is often stated as though it were a self-evident fact. So far as the soil is concerned, the oppo- site is essentially true, — that, in nature, there is no equilibrium. Thus an ancient forest land now lies from 10 to 300 feet beneath the Illinois black prairie, which covers the unweathered glacial drift of the most southern lobe of the Wisconsin glaciation. It is of first importance that the man who controls land, and who is thus responsible for its future productive power, should have sufficient fundamental knowledge concerning the composition of common soils and the plant-food requirements of common staple crops to furnish him a foundation of absolute facts on which to build possible systems of permanent agriculture. Because of this need, considerable space is devoted to the ultimate composition of soils as they exist on the earth to-day. SOIL COMPOSITION First, in comparison with the average composition of the earth's crust, and as a good basis of comparison for all other soils, let us consider the total nitrogen, phosphorus, and potassium in the unmanured land on the Rothamsted Experiment Station at Har- penden, England. In 2 million pounds (6f inches per acre) of the surface soil where a four-year crop rotation of wheat, turnips, barley, and clover (or beans) has been followed for 60 years, there are found, in round numbers, 2500 pounds of nitrogen, 1000 pounds of phosphorus, and 35,000 pounds of potassium. These are num- bers worth keeping in mind. In Table 9 is given the composition of four different soils, of which two (from Holland and Scotland) are extremely productive, and the other two (from Germany and Maryland) are nonpro- ductive soils from barren lands. The first is an analysis by Baumhauer of a fertile alluvial soil near the Zuider Zee, and the second is Anderson's analysis of rich wheat soil of Midlothian. The third analysis, by Johnson, is said to represent "the most sterile soil in Bavaria," and the last, by Veitch, represents the " barrens " of southern Maryland. TABLE 9. COMPOSITION OF SOILS Pounds in 2 Million (per Acre about 6| Inches Deep) PLANT FOOD VERY PRODUCTIVE SOILS NONPRODUCTIVE SOILS Holland Alluvium Scotland Wheat Soil German Barrens Maryland Barrens Phosphorus . . . . 4100 17040 1560 58460 132340 7160 3780 5880 12980 17560 72420 360 (?) trace none trace 1380 22960 180 2000 840 580 17500 Potassium Magnesium Calcium Iron Sulfur These analyses are given to show that the supply of plant food in the soil is sometimes the great factor of difference between productive and nonproductive land; but the fact should not be overlooked that in other cases other factors may also be important (as excess or deficiency of moisture, poor physical condition, 64 SCIENCE AND SOIL absence or inactivity of soil organisms, or the presence of injurious substances) . Compared with the average crust of the earth (Table 7), these two fertile soils are both characterized by their high phosphorus content. The Holland soil is low in magnesium, and the Scotland soil is low in potassium, when compared with the earth's crust, although when compared with the phosphorus supply and with crop requirements, a somewhat different view is presented. It is fair to raise the question whether the sulfur reported for the Scotland soil represents the total or only the nonvolatile, because this soil contained more than 10 per cent of organic matter, and it is now known that most of the sulfur may be lost in the ignition of such a soil. The analysis of the German soil reports " insoluble silicates," and probably the amounts given are for plant food soluble in strong acid, but Veitch's analysis of the Maryland soil represents total amounts, determined by the fusion process. It will be seen that the Holland soil contains eight times as much potassium, twenty- three times as much phosphorus, and a hundred times as much calcium as the Maryland soil. The first requisite for a good soil is that it shall be rich in plant food, but it should always be remembered that that provision alone does not insure large crops, nor does a large stock of goods in the merchant's store to-day insure a good business for him to-morrow. In this connection, we may refer to the analysis of residual clay from slightly phosphatic limestone, shown in Table 5, with i.n per cent of phosphorus, which amounts to 22,200 pounds, or more than ii tons, of phosphorus per acre in a 6f-inch stratum of 2 million pounds' weight. Other soils abnormally high in phosphorus are found in the phosphate regions of Tennessee and Central Ken- tucky. Thus, Mooers (Tennessee Bulletin 78) reports the analysis of an upland soil and a bottom-land soil, from near Pulaski, Giles County, Tennessee, showing 13,200 and 14,800 pounds, respec- tively, of phosphorus per acre in a 6f-inch stratum (2 million pounds) ; and analyses by Peter and A veritt (Kentucky Bulletin 1 26) show 12,100 pounds and 12,400 pounds of phosphorus in 2 million pounds of the surface and subsurface, respectively, of soil from SOIL COMPOSITION near Midway, Woodford County, Kentucky, and 15,330 pounds and 14,800 pounds of phosphorus in 2 million pounds of surface soil from two fields nearTebb's Station, Clark County, Kentucky. On the farm of the Kentucky Agricultural Experiment Station at Lexington, Fayette County, the surf ace soil contains 15,000 pounds of phosphorus per acre in a 6|-inch stratum, and the lower subsoil contains 100,000 pounds of phosphorus in 2 million of earth. In other words, the lower subsoil between 40 and 80 inches contains, as an average, about 5 percent of the element phosphorus, equivalent to 25 per cent of tricalcium phosphate. Notwith- standing the occasional existence of such abnormal soils, the more common soils even of Kentucky and Tennessee, outside of the limestone or phosphate regions, are very deficient in phosphorus. Four samples of residual limestone soils from tobacco planta- tions about 35 miles southwest of Havana, Cuba, were found to contain as an average 4790 pounds of acid-soluble phosphorus in 2 million pounds of soil. (Frear, Penn. Report, 1901.) In Table 10 is shown the composition of adobe soil from New . Mexico and " the characteristic red earth from the decomposition of coralline limestone on the Islands of Bermuda " (Merrill). TABLE 10. COMPOSITION OF ADOBE AND CORAL LIMESTONE SOILS Pounds in 2 Million of Soil (per Acre about 6f Inches Deep) CONSTITUENTS ADOBE SOIL, NEW MEXICO CORAL LIMESTONE SOIL, BERMUDA ISLANDS Phosphorus 8200 ^400 Potassium 28600 2OOO Magnesium 35600 ?8oo Calcium 198800 5O2OO Iron 71600 I 724OO Sulfur C2OO Silicon 4I92OO 376800 Aluminum I 3Q6OO 14=; 600 Sodium 8800 80 Chlorin 28OO M^ansjanese 2OOO Carbonate carbon 46600 12300 Water of hydration 76800 324600 Volatile 68600 224200 Oxygen, etc. . 887600 680780 66 SCIENCE AND SOIL These abnormal soils are likewise characterized by a high phos- phorus content. The coral soil is also abnormal in its extremely low potassium content, when compared with ordinary soils. Leather reports the average acid-soluble phosphorus of the " black cotton soils " of India as 520 pounds in 2 million of soil, and the analyses of eighteen other types of Indian soils show the phosphorus as varying from a " trace " to 790 pounds; while among the other four types described by him, one abnormal soil (essentially an iron ore) contained 34.10 per cent of iron and .28 per cent of phosphorus, corresponding to 5600 pounds of phos- phorus per acre in a 6|-inch stratum. Von Ugrimov's analyses 1 of the cultivated " black earth " soil of southwest Russia shows only 260 pound's of acid-soluble phos- phorus in 2 million of soil; while Hilgard 2 gives .13 per cent of P2O5, corresponding to 1130 pounds of phosphorus in 2 million of cultivated soil, and .14 per cent of P2O5, or 1220 pounds of phos- phorus, in 2 million of virgin soil. The fact that the samples se- cured upon his request and analyzed by Hilgard showed 5.54 per cent of humus in the cultivated soil and only 5.11 per cent in the virgin soil, leads one to question whether the sample referred to as cultivated soil, containing 4800 pounds of nitrogen, and acid- soluble minerals amounting to 1130 pounds of phosphorus, 8600 of potassium, 9000 pounds of magnesium, and 18,300 of calcium (in 2 million of soil), can fairly represent the black earth soil of Russia whose average yield of wheat for 20-year periods is less than 10 bushels per acre in a three-year rotation, including one year of green fallow. (See Bulletin 42, Bureau of Statistics, United States Department of Agriculture.) The report of Von Ugrimov's investigations states that "pot and field experiments with wheat, and analyses of the crop produced, bear out the chemical analysis in indicating that phosphorus is the element of plant food especially needed in this soil." Analysis of " typical soils " of British East Africa shows that they are fairly well supplied with nitrogen and potassium, but deficient in phosphorus, — "a deficiency which is stated to .be common throughout East Africa." 3 1 Experiment Station Record, 10, 1015. 2 "Soils," page 364. 3 Experiment Station Record (1908), 10, 1015. SOIL COMPOSITION 67 Investigations by Ingle 1 (as chief chemist for the Transvaal Department of Agriculture) showed that " analyses of Transvaal soils indicate that they are, as compared with English soils, very poor in phosphorus, nitrogen, and lime, but usually rich in po- tassium." The Massachusetts Experiment Station (Bulletin 117) reports the following analysis of soil from Turkey, Asia, the amounts per acre being computed for 2 million pounds of surface soil (about 6f inches deep). Nitrogen 06 per cent, or 1200 pounds per acre. Phosphorus " none " Potassium 51 percent, or 10,200 pounds per acre. Calcium 72 per cent, or 14,400 pounds per acre. The 10,200 pounds of acid-soluble potassium is probably much below the total potassium present. Professor J. B. Harrison has recently reported 2 that the soil of the Experiment Station Farm in British Guiana, South America, contains 43,600 pounds of total potassium in 2 million of soil. The amount of phosphorus is not reported. In the volcanic ash ejected from Vesuvius during the eruption of April 4 and 5, i9o6,Comanducci found .33 per cent of phosphorus and 3.87 per cent of potassium, — amounts which correspond to 6600 pounds of phosphorus and 77,400 pounds of potassium in 2 million pounds of the volcanic material. The surface of the United States may be divided into two areas, the glaciated and the unglaciated, as shown on the accompanying map. In general, the great ice sheets moved from north to south, and as they flowed slowly over the face of the earth, they caused enormous erosion of the surface. The eroded material was carried forward in the ice, and much of it was ground to powdered form, while some was reduced only to the form of rounded bowlders, pebbles, and sand grains. This mixture embedded in silt and clay is called glacial drift, or till, or bowlder clay. As the ice melted, the drift material was deposited, sometimes in moraines, or ridges, where for a long period of time the forward movement of the gla- 1 Journal of Agricultural Science, December, 1908. 2 West Indian Bulletin (1908), 0, 9. 68 SCIENCE AND SOIL cier was practically equaled by the rapidity with which the ice melted at the terminus, and sometimes over broad inter-morainal tracts, where the melting proceeded more rapidly. Many preglacial valleys were filled with the drift (in places 300 feet deep) , and com- monly glacial drift was deposited over the general level of the glaciated area to a depth of 10 to 100 feet. We are especially interested in four of the important ice sheets that occurred during the glacial epoch. Where the drift from the first of these was not covered by a subsequent glacier, the area is termed the Kansan glaciation; where the drift from the second gla- cier was not covered by a subsequent glacier, the area is termed the Illinoisan glaciation; where the drift from the third glacier was not covered by a subsequent glacier, the area is called the lowan glaciation; and the area covered by the drift from the fourth glacier is termed the Wisconsin glaciation, where not covered by a subse- quent glacier. As will be seen from the glacial map, these respective areas are not confined to the states named. It should be understood that, notwithstanding the extensive glaciated regions, glacial soils are not common in the older glaciated areas. The most common soil material between the Alleghanies and the Rocky Mountains, and between the Great Lakes and the Gulf, is loess. Loess is a very fine material 1 consisting of grains of quartz, felspar, mica, hornblende, and other granitic or silicate minerals, with more or less limestone, dolomite, magnetite, pyrite, etc., and some clay. Loess has been derived in large part from glacial drift, having been transported by the action of wind and flowing water, probably from deposits of exposed till before it was protected by vegetation (and to some extent from the melting or evapo- rating glaciers), and deposited over all other soil formations and over older glaciated areas. Many of the residual soils in the drift- less, or unglaciated, areas in the Mississippi Basin are now covered with loess. Even the tops of theOzark Hills of southern Illinois, beyond the most southern point of the glacial lobe, and high above 1 In some places the loess is more or less mixed with the underlying residual or glacial materials, through the action of crawfish, burrowing animals, etc., and occasionally loess deposits are subsequently covered by mixed alluvial materials, which may include sand and gravel with silt, clay, and organic matter. 1«7 157 147 137 127 117 107 87 87 77 07 67 47 87 I Ionian Drift Wisconsin and later drift sheets I Existing glaciers and ice sheets \ Quaternary lakes (Bonnevillc. Latimtan ami Uladal Lute .!,/«»««) J I 107 Longitude 97 West from 87 Greenwich 77 L.L.PI MAP OF THE GLACIATED AREA OF NORTH AMERICA From Upham's map as modified by Leverett, United States Geological Survey SOIL COMPOSITION 69 the nearest glacial ridges, are covered with several feet of loess. The older glacial drift is usually loess-covered. The depth of loess varies from three feet or less in the somewhat recent glaciations, and in driftless areas, remote alike from the glacial borders and from large stream courses, to eight or ten feet in areas near the borders of the greatest glacial action; while in the "deep loess" areas covering the bluff lands along some of the large streams the depth of loessial material may be from ten to fifty feet or more. Some very complete analyses have been made of samples of loess from widely separated areas. The results given in Table 13 are reported by the United States Geological Survey. The first three represent loess deposits covering the bluffs at Galena, Illi- nois; Vicksburg, Mississippi; and Kansas City, Missouri; while the fourth " was taken from the summit of a ridge in the suburbs of Dubuque, Iowa, at a point about 300 feet above the Mississippi River." TABLE n. COMPOSITION or LOESS DEPOSITS Pounds in 2 Million of Loess CONSTITUENTS GALENA, ILLINOIS VICKSBURG, MISSISSIPPI KANSAS CITY, MISSOURI DUBUQUE, IOWA Phosphorus .... 600 I2CO 800 20OO Potassium 34400 I80OO 30600 35600 Magnesium .... 44400 55°°° 13400 13400 Calcium 77200 128000 24200 22800 Ferric iron .... 36600 36600 45400 49400 Ferrous iron .... 8000 10400 1800 I5OOO Sulfur 800 IOOO 4OO 4OOO Silicon 606800 "? 60800 *f w 699600 T^ 6826OO Aluminum 112600 k} wv 84200 I 29800 I274OO Sodium 2OOOO 1 74.00 2I2OO 242OO Manganese .... 800 * / Hw^ 1800 4OO IOOO Titanium 4800 6200 1600 8600 Carbonate carbon . . 34400 52600 26OO 220O Organic carbon . . . 2600 3800 24OO I8OO Water and organic hy- drogen 41000 22800 54000 5OOOO Oxygen, etc Q7COOO 99I2OO 0722DO 96OOOO V/JWW V / 70 SCIENCE AND SOIL These analyses show the general range in composition of the mineral material constituting the bulk of our most common, most extensive, and most valuable soils, in the central part of the United States, both north and south (along the Mississippi Valley), where, as a very general rule, the surface is covered by a blanket of loess two feet or more in depth. In the fresh condition, as in the deeper strata, loess usually contains considerable amounts of calcium carbonate and more or less magnesium carbonate, as is the case with the samples from Galena and Vicksburg, both of which are known to represent strata of considerable depth. The Dubuque sample was evidently taken from the surface, and this may be the case with the Kansas City sample, in both of which the carbonates have evidently been greatly reduced by leaching. As an average, the phosphorus content amounts to 1150 pounds per acre for a stratum of 6f inches (2 million pounds) , including the Galena sample, which is decidedly low, and the Dubuque sample, which is abnormally high. The average of nine different composite samples of subsoils from different places in the deep loess areas in Illinois shows mo pounds of total phosphorus in 2 million pounds of loess, the extreme variation being from 740 to 1540 pounds. (See Illinois Experiment Station Bulletin 123, pages 288-289.) While phosphorus is, in a sense, an incidental substance in loess deposits, the element potassium is an important constituent of the most common original minerals, and any marked variation in potassium content must be accounted for largely by decomposition and loss by weathering of the particles, the chief losses having occurred probably before the accumulation into the present loessial deposits. The average of the two northern samples (Galena and Dubuque) shows 35,000 pounds of potassium, while the sample from the south- west (Kansas City) shows 30,600 pounds, and the southern loess contains only 18,000 pounds of potassium in 2 million. If loess is derived chiefly from glacial drift, as viewed by the United States Geological Survey (38th Monograph, page 159), then it would be expected that the southern loess, transported far from glacial de- posits, would be lower in potassium than the northern loess, which has been less exposed to weathering. As an average of the nine SOIL COMPOSITION 71 composite samples of subsoil from different places in the deep loess areas of Illinois, 35,070 pounds of potassium were found in 2 million of loess. In the loess which was not mixed with carbonates, or from which most of the carbonates have been removed, presumably by leach- ing (Dubuque and Kansas City samples), the total supply of mag- nesium and calcium is markedly smaller than the supply of potas- sium; but when compared with the average requirements of a general crop rotation (see Table 13) the supply of magnesium and calcium is still somewhat more ample than that of potassium. While the average supplies of sulfur and phosphorus are about equal, the requirement for phosphorus is five times as great as for sulfur in the total produce of the average crop rotation, and forty times as great if the grain only is removed and not returned. A partial analysis of loess from Cheyenne, Wyoming, reported by Eakin, shows, in 2 million pounds of loess, 960 pounds of phos- phorus, 44,500 of potassium, 14,880 of magnesium, 69,700 of calcium, and 20,000 pounds of carbonate carbon. It is suggested that most of the carbonates contained in deep loess deposits may have a different origin than the silicates, which constitute the bulk of the material. In many deep loess deposits, pieces of limestone shells (usually of light weight) are a characteris- tic, indicating that a part of the loessial material may have come from areas which were at times covered with water and at other times dried on the surface and exposed to wind action. Thus, while such loess may have been derived from glacial drift, more or less of it has had some intermediate resting place where carbonates tend to accumulate, as in ponds, shallow lakes, swamp areas, bottom lands, or on seepy slopes; and from the dried surfaces of such areas much of it has been transported .by wind action over bluffs and upland plains. It is noteworthy that the broadest deep loess deposits along the river bluffs are found where the valley is correspondingly wide (Illinois Bulletin 123, page 238). The most definite lesson to be drawn from these analyses of this most important soil material is, that phosphorus is clearly the most limited element of plant food; whereas among the four elements, potassium, magnesium, calcium, and sulfur, it is difficult to de- termine which is likely to be the most limited. 72 SCIENCE AND SOIL SOME EASTERN RESIDUAL SOILS InTable 12 is given the ultimate chemical composition of residual soils derived from ten different geological formations, and as a rule the results are the average from several samples of the same type of soil. The soils were collected in Maryland, but in most cases the same soil types extend into other states and may be con- sidered as more widely representative of these soil formations. The soil analyses 1 were made by Mr. F. P. Veitch of the United States Department of Agriculture, under the direction of Professor Milton Whitney, Chief of the Bureau of Soils, and with the indorse- ment of Doctor H. W. Wiley, Chief Chemist of the Department of Agriculture. In all cases the samples analyzed were taken " immediately under the top soil," and thus represent the upper stratum of the subsoil. The results are given in Table 12 on the basis of pounds of the different elements present in 2 million pounds of the soil, corresponding approximately to the amounts per acre in a" 6f -inch stratum. " Volatile " means loss on ignition, and includes organic matter, combined water, probably some sulfur, which may be oxidized from organic matter or from pyrites, and possibly some carbon dioxid (not completely replaceable if derived from magne- sium carbonate). Oxygen may be lost or gained during ignition, depending upon the compounds of iron, sulfur, etc., the amount of organic matter, and the stage to which the ignition is carried. The oxygen is estimated by difference, which really includes errors and all undetermined elements not otherwise reported. It should be kept in mind that sandstone does not mean quartz. It means a stone with sand grains cemented together. The sand grains may consist of quartz (silicon dioxid), but more commonly they are grains of silicate minerals, including much aluminum and iron and more or less of the other abundant mineral elements. Residual soils resulting from the disintegration, decomposition, and leaching of the previous geological formations vary with the character of the original rock, and with the loss by leaching. In the case of the limestone formations, it is apparent that the residual soil consists of impurities contained in the original limestone, the 1 Maryland Agric. Expt. Station Bulletin 70. SOIL COMPOSITION 73 c3 ^ El" O O 00 O H 00 Jo o o o o o 80 o o o o o O 0 O NO •2u H IO •^•00 ON to O ON H O CNI o CS 10 ON CO Tj- PI OH to CNI H ON _« O O 00 O O O 0 O o o o t JO O O O O O O 0 3 Jj o CO'HH M M ON J~~» NO CO ON "fr w.s H H OO M NO NO H H ON M slderberg imestone 80 o M O "T O O O O O O O . M OO H H t>> 80 o o o o o O O £<& oo oo O co KJ NO M H ON II O O 0 O 0 O O O O CN] o o o o o o o o O O o t C ^ CO NO rf CNI O OO NO ^3" **}* O oo •* ONOO 00 M "^ GO W M 8 CS Tf- ^ 0* ON •t en >O H ON 0 O 0 £. o O O O O O O O ^3" 8 8 8 g 0 0 O 't 1 to O-*» fi •*" Tj- OO CS O 0 w to •* O M H CN W IO H £ r^* oo . IO (N w OO „ 0 0 •^f O O Q Q O §O O 00 Q Q O O o o o o Q Q O 00 'S t~~ CNI VO Tf" CO •* o o NO O NO a CN| t>- IO *^ M ON t°"~ t""* M Tj* NO O CN| M OO CN| CN| NO M M ON CONSTITUENTS Phosphorus . Potassium Magnesium Calcium . . Iron . . . Sulfur . . . c.S £ ^ o (-1 ^3 M) 8 S .2 c •3 3 'O CN? "5 r3 o ^ C/2 < C/D <5 . O U OJ C ~ on^y aoout one sixth of the total potassium contained in old soils is thus dissolved, as a general average, although the proportion varies greatly with different types of soil. Doctor Frear has subsequently furnished data showing that 2 million pounds of the fine earth in the surface soil on the Pennsylvania State College farm contain 50,700 pounds of total potassium. SOIL COMPOSITION 77 SOILS OF THE CENTRAL STATES The accompanying soil map of Illinois and Tables 15, 16, and 17 serve to illustrate in a very trustworthy manner both the uniform- ity and variation that may be expected among the most important soil types in the North Central States. This detailed information from Illinois applies with almost equal value to similar soils in many other states. With a north and south extension of nearly 400 miles in the center of the greatest agricultural region of the United States, Illinois occupies a unique position. In latitude it reaches almost from Vermont to North Carolina, Cairo being farther south than Richmond, and Beloit farther north than Boston. Cairo is within 35 miles of the Tennessee line, and 150 miles south of Covington, Kentucky. From Cairo to Mobile on the Gulf is no farther than from Beloit to the 49th parallel, which marks the northern boundary of the United States. The soils of Illinois are in large measure representative of the soils of the wheat belt, of the corn belt, and, in part, of the cotton belt. Cotton growing has been a commercial success in southern Illinois, and much spring wheat has been produced in the north end of the state, while central Illinois is the heart of the corn belt. Fourteen great soil areas are recognized in Illinois, including the extensive unglaciated regions in the southern and northwestern parts of the state, the lower, middle, and upper Illinoisan glacia- tions, the pre-Iowan and lowan glaciations, the early and late Wisconsin glaciations, with numerous moraines and intermorainal tracts, the deep loess deposits, and the early and late swamp and sand areas, and extensive and widely distributed bottom lands and terraces. As already explained, the material called loess constitutes the chief basis for nearly all of the upland soils of central United States. The principal differences among these soils of loessial origin are due to difference in age, topography, and climatic conditions. Some additional or subsequent differences have been brought about by variation in native vegetation and in systems of farming. Prairie and timber soils. The upland soils may be divided into prairie soils and timber soils, according to the character of the original vegetation; and with similar topography the difference in 78 SCIENCE AND SOIL vegetation is not due to original differences in the soil materials ; but rather the difference between prairie land and timber land is due to the influence of the vegetation upon the soil. The existence of prairies over areas naturally well surface-drained is due very largely, if not entirely, to the prairie fires, which were, as a rule, of annual occurrence and often a source of danger to the early settlers in prairie regions. The annual destruction of any seedlings that may have started effectually prevented the growth of forests, on the prairie lands, and it is noteworthy that level areas or valleys on the northeast side of streams were usually timbered, while corresponding areas on the southwest were usually prairie, because of the prevailing southwest winds during summer and autumn. (See " Soils of Clay County, Illinois.") Prairie fires have no tendency to run down hill, and they make but little progress against the wind. The wild prairie grasses and weeds, including native legumes, developed an abundant root system to an average depth of 16 to 20 inches, varying somewhat with the latitude or length of season, the depth being greater in the latitude of central Illinois than in northern Illinois; and, with the partial decay of these roots, followed the marked accumulation of humus which characterizes the " black soil " of the prairie; while the smaller amount of humus is the chief characteristic of the timber soils. Rotting tree roots are subject to very complete decay, because of the large cavities and ready admission of air. Boring insects and burrowing animals also hasten the destruction, so that the small amount of leaf mold that remains constitutes the main source of humus for timber soils, and even this is exposed to rapid decay. Being poorer in organic matter, the upland timber soils are correspondingly poorer in nitrogen than the prairie soils. The prairie lands may be classified according to topography, as undulating prairies and flat prairies. The undulating prairie soil covers the nearly level or gently rolling prairie lands that were naturally fairly well surface-drained. It is usually markedly uniform in a given formation, and consti- tutes the most important soil of the corn belt. More or less of the clay and finer silt has been carried downward from the surface and accumulated in the subsoil, and some has been carried away by SOIL COMPOSITION 79 surface washing during many centuries and collected in lower lying flat areas. The exposure of the surface after the annual prairie fires permitted some slight surface washing which other- wise would not have occurred. The flat prairie soils occupy the lower lying level areas that were naturally poorly surface-drained and inclined to be swampy, espe- cially during the wet season of the year. This soil has been formed in part from deposits of fine earth and vegetable matter washed in from the surrounding higher land. The rank growing swamp grasses have, from the partial decay of their roots (and of more or less of their tops), added much organic matter to this soil. The undulating prairie soils vary from a gray silt loam on tight clay in the older areas, to a dark brown silt loam, in the later forma- tions, and the common flat prairie soils vary with age from drab silt loam to black clay loam. Many other less extensive soil types occur here and there on the prairies, including, as extremes, sand dunes formed of wind-blown material from old shallow lake beds, gravel points, or exposed glacial till, bogs of peat or muck, and sometimes adjoining strips of plastic clay. Some intermediate types include deep silt loam, sandy loam, silt on clay, etc. These are of small importance compared with the very extensive and most common prairie soils reported in Tables 15, 16, and 17. There are three principal types of upland timber soils in most of the great loess-covered areas. One, a light gray silt loam, occupies the flat areas ; a second type, yellow-gray silt loam, covers the undulating or gently sloping lands; and the third (yellow silt loam) is hilly or steeply sloping and consequently subject to serious ero- sion, or surface washing, especially when under cultivation. In addition, there are the areas of deep loess (yellow fine sandy loam) covering the bluffs in many places along the larger river valleys; and other less extensive types are sometimes found. In northern Illinois and southern Wisconsin, in what is termed the lowan glaciation, considerable areas are found of a brown sandy loam, occupying in the main the undulating uplands. The top soil con- sists of brown sandy loam, containing some gravel in places and occasionally pieces of stone. The subsoil at a depth of three feet or more frequently contains much stone, the proportion increasing 8o SCIENCE AND SOIL with the depth, the disintegrating bed rock being found at 4 to 10 feet beneath the surface. The bed rock and some of the pieces found in the soil and subsoil consist of impure limestone. The soil is commonly recognized as drift, but it is certainly much modified by the residual material, and in places there is but little evidence of glacial or loessial deposit. Sand, swamp, and bottom lands. In the older formations, as in the Illinoisan glaciations and still farther south, the soil of the smaller river bottoms is chiefly a deep gray silt loam; while, in the more recently formed great soil areas, the principal bottom land soil is a brown loam. In both cases the bottom land resembles somewhat the top soil of the adjoining upland (which has con- tributed much to its formation) , modified by additions of humus and alluvium from other sources. Many other types of bottom land are also found, but usually they are less abundant. Extensive swamp regions are found in most of the Northern States, especially in Michigan, Wisconsin, and Minnesota, and in the northern parts of Ohio, Indiana, Illinois, and Iowa. These swamp soils vary almost from pure sand to pure clay, and almost from 100 per cent mineral matter to 100 per cent organic matter; and they also vary from moderately acid soils to marls containing more than 50 per cent of calcium carbonate, and not infrequently magnesium carbonate is present in sufficient amount to render the soil non-productive and place it in the alkali class. These different constituents vary quite independently, sand, peat, clay, peaty sand, sandy peat, peaty clay or clayey peat (muck), sandy clay, clayey sand, loam, sandy loam, clay loam, and peaty loam being among the possible soil types; and any of these may be acid or may contain " alkali." In addition we find such variations as deep peat, medium peat, and shallow peat, with sand or clay or sandy clay subsoil. In places there are broad, level, and very uniform areas of deep peat, of peat on sand, or of nearly pure sand; and in other places peat bogs and sand ridges alternate every few rods. Not infre- quently sand dunes (still subject to more or less wind action) are found in or adjoining these swamp regions; and in some sections there are more extensive sand regions, including considerable parts of counties in Ohio, Indiana, Illinois, and Wisconsin, and an area of several counties in the north central part of the lower peninsula of Michigan. SOIL COMPOSITION 81 In case of the most important soil types the averages reported in Tables 15, 16, and 17 are based upon analyses of a large number of composite soil samples. The two most extensive soil types in Illinois are the gray silt loam prairie (330) of the lower Illinoisan glaciation (the so-called hard-pan soil of " Egypt "), and the brown silt loam prairie (1126) of the early Wisconsin glaciation. The averages reported for the gray silt loam on tight clay of the lo\ver Illinoisan glaciation represent 57 different soil samples. In 2 million pounds of the surface soil the potassium (the most constant constituent) varied from 23,120 to 26,440 pounds, the phosphorus varied from 700 to 1000 pounds, and the nitrogen varied from 2140 to 3500 pounds; and in every case the surface, subsurface, and subsoil were found to be acid. The data reported for the brown silt loam of the early Wisconsin glaciation are averages obtained by analyzing 90 different samples of soil collected in ten different counties, and representing more than 500 different borings. In 2 million pounds of the surface soil of this type the potassium varied from 31, 980 to 43,100 pounds, the phosphorus varied from 980 to 1620 pounds (or, if we disregard four samples, from 1020 to 1340 pounds), and the nitrogen varied from 3980 to 7520 pounds (or from 3980 to 6340, if we disregard two samples). The limestone has not been leached out of the early Wisconsin brown silt loam to such a depth as in the older gray silt loam. In one case limestone was present in the surface of the brown silt loam, and in three cases it was found in the subsurface, while it was more often present in the subsoil, although in many cases even the subsoil was found to be acid, but never to such a degree as is common for the subsoils of the older brown silt loams (middle and upper Illinoisan, pre-Iowan, and lowan). With unimportant exceptions, all samples of surface, subsur- face, and subsoil of the yellow silt loams (and the yellow fine sandy loam) of the hill lands were distinctly acid, the degree varying with the age of the soil. In the late Wisconsin glaciation both the brown silt loam and the yellow-gray silt loam samples were almost invariably slightly acid in the surface and subsurface, but exceedingly well supplied with carbonates in the subsoil. 82 SCIENCE AND SOIL TABLE 15. FERTILITY IN ILLINOIS SOILS Average Pounds per Acre in 2 Million Pounds of Surface Soil (o-6f Inches) ' Son, TYPE NO. SOIL AREA OR GLACIATION SOIL TYPE TOTAL NITRO- GEN TOTAL PHOS- PHOR- US TOTAL POTAS- SIUM LIME- STONE PRESENT LIME- STONE RE- QUIRED PRAIRIE LANDS, UNDULATING 33° Lower Illinoisan Gray silt loam on tight clay 2880 840 24490 1160 426 Middle Illinoisan Brown silt loam 437° 1170 32240 70 526 Upper Illinoisan Brown silt loam 4840 1 200 32940 70 626 Pre-Iowan . . Brown silt loam 4290 1190 35340 no 726 lowan . . . Brown silt loam 4910 1 220 32960 90 1126 Early Wisconsin Brown silt loam 5050 1190 36250 60 1026 Late Wisconsin Brown silt loam 6750 1410 45020 60 PRAIRIE LANDS, FLAT 329 Lower Illinoisan Drab silt loam 2800 710 26260 1300 420 Middle Illinoisan Black clay loam 54io 143° 31860 830 520 Upper Illinoisan Black clay loam 6760 1690 29770 30 II2O Early Wisconsin Black clay loam 7840 2030 3SI40 32530 I22O Late Wisconsin Black clay loam 8900 1870 37370 980 TIMBER UPLANDS, ROLLING OR HILLY 135 Unglaciated Yellow silt loam 1890 950 3*450 80 335 Lower Illinoisan Yellow silt loam 2150 950 3l85o 310 435 Middle Illinoisan Yellow silt loam 1870 820 33470 40 535 Upper Illinoisan Yellow silt loam 20IO 840 34860 130 635 Pre-Iowan . . Yellow silt loam 2390 850 37180 30 735 lowan . . . Yellow silt loam 1910 910 3578o 30 "35 Early Wisconsin Yellow silt loam 1890 870 32720 60 864 Deep loess . . Yellow fine sandy loam . 2170 960 35640 70 1 The numbers given in Table 15 represent the total amount contained in 2 million pounds of the surface soil on the dry basis, with the exception of deep peat swamp soil, for which the amounts in i million pounds are used, because its spe- cific gravity is only one half that of ordinary soil, and of sand soil for which -2\ million pounds are used, because it is about one fourth heavier than ordinary soil. SOIL COMPOSITION TABLE 15. FERTILITY IN ILLINOIS SOILS — Continued SOIL TYPE NO. SOIL AREA OR GLACIATION SOIL TYPE TOTAL NITRO- GEN TOTAL PHOS- PHOR- US TOTAL POTAS- SIUM LIME- STONE PRESENT LIME- STONE RE- QUIRED TIMBER UPLANDS, UNDULATING 1034 Late Wisconsin Yellow -gray silt loam . . . 2890 810 47600 40 760 lowan . . . Brown sandy loam 3070 850 26700 100 TIMBER UPLANDS, FLAT 332 Lower Illinoisan Light gray silt loam on tight • clay . . . 1890 810 27280 45^ SAND, SWAMP, AND BOTTOM LANDS r33! Old bottom lands Deep gray silt loam . . . 3620 1420 36360 440 i45i Late bottom lands Brown loam 4720 1620 39970 2090 1481 Sand plains and dunes . . . Sand soil . . 1440 820 30880 200 1401 Late swamp . Deep peat . . 34880 1960 2930 130 1415 Late swamp . Drab clay . . 5760 1900 48080 36300 1400 Late swamp . Marly peat . . 20900 1520 920 1278000 A careful study of the mass of evidence recorded in Tables 15, 16, and 17 clearly reveals the fact that the most important and most extensive areas of Illinois soils are poor in phosphorus. The only soils well supplied with phosphorus are the black clay loams, the bottom lands, and the clay and peaty swamp soils. On the other hand, the supply of total potassium is very great in all of the soils reported upon, with the exception of the deep peat and the abnor- mal marly peat, which are markedly deficient in that element. It should be kept in mind that small bodies of peat soil surrounded by normal upland soils rich in potassium are likely to have received deposits of silt or clay by overflow from time to time, and as a rule they are not deficient in potassium, and shallow peat bogs with clay SCIENCE AND SOIL TABLE 16. FERTILITY IN ILLINOIS SOILS Average Pounds per Acre in 4 Million Pounds of Subsurface Soil (6f-2o inches) Son. TYPE No. SOIL AREA OR GLACIATION SOIL TYPE TOTAL NITRO- GEN TOTAL PHOS- PHOR- US TOTAL POTAS- SIUM LIME- STONE PRESENT LIME- STONE RE- QUIRED PRAIRIE LANDS, UNDULATING 33° Lower Illinoisan Gray silt loam on tight clay 3210 1500 5357° 6350 426 Middle Illinoisan Brown silt loam 5800 1920 62590 no 526 Upper Illinoisan Brown silt loam 6480 2090 64820 1 20 626 Pre-Iowan Brown silt loam 4650 2060 72370 570 726 lowan . . . Brown silt loam 5*40 1940 6622O 360 1126 Early Wisconsin Brown silt loam 6560 2OOO 72780 90 1026 Late Wisconsin . Brown silt loam 6870 1960 96420 150 PRAIRIE LANDS, FLAT 329 Lower Illinoisan Drab silt loam 3160 1230 54420 3980 420 Middle Illinoisan Black clay loam 6180 2260 64070 2940 520 Upper Illinoisan Black clay loam 7380 2690 60760 70 1 1 20 Early Wisconsin Black clay loam 7200 3090 71670 49300 1220 Late Wisconsin . Black clay loam 9100 2860 78840 1310 TIMBER UPLANDS, ROLLING OR HILLY i35 Unglaciated . . Yellow silt loam 2030 21 2O 67320 4850 335 Lower Illinoisan Yellow silt loam 2170 2OOO 67380 6630 435 Middle Illinoisan Yellow silt loam 1980 1510 65370 37° 535 Upper Illinoisan Yellow silt loam 1900 1610 72570 222O 635 Pre-Iowan . . Yellow silt loam 2290 i75o 76150 650 735 lowan .... Yellow silt loam 2120 1960 71180 1500 "35 Early Wisconsin Yellow silt loam 1870 159° 68690 3910 864 Deep loess . . Yellow fine sandy loam . 26lO 1600 71760 830 TIMBER LANDS, UNDULATING 1034 Late Wisconsin . Yellow-gray silt loam . . . 2710 1390 IIIIOO 340 760 lowan . . . . Brown sandy loam 3920 T59° 543°° 62O SOIL COMPOSITION TABLE 16. FERTILITY IN ILLINOIS SOILS — Continued SOIL TYPE No. SOIL AREA OR GLACIATION SOIL TYPE TOTAL NITRO- GEN TOTAL PHOS- PHOR- US TOTAL POTAS- SIUM LIME- STONE PRESENT LIME- STONE RE- QUIRED TIMBER UPLANDS, FLAT 332 Lower Illinoisan Light gray silt loam on tight clay .... 1920 1240 58480 7200 SAND, SWAMP, AND BOTTOM LANDS I,w Old bottom lands Deep gray silt loam .. . . 2250 1830 68090 449° 1451 Late bottom lands Brown loam . . 6660 2160 77540 980 1481 Sand plains and dunes . . . Sand soil . . . 2070 1480 62690 5o 1401 Late swamp . . Deep peat . . 64980 2940 7010 2IO subsoils are also well supplied with potassium, or may be by deep plowing; whereas broad_ areas of deep peat or of shallow or medium peat on sand are as a rule deficient in potassium. Most of the older soils (chiefly in southern and western Illinois) are markedly acid in the surface and subsurface, and exceedingly acid in the subsoil. The rolling or hilly timber uplands and the sand soil are very deficient in nitrogen, while the undulating prairie lands (except in the late Wisconsin glaciation), the undulating timber lands, and even the flat prairie lands in the oldest forma- tion, are only moderately well supplied with nitrogen. The black clay loams (especially in the more recent formations) are rich, and the peaty soils exceedingly rich, in humus and nitrogen. In the main the soils of central and northern Illinois are com- parable with similar soil types in Indiana and Ohio on the east, and with Iowa and eastern Nebraska soils on the west; and the soils of southern Illinois are comparable with similar types in Missouri and eastern Kansas on the west, and also with the loess- covered areas in southern Indiana, Kentucky, Tennessee, and northwest Mississippi; while some of the same soil types that are 86 SCIENCE AND SOIL TABLE 17. FERTILITY IN ILLINOIS SOILS Average Pounds per Acre in 6 Million Pounds of Subsoil (20-40 Inches) SOIL TYPE No. SOIL AREA OR GLACIATION SOIL TYPE TOTAL NITRO- GEN TOTAL PHOS- PHOR- US TOTAL POTAS- SIUM LIME- STONE PRESENT LIME- STONE RE- QUIRED PRAIRIE LANDS, UNDULATING 330 Lower Illinoisan Gray silt loam on tight clay . . 3240 2400 84300 21580 426 Middle Illinoisan Brown silt loam 3440 2680 90040 20O 526 Upper Illinoisan Brown silt loam 344° 2790 98580 460 626 Pre-Iowan . . Brown silt loam 3940 3380 102620 1650 726 lowan .... Brown silt loam 3540 2780 99780 1940 1126 Early Wisconsin Brown silt loam 3420 2620 117880 66600 1026 Late Wisconsin Brown silt loam 363° 2630 160140 728000 PRAIRIE LANDS, FLAT 329 Lower Illinoisan Drab silt loam 3400 1690 80830 15770 420 520 II2O 1220 Middle Illinoisan Upper Illinoisan Early Wisconsin Late Wisconsin Black clay loam Black clay loam Black clay loam Black clay loam 3020 3J4° 349° 3180 3°3° 3640 363° 393° 94900 96220 111280 125370 149200 I2IO II7500 547° TIMBER LANDS, ROLLING OR HILLY 135 Unglaciated . . Yellow silt loam 1970 3280 10543° 20660 335 Lower Illinoisan Yellow silt loam 2480 3170 99670 21500 435 Middle Illinoisan Yellow silt loam 2820 2810 99000 3700 535 Upper Illinoisan Yellow silt loam 2280 3270 100950 62IO 635 Pre-Iowan . . Yellow silt loam 2380 3400 IO2IOO 5480 735 lowan .... Yellow silt loam 2490 3900 105030 3750 "35 Early Wisconsin Yellow silt loam 2450 2660 103830 44OO 864 Deep loess . . Yellow fine sandy loam .... 2730 3320 IO52IO 3620 TIMBER UPLANDS, UNDULATING 1034 Late Wisconsin Yellow gray silt loam .... 3240 2400 156740 1034000 760 lowan .... Brown sandy loam .... 4160 2440 81180 49700 SOIL COMPOSITION TABLE 17. FERTILITY IN ILLINOIS SOILS — Continued SOIL TYPE No. SOIL AREA OR GLACIATION SOIL TYPE TOTAL NITRO- GEN TOTAL PHOS- PHOR- US TOTAL POTAS- SIUM LIME- STONE PRESENT LIME- STONE RE- QUIRED TIMBER UPLANDS, FLAT 332 Lower Illinoisan Light gray silt loam on tight clay. . 2IOO 2230 9055o 19750 SAND, SWAMP, AND BOTTOM LANDS 1331 Old bottom lands Deep gray silt loam .... 2280 2620 101610 9060 1451 Late bottom lands Brown loam . . 4150 2410 119520 4620 1481 Sand plains and dunes . . . Sand soil . . . 3100 2230 94030 80 1401 Late swamp . . Deep peat . 97730 3740 11510 290 found in the "Great American Bottoms" in southwestern Illinois, are found in the Mississippi Delta farther south. The terminal moraine of the Wisconsin glaciation extends from southern Edgar County, Illinois, to the center of Parke County, Indiana, thence in a southeast direction to the north line of Jen- nings County, thence northeast to the south line of Fayette County, and thence to the southeast corner of Franklin County. About two thirds of the state lies north of this line and resembles the timber uplands of northeastern Illinois, with a smaller propor- tion of prairie lands and considerable areas of swamp, especially in the Kankakee river basin. South of the terminal moraine the state is largely loess covered, and resembles the timber lands of southern Illinois, except for a central area of residual soils in Monroe, Lawrence, Martin, Orange, Washington, Warrick, DuBois, Crawford, and parts of most adjoining counties. In the 1907 Report of the Indiana Geological Survey, Mr. Robert E. Lyons gives data from which the following table is derived : SCIENCE AND SOIL TABLE 18. COMPOSITION OF SOUTHERN INDIANA SURFACE SOILS Pounds per Acre in 2 Million of Soil (About 6f Inches Deep) SOILS NUMBER OF ANALYSES TOTAL NITROGEN ACID-SOLUBLE PHOSPHORUS Limestone soil (Residual rolling) . . . 5 3080 *33° Limestone and shale soil (Residual rolling) i 3660 4990 Volusia silt loam (Residual shale, rolling) i 2300 1160 Miami silt loam (Loessial, level) . . . i 1 1 60 1340 Waverly silt loam (Stream valleys) . . . 2 4450 2162 Ohio river bottom land I 1840 2430 Aside from the bottom lands, these soils are as a rule markedly acid. The residual soil derived from limestone and shale is rich in phosphorus, and the bottom lands are also well supplied with that element. On the other soils much commercial fertilizer, chiefly bone meal and acid phosphate, is already being used. The upland silt loams are becoming very deficient in nitrogen and organic matter. Two analyses of the shale underlying some of the residual soils show an average potassium content of 64,500 pounds and only 3000 pounds of total calcium, in 2 million of shale. The rolling residual soil derived from shale (Volusia silt loam) and the level or gently undulating loessial soil (Miami silt loam) are very acid. A residual soil of sandstone origin is found in Martin, Lawrence, and some adjoining counties. In his discussion of Indiana soil types, Mr. Charles W. Shannon makes the following suggestive statement : " Large amounts of planer dust from the stone mills are being used as a lime application on the various soils with good results. The most noted of these experiments are in cases where from 1000 to 2000 pounds per acre of the dust was applied to fields of alfalfa and clover, and as a result much better stands were secured than in parts without the lime. This is a cheap source of lime for those who have access to the mills." As an average of 21 analyses the Ohio Experiment Station (Bulletin 150) finds, in 2 million pounds of the surface soil on the Station farm at Wooster, 1880 pounds of total nitrogen and 920 pounds of acid-soluble phosphorus; and Ohio Circular No. 79 re- SOIL COMPOSITION 89 ports data showing 31,000 pounds of total potassium in the same stratum. The average of 161 soils from various parts of Ohio shows 960 pounds of acid-soluble phosphorus in 2 million of soil. As a general average about 85 per cent of the phosphorus in such soils is soluble in the acid used, so that the total phosphorus probably amounts to about noo pounds. The average composition of three samples of surface soil from the loess-covered uplands at the Missouri Experiment Station at Columbia, in central Missouri, shows 2710 pounds of total nitro- gen, 690 pounds of total phosphorus, and 28,500 pounds of total potassium, in 2 million pounds of soil (Schweitzer, Missouri Bul- letin No. 5). These amounts correspond closely with the average composition of the most common upland soils of southern Illinois; and the more highly productive corn belt soils of north central and northwest Missouri are more nearly comparable with the brown silt loams and black clay loams of the middle Illinoisan glaciation. An analysis * of the worn upland soil near St. Louis, Missouri, shows 1160 pounds of nitrogen, 700 pounds of total phosphorus, and 35,200 pounds of potassium in 2 million of soil. This is about the average composition of the subsurface soil of the deep loess area in Illinois, and indicates previous loss of surface soil by washing. Professor Keyser has kindly furnished the author with some unpublished data concerning the soils of Nebraska, showing that the glacial silt loam of eastern Nebraska, which has been formed evidently from the weathering of the till of the Kansan glaciation, contains, in 2 million pounds of the surface, 3940 pounds of nitro- gen, 660 pounds of total phosphorus, and 23,000 pounds of potas- sium; while the ordinary loessial soil representing the most common corn belt type in the southeast part of the state (and probably of northeast Kansas as well) contains 5160 pounds of nitrogen, 1060 of total phosphorus, and 29,000 pounds of potassium, correspond- ing very closely to the brown silt loams in the loess-covered middle and upper Illinoisan glaciation. The common silt loam of the less humid region of central Nebraska contains 3680 pounds of nitrogen, 1520 of phosphorus, and 48,000 of potassium. A preliminary general soil survey of Iowa (Stevenson, Iowa 1 Reported by Doctor R. O. Graham, Bloomington, Illinois, as a commercial analysis. go SCIENCE AND SOIL Bulletin 82) shows five important soil areas in Iowa, which closely resemble similar areas in Illinois. (1) About one tier of counties bordering the Mississippi (with a western projection which includes most of Cedar, Johnson, Iowa, Poweshiek, and Jasper counties) is termed the Mississippi loess area, and resembles the loessial soil on the Illinois side. (2) Similarly, along the Missouri and Big Sioux rivers the two western tiers of counties are chiefly in the Missouri loess area. (3) The Kansan glaciation (overlaid with shallow loess on the less rolling lands) covers the southern third of the remainder of Iowa, and this resembles closely the lower Illinoisan, except that the older Kansan is more broken and has much more exposed till on the eroded hillsides. (4) The eastern part of the remainder of the state is covered by the lowan glaciation, and (5) the somewhat larger western part by the Wisconsin glaciation, both of which are also found in Illi- nois. In both states the lowan glaciation is characterized by its rolling topography and perfect surface drainage, and the Wiscon- sin by its level prairies which require much artificial drainage by tile and open ditches. Eight analyses of lowan soils, reported to the author by Doctor J. B. Weems while professor of agricultural chemistry in the Iowa State College, showed 900 pounds of acid-soluble phosphorus in 2 million of soil, as a general average. The several soil types repre- sented varied considerably, however, as would be expected from comparison with similar Illinois soils, the highest amount reported being 1600 pounds of acid-soluble phosphorus per acre in a 6|-inch stratum, corresponding to 1880 pounds of total phosphorus, if 85 per cent were acid-soluble. (This method of estimating total phosphorus from the acid-soluble phosphorus is never safe for application to individual soil samples, but it is approximately cor- rect for large averages of most common soils of central United States.) The acid-soluble potassium (which varies from less than one sixth of the total in old soils to more than one third of the total in more recent, less weathered soils) amounted to 4670 pounds as an average of the eight soils (the highest being 7800 pounds) in 2 mil- lion of surface soil, corresponding probably to 30,000 to 40,000 pounds of total potassium. The analysis of loess from Dubuque, SOIL COMPOSITION Iowa, shows 35,600 pounds of total potassium in 2 million (see Table u). • Since the above was written the author has secured, through the kindness of Professor Stevenson, the unpublished data shown in Table 18.1 which present the average results of from one to six analyses of the most important soil types in the great soil areas of the state. Professor Stevenson writes: " The samples are believed to represent the most widely distributed type of the respective areas. We did not determine total potassium." (Two types, the uncovered glacial till, and the shallow loess, are included for the Kansan area.) TABLE 18.1. PLANT FOOD IN SURFACE SOILS OF IOWA Pounds per Acre in 2 Million (about 6f -inch Stratum) FORMATION OR SOIL AREA NUMBER OF ANALYSES TOTAL NITROGEN POUNDS TOTAL PHOSPHORUS POUNDS Kansan glaciation (exposed till) . . . lowan glaciation 31 e 2380 •2OAO 860 1160 Wisconsin glaciation I 7C2O 1460 Loess on Kansan glaciation .... Mississippi loess 6 -2 3400 2IOO 1020 IO2O Missouri loess IO AA2O I42O 1 Only two analyses for phosphorus in the Kansan till The Kansan drift is the oldest and the poorest in both nitrogen and phosphorus, while the Wisconsin is the newest and the richest, with the lowan intermediate in both respects. The Mississippi loess and the shallow loess on the less rolling parts of the Kansan glaciation are similar in composition and probably of similar origin (of the lowan age) , but the Mississippi loess is much deeper and of a more rolling topography, which insures a much better subsoil, physically, and may also account for the somewhat lower nitrogen content of the surface, through loss of organic matter by washing. The higher phosphorus content of the Missouri loess suggests that it owes its origin in part to deposits from the semi-arid plains of the northwest, the richer mineral soil having encouraged the more recent accumulation of nitrogen. 92 SCIENCE AND SOIL SOILS OF THE SOUTHERN STATES The average composition of twelve samples of soil from seven different counties in Kentucky outside of the Blue Grass Region shows 550 pounds of acid-soluble phosphorus in 2 million pounds of surface soil; and within the famous Blue Grass Region the supply of acid-soluble phosphorus amounts to 5200 pounds (nearly ten times as much) in 2 million of soil, as the average of 30 soil analyses, collected from five counties, these residual soils having been formed in part at least from the weathering of phosphatic limestone. The state of Tennessee may be divided geologically into five principal sections: (1) The great loess-covered undulating upland area of west Tennessee, lying chiefly between the Mississippi and Tennessee rivers. (2) The Central Basin, including most of ten counties (David- son, Trousdale, Jackson, Smith, Wilson, Williamson, Rutherford, Bedford, Marshall, andMaury) and parts of several adjoining coun- ties. The Central Basin includes much of the great phosphate beds of Tennessee. It resembles in some respects the Blue Grass region of Kentucky, and by some the Central Basin of Tennessee is claimed to be the original home of Kentucky blue grass. (3) The Highland Rim, surrounding the Central Basin. (4) The Cumberland Plateau, farther east. (5) The East Tennessee Valley, lying between the Cumberland Plateau and the Unaka Mountains on the eastern border of the state. Table 19 shows the plant food in representative soils of each of these great sections, the averages of several soil analyses being reported for the more important areas (Mooers, Tennessee Bulletin 78). The average composition of the yellow silt loam soil on the loess- covered Ozark Hills of southern Illinois shows 1890 pounds of nitrogen, 950 pounds of phosphorus, and 31,450 pounds of potas- sium per acre in 2 million pounds of surface soil, which is practi- cally the same as for the west Tennessee soil. It may be kept in mind, too, that the north line of west Tennessee is only 35 miles SOIL COMPOSITION TABLE 19. COMPOSITION OF SURFACE SOILS OF TENNESSEE Average Pounds per Acre in 2 Million of Soil (about 6|-inch Stratum) SECTION OR AREA ORIGIN OF SOIL NITROGEN (Total) PHOSPHORUS (Total) POTASSIUM (Total) West Tennessee . . Loess deposit . . . 1890 890 31020 Highland Rim . . . Limestone .... 2100 660 24600 Central Basin . . . Limestone and phos- phate 2350 2030 18160 Cumberland Plateau . Sandstone .... I7OO 380 7840 East Tennessee Valley Limestone and dolomite 2080 980 12130 Bottom land Alluvial 262O 1840 ?4l6O from the southern point of Illinois, and that much of the upland soil of northwest Mississippi is essentially of the same character. It will be noted that the average soil of the Central Basin is comparatively rich in phosphorus, while the soils of the Highland Rim, and more especially the sandstone soils of the Cumberland Plateau, are extremely deficient in phosphorus, and the latter is also very poor in potassium, as might be expected from its origin. The number of soil analyses entering the averages in Table 18 is not sufficient for final data, but in the main they are supported by larger numbers of analyses for acid-soluble plant food. Thus, the averages of 25 analyses of soils from eight counties in the Central Basin show 2020 pounds of acid-soluble phosphorus in 2 million of soil, while 700 pounds of phosphorus is the corresponding aver- age for 16 analyses of loessial soil from eleven counties in west Tennessee. The acid-soluble phosphorus in the samples whose total phosphorus content was determined (and thus afforded for use in Table 18) was 1710 pounds for the Central Basin and 750 pounds for the west Tennessee soil, 84 per cent of the total having been dissolved by strong hydrochloric acid, in either case. On this basis, the general average of all samples would show 830 pounds of total phosphorus for west Tennessee and 2400 pounds for the Central Basin, in 2 million pounds of surface soil. Hilgard reports the average of 97 analyses of Mississippi soils showing 790 pounds of acid-soluble phosphorus in 2 million pounds of surface soil, but the more abundant upland soils average about 94 SCIENCE AND SOIL 700 pounds, while 4100 pounds of phosphorus per acre (in a 6f -inch stratum) have been found in an upland soil of limestone origin. The " black prairie " limestone soils found in limited area in northeastern Mississippi and in northwestern Alabama are as a rule well supplied with phosphorus. This formation is apparently an extension of that found so commonly in the broad Central Basin of Tennessee, which also extends into Kentucky, where it again expands into the great Blue Grass Region. An analysis made by Doctor H. C. White of a sample of soil representing the University farm of Georgia, collected in March, 1884, from land that " had been cleared in December (1883) of a second growth of oak and hickory," gave the following amounts per acre based upon the surface foot, which was assumed to weigh 3,528,000 pounds. TABLE 19.1. COMPOSITION OF GEORGIA SOIL (UNIVERSITY FARM) PLANT-FOOD ELEMENTS POUNDS PER ACRE FOOT POUNDS IN 2 MILLION Nitrogen T.6%2 2OOO Phosphorus "^O •3.OO Potassium 2IOOO I2OOO Magnesium <7OO 3200 Calcium . . 74OO 4100 Iron . . .... 2Q7OO 16100 Sulfur IIOO 600 The original statement reports " sand and clay," and it must be assumed that the mineral elements as given above represent the amounts soluble in strong acid. The Texas Experiment Station has analyzed a considerable number of soil samples collected to represent six general groups or " series," as mapped and named by the United States Bureau of Soils. The following descriptions are taken from Texas Bulletin 99 (1907) = "Norfolk soils. These are light-colored upland sandy soils, with a yellow clay or sandy clay subsoil, usually with good drainage. " Of the areas under study, the Norfolk soils are found in Anderson, Houston, and Bexar counties. They are widely distributed in the eastern part of the state. SOIL COMPOSITION 95 "Orangeburg soils. The Orangeburg soils are gray to brown upland soils, with a red or yellowish clay sandy subsoil. The red color of the sub- soil distinguishes the Orangeburg soils from the Norfolk soils. The red soils appear to be more productive, and are generally stronger than the correspond- ing soils of the Norfolk series. The Orangeburg soils are widely distributed, especially in East Texas. "Lufkin soils. The Lufkin soils are gray, with heavy, very impervious plastic gray and mottled subsoils. These soils are generally lower in agricul- tural value than the Norfolk and Orangeburg soils, perhaps on account of the nature of the subsoils. These soils are found in Houston, Lamar, and Travis counties, of the areas studied. " Susquehanna series. These are gray and brown surface soils with heavy plastic mottled subsoils. They differ from the Lufkin series in the color of sub- soil. They are generally of low productiveness. "Houston series. These are black calcareous prairie soils, very productive and durable. They are among the best soils of the state. Some of them have been in cultivation forty or fifty years without fertilizer, and though some of them have decreased somewhat in fertility, they are still productive. They are found, in areas surveyed, in Lamar, Hays, Travis, and Bexar counties. They are of general occurrence in the east-central portion of the state. "These soils appear to owe their productiveness to their content of lime and organic matter, and nitrogen. Some of these soils will become deficient in phosphoric acid. "Yazoo soils. These soils are bottom land, generally subject to overflow and very productive. The soils are mapped in only two areas, Anderson and Travis counties." The following tabular statement gives the average amounts of total nitrogen, acid-soluble phosphorus, and acid-soluble potassium in 2 million pounds of the surface soil of these different soil types, based upon the analyses reported by Doctor Fraps. TABLE 19.2. AVERAGE COMPOSITION OF SOME TEXAS SOILS Pounds in 2 Million of Soil CONVENTIONAL NAME Son. TYPE TOTAL NITRO- GEN ACID- SOLUBLE PHOS- PHORUS ACID- SOLUBLE POTAS- SIUM Norfolk soils . . Orangeburg soils . Lufkin soils Susquehanna series Houston series Yazoo soils . . . Light sandy upland .... Gray-brown sandy upland . . Gray silt loam on tight clay . . Gray-brown silt loam on tight clay Black prairie IOOO I2OO IOOO 1400 2800 1600 1 80 440 180 260 53° 960 2OOO 62OO 1800 3700 5600 6700 Bottom land 96 SCIENCE AND SOIL With the exception of the bottom land, these soils are extremely poor in phosphorus; and, aside from the black prairie, they are also very poor in nitrogen. Hilgard reports 940 pounds of acid-soluble phosphorus in 2 million pounds of Louisiana soil, as an average of 35 analyses. The Geological Survey of Louisiana has furnished the data for the following statement, showing the total nitrogen, acid-soluble phosphorus, and acid-soluble potassium in 8 surface soils and 2 subsoils in Madison Parish, opposite Vicksburg. These are all alluvial soils deposited from the overflow of the Mississippi River. TABLE 19.3. COMPOSITION OF LOUISIANA SOILS IN MADISON PARISH Pounds in 2 Million of Soil DESCRIPTION OF SOIL TOTAL NITROGEN ACID-SOLUBLE Phosphorus Potassium Black soil on buckshot clay; virgin swamp, but good land when cleared Subsoil of same type 3986 1286 1450 1044 6562 4976 Dark gray soil on buckshot clay; once swamp; cultivated o years 3334 950 1174 928 5466 3570 Subsoil of same Dark gray clay ; virgin soil ; wooded swamp; overflow land Black waxy soil on dark gray clay ; very fer- tile when thoroughly drained; i£ to 2 bales of cotton per acre 2286 2456 2662 1562 1244 1306 2000 2024 1760 1234 1474 I2O6 8412 5134 4748 5084 4IOO 3806 Black soil on buckshot clay Brown silty loam ; old land . ... Brown sandy loam ; cultivated 50 years . . Light brown sandy loam; once productive, but now worn out for corn and cotton In phosphorus content these soils resemble the late bottom lands of Illinois, but the nitrogen content is, as a rule, much lower in the Louisiana soils, and extremely low in the old, worn soils. It may be mentioned here that the Mississippi Experiment Station has conducted some field experiments on delta lands, concerning which the following is reported in Mississippi Bulletin 119: SOIL COMPOSITION 97 "The land on which the tests were made had been cropped in cotton for many years. A part of it is loam soil and is well drained. A test was also made on stiff buckshot land." "The average increase from 300 pounds of high grade cotton seed meal per acre for the three years (1906-8) has been 106 pounds of lint cotton. We have not been able to increase the size of the crop nor its earliness by the use of either phosphorus or potash." "The increase from the application of 300 pounds of cotton seed meal to the stiff buckshot land was 36 pounds of lint cotton per acre. This is not sufficient to make the use of the meal profitable on this character of land." As an average of 38 analyses of Arkansas soils, Hilgard gives 1400 pounds of acid-soluble phosphorus in 2 million of soil, sug- gesting bottom-land soils or some connection with the phosphate deposits of that state. SOILS ,OF THE NORTHERN STATES In the northern tier of states, where the soils are of more recent origin and where the climate of winter offers less exposure to weath- ering and leaching, the normal soils are, as a rule, richer in mineral plant food, as is indicated, for example, by comparing the soils of the late Wisconsin glaciation in northern 'Illinois with similar types in the older middle Illinoisan glaciation. Of course, the up- land timber soils are not comparable in nitrogen content with the black prairie soils. The late Doctor Robert C. Kedsie, one of the few great scientists who, with Doctor E. W. Hilgard and Doctor S. W. Johnson, helped to lay firm foundations for the American Agricultural Experiment Stations, reported analyses of 28 Michigan soils grouped in accord- ance with a general survey or classification of the soils of that state (Michigan Bulletin 99) : 1. The four southern tiers of counties are classed as theMichigan "wheat belt." 2. The area along the eastern shore of Lake Michigan, including especially the light porous soils upon which peaches of the finest quality are extensively produced, is termed the " peach belt." 3. Several counties in the Traverse Bay region, including much soil of the sandy-loam type, constitute the " potato district." 98 SCIENCE AND SOIL 4. The large tract of light, sandy lands in the north-central part of the Lower Peninsula is called the " Jack Pine Plains." 5. The peaty swamp soils, used especially for the growing of celery, peppermint, etc., are so designated. In Table 20 is given the average composition of the soils from each of these sections. The nitrogen reported is total, but the data for phosphorus and potassium represent only the amount soluble in acids, probably much stronger, however, than now commonly used. The amount closely approaches the total in case of phos- phorus, but probably represents less than one half of the total potassium actually present in the soil. (A larger proportion of the total potassium present in the newer, less- weathered soils is soluble in acid than of the total potassium in the older leached soils found in the states farther south.) TABLE 20. AVERAGE COMPOSITION OF SOME MICHIGAN SOILS Pounds per Acre in 2 Million of Soil (about 6§ Inches) ACID- ACID- AREA Son, SECTION SOLUBLE SOLUBLE PHOSPHORUS POTASSIUM (i) Southern Michigan Wheat belt . . 4600 3600 27360 (2) Lake shore region : . Peach belt . . 1860 2330 22600 (3) Traverse Bay region . Potato district . 1260 1520 16400 (4) North-central tract . . Jack Pine Plains 740 290 4600 (5) Swamp areas .... Deep peat ' . . 22OOO 2980 2660 1 The amounts given for deep peat represent the plant food in i million pounds of the material, the specific gravity of which is about one half that of ordinary soils. These striking and valuable results obtained in a very prelimi- nary general survey of Michigan soils clearly indicate the much greater possible value of an extended and detailed investigation of the soils of the state. The high phosphorus content of the Michigan soils, especially of the great area of the wheat belt (3600 pounds), is in marked contrast, not only with the extremely low phosphorus content of the Jack Pine Plains (290 pounds, — less than one tenth as much), but also with the small supply of phosphorus in the common upland soils of southern Illinois, central Missouri, and the western parts of Kentucky, Tennessee, and other southern states. SOIL COMPOSITION 99 In commenting on his study of Michigan soils, Doctor Kedsie said: "Chemical analysis of the soil is of value in determining whether the soil is capable of fertility or the contrary; also in determining the measure of its possible fertility.1 There are certain ash elements which are absolutely nec- essary for plant growth, in the absence of any one of which vegetable growth is impossible; if the supply is relatively limited, plant growth will be limited correspondingly. Hence, chemical analysis of a soil is of importance in deter- mining possibility of fertility and of the relative fertility which may be secured under favorable conditions. . . . Chemical analysis will not always dis- tinguish between a fruitful and an unfruitful soil. A soil may be unproductive for physical reasons, though it may still contain all the chemical elements of fertility." The Michigan wheat-belt soils include several different soil types, but among the nine soil samples analyzed from that area, the poor- est contained 2600 pounds of acid-soluble phosphorus, or 500 pounds more than the average of the best Illinois soil. More than four times as much phosphorus is contained in the average Michigan wheat-belt soil (when these samples were taken) as is now con- tained in the common soils of southern Illinois. A preliminary general survey of Wisconsin soils (Whitson, Wisconsin Agricultural Experiment Station, Annual Report for 1905, pages 262-270) outlines seven different great soil areas in that state: (i) The unglaciated area of the southwest quarter (extending into northwestern Illinois) with three subdivisions in which residual sands, sandy loams, and clay loams, respectively, predominate; (2) the early, and (3) the late glaciations (each in two divisions based upon the underlying rocks), occupying largely the remaining three fourths of the state, and covered with glacial till, with little or no loess deposit; (4) separated sand areas of glacial origin, as in the south-central, extreme northern, and northwest parts of the state; (5) a loessial area covering a strip of upland along the Mississippi ; (6) " red clay " areas of lacustrine origin between Green Bay and Lake Winnebago and on the Lake Superior shore ; and (7) the scattered swamps of muck and peat. But few analyses of Wisconsin soils have been reported. An 1 Italics by C. G. H. ioo SCIENCE AND SOIL ultimate analysis of virgin soil from the Wisconsin Experiment Station Farm at Madison, in the late glaciation, shows 3600 pounds of nitrogen, 1500 pounds of phosphorus, and 36,300 pounds of potassium, in 2 million pounds of surface soil. Where the same soil had been heavily cropped in pot cultures (19 crops having been removed) the nitrogen was reduced to 2200 pounds and the phos- phorus to 1140 pounds, with no determinable change in total potassium content. The analysis of another soil from the late glaciation from northern Outagamie County showed 1400 pounds of nitrogen and 2380 pounds of acid-soluble phosphorus. This glaciation Professor Whitson regards as the best soil area in the state. Residual sand from Jackson County contained 1000 pounds of nitrogen, 870 of phosphorus, and 5100 of total potassium in 2 million of surface soil, and the glacial sand from Vilas County con- tained 1000 pounds of nitrogen, 1580 pounds of phosphorus, and 30,000 pounds of potassium, indicating that the residual sand is more largely quartz, while the glacial sand consists chiefly of sili- cate minerals. (Compare with Tennessee soils.) Red clay from Ashland County contained 1400 pounds of acid- soluble phosphorus in 2 million of soil; and peaty swamp soil from Sauk County contained 32,000 of nitrogen, 1230 pounds of acid- soluble phosphorus, and only 910 pounds of acid-soluble potassium, in i million pounds of the surface soil. The acid-soluble plant food has been determined in many Minne- sota soils (Snyder, Minnesota Bulletin 41), and a few analyses are reported showing the total plant food in representative soils. The average prairie soil of the Red River Valley in northwestern Minnesota contains 8200 pounds of nitrogen, 3340 pounds of phos- phorus, and 45,100 pounds of potassium in the surface 2 million pounds; and the average prairie soil in west-central Minnesota contains 5300 pounds of nitrogen, 1760 pounds of phosphorus, and 63,300 pounds of potassium. A general average for the soils of the east-central part of the state is 5600 pounds of nitrogen, 2460 pounds of phosphorus, and 29,000 pounds of total potassium; while the average southeastern Minnesota soils contain 4400 pounds of nitrogen, 1910 pounds of phosphorus, and 30,200 pounds of potassium, in 2 million pounds of surface. SOIL COMPOSITION 101 From the general glacial map of the United States it will be seen that southeastern Minnesota lies in the older lowan glaciation, while most of the remainder of the state is covered by the late Wisconsin glaciation. This may account for the marked difference in potassium content between the soils of eastern and western Minnesota. (See also Tables 15, 16, and 17 for a comparison of these areas in Illinois.) It will be noted that the Red River basin lies within the boundaries of the old glacial lake, " Agassiz." (See also Canadian soils.) The analysis of a sample composed of equal parts of two hundred representative soils from various parts of Minnesota showed 2360 pounds of acid-soluble phosphorus in 2 million of soil. The soils of the arid plains are, as a rule, rich in mineral plant food and poor in nitrogen, doubtless due to the fact that with but little rainfall there has been practically no loss of minerals by leaching, and but small accumulation of vegetable matter, in which the supply of nitrogen is contained. Headden (Colorado Bulletin 65) reports ultimate analyses of four Colorado soils, showing as an aver- age 2900 pounds of phosphorus and 39,500 pounds of potassium; but the average of six soils shows only 2000 pounds of nitrogen, in 2 million of surface soil. Widtsoe (Utah Bulletin 52) shows 37 analyses of Utah soils averaging 1850 pounds of acid-soluble phos- phorus and 2450 pounds of total nitrogen, in 2 million of soil, in the most fertile valleys. On the arid plains the supply of nitrogen is usually very much less. Doctor Widtsoe states that he, " in common with those who have traversed the wastes of western America, has traveled for days without seeing a trace of vegeta- tion, and such soils are almost devoid of organic matter and humus, and contain but small quantities of nitrogen." Hilgard gives 600 pounds of total nitrogen and 2000 pounds of acid-soluble phosphorus in 2 million of soil, as the average of 16 analyses of the arid soils of Colorado. As a rule, the soils of the arid region contain about 3 per cent of lime (CaCO3), or 30 tons of calcium carbonate in 2 million pounds of soil; and the Utah Station reports 18 soil analyses from one county, containing as an average more than 20 per cent 102 SCIENCE AND SOIL of calcium carbonate corresponding to 200 tons per acre to a depth of 6| inches. For 2 million pounds of surface soil, Hilgard's " Soils" gives the following amounts of acid-soluble phosphorus as the average of many analyses: Nevada, 2800; Wyoming, 1570; Montana, 1920; Idaho, 1400. Mr. E. E. Hoskins, while a student in the University of Illinois, analyzed a sample said to represent good " bench " land near Boise City, Idaho, finding 2350 pounds of nitrogen, 1320 of total phosphorus, and 41,400 pounds of potassium, in 2 million of soil. In the Pacific coast states, Hilgard reports (as acid-soluble) 1830 pounds of phosphorus and 10,790 of potassium for Washing- ton, 960 of phosphorus and 8960 of potassium for Oregon, in 2 million pounds of surface soil; also 2800 pounds of total nitrogen, 875 of acid-soluble phosphorus, and 10,100 pounds of acid-soluble potassium, in 2 million pounds of the surface soils of California, as an average of 262 analyses. THE SOILS OF CANADA It will be kept in mind, of course, that Canada comprises an immense territory (3,300,000 square miles, compared with 3,600,000 square miles in the United States, including nearly 600,000 square miles in Alaska) and includes vast areas that are uninhabited, and in part uninhabitable, although the more favored regions are sufficiently extensive ultimately to support a mighty nation. The Canadian Agricultural Experiment Station (known officially as the Dominion Experimental Farms, with headquarters at Ottawa) has made analyses of Canadian soils collected from Van- couver Island on the west to Nova Scotia on the east. Perhaps the most complete investigation has been made of the valley lands and Piedmont soils of British Columbia. In general, these soil investigations have been conducted with reference to uncultivated lands or lands put under cultivation in recent years. They do not represent the comparatively small areas of Canadian soil that have been long cultivated, some of which has already been much de- pleted or much fertilized. The samples have been collected with EUGENE WOLDEMAR HILGARD, A MAN OK SCIENCE Born at Zweibriicken, Bavaria, January 5, 1833 ; educated at Belleville, Illinois, and Heidel- berg, Germany ; state geologist and professor of chemistry of the University of Mississippi, 1855-1873 ; professor of geology and natural history of the University of Michigan, 1873- 1875; professor of agriculture and director and chemist of the agricultural experiment station, University of California, 1875-1906 ; chemist since 1906 ; author of " Soils " (1906) SOIL COMPOSITION 103 sufficient purpose, method, and discrimination to give much im- portance to the results, which are summarized as follows: TABLE 20.1. COMPOSITION OF CANADIAN SURFACE SOILS Pounds per Acre in 2 Million of Soil (about 6§ Inches) ACID-SOLUBLE NUMBER OF TOTAL NITROGEN Phosphorus Potassium Calcium 21 British Columbia .... 5240 2360 6970 16700 6 Northwest Territory . . . 9180 1520 5670 4I30 I Manitoba 2OIOO 2OO 17100 27000 6 Quebec 4<2O 17^0 73OO 74OO 6 Ontario (Muscoka only) 27OO 1250 3650 6300 5 Maritime Provinces . . . 26OO 960 7300 1570 In referring to the averages represented in this tabular state- ment, Professor Frank T. Shutt (Chief Chemist for the Dominion Experimental Farms since 1887) says: "They are not provincial averages; they are rather averages from large untilled areas in the several provinces, and may therefore serve to indicate the general character of much of the yet unoccupied lands of Canada." (Do- minion Experimental Farms Reports, 1897, page 169.) A study of the details shows much variation, but in the main these are counterbalanced so that the averages have much meaning. Professor Shutt states that the one sample from Manitoba " repre- sents the unfertilized and uncropped prairie soil of the Red River Valley, Manitoba," and adds: "It was taken from section 31, township 4, range i west. The uniformity in character of the soil over a very large area in Manitoba makes the data here presented of more than ordinary importance. " "We may safely conclude that there is here ample scientific proof of the well- nigh inexhaustible stores of plant food, and that this prairie land, as regards the elements of fertility, ranks with the richest of known soils. " Doctor George M. Dawson, Director of the Geological Survey of Canada, is quoted as follows: "Of the alluvial prairie of the Red River much has already been said, and the uniform fertility of its soil cannot be exaggerated. . . . The area of this lowest prairie has been approximately stated as 6900 square miles. " 104 SCIENCE AND SOIL Of course, these averages from the various Canadian provinces must be considered as tentative and very preliminary, but they must also be accepted as giving a reasonably correct general view of the invoice of soil fertility in the most extensive types of soil in those sections. Some other analyses made in part in connection with special investigations are discussed in another place. REVIEW OF SOIL COMPOSITION A general summary of the mass of evidence contained in the preceding pages concerning the composition of soils clearly sets forth a number of definite and assured facts bearing significant rela- tions to systems of permanent agriculture. One of these most clearly established facts is that potassium as an element of plant food belongs in the class with calcium and magnesium rather than with phosphorus and nitrogen. In all normal soils the supply of potassium is enormous. Thus, as an average of the Maryland soils reported in Table 12, representing ten different geological forma- tions, more or less abundant in most of the Atlantic states, we find 37,860 pounds of potassium and only 14,080 pounds of magnesium, 7840 pounds of calcium, and uoo pounds of phosphorus, in 2 million pounds of soil. Measured by the total requirements of approximately maximum crops in a rotation of wheat, corn, oats, and clover (Table 13), the potassium is sufficient for 473 years, the magnesium for 828 years, and the calcium for 187 years; while the total phosphorus is suffi- cient for the same crops for only 57 years. If we consider the plant food removed in the grain alone, assuming that the coarse products will remain on the farm, and also disregard the one abnormal mag- nesium soil (from serpentine), the relative plant- food supply is represented by 105 years for phosphorus, 3060 years for potassium, 2828 years for magnesium, and 6970 years for calcium. The complete analyses of the loess deposits at Dubuque, Iowa, and Kansas City, Missouri, which contain only moderate amounts of carbonates, show, as an average, 33,100 pounds of potassium, 13,400 pounds of magnesium, 23,500 pounds of calcium, and 1400 pounds of phosphorus, in 2 million of loess. In the following summary are reported the total phosphorus, total potassium, total magnesium, and total calcium in the surface soil of the five most extensive soil types of Illinois. TABLE 20.2. CERTAIN PLANT-FOOD ELEMENTS IN ILLINOIS SURFACE SOILS Pounds per Acre in 2 Million of Soil (about 6f Inches) Son, TYPE No. SOIL AREA OR GLACIATION Son. TYPE PHOS- PHORUS (Total) POTAS- SIUM (Total) MAGNE- SIUM (Total) CALCIUM (Total) 135 Unglaciated Yellow silt loam 95° 3*450 7220 5340 33° Lower Illinoisan Gray silt loam 840 24940 6740 14660 426 Middle Illinoisan Brown silt loam 1170 32240 8800 15940 526 Upper Illinoisan Brown silt loam I2OO 32940 8340 19220 1126 Early Wisconsin Brown silt loam IIQO 36250 9400 1 1 120 General average IO7O 31^60 8lOO 1^260 Number of years' supply : (fl) For total rrons . .... c6 3QS 476 *i6 (b) P 'or grain only 102 2475 2025 11780 It should be kept in mind that these data are based upon the average composition of many soil samples from every type, and that these are widely representative of the most extensive and important soil types in the Central states. They signify determined facts. As a general average of these soils, potassium is better supplied than magnesium for grain farming (in which all coarse products are returned to the land), and potassium is better sup- plied than calcium for a system in which all of the produce is re- moved. In normal soil types the relations existing among these four elements in the subsoil are not essentially different from those in the surface, except in subsoils that are rich in calcium and mag- nesium carbonates. Even in the almost unweathered glacial sub- soils of the principal types of the Late Wisconsin glaciation (brown silt loam and yellow-gray silt loam ; see Table 17), 2500 pounds of phosphorus and 158,000 pounds of potassium are the relative and total amounts of those two elements per acre in a 20- inch stratum. Measured by the average losses from the farm by selling maximum crops of corn and wheat, our two principal grains, the phosphorus in 6 million pounds of this subsoil is sufficient for 173 years, or one half as long as from 1565 to 1911, and the potassium is sufficient io6 SCIENCE AND SOIL for 9900 years, or as long as from 8000 years before Christ to 1900 years after Christ. In other words, on the absolute mathematical basis there is no more reason for applying potassium to normal soils as plant food than there is for applying magnesium and cal- cium for the sake of adding them also as plant food. The use of potassium on soils actually deficient in that element (as peaty swamp soils) , and on some other soils as a soil stimulant, is discussed in the following pages. In brief, it may be said, of the plant-food elements supplied by the soil, that nitrogen and phosphorus are in one class; that potassium, magnesium, and calcium are far removed in a second class; and that iron is distinctly in a third class; while the nitrogen of the air, so far as concerns the supply for permanent agriculture, should be classed with carbon, hydrogen, and oxygen. The place of sulfur is not so easily determined. Measured by the soil's supply, sulfur would be classed with phosphorus and nitrogen; or measured by the crop demands, it would be classed with iron; but, if both supply and demand are considered, it must be classed with potassium, magnesium, and calcium. It is, however, known with certainty that more or less sulfur is carried into the air with the products of combustion and of decay, and some sulfates are also carried into the atmosphere in the dust from evaporated ocean spray. Long- continued investigations at Rothamsted and elsewhere have shown that as an average rainfall brings to the soil, chiefly in the form of sulfates, about 7 pounds of sulfur per acre per annum, or i pound more than would be required for a zoo-bushel crop of corn. CHAPTER VII AVAILABLE PLANT FOOD "AVAILABLE plant food" is an expression much used in connec- tion with commercial fertilizers, and the argument is commonly made that because the soil does net contain available plant food, we should therefore apply available plant food in commercial fer- tilizers. Instead of following this advice, however, the farmer should, as a very general rule, adopt a system of farming that will make available the plant food in the soil so far as practicable, and, if any element is actually deficient in the soil, apply that element in cheap form and in positively larger quantities than will be re- moved in large crops; and then make it, too, available by his method of farming. There are three methods of determining with some degree of satisfaction which elements, if any, are deficient in the soil: First, we may compute from the composition of the soil and the requirements of crops the probable durability of a soil with reference to any element of plant food. Thus, we may determine that the unglaciated yellow silt loam surface soil of Illinois, Kentucky, Tennessee, and other adjoining states contains sufficient nitrogen for less than 20 large corn crops if only the grain were removed; while the potassium in the late Wisconsin brown silt loam is suffi- cient for more than 2300 such crops. Second, we can assume for a rough estimation that the equiva- lent of 2 per cent of the nitrogen, i per cent of the phosphorus, and I of i per cent of the total potassium contained in the surface soil can be made available during one season by practical methods of farming. Of course, the percentage that can be made available will vary very much with different seasons, with different soils, and for different crops; and yet with normal soils and seasons and for ordi- nary crops the above percentages represent roughly about the 107 io8 SCIENCE AND SOIL proportion that is liberated from our common soils of the element that limits the yield of the crop. In Table 21 are given the amounts of annually available plant food in Illinois soils as roughly estimated by this method of com- putation. Of course, these amounts would be smaller and smaller year by year in proportion as the total supply is decreased, and accordingly complete exhaustion is not only impracticable and unprofitable because of the continual reduction in crop yields, but it is mathe- matically impossible, just as it would be impossible to exhaust a bank account if only one per cent of the remaining deposit could be withdrawn each week. A peaty swamp soil containing 2930 pounds of total potassium per acre in the first 6f inches would liberate during the season, according to this estimate, about 7 pounds of potassium, which would be equivalent to a crop of 10 bushels of corn, which represents roughly about the average yield from such land when not treated with potassium, as is shown in the following pages. The common brown silt loam prairie soil, when well farmed, will average about 50 bushels of corn per acre, which would require n^ pounds of phosphorus and 74 pounds of nitrogen, while 12 and 96 pounds represent i per cent of the phosphorus and 2 per cent of the nitro- gen, respectively, in the surface soil, where phosphorus is the first limiting element and nitrogen the second. These illustrations are given not to prove that this rough esti- mate is applicable, but rather to show the basis which suggests such a computation. It has some value, chiefly, perhaps, in that it helps one to understand why it is that with phosphorus enough in the surface soil for 50 crops, we obtain only half a crop as an average. . On this basis we should try to keep sufficient phosphorus in the surface soil for 100 large crops, of which one per cent would then be sufficient for one large crop. This would require about 2300 pounds of phosphorus per acre, or but little more than is actually contained in the most productive Illinois corn-belt soil, as the early Wisconsin black clay loam in such counties as McLean, Champaign, Edgar, et al. While there are several agencies that tend to convert insoluble AVAILABLE PLANT FOOD 109 plant food into available forms, such as the products of decaying organic matter, including carbonic, nitric, and various organic acids, the different forms of lime, and most soluble salts, each of which is more fully discussed in its proper place, it is also known that the plant roots themselves influence the availability of plant food, probably by means of the carbonic acid or other substances which they excrete. The juices of plants are commonly distinctly acid, and the roots have some power to exude moisture, which certainly contains carbonic acid and very possibly contains no other solvents, although this question is not fully settled. Where growing plant roots lie in contact with the polished surface of marble (calcium carbonate) and some other materials (as prepared slabs of calcium phosphate), distinct etching occurs, as was early shown by Sachs and Czapek. Kossowitsch conducted an experiment in which he grew plants (peas, flax, and mustard) in two pots of sand which differed only by the addition of fine-ground rock phosphate to one. The plants were watered with the slow and constant application of like amounts of a dilute solution containing all essential plant-food ele- ments, except phosphorus. For the pot which contained no phos- phate this solution was, in this continuous process of watering, passed through a third pot of sand to which the fine-ground rock phosphate had also been added. Kossowitsch found that the plants made a much better growth in the pot where the roots were in direct contact with the phos- phate, thus showing that they exert a solvent action in addition to any that may be exerted by the nutrient solution. If, for example, the plant roots come in contact with, or exert an influence upon, the equivalent of only one per cent of the sur- face of the soil particles within the root range, this would offer an explanation of the relationship (which is very irregular in different soils and seasons) that exists between the total amount of any plant-food element in a given soil stratum and the amount secured by a given crop during the growing season. It is well known that soluble phosphates and soluble potassium salts, when applied to normal soils, are almost immediately con- verted into insoluble forms; and the higher availability of such soluble fertilizers is now believed to be largely due to the fact that no SCIENCE AND SOIL TABLE 21. ANNUALLY AVAILABLE FERTILITY IN ILLINOIS SOILS, ROUGHLY ESTIMATED Pounds per Acre Son. AVAIL- AVAIL- AVAIL- TYPE No SOIL AREA OR GLACIATION SOIL TYPE ABLE NITRO- ABLE PHOS- ABLE POTAS- GEN PHORUS SIUM PRAIRIE LANDS, UNDULATING 33° Lower Illinoisan . . Gray silt loam on tight clay 58 8 62 426 526 626 726 Middle Illinoisan . . Upper Illinoisan . . Pre-Iowan .... lowan . ... Brown silt loam . . . Brown silt loam . . . Brown silt loam . . . Brown silt loam . 87 97 86 08 12 12 12 12 81 82 88 82 1126 1026 Early Wisconsin . . Late Wisconsin . . Brown silt loam . . . Brown silt loam . . . 101 135 12 14 9i "3 PRAIRIE LANDS, FLAT 329 Lower Illinoisan . . Drab silt loam . . . 56 7 66 420 Middle Illinoisan . . Black clay loam . 1 08 14 80 520 Upper Illinoisan . . Black clay loam . . . 135 i? 74 1 1 20 Early Wisconsin . . Black clay loam . . . 157 20 88 I22O Late Wisconsin . . Black clay loam . . . 178 19 93 TIMBER UPLANDS, ROLLING OR HILLY 135 Unglaciated .... Yellow silt loam . . . 38 10 79 335 Lower Illinoisan . . Yellow silt loam . . . 43 10 80 345 Middle Illinoisan . . Yellow silt loam . . . 37 8 82 535 Upper Illinoisan . . Yellow silt loam . . . 40 8 87 635 Pre-Iowan .... Yellow silt loam . . . 48 9 93 77C lowan Yellow silt loam . 78 Q 80 "35 Early Wisconsin . . Yellow silt loam . 38 9 82 864 Deep loess .... Yellow fine sandy loam 43 10 89 TIMBER UPLANDS, UNDULATING 1034 760 Late Wisconsin . . . lowan .... Yellow-gray silt loam . Brown sandy loam 58 61 8 119 67 TIMBER UPLANDS, FLAT 332 Lower Illinoisan . . Light gray silt loam on tight clay .... 38 8 68 AVAILABLE PLANT FOOD in TABLE 21. ANNUALLY AVAILABLE FERTILITY IN ILLINOIS SOILS, ROUGHLY ESTIMATED — Continued SOIL AVAIL- AVAIL- AVAIL- TYPE No. SOIL AREA OR GLACIATION SOIL TYPE ABLE NITRO- ABLE PHOS- ABLE POTAS- GEN PHORUS SIUM SAND, SWAMP, AND BOTTOM LANDS ^S1 Old bottom lands . . Deep gray silt loams 72 14 Qi 145 1 Late bottom lands Brown loam .... 94 16 100 1481 Sand plains and dunes Sand soil 2Q 8 77 1401 Late swamp . . . Deep peat (?)' 20 7 1415 Late swamp . . . • Drab clay us *9 120 1 The nitrogen in peat is so very slowly available that not even a rough estimate can be made here. they first dissolve in the soil water and spread over the surface of the soil particles before becoming insoluble, and thus they offer a much more extensive surface for contact with plant roots than would plant-food particles applied in insoluble form, unless very finely ground. Third, we may apply different elements of plant food to the soil and note the effect, if any, in increasing the yield of crops, and thus sometimes discover what element is most deficient in the soil. One might suppose that this would be the best method, but such is not the case. This method frequently gives erroneous results which, if followed, may lead to land ruin, because the substance applied may produce little or no benefit on account of the special plant-food element it contains, but it may act as a powerful soil stimulant and thus liberate from the soil some other entirely differ- ent element in which the soil is already becoming deficient. Thus have many lands been practically ruined by the use of land-plaster and salt, by the improper use of lime, and even by the use of clover merely as a soil stimulant. Some good illustrations of this action of soluble salts are shown in the following pages. "In considering the general subject of culture experiments for determining fertilizer needs, emphasis must be laid on the fact that such experiments should never be accepted as the sole guide in determining future agricultural practice. 112 SCIENCE AND SOIL If the culture experiments and the ultimate chemical analysis of the soil agree in the deficiency of any plant-food element, then the information is conclusive and final; but if these two sources of information disagree, then the culture experiments should be considered as tentative and likely to give way with increasing knowledge and improved methods to the information based on chemical analysis, which is absolute. " 1 The plant food in the subsurface and subsoil is unquestionably of some value, but even the total supplies of nitrogen and phos- phorus that are held within the feeding range of ordinary plant roots are not unlimited when measured by crop requirements in permanent agriculture. However, the thing of first importance is to maintain a rich surface soil, for no subsoil is of much practical value if it lies beneath a worn-out surface. On the other hand, if the subsoil will act as a reservoir for moisture, then a rich top soil will produce large crops. Manures and fertilizers are applied to, and incorporated with, the plowed stratum only. On the Rothamsted fields where chalk exceeding 100 tons per acre was applied to the land a hundred years ago, practically no calcium carbonate is found below the plowed soil even after a centur^ of cultivation, although 50 tons of the chalk applied still remain in the surface soil ; and the land fertilized with nitrogen, phosphorus, and potassium, which has yielded more than 30 bushels of wheat per acre as an average of fifty years, contains, as an average, no more plant food in the strata below the surface 9 inches than is found in the same strata where the land has been unfertilized and has produced an average yield of only 13 bushels of wheat for the same fifty years. The supply of nitrogen in soils is contained only in the organic matter; and thus the amount of nitrogen in the subsoil of normal soil types is relatively small, as will be seen from a study of Tables 15, 16, and 17, six million pounds of subsoil containing, as a rule, less nitrogen than two million pounds of surface, except where the surface is much worn. The small amount of humus in the subsoil is also quite inactive, and the liberation of nitrogen from its decom- position is very slight. Furthermore, in all humid regions there is large loss of nitrogen in drainage waters; so that in practice the addition of nitrogen to the surface soil must be somewhat greater 1 "Cyclopedia of American Agriculture," Vol. i, page 475. AVAILABLE PLANT FOOD 113 than that removed in crops, if the productive power of the soil is maintained. The phosphorus of the soil exists in both organic and mineral forms. While the supply of mineral phosphorus is likely to be smaller in the surface and subsurface than in the subsoil, the organic phosphorus, like nitrogen, varies with the organic matter, which, as a rule, decreases rapidly below the plowed soil. As a consequence, there is usually less phosphorus in the subsurface than in the sur- face, unless the surface is very poor in organic matter; and also there is less phosphorus in the subsurface than in the subsoil, unless the subsurface is much richer in organic matter. A larger supply of phosphorus in the surface than in the subsurface is suffi- cient to prove that plants secure some phosphorus from below the surface soil, because the excess of phosphorus in the surface is contained in the decomposition products of plant residues from the centuries gone by. Where the subsoil is richer in phosphorus than the subsurface, it indicates either that phosphorus has been lost from the subsurface by leaching or that the plant roots have with- drawn phosphorus from the subsurface to a greater extent than from the subsoil. With the brown silt loams and black clay loams of the corn belt, the surface stratum rich in organic matter is always richer in phos- phorus than the subsurface; and, with one exception, the subsur- face, which is also quite well supplied with organic matter, is richer in phosphorus than the subsoil, but the other upland soils contain less organic matter in the surface and much less in the subsurface, and the subsoil, with a single exception, is richer in phosphorus than the subsurface, equal weights of soil always being considered. The potassium is contained almost solely in the mineral part of the soil, and the supply regularly increases with depth, the sub- surface being richer than the surface and the subsoil still richer. SOIL SURVEYS BY THE UNITED STATES BUREAU OF SOILS THE Bureau of Soils of the United States Department of Agricul- ture was organized in 1895, with Professor Milton Whitney as Chief. The energies of this Bureau have been devoted largely to making surveys of the soils in certain localities in most of the different states; and, second, to laboratory investigations in sup- port of a theory early announced by Whitney and Cameron, to the effect that practically all soils contain sufficient plant food for good crop yields, and that this supply will be indefinitely main- tained. The soil surveys are of general interest but of doubtful value to the local farmers and landowners, because they are reported with practically no information concerning valuable methods of soil improvement other than that based upon the actual practice al- ready in vogue, the Bureau having conducted no systematic field experiments and having reported practically no chemical analyses of the various soil types identified. The mechanical analyses which are almost invariably reported give little information of value fur- ther than to support the soil surveyor's classification of the soils into sandy soils, silty soils, clay soils, etc. Even the soil sur- vey, as conducted by the Bureau, is often too general or superficial in character to be of local use, differences in soils which are clearly recognized by the farmers being often ignored or overlooked. This will be better understood by examination of concrete illus- trations ; such, for example, as a comparison of the Bureau's map of Tazewell County, Illinois, published in the Report of Field Operations for 1902, with that of McLean County, which joins Tazewell on the east, and which accompanies the Report for 1903; or by a comparison of the Bureau's map of Clay County, Illinois 114 SURVEYS BY THE UNITED STATES BUREAU 115 (1902), showing all of the upland, comprising 85 per cent of the county, as one soil type (Marion silt loam), with the detail soil map published by the University of Illinois Agricultural Experi- ment Station (1909), showing eleven different types of upland soil, most of which are commonly recognized by the local farmers, the most extensive type (gray silt loam on tight clay, or " typical " Marion silt loam) comprising only 37 per cent of the county. These upland soil types vary in agricultural value from $15 to $60 an acre. They vary in average composition from noo pounds of nitrogen and 400 pounds of phosphorus to 3890 pounds of nitro- gen and 820 pounds of phosphorus in 2 million pounds of the sur- face soil. On more than 15,000 acres of the level upland prairie soil, surface drainage can be provided only with much diffi- culty, while 40,000 acres of eroded timberlands are so rough that the soil ought not to be kept under cultivation. About 30 per cent of the upland is not underlain with tight clay, while the remainder has the subsoil sometimes called " hardpan " by the resident farmers. These facts are mentioned in order that the student may under- stand that the soil survey as made by the Bureau of Soils is not intended to be in sufficient detail for local, specific use. To quote Professor Whitney's language from a letter to the author under date of March 26, 1903: "In the work on the scale in which the Bureau is engaged, we cannot recog- nize differences that might and should be recognized in a more detailed survey of a limited area. It is necessary for us to show only important differences in the soils which will be of value to the people of large areas. " Nevertheless, the soil surveys of the Bureau have large value as a source of general information concerning the soils of the United States. The author has very great respect for the art of surveying soils, whether for general information over broad areas or for spe- cific use where the details are mapped. The accompanying map of. the United States, showing " Soil Provinces," as published by the Bureau of Soils, is based in part upon the work of the United States Geological Survey. Within these 14 great soil provinces, the Bureau of Soils had recognized (previous to January i, 1908) different soil types to the number of n6 SCIENCE AND SOIL 715, most of which have been grouped into 86 soil series; and the following extracts from Bureau of Soils Bulletin 55 (1909), de- scriptive of these soils, cannot fail to be of interest and value to the student of American soils. CLASSIFICATION or SOILS "The texture of the soil is expressed in the mechanical analysis by a separa- tion into seven grades, the sizes of which are arbitrarily fixed. The results of the analysis show the percentages of sand, silt, and clay. " "When, aside from texture, the physical and chemical properties of the soil and its method of formation are alike, we have what we call a soil series, extending from the coarse gravelly or sandy soils on the one side to the finer silt and clay soils on the other, and in such a series the texture of the soil deter- mines the distribution of crops. " It would be a comparatively simple matter to compare and classify soils according to the mechanical analysis or texture, but this standard alone is not sufficient, and the problem is in reality a very difficult thing, for in working out the relation of the soils to crops, other factors enter which must be recognized in the correlation. One of the most important of these is the structure or the arrangement of the mineral matter. In some soils the mineral particles have a granular arrangement of flocculated masses, making the soil loose and porous. In others the grains appear to have no such coherency, but exist in a compact form, making the soil hard and compact. We also have the gumbo and adobe soils and others that are exceedingly plastic. Then, again, the amount and character of the organic matter influences not only the productive capacity of the soil, but its adaptation to crops, while the color of the soil has to be considered as indicative of certain obsure chemical or physical relations that influence the adaptation and productivity. The drainage features also come in, often with material influence on the organic constituents, on the aera- tion, and on oxidation processes. " "The experienced soil-survey man can judge very accurately of the texture of the soil material, but even his judgment, before being accepted, is always confirmed by mechanical analysis. " Where soils have a common origin and differ only in texture and are alike in color and in physical properties other than those affected by texture, they are arranged in what we call a series having the soil generic name with qualifying textural terms. We have, for example, the Miami gravelly loam, the Miami fine sand, the Miami sandy loam, the Miami silt loam, and the Miami clay loam as prominent types in the Miami series. In this particular series we have fourteen types, and possibly two or three other types will be encountered. In the Norfolk series we have twelve types." "If the texture and structure of two soils is the same, and one differs in a marked degree from the series color, and that departure is fairly constant and Rocky Mountain Valleys and Plains Northwestern Intermountain Regions U L F OF UNITED STATES SOIL PROVINCES Reduced from maps of U.S. Geological Survey and Bureau of Soils SCALE OF MILES 100 200 300 400 500 from HO Greenwich 85 SURVEYS BY THE UNITED STATES BUREAU 117 typical of the area covered by the soil, this soil likewise is thrown out of the series, because we have reason to know, by observation of the growing crops, that this color difference stands for a difference in the chemical changes which go on in the soil and which are necessary for the welfare of certain crops. " In the classification of soils, therefore, the texture is used to determine the place in the series; the structure and color to determine what series the soil can be correlated with. " The following table gives the name and area of the soil provinces and the proportion of each that has been covered by the soil survey. It is not unlikely that as the work progresses and as our knowledge of the soils increases it will seem advisable to divide some of these provinces into two or more parts. SOIL PROVINCES OP THE UNITED STATES PROVINCE ESTIMATED AREA AREA SURVEYED Atlantic and Gulf coastal plains . . . River flood plains Acres 233000000 64000000 48000000 72OOOOOO 68OOOOOO 455OOOOOO 4IOOOOOO 107000000 lOQOOOOOO 76000000 365000000 58000000 2IOOOOOO 2IIOOOOOO Acres 25613666 8061247 7271798 6367009 6052926 22417832 5091882 1825850 1005600 I4S5428 2939840 "3H55 459388i Per Cent 10 J3 15 9 9 5 12 2 I 2 I 2 22 Piedmont Plateau Appalachian Mountains and plateaus . Limestone valleys and uplands . . . Glacial and loessial Glacial lake and river terraces .... Residual soils of Western prairie . . . Great Basin Northwestern intermountain region . Rocky Mountain valleys and plains . . Arid Southwest Pacific coast Total I9280000OO 93828114 ATLANTIC AND GULF COASTAL PLAINS The Atlantic and Coastal plains together constitute one of the most impor- tant physiographic divisions of the United States. The Atlantic Coastal Plain extends from the New England states southward to the Florida Peninsula, where the Gulf Coastal Plain begins, and extends thence westward to the Mexican boundary line. It is, however, discontinuous, being interrupted by the alluvial bottoms of the Mississippi River. From the coast the Atlantic Plain extends inland to the margin of the Piedmont Plateau; that is, to a line passing through Trenton, Baltimore, Washington, Richmond, Raleigh, Columbia, Augusta, and Macon. In its northern extension it is represented by a narrow belt, but widens in New Jersey, and attains its maximum breadth n8 SCIENCE AND SOIL of about 200 miles in North Carolina. The Gulf Plain extends up the Missis- sippi to the mouth of the Ohio, its inner boundary line passing through or near Montgomery, luka, Cairo, Little Rock, Texarkana, Austin, and San Antonio. The surface is that of a more or less deserted plain marked by few hills, slightly terraced with bluffs along streams. The inner margin of the Coastal Plain is usually from 200 to 300 feet above tide water, but sometimes rises to 500 feet. The drainage here is usually well established, and the surface is rolling to hilly, and consequently carved and eroded. There is a wide belt border- ing the coast where the elevations are mostly under 100 feet. North of the James River, where the Coastal Plain is narrow and deeply indented with tidal estuaries, drainage is usually well established and the surface is rolling, but in the broad southern extension, where the seaward slope is hardly more than i foot to the mile, drainage is apt to be deficient. Here rain water often remains upon the surface for a considerable time, although the conditions, are not comparable with those of a true swamp. The soils in this level section, while composed largely of sand, are compact, usually deficient in organic matter, and not very productive. Many of the flat interstream areas possess such poor drainage that true swamps, such as the Dismal and Okefenokee, have been formed. Near the coast and along the tidal estuaries, extensive marshes, separated from the ocean by sand barriers, are found. The Coastal Plain is made up of unconsolidated gravels, sands, and sandy clays, with less frequent beds of silts and heavy clays. The desposits on the Atlantic coast have been derived mainly from the erosion of the Piedmont Plateau and other inland areas, while the deposits on the Gulf coast have been derived mainly from transported glacial material and from western plains. The materials have been transported and deposited beneath the sea and subsequently exposed by the uplift of the ocean floor. In the more northern parts of the Coastal Plain, and even as far south as Virginia, the character of the deposits has been modified by glacial action and the flooded condition of the streams resulting from the melting of the ice. The Coastal Plain deposits range in age from Cretaceous to Recent. Al- though extensive areas of the older sediment are exposed at the surface to form soils, still by far the greater part of the materials is Quarternary or Recent in age. The soils are for the most part composed of sands and light sandy loams, with occasional deposits of silts and heavy clays. The heavy clays are found principally near the inner margin of the Coastal Plain. The silts, silty clays, and black calcareous soils, upon which the rice and sugar-cane industries of southern Louisiana and Texas are being so extensively developed, have no equivalents in the Atlantic division. Bastrop series. Brown soils with reddish brown to red subsoils occurring as nonoverflow terraces. Cotton, corn, sorghum, alfalfa, melons, and potatoes are successfully produced. Crockett series. Dark gray prairie soils underlain by mottled red sub- soils. Derived from slightly calcareous material, the soils of this series are SURVEYS BY THE UNITED STATES BUREAU 119 productive. Cotton and the general farm crops are the leading products. The gravelly soil is early and adapted to early truck. Elkton series. Light gray to white surface soils, with mottled whitish gray and yellow subsoils, overlying gravel and coarse sands. Gadsden series. Gray soils, with subsoils of similar texture occupying gentle slopes and depressions and formed by wash or creep from higher areas.' This series occurs in rather local development, particularly in Florida, southern Georgia, and Alabama. They are very productive soils and well adapted to tobacco. No heavy members of the series have been encountered, and it is doubtful if any exist. Guin series. Gray soils with brown to yellowish red subsoils, occurring as rolling and hilly lands. Intermediate series between the Norfolk and Orangeburg series. Owing to the rather rough topography, these soils have not been developed as much as either of the other series, although they seem capable of producing better crops than they do now. Houston series. Dark gray or black calcareous prairies. One of the most productive series for Upland cotton and well adapted to alfalfa and other forage crops. Laredo series. Gray to light brown calcareous soils with gray subsoils. Good cotton, corn, and sugar-cane soils, and especially adapted to the early production of vegetables — cabbage and onions in particular. Lufkin series. Light-colored soils with heavy mottled gray and yellow subsoils. The soils of this series have only a moderate degree of productivity. Montrose series. Gray soils with heavy plastic mottled yellow subsoils. These soils are in part poorly drained, but where cultivated they produce moderate yields of cotton and corn. Myatt series. Gray soils with mottled yellow, gray, and whitish subsoils occurring in poorly drained areas around heads of streams and intermediate between uplands and bottom lands. The series seems to be of local extent and but little developed. Norfolk series. Light-colored soils with yellow sand or sandy clay sub- soils. This series contains some of the most valuable truck soils of the Atlantic and Gulf Coast states, and certain members of the series are adapted under certain climatic conditions to wheat, grass, tobacco, and fruit. Oktibbeha series. Gray soils with brown to yellowish brown heavy sub- soils related to Houston series in origin. The soils of this series are distinctly inferior to the soils of the Houston series and, as they appear to cover large areas in Mississippi and Alabama, present a difficult problem in soil improve- ment. Orangeburg series. Light-colored soils with red sandy clay subsoils. This series constitutes some of the best cotton soils of the South, and certain members of the series are particularly adapted to tobacco. Portsmouth series. Dark-colored soils with yellow or mottled gray sand or sandy clay subsoils. Where drainage is adequate, this series is adapted to some of the heavier truck crops, to small fruits, and to Indian corn. 120 SCIENCE AND SOIL Sassafras series. Yellowish brown surface soils with reddish yellow to light orange subsoils overlying gravel beds. Susquehanna series. Gray soils with heavy red clay subsoils which become mottled and variegated in color in the deep subsoil. Only one member of this series, the sandy loam, has been developed to any considerable extent. This one is used for fruit and general farm purposes, but the other members are particularly refractory and difficult to bring into a productive state. Webb series. Brown to reddish brown soils with reddish brown to red sub • soils, a semiarid prototype of the Orangeburg series. The soils of this series have not been used to any great extent, owing to lack of irrigation facilities. Wickham series. Reddish or reddish brown terrace soils overlying reddish, micaceous heavy sandy loam or loam subsoils. The soils of this series have a relatively high productivity for general farm crops. Wilson series. Dark gray prairie soils with mottled gray subsoils. The clay member of this series is a strong soil devoted to general farming, with cotton as the leading crop. The other members are used for cotton, but are inclined to be droughty. RIVER FLOOD PLAINS An extensive and characteristic group of soils, usually known as " bottom lands," is found in the flood plains of numerous rivers and streams of the United States. The largest development of this group occurs along the Mis- sissippi River, where the bottoms are often many miles in width. The soils have been formed by deposition from stream waters during periods of overflow. The texture of the material depends upon the velocity of the current at the time of the deposition. Where the current is very rapid, large stones and bowlders are borne along, and beds of gravel and sand are formed. Along the swift-flowing streams the texture of the soil changes often within short distances, but in wide bottoms large areas of very uniform soils are often formed. The soil material has usually been derived from various kinds of rocks, but in some instances is closely related to the surrounding geological formation. The red soils along the Red and other rivers in the Southwest have been formed by the reworking of the Permian Red Beds. In general, the soils along the streams which flow through the prairie region have a darker color than those along the streams which run only through the timbered sec- tions of the country. The difference in the origin, drainage, color, and organic-matter content has given rise to several series of alluvial soils in the humid portion of the United States. Congaree series. Brown or reddish brown soils found along Piedmont streams and representing wash from Cecil soils. Valuable and dependable corn soils, but too low and moist for cotton. Huntington series. Dark brown to yellowish brown soils occurring along streams in the Alleghany plateaus. Both the general farm crops and truck crops thrive on these soils. SURVEYS BY THE UNITED STATES BUREAU 121 Miller series. Brown to red alluvial soils formed from the reworking of materials derived from the Permian Red Beds. Very productive soils suit- able for cotton, corn, sugar cane, alfalfa, and vegetables; especially adapted to peaches. Ocklocknee series. Gray to yellowish brown soils found along streams in Coastal Plain Georgia, Alabama, and Mississippi. Cotton, corn, and pastur- age are the leading products. Wabash series. Dark brown or black soils subject to overflow. Very pro- ductive soils used for cotton, sugar cane, corn, wheat, oats, grass, alfalfa, sugar beets, and potatoes and other vegetables. Waverly series. Light -colored, alluvial soils subject to overflow. Less pro- ductive than the Wabash soils, but adapted to the same wide range of crops. Wheeling series. Brown to yellowish brown soils occurring on gravel ter- races along streams issuing from glaciated regions. Excellent soils for general farming, and fruit and truck growing. PIEDMONT PLATEAU Lying between the Atlantic Coastal Plain and the Appalachian Moun- tains and extending from the Hudson River to east-central Alabama is an area of gently rolling to hilly country known as the Piedmont Plateau. On the Atlantic side it is closely denned by the "fall line," which separates it from the Coastal Plain, but on the northwestern side the boundary is not sharp, although in the main distinct. In its northern extension the Piedmont Plateau is quite narrow, but broadens toward the south, attaining its greatest width in North Carolina. The surface features are those of a broad rolling plain that has been deeply cut by an intricate system of small streams, whose valley walls are rounded and covered with soil, although many small gorges and rocky areas occur. The altitude varies from about 300 feet to more than 1000 feet above sea level. The extreme northern part of the Piedmont region, in New Jersey, has been glaciated, but elsewhere the soils are purely residual in origin and have been derived almost exclusively from the weathering of igneous and metamorphic rocks. The chief exception is the detached areas of sandstones and shales of Triassic age. Marked differences in the character of the rock and the method of formation has given rise to a number of soil types, those derived from crys- talline rocks being the most numerous and widely distributed. Among these the soils of the Cecil and Chester series predominate. The principal types formed from the sandstones and shales are included in the Penn series. Cecil series. Gray to red soils with bright red clay subsoils, derived from igneous and metamorphic rocks. Constituting by far the larger portion of the province, these soils are well adapted to, and used for, cotton, export tobacco, and fruit, and the lighter members for truck crops. As a rule, they are not highly developed, but where properly handled the heavier members produce excellent crops of corn and grazing and hay grasses. 122 SCIENCE AND SOIL Chester series. Gray to brown surface soils with yellow subsoils, derived principally from schists and gneisses. The most valuable soils of the province for wheat and corn, and good for certain fruits. The most highly developed soils of the Piedmont Plateau. Penn series. Dark Indian red soils with red subsoils, derived from red sand- stones and shales of Triassic age. Excellent soils for general farm crops, particularly wheat, corn, and hay. APPALACHIAN MOUNTAINS AND PLATEAU The Appalachian Mountains are made up of a number of parallel ranges and intervening valleys, which extend in a general northeast and southwest di- rection from southern New York to northern Alabama. The elevation ranges from about 1500 to nearly 7000 feet above sea level, the highest point being attained in western North Carolina. Immediately east of the Appalachian Mountains, and usually separated from them by a valley, is a wide stretch of country know as the Alleghany Plateau. In a broad way this plateau is carved out of a great block of sedimentary rocks tilted to the northwest from the mountains. It is crossed by numerous streams. As they run in deep channels (all the larger ones being from 200 to 1000 feet in depth), the dissection of the plateau block is often minute. The rocks of the eastern ranges of the Appalachian Mountains are igneous or metamorphic in origin, while the western ranges, as well as the Alleghany Plateau, are made up of sedimentary rocks. Different series of soils have, therefore, been formed in different parts of these mountains and plateau. The igneous and metamorphic rocks give rise to the soils of the Porter series, while the Dekalb and Upshur series are formed from the weathering of the sandstones and shales of sedimentary origin. The character of the topography in the mountain and much of the plateau region is such that general farming is not practicable. These areas are, how- ever, well suited to grazing and fruit growing, and these are very important industries. Dekalb series. Brown to yellow soils with yellow subsoils, derived from sand stones and shales. Soils of this series are used, according to texture, elevation, exposure, and character of surface, either for the production of hay, for pasture, or for orchard and small fruits. Fayetteville series. Grayish brown to brown soils with yellowish or reddish brown subsoils. Adapted to apples, grapes, and small fruits, and give mod- erate yields of general farm crops. Porter series. Gray to red soils with red clay subsoils, derived from igneous and metamorphic rocks. This is the greatest mountain fruit series of the eastern United States. It is also used for general farming. Upshur series. Brown to red soils with red subsoils, derived from sand- stones and shales. Somewhat more productive than the Dekalb soils. Used for cotton, corn, wheat, and forage crops. SURVEYS BY THE UNITED STATES BUREAU 123 LIMESTONE VALLEYS AND UPLANDS The limestone soils are among the most extensively developed of any in the United States and occur in both broad upland and inclosed narrow valley areas. The greatest upland development is seen upon the Cumberland Plateau in eastern Tennessee and Kentucky and upon the Carboniferous formation in central Tennessee and Kentucky, northern Alabama and Georgia, and in Missouri. The valley soils are found principally in Pennsylvania, Maryland, and Virginia, and in the mountain section of eastern Tennessee and Kentucky and northern Alabama and Georgia. The topography of the plateau soils varies considerably. In the Cumberland Plateau and Highland Rim the sur- face is undulating ; in the region of the Ozark uplift in Missouri and Arkansas it is quite rough and hilly, and where there is an elevation of the surface, or where the plateau is deeply dissected by erosion, it presents a quite mountain- ous topography. The valley soils of the Appalachian region also show consider- able topographic relief, sometimes exhibiting mountainous surface features. The limestone soils are residual in origin, being derived from the weather- ing in place of limestone of differing age and composition. This is accom- plished by the removal through solution of the calcium carbonate of the lime- stone, leaving behind the more resistant siliceous minerals. These soils are remarkable for the fact that they contain but a very small percentage of the original limestone rock, the larger part having gone into solution. It has thus required the solution of many feet of rock to form a single foot of soil. Thus far the limestone soils east of Kansas and Texas and north of central Alabama and Georgia have been grouped in two important series, known as the Hagerstown and Clarksville. Clarksville series. Light gray to brown soils with yellow to red subsoils, derived mainly from the St. Louis limestone. While not as strong as the Hagerstown soils, this is a valuable series. Apples and peaches are commer- cially important. Tobacco is a leading product. General farming is firmly established in many extensive regions. Cumberland series. Brown surface soils, derived from thin deposit of sedimentary material overlying residual limestone subsoils. Used for cotton and other general farm crops, truck and fruit. Decatur series. Reddish brown to red soils with intensely red subsoils. Intermediate in value between the two series just described. Cotton, corn, wheat, oats, forage crops, blue grass, and peaches are the leading crops. Hagerstown series. Brown to yellowish soils with yellow to reddish sub- soils, derived from massive limestone. Among the most productive soils of the eastern United States. Fine wheat and general farming soils, and the seat of important apple orcharding interests. Blue grass is indigenous. GLACIAL AND LOESSIAL REGIONS The soils of the glaciated part of the country constitute one of the most important groups in the United States. The group includes all soils derived 124 SCIENCE AND SOIL directly from till or loess. The soils formed from the till are confined to that part of the country lying north of the southern limit of glacial action, but the loess soils occur also south of this line, especially along the Mississippi and Ohio rivers and in Kansas and Nebraska. The line of the southern extension of the ice sheet touches the Atlantic coast about New York City, passes through northern New Jersey, southern New York, and northwestern Pennsylvania, swings southward through Ohio to Cincinnati, crosses the Mississippi River at St. Louis, and follows the south side of the Missouri River into Montana, where it crosses the Canadian boundary line, then dips southward into Idaho as a long lobe in the mountainous nonagricultural region, and crosses the northwestern part of Washington, including the Puget Sound region. Practically all of the United States north of this line was covered in recent geological time by a great continental glacier, many hundreds, and even thou- sands, of feet in thickness. This great ice sheet, moving in a southern direction, filled up valleys, planed off the tops of hills and mountains, ground up the underlying rocks, carried the derived material both within and upon the ice, and finally deposited the gravel, sand, silt, and clay, as a mantle, varying in thickness from a few feet to more than 300 feet. Often this material has been transported hundreds of miles, and is wholly unrelated to the underlying rocks, but in some places the movement has been slight, and the drift consists very largely of the ground-up underlying rock. Over a large porportion of the area covered by the drift and also along the Ohio and Mississippi rivers and in Kansas and Nebraska, the surface material consists of a fine silty deposit, known geologically as "loess" and "Plains marl." In the classification of the glacial soils, three important series — Miami, Marshall, and Volusia — having distinct characteristics have been recognized and, in addition, quite a number of miscellaneous soils which cannot be put in any series. Marshall series. Dark-colored upland prairie soils. The principal soils of the great corn belt belong to this series, while in the Northwest the finest wheat soils are found in this group. They are among the best general farming soils of the entire country. Miami series. Light-colored upland timbered soils. The different mem- bers of this series are considered good general farming soils and have in ad- dition special adaptations for truck, fruit, small fruit, and alfalfa. Volusia series. Light-colored soils with yellowish subsoils, derived by feeble glacial action from sandstones and shales. The soils of this series are adapted to the production of potatoes, grass, oats, buckwheat, and, in the less elevated positions, to corn. GLACIAL LAKE AND RIVER TERRACES Another important group of soils occurs in the glacial region, principally as terraces around lakes, or along streams, or as deposits in areas which were formerly covered by water. At the close of the glacial epoch the lakes in this part of the United States were not only more numerous, but the waters of those which remain reached a higher level and covered areas that are now far above SURVEYS BY THE UNITED STATES BUREAU 125 their present shore lines. In some cases several distinct terraces, each one marked by an old shore line, are easily discernible, and represent successive stages in the lowering of the water level. The elevation above the lake varies from a few feet to more than 200 feet. The surface of each terrace is usually rolling to level, with a gradual slope toward the lake, but sometimes areas of a rough and broken character occur. The streams which cross these terraces have frequently, by their cutting, produced deep, steep-sided valleys, especially near the lakes. The soils of this group vary from typical beach gravels to offshore deposits of heavy clays. The material from which they are derived consists of glacial debris reworked and redeposited in the lakes or along streams. While this glacial material is made up of rocks of widely varying origin, a large proportion of it often consists of the country rock. In the eastern part of the Great Lake region the percentage of sandstone and shale fragments is usually very high, while in the western part more of the igneous rocks are present. This fact, together with differences in drainage conditions, has given rise to several series of soils. Clyde series. Dark-colored swamp soils formed from reworked glacial material deposited in glacial lakes. A special use for these soils is the pro- duction of sugar beets, while general farm crops, truck and canning crops, are grown extensively. Dunkirk series. Light-colored reworked glacial material occurring as terraces around lakes and along streams. Good general farming soils and especially adapted to grapes and other fruits. Fargo series. Black calcareous soils rich in organic matter formed by deposition of material in glacial lakes. This is the most important group of soils in the Red River Valley, and includes exceptional soils for the production of wheat, barley, and flax. While these are the chief crops at present, the soil adaptations are by no means limited to small grain production. Timothy and vegetables may become more important products with the development of markets. Hudson series. Light brown to yellowish brown soils, with drab to yellow- ish subsoils. Merrimac series. Brown terrace soils underlain by gravel, formed prin- cipally of reworked glaciated crystalline rocks. Leachy soils of low general farming value, but especially adapted to trucking and apple orcharding in some sections. Sioux series. Dark-colored soils resting on dark or light-colored subsoils, with gravel beds usually within 3 feet of the surface. The crops produced on soils of this series range from early short-seasoned truck crops through special crops like alfalfa and sugar beets to the wide variety of general farm crops produced in the Central West. Superior series. Gray and red soils with red subsoils, formed from reworked glacial material deposited in glacial lakes. Not extensively developed, but known to include fine types for clover, timothy, and small fruits. 126 SCIENCE AND SOIL Vergennes series. Light-colored soils, with gray or whitish subsoils, derived from Champlain clay, or lighter deposits over these clays. This series includes the best hay and apple soils of the Champlain Valley. A wide variety of tillage crops is grown, but cultivation of the heavier members of the series is very difficult. RESIDUAL SOILS OF THE WESTERN PRAIRIE REGION This region consists of the nonglacial part of the prairie plains bounded on the north by the Missouri River, the southern limit of glaciers, and extending southward through Texas to the Rio Grande. On the west it merges into the Plateau region at very near the 2ooo-foot contour, and on the east is limited by the Gulf Coastal Plain and the Ozark Plateau. Its surface is gently rolling, with occasional low hills, and is cut by numerous stream channels. The rocks are of Carboniferous age, and consist of sandstones, shales, and limestones more or less interbedded. These rocks give rise to three series of soils, viz. Oswego, Crawford, and Vernon, together with a number of miscellaneous soils. In Kansas and Texas these soils are in some instances more or less modified by the admixture of gravel and sand from Tertiary deposits brought down from the higher areas farther west occupied by crystalline rocks. Crawford series. Brown soils with reddish subsoils, derived from lime- stones. The soils of this series range from rough areas suited mainly for pastures to fertile general farming, fruit growing, and trucking soils. Oswego series. Gray or brown soils, derived from sandstones and shales. The lighter members of this series are adapted to corn, oats, potatoes, truck, and fruit ; the heavier to these crops and wheat. Vernon series. Brown to red soils typical of the Permian formation. Soils of this series show a wide adaptation according to texture. General farm crops, including cotton, corn, wheat, Kafir corn, and sorghum are the leading prod- ucts. Small fruit, peaches, and truck are grown to some extent and are capable of marked extension. GREAT BASIN With the exception of one soil type recognized in the Laramie area, Wyoming, the soils in this group, so far as mapped, are confined to the Great Interior Basin region. They are derived from a great variety of rocks, and consist of colluvial soil of the mountain slopes, deep lacustrine and shore deposits of the Bonneville period, and of recent stream-valley sediments and river-delta deposits. When not situated above or outside the limits of irrigation, or rendered unfit for cultivation by accumulation of alkali or seepage waters, they are of great agricultural importance, and are devoted mainly to the production of grains, sugar beets, alfalfa, stone or other tree fruits, and vegetables. Bingham series. Porous dark or drab colluvial and alluvial soils under- lain by gravel or rock, occupying lower mountain slopes. The lighter types SURVEYS BY THE UNITED STATES BUREAU 127 when irrigable are devoted to orchard fruits, the heavier types to alfalfa and sugar beets. Jordan series. Light to dark-colored lacustrine desposits. These soils are utilized principally in the production of alfalfa, sugar beets, truck crops, and grains under favorable conditions for irrigation and drainage, but consider- able areas covered by some of the members of this series are not utilized on account of the accumulation of alkali, poor drainage, or because of their drift- ing character. Malade series. Dark-colored alluvial soils underlain by light-colored sands, sandy loams, or heavy reddish material. These soils are devoted chiefly to sugar beets, alfalfa, grain, and some orchard fruits. Redfield series. Red soils consisting of colluvial and alluvial materials derived from red sandstones and other rocks. The lighter members are adapted to the production of alfalfa, grain, and general farm crops when irrigable and well drained. The heavier members, as far as encountered, are poorly drained and have not been developed. Salt Lake series. Dark-colored soils underlain by stratified sediments of lacustrine origin. These soils, as far as encountered, occupy very low, flat positions around the lake, and have not been developed to any extent. NORTHWESTERN INTERMOUNTAIN REGION The most extensive and uniform soil types of this region consist of residual materials overlying and derived from extensive basaltic lava plains and in some cases from granite rocks or of ancient lacustrine sediments or extensive lake beds now more or less modified by erosion or aeolian agencies. Owing to erosion by streams and to movements of the earth's crust, these soils now generally' occupy more or less elevated sloping or rolling plains. About the margins of the lacustrine or residual deposits they are covered by sloping plains and fans of colluvial wash from the adjacent mountain borders, while in the vicinity of the larger streams, which have carved and terraced the lacustrine beds and residual soils, occur other series of recent alluvial stream sediments derived from reworked materials of the lake beds or from the weathered prod- ucts of the mountains. It is the soils of this region that constitute a large portion of the great grain -producing lands of the Northwest. Bridger series. Dark-colored soils with sticky yellow subsoils, of colluvial and alluvial origin. These soils generally occupy elevated foot slopes or sloping valley plains and have not been developed to a great extent. They are most extensively used for the production of grain, and, when irrigated, are utilized in the production of alfalfa and other hay crops and, under favorable climatic conditions, are adapted to fruits. Gallatin series. Light to dark-colored soils with yellowish to dark com- pact subsoils, of recent alluvial origin from basaltic and volcanic rocks. These soils generally occupy low positions, very frequently poorly drained, often subject to overflow, and have not been extensively developed for agricultural 128 SCIENCE AND SOIL purposes. They are used chiefly for grazing and, to some extent, in the pro- duction of hay, grains, and in some sections for vegetables. Yakima series. Ash-gray to light brown soils derived principally from ancient lake sediments consisting of an admixture of volcanic dust and ba- saltic, andesitic, and granitic materials. Certain members of this series have been very successfully developed for hop culture, alfalfa, grass, grain, and fruit, while other members of the series, owing to their elevated position and general rough character, have not been developed at all. ROCKY MOUNTAIN VALLEYS, PLATEAUS, AND PLAINS The soils of the Rocky Mountain valleys, plateaus, and plains are derived from a wide range of igneous, eruptive, metamorphic, and sedimentary rocks. The plateau and plain types occupy a more or less elevated position, and have sloping, undulating, or irregular surface features. They are derived from underlying sedimentary rocks or consist of the remains of the ancient extensive mountain foot-slope material or of alluvial deposits along streams trenching and terracing the sedimentary rocks of the plateaus and plains. The mountain slope and intermountain types consist of residual and colluvial deposits or of ancient lacustrine or later stream sediments, occupying mountain foot slopes and narrow valleys. The soils of the mountain slopes are usually of little agricultural value, owing to their rough surface, elevated position, and the consequent imprac- ticability of irrigation. Those of the plateaus, valleys, and plains vary widely in economic importance, depending largely upon climatic features, topography, position, and water supply for irrigation. They range from grazing lands of nominal value to soils adapted to the most important and intensively culti- vated fruit, melon, sugar beet, and other special crops. Billings series. Compact adobe -like gray to dark or brown soils and sub- soils, formed mainly by reworking of sandstones and shales, and occupying old elevated stream terraces. This is an important series adapted to alfalfa and general farm crops and stock raising ; also used to a considerable extent in the production of sugar beets. Colorado series. Light gray to reddish brown soils and subsoils, derived from colluvial wash. Where irrigable, these soils are important soils in the production of alfalfa, sugar beets, melons, and, to a limited extent, fruits. A number of the soils of the series, however, are so situated as not to be suscep- tible to irrigation, and have not been developed for agricultural purposes. Finney series. Brown to nearly black soils derived from glacial material underlain by lighter-colored subsoils. The heavier soils may be dry farmed to advantage, and would become very productive with irrigation. The lighter soils have a broken surface, are porous, and easily drifted by the wind. They are best adapted to grazing. Fruita series. Reddish brown soils, formed by reworking of sandstones and shales, occurring as stream terraces. When well drained and free from SURVEYS BY THE UNITED STATES BUREAU 129 alkali, the members of this series are admirably adapted to the production of choice fruits, alfalfa, sugar beets, grains, and truck crops. Laramie series. Dark-colored soils, with light-colored gravelly subsoils, de- rived from colluvial mountain wash. These soils have not been extensively developed, owing to their elevation, and are used principally for grazing pur- poses. Laurel series. Light gray to black soils, underlain by river sands or gravels, occurring in flood plains along streams. Under favorable moisture conditions, these are fertile soils, adapted according to locality to corn, alfalfa, sugar beets, and truck crops, but the areas are often subject to overflow, and in some cases cannot be drained. Mesa series. Light gray to brown soils derived from old flood -plain de- posits, now elevated to form mesa lands. Where these soils have been devel- oped and are susceptible of irrigation, they are used mainly for alfalfa and sugar beets. One member of the series has been quite extensively and very successfully used for the production of apples and peaches. Morton series. Brown residual soils, derived from sandstones and shales- The soils lie in the semiarid region, and give good yields of wheat, flax, oats, and potatoes, when rainfall is sufficient. San Luis series. Reddish brown gravelly soils, formed from lacustrine sediments of volcanic rock materials. On account of the position and the danger from alkali, these soils have not been successfully developed, but have been used mainly for pasturage and forage crops. Wade series. Brown to dark brown alluvial soils, formed by reworking of sandstones and shales. Used for oats, flax, millet, and wheat. ARID SOUTHWEST The soils of the arid Southwest are mainly of colluvial, alluvial, and lacus- trine origin. They occupy mountain foot slopes, alluvial fans, debris aprons, or sloping plains of filled valleys, sloping or nearly level plains, and bottoms of stream valleys or sinks and drainage basins. The principal colluvial soils of this region are also common to the Pacific coast. The climate of the arid Southwest is characterized by semitropical desert conditions, and where the soils are not capable of irrigation, they have little or no present agricultural value. Gila series. Light to dark brown soils of flood-plain alluvium, underlain at varying depths by coarse sands and gravels. Under favorable irrigation and drainage conditions, the members of the Gila series are adapted chiefly to the production of alfalfa, potatoes, truck, and root crops. Imperial series. Light-colored or reddish soils formed from old marine or lacustrine sediments modified by more recent deposits, and underlain to great depths by heavy material. These soils are particularly adapted to alfalfa, sorghum, and other forage crops. India series. Light -colored soils usually underlain by coarser sands and 130 SCIENCE AND SOIL gravels, formed by colluvial and alluvial wash from granitic rock, mingled with some shale and sandstone. These soils are adapted to fruit, truck crops, sweet potatoes, melons, and alfalfa, under favorable conditions of irrigation and drainage. PACIFIC COAST The soils of the Pacific coast, including those of the coastal and interior mountain ranges, foothills, and valleys, have been classified into a number of series, varying in field characteristics, topography, origin and mode of formation, and agricultural importance. They range from residual and colluvial soils of the mountain sides, foot slopes, and foothills, to deep and extensive river flood plains and delta sediments, and ancient and modern shore and marine lacustrine deposits. While some of these series are confined to a single coastal or interior mountain range or valley, others are of wider range and extend over several different physiographic regions. The value of these soils and their adaptation to crops is dependent largely upon the possibilities of irrigation and upon local conditions of rainfall and temperature, all of which are to great extent dependent upon topography. They range in agri- cultural importance from those devoted only to extensive grain farming to the most valuable and intensively cultivated lands devoted to citrus and decid- uous fruits, vines, small fruits, and other special crops. Anderson series. Reddish gray or light red alluvial soils occupying prin- cipal valley plains and the bottoms of intermittent streams. Generally gravelly. The. soils of this series, when not too gravelly, are adapted to the production of peaches, pears, prunes, and small fruits, but are, in so far as mapped at present, inextensive types of secondary agricultural importance. Fresno series. Light-colored soils with light gray, ashy subsoils, and alkali -carbonate hardpan, derived from old alluvial wash. Where protected from alkali accumulations, these soils have been very successfully used for vineyards and raisin grapes, and are particularly adapted to almonds, peaches, and apricots. Hanford series. Recent alluvium of flood or delta plains derived from a variety of rocks. The light -textured soils are light in color, and the heavy tex- tured soils are dark in color. The lighter members of the series are adapted to the same class of fruits and raisin grapes as the Fresno series. The heavier members of this series, however, are better adapted to alfalfa, sugar beets, celery, asparagus, and other truck crops. Maricopa series. Loose, dark-colored soils derived from unassorted col- luvial or partially assorted alluvial materials, generally derived from granitic or volcanic rocks. There are two heavy members of this series upon which alfalfa, grain, and sugar beets are important crops. The lighter members, when occupying positions so that they can be irrigated, are adapted to citrus and deciduous fruits ; also vines. Oxnard series. Dark-colored alluvial or colluvial soils derived from higher lying areas of sandstones and shales. Members of this series are used to a very SURVEYS BY THE UNITED STATES BUREAU 131 large extent for sugar beets and lima beans, and, where irrigation is not practi- cable, extensively used for grain. Placentia series. Reddish soils derived largely from the weathering of allu- vial and colluvial deposits, generally underlain by heavy compact red material with an impervious adobe structure. Large areas of these soils are devoted to dry farming of grain, and occur throughout southern California and in some of the coastal valleys, viz., Bakersfield, Salinas, and San Jose. These are exten- sive areas under irrigation, which are valuable for producing both deciduous and citrus fruits. The heavier members of this series have been more success- fully used for grain production in southern California. They seem particularly well adapted to English walnuts and olives. The soils are usually well drained. The English walnut does not thrive on poorly drained soils. Redding series. Ancient alluvial valley deposits of red and deep red color, generally gravelly. Heavy red subsoils with hardpan. The soils of this series, when not carrying an excess of cobbles or underlain at shallow depths by hardpan, are excellently adapted to the production of choice peaches and small fruits. Sacramento series. Gray alluvial soils consisting of recent stream sedi- ments. The lighter members of this series are used mainly for the production of prunes, pears, and peaches. The members of the series having a medium texture are adapted to sugar beets, alfalfa, and prunes. The heavier members are at present poorly drained, and have not been highly developed, being used mostly for grain and grazing. Salem series. Residual, alluvial, or colluvial soils, either red or dark in color, derived from rocks of basaltic, schistose, crystalline, or arenaceous character. These soils, so far as they have been encountered, seem particularly adapted to hops, potatoes, and have been used to some extent 'for apples, peaches, and grain. They have not been very highly developed in the areas in which they have been encountered. San Joaquin series. Compact red soils and subsoils derived from old marine sediments, usually underlain by red hardpan. These soils have been used almost exclusively for dry farming to grain on account of the general occurrence of hardpan and very stiff and impervious subsoils. Recently, in the Sacramento area, some members of this series have been very successfully used for the production of the Tokay grape and strawberries. Sierra series. Light gray to red and frequently gravelly soils, often under- lain by red adobe. Members of this series constitute some of the most valu- able deciduous fruit soils of the foothills in northern California. Sites series. Residual and colluvial soils of reddish gray or dark brown color, derived from sandstones, shales, conglomerates, and volcanic or altered ma- terial occupying low, rolling foothills and their valley slopes, usually underlain at shallow depths by sandstones, conglomerates, or heavy subsoils. The Sites loam and clay loam adobe are the important soils of this series and are pro- ductive, but, owing to their positions, are generally unirrigable and adapted to dry farming to grains. 132 SCIENCE AND SOIL Stockton series. Brown to black soils with heavy yellow subsoils, derived from old alluvial sediments. These soils have been used principally for the production of grain. The lighter members of this series have been adapted to fruit. Willow series. Brown soils consisting of wash deposited by intermittent foot- hill streams. These soils have been used almost exclusively for dry farming grain crops. Large ranches are being broken up and brought under irrigation, and alfalfa and sugar beets are likely to prove the most important crops. The following additional quotations from Bureau of Soils Bulle- tin 55 will serve to acquaint the student with the general charac- ter of the more detailed descriptions which are given of the soil types singly and in series: Leonardtown loam1 (Maryland, Virginia, Kentucky, — 196,834 acres). "The Leonardtown loam is a valuable upland soil of Maryland and Virginia. The surface is slightly rolling, the drainage in most areas is good, and altogether the land is well suited to general farming. The soil has a special value in the production of wheat and grass." Marion silt loam (Illinois, Missouri, — 695,040 acres). "A large pro- portion of southern Illinois is occupied by Marion silt loam. The type occupies level prairie land and is characterized by hard silty clay subsoil locally known as ' hardpan. ' It is low in organic matter, and this, combined with the impervious nature of the subsoil, causes crops to suffer in wet as well as dry seasons. Wheat, corn, and grasses are the principal crops, but the average yields are considerably lower than upon the black prairie soils. It seems especially well adapted to apples, and many large orchards have been planted. Strawberries also do well." Marshall series (glacial and loessial regions). "The Marshall series in- cludes the dark-colored upland glacial and loessial soils, which cover almost all of the great prairie region of the Central West. The soils of this series are characterized and distinguished from those of the Miami series by the relatively large quantity of organic matter in the surface soils, which gives them a dark brown to black color. The topography is level to rolling, and artificial drainage is necessary on many level and low -lying areas to secure the best results. The soils of this series are very productive and constitute the great corn soils of the country. "The Marshall silt loam, loam, and clay loanf constitute the principal soil types throughout the great corn belt, and rank among the most productive of our general farming soils. In Iowa, Illinois, and Nebraska, corn, oats, clover, and timothy are the leading crops, while in Minnesota and the Dakotas wheat becomes of primary importance. The Miami (Marshall) black clay loam, when drained, is also an exceedingly fertile soil, being particularly well adapted 1 The Leonardtown loam and Leonardtown gravelly loam in the Norfolk, Vir- ginia, report are the Portsmouth silt loam. SURVEYS BY THE UNITED STATES BUREAU 133 to corn. The sandy loam and fine sandy loam, while not so well adapted to general farming as the heavier soils, are quite productive and have a wide crop adaptation. The sand and fine sand are well suited to truck crops, but give rather uncertain yields of general farm crops. "The acreage of the types so far encountered is as follows : AREA AND DISTRIBUTION OF THE SOILS OF THE MARSHALL SERIES SOIL NAME STATES IN WHICH EACH TYPE HAS BEEN FOUND TOTAL AREA (Acres) Marshall stony loam Gravel * . • . . Gravelly loam . Sand .... Fine sand . . Sandy loam . . Fine sandy loam Loam .... Silt loam 3 Clay loam Black clay loam 3 Clay .... Total . North Dakota, South Dakota .... Minnesota, North Dakota Kansas, Minnesota, NorthDakota, Wisconsin Indiana, Iowa, Wisconsin Indiana, Iowa, Minnesota, Nebraska, North Dakota Illinois, Indiana, Kansas, Minnesota, South Dakota . . ' Indiana, Minnesota, Nebraska, North Da- kota Illinois, Indiana, Iowa, Michigan, Minnesota, Nebraska, North Dakota, South Dakota, Wisconsin Colorado, Illinois, Indiana, Iowa, Kansas, Louisiana, Minnesota, Missouri, Nebraska, North Dakota, Wisconsin Iowa, Minnesota, North Dakota, Wisconsin Illinois, Indiana, Iowa, Michigan, Minnesota, North Dakota, Ohio, South Dakota, Wis- consin North Dakota . 84096 2560 106816 52736 261440 1680832 4454470 600320 572176 76800 8057686 1 The soil mapped as Marshall gravel in Pontiac area, Michigan, is Miami gravelly sand. 2 Mapped as Miami silt loam in Clinton and St. Clair counties, Illinois, and as Fresno fine sandy loam in Lower Arkansas Valley area, Colorado. 3 The soil mapped as Miami (now Marshall) black clay loam in the Toledo area, Ohio, is Clyde clay. Miami series (glacial and loessial regions). "The Miami series is one of the most important, widely distributed, and complete soil series that has been established. The series is characterized by the light color of the surface soils, by derivation from glacial material, and by being timbered either now or originally. The heavier members of the series are better adapted to wheat I34 SCIENCE AND SOIL AREA AND DISTRIBUTION OF THE SOILS OF THE MIAMI SERIES Son. NAME STATES IN WHICH EACH TYPE HAS BEEN FOUND TOTAL AREA Miami stony sand . Michigan, New York, Washington, Wiscon- sin 100278 Stony sandy loam New York, Rhode Island, Vermont, Wash- ington 267^28 Stony loam . . Gravel Michigan, Minnesota, New York, Ohio, Rhode Island, Washington, Wisconsin Illinois, Wisconsin 879094 21^76 Gravelly sand * Michigan Washington QI242 Gravellysandy loam Indiana, Michigan, Minnesota, Washing- ton . 58624 Gravelly loam 2 . Indiana, Michigan, Ohio 71232 Sand .... Fine sand . . . Sandy loam 3 Indiana, Kansas, Michigan, Minnesota, Nebraska, Ohio, Wisconsin .... Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, New York, Wisconsin Indiana, Iowa, Michigan, Minnesota, Ohio, Washington, Wisconsin 795720 263564 74X46O Fine sandy loam 4 Loam 5 . . . . Silt loam 8 . . Indiana, Michigan, New York .... Indiana, Michigan, Wisconsin .... Illinois, Indiana, Iowa, Kentucky, Mis- souri, Nebraska, Rhode Island, Wiscon- sin 130816 2I472O 10^1488 Clay loam 7 . . Indiana, Iowa, Michigan, Ohio, Washing- ton, Wisconsin . . . . '. . . . l83l8l8 Total . . . 742276O 1 Mapped as Marshall gravel in Pontiac area, Michigan. 2 The soil mapped as Miami gravelly loam in the Big Flats area and Syracuse area, New York, is the Dunkirk gravelly loam. 3 The soil mapped as Miami sandy loam in the Grand Forks area, North Dakota, is the Clyde fine sandy loam; in the Montgomery County area, Ohio, is Wabash sandy loam; and in Posey County area, Indiana, is Wabash fine sandy loam. 4 The soil mapped as Miami fine sandy loam in Posey County, Indiana, and Union County, Kentucky, is Waverly fine sandy loam; in the Boonville area, Indiana, is the Norfolk fine sandy loam; in the Lyons and Syracuse areas, New York, is Dunkirk fine sandy loam; and in St. Claire County, Illinois, is Memphis silt loam. 5 The Miami loam in the Auburn, Lyons, and Syracuse areas, New York; the Columbus, Coshocton, Montgomery, Toledo, and Westerville areas, Ohio; the Fargo and Grand Forks areas, North Dakota; the Marshall, Minnesota, and Pontiac areas, Michigan; and the Viroqua area, Wisconsin, is Wabash loam. The soil mapped as Miami loam in Tazewell County, Illinois, is Sioux loam, and that mapped as Miami loam in the Janesville area, Wisconsin, is the Sioux sandy loam. 6 The soil mapped as Miami silt loam in the Syracuse area, New York, is Dunkirk SURVEYS BY THE UNITED STATES BUREAU 135 than the corresponding members of the Marshall series, but they do not pro- duce as large yields of corn. "The clay loam is the most important for general fanning, and forms the principal type of soil in western Ohio and central and eastern Indiana. It is especially well adapted to small grain and grass crops. The silt loam is more rolling and hilly than the clay loam and is not so well suited to general farming. Wheat does better upon it than upon the Marshall silt loam, with which it is closely associated, but the yields of corn are considerably less. It is also well adapted to fruit, especially apples. The sandy loam and fine sandy loam are used for general agriculture, but are especially adapted to me- dium and late truck crops and fruit. The loam is suited to corn and potatoes, while small grain and grass are grown, but with less success than upon the clay loam. Strawberries and raspberries, as well as other small fruits, do well on this type. The stony sand, gravelly sand, and gravel are not of much agricul- tural value under present conditions. The stony loam is a good general farm- ing soil, is also well adapted to apples, and furnishes excellent pasture, while in New York alfalfa is grown upon it very successfully. The stony sandy loam and gravelly sandy loam are not strong soils, but are fairly well suited to light farming, fruit, and truck. The sand and fine sand are not adapted to general farming, but are the best early truck soils of this section. "The acreage of the different types so far encountered is shown in the preceding table." silt loam, and that mapped as Miami silt loam in Clinton and St. Claire counties, Illinois, is Marshall silt loam. 7 The soil mapped as Miami clay loam in Toledo area, Ohio, is Dunkirk clay loam, and that mapped as Miami clay loam in the Stuttgart area, Arkansas, is Crowley silt loam. CHAPTER IX SOIL ANALYSIS BY THE UNITED STATES BUREAU OF SOILS THE United States Bureau of Soils Bulletin 54 (December, 1908), on "The Mineral Composition of Soil Particles," contains data from which can be computed 1 accurately the total amounts of phosphorus, potassium, magnesium, and calcium, in the ignited surface soil of twenty-seven important soil types of the United States. The loss on ignition of ordinary soils usually approaches 10 per cent, and includes chiefly the combined water, organic matter, and more or less carbon dioxid, if carbonates are present; consequently, the results given on the basis of ignited soil are, as a rule, about one tenth higher than if given on the usual basis of dry soil. The following general statements regarding these soil samples are made by the Bureau of Soils (Bulletin 54, page 15) : "Our extensive collection of soils from important and well-marked soil types enables us to select samples fully representative of the soils of the country. Accordingly, agricultural soils of known character were selected so as to include those from various geographical sections and from a number of soil provinces. Thus we have taken soils from the Coastal Plains, the Piedmont region, glacial soils, nonglacial soils of the interior, and those of the arid region, the list comprising 27 soil types." "We have but two soils of the arid region to compare with the twenty-five of the humid region. The latter were collected to represent soils of all classes — those of low, of medium, and of high productivity ; sandy soils, clay soils, calcareous soils, and those intermediate between these extremes. They may be taken as fairly well representing the humid soils. The two arid soils cannot be considered to represent so well those of the region because of their limited number and similarity of texture, both being fine sandy loams." "The Coastal Plains soils have resulted, to a large extent, from material washed from the Piedmont Plateau and deposited in water at lower levels. 1 The Bureau of Soils Report shows, for example (Bulletin 54, page 19), that Leonardtown loam contains 29.5 per cent of sand, 55 per cent of silt, and 15 per cent of clay, and that the total PaOs which these particles contain is .01 per cent in the sand, .02 per cent in the silt, and .03 per cent in the clay. 136 ANALYSIS BY THE UNITED STATES BUREAU 137 They have suffered from decomposition and solution more than have the Piedmont and Appalachian soils, and there has often been a greater .separation of the finer from the coarser particles. " "The tables show that the residual soils, Chester mica loam, Porter's black loam, and Cecil clay, contain more plant-food constituents than do the Coastal Plains soils. This is especially true of the phosphorus, the potassium, and the magnesium." "The sandy and the silty glacial soils are somewhat similar in percentage composition. Owing to the latter consisting to so large an extent of such fine particles, it might have been supposed that decomposition and leaching would have affected them more . . . but the silty soils are loessial for the most part, and were formed from material blown by winds from glaciated areas and deposited where now found, or of material that has since been reworked by water. Minerals rich in alkalis and alkaline earths, being relatively easily crushed, would form a larger percentage of these silty soils than they do of the original glacial soils ; so that even if there has been a tendency to impoverish them by leaching, their originally greater richness enables the loessial soils to compare well with those strictly glacial." In Table 22 are reported the total amounts of phosphorus, po- tassium, magnesium, and calcium found by the Bureau of Soils in 2 million pounds of ignited soil for the surface soil of each of the 27 type soils, and also the amounts in the acid-soluble portion of one subsoil, or underlying greensand marl. While these soils " were selected to represent all classes — those of low, of medium, and of high productivity," Bulletin 54 gives no information as to the agricultural value of the different soils. Fortunately, the Annual Reports of the Bureau of Soils contain the descriptions made by the soil survey men concerning the common crops and normal crop yields produced on each of these soils, and thus a correlation is made possible between chemical composition (as recently determined by actual ultimate analysis) and produc- tive capacity, of these important and extensive types of soil (as reported in previous years from field investigations). Even here the student is advised not to accept opinions expressed, predictions made, or conclusions drawn, unless clearly supported by chemical facts or by long-continued agricultural experience. In each of the following descriptions the first paragraph is quoted from Bureau of Soils Bulletin 54 (December, 1908), and the second paragraph is quoted from the Annual Report of the " Field 138 SCIENCE AND SOIL <; U 88 H 0 O vo ^ O 0 O 0 O vo co O 00 rj- H M CO O 0 O O CM VO Tt- 1^00 l-l CM M 0 0 OO O CM l^. 00 vO CM o o o o CO O CM ro M M 0 O o o CM O O 00 •0 00 NO 00 <-O\O * H CM 0 CO •* M ("C o o M \O t--00 VO>O M o o o o o o O i-i O O O ON ON r^ co O t-» CM rt- 10 O CM co ro •o O Tj- H 0 0 t^ to ON O 0 O 0 \o t^oo CO >O ^t O 01 c> O O 10 Tj- O H O O O O O O (ITN vo VO\O CO CO ^- O M O\vo H o o \o o CM \O CM ^ •^ 'rt Q PH . c 13 S Q, . T3 rt en H fc H j 1 is ^s t^—1 •ss is 1) tn C CU 43 C 0« S.2 1) T3 U r- fehH Leonardtown loam . . Norfolk sand .... Norfolk loam .... Orangeburg sandy loam Crowley silt loam (prairie) Orangeburg fine sandy loar Shelby silt loam (prairie) Marshall loam (prairie) Marshall silt loam (prairie) Miami silt loam (timber lai Marshall black clay loam ( Marion silt loam (prairie) •^? •3J ^ x a C^ 3 — ~rt^ a, > g 3 '^x d li = 5 w .--S.5 a rt en ll| ^^> E rt _O > -o C 1 1) •-S ,u 1 ^7 t. t) '9 ^ t-, rt x&xL ^^ u •*-» *-* 'en en o-c ^c? % -° en ^5 0^ ^^ C -C fll • • s rt^S o. . .a •a^TS 3 •S-^l • 6 c >,6 o .If rt rt o, — 0 "u^ .^l- ^* (J Cy a c rt J5 u I 111?! s e 2 ^ ">- « -. T3 eu C & $2 a c o o bCW) "o "o OO ^-.^-v cc! rt 3 3 "rt 3 32 'en 'en Jfl 2 0 — •— S H < ^oi? ~ U 0 C C rt rt rt --^ —^ oo rt as O fe 'g & rt N*_X- rt "rt 13 "rt 13 13 J2 >. en en eu CD 3^3^-a *j en tn H .S >a C rs — ri Xa C C c c c c en en en en en en en en en en en en T3 _0 H ti S £ £ S cu M ^s 1 03 '&, rt ,rt rt 3 5 3 8 8 0 0 0 0 rt 3 13 13 eu (D D ^"' C T3 TJ u _ •3 2 CX &, QH rt rt rt DH &, PH "3 13 "d "rt "rt "rt "rt ~ci "rt rt cd ctj "rt d C rt rt 2 2 S o o o ooo CJUU OO oo OO ooo 0 C/} CO j j jCEO 00 .c^o 0 u-c 5' ^3 o . U u 8 n fe^'S hrt ' ' ' ^ '0 tu CJ LOCALITY lifornia, Fresno lifornia. Indio . :x3- >-, v_; „, rt a, cu ^§S *5'J3> asfi*5 131313 C C C rt rt rt (-1 »-• j-- ibama, Montgome uisiana, Acadia P xas, Nacogdoches ssouri, Shelby Co nnesota, Marshal sconsin, Janesville sconsin, Janesvilh O U C 0 go J >, u d So en" en" 11 lio, Toledo . lio, Toledo . . . lio, Wooster nnecticut, Conn. 1 £o § «" £S C/T M* ^) c/l a c ntucky, Scott Co. nnessee, Pikeville ibama, Dallas Co rth Carolina, Stal -ginia, Albemarle iryland, Hartford 6 S _CU _4) "rt "rt C/3 cV} ^ P^ CU D 22 e cu cu i — it— > £ £ OO ^^^ CH f* f^\ S ££ ANALYSIS BY THE UNITED STATES BUREAU 139 Operations of the Bureau of Soils " for the year designated. For convenient reference, the number of pounds of phosphorus shown in Table 22 is given at the beginning of the second paragraph in each description, not that the supply of the one element always correlates with productive power, but because it does so more frequently than any other. SOILS OF THE ARID REGION Fresno fine sandy loam (California) "is composed largely of silt to fine sand. It is locally known as 'white-ash land,' from its color and its physical character. The soil has probably been derived from volanic ash, but light- colored loams and sands have also contributed to it. The soil lies flat and works well, unless it be puddled, when water penetrates it slowly and hard clods or lumps form on drying. The lower subsoil is heavier, a blue clay being encountered at the depth of a few feet. Because of the poor drainage or light rainfall, this soil generally contains alkali. " (1830 Ib. P.) " Was originally considered extremely productive, and is now, where the drainage is good. Some of the first colonists settled on this land through choice. " (Report for 1900, page 46.) India fine sandy loam (California) "is made up of clay, silt, and the fine grades of sand. The clay is so flocculated that the soil in its field condition is lighter than the mechanical composition would indicate. The soil was mainly formed by erosion from adjacent mountains, the material being deposited in a bay or arm of the sea, but it has been greatly modified by wind action. It contains micaceous grains and minute shells. The soil ranges in depth from 2 z to 5 feet and is underlain with sandy loam or sand. The surface usually has a uniform slope and is generally well drained, but its high capillary power draws much water to the surface, causing an accumulation of alkali by its evaporation. In the lower levels the alkali is present in injurious amounts. Owing to insufficient rainfall, the salts are not washed out of this soil so well as might be expected from its physical character. " (2090 Ib. P.) "Where not too strongly alkaline, it will produce in abundance any of the crops suited to the climate." (Report for 1903, page 1255.) COASTAL PLAINS SOILS Leonardtown loam (Maryland) "consists of a yellow silty loam, fine and powdery when dry, but puddling to a plastic mass when thoroughly wet. The subsoil consists of a brittle mass of interlocking clay lenses, lumps, and fragments, separated by seams and pockets of medium to fine sand. This subsoil is as impervious as clay, owing to its peculiar shingle-like structure. It is an upland soil, and is generally slightly rolling. " (160 Ib. P.) "Covers about 41 per cent of St. Mary County. . . . This 140 SCIENCE AND SOIL soil has been cultivated for upward of two hundred years, but it is now little valued and is covered with oak and pine over much of its area. It is worth from $i to $3 an acre. The cultivated areas produce small crops of corn, wheat, and an inferior grade of tobacco." To this statement of facts is added the opinion that " the generally low estimation in which land is held is probably wholly unjustified. ... In texture, in chemical composition/ and in general agricultural value (when carefully and intelligently farmed) these lands compare favorably with the Hagerstown loam of western Maryland and Lancaster County, Pa., which are con- sidered the most valuable soils of the Atlantic States for general farm crops." (Report for 1900, page 33.) The Bureau of Soils also reports that 45,770 acres of this type of soil are found in Prince George County, which borders the District of Columbia on the east and south, concerning which the Bureau's Report for 1901 contains the following statements: "The soil is not adapted to tobacco, and has consequently been allowed to grow up to scrub forests, so that large portions of it are at present uncleared. Such unimproved lands can be bought for $1.50 to $5.00 an acre, even within a few miles of the District line. The soil has been badly neglected, and, when cultivated, the methods have not been such as to promote fertility. It is fre- quently acid, and needs lime and manure, or green crops turned under. When properly handled, as it is in a few places, good yields of wheat, corn, and grass are obtained." And to this statement of facts is also added the opinion that " upon the whole it is one of the most promising soils 2 of the local- 1 See Table 22 for chemical composition of the Leonardtown loam and the loam and clay of the Hagerstown series. — C. G. H. 2 Determinations of the water-soluble constituents in 36 samples of Leonardtown loam are included in the data which led Whitney and Cameron to draw the very erroneous conclusions that "practically all soils contain sufficient plant food for good crop yields, that this supply will be indefinitely maintained, and that the actual yield of plants adapted to the soil depends mainly, under favorable climatic conditions, upon the cultural methods and suitable crop rotations." (Bureau of Soils Bulletin 22, page 64.) The following quotations are taken from page 34 of Bureau of Soils Bulletin 22 (1903). They furnish some information as to what is done when this soil is "properly handled": "There is no apparent relation between the yield of crops and the soluble salt content of soils, even where the yields per acre differ as much as from 4 bushels to 25 or 30 bushels. ANALYSIS BY THE UNITED STATES BUREAU 141 ity, although it is not so considered by the resident farmers." (Report for 1901, page 45.) Norfolk sand (Maryland) "is a coarse to medium orange or yellow sand, having a depth of about 10 inches. The subsoil is coarse to medium, becom- "As bearing upon this point of the association or nonassociation of high analyti- cal figures with large crop yields, no more striking evidence occurs to us than the following letter written by Mr. Taylor, May 26, while in the field in St. Mary County, Md., and forming a part of his regular reports to the Bureau at Washington: "'At Park Hall, upon the farm of Mr. S — , who is recognized as one of the best farmers of the community, I secured some samples of the Leonardtown loam from a wheat field which will produce from 30 to 35 bushels per acre this season. The land was in tobacco last season, upon which barnyard manure and 400 pounds of fertilizer had been used. Nothing was added when the wheat was sown. The land was plowed about 8 inches deep. The soil lacked the usual grayish, ashy appearance of the Leonardtown loam, and, owing to cultivation, was loose and mellow to a depth of over two feet. One of these samples was compared with one taken from another wheat field upon the same type, where the yield would not be over 6 or 8 bushels. This latter land was farmed by negroes, was in wheat last year, and produced a fair crop, so it is said. No manure but a little guano was used last fall. The ground was uneven on the surface, and below the first 4 or 5 inches the soil was hard and compact. A comparison of the analyses is given below: PARTS PER MILLION OF OVEN-DRIED SOIL CONDITION PER CENT OF MOISTURE PHOSPHORIC ACID (P04) NITRIC ACID (NO.) CALCIUM (Ca) POTAS- SIUM (K) Good wheat: First foot 14.2 2.QO IT.. 22 14.27 24.^6 Second foot IQ.Q 3.72 lO.QI 12.^2 24.80 Poor wheat : First foot 14.7 A 72 ie T.A 7 01 7C.4O Second foot IQ Q A T,A ii 16 4 I1? 70.28 It will be noted that the poor soil shows more water-soluble plant food with all elements except calcium. Other data reported show that the pounds of water- soluble calcium per million pounds of oven-dried soil of the Leonardtown loam varied from 2.66 to 29.52 in the first two feet where the soil was in "good condition," and from 3.95 to 24.99 where the soil was in "poor condition." It appears, however, that the conclusions of Whitney and Cameron even con- cerning the nonrelationship between crop yields and water-soluble plant food are wrong. Professor F. H. King, a most careful investigator of the highest integrity, as the result of two years' experiments, including many determinations made during the crop season, before severing his connection with the Bureau of Soils, was led to the following conclusions: 142 SCIENCE AND SOIL ing loamy at about 3 feet. It is a common type of soil in the Atlantic and Gulf Coastal plains. The surface is level to rolling, and the soil is well drained." "Our own observations, published by the Bureau of Soils (Bulletin No. 26), have demonstrated that four good soils, observed to produce two and a half times the yield per acre of corn and potatoes that four poorer soils did under identical treatment, also gave up, when washed three minutes in five times their weight of pure water, 2.58 times as much plant food. Not only was there this difference in the amount of plant food carried in water-soluble form in the best and in the poorer soils, but the amounts of this same plant food taken out of like areas of field by like numbers and like kinds of plants during the same time was 3.2 times as great in the sap of the plants which gave the highest yields. " (Proceedings Jamestown Con- gress of Horticulture, 1907, page n.) The following tabular statement is a summary of Professor King's data secured under known conditions from the eight soils mentioned. (See Bureau of Soils Bulletin 26, page 120.) It should be stated that each value recorded for plant food determined is the average of 28 different determinations. These data are certainly far more trustworthy than the selected results from such miscellaneous samples as are referred to by Professor Whitney in Bulletin 22 (see above quotation). AVERAGE CROP YIELDS AND MEAN AMOUNT OF WATER-SOLUBLE PLANT FOOD IN FOUR POOR SOILS AND FOUR GOOD SOILS FOR THE SEASON OF 1903. — By F. H. King POOR SOILS STATE .... North Carolina Maryland Pennsylvania Wisconsin SOIL TYPE . . . fj e jr rolk Sand OJ eg TO O £|^ 1^ a^ 2S D o 11 n o o gerstown ly Loam g iS _u ^ e 0 g 1 *& c/5 o X «1 (/J It <°, rt--S W0 rt m i— i 9 GOOD SOILS CROP YIELDS PER ACRE: AVERAGE OF FIVE PLOTS FOR EACH SOIL Corn, bu. Potatoes, bu. . 36.3 47-7 38.9 70.4 29.6 102.7 29-5 93-4 33-6 78.6 64-3 213.2 52-9 168.0 54-7 iS7-o 80.4 290.5 69-3 237.1 POUNDS OF WATER-SOLUBLE PLANT-FOOD ELEMENTS IN 4 MILLION POUNDS OF SOIL: AVERAGE OF 28 DETERMINATIONS FOR EACH SOIL Nitrogen . . . 8-3 6.0 8.1 7-7 7-5 21-3 15.0 23-7 34-6 11.9 Phosphorus . . 10.4 10.8 15-2 11.9 12. 1 22.2 18.3 15.6 30.0 24.7 Potassium . . . 47-5 42.7 46.5 47.1 46.0 69-3 48.9 60.4 99-9 68.0 Magnesium . . 44-7 45-i 42.2 47-5 44-9 9I.I 78.0 69.1 115.2 IO2.I Calcium .... 91.6 114.9 94-5 100.8 100.5 264.8 264.4 223.4 293.6 277.7 Sulfur .... 62.3 71.9 50.2 83.1 66.9 156.9 121. 6 104-9 217.9 I83.I ANALYSIS BY THE UNITED STATES BUREAU 143 (520 Ib. P.) " Very poorly adapted to general farm crops, and little success is attained with either corn or wheat, and none of the grasses do well. " (Report for 1901, page 46.) Norfolk loam (Maryland) "is a mellow brown sandy loam to a depth of about 9 inches. The subsoil to 36 inches is a medium to heavy loam, which is often underlain by fine yellow sand. In the area from which the samples were taken it has a slight elevation and is gently rolling. It is usually well drained. " (610 Ib. P.) "The principal crop grown is wheat, which yields 20 to 30 bushels per acre on the heaviest phase of the type in fair seasons and from 15 to 20 bushels on the lighter areas, these yields depending largely on the amount of fertilizer used. ... It responds readily to applications of fertilizer and lime." (Report for 1903, page 172.) Orangeburg sandy loam (Alabama), "locally called 'red lands,' is a brown to a reddish brown light sandy loam, 4 to 15 inches deep, resting on a friable brick-red sandy clay subsoil. The surface is rolling. It is generally well drained, although there is a tendency to form a ' plow sole ' or ' hardpan.' " (520 Ib. P.) "Practically all of this type is under cultivation, and is highly prized for the production of cotton. The yields are not so high in some in- stances as on the Houston clay and other prairie types, but it is considered a safer soil from year to year than the prairie type. Cotton yields from one half to i bale per acre. As much as ij bales per acre has been produced where the land has been heavily fertilized. Very little corn is grown, as it is claimed that the yields are generally light. The difficulty here, as with the Orangeburg fine sandy loam, is that the soil proper is shallow." (Report for 1905, page 436.) In -the same report (page 438) the Orangeburg fine sandy loam, mentioned above, is described as follows: " Cotton is the principal crop grown on the Orangeburg fine sandy loam. The yields range from one fourth to i bale per acre, depending upon the amount of fertilizers used and the methods of cultivation. It is not considered a good corn soil, and, as a result, not much corn is planted. Corn yields range from 10 to 20 bushels per acre." Crowley silt loam (Louisiana) "usually has a surface of about 16 inches. It is of a brown color when wet, but ash gray when dry. It is composed of fine sand and silt, with sufficient clay to render it rather impervious. If stirred when wet, it puddles somewhat. This soil is underlain by a clay of mottled brown and yellow color, with brick-red streaks and blotches. The subsoil is highly impervious and the surface level, so that the soil has very poor drainage. The samples analyzed are from level prairies in southern Louisiana." (1220 Ib. P.) "From the time that the Crowley silt loam was first cul- tivated, rice has been the only crop to receive attention. Nothwithstanding this continued annual cropping with the same crop, without attempting in any way to maintain the productiveness of the soil, there has as yet been no de- crease in yields." (Report for 1903, page 470.) I44 SCIENCE AND SOIL Orangeburg fine sandy loam (Texas). "Varies in color, being red, brown, or gray. It is a light sandy loam, generally carrying iron concretions. The subsoil is red, friable, sandy clay. The type occupies the upland and has good natural drainage." (960 Ib. P.) " Cotton is the principal crop raised upon this soil. Yields from one half to three fourths of a bale per acre are secured. Corn does fairly well. " (Report for 1903, page 495.) GLACIAL OR LOESSIAL SOILS Shelby silt loam (Missouri) " is a silty soil of medium depth and of a light gray color when dry; dark gray when wet. It grades into a stiff, impervious silty clay, plastic and waxy when wet, friable and loamy when dry. The sub- soil is a dark mottled clay. It is level or gently rolling. The original growth on this type of soil was the prairie grasses. " (1920 Ib. P.) "The following yields are secured on this soil in good seasons: Hay, from 2 to 3 tons ; corn, from 35 to 40 bushels ; oats, 30 to 60 bushels ; wheat, 15 to 20 bushels, but uncertain; Kafir corn, 20 to 40 bushels; mil- let, 30 to 40 bushels of seed per acre. The Shelby silt loam is a typical grass soil." (Report for 1903, page 884.) Marshall loam (Minnesota) "is a somewhat heavy loam from 10 to 12 inches in depth and of a dark brown color. Under this is a stiff, sticky yellow subsoil to a depth of about 3 feet. Below this is a stiff bowlder clay, mottled yellow and gray. The type is generally rolling and well drained. Bowlders and glacial gravel occur to some extent over this soil." (1830 Ib. P.) "The Marshall loam is the safest soil in the area, as it is the surest to produce at least an average crop. . . . The Marshall loam, taken as a whole, excels all other soil types of the area in the production of wheat, on account of the superior quality of the grain produced." (Report for 1903, page 820.) Marshall silt loam (Wisconsin) "is a mealy, chocolate -colored silt loam with a dark brown tint when moist. It contains a large amount of silt, and becomes somewhat sticky when wet. It is about 10 inches deep. The sub- soil is a sticky, reddish-yellowish silty clay, about 3 feet deep, and rests upon a glacial gravel or the disintegrating limestone of the region. The soil probably owes some of its distinguishing characteristics to the influence of this limestone. The type is rolling and well drained. It was originally covered with the prairie grasses of the region. " (2450 Ib. P.) "It is one of the strongest and most fertile soil types of the region, forming the larger portion of the original rolling prairie of southern Wisconsin. It produces, under average seasonal conditions, from 50 to 60 bushels of corn, from 40 to 50 bushels of oats, about i£ tons of hay, and 1200 pounds of tobacco. " (Report for 1902, page 557.) Miami silt loam (Wisconsin) "is a very silty loam, light brown when wet, and light gray when dry. Its depth is about 8 inches. It is underlain by several ANALYSIS BY THE UNITED STATES BUREAU 145 feet of stiff, yellow, silty clay that is always mottled with gray, showing poor drainage and aeration. This type originally consisted mainly of timber lands and oak openings." (2100 Ib. P.) "The crop yields on the Edgerton (Miami) silt loam average from 45 to 50 bushels of corn per acre, about 40 bushels of oats, from i to i\ tons of hay, and from noo to 1200 pounds of tobacco." (Report for 1902, page 55 7-) Marshall black clay loam (Illinois) "is a heavy, somewhat sticky, granular clay loam, containing a large percentage of silt and organic matter. It has a depth of about 18 inches. The subsoil is a mottled yellow or drab-colored sticky, silty clay. This soil type has formed where the natural drainage was poor. The surface is level. In its original condition it was wet and swampy and required thorough drainage. " (2970 Ib. P.) "There are few soils more productive than the Miami (Mar- shall) black clay loam. Some areas have been cropped almost continuously in corn for nearly fifty years without much diminution in the yields, but the effect will undoubtedly be seen if the practice is continued much longer." (Report for 1903, page 787.) Marion silt loam (Illinois, gray silt loam on tight clay) "consists of a light brown to whitish very silty loam, containing very little organic matter. Its depth averages 12 inches. The soil cakes on drying, but breaks down into flourlike dust when pulverized. The subsoil is heavier, and contains more clay. It is so impervious to water as to be locally called hardpan. The lower subsoil is a hard, silty, mottled yellow clay, often containing iron concretions. Below 4 or 5 feet, more or less gravel is found. The type is level or slightly rolling. The soil has very poor natural drainage, owing to the rather impervious subsoil and the level surface. While of loessial origin, this soil has been largely formed from sandstones and shales ground up by glaciers." (1050 Ib. P.) "The average yield of corn is not much more than 15 bushels per acre. . . . The Marion silt loam is not a strong soil, and is not well adapted to general farming purposes. The small yield of corn indicates that it is not a good soil for that crop, although the profit from corn, according to many farmers, is as much as from other crops. " (Report for 1902, page 542.) Miami sand (Ohio) "is a coarse to medium loose and deep yellowish sand. It is underlain by a yellow sand of about the same texture. It is level to rolling, and consists of glacial material somewhat modified by wind action. It occupies elevated positions, and is well drained. " (2360 Ib. P.) "Grass, corn, wheat, truck, and fruit are grown on 'this soil. The quality of these is good, and in some cases better than the produce grown on the other soils in the area, but the yield is usually 15 to 30 per cent less, and crops sometimes are cut short or fail because of susceptibility to drought. The yield of wheat ranges from 10 to 20 bushels per acre, and of corn from 20 to 45 bushels per acre. . . . This soil yields from 75 to 120 bushels per acre of an excellent quality of potatoes. " (Report for 1902, page 394.) W abash loam (Ohio) "is a dark brown to black soil of good depth, and con- 146 SCIENCE AND SOIL taining a small proportion of the coarser grades of sand. The subsoil is a heavy brownish yellow loam overlying a fine gravelly loam. It is a bottom land, frequently occurring as terraces. It is generally well drained. It con- sists of glacial drift redeposited by stream action. " (1570 Ib. P.) "One of the more fertile soils of the area. Some of the fields, tilled for more than half a century and only moderately manured, still produce abundantly. . . . Corn yields from 40 to 100 bushels per acre, with the aver- age production probably about 75 bushels, and wheat from 20 to 35 bushels per acre." (Report for 1902, page 395.) Volusia silt loam (Ohio) "is a gray to brown silty loam with an average depth of 8 inches. The subsoil is a light yellow silty loam, mottled with gray in its lower portions. It has resulted in most part from the glaciation of shales. Its mechanical constituents closely resemble in size those of the soils derived from the loess, being composed largely of silt. This is doubtless due to the silt in the shales from which this soil type comes in large part. " (1480 Ib. P.) "The average yield of wheat is about 20 bushels per acre, and yields as high as 30 bushels are not uncommon. Corn, under the best cultural methods, will average 40 to 45 bushels per acre. Oats will yield an average of 50 bushels per acre, although larger yields are often reported. From 100 to 150 bushels of marketable potatoes per acre is the average produc- tion of this crop. " (Report for 1904, page 559.) Podunk fine sandy loam (Connecticut) "is an alluvial soil, formed by the reworking by running water of glaciated granites, gneisses, and schists. It contains an abundance of micaceous mineral particles visible to the eye. It is underlain by fine sand. The soil is of a dark brown color and is well drained. The tobacco field from which the sample came had been heavily fertilized for years. " (1920 Ib. P.) " The type is entirely under cultivation and produces good crops of corn, late truck, cucumbers for pickling, and tobacco. The area in the latter crop is large, and the yields range from 1700 to 1900 pounds in the open field." (Report for 1903, page 54.) RESIDUAL SOILS Oswego silt loam (Kansas) "consists of a dark gray silty loam, varying from very shallow to 10 inches deep, which grades into a stiff clay, becoming more impervious with depth. It becomes hard and compact on drying, but it is easily broken up into a mellow loam if plowed when in proper condition of moisture. This is an upland type, and occupies gently rolling prairies. Owing to the topography of the country, the type has good surface drainage. The Oswego silt loam is derived from the weathering of the underlying rock, this usually being shales, with occasional interbedded layers of sandstone and limestone. (1050 Ib. P.) "The Oswego silt loam is not a strong soil It is better adapted to wheat than to any of the other crops grown in the area, but, even with wheat, commercial fertilizer costing about $1.25 an acre is used on ANALYSIS BY THE UNITED STATES BUREAU 147 this soil, while none is deemed necessary on the other soils." (Report for 1903, page 897.) W abash silt loam (Kansas) "varies from 12 to 24 inches in depth and con- sists of a dark brown to black heavy silt loam. It is easily cultivated and readily kept in good tilth. The subsoil consists of a compact and rather heavy brown or yellowish silt loam. It occurs as long, narrow, tracts in the creek valleys and along the outer edges of the river valleys. The type occupies a rather low position in stream valleys and on gentle slopes. Its surface is nearly level or gently sloping. It forms first bottoms of smaller streams and second bottoms of larger ones. It is well drained naturally. The type has been deposited by water, the surface consisting largely of material washed from the surrounding hills, which are made up of shales and limestones. This wash from the hills is continually adding to the type. " (1140 Ib. P.) "Corn is the most important crop, and yields from 30 to 75 bushels per acre, 40 to 45 bushels being an average yield in ordinary seasons. Alfalfa, a very important crop on this type, yields 3 to 5 cuttings a year, and averages about i ton per acre for each cutting. The average an- nual yield is probably 3 or 4 tons of cured hay per acre. Wheat yields from 20 to 35 bushels per acre. . . . The land is cropped constantly, but as yet the yields have not diminished greatly, although no fertilizer and very little manure is used. The soil is naturally rich in organic matter, which may account for its continued productiveness. Corn is often cropped year after year on this type, and no system of rotation is used." (Report for 1906, page 932-) Hagerstown clay (Kentucky) "has a heavy texture, and varies from 3 to 12 inches in depth. It is yellow or brown in color. The subsoil is a heavy yellow clay, extending to a depth of 3 or more feet. This soil type is derived from limestones and shales. These rocks offer considerable resistance to disinte- gration, and the soil may therefore be more thoroughly leached than would be the case were the rocks more readily decomposed. The surface is rather rough, rounded hills being dominant features. Surface washing has been great, and the soil is generally shallow, the depth depending on its position. This is a residual soil, being formed from the breaking down in place of the underlying limestones and shales." (3490 Ib. P.) "Tobacco yields from 800 to 1200 pounds; wheat from 25 to 35 bushels; corn from 25 to 40 bushels; and hay from i^ to 2 tons to the acre. . . . On the stony phase of this soil the same crops are produced, but the yields are lower — tobacco, 500 pounds; corn, about 25 bushels; wheat, less than 12 bushels. . . . The Hagerstown clay is a good grain and grass land, but it is rapidly deteriorating from continuous surface washing. . . . Unless better methods are speedily adopted, this soil type will soon reach the condition of its stony phase, locally known as the ' barren limestone' land. " (Report for 1903, page 626.) Hagerstown loam (Tennessee) " consists of brown or yellowish brown mellow loam from 9 to 12 inches deep. It is underlain by a yellow to reddish yellow 148 SCIENCE AND SOIL stiff loam or light clay loam, which becomes a more pronounced red in depth. Traces of chert are found in both soil and subsoil. This type was formed by the slow weathering of limestones. In this soil the weathering has been so complete and the leaching so excessive that the lime of the disintegrated stone has been largely washed from the soil. The type has a moderately rolling sur- face, and has good surface drainage, but the subsoil is rather impervious. The underlying limestone comes near the surface in some places, owing to erosion. " (1050 Ib. P.) "The Hagerstown loam is all used in the extensive system of general farming which is practiced throughout the area. Corn yields from 15 to 30 bushels, with a probable average of 22 bushels per acre. Wheat yields from 5 to 20 bushels, with an average of 10 bushels, and the compara- tively small amount of hay which is grown yields an average crop of i ton per acre." (Report for 1903, page 584.) Houston clay (Alabama) "has resulted from the weathering of rotten limestones or chalks of Cretaceous time. Owing to its proximity to the soft and easily broken down lime rock, this soil is highly calcareous, and often contains lime concretions, especially in the subsoil. It may be considered to be of comparatively recent origin and as a residual Coastal Plains soil. The soil is a gray, brown, or black loamy clay, 6 inches deep. This is underlain with 3 feet or more of heavy gray or mottled yellow clay. The surface is gently rolling and the drainage very good. Agriculturally, the soil is lighter than would be expected, from its high clay content. This may be due to floccula- tion by the high percentage of lime present. " (5150 Ib. P.) "The Houston clay, while clodding badly when plowed too wet, and requiring care in its management, is a very strong and productive soil." (Report for 1905, page 464.) Cecil clay (North Carolina) "is found on uplands, gentle slopes, and roll- ing lands of the Piedmont Plateau. The Cecil clay is a residual soil, result- ing from the disintegration of a number of rocks, differing in mineralogical characters. Granites, gneisses, schists, and other somewhat similar rocks have contributed to the formation of this type, and so thorough have been the dis- integration and decomposition that the same red clay results from all. There is such a gradual change from soil to the parent rock that there is generally no sharp line between the two. The soil consists of a heavy red loam, contain- ing many sand grains of the original minerals forming the rocks from which the soil is derived. It is shallow, averaging about 5 inches. The subsoil is a stiff, tenacious red clay to a depth of 3 or more feet. It becomes heavier at greater depths. Natural drainage is fairly good, probably due to the sand and rock fragments contained in soil and subsoil. " (960 Ib. P.) "The soil is generally thin, but can be deepened by proper methods of cultivation and by green manuring. When so deepened, it assumes the properties of a heavy clay loam, and is very productive. It requires, however, considerable care and labor to maintain its fertility." (Report for 1901, page 55.) ANALYSIS BY THE UNITED STATES BUREAU 149 The average yields of corn, wheat, and oats are reported as 18, 12, and 20 bushels, respectively, per acre. In the description of this same type of soil for the Leesburg area of Virginia, the follow- ing statements were recorded by the field men of the Bureau of Soils: "The soil responds readily to applications of lime, and is much benefited by its use. Much commercial fertilizer, as well as lime and barn -yard manure, is used on this soil. In fact, so much acid phosphate has been added of late years that the land has become quite sour, and it is hardly possible to obtain a stand of grass or clover without the use of lime. " (Report for 1903, page 221.) Porter's black loam (Virginia) "is a loose, mellow black loam, averaging about 12 inches deep. The subsoil is slightly heavier and of a light brown to yellowish color. In depressions and coves, where wash from the higher ground has accumulated, there is no sharp distinction between soil and subsoil, the loose black loam being several feet deep. Both soil and subsoil contain frag- ments of the rocks whose decomposition has formed the soil — granites, gneisses, and schists. This type occurs principally in the coves of the Blue Ridge Mountains, but is also found upon the tops and upper slopes. " (4630 Ib. P.) "Locally the Porter's black loam is called 'black land' and ' pippin land,' the latter term being applied because, of all the soils in the area, it is preeminently adapted to the production of the Newtown or Alber- marle Pippin. This black land has long been recognized as the most fertile of the mountain soils. It can be worked year after year without apparent im- pairment of its fertility." (Report for 1902, page 210.) Chester mica loam (Maryland) "as its name indicates, is characterized by a great quantity of micaceous particles. It is derived from granites, gneisses, and other micaceous rocks over which the type lies. It is strictly a resid- ual soil and consists of a brownish loam 10 to 15 inches deep, underlain by a lighter colored, heavier loam, also containing mica. The surface varies from gently rolling to somewhat hilly. " (1130 Ib. P.) "It is not naturally a strong soil, but is susceptible of being made quite fertile and productive through intelligent tilling and manuring." (Report for 1901, page 222.) Collington sandy loam (New Jersey) "has resulted from the weathering of the greensand, or glauconite, of New Jersey. The subsoil, which comes within 6 or 8 inches of the surface, is a sticky, tenacious, claylike material, yel- lowish or greenish in color. Owing to its relations to the greensand deposits, this type differs from the other Coastal Plains soils. " (From 260 Ib. P. in surface to 27,600 Ib. P. in lower subsoil.) " Since millions of tons of this greensand marl have been employed as fertilizers, it is at once evident that any soil possessing a subsoil of this material will contain more than the ordinary amounts of potash and lime. When, in addition to this, its phys- ical structure is also well adapted to crop production, it would seem that a particularly valuable soil was formed. . . . The marl specimen was collected SCIENCE AND SOIL as a sample to show the amounts of plant foods in the material actually used as a fertilizer. The potash content is not high for a greensand marl, but the phosphoric -acid content is unusually high. The subsoil analysis (by acid digestion) reveals the fact that the lime, potash, and phosphoric acid of the original material have been extensively dissolved and removed, though fair amounts still remain. " (Report for 1901, page 139.) Unless otherwise stated, the above quotations from the soil survey field men and from Bureau of Soils Bulletin 54 refer specifi- cally to the areas in which the samples analyzed (Table 22) were collected. In general, there is very distinct correlation between the compo- sition of these extensive soil types and their natural productive- ness as recorded by the soil surveyors themselves some years be- fore the chemical analyses were made. It should be kept in mind that the data reported in Table 22 are for amounts in 2 million pounds of ignited soil, and are thus somewhat higher than, and not strictly comparable with, the results of analyses of the ordinary dry soil. It is important, also, to know that most of the 27 type soils described are found not only in the state and area in which these analyzed samples were taken, but are widely distributed throughout the respective formations, as the Coastal Plains, glacial areas, Piedmont, or other regions. Thus the 1903 Report of Field Operations of the Bureau of Soils mentions that Norfolk sand was found that year in New York, Delaware, Maryland, Virginia, North Carolina, Georgia, Florida, and Alabama; and Marshall black clay loam has been reported for Ohio, Indiana, Michigan, Wisconsin, Illinois, Iowa, Minnesota, South Dakota, and North Dakota. The Bureau of Soils includes in the one soil type (Marshall silt loam) the common brown silt loams of the Middle and Upper Illinoisan glaciations, of the Pre-Iowan, lowan, and Early Wiscon- sin glaciations, in Illinois, as well as soil in the Janes ville and Viroqua areas of Wisconsin, in the Grand Island and Staunton areas of Nebraska, and in the Jamestown area of North Dakota, with other areas in Colorado, Minnesota, Kansas, Missouri, Iowa, Indiana, and Louisiana; but it is apparent that the ultimate chemical composition of the soil is not considered among the characteristics required by the Bureau for a soil type. Thus the Marshall silt loam (brown silt loam) of the Middle Illinoisan glacia- ANALYSIS BY THE UNITED STATES BUREAU 151 tion contains 1170 pounds of total phosphorus in 2 million of dry soil (see Table 15), while 2450 pounds are reported in 2 million pounds of ignited soil of the Wisconsin area (see Table 22). Doctor Fraps finds 480 pounds of acid-soluble phosphorus in 2 million pounds of the Houston clay of Texas- from samples furnished him by the Bureau of Soils, while Table 22 shows 5150 pounds of total phosphorus in 2 million pounds of ignited Houston clay of Alabama. The Texas soil is evidently very deficient in phosphorus, but this is certainly not the case with the Alabama soil, which, it will be seen, outranks in phosphorus content every other soil reported in Table 22. The Bureau of Soils has not reported the ultimate chemi- cal analyses of different samples of the same type soil from different areas, so that it is impossible to make any such comparative study from the Bureau's data alone. In the author's opinion, the exact chemical data from which Table 22 is derived, and the careful descriptions given of the type soils analyzed, constitute the most valuable contribution of the United States Bureau of Soils to American agriculture. This absolute invoice of plant food, together with the description of physical properties, crop adaptations, and topographic features, furnishes a basis for the intelligent consideration of possible permanent and profitable systems of agriculture. Actual field experiments, to determine the rate at which the plant food can be made available, are lacking, and no report is made of the nitrogen content of the soils or of the limestone present or required. The percentages of " lime " (CaO) and magnesia (MgO) are given in Bulletin 54, but these signify little or nothing in relation to lime. Even the very acid Marion silt loam of Clay County, Illinois (gray silt loam on tight clay), is reported by the Bureau to contain .56 per cent of CaO (5.6 tons in 2 million pounds of soil), whereas it contains neither calcium oxid nor calcium carbonate, the calcium present existing usually in acid silicates. In general, the work of the Bureau of Soils has been directed toward a study of crop adaptation, in accordance with a somewhat prevalent notion that every soil is intended to grow some definite crop or crops, and that success will be attained if the proper crop is found for the special soil. While all must recognize that the natural adaptation of soil and crop is an important factor in many 152 SCIENCE AND SOIL cases, in the author's opinion it is a matter which has been given undue consideration in comparison with other extremely important factors. Even in the common practice of agriculture, soils at first well adapted to the growing of a certain crop do not remain so adapted. The fact is too well known to need illustration that specific crops are often grown with success for years finally to fail and be aban- doned for some other successful crops, which in turn finally give way to others. Thus good wheat land finally becomes poor wheat land, but still remains good for timothy hay, which in turn gives way to red top, and this may be followed by partial abandonment of the land for crop production. At any stage in this process of soil depletion, the land may be restored to its original power to produce wheat, by adopting the proper systems of soil enrichment. When land refuses longer to grow any crop which it has formerly produced with satisfaction and profit, the landowner should, as a very general rule, find out what the trouble is, and then proceed to remedy it; but, instead of meeting and overcoming such diffi- culties, the American farmer has literally run away from them; either by seeking newer lands or by adopting any other crop which the land would still produce. The most common staple crops can be grown on almost any soil if it is well drained, well watered, and sufficiently rich. Of course, the matter of crop adaptation must not be ignored, but if we would grow either plants or animals, we must not neglect the food supply. CHAPTER X CROP REQUIREMENTS FOR NITROGEN, PHOSPHORUS, AND POTASSIUM A STUDY of Table 23 is sufficient to make one familiar with the requirements of the more important crops of the United States for the three plant-food elements that are now recognized as having money values in commercial fertilizers. Information is also given regarding the amounts of these three elements in different parts of the crop, as in grain, straw, corn stover, and cotton stalks and lint, in order that it may be known with some degree of accuracy how much of each element is removed from the soil in crops and how much is sold from the farm in different kinds of farm produce. The ideal practice is to return to the soil, either directly or in farm manure, all plant food not sold from the farm. The data given in Table 23 are on the basis of pounds per acre for crop yields which are large, but which, when the best condi- tions are provided, have been and may be produced with very great profit, — yields that may well stand as ideals, desirable and pos- sible to be attained. Approximately proportionate amounts of plant food would be required for any other yields. Thus, if it is preferred to plan to make possible yields only one half as large, then the amounts given in Table 23 may be divided by two. (In Section 3 of the Appendix, data are reported showing the more com- plete composition of a much larger number of crops, but the re- sults there given are derived from a smaller number of analyses than are represented for the crops reported in Table 23; and, consequently, some differences are to be expected.) The value of the elements is computed on the basis of the present market prices for plant food from the most abundant natural deposits, delivered in car-load lots to central Illinois, and in suit- able condition for direct application to the land. Nitrogen in sodium nitrate 15 cents a pound. Phosphorus in ground raw phosphate ... 3 cents a pound. Potassium in kainit 6 cents a pound. 154 SCIENCE AND SOIL TABLE 23. FERTILITY IN FARM PRODUCE Approximate maximum amounts removable per acre annually PRODUCE BOUNDS MARKET VALUE Kind Amount Nitro- gen Phos- phorus Potas- sium Nitro- gen Phos- phorus Potas- sium Total Value Corn, grain 100 bu. IOO 17 19 ^15.00 $ -51 $ I.I4 $16.65 Corn stover 3T. 48 6 52 7-2O .18 3.12 10.50 Corn crop l . 148 23 71 22.20 .69 4.26 27-I5 Oats, grain 100 bu. 66 II 16 9-00 •33 .96 11.19 Oat straw . . HT. 31 5 52 4-65 •15 3.12 7.92 Oat crop 07 16 68 14.1;? .48 4.08 10. II Wheat, grain . 50 bu. y / 71 12 13 " J*J 10.65 »JM .36 T..V^ -78 y- ••• 11.79 Wheat straw . 2}T. 25 4 45 3-75 .12 2.70 6-57 Wheat crop . 96 16 58 14.40 .48 3-48 18.36 Soy beans . . 25 bu. 80 J3 24 I2.OO •39 1-44 13-83 Soy bean straw 2jT. 79 8 49 11.85 .24 2.94 15-03 Soy bean crop . i59 21 73 23-85 •63 4-38 28.86 Timothy hay . 3T. 72 9 7i IO.80 .27 4.26 15-33 Clover seed 4 bu. 7 2 3 1.05 .06 .18 1.29 Clover hay 4T. 160 20 1 20 24.OO .60 7.20 31.80 Cowpea hay . 3T. 130 14 98 I9-50 .42 5-88 25.80 Alfalfa hay 8 T. 400 36 192 60.OO i. 08 11.52 72.60 Cotton lint . . 1000 Ib. 3 0-4 4 •45 .01 .24 .70 Cotton seed 2000 Ib. 63 II 19 9-45 •33 1.14 10.92 Cotton stalks . 4000 Ib. IO2 18 59 15-30 •54 3-54 19.38 Cotton crop . . 1 68 29.4 82 25.20 .88 4-92 31.00 Potatoes . . 300 bu. 63 *3 90 9-45 •39 5-40 15-23 Sugar beets 20 T. IOO 18 i57 15.00 •54 9.42 24.96 Apples . . . 600 bu. 47 5 57 7-o5 •15 3-42 10.62 Leaves . 4T. 59 7 47 8.85 .21 2.82 11.88 Wood growth . rirtree 6 2 5 .90 .06 •30 1.26 Total crop . 112 14 109 1 6. 80 42 6 ^4 21 if) Fat cattle . . 1000 Ib. 25 Ait 7 i 3-75 •'+*• .21 w" JT- .06 fj /" 4.02 Fat hogs . . 1000 Ib. 18 3 i 2.70 .09 .06 2.85 Milk .... 10000 Ib. 57 7 12 8-55 .21 •72 9.48 Butter . . . 400 Ib. 0.8 0.2 O.I .12 .01 .01 .14 1 To this might also be added 1000 pounds of corncobs, containing 2 pounds of nitrogen, less than \ pound of phosphorus, and 2 pounds of potassium. CROP REQUIREMENTS 155 The figures given in Table 23 are based upon averages of large numbers of analyses of normal products, of which some have been made by the author and his associates, and many others by various chemists in America and Europe. These averages are trustworthy for large crops of good quality. Abnormal or special crops may vary considerably from these averages. Thus, we have high-protein corn and low-protein corn, one strain requiring 50 per cent more nitrogen, and somewhat more phosphorus, than the other (Illinois Bulletins 87 and 128); and it has been shown, for example, that alfalfa and cowpeas are not only much more productive, but also much richer in nitrogen, when grown on normal soils with the proper root-tubercle bacteria than without bacteria. On the whole, however, it is as nearly correct to say that a fifty-bushel crop of wheat requires 96 pounds of nitrogen and 16 pounds of phosphorus as it is to say that a measured bushel of wheat weighs 60 pounds. It may be said that other similar crops resemble somewhat closely those given in Table 23 as to plant-food requirements. Thus rye and barley are not markedly different in requirements from wheat and oats, considering equal yields in pounds of grain and straw. Other root crops may be compared with sugar beets, other grasses with timothy, hay from other annual legumes with cowpea hay, and other biennial and perennial legumes may be compared in a general way with red clover and alfalfa. How many years would be required to sell as much phosphorus from the farm in cotton lint yielding 2 bales (of 500 pounds each) per acre as in 4 tons of clover hay, which may be produced in the two cuttings in one season? Compare the nitrogen and potassium contained in 100 bushels of corn and in 20 tons of sugar beets. Compare wheat and clover in plant-food requirements. Assuming that two thirds of the nitrogen used by the clover plant is deposited in the tops and only one third in the roots, and that a given soil will furnish as much nitrogen to a growing clover crop as to a growing wheat crop, what is the effect upon the total nitrogen content of the soil of growing clover if all of the tops are removed ? Compute the cost of commercial nitrogen for a 5o-bushel crop of corn, assuming that 40 per cent of the nitrogen applied will be lost in drainage waters. CHAPTER XI SOURCES OF PLANT FOOD IF the productive capacity of American soils is to be maintained, elements of plant food which are present in such small amounts as to limit the crop yields even under good systems of farming must be returned to the soil as needed, and information is given in Table 24 to show the average quantities in pounds of the different valu- able elements of plant food contained in one ton of average fresh farm manure, rough feeds and bedding, and other fertilizer ma- terials. In computing the value of plant food in these materials, nitrogen is counted at 15 cents a pound and potassium at 6 cents a pound; while phosphorus is counted at 3 cents a pound in raw rock phos- phate, at 10 cents a pound in bone meal, and at 12 cents a pound in acid phosphate, these prices being based upon the usual average market values for the standard fertilizing materials in such quanti- ties as ought to be purchased by farmers, either singly or by two or more uniting. From the data given in Tables 23 and 24 it is a simple matter to compute the amounts of average manure or other fertilizers neces- sary to be applied to the land to replace the plant food removed in any rotation of crops. Observe, for example, that a four-year rotation, including corn for two years, oats with clover seeding the third year, and clover for hay and seed crops the fourth year, would require 39 tons of manure to supply the nitrogen, 41 tons to supply the phosphorus, 40 tons to supply the potassium, assum- ing the yields given in Table 23, and counting that the clover se- cures from the air as much nitrogen as is removed in the hay and seed crops. Observe that one ton of raw rock phosphate or one ton of steamed bone meal contains more phosphorus than 100 tons of average manure. Observe that 250 pounds of phosphorus can be purchased for $7.50 in ground natural rock phosphate, for $25.00 156 SOURCES OF PLANT FOOD in steamed bone meal, for $30.00 in acid phosphate, and for $65.00 in " complete " fertilizer. TABLE 24. FERTILITY IN MANURE, ROUGH FEEDS, AND FERTILIZERS NAME OF MATERIAL POUNDS PER TON MARKET VALUE PER TON Ni- tro- gen Phos- phor- us Po- tas- sium Nitrogen Phos- phorus Potassium Total Value Fresh farm manure . . 10 2 8 $ 1.50 $ .24 $ .48 $ 2.22 Barnyard manure l . . 10 3 8 I.SO •36 .48 2-34 Corn stover .... 16 2 17 2.40 .24 I. O2 3.66 Oat straw 12 2 21 1. 80 24 1.26 33O Wheat straw .... 10 2 14 I.SO •* *f .24 '.84 •ou 2.58 Clover hay .... 40 5 3° 6.00 .60 1. 80 8.40 Cowpea hay .... 43 5 33 6-45 .60 1.98 9-°3 Alfalfa hay .... 50 4 24 7-50 .48 1.44 9.42 Dried blood .... 280 42.00 42.OO Sodium nitrate . . . 310 46.50 46.50 Ammonium sulfate . . 400 60.00 6O.OO Raw bone meal . . . 80 1 80 I2.OO iS.OO 30.00 Steamed bone meal . . 20 250 3-00 25.00 28.00 Acidulated bone meal . 40 140 6.00 1 6. 80 22.80 Raw rock phosphate 250 7-5° 7-5° Acid phosphate . . . 125 15.00 15.00 Double superphosphate 400 48.00 48.OO Basic slag phosphate 1 60 16.00 16.00 Potassium chlorid . . 850 51.00 51.00 Potassium sulfate . . 850 51.00 51.00 Kainit 200 I2.OO I2.OO Wood ashes 2 . . . . 10 IOO i. 20 6.00 7-2O "Complete" fertilizer3 33 88 33 (?) (?) (?) 23.00 (?) 1 About two tons of fresh farm manure are required to produce one ton of com- mon barnyard manure six months old, with losses about as indicated. 2 Wood ashes also contain about 1000 pounds of lime (calcium carbonate) per ton. 3 This is the average composition and the average selling price of twelve brands of so-called "complete" fertilizer offered for sale in Illinois. Only 70 pounds of the phosphorus is guaranteed available, 18 pounds being insoluble. The cost of 88 pounds of phosphorus in raw rock phosphate would be $2.64. 158 SCIENCE AND SOIL If the element calcium becomes deficient in the soil (and it does in some cases), the most economic source is ordinary limestone; and, if magnesian limestone is applied, both calcium and mag- nesium are thus added to the soil. Kainit also supplies magnesium. Sulfur would be furnished in applications of acid phosphate, land-plaster, potassium sulfate, or kainit, as well as in magnesium sulfate and sodium sulfate, both of which are sometimes to be had as waste products or by-products. Iron sulfate (FeSO^ is a common by-product in certain manu- facturing processes, and strenuous efforts have been made from time to time to encourage its use as a fertilizer. Since numerous investigations have been conducted both in Europe and America to ascertain its fertilizing value, it is easily possible to select, from the many results thus secured, some few which indicate appre- ciable or even marked benefit. These results, however, have failed of verification. As a general average, iron sulfate produces less benefit than land-plaster, and sometimes detrimental effects are shown. A fair consideration of all results of carefully con- ducted experiments certainly leads to the conclusion that the use of iron sulfate as a fertilizer cannot be recommended in systems of soil improvement; although, like common salt (NaCl), it may sometimes produce a stimulating action sufficient to cover the cost where it can be secured at less expense than land-plaster, common salt, or other soluble salts. PART II SYSTEMS OF PERMANENT AGRICULTURE FOR practically all of the normal soils of the United States, and especially for those of the Central states, there are only three con- stituents that must be supplied in order to adopt systems of farm- ing that, if continued, will increase, or at least permanently maintain, the productive power of the soil. These are limestone, phosphorus, and organic matter. The limestone must be used to correct acidity where it now exists or where it may develop. The phosphorus is needed solely for its plant-food value. The supply of organic matter must be renewed to provide nitrogen from its decomposition and to make available the potassium and other essential elements contained in the soil in abundance, as well as to liberate phosphorus from the raw mineral phosphate naturally contained in or applied to the soil. Other fertilizer materials have some value, and sometimes great value, on uncommon or abnormal soils, and certain other substances are powerful soil stimulants, especially on soils deficient in organic matter; and, if applied with intelligence, they may sometimes be used temporarily with advantage and justification, but they are unnecessary and, as a very general rule, they are unprofitable, in good systems of soil improvement. There are, of course, numerous and more or less extensive areas of abnormal soils, such as the residual sands and the peaty swamp lands (both of which are very deficient in potassium), and also soils exceedingly rich in phosphorus, as in the geologic neigh- borhood of the natural phosphate deposits in the Central Basin of Tennessee and the Blue Grass Region of Kentucky. 159 CHAPTER XII LIMESTONE CALCIUM carbonate, in the form of chalk or marl, has been used for soil improvement since the beginning of agricultural history. Large use has been made of these natural materials in England and France, especially. An English record of 1795 mentions the " pre- vailing practice of sinking pits for the purpose of chalking the surrounding land therefrom," and states that " the most experi- enced Hertfordshire farmers agree that chalking of lands so circum- stanced is the best mode of culture they are capable of receiving." On the famous Rothamsted Experiment Station it has been found that the fields that had received liberal applications of this natural limestone a century ago are still moderately productive, while certain fields remote from the chalk pits which show no evidence of such applications are extremely unproductive. Director Hall of the Rothamsted Experiment Station states that many of the farmers in that vicinity are still reaping profitable crops from lands enriched by the heavy applications of chalk made by their ancestors many years ago. There appears to be no record that these easily pulverized lime- stone materials have ever been burned in order to increase their agricultural value. The productive power and durability of the natural limestone soils is indicated by the time-honored truth, " A limestone country is a rich country." Where such natural materials as chalk and marl have not been accessible, more or less use has been made of water-slacked or air- slacked lime; because, by burning and slacking, limestone rock may be reduced to powdered form and thus distributed over the land. With the development of rock-crushing and rock-grinding machinery, fine-ground natural unburned limestone can be had, and where this material can be gotten at reasonable cost, it replaces all other forms of lime used for the improvement of nor- mal soils. 160 LIMESTONE 161 In the " Georgical Essays " (1777 edition), we find an article by T. Henry, F.R.S., on the " Action of Lime and Marl as Manures," in which the following statements occur: "The lime, that we may come nearer to nature in our imitation, should not only be slacked, but be exposed to the open air, and often turned for several months, that it may recover its air; for it requires a long series of time be- fore it recovers the whole of which it has been deprived in calcination. . . . "I find that Doctor Home thinks that lime produces little effect on vegeta- tion till it is become effete. It may be known to have recovered its air by its no longer forming lime water, and by effervescing violently with acids without growing hot. If, however, the method described in the last note be used, it will be sufficient, if the lime be fallen, without waiting for the recovery of its air, as this point will be acquired during the long time which the mixture is to be exposed to the action of the atmosphere. . . . "Upon the whole, may we not conclude that lime, in most cases, is a stronger manure, when it has recovered the air of which it has been deprived in calci- nation, than it is when brought fresh from the kiln ; and that when procured for the purposes of agriculture, its efficacy and permanency will in general be increased, by mixing it, in its effete state, with the other ingredients which enter into the composition of marl?" When limestone is burned, the calcium carbonate (CaCO3) is decomposed, the carbon dioxid (CO2) passes off as a gas, leaving the product calcium oxid (CaO), which constitutes 56 per cent by weight of the limestone used. When exposed to the moisture of the air or soil, the quicklime (CaO) quickly takes up water and forms calcium hydroxid, Ca(OH)2, sometimes called hydrated lime, which means merely water-slacked lime. The product is the same whether the slack- ing (hydrating) is performed by the manufacturer at large expense, or by the farmer at little or no expense. When slacked lime is exposed in the air or soil, carbon dioxid is gradually absorbed, and the calcium carbonate is thus reformed. Thoroughly air-slacked lime is exactly the same material as fine- ground limestone. In other words, no matter what form of lime we apply to the soil, the benefit derived during the subsequent months or years is due to one and the same compound, calcium carbonate. These facts alone would be sufficient, perhaps, to lead one to use ground natural limestone in preference to the disagreeable caustic 1 62 SYSTEMS OF PERMANENT AGRICULTURE lime, but there are other facts worthy of the most careful consid- eration. Burned lime, whether fresh or hydrated, is known always as caustic lime. According to Webster's Dictionary, the word caustic means "capable of destroying the texture of anything or eating away its substance by chemical action." This definition well describes the action of caustic lime upon the organic matter of the soil. The lime breaks down the organic compounds and unites with the liberated carbon dioxid or other acid products. Not all of the reactions involved are understood, but the general effect is well known, and its long recognition in European countries has given rise to the proverbial expressions, "Lime, and lime without manure, Will make both farm and farmer poor," and " Kalk macht die Vater reich, aber die Sohne arm." (Lime makes the fathers rich, but the sons poor.) Caustic lime is not only a powerful agent in hastening the de- struction of organic matter, but it also has some power to increase the solubility of phosphorus and potassium, all of which may be of special help to legume crops; and if such crops are grown and removed from the land and the decaying roots and residues used as a further stimulant for the production of wheat, corn, or other crops, more rapid progress can be made toward land ruin than where no lime is used. On the other hand, even caustic lime can be used with profit if ample provision is made to replace the organic matter destroyed and also to restore the phosphorus (and potassium if necessary) removed in the crops. The caustic action of slacked lime on the skin or flesh is familiar to all, but a child can play in ground limestone as safely as in the soil of the garden. The chief reason, and usually the only justifiable reason, for applying lime to soils is to correct, or neutralize, soil acidity. The fermentation and decay of nearly all forms of organic matter is accompanied by the formation of acids, including carbonic acid, nitric acid, and various organic acids, such as the well-known lactic acid of sour milk, acetic acid of vinegar from apple juice, various LIMESTONE 163 acids in ensilage and sauerkraut, etc. Souring is usually the first stage in the process of decay of organic matter. Thus, there are two principal effects produced by applying lime to soils: one of these is to furnish a base for neutralizing the acids that may exist in the soil or that may form in such necessary processes as nitrification, and the other is a more active decomposi- tion or destruction of the soil itself, especially of its organic matter or humus content. To correct the acidity of sour soils is certainly a very desirable and profitable use of lime. Clover, alfalfa, alsike, cowpeas, soy- beans, and most other valuable legumes will not thrive on soils that are strongly acid. To be sure, such crops can be made to grow on acid soils by liberal applications of farm manure or other fer- tilizers, but the nitrogen-gathering bacteria of such legume plants do not properly develop and multiply in acid soils, and consequently the legumes do not have the power which they should have to accumulate large quantities of atmospheric nitrogen by means of the root-tubercle bacteria. Furthermore, the process termed nitrification by which the nitrifying bacteria transform the in- soluble organic nitrogen, in farm manure and plant residues, into soluble nitrate nitrogen, the form in which it becomes available as plant food, is greatly promoted by the presence of limestone and retarded by acid conditions. The use of some form of lime for correcting the acidity of soils, and thus encouraging nitrification and the growth of clover and other legumes with their wonderful power to enrich the soil in nitrogen, is certainly good farm practice. Any form of lime which is finely divided and can be thoroughly mixed with the soil will serve this purpose, whether it be ground limestone, marl, or chalk, or fresh-burned lime, water-slacked lime, or air-slacked lime. The one effect of lime, due to its basic property, results in a building-up process, through the increased growth of legumes and nitrogen-gathering bacteria; while the other effect, the decompo- sition of the soil, produced by its caustic property, is in all respects a destructive process, serving only to destroy humus and to liber- ate and reduce the stock of plant food stored in the soil. Whether this second effect is desirable, will depend upon the soil itself. On soils which are exceedingly rich in organic matter, such as peaty 1 64 SYSTEMS OF PERMANENT AGRICULTURE soils and other swamp soils, it would seem altogether rational to make temporary use of caustic lime to hasten the decomposition of the soil and consequent liberation of nitrogen, if such treatment is necessary, which is not usually the case. There may possibly be conditions under which soils contain large amounts of phosphorus and potassium which are too slowly available for profitable crop production, and in such cases it might be good farm practice for a time to make use of lime to hasten the liberation of these mineral elements of plant food. We should bear in mind, however, that this use of lime on a soil which is already deficient in nitrogen, or other plant food, only serves to still further exhaust the soil of its meager supply of these elements. Without a doubt, this is the most common condition and the most common effect of the continued use of caustic lime. It is true that the immediate effect is usually somewhat increased crops, but it should be borne in mind that when a farmer pays out money for lime to be used for this purpose, he is purchasing a stimulant which will ulti- mately leave his land in worse condition than before, especially in the loss of nitrogen and organic matter. Of course, the landowner must be governed somewhat by the cost of the material. As a rule, fine-ground limestone will be both the best and the most economical form of lime to use, wherever it can easily be obtained. If caustic lime be used, we should make special provision to maintain the humus in the soil by making even larger use of farm manure, legume crops, and green manures. It might be expected that burned lime would produce a greater increase in the crops for the first year or two than would be pro- duced by the ground limestone, more especially where the mineral elements, phosphorus and potassium, are not applied; for the reason stated, that ground limestone produces only the milder ac- tion, chiefly of correcting the acidity of the soil and thus encourag- ing the multiplication and activity of the nitrogen-gathering and nitrifying bacteria; whereas, .the burned lime not only produces this same effect, but it also acts as a powerful soil stimulant, or soil destroyer, attacking and destroying the organic matter and thus liberating plant food from the soil, usually resulting in more or less waste of valuable nitrogen and humus. The most extended investigation ever conducted relating to the LIMESTONE 165 use of burned lime and ground limestone in comparative tests is reported by the Pennsylvania Experiment Station (Report 1902). Four plots were treated with burned lime (slacked before being spread) at the rate of two tons per acre once in four years. Four other plots were treated with ground limestone at the rate of two tons per acre every two years. A four-year rotation was practiced, consisting of corn, oats, wheat, and hay, the hay being mixed timothy and clover, seeded on the wheat land in the spring. By having four sets of plots, each crop was grown every year. Seven products were obtained and weighed each year; namely, corn, corn stover, oats, oat straw, wheat, wheat straw, and hay. TABLE 25. PENNSYLVANIA EXPERIMENTS WITH BURNED LIME AND GROUND LIMESTONE Twenty Years' Produce per Acre CORN OATS WHEAT HAY, Grain Stover Grain Straw, Grain Straw (19 yr-) (Bushels) (Tons) (Bushels) (Tons) (Bushels) (Tons) None 819 18.8 678 14-3 279 13.2 24.9 Burned lime .... 699 I6.5 6l7 I7.8 3*8 14.6 23.6 Ground limestone . . 798 18.6 733 20.4 331 16.6 29.2 Thus, after twenty years' results had been obtained (1882 to 1901), the Pennsylvania Station reports data showing that with every product a greater total yield had been obtained from the plots treated with limestone than from those treated with caustic lime. Furthermore, with every product whose total yield for the last eight years was greater than the total yield of the first eight years, the limestone produced a greater increase than the caustic lime; and with every product whose total yield for the last eight years was less than the total yield of the first eight years, the decrease was less where limestone wras used than where caustic lime was applied (oat straw alone excepted). This is significant, in that it demonstrates the tendency of caustic lime with continued use to exhaust or destroy the fertility of the soil. In discussing these investigations, Doctor Frear of the Pennsylvania Station says: 1 66 SYSTEMS OF PERMANENT AGRICULTURE " In each case the yields with the carbonate of lime showed superiority under the conditions of this experiment over those following an equivalent application of caustic lime. " After these experiments had been in progress for sixteen years, the soil of each of the four plots in each test was sampled for analy- sis. The average nitrogen content for the four plots receiving ground limestone was found to be 2979 pounds per acre to a depth of 9 inches, while only 2604 pounds were found in the soil treated with caustic lime. This difference of 375 pounds of nitrogen is equal to the nitrogen contained in 37^ tons of farm manure. In other words, the data indicate that the effect of caustic lime as compared with ground limestone was equivalent to the destruction of 37^ tons of farm manure in 16 years, or more than two tons a year to the acre. Or, if we count the soil nitrogen worth 15 cents a pound (a fair market price), there is a liberation of more than $7.00 worth of nitrogen for every ton of burned lime used during the 1 6 years. The estimation of humus in these soils, based upon the determi- nation of organic carbon (multiplied by Wolff's factor, 1.724), showed the soil receiving limestone to contain 37.9 tons of humus per acre to a depth of 9 inches (counting 300,000 pounds of soil to the acre-inch), while only 34.2 tons of humus remained in the soil treated with caustic lime. If 4 tons of farm manure contain only i ton of dry matter (average fresh farm manure contains about 75 per cent of water), and if 2 tons of dry matter would be re- quired to make i ton of humus (when exposed to the weather, manure usually loses half of its dry-matter content within one year or less), then this difference of 4.7 tons of humus would be equal to 37.6 tons of fresh farm manure, which represents the loss from the destructive action of caustic lime as compared with ground limestone. During the 20 years, the land treated with ground limestone produced per acre 99 bushels more corn, 116 bushels more oats, 13 bushels more wheat, and 5.6 tons more hay, than the land treated with caustic lime. Counting 35 cents a bushel for corn, 30 cents for oats, 70 cents for wheat, and $6.00 a ton for hay, the value of the produce from the limestone treatment was $112.15 more than that from the land treated with caustic lime. The total ultimate LIMESTONE 167 effect of the caustic lime for the 20 years was an actual decrease in the yields of all crops except wheat; while the ground limestone produced an increase in all crops except corn, on which the de- crease was only one sixth as much as with caustic lime. If it is true, as indicated by the Pennsylvania experiments, that 8 tons of burned lime, applied during 16 years, released 375 pounds of nitrogen and destroyed organic matter equivalent to 37 tons of farm manure, or more than $7.00 worth of nitrogen and 4| tons of manure destroyed for each ton of burned lime used, as compared with ground limestone ; and if larger crops were obtained where limestone was used, especially where the practice is extended over several years, and if the ground limestone is sustaining the productive capacity of the soil much better than the burned lime ; then, as a very general rule, we should avoid applying caustic lime to the land, but make liberal use of ground limestone where needed to correct the acidity of the soil and to furnish a natural base, although, as used in these Pennsylvania experiments, without manure and with no return of plant food, the increase in crop yields produced by ground limestone has not been sufficient to pay for the heavy applications. The Maryland Experiment Station has recently reported experi- ments with different kinds of lime, covering eleven years, with a rotation of corn, wheat, and hay (timothy and clover), 1400 pounds of calcium oxid (burned lime) and equivalent amounts of calcium carbonate (ground oyster shells and shell marl) having been applied per acre at the beginning. Four crops of corn, three of wheat, and four of hay were harvested during the eleven years, with the follow- ing total results per acre: TABLE 25.1. MARYLAND EXPERIMENTS WITH LIME PRODUCE IN ELEVEN YEARS KINDS OF LIME USED Corn (Bushels) Wheat (Bushels) Hay (Tons) 4 Crops 3 Crops 4 Crops None . 08 32 2.6o Caustic lime burned from stone ' . . 128 32 3-°9 Caustic lime burned from shells * . . I2Q 34 3.82 Calcium carbonate in ground shells 148 42 3-97 Calcium carbonate in shell marl . . 145 43 4.29 1 Average of two plots. 1 68 SYSTEMS OF PERMANENT AGRICULTURE In commenting on these results, Director Patterson of the Mary- land Experiment Station says: " It will be noted that the car- bonate of lime gave decidedly better results than the caustic lime." Neuffer, of Heilbronn, Germany, has published a book entitled, " Das Kalksteinmehl im Dienste der Landwirtschaft " (The Use of Ground Limestone in Agriculture), in which he advises that ground limestone, and " not burned lime," should be used in the improvement of soils deficient in lime. Porter and Grant, in a recent Farmers' Bulletin issued by the Agricultural Department of the County Council of Lancaster, England, report experiments on manured and unmanured meadow lands, showing that ground limestone is more profitable as an application to grass lands than burned lime, and that it can be economically used on grass lands which are in need of lime. No trustworthy investigations support the use of burned lime in preference to ground limestone; although we have ample in- formation showing that on many soils a moderate use of burned lime in connection with a liberal use of farm manure and green manures yields profitable returns, which would, no doubt, be still more profitable if the burned lime were replaced with ground limestone. The most abundant impurity of limestone is magnesium car- bonate, which sometimes occurs in equal molecular proportion with calcium carbonate, in what is called dolomite (CaCO3 MgCO3). Limestones containing considerable amounts of magnesium car- bonate are also called magnesian limestones, even though the proportion of magnesium may be less than in dolomite. Dolomitic limestone is usually slightly heavier than ordinary limestone, and it is scarcely attacked by cold hydrochloric or acetic acid, while pure calcium carbonate is rapidly decomposed, the carbon dioxid being liberated as a gas. The molecular weights are 100 for calcium carbonate and 84 for magnesium carbonate, and consequently 84 pounds of the latter has the same power to correct soil acidity as 100 pounds of the former; or 92 pounds of dolomite will correct as much soil acidity as 100 pounds of pure ordinary limestone. To determine the amount of limestone present in the soil, or to determine the value of a sample of limestone for use on acid soils, LIMESTONE 169 it is usually sufficient to determine the content of carbonate carbon (or carbon dioxid) and compute from this the equivalent amount of calcium carbonate. Of course, this computation would show that 100 pounds of pure dolomite would be equivalent to about 109 pounds of pure limestone. Agricultural writers have placed upon record the general opinion that magnesian lime is very likely to produce injurious effects when used upon the soil. In his comprehensive and very valuable treatise upon "The Agricultural Use of Lime," Doctor William Frear includes the fol- lowing comments (Report Pennsylvania State College, 1899-1900, pages 14 to 176) : "Lloyd states that lime (CaO) is the only material of value in burnt lime and applies the adjective 'bad' to a lime containing 60 per cent of lime (CaO) and 30 per cent of magnesia (MgO). Low says of the magnesian limestone of England: 'If applied after being calcined, in the same quantity as other limes, it produces a temporary sterility, burning, as it were, the soil ; hence, it is termed hot lime and is applied in much smaller quantity than other kinds of lime. ' This action he attributes, after Sir Humphry Davy, to the longer period of causticity commonly supposed to occur with magnesia. In the form of car- bonate, he says, 'magnesia seems to exercise a highly favorable action; and magnesian limestone may perhaps be regarded as the most valuable of any, since a smaller quantity of it suffices for the ends proposed.' "The subject is quite fully discussed by Storer (Agriculture, 1897, Vol. 2, page 135), who notes that it was early observed by English chemists that certain limestones which had sometimes been found in practice to injure crops, contained magnesia, and that Tennant, on applying calcined magnesia to various soils with different crops, found that his plants either died, were un- healthy, or vegetated very imperfectly ; also, that Knop found, in growing plants by water culture (i.e., in very dilute solutions of plant foods), that magnesium salts are distinctly harmful unless accompanied by abundance of lime, potash, or ammonia salts; by themselves, the magnesium salts caused peculiar mal- formations of the plant roots, followed shortly by the death of the plants. "Storer notes, on the other hand, that Sir Humphry Davy found that the very magnesian limestones to which objection was made, gave very beneficial results on certain soils, and that magnesia, though injurious when present in caustic condition in considerable quantity in ordinary soils, may be beneficial when mixed with peat or where present as carbonate. " In a recent investigation at Rothamsted, Ashby reports a larger fixation of nitrogen by Azotobacter when magnesium carbonate was present than when calcium carbonate was used. For each 1 70 SYSTEMS OF PERMANENT AGRICULTURE gram of carbohydrate (mannite) consumed, 8.92 milligrams of nitrogen were fixed in 6.6 days with magnesium carbonate, and 5.80 milligrams in 4.6 days with calcium carbonate. Ashby says (Journal of Agricultural Science, January, 1907, pages 46, 47) : "With magnesium carbonate there was 50 per cent more nitrogen fixed, and a delay of two days in development. . . . One must conclude, therefore, the magnesium carbonate not only neutralizes more effectually than calcium car- bonate any trace of acidity due to foreign organisms in the early stages of culture, but also prevents butyric fermentation; but at first it inhibits the growth of Azotobacter itself. " Table 26 gives the results of an investigation concerning magne- sium carbonate conducted with the assistance of the author's students and associates at the University of Illinois. The experi- ment bears upon two lines of inquiry — (i) the value of magne- sium carbonate for soil improvement, and (2) methods of correcting this " alkali" when present in injurious amounts. Several series of 4-gallon pots were filled with the common brown silt loam prairie soil from the University farm, and to five of the six pots in each series was added magnesium carbonate in amounts varying from .4 per cent to 2 per cent of the dry soil. In addition, Series C and F received calcium sulfate in such an amount as to maintain the ratio of MgO to CaO = 4 to 7, in ac- cordance to Loew's advocated optimum ratio. After the crop of 1904 was harvested, the pots in Series F were thoroughly leached in order to remove magnesium, more or less of which was expected to react with the calcium sulfate, leaving the harmless calcium carbonate. The data recorded in Table 26 show a distinct and persistent benefit from the use of magnesium carbonate up to .8 per cent of the soil, while with 1.2 per cent the plants are very seriously injured and with 1.6 per cent they are usually so nearly killed as to produce no grain, and they are practically all killed with 2 per cent of mag- nesium carbonate. The application of i per cent of magnesium carbonate would require 10 tons per acre for the surface 6§ inches, but if the material were applied and mixed with only the surface inch by a light har- rowing, it would require only i^ tons per acre for i per cent. Since pure dolomite would contain only 46 per cent of magnesium LIMESTONE 171 TABLE 26. ILLINOIS POT-CULTURE EXPERIMENTS Magnesium Carbonate in Brown Silt Loam Prairie Soil POT No. " ALKALI " APPLIED AMEND- MENT APPLIED ADDITIONAL TREATMENT GIVEN YIELDS OF WHEAT GRAIN, GRAMS PER POT Kind Per Cent 1904 1905 1906 1907 1908 O. i9o8R. SERIES A: MAGNESIUM CARBONATE I None .0 None None 15-23 10.93 11.50 6.30 8.60 9.6l 2 MgCOs •4 None NC= ^(y 0 H— O\ /> >C=0 = Ca< K—0' ^O =0 H- 176 SYSTEMS OF PERMANENT AGRICULTURE In all humid regions where water passes through the soil, there is loss of calcium carbonate, leached out in the form of soluble bicarbonate. The " lime " carried in solution in " hard " waters from surface wells appears as a crust or scale in the teakettle, the soluble bicarbonate being decomposed by heat and the insoluble normal carbonate thus precipitated. Even virgin soils in old soil formations are often not only deficient in limestone, but they are sometimes found to be exceedingly acid, and thus require heavy applications of limestone to correct or neutralize the acids in the soil. Usually these soil acids exist in part, at least, as organic acids (humic acid etc.), but it is very evident that they are not always entirely organic, because the acidity often markedly increases while the organic matter decreases, with depth of soil, as will be seen from Tables 15, 16, and 17 (see soil types 330, 135, and 335), in which the measure of acidity is shown by the " limestone required," and the organic matter is indicated roughly by the nitrogen. Thus, in the Lower Illinoisan yellow silt loam, limestone required to correct the acidity increases from 310 pounds in the surface to 3315 pounds in the subsurface, and to 7200 pounds in the sub- soil, considering 2 million pounds of each; while the nitrogen de- creases from 2150 pounds in the surface to 1085 pounds in the subsurface, and to 827 pounds in the subsoil; and, as a matter of fact, the organic carbon decreases from 23,400 pounds in the sur- face, to 9710 pounds in the subsurface, and to 6190 pounds in the subsoil, 2 million pounds of each being considered. Detmer assigns to humic and ulmic acids the molecular formula C60H54O27, and to their salts such formulas as Ag8C60H46027, and Ca3(NH4)2C60H46O27. These would correspond to Ca4C60H46O27. On this basis, if all of the organic carbon in the subsoil were in the form of humic acid, it would be equal to less than one half of the acidity found. These computations are based upon the average of many analyses of soil samples from this type. Individual samples show as high as six times as much acidity as could be accounted for from the total organic carbon if in the form of humic acid. Acid silicates (see acid salts), formed from polysilicates (see under silicon), from which some basic elements may have been removed and replaced with acid hydrogen, by reaction with soluble LIMESTONE 177 organic acids, or possibly by the long-continued weak action of drainage waters charged with carbonic acid, do exist in the soil, and the evidence thus far secured indicates that they account for most of the acidity of soils that are at the same time strongly acid and very deficient in humus. Calcium bicarbonate may be formed by the action of carbonic acid on silicates containing calcium, even though no limestone is present. It is well known that plants have power to secure calcium, as plant food, from acid soils containing calcium in silicates but not containing limestone. Of course, it is not necessary to apply limestone to soils that already contain abundance of calcium carbonate, but it should be applied to soils that show acidity in the top soil and subsoil. Not infrequently slight acidity exists in the surface, and sometimes in the subsurface also, where the subsoil contains very large amounts of limestone. From present information we cannot strongly advise the application of limestone to such soils, although it would cer- tainly do no harm, and for some crops might be beneficial. But where the subsoil also is strongly acid, liberal applications of lime- stone should be made. While a small amount of aridity in the sur- face may not be a serious injury when the rainfall is abundant, there is apparently in humid regions some rise of acidity from strongly acid subsoils in times of partial drouth, corresponding somewhat to the "rise of alkali " in arid regions, where the water leaves the soil only by evaporation from the surface. If, however, the subsoil contains abundance of limestone, some calcium bi- carbonate will be brought upward into the subsurface or surface soil with the capillary rise of the soil moisture, and this will be left as normal carbonate when the water evaporates, and may serve to reduce the acidity of the subsurface or surface soil, at critical times, as in time of drouth. Clover and alfalfa are plants that are very sensitive to acid conditions when dependent for most of their nitrogen upon the bacteria, Pseudomonas radicicola, but these crops are grown very successfully upon such soils as the brown silt loam of the Early Wisconsin glaciation and the brown silt loam, yellow-gray silt loam, and yellow silt loam, of the Late Wisconsin; whereas, they are complete failures on the Lower Illinoisan gray silt loam prairie, 178 SYSTEMS OF PERMANENT AGRICULTURE and very unsatisfactory crops for all soils with strongly acid sub- soils, although, as already stated, such crops can be grown for a time on such soils if liberally fed with farm manure or other fer- tilizers. The legume plants, themselves, are not so sensitive to acid conditions, but, rather, the bacteria depended upon to furnish nitrogen; and while these will sometimes live and even form tuber- cles, they seem to develop but little power to fix nitrogen under such unfavorable conditions. THE TIME TO APPLY LIMESTONE The answer to this question can be no more definite than to a similar question concerning farm manure. We should consider the matter of hauling and spreading limestone in relation to the other necessary farm work, keeping in mind conditions of weather, roads, and land. It is applied but once during the crop rotation and for the benefit of all crops, although its most direct benefit is for the legumes, the other crops receiving large indirect benefit if the legume crops are returned to the soil. It is sometimes applied in winter or spring, but, as a rule, it is more satisfactory to apply it during the summer or early fall, when the land is dry, the roads are good, and the days are long. It is not best to apply it in intimate connection with phosphate, because the limestone will retard the availability of the phosphorus, although this effect is temporary, and in any case the two materials must ultimately become mixed if applied to the same land. The phosphate may well be applied with organic matter (manure or clover), mixed with the surface soil by disking, *and then plowed under, and the limestone may then be applied after plowing and well mixed with the surface soil in the preparation of the seed bed, where wheat and clover are to be seeded, or where corn is to be followed by oats and clover, the oats being disked in without re- plowing. Thus the limestone is well distributed in the first 3 or 4 inches of the soil where the atmospheric nitrogen enters and where the nitrogen-fixing bacteria do much of their work, while the phos- phate is mixed with the decaying organic matter in the next 3 or 4 inches of soil where the plant roots feed in large degree. Another good way is to apply the phosphate for corn and the limestone for wheat about three years later. LIMESTONE 179 METHODS OF APPLYING LIMESTONE No single method need be followed in applying limestone to the land, but it should be spread as evenly as practicable. This may be done by hand with a light shovel, either from the wagon or from small equal-sized piles placed at regular intervals. Thus, a pile of 100 pounds every 33 feet each way makes two tons to the acre. It can easily be thrown 16 or 18 feet with a shovel. A spreader made for the purpose of applying ground limestone or rock phosphate is very useful. There are some fairly satisfactory machines on the market at the present time. Several spreaders are manufactured that serve well for applying ashes, slacked lime, or other light materials, but most of them are not suited for han- dling such heavy materials as limestone and rock phosphate. The directions given below are similar to those published by the Ohio Experiment Station for a " home-made " spreader which any farmer can have made, and which is more satisfactory for spreading these heavy materials than some of the machines on the market. Make a hopper similar to that of an ordinary grain drill, measur- ing inside 8| feet or n feet long with sides about 21 inches wide and about 20 inches apart at the top. The sides may be trussed with f-inch iron rods running from the bottom at the middle to the top at the ends of the hopper. Let the bottom be 5 inches wide in the clear, and cut in it crosswise a row of diamond-shaped holes, 2 inches wide, 2\ inches long, and 4 inches apart (6 inches between centers). Make a second bottom with holes in it of the same size and shape as those of the main bottom, and so shaped that they will register. Let this second bottom slide loosely under the first, moving upon supports made by leaving a space for it above bands of strap iron 12 inches apart, which should be carried from one side to the other under the hopper to strengthen it. The upper bottom piece may be of about 8-inch sheet steel, and the lower one may be of smooth, seasoned hard wood, about i inch thick and 7 inches wide, reenforced with strap iron if necessary, and well oiled or painted. To this under strip, attach a V-shaped arm, extending an inch in front of the hopper, with a half-inch hole in the point of the V, in which drop the end of a strong lever, bolting 180 SYSTEMS OF PERMANENT AGRICULTURE the lever loosely but securely to the side of the hopper, and fasten to the top of the hopper a guide of strap iron, in which the lever may move freely back and forth. The object of this lever is to regulate the size of the openings by moving the bottom board. Make a frame for the hopper, with a tongue to it, similar to the frame of an ordinary grain drill. Get a pair of old mowing-machine wheels with strong ratchets in the hubs, and with pieces of round axle of sufficient length to pass through the frame and into the ends of the hopper, which are to be welded to a square bar of iron about if inches in diameter and the length of the inside of the hopper. The axles should be fitted with journals, bolted to the under side of the frame. Make a reel to work inside of the hopper by securing to the axle, 12 inches apart, short arms of |-inch by i-inch iron, and fastening to these arms four beaters of f-inch square iron, about an inch shorter than the inside of the hopper, the reel being so adjusted that the beaters will almost scrape the bottom of the hopper, but will revolve freely between the sides. The arms may be made of two pairs of pieces, bent so as to fit around the axle on opposite sides, and secured by small bolts passing through the ends and through the beater, which is held between them. The diameter of the completed reel is about 5 inches, and it serves as a force feed. Two pieces of oilcloth may be tacked to the bottom of the hopper, one in front and one behind, of sufficient width to reach nearly to the ground, in order to reduce the annoyance of the flying dust to man and team. Another piece may be buttoned across the top of the hopper in windy weather, if desired; but the dust of limestone or of natural phosphate is certainly no worse than the dust of the field. A sort of second force feed has been evolved from the extensive experience of Illinois farmers in building home-made machines: Two pieces of sheet steel, each about 6 inches wide and the length of the machine, are used as a V-shape bottom for the hopper, forming nearly a right angle at the lowest point. One piece is stationary and the other is given an endwise motion back and forth by means of a small wheel with a heavy rim waving in and out horizontally and running through a slotted piece firmly attached to the movable sheet steel. Two very small wheels LIMESTONE 181 forming the sides of the slot serve to reduce the friction, and a lever is arranged to throw this mechanism out of gear. One of the pieces of sheet steel is provided with an adjustment by means of which a crack is opened of any desired width, the entire length of the bottom. Thus the stone falls, not through holes or in streaks, but in a perfect broadcast. Several of these home-made machines are in use. The draft is more than with the reel alone, but they are undoubtedly more satisfactory than anything on the market. The cash expense for such a machine, aside from the mower wheels with axle and ratchets, has varied from less than $10 to more than $20, depending on cost of material and labor. Farmers with some mechanical skill hire only the necessary blacksmithing. HINTS ON SPREADING LIMESTONE (AND PHOSPHATE) In hauling and spreading limestone it is of first importance to save time and labor. As a rule, it is more economical to purchase both limestone and raw phosphate in bulk, and have it shipped in paper-lined box cars. Wetting will do no harm except to give trouble in spreading. Bags are expensive and easily damaged, and with tight wagon boxes the use of bags is wholly unnecessary. As a rule, the plan should be to haul the limestone direct from the car to the field, and spread it at once. Only two days are allowed to unload a car, although an extra day's car service costs only one dollar. With a haul of two miles or less, and with two men, one boy, and two teams, with three wagons and one spreader, 30 tons of ground limestone can be taken from the car and spread over 10 to 15 acres of land in two or three days, provided the roads and other condi- tions are favorable. One man is kept in the car loading the limestone into a wagon. The boy with one team hauls the loaded wagon to the field, leaves it there, and takes an empty wagon back to the car, hitching at once to the loaded wagon and leaving the empty wagon to be loaded. The other man and team remain in the field with the spreader, spreading one load while the boy is gone for the next. If an extra team is at hand, the man at the car may drive to meet the empty wagon and thus save some time. 182 SYSTEMS OF PERMANENT AGRICULTURE When spreading across a forty-acre field, the loaded wagon should either be hauled to the middle line of the field, or half of the loads should be hauled to one side and the other half to the oppo- site side of the field, using an extra wagon. The spreader hopper should hold at least 1000 pounds on the half -rod machines, or 1333 pounds on the n-foot machines, so that by driving 80 rods, the load will amount to at least two tons per acre. Starting from the middle of the field, one hopperful will spread to the side (40 rods) and back, when the spreader must be backed up to the wagon and refilled. Four such drives (320 rods) with the half-rod machine, or three drives (240 rods) with the n-foot machine, will spread a two-ton load over an acre. If the roads are good, two tons can be hauled at a load with a good team and wagon. If necessary to draw the loaded wagon to the middle line of the field, a four-horse team is provided by adding the spreader team. For making applications from one half ton to two or three tons per acre of limestone or rock phosphate, an arrangement of this sort is very satisfactory. For heavier applications one can go over the ground twice, or it can be spread by hand. For longer dis- tances, one or more additional teams are needed on the road. Where manure is to be. spread, rock phosphate may well be spread with it. The phosphate may be sprinkled over the manure from day to day as it is being made in the stall or covered feeding shed, or the manure spreader may be partly filled with manure, phosphate then being sprinkled on sufficient for the load, the load com- pleted, and then spread on the land. It should be kept in mind, however, that, if any leaching occurs after the phosphate is mixed with the manure and before the manure is spread on the land, some loss may ensue of the added phosphate; while if the phosphate is taken directly from the car and spread on the land where manure has been or is to be applied, it can later be plowed under with the manure with no danger of loss of phosphate. CHAPTER XIII PHOSPHORUS PHOSPHORUS is the only element that must be purchased and returned to the most common soils of the United States. Phos- phorus is the key to permanent agriculture on these lands. To main- tain or increase the amount of phosphorus in the soil makes pos- sible the growth of clover (or other legumes) and the consequent addition of nitrogen from the inexhaustible supply in the air; and, with the addition of decaying organic matter in the residues of clover and other crops and in manure made in large part from clover hay and pasture and from the larger crops of corn and other grains which clover helps to produce, comes the possibility of liberating from the immense supplies in the soil sufficient potas- sium, magnesium, and other essential abundant elements, supple- mented by the amounts returned in manure and crop residues, for the production of large crops at least for thousands of years; whereas, if the supply of phosphorus in the soil is steadily de- creased in the future, in accordance with the past and present most common farm practice, then poverty is the only future for the people who till the common agricultural lands of the United States. And this does not refer to the far-distant future only, for the turning point is already past on most farms in our older states and on many farms in the corn belt; and lands that have passed their prime with sixty years of cultivation will decrease rapidly in pro- ductive power and value during another sixty years of similar exhaustive farm practice. The world's supply of phosphorus exists in three principal sources: First are the supplies in the various soils, concerning which the reader of the preceding pages will have sufficient posi- tive knowledge for intelligent thought. Second are the natural beds of calcium phosphate, varying in purity from a few per cent, to as high as 80 per cent, of tricalcium phosphate, Ca3(PO4)2. 183 184 SYSTEMS OF PERMANENT AGRICULTURE Third are the extensive deposits of phosphatic iron ores con- taining more or less ferric phosphate, FePO4, the phosphorus being recovered in the slag produced in the conversion of pig iron into steel. About three fourths of the phosphorus taken from the soil by crops of corn, wheat, or other cereals, is deposited in the grain or seed, about one fourth remaining in the straw or stalks. If the grain is sold, three fourths of the phosphorus required for the crop is sold with it; and, likewise, when grain is bought and brought to the farm, a like proportion of phosphorus is brought with it. When crops are fed to animals, as a general average about three fourths of the phosphorus, three fourths of the nitrogen, and practically all of the potassium are returned in the manurial excrements. Thus, if sufficient grain is bought and fed, and if the manure is saved and applied to the land, the soil can be made richer in phosphorus year by year, and in most sections some instances can be found of farmers who succeed in maintaining or increasing the fertility of their soil by this practice. If they have the neces- sary knowledge and skill and material equipment and sufficient capital, they may feed stock for the open market, or if this is not profitable, they may produce pure-bred stock to sell at higher prices for breeding purposes. In any case, live-stock farming can never be permanently profitable to a large proportion of the farmers in a great agricultural country, because the world cannot live on meat and dairy products only, and the relative supply and demand always compels the sale of much grain from most farms. Conse- quently, this system of adding phosphorus to one farm by taking it from other farms must be of limited application; and live-stock farmers who feed only the produce from their own land gradually reduce the phosphorus of the soil at least by the amount sold in the animal products. A still more limited supply of phosphorus is secured for use in soil improvement by utilizing the bone meal prepared by the pack- ing houses. This, of course, also comes from the soil originally. It is made chiefly from bone scraps which have no value for other uses. The best bone is worth several times as much for the manu- facture of buttons, cutlery, toilet articles, etc., as for fertilizer purposes. Probably not more than one tenth of all the phos- PHOSPHORUS 185 phorus shipped off from American farms in animal products is returned to the soil in bone fertilizers. The mineral matter in bone consists chiefly of tricalcium phosphate, with a small amount of calcium carbonate. There is practiced more or less adulteration of bone fertilizers by admixture of raw rock phosphate or acid phosphate. There are three principal forms of bone meal offered for sale — (i) raw bone, (2) steamed bone, and (3) acidulated bone. Raw bone meal. Raw bone meal contains about 9 per cent of phosphorus, 4 per cent of nitrogen, and much organic matter, including more or less fat, which tends to retard decomposition. The most common application of bone or other ordinary commercial fertilizer is 200 pounds per acre. Since 9 per cent means 9 pounds per hundred, this application would amount to 18 pounds of phosphorus per acre, or one pound more than is contained in 100 bushels of corn. Since 200 pounds is one tenth of a ton, raw bone meal contains about 180 pounds of phosphorus per ton. Hence, the rule: To convert per cent into pounds per ton, double the per cent and add one cipher. It is always advisable to memorize pounds per ton and to think in those amounts, rather than in per cent. At 10 cents a pound for phosphorus and 15 cents for nitrogen, a ton of raw bone meal costs about $30, which is $18 for the phosphorus and $12 for the nitrogen. Nearly i| tons of raw bones are required to make one ton of steamed bones, the loss in weight consisting of fat, flesh, glue, and other organic matters rich in nitrogen. Steamed bone meal. Steamed bone meal contains from 12 to 14 per cent of phosphorus, and it should average at least 12^ per cent, or 250 pounds of phosphorus per ton, costing $25 at 10 cents a pound for the element phosphorus. Thus, 200 pounds of steamed bone per acre supplies 25 pounds of phosphorus, or two pounds more than is required for a hundred-bushel crop of corn (grain and stalks). By steaming bones the nitrogen is largely removed in the organic matter, only about .8 per cent, or 16 pounds per ton, being found in good steamed .bone, an amount within the legal limits of error in some fertilizer laws, and too small to justify consideration in the purchase price, especially when nitrogen can be secured from the inexhaustible supply in the air by using leg- 1 86 SYSTEMS OF PERMANENT AGRICULTURE umes in crop rotations. To supply sufficient nitrogen for a hun- dred-bushel crop of corn would require 9 tons of steamed bone meal, costing about $225. The phosphorus in raw bone and steamed bone exists in the form of the insoluble tricalcium phosphate, but because of the porosity and fine division of the bone particles and the presence of decom- posing organic matter in intimate contact with the extensive sur- face within the pores, phosphorus is liberated quite readily from bone meal, steamed bone being more active because of the removal of the fat and because it is usually more finely ground than raw bone. Acidulated bone meal. Acidulated bone meal (" acid bone ") is made by adding to a ton of bone meal sufficient sulfuric acid to convert a part of the insoluble tricalcium phosphate into the sol- uble monocalcium phosphate, or at least into the more readily available dicalcium phosphate. The bone meal thus treated is said to be mildly acidulated. As an average, it contains about 140 pounds of phosphorus and 40 pounds of nitrogen per ton. Much of the so-called " dissolved bone " sold in the fertilizer trade is made from phosphate rock, and this is no detriment to the product so far as the soluble portion is concerned, but the insoluble portion is more rapidly available if derived from bone than from rock. In the acidulated and most readily available form, phosphorus sells at about 12 cents a pound. Other bone products include bone black, dissolved bone black, and bone ash. Tankage from the packing houses varies from nearly pure bone to a high percentage of nitrogenous organic matter, including dried blood, meat, and mixed offal. Some further data will be found under nitrogen fertilizers. Three principal kinds of phosphorus fertilizer are derived from phosphate rock. These" are (i) the fine-ground natural rock, (2) acid phosphate, and (3) double superphosphate. Natural phosphates. Natural phosphate beds are widely dis- tributed over the earth, some of the most important deposits being in Tennessee, South' Carolina, Florida, and Canada, also in France, Belgium, Norway, Spain, and North Africa. The present annual production of the world amounts to about three million tons, of which two million tons are produced in the United States, about PHOSPHORUS 187 one million for home consumption, and an equal amount for exportation, chiefly to Great Britain, Germany, and other parts of Europe. It is estimated that the total phosphate deposits of the world thus far discovered will still furnish somewhere from 200 million to 500 million tons of high-grade phosphate rock. Some phosphate deposits have recently been found in Wyoming, Idaho, and Utah, and doubtless still other extensive deposits will be discovered in various parts of the earth; but, nevertheless, the world's total supply of high-grade phosphate is apparently very limited when measured by crop requirements, as evidenced by the enormous shipment of phosphate from America to Europe, despite the exten- sive and long-continued search by geologists for any undiscovered European deposits. (See also Appendix.) Facts worthy of careful consideration are that the Chilian gov- ernment derives a large revenue from export duties on sodium nitrate, from the world's greatest natural deposits of combined nitrogen, an element which the Chilian landowners can always secure, however, from the inexhaustible atmospheric supply; whereas, from the United States we are exporting half of our total production of phosphates with no restrictions, although we are thus shipping away from our lands the only element we shall ever need to purchase in order to maintain the fertility of our own soils. The laws of Norway greatly restrict the exportation of phosphate from that country. To restore to the soils of the United States the phosphorus removed by the corn crop alone, would require the annual applica- tion of our total annual production of phosphate rock, counting 23 pounds of phosphorus for a hundred-bushel crop of corn, and 2\ billion bushels as the average corn crop of the United States. The Florida phosphates are classed chiefly as hard rock and soft rock, the South Carolina phosphates as land rock and river rock; and the Tennessee phosphates as brown rock and blue rock. The quality is usually expressed as percentage of purity; that is, percentage of tricalcium phosphate. The South Carolina land rock is the lowest in phosphorus, averag- ing less than 50 per cent calcium phosphate, or less than 10 per cent of phosphorus. The South Carolina river rock and the Florida 1 88 SYSTEMS OF PERMANENT AGRICULTURE soft rock average 50 to 60 per cent pure; while the Florida hard rock and the Tennessee brown rock contain from 60 to 75 per cent of calcium phosphate. The Tennessee blue rock varies from less than 50 to more than 70 per cent, or from 200 to 300 pounds of phosphorus per ton of rock. The Florida soft rock contains chiefly phosphates of iron and aluminum, while in the other rocks the phosphorus is largely in the form of tricalcium phosphate. Aside from the deposits of high-grade phosphate, containing 45 or 50 to 75 or 80 per cent of calcium phosphate, there are known to exist very much more extensive deposits of lower grade phos- phates and phosphatic limestones containing from less than 10 per cent to 40 per cent or more of calcium phosphate, correspond- ing to from 2 to 8 per cent of phosphorus, or from 40 to 160 pounds of phosphorus per ton of rock. At present, these deposits have no market value, because, if the phosphate costs $4.00 per ton fine- ground and on board cars at the mine, and if the freight charges are $3.00 per ton, the freight on two tons of low-grade rock would amount to $6.00; while the delivered cost of one ton of high-grade rock supplying the same amount of phosphorus would be only $7.00, leaving but 50 cents a ton for the low-grade rock, which would barely pay for the expense of easy quarrying and grinding. As the supplies of high-grade phosphate become exhausted and prices advance, the lower grades will no doubt be utilized in this country as they are in Europe, where 35 to 40 per cent Belgian phosphate is now one of the chief commercial grades. About 62^ per cent calcium phosphate, or 12^ per cent of phos- phorus, is the average grade of the fine-ground natural rock phos- phate now used in Illinois, and to some extent in other states, for direct application to the soil in intimate connection with abundance of decaying organic matter, as farm manure, clover, or other green manures. In this form the element phosphorus costs the farmer about 3 cents a pound. The information thus far secured amply justifies the adoption of a system of farming in which fine-ground natural phosphate rock should be applied at the rate of 1000 to 2000 pounds per acre every three to six years, for three or four successive crop rotations, after which the application may be reduced one half, or to 200 pounds per acre for each year in the rotation, which would still PHOSPHORUS 189 insure a small increase rather than a decrease in the future years. More specific data concerning the use of raw rock phosphate, the results of the most careful experiments, and the comparative value of different forms of phosphorus are more fully discussed in the following pages, after some consideration of organic matter. Acid phosphate. Acid phosphate is the name of a manufactured product, not of a chemical' compound. Chemically, there are two acid phosphates of calcium, (i) the monocalcium phosphate, CaH4(PO4)2, and (2) the dicalcium phos- phate, Ca2H2(PO4)2. These chemical compounds, together with tricalcium phosphate and tetracalcium phosphate, phosphoric acid, phosphorus pentoxid, and phosphorus, itself, form a very impor- tant and interesting series. For the sake of simplicity and uni- formity, two atoms of phosphorus are given in each case, this being necessary in some cases : MOLEC- PHOS- NAME FORMULA ULAR PHORUS WEIGHT (Per Cent) Phosphorus P2 62 IOO.OO Phosphorus pentoxid P2O5 142 43 66 Phosphoric acid Hfi(PO4)o 196 -21 63 Monocalcium phosphate .... CaH4(PO4)2 234 26.50 Dicalcium phosphate Ca2H2(PO4)2 272 22.43 Tricalcium phosphate Ca8(PO4)8 3IO 20.00 Tetracalcium phosphate .... Ca3(CaO)(P04)2 366 16.94 Of these substances, tricalcium phosphate is the only one that occurs in nature. The element phosphorus takes fire when exposed to the air, two atoms of phosphorus uniting with five atoms of oxygen to form phosphorus pentoxid, sometimes called phosphoric oxid. This compound is the most powerful dehydrating agent known in chemistry, having power to abstract water from many other substances. It unites with water to form true phosphoric acid, H3PO4, or H6(PO4)2. This is one of the strong acids, and if it comes in contact with calcium carbonate, for example, it takes up one, two, and, finally, three bivalent atoms of calcium in place of the univalent hydrogen atoms, and thus forms acid monocalcium igo SYSTEMS OF PERMANENT AGRICULTURE phosphate, acid dicalcium phosphate, or the neutral tricalcium phosphate, respectively, carbonic acid being liberated, which promptly decomposes into water and the gas, carbon dioxid. Tetracalcium phosphate is thought by some to be the compound in which phosphorus exists in basic slag phosphate, being essen- tially tricalcium phosphate loosely united with the CaO group (see under basic slag phosphate). In the manufacture of commercial acid phosphate, the phosphorus material most commonly used in mixed commercial fertilizers, one ton of ground raw rock phosphate is treated with about one ton of sulfuric acid, and the resulting material consists chiefly of mono- calcium phosphate and calcium sulfate (land-plaster), together with all of the impurities contained in the original materials, and this mixture is the ordinary acid-phosphate fertilizer: Ca3(P04)2 + 2 H2S04 = CaH4(P04)2 + 2 CaSO4. This equation shows only the general reaction between the chemical compounds, tricalcium phosphate and sulfuric acid, but impurities are always present, and both the impurities and the calcium sulfate are included in acid phosphate, in which the phos- phorus is held chiefly in the water-soluble compound, monocalcium phosphate. The reaction may be expressed in two equations, the two molecules of sulfuric acid being added separately, thus showing dicalcium phosphate as an intermediate product. Small amounts of both dicalcium phosphate and tricalcium phosphate usually remain in acid phosphate, and a considerable part of the sulfuric acid used reacts with impurities which consist chiefly of silicates of the abundant metals, aluminum, iron, calcium, magnesium, potassium, and sodium. Sometimes calcium carbonate is among the impurities. As a rule, about one fourth of acid phosphate consists of phosphates (chiefly monocalcium phosphate), while three fourths consist of calcium sulfate and impurities. The readily available phosphorus in acid phosphate has a market value of about 12 cents a pound. This includes the phosphorus soluble in water and also that dissolved by ammonium citrate solution, which is sometimes called the " citrate-soluble " or the " reverted." The term reverted is properly applied to dicalcium PHOSPHORUS 191 phosphate formed from monocalcium phosphate by reaction with tricalcium phosphate: CaH4(P04)2 + Ca3(P04)2 = 2 Ca2H2(PO4)2. On long standing, this sort of reaction evidently takes place if an excess of tricalcium phosphate was left in the original product, and consequently the percentage of water-soluble phosphorus may be greater in fresh acid phosphate than in that which has been stored for some time, the dicalcium phosphate, or " reverted," being sol- uble in citrate solution, but not in water. If the raw phosphate rock contains 12 per cent of phosphorus, the acid phosphate made from it will contain about '6 per cent of phosphorus. The most common grade is known as 14 per cent acid phosphate, which the fertilizer agent would say means that the acid phosphate contains 14 per cent of " phosphoric acid," by which, however, is meant not 14 per cent of true phosphoric acid, H3PO4, but 14 per cent of phosphorus pentoxid, P2O5, which is equivalent to 6.1 per cent of the element phosphorus, corresponding to 122 pounds of phosphorus per ton of acid phosphate, which sells at about $15 a ton. Where 250 pounds of phosphorus cost $7.50 in fine-ground natural rock phosphate, the same amount of phosphorus will usually cost $30 in the two tons of acid phosphate.1 Double superphosphate. Double superphosphate consists chiefly of monocalcium phosphate, CaH4(PO4)2, and a moderate amount of impurities. It is richer in phosphorus than any other fertilizer material. It is made (i) by treating low-grade phosphate rock with an excess of sulfuric acid, by which true liquid phosphoric acid is liberated. This is leached out of the mass, and (2) this true phosphoric acid is applied to high-grade phosphate rock, thus: (1) Ca3(P04)2 4- 3 H2S04 = H6(PO4)2 + 3 CaSO4; (2) Ca3(P04)2 + 2 H6(P04)2 = 3 CaH4(P04)2. 1 Both acidulated bone and acid phosphate are sometimes called superphos- phate; and in England "super" (meaning literally over or higher) is the common term for acid phosphate, somewhat as photographers use the term "hypo" (mean- ing under or lower) for sodium thiosulfate, formerly incorrectly called hyposulfite of soda. 192 SYSTEMS OF PERMANENT AGRICULTURE By this means the impurities of the low-grade phosphate and the calcium sulfate formed in the first reaction are left behind, and the monocalcium phosphate is then formed as a condensation prod- uct, with only the impurities of the high-grade phosphate and a small amount of calcium sulfate made from the excess of sulfuric acid which is carried with the phosphoric acid. In practice, double superphosphate is made to contain about 20 per cent of the ele- ment phosphorus, corresponding to about 75 per cent of mono- calcium phosphate, and to 400 pounds of phosphorus per ton of product. This material is not made in the United States, but is produced to a considerable extent in Germany. It has advantage over ordinary acid phosphate in long-distance shipping, and it also permits the use of phosphate rock containing more iron and alumi- num than can be used for the manufacture of common acid phos- phate on account of the deliquescent properties of the products. Slag phosphate. Basic slag phosphate results as a by-product when pig iron, made from phosphatic iron ores and thus contain- ing considerable phosphorus, is converted into steel by the basic process in which an excess of lime is used. By proper methods a slag is produced which may contain about 8 per cent of phosphorus, or 1 60 pounds per ton. It is commonly held that the phosphorus is in the form of tetracalcium phosphate, Ca4O(PO4)2, whose struc- tural composition might be represented thus: Tetracalcium Tricalcium Phosphoric phosphate phosphate acid /°\ /°\ H-°\ Ca-O-7P=O Ca— cAp=O H— O^P=O Ca/0 /° H-° cl>° < XOX O H— (X Ca-cAP=O Ca*-(Ap=O H— O~P=O XCK ^0' H— CT Whether these formulas express the relationship of tetracalcium phosphate to tricalcium phosphate and to phosphoric acid, is not fully known, and it is even questioned if the phosphorus in basic PHOSPHORUS 193 slag exists in the form of tetracalcium phosphate. However, an excess of calcium oxid is present, and the phosphorus in slag under suitable conditions can be made available. No doubt, the lime produces some benefit for its own sake on certain classes of soil. In value, the phosphorus is rated at 10 cents a pound, the same as in bone meal. The iron ores from the Lake Superior region, which are used in the Illinois Steel Works, contain too small an amount of phosphorus to give value to the slag produced, but some phosphorus-bearing ores are used in Pennsylvania, and slag phosphate has been pro- duced and used in that state to a limited extent for several years. In Europe very large quantities of slag phosphate are produced and sold under the name of Thomas slag, although Jacob Reese, who for many years controlled the production in Pennsylvania, claimed priority over the European discovery. The conditions under which the different forms of phosphorus should be used are discussed in the following pages. CHAPTER XIV ORGANIC MATTER AND NITROGEN THE organic matter of the soil may be considered in two classes, active and inactive, although no very sharp line can be drawn between them. The most active organic matter consists of such substances as decaying plant roots and crop residues, green manures and animal manures, incorporated with the soil. These products decay rapidly in the soil and in the process of decomposition liberate not only plant food which they contain, including nitrogen, phosphorus, and potassium, but they also set free other decomposition products, such as carbonic acid, nitric acid, and organic acids, which have power to dissolve more or less additional plant food from the mineral part of the soil. The inactive, or less active, organic matter consists of the more resistant organic residue that remains after several years and that decomposes very slowly. If present in large quantity, its gradual decomposition may still supply sufficient nitrogen to meet the needs of good crops, although its power to liberate mineral plant food from the soil may not provide adequate supplies of available phosphorus, potassium, etc. Thus, we find that one soil may at the same time be richer in organic matter and less productive than another soil, even though the two soils are alike in other respects. Three tons per acre of fresh, actively decaying organic matter may be more effective for a year or two than thirty tons of old and less active humus. The term humus is not synonymous with organic matter. Humus includes only that part of the organic matter that has passed the most active stage of decomposition and completely lost the physical structure of the materials from which it is made, and has thus become, as a rule, thoroughly incorporated with the soil mass. 194 ORGANIC MATTER AND NITROGEN 195 It is the decay of organic matter, and not the mere presence of it, that gives " life " to the soil. Partially decayed peat produces no such effect upon the productive power of the soil as follows the use of farm manure or clover residues. DECAY OF ORGANIC MATTER A matter that has led to much confusion and misunderstanding is the common talk of " available plant food," as distinct from the total supply, whereas there is no line of distinction. The question as to the amount of available plant food contained in the soil at any given time is very insignificant in comparison with the ques- tion how to make plant food available. The plant food removed from the soil by a crop is not available when the crop is planted, but it must be made available during the growing season. Plant food is made available by chemical and biochemical processes, of which ammonification and nitrification are among those best understood. For the exact information we now have regarding these processes, we are indebted to the researches of Pasteur and Schlosing and Miintz of France, Winogradsky of Russia, Warington of England, and others. The nitrogen in the soil is almost entirely in organic compounds; that is, the nitrogen is united or combined with other elements, notably carbon, hydrogen, and oxygen, with small amounts of phosphorus, and sulfur, in the form of partially decayed organic matter; but plants cannot use these insoluble organic compounds of nitrogen occurring in the soil. There are at least three different kinds of microscopic organisms (called bacteria), and also three different steps, or stages, involved in the process of nitrification, the nitrogen being changed from the organic compounds, first into the ammonia1 form (NH3), second into the nitrite form, as Ca(NO2)2, and third, into the nitrate form, as Ca(NO3)2. During the process the nitrogen is separated from the carbon and other elements composing the insoluble or- ganic matter, and is united or combined with oxygen and some alka- line element to form the soluble nitrate, such as calcium nitrate, 1 Technically this first step (ammonification) is preliminary to, and not a part of, nitrification proper. 196 SYSTEMS OF PERMANENT AGRICULTURE which is one of the most suitable compounds of nitrogen for plant food. This, then, is the general process of nitrification (including am- monification and nitrification proper) , in which the ammonifying and nitrifying bacteria transform or transfer the nitrogen from insoluble organic compounds into soluble nitrate compounds in which it may serve as available plant food. Each specific class of bacteria performs a distinct function. Thus, the ammonifying bac- teria serve only to convert organic nitrogen into ammonia nitrogen; the nitrite bacteria (also called nitrous bacteria) serve only to convert ammonia nitrogen from ammonia or ammonium salts into nitrous acid (HNO2) or nitrites; and the nitrate bacteria (also called nitric bacteria) serve only to convert nitrous acid or nitrites into nitric acid (HNO3) or nitrates. While we may assume that the nitrogen passes through the forms of nitrous and nitric acid, those acids are never present in detectable quantities, the presence of a salifiable base being essential for the progress of these biochemical reactions. The final product is al- ways a nitrate, except under artificial conditions, when nitrites may be obtained in quantity in the absence of the nitrate bacteria. Under the natural conditions existing in normal soils, even nitrites can scarcely be detected, because of the quickness with which they are converted into nitrates. The nitrate that is formed may be calcium nitrate, magnesium nitrate, potassium nitrate, sodium nitrate, or even ammonium nitrate, depending upon which base is present in the most suitable form. In the nitrification of ammonium carbonate, (NH4)2CO3, the reaction will stop when only one half completed if no other base is present, the final product being ammonium nitrate, NH4NO3. (NH4)2 CO3 + 3 O = NH4NO2 + CO2 + 2 H2O. NH4NO2 + O = NH4NO3. To continue the process beyond this point would require the formation of appreciable amounts of free nitric acid, of which the bacteria seem incapable. In the production of lactic acid in the souring of milk, the lactic bacteria are capable of continuing the ORGANIC MATTER AND* NITROGEN 197 process until the solution contains about .7 per cent of free lactic acid, beyond which they become inactive; but, if the free lactic acid is neutralized by the addition of some base, the bacteria again become active. In the process of nitrification there is required, not only the presence of calcium or some other alkaline element or group, in suitable form (as in carbonates) , but also a good supply of the ele- ment oxygen; for calcium nitrate contains but one atom of calcium (Ca) with two atoms of nitrogen (N)2, and six atoms of oxygen (O3)2, in each molecule, as indicated in the formula Ca(NO3)2. Magnesium nitrate, Mg(NO3)2, potassium nitrate, KNO3, and all other nitrates also contain oxygen. The supply of oxygen for the formation of nitrates in the process of nitrification comes from the air, and, aside from the killing of weeds, one of the most important effects of cultivation, or tillage, is that it permits the air more freely to enter the soil, and thus promotes nitrification. Another absolute requirement for the process of nitrification is the presence of phosphorus and probably of other mineral food supplies necessary to the growth and multiplication of the bacteria themselves. It is known that without phosphorus there can be neither growth nor life. These minute forms of plant life do not utilize the carbon dioxid of the air by means of the sun's energy; but they derive energy from the oxidation of the nitrogen com- pounds, and by means of this energy they are able even to decom- pose carbonates, if necessary, and to derive their supply of carbon from this source for the formation of their own organic bodies; but for all of this the mineral plant food must be supplied. (As a rule, the carbohydrates furnish the necessary carbon for bacterial growth.) An important consideration in this general connection is the fact that in the conversion of sufficient organic nitrogen into nitrate nitrogen for a hundred-bushel crop of corn, the nitric acid, if formed, would be alone sufficient to convert seven times as much insoluble tricalcium phosphate into soluble monocalcium phosphate as would be required to supply the phosphorus for the same crop. While this specific reaction could not occur in quantity, because the acid monocalcium phosphate would prevent nitrification, the suggestion is of interest in that it affords a quantitative comparison 1 98 SYSTEMS OF PERMANENT AGRICULTURE between one of the decomposition products of organic matter and the process of making insoluble plant food available, thus: Ca3(P04)2 + 4HN03 = CaH4(PO4)2 + 2 Ca(NO3)2. In accordance with this equation, 56 parts of nitrogen are equiva- lent to 62 parts of phosphorus in the reaction; whereas, when measured by the requirements of the corn crop, 56 parts of nitro- gen are equivalent to less than 9 parts of phosphorus, or only one seventh of 62. Even though the nitric acid may be at once neutralized by reac- tion with calcium carbonate, it is known that the liberated carbonic acid exerts an influence in the conversion of insoluble phosphates and potassium salts into soluble compounds. Of course, the quantity of organic acids and carbonic acid other- wise produced in the decay of organic matter is many times as great as that of nitric acid. Recent investigations of Hall, Miller, and Gimingham (Pro- ceedings Royal Society, 1908) seem to prove that nitrification proper does not occur in acid soils, and that crops growing on such soils must take up their supplies of nitrogen in the form of ammonium salts, formed in the process of ammonincation. It is shown, however, that there may be a small amount of nitri- fication in soils that are, on the whole, acid, but which contain here and there particles of calcium carbonate within whose limited sphere of influence the soil is alkaline and nitrification takes place. Under certain abnormal conditions, as under a slime or scum from sewage which prevents access of air, some denitrification may occur. In this process the denitrifying bacteria may even decompose nitrates in order to secure oxygen, and the element nitrogen may be liberated as free gas. Such loss may readily occur in the decay of manure in piles, but in normal soils there is prac- tically no denitrification. METHODS OF SUPPLYING ORGANIC MATTER There are three general methods of supplying organic matter to the soil in practical agriculture: (1) By green manures and crop residues. (2) By accumulations in pasturing. (3) By applications of farm manure. So much has been said and written regarding the value of farm manure that it is common talk that the manurial value of the food is almost wholly recovered in the manure; and there is even a vague notion in the minds of some that the manure is worth more for soil improvement than is the food from which the manure is made; while it is very generally believed that pasturing land increases the fertility of the soil. The fact is that the most important and least appreciated method of maintaining or increasing the supply of organic matter in the soil is by the use of green manures and crop residues. This is best understood by considering the digestibility of common food stuffs and by applying mathematics to the data (see Table 29). TABLE 29. AVERAGE DIGESTIBILITY OF SOME COMMON FOOD STUFFS FOOD STUFFS PER CENT DIGESTED OF TOTAL IN FOOD DRY MATTER OF FOOD RECOVERED IN MANURE Dry Matter Nitrogen Per Cent Pounds per Ton Pasture grasses 71 66 67 61 61 60 48 43 60 63 79 64 70 91 61 70 67 81 57 62 74 3° ii 45 42 52 49 78 76 79 29 34 33 39 39 40 52 57 40 37 21 36 3° 9 39 580 680 660 780 780 800 1040 1140 800 740 420 720 600 180 780 Red clover, green Alfalfa, green Mixed meadow hay Red clover hay Alfalfa hay Oat straw Wheat straw Corn stover Shock corn Corn-and-cob meal Corn ensilage . . Oats Corn Wheat bran 200 SYSTEMS OF PERMANENT AGRICULTURE Thus, when pasture grasses containing one ton of dry matter are eaten, only 580 pounds of the dry matter consumed will be re- turned to the land in the droppings; and the manure made from one ton of dry clover hay contains only 780 pounds of dry matter instead of the 2000 pounds taken from the field. In other words, a ton of clover plowed under will add nearly three times as much organic matter to the soil as can possibly be recovered in the ma- nure if the clover is fed. In the case of oat straw, about one half is digested and one half recovered in the manure, while only one tenth of the dry matter of corn is found in the manure. It must be kept in mind, furthermore, that to return even these proportions of organic matter to the land requires that the manure shall be applied to the soil before losses occur by fermentation and decay. The Maryland Experiment Station allowed 80 tons of manure to lie in an uncovered pile exposed to the weather for one year, during which time the amount was reduced to 27 tons. Professor Shutt, Chief Chemist for the Experiment Stations of the Dominion of Canada, exposed two tons of manure containing 1938 -pounds of organic matter, from April 29 to August 29, four months, during which time the organic matter was reduced by fermentation and decay to 655 pounds. During the same time the nitrogen was reduced from 48.1 pounds to 27.7 pounds. In ordinary farm practice more or less loss of organic matter is almost certain to occur unless the manure is applied to the soil within a day or two after it is produced. Because the nitrogen of the soil is contained in the organic matter and must be applied in that form in general farming, and because the figures are available, the per cent of nitrogen digested is shown in Table 27 for the common food stuffs named. The fact that 62 per cent of the nitrogen in red clover hay is digested means that only 38 per cent of the nitrogen in the food consumed will be recovered in the solid excrement. If the food ration consists of equal parts of corn and clover hay, the solid excrement will contain 31 per cent, or less than one third, of the nitrogen in the food. Of the remaining 69 per cent, about one third will be retained by the animal (or secreted in milk) and two thirds excreted in the liquid manure, as a general average in live-stock farming for animal ORGANIC MATTER AND NITROGEN 2OI products. Mature work animals excrete practically as much nitrogen as they consume. These facts certainly emphasize the importance of saving all liquid manure and the danger of loss of nitrogen in that form. In a series of digestion experiments (not yet published) con- ducted by the Illinois Experiment Station, with six milk cows, dur- ing a period of 15 days, the average daily consumption of food per cow was 19.67 pounds of dry matter contained in a ration of clover hay, corn silage, and mixed concentrates, including corn, oats, wheat bran, gluten meal, and linseed meal. The total dry matter recovered in the dung and urine amounted to 8.1 1 pounds, or 41.23 per cent. (With heavy feeding the digestibility is appreciably less than with lighter feeding.) Of the nitrogen consumed, 80.32 per cent was recovered in the dung and urine, and 20.12 per cent in the milk, indicating a slight loss from the animal bodies. Of the phosphorus consumed, 73.34 per cent was recovered in the manures and 22.28 per cent in the milk, only 4.38 per cent being retained by the animals. Of the potassium taken in the food, 76.02 per cent was recovered in the manures and 13.69 per cent in the milk, the balance, 10.29 per cent, probably having been largely excreted through the skin. The experiments were conducted the last half of June. (Consider- able amounts of commercial potassium salts were once regularly obtained from the washing of sheep wool.) Table 30 shows these results in more detail for ready comparison: TABLE 30. PLANT FOOD RECOVERED FROM FOOD CONSUMED BY MILK Cows Illinois Experiments: Average of 90 Days PLANT-FOOD ELEMENTS RECOVERED IN MILK RECOVERED IN DUNG RECOVERED IN URINE NOT RE- COVERED IN TOTAL MA- NURES Nitrogen, per cent 2O.12 T.Z.%6 44.76 10.68 Phosphorus, per cent 22.28 72. 33 I.OI 26.66 Potassium, per cent I?. 60 l6.7O ZQ.1,2 23.08 As an average of the best three cows, the plant food not recovered in the total solid and liquid manures was 25.03 per cent of the 202 SYSTEMS OF PERMANENT AGRICULTURE nitrogen, 28.07 Per cen^ °f the phosphorus, and 28.45 Per cent °f the potassium, of the food consumed, the differences being prac- tically accounted for by the larger amounts of milk produced by the best cows. However, the poorest cow of the six, in milk produc- tion, digested, during the three successive 5-day periods, 47.18 per cent, 44.77 per cent, and 52.54 per cent, respectively, of the total phosphorus consumed in the food; or, as an average of the 15-day period, only 52.94 per cent of the phosphorus taken in the food was recovered in the dung and urine from this cow. The Pennsylvania Experiment Station (Annual Report for 1899- 1900, page 321) reports digestion experiments with two milk cows during a period of 50 days, with the results shown in Table 31. The rations fed were three fifths mixed clover and timothy hay, and two fifths concentrates, including corn meal, buckwheat middlings, cotton-seed meal, and linseed meal. TABLE 31. PLANT FOOD RECOVERED FROM FOOD CONSUMED BY MILK Cows Per Cow for 50 Days: Average for 2 Cows: Pennsylvania Experiments NOT RECOV- PLANT-FOOD ELEMENTS CONSUMED IN 50 DAYS RECOVERED IN MILK RECOVERED IN DUNG RECOVERED IN URINE ERED IN TOTAL MANURES Nitrogen, pounds . . 67.96 IJ-39 21.46 36.07 10.43 Phosphorus, pounds . 9-73 2.06 6-75 •13 2.8S Potassium, pounds 37.68 3-76 5-93 28.38 3-37 Nitrogen, per cent . . TOO 16.76 3i-53 53-0° I5-36 Phosphorus, per cent . IOO 21.17 69-37 1-34 29.29 Potassium, per cent . 100 9.98 iS-74 75-32 8-94 In the Pennsylvania experiments both the nitrogen and potas- sium are slightly more than accounted for, but 8.12 per cent of the phosphorus in the food consumed was retained by the animals, probably in part for the formation of bones in unborn calves. As an average of both the Pennsylvania and Illinois experiments, only one third (exactly 33.57 per cent) of the nitrogen consumed was recovered in the dung, and nearly one half (48.91 per cent) was excreted in the urine. These facts are worth remembering, and also that 28 per cent of the phosphorus consumed was not recov- ered in the total manurial excrements. ORGANIC MATTER AND NITROGEN 203 Pennsylvania Bulletin 63 reports an experiment covering two months, with four steers, two fed on a cement floor with the manure kept tramped under their feet, and two on an earth floor from which the manure was piled into an adjoining stall and kept under cover. If we assume no loss from the litter used, the following percentages were recovered from the food consumed: PERCENTAGES RECOVERED OF PLANT FOOD IN FEED METHOD OF KEEPTNG MANURE NITROGEN PHOSPHORUS POTASSIUM On cement floor, tramped . . On earth floor, piled . . . Average per cent recovered 84.8 54-0 81.3 69.0 91-5 7I.O 69.40 75-15 81.25 Of the total dry matter used for feed and bedding, 40.35 and 31.03, or, as an average, 35.69 per cent was recovered in the manure. In Table 32 are given data (in part estimated) from an experi- ment by the Ohio Station (Bulletin 183) in which 28 steers were fed on a cement floor from December i, 1904, to June i, 1905, a period of six months, or 182 days, during which time the average weight of the steers increased from 872 to 1230 pounds. At best, these results can be considered only as approximations, except as to the composition of the manure and the phosphorus added in the raw rock, but they are of interest and of some value as indicating what can be accomplished under the conditions. The amounts of feed and bedding used were accurately weighed, but their plant-food content was computed from accepted averages from each material. "The steers were turned out of the stable once a day to get water, and were allowed to run in the yards from one to two or three hours, consequently some manure was left in the yards." One would assume from this that one tenth or more of the excrements were voided in the yards. An experiment with 100 sheep (averaging 84 pounds each) for a feeding period of 112 days was conducted by the Ohio Station, in which more definite data were secured. Of the feed, 26,936 pounds consisted of hays which were analyzed; while accepted averages were used only for the standard concentrates, including 20,057 pounds of corn, 905 pounds of cotton-seed meal; and 905 pounds 204 SYSTEMS OF PERMANENT AGRICULTURE TABLE 32. RECORD OF Six MONTHS' FEEDING OF 28 STEERS ON CEMENT FLOOR Ohio Experiment Station MATERIALS TOTAL AMOUNT (Pounds) DRY MATTER (Pounds) NITROGEN (Pounds) PHOSPHORUS (Pounds) POTASSIUM (Pounds) Wheat bran .... Corn meal .... Linseed meal . . . Cotton-seed meal . Corn silage . . . . Corn stover . . . .. Mixed hay . . * . Total feed .... 9448 48128 5593 5097 63231 4896 31814 8324 40909 5083 4685 15808 4406 26946 252-3 875.9 304.0 346.1 177.0 5°-9 448.6 120. 1 148.2 40.9 64.6 30.6 6.2 37-8 126.2 159.8 63-7 36.8 194.2 56.8 409-3 106161 2454.8 448.4 1046.8 Straw bedding . . . Raw rock phosphate . 39033 4753 35I31 4753 230.3 20.6 564.6 165.2 Total supplied ... ... 146045 2685.1 1033.6 1212 Total manure .... 255203 49698 2006.0 799.0 1064 PERCENTAGES RECOVERED From total supplied T.A O 74.7 77. c 87.8 With phosphate excluded . . . 31.8 74-7 50.0 87.8 Corrected for loss in yards . . 35-3 83.0 55-6 97.6 of linseed meal; and for 3020 pounds of oat straw used for bedding. It is understood that the sheep were kept confined in the stables during the entire time. Eight different analyses were made of the manure, and the Ohio Station computes that, of the total plant food contained in the feed and bedding, 64 per cent of the nitrogen, 79 per cent of the phos- phorus, and 97 per cent of the potassium, were recovered in the manure. Wood, of the University of Cambridge, England, reports an experiment with four heifers to determine losses in making and storing farm manure (Journal of Agricultural Science, April, 1907). The experimental feeding began on January 31, 1906, and ended on April 25, 1906, a period of 84 days. During this time two of the animals consumed 13,720 pounds of mangels (containing 1784 ORGANIC MATTER AND NITROGEN 205 pounds of dry matter), 1176 pounds of hay, and used up 1963 pounds of straw as food and litter. The other two animals in an adjoining stall consumed exactly the same amounts of mangels and hay, 100 pounds less straw, and, in addition, 672 pounds of oil cake made from hulled cotton seed. The stalls in which the animals were housed during the experi- ment were bricked up to the highest level reached by the accumu- lated manure. The floors were not cemented, but were made of clay which was well rammed, and through which, according to Wood's statement, " there could be little leakage of soluble con- stituents." The manure was kept tramped under the feet of the animals, sampled for analysis at the end of the feeding period without disturbing the mass, then left in the compact condition for six months (till November 6, 1906), when it was sampled and weighed (8075 pounds from lot i and 8106 from lot 2) and applied to the soil. Following are the essential results: LOT i. (CAKE NOT FED) CONSTITUENTS DRY MATTER NITROGEN PHOS- PHORUS POTASSIUM In total feed and bedding, pounds . . 4421.0 47-9 5-5 93- 1 Percentage found * in fresh manure . 58.6 75-2 67-5 86.4 Percentage applied to the soil . . . 42.4 64.6 67-5 72.8 LOT 2. (CAKE FED) In total feed and bedding, pounds . . Percentage found l in fresh manure . Percentage applied to the -soil . . . 4942.0 60.0 41.6 9°-3 78.5 51.6 14.8 69-3 69-3 105.0 71.1 1 Assuming no loss of phosphorus during storage. Wood computes that the following percentages from the oil cake fed were recovered and applied to the soil: Dry matter .... 29 per cent. Nitrogen 37 per cent. Phosphorus .... 70 per cent. Potassium 52 per cent. 206 SYSTEMS OF PERMANENT AGRICULTURE As a general average for dairy fanning, cattle feeding, and sheep feeding, it is shown that practically one third of the organic matter, three fourths of the nitrogen, and three fourths of the phosphorus contained in the feed and bedding are recovered in the total manures. Nearly all of the potassium may be recovered except that sold in milk. (Some potassium may be excreted through the skin, especially in hot weather, but even this is washed off in the pastures by summer rains.) Emmet and Grindley have reported the following suggestive data from digestion experiments with swine (Journal American Chemical Society (1909), 31, 577): COEFFICIENTS OF DIGESTIBILITY OF THE CONSTITUENTS IN THE FEEDS CONSUMED Per Cent Digested FOOD RATION ANIMAL DRY SUBSTANCE ORGANIC MATTER PROTEIN (Nitrogen) PHOS- PHORUS Ground corn Pig A 87.1 86.7 80.6 64.7 Ground corn Pig B 86.5 86.2 76.0 6^.4 Ground corn and middlings Ground corn and middlings Pig A PigB 87-3 86.8 87.6 87.2 84.4 82.4 74.6 77-7 Average of four .... 86.9 86.9 8l.2 70.6 It is common knowledge among farmers that swine fed largely on grain produce but little solid manure; and in these experiments only about 13 per cent of the organic matter, 20 per cent of the nitrogen, and 30 per cent of the phosphorus were recovered in the solid excrement. However, the existing data are not sufficient to justify the adoption of these determinations as representing the average digestibility by swine of the phosphorus contained in the grain rations. That the normal coefficient is high, is evidenced by the fact that, unlike most animals, swine normally excrete very appreciable amounts of phosphorus in the urine. The production, composition, care, and value of farm manure are discussed in a later chapter. ORGANIC MATTER AND NITROGEN 207 THE FIXATION OF FREE NITROGEN As already stated, the nitrogen naturally in the soil is contained essentially in the organic matter. Any process which tends to decompose or destroy this organic matter, such as nitrification or other forms of oxidation, will also tend to reduce the total stock of nitrogen in the soil, whether removed by cropping or lost by leach- ing. Because of this fact, the matter of restoring nitrogen to the soil becomes of very great importance. Of course, a part of the nitrogen removed in crops may be returned in the manure produced on the farm; and nitrogen may also be bought in the markets in such forms as dried blood (14 per cent), sodium nitrate (15^ per cent) , and ammonium sulfate (20 per cent) ; but when we bear in mind that such commercial nitrogen costs from 15 to 20 cents a pound, and that one bushel of corn contains about one pound of nitrogen, it will be seen at once that the purchase of nitrogen cannot be considered practicable in general farming, although in market gardening, and in some other kinds of intensive agriculture, commercial nitrogen can often be used with very marked profit. Considering all of these facts, and the additional facts that there are about seventy-five million pounds of atmospheric nitrogen resting upon every acre of land, and that it is possible to obtain unlimited quantities of nitrogen from the air for the use of farm crops, and at small cost, the inevitable conclusion is, that the inex- haustible supply of nitrogen in the air is the store from which we must draw to maintain a sufficient amount of this element in the soil for the most profitable crop yields. . It is often stated that legume plants, such as clover, have power to obtain free nitrogen from the air. This is not strictly true. Red clover, for example, has no power in itself to get nitrogen from the air. It is true, however, that certain microscopic organisms 1 which commonly live in tubercles upon the roots of the clover plant do have the power to take up free nitrogen and cause it to unite with other elements to form compounds suitable for plant food. 1 Among the scientists who were prominent in making these discoveries regard- ing the action of bacteria in the fixation of free nitrogen were Hellriegel, Willfarth, and Nobbe in Germany, Atwater in America, Lawes and Gilbert in England, and Boussingault and Ville in France. 208 SYSTEMS OF PERMANENT AGRICULTURE The clover plant then draws upon this combined nitrogen in the root tubercles, and makes use of it in its own growth, both in the tops and in the roots of the plant. These nitrogen-fixing bacteria live in tubercles upon the roots of various legume plants, such as red clover, white clover, alfalfa, sweet clover, cowpeas, soy beans, vetch, field peas, garden peas, field and garden beans, etc. The tubercles vary in size from smaller than a pinhead to larger than a pea, varying somewhat with the different kinds of plants, being especially small upon some of the clovers, and large upon cowpeas and soy beans. The tubercles are, of course, easily seen with the eye, but the tubercle is only the home of the bacteria, somewhat as the ball upon the willow twig is the home of the insects within. The bacteria themselves are far too small to be seen with the unaided eye, although they can be seen by means of the powerful microscope. Several million bacteria may inhabit a single tubercle. It is not necessary to see the bacteria, because if we find the tubercles upon the roots of the plant, we know that the bacteria are present within, otherwise the tubercle would not be formed.1 It has also been demonstrated that, as a rule, there are different modifications of nitrogen-fixing bacteria for markedly different species of legume plants. Thus, we have one kind of bacteria for red clover, another for cowpeas, another for soy beans, and still a different kind for alfalfa. There are some noteworthy exceptions to this rule. Thus, the bacteria of alfalfa (Medicago saliva) and of common sweet clover (Mellilotus alba] are interchangeable, and apparently identical, as are also the bacteria of cowpeas (Vigna unguiculata) and the widely distributed native partridge pea (Cassia chamaecrista) , relationships of much importance in connection with soil inocula- tion for alfalfa and cowpeas. There is evidence that, by a compara- tively long process of breeding, or evolution, the bacteria which naturally live upon one kind of legume may gradually develop the power to live upon a distinctly different legume to which they were not at first adapted. This change which has been brought about 1 A few plants form starchy nonbacterial tubers, which may be of large size, like the potato and artichoke, or of smaller size, as on the rootstalks of nut grass (Cyperus rotunda). ORGANIC MATTER AND NITROGEN 209 with some certainty in artificial cultures, and which very possibly occurs to some extent in farm manure from legume hay, may fur- nish bacteria with feeble action for a time, but ultimately, no doubt, with full power. Of course, this process of forcing bacteria to live upon a legume to which they are not naturally adapted has little or no practical value, because it is unnecessary, if there is a species of bacteria which naturally live upon the same legume. On the other hand, if, by any such process of breeding, or evolution, a species of nitrogen-fixing bacteria could be developed which could live on a nonleguminous plant, as corn, for example, it would be of incalculable value. As yet, the efforts of bacteriologists, working on this problem, have given only negative results, so far as known to the author. Attention is called to the fact that there are numerous instances where two different kinds of plants live together in intimate part- nership relation. If only one of the two plants receives benefit from this relationship or association, then the plant receiving the benefit is called a parasite. Thus the mistletoe is a parasite upon the elm or gum or other tree on which it lives. The mistletoe draws its nourishment from the tree. The tree is injured rather than benefited by the mistletoe. Dodder is also a parasitic plant, living upon other plants, except during the early part of its growth. Ticks and lice are common examples of animal parasites, living upon other animals. In some cases a relationship exists which is not parasitic, but symbiotic. The term symbiosis, Which is commonly used by biologists to define this relationship, means living together in mutual helpfulness. The association of bees and flowers may serve to illustrate this mutual helpfulness, although this is not an example of intimate symbiosis. Thus, the bees obtain their food from the flowers and, in turn, the flowers, many of them, are in-, capable of producing seed or fruit unless the pollen is carried from the male flower to the female flower by bees or other agencies. It is well known that plant lice and ants are mutually helpful. Likewise, the association of nitrogen-fixing bacteria and legume plants is a relationship of mutual helpfulness, and this is one of the best illustrations of what is meant by symbiosis. The legume furnishes a home for the bacteria and also furnishes in its juice or 210 SYSTEMS OF PERMANENT AGRICULTURE sap most of the nourishment upon which the bacteria live. The bacteria, on the other hand, take nitrogen from the air, contained in the pores of the soil, and cause this nitrogen to combine with other elements in suitable form for plant food, which is afterward given up to the legume for its own nourishment. Another illustration of remarkable parasitism, if not, indeed, one of true symbiosis, is found in the common lichens living upon rocks and trees. The lichen is not a single plant, but two plants, — one an alga, which lives upon the wood or stone, and the other a fungus, which lives upon the alga. Algae also live in the free state separate from fungi, and the present opinion of botanists seems to be that when the two are associated in the form of lichens, this association is not detrimental, but rather beneficial, to the alga, as well as to the parasitic fungus. If this is true, then it is another case of true symbiosis. (It is now known that some fungi have power to feed upon atmospheric nitrogen, and probably those in lichens furnish combined nitrogen to the algae upon which they live.) In the symbiosis of legume plants and nitrogen-fixing bacteria we have a partnership or relationship of immeasurable value to agriculture. Here is a class of plants (legumes) that are capable of consuming or utilizing nitrogen in quantities larger than could possibly be obtained from ordinary soils for any considerable length of time. They have no power in themselves of taking nitrogen from the atmosphere, and to them the symbiotic relation with this low order of plants (the nitrogen-fixing bacteria, Pseu- domonas radicicola), is especially helpful, and for the best results it is absolutely necessary. INOCULATION FOR NITROGEN FIXATION While it is true that nitrogen-fixing bacteria are essential to the most successful growing of legumes, it is also true that, as a very general rule, the proper bacteria for the ordinary legumes are already present in the most common soil, especially where the particular legume has been grown in the vicinity for several years, or where manure made from the legume has been applied. This applies especially to alfalfa in the alfalfa country of the West, to ORGANIC MATTER AND NITROGEN 211 cowpeas in the cowpea country of the South, and to the clovers throughout the Central and Eastern states. Where the special legume has not been grown successfully in the vicinity; or even on fields where the legume has not been grown for many years, and where neither manuring, overflow, nor dust storms have brought the bacteria from other fields, it is worth while to consider inocula- tion. The bacteria for clover, cowpea, and vetch are now very widely distributed over the United States (in part because of the par- tridge pea and wild vetches); but for alfalfa (except in alfalfa regions) and for soy beans, the question of inoculation should always be considered. For inoculating alfalfa, either alfalfa soil or sweet-clover soil can be used, care being taken to use only well- infected soil, collected where the plants have been growing for several years, well provided with root tubercles. The accumulated practical experience of the past twenty years, and the data thus far reported from many comparative experi- ments, combine to prove that the simplest and surest and most economical method of inoculation is by means of well-infected natural soil, collected where the plants are thrifty and free from noxious weed seeds, although the danger of carrying weed seeds or plant diseases by overflow, by wind storms, and in purchased ma- nures and farm seeds is probably a hundred times greater than by using infected soil for inoculation. The amount of soil used varies from 100 pounds to a wagon load to the acre. It may be applied broadcast with some degree of uniformity, and it should be mixed with the surface soil without delay, as by harrowing or disking, because exposure to the sunlight tends to destroy the bacteria. Successful seed inoculation can be performed with fresh, properly prepared artificial cultures, but, as a rule, this method has proved unsatisfactory. Ten years ago German promoters undertook to establish the business of selling nitrogen bacteria for seed inocu- lation, and more recently American promoters have widely ad- vertised similar products, but failure is the most common report from their use. For large seeds, such as soy beans, a very satisfactory method of inoculation, suggested by the Illinois Experiment Station, is to thoroughly moisten the seed with a 10 per cent solution of glue, 212 SYSTEMS OF PERMANENT AGRICULTURE immediately sift over them sufficient dry, pulverized, infected soil to absorb all of the moisture, thus furnishing a coating of infected soil for every seed. The seed should be shoveled over a few times, then screened, and planted within a day, or spread out to dry, after which they may be kept as long as though not covered with dust. The coat of thoroughly infected soil provides a much better inoculation than is common from the use of artificial cultures, and it does not interfere with drilling the seed immediately after treatment. If this method is used for inoculating small seeds, such as alfalfa, greater care must be taken to screen them afterward to prevent clusters of seeds from remaining glued together. If seeds are moistened, they should either be planted very soon thereafter or spread out and thoroughly dried, otherwise they are likely to mold and lose vitality. Infected soil should never be long exposed to bright sunshine, which is very destructive to all forms of bacteria. There has been much discussion during recent years concerning the development of unusually virile bacteria, but even if it were possible to develop and maintain in the soil bacteria of greater nitrogen-fixing power, it is a question whether the discovery would have great practical value (especially after the first year) , for the simple reason that bacteria multiply with such tremendous rapid- ity that we may soon have many times the number of bacteria that are really needed to do the work. In other words, the increase in numbers may result in just as great efficiency as would result from any increased power of the individual bacteria. One who carefully studies the formation of root tubercles on plants growing on soils in varying conditions or degrees of infection will observe that on plants sparsely infected the individual tubercles or clusters develop to enormous size, comparatively speaking; while in well-infected soils the individual tubercles are much smaller, and clusters scarcely form. It is also observed that the marked effect on the growth, color, and composition of the plant is produced even though only a half-dozen large tubercles form on the roots. It is very evident that the relationship between the bacteria and the host plant is such that if the soil is sparsely infected, so that the roots come in contact with but few bacteria, and but few tubercles are started, those few tubercles will be so enlarged, either in individuals or as ORGANIC MATTER AND NITROGEN 213 clusters, that the multiplication and activity of the bacteria are sufficient to meet the needs of the host plant so far as nitrogen is concerned. Of course, as soon as the soil becomes well infected, the plant roots come in contact with large numbers of bacteria, and many tubercles are formed, but most of them remain small, and no large clusters are formed, because the bacteria in the large num- ber of small tubercles are apparently capable of furnishing all the nitrogen needed by the host plant. If the other elements were provided in greater abundance, the tubercles would undoubtedly become enlarged, as much as necessary to supply the nitrogen needed to balance the supply of the other plant-food elements util- ized by the plant. NITROGEN FROM SOIL AND Am Experiments or demonstrations almost without number have been performed to determine the amounts of nitrogen taken from the air by various legume plants when grown in sand cultures essentially free of combined nitrogen, but there are much less data concerning the relative amounts of nitrogen taken from the soil and from the air by legume crops grown on normal cultivated land. There are two methods by which such information can be se- cured with a fair degree of satisfaction. One of these is to deter- mine the amounts of nitrogen in infected plants and in similar plants not infected, grown on the same type of soil; and the other is to compare the total nitrogen content of a nonleguminous crop with that of a crop of infected legume plants, grown at the same time on similar soil. Though not strictly exact, these methods furnish practically correct information. In Table 33 are shown the results of a field experiment to de- termine the amount of nitrogen taken from the air by alfalfa when grown on the common corn-belt prairie land (Illinois Bulletin 76). The difference between the amount of nitrogen contained in the crop from the inoculated soil, on the one hand, and in the crop from the uninoculated soil, on the other hand, represents the amount of nitrogen secured by the bacteria. In no case will this give too much credit to the bacteria; but, if any unavoidable cross inoculation 214 SYSTEMS OF PERMANENT AGRICULTURE TABLE 33. FIXATION OF NITROGEN BY ALFALFA IN FIELD CULTURE Illinois Experiments on Common Prairie Land PLOT No. TREATMENT APPLIED DRY MATTER IN CROES (Pounds) NITROGEN IN DRY MATTER (Per Cent) NITROGEN IN CROPS (Pounds) NITROGEN FIXED BY BACTERIA (Pounds) ia ib 2d 2b 3a 3* None 1180 2300 1300 2570 1740 3290 I.8S 2.70 2. 02 2.65 2.03 2.71 21.81 62.04 26.2O 68.02 35-40 89.05 Bacteria 40.23 Lime Lime, bacteria 41.82 Lime, phosphorus .... Lime, phosphorus, bacteria 53-65 occurred during the progress of the experiment, these amounts might understate the effect of the bacteria. It is very probable, however, that the increased root development, induced by remov- ing the nitrogen^limit in plant growth, would make it possible for the infected plants to secure somewhat more soil nitrogen than otherwise. (Note the effect of phosphorus in the record.) How- ever, this, too, should perhaps be placed to the credit of the bac- teria, even though it is not atmospheric nitrogen, because if such nitrogen existed in the soil solution, it would soon have been lost in drainage waters if not taken up by the enlarged root system of the growing crop. Slightly more than one third of the total nitrogen contained in the crop from the inoculated unfertilized plot was secured from the soil, with larger proportions and larger actual amounts for the other plots. It should be borne in mind that nitrogen is required for root growth as well as for growth above ground, and that three other crops of alfalfa were cut from these plots during the season, this cutting having been made on May 28. Evident cross inoculation occurred before a second cutting was obtained; but the data given in Table 33 indicate that plot ib secured about 172 pounds of nitrogen from the air during the season, the yield of air-dry hay having been 2563 pounds for the first cutting and 10,980 for the ORGANIC MATTER AND NITROGEN 215 season. The value of this " gathered " nitrogen amounts to $25.80 l per acre at 15 cents a pound. Similarly, plot 36, yielding 3625 pounds of air-dry hay in the first cutting and 17,060 per acre for the season, ''gathered" 252 pounds of nitrogen from the air, worth $37.80 at 15 cents a pound. The Illinois Station also conducted a series of pot cultures in- cluding 12 inoculated pots and 12 similar uninoculated pots, the results of which support very well the field experiments reported in Table 33. (See Illinois Bulletin 76.) The Dominion of Canada Experiment Stations (Report for 1905) as an average of twenty-one pot cultures increased the nitrogen content of the soil from .0392 per cent to .0457 by growing mam- moth clover for two successive seasons, and turning it all back into the soil. This amounted to 179 pounds' increase of nitrogen per acre to a depth of 9 inches, but it should be noted that the soil was extremely poor in nitrogen, containing only 784 pounds in 2 mil- lion at the beginning. In a similar plot experiment for two full seasons, two cuttings of mammoth clover and all residues being returned to the soil each season, the nitrogen content was increased from .0437 to .0580 per cent, making a gain of 175 pounds per acre to a depth of 4 inches; but only 874 pounds of nitrogen were contained in 2 million of soil at the beginning; so that in both of these experiments the results are not very different than would be secured in sand cultures. The clover was reseeded each year and grown without a nurse crop. The average annual fixation reported amounts to less than 90 pounds per acre. In another experiment by the Illinois Station (Bulletin 94) six sets of immature cowpea plants (10 plants in each set) were carefully collected, tops and roots. Three sets were infected, the others not infected. The plants were taken from a catch crop grown after oats had been harvested, on land that had been heavily cropped with corn and oats until nitrogen had become a limiting element, especially for a catch crop grown after oats. As a general average, the infected plants contained 86 parts of nitrogen in the tops, 5 parts in the roots, and 9 parts in the tubercles, while in direct comparison the noninfected plants contained 25 parts of 1 "They not only work for nothing and board themselves, but they pay for the privilege." — DAVENPORT. 2i6 SYSTEMS OF PERMANENT AGRICULTURE nitrogen in the tops and 2 parts in the roots, thus indicating that 73 per cent of the nitrogen contained in the infected plants was secured by the bacteria. The nitrogen in the dry matter of the infected plants varied from 4.09 to 4.33 per cent in the tops, from 1.45 to 1.53 per cent in the roots, and from 5.76 to 6.05 per cent in the tubercles; while the nitrogen in the dry matter of the noninfected plants varied from 2.32 to 2.69 per cent in the tops, and amounted to .88 per cent (in each of three lots) in the roots. From an experiment with soy beans by the Wisconsin Station (Report for 1907), it is computed that only 14 per cent of the nitrogen contained in well-infected plants was secured from the air. The yield of dry matter was practically the same, but the infected plants were richer in nitrogen and protein, and thus of better quality. "The soy beans were grown on low, rich soil in these experiments." The Michigan Station (Bulletin 224) reports data showing that 33 per cent of the nitrogen in soy beans was' secured by the bacteria, on well-infected plants. As an average of 20 untreated plots in one test, and of 16 plots treated with phosphorus and potassium in another test, both over a period of 25 years, in a four-year rotation of corn, oats, wheat, and hay (mixed clover and timothy), the Pennsylvania Experiment Station reports the following yields in pounds per acre per annum : POUNDS PER ACRE TREATMENT CORN OATS WHEAT HAY Ears Stover Grain Straw Grain Straw None 2956 3783 1955 2801 I°33 1279 1403 1762 8lS 1108 1265 1776 2783 4138 Phosphorus, potassium . If we compute the nitrogen in the three uncultivated crops (see Table 23), adopt the estimate that the hay was three fourths clover and one fourth timothy, and assume the soil nitrogen for the hay crops to be as indicated by a curve projected from the amounts furnished to the oats and wheat crops, then the clover must have secured from the air 66 per cent of its nitrogen when grown on ORGANIC MATTER AND NITROGEN 217 untreated land, and 64 per cent on land treated with phosphorus and potassium, the average annual yields of nitrogen per acre being 29.1 pounds for oats, 25.9 for wheat, and 50.1 for hay, on untreated land, and 36.15 pounds for oats, 35.5 for wheat, and 75.5 pounds for hay, on treated land. While the calculation of 65 per cent is probably near the truth for the treated land, where the nitrogen is likely to be the limiting element in crop production, the marked reduction in yield of nitrogen between the oats and the wheat crops on the untreated land is probably not a true index of the change in available soil nitrogen, because on these plots phos- phorus is certainly the limiting element for wheat, as will be seen from later discussion. In any case, we are safe in concluding that soil which will fur- nish from 26 to 36 pounds of available nitrogen for a crop of oats or wheat will also furnish approximately as much for the hay crops, whether timothy or clover. Clover and other legumes take available nitrogen from the soil in preference to the fixation of free nitrogen from the air, the latter being drawn upon only to supplement the soil's supply and thus balance the plant-food ration. In other words, the legumes have no nitrogen limit in yielding power when properly infected, but with abundance of available soil nitrogen constantly provided to fully balance other essential elements or factors, there is little or no development of root tubercles, and little or no fixation of free nitro- gen occurs. From the experimental data here presented or referred to, and from many other calculations approximating exactness, the con- clusion may be drawn that on normally productive soils at least one third of the nitrogen contained in legume plants is taken from the soil, not more than two thirds being secured from the air. This proportion would apply to the nitrogen content of the roots as well as to the tops; so that, if one third of the nitrogen of the entire plant is in the roots and stubble, and two thirds in the crop harvested, the soil would neither gain nor lose in nitrogen because of the legume crop having been grown, the soil having furnished as much nitrogen to the plant as remains in the roots and stubble. When grown on richer soils, such legume crops leave the soil poorer in nitrogen; but on poorer soils, furnishing less than the 218 SYSTEMS OF PERMANENT AGRICULTURE normal amount of available nitrogen, the growing of such legumes would enrich the soil in proportion to its poverty. In other words, to the soil that hath not, shall be given; but, from the soil that hath, shall be taken away. When properly infected, legume plants have power to make nor- mal growth and full development on soils absolutely devoid of nitrogen, if available mineral plant food, limestone, moisture, aeration, and all other essential factors are provided in abundance or perfection; and the statement sometimes made that the pres- ence of soluble nitrogen is necessary, in order to give clover a start, is not correct, as witness the accompanying illustrations of clover growing in purified quartz sand void of nitrogen, with all plant food provided except nitrogen, the culture on the right marked "Bac- teria " having been well inoculated with the clover bacteria; while the middle culture was started in exactly the same manner, except that it was not inoculated. In the culture on the left, all plant food was provided, including nitrogen. NITROGEN IN TOPS AND ROOTS OF LEGUMES From data already given it will be seen that in the study of immature cowpeas at the Illinois Station, the infected plants contained only 14 per cent of their total nitrogen in the roots with more than half of this in the tubercles themselves, at that stage of growth. As the plants approach maturity, the tubercles decay, and only the shell, or outer coat, remains, the nitrogen being absorbed largely by the host plant, but in some part evidently by companion ORGANIC MATTER AND NITROGEN 219 plants, as timothy or blue grass, whose roots may come in contact with the decomposing tubercles. In case of the noninfected cow- peas, only 7 per cent of the nitrogen was found in the roots at that stage of growth. Pot-culture experiments by the Dominion of Canada Experiment Station, with plants planted May 20 and harvested August 4, showed that very poorly infected horse beans contained 19 per cent of their nitrogen and 18 per cent of their organic matter in the roots, while better infected plants made a larger yield and contained 25 per cent of their nitrogen and also 25 per cent of their organic matter in the roots; whereas well-infected mammoth red clover contained 40 per cent of its nitrogen and 35 per cent of its organic matter in the roots. In a field experiment with mammoth clover, seeded with barley in the spring and harvested May 25 the following year, the Cana- dian Station found, per acre, 123.8 pounds of nitrogen in the tops and 48.5 pounds in the roots, to a depth of four feet, corresponding to 72 per cent in the tops and 28 per cent in the roots. As an average of four determinations with red clover, the Con- necticut Station found 28 per cent of its nitrogen, 35 per cent of its phosphorus, and 21 per cent of its potassium in the roots and stubble. As an average of two determinations by the Illinois Station, the red-clover roots found in the surface soil (o to 7 inches) contained 25 per cent of the total nitrogen of the plants, while only one per cent of the total was contained in the roots in the subsurface stra- tum (7 to 20 inches). In the case of nearly mature cowpeas, 12 per cent of the total nitrogen was found in the surface roots (o to 7 inches) , and i per cent in the subsurface (7 to 20 inches) ; and the corresponding figures for nearly mature soy beans were 8 per cent and i per cent. In Table 34 are recorded the data from an Illinois investigation of sweet clover, in which determinations were made of the total dry matter and nitrogen; (i) in the tops as they would ordinarily be cut with a mower, (2) in the surface residues, consisting of stubble and fallen leaves and old stems, (3) in the large roots in the plowed soil to a depth of seven inches, (4) in the smaller roots in the plowed soil, and (5) in the roots of the subsurface stratum from 7 220 SYSTEMS OF PERMANENT AGRICULTURE to 20 inches in depth. The investigation was made when the sweet clover was full grown and nearly mature. The crop was started the previous season, sweet clover being a biennial plant. TABLE 34. ILLINOIS INVESTIGATIONS OF SWEET CLOVER (MELLILOTUS ALBA) PARTS OF PLANT DEPTH (Inches) DRY MATTER PER ACRE NITROGEN PER ACRE Pounds Per Cent of Total Pounds Per Cent of Total Tops harvested . . . Surface residues . . Total tops .... 9029 1338 174 23 10367 81 I97 86 Large surface roots Small surface roots Total surface roots Subsurface roots . . Total roots .... o to 7 o to 7 o to 7 7 to 20 o to 20 I568 241 17 5 1809 60 1 M 5 22 9 IO 4 2410 J9 31 14 Total tops and roots .... 12777 IOO 228 IOO It will be seen that the yield of sweet clover is very large, amount- ing to 6.4 tons of total dry matter, of which, however, the roots contain only 1.2 tons per acre, or less than one fifth of the total. The tops of sweet clover are nearly as rich in nitrogen as full- grown red clover (40 pounds per ton), but the roots contain only one seventh, or 14 per cent, of the total nitrogen. Nearly 24 per cent of the total nitrogen was found in the roots, stubble, and sur- face residues (largely of the previous season's growth). The sweet clover used in the investigation was well infected; but, in a previous experiment on the same soil (brown silt loam prairie of the early Wisconsin glaciation), it was found that the yield of sweet clover was almost exactly doubled by thorough inoculation, and the percentage of nitrogen in the infected plants was also about one half more than in the noninfected plants, showing that on this soil about two thirds of the nitrogen required for this large crop was secured from the air. While sweet clover makes a fair quality of hay, if cut sufficiently early in its growth, and is also used for pasture with some success ORGANIC MATTER AND NITROGEN 221 when nothing better can be had, it is not to be compared with red clover or alfalfa for either purpose, but it does give promise of great value as a green manure crop, and it seems appropriate to emphasize the fact that the 6.4 tons of dry matter furnish as much humus-forming material and as much nitrogen as would be furnished by 25 tons of average farm manure. In the Wisconsin experiments above referred to, the infected soy beans contained in their roots about 4 per cent, 6 per cent, and 5 per cent, of their nitrogen, phosphorus, and potassium, respec- tively; and in the Michigan experiments the corresponding figures are about 4 per cent, 6 per cent, and 6 per cent, respectively. From an exhaustive investigation of the crimson-clover plant (Trifolium incarnatum) , Penny (Delaware Bulletin 67) reports the following average results for fall-seeded crops harvested about May 15, when nearly in full bloom: TABLE 35. COMPOSITION OF CRIMSON CLOVER IN BLOOM Delaware Experiments : Pounds per Acre PARTS OF PLANT AIR-DRY MATTER NITROGEN PHOS- PHORUS POTASSIUM Tops, pounds 4CI2 IO3 7.6 70 Roots, pounds 2022 41 3-1 15 Total, pounds 6^34 144 IO.7 85 PERCENTAGE OF TOTAL Tops, per cent 60 72 71 82 Roots, per cent 71 28 2O 18 The proportions were found to vary considerably, but this gen- eral average shows the crimson-clover roots (to a depth of 24 inches) to contain less than one third of the organic matter, nitrogen, and phosphorus, and less than one fifth of the potassium of the entire plant. It was found that 77 per cent of the roots were in the first 6 inches of soil, and 1.3 per cent in the second 6 inches, 7 per cent in the third, and 3 per cent in the fourth 6 inches. In Table 36 are recorded much additional information concern- 222 SYSTEMS OF PERMANENT AGRICULTURE TABLE 36. COMPOSITION OF PLANTS (Tops AND ROOTS) Delaware Experiment Station : Crops seeded July 22 CROP, AND DATE OF HARVEST PARTS OF PLANT POUNDS PER ACRE AND PER CENT IN ROOTS Air-dry Matter Nitro- gen Phos- phorus Potas- sium Cowpeas Nov. 7 Tops 3718 3OI 9 65.2 4.2 .1 7.2 1.0 .1 39-2 1.9 .1 Roots o to 8 inches . ... Roots, 8 to 12 inches .... Per cent in roots 8 6 13 8 Soy beans Nov. ii Tops 6790 717 39 130.9 8.8 •5 I6.5 1.0 .0 38.3 1.4 .1 Roots, o to 8 inches Roots, 8 to 12 inches .... Per cent in roots 10 6| 5* 4 Vetch Nov. 19 Tops 3064 584 16 108.0 12.8 •4 9.8 2.0 .1 65.1 5-7 .2 Roots o to 8 inches Roots, 8 to 12 inches .... Per cent in roots 17 ii 18 8 Crimson clover Nov. 20 Tops 5372 38i 32 128.2 5-7 •5 25-9 .8 .1 69.7 3-2 •3 Roots, o to 8 inches Roots, 8 to 12 inches .... Per cent in roots 7 6 3* 5 Alfalfa Nov. 20 Tops 2267 1972 8 54-8 40.2 .2 5-7 3-7 .0 26.7 7-9 .0 Roots, o to 8 inches Roots, 8 to 12 inches .... Per cent in roots 47 42 39 23 Red clover Nov. 22 Tops ; 2819 1185 27 69.8 32-5 •7 8-3 4-3 .1 38.6 8.0 .2 Roots, o to 8 inches Roots, 8 to 12 inches .... Per cent in roots 3° 32 35 18 Cow-horn turnip Nov. 15 Tops . 2565 2902 64.4 44-7 6.2 5-i 66.6 t- COCO o ^p NO oo M NO CN) ON ON M NO tO IH IH ON H IH H 0 CNI s-i IH NO M co t^ to CN) tONO M tO H NO CNI CN) to t tO CO M O co to tM *0 si| M 00 CN) CNI ON f *-* t~~ M t CN) CO CO CN) CO CN) oo t CN) to CNI CN| M t— to co CM IO CO M CNI Q ~ ° cor- ON H M 00 O tH ON M o o o O\ t- O M r- M o o_ < B ^£L, >< < 9 oo NO M CN! t CN) H t— H IH NO CN) rf CN) ONNO CN) NO IN M CO CS to CN) t CN! CN NO ON CN! W Z IH CNI oo NO W \f. CN t t CN! to ON ON ON CN) M NO 0 Z, ON CO HI •^ CN! t ON M H NO CO M IH w O IN -H 0 IN art (N ON co l~» O CO H Tf I/~ CO oo co NO t~~* ON t"** CO CN! ON O *"~ CNI t~» m % 5/5 ^ < CN! ON CO CM !>• M CNI NO CO IO CNI CO NO CN) r- CN) t^ t CN! to CNI CO w NO to CO H IO H CO e«i H OO t^* to f^* O tooo' to 0 ON ON CO to O c<- COCO IH O co CNI ON ^11 O Is- CO CN! t*^ CO HH NO H t^ M (N Tf CNI •*• t cs t^ CN) "3- M CNJ to CS co too t CNI M NO co to M r— ON CN! 1-— O ON NO H OO NO CNI t- CO CN) M CON CNI O ^ O Z M H 00 t . NO ON H CN! 1^ Tf H CNI M ! ^ M M CN IH to CNI j. H coco co O O 00 Is- to rr t to O CN! NO O NO NO < ho O to H H CO CN t— ON M H ON to M CN) IH IO O H CN! H CN| PI NO H iJ 06 ij a O to NO NO CO CN! oo 00 NO M f^ to co NO 00 ON CN! NO CO * H Z 00 co CNI CN! 00 O H NO M O i-i co CO 1^ CN! l>- W CNI O CNI to CNI NO t t CN) >0 w 3 D ON NO co «>. to o CN! H M f— M toco t^ £-N£~ « .5 5 NO t CN) CNI NO t^ O CN! CN) CN) O *— M to CNI M ON W t M CO NO CO M ^ ON M O (N I-I NO CN) CN) 10 tr IT CO J^* O (N tNO O CO * ^o NO r— NO CN) t O IH 00 NO M M H M M CN! M M CNI • 73* en 6 H 3 fi SOIL TREATMENT . 3° J2 g A- . f» CvJ *• QJ r— OO ON ON ON ON 00 CO CO en _rtd a> G O I-I IN O O O ON ON ON en CO t to O O O ON ON ON en -t-T «-T C . NO £— OO o o o ON ON ON 4 years Corn, av. bu. 4 years Wheat, av. bu. 3 years Hay, av. tons Value of 3 av. crops Gain for treatment Gain for phosphorus Cost of phosphorus Profit from phosphori Net profit, per cent 248 SYSTEMS OF PERMANENT AGRICULTURE pq O T3 O. £ u O „ H * PI ONOO NO M in 10 rO t^ CNI O M M ONQO CO M pq 1 f- CO co CONO ' ** NO NO H 10 M co 04 O 00 ' ^ IO N W w PI O 00 O Tf COOO 00 CO 10 *^« t H NO r^ ^ t~- ON C4 M M TT O O H TtNO ^ m 111! s £ H Tf CN) NO P« Pi NO H t-» N NO CO M 00 co fO 0 O 00 w t^ CNl VOOO co tO CN1 M ^ P) O PI M Tf rr M w ^ Q « Q 3 OO co COOO O CO PI ON * NO *» -H << 0 s £ t^- M H >O CNl M ON H t^ N co ON H oo 0 O O H t^« co NO t^. *f IO O M ^ P) P) ON M M »0 •g oo 00 M O NO J^ H 'O IO N NO •* O^ NO t^. ONOO ON CN) H 00 00 TJ- O H PI N ro CQ ^g|£ £n H ON PI NO CM O O N t*« r*j l^ CO H 00 ro 8 M O M r^ co t^» t^- cs VO CN) HI <& H IH T}- "S H W S a « > 5 NO oo \O r^ NO N ON O NO M CO CO o co IO ON r^ •*NO H ONOO 00 O M PI NO ra S5<^ £<*o s £ t^* H H 10 o ^f O »o u ,j » •J D t PI co t^ NO 00 H CO N H Tt CO ">C t^ 10 ON •* CO M •t •* n £3 C/3»H <*=5 O 00 H NO M f" ^ H IO «>l M' \o H 00 > i/". rf M M NO •*t M e# m H °g « 5 t^ O ID re •* « H O NO M CO t^- ^t P) ON IO •* o n < 3 ^§ •* H H 10 K H LO H O CN! 00 rf H r^ es t-«. »o P< O M NO O M t 'f H •* M O MD NO H H (N CO H ^- W 00 PI <& 'S ™ • w C 1 PM Soil, TREATMENT . en Jjil -° *TK- ~ nJ o££ U^U O O H Ov O O 00 ON ON CO 3 C 3^1 _Q ^ „ -i-J Ui E~S gJ o£^2 0^0 PI ro Tf O O O ON ON ON 0 i*t •^ +j- & r-rt_g fl 5 •" o^ ^ r N i^ O u ^ in lONO t^- ao o ON ON 3 jQ a" G ^3 O 0 •* >2 > ^ > rt -w" ctf " CNJ . all 2 52 JS oj c^ rt >* >* >^ n- co rrs S.-E 2 go u So- v/ *-• CO > J3 0 rt «j^ cois cx <4-l »-. Ul °^^ § C C 3r3r3 aJ« D *S C i_. OH a; i| S co fXft -C S U- CX O^c ^-i; o o — g «• H O O CNl !>• CO ONNO CN! Tf CNl O CN! NO CO •^-NO CN! NO CN) to CO to IO CO M f£f CN! CONO M •* 10 a si moo M 1^. ON w M to t^ 4s* 10 HI CNl 00 ON TJ- cxD 't r^ H rj- rr* O r~- ^r ON O « SET < 85*7 >-,«! VO N t^ M O 10 M t^ M t^- f H •CO CNl IO CN to co ON CO Hi to CN] NO IO HI Tf CNI Hi t& CNI oo •* t>- CO H O *>• l^. o 10 J tooo oo O 10 f^ co CONO •* <>. en OJ in u O % H N IO t 0 CO M (N 00 M 10 H HI HI 00 00 CNl ON H £5= D G O all 00 H *• ro I^NO CO !>• co TJ- M rj- Tj- tONO O Tf ON CN! NO co co O CNI CN) H 60 C U • w •* ON to to NO M r)- ON t^-NQ o •* 5) < *& ro H t^ M oo oo H NO tN CO Tt M CO CN) 00 co •. co to H NO NO NO NO co "ON H O t/> U o fc ON NO 10 NO •* M ^ H t^ W H (N M ONOO CNl HI O cq HI •3- H H CO &} 3 t« ^ u (M 00 oo rf ro Tj- Tf Tf IN t^ •* o H ONNO NO 0 CN! M .52 u 1 ro N 10 ro *-— * M t^- W HI CO CNl M ^r t^- CO HI CN! <£f T3 C NO H dl CN1 J>. •* Tf H NO * HI CO IN CNl NO TT CNl ON O ON t^ M O "rt U s z *3 6 NO t^ O M t^ CNl NO NO HI CO H CNl NO to CN! r^ t^. to H 00 00 CO H <& i C o >o M ss 10 NO CNl to to Tj- O ON ^ O 00 •* HI Tf 10 t~» o *^ H! O rt tn U y Tt- 1- t-^. M t^ 10 M t^-NQ Hi CS M CO Tj- Tf CN! ONNO •0 10 NO CO VI P-JI -Q ^ ~ — ^ _~ rt o F o£^2 U^U M N CO O O O ON ON ON c« -i c 3^2 -0 ^r -• ~ as o^-2 U^O Tf 10 NO O O O ON ON ON a 9-P -O - E"2 °> u^ r^-oo &o ON *ti* > a ^ j*s. C -. ^ >, •* •* N Value of 3 av. crops Gain for treatment Gain for phosphoru * 2 ~ 3 -^ C i_ O, oj IIS g PHC. J3. S *T ^8* "o^ 2 £s ^ S 82 UPL,£ 2 50 SYSTEMS OF PERMANENT AGRICULTURE comparison, will, as a rule, give more trustworthy results. Thus, we may subtract the average yields of the adjoining untreated plots, 14 and 17, from the yields of the manured plots, 15 and 16, to determine the increase produced by manure alone. Then we may subtract the yield either of plot 4 or the average yield of the three untreated plots, i, 4, and 7, from the yields of the plots which receive both manure and phosphorus, to determine the in- crease produced by manure and p'hosphorus combined, subtract- ing from these figures the increases for manure alone, to determine the effect of phosphorus. Still another method would be to average all of the untreated plots whose results are in satisfactory agree- ment, and discard the results of those that differ so widely as to be clearly abnormal. By this method probably the results from plots Ai, 67, and Ci would be discarded. By any of these methods of comparison, direct or indirect, it will be found that, as a general average of all tests on all series, the raw phosphate has produced practically the same gross increase as the acid phosphate, although the acid phosphate applied cost twice as much as the raw phosphate. Yet another indirect method of comparison can be made and this one is preferred by the Ohio Experiment Station. This method assumes that naturally the land varies somewhat uniformly from one untreated plot to the next untreated plot, so that plot 15, for example, if it had remained unmanured, would have produced a yield equal to the sum of two thirds of the yield of plot 14 plus one third of the yield of plot 17, and that this computed yield for plot 15 (untreated), subtracted from the actual yield of plot 15, gives the increase produced by the manure. The effect of the ma- nure and phosphate is computed in the same manner, and the difference gives the effect of the phosphorus. This method would be correct if the assumption upon which it is based were correct; but considering that the change in the direc- tion of such a curve is just as likely to occur on any other plot as on the plots that happen to be numbered i, 3, 4, 7, etc., its appli- cation may be of questionable value. However, in Tables 37, 38, and 39, the actual yields are reported for the twelve years, and from these any one can make his own deductions. The total value of the three crops, based upon the average yields, USE OF PHOSPHORUS IN DIFFERENT FORMS 251 is given for each series, counting 35 cents a bushel for corn, 70 cents for wheat, and $6.00 a ton for hay. These prices 1 are based upon the ten-year average farm price for Illinois, as reported by the United States Department of Agriculture for the years 1899 to 1908, for which the reported averages are 40.1 cents a bushel for corn, 76.5 cents for wheat, and $9.32 a ton for marketable hay. The differences between these averages and the prices used in the tables will probably cover the cost of husking corn and threshing wheat, stacking and baling the hay, and marketing the increase. The value of the increase in corn stover and wheat straw may per- haps cover the extra cost of handling (binding twine, etc.) and occasional losses for poor quality of grain and hay. The prices used are intended to be sufficiently conservative to guard against financial exaggeration. Other prices should be used to suit local conditions. The special purpose of reducing all results to the basis of value is to make possible a more simple comparison. From these total values of the three crops by plots, the gain for treatment is computed by the Ohio method, except that in Series C the results from plot i are discarded 2 as being plainly abnormal and untrustworthy. A comparison of the value of crops grown on plots A2, A3, and A5, on plots B2, 63, and 65, and on plots C2, 03, and €5, plainly indicates that plot C2 is normal; and the effect of manure and phosphorus on plots C2 and 03 is determined by subtracting the average results of plots €4 and 07. It will be noted that the cost of raw phosphate is reckoned at $7.50 per ton and the cost of acid phosphate at $15 per ton, or $1.20 for 320 pounds of raw phosphate and $2.40 for 320 pounds of acid phosphate, applied with the 8 tons of manure every three years. Three important facts are clearly established by these data: (i) the value of manure, (2) the superiority of stall manure over 1 Elsewhere the author uses 30 cents a bushel for oats, the lo-year average price for Illinois being 32.2 cents; 40 cents a bushel for barley, 44.7 cents being the lo-year average price for Minnesota and Wisconsin, leading barley states; and 50 cents a bushel for potatoes, the New York lo-year average farm price being 57.6 cents. 2 A personal communication from Director Thorne states that this plot occupies a depression, running lengthwise of the plot, with higher land on each side, Evi- 252 SYSTEMS OF PERMANENT AGRICULTURE yard manure, and (3) the value of phosphorus when applied in connection with manure. The first two will be further discussed under the subject of farm manure. As an average of the results from the three series of plots, the value of the increase from 320 pounds of raw phosphate was $10. 19 with yard manure and $10.23 with stall manure; and the value of the increase from 320 pounds of acid phosphate was $11.77 with yard manure and $12.01 with stall manure, the value of the ma- nure alone having been deducted in all cases. If we subtract from these gross gains the cost of the phosphorus, we have average net profits of $9.01 for raw phosphate and $9.49 for acid phosphate; or, on the basis of money invested, we have net profits of 751 per cent from raw phosphate and 395 per cent from acid phosphate. With double the investment the profit per acre is slightly greater from acid phosphate; but, on the basis of money invested, the profit from raw phosphate is almost double that from acid phosphate. dently more or less surface wash has accumulated in this depression in times past. (See the accompanying contour map of this Ohio field.) SECTION A NORTH SECTION B SECTION C 7 8 9 10 13 U 15 16 W 18 19 20 11 12 13 H 15 16 17 18 19 20 11 12 13 14 16 16 17 18 19 2 TOPOGRAPHY OF LAND: OHIO EXPERIMENT FIELD One-foot contour lines; highest land at south end of section C. USE OF PHOSPHORUS IN DIFFERENT FORMS 253 The student or landowner must draw his own conclusions as to which is the better basis upon which to compute the profit. It should be kept in mind that 320 pounds of raw phosphate contains 40 pounds of phosphorus, while 320 pounds of acid phosphate con- tains about 20 pounds of that element, so that the raw phosphate is enriching the soil in phosphorus twice as much as the acid phos- phate, while the removal in crops is practically equal. An examination of the values of the three crops by plots suggests that the use of the data from plot Ai ($25.02) is unfavorable to the raw phosphate on that series, because of the lower uniform values from plot A4 ($20.61) and plot Ay ($20.03). On the other hand, the use of the data from plot By is favorable to the acid phosphate, by the Ohio method of comparison. By the method of direct comparison, by which the total values from plots 15 and 16 (manure alone) are subtracted from the total values from plots 2 and 3 and from plots 5 and 6, respectively, it will be seen that the average value of the increase is $10.59 fr°m 320 pounds of raw phosphate, and $10.61 from 320 pounds of acid phosphate, the net profit per acre being $9.39 for the raw phosphate and $8.21 for the acid phosphate; or, on the basis of money in- vested, the net profit is 783 per cent for raw phosphate and 342 per cent for acid phosphate. For convenient reference, the aver- age actual yields and values are summarized in the accompanying tabular statement. TABLE 396. OHIO EXPERIMENTS WITH MANURE, RAW ROCK PHOSPHATE, AND ACID PHOSPHATE Average of Twelve Years, with Duplicate Tests on Each Field SOIL TREATMENT FIELD A FIELD B FIELD C AVERAGE Manure alone 47.2 6?. 6 '?'?.'* l?4.7 Manure and rock phosphate ^6.4 60. <; S8.2 61.4 Manure and acid phosphate ^4.6 70.8 62.O 6?.i WHEAT, BUSHELS PER ACE E Manure alone 2O 4 21 7 17 2 19.8 Manure and rock phosphate Manure and acid phosphate 23-4 23.8 3°-4 29.4 24.6 25-1 26.1 26.1 254 SYSTEMS OF PERMANENT AGRICULTURE TABLE 396. OHIO EXPERIMENTS WITH MANURE, RAW ROCK PHOSPHATE, AND ACID PHOSPHATE — Continued SOIL TREATMENT FIELD A FIELD B FIELD C AVERAGE CLOVER HAY, TONS PER ACRE Manure alone i.oq 1.34 •83 1.30 Manure and rock phosphate .... 2.47 I.QO 1.70 2.05 Manure and acid phosphate 2.23 1.76 1. 02 1. 07 TOTAL VALUE OF THE THREE CROPS PER ACRE IVIanure alone .... $42.60 $45.47 $35. 6q $41.27 Manure and rock phosphate .... Manure and acid phosphate 50.86 40.12 56.42 ^.76 48.30 5:0.76 51.86 51.88 Cost of rock phosphate for the three crops . $I.2O Cost of acid phosphate for the three crops . 2.4.O It is worth while to note that the first corn crop on Series C (Table 39) was not benefited by raw phosphate, and the first corn crop on Series B (Table 38) was increased only 2.7 bushels, as an average, by raw phosphate, while other instances appear in which phosphorus produced no apparent benefit, as, for example, with stall manure for corn in 1901 and 1906, and with either manure for wheat in 1907, all of which emphasizes the fact that one field trial with one crop for one year may have almost no value in de- termining the effect of additions of phosphorus to the soil. Director Thorne has expressed some disappointment l because 1 Considering its source, the following statement by Director Thorne, taken by itself, probably constitutes the strongest "evidence" that can be quoted in favor of acid phosphate and against the use of raw rock phosphate. In referring to the general averages of all results secured in these manure-phosphate experiments from 1897 to 1907, he says (Ohio Agricultural Experiment Station Circular 83, page 23) : ''While the treatment of manure has in every case increased its effectiveness, the gain per acre produced by reenforcing the manure with acid phosphate has been so much greater than that from any other treatment that it has not been profitable to use anything else, even though the other materials had cost nothing. " Of course, in this statement, Director Thorne refers not to profit on investment, but to profit per acre regardless of the amount invested, and he includes the data CHARLES E. THORNE, DIRECTOR OF OHIO AGRICULTURAL EXPERIMENT STATION USE OF PHOSPHORUS IN DIFFERENT FORMS 255 the 40 pounds of phosphorus applied in raw phosphate has not pro- duced markedly greater benefit than the 20 pounds in acid phos- phate, these applications having been repeated in connection with manure every three years for twelve years. This is an important and interesting question. It may be best considered in connection with the general average yields recorded in Table 40, which, it may be observed, includes results from the use of kainit, gypsum, and " complete " fertilizers. The amounts of kainit and gypsum used are the same as raw phosphate and acid phosphate; namely, 320 pounds with 8 tons of manure per acre every three years. The kainit costs about $15 per ton and the gypsum about $6 per ton. The footnotes to Table 40 give further data, so that any one may make his own computations concerning the increase in yield or profit from every kind of treatment. Attention is called to the fact that plot n is a continuation of plot i, and on Series C it is so abnormal that its influence is seen in the general average of every crop. By computations from Tables 23 and 40 it is a simple matter to construct most of Table 41, in which a balance is shown for the elements nitrogen and phosphorus supplied and removed in these experiments with manure and phosphates. The figures given in Table 40 may be considered as approxi- mately correct, but the amounts of nitrogen furnished by the soil and by the clover residues are roughly estimated. This estimate is based upon the assumption that the total clover tops, aside from the clover hay harvested, will be equivalent to one half of the regu- lar hay crop. These residues consist of (i) the first season's growth, chiefly after the wheat harvest; (2) the fall growth the second season ; and (3) the following spring growth before plowing for corn. These estimates are added to Table 40, not as well-established facts, but rather as suggesting methods of study that deserve further investigation. To one familiar with field conditions it seems certain that the clover is given at least all credit due for from plot Ci in computing the increase produced by the raw phosphate by the Ohio method of comparison. As a suitable topic for a debating society, the author suggests the question: Shall we use acid phosphate or raw rock phosphate in systems of permanent agriculture? 256 SYSTEMS OF PERMANENT AGRICULTURE TABLE 40. CROP YIELDS PER ACRE IN OHIO EXPERIMENTS WITH MANURE, PHOSPHATES, KAINIT, GYPSUM, AND "COMPLETE" FERTILIZERS Average of Three Series PLOT1 No. TREATMENT APPLIED CORN, 12 YEARS WHEAT, ii YEARS HAY, 8 YEARS (Tons) Grain (Bu.) Stover Tons) Grain (Bu.) Straw (Tons) I 2 3 4 5 6 7 8 9 10 37-2 59-4 63.3 31.0 60.3 64.4 30.8 54-6 60. i 32-9 I. II 1.66 1.78 I.OO 1.64 1.74 •99 1.58 i-7S I.OO 12. 1 25.0 26.4 IO-4 25-3 26.2 9-7 21.3 23-4 10.3 •74 1-35 i-44 .61 1.36 1.44 •58 1. 20 i-35 .60 I.I9 1.90 2.19 .89 1.77 2.17 •85 1.49 1.94 •95 Yard manure and raw phosphate . . Stall manure and raw phosphate . . Yard Manure and Acid phosphate . Stall manure and acid phosphate . . Yard manure and kainit Stall manure and kainit None ii 12 13 14 IS 16 17 18 J9 20 None 36.8 58.0 60.7 31.6 51.2 58.2 36.6 43-1 44.4 34-i 1.17 1.68 1.78 I.OO 1.44 1.65 1-15 1.29 1.23 1. 01 i3-x 23-4 23-3 9-9 18.8 20.4 IO.2 13-4 14.9 IO.O .81 1.32 I-3I •57 i. 06 1.14 .62 .78 .88 .62 1.30 1.63 1.65 •85 1.28 1.63 I.OO 1.36 i-43 I.IO Yard manure and gypsum .... Stall manure and gypsum .... None Yard manure alone Stall manure alone None " Complete " fertilizer 2 "Complete" fertilizer3 None 1 In the field, plots i and ii, 2 and 12, etc., lie end-to-end, plot n being essen- tially a continuation of plot i, etc. 2 The "complete" fertilizer applied to plot 18, every three years, contains 160 pounds sodium nitrate, 80 pounds acid phosphate, and 80 pounds potassium chlorid. 3 The "complete" fertilizer applied to plot 19, every three years, consists of too pounds of slaughter-house tankage (containing 6 pounds of nitrogen and 6 pounds of phosphorus), 80 pounds of acid phosphate, and 10 pounds of potassium chlorid. USE OF PHOSPHORUS IN DIFFERENT FORMS 257 TABLE 41. BALANCE SHEET FOR NITROGEN AND PHOSPHORUS IN MANURE- PHOSPHATE EXPERIMENTS Totals for Three Years, Pounds per Acre: in Part roughly Estimated Plot No 3 3 5 6 TREATMENT APPLIED YARD MANURE, RAW PHOS- PHATE STALL MANURE, RAW PHOS- PHATE YARD MANURE, ACID PHOS- PHATE STALL MANURE, ACID PHOS- PHATE Nitrogen removed in three crops Nitrogen supplied in manure 211 80 232 80 207 80 23I 80 Nitrogen difference 131 1^2 127 ICI Nitrogen in clover hay 76 88 7* 87 Nitrogen from soil and clover residues . . . Nitrogen from clover residues (estimated) . . 55 38 64 44 56 36 64 44 Nitrogen furnished by soil (estimated) . . . 17 20 2O 20 Phosphorus removed in three crops .... Phosphorus supplied in manure and phosphate 31 64 34 56 31 44 34 36 Phosphorus added in excess ?7 22 1 3 2 nitrogen fixation. In other words, that the draft upon the soil by the crops grown is likely to be greater rather than less than 20 pounds above that supplied by the manure and clover, and, in addition to this, there are losses of nitrogen in drainage waters probably exceeding all other additions (as in rainwater, by azo- tobacter, etc.). The loss of nitrogen by drainage is no doubt much greater from the best-treated plots than from the untreated plots. On the whole, it seems clear that nitrogen must limit the crop yields on these four plots treated with manure and phosphate. On the other hand, in every case the phosphorus applied exceeds the amount removed in the crops, so that, instead of there being any draft upon the soil, there is a positive increase in the phosphorus content of the soil above the crop requirements. This increase varies from 2 and 13 pounds with acid phosphate to 22 and 33 pounds with raw phosphate. If nitrogen is the limiting element on all of these manure-phosphate plots, it is plain to see why the raw phosphate gives practically no larger yields than the acid phos- 258 SYSTEMS OF PERMANENT AGRICULTURE phate, even though twice as much phosphorus is applied in the raw phosphate as in the acid phosphate. If more clover were plowed under or if more manure were returned, so as to remove the nitro- gen limit, the comparative value of the two forms of phosphorus could, perhaps, be more definitely determined. Such additional supplies of decaying organic matter would tend to make avail- able still larger supplies of potassium, magnesium, etc., and thus to avoid their becoming limiting factors. It is possible that the use of acid phosphate tends to prevent loss of ammonia from the manure during the few weeks that elapse between the mixing of the phosphate with manure and the application to the land. Two important facts are well established by these Ohio experi- ments: First, that fine-ground natural rock phosphate is a material that can be employed with very large profit as a phosphorus fertilizer, when used in connection with liberal amounts of decaying organic matter; and, second, that, under the conditions of these experiments, the raw phosphate gave practically the same profit per acre, and twice as much profit for the money invested, as the acid phosphate. In the Ohio Farmer for August 22, 1908, Director Thome re- ports some interesting and valuable results showing the effect of raw phosphate on clover grown in 1908 on the Strongsville Experi- ment Farm, located between Wooster and Cleveland, on a heavier type of soil of nearly level topography. The author has also been given the figures for the 1908 oat crop. Director Thorne states that lime and raw phosphate were ap- plied across the plots in the five-year rotation, " dividing the sec- tion of plots into 4 divisions, using one ton of lime per acre on the first, two tons of lime on the second, one ton of floats on the third, and two tons of floats on the fourth, applying the lime and floats across all the plots, fertilized and unfertilized alike." The crops grown in the five-year rotation experiments at Strongs- ville are corn, oats, wheat, clover, and timothy, and the fertilizers applied are similar to those in the older five-year rotation at Woos- ter as reported in Table 82, with ten additional plots in'each series. It is understood that during the course of five years all of the series have received (or will receive) the treatment with lime and raw phosphate, as above described. USE OF PHOSPHORUS IN DIFFERENT FORMS 259 The data thus far reported concern only the clover and oats for 1908, and they almost certainly show more marked differences than will appear from long-continued and more general experiments : OHIO EXPERIMENTS WITH LIME AND RAW PHOSPHATE SPECIAL SOIL TREATMENT APPLIED, TONS PER ACRE YIELDS, PER ACRE, 1908 Clover Hay (Lb.) Oats (Bu.) (a) Average of 14 Otherwise Unfertilized Plots One ton of lime 2220 72 7 Two tons of lime 2670 7X 7 One ton of raw phosphate 5040 47-8 Two tons of raw phosphate CIQO "^.O (b) Average of 26 Otherwise Fertilized Plots One ton of lime 2700 43 8 Two tons of lime 3880 4.C.Q One ton of raw phosphate ^460 $2.7 Two tons of raw phosphate "^o 62 2 Even if we assume that the lime produced no increase, the effect of the raw phosphate is very marked. On the " otherwise unfer- tilized " land one ton of raw phosphate per acre produced 1.41 tons more clover and 15.1 bushels more oats, the value of which, at $6 per ton and 30 cents a bushel, respectively, would amount to $12.99, or enough to pay for two tons of raw phosphate at $3.50 per ton where the freight rate does not exceed $3 (as in southern Illinois). The two-ton application produced markedly greater effect on the oats, but not on the clover. These results include two entirely separate series of plots, and each reported yield is the average from a large number of plots. The data must not be considered as ex- ceedingly trustworthy, but we must agree with Director Thome's statement that " it looks as though we were getting something worth while there." Knowing that the effect of lime where needed 260 SYSTEMS OF PERMANENT AGRICULTURE is usually marked on both clover and oats, the " increase " given for phosphate must be regarded as quite unusual. Not more than one tenth of the phosphorus applied would be removed in the two crops. These results point toward the possibility of adopting profitable systems of permanent agriculture; and yet the most common fertilizer practice among the farmers of Ohio is to apply about 7 pounds of soluble phosphorus per acre, with 3 or 4 pounds each of nitrogen and potassium, in 200 pounds of "complete" fertilizer about twice during a four-year or five-year rotation. In the Rural New Yorker for June 5, 1909,' Director Thorne reports an average yield of 2440 pounds of clover hay with lime applied, and 5112 pounds where rock phosphate is used, with no other fertilizers; and where nitrogen or potassium had been applied, the yield with lime was 2606 pounds and with rock phosphate, 5488 pounds, although where lime and a " complete " fertilizer, including nitrogen, phosphorus, and 480 pounds of acid phosphate, was used, the total yield of clover hay was only 4259 pounds. The following comment is made by Thorne: "This experiment thus indicates that floats (raw phosphate) may be very usefully employed for the combined purpose of carrying lime and phosphorus, the increase over the limed land being more than enough in this one crop to pay for one ton of floats per acre, which quantity has seemed to be as effective as the larger quantity, although the two tons of lime have produced a larger yield than the one ton, though not enough larger to pay the additional cost. This experiment, therefore, is confirming those of the Maryland and Illinois stations in showing that floats may be profitably used as a carrier of phos- phorus on acid soils well stocked with organic matter, but the meager effect produced upon cereal crops preceding clover would call for caution in de- pending upon floats alone. " Of course, more clover means more humus and more nitrogen if the clover is plowed under either directly or in manure; and, while the cereal crops preceding clover are quite certain to be un- satisfactory, they are sure to be increased after plowing under the larger amount of clover or manure with phosphate. The data reported by Thorne in these experiments do not show what increase was produced by lime alone, all plots having been treated either with lime or with raw phosphate, and consequently there seems to be no support for the suggestion that the effect of the raw phos- phate is in part due to the lime which it carries. In fact, the ordi- USE OF PHOSPHORUS IN DIFFERENT FORMS 261 nary high-grade raw phosphate carries very little, if any, lime, about 7 per cent of calcium carbonate being the largest amount in any high-grade phosphate known to the author. Tricalcium phosphate is a neutral substance which has practically no power to correct soil acidity, except as the phosphorus is converted into the dicalcium or monocalcium compound and removed from the soil by the growing crop. Furthermore, the most marked benefit from the use of raw rock phosphate in Illinois is not on markedly acid soils, but on the most common corn-belt prairie land, as on the Urbana and Gales- burg experiment fields on brown silt loams, which are practically neutral soils valued at $150 to $200 an acre. The Ohio investigations with raw and acid phosphates are in a class by themselves. No others have been conducted anywhere in the world that can compare with these in agricultural value. Many experiments with various phosphates have been carried on for a single season, and some for several years, but, as a rule, no farm manure has been used, and no adequate provision made for a supply of decaying organic matter. Where nitrogen has been sup- plied, it has usually been in some commercial form, such as sodium nitrate. One exception to this is found in the Maryland experi- ments, which have been conducted on one field since 1895, eleven years' results having been reported by Director Patterson (Mary- land Bulletin 114). Aside from single plots treated with different acid phosphates and reverted phosphates in the Crimson Clover Series, there is a strictly comparable triplicate test with (i) raw bone meal, (2) slag phosphate, (3) no phosphate, (4) South Caro- lina raw rock phosphate, and (5) Florida soft rock phosphate (containing phosphates of iron and aluminum). Equal amounts of phosphorus were used in all tests, 65^ pounds of phosphorus per acre having been applied only at the beginning of the experi- ment. The surface soil contained 1300 pounds of phosphorus in 2 million pounds of the soil. The following crops were grown: Corn in 1895. Wheat in 1899. Corn in 1903. Corn in 1896. Hay in 1900. Wheat in 1904. Corn in 1897. Hay in 1901. Hay in 1905. Crop failure in 1898. Corn in 1902. Corn in 1906. 262 SYSTEMS OF PERMANENT AGRICULTURE In one series of tests, crimson clover was regularly seeded in the corn to be plowed under later as a green manure, and in another series rye was employed in a similar manner, the third test being made without special provision for organic matter. Table 42 gives a summary of these investigations for all forms of phosphorus that were used under the three different conditions: TABLE 42. MARYLAND EXPERIMENTS WITH DIFFERENT FORMS OF PHOSPHORUS Twelve Years' Work : Yields per Acre : Average of Three Plots Six CORN CROPS, Av. Two WHEAT CROPS, Av. THREE HAY TOTAL AVER- TW XTV\ T}rrf\nm-rS\Ttrtr- A Tin* -TK>T*. Av. YIELD Grain Stover Grain Stover (Tons) (Tons) (Bu.) (Tons) (Bu.) (Tons) 8, 13, 18 Raw bone meal . . . 39-6 1.25 23.6 1.22 I.8S 6.41 9, 14, 19 Slag phosphate . . . 39-i 1.22 22.6 1.24 1-95 6.46 10, 15, 20 No phosphorus 40.0 I.I7 12. 1 •73 1.44 5.10 II, l6, 21 S. C. rock phosphate 39-7 1.95 2O. I 1.07 1-95 6.26 12, 17, 22 Florida soft rock . . . 42-5 1.27 19.9 •94 1.89 6.19 Here are represented thirty-three separate tests (three plots for eleven years) for each form of phosphorus. As an average, the raw rock has given nearly the same results as the bone and slag. The average increase in yield is very marked with wheat, less marked with hay, and practically no effect is seen with corn. The value of the total increase in twelve years is about ten times the cost of the raw rock phosphate, at Illinois prices, and still more at Mary- land prices, for farm produce. In commenting upon his experi- ments, Director Patterson says: "The results obtained with the insoluble phosphates has cost usually less than one half as much as that with the soluble phosphates. "The results show decidedly that plants are able to utilize insoluble rock phosphates. "The use of an abundance of organic matter in the soil when insoluble phosphates are applied was evidently a necessity for their best effects. "Soluble phosphates produced the best yield of wheat. 'Florida soft phosphate produced the best yield of corn. "Reverted phosphates produced the best yield of hay. USE OF PHOSPHORUS IN DIFFERENT FORMS 263 "Insoluble South Carolina phosphate rock produced a higher total average yield than dissolved South Carolina rock. "Florida soft phosphate is chiefly an aluminum-iron phosphate which occurs in large quantities deposited in many parts of that state. It is not well adapted to treatment with acid for making soluble phosphates, as the aluminum and iron make a sticky mass which is hard to dry and keep in a good mechan- ical condition. The Florida soft rock has been largely used as a fertilizer in its natural condition in some parts of that state on the light, sandy land, giving good results. When used in this way, there has been applied at the same time heavy dressings of the native mucks from the swamps and lakes. This muck furnishes nitrogen as well as the much-needed organic matter. In order to have a complete fertilizer, there is also applied some German potash l salt. " The Pennsylvania Experiment Station has reported the results of an experiment extending over twelve years (1884 to 1895), in which four different kinds of phosphorus were used in a four-year rotation of corn, oats, wheat, and hay (clover and timothy). Only one field was employed, so that each crop was grown only three times during the twelve years. The four forms of phosphorus were (i) acid phosphate made from bone black, (2) " reverted " phosphate made by mixing equal weights of dissolved bone black and quicklime twelve hours before application, (3) fine-ground bone meal (containing 8 pounds of nitrogen and 35 pounds of phosphorus in 300 pounds of bone), and South Carolina ground raw rock phosphate. The amounts applied per acre in each four years were 28 pounds of soluble and " reverted " phosphorus, and 35 pounds in bone and raw rock. No special provision was made for supplying decaying organic matter, but 94 pounds of nitrogen (102 pounds on the bone-meal plots) and 83 pounds of potassium (in potassium chlorid) were applied per acre, each four years, to all phosphorus plots and also to two comparison plots that received no phosphorus. In addition, there were two plots that received no application of plant food. The entire experiment was carried on in duplicate. One half of the fertilizer for the rotation was applied to the corn crop and the other half to the wheat crop. Table 433 gives the average yields of all products harvested dur- 1 Most peat soils and some sands are extremely deficient in potassium, and it is also difficult to liberate potassium that may exist locked up in coarse sand grains. — C. G.H. 264 SYSTEMS OF PERMANENT AGRICULTURE ing the entire twelve years, and Table 43& gives the average re- sults for the last four years. TABLE 43. PENNSYLVANIA EXPERIMENTS WITH DIFFERENT FORMS OF PHOS- PHORUS: YIELDS PER ACRE: AVERAGE OF DUPLICATE PLOTS (a) Average of Twelve Years' Work CORN, OATS, WHEAT, ^ 3-YEAR 3-YEAR 3-YEAR AVERAGE AVERAGE AVERAGE >H— ' PLOT PLANT FOOD APPLIED *>g Grain Straw Grain Straw Grain Straw - CO oo to H coso sO H H III to to CO Os H O sO Os H O H CS) tosO CO CO H so O to sO sO M" to t O O co rj- coso >-< ^O ^O .-: H CS CS) H H O IO ^ H t>^ H sO Os Os IO t-» Csl O W H) III 00 cs) so O 00 OssO to t^ Os CS) Csl csi O CO CS) Os Os to l^ H H T}- H CO OssO to O co H H H 111 H H OO so rj- OO csi to t^- H so OO Os O\ ^t co H to Os ^J- TJ- H O O sO O O co to H CS) CS] H H O CO to sO sO IO H ^j1 H Os CO H H 'T. ( > • . . . nj 1 3 • ' 4 3 *> I • 1 i ' 'i ) I •<* • • (3 • • • • T a> 3 • • •} 5 E K . . t, ;l •] n u^ a> ... "o <3 • OTJ ' ' < r*r a3 u '"~s s"*- I I •>. 5 vli'g FORM OF PHOSPHOR: O ^ ^-^ ij Os Q "K si S5 ? %£ W U OO o w (3 1" \f^ to so O O H H HI H H H H •lll^ll 3 o, cu HV! tn^3SC3^.Ui Ipsji^ 3'fi1 ™ ill- ftp si CJCJcS USE OF PHOSPHORUS IN DIFFERENT FORMS 269 H § « £ w o & S Q •« I— 1 • M MM PO M M O M M M life to ^3" NO ^" NO to (N ON ^}" t^^ ^j" lo t^* xo MM ^ O ^ to M OO ON to W mNO r\ ON fqrotoO^m- '^- M N 0) M M ON M ON ON m O \o moo 1 y OO t"^ M ON ON ONOO ^O ON ON CS M to 00 M O tONO O N ^OO M T}~ fO t*^ ON to PO • to M CS M M M **"* M M IO O *^- H 00 to O NO M M Q s§< < o w C/} M OO OO ON OO OO OO CS M *J- M rfr 10 M MTfl^-MONONro M ro^fOMCNirf . \O M CS M M M O M M OO NO M IO M M agS MM (N M M „ M 00 M Tj- t^» Tj- ON w j ON <"O M O ^ W CS \o oo o oo to *^t* ci CS M M Tf OO ^ IO NO O ON "^J" O M 1O cfl Tj-otonoitoro OO M N CN M M fq M o » m ONOO oo oo oo lOOO ON «>• N ON N M CS1O'4'rl'P)NOpJ NO t*» ONNO M 00 ^t O M M FORM OF PHOSPHORUS APPLIED V 2 "H ' 5 ^ .^ ^O . . . ! I ! . -o • 0 <^ bD v •^X, , s ™ Q C/3 s ^ ^_j 0s1 O C/3 cd O ^ ^ c^ O WU OO o wu Tt Tf to to ^O O O ON ON ON ON ON O O 00 00 00 00 00 ON ON 270 SYSTEMS OF PERMANENT AGRICULTURE A study of Table 44 shows a decrease in ear corn and a small increase in corn stover from the use of raw calcium phosphate the first year, and some increase in both oats and oat straw the second year. During the next four years this raw rock phosphate produced a larger average increase in the yield of hay than any other form of phosphorus applied, except steamed bone meal. This suggests that the longer growing biennial and perennial plants, such as clover and timothy, may be better able to utilize the raw phosphate than the short-lived annuals. It is important to keep in mind also that these four years con- stitute a considerable part of the entire time of the experiment, and that it was only during these four years that the investigations have the greatest practical significance, because the first two years would be required to get the phosphates thoroughly incorporated with the soil and get well under way the action of the various agencies that help to make the raw phosphates available, and it was only during the first six years that equal money values of the different phosphates were used. As stated by the Rhode Island Station, " Deherain and other French writers recommend that, upon acid soils, such untreated phosphates should be applied several months or a year before liming is resorted to, so as to secure as great a decomposing action upon them by the soil as possible." In 1900, the three largest yields of ear corn were produced by steamed bone meal, raw calcium phosphate, and roasted aluminum phosphate, in the order named. The results from the several crops grown in 1901 are reported as the weight of the fresh or green crops. It will be seen that the raw calcium phosphate produced some increase in eleven of the twelve crops reported, and the average increase from this raw rock phosphate is more than three fourths as much as from the common acid phosphate costing two or three times as much for the appli- cations made previous to 1901. The relative effects of the different phosphates are about the same on the unlimed land as where some lime had been applied, except that the superiority of the slag phosphate, steamed bone meal, and common raw rock phosphate (calcium phosphate) over the four acid phosphates (including superphosphate) was even more marked in the four years' hay crops on the unlimed land. USE OF PHOSPHORUS IN DIFFERENT FORMS 271 As an average, the raw calcium phosphate produced more than 90 per cent as much increase as the common acid phosphate in the various crops grown in 1901 on the unlimed land. The value of the increase produced by the raw calcium phosphate in the hay crops alone is twice the cost of all the phosphorus ap- plied in this form during the eight years. The value of the lime in the slag phosphate is indicated espe- cially in the increase in hay on the unlimed land. The aluminum phosphates (which also contain some iron phosphate) gave much poorer results than the raw calcium phosphate; but no final con- clusions should be drawn regarding this, because the aluminum phosphate may not have been as finely ground as the common rock phosphate, which in these experiments was applied as " floats," the dust that collects about phosphate mills. There is evidently no advantage from roasting the aluminum-iron phosphate (Re- dondite). The somewhat poorer results obtained with the double super- phosphate, as compared with the other three acidulated phosphates, is probably due to the manufactured land-plaster (calcium sulfate), which is a powerful soil stimulant, and which as already explained constitutes about 50 per cent of ordinary acid phosphate. It will be noted that the lime itself more than doubled the yield of hay as an average of all plots, and also increased the yield of corn. This Rhode Island soil is acid, and for most crops is markedly improved by liming. In commenting upon these experiments, Director Wheeler says (Rhode Island Bulletin 114): "With the pea, oat, summer squash, crimson clover, Japanese millet (on the unlimed land), golden millet, white-podded Adzuki bean, soy bean, and potato (on the unlimed land), floats (raw calcium phosphates) gave very good results; but with the flat turnip, table beet, and cabbage they were relatively very inefficient." "The use of fine-ground bone, basic slag meal, and floats has tended con- tinually to make the unlimed land more favorable to clover, as is well shown by its appearance only upon those plots of the unlimed series where these phosphates had been used, while it was absolutely lacking where the raw and roasted Redondite and the soluble phosphates had been applied. Upon the limed land, clover has been uniformly common upon all the plots." "Floats can probably be used to best advantage on moist soil, rich in decay- 272 SYSTEMS OF PERMANENT AGRICULTURE ing vegetable matter, and for such crops as certain legumes, Indian corn, millet, and possibly wheat and oats, which seem far better able to make use of them than certain vegetables." In Tables 440 and 4$c are recorded the results obtained in the continuation of these Rhode Island experiments, with soy beans in 1902, with nineteen different kinds of plants in 1903 (varying from i row of spinach and 2 rows of lettuce to 10 rows of barley and 16 rows of oats), and with oats in 1904. The heavy applications made in the spring of 1902, amounting to 1426 pounds of acid bone black, 1738 pounds of acid bone meal, and 1771 pounds per acre of acid phosphate, with no additional application of raw calcium phosphate, render the subsequent crop yields of less economic importance, in the author's opinion, but they are of interest because of the great variety of plants repre- sented, although the data are not sufficient to justify very definite conclusions. The following comments on the results of 1902, 1903, and 1904, are given in Rhode Island Bulletin 118, page 84: "Floats (raw calcium phosphate) gave very good results with the soy beans, peas, crimson clover, mangel-wurzel (on limed land), barley (on limed land), potato (on unlimed land), Japanese millet, oats, and golden millet; but they proved highly inefficient, especially for Hubbard squash, rutabaga, crookneck squash, flat turnip, cabbage, mangel-wurzel (on the acid unlimed land), tomato, lettuce, New Zealand spinach, and red valentine bean." One of the oldest known facts concerning plant nutrition is the weak power of turnips and other plants of the cabbage family (Crucifferae) to secure phosphorus from insoluble forms. Thus, almost the first important result of Sir John Lawes' agricultural experiments was the discovery, seventy years ago, that dissolved bone black was very much more efficient than the untreated ma- terial for the production of turnips. In no case in the Rhode Island results for 1902, 1903, or 1904, with soy beans, crimson clover, millet, or oats (representing the farm grains, grasses, and legumes) was the increase from acid phos- phate double the increase from raw calcium phosphate, and as an average of the results with these crops (on limed or on unlimed plots) the increase from acid phosphate was not more than i| times that from the raw calcium phosphate, although the cost of USE OF PHOSPHORUS IN DIFFERENT FORMS 273 |Jil O\O 00 \O O w H O OO^OMOOVOOOO O'O ~ OOOOOO OONOM 1^-t^-OsOO OLO Os O H t^-OO O\ M O\ O 00 O OOvtocMO CMOO^OMPO fOvO •* CM t^ O\ CMO\^-O\<^ vOMO^CMtO^CMM M M MMMMCMMM , " USE OF PHOSPHORUS IN DIFFERENT FORMS 275 the acid phosphate applied is three times that of the raw rock; and it should be kept in mind that no provision was made to keep the soil supplied with decaying organic matter, although nitrogen and potassium, in commercial form, were applied to all plots alike. If we add together all of the grain and hay produced during the decade following the first year of the experiment, including the oat grain in 1895, the hay in 1896, 1897, 1898, 1899, and 1904, the ear corn in 1900, and the soy beans in 1902, we secure the following totals for the plots designated : SOIL TREATMENT No PHOSPHATE ROCK PHOSPHATE (CA) ACID PHOSPHATE Unlimed Limed Unlimed Limed Unlimed Limed Pounds per acre . . . Gain for phosphorus Gain for lime .... 83IO 27470 22890 14580 35340 7870 12450 22860 14550 37000 9530 14140 19160 These figures present in very concise form an economic summary of the Rhode Island experiments with " floats" and acid phosphate, as applied to the more valuable produce of the general farm crops grown during the ten years. The acid phosphate gave slightly poorer results than the raw rock on the unlimed land and 20 per cent better results on the limed land. The value of lime is also strikingly shown. It should be kept in mind, however, that the more abundant growth of clover upon the limed land during the four years, 1896 to 1899, would likely benefit succeeding crops, irrespective of the lime itself. The Maine Experiment Station reports two series of experiments with different phosphates, one covering a period of nine years with all tests in triplicate on 2oth-acre plots where equal amounts of phosphorus were compared, and the other for five years (1890 to 1894) on 2^-acre plots where equal money values of phosphorus were compared. In the nine-year experiments, fertilizers were applied five times — in 1886, 1887, 1889, 1893, and 1894. When applied, the amounts per acre were 200 pounds of ammonium sulfate, 100 pounds of potassium chlorid, 360 pounds of fine-ground bone, 300 pounds of 276 SYSTEMS OF PERMANENT AGRICULTURE fine-ground South Carolina raw rock phosphate; and, for soluble phosphorus, 400 pounds of acidulated bone black were used for 1886, 1887, and 1889, and 500 pounds of acid phosphate made from South Carolina rock for 1893 and 1894. The stable manure was applied five times at the rate of 20 tons per acre. The results are reported in Table 46 for each of the eight crops harvested. TABLE 46. MAINE EXPERIMENTS WITH DIFFERENT PHOSPHATES Pounds per Acre of Air-dry Produce : Average of Three Plots in Each Case TREATMENT APPLIED NONE NK AND ACID PHOS- PHATE NK AND BONE MEAL NK AND RAW PHOS- PHATE NK ONLY STABLE MANURE 1886 Oats, grain . . . 1886 Oat straw . . . 1670 1994 2486 34M 2286 3134 2l66 2886 1936 2564 22l6 3050 1887 Oats, grain . . . 1887 Oat straw . . . 800 I20O 1160 2240 956 1610 1064 2036 1052 1648 1014 2060 1888 Hay 1889 Fallow .... 1890 Peas, grain . . . 1890 Pea straw . . . 2566 2434 2800 2566 2234 4010 742 664 902 948 946 976 848 914 762 660 1360 1284 1891 Oats, grain . . . 1891 Oat straw . . . 1166 726 1346 986 1376 1090 1160 776 1296 704 1542 1746 1892 Peas, grain . . . 1892 Pea straw . . . 468 748 368 756 376 696 308 552 440 740 588 1388 1893 Corn, total . . . 1894 Corn, total . . . 395 749 1415 2926 1326 3038 1076 2631 905 1879 2931 3562 Total yield, 8 crops . . Total increase . . . Gain for phosphorus 13888 21381 7493 456i 20610 6722 379° 18983 5095 2163 16820 2932 26751 12863 Of the different forms of phosphorus, the acid phosphate gave the best results for the first two years, especially in oat straw, but afterward the bone meal gave the best results. The raw phosphate produced only about one half as much increase as the other forms, but at less than one third the cost. USE OF PHOSPHORUS IN DIFFERENT FORMS 277 In the other phosphate experiment by the Maine Station, cover- ing five years on a lo-acre field divided into four 2^-acre plots, the fertilizer applications were made but once (in 1890). The amounts applied per acre were 20 loads of stable manure on plot i, and 66 pounds of sodium nitrate, 16 pounds of ammonium sulfate, and 100 pounds of potassium chlorid (supplying only 14 pounds of nitro- gen and 42 pounds of potassium) on plots 2 and 3. In addition, plot 2 received 1000 pounds of raw rock phosphate (containing 107 pounds of phosphorus), and plot 3 received 500 pounds of acid phosphate (containing 35 pounds of phosphorus), per acre. Plot 4 received no fertilizer. Table 47 gives the results obtained for the five years of the experiment, and also the average yields of hay for two years before the fertilizers were applied. TABLE 47. MAINE EXPERIMENTS WITH EQUAL MONEY VALUES OF RAW PHOSPHATE AND ACID PHOSPHATE Pounds per Acre of Air-dry Produce PLOT No. . ..... 1 2 3 4 TREATMENT APPLIED, 1890 ONLY STABLE MANURE NK AND RAW PHOS- PHATE NK AND ACID PHOS- PHATE NONE 1888 and 1889 Hay, average yield . 2542 2416 2082 2510 1890 Barley and peas, total . . . 1891 Oats, grain 2208 I "^6 1712 1447 1422 1^22 in8 1 1QA 1891 Oat straw 2282 ie 74 I44Q 1176 1892 Barley hay .... 2444 2724 TO 2O 1161 189"? Fallow 1894 Oats, total 1804 24^7 1734 QCT7 Total yield, 1890 and 1891 . . . Total yields, 1892, 1894 .... 6O26 5338 4693 4777 4394 3664 3598 2IJ.8 278 SYSTEMS OF PERMANENT AGRICULTURE These data show that the first two years after application the acid phosphate gave about the same results as the raw phosphate, but the last two crops gave better results from the raw phosphate, even when compared with the original apparent difference in the productive power of the two plots, — a difference which may or may not hold for other crops in other years. In commenting on these experiments, Director Jordan said (Maine Report, 1894, page 31): "With the exception of the oat crop of 1891 the production of plot two has largely exceeded that of plot three. Especially is this true of the 1894 crop after the exhausting effect of three years of cropping. . . . This is certainly one instance of the unmistakable persistent influence of a crude phosphate in increasing the growth of a field crop." According to Doctor Jordan, the 20 loads of stable manure con- tained 172 pounds of nitrogen, 50 pounds of phosphorus, and 146 pounds of potassium. The Massachusetts Experiment Station has reported, with the following explanations, an experiment with different kinds of phosphates, extending over n years, 1890 to 1900 (see 9th, loth, and i3th Annual Reports) : "This series of experiments was begun by Doctor Goessman in 1890, with a view of determining whether it is not more profitable to employ one of the cheaper natural phosphates than to use the more costly acid phosphate." "The field was first divided into five plots, containing about 6600 square feet each. These plots received equal money's worth (on the basis of prices in 1890) of the phosphates used, as follows: Plot i. Phosphatic slag. Plot 2. Mona guano. Plot 3. Apatite at first; later Florida phosphate. Plot 4. South Carolina phosphate. Plot 5. Dissolved bone black. "Plot 3, as above stated, received an application of ground apatite in 1890. In 1891 it was found impossible to obtain this material, and no phosphate of any kind was applied to the plot. In 1892 and 1893 ground hard Florida phosphate was applied to this plot. It is not believed, however, that it is fair to this phosphate to compare it with the others, since it has been used only two years, while the others have been applied for four years. "From the beginning, each of these five plots has received the same applica- tion of nitrate of soda and potash-magnesia sulfate. The quantities of these applied per plot during the first four years were about 44 pounds of the former and 66 pounds of the latter. USE OF PHOSPHORUS IN DIFFERENT FORMS 279 "Since 1894 no phosphate of any kind has been applied to these plots, but the quantity of nitrate of soda and of potash-magnesia sulfate has been used in one half greater quantities. "At first Doctor Goessman included no plot on which phosphate was not used for comparison with the others. Later such a plot was added, but it was left entirely unmanured until 1896. During 1896 and 1897 it has received the nitrate of soda and potash-magnesia sulfate at the same rate as the other plots." The data (excepting from plot 3) are recorded in Table 48. TABLE 48. MASSACHUSETTS EXPERIMENTS WITH EQUAL MONEY VALUES OF DIFFERENT PHOSPHATES Pounds per Plot (about 6600 Square Feet) PLOT No PHOSPHATE APPLIED NONE SLAG PHOS- PHATE GUANO PHOS- PHATE S.C. RAW PHOS- PHATE ACID BONE BLACK 1890 Potatoes 1600 1891 Wheat, grain 67 1891 Wheat straw 313 1892 Serradella (air-dry) . 1893 Ear corn 470 1893 Corn stover 1190 1894 Barley, grain 169 1894 Barley straw 221 1895 Rye, grain 195 1895 Rye straw . . 1896 Soy beans, grain 1896 Soy bean straw 426 1897 Swede turnips (roots) .... 830 1870 1898 Corn . . . 1899 Oats .... 1900 Cabbage (heads) 103 900 1415 73 267 682 810 148 1 66 465 540 3655 830 1830 78 302 622 580 890 144 216 189 57° 1965 1500 2I2O 346 584 542 780 118 272 440 247 495 1619 Pounds per Acre 1890-1893 Phosphorus applied . . 1890-1900 Phosphorus removed None 278 124 207 121 4l6 122 142 Il6 1900 Balance not removed Phosphorus unused, per cent . 55 86 43 294 7i 26 280 SYSTEMS OF PERMANENT AGRICULTURE The record for yields for 1898 and 1899 appears not to have been published, but the report states that in 1898 the yield of corn was good upon all of these phosphate plots, and that there was but little difference between the yields of oats on the different plots in 1899. In the Report for 1900 the following summaries are made by Professor Brooks: "Taking into account all of the crops which have been grown upon this field, except the Swedish turnips (rutabaga), which were affected by disease not apparently due to the fertilizer which had been used on a portion of the plots, and the yields of which, therefore, as expressed in figures, would be mis- leading, and representing the aggregate yield which stands highest, by 100, the efficiency of the different phosphates is as follows : Phosphatic slag . . . 100.0 Ground South Carolina rock . 92.3 Dissolved bone black . . 90.7 Mona guano .... 88.3 "There was at first no no-phosphate plot used in the experiment, but we have had a no-phosphate plot since 1895. Taking into account the yields of the several plots since 1895, and excepting the Swedish turnips, which were grown in 1897, for reasons above stated, the phosphates have the following relative rank: South Carolina rock phosphate 100.0 Phosphatic slag . . .99.0 Dissolved bone black . . 97.7 Mona guano .... 95.4 No phosphate . . . 55.4 "The following conclusions appear to be justified by the results which we have obtained: "It is possible to produce profitable crops of most kinds by liberal use of natural phosphates, and in a long series of years there might be a considerable money saving in depending, at least in part, upon these rather than upon the higher priced dissolved phosphates." "Between ground South Carolina rock, Mona guano, and the phosphatic slag there is no considerable difference in the economic result." It will be seen that the South Carolina rock phosphate produced larger yields than the dissolved bone black with all of the fourteen different crop products reported, excepting potatoes the first year, wheat straw the second year, and barley straw the fifth year. It should be kept in mind, too, that no adequate provision was made for supplying decaying organic matter to this soil during the eleven USE OF PHOSPHORUS IN DIFFERENT FORMS 281 years of the experiment, and we have the following statement from the Massachusetts Report for 1896, page 190, concerning the earlier history of this field: "Previous to 1887 it was used as a meadow, which was well worn out at that time, yielding but a scanty crop of English hay. During the autumn of 1887 the sod was turned under and left in that state over winter. It was decided to prepare the field for special experiments with phosphates by systematic ex- haustion of its inherent resources of plant food. For this reason no manurial matter of any description was applied during the years 1887, 1888, and 1889. "The soil, a fair sandy loam, was carefully prepared every year by plowing during the fall and in the spring, to improve its mechanical condition ; during the same period a crop was raised every year." A second series of experiments with different phosphates was begun by the Massachusetts Station in 1897, upon thirteen plots of land that had all received 600 pounds of bone meal per acre in 1896. In this series equal amounts of phosphorus are being applied in ten different phosphates. The results thus far reported are variable and inconclusive. In some cases soluble phosphates have produced the best yields, especially upon garden vegetables, while in some other cases the raw phosphates have given better results. Of course the 600 pounds of bone meal applied to the entire field in 1896 greatly reduced the need for phosphorus for some years. The published data are given in Table 49. (See Massachusetts Reports 1898 to 1907.) In the report for 1903, Professor Brooks makes the following comments concerning the cabbage crops: "Apatite and soft Florida phosphate are the least effective among the phos- phates employed. "South Carolina rock gives a surprisingly good return, being exceeded in yield of hard heads by only one plot, — the one receiving dissolved bone, — while in total yield it is materially exceeded by but few. "The phosphatic slag ranks among the best of the phosphates." Professor Brooks also makes the following general statements concerning these phosphate experiments: "In estimating the significance of the results upon this field, it is important to keep in mind the facts as regards the character of the soil. It is what would be called a strong and moderately heavy loam, and has great capacity to retain moisture. The relatively insoluble phosphates are known to give better results on soils of this character than on those which are lighter and drier." 282 SYSTEMS OF PERMANENT AGRICULTURE 8 ' M 1 \oo cooooooo oo 00 N N O • N t 3 « S§ M fo ' • N N 8P ^ P vo O ^^ O O^ O O O O ON §, ^ ^ Q 6 9 VO O HI O OO O ^O N Tt N to HH u o W O 't N ^O co ^t* ^* cooO ^O ^o H ' — N P <«3 HI N rf 8* ' °° Si vO O NQHiTfOOO OO u iffc H H 05 IH OO ^O to '^tO'tO'tO NO co to N \O co M OO ^tOO HI o £^ S | jj •" t/i &( (7 H HI CO ^t t. " >-. K M W 00 ggg vo^t OO'tONOO OO r}- OO M O *"*" '-I" ON N OO *O 't ^^ « "8 0 ^ o\ s l(gj Tf ^f IOOO HI O ^O N co co « S M i^ "*"* ^ ft 1 2; \O ^O HOHiOOO't OO TJ- co ^t O to^O O N 't O O tj ^] Q «J5 S ££ ti en 0 O H CO HI O ^ ^" HI CO O N HI CO ^ c ^ Q u CX rt 13 -^ V PH < ' W vOO ^OOtoOOOO O'* ^ a! iV ^3 ,^ > t^- IN 0 izill ^•00 N t^^^^-M M H VO S 8-2 O >0, A, N CO C> 1) U ? >-i . M g H o OO^M OO NONOOOO OO ^^ 1 ^ K ^ 10 tn <3 O ^" < Hi K g O*\ ON N HI N ^O ON O 't NO *O 1/1 4S bo Tj" 't HI to ONOO NO f^* HI i « • ^ N CO •2 — « 8 S ' W OO wQNOOOO O't O -5 to ^2 « siii O NO IOOO »OOO ^ 't ^ O O OOON HN OOOOOON NOlO lO't CN1NOCOHIOO l^M m ^ -o 2 ^ _^ s o c .2 ^£'g.s I . . > « >, « 41 — "T3 00 1> 1) O 4,- -C 60 C/3 ^ ^. in 3 CO 5n _> M uT -j J3 M • • _d • _o . n3 O OT "j en rt . pH .„ S £ C PnT3 13 -o O '"" •*-'' H D *^!X "f. O i^ O cvl *^ fa -4-* O ^^ QJ rt j U (A H «U < P ^^ *T3 5 5 *~^ ^ (j ^ cj H -4-» j? ^_» S S GO ; U « U §" W H WUUUWOwOc/iUUWWffiffi 0 -^ H rt 1 £ ^* t^OO ON O H M N N CO co Tf IONO NO ONONONONOOOOOOOOOOO CO OO 00 00 ON ON ON ON ON ON ON ON ON ON ON ~ •* w ^"3^1) *tt -^ O »j USE OF PHOSPHORUS IN DIFFERENT FORMS 283 "It appears reasonable to believe that on soils of the character of this field the farmer may safely depend for a considerable portion at least of the phos- phorus needed by his crops upon the cheaper natural phosphates, such as finely ground South Carolina rock and finely ground bone, while phosphatic slag also promises to give a most useful fertilizer upon soil of this character." This entire field, including the no-phosphate plot, has received 52 pounds of nitrogen and 126 pounds of potassium per acre per annum for the ten years. The phosphate plots have each received 42 pounds of phosphorus per acre each year; but no provision was made for maintaining organic matter in the soil. It should also be kept in mind that the raw phosphates that gave poor results may not have been ground to a sufficient degree of fineness. The Illinois Experiment Station is conducting much more ex- tensive experiments than any other state with the use of fine- ground natural rock phosphate, but these investigations were begun too recently to furnish information from which such final conclu- sions can be drawn as from the Ohio work, for example. In Table 50 are reported results obtained from the University of Illinois soil experiment field near Galesburg, Knox County, on the ordinary brown silt loam prairie soil of the Upper Illinoisan glaciation, which, in 1903, contained in 2 million pounds of the surface soil 5020 pounds of nitrogen, 1160 pounds of total phos- phorus, and 31,700 pounds of potassium. A six-year rotation is under way on this field, including corn for two years, oats the third year, and wheat the fourth, followed by two years of clover and timothy. (After the first six years the rotation will be corn, corn, oats, clover, wheat, clover.) There are three independent series of plots, so that every year corn is grow- ing on one series, oats or wheat on another series, and clover and timothy on the other. The land was timothy sod at the beginning, and Series 300 was not broken during the first two years, \ ton of phosphate per acre having been applied at the beginning as a top dressing, which, as was expected, produced practically no effect. A ton of phosphate per acre applied in the beginning to Series 200 produced no effect on the oats seeded on timothy sod in 1904, and but little effect on the wheat which followed in 1905. The regular plan is to apply i^ tons of raw rock phosphate per acre to the clover and timothy 284 SYSTEMS OF PERMANENT AGRICULTURE sod before plowing for corn, and this application will probably be repeated every six years until the total phosphorus content of the plowed soil is about doubled, after which the amounts applied for each rotation will be reduced to supply only about as much phos- phorus as is removed in the crops grown. The heavy applications of phosphorus that will thus be made during the first three or four rotations cost about $1.88 per acre per annum, which is less than is commonly expended for " com- plete " fertilizers in the older states, in a system that supplies less phosphorus than is removed in the crops grown and that thus leaves the land poorer year by year. (An application of 200 pounds of " 2-8-2 " fertilizer J would furnish less than 9 pounds of total phosphorus and at an average cost of at least $2.) Different systems of supplying organic matter are followed upon the different plots numbered in Table 50 (legume catch crops, crop residues, and farm manure), so that the same yields are not to be expected upon plots 2, 3, 4, and 5, for example; but these four plots differ from the next four only by the application of phos- phorus to plots 6, 7, 8, and 9. For the student of details, it may be said that, aside from the phosphorus applied, plot 5 is treated the same as plot 6, while plots 2, 3, and 4 are treated the same as plots 7, 8, and 9, respectively. Of course, the benefits of the crop rotation, including the use of different methods of supplying organic matter and nitrogen, cannot be determined before even the first rotation is completed; and the results thus far secured from the phosphorus applied are to be considered very preliminary. They show but little of what it is reasonable to expect from the system when fully under way after the benefit of one or two full rotations is felt. In the last column of Table 50 are given the values of the in- creases produced by the raw rock phosphate, including the yearly totals from the three crops; that is, from three acres. By keeping in mind that the annual cost of the phosphate for three acres is $5-63 (while the heavy applications are being made), the financial progress of the experiment during the first five years is seen at a glance. In round numbers, the increase paid 50 per cent interest 1 This means 2 per cent of ammonia (NHs), 8 per cent of available "phosphoric acid " (P2O6), and 2 per cent of potash (K2O). USE OF PHOSPHORUS IN DIFFERENT FORMS 285 o WJ W « a -c /ALUE OF AVERAGE asvaaoisi 00 ON M to O to HI 00 00 O •3" CM NO ON O 00 NO CO O to 00 CO l^-OO O to H 00 CM to t*~* NO CM Tj- to 4 1 4" 1 HI HI M M ' 1 2 1 fOco to to (M rj- M HI o to o o o HI ON - O H 00 •* 1000 CM O CM H H H to NO O to NO^H Tf 10 t^- to Tf M HI 00 t to f^ O CM HI (0 M CM IO CM to t-ON^5 00 Tj- 0\ ON NO t^ H 00 CNI HI NO HI H NO ^3" ON CO Hi NO to Tf HI HI oo •* ON CM CM ON CO O i~- to CNI ON CN1 f^* CN to CO CS ON CO H rj- M CO to O ££-" T}- CS M 0 • H 00 E 1 10 to HI NO IO NO O CNI t°.^ CO C» H 00 co CM NO HI 0 ; uOffi Corn, bushels Wheat, bushels Hay, tons en co en 3 3 G O c3 c3 UOffi 0 < 8 o o o to to tr- O O O 0s- O^ 0s 'O ^O VO 000 ON ON O 0 0 O ON ON O oo oo oo \o o o ON ON ON o U I 888 W CM C<- 888 W CNI C<- 8 8 8 CO HI CNI O O O o o o CO HI CN) 888 CM CO H 286 SYSTEMS OF PERMANENT AGRICULTURE on the investment in the first three years, and during the next two years it paid the annual cost and 40 per cent net profit on the same. The results of the Galesburg field are in harmony with those thus far secured from many other University of Illinois soil experiment fields in different parts of the state, and they are also in harmony with numerous practical tests by progressive Illinois farmers who make adequate provision for supplying the soil with decaying organic matter. Thus, as an average of four independent tests on each experiment field, Tennessee raw rock phosphate increased the yield of corn in 1908 by 12.1 bushels per acre on the Galesburg field, by 11.9 bushels on the Myrtle field for first-year corn and 9.3 bushels for second-year corn, by 16.0 bushels on the Rockford field for first-year corn and 7.6 bushels for second-year corn, by 3.5 bushels on the Antioch field, by 9.1 on the Auburn field, and by 8.4 bushels on the Urbana field. These experiment fields are in six different counties, and they have been in operation from four to six years. The average yield of corn in 1908 was 67.3 bushels where raw phosphate has been applied and 57.5 bushels without phosphorus. The phosphate applied thus far adds phosphorus to the soil at the rate of 60 pounds or more per year, while 16 pounds are required for a 68-bushel crop and about 2 pounds for the lo-bushel increase. Thus, the value of the increase ($3.43) will pay the cost of the phosphate (less than $2) and leave 50 per cent net profit, and with 70 per cent of the phosphorus left in the soil. The effect on wheat and clover is almost as marked as on corn. Of course, more clover means more nitrogen secured from the air, and it may also mean more manure to return to the soil. Mean- while, the untreated land grows poorer year by year. In Table 51 are given the results reported by the Illinois Experi- ment Station (Circular 97) from a series of pot cultures conducted for the purpose of comparing equal money values of raw rock phos- phate and steamed bone meal. In the Illinois field experiments, the standard annual application of phosphorus is 25 pounds per acre in 200 pounds of steamed bone meal and nearly equal money values of other forms of phosphorus. The 25 pounds is based upon the requirements of a zoo-bushel crop of corn, with i or 2 pounds for loss in drainage. In pot cultures very large crops are commonly USE OF PHOSPHORUS IN DIFFERENT FORMS 287 produced, and to meet the needs of such crops the applications of plant food are made three times as large as in field experiments. The soil used was from the gray silt loam prairie of the lower Illinoisan glaciation, and wheat was the crop grown in the pots. The phosphate used is known as the Tennessee blue rock. In cer- tain pots the phosphorus was turned under with a good growth of clover; in other pots with farm manure, and in others with both clover and manure. TABLE 51. COMPARATIVE EFFECTS OF STEAMED BONE MEAL AND RAW ROCK PHOSPHATE, IN CONNECTION WITH CLOVER AND MANURE PARTIAL TREATMENT APPLIED SERIES 100, WITHOUT POTASSIUM, WHEAT YIELDS SERIES 200, WITH POTASSIUM, WHEAT YIELDS Grams per Pot Increase in Grams Grams per Pot Increase in Grams None . IO.O 16.3 14.7 14.2 22.2 23-3 I6.5 22-7 19.4 19-5 24.1 23-3 6-3 4-7 4-2 12.2 J3-3 6-5 12.7 9-4 9-5 J3-1 J3-3 II.3 18.4 18.4 18.2 21.9 21.9 18.1 19.1 19-3 19.0 25-3 25-3 J-3 8.4 8.4 8.2 11.9 11.9 8.1 9.1 9-3 9.0 iS-3 15-3 Clover Bone meal Rock phosphate Clover, bone meal Clover, rock phosphate Manure Clover, manure Manure, bone meal Manure, rock phosphate .... Clover, manure, bone meal . . . Clover, manure, rock phosphate . . It will be seen that the untreated soil (pot 101) yielded 10 grams of wheat. Where clover was 'turned under (102), the yield was in- creased by 6.3 grams, and where bone meal was turned under with clover (105), the yield was 22.2 grams, the increase of 12.2 grams being nearly double that produced by clover without bone meal. Where raw rock phosphate was turned under with clover (106), the wheat yielded 23.3 grams, making a total increase of 13.3 grams over the yield of the untreated soil. Of this increase 6.3 grams should be credited to the clover and 7 grams to the rock phosphate, by one computation; or 4.2 to the phosphate and 9.1 to the clover, by the other route. Thus, rock phosphate used alone produced an 288 SYSTEMS OF PERMANENT AGRICULTURE increase of only 4.2 grams, which, added to the increase of 6.3 grams due to clover alone, makes only 10.5 grams. In other words, the sum of the gains which they make when used separately was 2.8 grams less than the increase produced when the rock phosphate and clover were turned under together. Somewhat similar results are produced with clover and bone meal when used separately and together; also with bone meal and potassium, and with rock phosphate and potassium. Such marked combined action does not appear, however, from other combinations, possibly because of other limiting factors. As a general average, the rock phosphate has made slightly better gains than the steamed bone meal. The pots used in these investigations are io| inches in diameter, consequently i gram per pot corresponds to i pound per square rod, or to 160 pounds per acre. The actual yields in grams per pot are given, but the results may also be computed to bushels per acre. It should be remembered that pot cultures constitute an intensive form of agriculture. They are carried on under almost complete control, except in very warm weather, when too much shade may be required to avoid too high temperature. The yields obtained are usually two or three times as much as can be expected in the field under ordinary weather conditions. They are not, however, larger than could be obtained in the field under perfect weather conditions. The largest yield reported in Table 51 is 25.3 grams per pot, or 67 bushels of wheat per acre. Pot culture yields have been produced corresponding to 142 bushels of wheat, and to 230 bushels of oats, per acre. Doctor Alfred M. Peter of the Kentucky Station has kindly furnished the author the following data secured by him with the cooperation of Mr. S. C. Jones of the Kentucky Geological Survey: KENTUCKY EXPERIMENTS: POT CULTURES TREATMENT APPLIED CROP YIELDS: GRAMS PER POT Kind Amount Tobacco (Leaves and Stalks) Wheat (Grain Only) Alfalfa Hay None .... none 4.0 grams 10.5 grams 4.8 10.0 12-9 6.8 9-5 I2.O 6-7 n-5 IO.2 Acid phosphate .... Raw phosphate .... USE OF PHOSPHORUS IN DIFFERENT FORMS 289 The soil used in these pot cultures (which were 2-gallon jars) was a residual limestone soil from Christian County, Kentucky, and contained 870 pounds of total phosphorus and 32,120 pounds of potassium in 2 million pounds of surface soil. The results are of interest; but, as Doctor Peter writes, " it must be understood that they are only single experiments carried out one season, and must be valued accordingly." The Wisconsin Agricultural Experiment Station (Bulletin 174) also reports a single year's experiment with field cultures showing that manure and raw phosphate increased the yield of rutabagas by 27 per cent and the yield of potatoes by 47 per cent above the yield from manure alone, and the opinion is expressed that " these results leave no doubt that the use of phosphate supplementing manure is beneficial." In describing Mr. J. F. Jack's Virginia farm, Joseph E. Wing, the well-known agricultural writer, makes the following state- ments (Breeders' Gazette, June 2, 1909) : "Proud as we are of Woodland Farm, I find acre after acre of alfalfa on Mr. Jack's farm as good as our best. I find it as good as the best that I ever saw in California. Is it all good? No. There are acres that are thin, stunted. What cause ? He is seeking that now. Doubtless there are areas that are too poorly drained, there are places yet sour, and some land needs more feeding. No doubt at all of that. He has not limed liberally at all times. Last year, for example, some men told him that they had a prepared lime that was two times as effective as ordinary lime. He had been using a ton to the acre ; he bought this lime, at a higher price, and used but 1000 pounds. Then he learned to his sorrow that the lime was simply slacked at the kilns, was so-called 'agricultural lime' and had only about half the strength of fresh burned lime. So it seems sure that much of his land has had too little lime. He finds that lime carbonate, that is, simply ground limestone, gives him as good results as anything, and that fortunately is cheap." "What an interesting thing it is to find this old Eastern land being newly dis- covered. . . . But here the soil must be fed, do not forget that ! The natives forgot it, hence their sorrow now." "Business methods apply to farming as well as to anything else. Farming is a business, and, with present prices for things, a paying business. It pays to buy lime (Mr. Jack is getting his ground limestone delivered to him for $2.90 this year) to make land sweet, to buy phosphorus, to sow legumes, to build soil. Alfalfa is as easily set in Virginia as in any other state, and it grows splendidly when the land is made sweet with lime, filled with decaying vegetable matter or humus, given inoculation and phosphorus." 290 SYSTEMS OF PERMANENT AGRICULTURE " Alfalfa will make land in Virginia yield good returns on a valuation of $200 per acre, or more, and land that was worth $30 per acre can be set in alfalfa at a cost of about $15 per acre, including lime, fertilizer, seed, and the growth of crimson clove'r. Enthusiasm and faith, with carbonate of lime, phosphorus, and clovers, can make a land beautiful to the eye, inspiring to the soul, and filling to the purse." " One most valuable result seen here is apparently that untreated Tennessee phosphate rock is giving as good results as phosphorus in any other form, using not the same amounts, but the same cost equivalents. In fact, there seems de- cided gain from the use of the raw phosphate." In a personal communication to the author, Mr. Wing adds the following : "This is not guesswork, since checks were left with no fertilizer, and the line is most marked. Apparently on that soil an application of 900 pounds of raw rock gave much better alfalfa than did 400 pounds of raw bone meal or a mixture of 200 pounds of bone meal and 300 pounds of acid phosphate." NOTE. On the author's Poorland Farm, including about 320 acres in the prairie section of southern Illinois (gray silt loam on tight clay), limestone is applied at the rate of 2 to 3 tons per acre, and raw rock phosphate at the rate of i ton per acre, every six years. After the phosphorus content of the plowed soil has been increased from 800 pounds per acre to about 2000 per acre, the application of that element will probably be reduced to an amount which will simply maintain the supply. Thus far the use of 1 1 car loads of raw phosphate and 17 car loads of ground limestone has given as satisfactory results as one could reasonably expect during the first crop rotation. For this special soil a six-year rotation is planned, as follows: First year Corn (and legume catch crop). Second year Part oats or barley, part cowpeas or soy beans. Third year Wheat. Fourth year Clover, or clover and timothy. Fifth year Wheat, or clover and timothy. Sixth year Clover, or clover and timothy. The plan may be a grain system where wheat is grown the fifth year, only clover seed being harvested the fourth and sixth years, or it may be changed to a live stock system by having clover for pasture and meadow the last three years, all manure produced being applied to the pasture land to be plowed under for corn. While the untreated land has produced about one third of a ton per acre of poor hay (part timothy and redtop, part foul grass, sorrel, and other weeds), the treated land has produced more than i£ tons per acre of clean clover and timothy hay. A car load of limestone or phosphate is purchased with less hesitation than a cow or a horse, and at about the same cost. A careful study of the literature of agricultural science from European countries reveals no investigations with the use of raw rock phosphate that compare in value with those conducted by USE OF PHOSPHORUS IN DIFFERENT FORMS 291 any one of the seven states, Ohio, Maryland, Pennsylvania, Rhode Island, Maine, Massachusetts, or Illinois. There are three essential points to be kept in mind concerning the use of raw rock phosphate: First, all rock is not phosphate rock, and the farmer should purchase only guaranteed material, and he should know how much phosphorus is contained in the ground rock he applies to the land, if necessary by taking 100 teaspoonfuls from 100 different parts of the car load (including different depths), thoroughly mixing, and sending half a pound of this to a reliable commercial chemist for analysis. Second, the rock should be very finely ground, and it should be purchased upon a guarantee that at least 90 per cent of it will pass through a sieve with 100 meshes to the linear inch (10,000 meshes to the square inch) , which is no finer than is required for slag phos- phate. Third, raw phosphate should not be expected to give marked benefits except when used in connection with adequate supplies of decaying organic matter. It has practically no value as a top dressing, but must' be plowed under and thoroughly incorporated with the soil where the roots feed. Of course, it will supply only the element phosphorus, and will not take the place of any other deficient element, nor act as a soil stimulant to liberate other plant food from the soil, although it sometimes contains small amounts of carbonate, and then has some tendency to correct soil acidity, but in this it is insignificant compared to the effect of ground limestone. The following interesting discussion concerning the use of raw rock phosphate, from the viewpoint of the fertilizer manufacturer, was published in pamphlet form and widely disseminated in 1908, by the National Fertilizer Association. It was also published in full in the American Fertilizer, August, 1908, and in part in Ar- mour's Farmer's Almanac for 1909. It is reproduced in complete form in the following pages, because it deserves to be read by every careful student of soil fertility. Its cautions against the use of raw phosphate as a source of immediately available plant food are commended. It also serves to emphasize the fact stated in the introduction, that, " if the independent farmer is to adopt and 292 SYSTEMS OF PERMANENT AGRICULTURE maintain permanent systems of profitable agriculture, he cannot accept ' parrot ' instruction," not even when offered by the fer- tilizer agent. The fact that acid phosphate and " 2-8-2 " fertilizers are still used in the Eastern and Southern states, largely as soil stimulants and for a single crop, or for one year's effect only, fairly raises the question whether agricultural practice in those states is not influ- enced more by the " arguments" of the fertilizer agent than by the established facts and principles from the experiment stations. RAW ROCK PHOSPHATE, "FLOATS" PUBLISHED BY THE NATIONAL FERTILIZER -ASSOCIATION For years the raw rock question has cropped out spasmodically, in different parts of the world, like the measles or some other affliction. Sometimes it was the result of the recommendation of some impractical theorist who occupied a position that brought him before the farmer — oftener it was foisted on an unsuspecting farming community by some one who was either directly or indirectly interested in an offgrade phosphate mine, and who used his official position to further the interests of the rock mine at the expense of the farmer. But no matter what started its use, the result has always been the same — no benefit derived from its use — a distrust of legitimate fertilizers, and a distinct set-back to agricultural interests which has taken several years to overcome. In the following pages we give you the opinions of foreign experiment station men as well as those of our own country. Both statistics and your good common sense tell you that the older the state or country, the more fertilizers are used and the greater the knowledge they have of their use. The mere fact that in these older communities, both abroad and in this country, the use of legitimate fertilizers has increased rapidly from year to year for a hundred years conclusively shows their value. The fact that wherever raw rock has been used, its use has been abandoned, shows its worthlessness. Read what authorities who know have to say on this subject. RESOLUTION passed by the Association of German Agricultural Experiment Stations, in congress assembled, September 14, 1907, at Dresden, Germany: "As a result of the extensive advertising which is done by certain parties ad- vocating the use of RAW ROCK PHOSPHATE, the association passed the follow- ing resolution: "THE ASSOCIATION HAS CONCLUDED, FROM FERTILIZER EXPERIMENTS AT HAND WITH RAW PHOSPHATE FERTILIZER, THAT THERE IS SHOWN NO PROFIT- USE OF PHOSPHORUS IN DIFFERENT FORMS 293 ABLE FERTILIZER EFFECT, APART FROM THOSE OF ACID SOIL. IN CONSEQUENCE THEREOF THE ASSOCIATION FEELS IT SHOULD' DISCOURAGE THE USE OF RAW PHOSPHATE ON OTHER SOILS." See 67th Volume (5-6), page 329, " LANDWIRTSCHAFTLICHER VERSUCHS STATION." The Association of German Agricultural Experiment Stations represents the highest authority on agricultural matters in Germany, and undoubtedly the best in the world. German investigators, particularly Dr. Von Liebig, were the authors of most of the fundamental principles underlying fertilization and agriculture, — and it is to them that we largely owe the progress made in this direction. In view of the well-known thoroughness of German agricultural investigators, and the fact that the Association of German Agricultural Experiment Stations is universally regarded as the world's highest authority on such matters, their opinion on the use of RAW PHOSPHATE as a fertilizer is of great importance to the American farmer. On account of the high price of land in Germany intensive farming is every- where practiced. The farmers must, of necessity, use fertilizer containing plant food in available condition. Their selection of fertilizers is based on in- numerable experiments covering over a hundred years. The difference in crop yields per acre in Germany as compared with the United States is conclusive evidence of the soundness of their methods of fertilization. The average wheat yield per acre in Germany for the ten years 1895 to 1904, inclusive, was 27.2 bushels, as compared with 13.4 bushels in the United States for the same period. On oats the yield per acre in Germany was 46.0 bushels, as compared with 29.2 bushels in the United States for the same period. (See pages 671 and 678, "Statistical Matter," reprint from Year Book of Department of Agriculture for 1905.) l The soils of Germany have been cropped for hundreds of years, while a large portion of those in this country are virgin or comparatively fresh. Proper fertilization is the secret of the higher yield per acre in Germany. If the United States is to maintain its supremacy in agriculture, farmers in this country will have to properly fertilize their crops, — and they can well take heed to the experience of their German brothers in this respect. Before using raw rock, therefore, you would do well to ascertain its true fertilizing value — the availability of the plant food it is supposed to contain — and especially to consider the decision of the German experimenters after years of careful testing. From the standpoint of furnishing available plant food, RAW ROCK PHOS- PHATE is not a fertilizer. The report of the twenty-fourth annual meeting of the Association of German Agricultural Experiment Stations, at which the 1 The ten-year average yield of corn in the great state of Georgia, where more manufactured acidulated commercial fertilizers are used than in any other state, is ii bushels per acre. — C. G. H. 294 SYSTEMS OF PERMANENT AGRICULTURE resolution quoted was passed, states that from "real exact experiments," con- ducted by such authorities as P. Wagner, Tacke, Bottcher, Lemmerman and others, "but little fertilizing effect was shown." Further experiments made by Czerhati, L. Key, Clausen, and others, led to similar results just stated. The same report states that, "From the present experiments it can be con- cluded with certainty that the general use of earthly phosphates (RAW ROCK PHOSPHATE) cannot be considered as phosphoric acid fertilization." Phos- phoric acid is the only element this material contains, and if IT is NOT available it is useless for fertilizing purposes. The experiment station officials of Germany have gone on record against the use of RAW ROCK PHOSPHATE in no uncertain tone. Their opinion is shared, with but one or two exceptions, by all the experiment stations in this country. If THIS material cannot be recommended for German soils, where proper fertilization has been studied for so many years, is it not folly to attempt its use on the comparatively fresh soils of this country? This report also refers to some recent experiments conducted by parties en- deavoring to promote the sale of raw rock phosphate in Europe. In comment- ing on the so-called tests or experiments, the German report states — that they "were carried out with but very little exactness." They further class these experiments as "entirely unfounded, have been rejected by scientific agricul- turists, and especially by Wagner, Tacke, and Bottcher, in a manner not to be misunderstood." The said representations of these promoters are classed as "A very serious deception," and misleading to the farmers. The efforts to promote the sale of RAW ROCK PHOSPHATE in this country — in the light of world-wide failure to show any appreciable fertilizing effect — can only be classed, in the language of the German experimenters, as "a very serious deception," and misleading to the farmers. Not alone in Germany have experiments with RAW ROCK PHOSPHATE proven very unsatisfactory. Professor F. H. Storer, in Volume I of his book "Agri- culture," in speaking of the value of raw phosphate USED IN CONNECTION WITH MANURE, as compared with superphosphate, says: "This question would seem to have been answered long ago, in so far as good land is concerned, by the common English practice of using superphosphates." Again, later, in comparing the effects of the same materials for fertilizing purposes in European countries, he says: "For Europe at least, i.e., for fertile districts, the question has been decided fully long ago and most emphatically in favor of superphosphate. It has been decided by the long-continued experi- ments of a multitude of farmers, and their conclusion has been plainly expressed by the ever increasing demand for superphosphate." 1 1 Director Hall of Rothamsted, in his "Fertilizers and Manures" (1909), page 118, says: "The mineral phosphates have been but little employed directly as manures, though there is plenty of evidence that when they are really finely ground, they are effective enough on soils retaining plenty of water." — C. G. H. USE OF PHOSPHORUS IN DIFFERENT FORMS 295 Coming down to our country, we find that experiments with RAW ROCK PHOSPHATE — with scarcely any exception — have proven unsatisfactory. Experiments conducted by the Maine Experiment Station, covering several years on various crops, designed to show the relative availability of phosphoric acid as supplied in acid phosphate, floats (raw rock phosphate), and redonda phosphate, were summarized as follows : "In every case the acid rock (Acid Phosphate) gave the best returns. The gain was especially marked with the family Gramineae, three members of which (barley, corn, and oats) yielded nearly double the amount produced by the Floats or Redonda. The effect upon sunflowers and buckwheat was equally marked. If we compare the amount of dry matter produced by the acid rock with that produced by the Floats for all crops grown, we find the balance in favor of the acid rock to be FIFTY-TWO PER CENT. In other words, the effect of the available phosphoric acid as compared with the insoluble phosphate was to increase the product MORE THAN ONE HALF." l The Georgia Experiment Station, commenting in Bulletin No. 2, concern- ing field experiments with phosphates and kainit applied to cotton, states: "Of phosphates, Acid Phosphate appears to lead, slag conies next, and the FLOATS ARE LAST." 2 A later Georgia Bulletin (No. 31), in reviewing a comparison of superphos- phate with Tennessee soft phosphate, states : "Superphosphate in a complete fertilizer was compared with one, one and a half and two times the same amount of Tennessee soft phosphate. The latter (Tennessee soft phosphate) was ap- plied in each case at a loss." 2 In the Annual Massachusetts Experiment Station Report for 1902, concern- ing an experiment with various kinds of phosphates which were applied in equal amounts of phosphoric acid, there is the following regarding raw rock phosphate : "Tennessee phosphate and Florida soft phosphate gave results very much inferior to all the others." This was an experiment on onions. In the Massachusetts Annual Report for the following year (1903), concern- ing the same experiment continued on cabbages, the previous year's results are confirmed: "That Tennessee phosphate and Florida soft phosphate proved very much inferior to all others."3 1 This quotation is taken from page 72 of the 1898 Report of the Maine Experi- ment Station; while on page 57 of the 1900 Report occur the following statements: " For the first year the largest increase of crop was produced by soluble phosphoric acid. For the second and third years, without further addition of fertilizers, better results were obtained from the plots where stable manure and insoluble phosphates were used. " — C. G. H. 2 These are single-year tests. The following quotation might also be made from page 161 of Georgia Bulletin No. 25 : "Florida Soft Phosphate appears to be equally as valuable as Acid Phosphate, the difference, if any, being rather in its favor. " — C. G. H. 3 On the other hand, the South Carolina raw rock phosphate produced a much larger yield than acid phosphate, especially of marketable cabbage, as shown in Table 49. — C.G. H. 296 SYSTEMS OF PERMANENT AGRICULTURE In Scott County, Indiana, an experiment to determine the relative value of raw rock phosphate and acid phosphate was started in 1904 and continued for four years. Equal values of rock phosphate and acid phosphate were ap- plied in ONE application the first year — corn and wheat alternating. The actual amount of plant food applied was 286 pounds of total phosphoric acid in the rock phosphate and 100 pounds phosphoric acid in the acid phosphate. There were three plots in the experiment — one fertilized with rock phosphate, one with acid phosphate, one unfertilized. Notwithstanding the fact that the first year's corn crop was a total failure on all plots, the results on wheat showed a gain of fourteen bushels per acre with acid phosphate as against only nine bushels for the rock phosphate over the unfertilized plot. The profit per acre in four years from rock phosphate was $11.55; ' the profit in four years from acid phosphate was $13.50. In Marion County, Indiana, another experiment for the same purpose was started, and crops harvested for two years.2 Only one application of fertilizer was made, the entire amount being applied the first season. As in Scott County, equal values of rock phosphate and acid phosphate were applied. The results speak for themselves, and they are given in the table below as taken from Circular No. 10 of the Indiana Experiment Station. The yields are given in bushels per acre: CORN WHEAT Amount per acre 1904 1905 Unfertilized 20 3 Rock phosphate, 1000 Ib 20 6 Acid phosphate, 715 Ib ._ 27 16 The value of the increase per acre, figuring corn at 35 cents and wheat at 80 cents per bushel, on the plot fertilized with acid phosphate, was $12.85 Deducting cost of acid phosphate 5.00 Net return on the increase $7.85 Value of the increase with rock phosphate 2.40 Deducting cost of rock phosphate 5.00 Or a net loss of $2-6o per acre On the total yields the results were as follows: Unfertilized . ^.40 per acre Raw rock phosphate 6.80 per acre 8 Acid phosphate I7.25 per acre : This is a very fair profit considering that about two thirds of the raw rock phosphate will remain in the soil after the acid phosphate is completely exhausted, .t should also be noted that the cost of the acid phosphate was figured at $14 per ton, and the cost of the raw phosphate at $10 per ton. — C. G H 3 Italics mine. — C. G. H. 8 Cost of rock phosphate and acid phosphate deducted. USE OF PHOSPHORUS IN DIFFERENT FORMS 297 These figures show that the rock phosphate was applied at a dead loss of $2.60 per acre — the unfertilized yield value being $2.60 per acre more than the rock phosphate. The yield with acid phosphate was $7.85 more than the UNFERTILIZED, and $10.45 Per acre more than the RAW ROCK PHOSPHATE. These results are from experiments primarily intended to show the value of raw rock as a fertilizer. They are self-explanatory, and show conclusively the FOLLY OF CONSIDERING THIS MATERIAL AS A FERTILIZER. Further, these results Were obtained from 100 pounds of phosphoric acid in acid phosphate as compared with 286 pounds of raw rock phosphate. The practical farmer, interested in the proper use of commercial fertilizers, can easily figure that where acid phosphate gave such remarkable returns on experiments covering a series of years, it will pay him a handsome profit to invest judiciously in fertilizers every year giving such good returns. Tennessee has some of the largest phosphate deposits in the world. In this state where the value of phosphate is so well understood, Professor C. A. Mooers, chemist and agronomist of the Agricultural Experiment Station at Knoxville, in a recent letter to THE AMERICAN FERTILIZER has the following to say with regard to the use of this material in its crude state : "A bill was introduced in the Legislature, just adjourned, to allow the sale of ground rock phosphate as a fertilizer. In presenting this matter to the Agricultural Committee, the Commissioner of Agriculture and myself took the position that it would not be desirable to tag this material, as that would, to a certain extent, stamp it with the State's approval. Our position is better under- stood when it is considered that a very large part of the fertilizers used in this State are for wheat, and, as is well known, RAW PHOSPHATE ROCK, as ordinarily used, GIVES NO RETURNS ON THIS CROP. Other large amounts are used, es- pecially in West Tennessee, by the truckers, and for garden crops also RAW PHOSPHATE WOULD BE INADVISABLE. Fertilizers have been used in this State for many years, but our farmers have not studied the matter to any great ex- tent, so that many of them would buy a fertilizer just BECAUSE IT WAS CHEAP, especially if it had the State's tag on it. "Our results on leguminous crops, which are supposed to be better able to make use of the so-called insoluble forms of phosphoric acid than others, do not warrant the general use of RAW PHOSPHATE. I have recently corresponded with a number of station men who are interested in the use of fertilizers, and I find that the general opinion is AGAINST THE USE OF THIS MATERIAL, although under special conditions, such as are found on a decidedly acid soil, its use may be advisable." The state of Alabama is one of the oldest of the states using commercial fertilizers. Bulletin No. 24, issued May 15, 1908, contains an article on "Raw Phosphate Rock as a Fertilizer." Following are extracts from this article: "Many parties have written to this office for information as to the relative fertilizing value of the raw phosphate, as compared with the acidulated phos- phates, and the writer has invariably advised caution in the employment of this particular kind of phosphate. 298 SYSTEMS OF PERMANENT AGRICULTURE "The samples of this material which have reached this laboratory have al- most invariably exhibited a poor mechanical condition, the particles being coarse and irregular in size. As the fineness of division of this phosphate has a most important influence upon its availability to the plant, purchasers of this material have been advised to only use the rock which has been pulverized to a state of practical impalpability, the material in this condition being commonly designated by the name of 'floats.' "A typical analysis of the raw phosphate rock sent to this laboratory this season is given herewith : Citrate-soluble phosphoric acid 0.68 per cent Acid-soluble phosphoric acid 23-S5 Per cent Total phosphoric acid 24.23 percent "It will be noted that nearly all of the phosphoric acid in this phosphate is in an insoluble or acid-soluble condition, and there is SCARCELY A TRACE or WATER SOLUBLE PHOSPHORIC ACID TO BE FOUND IN THIS RAW PHOSPHATE. "With regard to the comparative availability of raw phosphate rock, it might be stated that the Experiment Station at Auburn has, during the past few years, carried out under its supervision more than one hundred cooperative soil and crop tests in a great many different localities in the State with a view to deter- mining the comparative efficiency of raw phosphate and acid phosphate for fer- tilizing purposes. These tests have been carried out upon quite a variety of soils, and upon most soils the RAW PHOSPHATE HAS FAILED TO GIVE ANYTHING LIKE AS GOOD RESULTS AS THE ACID PHOSPHATE. "In the case of acid phosphate, the ready solubility of most of the phos- phoric acid contained therein promotes its rapid and thorough distribution through the top layer of the soil, and hence the plant food is so well dissemi- nated that it is brought within easy reach of the root system of the plant, whereas in the case of the crude insoluble phosphate the diffusion and dis- tribution of the phosphoric acid is necessarily slow, and much of the phos- phate is left unutilized at the end of the season in which it is applied. "For the above reasons IT is DEEMED INADVISABLE TO EMPLOY THE CRUDE PHOSPHATE to any great extent upon any given soil until comparative tests of the crude rock and acid phosphate have been made upon that soil, and, even under these conditions, it will probably be found necessary to use much larger amounts of phosphate rock .than are ordinarily employed to secure a satis- factory return from its application." While the experience of the German Experiment Stations, combined with a majority in this country, show emphatically that raw rock phosphate has little or no fertilizing value, in addition the method of applying followed by users of this material in this country is MOST EXTRAVAGANT AND WASTEFUL. The method followed would soon exhaust the known or visible supply of phosphate rock. Further, the enormous quantities necessary to apply per acre, instead of being scattered over and benefiting millions of acres, would be wasted on comparatively few. USE OF PHOSPHORUS IN DIFFERENT FORMS 299 On the other hand, this crude material, when properly treated with sulfuric acid and converted into acid phosphate, to be used either as straight acid phosphate or in mixed fertilizers, becomes a source of available plant food of greatest value. Raw rock phosphate, as mined and sold by certain operators, does not contain plant food immediately available to growing crops. It is only by proper handling and treatment with sulfuric acid that this material is con- verted into fertilizer furnishing plant food available to various crops and soils. Reputable fertilizer manufacturers decry the use of raw rock phosphate as a fertilizer, knowing that it will NOT prove satisfactory, as borne out by exten- sive experiments of the world's best agriculturists. They have gone on record against its use, and any lack of results on the part of those using this material should not vitiate against the use of commercial fertilizers rightly prepared, furnishing available, nourishing plant food for all crops. (Signed) THE NATIONAL FERTILIZER ASSOCIATION. The author feels that no further comment is necessary regarding this statement from the National Fertilizer Association. The facts are presented in very complete form in the preceding pages, and the reader must draw his own conclusions. For other illustra- tions of the possibility of erroneous conclusions being drawn by such German investigators as advance theories or reach conclu- sions without sufficient facts, reference may be made to Chapter 31, and also to Sir Henry Gilbert's very interesting and complete discussion 1 of the sources of fat in the animal body, based upon Rothamsted investigations in which 327 different animals were dissected, 10 different selected carcasses having been subjected to chemical analysis, following the analysis of the food stuffs provided during long feeding periods. 1 Pages 231 to 282, Bulletin 22, Office of Experiment Stations, United States Department of Agriculture, "Agricultural Investigations at Rothamsted dur- ing a period of Fifty Years," containing a series of six lectures delivered in America in 1893 by Sir Joseph Henry Gilbert, under the provisions of the Lawes Agricultural Trust. CHAPTER XVIII THEORIES CONCERNING SOIL FERTILITY ABOUT three hundred years ago Van Helmont, a Flemish alche- mist, planted a five-pound willow tree in 200 pounds of dry soil. He watered it with rain water for five years, and then found that the tree had gained 164 pounds and that the soil had lost only 2 ounces, in weight. Therefore, he concluded, water is the source of plant food. While it seemed to him that his evidence was strong and positive, all know now that his conclusion was wrong, and that the air, the water, and the soil are all essential sources of plant food. It will be noted that 2 ounces removed from the 200 pounds of soil correspond to 1250 pounds from 2 million pounds of soil. Later, Bradley, in his "General Treatise of Husbandry and Gardening," argued that water could be distilled or evaporated, which was not the case with willow trees; and, hence, that water is not the food of plants. He held that air must be the food of plants. About two hundred years ago, Jethro Tull, the inventor of the first seed drill, taught that neither water nor air could be the food of plants because they were furnished alike to all plants; whereas, two adjoining fields produced very different yields because one was impoverished soil while the other had been enriched. Tull wrote as follows: "It is agreed that all the following materials contribute in some manner to the increase of plants, but it is disputed which of them is that very increase of food: (i) Niter, (2) Water, (3) Air, (4) Fire, (5) Earth. . . . " Niter is useful to divide and prepare the food, and may be said to nourish vegetables in much the same manner as my knife nourishes^me, by cutting and dividing my meat; but when niter is applied to the root of a plant, it will kill it as certainly as a knife misapplied will kill a man ; which proves that niter is, in respect of nourishment, just as much the food of plants, as white arsenic is the food of rats, and the same may be said of salts. 300 THEORIES CONCERNING SOIL FERTILITY 301 "Water, from Van Helmont's experiment, was by some great philosophers thought to be it. But these were deceived, in not observing that water has always in its intervals a charge of earth, from which no art can free it. "Air, because its spring, etc., is as necessary to the life of vegetables as the vehicle of water is, some modern virtuosi have affirmed, from the same and worse arguments than those of the water philosophers, that air is the food of plants. . . . "Fire. No plant can live without heat, though different degrees of it be necessary to different sorts of plants. Some are almost capable of keeping company with the salamander, and do live in the hottest exposures of the hot countries. Others have their abode with fishes under water, in cold climates; for the sun has his influence, though weaker, upon the earth covered with water, at a considerable depth, which appears by the effect the vicissitudes of winter and summer have upon the subterraqueous vegetables. " But that fire is the food of plants, I do not know any author has affirmed, except Mr. Lawrence, who says: ' They are true fire-eaters '; and even he does not seem to intend that this expression of his should be taken literally." "Earth. That which nourishes and augments a plant, is the true food of it. "Every plant is earth, and the growth and true increase of a plant is the addition of more earth." "Too much earth, or too fine,. can never possibly be given to roots . . . and earth is so surely the food of all plants, that with the proper share of the other elements, which each species of plants requires, I do not find but that any common earth will nourish any plant." "The mouths, or lacteals, being situate, and opening in the convex super- ficies of roots, they take their pabulum, being fine particles of earth, from the superficies of the pores, or cavities, wherein the roots are included. . . . These particles, which are the pabulum of plants, are so very minute and light, as not to be singly attracted to the earth, if separated from those parts to which they adhere, or with which they are in contact (like dust to a looking glass, turn it upwards, or downwards, it will remain affixed to it), as these particles do to those parts, until from thence removed by some agent. "A plant cannot separate these particles from the parts to which they adhere, without the assistance of water, which helps to loosen them. "As to the fineness of the pabulum of plants, it is not unlikely that roots may insume no grosser particles than those on which the colors of bodies depend; but to discover the greatness of those corpuscles, Sir Isaac Newton thinks, will require a microscope that with sufficient distinctness can represent objects five or six hundred times bigger than at a foot distance they appear to the naked eye." In general, Jet.hro Tull taught that the soil particles are the food (pabulum) of plants, and that, if the soil were made sufficiently fine by cultivation, the plants could then absorb these fine particles of earth and produce large crops continuously. In answer to the 302 SYSTEMS OF PERMANENT AGRICULTURE arguments of his critics that the agricultural practice of his time was the result of long experience and consequently must be cor- rect, he expressed a fundamental truth in the following words: "The experience of 1700 years no more proves this practice to be right, than the long experience of cattle drawing by their tails proved that practice right, before drawing by traces was by experiment proved to be better: for nothing can be depended on as experience, which has not been tried by experi- ment." He also classes himself with those who " cannot believe that a man will become bald by being shaved at the wrong time of the moon, without more experience than has been made for it these 1700 years past." Another century passed, during which the humus theory ad- vanced by Thaer and others gained some recognition. The humus of the soil was held to be the source of carbon and carbonaceous matter for the plant. Humus and water were considered the only sources of plant food, and the productive power of the soil was believed to depend solely upon its humus content. In the "Georgical1 Essays" (Edition of 1777) by Doctor A. Hunter, which also includes many essays or reports by other " philosophical farmers," we find the following interesting state- ments : "The ancient writers gave us excellent comments upon the husbandry of their times. Hesiod wrote very early upon Agriculture. Mago, the Carthagin- ian general, composed twenty-eight books upon the same subject, which were translated by order of the Roman Senate. Upon these models Virgil formed his elegant precepts of husbandry. "Cato, the Censor, wrote a volume upon Agriculture. Columella has left us twelve books upon rural matters. Varro's treatise will ever be esteemed. . . . "The celebrated Sully calls Agriculture one of the breasts from which the State must draw its nourishment. That great man could not possibly have given us a more happy simile. ..." "Colbert entertained a different notion of policy. Esteeming manufacturers and commerce as the sinews of the State, he gave all possible encouragement to the Artisan and the Merchant, but forgot that the manufacturer must eat his bread at a moderate price. The farmer being discouraged, the necessaries of 'Georgical, like the proper name George (Latin Georgius), meaning husband- man or farmer, is derived from the Greek -yi) (ge-, as in geology), the earth, and tpyttv (ergein, as in energy), to work. The Georgics of Virgil are poems on agri- cultural affairs. THEORIES CONCERNING SOIL FERTILITY 303 life became dear; the public granaries became ill stored; manufactures lan- guished; commerce drooped; a numerous army soon consumed the scanty harvest; and, in a short time, Industry fell a sacrifice to the ill-judged policy of the minister. "From that period to the present, the French nation have constantly been availing themselves of their mistake. Under the genial influence of the King, Societies are erected in every province. Men of the first distinction do not dis- dain the cultivation of their own lands. M. de Chateauvieux and Duhamel are the greatest ornaments of their country. — Let us imitate the virtues of that fash- ionable nation. . . ." "The art of husbandry boasts an origin coeval with the human race. Its age, however, seems to have contributed but little towards its advancement, being at present extended but a few degrees beyond its primitive institution. Until the philosopher condescends to direct the plow, Husbandry must remain in a torpid state. . . . " I take it upon me to say, that, to be a good husbandman, it is necessary to be a good chymist. Chymistry will teach him the best way to prepare nourish- ment for his respective crops, and, in the most wonderful manner will expose the hidden things of nature to his view. The principles of Agriculture depend greatly upon chymistry; and without principles, what is art, and what is science ? " Directed by instinct, the animal seeks its own proper food ; but the vegetable, not being possessed of the power of motion, must be satisfied with the nourish- ment we give it. To direct this upon rational principles, is the business of the ' philosopher. The practical farmer will suffer himself to be instructed as soon as he perceives the practice correspond with the theory laid down to him. Let us expect no more of him. Men of limited education commit great errors when they attempt to reason upon science. In husbandry, effects are constantly applied to improper causes. Hence proceed the errors of our common farmers. To overcome these is the peculiar province of the philosopher; who, in his turn, must support his reasoning by facts and experiments. "I lay it down as a fundamental maxim, that all plants receive their principal nourishment from oily particles incorporated with water, by means of an alka- line salt or absorbent earth. ... It may be asked, whence do natural soils receive their oily particles? I answer, the air supplies them. During the summer months, the atmosphere is full of putrid exhalations arising from the steam of dunghills, the perspiration of animals and smoke. Every shower brings down these oleaginous particles for the nourishment of plants." " The ingenious Mr. Tull, and others, have contended for earth's being the food of plants. If so, all soils equally tilled would prove equally prolific. Water is thought, by some, to be the food of vegetables, when in reality it is only the vehicle of nourishment." After pointing out the great value of oil meals, rape cake, etc. (and later of fish scrap), for soil improvement, and after noting 304 SYSTEMS OF PERMANENT AGRICULTURE that all seeds contain oil, and that hemp, rape, and flax (rich in oil) are very exhaustive crops, Doctor Hunter adds, much to his credit as a scientist: "As I have not the vanity to think my experiments sufficiently conclusive, I embrace this opportunity to request assistance of the practical farmer, in order that the merits of the invention may be fully determined. Should my theory concerning the food of plants be thought erroneous, the compost (made in part of crude whale oil, 'train oil') will of course be disregarded. But, on the con- trary, should it be agreed to that oil, made miscible with water, constitutes the chief nourishment of vegetables, then the invention will probably become the subject of future experiment. "Though theory may direct our inquiries, yet experience must at last deter- mine our opinions, for which reason I propose to enlarge my experiments; and as I have no other view but the investigation of the truth, I shall lay them faith- fully before the public, whether they prove successful or not." Among the " Georgical Essays," the two reports which follow are of special interest. The first bears upon the oil theory, and both show evidence of the search for truth, and indicate the approach- ing dawn of chemical science. The editor says that the 1777 edi- tion is a reprint, and that " this volume contains several additional papers"; so it is not clear that Doctor Hunter knew of these experiments. "A COMPARATIVE VIEW OF MANURES "By A. YOUNG, ESQ. "In the year 1771, 1 marked out a rood of land into divisions, and sowed them with oats. The variety of manures made use of in this experiment are marked as follows: PRODUCE PER ACRE Na B. P. 1. 40 cubical yards of farmyard compost, and dung ... 40 2! 2. 20 ditto . -j j 3- 10 ditto 45 0 4- 10 ditto 46 i 5. 10 loads of bones, each 40 bushels ..." 63 i 6- 20 ditto .....'.'. 57 o 7. 200 bushels of lime ~g Ta 8. 40 yards of chalk „ 9. No manure 2i 10. 80 yards of chalk . . . 25 21 n. 120 ditto THEORIES CONCERNING SOIL FERTILITY 305 PRODUCE P.ER ACRE No. B. P. 12. 40 yards of chalk, earth mixed with train oil, six months ago, and often turned 33 oj 13. 40 ditto, earth mixed with urine, four months ago, and often turned 37 2 14. 40 ditto, earth alone 33 o£ 15. 40 ditto, earth from the farmyard 37 2 16. 1 20 ditto, red gravelly loam 29 i£ 17. 160 ditto 31 i " N.B. The season was remarkable dry, which circumstance certainly had a considerable effect upon the different crops." "ON BONES USED AS A MANURE "Bv ANTHONY ST. LEGER, ESQ. "During a long course of speculative and practical Agriculture, in which, with critical exactness, I employed myself in making experiments upon almost every kind of manure, I was fortunate enough to discover that bones are su- perior to all the manures made use of by the farmer. "Eight years ago I laid down to grass a large piece of very indifferent lime- stone land with a crop of corn (Wheat, presumably); and, in order that the grass seeds might have a strong vegetation, I took care to see it well dressed. From this piece I selected three roods of equal quality with the rest, and dressed them with bones broken very small, at the rate of sixty bushels per acre. Upon lands thus managed, the crop of corn was infinitely superior to the rest. The next year the grass was also superior, and has continued to preserve the same superiority ever since, insomuch that in spring it is green three weeks before the rest of the field. "This year I propose to plow up the field as the Festuca sylvatica (prye grass) has overpowered the grass seed originally sown. And here it will be proper to remark that, notwithstanding the species of grass is the natural prod- uce of the soil, the three roods on which the bones were laid have hardly any of it, but, on the contrary, have all along produced the finest grasses, "Last year I dressed two acres with bones, in two different fields prepared for turnips, sixty bushels to the acre, and had the pleasure to find the turnips greatly superior to the others managed in the common way. I have no doubt but these two acres will preserve their superiority for many years to come, if I may be allowed to prognosticate from former experiments most carefully con- ducted. "I also dressed an acre of grass ground with bones last October (1774) and rolled them in. The succeeding crop of hay was an exceeding good one. However, I found from repeated experience that, upon grass ground, this kind of manure exerts itself more powerfully the second year than the first. "It must be obvious to every person, that the bones should be well broken 306 SYSTEMS OF PERMANENT AGRICULTURE before they can be equally spread upon the land. No pieces should exceed the size of marbles. To perform this necessary operation, I would recommend the bones to be sufficiently bruised by putting them under a circular stone, which, being moved round upon its edge by means of a horse, in the manner that tan- ners grind their bark, will very expeditiously effect the purpose. At Sheffield it is now become a trade to grind bones for the use of the farmer. Some people break them small with hammers upon a piece of iron, but that method is in- ferior to grinding. To ascertain the comparative merit of ground and unground bones, I last year dressed two acres of turnips with large bones, in the same field where the ground ones were used ; the result of this experiment was, that the unground material did not perform the least service ; while those parts of the field on which the ground bones were laid were greatly benefited. "I find that bones of all kinds will answer the purpose of a rich dressing, but those of fat cattle I apprehend are the best. The London bones, as I am in- formed, undergo the action of boiling water, for which reason they must be much inferior to such as retain their oily parts ; and this is another of the many proofs given in these essays that oil is the food of plants. The farmers in this neighborhood are become so fond of this kind of manure, that the price is now advanced to one shilling and fourpence per bushel, and even at that price they send sixteen miles for it. "I have found it a judicious practice to mix ashes with the bones; and this winter I have six acres of meadow land dressed with that compost. A cart load of ashes may be put to thirty or forty bushels of bones, and when they have heated for twenty-four hours (which may be known by the smoking of the heap) let the whole be turned. After laying ten days longer, this most excellent dress- ing will be fit for use." In 1822, William Corbett, in his compilation of the writings of JethroTull, made the following statements: "Mr. TulPs main principle is this, that tillage will supply the place of manure; and his own experience shows that a good crop of wheat, for any number of years, may be grown every year upon the same land without any manure from first to last." "Mr. Tull continued his wheat crops to the harvesting of the twelfth upon the same land without manure; and when he concluded his work, he had, as he in- forms us in a memorandum, the thirteenth crop coming on, likely to be very good." It may be stated, however, that, after the time of Jethro Tull and before Corbett's republication of the Tullian methods and theories, some truly scientific facts had been discovered. In fact, chemistry had begun to assume the character of an exact science. Priestly had discovered oxygen and also identified as oxygen the gas which others had previously observed is given off from the THEORIES CONCERNING SOIL FERTILITY 307 leaves of plants under the influence of sunlight;1 Senebier had shown that the carbon of the plant is derived from the carbon dioxid of the air; and De Saussure had analyzed the ash of many plants, had shown that these ash constituents were derived from the soil, and that, though small in quantity as compared with the amount of material furnished to the plant by the air and water, the ash constituents were also essential to plant growth. De Saussure's publication in 1804 of his " Reserches Chimique sur la Vegetation" gave to the world the first definite and approxi- mately cotrect statement concerning the requirements and sources of plant food. While Davy's lectures on Agricultural Chemistry (first published in 1813) did much to extend the existing knowledge, and the investigations of Bousingault and Lawes began to develop (about 1835), it remained for Liebig to bring together the work of all and present it in a more comprehensive form in his " Organic Chemistry in its Application to Agriculture and Physiology," published in 1840. Thus, while Liebig is popularly known as the "Father of Agri- cultural Chemistry," the more fundamental contributions to knowl- edge concerning soil fertility and plant growth have been made by Senebier- (of Switzerland), De Saussure (of France), Lawes and Gilbert (of England), and Hellriegel (of Germany), the last being the discovery (in 1886) of nitrogen fixation by the root-tubercle bacteria of legumes. Liebig devoted much effort toward the proof of his theory that the ammonia of the air is the source o*f nitrogen for plants; but in this he failed, and Lawes and Gilbert's laboratory and field investigations at Rothamsted, which were in part planned for the purpose of disproving Liebig's nitrogen theory, clearly established the fact that in the main the soil must furnish nitrogen as well as the mineral elements of plant food. The following quotations from Liebig's writings are interesting; and they are also instructive, in that they well illustrate the weak- ness of drawing quantitative deductions and specific conclusions from qualitative data and general observations. Thus wrote Liebig: 1 Any one may observe the bubbles of oxygen formed upon fresh leaves placed under water in the sunlight. 308 SYSTEMS OF PERMANENT AGRICULTURE "Let us picture to ourselves the condition of a well-cultured farm, so large as to be independent of assistance from other quarters. On this extent of land there is a certain quantity of nitrogen contained both in the corn and fruit which it produces, and in the men and animals which feed upon them, and also in their excrements. We shall suppose this quantity to be known. The land is culti- vated without the importation of any foreign substance containing nitrogen. Now, the products of this farm must be exchanged every year for money, and other necessaries of life, for bodies, therefore, which contain no nitrogen. A certain proportion of nitrogen is exported with corn and cattle ; and this ex- portation takes place every year, without the smallest compensation ; yet after a number of years, the quantity of nitrogen will be found to have increased. Whence, we may ask, comes this increase of nitrogen ? The nitroge*n in the ex- crements cannot reproduce itself, and the earth cannot yield it. Plants, and con- sequently animals, must, therefore, derive their nitrogen from the atmosphere. "The last products of the decay and putrefaction of animal bodies present themselves in two different forms. They are in the form of a combination of hydrogen and nitrogen, — ammonia, in the temperate and cold climates, and in that of a compound, containing oxygen, nitric acid, in the tropics and hot climates. The formation of the latter is preceded by the production of the first. Ammonia is the last product of the putrefaction of animal bodies; nitric acid is the product of the transformation of ammonia. A generation of a thousand million men is renewed every thirty years: thousands of millions of animals cease to live and are reproduced in a much shorter period. Where is the nitrogen which they contained during life? There is no question which can be answered with more positive certainty. All animal bodies, during their decay, yield the nitrogen, which they contain to the atmosphere, in the form of ammonia. Even in the bodies buried sixty feet underground in the church- yard of the Eglise des Innocens, at Paris, all the nitrogen contained in the adi- pocere was in the state of ammonia. Ammonia is the simplest of all the com- pounds of nitrogen ; and hydrogen is the element for which nitrogen possesses the most powerful affinity. " The nitrogen of putrefied animals is contained in the atmosphere as ammonia in the form of a gas which is capable of entering into combination with carbonic acid, and of forming a volatile salt. Ammonia in its gaseous form as well as all its volatile compounds are of extreme solubility in water. Ammonia, there- fore, cannot remain long in the atmosphere, as every shower of rain must con- dense it, and convey it to the surface of the earth. Hence, also, rain water must, at all times, contain ammonia, though not always' in equal quantity. It must be greater in summer than in spring or in winter, because the intervals of time between the showers are in summer greater; and when several wet days occur, the rain of the first must contain more of it than that of the second. The rain of a thunderstorm, after a long-protracted drought, ought for this reason to contain the greatest quantity, which is conveyed to the earth at one time. . . ." "If a pound of rain water contain only one fourth of a grain of ammonia, THEORIES CONCERNING SOIL FERTILITY 309 then a field of 40,000 square feet must receive annually upwards of 80 Ib. of ammonia, or 65 Ib. of nitrogen; for, by the observations of Schubler, which were formerly alluded to, about 700,000 Ib. of rain fall over this surface in four months, and consequently the annual fall must be 2,500,000 Ib. This is much more nitrogen than is contained in the form of vegetable albumen and gluten, in 2650 Ib. of wood, 2800 Ib. of hay, or 200 cwt. of beet root, which are the yearly produce of such a field, but it is less than the straw, roots, and grain of corn which might grow on the same surface, would contain. "Experiments, made in this laboratory (Giessen) with the greatest care and exactness, have placed the presence of ammonia in rain water beyond all doubt. It has hitherto escaped observation, because no person thought of searching for it.1 All the rain water employed in this inquiry was collected 600 paces southwest of Giessen, whilst the wind was blowing in the direction of the town. When several hundred pounds of it were distilled in a copper still, and the first two or three pounds evaporated with the addition of a little muriatic acid, HC1, a very distinct crystallization of sal-ammoniac (NH4C1) was ob- tained : the crystals had always a brown or yellow color. "Ammonia may likewise be always detected in snow water. Crystals of sal-ammoniac were obtained by evaporating in a vessel with muriatic acid several pounds of snow, which were gathered from the surface of the ground in March, when the snow had a depth of 10 inches. Ammonia was set free from these crystals by the addition of hydrate of lime. The inferior layers of snow, which rested upon the ground, contained a quantity decidedly greater than those which formed the surface. "It is worthy of observation, that the ammonia contained in rain and snow water possesses an offensive smell of perspiration and animal excrements, — a fact which leaves no doubt respecting its origin. . . ." "We find this nitrogen in the atmosphere, in rain water, and in all kinds of soils, in the form of ammonia, as a product of the decay and putrefaction of pre- ceding generations of animals and vegetables. We find, likewise, that the pro- portion of azotized matters in plants is augmented by giving them a larger supply of ammonia conveyed in the form of animal manure. "No conclusion can then have a better foundation than this, that it is the ammonia of the atmosphere which furnishes nitrogen to plants." As an average of 15 years, the total amount of nitrogen brought to earth in rain and snow was found to be 3.97 pounds per acre per annum, at Rothamsted. Other records, varying from 3 to 7 years, have shown 3.45 pounds per acre per annum on the Barbados Islands, 3.54 pounds in British Guiana, 3.69 pounds in Kansas, 5.42 pounds in Utah, and 3.64 pounds in Mississippi; while the records from Paris show 8.93 pounds, and those from Gembloux, 1 " It has been discovered by Mr. Hayes in the rain water in Vermont." — W. 3io SYSTEMS OF PERMANENT AGRICULTURE Belgium, 9.20 pounds, both of which are doubtless influenced by the atmosphere from the cities with their numerous factories and other sources of pollution. Professor Shutt reports 4.32 pounds of nitrogen per acre in one year's precipitation at Ottawa, Canada, in 37.35 inches, of which 3.24 pounds were found in 24.05 inches of rain and 1.08 pounds in 13.3 inches of snow water (corresponding to about 133 inches of snow), the average composition being based upon analyses of 46 samples of rain water and 32 samples of snow water. Of the nitro- gen found in rain water, 61 per cent existed in free ammonia, 22 per cent in nitrate (and nitrite) form, and 17 per cent as organic nitrogen, the corresponding percentages for snow water being 56, 34, and 10. Liebig also discussed very interestingly and, in the main, very erroneously, the reasons for the value of crop rotation. In 1840 he wrote as follows: " Of all the views which have been adopted regarding the cause of the favor- able effects of the alternations of crops, that proposed by M. Decandolle alone deserves to be mentioned as resting on a firm basis. " Decandolle supposes that the roots of plants imbibe soluble matter of every kind from the soil, and thus necessarily absorb a number of substances which are not adapted to the purposes of nutrition, and must subsequently be expelled by the roots, and returned to the soil as excrements. Now as excrements can- not be assimilated by the plant which ejected them, the more of these matters which the soil contains, the more unfertile must it be for plants of the same species. These excrementitious matters may, however, still be capable of as- similation by another kind of plants, which would thus remove from the soil, and render it again fertile for the first. And if the plants last grown also expel substances from their roots, which can be appropriated as food by the former, they will improve the soil in two ways. "Now a great number of facts appear at first sight to give a high degree of probability to this view. Every gardener knows that a fruit tree cannot be made to grow on the same spot where another of the same species has stood; at least not until after a lapse of several years. Before new vine stocks are planted in a vineyard from which the old have been rooted out, other plants are cultivated on the soil for several years. In connection with this it has been ob- served, that several plants thrive best when growing beside one another; and, on the contrary, that others mutually prevent each other's development. Whence it was concluded, that the beneficial influence in the former case depended on a mutual interchange of nutriment between the plants, and the injurious one in the latter on a poisonous action of the excrements of each on the other respectively. THEORIES CONCERNING SOIL FERTILITY 311 "A series of experiments by Macaire-Princep gave great weight to this theory. He proved beyond all doubt that many plants are capable of emitting extractive matter from their roots. He found that the excretions were greater during the night than by day, and that the water in which plants of the family of the Leguminosa grew, acquired a brown color. Plants of the same species, placed in water impregnated with these excrements, were impeded in their growth, and faded prematurely, whilst, on the contrary, corn plants grew vigorously in it, and the color of the water diminished sensibly ; so that it appeared, as if a certain quantity of the excrements of the Leguminosa had really been absorbed by the corn plants. These experiments afforded as their main result, that the characters and properties of the excrements of different species of plants are different from one another, and that some plants expel excrementitious matter of an acrid and resinous character; others mild (douce) substances resembling gum. The former of these, according to Macaire-Princep, may be regarded as poisonous, the latter as nutritious. "The experiments of Macaire-Princep are positive proof that the roots, prob- ably of all plants, expel matters, which cannot be converted in their organism either into woody fiber, starch, vegetable albumen, or gluten, since their ex- pulsion indicates that they are quite unfitted for this purpose. But they cannot be considered as a confirmation of the theory of Decandolle, for they leave it quite undecided whether the substances were extracted from the soil, or formed by the plant itself from food from another source. It is certain that the gummy and resinous excrements observed by Macaire-Princep could not have been contained in the soil ; and as we know that the carbon of a soil is not diminished by culture, but, on the contrary, increased, we must conclude that all excrements which contain carbon must be formed from the food ob- tained by plants from the atmosphere. Now, these excrements are compounds, produced in consequence of the transformations of the food, and of the new forms which it assumes by entering into the composition of the various organs. "M. Decandolle' s theory is properly a modification of an earlier hypothesis, which supposed that the roots of different plants extracted different nutritive substances from the soil, each plant selecting that which was exactly suited for its assimilation. According to this hypothesis, the matters -incapable of as- similation are not extracted from the soil, whilst M. Decandolle considers that they are returned to it in the form of excrements. Both views explain how it happens that after corn, corn cannot be raised with advantage, nor after peas, peas; but they do not explain how a field is improved by lying fallow, and this in proportion to the care with which it is tilled and kept free from weeds ; nor do they show how a soil gains carbonaceous matter by the cultivation of certain plants, such as lucern and esparsette. "Theoretical considerations on the process of nutrition, as well as the ex- perience of all agriculturists, so beautifully illustrated by the experiments of Macaire-Princep, leave no doubt that substances are excreted from the roots of plants. . . ." 3i2 SYSTEMS OF PERMANENT AGRICULTURE "It is scarcely necessary to remark that this excrementitious matter must undergo a change before another season. During autumn and winter it begins to suffer a change from the influence of air and water; its putrefaction, and, at length, by continued contact with the air, which tillage is the means of procur- ing, its decay are effected; and at the commencement of spring it has become converted, either in whole or in part, into a substance which supplies the place of humus, by being a constant source of carbonic acid. " The quickness with which this decay of the excrements of plants proceeds, depends on the composition of the soil, and on its greater or less porosity. It will take place very quickly in a calcareous soil ; for the power of organic ex- crements to attract oxygen and to putrefy is increased by contact with the al- kaline constituents, and by the general porous nature of such kinds of soil, which freely permit the access of air. But it requires a longer time in heavy soils consisting of loam or clay. "The same plants can be cultivated with advantage on one soil after the second year, but in others not until the fifth or ninth, merely on account of the change and destruction of the excrements which have an injurious influence on the plants being completed in the one, in the second year; in the others, not until the ninth. "In some neighborhoods, clover will not thrive until the sixth year; in others not till the twelfth; flax in the second or third year. All this depends on the chemical nature of the soil ; for it has been found by experience that in those districts where the intervals at which the same plants can be cultivated with advantage, are very long, the time cannot be shortened even by the use of the most powerful manures. The destruction of the peculiar excrements of one crop must have taken place before a new crop can be produced. "Flax, peas, clover, and even potatoes, are plants the excrements of which, in argillaceous soils, require the longest time for their conversion into humus; but it is evident, that the use of alkalies and burnt lime, or even small quantities of ashes which have been lixiviated, must enable a soil to permit the cultivation of the same plants in a much shorter time. "A soil lying fallow owes its early fertility, in part, to the destruction or con- version into humus of the excrements contained in it, which is effected during the fallow season, .at the same time that the land is exposed to a further disinte- gration." In the first American edition of Liebig's book, published in 1841, Doctor John W. Webster, then Professor of Chemistry in Harvard University, inserted an appendix, in which he wrote as follows: ' "It should be stated that the accuracy of the experiments of Macaire-Princep adduced by the author (Liebig) is not generally admitted. Other chemists have been unable to obtain similar results, or, if they do, are inclined to ascribe them to injury of the roots of the plants examined. Professor Lindley has in his notice of Liebig's work remarked that he has no fixed opinion on the sub- ject, it being a question of facts and not of induction." THEORIES CONCERNING SOIL FERTILITY Liebig so emphasized the importance of the mineral plant food, as established by De Saussure's careful work, that it has ever since been referred to as " Liebig's mineral theory of plant nutrition." In recent years, Whitney and Cameron have revived Decandolle's theory of toxic excreta from plant roots, in support of another more radical theory announced by them, to the effect that soils do not wear out or become depleted by cultivation and cropping. While this theory is advanced with no adequate foundation and in direct opposition to practical experience and to so many known facts of mathematics, chemistry, and geology, that it is in itself quite unworthy of further consideration, the fact is that it has been pro- mulgated by Professor Whitney as Chief of the United States Bureau of Soils, and by Doctor Cameron, as the chief chemist of the same Bureau; and, consequently, it cannot be ignored. The author finds practically no support for these radical theories, either in the American Experiment Station bulletins or in the publications from the older scientific bureaus at Washington, such as the United States Geological Survey, the Bureau of Chem- istry, and the Bureau of Plant Industry; while they are directly contrary to the teachings of all recognized European authorities. But even above any so-called authorities, we must recognize facts, if there are any, for an opinion contrary to the facts is of no per- manent value by whomsoever it may be held. The following statements from Whitney and Cameron will give a clear idea of the plain teachings of the Bureau of Soils, so far as represented by its leaders. Thus, on page 64 of Bulletin 22 of the Bureau of Soils, published by Whitney and Cameron in 1903, we read: " That practically all soils contain sufficient plant food for good crop yields, that this supply will be indefinitely maintained," etc. Again, on pages 21 and 22, Farmers' Bulletin 257, published in 1906, we have the following definite statements from Professor Whitney: "There is another way in which the fertility of the soil can be maintained, viz., by arranging a system of rotation and growing each year a crop that is not injured by the excreta of the preceding crop; then when the time comes round for the first crop to be planted again, the soil has had ample time to dispose of the sewage resulting from the growth of the plant two or three years before. This, I 3i4 SYSTEMS OF PERMANENT AGRICULTURE think, is the basis or reason in many cases for our crop rotation, viz., that these excreted substances are not toxic alike for all plants, and the soil has time to recover its tone and cleanse itself. I have told you that barley will follow po- tatoes in the Rothamsted experiments after the potatoes have grown so long that the soil will not produce potatoes. The barley grows unaffected by the excreta of the potatoes, another crop follows the barley, and the soil is then in condition to grow potatoes again." Again in the report of the Hearings before the Committee on Agriculture of the United States House of Representatives, under date of January 28, 1908, page 428, we find the following statements by Professor Whitney: "The investigations of the Bureau of Soils, as to the cause of the deterioration of soils, and the causes that limit crop production, have changed the viewpoint of the entire world." On pages 445-449 of the same publication, Doctor Cameron makes the following statements: " All soils contain practically all the common rock-forming minerals. Now, it is a principle of chemistry that when a solvent is brought in contact with a substance, that substance will go into solution until there is a state of equilib- rium between the quantity of the substance outside and inside ; in other words, we get sc saturated solution. If these rock-forming minerals were in all soils, we should have the same solution in every soil, and that has been shown to be the case. There are various variations, due to absorption, perhaps, of the soil. In the first place, I must ask you gentlemen to remember that the soil and the plant and the water in the soil is moving. The soil grains are constantly moving, and the solution in the soil is constantly moving, and the growing plant is constantly moving. If a plant stops for a moment, it dies. The soil solution cannot stop for a moment, because it has to be moving all the time. When water falls on the soil, part of it runs off the surface, and part of it runs through the surface by gravitation and comes out in the subsoil, and part of it starts and rises as soon as we get sunlight on the surface, and this part comes up in films over and through the finer spaces, and is bringing with it dissolved material from below. "The water that falls and goes through down and out goes rapidly through larger openings, and gets very little of the soluble material, because it is not long in contact with the soil grains. It gets some by reason of the fact that, as we know, our springs and rivers and wells are all soil solutions, and carry mineral matter. Now, water rising by capillarity cannot get very concentrated be- cause it gets saturated with the minerals, and any excess that is contained in it is thrown out, except in extreme conditions, as in the West, and then we get alkali conditions; but under ordinary humid conditions we cannot have an excess of it, and the soil solution is bringing materials from below which the THEORIES CONCERNING SOIL FERTILITY 315 plant gets, and, as a matter of fact, the most important discovery of the Bureau of Soils in recent years is that plants are feeding on material from the subsoils, far below where the roots go." Subsequent to this statement, the following dialogue is recorded : The Chairman. "When you say that all soils contain all the elements of plant food, and there is in those soils at all times a saturated solution of which all these elements of plant food make a part, do you not practically say that all soils have all the plant food they need, and that it is at all times available for the plant ; or is it not available for the plant if it is in a saturated solution ? " Mr. Cameron. "Certainly, if there is water enough; if the soil is moist." The Chairman. "Is it not therefore a justifiable inference from what you have said, that there is all the time in all soils enough plant food available for plant life?" Mr. Cameron. "True; perfectly true as regards mineral nutrients." The Chairman. "Then I come back again to the question, why is it neces- sary, or is it in your judgment necessary, ever at any time to introduce fertiliz- ing material into any soil for the purpose of increasing the amount of plant food in that soil." Mr. Cameron. "Not in my judgment." The Chairman. "Then in your judgment the only reason for the introduc- tion of fertilizers is for the antitoxic effect or the mechanical effect they may have on the soil." Mr. Cameron. "Mainly that, but there are other functions of fertilizers that we know comparatively little about. We know that certain kinds of life, bacteria, molds, can grow in certain solutions of salts, and cannot in others. It may be that fertilizers affect them. But all that is an unexplored field, and little is known about it. ... If you will allow me to say one more word about fertilizers: What are fertilizers? What are the characteristics that a substance must have in order to be a fertilizer? It must be obtained in large quantities. It must also be cheap. Now, the substances which are used as fertilizing mate- rial are substances which can be obtained in large quantities. They are sub- stances, and are the only substances, which we can get hold of that we can get in large quantities, that we can get cheap, and with one exception — that is, sodium chlorid — common salt. It has not been much used as a fertilizer, because it has not any so-called plant food in it ; and yet it has been used in quite a large number of experiments on quite a large scale, and wherever it has been used, it has generally been found to be quite a good fertilizer. In the investiga- tions of the Bureau we have used pyrogallol. It contains no plant food, but carbon, hydrogen, and oxygen, yet, nevertheless, it is a powerful fertilizer ; ' but 1 Director Wheeler of the Rhode Island Agricultural Experiment Station reported to the Graduate School of Agriculture held at Cornell University, July, 1908, that a thorough investigation under field conditions at the Rhode Island Station showed practically no benefit from the use of pyrogallol as a fertilizer; whereas, very marked effects were produced by manures and commerical fertilizers. — C.G.H. 3i 6 SYSTEMS OF PERMANENT AGRICULTURE cannot be obtained cheaply. It is worth over $2 a pound, and nobody would think of recommending it as a fertilizer. . . ." "There has not been a publication on the subject of soil fertility going out from the Bureau of Soils — and I think I can speak advisedly, for every one has gone through my hands — in which we did not have the experimental proof long before the publication went out, and that this is being recognized I think I can claim by the fact that a number of agrkultural colleges in the country are using our bulletins as text-books. I have recently come from a lecture trip extending from Louisiana to Michigan, and I found everywhere that this is being taught, and, as I say, our publications are being used for text-books." On page 5 of Farmers' Bulletin 257, Professor Whitney makes the following statements: "I shall be glad, however, to speak of certain general features of the essential and broadly applicable laws of soil fertility that the Bureau of Soils, with its large force of field men and its large force of chemists and soil physicists, has investigated in the last twelve years. We think that as a result of this work we understand far more of the principles of soil fertility now than we ever have before, and I wish to give the results in words as simple as possible. You need not necessarily believe everything I say (because I cannot say truly that I believe everything myself, but only that our opinions seem reasonable deduc- tions)." In general, the soil fertility theories of Whitney and Cameron may be briefly summarized in the following statements, all of which are direct quotations: 1. "It appears further that practically all soils contain sufficient plant food for good crop yields, that this supply will be indefinitely maintained, and that the actual yield of plants adapted to the soil depends mainly, under favorable climatic conditions, upon the cultural methods and suitable crop rotation." —WHITNEY and CAMERON, in Bureau of Soils Bulletin 22, page 64. 2. "In all soils there are rock particles or minerals containing phosphoric acid and potash, and in all the soil solutions that we have ever examined — and we have examined hundreds of them from all parts of the country — you will be astonished to learn that the composition and concentration of the soil moisture, which is the nutrient solution spread throughout the surface soil of the earth for plants to grow in and to gather their food from,— you will be astonished to learn that the concentration of this soil moisture is sensibly the same whether we examine your sandy truck soils on your river necks, your sandy clay wheat soils on the uplands, the Hagerstown clay in the valley of the Shenandoah, or the black prairie soils of the West. These minerals are contributing to the solution in which the plant feeds. As I have said, these minerals are difficultly soluble, but they are appreciably soluble. They are oluble enough to maintain a solution which is amply sufficient for the plants to THEORIES CONCERNING SOIL FERTILITY 317 gather their food from. All soils having, broadly speaking, all of these minerals in them, have approximately the same composition in their soil moisture. "This is a very astonishing fact, but, looked upon in the light of our experi- ments, it is an actual fact that all soils contain sufficient plant food for the sup- port of plants. Further, when the plant takes into its substance some of the mineral matter from the solution, the solid minerals in contact with the solution immediately dissolve, and the solution is restored to its former concentration. The exhaustion of the soil, therefore, is merely a relative phrase and resolves itself into the question of the rate at which the solution can recover itself. I may state to you that the rate is as fast on an acre planted in our ordinary crops as the demand made upon it by the plant." — WHITNEY, in Farmers' Bulletin 257, pages 10, n. 3. "It is not to be denied that plants will not infrequently do better when they are growing in a soil, a nutrient solution, or a soil solution many times stronger than they actually need. . . . "If we take a plant and grow it in a water culture, the plant does better if we have a solution containing several times more phosphorus and potash than it actually needs to feed on. Why it is we do not know, but granting that the plant does better in a solution stronger than it actually needs as a food, we still have a solution in the soil apparently strong enough for any need the plant may have. "Now we come to a very interesting thing to the farmer. If soils have suffi- cient food for the needs of plants and if this supply is constantly maintained, as I say, by the solution of these minerals in the soil, then what is the function of fertilizers, and what do we mean by worn-out lands or exhausted lands ? . . . The chemical idea of the exhaustion of a soil is not logical in the light of the ex- perience which all of us have seen, that when fertilizers are applied, the soils are not always made immediately productive. You can go into many of the regions of the worn-out soils of our Eastern states and reclaim those soils or make them productive, but not with any amount of fertilizers you can apply." "I should say that the soil ought to take care of the excrement of plants. It is its business to do so. It is its proper function. Whether it does this through the agency of bacteria, whether it is due to the abnormal absorptive power of the soil or to direct oxidation, we do not know. It is probably due in part to each. Take a natural soil, a prairie sod; the sanitary conditions in that soil are almost perfect." * — WHITNEY, in Farmers' Bulletin 257, pages n, 12, and 15. 4. "Apparently these small amounts of fertilizers we add to the soil have their effect upon these toxic substances and render the soil sweet and more healthful for growing plants. We believe that it is through this means that our fertilizers act rather than through the supplying of plant food to the plant." — WHITNEY, in Farmers' Bulletin 257, page 20. 1 See Table 70 for effect of plant food on permanent grass park more than 250 years old. — C. G. H. 3i8 SYSTEMS OF PERMANENT AGRICULTURE 5. "I have attempted to show you the way I believe fertilizers act and the reason we use them. I think that this is the way stable manure and green manures act. I think that is the principal office of nitrate of soda, potash, and phosphoric acid ; but they do not all act alike on the same soils. We are work- ing now on a soil in Iowa which with stable manure every time produces a smaller crop than without. . . . 6. "There is another way in which the fertility of the soil can be maintained, viz., by arranging a system of rotation and growing each year a crop that is not injured by the excreta of the preceding crop; then when the time comes round for the first crop to be planted again, the soil has had ample time to dispose of the sewage resulting from the growth of the plant two or three years before." — WHITNEY, in Farmers' Bulletin 257, page 21. 7. "The soil solution is bringing materials from below which the plant gets, and as a matter of fact the most important discovery of the Bureau of Soils in recent years is that plants are feeding on materials from the subsoils, far below where the roots go." — CAMERON, in the Hearings before the Committee on Agriculture of the United States House of Representatives, page 446 (1908). 8. The Chairman. "Then I come back again to the question, Why is it necessary, or is it in your judgment necessary, ever at any time to introduce fertilizing material into any soil for the purpose of increasing the amount of plant food in that soil?" Mr. Cameron. " Not in my judgment." — Hearings before the Committee on Agriculture of the United States House of Representatives, page 446 (1908). 9. " In the truck soils of the Atlantic coast, where 10 or 15 tons of stable manure are annually applied to the acre, in the tobacco lands of Florida, and of the Connecticut Valley, where 2000 or 3000 pounds of high-grade fertilizers carrying 10 per cent of potash are used, even when these applications have been continued year after year for a considerable period of time, the dissolved salt content of the soil as shown by this method is not essentially different from that in surrounding fields that have been under extensive cultivation. "In England and in Scotland it is customary to make an allowance to tenants giving up their farms for the unused fertilizers applied in the previous seasons. The basis of this is usually taken from 30 to 50 per cent for the first year, and at 10 to 20 per cent for the second year after application, but in the experience of this Bureau there is no such apparent continuous effect of fertilizers on the chemical constitution of the soil." - WHITNEY and CAMERON, in Bureau of Soils Bulletin 22, page 59. The question may be asked if the plant food brought to the sur- face by capillary moisture in humid sections is greater than that lost by leaching. Compare, for example, the composition of the old prairie soil (gray silt loam) in the lower Illinoisan glaciation THEORIES CONCERNING SOIL FERTILITY 319 and the more recent prairie soil (brown silt loam) of the late Wis- consin glaciation. Compare also the amounts in the surface and subsoil (in 2 million pounds of each) of potassium, or any other ele- ment which does not accumulate in plant residues. Note whether the calcium carbonate on Broadbalk and Hoos fields at Rothamsted is steadily accumulating at the surface. There are abundant sup- plies in the subsoil " far below where the roots go." Note the com- plete absence of calcium carbonate in very many Illinois soils. (See also Tables 4, 5, and 21 in the preceding pages.) Attention is called to the fact that nitrification is a process of biochemical action and not one of mere solution. Director Hall of the Rothamsted Experiment Station makes the following statement in his "Fertilizers and Manures" (1909), page 288. " When the Rothamsted soils, with their long-continued differences in fertilizer treatment, are extracted with water charged with carbon dioxide — the nearest laboratory equivalent to the actual soil water — the amount of phosphoric acid going into solution is closely proportional to the previous fertilizer supply, and this proportionality is maintained if the extraction is repeated with fresh solvent, as must be the case in the soil." It should be kept in mind, of course, that a one-crop system followed year after year upon the same land usually encourages the growth of certain weeds whose " habits " are similar to those of the crop grown, that it also tends toward the breeding of insect enemies and to the development of fungous diseases peculiar to that crop, such as " flax sickness," investigated by the North Da- kota Experiment Station, and " clover sickness," which has long been thought to be an actual fact in practical agriculture, concern- ing which the Tennessee Station has recently reported some prom- ising results. The legume plants appear to be especially suscep- tible to such fungous diseases, the cowpea wilt being well known, and " bean-sick " soil is a common expression. It seems probable that bacterial as well as fungous diseases may develop under suit- able conditions. While it is possible that inanimate toxic substances may also be formed in the soil from possible plant excreta, or less improb- ably from the decomposition of the crop residues, there is no knowl- edge or evidence sufficient, in the author's opinion, to justify a theory that fertilizers act primarily as antitoxins. It should be 320 SYSTEMS OF PERMANENT AGRICULTURE remembered that well-fed plants are usually better able to resist or overcome the attacks of insects and diseases. It is well known that there are some exudations from germinating seeds, and it seems evident that water used repeatedly for 2O-day cultures with seedling plants becomes stagnant, putrid, or toxic, but can we correlate this with field conditions? Alkaline slag phosphate, acidulated rock phosphate, neutral steamed bone meal, and insoluble raw rock phosphate are very different chemical substances, and the very complete data already presented show that any one of these forms of phosphorus may be used to increase crop yields. Sodium nitrate, ammonium sulfate, and dried blood are exceedingly different substances, but they all contain nitrogen, and where nitrogen is deficient in the soil, any one of these materials will benefit the crop. Moreover, with legume plants, essentially the same results are secured whether nitrogen is supplied in dried blood or provided by the nitrogen-fixing bacteria without fertilizer application. It may be noted that while Whitney and Cameron in Bulletin 22 (1903), of the Bureau of Soils, included nitrogen as distinctly as . phosphorus, potassium, and calcium, as being contained in prac- tically all soils in an ample supply which " will be indefinitely maintained," and while Professor Whitney also asserts, in Farmers' Bulletin 257 (1906), that the correction of toxic substances is " the principal office of nitrate of soda, potash, and phosphoric acid," and while Cameron admits in the Hearings before the Com- mittee on Agriculture (1908) that it is never necessary at any time to introduce fertilizing material into any soil for the purpose of increasing the amount of plant food in that soil; nevertheless, Whitney and Cameron are beginning to qualify their theories by saying " mineral elements " or " mineral plant food," presumably because the mathematical opposition is too strong, considering that the soil contains but very small amounts of nitrogen " far below where the roots go." On December 9, 1908, the National Conservation Commission presented its report (prepared for the President) to the Conference of Governors and State Conservation Commissioners assembled in Washington, in which great emphasis was laid upon the impor- tance of conserving the supply of natural phosphates, as a result THEORIES CONCERNING SOIL FERTILITY 321 of which the President soon afterward withdrew from entry the re- maining government lands that were known to contain phosphate deposits, acting upon the advice of the United States Geological Survey and the National Conservation Commission; while on December 10, 1908, the daily press of the country very generally published a Washington dispatch headed " SOIL WON'T WEAR OUT," in which Professor Whitney was credited with the following statements: "There is a general impression among economists that soil fertility is declin- ing through loss of mineral plant food, but the Bureau of Soils, through the extensive soil surveys and investigations made in the laboratories and from the study of world-wide records, has determined that this impression of the decline of soil fertility is erroneous. "It is not unreasonable to expect that as this country becomes more thickly settled and our people are forced to cultivate smaller areas, with more intelli- gent and more intensive methods, the actual amount of crops obtained from the land now in crops can be increased two and one half times over what we are now producing. "But the amount of land in crops is only about one fourth of the amount in farms. Applying this ratio to the whole amount in farms, it is apparent that the land in farms at present can be expected to produce in time something like ten or twelve times the amount of crops that are now produced on these farms. "So far as the present outlook is concerned, the nation possesses ample resources in its soils for any conceivable increase in population for several centuries. "The Bureau of Soils finds that the decline in yield is due generally to the accumulation of organic products in the soil which are not eliminated through proper cultural methods as fast as they have accumulated, and that the failures that are reported are, therefore, due to improper methods of cultivation and crop rotation. "Our own government statistics show that during the last forty years the yields per acre of all our cereal crops has shown a tendency to increase. Statis- tics of all the European countries show that the yields in recent years have con- sistently increased." Of course this press dispatch would not be quoted here except that it is in strict accord with the persistent teaching of Whitney and Cameron, which will be found of greatest interest for compari- son with that of Jethro Tull or Doctor Hunter, and with Liebig's nitrogen theory. Since the above was written, Bulletin 55 of the Bureau of Soils, " Soils of the United States," by Milton Whitney, has been pub- 322 SYSTEMS OF PERMANENT AGRICULTURE lished (February, 1909), from which the following statements are quoted : "The soil is the one indestructible, immutable asset that the nation possesses. It is the one resource that cannot be exhausted ; that cannot be used up. The general conception of the exhaustion of soils is that the crop removes plant food, and that unless we return some considerable portion of plant food to the soil it eventually becomes incapable of longer producing adequate crops. We quote from a recent article in one of the agricultural journals : " 'We have warned our readers for the last ten years of what is coming if they continue to grow grain crops and sell them off the farm continuously from year to year. . . . Don't imagine for one minute that your soils are of inexhaust- ible fertility. No such soils were ever made in the Western Hemisphere, except, perhaps, such as are enriched by overflow every three or four years.' "The impression prevails that our crops take out phosphoric acid, potash, and nitrates to such an extent that the soil becomes incapable of longer supply- ing these plant-food constituents for a satisfactory yield." "As we see it now, the main cause of infertile soils or the deterioration of soils is the improper sanitary conditions originally present in the soil or arising from our injudicious culture and rotation of crops. // is, of course, exceedingly difficult to work out the principles which govern the proper rotation for any par- ticular soil.1 "The important thing is that we now understand the nature of the soil ; how it supplies the nutrient constituents for the crops and how it maintains the supply ; how crops may affect each other when grown in succession on the soil ; how cul- tivation affects the conditions resulting from the crop, and, lastly, we are begin- ning to understand how fertilizers come into this scheme and themselves act on or change toxic conditions in the soil, rendering the soil again sweet and healthy for the growing crop." "It has been shown that in southern Maryland and in middle Virginia the cause of the recent depression in agriculture and of the low yield of crops is due to methods which have prevailed rather than to any exhaustion of the soil, and that with improved methods these areas are coming up and will again be made to produce satisfactory crops. The soils are not wearing out in the sense that they are unable longer to provide mineral nutrients, but the yields are low because through the prevailing methods the soils have not been maintained in proper condition. In these latter instances the yields have actually declined, but not from the cause which has been generally ascribed. "It has been shown that from the modern conception of the nature and pur- pose of the soil it is evident that it cannot wear out, that so far as the mineral food is concerned, it will continue automatically to supply adequate quantities of the mineral plant foods for crops, but it has also been shown that the soil can be abused and its fertility temporarily impaired by improper methods of handling. 1 Italics mine. — C. G. H. THEORIES CONCERNING SOIL FERTILITY 323 "Lastly, it has been shown from the statistics of European countries that the soils of the world are not wearing out, but that, on the contrary, after a thousand years of cultivation, with the introduction of better methods, with the necessity of raising larger crops, these soils are responding with an increased yield even over what they produced at the beginning of the last century. "As a national asset the soil is safe as a means of feeding mankind for untold ages to come. So far as our investigations show, the soil will not be exhausted of any one or all of its mineral plant-food constituents. If the coal and iron give out, as it is predicted that they will before long, the soil can be depended on to furnish food, light, heat, and habitation not only for the present population, but for an enormously larger population than the world has at present." This general outline of soil-fertility theories has been introduced at this point in order that the reader may note their application in the following pages; and it is hoped that the preceding and suc- ceeding data are sufficient to enable him to form his own opinion. It is well to keep in mind a few general facts: e.g., that the total corn acreage of Rhode Island and Connecticut combined averages less than three townships (about one sixth of one average Illinois county, of which there are 102); that the total corn acreage of Maine, New Hampshire, Vermont, Massachusetts, Connecticut, Rhode Island, New York, New Jersey, Pennsylvania, Delaware, and Maryland, all combined, is less than the average corn acreage of Georgia, whose ten-year average yield is n bushels per acre, and less than one half the corn acreage of Illinois; that Illinois produces the same amount of corn per annum as the aggregate production of the six New England states, the six Middle Atlantic states, and the six South Atlantic and Gulf states — eighteen in all — extending from Maine to the mouth of the Mississippi, although Georgia, one of these states, is larger than Illinois ; that during the last ten years the average corn acreage of Illinois has been increased from 7 million to 10 million acres by putting under cultivation old blue-grass pastures and drained swamp areas representing the richest soil of the state; that in the Eastern states manure, made in part from food stuffs shipped from the newer states, is worth about $2 a ton; that level or gently undulating farm lands in Maryland and Virginia sell for less than $5 an acre, while those of Illinois and Iowa are worth $100 or $200; that, while England produces 32 bushels of wheat per acre with a total production of 50 million bushels, 324 SYSTEMS OF PERMANENT AGRICULTURE England imports 200 million bushels of wheat, 100 million bushels of corn, nearly a billion pounds of oil cake, and much phosphate and other fertilizing material; that Germany produces 125 million bushels of wheat, and in addition imports 75 million bushels of wheat, 40 million bushels of corn, a billion pounds of oil cake, and much phosphate, etc., and that Germany's chief export is 2 billion pounds of sugar (C^H^Ou); that Den- mark produces 4 million bushels of wheat, imports 5 million bushels of wheat, 15 million bushels of corn, 800 million pounds of oil cake, phosphates, etc., and exports 175 million pounds of butter; that Belgium produces 12 million bushels of wheat and imports 60 million bushels, etc. It is interesting also to keep in mind the following statement1 by Doctor Bernard Dyer in his American lectures on " Results of Investigations on the Rothamsted Soils," in connection with his discussion of the Broadbalk wheat plot that has received an annual application of 15.7 tons of farm manure since 1844: " It is to be borne in mind, however, that the quantity of dung used in these continuous wheat -growing experiments is, on the yearly average, far less than would be used in practical agriculture on any of the rotation systems." As early as 1855, England was importing annually more than 200,000 tons of guano from the west coast of South America and from the islands of the sea. The guanos vary in composition from about 15 per cent of nitrogen and 5 per cent of phosphorus to less than i per cent of nitrogen and more than 1 5 per cent of phosphorus. Aikman writes in "Manures and the Principles of Manuring" (1894) as follows concerning the use of bones in England: "Employed first in 1774, their use has steadily increased ever since, and their popularity as a phosphatic manure is among farmers in this country quite un- rivaled. . . . Soon their use became so popular that the home supply was found inadequate. . . . So largely were they used by English farmers that Baron Liebig considered it necessary to raise a warning protest against their lavish applica- cation : 'England is robbing all other countries of the condition of their fertility. Already in her eagerness for bones she has turned up the battlefields of Leipzig, of Waterloo, and of the Crimea; already from the catacombs of Sicily has she carried away the skeletons of many successive generations. Annually she recovers from the shores of other countries to her own the manurial equivalent of three millions and a half of men.' " 1 Page 50, Bulletin 106, Office of Experiment Stations, United States Depart- ment of Agriculture. THEORIES CONCERNING SOIL FERTILITY 325 Aikman states that at the present time about 100,000 tons of bones are used annually on English soils, and that bone ash is still imported from South America. The East Indians complain that England has robbed India of bones. The importation of mineral phosphates into England exceeded 250,000 tons in 1885, when more than a dozen countries were being drawn upon for this material, representing three continents and Australia; and as early as 1892 the United States was furnishing Great Britain more than 200,000 tons of phosphate a year. Besides this, England has her own phosphate deposits in the form of coprolites or phosphatic nodules, which, according to Aikman, " have been found in great abundance in the greensand formation, in the crag of the eastern counties, and in the chalk formations of the southern counties." He adds: "They are found in large quantities in Cambridgeshire. . . . They were also found in enormous quantities in Suffolk, Norfolk, Bedfordshire, and Essex, and were for a long time largely used in the manufacture of superphosphate (acid phosphate), but of late years have not been used to anything like the same extent, owing to the fact that there are richer and cheaper sources of phosphate of lime available." In addition to all this, England produces and supplies. to her soils large quantities of slag phosphate, the amount of which ex- ceeded 100,000 tons a year before the close of the last century, and her annual production has since risen to 300,000 tons per annum. France, Germany, and other small European countries are not far behind England in the matter of increasing the fertility of their soils. By 1890 France was using about 400,000 tons of phosphate annually, and this was supplemented by slag phosphate, the amount of which exceeded 200,000 tons in 1899, while Germany applied 800,000 tons of slag phosphate to her soils the same year. The application of phosphates to the soils of Europe has largely in- creased during the years of the present century. Thus, in 1907, Italy, with a total area of less than 115,000 square miles (about twice as large as Illinois), used 950,000 metric tons of phosphate (also 82,000 tons of nitrogen fertilizer, and 7000 tons of potassium salts; and during the five years, 1904 to 1908, more than i^ million long tons of Florida phosphate were shipped to Germany. 326 SYSTEMS OF PERMANENT AGRICULTURE Since the promulgation of much definite knowledge during the first half of the last century, by such teachers as De Saussure, Davy, Bousingault, Liebig, and Lawes and Gilbert, the increasing appli- cations of phosphates, manures made in part from imported food stuffs, and other fertilizing materials, including more or less po- tassium salts and nitrates, and in more recent years a larger use of legumes, are found to bear fruit in the corresponding increase in the crop yields of western Europe, as will be seen from the follow- ing crop statistics, compiled by Professor Wilhelm Kellerman (Landwirtschaftliches Jahrbuch, 1906, page 289) and republished by the United States Bureau of Soils (Bulletin 55) for the purpose of showing that soils do not wear out. The data from Schmatzfeld are of interest because jof the old records, but they appear to represent in the main single years, and in part selected years. Even the tenth-year records from 1830 to 1870 may signify but little. Thus the rye and oats for 1870 average less than for 1830. The late averages are, of course, very significant. The Trebsen records have much value because they include several lo-year averages which show no advancement prior to the publication of De Saussure's work, which gave to the world the YIELDS or CEREALS IN SCHMATZFELD, GERMANY YEAR WHEAT (Bu.) RYE (Bu.) BARLEY (Bu.) OATS (Bu.) I "5 52-1 5 5 7 . 12 < 122 I/I 2 14 8 1660 12 8 8 ? 1670 . 14 6 °-o 16 i iZ-6 T7 A 1822 18 7 i /-4 182"? . 18 i z4-6 66- 1 ?R T 1810 . 18 7 2? 6 32-5 1840 or 6 35-u 21 6 1850 . . 28 7 31 ON H PI O t~- PI 0 CO 't in 0 X r«5 C! i/) in « •*• X O tJ-00 00 O 00 PI in Tt- 0 "H " t^O r- co •H U PH Ul D NITROGE £ Turnips Removed M •* •* O O M W \O O ^ M t^I^OO •* ^- invO O> O Tj- rf CO PI y; X; to m r^ Tj-OO 00 00 O Pi M O *fr r^ •* O CO CO to ro NITROGI -t o » 1^ IB M •t 14 P*3 O O O « o o •* o ro mO "~> 00 00 00 Tj- Tj- m CO m TURNIPS V~l Hi 0 -1- * •* "" a, ts § 5= 00-0 T3 C O PH NERALS AN 1 IS S 00 PI O •* « •t CO PI O t^ O Pi P" O 00 "1 COOO OO OO PI Tl- CO M « '0 X •c>! OO O vO COOO O N O OOO CO Tf TJ- [NERALS A^ * -t -1- H |»J -T H) 0 0 Tt TJ-O o t^ O m M O t^ m m rt- ro co m REMOVED X M 00 0 M 00 ct •2,-v sA c^ tn "o O 51 en s | 3 Turnips Removed O 00 O O O M O 00 O •* T)- -t 2 0 i«5 0 \r, M iq 00 O •* O Tf t-- Tt O O O 1- "1 « W 00 " to lO ro w> TURNIPS ft n •* in CO 2S u " It ' £ £ P« 'e l§ O Tt- O •* Tf TJ- O O ^O O n X O •* O O PI Tj- O cr 01 <>< O N N OM^- "-/O Q a > •* 3 1^ CH 37, 00 O " 'too 0 OO COO 9 i^ oo i^ ^ « o ^ o i o Hi C & r>- r^ n c * O ^s T3 CO 10 CO PI M o> -t 0 JO c: PI • AMSTED. 8 i A 0 .|§ gT3 H£ •"tOO PI O O PI PI "•> >t PI «~- O O coco Tj-OO IH TOO PI PI PI 0 i/; "^ 0 n c ' O Tt O COOO O CO N O PI PI PI 5 « t^ 0 W5 0 in o H M c O rf- ••to « O O >0 O "1 OOO t^oo o o M « M ro REMOVED 0 \r 1 - O U .* ±"T3 W H O « p" M 3 Turnips Removed Tt Tj- PI VO ^O 00 O CO t^ M PI O PI P< O IT) 10 IT5 CO t^ P« Pt M c C O -s- o •* •* M 00 Tf •H O CO O i-1 PI M PI PI 19404 go 90 co •o 1 00 *t"G >O O O> fOOO M VO M O ro N •* O «««!»• TURNIPS a M U^ i-l o PI §5 t/1 ••-> fi &9 JJ M hj H £ M H] 1 |§ 11 00 PI OO 00 00 O >0 O ^O O O O 00 M O (^ ro ro M Oi DO \r , r. O P« Tj- M M OO mo PI 10 co TJ- DC C 'i n 0 f^ ", a vO O Tj- « r^OO •* fO in to t^ •* M M « 1O REMOVED y. a o 6 u D i- Eirt ^fl O •£"0, H 0 O < fc riLIZED • UH Turnips Removed 00 ^HO vO O PI *t O O Tf >O M O M 00 O 't »O •x O i/- If- O oo •* M PI U^ oo mo »O CO co 'ILIZED 2OO2 0 a o 0 CO O O 00 O OO O O N vO w r~.\o M M M •«• TURNIPS -/. M t PI C fe o § . cc =; ^ •* rt O 00 «-i J M 4, ft, o en i B C) a i £ O I i 8g S-g E-'fc 00 "*00 PI PI O 00 ^O IH 10 PI M P| M OV PI PI PI •/: M O^ PI t^oo t~- TURNIPS C <" S u ^ =1 «! *-• •P--SSI 0 (VI 111 **C/3 g 2^ S « >o S '4 m 1 Ss H >< S> > r^ •J I &T3 S Hi pq 3 • _Q, ,* a •_. - '„ n \ G D g ear a efl 5 u U j ear a I'^s T5 o O X tn I i - c C I H 00 PI *O O t rj- 10 mo O oo oo oo oo oo 0 o « 00 P) O O O t-*. r^oO oo oo oo oo 25 1- DQ N 6 n •a 0 S X) 00 « O O CO O O O 00 00 00 O 14 -t 0 - HI 0 n ^3 m 00 o 0 M ^^.S " 4J Xi *J THE ROTHAMSTED EXPERIMENTS 347 •lo O CO O •* •* CO 01 O "tO -J Ol "tO 0 rsi X O O O O Q £ 01 01 "to O U". r^ 10 N Qv Os CO CO "t o 3 "g U-) ^J- HI HI cs t^- CO CO HI 10 HI B o Hfa ns B •^ « 0 -t co 8 os S3 3 HH Cfi "O .2* > O O oo 01 CO OO OO 00 O O HI ^t 01 CO O 0 ^r -0 co "t HI CO HI p Q H c^ a H « n ^ o o o •* o 00 O O 5 0 O 00 O •* S bj "S ^ "t O HI o 00 — t- 3 -t -r CO cOO O > fc OsO co 10 C- (V, co O co IO IO QsOO l~~ o OS ^ S O CO HI CN ^" '*"•• ^" os r» ^> M g Hfa H -t (4 g o (§ o CO 2 TJ g en co h-5 p. QJ O 00 O •* O (1) Tt" 00 "t ^f -t •<*- "t O 0 M 'So o-oo o o oo 0 O 00 •* O M 00 6 01 -t 0( •* 01 CO Hi CO O t^ os 3 u O "t HI C) t^ -t IO HI O H! ^oj M H .ss 01 «N 01 "tOO o O 00 01 •* •<*••* 0 0 1- 0) "t O 00 S J_ Tj- Tj- CO Ol *O OQ OsOO Tt 1- t-» IO 01 O o ^ o M Hl HI HI IN M HI HI t Hfc, n I/-, H 00 IN l-r p^ 00 "c3 _. *• •t tfl a PH O 00 O "t ^f 00 M 00 "t O 01 co 00 01 o ^O B 11 O\ IOOO 01 >0 <3 co O O 0 0 -t O O 00 HI r^ "t OsO z OS 3 as w HI HI HI J 01 H HI p o B £•"<* a H O i OH C 00 O Ol o\ ^ O ^"O M IO QO *t 0 01 00 00 co HI O Ol I -t -r O Ol Ol O 10 "t O CO o 1 o I Hfc 0 01 HI H 0 01 CO M " M Os 01 01 g ^. 00 o PH \o o S «13 ^~ !g " U £1 O 00 O 00 01 Qs Ol 00 HI IO 0 -t O Ol O OS HI CO BO 0 M -t -t 00 "to ^ 3- ^ foco* 2 i 3 H M « H 01 HI ^rt H H •1 O 00 "tOO ^t Os •* 01 01 00 Ol Tf Ol M o 01 "tOO 00 WTf -t aa B "t 01 00 O 10 HI ^" CO 00 HI ^"O Q o § Ht, H H -t 00 03 01 -3 o LT. £ co "t H-4 « ^ 01 "t 01 T)- •o 01 OO O vO HI a OO O O 2 o 1/5 -T: O IOOO co g H> 1O CO HI H LT OO *O CO p IO * CO Ol •* o 1> £ hfe o ' H 0 §0 3 tn T3 •2* > "tO O O "t o O O Ol 0 00 rj- O 0 00 0 1 01 SO 01 t— OO 00 M u- rj *H ioo r~- i Os IO Ol 0) H •^t ?l 10 D -J 2 'JJ 1" bo 1* > j; § .• .• 3 ^ H a u ri u rt 1- 0 5 u S H 00 Ol O O ^t 0 S D A' IN 00 Ol O O S 5 -t H 0 rt 0 00 Ol o O g, „ i u 6 as f-i £ m 00 00 00 00 00 S S oo oo oo oo io -r -a OO Os Os O 00 OO 00 Os 0 o- -a co 348 INVESTIGATION BY CULTURE EXPERIMENTS .»8 H H NO 00 NO •* •«t O O O\ --0 O VO ON t Q Q z S n 11 Tt r^vo I^NO Tf CONO vo t 0 in o 1 00 t"- •* «^ ro •* ^r ^ fi cs rO IN ^r 10 co 10 10 N (N CO M g 1 M M c o • ~S kl 8,| O ON vo ON NO 0 J4 oo toco •* 8 p 0 0- ro O NO co 01 e ^ 01 N £ 0 B a H| r~- i-^- Is- O t ro co tNO t 1/1 -i- ON M O ro CO ro ro rO X Q ON M o (*J O 00 M t \O CO CO 01 C H rO CM Sc £j 00 00 00 O 1s- M rO ^tvO •^f * V d HI Tj- ir> •T <*}<*><*! -t- (*! ir, rC O O O N "Fl n O\\O lo^O fO « **J oo oo cr> xo ON t>- 00 0) 00 ON vo u-j •^- rj- fi PO "*• r*i ro tr, »*5 M o CD o- N P) fN M 1 J »o H M M VO ON HI M HH M ^ 5 h H IK 2; F O O\ H vO 00 Ox \O ON\O O- i: H 0 '/T NO COOO c P i 3 hi oo oo o o Is- •\O O M ro \o >o IOQO M hH ' B C >> ~ IH U b >-< U h w i u U >, 3 o >, & 0 -, s a h 1 1 ;/, ON PC 1s- H vr ^t vo loO *O oo oo oo oo oo 6 C) 'x 0 n x ON CO Is- H vO 1s- t^OQ oo oo oo oo X V CO or, 0 01 -r 6 0) •a ON CO 1s- 00 ON ON M a M X IO & M o CN! 73 co THE ROTHAMSTED EXPERIMENTS 349 I || ON CO O to t l~» CO to HI CM CO co CO t co o w HI (N t NO NO co t ON t ON CO CM CM CM * X M X 0 NO ON O t^ ON to HI H CM 0_ REMOVED 0) M 00 HI D NITROGI £ || CM to O O 00 t ON O CM ON OO to t ON CO CM CM CM. CO CM M so' t NO to to NO CM CM IO O NO NO t-» CO HI HI M D NITROGI 00 CM to Ht §N 01 HI l^. tO CO ONNO CM IO t HI HI HI M TURNIPS 10 M HI [NERALS AN 0 fo NO H to O 00 too O t to CO CM CO CO CM 00 co co ON 10 ONNO tOOO CM IO CM t^ CM t HI O CO CM CO CO [NERALS AN NO X co co to 10 to O O co co O to O M co CM CM CO % REMOVED HI 10 NO to be 3 Turnips Removed OO t to O to CO O CO t ON CM CM CM CO CM CN) CO CN] CO ON COOO co O CM M to CO t^ ONOO CO M HI HI NO CM to cs 0) to ONOO CO CO CM NO NO co M M CM H TURNIPS CM CN! k o !•? ON ON to to CO en r->- 10 t O CO CM CM CM CM 0 (N (N CO 0 10 M NO 0 NO M O O CM HI HI HI O H cs M M HI CO t CM HI CM t H HI HI 0) REMOVED NO 00 co HORUS 3 Turnips Removed O CO IO t ON t^ O t to O 00 O to ON 10 HI CM M HI HI Hi HI CO O t O- t*~ t — to CO OO CO O CM HI M HI HI Z K CO 0 Ht co M H IO CO O~ NO O NO ON CM O- H HI TURNIPS H 0 £ 0 S ^13 r^ NO O co t CM tooo to t CO t^ l^. to CM CM (*; CO HI HI t O O t ON co tOO ON t CM H HI HI 00 to CO 0) IO 8 CN! o- CONO t HI NO ON NO t t^- M HI M to REMOVED ,^ 00 ON M a> E"S HI co to O to HI OO t O NO to M 00 CM IO to t O CM *O f*^ IO O to HI CM M jj 1 M HI ON O M CO ON CM CO l^ H JRNIPS M HI ^£ P c •|° 3"^ ON O O OO ON CO HI CO HI O HI CM t O 00 CO CM CM CM HI HI t CO CM t -r oo M toco Tj" M t^*NO NO CO CM to HI HI HI HI CO 0 HI 10 H O a: NO ONOO ON co to ONNO HI HI HI o- Hl REMOVED CO CM I>- rlLIZED h Turnips Removed 8t^ O O 00 CO CO ON CM CM i-H CO HI CO CM CM CM CM HI HI H CO t t tO M t ONNO ON HI M TURNIPS M HI UNFER- a I"8 to «>• CM o O CM t"- HI t^NO CM O CO ON t CO CM CM HI HI 0 0 co t to HI co t ON t NO ON t co t M HI HI HI H fc •/. CO HI to 01 to l-l ^ tooo NO NO tooo 00 co ON H NO REMOVED NO a CM M J •S" > £ ° OO CO O CM to ON t NO to HI CM M CM CM CM 00 10, 0) 00 CO HI t t t ONOO ON co CM t HI HI HI HI 0^ 01 HI t M H HI O HI CO t to ON t CM HI HI H TURNIPS HI ^ s*. £3 V rt rt § rrt m tr. c rt B H 1 >- et 0 > l_! ed 0) U 1 u a) 0 1 e rt S, H H M H S rn ON co t~- HI to Tf IO IO NO NO 00 00 00 00 00 0 (M 0 CM ON CO l^ M 00 00 OO GO to X CO 0 01 "C 8 -a ON CO Is- CO ON O* CO 00 00 M 8- HI § M 8 CO 350 INVESTIGATION BY CULTURE EXPERIMENTS E-3 I! 0H !? n3 O • «.§ o U Mia c t f° oo to o to to 00 TJ- o O O w 00 00 0 — ON O 00 1000 •* i r: I^-NO r^ ON M o ^•00 co C o 0 "• Q FROGEN over or St H§ NO M M Pt « "' ^ H ON TROGEN s,' c c< 2 E 0 T3 PI PI 3 PONO -a 1^.00 g NO TJ- g 00 ^t- eu " 'S o NO to O O O M IO PI 00 ON NOO^NO 00 -t M 00 •* O ON •* •/ EJ SS ON CO «0 PI ON o 3NNO 00 c< H - -r ^ p (£4 Q B Z «tti R c 3 < 1° •^-00 w OO CO -4" " •* c 5 9 90 0 PO B 0 3 n 5 pq L_ | -^ 2 H "S M ON ^ i Pi ^ co O £ | • a| ON^*^- 0 pp * M E 3 P» M n £ 'S 0 O^ C* fT> 6 4 6 IO 4 • a; 3 S M Tf Cl Cl PI M (N U ^(2 H ag o oo o ^ ^o 00 00 O •* ^, 0 •« Tf | D 10 PI 6 r. -t 2 'T. 3"T3 IO — — PN) — X • •- H CO ON '-t c' M 1 Q M « n T3 O PI •g PI ON 5 J — I- e o NO e r- 10 E .^7 > 00 ro m O 00 00 N£ 3 to PI X NO 0 PI r~ M o M NO PI -f | -t p w^ OH O M M D c U || M o *o '^ *^- \O W M W IO M NO 3 n c- -T • H 0 r^ £/ 3 h = ~ - C £ 3 I! 6 GO 6 6 v> CO — X q •* ft CO 1 pq f~^ fr^ i JB « v o to | i &"S ^ toco "t e 00 c 1 M m 11 ino 6s t^ -t H • • 3 C o M M CO ^OH H en JQ £ h ^- ro 10 10 in O ' vO i r^ c O loO ^ O >J^ ON WJ r, f\ t?T3 > co a •7 ^ 5 10 M 1 1 PI M H ^. I y Hfe M H K; O "c Njro 0 2 S TILIZED 5 Turnips Removed « 10 O O PO ci 10 O Tt M O O >- 00 O 00 Pi 0 0 t^oo r- ON P4 PI riLIZED 90 90 CO O 00 O to co NO *•"- CO M M H TURNIPS -t £ *M § n °" OH P S.* M • . ^ •2*o to M O 00 gi ^j. E CO ON r^ X n I-g 10 10 t^-oo '_- «*• K P --TJ n > c u _ ^ 00 co P tn "O i-< co O co -t 00 o -t E M " « ll 3 ~ 100 ONO fj 00 -t 8 D ^tti H o i a §1^^, fe w 1 i > 2 to CO >. TMENT f > d c c c Q rt rt cd cd r— i QJ 1> QJ (U 0 .-: «C «U u 1 Clover S g C ^ O ?? CQ CJ CQ U _c o X t! c U £ 0? ON -oo tn 00 00 S " " ffl -2 2 r M i H s O ^-00 N ^O CO CO 00 CO 00 H M H H H O ^00 PI 00 00 00 00 c « 1C O -^-00 ON ON ON 00 00 00 0 c- M 0 0 c O PI 1000 00 00 00 00 THE ROTHAMSTED EXPERIMENTS 351 .§* M •* 10 O O-. DC fO O •* fO 0 o co co O | X 6 I2 r^ x^. o O O (M oo •* •* M ft CO q c^ O O c> M CO H CO s^ r)- ts Tj- 1O M O •^- CO <0 NO H A H H c 0 M 3 b s.| M co TJ-O •<*• M CO ^- M -^10 G 0 a >o II « O O CO r^ PI H P) CO •* co co K D h << a BS OO 00 OO oo 10 10 H O 10 < t») ^ O O •* 0 i [NERAL V a 3^ H£ M O 00 "I- " "5 'fr co •* « fO CO 10 O •* O INERA1 rfl -t a 10 O 0 O " •* ^t •* a* 1 K M CO •* 3 M J .£•1 CN •*!» M 00 n "-. O O O H S co <0 HJ •* o a g C1 CO 3 6 H£ 00 J^ O>O ro M ro ro TJ- IN c^ Tt IH CO 10 M CO H Tj- -t TO rf ^ PI •* CO t K D h .SS CKVO VO H H PO T^-vO W 1O cp 04 O O 00 C K 3-0 O fOOO C>00 O fj O fO Tt lo H5 N !*• CO O O COOO TJ-\O n ~ PI r^ O 3 I-1* H l§ ro ro 10 O •* o IO H OO 0 •'l-OO O M M 4 « 10 ^t w IO a; o 01 O OO 00 PI S I 2 le h£ O t>- »o 10 t^ CO ro «»5 t « 10 to •H Tf 0 CO S N JJ -t co cs n P) IH VO co Pi Pi K D h K E SS O\ <*5 ro »O 10 to •*>O C>00 « H »• ro vO IO IO O g P 0) a K H£ O TJ- O O 10 f*5 fO fO fO « 00 CN 'O 11 OVOO 10 MM « P f^ N O Pi H O Pi O Pi PI co a 0 S H « •* CO 3 E| 1O <*} ro M o 00 N O O tf •* O 10 « 10 co co S H 8 3 Ui id u ^ U s h (S IH • u !5 u H .£3 > I o o
  • 10 10 10*0 o oo oo oo oo oo o « n 0 ^ H IO O CO oo oo oo oo f. X 0 -s O « -c M 10 O Ov O^ O^ 00 00 00 co 0 e> H o" 0^ o « T3 CO CO N~ C N I* •^ a .2 l^ "S « S.n tc « "03 .2 •*> 1| o w cT .y s . w « •£ «* J'^w 'S — "S 3 S « **** J "o K w "aS & 352 INVESTIGATION BY CULTURE EXPERIMENTS SE 6 ll ON t-» lOOO to ON M tONO ON re to »o to CM PC t^NO O O *3~ ONOO M ON *~» Gf 0 -t oo O vo NO CM re tOOO 10 re CM NO OO ^t 00 00 re CM 1 re re fa P TJ- f\) re^O ^"* BO M CM ONNO en )J3 O H toOO •<*• NO IO to O CM r-~ re re o to re re re re H 8 o re to re ON O IO IO M NO NO re CM O CM CM H P) a Q z t3 -T 00 H •0 o cs 00 ONM M M OO MOVED ^ I F* hik c> n •* M to to o g \o r/) H p^ re 5 &1 •H ON O OO re oo H re ON O •<*• O ON M 0 01 OO reoO 12 £ u CM e_i E o ON re CM oo r/5 ON tH M X TtNO O Tf H rf. f" 't re CM ^0 M to r— re M r^ £^ re re Tf re CM re CM CM H P) M CM 00 . ^^ ^_ e d rt cd a a tt (Tt c > 2 >x J^ q r- P* S 3 03 u u OJ X, - 1 p 0 V 0 (U H u 1 C/3 H 10 ON re t^ IO IO IONO *O oo oo oo oo oo 0 01 R w M 10 ON re oo oo oo oo ® 1^ 00 30 o na o SM oo n ro M so M ro r-~ H M ro Tfr H M M W O O "O W N f*5 ro N PI 00 ro ro •» c< M T}-\O •O W ("O ^" fO PO ro •* ro ro 2 ror? ^2 il| t <3 ^O so •* > 00 00 00 OO PI H 10 10 00 £ g" „ g-^ ^ H-l g, ro w N W (N O O O P) ro pT?"ro - § *i ** ft efl o -^ >S v u •« o ioto O ro t-~ to "» r>* Q\ vo ^- o to PI so .Ot^ o jg — -u o £ H csj ^ 3 • a, Q O* H ? 00 ro ro M ro P) •* PI O O M Pi ro •* Tt- 3 \O OO O O^O ro ro PI ro PI sO 00 PI l! .i1^ i 21 Pi ^^L ; ^;^o ^ 1J laf § !> ? PI oo t^so 10 M PI ro s ro 10 t^so l^ O *O P) PI to 1000 i: o a 50 2 S ^ ^ 8 ^ « " S rooO PI •* PI 00 ? "1" ^r •* T?°° N1? ro Ti- •a °° "o 6 -g >• Q><^ ~ 2~° S ' M M 8 rt 3 XJ-p'+H S r M o 0 0 ro r^ PI iO| ISO SO 00 IH T(- Nl 1 •go -go 5 4j o^oo H a oo PI ro M TJ- ro rood M Pi ro rol 00 6 t^ M t^. I M P> PI ro PI 1 II °- cs o S -o «> r M " lOsO ro O 10 Os| Os O Os 10 Tf I I 'g «T S t^ 'o ^ roj^ 00 H 00 ro ro ro w IO TJ- PI IO M (si ro rol M SO 0 1-1 M PI PI ro ro ro 1 1 2 ^-rt ^ ° ^ 0 O - *J J3 r « so 10 Os OsCO 00 M rj- 10 to rosO J^ •* •* — • JH KH" 4J g t <££oc'° S. 1O P* rOM tj- rOPI t^. M CN ro ro S c?pTrorT O to o 8 | oni 1 r « so IH 0 10 M 10X) s! 3 •* 10 •* TJ-SO IO ro 0 r- 0 8Ji H « ^ ^ •« « 8, H 8> 10 O • O f* OO O P. ro Os PI 00 M OO 10 1 'i- ro M od si w Pt ro ro 3 00 OssO 6s 10 SI IH M PI PI M M ro M PI «5 M PI PI rt "X f Si •» *-• •al jw§ -a > o ci°o o c* PI sq 00 Pi ro M 00 10 O •* ro M- lood M (si ro ro 6s PI 6 PI i- \ 00 10 §1 &-« s •S -4J *-^ HH > O K r^ O 00 ^ t 00 0 PI H M M Os PI P! 6ssd M M M P) ro ro f^ M 00 rl " PI PI PI PI PI £ l|l ' > 0 ''so O ^ 1010 10 r-. qsio M Os rooo Tf N C1 N rt "X w^ rt < ">*<% HC2 ro M IH Pi ro t PI PI ro ro ro *? 00 so 15 ™ D 4J*o >2«~|>8| 4* rOM ^" PI !-•» M " PI ro ro rood ro P! ro PI PI ro ro ro '"^ • M 13 rt fi t) £'g S^o g «£ °° ^ OO H CO •8^ 1 1 \£ 1 .1 r, 3 •« * « ^ »« ° C C g ^TJ _0 E O ||S|i u c o PI PI PI Pi 1 t- Os M ro 8 PI •HT PI PI Os ro Os J^ 3 Tj- t>* IH Os O || PI O Os O 1^ t g £ 3 ^ *« *^ K "^ ._ ^rt £8 • | •|2 ^&^ 2 ^8s|l ^"* c rH rosO Os r § * ' •O so so so so so t 00 00 00 00 00 00 00 00 00 Os S^ >". go fc 3T3 C JH f3 OJ g.0^ rt-0 >^ |§£?f||| 5 "§ «"§ ^'1 •§ ^ Pn xi rt 13 - fr, 1 £r£~ M Sa 2-o§ THE ROTHAMSTED EXPERIMENTS 365 w £ >" CO W O en 5 3 9 00 1 O NO NO CO o oo N r^ CM CM 00 w CM HI PO ON M CM PO ^t -> 3 rt* PO O to r*~ ^ H t^ M M ON ^ r^ HI r^ PO co N M CM CM PO CM 00 0 00 O N l^ Tj- rj- PO t~~ H CM PO HI PO CM NOTE. Average yields of straw for the eight years, 1844 to 1851, were 2979 Ib. from plot 2, also 1736 Ib. from plot 3, and 2654 Ib. from plot 10. 1 Applied in alternate years. 2 This is three fourths of actual yield on half plots after fallow. o ONOO CO ON CO O HI 1^ O ON ON HI IO HI NO 10 CM O HI CONO 00 N t~~NO NO 00 CO O O CO Tt" HI IO o PO 't' IONO 10 POCO Tf 10 f co to t^ NO CM CM IONO O •* ON rt- NO 1 M tO O CM ON TT O ^J- CM O HI CM PO •* •O ON CM ON CM H H VOOO ON CO H 3 00 ONNO O 00 •O H M CM *J- CM CM CM CO H O CO CM CO tO ^- 10 M CM CO CO I IO HI 1000 O NO 1 0] Tj- 10 10 •^ 00 CM O CO H 3 CO H M ON PO * CM CO TT •* CO 0 H t^. CM •rj- to 10 0 H PO HI 1^. O O co PO •<* PO o IO CM CM co to r^oo t^. ^t to HI i HI CM PO > ONNO O NO OO D 00 CM Tj- Tj- CO O CO ON CM ON ON S M H CM CM H 00 O r^ H CM PO Tf COOO ON CM HI 10 ON PO CM CNI I IO Tt O ON Tf IO CO ON CO HI ON CM 00 00 T}- 00 O HI r^. NI CM PO Tf N 00 CO O ONOO -O OO O t^ t^ CM D CM CO t^ ON H S H H M CM CM ONNO ON Tj- M CO M PO t^> t^- ON O IO CM CO AVERAGES ^fsl n-cQ H ON HI CM "3- PO TT CM 00 CM CM CONO r— HI CM CO Tj- CO IO IONO M r-~ CM oo •'too HI CM CM CO CM o (*• tT v-> Q 00 £00 -HOO 0 't 00 w to PO PO HI 00 •* HI HI t^. ON HI ON HI CM CO ^t O CO ON CM IO O NO ^" NO PO CO t~*. CO *O 't' CM CM PO PO PO M|f| 00 H O NO ONNO t^ w ON ON PO ^NO to HI CM PO "t 10 Tf- ^t CO Tf NO HI PO CM t~-> O IO M IO HI CM CM PO PO CO 0 Tf co CO M CO CO CM •* co M CO AVERAGE YIELDS w C 0 0 0 >* 2"" M 10 NO ONNO f^* co < O CO H 10 ( ^J- tooo i>- H CM PO 't > ONNO 1OOO ON > •* O 00 CM 10 -1 t^ CM 00 CO NO O M CM CM CO CM NO NO CM HI to O CO ^t- <£j? J M CO CO 24I2 1 n- ON CO O oo o PO o 00 'too t^ i CM CM rfNO C M CM PO Tj- 0 HI O 00 O CM > 00 HI NO CM NO M H CM CM CO CM Mlsl OO CM ON to 00 ON CO ONNO ON O O ON W NO HI CM CO f» M CM PO Tf O O 00 co co £•*» PO tO t^. O t^. PO O to CM HI CM PO PO PO S4f2l NO CM PO >0 HI ON NO CS TJ- ON CO 00 H ON O ON CM O H HI CO Tt CNI NO O HI NO O OO O ON TT r~» cooo O ON M M CM CO CNI 24l2t 00 00 0 00 00 CM CO HI Tj- •* CM 00 CO to *t CO rf 1000 00 HI CM CO •* CO TT NO HI CO IO Tj- HI Tf PO M CM co PO CO 0 £ oo -^ oo ON O CO HI ON ON O IO •^•oo O PO O TJ- IO CO to CM r^. HI CO OO ON CM CO CO PO PO |"eld » A 38.1 20. 6 9-9 13.6 4.9 1881 IX-5 T> 8 5-3 3°-4 35-4 1882 1883 13.0 II. O I3-9 2-3 1.8 18.1 (fallow) 33-5 (clover) 29.4 (clover) 36.S 3°-3 32-8 35-3 30.8 37-o 41.9 J3-S 10.8 iS-9 (Note Changes) Minerals Minerals & 86 Ib. N 1884 jggr 13.0 20.3 32-5 43-5 35-° 1886 1887 1888 J5-3 9.0 14.9 9-3 19.0 12 8 (fallow) 34-8 (clover) 25.6 (clover) 42-3 36.5 34-S 30.8 42.4 34-5 37-9 44-6 39-6 1889 35'3 33-9 1890 1891 14.0 13.8 t7.8 23.1 (fallow) 32.0 (beans) 29-5 (beans) 42.3 43-o 48.5 35-5 37-6 40.0 37-3 42.1 1 1 8 ->R T rSn-3 9.4 n 8 33-4 38.1 OT 8 31.8 1894 1895 1806 18.0 10. 0 16 8 r3-5 15-5 !5-5 TA T (fallow) 21.8 (clover) 23-3 (clover) 37-o 34-3 45-5 43-9 49-0 40.O I9-5 47.0 32.6 rXn'? o ~ 44.0 37-8 1898 1899 0.9 2.0 2.0 7.0 20.3 15.8 (fallow) 26.8 (beans) 3°-3 (beans) 40.3 37-3 38.0 42-5 37-° 29.4 39-i 27-S 23-8 37-5 IgOO 2-3 T R 11.9 33-3 34-9 I9O2 1903 3-3 7.6 14.7 22.4 14.0 X i (fallow) 20.3 (clover) 18.9 (clover) 28.9 39-6 4i-5 29.7 45-2 3S-8 3°-S 33-5 26.8 1 8 o ' 22.3 18 f 1906 1907 15. 2 1 9.1 12.9 13-4 14-3 (fallow) 16.3 (clover) 21.4 (clover) 36.8 3s-5 43-6 33-7 47-5 34-7 34-2 43-1 34-7 12.4 7-2 38.0 47-5 3«.I 1 Actual yield on half -plot after fallow (in making average, only | of these yields are used). 374 INVESTIGATION BY CULTURE EXPERIMENTS yield for the succeeding years is greater than the average of all, excluding, of course, the yields eliminated. The data recorded will be especially useful for working out as- signed problems, and it has been brought together from several different publications, and the complete records are made possible only through the kindness of Director Hall of Rothamsted, who has furnished the author with some unpublished data. The last column in Table 62 shows in greater detail about the same fact as is well illustrated in the data from the twin plots, 17 and 18, in Table 60; namely, that commercial nitrogen must be utilized by the crop for which it is applied, or it will be largely lost in drainage water. While plot 1 6 received 172 pounds of nitrogen in 800 pounds of ammonium salts per annum for 13 years (1852 to 1864), and pro- duced 39.5 bushels of wheat per acre as an average for those years, there is apparently but little residual effect except for one year after the application was discontinued, the average yields of the 19 years without fertilizers being 14.6 bushels of wheat and 1400 pounds of straw per acre. The following statement will be of some interest in this connec- tion: TABLE 63. WHEAT YIELDS ON BROADBALK FIELD, ROTHAMSTED Thirteen Years' Average, 1852-1864 TOTAL YIELDS PER ACRE PLOT No. Son. TREATMENT APPLIED EVERY YEAR PER SALTS PER (Lb ) ACRE Wheat Straw (Lb.) (Bu.) (Lb.) 5 6 7 8 16 Minerals (P, K, Mg, Na, S) . . Minerals and ammonium salts . . Minerals and ammonium salts . . Minerals and ammonium salts . . Minerals and ammonium salts . . None 43 86 129 172 7Q2 992 1192 1392 1592 I8.3 28.6 37-i 39-o 39-5 1862 3038 4270 4788 5222 While the second addition of nitrogen produced almost as large an increase as the first, the third addition gave but little increase of grain, and the fourth still less, although the yield of straw was very appreciably increased, even by the fourth increment of nitrogen. For convenience a general summary of some of the more impor- THE ROTHAMSTED EXPERIMENTS 375 tant data relating to wheat yields at Rothamsted is given in Table 64: TABLE 64. WHEAT YIELDS AT ROTHAMSTED Wheat, Bushels per Acre, Averages FIELD . . . . BROAD- BALK Hoos AGDELL AGDELL AGDELL BROAD- BALK BROAD- BALK BROAD- BALK Crop System . . Wheat Every Year: Plot3 Wheat and Fallow Alternating Turnips, Barley, Fallow, Wheat Turnips, Barley, Legume, Wheat Turnips, Barley, Legume, Wheat Wheat Every Year: Plot 2 Wheat Every Year: PlotS Wheat Every Year: Plot 16 Soil Treatment None None None None Phos- phorus Farm Manure Minerals and 129 Ib. N Minerals and i72lb.N T&A A ffifT 17.4 15-9 14.8 15-4 28.0 29.6 3S-o 35-6 38.3 38.1 30.8 1044 1051 1851 . . . Tp.r'o -I'-V,- (fallow) 25-2 ' 23-5 3°-5 28.5 28.0 39-5 2 41.1° 1855,' 59/63, '67 36.3 3J-4 31.2 (Note Change) None 1868-1883 . ^i, '75, '79, '83 11.4 9.2 !3-7 12.4 32.0 29.8 34-i 30.1 13-3 II. I 19.9 20.5 25.8 (Note Changes) Minerals Minerals and 86 Ib. N TQQ. TSrM-> 12-5 12.7 10.4 8.4 iS-9 18.4 14.0 14.2 39-6 42.4 35-3 3r-7 37-8 38.4 39-4 35-3 34-9 38.0 32.7 30.8 1887/91/95, '99 1900-1907 1903-1907 . 28.9 27.2 40.5 I8.3 20. 2 32.9 1844-1875 . 1876-1907 14.8 "•5 21.4 14.7 30.2 24-5 28.1 23.6 29-3 34-8 33-4 35-8 37-5 36.8 1844-1907 . *3-J 17-5 27.2 25-7 32-2 34-6 37-i 1 Average of 15 crops. 2 Average of 13 crops (1852-1864). 3 Average of 3 crops (1855, '59, '63). In Table 64, the average wheat yields from the plots indicated are grouped in two ways. First are given the averages of all years for those plots or twin plots (on Hoos field) which furnish a con- tinuous record; and, second, the averages are given only for those years when wheat was grown on Agdell field. 376 INVESTIGATION BY CULTURE EXPERIMENTS There are a preliminary and a final period of 8 years each, and three i6-year periods intervening. These figures show that the middle i6-year period (1868-1883) gives averages clearly below the normal, and that the average of the four years within that period are still lower, thus proving that even two i6-year periods may not positively establish by crop yields whether land is growing better or poorer. A comparison of the first and second 1 6-year periods indicates that all plots are growing poorer; while a com- parison of the second and third 1 6-year periods indicates that all plots are growing better. In the lower part of Table 64 are recorded the average yields for all wheat crops grown in two 32-year periods, and these figures are the best that can be secured. They show decreases of 6.7 bushels with the wheat and fallow plot (Hoos field), 5.7 bushels with the fallow system, and 4.5 bushels with the legume system, on Agdell field, and 3.3 bushels decrease with unfertilized continuous wheat, which, however, is a greater percentage decrease than on either of the Agdell plots. It should be kept in mind, however, that wheat is the only profitable crop now grown on the unfertilized Agdell plots. The yields increased slightly on the farm manure plot and very considerably where minerals and legumes were used on Agdell field. Finally, in the last line, are recorded the general average of all wheat crops grown on these plots since the experiments were be- gun, with extremes differing by 24 bushels, a difference which in 64 years amounts to 1500 bushels more wheat from the applica- tion of plant food than could be obtained without it, in the same system of cropping. Table 65 gives, in brief, some of the very interesting and valuable weather records of Rothamsted, and for comparison is given the very trustworthy average rainfall records for northern, central, and southern Illinois, and Tennessee, as representing a wide range of latitude in central United States, with the average precipitation (including snow measured as water) varying from 33.48 inches in northern Illinois to 53.69 in Tennessee. (See also map showing average annual precipitation in the various parts of the United States.) The 5o-year record gives practically 28 inches as the average THE ROTHAMSTED EXPERIMENTS 377 00 co d TJ- r-» OO O Tf •* NO CM NO P) IO HI PI IO H 00 S H PI P) H H H H ^ ON O H to OO ONNO PI CO CO CM O O COOO O O 00 ON CM NO n § H CM M CO M H i NO i-i 00 *t to NO to to O Ml — IO M O to O O O IH IH rj- O ONOO CO 00 1 o> CO H M CM CO P< NO i 0 £:p?££ t^OO H CM O CO CO ON Tf CM to ON t^- CO t^- ON £ 9 1 o ti H P) HI H O M •* co O CM 1000 •* Tj- CO ONOO OO O r^ O ON IO o> PI P< H rj- H l-l IO H CM CO CM "so ^ .1 Tt O to O to 10 10 O CM tO^O J>«NO co n-oc i/-. 00 00 M M CO ">|^£? P) H H H PI CM nt o z 2 ^ „- i. * CO ONOO 00 M NO ro t^* M CO f^ ^O CM O NO CM IO CM OO IO % W ^ojsfs1 CM H H H CM CM CM CM CM CO CM CM co CM H I % 00 Tf O 00 00 NO •* O NO NO CM t^ 10 CM 00 O HI CO u o> H H CO CO H H CM CO H CM CM to CM s 1000 M IO CM Tl- COOO O M H O T}* ^O CM 00 00 ON Tt- ON J^-00 "* CO 5- 0 H H H CM P» CM CM H TJ- CM CO pT 8 1 o" CONORS 00 H CM ^ CONO ^}" CM to O CM ON i-i co CM t^ to 0 § t-H M- P) H H CO M H IO ^- PI 00 CM 1 CO ON to CM co to r-~NO M O •* •* to t-» CO O CM NO CM M 5 •3 M $ M CO CM M Tj- H CO CM HI CO H NO CM to O ^t- to CM CM ONNO CM H O ON ; 1OOO tO ON O 00 ON CO NO CM CM M COM CM NO •<*• CO 0 0 M M HI H CO b!J« O NO •* CM CM O O 00 CM O O CO NO ONNO CO •* ON O NO 00 O P 2 <» PM CO M H CO CM H CM CO P) HI PI ~<" A - 1 N o !3 v u f2 o to ONOO r^ CM CM IO Tt" CM CO t/-,NO M O CO HI IO M IO PO P) o & g ? bc^. g, CM H M M CM CM N PI CM CO PM CM CO c v oo "2 fe « M to CM O M IO M tO O r^ w O 00 O r^ ON NO NO O PI ON NO h 2 ">i< rt VO *ONO ^J" ^^^-TJ- CO CM Tj- Tj- CO to I? " i u H § O ONOO OO co O •* t^ M M to r~- M ON P) CM HI O 00 CM ON CO J3 £<<< "£ CM CM CO CO rh rj- CO CM CO CM CM CM CO ffS tO CM CM CO OO OO TJ- to NO ON t^*OO O CM ro to t^- CM IOOO ON 3 'o PM PI ^-PO Tj- Tj- CO P* CM CM CO CM CM 3 •._* CM CM O CO NO P< CO ON CM CM IONO t^- Tt ON Tt- CM O 00 CM M O M o^cw CM PM COPO ^- T^- ro CM CO PI CM CM 00 co £e« J^ Tf P) W ON ON t^» to to to r- i~» O OO NO ON M rr co M CM O CM 00 00 't £ "^ M H P) CM rf co co CM PO ci P< M CO CO in B S S3 P I 11 II A •—»(—>< B j fl'l ii O li *> ~8 O H >~> co I VJ CM J3 4-. .2 378 INVESTIGATION BY CULTURE EXPERIMENTS annual rainfall at Rothamsted, and of this amount 50 per cent passes off in drains, at a depth of 40 inches, and 50 per cent is evaporated, from a soil kept free of vegetation. Roughly, the evaporation from a bare soil may be regarded as a constant, to be subtracted from the rainfall to find the drainage (and run-off, if any). Thus, if we regard 14.25 inches as the constant for evaporation at Rothamsted, the drainage should be 6.24 inches for 1898 and 24.44 inches for 1903, while the actual records show 7.90 and 23.59 inches, respec- tively. Of course the evaporation can be markedly reduced by culti- vating the surface as soon as practicable after each rain, in order to destroy the capillary connection and to maintain a dust mulch, and thus largely preventing the rise of moisture to the surface. On the other hand, evaporation is greatly increased by growing crops, so that during the growing season the drainage would be less on the ordinary field than from the bare soil.1 BARLEY EVERY YEAR ON Hoos FIELD, ROTHAMSTED Table 66 presents in summarized form the data secured from Hoos field, where barley has been grown every year since 1852. These experiments help to answer some important questions con- cerning which neither Agdell nor Broadbalk give any information. The yields, as an average of 55 years, vary from 14.8 bushels on the unfertilized land, and 15.7 bushels where only the sulfates of potassium, magnesium, and sodium were used, to 43.9 bushels with sodium nitrate and acid phosphate, and 47.7 bushels with farm manure (15.7 tons a year). As an average of the 30-year and 25-year periods, the yields have decreased nearly ro bushels per acre on all plots receiving nitrogen, undoubtedly because the 43 pounds of nitrogen was not sufficient for larger crops, after deducting losses by leaching. It will be remembered that the second addition of 43 pounds of nitro- 1 Ingle reports some computations in his "Manual of Agricultural Chemistry," page 76, in which the drainage is reckoned at about 86 inches; but probably the intention was to use 8.6 inches, which would reduce his estimated "enormous loss of phosphoric acid" to a very insignificant amount quite in harmony with other data, such as he gives on page 77. THE ROTHAMSTED EXPERIMENTS 379 gen on Broadbalk produced 9.0 bushels of wheat per acre. A 40 bushel crop of barley would remove in the grain and straw about 56 pounds of nitrogen, in accordance with the average of many analyses; so that, where 4o-bushel crops are produced with only 43 pounds of nitrogen supplied, the soil is now being exhausted of its nitrogen content about as rapidly as on the unfertilized land. According to the analyses reported by Dyer, the nitrogen content of the soil to a depth of 27 inches decreased by 528 pounds per acre on plot A4 and by 841 pounds on plot N4 during the 14 years from 1868 to 1882, while the nitrogen content of plot 04 actually in- creased by 8 1 pounds per acre. This problem is complicated by the fact that there is often con- siderable growth of leguminous weeds (especially of yellow trefoil) on plot 04. The decrease in yield from 24.2 to 15.5 bushels cer- tainly does not harmonize with any actual increase in the nitrogen content of plot 04, but it seems very certain that the nitrogen content of plots A4 and N4 was drawn upon during the 14 years at the rate of 40 to 50 pounds a year, of which probably one half is lost in drainage, as an average. In the lower part of Table 66 are recorded some computed effects for different elements under different conditions. Of course, many other similar computations could be made from the data. In computations of this sort, the first effect should be determined for the most limiting element, the next effect for the second limiting element, etc. While it is of interest to compute the effect of apply- ing the most limiting element where all others have been applied, the result has no practical significance, because every application should pay for itself. It is evident that nitrogen is the most limiting element for barley on Hoos field, because the ammonium salts produce a greater in- crease alone than either acid phosphate or alkali salts. Phosphorus is as clearly the second limiting element. While the alkali salts alone had some power to increase the yields during the earlier years (probably due to their power to liberate phosphorus or encourage nitrification), their stimulating action during those years is indicated by reduced yields during the later 25-year period when plot 03 produced less than Oi. Exactly the same conditions appear where alkali salts have been added to acid 380 INVESTIGATION BY CULTURE EXPERIMENTS H M pq « « 00 « oo r^ M tovO •*'« PO m t^ N t^. Tj- M H t^O PO w PI O "> « i M t^ T)- CA rO 00 P) M O w H Ov t^ rooO P) O W M « « n PO ood « 11 M « « PO to « H Tj- Pi 00 O t^ M S "8 O «-~io « t~~ 10 M H M O O P< -O 00 00 O •*00 1^.00 O IO 't j 1 M OO •* O H M M M Tl- M 00 P< O O H O HI PI O\*C PI •* POO OOO w o in ON t^oo PO M O^>A PO O PO Pi O >O PO Pi « TC 00 00 O OO O s 1 l vO « t^vO M P) M PO IH PO M PO Pi Pi Pi PO M PO .OOJOO^ >< I ^o >no o t^ 0 «^ P< VO M \O 'H PO t-i t^. t-i 2| « P« O HI rj- .5 <* S CO ^"1 O^ O^ O "^ C-*C POOO r^ r^ PI 0s *O IT) PO O r>» 10 »o fo O £>< 21 > " fO co 1^-00 O O M oo 00 10 PO O>O Pi 00 'tO « •* PO «^oo PI O Q;*^ °° M ^ w IH M PO w PO w « Pi PO Pi PO N t ^ "2 ?M - w \o »^ C oo o O «£oO S Ov K*1 M M M >o O m O PO POOO M PO « PO •* O O 00 P) 't PI PO POO Pi *O PO PO PO PO PI O « t- O PI >0 t^ M 3 « 0 ^ *?0°0 C> O OO N POOO ^ m O 00 O> Pi Pi 0 -n.0 o T)- 1000 O H M « M N rt M M W < - W so U"J «: 10 o O iOj^C30 ** 0 Tf O 10 O vO ^00 " Pi PO « •* O PO O PO PI Tf PO Tj- oj M ti 6 ro **• PO 't «-» •* M PO Tt Tj- - 0* \O WOOD o -y? o « •* TT t^ Tt O "1 Pi O M Pi 10 pi o O O >< 2"" 21 O 00 O 00 o "•< o o « PO P< 't PO PO Pi O Pi 't PI PO Tj- t^ »O O PO PO PO PO O t^ PI •* O PO O O O M M PI M O t,' O 0 O ">H«« » O to O w o oo pt \o « O ^J-O « PO O Pi 0 •* O M •sfOO O 3 D £ o t; oo o °> M£oO —00 r-. t» •* O pj r^. ci \6 lOvd ui 6 IO l>» M ^ od Pi c> o M- "too O 1000 POO •^•vO o tA. PO r^« PI O t^ pj M 4o' PO ^VERAGI O .• r^ ooo "^•"'S PO J^- 4 t~ M >O M O vd o 6 M P) •* PO •* Pi O O Pi 00 ci ON PJ Pi •* Pi T)- q 10 •* q> 6> M t^. H PO <*• PO1* M PI 4 o PO «o 10 POOO 10 >o Pi TJ- PO M 6 M M M Pi » N H O u ^O o *"- w^ CO 4J 00 K*1 HI H «~- PO O -t " M M Pi « 00 ui\O ro •* PO O O O M ro 1^2 »A>>O PO •* PO t t>- <>o O ood od o PO •* PO Tf q oc q f •*•*•*•* q q 10 IO •* •* PI q P; PO « PI O W5 i i ;H«g7 >J"I3 - c/3 %s^ O P) O Pi O O O- PO -^- <^ O 01 o P« oooo P" lOvO O> 10 t>» ui f^ r^O t^o Ci vO VO O O > • 1? ( i \ >*•,«-? e; Z OS ~ .J 4 W W o i ; ° ^^d o o o o PO PO PO PO * t •* t PO PO PO PO •* Tt- ri- rr OOOO •* •* Tf •* ^i^ " 1 P) N 10 "^ rivX) M • • • -o • 4) ™ * M .... Son, TREATMENT APPLIED EVERY YEAR Unfertilized .... Acid phosphate . . . Alkali salts (K, MR, Na, S Minerals (P, K, Mg, Na, S rt "c3 « •x «13 • S'-a S ^^.s en &l'^ ^ iiTJ-aTJ - >- rt cc *fci ;en effect : A i over O i horus effect: A 2 over Ai horus effect : N2 over N i n effect : N2 over A2 . . aal effect of manure: 7-1 01 o 6 £* H M CO Tf oooo 1-1 P) PO •* «« M Pi PO t £££2 M N PO •* uooo W Pi ^ aa 3.5 P 01 5f.2 .§ il O O-r) tn — J3X! o u KAP^coff THE ROTHAMSTED EXPERIMENTS en £ ^ 5 § P4 O * I" WJ5 l_ b- rt W ^ H) M M M 2 N > H 3 § 8 H >< 00 1 O t^NO OO t^ 10 ro 1-1 t^ rf-NO ro OO PI CO CO i^. ON PO to t^ I--. O M rt\O N tooo to O 00 NO ON M 10 M H ro r^ M PI ro ON t~ 8. o O oo N NO ro IO W r^. O NO <^- M M N O O O vO ON ^- l>- lO IO M t^OO M N M N NO tO M NO PI -*00 Tl- ONOO ^ O H pi M ro tj- ON ON H t^ H PI PI ro TJ- Pi NO Pi Pi PI PI M NO NO to O PI PI Tj- 00 O ON 10 ON PI to M o ON OO NO ONNO P« t-t ! ONOO to t~- ro " OO r~- r~- ONOO N O t^- M oo ON ONO vO IO ro O ON H N N N rooO ro ro ON ^ r^ PI 00 N M ON M ro PI p| ON tooo O M NO ON O Tf tO Tj-00 PI PI PI PI •* ro rONO •* f- H ro H t^ IO M IO to O to to ON 00 00 rooO NO M 10 8, •* O Tj- ro O tooo ro to t^ r-» ro H rooO 00 O ro N vO 00 ro M ir> ^j- H CS M M PI O OO NO O ON tONO NO 00 NO t^ H. PI H PI if) Pi O t^ l-l OO rooO ON ON ro to M M PI PI ro to NO PI ON M ro ON lOOO PI ON ro ONOO NO NO t^ t^ PI t~» ro H 1 o> HI PI ^t- ON IO r}-CO N 00 O t^OO H TtOO NO OO (NI O ro t^- ^O Tf t^-00 M N M P) H J-^ tO t- ON tONO to NO ON IO ON M PI PI PI 00 PI tooo PI NO tj- ON H H 00 •* PI PI PI PI H ON Tj- 10 ON Pi ro ro -^-NO ON O t^-OO NO •* ON t^ t^ PI IO M f> O o> *•- M ro ON PI PI t^ t^ NO Pi f- to H H Tj- 1^ IO t^ TJ-00 W NO ro lOOO 00 HI N M O r)-OO ro M ro M ro rj- ONOO to 00 00 ON. Tf- H t^ -^-00 PI PI PI PI ••too PI 00 NO O M Tf r^ POND M- t^ M- Th P) t» ON t^ PI 00 00 ON AVERAGES j2 N NO "> rt oo o o W 0> 00 -^ O >< « « OO •* •* Tj- ro M rj-oo r~ ON f- ON N 00 O vO t^ H M Tj- ro M \o to M CNl M M OO •* t^ ^t- ro ro ON ro r^ t^-oo t-« M P) H PI OO ro ON ro w o t-» to M ro M ro PI Pi PI PI 00 Tf 0 rf to ro H PO Tj-NO NO NO O ro •* ON HI t^ NO t^ ONNO t-» o a "i o °o m moo « oo >< « « ON H ro O ro M 00 00 M ro H ro M H H M t~» CO N M ON ro ro O 00 CO M O M PI « ro H ro PI NO PI ro Tf PI PI H Tf PO PI ro PI ro ro ro O to ON to O ro r— ONOO O PI PI PI PO O ro O ro 00 N Pi ro OO NO PI O H to ro HI O NO t-~ ON ON ro NO HI EN NO w d "1 O O v> 0; 00 •*-• O* ^ M » t>- M ro lr> to roOO O ON M ON M M M OO 00 to to 10 O ON ON NO lOOO t^ M N M O H PI ^- 1^ 8 to ON to ON M 0 PI PI PI ro NO 00 00 to oo to H PI rfNO to t^ PI PI PI PI ro ro M 00 PI PI Pi ro HI O H •* NO O to to TJ- to I>-OO ON Tj- PI HI AVERAGE YIELDS 2 « NO in rt 0 o 0 J 0; Ov-*-» Ov >1 M " 00 00 \O 10 00 O 00 rh O O t-- N M M NO 00 r^ O ON NO NO NO ro rONO r^ 1-1 N M CN\ O 00 O ON OO ON t^ H NO O 00 ON H ro M PI 00 •* M to OO **• •*• l-l M Tj- tO t^. PI PI PI PI ro w H NO ro •*• H PO 00 Pi 00 O to O t~» M ro PI f» ON ^ l-~ NO H 0 rt So 0 M u 00 •*-» O* >H « " ON ^vO ^O ro t^ O ON r^oo t^oo M **• r^ ro ro O 'ONO CNI ON to ro M H H M 00 O r^ PI to M ro ro NO NO 00 NO M PI M PI rooO H ON 00 i-i NO PI O Pi O PI PI PI PI PI O ON ON rt- tj- ro M ro ro PI PI NO M ON t^- to O >O rl-NO ON t^ (^ O rt 00 o O> >-. 5 oo i>oo >H " N t-~ M M vO O >O •* t^- O\ t-- ON M NO to H O O ro M IO CNI N M o S "~- o oo M £oO —00 >H " " ro w 10 M t^ Tj- ON to 1^ ON t^ ON rONO r~- N P) ON rj- ro to P< I^.NO M PI M CN! to Tj- O ro PI tONO ro r^- to ON t^- M PI M PI rooO O t- NO ON roOO ro rt- ro to PI PI PI PI PI M i^ r^ O ro PI ro O ro ONOO ON IO r^ PI IO ON {^ «>.00 PI PI H O c3 >O O «- « 4J 00 -£ 00 >H « " M N 00 W •3- (M ON M M ro w •<*• M M M )H ON O NO NO ^J-OO O ro ON O N M M ro PI ro H O tO M to r-. r-» r^ PI Pl TJ- ro PI ro PI ro PI PI Tj- Tj- PI M ro H r^. ONOO w PI PI PI ro ON ON Tf Tt ro ro ro ro l^ HI ON O t^ O ro HI ON O OO M O M PI S2 « « 22<£2S >. i-i M ON t~» ro O vO to O IOO 1OOO M M M M OO to ro l^ M PI to ro PI M Tj- PI p< ro Pi ro 00 fO O >" OO f- ON '^ NO lOOO 00 PI ro PI ro ro O t^NO ON to ro O PI Tj- PI IO ro ro ro ro ON ON r^- t^ ON ON PI PI t^- t^ tooo 00 HI O oo rj- r- t~- ONOO Tj- Tj- Son, TREATMENT APPLIED EVERY YEAR . • (_( • H H 1 PI p, ^ cd ° •^^^'BJ0 ^PnPMCOPi CO CO nT nT . .ZZ be s£ •as "MK lf.§~ .2 0 o3 « l-a*fi -S^lc c 3 3 £ ^«S ill •III »a«s H *O T3 T3 rt C C C tn s3 n3 d g ^/J tn tn .5 "rt "rt "d C • o 00 ON PI PI to w> CO 00 s a 3 3 G C a nj S§ H . O O &* H N ro "t ooco H PI ro rf «« M PI ro Tt ££££ M PI ro -"t uouu H PI 1 1 t^. t^ 382 INVESTIGATION BY CULTURE EXPERIMENTS phosphate (compare O2 and 04 for the 3o-year and 25-year periods). Where nitrogen has been applied without phosphorus, the stimu- lating effects of the alkali salts is still apparent, probably because they continue to liberate some phosphorus from the soil. Where both nitrogen and phosphorus are provided, the effect of the al- kali salts is most marked, and. here it is increasing, very possibly because all of the potassium needed by the larger crops is not liber- ated from the soil on account of lack of decaying organic matter. Here it will be seen, however, that the sodium in sodium nitrate without potassium (plot N2) produces even better results than the alkali salts, including potassium (plots A4 and N4), but this com- parison is complicated by the fact that ammonia nitrogen and nitrate nitrogen may have different effects, and the chlorin and sulfate radicle may also produce some effect. It is of special interest to compare the marked residual effect of farm manure on plot 7-1, Hoos field, with the absence of such an effect from the heavy applications of commercial fertilizers (in- cluding 172 pounds of nitrogen) on plot 16 of Broadbalk field. (See Table 62.) However, it should be kept in mind that plot 7-1 re- ceived 314 tons of manure during the 20 years (1852 to 1871), which is equivalent to almost 6 tons per acre a year for the entire 55 years- At 40 cents a bushel for barley, the manure applied to plot 7-2 has been worth about 85 cents a ton, while that applied to plot 7-1 has already paid $1.36 a ton for itself, not deducting interest on investment or counting the remaining residual effect, plot Oi being used as the basis for comparison. A comparison of plots N2 and 7-2 shows the marked superiority of the farm manure in a dry season (1893), while the commercial fertilizers give nearly as good results in normal or wet seasons, and probably would surpass the farm manure if the nitrogen were in- creased sufficiently. If the 43 pounds of nitrogen cost $6.45 and the 29 pounds of phosphorus $3.48, and if barley is worth 40 cents a bushel, the ammonia nitrogen has left a deficit of $1.97 a year for the 55 years, while phosphorus, in addition to nitrogen, has overcome $1.92 of the deficit, leaving a net loss of 5 cents per acre per annum. THE ROTHAMSTED EXPERIMENTS 383 The nitrate nitrogen practically paid for itself as an average of the first 30 years, but left a deficit of about $i a year for the subsequent 25-year period, of which, moreover, the last 15 years show an annual loss of $1.49. Phosphorus added to nitrate has paid for itself and 60 per cent net profit as an average of the 55 years, and the effect of phosphorus is apparently increasing where applied in this connection, which practically amounts to using it in addition to both nitrogen and potassium, assuming that the sodium has power to liberate potas- sium from the soil. If the nitrogen were secured from the air by clover, and if the potassium were liberated from the soil also by clover, plowed under directly or in manure, it is easy to see that applied phosphorus would be still more profitable, especially if the 29 pounds were applied in raw natural phosphate at a cost of 87 cents instead of in acid phosphate costing $3.48. It should be remembered always that computations based upon increases compared with the yields from unfertilized land may indicate profits that would not be wholly realized if the total yield of the unfertilized land is not sufficient to pay for its own cost. In other words, if it costs more than the value of 14.8 bushels of barley to secure that yield, then the financial deficit from the un- fertilized land must also be overcome before any profit can be had from the use of fertilizers. Furthermore, in planning systems of permanent agriculture, we must also consider whether the apparent increasing gains are due solely to improvement resulting from soil treatment, or in part to the general depletion of the unfertilized land. Probably nothing is more difficult for the average landowner to realize than that what appears to be profit is in part at least taken from his own capital. This is very clearly illustrated in the Hoos barley experi- ments. Thus, with nitrogen on plot Ai, during the 15 years (1892 to 1906), there appears to be an average increase in yield of nearly 8 bushels per acre above the unfertilized yield; but, by referring to the average for the first 10 years (1852 to 1861), it will be seen that the unfertilized yield has decreased by more than 12 bushels. On this basis, as an average of the last 25 years, the apparent in- crease from nitrogen is wholly represented in the decrease in pro- ductive power, and consequently in the decrease in value, of the unfertilized land. 384 INVESTIGATION BY CULTURE EXPERIMENTS POTATOES EVERY YEAR ON Hoos FIELD, ROTHAMSTED On another part of Hoos field potatoes were grown every year for 26 years (1876-1901). There were several changes in the va- rieties grown, so that but little importance, at most, should be attached to the yields in successive periods as indicating decreas- ing or increasing fertility, except in those cases where the change is so regular and so marked as to leave no room for doubt. It is especially to be kept in mind that the variety " White Beauty of Hebron " was grown only during the last five years (1897-1901). During the previous 21 years the varieties grown were " Rock " for 4 years, "Champion " for n years, " Button's Abundance " for 5 .years, and " Bruce " for one year, and, in this order, from 1876 to 1896. Thus, the two five-year periods from 1882 to 1891 should be comparable, but, of course, seasonal variation renders even that possible comparison of doubtful value. The special object of the experiment was to ascertain the effect upon the yield of potatoes of different fertilizing materials, as indi- cated in Table 68, which shows the general plan, the treatment applied, and the yields obtained each year. One of the points most clearly indicated by the data in Table 68 is that " White Beauty of Hebron," grown from 1897 to 1901, was a very poor yielding variety. It may be said that 1879 was an exceedingly wet year at Rotham- sted, the rainfall being 2.79, 3.48, 5.55, 4.24, and 6.56 inches for the respective months April to August. In any consideration of these potato experiments, it should be kept in mind that potatoes are a market-garden crop, and constitute one form of intensive agriculture. An annual investment of $25 to $40 an acre for fertilizing materials is not beyond consideration for a crop that may yield 300 bushels, that may be worth $150 an acre. In the last lines of Table 68 are given the average yields for the first 6-year period and for the four successive 5-year periods, and, finally, the average for the 26-year period, followed by the several averages for the value of the increase and the cost of treatment. Since New York leads in the production of potatoes, the price THE ROTHAMSTED EXPERIMENTS 385 used in these computations is 50 cents a bushel (57.6 cents being the lo-year average farm price for New York State, and also for Ohio), and the cost of manure is figured at $2 a ton; but these figures should always be modified to meet average local conditions. They only help to summarize the results so as to bring to mind their economic importance. Thus, at the prices named, the treatment applied to plot 4 has cost $35.32 a year, and the increase produced has been worth $70 a year, or sufficient to pay the cost and leave practically 100 per cent net profit. The ammonium salts on plot 5 have paid but half their cost, and the sodium nitrate alone has but slightly more than paid for itself. By far the largest returns for money invested has been from acid phosphate on plot 9, which has paid for itself and added more than 600 per cent net profit as an average of the 26 years. Indeed, the acid phosphate alone exactly doubled the average yield of 26 years. The alkali minerals, including 300 pounds of potassium sulfate, 100 pounds of magnesium sulfate (Epsom salt), and 100 pounds of sodium sulfate (Glauber salt), have not paid their cost when used in addition to acid phosphate, the average annual increase of plot 10 over plot 9 being only 7 bushels, and the annual cost $7.90. The largest average yield and the largest net profit per acre is from plot 8, which produces as much on one acre as were grown on four acres of untreated land. It should be noticed, however, that, during the last 10 years of the experiment, the farm manure plots, 3 and 4, have forged ahead of the complete chemical fertilizers on plots 7 and 8. Director Hall makes the following statements in his book on " Rothamsted Experiments " (1905) : "In the Hoos field, experiments upon potatoes were begun in 1876, and con- tinued for twenty-six years ; they were then discontinued, because the crop on the plots receiving no organic manures had fallen to a very low ebb in conse- quence of the deterioration of the texture of the soil. But on the plots receiving farmyard manure, and even on those receiving only a complete artificial manure (plots 7 and 8), the crop was maintained in favorable seasons. No falling off was observed which could be attributed to the land having become 'sick' through the continuous growth of the same crop, or through the accumulation of disease in the soil." 386 INVESTIGATION BY CULTURE EXPERIMENTS -a d OH O H 0 « -"3.n a § z 1 H H CO O CO W CO ^J" VO ^t ^J" M OHM H W MINERALS 00 vo ON H vo vO 00 ^" OO OO M H H M vo VOOO d ONVO 00 H O H H • ACID PHOSPHATE (29 Ib. P) ft c J e d d "3" ^t" T(- O OHM H M ACID PHOSPHATE ONVO OO ^ t~» t^-00 ^t* 00 ^ H M H t^ co ^00 t>- t>* vO t^» ON ON oo SODIUM NITRATE AND MINERALS 1 g CO -t OO ^ ^ W d CO d d ^ ONOO l~- SODIUM NITRATE AND MINERALS vO O OO vo co d CO M M H vo VO co w TJ- ON T}- co co O M H M d d 3 $ D N TJ- H Tf O CO O ON co O vo O 3§ii ON Tf ONOO ON M co O vO co d ON ONOO O vO O d co ON - H r^-vo Tf H 00 *-- O d vo 0. vovO ONOO • HI > t>» t^oo oo oo oo oo oo oo oo n < o CO d co TJ- vovO oo oo oo oo oo oo oo oo oo oo 1^00 ON O M 00 OO OO O. ON 00 oo oo oo oo H d c a THE ROTHAMSTED EXPERIMENTS 387 "JO 800 0 W N IO CM NO OO O t-~ CO 10 H O O CO co CONO to to TT CM 00 10 CO ON O •* 0 -^ . to o\ H |i H H M M " ^3" NO co co o oo to to ^ CM OO NO 00 a OO CM to O O 00 w O CO 0 -. . N * ^ C/2JZ "* " CM HI CM H M M H CM CM l-i CM H CM Ht f^ ^ J5 S t* 05 800 r- i^i« O NO to CO CO NO ^ t~- CM H H oo rJ-NO O O CONO O ON ON „ ^ co N 5J- 12 ^ ^ < s CNl M CM H CM i-H M CM CM H CM H M f>» N 'o a + nj w~ S 5 '4-1 P <-° OO ONOO t^NO •3" CM NO COOO O NO ^f CM **» ON oo 9 JN U co £? oj ~T t^>NO H tOOC •3" ^ co CM co t^ t^OO CO t^ CM ^ N ^ 0 " K H H H H 2 w;z; — ** 1/5 o 6a«2 ONOO •'tOO t~~ O ON O ON I-H H IO *^- O Tfr ^aa 3 en 10 ^ ^ |J O CO CM CM hi IO ON, oo to to co NO H \O N cd *"3S «» H P S|H Tt-00 to Tf O OO CONO t~^» NO NO CM 00 O •* H 0 o 9 N •o 1 ^i^ CM CM CO CM CM *^" ^J" NO £"*» OO H H NO CN! rj- r^ to O H H CM M ON M •* O to •-H t*-^ CO U i-| w wo Sgh < iz i- f^ < vi H r^ o i-i M ON CO CM O NO H CM CO CM CM t^ Tj" NO "^t* CM rf COOO ONOO M H 0 CNl ON CO CM ON tONO ^- O I-H H CM M CO H o •* ^_ to co CM co CM' M NO co 0, in 1 O. rr~ bo ^ -2 O to y • ||| CM M CONO NO •^- 'if CM CM CM NO 0 M ON to O •* i-t O M H to CM to to t^. CM n) _ s" ON H w 10 ON 10 t^-NC CM NO to NO t^» If. NO r^ O w _O 14 0 CO co ^ co •*} CM CM H M CM oo NO CO Tt" CM 10 t^. ^ ON 00 M l-i , , O a I-H 1 pCf H 3 w M NO M NO I-H O J « OO oo H OO ON ON O 00 00 00 O H M M M ON H 1 g • ; 5 | | | | | NO ai Q "* g « NO CM t-- CM t^ m c§ S w & 00 OO OO ON ON oo oo oo oo H H H H M £ § \ 53 ^^ • y c i i 5 W H) ^ > > S* s> et .S •- ; j b HI d rt rt 03 eS (H cj SD'OU H ^ S fi I5 CM CO Tt" IO NO ON ON ON ON 0s t^OO ON O M ON ON ON O O {J 388 INVESTIGATION BY CULTURE EXPERIMENTS It should be noted that the average yields on plots 3, 4, and 8 increased during the fifteen years previous to the last five, when the " Beauty of Hebron " variety was introduced; and, as an average, the farm-manure plots yielded higher during the five years ending 1896, than during the six years beginning 1876, notwithstanding the addition of acid phosphate during the earlier period. During the first six years the use of $150 worth of plant food on plot 8 produced $615 worth of potatoes, above the 85-bushel yield on the untreated land, which is also the lo-year average yield of potatoes for New York State. Even when used in addition to manure, during the first six years, acid phosphate, as well as phos- phate and nitrate, paid 100 per cent net profit on the investment; but no test was made with manure and nitrate without phosphate. These Rothamsted data furnish no information concerning the effect of potassium, except that it failed to pay its cost on plot 10. It might be said that all but 9 bushels of the i96-bushel increase on plot 7 should be credited to the minerals (compare plot 5) , but how much of this increase would have been produced by acid phosphate and ammonium salts is not revealed; on the other hand, nitrogen must be credited with the increase from plot 7 above plot 10; all of which means that phosphorus is the first limiting element and nitrogen the second, for the growth of potatoes on this normal soil. To maintain satisfactory soil texture and to provide for the liberation of potassium, magnesium, etc., from the immense supply in the soil, liberal applications of manure should be made, and for the improvement of the subsoil the growing of clover in rotation will produce benefits that manure cannot produce. On the other hand, in such intensive agriculture, there is large profit in a moderate use of commercial nitrogen, especially in such form as sodium nitrate, which also furnishes sodium as a soil stimulant. Whether one should use raw phosphate or acid phosphate, in connection with the manure, clover, and sodium nitrate, is not established, but the Rhode Island and Wisconsin data indicate that potatoes are able to utilize the raw phosphate to some extent, and (in Rhode Island) even without adequate provision for decaying organic matter. It would seem advisable, however, to use the acid phosphate until the raw rock has been more thoroughly tested for potatoes, especially considering that the expense for phosphorus, THE ROTHAMSTED EXPERIMENTS 389 even in acid phosphate, is one of the smallest items in the produc- tion of this expensive and valuable crop. RESIDUAL EFFECT OF FERTILIZERS ON Hoos FIELD Any one who has made himself acquainted with the 26-year potato experiments on Hoos field will naturally be interested in the further history of those plots. The data reported since 1901 are given in Table 69, following a summary of the soil treatment and potato yields. The barley yields for 1902 are in harmony with the common ex- perience that potatoes leave an excellent seed bed for a succeeding crop of barley or wheat; and the residual effect for one year is also very marked where nitrogen has been applied, as was the case with continuous wheat on plot 16 of Broadbalk field. Even the first barley crop on plots 9 and 10 are no better than on plots i and 2, clearly showing that nitrogen was the limiting element for the quick- growing barley crop. Aside from the farm-manure plots, much less residual effect is apparent after 1902; and, in all cases where the treatment is comparable, the barley yields of these plots in 1903 were less than on corresponding plots in the same field (Hoos) where barley had been grown every year for more than half a century. If we keep in mind that nine of the eighteen plots of continuous barley produced more than 36 bushels per acre in 1902, also that four of the ten plots where potatoes had been grown for 26 years produced less than 36 bushels of barley in 1902, and that the largest average yield of potatoes from the farm-manure plots (3 and 4), either for one year or for five years, was secured after pota- toes had been grown on the same land every year for more than fifteen years, then the following statement by Whitney seems clearly inapplicable: "One of the most interesting instances going to show that toxic substances are formed and that what is poisonous to one crop is not necessarily poisonous or injurious to another is a series of experiments of Lawes and Gilbert — the growing of potates for about fifteen years on the same field. At the end of this period they got the soil into a condition in which it would not grow potatoes at all. The soil was exhausted, and under the older ideas it was necessarily deficient in some plant food. It seems strange that, under our old ideas 390 INVESTIGATION BY CULTURE EXPERIMENTS o PH • . M _ S Hill fc HW Potato Potato 00 00 NO •4 H cs O N \O t>-NO (N NO 0) CO •* rf Q Tj- T}- t^ 00 >0 vO O )lications H N H ON ON ^fr VO W « , - cfl n -2 § U 6, — < Tt ON 10 rO 10 H O o £ N ^t H t» TfvO O O ON > O O 10 ro O ON P> c O M NO VO t^ Tf If) 00 \O 00 O N ro M O M CO 1-1 N t-* t^* O ^* M M T3 P) \O M VO N N OO en CO P4 ro O fO tO N TJ- OMD \O «3 cn t/3 «3 t« t/3 ll| 3 3 42 X2X1 g ^^ ££ n -> •*•' ei d rt PQPQO O eg | H 1O M *?« q o B H O t^. O CO H 00 t CO M CO O M 8, S3SSTM{) O t co t IO t cooo M O M 10 H H 00 N t^OO t^ t t 0. M CO ^ ON oo o oo "MM SN&ONO^ 00 OO 00 CO ONOO O M CO M O t S o ssuinSsq bo o A rt r t^ CO ON 00 M t oo oo t H M O O co coi 1 t co too 10 coo 0) ON H.l 1 tOM U>0 ^£ S9SSBJO 00 1000 O M 1OI 1 O O t— t^OO IO ON COO ON O t^ 10 t OO t^.O 00 00 O M O O OO OO 1 O !>• 00 O IO t^. OO ON ON ON t M O O ON ON oo r^- TWO CUTTINGS1 00 s § OO t^ t O ON 10 CO H H 00 O 10 H t M t^ ON O O M t t H CO O t 10 COOO o too oo ON 1000 COO 00 H ONOO 01 COO •- O OO M 01 OO t N O 3m " .9§ o •£>< O O r^ t t^oo tO ON 11 10O i-c 00 M 10O 00 O O M M 01 10°M t M 00 01 MHO O M CO 00 0 t- 0 55 I too t ON IO t O t^ co M H H COO H H IO M O OO M 10 10 10 CO M M CO t M too OO t OO co r^ ON t CO IO t^> t ON M M tOO t IO M OO CO M CO t CO •M r co u >_ o ONO M t o-oo CO 01 OO CO 01 H ON t O ON O H M IO CO ONOO 00 00 H M ONOO 10 oo r^ t^ M COO H O £>. IO M M OO H O O cooo t H ON M OO 0- O«M O O co S^ OO too cooo t CO too 00 O M M O t M CO M to o t co 10 H t M OO 00 H CO 01 t t 0 oT2 M CO t tO M H ON t 100 oo CO CO M 1 . -NO »O 6 i-. >^> o o t-000 oo 10 o IO COO M OO O M N °O t^ 10 coO O IO H t COOO M t ON H CO t "S CO C? t t co H M IO M Two CUTTINGS * bC > o^ o CO N O O too N H H O O ~*O t H 00 H CO O 01 ro CO O M CO IO t^ t^ IO ON O O O M O O 10 COOO t O '10 t to oo O ON CO M IOO IO t^. IO tO M ON O IO t 10 CO 1 . -NQ ID oi o oo MO M MOO too oo o 00 O 01 t 01 O N;>' NO O f^ ">(^00 ^00 00 t O M 00 t H IO M t CO M O O t OO H O CO t t « CO M 00 O O 00 M t CO H ON CO t M 0) 00 O O CO co O H t t O O- 10 tOO M OO O 100 r^ MOO O t M M CO IO ON CO t-» CO IO CO 6 ^ r>S) *" 00 M O HOW to 10 tO -t 01 t t O O ON H O t** t 01 oo co H O-O CO co co O 00 O t 0 0 H 00 O ON ON O O IOO •>*> O 00 0 O 00 10 OO M ON M O IO o oo o tO co O O 00 t 10 co (•qi) 3Hoy aa 4) oca MOO c a MM "Q ON Ov O M M M 0 O J^ 0) M M (i ON t ON ON t H O O H H H IN M M C ON t o to 01 f^ IO O CO M H H (•q-i) *c a go o §CO 00 a 1) O 1) c a c o o o c c c O ca O O ON ON 00 00 0) M goo §0000 a c ^^ o 1 1 KOHJ saxva K31SAS IVNI..J t to O O 10 00 00 00 H H H O O 00 IO IO ON 00 00 00 ONO 01 O loO OO 00 OO O M O M 1OO 1OO oo oo oo oo O O 00 oo oo oo o o o i^ IO IO oo oo oo Son, TREATMENT APPLIED PER ACRE EVERY YEAR (Excepting Changes as Noted) & (Manure, amm. salts) ; then amm. sa (Manure); then unfertilized . . Unfertilized . . 3 . 'II ' •fl • fill •— • fl r* p, (J G G (Amm. salts); then minerals . Minerals (P, K, Mg, Na, S) . . (Minerals) ; then minerals, except K • o '.H HI § C C CJ2" 111J c a a'c I'i'i'i sses BBSS Unfertilized Amm. salts, minerals, and straw3 Nitrate and minerals .... (Nitrate); then minerals . . . Nitrate and minerals .... Sodium nitrate So H M CO It"* O t^.00 ONO V ? H H M M CO t IOO t^ >*> l) r^. C 3 H > i-11 oo - in O . 00 fJH e . » % « * - 00 -C 394 INVESTIGATION BY CULTURE EXPERIMENTS mind, also, that soluble acid phosphate is almost immediately converted into an insoluble form when brought in contact with ordinary soil, and that alkali salts have more or less power to make phosphates soluble. The yields harvested for the first and second lo-year periods are comparable for most plots, and this is also true for the following 2o-year and lo-year periods; although the yields of first crops only (1856 to 1875) cannot be compared with the yields of two cuttings (1876 to 1905). The double comparisons plainly indicate that the yield of hay is decreasing on all plots except those to which minerals are applied without nitrogen (plots 5, 6, 7, and 15) orwith organic matter (plot 13). The largest percentage decrease during the last thirty years has occurred on the unfertilized land (plots 2, 3, and 12) and on plot i, where ammonium salts and heavy applications of farm manure were used during the eight years, 1856 to 1863, and ammonium salts alone thereafter. Marked decreases have also followed the use of acid phosphate and ammo- nium salts, either separately or together; while the addition of alkali salts with both nitrogen and phosphorus has lessened the decrease, but not entirely prevented it. Plots 6 and 7 appear to have reached an equilibrium, having produced about the same yield during the last lo-year period as during the previous 2o-year period, and plot 15 appears to be in the same class during the last lo-year period. A most striking fact is the controlling influence of the alkali salts; but there is no plot receiving alkali salts alone, and the question again arises whether the effect of the alkali salts is more largely direct or indirect. Here, as on the Broadbalk field, the mag- nesium and sodium salts have produced a marked effect, as will be seen from plots 8 and 10 in comparison with plots 4-1 and 4-2. Thus, as an average of the thirty years, 1876 to 1905, the addition of 250 pounds of sodium sulfate and 100 pounds of magnesium sul- fate increased the yield of plot 10 over that of plot 4-2 by 1243 pounds of hay per acre per annum; but increasing the application of alkali salts from 350 pounds to 700 pounds, by substituting 500 pounds of potassium sulfate for 150 pounds of the sodium sulfate, produced a further increase of only 1009 pounds of hay on plot 9; while the further addition of 400 pounds of sodium silicate on plot THE ROTHAMSTED EXPERIMENTS 395 1 1-2 produced an increase of 845 pounds of hay over plot u-i, as a 3o-year average. When we remember that the sulfates of mag- nesium and sodium contain large amounts of water of crystalliza- tion, and that potassium sulfate is an anhydrous salt, the value of potassium for its own sake is still more questionable. Attention is called to the fact that the total weight of salts applied to the best-yielding plot (11-2) is greater than the total weight of field-cured hay produced on the unfertilized land, as an average of the last zo-year period. It seems very probable that the benefit of the alkali salts is due in part at least to their power to increase or maintain the solu- bility of the phosphorus, and thus provide a means by which that element is carried deeper into the soil, where it may be taken up by the plant roots. Even then it is probable that a very consider- able part of the phosphorus applied to The Park plots during the past half-century still remains within an inch or two of the surface. The botanical composition of the herbage (first crops only) is given in the last four columns of Table 70; first for the average of nearly fifty years, and second for the season of 1902. It is espe- cially interesting to note the large percentages of legumes on plots 6, 7, and 15, which receive the minerals alone and consequently must depend upon legumes for a supply of nitrogen. Plot 8 (miner- als, except potassium) shows the next highest percentage of leg- umes in 1902; and, in proportion to the actual application of anhydrous alkali salts, this is relatively higher than the figures indicate. Plot 16, which receives the minerals and the smaller application of nitrate, shows about the same percentage of legumes as the unfertilized plots and the acid-phosphate plot. Where heavy applications of nitrogen are used, the legumes are almost lacking, and entirely so in a few cases. On some plots the herbage is largely weeds. Thus, the 1902 crop of plot 2 (unfertilized since 1864) consisted of 30 per cent of grasses and legumes and 70 per cent of weeds, so that the produce is deteriorating in quality as well as in yield. The following state- ment by Lawes and Gilbert was published in 1900: "The total number of species that have been observed on the plots is 89, com- prised in 63 genera, and 22 orders; whilst, to take some of the more important 396 INVESTIGATION BY CULTURE EXPERIMENTS orders, there have been found — of Gramineae (grasses) 20 species, of 5 genera; of Leguminosae 10 species, of 5 genera; of Composite 13 species, of 12 genera; of Umbelliferae 5 species, of 5 genera; of Polygonaceas 3 species, of i genus; of Ranunculaceae 5 species, of i genus; and of Plantaginaceae 2 species, of i genus. The majority of the 22 orders are, however, represented by only one, two, or three species, and only one genus each. To take an example, it may be stated that the herbage of the unmanured plot comprises about 50 species, and that any kind of manure — that is, anything that increases the growth of any species — induces a struggle, greater or less in degree, causing a greater or less diminution, or a disappearance, of some other species; until on some plots, and in some seasons, not more than 15 species have been observable; indeed, on some, after a number of years, no more than this are ever traceable." Director Hall reports that in 1903 about 97 per cent of the prod- uce from plot n-i (ammonium salts and minerals) consisted of three species: false oat grass (Arrhenatherum avenaceum), meadow foxtail (Alopecurus pratensis}, and meadow soft grass (Holcus lanatus}. On plot 14, which receives nitrate and minerals, the herbage is quite similar except that about 45 per cent of meadow soft grass is replaced by 23 per cent of soft brome grass (Bromus mollis], 9 per cent of blue grass (Paa pratensis}, 3 per cent of meadow pea (Lathyrus pratensis), and 10 per cent of wild beaked parsley (Anthriscus sylvestris), a weed practically never found on any other plot. The herbage of plot 7 (minerals) in 1903 included 4.27 per cent of white clover, 6.41 per cent of red clover, .43 per cent of bird-foot trefoil (Lotus corniculatus) , and 22.04 per cent of meadow pea; while plot 8 (minerals except potassium) showed 1.25 per cent of white clover, 1.38 per cent of red clover, 12.24 per cent of bird-foot trefoil, and 3. 70 per cent of meadow pea. Yarrow (Achillea milk- folium) is a common weed (i to 10 per cent) on plots 6, 7, 8, and 15. The produce of plot 6 (changed from ammonium salts to min- erals in 1869) contained sorrel (Rumex acetosa) to the extent of 12. 1 1 per cent in 1862 and 24.27 per cent in 1867, which dropped to 7.51 percent in 1872 and to 5.24 percent in 1903. Plot 5 showed 14.84 per cent of sorrel in 1903. Lance-leaf plantain was found to the extent of 1.98 per cent on plot 3 (unfertilized), 2.49 per cent on plot 4-1 (acid phosphate), 5.85 per cent on plot 8 (minerals except potassium), and 10.70 per cent on plot 17 (sodium nitrate), in 1903. THE ROTHAMSTED EXPERIMENTS 397 The number of species found in 1903 varied from 10 on plot n-i to 47 on the unfertilized plot 3. On plot 3 there were 16 species varying in amount from .59 per cent to 5.98 per cent, while 3 species were present in large quantity; namely, 20.15 Per cent °f quaking grass (Briza media), 17.45 per cent of sheep's fescue grass (Festuca ovina), and 13.81 per cent of the burnet weed (Poterium sanquisorbia) . For a more complete discussion of the produce from The Park, see pages 150 to 189 of Director A. D. Hall's book, " The Rothamsted Experiments." In considering the financial aspect of these experiments, probably we cannot do better than to take 2600 pounds of hay, the average of plots 3 and 12 for the fifty years, as a general basis of comparison, and then figure the increase in the yield of mixed hay at $3 per 1000 pounds, or $6 per ton, which allows more than $3 per ton for the extra expense of harvesting, stack- ing, baling, and marketing, and for loss, based upon the lo-year average price for central United States. On this basis the top-dressing with $3.48 worth of acid phosphate produced practically no effect, the average increase of 18 pounds of hay per acre being worth about 5 cents. The use of $12.90 worth of ammonium salts on plot 5 produced $1.19 worth of hay; but with both ammonium salts and acid phosphate (plot 4-2) the increase was worth $3.98 (cost $16.38). The addition of alkali salts on plot 7 has increased the yield over plot 4-2 by 2197 pounds of field-cured hay, worth $6.59, but the average cost of the potas- sium itself is more than $10. The total increase on plot n-i over the unfertilized land is 4900 pounds, or $14.70, while the total cost amounts to more than $33. As an average the minerals on plot 7 paid less than half their cost, but as an average of 40 years the wheat straw was worth about $2.60 a ton as a fertilizer for the increase it produced on plot 13 above plot 9; or as a substitute for nitrogen, at 15 cents a pound, the straw was worth $4.85 a ton. (See plot n-i.) An investment of $6.45 in sodium nitrate, applied alone to plot 17, returned $4.14; but, if the hay were figured at $10 a ton net, it would have been worth $6.90, thus showing an average profit of 45 cents per acre per annum, if we disregard the gradual decrease in yield of the unfertilized plots, which, however, cannot be ig- 398 INVESTIGATION BY CULTURE EXPERIMENTS nored in planning systems of permanent agriculture. With hay at $15 to $25 a ton, which are common prices near the large Eastern markets, very satisfactory profits may be made by top-dressing timothy meadows with 200 pounds or more of sodium nitrate, or with perhaps 300 pounds each of sodium nitrate, acid phosphate, and kainit. As a rule, smaller applications will give the greater profit for the money invested in fertilizers, but larger amounts may yield still greater profit per acre, especially when the price of hay is $20 or more. On the other hand, at the average prices that can be counted on for the Central states, the data from the Rothamsted investigations afford no evidence of profit from the use of commercial nitrogen or potassium salts or acid phosphate or any combination of these ma- terials, for top-dressing permanent meadows. ROOT CROPS ON BARN FIELD, ROTHAMSTED While some important experiments with turnips were made by Sir John Lawes, even before 1840, the principal individual plot records date from 1845; and, with the exception of three years when barley was grown without the annual fertilizing (1853- l855)> root crops have been grown every year on this part of Barn field. These experiments were made more extensive in 1856, as will be seen from Table 71, which gives certain average yields in four periods, from 1845 to 1870, and the detailed records of sugar beets grown on these plots from 1871 to 1875, the last two years without the full yearly application of fertilizers. The last column shows the percentage of sugar in the beets in 1873, which was apparently a normal season and the last in which the fertilizers were applied in full for the sugar beets. From these data, the sugar per acre can be computed, but it should be kept in mind that the yield of beets is given in tons of 2240 pounds and for roots with only the leaves removed. The fertilizers applied were in general the same as those specified in Table 7 1 b. It will be seen that the first year sugar beets were ever grown on thisfield the yield varied from5.o5tons to 28.90 tons,— a fact which THE ROTHAMSTED EXPERIMENTS 399 TABLE 71. ROOT CROPS ON BARN FIELD, ROTHAMSTED Yield per Acre of Roots, in Long Tons (2240 Ib.) IN I h 0 0 SUGAR BEETS (Vilmorin's) SOIL TREATMENT EVERY YEAR >"in > g « 5S 1 (About 43 Ib. N till 1860 ; then 86 < g "3 o o7 Ib »•? PLOT Ib. Nothing applied for the Bar- iz M Z f, .5. £ o a c a e g tj « No. ley, and no Manure, Rape Cake, or B oo" Boo < in Boo" o o 0 o n ,. Nitrogen applied for 1874 or 1875. P— 1 ~ in h V H t""1 L1 H r~ ' About 300 Ib. Potassium Sulfate ~ 1 w \2 ^ 1 w £ W " fc 1 H N f) ^ m •-£l till 1871 ; afterward 500 Ib.) H ^ w™ H) 11 00 00 00 00 00 rt fo £f «^ £ GO m GO 300 GO " Oi Farm manure (14 long tons) . . . _ 6. 20 18.15 15-65 15.10 10.80 I7-25 12. 1 O 2 Manure and phosphate .... — 6-35 14.65 16.00 14.30 13-15 15-55 2-3 03 Unfertilized since 1845 .... 1. 2O 2.30 18.8 •55 7-55 7-85 5.05 5.10 5-45 3-1 04 Minerals (P, K, Mg, Na , S, Cl) . . 8.05 7.85 20.8 2.80 7-55 6.70 5.10 6.50 S-45 3-1 os Acid phosphate 8.80 7-45 21.0 2.60 5.60 6.85 5.25 5-9S 5-55 3-5 O6 Phosphate and pot. sul 8.00 6.80 18.8 2-35 5-05 6.30 4.60 5-55 5-20 3-6 07 Phos., pot., and amm. salts (8 Ib. N) — — — • 2.60 5-90 6.75 5-95 6.70 5-55 3-7 08 Unfertilized since 1853 .... — — • — I-I5 7-50 5.20 4-55 5.00 4-75 3-9 Ni Nitrate and farm manure — — — 7-45 27-65 23-45 20.25 11.70 19.90 0.6 N2 Nitrate, manure, and phosphate . — — — 7-65 25-80 24-30 21.50 7-45 19.90 0.2 N3 Sodium nitrate — — — •95 22.15 21.35 14-25 3.10 9-25 1-3 N4 Nitrate and minerals — — — 5.10 22.75 20.10 16.45 8.80 9.40 1-4 N5 Nitrate and phosphate .... — — — 4-65 20.95 19.30 18.40 7-50 9-95 0.9 N6 Nitrate, phosphate, and pot. sul. . — — • — 4-55 21.25 16.80 15-85 8.05 8.20 1.8 N7 Nitrate, phos., pot., amm. salts . . — — — 4-65 20.95 17.00 16.70 9-25 8.10 i.i N8 Sodium nitrate — - — — 1-65 21.65 15-30 12.45 7-65 7.20 0.3 Ai Amm. salts and farm manure . . . . — 8.40 22.05 22.70 22.10 n-35 2I.OO 0.7 A2 Amm. salts, manure, phosphate . — — — . 8.25 21-75 22.00 I9.2O 9-25 i8.8s I.O A3 Ammonium salts i-35 3-85 20.5 •65 15-30 15.15 9-15 3-35 8.00 2.4 A4 Amm. salts and minerals . . . 9-75 9-45 22.5 4.60 I7-50 15-50 I2.5O 7-50 7.80 2.4 AS Amm. salts and phosphate . . . 9.90 8.70 23.0 3-8o 15.20 14-25 10.95 7-30 7.8o 2.5 A6 Amm. salts, phos., and pot. sul. 9.80 8.7o 20.5 4.25 17.20 M-3S 12.90 8.05 7.05 2-5 A7 Amm. salts, phos., and pot. sul. : — • — 4.60 18.40 15.45 13-00 8-75 7-30 3-o A8 Ammonium salts — — 1. 10 16.10 13-50 8.40 6.50 6.05 2-5 AC i Amm. salts, rape cake, and manure — — . 8.80 26.20 26.40 22.75 13-35 2-35 9-7 AC 2 Amm. salts, cake, manure, phos. . . — — 8.70 25.10 25-45 23-35 12.25 0.45 9.8 AC 3 Amm. salts and rape cake . . . 5-50 7.OO 24-5 3-30 19.90 20.40 15.60 2-55 4-05 0.7 AC 4 Amm. salts, cake, and minerals . . 10.25 13-05 25-0 6.60 22.75 23.40 20.15 1 0.60 2.70 0.6 AC 5 Amm. salts, cake, and phosphate IO.O5 II. 20 26.8 5-8o 19.90 18.55 14-75 7-75 3-85 I.O AC 6 Amm. salts, cake, phos., and pot. . IO-3S 12.40 25.0 6-30 23-55 22.80 2O. IO 9-50 2.4O 1.3 AC 7 Amm. salts, cake, phos., and pot. . — 6-75 21.00 23-45 19.80 11.70 I.8S 1.5 ACS Amm. salts and rape cake . . . — — — 3-05 17-95 19.60 15.10 7-30 2.IO Q-3 Ci Rape cake and farm manure . . — — — 8.00 28.90 22.25 23-50 14.50 9.65 I.O C 2 Rape cake, manure, and phosphate — . — — 7.80 25.20 20.75 2I.9O 13-05 8-50 0.9 C3 Rape cake '. . 6.55 7.70 25-9 3-40 2O.8O 16.15 14.65 3-95 1.85 3-5 C4 Rape cake and minerals .... II. IO 12.35 25.2 5-40 21-35 17.90 16.05 8.10 0.15 2-S cs Rape cake and phosphate . . . 10.90 10.50 27.0 S-oo 18.95 15.90 13-95 5-8s 1. 10 2.8 C6 Rape cake, phos., and pot. sul. . . 10.85 11.70 25.0 5-15 21.00 15-85 14.70 7.65 0.10 2.3 C7 Cake, phos., pot., amm. salts . . — — 5-45 21-35 15-50 15.85 8.20 0.30 2.4 C8 — — — 3-70 20.35 15.00 12.10 3-60 1. 60 2.4 does not suggest that the principal office of farm manure and rape cake is to destroy toxic excreta from the roots of sugar beets. In Table 716 are recorded the yields of mangel roots since 1876, in averages of 5-year periods for 30 years, and for single years subsequently. (Swede turnips were grown in 1908, after the man- gels failed, the yield of turnips varying from 1.34 to 13.01 tons.) 4oo INVESTIGATION BY CULTURE EXPERIMENTS TABLE 713. MANGEL-WURZEL ON BARN FIELD, ROTHAMSTED Yield per Acre of Roots, in Long Tons (2240 Ib.) SOIL TREATMENT EVERY w o! •T YEAR i < AVERAGE YIELDS LATE YIELDS (Except no Nitrogen Salts q-O ^ 'M' ^ Td PLOT in 1885 and 1901', 500 Po- tassium Sulfate applied to P. C 3 /. c H 3 No. Plots 2 for 1895 and since, and Minerals with No Potas- *£ "j^ J O in bushels per acre, from 19.9 (unfertilized) and 24.5 (minerals) to 47 (ammonium salts) and 59 (ammonium salts and minerals) ; and for the other four years from 13.1 (minerals) and 13.8 (unfer- tilized) to 28.9 (ammonium salts) and 38 (ammonium salts and minerals). No oats were grown on this field from 1847 to 1868, and the first crop of oats (1869) varied, in bushels per acre, from 36.6 (unfer- tilized) and 45 (minerals) to 56.1 (ammonium salts) and 75.2 (ammonium salts and minerals). The records for the 9 years, 1860 to 1868, are: fallow, wheat, wheat, fallow, beans, wheat, beans, wheat, wheat; with no fertilizers applied during those years except farm manure for the beans in 1864. Experiments with legume crops, especially with beans and clover, have been in progress on Geescroft or Hoos fields (or both) most of the time since 1847. In summarizing their experimental results after more than fifty years, Lawes and Gilbert recorded the following statements (Rothamsted Memoranda, published in 1901) : "When the same description of leguminous crop is grown too frequently on the same land, it seems to be peculiarly subject to disease, which no conditions of manuring that we have hitherto tried seem to obviate." "The general results of the experiments on ordinary arable land in the field has been that neither organic matter rich in carbon as well as other constitu- ents, nor ammonium salts, nor nitrate of soda, nor mineral constituents, nor a complex mixture, supplied with manure, availed to restore the clover-yielding capabilities of the land; though, where some of these were applied in large quantity, and at considerable depths, the result was better than when they were used in only moderate quantities, and applied only on the surface. THE ROTHAMSTED EXPERIMENTS 407 "On the other hand, it is clear that the soil in the garden, which at the com- mencement contained in its upper layers about four times as much nitrogen as the arable land, and would doubtless be correspondingly rich in other constitu- ents, has supplied the conditions under which clover can be grown year after year on the same land for many years in succession. "The results obtained on the soil in the garden seem to show that what is called 'clover sickness,' cannot be due to the injurious influence of excreted matters upon the immediately succeeding crop. "That clover frequently fails coincidently with injury from parasitic plants or insects cannot be disputed; but it may be doubted whether such injury should be reckoned as the cause, or merely the concomitant, and an aggravation, of the failing condition." "When land is not what is called 'clover-sick,' the crop of clover may fre- quently be increased by top dressings of manure containing potash and super- phosphate of lime ; but the high price of salts of potash, and the uncertainty of the action of manures upon the crop, render the application of artificial manures (as top dressings) for clover a practice of doubtful economy. "When the land is what is called 'clover-sick,' none of the ordinary manures, whether ' artificial ' or natural, can be relied upon to secure a crop. "So far as our present knowledge goes, the only means of securing a good crop of red clover is to allow some years to elapse before repeating the crop upon the same land." In his book on the "Rothamsted Experiments" (page 146), Director Hall gives the complete data and the following summary of the clover grown year after year on a small plot of rich garden soil at Rothamsted: RED CLOVER ON RICH GARDEN SOIL, ROTHAMSTED Pounds per Acre YEARS AIR-DRY HAY DRY MATTER NITROGEN IN CROPS Average of 25 Average of 25 years (1854-1878) . years (1879-1903) . 7664 3924 6387 3270 179 101 During the fifty years there have been only two crop failures (1895 and 1900); but the plot required seeding only five times during the first twenty years (1854, 1860, 1865, 1868, and 1871), whereas since 1874 it has been seeded or reseeded almost every year, and sometimes two or three seedings in one year have been required to secure a stand. Late yields of dry matter are: 2887 pounds in 1901, 1169 pounds in 1902, and 1589 pounds in 1903; and Hall's book contains the following: "In March, 1897, and in July, 1899, all the plants were removed by hand, burnt and their ashes returned, and the soil was carefully picked over by hand for the Sclerotia of the fungus, Sclerotinia trifoliorum, many of which were found. The soil was also dressed with carbon bisulfid as a fungicide, before fresh seed was sown. In 1903, which was a favorable year for the growth of clover, a fair plant was obtained by reseeding, and in the spring of 1904 the best crop for many years was cut from this plot." Director Hall expresses the opinion that the fungus named is not the only cause of " clover sickness." Finally, it should be understood that, while the Rothamsted field experiments have been conducted with extreme care, there are some possible sources of error, and the Rothamsted Station has been very careful to point these out where they are of probable consequence. Warrington, in his Rothamsted lectures (Bulletin No. 8, Office of Experiment Stations, United States Department of Agriculture) , delivered before the Association of American Agri- cultural Colleges and Experiment Stations, in 1891, under the pro- visions of the Lawes Agricultural Trust, makes the following statements : "The earlier experimental fields at Rothamsted were not arranged as skill- fully as the later ones; thus, Broadbalk wheat field has long, narrow plots, and the influence of the manure of neighboring plots is in some cases distinctly felt. The barley experiments in Hoos field are the best laid out ; here the plots are nearly square; they have each an area of one fifth of an acre." (In the author's opinion, tenth-acre plots, 2 by 8 rods or i by 16 rods, or fifth-acre plots, 4 by 8 rods or 2 by 1 6 rods, are more satisfactory than square plots for field experiments, because greater uniformity between plots is thus secured; but in all cases a pro- tecting border of at least one fourth rod should completely surround every plot, the same crops being grown upon the border as upon the plot proper. This requires a half-rod division strip between plots, and wherever needed, an additional uncultivated strip of grass sod should be left between the plots.) On the Grass Park at Rothamsted an imaginary line is the only division between the plots, but the ground is never broken, and the fertilizers are applied as top dressings with exactness (a cloth screen A. D. HALL, DIRECTOR OF ROTHAMSTED EXPERIMENT STATION Author of "The Soil," " Fertilizers and Manures" THE ROTHAMSTED EXPERIMENTS 409 being placed on the line), and Director Hall states that the influ- ence of the fertilizers can scarcely be detected six inches over the line, either in the yield or in the character of the herbage, despite the exceedingly marked differences that have developed between the plots. THE CHEMISTRY OF ROTHAMSTED FIELD EXPERIMENTS While much chemical work has been carried on from the begin- ning by the Rothamsted Experiment Station in connection with the field experimentation, it has been directed more largely to investigations concerning the composition of the crops produced than to soil analyses. From most of the fields few soil analyses have been reported; but in the case of Broadbalk field some very complete and thorough investigations have been made of several plots. The results are briefly summarized in Table 73. The soil samples upon whose analysis the data in Table 73 are chiefly based were collected in 1893, fifty years from the beginning of definite plot experiments on Broadbalk field, although on several plots the final systems of treatment were not fully settled until 1852. For this reason the average yields are given for the forty- two years, 1852 to 1893, but the plant food removed and applied is computed for the fifty years; and, in the main, estimation of plant food removed is based upon the analysis of the actual crops harvested. In computing from percentages found by analysis to pounds per acre, Doctor Dyer has used as the weight of fine dry soil per acre 2,590,000 pounds for the first 9 inches, 2,670,000 pounds for the second, and 2,790,000 pounds for the third 9 inches. The cor- responding weights, including stones, are 3,120,000, 3,040,000, and 3,000,000 in round numbers. (For the common silt loam soils of Illinois, we have found 300,000 pounds per acre-inch to be practi- cally correct. This would correspond to 2,700,000 pounds per acre for a 9-inch stratum, or 2 million pounds for a 6f-inch stratum.) In considering the composition of the soils represented in Table 73, it should be kept in mind that the nitrogen reported is total, while the phosphorus and potassium are the portions soluble in strong acid. In the case of phosphorus, this usually represents 4io INVESTIGATION BY CULTURE EXPERIMENTS nearly the total, but in different soils it may vary from the total to as low as 75 per cent of the total; while only from 15 per cent to 30 per cent of the total potassium is acid soluble, although in some abnormal soils, as certain peaty soils, it may reach 60 per cent or more of the total. Potassium varies greatly in this respect at different depths in the same field. Thus, on the gray silt loam prairie of the lower Illinoisan glaciation the percentage of the total potassium that is soluble in hydrochloric acid (specific gravity 1.115), during ten hours' digestion at the temperature of boiling water, varies from as low as 14 per cent in the surface soil to as high as 38 per cent in the subsoil of the same field. Because of these facts the determinations of potassium reported in Table 73 must not be considered as the basis for any final con- clusions, but the phosphorus data must be approximately correct, and the results for nitrogen are practically exact, except for pos- sible variation (from the field average) of the samples of soil col- lected. The data are all reported for 9-inch strata of soil, corre- sponding to the depths to which the samples were taken. Table 73 contains much information, but it is self-explanatory. Thus, plot 7, which has received both ammonia and the regular minerals (as more fully explained in the previous pages), produced an average yield of 32.8 bushels of wheat and 3668 pounds of straw, and 2450 pounds of nitrogen, 482 pounds of phosphorus, and 2117 pounds of potassium were removed in the crops during the fifty years; while there were applied 4300 pounds of nitrogen, 1336 pounds of phosphorus, and 4181 pounds of potassium. The appli- cations have been nearly double or more than double the amounts removed. If we compare plots 7 and 3, we find in the first 9 inches about 23 per cent more nitrogen, 71 per cent more phosphorus, and 19 per cent more potassium in plot 7 than in the unfertilized plot 3. On the other hand, in the lower strata, plot 7 contains distinctly less phosphorus than plot 3 or 4, but this difference is much less marked if plots 12, 13, and 14 be considered. The variations in the lower strata are too great to draw conclusions from any one plot, and this is more especially true as regards potassium. In the lower part of Table 73 are recorded some average results that should be more significant, at least for nitrogen and phos- THE ROTHAMSTED EXPERIMENTS 411 8 • en 0 § o 01 PM w s~x Jr1 bo < .2 < OJ PQ > § I & * " \-) & « PH H POTASSIUM (Lb.) 1 1 6Sp?Mi so Os t— M SO Tj- Os M M sO ^O O t— fO co to ^ 0 O PI 1-1 to O to to \O to O ^" O 11 O a tO Tj- to PI O 11 o 1 Os -t t/) cfl TJTJ C C 3 3 O O ft ft Os Os so to Os P) e PI t- O •* M O ^-so O 0 0 t- 0- osoo oo t- sO so to t— M PI O tO Os PI Tj- OS t— 00 00 00 Os to OS Tj- * to OO 00 0 M 6 jsaij M Tf CO PI to « O so n O Os Os IO Tf to Os t— so PI t-00 >-. 'f Tt~ IO IO OssO CO to 00 00 to rj- to 0 o oS ui p9i|ddy ^ C " Wl "PI 00 00 "b 11 oo oo t- •* «o * Os 11 tS! PI o X to sdoj3 oS ui pgAOUiay; to co O CO t— IT) O •* M o O Pi M 00 t- 0 11 00 T)- o PI Os ^ O to t-00 OO PI sO O 11 hH Os so' PHOSPHORUS (Lb.) 1 .S| 1 6 P^RX 11 oo oo O M 11 vO O 00 to co Os P* to O OsOO t- M Os Os to to O H" 10 oc 0 •3-3 d c a 3 0 0 ftft SO 00 PI 'a a o to S9U3UJ 6 pUO39g 1000 «- Os M tO PI CO PI OO CO IO IO T}- o Os co PI O PI ^f o to to to PI H P| CM so M Os PI 1 to Csl | S3U3UJ 6 jsaij 11 M 00 PI 11 M OsOO cot- t— O Os PI PI PI 11 11 0 to M o tO t— P) M PI PI to CO PI PI CS! PI O sO co co 0 oS ui p9i[ddv ^ a N O to to Os Tj- CO 11 11 sO 11 OO PI to r— so oo to to to to Ov so SO 00 to -s^ : sdojj OS Ul p3AOUI3^[ Os « t- 10 O PI O PI O Os O PI so OO t— to w> to i- C- 11 to n NITROGEN (Lb.) Found in Soil, 1893 6 PWI oo oo t- 00 so Os Os oo o oo oo O to «— o oo oo 3 GO en m -a-o c a 3 3 ft° r- io « IO CS! •M U o kl V ft CO SO S91J3UJ 6 pUO39g Os O PI O Os O Tf Os PI O O-OO 11 M Os Tf PI 00 SO <3" PI Tj- M P, 0 H CS! PI CM PI M O PI PI 0 S9l[3Uf 6 JSJIJ 11 IO 10 so 00 PI Os PI co Pi PI PI so co 11 to Os w ^ Os O O PI PI to tO PO OsOO Os IO PI P) •o o£ u paijddy D 1> £§§ c c H o o o S o o o O co co co O O O 0 o o o o CO CO CO CO o o 0 CO o 0 c«5 -t -t CO CO SdOJQ OS UI pOAOUI9>J o o to O o o o o O to O to 11 PI M 1 0000 Os O CS! PI sO O co to o o PI M X 9 WHEAT, AVERAGE ~T s> /'•^•T^ Mrno a oo (. nn; AVT-aig [-1 1 M Os M CO M 11 t— so oo ^- coso Os PI M CO « PI 10 M rh PI Pi co to co X 10 /. 00 (. ifl)in«0 SO 00 IO oo oo »o os 10 i- PI OO •^t PI co •* PI OS u M tO « PI Tf o n O PI co to to SOIL TREATMENT APPLIED EVERY YEAR FOR 50 YEARS (OR LESS) ' B 1 C/3 • '.3'ffl (u-3 o a rt sD ft C fill rt T3 13 "cs tf tfl W W3 OT Farm manure (15.7 tons) . . Unfertilized Unfertilized since 1852 . . Minerals (P, K, Mg, Ca, Na, Amm. salts and minerals Ammonium salts . . . . Amm. salts since 1850 Average of plots receiving . . 1 Average of plots not receiving Excess Balance Apparent loss in 50 years . . Apparent loss, average per year •d "ft ft 'a 3 o ciS "o 1 d o ll d CO ^~ ^8-S H PI tO •* 4i2 INVESTIGATION BY CULTURE EXPERIMENTS phorus. The line marked " average of plots receiving " includes the average of plots 7 to 14 for nitrogen, of plots 5, 7, and n to 14 for phosphorus, and of plots 5, 7, and 13 for potassium; while in the next line are given the averages for the plots as indicated (') for the respective elements. By subtraction we find the excess or deficiency (— ). The difference between the sum of the ex- cesses found in the three soil strata and the balance with respect to applications and removal in crops gives us the apparent loss in 50 years of the respective elements, and indicates an annual loss per acre of 55 pounds of nitrogen, 8J pounds of phosphorus, and 59 pounds of potassium, — losses besides those which are ac- counted for in the crops removed. In terms of plant food applied, these losses amount to 63 per cent of the nitrogen, to 31 per cent of the phosphorus, and to 69 per cent of the potassium. There are two principal ways in which plant food may be lost from the surface soil, aside from removal in crops; namely, by leaching and by erosion (including erosion by wind action as well as by water). In addition, some mechanical mixing of surface and subsoil may occur, because of burrowing animals and insects, soil cracking, etc., and losses of nitrogen by dentrification are possible, though not probable to any important extent under normal conditions. In Table 74 is recorded the average composition of waters col- lected from the tile drains of Broadbalk field during the years 1866, 1867, 1868, and 1869. These averages represent the mean of a large number of analyses made by Doctor Augustus Voelcker. The results are given in Table 74 on the basis of 3 million pounds of water, which corresponds to a drainage of 13^ inches per acre, which is less than the average annual drainage (14.73 inches) from the uncropped bare soil of the Rothamsted drain gauge (see Table 65), and more than Dyer's estimate (10 inches) for the ordi- nary cropped soils at Rothamsted, but probably not more than the average for the cropped soils of central United States. The actual amounts found in pounds per million of drainage water will be secured by dividing these data, by three. Some apparent relationships may be noted between the appli- cations and losses of certain elements, and also between certain elements in the drainage water, such as calcium and sulfur, but THE ROTHAMSTED EXPERIMENTS 413 S, I I o ^j- ro l-l H 10 H H H < r . E » CJ U & WW rt O J O ow < ^ O 10 ON (» t^ IO 00 O H S£ -I * A S3 ON M \O co O M t^ TJ- N O 00 O CO NO <» ro §z«s 00 ON M r)- IONO 0 ro m N vO IH •* ON E S 21 < So 10 p, VO O P> t^ W t-~ IO 1^. N IO M O »- c « s§- (X -71 n? •* co w t— NO 0 ^- IN \O OO N C^NO 8,1 § 1 S 0 «£• 1 «J MONO 1) ON P) t^. rj- tf) O M HI& a 22J ^£ 3 M N M 5 M W H H H O H £§I ri 2 i o zo 00 w t^ M PI IO Os ON Tj- P) 1^ ON ON co S^ o^g5 00 N ^i" H IONO N M IO M \O NO M f» CN) H ^5'S . S . 6* CO t~- co 10 O r— Cl f- ON ro N vO NO O ^0 I •*e o •:- & <•£ as W (^H « ^H H ^ 00 H •5J- M 10 10 Pi O M PI •* IO to 10 M 10 10 N t^- r^- Tt Tj- 10 10 M f^ CN1 ££>-o *-%% i< io2: ss^gS »0 rj- •o 10 O M PI co *t 10 t^ Tj- Tt Tf M N ro ro ^ eC^4 E'C a o "3s u, j;^ § £ J ^- 0 PI PI PI M ON ON ON ON t^ ON M co M M IO O rj H O M •<*• cs « O tJ NO " « J2 °o « . -S "ri J 6C s • JJ CO T3 00 be J» H rt u TREATMENT EVERY \ Except as noted here 01 elsewhere) manure (15.7 tons) tilized .... -als(P, K,Mg, Ca,S salts and minerals salts and minerals salts and minerals e1 and minerals . |-1ai x r r - o_ <" 55 M ''Or§_§ ° a Mil ?i«Y§ C tn ^tft1 en en *jU c^ cd rt crj c tn tn en w .5 oo £•£ ^2 d to —"NO S"S 1^ ^^ cn .— •C2S^ ?s>*.2 «'S£-S li§l ^^oT2 IffJl 2 ^ -. 3 D : o •- > IJH fe in < H . O O iJ>- fi? a co -3- >o\o t^oo o O H « ro •* lO^O rt « Oj „ S, 4i4 INVESTIGATION BY CULTURE EXPERIMENTS the points of chief interest are that from the four or five best yielding plots the annual losses per acre are probably not more than 50 pounds of nitrogen, i^ pounds of phosphorus, and 7 pounds of potassium. Dyer assumes an average drainage of 10 inches per annum for Rothamsted, which would reduce these figures by one fourth, but he also suggests that the losses in drainage are probably greater now than they were in 1866-1869. While the drainage certainly accounts for most of the loss of nitrogen, there remains not accounted for an annual loss of about 7 pounds of phosphorus and 50 pounds of potassium per acre. Dyer suggests that these losses are to be accounted for by descent into the subsoil. The data for potassium, representing the " acid- soluble" only, are too uncertain to warrant any conclusion. In the author's opinion it is not improbable that some of the potas- sium, applied as soluble potassium sulfate, may have reacted with silicates and formed compounds that are not dissolved by strong acid. This seems less doubtful when we consider the proper- ties of cement, and the changes that occur even in a short time in the " setting " of that material. The data afford practically no evidence for the descent into the subsoil of either phosphorus or potassium. The phosphorus determinations so nearly represent the total amounts that they serve satisfactorily for general computations, and as an average they show less phosphorus in the subsoil of the plots where phos- phorus has been applied, although plots 5 and 14 are exceptions. At least most of the unused phosphorus remains in the plowed soil. Thus plots 4 and 5 have produced almost the same average yields, and plot 5 contains 1121 pounds more phosphorus in the first 9 inches, but only u pounds more in the second depth, than plot 4. The third depth shows a different relation, but this is reversed in the case of plots lob and u, whose average yields are not markedly different. With 2250 pounds of phosphorus in the surface 9 inches, it would require about one inch of erosion in 35 years to account for an annual loss of 7 pounds of phosphorus. This would also ac- count for 10 pounds additional loss in nitrogen, and it seems the most probable explanation. Land that has sufficient slope to provide any surface drainage will suffer some erosion if such drain- THE ROTHAMSTED EXPERIMENTS 415 age occurs when the land is not covered with vegetation. When- ever roily water leaves a field, some soil goes with it; and the loss of a tenth of an inch in three or four years is not improbable, even for nearly level land, if annually cultivated, especially if torrential rains sometimes occur (see record of Barn field) Whether one assumes 10 inches or 13^ inches of drainage, there is some degree of correlation between the computed calcium carbonate equivalent to the calcium found in the drainage water, as shown in Table 74, and the loss of calcium carbonate from the sur- face soil of Broadbalk field, as recorded in Table 27. While there are marked discrepancies, both methods agree that, as an average, more calcium is removed from the plots receiving ammonium salts. Analyses made of surface soil from the barley plots on Hoos field in 1889 show in 2 million pounds of soil 960 pounds of phos- phorus as an average in the 8 plots receiving no phosphorus, 1560 pounds as an average in the 8 plots receiving acid phosphate with- out rape cake, 1900 pounds as an average in the 2 plots receiving acid phosphate and rape cake, and 1540 pounds in the farm manure plot (7-2). Table 75 shows the nitrogen content of the surface 9 inches of the different plots on the Agdell rotation field. From the data thus far reported, the nitrogen content of the soil on Agdell field appears to be decreasing about 10 pounds a year, except on the legume plots which receive rape cake and ammonium salts, where an increase is shown on the " fed " plot amounting to 212 pounds in 16 years. While the individual variations are great, the results indicate a slightly larger loss of nitrogen in the legume rotation than with fallow, but where nitrogen is applied, the op- posite is shown. The factors of erosion and deposition and of difficulty in securing samples (by the method used) which fairly represent the average of the plot are sufficient to account for any of the changes indi- cated by these analytical data; and it may be stated that the to- pography of Agdell field suggests the possible influence of such factors. On the other hand, the indicated gain of 180 pounds of nitrogen per acre during seven years with the legume rotation on the unfertilized land, with all crops removed, has actually been 4i 6 INVESTIGATION BY CULTURE EXPERIMENTS TABLE 75. AGDELL ROTATION FIELD, ROTHAMSTED Nitrogen in Surface 9 inches; Pounds per Acre SYSTEMS AND SOIL TREATMENT Nov., 1867 (After Wheat) OCT., 1874 (After Clover or Fallow) NOV.-JAN., 1883-1884 (After Wheat) Legume, unfertilized, turnips removed Legume, unfertilized, turnips fed off . Legume, phosphorus, turnips removed Legume, phosphorus, turnips fed off . Average of four plots .... 3127 3"3 3-I85 33 i 2 33°7 2849 2978 3170 3J39 2892 2897 3110 3^5 3076 3010 Loss in 1 6 years 175 Fallow, unfertilized, turnips removed . Fallow, unfertilized, turnips fed off Fallow, phosphorus, turnips removed . Fallow, phosphorus, turnips fed off Average of four plots 3127 2959 2938 2976 3"3 2976 2753 2702 2952 2724 2786 2947 3000 2842 2853 Loss in 1 6 vears 147 Plots receiving Minerals, Rape Cake, and Ammonium Salts Legume, turnips removed Legume, turnips fed off 3038 3X94 3096 3293 3012 3408 Average of two plots . . ... 3116 3IQC T.2IO Gain in 16 years (Q4) Fallow, turnips removed Fallow, turnips fed off 3010 3IQ7 2887 2OAO 2918 2986 Average of two plots ?IO4 2QI4 2Q 00 O> 0 00 00 Tf CO •^ ON PI 0s co co *t co ^^ e ig IO CO t^- tO HI r-pi HI ON PI COO O 00 0 M OO t~-« ^ -tpl 0^ CO CO M PI HI HI PI PI ^ PI HI M PI CO co co PI co t/3 HI O O HI 00 CO PI CO 00 00 M PI HI CO O 00 O O HI M PI HI N 10 HI « £fi 0 O w * IOW 't Tj-T)- oo PI H PI M t^o- r-.-. DO O 00 O O> H. ^f O to co 00 O •*«- 10 to to oo 10 to \r 00 00 ONOO ONO o t Pi tO PI PI CO HI 8£ TfOO O t- ON N PI ON PI O O PI r-~ pi NO >r i ^ ^^ c 0 9.\ 5 6 7 8 NP . . NK . . PK . . 48 48 Dried blood . Dried blood . 42.8 34-7 48.4 45-o 38.9 33-5 40.8 35-4 18.9 12.8 17.7 iS-i 36lO 2690 4340 3560 50-71 39-23 54-59 47.62 33-41 33-84 34-27 34-70 17-30 5-39 20.32 12.92 12.24 17.16 15.00 (?) 5.06 (-11.77) 5-32 12.92 Manure for 10 years prior to 1882 9 10 II 12 NPK . . NPK . . NPK . . NPK . . 48 96 144 6° Dried blood . Dried blood Dried blood Blood and bone 47.6 47-4 48.5 40.6 41.9 42.9 42.1 41.1 20.9 23.0 24.6 20.5 4380 4140 422O 4220 57.00 57.98 59-49 56.70 35-13 35-56 35-99 36-42 21.87 22.42 23-50 20.28 22.20 29.40 36.60 23.16 (-.33) (-6.98) (-13.10) (-2.88) 11 14 15 16 Land-plaster (CaSO4), 640 Ib. . 36.2 3Q-6 29.4 37-8 39-o 37-8 41.0 40.1 40.8 40.4 12.9 2480 240O 4290 4030 3660 4290 4I2O 4300 4000 38.32 36.84 1.48 1. 60 (-.12) None . . PK . . 3S-S 47-8 49-5 41.1 12.6 17.8 22.5 20.3 37-27 53-40 56.87 50.92 37-27 37.20 37-12 37-05 16.20 19-75 13-87 15.00 3.60 22.20 1.20 I6.I5 (-8-33) Yard manure, 12 tons .... 17 18 19 20 21 22 23 NPK .' . 48 | Dried blood . Yard manure, 16 tons .... 46-S 47.8 50. i 48.8 23-7 22.9 24.1 24.8 58.04 57-15 59-55 58.56 36.97 36.90 36.82 36.75 21.07 20.25 22.73 21.81 4.80 29.40 6.00 36.60 16.27 (-9-15) 16.73 (-I4.79) NPK . . 96 IDried blood . Yard manure, 20 tons .... NPK . . 144 [Dried blood . Lime (CaO), 2 tons ; yard ma- nure, 12 tons Si-4 27.4 40.6 27.2 22.5 14.4 433° 2440 58.91 35-15 36.67 36.60 22.24 (-1.45) 12.60 o.oo 9-64 ( — 10.45) 24 25 26 27 28 None . . PK . . NPKNa . NPKNa . NPKNa . 30.4 47-5 49-3 49-5 40.6 3°-9 40.2 40.3 41.1 4I-S 13-4 18.5 21.9 23.7 24.5 2410 4230 4330 4370 437° 36.52 54-33 57-67 59-36 60.01 36.52 36.74 36.96 37-iS 37-40 17.59 20.71 22.18 22. 6l 15.00 22.20 29.40 36.60 2-59 (-I-49) (-7.22) (-13.99) 48 96 144 Sodium nitrate Sodium nitrate Sodium nitrate 2Q 30 31 32 PK . . NPK . . NPK . . NPK . . 48 96 144 Amm. sulfate . Amm. sulfate . Amm. sulfate . 42.7 46.4 46.9 40.1 38.6 39-8 41.1 4°-3 17.4 21.6 23.5 22.5 4040 4020 3630 3280 50.83 55.36 56-09 51-72 37-62 37.84 38.06 38.28 13.21 I7-52 18.03 13-44 15.00 22.2O 29.40 36.60 (-1-79) (-4-68) (-11-37) (-23-16) 33 34 35 3f> Land-plaster (CaSO4), 640 Ib. '. Ground limestone (CaCO3), 4 tons 31-5 34-9 48.5 33-6 31-2 32-4 40.4 3I-I 13.1 15.5 21.7 14.1 2570 2880 4690 2730 37-27 41-43 58.36 39-iS 38.50 38-72 38.94 39-15 (-1-23) 2.71 19.42 1. 60 6.00 23.16 (-2-83) (-3.29) (-3-74) NPK . . None . . 60 Blood and bone 1 Where used, potassium is always applied at the rate of 166 Ib. per acre in potassium chlorid, and phosphorus always at the rate of 42 Ib. per acre in acid . bone black except on plots 12 and 35, where ground bone is used. One half of the application is made for corn and the other half for wheat, except the burnt lime, which is all applied for corn. In Table 78 are recorded the average results in actual yields of corn, oats, wheat, and hay, for the 24 years, 1885 to 1908. In order 424 INVESTIGATION BY CULTURE EXPERIMENTS to eliminate so far as possible the influence of seasonal variation' .in individual crops and to simplify comparison, the aggregate value of the four crops has been computed, so that in all cases the finan- cial statement refers to values for four acres. No value is allowed for the corn stover or straw, and the prices used are 35 cents a bushel for corn, 30 cents for oats, 70 cents a bushel for wheat, and $3 per 1000 pounds for hay. While these prices should be modified to suit local conditions, they are as high as can safely be used as a basis for planning profitable systems in the center of the principal grain- growing section of the United States, especially if we must allow for some shrinkage (particularly in the yield of hay) and for occasional unavoidable losses from damaging storms. In the column headed " Value of the four crops," it will be seen that the figures range from $31.69 (plot i, untreated) to $60.01 (plot 28, receiving phosphorus, potassium, and the heaviest appli- cation of sodium nitrate). The four untreated plots show $31.69, $37.27, $36.52, and $39.15, making a very considerable variation; and the problem presents itself, How shall we determine the increase produced by the different kinds of treatment? Manifestly, we must adopt some method of estimating what would have been the yield of the fertilized plots if they had not been fertilized. The average of the four untreated plots would be the most satisfactory under some conditions, but plainly this is not correct for these conditions, because this would show an injurious effect from the nitrogen alone, whereas positive and very appreciable gains are produced in every crop on plot 2 in comparison with the immediately ad- joining unfertilized plot (No. i). In the absence of specific objec- tions it seems best to assume that the productive power of the land, if unfertilized, would vary in uniform graduation from one check plot to the next, and the figures given in the column headed "Value if unfertilized" are computed on this basis. While this seems fair to plots near No. i, a comparison of duplicate plots shows some marked differences in " Value of increase," especially be- tween plots 9 ($21.87) and 17 ($13.87), and between 7 ($20.32) and 29 ($13.21), although in the main the duplication is sufficiently harmonious to justify full confidence in important average results. Thus, the four plots receiving phosphorus and potassium show "Value of increase" amounting to $20.32, $16.20, $17.59, and PENNSYLVANIA FIELD EXPERIMENTS 425 $13.21 (average $16.83); and with 48 pounds of nitrogen in addi- tion, the increase is $12.87, $I3-87, $20.71, and $17.52 (average $18.49). Here we have an average increase of $1.66 resulting from the application of $7.20 in 48 pounds of nitrogen. Further addition of nitrogen produces some additional increases, but always far below the cost of the nitrogen applied. After subtracting the cost of treatment (counting nitrogen at 15 cents a pound, phosphorus at 12 cents in acid bone black and at 10 cents in ground bone, and potassium at 6 cents a pound), we find the greatest net profit from commercial plant food is in the use of phosphorus alone. While $5.04 worth of phosphorus used alone produced $12.17 increase (plot 3) , when applied in addition to other treatment, the same amount of phosphorus produced $14.28 over nitrogen (plot 5 over plot 2), $17.64 over potassium (plot 7 over plot 4), and $16.48 over nitrogen and potassium (plot 9 over plot 6). Plots 12 and 35 also show marked increases from the use of ground bone. Plot 17 appears to give too low results compared with the general averages or with plot 15, although the increase from plot 17 (NPK) is $8.48 more than that from plot 6 (NK). Thus, under every condition phosphorus has much more than paid its cost, the average effect being a net profit of about 200 per cent for phosphorus if we dis- regard the cost of the other elements. While phosphorus and nitro- gen together more than paid the combined cost and produced dis- tinctly better crops, this system yields less net profit than the phosphorus alone. Similarly, phosphorus and potassium gaVe larger increases, but less profit than phosphorus alone. In no case has either nitrogen or potassium paid their cost. It is noted that $9.96 worth of potassium alone produced only $2.68 increase, but when applied with phosphorus the average increase ($16.83) was $4-66 more than that from phosphorus alone ($12.17). Surely we should try to secure this increase by some means. If kainit at one third the cost would produce the same increase, it could be used with profit, and if farm manure or clover as green manure would produce still greater increase at still less cost, we should plan accordingly. Where manure was applied at the rate of 12 tons per acre in four years (6 tons for corn and 6 for wheat) , the value of the increase is 426 INVESTIGATION BY CULTURE EXPERIMENTS $19.75, as an average of the 24 years. With 16 tons the increase was $21.07, and with 20 tons li was $22-73- Tnus> tne I2 tons were worth $1.65 a ton, 16 tons were worth $1.32 a ton, and 20 tons were worth $1.14 a ton. Thus, we may say that the first 12 tons were worth $1.65 a ton, the next 4 tons were worth 33 cents each, and the last 4 tons were worth 42 cents each, or, as an average, the 8 tons of manure applied after the first 12 tons were worth 37 cents a ton. The " cost of treatment " for the manure applied may be de- termined in at least three different ways: First, we may consider the manure as a by-product of the farm and only allow for the cost of hauling and spreading, for which 30 cents a ton is sufficient, as a rule. This is the figure used in the tables under discussion. Second, we may estimate the cost of shipping manure from some fairly large source of supply, such as the stock yards of Chicago or other cities. This cost would probably amount to $i to $2 per ton, including the hauling from the railway station and spread- ing on the land. Third, we may purchase feed and thus produce manure on the farm and allow for the manure whatever is necessary. Table 78 shows the average value of the manure applied at differ- ent rates, and also the profit from using the manure that is regularly produced on the farm. There is no record of the amounts of manure applied to plot 8 (on the four series) previous to the beginning of these experiments; but its residual effect is very apparent, the average increase amounting to $3.23 per acre per annum for the 24-year period, in comparison with the unfertilized plots. Caustic lime alone decreased the crop yields as an average, but when used with manure it produced an average increase of $2.49, or about 25 per cent of its cost at "$4. 50 per ton. As an average of the two tests, the light application of land-plaster produced practically no effect. The heavy applications of ground limestone produce an average increase of $2.71, or not quite half its cost at $1.50 per ton. In the last four or five years the effect of ground limestone alone is apparently decreasing, — a result to be expected sooner or later where no manure or plant food is returned to the PENNSYLVANIA FIELD EXPERIMENTS 427 land. It should be kept in mind, too, that this soil is not very acid, and, consequently, neither burned lime nor ground limestone would be expected to produce marked effects. Since phosphorus and manure were both used separately with marked effect and profit, it seems probable that phosphorus and manure together would have produced still more satisfactory re- sults; and, if the action of the ground limestone were modified by being used with manure as much as was that of caustic lime, then it, too, would have produced increases above its cost. At least, the facts suggest that manure, phosphorus, and limestone .would make a very profitable combination ; and green manures and other crop residues could of course be used in place of animal manures. - In Tables 79 and 80 are recorded the 24 years' data arranged in two periods of 12 years each. While 24 years is too short a period to furnish very trustworthy data concerning the tendency of a sys- tem of farming toward increase or decrease in crop yields on one piece of land or with one crop, probably the average results from all crops on the four series of plots in these Pennsylvania Experi- ments furnish almost, if not quite, as satisfactory information along this line as any of the Rothamsted fields. Such results are recorded in the columns headed " Value of the four crops." It should be remembered that a poor year for oats may be a very good year for winter wheat, corn, or hay, and that four crops every year for 27 years furnish almost twice as much data as the single system on Agdell field, even though continued for 60 years. As an average of the 20 plots that have received no treatment since 1882 (including the No. 8 plots), the yields have decreased as shown below: AVERAGE YIELDS PER ACRE ON 20 UNFERTILIZED PLOTS CROPS 12-YEAR AVERAGE, 1885 TO 1896 1 2- YEAR AVERAGE, 1897 to 1908 AVERAGE DECREASE Corn, bushels 41 7 27 7 14 o Oats, bushels 36.7 2Z O II 7 Wheat, bushels I -}. •} 128 5 Hay, pounds 3070 2l8o 800 Average value .... $11 05 $8 18 $2 87 428 INVESTIGATION BY CULTURE EXPERIMENTS TABLE 79. PENNSYLVANIA EXPERIMENTS: FOUR-YEAR ROTATION Records per Acre for Three Complete Rotations, 1885 to 1896 TREATMENT FOR EACH FOUR YEARS AVERAGE OF 12 YEARS FROM FOUR ACRES 6 X c £ i 2 3 _4_ 5 6 7 8 g 10 II 13 Important Elements Applied Ni- tro- gen per Acre (Lb.) Form of Nitro- gen Applied ^orn Av. Bu. per Acre Oats Av. Bu. per Acre Wheat Av. Bu. per Acre lay Av. Lb. per Acre Value of the Four Drops Value f Un- fertil- ized Value of In- crease Cost of Treat- ment Profit or — Loss) >Tone . . N (48 lb.) P (42 lb.) K(i661b.) 34-7 4°-5 45-o 40.6 48.0 40.6 51.6 5i-4 34-7 35-4 39-1 37-5 ii. i 12.8 14.6 ii-5 2480 2640 3400 2900 37-77 41.68 47.90 42.21 $37-77 38.15 38.53 38.91 K 7.20 5-04 9.06 $ 48 Dried blood . 3-53 9-37 3-30 (-3-6?) 4-33 (-6.66) NP . . NK . . PK' . . 48 48 Dried blood . Dried blood . 42.4 38.7 44.6 40.8 18.2 13-4 15-7 15-5 374° 3050 4320 4050 53-48 44-35 55-39 53-23 39-29 39-67 40.05 40-43 14.19 4.68 15-34 12.80 12.24 17.16 15.00 (?) i-95 -12.48) • 34 12.80 Manure for 10 years prior to 1882 NPK . . NPK . . NPK . . NPK . . 48 96 144 60 Dried blood . Dried blood . Dried blood . Blood and bone 50.7 50.0 50.8 Si-4 46.1 47-3 45-4 44-7 19-3 20.7 22.1 l8.4 4420 4230 413° 4110 58.35 58-87 59.26 56-61 40.81 41.19 41-57 41-95 17-54 17.68 17.69 14.66 22.20 29.40 36.60 23.16 (-4-66} — 11.72) -18.91) (-8.50) 13 Land-plaster (CaSO4), 640 lb. . 42.8 36.9 13-9 2810 44.21 42-33 1.88 1.60 .28 14 IS 16 None . . PK . . 4-'- 5 51.8 Si-9 43-4 35-3 41.8 42-5 41.8 I3-I l6.4 19-5 18.0 2690 4140 3910 3920 42.71 54-57 56-3° 52.09 42.71 42. 6/ 42-57 42.50 n-93 13-73 9-59 15-00 3-6o 22.20 ( ?07) 10.13 ( — I2.6l) Yard manure, 12 tons .... 17 NPK . . | 48 | Dried blood . iS 10 20 Yard manure, 16 tons .... 47-8 50.8 49-9 51-6 44.1 43-7 43-2 44-3 20.4 2O.6 20.3 21.9 4140 4210 3890 4030 56.66 57-94 56-31 58-77 42.43 42.36 42.29 42.22 14-23 15.58 14.02 16.55 4.80 29.40 6.00 36.60 9-43 (-13.82) 8.02 ( - 20.05) NPK . . 96 I Dried blood . Yard manure, 20 tons .... 2 I NPK . . 144 |Dried blood . 22 33 Lime (CaO), 2 t nure, 12 tons Lime (CaO), 2 to ons; yard ma- 50.2 31-4 44.0 32-4 19.7 14.6 4010 2540 56.59 38- 5 42.1. 42.08 14.44 (-3-53 12.60 9.00 1.84 (-12.53) ns 24 25 2>, aj 2.S None . . PK . . NPKNa . NPKNa . NPKNa . 37-5 50.8 50.8 50-7 50.3 36.6 44.6 43-7 44.8 44-4 13-6 16.4 2O-3 21.4 22.1 2790 4260 4250 4280 4300 42.00 55-42 57.85 59.01 59-30 42.00 42.26 42-5 42.78 43-04 13.16 15-33 16.23 16.26 15.00 22.20 29.40 36.60 1 i 9.t\ 48 96 144 Sodium nitrate Sodium nitrate Sodium nitrate (-6.87) (-13.17) (-20.34) 39 3« 3i |i PK . . NPK . . NPK . . NPK . . 48 96. 144 Amm. sulfate . Amm. sulfate . Amm. sulfate . 44.8 48.1 Si-3 43-3 43-0 44-3 46.2 43-6 15-9 IQ.9 22.7 23-3 424 43° 425 359 52.43 56.96 60.46 55-32 43-30 43-56 43-8 44-0 9-13 13.40 16.64 11.24 15.00 22.20 29.40 36.60 (-5-87) (-8.80) (-12.76) (-25.36) 33 34 3i Land-plaster (CaSO4), 640 lb. . Ground limestone (CaCO<,). 4 tons 38.4 41.0 46.8 42.6 36.3 39- 43-6 37- 13-3 15-3 18.8 13-0 300 328 469 333 42.91 46.63 56.69 45.13 44-3 44.6 44-8 45-1 (-1-43 2.03 11.83 1. 60 6.0 23.1 (-3.03) (-3-97) (-11-33) NPK . . None . . 60 Blood and bon With every crop there has been a decrease in yield, varying from £ bushel of wheat, or 4 per cent of the crop, to 14 bushels of corn, or 33 per cent of that crop. The average yearly value of produce from one acre has decreased from $11.05 to $8.18; and this $2.87 represents the total decrease, not for a 24-year period, but for a PENNSYLVANIA FIELD EXPERIMENTS 429 TABLE 80. PENNSYLVANIA EXPERIMENTS : FOUR-YEAR ROTATION Records per Acre for Three Complete Rotations, 1897 to 1908 TREATMENT FOR EACH FOUR YEARS AVERAGE OF 12 YEARS FROM FOUR ACRES * w » H | Plot No. 'mportant Elements Applied Ni- tro- gen pei Acre (Lb.) Form of Nitro- gen Applied ^orn Av. Bu. per Acre Oats Av. Bu. per Lcre Wheat Av. Bu. per Acre lay Av. Lb. per Acre Value of the Four Crops Value if Un- fertil- ized Value of In- crease Cost of Treat- ment Profit or (- Loss) ^Jone . . N (48 Ib.) P (42 Ib.) K (166 Ib.) 23-6 25-8 35-8 25-5 21. 0 23-5 30.8 24-9 35-5 28.3 37-2 29-9 9.1 10.7 iS-5 10.4 SSo 6go 2950 820 3470 2340 4360 3070 125.58 28.64 41.47 29.14 $25.58 26.06 26.54 27.03 S S 48 Dried blood . 2.58 14-93 2. II 7.20 5-04 9-c6 (-4-62) 9.89 ( T «r^ 5 6 7 8 NP . . NK . . PK . . 48 48 Dried blood . Dried blood . 37-6 28.8 45-1 38.6 19.6 12.3 19.7 14-5 47-94 34-20 53-82 41.84 27-51 27.90 28.48 28.96 20.43 6.21 25.34 12.88 12.24 17.16 15.00 (?) 8.19 (-10.95) 10.34 12.88 Manure for 10 years prior to 1882 9 IO II 12 13 NTPK . . NPK . . NPK . . NPK . . 48 96 144 60 Dried blood . Dried blood . Dried blood . Blood and bone 44-5 44.8 46.S 47-9 37-7 38.S 38.7 37-5 22.6 25-3 27.2 22.7 4340 4060 4320 4330 55-73 57-12 59-89 56.90 29.44 29-93 30.41 30.80 26.29 27.19 29.48 26.01 22.20 29.40 36.60 23.16 4.09 (-2.21) (-7-12) 2.85 Land-plaster (CaSO4), 640 Ib. . 29-6 24.4 12.0 2140 32-50 31-38 1. 12 1. 60 (-.48) M IS ^one . . PK . . 28.5 43-9 47-2 38-7 23-6 33-9 35-5 33-8 12. 1 19-3 25-4 22.6 2110 4440 4150 3300 31.86 52.37 57-40 40.68 31-86 31-77 31.68 31-59 2O.6O 25.72 l8.O9 15.00 3.60 22.2O 5-60 22.12 (-4-11) 16 Yard manure, 12 tons .... I? NPK . . 48 IDried blood . 18 19 Yard manure, 16 tons .... 45-1 44-7 50.2 46.1 37-9 36.S 38.5 36.5 27.0 2S-3 27.9 27.7 4440 4020 4710 398o 59.38 56.37 62.78 58.42 31-50 31-42 31-33 31-24 27.88 24-95 31-45 27.18 4.80 29.40 6.00 36.60 23-08 (-4-45) 25-45 (-9.42) NPK . . 96 |Dried blood . 2O Yard manure, 20 tons .... 21 NPK . . 144 IDried blood . 22 23 Lime (CaO), 2 t nure, 12 tons Lime (CaO), 2 to ans; yard ma- 52.6 23-3 37-3 21.9 25-3 I4.I 4650 2340 61.26 31-62 3i.i5 31.06 30.11 •56 12.60 9.00 17-51 (-8.44) ns 24 25 26 27 28 29 30 31 11 33 34 None . . PK . . NPKNa . NPKNa . NPKNa . 23-3 44.1 47-7 48.2 49.0 25-2 35-7 36.8 37-4 38.2 I3-I 20.5 23.6 2O.2 27.1 2030 4200 4410 4460 4450 30.98 53-10 57-49 59.81 60.93 30.98 3i-i7 31-36 31-55 31-74 21.93 26.13 28.26 29.19 15.00 22.20 29.40 36.60 6.93 3-93 (-1.14) (-7-41) 48 96 144 Sodium nitrate Sodium nitrate Sodium nitrate PK . . NPK . . NPK . . NPK . . 48 96 144 40-5 44.6 42.6 36.8 34-2 35-3 36.0 37-c I8.7 23-3 24-3 21.6 3850 3740 3010 2960 49.08 53-73 51-75 47.98 31-93 32.12 32.31 32.50 I7-I5 21. 6l 19.44 15.48 15.00 22.20 29.40 36.60 2.15 (--59) (-9.96) (-21.12) Amm. sulfate Amm. sulfate Amm. sulfate Land-plaster (CaSO4), 640 Ib. . Ground limestone (CaCO3), 4 tons 24.7 28.8 50.1 24.6 26.1 2S-7 37-1 25.2 12.9 IS.8 24.6 15-3 2050 2470 4600 2130 31.66 36.26 59-96 33-27 32.69 32.88 33-07 33-27 (—1.03) 3.38 26.89 i. 60 6.00 23.16 (-2.63) (-2.62) 3-73 3J 3^ NPK . . None . . 60 Blood and bone 12-year period. Of course, the actual decrease will grow less and less as soil depletion continues, even though the per cent of decrease remain constant, for, as already stated, it is as impossible to com- pletely exhaust a soil as it would be to exhaust a bank account under a contract that only 2 per cent of the remaining deposit 430 INVESTIGATION BY CULTURE EXPERIMENTS could be withdrawn at any one time. (Two per cent per annum is approximately the rate of decrease in crop values from the 20 un- fertilized plots in these experiments.) There are only four combinations of commercial plant food that have maintained the productive power of the soil, and these are all combinations of the three elements; plot n (with dried blood, 144 Ib. N), plots 27 and 28 (with sodium nitrate, 96 and 144 Ib. N), and plots 12 and 35 (with bone and blood, 60 Ib. N). However, as an average of the 24 years, the increase was not sufficient to pay for the cost of treatment on any of these plots. It appears, however, that during the second 1 2-year period plots 12 and 35 have, as an average, paid the cost of the plant food and left a net profit of $3.29 from the four acres. Thus it will be noted that the only plots receiving commercial plant food that has paid its cost even for the second 1 2-year period and that has also fully maintained the crop yields are those treated with ground bone. The difference in favor of ground bone is not sufficient to show that it is distinctly better than acid phosphate, but we may surely conclude that the in- soluble bone is at least as good as the acidulated form. During the first 1 2-year period the net profit from the use of manure decreased as the amount of manure increased above 12 tons, but during the second 1 2-year period the value of the 1 2-ton application is more than twice as much as during the previous 12 years, and the greatest net profit per acre is where the heaviest applications are made (counting 30 cents a ton for manure), but the value per ton is still greatly in favor of the lighter application, the first 12 tons being worth $2.14 a ton and the next 8 tons only 72 cents, compared with $1.14 and 4 cents, respectively, for the first 12 years. While the larger amounts of manure show a distinct cumulative effect ($56.31 to $62.78, or $6.47 a year from the four crops), the lightest application has but little more than maintained the earlier crop yields, the markedly greater apparent profit dur- ing the second 12 years being due to the decreased yields of the unfertilized land. In Table 81 is given a summary of the effect of treatment over the 24-year period, and also a concise statement showing the actual or absolute profit or loss from every system, based upon the aver- age yields secured during the second 1 2-year period in comparison PENNSYLVANIA FIELD EXPERIMENTS 431 TABLE 81. PENNSYLVANIA EXPERIMENTS : FOUR- YEAR ROTATION Summary of Financial Results from Soil Treatment in Four-year Rotation TREATMENT FOR EACH FOUR YEARS FROM FOUR ACRES, ONE EACH or CORN, OATS, WHEAT, AND HAY 1 & i 2 3 4 6 7 8 P 10 II 13 13 14 15 Important Elements Applied Ni- tro- gen per Acre (Lb.) Form of Nitrogen Applied Effect of Treatment (Av. of 24 Years, 1885 to 1908) Value of the Four Crops Av. of 2d 12 Yr. Value if Un- fertil- ized Av. of ISt 12 Yr. Value of Increase of 2d over ist 12 Yr. Cost of Treat- ment For Permanent Systems Profit Loss Profit Loss None . N (48 Ib.) P (42 Ib.) K(i661b.) 825-58 28.64 41.47 29.14 S37-77 38-15 38.53 38.91 $12.19 16.71 2.46 19-73 48 Dried blood . . 7-i3 4.18 ( $12.19) (-9-5i) 2-94 7.20 5-40 9.96 7.28 ( 9-77) NP . . NK . . PK . . 48 48 Dried blood . . Dried blood . . 5-06 5-32 12.92 11.77 47-94 34-20 53.82 41.84 39-29 39-67 40.05 53-23 8-65 (-5-47) 13-77 (-11.39) 12.24 17.16 15.00 3-59 22.63 1-23 u-39 Manure for 10 years prior to 1882 . NPK . NPK . NPK . NPK . 48 96 144 60 Dried blood . . Dried blood . Dried blood . . Blood and bone . • •33 6.98 13-10 2.88 55-73 57-12 59-89 56.90 40.81 41.19 41-57 41-95 14.92 15-93 18.32 14-95 22.20 29.40 36.60 23.16 7.28 13-47 18.28 8.21 Land-plaster (CaSO4\ 640 Ib. . . .12 32-50 42-33 (-9-83) i. 60 n-43 None . PK . . -... i. 20 16.15 31-86 52-37 57-40 40.68 42.71 42.64 42-57 42.50 (-10.85) 9-73 14.83 7.18 15.00 3-6o 22.20 10.85 5-27 15.02 8-33 11.23 1 6 i? 1 8 Yard manure, 12 tons . . . . . NPK . 48 Dried blood . . Yard manure, 16 tons 16.27 16.73 9-15 14.79 50.38 56.37 62.78 58.42 42-43 42.36 42-29 42.22 16.95 14.01 20.49 16.20 4.80 29.40 6.00 36.60 12.15 14.49 15-39 20.40 10 20 21 22 23 NPK . 96 Dried blood . . Yard manure. 20 tons NPK . 144 \ Dried blood . . Lime (CaC 12 tons Lime (CaC >), 2 tons; yard manure, 9.64 10.45 61.26 31.62 42-15 42.08 19.11 (-10.46) 12.60 9.00 6.51 19.46 ), 2 tons 24 25 26 27 28 20 30 31 32 33 34 35 36 None . PK . . NPKNa NPKNa NPKNa 2-59 30.98 53-10 57-49 59-81 60.93 42.00 42.26 42.52 42.78 43-04 (-11.02) 10.84 14.97 17-03 17.89 15.00 22.20 29.40 36.60 1 1. 02 4.16 7-23 12-37 18.71 48 96 144 Sodium nitrate . Sodium nitrate . Sodium nitrate . 1.49 7.22 13-90 PK . . NPK . NPK . NPK . 48 96 144 Ammonium sulfate Ammonium sulfate Ammonium sulfate — 1.79 4.68 n-37 23.16 49.08 53-73 Si-75 47-98 43-30 43-56 43-82 44.08 5.78 10.17 7-93 3-oo 15.00 22.20 29.40 36.60 9.22 12.03 21-47 32-70 Land-plaster (CaSO4), 640 Ib. . . Ground limestone (CaSO3), 4 tons 2.83 3-29 3-74 3'-66 36.26 59-96 33-27 44-34 44.60 44.86 45-13 (-12.68) (-8.34) 15.10 1. 60 6.00 23.16 - 14.28 14-34 8.06 11.86 NPK . None . 60 Blood and bone . with yields of the unfertilized land during the previous 12 years. By this means only are we able to avoid the exaggerated influ- ence which is always credited to the soil treatment when com- parison is made with the decreasing productiveness of unfertilized check plots. 432 INVESTIGATION BY CULTURE EXPERIMENTS For this purpose these Pennsylvania data are probably the most valuable the world affords; and, in the author's opinion, this volume presents no more significant facts than are contained in Table 81. The decrease in productive value of the unfertilized plot is markedly uniform, notwithstanding the variation among those plots, the average decrease in value per acre per annum being $2.87 and the widest variation from that average being 18 cents. Plot 8, which had received manure during the 10 years previous to 1882, shows the same decrease as the other four unfertilized plots, the average for the four others being $i 1.46, while plot 8 shows $11.39. The average yield of the No. 8 plots during the second 1 2-year period is slightly less than the average yield of the four other unfertilized plots during the first period. From Table 81, it will be seen that, for permanent systems of farming, no form or combination of commercial plant food has been used with profit, the annual loss from four acres varying from $2. 46 with phosphorus alone, $3.59 with phosphorus and nitrogen, and $5 (as an average) with phosphorus and potassium, to $20.40 and $32.70 with the complete fertilizer carrying the largest amounts of dried blood and ammonium sulfate, respectively. Manure costing 30 cents a ton shows net profit in all cases, but the profit is greatly reduced by the addition of caustic lime at $4.50 a ton; although the lime produced sufficient increase to pay $2.14 a ton for it for use with manure, and the effect of the lime- manure treatment is distinctly cumulative, especially upon the clover and timothy, the yield of hay from the lime-manure plots being 640 pounds higher during the second 12 years than during the earlier period, and 500 pounds more than from the manure alone during the second period. Would ground limestone at less cost produce a greater benefit, and would the use of phosphorus also with farm manure or green manure produce still greater net profit? The Ohio investigations answer the latter question with a most emphatic affirmative. (See Tables 37, 38, 39, and 396.) Thus it will be noted that, as an average of the same 12 years (1897 to 1908), the value of the produce per acre per annum is 1-35 where 12 tons of manure are used in the Pennsylvania PENNSYLVANIA FIELD EXPERIMENTS 433 four-year rotation, $13.76 where 8 tons of manure are used in the Ohio three-year rotation (corn, wheat, and clover, — see Table 396), and 17.29 where 40 cents worth of raw phosphate, or 80 cents' worth of acid phosphate, was used in connection with 8 tons of manure in the Ohio rotation. From Table 81 it can easily be determined that the absolute value per ton of manure for permanent systems is $1.24 for the smallest amount used, $1.06 for the medium amount, and $1.02 for the heaviest application. For the additional 8 tons (12 to 20) the manure was worth 71 cents a ton. Based upon comparison with the yields from the untreated land during the last 12 years, the 12 tons of manure for the Pennsyl- vania four-year rotation were worth $2. 14 a ton; while, for the same 12 years, the 8 tons of manure, in the Ohio three-year rotation, were worth $1.82 a ton for the yard manure and $2.41 a ton for the stall manure. (See Tables 37, 38, and 39.) Director Thorne has emphasized the fact that the Ohio experi- ments at Wooster were started on fields that had for many years been under exhaustive tenant husbandry, and the unfertilized plots at Wooster during the last 12 years are more nearly compar- able with those at State College during the same 12 years than dur- ing the first 12 years. Thus the average annual produce per acre for the same three crops, corn, wheat, and clover, was $10.35 f°r the first 12-year period and $7.77 for the second 12-year period, in Pennsylvania; while for the last 12 years the average value in Ohio has been $8.06, these values being based upon the normal unfertilized plots, the No. 8 plots at State College, and the No. i and No. 1 1 plots at Wooster (see Table 40) not being included. If the oats are included, the Pennsylvania figures would be $10.47 for the first 12 years and $7.61 for the second period. In Pennsylvania Bulletin 90 (1909), Director Hunt summa- rizes the results of the first 25 years covered by these experiments. The following tabular statement, containing figures based upon Pennsylvania values, may be of special interest to the student of Eastern conditions. The upper part of this table shows the total weights of the seven products harvested, including ear corn, corn stover, oats, oat straw, wheat, wheat straw, and hay; and the lower part shows the total 434 INVESTIGATION BY CULTURE EXPERIMENTS values at 75 cents per TOO pounds of ear corn, 32 cents a bushel for oats, $1.33 per 100 pounds of wheat, $2.50 a ton for corn stover and straw, and $10 a ton for hay. TABLE SIP. PENNSYLVANIA FIELD EXPERIMENTS PLANT FOOD APPLIED UN- FERTIL- IZED NITRO- GEN (48 Lb.) PHOS- PHORUS (42 Lb.) POTAS- SIUM (166 Lb.) NITRO- GEN AND PHOS- PHORUS NITRO- GEN AND POTAS- SIUM PHOS- PHORUS AND POTAS- SIUM BLOOD (48 Lb. N), PHOS- PHORUS, POTAS- SIUM BLOOD (96 Lb. N), PHOS- PHORUS, POTAS- SIUM BLOOD (144 Lb. N), PHOS- PHORUS, POTAS- SIUM YEARS POUNDS OF TOTAL PRODUCTS FROM FOUR ACRES. (Averages) 1882-86 1887-91 1891-96 1897-01 1902-06 14679 14339 12611 9562 9848 14479 14060 11461 8326 8955 14628 15204 14647 12229 12907 14598 14476 12404 8780 9581 16176 16469 16622 14038 14358 15031 15959 12840 10450 11778 16577 17090 17764 15440 16368 16889 17492 18352 15867 16335 17994 18706 I94I5 16981 17780 17933 19210 19786 17221 18908 1882-06 I22IO H457 13922 11967 15534 12814 16647 16986 I8I37 18653 YEARS VALUES OF TOTAL PRODUCTS FROM FOUR ACRES. (Averages) 1882-86 1887-91 1892-96 1897-01 1902-06 $75-35 75-46 64.29 49.16 50.88 $73-61 74-13 58-57 42.36 45-5° $74-76 79.66 75-58 61.80 67.28 $73-9° 74.61 61.34 43-63 47-3° $83.3"; 86.15 85-75 71.91 74.83 $74.82 77-85 62.88 5I-85 54-13 $85.34 87.56 89.08 76.81 83.73 $85.20 87.81 90-75 77.92 83.87 $89.21 93-49 95-59 83.07 90.92 $91-53 94.70 96.56 85-25 95-87 1882-06 $63.03 $58.84 $71-79 $60.16 $80.40 $64.31 $84.51 $85.10 $90.47 $$02.79 1887-96 1897-06 $69.88 50.02 $66.35 43-93 $77.62 65-54 $67.98 45-47 $85.95 73-37 $70.37 52.99 $88.32 80.27 $89.28 80.90 •$94-54 87.00 $95-63 90.56 The last two lines in the table are lo-year averages computed by the author, all other figures being copied from Pennsylvania Bulletin 90. In his discussion of these experiments, Doctor Hunt makes the following comments (Bulletin 90, page 14) : "The most striking fact brought out by this table is that the application of 48 pounds of phosphoric acid and 100 pounds of potash in alternate years to a rotation consisting of corn, oats, wheat, and mixed hay (timothy and clover), namely, to the corn and wheat, has, during twenty-five years, maintained the crop-producing power of the soil. There is no evidence thus far to show but what the supply of nitrogen can be indefinitely maintained on this limestone soil by means of a rotation containing clover, provided the mineral fertilizers are abundantly supplied." PENNSYLVANIA FIELD EXPERIMENTS 435 These statements, if true, are of tremendous significance to American agriculture, for they refer to the oldest experiments of the kind in the United States; furthermore, the phosphorus- potassium plot is repeated four times in every series, so that the average results are from 16 different plots of normal soil every year for twenty-five years, and they must be considered highly trust- worthy. The small amount of limestone contained in this Penn- sylvania soil can very easily be supplied to any other soil by the direct application of ground limestone. It will doubtless be agreed by all that the results of the first few years at the beginning of a rotation and fertilizer experiment are not to be considered as comparable with the subsequent results. There are several reasons for this; but, for the present purpose, it is sufficient to consider that nitrogen may not have been a limit- ing element for all crops at the beginning. The data given in Table 8iP are not satisfactory for making any study of this special point, because the averages for the unfertilized land include the results from plot 8 which is represented to have received annual applications of manure during the ten years previous to 1882, because of which the addition of nitrogen alone appears (from Table 8iP) to have actually decreased the crop yields, which is not the case if we accept the system of comparison adopted for Tables 78 to 81. It must be evident from every point of view that nitrogen was not the limiting element for all crops at the beginning of these experiments. It is evident, however, that phosphorus was the principal limiting element at the beginning. Now, for the sake of simplicity, let us assume that from a given type of very uniform soil (see Table 87) sufficient phosphorus will become available during the season (1903) to meet the needs of a 54-bushel crop of corn (plot 102), while sufficient nitrogen will be liberated for a 62-bushel crop. The application of nitrogen with- out phosphorus could not be expected to appreciably increase the yield (plot 103), while the addition of sufficient phosphorus with- out nitrogen should increase the yield from 54 to 62 bushels, but unless nitrogen was also supplied, the yield could not be expected to go above 62 bushels. However, by applying nitrogen in addition to phosphorus, the yield might be still further increased (as to 69 436 INVESTIGATION BY CULTURE EXPERIMENTS bushels on plot 106) to a point where perhaps the supply of avail- able phosphorus again becomes the limiting factor. In other seasons or in later years, these conditions may become reversed, with nitrogen as the most limiting element, and phosphorus with little or no effect except in addition to nitrogen. (Note the results for 1907 and 1908, in Table 87.) We can conceive of conditions under which the supply of nitro- gen naturally liberated from the soil, when supplemented by that secured from the air by clover grown in the rotation, will meet the needs of the crops grown for several years, during which the nitro- gen does not become the limiting element to any marked degree, and it must be plain that in such case the crop yields give little or no information concerning the maintenance of nitrogen in the soil. Thus, it is only after nitrogen becomes the limiting element, in any given system, that the crop yields become an index as to the possible permanency of the nitrogen supply. In soils that are markedly deficient in phosphorus, that element may still remain the limiting element after the first small appli- cation has been made, provided the increased supply of available phosphorus is not sufficient to raise the crop yields to the point where nitrogen, for example, becomes the limiting factor; and it is easily conceivable that the increase produced by supplying po- tassium in addition to phosphorus, in the Pennsylvania experi- ments, was due, in part at least, to the power of potassium salts to hold the phosphorus in available form. Even where heavy applications of potassium were made, the sodium nitrate was more effective than dried blood, and, if only sodium nitrate had been added with phosphorus, the sodium would very probably have produced nearly as marked results as were produced by potassium. There are too marked variations among duplicate plots on the Pennsylvania field to justify fine distinctions, and even on more uniform land there are many factors involved with different crops and different seasons; but we dare not ignore the fact (Table 8iP) that the average value of the crops from four acres receiving phos- phorus-potassium treatment decreased from $88.32 to $80.27 during ten years, from 1891-1892 to 1901-1902, which are the middle points of the two lo-year periods. Whether we consider the values or the pounds of products, the apparent decrease is PENNSYLVANIA FIELD EXPERIMENTS 437 approximately 10 per cent in 10 years, and if this rate of decrease continues, we may expect the average values to drop during suc- cessive lo-year periods from $80 to $72, to $65 to $59, to $53, and to $48, in the next 70 years. It will be noted that the dividing point between the two zo-year periods in Table 8iP is exactly the same as between the two 12- year periods referred to in Tables 79 and 80; and it may also be noted that, as an average of the four phosphorus-potassium plots, the average yields during the second 1 2-year period show 6.4 bushels less corn, 8.3 bushels less oats, and 25 pounds less hay, but with 3.5 bushels more wheat, than during the first 12 years. These figures mean that for each rotation (four years) the yields have decreased by 2.1 bushels of corn, 2.8 bushels of oats, and 8 pounds of hay, while the yield of wheat shows an increase of 1.2 bushels. The algebraic sum shows, as an average, that each re- curring rotation produces $2.36 lower crop values from an acre of land than during the preceding four years. All this must remind us of the mineral plot on Agdell field, where the yields of turnips and legumes are still well maintained, and the wheat yield has appreciably increased, while only the barley has very markedly decreased. Mathematically, it is not possible for the roots and stubble of the clover crop to furnish sufficient nitrogen for the other four crops, — timothy (associated with the clover), corn, oats, and wheat; but the question again arises, whether important amounts of atmospheric nitrogen may not be fixed that are not thus ac- counted for. It is fully established that the azotobacter (and possibly other similar bacteria) fixes measurable quantities of free nitrogen under favorable conditions; and it is also fully es- tablished that the bacteria which commonly live in symbiotic relationship with legume plants can fix appreciable amounts of free nitrogen, under suitable artificial conditions, and entirely independent of legume plants. It is thus conceivable that these may fix nitrogen to a greater or less extent while they continue to live, not in the tubercles of growing clover, but upon the dead and decaying residues; and, if such is the case, it is exceedingly probable that the presence of carbohydrate matter (as in plant residues) and a liberal supply of available mineral plant food in a 438 INVESTIGATION BY CULTURE EXPERIMENTS limestone soil, will furnish the very favorable conditions, although the data thus far reported from Agdell field (Table 75) show a greater average loss of nitrogen (245 pounds from the surface 9 inches only) from the phosphorus plots than from the untreated plots (105 pounds) in the legume rotation; while, as an average of the four plots, the fallow rotation lost less nitrogen than the legume system. It should be kept in mind, also, that the organic matter of the soil contains nitrogen as well as carbon, and that the amount of combined nitrogen liberated from this organic matter may be nearly or quite sufficient to meet the needs of the bacteria that can be supported by the carbonaceous food. In laboratory cultures the fixation of nitrogen amounts to about 10 milligrams for each gram of sugar (mannite) consumed by the "free-living" bacteria.1 Thus the amount of nitrogen fixed is equal to about i per cent of the carbonaceous food consumed; whereas the organic matter of the soil contains, as a rule, more than 2 per cent of nitrogen. On Broadbalk field the nitrogen 2 content of the surface 9 inches decreased during 28 years (1865 to 1893) by 285 pounds (from 2722 to 2437) on plot 3 (unfertilized), by 265 pounds (from 2782 to 2517) on plot 5 (minerals), and by 63 pounds (from 3034 to 2971) on plot 7 (minerals and 86 pounds of nitrogen) ; while the only in- creases shown are 633 pounds (from 4343 to 4976) on plot 2 (farm manure), and 131 pounds (from 2991 in 1865 to 3015 in 1881 and to 3122 in 1893) on plot. 14, which receives ammonium salts (86 Ib. N), acid phosphate, and magnesium sulfate. (The possibility of erosion or deposit from surface washing should not be over- looked. Compare the nitrogen content of plots u, 12, 13, and 14 with respect to each stratum, as shown in Table 73.) 1 In this connection attention is called to the point that if increased growth of plants is caused by the use of pyrogallol, as reported from the unverified experi- ments of Whitney and Cameron, it may be due to the fixation of free' nitrogen by the nonsymbiotic bacteria that find in pyrogallol a suitable carbonaceous food supply. It is known that the addition of sugar to ordinary soil deficient in nitrogen will increase the growth of nonleguminous plants because of the increased nitrogen fixation by the "free-living" bacteria. JAll of these determinations were made by the older soda-lime method and are considered trustworthy for comparison, but the 1893 analyses reported in Table 73 were made by the newer Kjeldahl method, which gives somewhat higher and more nearly correct results. PENNSYLVANIA FIELD EXPERIMENTS 439 As already stated, a study of the present nitrogen content of the soil of Agdell field will probably furnish more satisfactory informa- tion than can be secured from any other source at this time. It is very evident that the loss of nitrogen in drainage water usually exceeds the addition in rainfall; and, unless there are sources of nitrogen other than can be found by the analysis of the legume plants (tops, roots, and tubercles), we must make provision to supply a sufficient excess of nitrogen in farm manure, crop resi- dues, or otherwise, to meet the needs of large crops and to overcome the loss in drainage from rich land. In Bulletin 221 of the New Jersey Agricultural Experiment Station, issued July, 1909, Voorhees and Lipman report in detail the results of ten years' investigations with twenty culture experi- ments (in triplicate) in which corn, oats, wheat, and timothy were grown in rotations in 60 cylinders, each 4 feet long and 23^ inches in diameter, set in the earth and open at both ends, so as to approach natural conditions for drainage. Cow manure, fresh and leached, and cow dung (solid excrement), fresh and leached, were used with and without sodium nitrate, ammonium sulfate, and dried blood, in various combinations. At the beginning of the experiment, in 1898, the surface soil (8 inches deep) contained 155.47 grams of nitrogen in each cylinder. The amounts of nitrogen applied during the ten years varied from 38.25 grams in the leached dung to 58.31 grams in the fresh manure and sodium nitrate combined. The total amounts of nitrogen removed in the sixteen crops harvested during the ten years varied from 21.88 to 36.70 grams; and the total loss of nitrogen, other than that contained in the crops removed, varied from 25.12 to 39.38 grams. Thus, in these long-continued and very carefully conducted experiments the absolute chemical control shows loss of nitrogen by leaching far in excess of possible additions by rain- fall, azotobacter, etc. After a full consideration of the data accumulated in these experiments with respect to their bearing upon the question of denitrification, the authors make the following statements: "We must conclude, therefore, that at least with cow manure, used at the rate of sixteen tons per annum for a period of ten years, no destruction of ni- trates takes place. In view of the long duration of the experiment, and of the 440 INVESTIGATION BY CULTURE EXPERIMENTS comparatively large amounts of manure used in the course of the ten seasons, we must assume that denitrification is not a phenomenon of economic importance, in general farming and under average field conditions. . . . We have no hesita- tion in emphasizing again the view expressed above that under the wide range of field conditions, denitrification is not a phenomenon of economic significance to the general farmer." With our present knowledge we should not do less than to base our practice upon the known mathematical and chemical facts concerning the nitrogen requirements of crops and the nitrogen content of manures, legume crops, and crop residues. CHAPTER XXI OHIO FIELD EXPERIMENTS ASIDE from the experiments outlined in Table 40, which deal especially with manure, alone and reenforced with different mate- rials, the Ohio Experiment Station has conducted, for 15 years, two very extensive and valuable investigations by means of plot experiments, relating to the maintenance of soil fertility. In one of these a five-year rotation is .practiced on five separate fields or series, each of which contains 30 tenth-acre plots about i by 16 rods, each of the five crops, corn, oats, wheat, clover, and timothy, being represented every year (excepting the clover and timothy for the first two years). In the other investigation, potatoes, wheat, and clover are grown in a three-year rotation on three separate series, each of which contains 34 tenth-acre plots of the same shape. Each of the crops is represented every year (ex- cept wheat the first year and clover the first two years). The detail plan of these experiments and the average results secured for the 15 years (1894 to 1908) are shown in Tables 82 and 83. Seasonal variations are too great to justify an attempt to de- termine from the data secured in fifteen years (only 13 years with clover and timothy) whether the productive power of the soil is increasing or decreasing. It will be recalled that Jethro Tull grew 13 crops of wheat in succession on the same land without the use of manure or fertilizers, and from the data secured the conclusion was drawn, " that a good crop of wheat, for any number of years, may be grown every year upon the same land without any manure from first to last." A more recent similar illustration is furnished by the Minnesota Experiment Station, showing average yields of 14.7 bushels of wheat from 1893 to 1898, and 17.2 bushels from 1899 to 1904, where wheat was grown every year without manure or fertilizer. 441 442 INVESTIGATION BY CULTURE EXPERIMENTS TABLE 82. OHIO EXPERIMENTS: FIVE-YEAR ROTATION Average Records per Acre for Three Rotations, 1894 to 1908 TREATMENT FOR EACH FIVE YEARS AVERAGE YIELDS PER ACRE FROM FIVE ACRES Plot No. | Elements Applied (Lb.) Forms of Plant Food, if not Stand- ard1 Q I o 17 18 19 20 21 Yard manure, 4 tons on wheat Yard manure, 8 tons on wheat 148.5 157-8 162.2 27.2 31-3 32-5 2450 3050 313° 2380 3300 .3200 2980 348o 3860 72.53 79.69 82.00 72-53 7I-03 69-53 8.66 13.46 i. 20 2.40 7.46 1 1. 06 9.69 14-54 »S 25 20 20 83 83 143.0 189.2 181.7 24-5 34-o 4.1 2660 3430 3120 2790 3MO 313° 68.03 90.85 87.74 68.03 68.64 69.25 22.21 18.49 11.13 11.13 11.08 7-36 19.94 14-34 Oil meal, 460 Ib. 22 »3 24 25 2.5 2O 20 83 83 Dried blood . . Amm. sulfate . 148.7 181.9 181.6 24.1 34-9 J4jj 2180 3220 3190 69-85 88.42 88.23 69.85 69-65 69.45 18.77 18.78 11.13 II. 15 7.64 7-65 14-45 14-57 as 2() «7 98 20 38 38 38 146.2 176.0 18^.8 151.2 184.8 24.7 35-6 37-0 2270 335° 3630 2430 3740 2700 3420 3250 2900 371° 69.25 87.98 91.29 69.25 70.00 70-75 10.93 14.82 20 20 83 8j Raw bone meal . Acid bone black . 17.98 20.54 12.68 13.0^ 5-30 7-46 2O 83 Basic slag . . . 24.9 37-6 71-49 02.,S( 71.49 72.24 20.65 12.68 7.07 14.09 30 Yard manure, 8 tons on potatoes 200. o 31-9 3090 3700 93-43 72.90 20.44 2.40 i, S.o.i 26.60 31 3f Yard manure, 16 tons on wheat 160.3 1 88.0 24. 33-4 2410 33io 2760 34 so 73-73 90-13 73-73 70.29 10.84 4.80 1.5-04 22.27 33 34 25 2C 8; Tankage .... 170.0 135-0 33-4 23.2 3080 1990 2650 2220 85.03 63.40 66.85 63.40 18.18 10.73 7-45 14-57 1 The standard forms include both dried blood (usually 50 Ib.) and sodium nitrate (usually 200 Ib.), acid phosphate, and potassium chlorid, the applications being divided, in most cases, between the potatoes and wheat. hay, $6.00 a ton; nitrogen, 15 cents a pound; phosphorus, 12 cents (10 cents in raw bone, slag, and tankage) ; and potassium, 6 cents. The price of potatoes varies greatly, and for that reason the figures OHIO FIELD EXPERIMENTS 449 given in the last column of Table 83 are based upon the price of 50 cents a bushel for potatoes, while 30 cents a bushel is the price used in the other computations. Of course the increase in crop values resulting from treatment is not computed at the delivered price for marketable potatoes, but sometimes this would be justi- fied, because the treatment may largely increase the percentage of marketable potatoes, and even with other crops the improvement in quality, as well as in quantity, may be a factor of some impor- tance. In any case, potatoes belong to the crops of intensive agri- culture, the largest average yield (200 bushels) amounting to $60 an acre at 30 cents, and to $100 an acre at 50 cents a bushel. Part of the field upon which these experiments have been con- ducted was virgin soil, cleared from forest for the purpose, and all of the land was fairly rich at the beginning. The average of 16 analyses of soil from the " East Farm," where the five-year rotation (Table 82) and reenforced manure experi- ments (Table 40) are conducted, and 5 analyses of soil from the " South Farm," where the potato- wheat-clover rotation experi- ments are under way, show that the South Farm soil contains about one half more acid-soluble phosphorus than the East Farm soil. It is also somewhat richer in acid-soluble potassium, while in total nitrogen the East Farm soil is slightly richer. By referring to the column headed " Value if unfertilized " (Table 83), it will be seen that the natural productiveness of the land varies markedly from plot i ($83.70) to plot 19 ($68.03) and plot 34 ($63.40); but the oft-repeated check plot (unfertilized) makes possible a comparison that could not be made without it. On the other hand, we can never be sure that the treatment applied to one plot (as phosphorus to plot 2, for example) has produced the same total increase as it would have produced if applied to some other plot (as to plot 20, for example) . Thus the actual total yield from plot 2 ($92.65) is greater than that from plot 20 ($90.85), but the computed increase from plot 20 is more than twice as great ($22.21) as that from plot 2 ($9.36). The fact is that more plant food is removed from plot 2 than from plot 20, but this is also true, of course, with respect to the adjoining unfertilized control plots. These difficul- ties are emphasized, however, by comparing plots n and 20, both of which receive the three elements, nitrogen, phosphorus, and po- 450 INVESTIGATION BY CULTURE EXPERIMENTS tassium, in the standard forms (the nitrogen chiefly in sodium nitrate). The treatment for these two plots differs only by the addition of 13 pounds more nitrogen to plot n (plot n receives 50 pounds of dried blood for wheat, and plot 20 only 25 pounds). The total yields (except potatoes) are greater from plot n, and the total value of the three crops is greater from plot n ($92.78) than from plot 20 ($90.85); but the increase from plot 20 ($22.21) is much greater than from plot n ($15.03), and the net profit from 20 ($11.08 or $19.94) is several times as great as from u ($1.95 or $6.91). In comparison with such plots as 14 and 27, it seems evident that plot 20 gives results above normal; while it is like- wise evident that plot u shows increases below normal, in com- parison with such plots as 8 and 24. These opposite abnormalities develop the striking discrepancy between n and 20. A study of the yearly details shows that for the first five years (1895 to 1899) the increase from treatment was greater every year in the wheat crops from plot u, the average difference being 3.6 bushels, whereas during the next five years (1900 to 1904) the in- crease in wheat was greater every year on plot 20, the average difference being 2.6 bushels. As an average of the four years, 1905 to 1908, the increase from plot 20 averaged 2.5 bushels more wheat than from n, although in two of these years the treatment gave about equal results on those plots. As an average of the first six years (1896 to 1901) plot n produced 79 pounds less clover than the unfertilized control plots, while on plot 20 the treatment produced an average increase of 630 pounds. These results and discrepancies serve to emphasize' the uncer- tainty of drawing correct conclusions from a single field experiment, even when continued for several years. On the other hand, most of the results from this potato-wheat-clover rotation are concord- ant, and justify confidence. Indeed, there is marked agreement in most cases where direct comparison is possible. Thus the increases from like amounts of plant-food elements on plots 21, 23, 24, and 33 vary only from $18.18 to $18.78. In harmony with the results from all other sources, the use of phosphorus on normal soils proves highly profitable, the increase produced by phosphorus, both alone and in addition to other OHIO FIELD EXPERIMENTS 451 elements, being sufficient to pay for the phosphorus (even in acid- phosphate) and leave a net profit of 200 to 300 per cent. The use of commercial nitrogen or potassium, alone or in combination, is of doubtful advantage with potatoes at 30 cents, but at 50 cents for potatoes the potassium has been a good investment, although, with sufficient manure or clover plowed under to supply the nitro- gen, it is very probable that abundance of potassium would have been liberated from the soil. Potatoes draw heavily upon potassium, and ultimately, on level land which neither receives deposits from overflow nor loses par- tially exhausted soil by erosion, potassium must become so defi- cient as to limit the crop yield, even with the best efforts to main- tain adequate supplies of active organic matter; but the total supply of potassium in 2 million pounds of this Ohio soil is suffi- cient for 200 bushels of potatoes every year for more than 500 years, and the land has sufficient surface drainage to insure some soil erosion. Another series of long-continued and very valuable experiments have been conducted by the Ohio Station on the Strongsville experi- ment farm, on a quite different type of soil, of nearly level topog- raphy, higher clay content, and less perfect physical condition. The surface acre-foot of Wooster soil contains, as a general average, about 2770 pounds of nitrogen, 1700 pounds of acid-soluble phos- phorus, and 7310 pounds of acid-soluble potassium, while the corresponding figures for the Strongsville soil are 6520, 1700, and 6300. Thus the Strongsville soil averages more than twice as rich in nitrogen, but somewhat poorer in acid-soluble potassium, while the phosphorus content is practically equal in the two soils. Table 84 gives the average results obtained from a series of 5- year rotation experiments (1896-1897-1898 to 1907). The plant- food materials are 440 pounds of sodium nitrate (and 50 pounds of dried blood), 320 pounds of acid phosphate, and 260 pounds of potassium chlorid. (One plot (No. 12) receives 680 pounds of sodium nitrate.) The more marked effect of phosphorus on the Strongsville soil is doubtless due to the larger supply of organic matter, the decom- position of which tends to furnish nitrogen and liberate potassium. As an average, the crops from the best-yielding plots have removed 452 INVESTIGATION BY CULTURE EXPERIMENTS 30 pounds of phosphorus during the five years, or 50 per cent more than was applied. TABLE 84. EXPERIMENTS AT STRONGSVILLE, OHIO Data per Acre for Five-year Rotation: Increase Only, except as Noted CLOV- TIMO- CORN, OATS, ER, THY, PLANT FOOD APPLIED I2-YR. AVERAGE II-YR. AVERAGE AVER- IO-YR. AVER- IO-YR. AVER- (Bu.) (Bu.) AGE AGE (Tons) (Tons) Nitrogen 76 Ib 1.18 -•°5 -.24 .08 .01 Phosphorus 20 Ib ... 0.151 IO.OQ 7.14 •45 .16 Potassium 108 Ib .20 .41 -.15 .05 — .OI Nitrogen phosphorus 10.46 14.38 9.72 •39 .13 Nitrogen potassium 1.40 2.24 1.67 ,06 -.02 Phosphorus potassium o 04 10. i-Q 26.8 70.8 64.7 407 Lime, nitrogen, potassium .... 70.0 72.9 S4-7 36. s 74-8 73-o 408 Lime, phosphorus, potassium . . . 49.8 39-6 36'9 i3-7 18.3 27.7 409 Lime, nitrogen, phosphorus, potassium 69. S 69.8 47.8 ^6.2 66.4 73-6 410 Nitrogen, phosphorus, potassium . . 57-2 66.1 50.0 26.5 66.0 71.9 ILLINOIS FIELD EXPERIMENTS 469 Plots i (especially) and 2 in this series were naturally more productive than the other plots, it being the regular custom of the Illinois Station to use the most productive land for the untreated check plots if any such differences are apparent when the field is established, as was the case in this instance. Plot i serves only as a check against the lime treatment, and the average of plots 2, 4, 5, and 8 gives a more reliable basis of comparison for ascertaining the effect of nitrogen. Potassium is evidently the second limiting element in this soil where decaying organic matter is not provided, but the limit of potassium is very far above the nitrogen limit. During the six years plot 7, receiving nitrogen and potassium, produced 291.3 bushels of corn (averaging 72.5 bushels a year), 54.7 bushels of oats, and 36.5 bushels of wheat, per acre. To pro- duce the increase of plot 7 over plot 5 would require about 75 per cent of the total nitrogen applied. Thus, there has been a loss of 25 per cent of the nitrogen applied, which is a smaller loss than usually occurs where commercial nitrogen is used. Without doubt, larger yields would have been produced, especially of corn, if 150 or 200 pounds of nitrogen per acre per annum had been used, which would have increased the cost of nitrogen to $22.50 or $30, re- spectively, per acre each year. It need scarcely be mentioned that commercial nitrogen is used in these and other experiments in Illinois only to help discover what elements are limiting the crop yields. It should never be purchased for use in general farming, but, if needed, secured from the atmos- phere by legume crops to be returned to the soil directly or in ma- nure. On three other series of plots on the Green Valley soil experi- ment field, a three-year rotation of corn, oats, and cowpeas is prac- ticed, every crop being represented every year. On plots receiv- ing lime and phosphorus and legume crops, as green manure, the yield of corn was 45.6 bushels in 1906 and 67.8 bushels in 1907, compared with 70.8 bushels and 64.7 bushels with lime, phosphorus, and nitrogen on plot 6 (see Table 90) and with 10.4 bushels and 13.1 bushels with no nitrogen on plot 4, for the respective years. On other plots receiving comparable treatment, where lime, phos- phorus, and potassium were used with nitrogen-gathering legume 47o INVESTIGATION BY CULTURE EXPERIMENTS crops as green manure, the corn yields in the three-year rotation were 54.6 bushels in 1906 and 51.5 bushels in 1907, compared with 66.4 bushels and 73.6 bushels on plot 9 with nitrogen applied, and compared with 18.3 bushels and 27.7 bushels on plot 8. with no nitrogen for the same years. The growing of legume crops and the use of farm manure (and possibly limestone) are the only recommendations made for the improvement of these well-drained sand soils, although further tests may show profit from potassium until more organic matter is supplied. As a rule, clover cannot be grown successfully on this land, but cowpeas and soy beans are well adapted to such soil, and they produce very large yields of excellent hay or of grain very valuable for feed and also for seed. Under the best conditions, with good preparation and heavy manuring, alfalfa can be grown on this sand soil, more than five tons of alfalfa hay per acre in one year having been grown on part of the Green Valley field. Both soy beans and alfalfa should be inoculated with the proper nitrogen-fixing bacteria. Heavy applications of ground limestone also may be especially beneficial in getting alfalfa started. (It should be kept in mind that residual sand soils, such as are found in the Coastal Plains soil province in the South Atlantic and Gulf States, are, as a rule, very deficient in mineral plant food, as well as in nitrogen.) Peaty swamp lands. Peat is chiefly of two kinds, one being known as moss peat and the other as grass peat. Moss peat consists largely of dead and decaying sphagnum moss, and grass peat of the residues of coarse swamp grass, sedge, flags, etc. Probably most of the beds in Ohio, Indiana, Illinois, and Iowa are grass peat, although there is some moss peat in northern Illinois. Indeed, in the detail soil survey of Lake County, Illinois, one swamp of several acres was found where the sphagnum moss is still growing luxuri- antly over a bed of moss peat. Where the soil consists very largely of decaying peat to a depth of 30 inches or more, it is called deep peat. As shown in Table 15, deep peat contains in one million pounds of surface soil about 35,000 pounds of nitrogen, 2000 pounds of phosphorus, and 2900 pounds of potassium. This shows in the ILLINOIS FIELD EXPERIMENTS 471 surface 6f inches of an acre about five times as much nitrogen as the early Wisconsin black clay loam prairie. In phosphorus con- tent these two soil types are about equal, but the peat contains less than one tenth as much potassium as the black clay loam. Thus, the total supply of potassium in the peat to a depth of 6| inches (2930 pounds) would be equivalent to the full potassium re- quirement (75 pounds) of a hundred-bushel crop of corn for only 39 years, or if the equivalent of only one fourth of i per cent of this is annually available in accordance with the rough estimate previously suggested, about 7 pounds of potassium would be liber- ated annually, or sufficient for about 10 bushels of corn per acre. In Table 91 are given all results obtained from the Manito (Mason County, Illinois) experiment field on deep peat, which was begun in 1902 and discontinued after 1905. The plots in this field were one acre l each in size, being 2 rods wide and 80 rods long, and untreated half-rod division strips were left between the plots, which, however, were cropped the same as the plots. TABLE 91. CORN YIELDS PER ACRE IN ILLINOIS SOIL EXPERIMENTS : MANITO FIELD: TYPICAL DEEP PEAT SOIL PLOT No. SOIL TREATMENT FOR 1902 (Per Acre) CORN, 1902 (Bu.) CORN, 1903 (Bu.) SOIL TREATMENT FOR 1904 (Per Acre) CORN, 1004 (Bu.) CORN, 1905 (Bu.) FOUR CROPS (Bu.) I 2 3 4 5 None 10.9 10.4 8.1 10.4 None 17.0 12. 0 12. 0 IO.I 48.0 42.9 None Limestone, 4000 Ib. Kainit, 600 Ib. . . Kainit, 600 Ib. ; acid- ulated bone, 350 Ib 3°-4 3°-3 31.2 32-4 33-3 33-9 Limestone, 4000 Ib. ; kainit, 1200 Ib. Kainit, 1200 Ib. ; steamed bone, 395 Ib Potassium chlorid, 400 Ib 49.6 53-5 48.5 47-3 47.6 52.7 159-7 164.7 166.3 Potassium chlorid, 200 Ib 6 Sodium chlorid, 700 Ib it. i I3-1 None 24.0 22.1 7°-3 7 8 9 Sodium chlorid, 700 Ib !3-3 36.8 26.4 14-5 37-7 25-1 Kainit, 1200 Ib. . . Kainit, 600 Ib. . Kainit, 300 Ib. . 44-5 44.0 4i-5 47-3 46.0 329 164.5 125-9 Kainit, 600 Ib. . . Kainit, 300 Ib. . . 10 None I4-92 14.9 None 26.0 13.6 69.4 1 In 1904 the yields were taken from quarter-acre plots because of severe insect injury on the other part of the field. 2 Estimated from 1903; no yield was taken in 1902 because of misunderstanding. 472 INVESTIGATION BY CULTURE EXPERIMENTS The results of four years' tests as given in Table 91 are in com- plete harmony with the information furnished by the chemical composition of peat soil as compared with that of ordinary normal soils. Where potassium was applied, the yield was from three to four times as large as where nothing was applied. Where approxi- mately equal money values of kainit and potassium chlorid were applied, slightly greater yields were obtained with the potassium chlorid, which, however, supplied about one third more potassium than the kainit. On the other hand, either material furnished more potassium than was required by the crops produced. The use of 700 pounds of sodium chlorid (common salt) produced no appreciable increase over the best untreated plots, indicating that where potassium is itself actually deficient, salts of other ele- ments cannot take its place. Applications of two tons per acre of ground limestone produced no increase in the corn crops, neither when applied alone nor in combination with kainit, neither the first year nor the second. Reducing the application of kainit from 600 pounds to 300 pounds, for each two-year period, reduced the yield of corn from 164.5 to 125.9 bushels. The two applications of 300 pounds of kainit furnished 60 pounds of potassium for the four years, or sufficient for 84 bushels of corn (grain and stalks). The difference between this and the 125.9 bushels obtained is 42 bushels, about what was obtained from the poorest untreated plot. The underdrainage provided for this experiment field was not sufficient for the best results, probably because of insufficient nitrification. In other experiments on peaty soil with imperfect drainage, the addition of $15 worth of nitrogen with potassium produced about 15 bushels more corn than where potassium alone was used. Peaty alkali soils. Aside from deep peat, there are many other types of peaty soil, as will be seen from the classification of Illinois soil types given in a previous chapter. Thus we find shallow peat and medium peat, underlain with clay, sand, rock, etc., and also sandy peat and peaty loam; and in some instances peaty soils also contain alkali, consisting chiefly of harmless calcium car- bonate with smaller amounts of injurious magnesium carbonate. In some cases these peaty soils actually contain a good percentage ILLINOIS FIELD EXPERIMENTS 473 of total potassium, more commonly in the subsurface or subsoil, but sometimes in the surface soil, also; and yet the untreated soil is unproductive, while the addition of potassium salts produces large and very profitable increases in the yield of corn, oats, etc. In pot-culture experiments the author has even been able by the addition of potassium sulfate to correct to a considerable extent the injurious property of magnesium carbonate that has been purposely applied to ordinary brown silt loam prairie soil which is known to contain abundance of available potassium. These facts are mentioned here because he recommends, in humid sections, trial applications of potassium salt to all classes of peaty and alkali soils that are unproductive after being well drained, whenever the supply of farm manure is insufficient. It should be understood that plenty of farm manure, preferably quick-acting, or readily decomposable, manure, such as horse manure, will supply potas- sium and thus accomplish everything that potassium salts can accomplish, and on some swamp soils manure produces good re- sults where potassium is without effect. In pot-culture experiments soils containing injurious amounts of magnesium carbonate have been treated with calcium sulfate (land-plaster) which brings about a double decomposition, or inter- change, forming the harmless insoluble calcium carbonate (lime- stone) and the very soluble magnesium sulfate, which is subse- quently leached out, leaving the soil productive. The new Manito experiment field is on alkali soil consisting of peaty, clayey sand with some gravel, and containing sufficient total potassium for normal crop yields. In Table 92 are recorded the treatment applied and results obtained in 1907. "TABLE 92. CORN YIELDS IN ILLINOIS SOIL EXPERIMENTS: NEW MANITO FIELD: PEATY ALKALI SOIL PLOT No. TREATMENT APPLIED FOR 1907 CORN (Bu. per Acre) 2OI 2O2 W. 202 E. 203 2O4 205 None 8.8 43-5 64.9 73-o 4-9 5-4 Manure 6 tons Manure, 12 tons Potassium sulfate, Calcium sulfate, 2 None .... 400 pounds to 1 6 tons 474 INVESTIGATION BY CULTURE EXPERIMENTS Plot 204 is divided into four equal parts and the calcium sulfate applied at the rate of 2 tons, 4 tons, 8 tons, and 16 tons per acre, at a cost of $6 per ton. It produced no benefit in 1907. Whether it will assist in the removal of the magnesium carbonate by double decomposition and leaching and thus improve the soil in time, time alone will tell. The 400 pounds of potassium sulfate are applied for a three- year rotation at an initial cost of $10. The increase of 66 bush- els of corn produced the first year, at 35 cents a bushel, amounts to more than twice the total cost of the potassium. The manure also gave very excellent results. In Table 93 are given all results obtained during six years' experiments on part of theMomence soil experiment field, located in Kankakee County, Illinois, near the Indiana line, on peaty swamp land which contains much decaying peat and coarse sand in the surface and subsurface, with a clayey sand subsoil resting on impure limestone, while the surface, subsurface, and subsoil contain more than half of the normal amounts of total potassium (19,000, 47,000, and 73,000 pounds, respectively, per acre). The soil contains but little alkali. After 1902 (when the corn was damaged by water) the land was tile-drained sufficiently well for ordinary years, but in the ex- tremely wet season of 1907 the corn was planted very late, and with the continued wet weather resulted in almost a complete failure. Potassium was not applied to plot 102 for 1902 and 1903, and was not applied to plot no for 1904. The untreated check plot 101 is naturally somewhat more productive than the other plots. These results from the new Manito field and from the Momence field, on abnormal swamp lands, emphasize the fact that, although some principles are well established and can be applied with normal results on normal soils and on some abnormal soils (as the deep peat and sand ridge soils), there are complex problems still unsolved relating to soils and soil fertility. These problems may be chemical, physical, or biological, and their solution may require the application of science yet unknown. Thus, some essential element of plant food may be present in abundance but held in unavailable form by chemical combination or physical absorption; or there may exist some still undiscovered ILLINOIS FIELD EXPERIMENTS 475 TABLE 93. CROP YIELDS IN ILLINOIS SOIL EXPERIMENTS: MOMENCE FIELD PLOT No. PEATY SWAMP LAND: SOIL TREAT- MENT APPLIED CORN (Bushels per Acre) 6 CROPS IQ02 1903 1904 1905 1906 1907 Bu. Value 101 102 None 6.6 5-.S 14.9 7-i 4.8 20. i 6.8 33-9 6.8 52.6 •3 14.9 40-5 $14.18 Lime (and potassium after 2 years) 103 104 I°S Lime, nitrogen Lime, phosphorus .... Lime, potassium 0.0 i-3 23-7 3-6 4.6 72.2 1-3 •4 34-6 4.1 1.8 41.4 5-3 1.9 50.0 •4 .2 16.2 14.7 IO.2 238.1 $ 5-15 3-57 83-34 106 107 108 Lime, nitrogen, phosphorus . Lime, nitrogen, potassium . . Lime, phosphorus, potassium . o.o 19.7 32.0 3-9 71.1 73- 1 .6 33-5 42.0 1.6 38.5 36-3 4-5 53-i 59-4 •4 16.5 19.9 II. O 232.4 262.7 $ 3-85 81.34 9i-95 109 no Lime, nitrogen, phosphorus, po- tassium Nitrogen, phosphorus, potassium 25.2 24.1 66.8 70.4 39-2 19.0 42.9 24.8 65.6 Si-3 25-1 23-4 264.8 $92.68 chemical substances injurious to agricultural plants or to necessary bacterial life, which may be corrected or destroyed by potassium salts or other materials; and the recent very extensive investi- gations by the United States Bureau of Soils indicate that condi- tions may be brought about, artificially at least, in which organic toxic substances develop that are injurious to plant growth. CHAPTER XXIII THE gray silt loam on tight clay is one of the common types of prairie land in the Kansan and lower Illinoisan glaciations. This or very similar prairie soil is found in many places, as in south- ern Illinois, northern Missouri, southern Iowa, and southeastern Kansas. In Illinois this soil type is found chiefly between the Kaskaskia and Wabash rivers in an area bounded on the south by the Ozark Hills and on the north by the terminal moraine of the Wisconsin glaciation, which passes through Shelby, southern Coles, and Edgar counties. This type of soil is well known and everywhere recognized by the farmers themselves as " hardpan land." It consists of a friable gray silt loam which commonly varies in depth from 6 to 12 inches, and below which is a light gray or nearly white layer, or stratum, of slightly loamy silt varying from less than one inch to more than 10 inches in thickness, and commonly referred to as " the gray layer." At a depth of 16 to 20 inches the soil is underlain by a tight clay subsoil, frequently termed " hardpan." It should be understood, however, that this subsoil is not true hardpan, which consists of sand or gravel cemented together with clay to form a substance which is practically impervious to water. The subsoil of this gray silt loam prairie is a tight clay, inclined to be gummy. Water passes through it, although quite slowly, and when wet it can be spaded without special difficulty, but when dry it becomes stiff and hard. Closely related to this prairie soil are level upland timbered soils underlain with tight clay, found in the southern part of Indiana, Illinois, and Iowa, and also in northern Missouri and western Kentucky. Where this soil is enriched by proper treatment, excellent crops are grown in seasons of normal rainfall, but they are likely to suffer in times of drouth more than would be the case with a better sub- 476 FIELD EXPERIMENTS IN THE SOUTH 477 soil. As a rule, the rainfall in southern Illinois is abundant and well distributed during the growing season, and where the top soil is kept fertile, severe injury from drouth is not common. From Table 15 it will be seen that the average surface soil of this type contains per acre 2880 pounds of nitrogen, 840 pounds of phosphorus, and 24,940 pounds of potassium, and it requires an application of 2 to 5 tons of ground limestone. Compared with the requirements for a practical crop rotation, this soil is very poor in phosphorus and very deficient in lime. Compared with the com- position of fertile soils, it is also deficient in humus as indicated by the total nitrogen. If by the best systems of crop rotations, with proper use of green manures, we can liberate, in favorable seasons, the equivalent of i per cent of the phosphorus contained in the surface soil, it would amount to about 8 pounds per acre for the first year for the type of soil under consideration. This would be sufficient for a 25- bushel crop of wheat. If with less perfect systems only half of i per cent is liberated, it would amount to 4 pounds, or enough for a i2-bushel crop of wheat, which is about the average yield for this soil. On the Illinois soil experiment field near Odin, Marion County, on this ordinary prairie land of the lower Illinoisan glaciation, wheat is grown in a four- year crop rotation with clover, corn, and cowpeas. By having four different series of plots, every crop may be grown every year. As an average of four years (1904, 1905, 1906, and 1907), wheat grown in this rotation produced 1 1| bushels per acre with no special soil treatment, all crops having been removed. Where one cowpea crop and some catch crops (as cowpeas seeded in the corn) had been plowed under during the rotation, the aver- age yield of wheat was increased to 14 bushels. Where lime or ground limestone had been applied and the cow- peas also plowed under, the average yield of wheat was 18^ bushels per acre. On this set of plots better cowpea crops and catch crops were produced and turned under as green manure, because the soil acidity had been corrected by the lime, applied for the special benefit of the legume crops. Where phosphorus was applied in addition to the use of lime and 478 INVESTIGATION BY CULTURE EXPERIMENTS green manure, the average yield of wheat during the four years was 27 bushels; and where potassium also was included, the average yield was 29! bushels of wheat per acre. These results are quite in harmony with what might be expected from the chemical composition of the soil. If, however, we con- sider the corn crops in the same rotation, we have a somewhat different set of results. The average yield of corn for the four years on the untreated rotated land was 38 bushels per acre; with legume treatment (cowpeas turned under), 41 bushels; with legume and lime treat- ment, 45 bushels; with legume, lime, and phosphorus, 46 bushels; and with legume-lime-phosphorus-potassium treatment the average yield of corn for four years was 61 bushels per acre. For more convenient comparison, these results are shown in Table 94. TABLE 94. CROP YIELDS IN ILLINOIS SOIL EXPERIMENTS: ODIN FIELD GRAY SILT LOAM PRAIRIE: LOWER ILLINOISAN GLACIATION AVERAGE OF EIGHT TESTS IN FOUR YEARS; Two TESTS EACH YEAR FOR EACH CROP (Bushels per Acre) Soil Treatment Applied Wheat Corn None (except rotation) n.6 13.8 18.5 27.1 29-5 38-3 40.8 45-3 46.2 61.3 Legume (cowpeas turned under) Legume, lime Legume, lime, phosphorus Legume, lime, phosphorus, potassium . . . These results are four-year averages. They were made in dupli- cate each year. They are representative and trustworthy. They have also been confirmed by results from other experiment fields on the same type of soil. The effects upon corn of the green manure alone and with lime are about the same as upon wheat, but the effects produced by phosphorus and potassium are very different with the two crops, phosphorus producing the largest increase in wheat, while potas- sium is much more effective with corn, although potassium without phosphorus (in other experiments) produces less increase in corn than when applied in addition to phosphorus. FIELD EXPERIMENTS IN THE SOUTH 479 A study of Table 23 will show that a 6 1 -bushel crop of corn re- quires more potassium than a 3o-bushel crop of wheat, which fact may account in part for the greater effect of potassium on corn, although about the same relation holds for phosphorus. A more important difference probably exists in the relative feeding powers of the two crops, influenced (i) by the difference in root system, including the different depths of feeding, (2) by the difference in seasonal conditions and consequent difference in decay of humus, in decomposition of other soil materials, and in activity of soil organisms during the principal period of growth, (3) by the sol- vent action of the carbon dioxid excreted by the bacteria and from the plant roots, and (4) possibly by different requirements as to the forms or combinations in which the plant-food elements can be absorbed and assimilated or utilized by corn and wheat. The Rothamsted data contribute much toward the solution of this practical problem, but the very important question recurs, whether more or less of the effect attributed to potassium may not be due to the stimulating action of the soluble potassium salt in liberating other substances from the soil instead of serving directly as plant food; and, if so, would it be advisable and more profitable to substitute some other less expensive material, such as kainit, for the concentrated potassium sulfate used in these experiments ? It can also be stated that as an average of 56 tests (including the use of twenty-five different varieties of corn) conducted in 1907 and 1908 on the Illinois experiment field near Fairfield in Wayne County, on the same type of soil, an application of 200 pounds per acre of potassium sulfate, containing 85 pounds of the element potassium and costing $5, increased the yield of corn by 5.4 bushels per acre; while 600 pounds of kainit containing only 60 pounds of potassium and costing $4, gave 9.9 bushels' increase. These applications are made but once for a four-year rotation. The kainit with 25 pounds less potassium produced 4.5 bushels more corn than the sulfate. At 40 cents a bushel for corn, the kainit has paid for itself. Kainit contains about 25 p'er cent of potassium sulfate together with some 16 per cent of magnesium sulfate, 12 per cent of magnesium chlorid, and 33 per cent of sodium chlorid, all of which are soluble salts; and the results plainly indicate that the effects produced are due not solely to the element potassium, 480 INVESTIGATION BY CULTURE EXPERIMENTS but in part at least, and probably in large part, to the stimulating action of the soluble salt. The soluble salts were applied in addition to phosphorus and the yields compared with the results obtained where the same amounts of phosphorus were applied without the soluble salts mentioned. Limestone was also provided in all cases. The soil is not well supplied with decaying organic matter, the action of which will largely, or, if provided in abundance, entirely take the place of the action of the soluble salts as such. Additional experiments on the Fairfield field include an equally complete test with kainit and potassium sulfate on land to which 8 tons per acre of farm manure had been applied. As an average of 56 tests with each material, 200 pounds of potassium sulfate increased the yield of corn by 1.6 bushels, while the 600 pounds of kainit gave 1.4 bushels' increase, as compared with 5.4bushels' and 9.9 bushels' increase, respectively, where these soluble salts were applied in the absence of manure, all other conditions being the same. Thus, where farm manure is supplied, the soluble salts produced but little effect and are not used with profit. On the other hand, phosphorus usually produces its greatest effect when used in con- nection with organic matter. In Table 95 are given the results obtained during seven years on the Du Bois experiment field, in Washington County, Illinois, on the same soil type (gray silt loam on tight clay). In this field there are two independent series of ten plots each, and the crop yields reported in the table are in all cases the average from two plots with like treatment. For convenient comparison it may be stated that at conservative prices the value of the seven crops on the untreated land is $34.30, while $99. 1 1 represents the corresponding value from an acre treated with lime, bone meal, and potassium sulphate, costing $46.25. The yellow silt loam is found in all glaciations, and much more abundantly (relatively) in the unglaciated areas in the South Central states. Like most of the soils of the Central states, it consists of a loessial deposit. It occupies much of the sloping lands or hillsides, not only in the original hilly sections (as in the un- glaciated, or driftless, areas from southern Illinois to northern Mississippi) , but also in the broken land regions along most of the FIELD EXPERIMENTS IN THE SOUTH 481 TABLE 95. CROP YIELDS IN ILLINOIS SOIL EXPERIMENTS: Du Bois FIELD GRAY SILT LOAM PRAIRIE: LOWER ILLINOISAN GLACIATION AVERAGE OF Two SERIES EACH YEAR (Bushels or Tons per Acre) Soil Treatment Applied Corn 1902 Oats 1903 Wheat 1904 Clover 1905 Corn 1906 Oats 1907 Wheat 1908 None 4-9 ,S-o J3-3 16.7 4.8 9.0 1.27 1.67 3J-4 34-4 16.0 26.3 2.6 9-5 Lime Lime, nitrogen 4-3 10.0 8-3 19.4 26.7 27.4 10. 1 26.7 15-5 1.79 2-35 2.19 34-9 34-2 48.2 34-i 37-9 41.8 n-5 18.5 i5-5 Lime, phosphorus Lime, potassium Lime, nitrogen, phosphorus . . . Lime, nitrogen, potassium .... Lime, phosphorus, potassium . . . 8.7 7.2 J3-3 29.4 25-5 27.8 32.0 21.8 29.9 2-37 2-43 2.91 3M 46.0 52.1 46-3 4i-5 47-2 19.7 i7-S 21.9 20.5 II.O Lime, nitrogen, phosphorus, potassium Nitrogen, phosphorus, potassium . . 10.4 3-4 3°-5 29-4 3r-9 27.8 2.86 2.69 49.0 45-3 44-4 36.1 interior streams in glaciated areas. Under ordinary methods of cultivation these lands are subject to serious loss from surface washing, and even when not under cultivation there is and has been more or less rapid erosion taking place. Where this soil has been under ordinary cultivation for several years, it is almost invariably poor in humus and nitrogen, and the dominant problem is to maintain or increase the organic matter in the soil, which will also increase the nitrogen. Of course the organic matter must in large part at least be grown upon the land, and legume crops are most suitable for this purpose because their growth is not limited by the small nitrogen content of the soil, and they also furnish green manures or animal manures rich in nitrogen. While these soils are not rich in phosphorus, that element is not the chief limit to the yield of crops because the nitrogen limit is so much lower as measured by crop requirements and by culture experiments. Furthermore, by surface washing, the nitrogen, which is contained only in the humus, is rapidly depleted, while the phosphorus is constantly renewed because of the supplies in the underlying materials. It is certain that for the highest crop yields phosphorus must be applied, and very probably it can ulti- mately be applied with profit in the best systems of soil improve- 482 INVESTIGATION BY CULTURE EXPERIMENTS ment and preservation, but, as stated above, the first requisite is an increase in humus and nitrogen. There is, however, a serious difficulty to the growing of legume crops, especially for clover and alfalfa. This type of soil, where it has been long under cultivation, is markedly sour or acid. This ap- plies to the Kansan glaciation in Missouri and to the lower Illi- noisan glaciation, and especially to the unglaciated yellow silt loam in the southern parts of Illinois and Indiana, and in the loess- covered areas of Missouri, Kentucky, Tennessee, and Mississippi. In the northern glaciations this type of soil is less acid than in the Kansan and lower Illinoisan, but it is usually more or less acid in the middle and upper Illinoisan, in the pre-Iowan and lowan, and even in the early Wisconsin glaciation, — and not only in the Central states, but also in New York and other Eastern states. In the unglaciated areas and in the lower Illinoisan and Kansan glaciations initial applications of at least two tons per acre of ground limestone are recommended for the yellow silt loam; and for the other glaciations two tons or more may well be applied where acidity is shown in the surface and subsoil and where difficulty is encountered in the growing of red clover or alfalfa. One of the very best crops, and probably the m9st satisfactory and profitable crop, to be grown on these yellow silt loam soils is alfalfa. Its power to secure nitrogen from the air, to root deeply, and to live for many years are all very great advantages for this soil. Furthermore, experiments have shown that where the land is properly treated with heavy applications of lime or ground limestone (five tons per acre) and thoroughly inoculated with the alfalfa bacteria and the alfalfa seeded on well-prepared and well- manured land at the proper time and given proper care, it grows luxuriantly and yields large and profitable crops on this soil, as in Illinois, Ohio, and New York. On the other hand, to sow 20 to 25 pounds of good alfalfa seed on this soil without special and proper treatment is much like throwing away about $4 an acre. Of course, if alfalfa is grown on this land, it should be fed on the farm, in part at least, and the manure returned to the soil, not only to help the alfalfa but also for other crops to be grown, such as corn and potatoes, which are a very profitable crop for this soil when properly enriched. FIELD EXPERIMENTS IN THE SOUTH 483 Table 96 gives the yields of corn, wheat, and clover obtained in 1907 on the Vienna soil experiment field in Johnson County, Illinois, located on the less rolling phase of yellow silt loam in the unglaciated area, and typical of more extensive areas of this type in other Southern states. (It should be remembered that geo- graphically and agriculturally one third of Illinois belongs with the South Central states. A straight line from the north point of Kentucky to the northeast corner of Missouri divides Illinois into two practically equal parts.) The land on which the Vienna field is located has been cropped for about seventy-five years. It had never had any soil treatment, so far as can be determined, and was badly run down when the Experiment Station came into possession of it in 1902. The field is divided into three series of five fifth-acre plots. A three- year rotation of corn, cowpeas, and wheat was followed for four years, then changed to corn, wheat, and clover, but, excepting the 1907 crop, the clover has failed. Cowpeas have been substi- tuted and the crop harvested or plowed under, as seemed practical, according to the yield and weather conditions. In 1902, oats were grown in the place of wheat. The soil treatment has been as follows: Plot i of each series, no treatment except as the cowpea stubble or the second growth of clover has been plowed under in the regular course of the rotation. Plot 2, legume crops and catch crops plowed under, except in 1905-1906-1907. Plot 3, legumes plowed under and lime applied. Plot 4, legume, lime, and phosphorus. Plot 5, legume, lime, phosphorus, and potassium. TABLE 96. CROP YIELDS IN ILLINOIS SOIL EXPERIMENTS: VIENNA FIELD YELLOW SILT LOAM HILL LAND: UNGLACIATED AREA SERIES IOO, CORN, 1907 SERIES 300, WHEAT, 1907 SERIES 200, CLOVER, 1907 VALUE OF THREE CROPS Plot Soil Treatment Applied Bushels or Tons per Acre Total Increase I 2 3 4 5 None 16.7 I7.8 30-3 37-i 38.1 4-3 6.1 13.0 13.6 15.6 •65 .81 1.92 2.56 2.23 $12.76 I5-36 3!-23 37-87 37-64 $ 2.60 18.47 25.11 24.88 Legume Legume, lime ..'.... Legume, lime, phosphorus . . Legume, lime, phosphorus, potas- sium 484 INVESTIGATION BY CULTURE EXPERIMENTS The primary object in applying lime is to correct soil acidity. In the spring of 1902 one ton per acre of slacked lime was applied; but, a method having been worked out by which it can be deter- mined by chemical analysis how much lime is equivalent to the soil acidity to any depth, it was found that the soil on this field was acid in the surface, more acid in the subsurface, and still more acid in the subsoil ; and in order to provide ample lime to correct this acidity, an additional application of eight tons per acre of ground limestone was made in the fall of 1902. From all informa- tion now available, it is believed that two to five tons per acre of ground limestone as an initial application will give very satisfac- tory results. Heavier applications may give more profit per acre, but less profit per ton of limestone used. Phosphorus has been applied at the rate of 25 pounds, and po- tassium at the rate of 42 pounds, per acre per annum, the present regular practice being to apply once in three years 600 pounds of steamed bone meal, containing 12^ per cent phosphorus, and 300 pounds of potassium sulfate, containing 42 per cent of potassium. Seven crops of corn, six of wheat, one crop of oats, and six of cowpeas and one of clover have been grown on the field since the work was begun in 1902. The yields of corn, oats, and wheat are given in Table 97. Counting only the crops removed, the limestone, at $1.50 per ton, has paid for itself and left a net profit of 34 per cent; and, assuming 1000 pounds' loss per acre per annum, more than half of the application still remains in the soil. Neither phosphorus nor potassium has been used with profit, but it is interesting to note that plot 5 has produced six times as much wheat as No. i. Seasonal conditions have very markedly influenced the yields of crops. Larger use of crop residues to increase the organic matter of the soil promises further improvement. Some very instructive results have been obtained from a series of pot-culture experiments which have been in progress since 1902 in the pot-culture greenhouse of the Illinois Experiment Station, and in which this yellow silt loam of the unglaciated hill land has been used. The soil was collected in the fall of 1901, and represents the old worn hill soil of Pulaski County, Illinois, only a few miles from Kentucky. It is much poorer in nitrogen and humus than the FIELD EXPERIMENTS IN THE SOUTH 485 TABLE 97. CROP YIELDS IN ILLINOIS SOIL EXPERIMENTS: VIENNA FIELD Corn, Bushels per Acre 1902 1903 1904 JPOS 1906 1907 1908 TOTAL SOIL PLOT TREATMENT APPLIED No. Series Series Series Series Series Series Series Seven 100 IOO 3°o 200 300 IOO 200 Years I None .... I"?.1? Q.T. 3O. < 37. "? 41.2 16.7 31?. 2 igc o 2 Legume 13.3 S.o 7C.C 42.O 40.6 17.8 3S.6 IQO.7 3 Legume, lime . . . 14.9 8-3 49-1 61.9 48.9 30-3 43-9 257-3 4 Legume, lime, phos- phorus 12. <: 7.4 40-4 "?7.2 4O. Q 77.1 42.0 247.4 5 Legume, lime, phos- phorus, potassium . 19.9 u.6 44-7 56.5 40.9 38.1 50.6 262.3 Oats or Wheat, Bushels per Atre Oats Wheat Wheat Wheat Wheat Wheat Wheat Wheat Series Series Series Series Series Series Series in 200 300 200 IOO 200 300 IOO 6 Years I None IQ.I •4 6.7 1.3 3.8 4-3 none i6.<: 2 Legume :/ 18.8 .6 / 7-i 10.8 O t;.4 T-'O 6.1 none L ^' j 30. o 3 Legume, lime . . . 19.8 •7 / IO.O 18.2 o ~ 17.9 13.0 4-5 ly*> 64-3 4 Legume, lime, phos- phorus 20. o 8.0 14.8 2S.6 11.3 13.6 8.3 81.6 5 Legume, lime, phos- J o *o O phorus, potassium . 3r-7 II. 0 17-5 30.0 15.0 15.6 9.8 98.9 average of the type, although large areas are to be found as badly worn as the field from which this soil was collected. This field has been under cultivation for about seventy-five years, and was still cropped when the soil was collected. During the earlier period of its cultivation the soil frequently produced 25 bushels of wheat an acre, but during the later years about 5 bushels has been the aver- age crop in normal seasons. Table 98 gives the results of seven years' experiments with pot cultures on this type of soil. It is seen that practically no gain has been made except where nitrogen was supplied, either directly in commercial form or indirectly by means of legume treatment. It should be borne in mind that no legume treatment preceded the 1902 wheat crop. The catch crop of cowpeas, which was planted after the 1902 wheat crop and turned under later in the fall, pro- duced a marked effect upon the 1903 wheat crop. This effect be- came more marked in 1904 and 1905, when every pot receiving legume treatment outyielded the pot receiving lime-nitrogen treatment. The last line in the table gives the yields from a pot of virgin soil collected from a piece of unbroken virgin sod land adjoining the cultivated field from which the soil in all the other pots was taken. It is seen that the yields from this pot are gradually de- creasing, doubtless due to the exhaustion of the more active or- ganic matter in the soil. TABLE 98. CROP YIELDS FROM PULASKI COUNTY (ILLINOIS) SOIL Pot-culture Experiments YELLOW SILT LOAM HILL LAND or THE UNGLACIATED AREA 1902 WHEAT !9°3 WHEAT 1904 WHEAT 1905 WHEAT 1906 WHEAT 1907 OATS Soil Treatment Applied (Grams) (Grams) (Grams) (Grams) (Grams) (Grams) •3 e A 4 4 6 Legume lime A IO 17 26 IQ 77 Legume, lime, phosphorus . . Legume, lime, phosphorus, potas- sium 3 3 14 16 19 2O 2O 21 18 19 27 30 Lime, nitrogen ...... 26 17 14 1C Q 28 Lime, phosphorus •2 6 4 6 4 8 Lime, potassium 3 3 3 5 5 10 Lime, nitrogen, phosphorus . . Lime, nitrogen, potassium . . Lime, phosphorus, potassium . 34 33 2 26 14 3 20 21 3 18 21 5 18 16 3 3° 23 7 Lime, nitrogen, phosphorus, po- tassium 34 31 34 21 20 26 Virgin soil (no treatment) . . 24 17 15 17 13 6 The results from the pot cultures bear out very conclusively the results obtained from the field tests; namely, that marked improve- ment can be made on this soil by turning under legume crops where lime has been applied. Very striking results appear in the oat crop in 1907. Of interest in this connection is another series of pot-culture experiments, with soil from the worn hill lands of Henry County, FIELD EXPERIMENTS IN THE SOUTH 487 in northwestern Illinois, which furnish additional information concerning the general need of nitrogen for these hill lands. The plan of these experiments, the soil treatment applied, and the results obtained are all shown in Table 98.1, and they require no further comment. TABLE 98.1. OAT YIELDS FROM HENRY COUNTY (ILLINOIS) SOIL Pot-culture Experiments YELLOW SILT LOAM HILL LAND: UPPER ILLINOISAN GLACIATION OAT YIELDS (Grams per Pot) Soil Treatment Applied None 5 4 Lime Lime, nitrogen 45 6 5 Lime, phosphorus Lime, potassium Lime, nitrogen, phosphorus 38 46 5 Lime, nitrogen, potassium • Lime, phosphorus, potassium Lime, nitrogen, phosphorus, potassium 38 3i Nitrogen, phosphorus, potassium None 5 The Mississippi Experiment Station has reported in Bulletin 108 one year's experiments (1906) at Holly Springs in the northwest part of that state, on similar worn hill land where fertilizers were used for cotton, corn, and cowpeas. The following comments are made: "Phosphates hastened the maturity of cotton. On land with some decaying organic matter in it, phosphate alone gave good results, good enough to make it profitable. Potash alone, or in combination with nitrogen and phosphates, gave no apparent results. Nitrogen (cotton-seed meal) alone gave good re- sults. Cotton-seed meal and phosphates mixed gave good results." Similarly, in referring to the corn and cowpeas, the following statements are made: "The land was thin upland. A drought of seven weeks obtained when the corn was young. Where the soil contained organic matter, phosphates alone gave good results. Potash alone, or in combination, failed to show any appre- 488 INVESTIGATION BY CULTURE EXPERIMENTS ciable benefit. Nitrogen (cotton-seed meal) alone gave good results. A mix- ture of cotton-seed meal and phosphates gave good results." "The fertilizer test with peas was interfered with somewhat by the October storm, but it was apparent that both acid phosphate and crude, finely ground rock increased the growth of peas in a marked manner — apparently doubling the crop." In Iowa Bulletin 98, 1908, are reported the following yields of clover hay from the Leon experiment field on the loess and till soils of southern Iowa. . TABLE 99. SOUTHERN IOWA FIELD EXPERIMENTS PLOT SOIL TREATMENT CLOVER HAY PER ACRE 401 a Loess 2800 pounds 402 d Loess Lime 2400 pounds Am (i Loess Manure . 4480 pounds 404 a Loess Phosphorus 5480 pounds AQC a Mixed ! Potassium . 2600 pounds 406 a Mixed Phosphorus 4750 pounds 4O7 ffl Mixed Potassium 2480 pounds 408 a Mixed Phosphorus, potassium 4680 pounds 409 a 410 a Till Till Lime, nitrogen, phosphorus, potassium . . Nitrogen, phosphorus, potassium .... 6560 pounds 4600 pounds 64 5-46 6 91 48.78 2O 06 ICXX . •} ?g A Q7 •7 OI 2 92 A2 77 1006 . 2 21 r o"? 417 6 48 /I/I 7/1 Averages, 1890 to 1906 .... 3.08 4.11 5-00 5-93 46.47 The yearly records for 1900 and 1904 are the extremes for the seventeen years. In Table 102 are recorded the results of a series of fertilizer ex- periments with cotton, as reported in Georgia Bulletin 63. For the 1904 corn crop fertilizers were applied uniformly to all plots (except No. n) as follows, in pounds per acre: SODIUM NITRATE COTTON- SEED MEAL ACID PHOSPHATE POTASSIUM CHLORID Section 4 17 r<6 I T.O 6.2 Section 5 21 IO< 162 8.0 In 1906 cotton was grown on at least part of the same land as in 1903, with the plan of experiment and results reported in Table 103. It is not clear whether these results may have been influenced 492 INVESTIGATION BY CULTURE EXPERIMENTS TABLE 102. GEORGIA FERTILIZER EXPERIMENTS WITH COTTON PLOT Nos., 2 Rows IN EACH PLOT (4 Rows for Plot n?) POUNDS APPLIED PER ACRE COST OF FERTIL- IZERS (Except Nitrate per Acre) YIELD OF SEED COTTON (Pounds per Acre) VALUE OF IN- CREASE AT 4 $ PER POUND 1004, CORN (Bushels per Acre) Sodium Nitrate (With Seed) Cotton- seed Meal Acid Phos- phate Potas- sium Chlorid NITROGEN TEST: DIVISION B, SECTION 4, EAST, 1003 1904 I, 6, 12, 17, 22 . 2, 7, 13, 18, 23 . 3, 8, 14, 19 . . 4, 9, 15, 20 . . 5, 10, 16, 21 . . ii (unfertilized) I5.6 15-6. I5.6 15-6 15-6 15-6 2OO 1 60 120 80 40 250 306 362 418 474 25.0 30.6 362 41.8 47-4 $4-15 4-15 4-15 4-15 4-15 1146 II2O IO72 I°57 1042 770 $15.04 14.00 1 2. 08 11.48 10.88 20.3 2O. I I8.5 19-5 19.6 2O. I POTASSIUM TEST: DIVISION B, SECTION 4, WEST, 1903 I, 6, 12, 17, 22 . 2, 7, 13, 18, 23 . 3, 8, 14, 19 . . 4, 9, 15, 20 . . 5, 10, 16, 21 . . ii (unfertilized) 19-5 19-5 !9-5 !9-5 i9-5 19-5 195 205 215 225 235 520 547 574 601 628 65 52 39 26 13 $6.50 6.50 6.50 6.50 6.50 1503 1438 1448 1488 I451 IOOO $22.13 19.71 20.16 21.96 20.29 POTASSIUM TEST: DIVISION B, SECTION 5, WEST, 1003 I, 6, 12, 17, 22 . 2, 7, X3> !8, 23 . 3, 8, 14, 19 . . 4, 9, 15, 20 . . 5,10,16,21 . . ii (unfertilized) i9-5 !9-5 i9-5 i9-5 i9-5 i9-5 *95 205 215 225 235 520 547 574 60 1 628 65 52 39 26 *3 $6.50 6.50 6.50 6.50 6.50 1556 1639 1635 1667 1693 IOO2 $24.93 28.66 28.48 29.92 31.09 • POTASSIUM TEST: AVERAGE, 1903 1904 1,6, 12, 17, 22 . 2, 7, i3> *8, 23 . 3, 8, 14, 19 . . 4, 9, 15, 20 . . 5, 10, 16, 21 . . ii (unfertilized) i9-5 i9-5 19-5 T9-5 !9-5 i9-5 J95 205 215 225 235 520 547 574 60 1 628 65 52 39 26 13 $6.50 6.50 6.50 6.50 6.50 1529 1538 1541 1578 1572 IOOI $23.76 24.16 24.30 25.96 25.96 25-3 25-5 25-3 2!. 7 24.0 22.2 by the residual effect of previous applications made in the " ni- trogen test," for example (Table 102). The use of fertilizers on corn in Georgia, if profitable at all, is evidently made possible only because the farm price of corn is very high (69 cents as a ten-year average), and the profit from the FIELD EXPERIMENTS IN THE SOUTH 493 TABLE 103. GEORGIA FERTILIZER EXPERIMENTS WITH COTTON (Division B, Sections 4 and 5, East, 1906) FERTILIZER FORMULA: Acid phosphate . . . 1000 Ib. Cotton-seed meal . . 498 Ib. Potassium chlorid . . 74 Ib. 1572 Ib. APPLIED PER ACRE RESULIS PER ACRE Mixed Fertilizer Nitrate (With Seed) (Lb.) Yield of Seed Cotton (Lb.) Increase Due to Fertilizer (Lb.) Value of In- crease l Amount (Lb.) Cost 12 plots of 3 rows each . 12 plots of 3 rows each . 12 plots of 3 rows each . 2 plots of 4 rows each . 4OO 800 1200 $4.00 8.00 I2.OO 22 22 22 22 1735 1890 2042 I4S4 28l 436 -00 55(3 $10.11 15.69 21.17 1 At 10 cents a pound for lint and 70 cents a hundred for seed. use of fertilizers is found in the cotton crop, which it should be remembered is the most valuable per acre of all the general field crops grown in the United States (potatoes and tobacco being considered as truck or garden crops) . As an average seed cotton is about one third lint and two thirds seed, and a hundred-bushel crop of corn is more difficult to produce than 3000 pounds per acre of seed cotton, which would yield 1000 pounds (or 2 bales) of cotton lint. At ten cents a pound for cotton lint and 70 cents per 100 pounds for cotton seed, such a crop would be worth $114 an acre, or about three times as much as 100 bushels of corn at the ten-year average prjce in the corn belt. Georgia produces less than 200 pounds of cotton lint per acre, on about 4^ million acres, the annual acreage being second only to that of Texas. Judging from the composition of the residual soils of Maryland, Tennessee, and Georgia, and from the statement by Director Red- ding concerning the value of farm manure reenforced with acid phosphate, it seems evident that large use of ground limestone and legume crops, the latter to be plowed under either directly or in farm manure, and liberal applications of phosphate, constitute the most essential factors for the permanent improvement of such land; although, under the present condition of most of the upland soils of Georgia and other Southern states, profitable use can no doubt be made of potassium, at least until the supply of active organic matter is greatly increased, and especially for the cotton crop, which pays such large returns for a comparatively small increase in yield per acre. 494 INVESTIGATION BY CULTURE EXPERIMENTS TABLE 104. ALABAMA FIELD EXPERIMENTS, 1905-1908 Averages per Acre per Annum LAU- DER- CULL- CHIL- Au- TAU- MONT- GOM- TALLA- POOSA MACON 0 DALE MAN TON GA ERY Cr\ Co., fe h s PLANT FOOD APPLIED COST Co., GRAY SILT Co., GRAY SANDY Co., GRAY SANDY Co., RED- DISH Co., BLACK OK RED GRAY SANDY UP- DARK GRAY SANDY p< LOAM UP- SOIL SANDY PRAI- LOAM SOIL LAND SOIL RIE No. of Years in Trial 4 3 4 3 3 4 3 SEED COTTON PER ACRE, POUNDS I Nitrogen (14 Ib.) l . . $2.50 649 442 647 888 464 635 468 2 Phosphorus (16 Ib.) . 1.68 678 571 578 789 484 587 625 3 Unfertilized .... 374 320 483 676 401 433 481 4 Potassium (20 Ib.) . . 1.50 796 452 646 700 576 610 634 5 Nitrogen, phosphorus . 4.18 1295 721 657 829 5°3 651 728 6 Nitrogen, potassium . 4.00 886 576 618 801 628 663 595 7 Phosphorus, potassium 3.18 695 614 570 716 644 626 610 Q T TVi ( o .-t ,' 1 1 1 nA ->ST ii — f\*ir\ •?QO 2t8 Coo o unieniiizeci .... 3s 7 291 437 079 3s2 35° 5-" () NPK 5.68 OI7 683 778 880 740 742 925 y TO NPK m . A Q3 V / 80 1 ****o 7IO / n* 718 808 / T1^ 689 / i 7QO " *J 893 t-yo / ^ / *-^ ^y / y :7\J INCREASE OF SEED COTTON PER ACRE, POUNDS T N $2. so 27C 122 161 212 62 2O2 ~I^ 2 P . . . i 68 1OA 2CT O/l I1 1 82 1^4 I41! t K I /i 87 D- 11 7 OS 7 3Q 6 NK 4 oo 16 14 L6-^^ 8 76 7 63 8 80 2 88 7 PK 2. 18 9OC 10 16 .^u •yu /•uo 8 24 8 09 3 08 0 NPK . . c 68 •yu T> in 12 89 TO NPK (i) . . . 402 **-5 / 8 98 9 81 12 84 ii 86 1^"*3 1 The nitrogen is regularly applied in 200 Ib. of cotton-seed meal (at $25 a ton), which also contains about 2 Ib. of phosphorus and 3 Ib. of potassium. The phos- phorus is applied in 240 Ib. of acid phosphate (at $14 a ton) and the potassium in 200 Ib. of kainit (only 100 Ib. on plot 10), costing $15 a ton. These are the prices reported in Alabama Bulletin 145, February, 1909. FIELD EXPERIMENTS IN THE SOUTH 495 Alabama field experiments. The Alabama Agricultural Experi- ment Station has reported the results of fertilizer experiments with cotton on the common soils in several different counties. In Table 104 are given three-year or four-year averages from seven different counties. In computing the value of the increase, Direc- tor Duggar allows $14 a ton for cotton seed and 10 cents a pound for lint, the average price for the five years, 1904-1908. He also assumes that the seed cotton is one third lint, and thus counts the cotton seed at 3.8 cents a pound, or at 3.2 cents a pound for the increase, after allowing .6 cent a pound for picking and ginning. These results especially emphasize two facts : first, that the soils are very poor, and second, that the cotton crop is so valuable that even small increases in yield justify large expenditures for fertil- izers. As an average of the 48 different tests, the yield of the un- fertilized land is less than 150 pounds of cotton lint per acre. With few exceptions, every kind of fertilizer has more than paid its cost, and as a rule every addition has increased the profit per acre, the largest profit being secured from the most heavily fer- tilized land. It should be kept in mind, however, that as an average a pound of Alabama seed cotton is worth five times as much as a pound of Illinois corn. Very probably the 200 pounds of kainit have been more effective than 50 pounds of potassium chlorid would have been, because these soils are as a rule very deficient in active organic matter, and under such conditions the larger quantity of soluble salt is likely to become more effective. The average annual rainfall of Alabama is given as 51 inches. The monthly rainfall from May to September averages more than 4 inches. Of the 20 records for these months during the four years, 1905-1908, the lowest was 2.42 inches, and only three others were below 3.44 inches. The highest was 8.50 inches, with only two others above 5.51 inches. Louisiana field experiments. The Louisiana Agricultural Experi- ment Station has conducted a series of field experiments since 1889, on the experiment farm at Calhoun, in the northern part of the state, on hill land originally covered with pine trees. The soil had become much exhausted from 70 or 80 years of previous cotton culture. The field consists essentially of six one-acre plots arranged in 496 INVESTIGATION BY CULTURE EXPERIMENTS three series of two plots each, one unfertilized and the other fer- tilized chiefly with compost made as described below. A three- year rotation has been practiced as follows: First year Cotton. Second year Com and cowpeas. Third year Oats followed by cowpeas. By having three series, each crop may be represented every year. For cotton, 30 bushels per acre are applied of a compost made by mixing 2 tons of acid phosphate with 100 bushels each of stable manure and cotton seed. For corn, 30 bushels per acre are used of a compost made with one ton of acid phosphate mixed with 100 bushels of stable manure and 100 bushels of cotton seed. After preparing the compost, it is allowed to ferment for two or three weeks, then thoroughly mixed, and after standing a few days longer is ready for use. The oats are fertilized with 200 pounds of cotton-seed meal and 100 pounds of acid phosphate per acre, and the cowpeas are also fertilized by applying 50 pounds of acid phosphate and 50 pounds of kainit per acre. The following average results are reported by Director Dodson in Louisiana Bulletin in, September, 1908: TABLE 104.1. LOUISIANA FIELD EXPERIMENTS AT CALHOUN: YIELDS PER ACRE: FROM 19 YEARS' RECORDS SERIES SEED COTTON (Lb.) CORN (Bu.) OATS (Bu.) Unfertil- ized Fertilized Unfertil- ized Fertilized Unfertil- ized Fertilized A .... 459 5°7 432 I5SS 1811 H75 9-7 8.9 9.6 3°-4 3°-5 33-5 22.1 12.4 14.9 40-3 32.2 44.1 B C . . . Average . . . 466 I5M 9-4 3J-4 16.4 41.8 Increase 1048 Ib. $39-82 5-5° 22.0 bu'. $11.00 6.00 25.4 bu. $11.43 2-95 Value 1 of increase . . . Cost of fertilizer .... 1 Computed at Louisiana prices, 3.8 cents a pound for seed-cotton, 50 cents a bushel for corn, and 45 cents a bushel for oats. FIELD EXPERIMENTS IN THE SOUTH 497 The cost of fertilizer is given as estimated by Director Dodson. No report is made of the yield of cowpeas. In 1889 the increases produced by the fertilizing were only 301 pounds of seed cotton, 4.7 bushels of corn, and 4.8 bushels of oats; but in the second year the increases were 1227 pounds of seed cot- " ton, 19.7 bushels of corn, and 25.3 bushels of oats, which are prac- tically as great as the averages for the entire period. The 1514 pounds of seed cotton would yield about 1000 pounds of cotton seed, or about 30 bushels, which would not be sufficient to make the compost for one acre of cotton and one acre of corn, counting the shrinkage in volume during the three or four weeks allowed for fermentation; and, besides the whole seed used in the compost, 200 pounds of cotton-seed meal are used for the oats. On the other hand, the corn, oats, and cowpea crops produced on the fertilized land would certainly make much more manure than was used in these experiments, so that, with little modification, this system could be made independent and permanent as well as more profitable. The following significant statements are made by Professor Dodson : "When we sell cotton lint, we sell cellulose, composed of hydrogen, oxygen, and carbon (CgHujOs), which was derived from the air and water. When we sell our seed, we sell the fertility of the land, as the Northern and Western farmer does when he sells his grain. The oil, however, has no fertilizing value, being, like the lint, composed of elements taken from the air and water, and cannot be used again by the cotton plant ; so if we sell only the lint and the oil, return- ing the hulls and the meal to the land, we have not reduced the fertility ap- preciably." With liberal applications of ground limestone where needed, and large use of the most suitable legume crops turned under, either in farm manure or in green manures, including not only cowpeas, but also red clover, alsike clover, crimson clover, Japan clover (Lespedeza), vetch, velvet beans, and even alfalfa under proper conditions, and with plenty of phosphorus, either as acid phosphate, steamed bone meal, or fine-ground raw rock phosphate, it seems very certain that the cotton and grain crops of the South could be increased even much above the yields maintained for 20 years in these valuable experiments by the Louisiana Station. 498 INVESTIGATION BY CULTURE EXPERIMENTS It is highly probable that a liberal use of kainit would also be profitable fora time in getting such systems under way on the more depleted soils. It must be kept in mind that crops must be grown before either farm manure or green manure can be plowed under. NOTE. On the Coastal Plains, especially from North Carolina to Florida, are some extensive areas of very sandy soils. For truck farming these be- come very productive where heavily fertilized, but they are commonly too poor to be used profitably for general farming. Thus, Bulletin 68 of the Florida Agricultural Experiment Station contains 40 chemical analyses of the ordinary very sandy loams upon which nearly all of the pineapples pro- duced in that state are grown, and in commenting upon these soils the authors say, " Few of the soils would be able to produce more than two or three crops of pineapples if all the plant food present were available." CHAPTER XXIV MINNESOTA SOIL INVESTIGATIONS BECAUSE they have been so widely quoted in the agricultural press of the central West, and even in text-books on soils and fer- tilizers, it seems especially important to give in some detail the re- sults of field and laboratory experiments conducted by the Minne- sota Agricultural Experiment Station since 1892. TABLE 105. MINNESOTA SOIL INVESTIGATIONS (a) Crop Yields per Acre in Bushels or Tons YEAR PLOT No. i WITHOUT MANURE, WHEAT GROWN CON- PLOT No. 2 8 LOADS1 OF MANURE IN FIVE-YEAR ROTATION PLOT No. 3 8 LOADS x OF MANURE IN FOUR-YEAR ROTATION TINUOUSLY 1893 12.3 Wheat . . 13.7 bu. Oats . 41.6 bu. 1894 8.9 Clover . . 2.16 tons Clover . 1.18 tons 1895 17-3 . Wheat . . 22.0 bu. Barley . . 42.5 bu. 1896 14.1 v. 14.7 Oats . . 31.4 bu. Corn . . . 66.7 bu. 1897 IO.2 Wheat . . 14.2 bu. Corn . . • 33-7 bu. 1898 25.2 Clover . . 1.41 tons Oats . . 76.4 bu. 1899 I7.6] Wheat . . 19.5 bu. Clover . 1.86 tons 1900 18.8 Wheat . . 24.4 bu. Barley . . 28.3 bu. 1901 16.2 . Oats . . 58.7 bu. Corn . . . 40.6 bu. 1902 18.3 Av. 17.2 Corn . . (?) Oats . . . 80.0 bu. 1903 18.6 Wheat . . 30.0 bu. Clover . 4.70 tons 1904 13-7 Clover . 3.98 tons Barley . 40.0 bu. Total clover in 12 years Clover . . 7-55 tons Clover . 7. 74 tons (b) Nitrogen 2 in Soil per Acre 9 Inches Deep YEARS TOTAL (Lb. ) Loss (Lb.) TOTAL (Lb.) Loss (Lb.) TOTAL (Lb.) Loss (Lb.) 1892 1896 1900 1904 5400 47T5 4230 3955 5400 5645 4840 4690 5200 53°° 4870 5480 685 485 275 245 (gain) 805 (loss) 150 (loss) 160 (gain) 490 (loss) 615 (gain) Total loss in 8 years, 1896 to 1904 . . . 760 955 (loss) 125 (gain) 1 Evidently these loads of manure weighed 1200 Ib. each, making 9600 lb., or less than 5 tons, per acre. — C. G. H. 2 Based upon the statement that .221 per cent of nitrogen is equivalent to 5400 lb. of nitrogen per acre. 499 500 INVESTIGATION BY CULTURE EXPERIMENTS Probably no agricultural investigations have ever been reported which have brought forth more error and confusion in the public mind than these experiments. While they are carried on in part to determine the effect upon wheat yields of continuous wheat culture upon the same land, the information secured only shows that some factor or factors, other than the continuous growing of wheat, have thus far exerted pre- dominating influence upon the production of wheat. The figures for nitrogen given in Table 105 are based upon the percentages reported from time to time by Professor Harry Snyder in Minnesota Bulletins 53, 70, and 89, and upon his later statement that all samples have been taken to a depth of 9 inches. Thus, in Minnesota Bulletin 53, June, 1897, we read: "Plots i, 2, and 3 were 4 rods by 5 rods." "On plot No. i, wheat was grown continuously. On plot No. 2, wheat was grown in 1893, and clover was sown with the wheat; a crop of clover was har- vested in 1894. In the fall of 1894 the clover sod was plowed under, and the next year a crop of wheat was grown, and in 1896 a crop of oats. It is the plan to apply manure at this point and produce a crop of corn, and to follow the corn with wheat and clover, the complete rotation being: (i) wheat and clover, (2) clover, (3), wheat, (4) oats, (5) corn and manure. "Plot No. 3. After the wheat crop in 1892, oats were grown, and clover was seeded with the oats, and in 1894 a crop of clover was harvested. The clover sod was fall-plowed and the next year barley was grown. After the barley crop the plot received 1200 pounds of manure, and the next year was seeded to corn, the complete rotation being: (i) oats and clover, (2) clover, (3) barley, (4) corn and manure." "In plots Nos. i and 2 there was originally present in the soil .221 per cent of nitrogen, equivalent to 5400 pounds of nitrogen per acre to a depth of 9 inches. After four years' continuous cropping of wheat, plot No. i yielded .193 percent of nitrogen, a loss of .028 per cent, equivalent to an annual loss of 171 pounds of nitrogen per acre." "In plot No. 2, where clover has been grown in a rotation, there has been a gain of nitrogen. At the end of the rotation there was .231 per cent nitrogen present in the soil. On this plot clover was grown, and the second growth of clover was plowed under for green manure. The total nitrogen removed in the crops amounted to 178 pounds. Notwithstanding the fact that larger crops have been grown on this plot than on No. i, there has been a gain of 245 pounds of nitrogen in the four years' rotation, in addition to the nitrogen removed in the crops." "The soil (of plot 3) originally contained .211 per cent of nitrogen. At the close of the rotation it contained .218 per cent of nitrogen. The amount of MINNESOTA SOIL INVESTIGATIONS 501 nitrogen removed in the crops during the four years amounted to 204 pounds. The gain in nitrogen has been at the rate of about 40 pounds per acre." Four years later, in Minnesota Bulletin 70, May, 1901, we find the following statements: "Plots Nos. i and 2 contained, at the beginning of the experiments in 1892, .221 per cent of nitrogen, while plots Nos. 3, 4, 5, and 6 contained .211 per cent. It is estimated that an acre of the soil of plots Nos. i and 2, to a depth of 9 inches, would contain approximately 7700 pounds of nitrogen, while the remaining plots would contain approximately 7400 pounds. At the end of the first four years of continuous wheat cultivation, plot No. i contained .193 per cent of nitrogen ; a loss of .028 per cent, equivalent to an annual loss of 1 71 pounds of nitrogen per acre. At the end of the second period of four years, the soil contained .173 per cent of nitrogen. "At the beginning of the experiment in 1892, plot No. 2 contained .221 per cent of nitrogen. At the end of eight years, after the removal of five crops of wheat, two of clover and one of oats, or six grain crops and two clover crops, the soil contained .198 per cent of nitrogen." "On plot No. 3, oats, clover, barley, and corn have been grown. The soil of this plot originally contained .211 per cent of nitrogen. At the end of eight years the soil contained .198 per cent of nitrogen." (See pages 254-256 in Min- nesota Bulletin 70.) In Minnesota Bull^in 89 (January, 1905) we find the following statements : "While 21.7 percent of the soil nitrogen was lost during the first eight years of continuous wheat culture, only 5.71 per cent was lost during the four years following." "On plot No. 2 a rotation consisting of wheat, clover, wheat, oats, and corn and manure has been followed, with some modifications because of climatic conditions. The soil of this plot contained originally about the same amount of nitrogen as plot No. i, namely, 7700 pounds per acre to a depth of one foot.1 At the end of twelve years the soil contained 6725 pounds." "The soil of plot number three originally contained about 7400 pounds per acre of nitrogen. At the close of the first period of four years, the soil showed a slight gain in nitrogen, and at the end of eight years, a slight loss. During 1 On page 380! Minnesota Bulletin 102 (September, 1907) a correction note states that this should read: "to the depth of three fourths of one foot, " and consequently it must be assumed that the "7700 pounds" should read "5400 pounds" (less than 75 per cent), and that corresponding corrections should be made throughout. According to the data (5400 pounds for .221 per cent) the soil of an acre to a depth of 9 inches would amount to about 2,450,000 pounds, which agrees with Professor Snyder's statement that the soil weighed about 75 pounds per cubic foot (page 9, Minnesota Bulletin 53). — C. G. H. 502 INVESTIGATION BY CULTURE EXPERIMENTS the third four-year period also there was a slight gain of nitrogen, and at the end of twelve years, the soil contained about 7800 pounds per acre, showing that where clover was grown once in four years in a rotation with grains, and one dressing of farm manure was applied to the corn at the rate of eight loads per acre, the nitrogen content of the soil has been maintained unimpaired." (See pages 193 to 195 in Minnesota Bulletin 89.) In his excellent and widely used text and reference book on " Fertilizers," Doctor Voorhees, Director of the New Jersey Ag- ricultural Experiment Station, makes the following statements: "Another source of natural loss of nitrogen is its escape from the soil as gas into the atmosphere. This is due to the oxidation of the vegetable matter, or to 'denitrification,' which takes place very rapidly where soils rich in vegetable matter are improperly managed. The possibilities of loss in this direction are strongly shown by investigations carried out at the Minnesota Experiment Station on ' the loss of nitrogen by continuous wheat raising' (Minnesota Bulle- tin 53). The results of these studies show that the total loss of nitrogen an- nually was far greater thaa the loss due to cropping. In other words, by the system of continuous cropping, which is universally observed in the great wheat fields in the Northwest, there was but 24.5 pounds of nitrogen removed in the crop harvested, while the total loss per acre was 171 pounds, or an excess of 146 pounds, a large part of which loss was certainly due to the rapid using up of the vegetable matter by this improvident method of practice. Whereas, on the other hand, when wheat was grown in a rotation with clover, the gain in soil nitrogen far exceeded that lost or carried away by the crop." These statements faithfully represent the teaching of Minnesota Bulletin 53, except as to the manner in which the nitrogen escapes. With the more recent accumulated information concerning soil bacteria, to which Doctors Voorhees and Lipman of the New Jersey Station have largely contributed, a revision of Voorhees' " Fer- tilizers" probably will not ascribe any large part of the loss to denitrification. In his own text-book on " Soils and Fertilizers," published in June, 1905, Professor Snyder makes the following statements (page 112): "A rotation of wheat, clover, wheat, oats, and corn with manure will leave the soil at the end of the period of rotation in better condition as regards nitrogen than at the beginning. These facts are illustrated in the following table : 1 1 Minnesota Agricultural Experiment Station Bulletin No. 53. MINNESOTA SOIL INVESTIGATIONS 503 "CONTINUOUS WHEAT CULTURE Nitrogen in soil at beginning of experiment 0.221 per cent Nitrogen at end of 5 years' continuous wheat cultivation . . 0.193 per cent Loss per annum per acre (in crop 24.5, soil 146.5) .... 171 pounds "ROTATION OF CROPS Nitrogen in soil at beginning of rotation 0.221 per cent Nitrogen at close of rotation 0.231 per cent Gain to soil per annum per acre 61 pounds Nitrogen removed in crops per annum 44 pounds "It is to be regretted that in the cultivation of large areas of land to staple crops, as wheat, corn, and cotton, the methods of cultivation followed are such as to decrease the nitrogen content and crop-producing power of the soil when this could be prevented." Unquestionably the greatest practical problem that confronts the average American farmer is to maintain the humus and nitro- gen content of the soil,1 and the author cannot be true to the stu- dent and neglect to present the determined facts in a matter of so vital consequence to American agriculture. It will be noted that the data just quoted relate only to the first four years (not five years or twelve years) of these Minnesota experiments, where no manure had been used. As a matter of fact, the published bulletins show that wheat (not corn) was grown on plot 2 the fifth year. The subsequent data show, however, that during the second four years (presumably with manure added) there was a loss of nitrogen from plot 2 amounting to .033 per cent (.231-. 198), or about 800 pounds per acre (counting only 2,450,000 pounds of soil for a depth of 9 inches) , and during the same four years the data for the other rotation, with manure applied, show a loss of 490 pounds of nitro- gen per acre from plot 3. The only point the author would emphasize is that these Min- nesota investigations have not yet furnished sufficient data to 1 It is a very simple matter to maintain or materially increase the phosphorus content. One ton of raw rock phosphate, costing from $7 to $10 (depending on distance of shipping), and containing, say, 250 pounds of phosphorus, will supply more of that element to an acre of land than would be removed in 12 years if the average crops were 100 bushels of corn (grain only removed), 100 bushels of oats, 50 bushels of wheat, and 4 tons of clover. 504 INVESTIGATION BY CULTURE EXPERIMENTS determine the conditions under which the supply of nitrogen will be maintained. Of course it requires no new investigations to show that sufficiently large applications of manure will maintain the supply of nitrogen, whether the crops are rotated or grown continu- ously, as at Rothamsted, with wheat, barley, or mangels. NOTE. In passing from this extended consideration of the field experi- ments conducted in various parts of the United States, the reader will per- haps be interested to note the following correspondence in relation to the application of science to practical farming: " OILMAN, ILLINOIS, November 23, 1909. " DEAR DOCTOR HOPKINS : — Am sending you a few comparative figures, which I trust may interest you. I have no doubt you can see more in them than I can, but I see much that gives encouragement for the future : "COMPARATIVE YIELDS OF CORN FROM TREATED AND UNTREATED LAND: 1909 CROP (Bushels per Acre) Corn on clover sod; land cultivated 30 years, with no manure and no pasture: Untreated Same kind of land: Treated with | ton raw rock phosphate Same : Treated with 4 ton phosphate and 3 tons limestone Second-year corn after clover : Untreated Same : Treated with J ton per acre of phosphate .... 65.1 bushels 81.9 bushels 84.1 bushels 70.0 bushels 77.6 bushels " On the clover sod there seems to be about a normal difference in yield. Much of the last field, including the check strip, has had two lo-ton appli- cations of manure in 6 years. " Kind regards, (Signed) " F. I. MANN." "UNIVERSITY OF ILLINOIS, URBANA, December i, 1909. " MR. F. I. MANN, Oilman, Illinois. "DEAR SIR: — I thank you for your letter of November 23, giving the 1909 results on your 200 acres of corn from the methods of soil improvement which you have been practicing for several years. I note that the cumula- tive effect of the system is apparently becoming evident. Where phosphorus produces a ton more clover per acre (as you reported last year), the increased clover and added phosphorus must increase the following corn crop. " Two of our old plots here at the University yielded exactly the same (64 bushels) as an average of three corn crops (1895-1897) before we began applying limestone and phosphorus to one of them. This year the untreated clover sod produced 32.8 bushels, and the treated land yielded 77.6 bushels, per acre. Where limestone without phosphorus was applied, the yield was 38 bushels, and, with limestone, phosphorus, and potassium, 83.7 bushels. "Very truly yours, (Signed) "CYRIL G. HOPKINS." CHAPTER XXV CANADIAN FIELD EXPERIMENTS THE government of Canada established an agricultural experi- ment station (Dominion Experimental Farms) in 1886, and a series of field experiments were begun by Director Saunders in 1887, which have been continued under his direction for more than 20 years. The following quotations taken from the Annual Report for 1897 gives general information concerning these experiments: "A piece of sandy loam, more or less mixed with clay, which was originally covered with heavy timber, chiefly white pine, was chosen for these tests. The timber was cut many years ago, and among the stumps still remaining when the land was purchased there had sprung up a thick second growth of trees, chiefly poplar, birch, and maple, few of which exceeded six inches in diameter at the base. Early in 1887 this land was cleared by rooting up the young trees and stumps and burning them in piles on the ground from which they were taken, the ashes being afterwards distributed over the soil as evenly as possible, and the land plowed and thoroughly harrowed. Later in the season it was again plowed and harrowed, and most of it got into fair condition for cropping." "The plots laid out for the experimental work with fertilizers were one tenth of an acre each, 21 of which were devoted to experiments with wheat, 21 to barley, 21 to oats, 21 to Indian corn or maize, and 21 to experiments with tur- nips and mangels. Owing to the difficulty and unavoidable delay attending the draining of some wet places, it was not practicable to undertake work on all the plots the first season. The tests were begun in 1888 with 20 plots of wheat and 1 6 of Indian corn; and in 1889 all the series were completed excepting six plots of roots, Nos. 16 to 21 inclusive, which were available for the work in 1890. In all cases the plots in each series have been sown on the same day." " In 1890 it was found that all the grain plots had become so weedy that the growth of the crops was much interfered with, and with the view of cleaning the land, one half of each of the wheat and oat plots was sown with carrots in 1891, and one half of each of the barley plots with sugar beets. In 1892 the other half of each plot in each of these series was sown with carrots. In 1893 it was thought desirable to continue this cleaning process, and carrots were again sown, on the half of the wheat and oat plots occupied with this crop in 1891, and also on the half of the barley plots cropped with sugar beets that year. In S°S 5o6 INVESTIGATION BY CULTURE EXPERIMENTS 1894, 1895, 1896, and 1897 the one half of the oat plots were sown again with carrots and the half of the plots devoted to wheat and barley were planted with potatoes." Other changes from the original plans, and also some general conclusions drawn by Doctor Saunders at the end of 20 years, are given in the following statements quoted from the Report for the year ending March 31, 1908. "These trials have shown that barnyard manure can be most economically used in the fresh or unrotted condition ; that fresh manure is equal, ton for ton, in crop-producing power to rotted manure, which, other experiments have shown, loses during the process of rotting about 60 per cent of its weight. In view of the vast importance of making the best possible use of barnyard manure, it is difficult to estimate the value of this one item of information. "When these experiments were planned, the opinion was very generally held that untreated mineral phosphate, if very finely ground, was a valuable fertilizer, which gradually gave up its phosphoric acid for the promotion of plant growth. Ten years' experience have shown that mineral phosphate, untreated, is prac- tically of no value as a fertilizer. " Sulfate of iron, which, at the time these tests were begun, was highly recom- mended as a means of producing increased crops, has also proven to be of very little value for this purpose. " Common salt, which has long had a reputation with many farmers for its value as a fertilizer for barley, while others disbelieved in its efficacy, has been shown to be a valuable agent for producing an increased crop of that grain, while it is of much less use when applied to crops of spring wheat or oats. Land-plaster or gypsum has also proved to be of some value as a fertilizer for barley, while of very little service for wheat or oats. Some light has also been thrown on the relative usefulness of single and combined fertilizers. "After ten years' experience had demonstrated that finely ground, untreated mineral phosphate was of no value as a fertilizer, its use was discontinued in 1898. Prior to this it had been used in each set of plots in Nos. 4, 5, 6, 7, and 8, in all the different series of plots, excepting roots. In 1898 and 1899, similar weights of the Thomas phosphate were used in place of the mineral phosphate, excepting in plot 6 in each series. In this plot the Thomas phosphate was used in 1898 only. "After constant cropping for ten or eleven years, it was found that the soil on these plots to which no barnyard manure had been applied was much depleted of humus, and hence its power for holding moisture had been lessened, and the conditions for plant growth, apart from the question of plant food, had on this account become less favorable. In 1899 the experiments were modified and an effort made to restore some proportion of the humus and at the same time gain further information as to the value of clover as a collector of plant food. In the spring of that year ten pounds of red clover seed per acre was sown with the CANADIAN FIELD EXPERIMENTS 507 grain on all the plots of wheat, barley, and oats. The young clover plants made rapid growth, and by the middle of October there was a thick mat of foliage varying in height and density on the different plots, which was plowed under. No barnyard manure was applied on plots i and 2 in each series from 1898 to 1905. "In 1900 all the fertilizers on all the plots were discontinued, and from then to 1905 the same crops were grown on all these plots from year to year without fertilizers, sowing clover with the grain each season. In this way some infor- mation has been gained as to the value of clover as a collector of plant food, and also as to the unexhausted values of the different fertilizers which had been used on these plots since the experiments were begun. In 1905-6-7 all the fertilizers were again used as in 1898." The corn plots and root plots were fertilized somewhat differ- ently from the others, and the corn was cut green and weighed in the fresh condition. The results with wheat, oats, and barley are of more general interest, and the most significant data from these crops are recorded in Table 106, in which all dated intervals are inclusive. In the author's opinion, we must question the conclusion of Doc- tor Saunders that nonacidulated mineral phosphate is of no value as a fertilizer. There are at least two important points to be con- sidered before drawing any final conclusion : First, does the land need phosphorus? Or, in other words, is phosphorus the limiting factor? Sandy loam soils are more likely to be deficient in either nitrogen or potassium than in phosphorus.1 Second, was any adequate means provided in the system of farming for liberating the phosphorus from the raw phosphate? From Table 106 we see that the raw phosphate used alone pro- duced practically no increase on wheat, oats, or barley, but this is also true as regards fine-ground bone during the first ten years, 1 Since the above was written, Professor Frank T. Shutt, Chief Chemist of the Dominion Experimental Farms, has kindly furnished the author unpublished analytical data from samples of soil collected in 1898, which show that 2 million pounds of surface soil contained, for plot 3 in the oats series, 2130 pounds of nitro- gen, 1950 pounds of acid-soluble phosphorus, and 3160 pounds of acid-soluble potassium ; while the corresponding figures for plot 3 of the barley series were 2600, 1850, and 2990, and for the wheat series, 2120, 1470, and 3240 pounds. Compare the following significant figures : NITROGEN PHOSPHORUS POTASSIUM Oats 9700 1600 6800 Soil 2130 1950 3160 508 INVESTIGATION BY CULTURE EXPERIMENTS 5< ^ t^ 01 toNO 00 Ol IO Tf ol *t t^ l-i 00 to 01 »O Ol H Ol ON Tt- 01 3 pq §, Tt-OO ON to to O 1 1°"" *O ^t" M to fO to Ol 1° J^J^ """ OO n t^- ON t^ M tO 1H M M 1 ON tor^ 00 00 NO M ON 01 to •* >o to tooo « oo 11 Ol to »O Ol to to fO fO fO NO tO Tf M to to to to V 1 o. 01 M NO M 01 to to to 01 01 to « 01 01 M Ol w 01 01 Ol 01 to *J- 01 t- 01 ^ oo »o r- r^ tooo 10 IO Ol fO t^ tO 't fO 01 IO 3 w o. M t^ tO Tj- to ts IO •* « O « 01 01 to to to NO 00 ON 01 01 01 01 01 fONO n Ol O 01 Ol Ol Ol 01 n 01 <0 r)- Ol Ol 01 01 tn 00 1^ ONOO NO NO ON to ^" OO fj- Ol NO «^> O OO NO OO O- t^OO 00 t^ W J rt 00 « 0. O O NO ON NO -3- IO 00 M Ol fO to Ol ^- 01 01 NO (N to to n O OO 01 •<* M oi 01 oi n 01 11 OO M N 01 01 0< Ol « < O 1^ to 4 4 to to M 10 f-~ 01 o r^ ^- ON OO tO O* ^ NO NO tO 11 11 00 IO •< 4 oi oi oi oo' w 01 01 01 « 30 00 O M 8, M 01 "I" NO 00 O tooO Ol NO CO ol 3 g -8 M to 01 NO IO *••* IOOO rfOO t^ 1-00 _ .a §- O to 10 IO IO to ON O 01 O NO n ON to to fO to to M 01 00 t^ ON tO to to to to ^t ON 0) O to ro -^- -^- -^~ >^- •* 10 -"t Ol OO ON »O IO fO t^ O O oo NO oo «^ " O ~S1 'JZ 'S3 ,1 ? {0^4 M 00 O to OO 10 10 NO 10 10 O ^00 tO IO >O to 01 >O O to ^ •* Tf IO Tf IOIO f^ fONO n Tt 10 't 10 kl f? t— ^t M O ^ O 00 ON r^oo ON t^ NO o •* t^ o O 11 t^ t^- M H Oi O to r^ t^OO O to IO 10 O 'J-NO GO ON t^ ^f W « S-, 1- t^OO NO f-~ 11 OO *i to to t 11 NO t^ to1* ONO O 01 10 •* M • ffj OO 1^00 3. °p £ £ £ W 00 1 Tt •*•*•* 01 NO fO t^ IO ON tO t- ON w 0 1 00 O O> 00—00 11 M NO to 11 Tf IO tO O 00 to ^f O IO 1ONO Ol to i^ 10 10 to rO rf~< ~*~ rO -t NO »O fO to ro O to •H O 10 01 "as °> r, fCoO '0 J §> *O NO' to 10 to »t to f «^o * fo to 10 oi 01 H 0, oi « 4 to oq oq oi TJ- NO to to 01 ON 01 00 NO Ol 00 00 00 00 H cd 01 01 M B oo r~ 0 M NO NO t^ 0\ 00 0- 00 11 00 O V3 rO I ~ 0 0 OO to O NO ^ < « °« O O O 0 0) t- « 0 11 fO to O M IO to >O 01 oi to to 01 . I a 2 2 & Ji ij ' 0 ' (U ' 2-^ ° ' . w . •" . """a Sj£ §i : a 0 0 < C3T ••o J. fl 1 ' crt t/j t^ „ ra n 22.2 -u 8 . >o 1 !f a> . 2 «" 1 1 1 REATMENT APPLIED tVERY (Except 1900 to 1904 (and no Trea Grain each Spring from 1899 to igc under in the Fall.) Mixed rotted yard manure, 15 ton Mixed fresh manure, 15 tons . Unfertilized Raw phosphate to 1897; slag since Sodium nitrate, 200 Ib., with raw 1897 and with slag since, 500 Ib. Rotting manure, 6 tons, with raw 1897 and with slag since, 500 Ib. Nitrate, 200 Ib., and ashes, 1000 phosphate to 1897 and with slag s Unleached wood ashes, 1500 Ib., w phate to 1897 and with slag since •If- • o « . ^:2 2 £%'&> • • o • » > •"o . ••& -2 o Xl"1 — ' 10 . " O 811 n O O .i: 10 to x tsS "> ^ t d O (1 § g'5 S 3 1 ^S-a'l i g".2 | S Iron sulfate, 60 Ib Sodium chlorid, 300 Ib. . . . Calcium sulfate, 300 Ib. . Acid phosphate, No. 2, 500 Ib. . S d ^ M 01 tO *~> NO ^ oo ON O M 01 tO M- >ONO t^ 00 ON O " CANADIAN FIELD EXPERIMENTS 509 plot 13 yielding only 1.2 Bushels more wheat, 2.3 bushels more oats, and .4 bushel less barley than plot 4. The 300 pounds of cal- cium sulfate on plot 20 produced .4 bushel more wheat, the same yield of oats, and .9 bushel less barley than the 500 pounds of acid phosphate on plot 21. By referring to Table 78, we find that as an average of six 4-year periods, 640 pounds of calcium sulfate pro- duced practically no effect (12^ cents per acre in four years), while dissolved bone black carrying 42 pounds of phosphorus produced an average increase of $i 2. 1 7. These results certainly indicate that phosphorus is not the limiting factor in crop yields on the Ottawa soil. The effect produced on oats by the fine-ground bone is probably due to the nitrogen contained in the bone. It is a common ob- servation that oats respond to nitrogen more rapidly than most other crops on the same soil, and it will be observed that sodium nitrate alone (plot 15) produced practically the same effect on oats as the nitrate and acid phosphate combined. Where nitrogen was provided, the raw phosphate (plots 5 and 7) produced a larger average increase on oats than did the acid phosphate (plots 10 and n), during the first nine years. A study of the results with wheat and barley indicate that potas- sium is the first limiting element for those crops. As an average of the first ten years, the largest yield of wheat was produced by po- tassium chlorid, aside from the farm-manure plots; and the second largest yield was with ashes (plot 14; compare with 13). Sodium chlorid also produced some increase in the yield of wheat, and with barley the 300 pounds of sodium chlorid produced the largest yield, aside from the two heavily manured plots. Even acid phosphate, containing much calcium sulfate and an acid salt of phosphorus, may liberate some potassium. It may be questioned whether potas- sium or nitrogen is most limiting for the barley crop, but it is plain that phosphorus is not the limiting element. Even during the sec- ond ten years, the 150 pounds of potassium chlorid or the 300 pounds of sodium chlorid rank higher than 500 pounds of bone or 500 pounds of acid phosphate, in either trial (plots 9 and 21), and also far above the slag phosphate (plot 4) . On plot 6 the raw phosphate was applied in connection with " actively fermenting " manure, and it may have produced some INVESTIGATION BY CULTURE EXPERIMENTS effect, but this cannot be known, because there is no comparison plot on which the same amount of untreated manure was used. If we average plots i and 2, we find that 6 tons of phosphated manure on plot 6 produced more than 70 per cent as much in- crease as 15 tons on plots i and 2. Comparison with the Penn- sylvania and Ohio experiments does not help much in trying to decide if the raw phosphate was effective when mixed with the manure at Ottawa, in part because the applications are propor- tionately different, and in part because the manure itself produces different effects on different soils. With the data presented, any possible comparisons can be made by the student. Except for the oat crops which, as stated, usually respond readily to nitrogen, ammonium sulfate with the 60 pounds of nitrogen produced a smaller average effect than the sodium nitrate with 30 pounds of nitrogen, thus indicating that the sodium in the nitrate exerted appreciable influence. Apparently, iron sulfate produced some small effect, but it is doubtful if it is greater than would have been produced by 60 pounds of sodium chlorid. For this and other comparisons in- volving few plots, the author calls attention to the fact that there are some marked natural variations among the individual plots in these series; and in such cases no final conclusions can be drawn. In the oat series, plot 3 produced 9 bushels more oats per acre than plot 12 as an average for the first nine years, and 14 bushels more for the next ten years. Plot u in the oat series is evidently a plot which yields below normal. The quotations from Doctor Saunders show that some parts of the field were naturally wetter than others, and on page 51 of the Annual Report for 1897 the statement is made that " plots 12, 13, and 14 were on a piece of rising ground on light soil." The records for 1903 and 1904 give the yields for the last two years of green manuring with catch crops of clover, while 1906 and 1907 furnish later records after the fertilizer applications were renewed, beginning with 1905. In Table 107 are given probably the most significant data from the root crops (and potatoes) that were grown on alternating half plots in these series during the seven years, 1891 to 1897. CANADIAN FIELD EXPERIMENTS ^v « 0. OJ_i sa Os CO CM \r> ir> M O O O 10 io CO CM CO H SO "^ Os Os O so so OsOO OssO OS-CM 00 Os O HI co M *gH <« g <« H^ >>< Ti- ll 0 -> 00 £ 2 . O OsOO so t^ OO HI SO co rj- t^ io ^ Os OO so ^t t*» £ " 60 ^"cSO 00 Os O M CO t^-so CO O CM HI SO t~- co ^3" u-j O O co co H ^ H V J fca CM CO *O Os ^ O ^1" CM ^ O O ON O Os O HI CM HI SO ON SO A 00* 3O IO M ro CM' t^- r^-sd TJ- Os ^ IO ON Os *• t~» t^ Os fOsO CO cs) T O t^ OO HI O Os '3' in i~~ *>• vo < S 2 S3« 1-vO rf H OO OssO CM IO ^^ HI CO CM 00 CO t^- O OsOO H ON 00 30 •^ S oo t^-oo t^ cosO H Os HI ro coco OO O »OOO HI O CM, w O c/jCQ H U O . , fea 00 t^ rr IO t^ "^ H CM 00 (N O O Tf O ON O ^ HI H CM ^- nP "So (OSO 10 so O 00 rj- O- t^ 10 t^ t^ SO CO ON t^ CM 10 f^OO *>- S CM CM w H CM CM CM H 1 HI tn z ^ . u| CO CO O OO VO CO IO '^- CO ^t Os co ^ ^ „ ^ ^ 1000 so HI 0 H 6g *fe CS| CM CM M IT) M OO H O H u-j ro H Csl CM (N HI TJ- CM' CM so CM CM CM CM HI Os O O HI HI CM CM CM - ' 10 ' 8 ' O dT M M 0^ 1) a H JH - o ai >i ji S'aja -X2 g 3*0,1 o .'R 8 . a G 10 o ^ io ° w s IO en _ H ^£ ^ • aT«i ' ^ ^ TREATMENT AP Mixed rotted yard manure, i Mixed fresh manure, 15 ton Unfertilized — • o< g 8 o- Qn C W* CX •^ o ^ "*"^ j C/5 2 Cy 03 vQ L5 TO r-i 1-1 '^ ro |_ OH •- aT •" •- c/3 • ^ _: • -a rt 2 """ ^ ^8 | 8 8 U ^ », W O 4J W) qj" >- *3 lit • co ^*O aT dp aT j I'i'J cu a< ex '-5 'G 'u 'o c . o . • .o "" H? • ~" ° Available Total 1.64 I.OO 2.50 .82 1.64 1.64 2.47 2.47 1.64 2.OO I.2S 3.00 I.OO 2.OO 2.OO 3.00 3-00 2.0O 8.00 8.00 8.00 8.00 8.00 IO.OO IO.OO 8.00 IO.OO IO.OO 14.00 IO.OO I2.OO 11.00 I I.OO IO.OO 11.00 13.00 13.00 IO.OO I I.OO I I.OO 15.00 I I.OO 2.OO I.OO 5.00 4-00 7.OO 2.00 3.00 IO.OO 2.OO 1 Complete fertilizer ..... 1 Sugar-beet grower 1 Onion, potato, and tobacco . . . 1 Kentucky tobacco grower . . . » 1 Special tobacco fertilizer .... 1 Vegetable and tobacco grower . . . Ammoniated phosphate Special phosphorus and potash . . . Diamond "S" phosphate .... n 1 Insoluble phosphoric acid derived from animal bone. "Nitrogen (ammonia) in all the above brands derived from Blood and Bone. "Purchasers of fertilizers should profit by the fact that the higher the guaran- teed fertilizer analysis, the less the cost of the plant food obtained, allowing for equal distances of shipment. It pays, therefore, to buy fertilizers on the basis of analyses and not simply by the price per ton for some certain brand. By making a careful comparison of our analyses when buying fertilizers, 's will lead. "For the convenience of our customers in making comparisons of values of different fertilizers, we suggest the following as being approximate values : "(i) Pure Bone Fertilizers. If any one of these grades is desired, the values for the essential compound ingredients may be limited as follows : Ammonia at 15 c. per Ib $3.00 per unit Phosphoric acid at 5 c. per Jb i.oo per unit Potash, actual, at 6 c. per Ib 1.20 per unit Ammonia at 15 c. per Ib 3.00 per unit "(2) Acidulated Fertilizers, which are more complex in their nature and manufacture, may be judged comparatively by the following: MANUFACTURED COMMERCIAL FERTILIZERS 523 Phosphoric acid, available, at 7 c. per lb $1.40 per unit Phosphoric acid, insoluble from bone, at 5 c. per Ib i.oo per unit Phosphoric acid, insoluble from rock, at i c. per lb 20 per unit Potash, actual, at 6c. per lb 1.20 per unit "Multiplying the minimum guarantees with the above valuations per unit, the total is the relative value per ton of 2000 pounds." Of these fertilizers, the "Special Bone Meal," containing 12 per cent of phosphorus, is a fair grade of steamed bone meal, and the "Garden City Phosphate" containing 6.1 per cent of available phosphorus and less than | per cent of insoluble phosphorus represents the most common grade of acid phosphate on the mar- ket. At the prices given, a pound of phosphorus would cost about ii cents in steamed bone meal and about 16 cents in acid phosphate. In some cases different brands have the same composition, even where sold by the same company. In fact, some of the larger fertilizer companies sell a dozen different brands of the same com- position, so that the total number of brands sold by all companies is very large, amounting to about 900 in the state of Indiana and to more than 1800 in Georgia, which only emphasizes the fact that most farmers purchase fertilizers by name rather than on the basis of plant food. Probably half of all the fertilizers bought by American farmers have as an average the " 2-8-2 formula," as in the " Superphosphate " and " Eagle Wheat and Corn Grower." In some states, as in Illinois, a deficiency of i per cent below the miminum guarantee is " not considered evidence of fraudulent intent," but greater deficiencies subject the dealer to a severe pen- alty if prosecuted. (See Model Fertilizer Law, in the Appendix.) In most states the burden of " fertilizer inspection and control " is placed upon the agricultural experiment station, and some- times this burden has almost prevented the stations from con- ducting investigations concerning the soils of the state or in other important lines where exact information is needed. In other states, as in Pennsylvania, Ohio, and Illinois, the enforcement of fertilizer laws is placed with the State Board of Agriculture, and the ex- periment station is left free to conduct agricultural experiments and investigations. It may be added that there is grave doubt if the agricultural investigator should be compelled to depend, either in large part or in small part, upon the income from tonnage tax 524 VARIOUS FERTILITY FACTORS or brand tax of commercial fertilizers sold in his state, as a source of revenue for the support of his department of investigation. FORMS AND SOURCES OF COMMERCIAL NITROGEN Aside from the free nitrogen of the air, there are four distinct "forms "of nitrogen: (i) organic nitrogen, (2) ammonia nitrogen, (3) nitrate nitrogen, and (4) cyanamid nitrogen. Aside from farm manure and crop residues, the chief sources of organic nitrogen are (i) dried blood and tankage from the slaughter houses or stock yards, (2) cotton-seed meal from the oil refineries in the South, and (3) fish-scrap in the Eastern and extreme West- ern states. (Near the coast seaweed often becomes the staple manure. It contains about as much nitrogen and phosphorus as farm manure, and nearly five times as much potassium.) Dried blood. A good grade of dried blood contains 14 per cent of nitrogen, while tankage is of various grades, ranging almost from blood to bone. It contains much of the offal, and may include the undigested contents of the stomach and intestinal tract. One common grade is "7 and 30" tankage, meaning 7 per cent of ammonia (NH3) and 30 per cent of " bone phosphate," Ca3(PO4)2, corresponding to 6 per cent each of nitrogen and phosphorus. A mixture containing about 2 parts of tankage, 3 or 4 parts of acid phosphate, i part of kainit, and i or 2 parts of filler will produce the "2-8-2 formula," with about 2 per cent of "insoluble phosphoric acid derived from animal bone." The annual pro- duction of tankage and blood amounts to about i million tons. Dried peat. Train loads of dried peat are shipped from the peat beds of Illinois and elsewhere to the fertilizer factories for use as a filler; and as a filler it is said to be superior to all other materials, because it is a very effective absorbent and thus keeps the acidu- lated fertilizers in excellent mechanical condition. Dried peat con- tains from 3 to 4 per cent of nitrogen which may be "found" by analysis, although the nitrogen in peat is at best no more active than that in the ordinary organic matter of the soil, which usually amounts to 3000 to 5000 pounds per acre in the plowed soil, so that 50 pounds of peat in 200 pounds of "complete" fertilizer would not appreciably affect the crop. MANUFACTURED COMMERCIAL FERTILIZERS 525 Cotton seed. Cotton-seed meal contains about i£ per cent each of phosphorus and potassium. Sometimes the whole cotton seed, containing about 3 per cent of nitrogen, is used directly as a fertilizer, but as practically all of the plant food remains in the hulls and cake (after the oil is expressed) , the meal is now largely used, and more profitably, of .course, unless the farmer pays more for his nitrogen than he received for the same amount in seed, which is likely to be the case if he buys ready mixed "complete" fertilizer. The annual production of cotton seed in the United States amounts to about 6 million tons. Fish scrap. Fish-scrap meal contains about 8 per cent of nitro- gen and 6 per cent of phosphorus. There are various other sources of organic nitrogen, some of which, like hoof meal, furnish avail- able nitrogen, while others, like hair, wool waste, and horn meal, are very slowly nitrified. Ammonium sulfate. Commercial ammonium sulfate is usually about 95 per cent pure, containing 20 per cent or more of nitrogen. It is obtained by washing coal gas through dilute sulfuric acid and concentrating the liquid until the ammonium sulfate crystallizes out. About 100,000 tons of ammonium sulfate is the present annual production from the gas plants and coke ovens in the United States, and the production is likely to largely increase, because most of the American coke ovens are still wasting the ammonia produced. Sodium nitrate. Sodium nitrate of commercial grade is about 95 per cent pure, and contains 15 per cent or more of nitrogen. It is obtained from the extensive nitrate beds of Chile, where it is found in very extensive deposits, thought to have resulted from the decomposition of seaweed in connection with sea salt. The impure material is leached and the nitrate secured by crystalliza- tion. The exportation began about 1830 and has quite steadily increased from 8000 tons in 1840 to about 2 million tons in 1908, the total exportation from 1830 to 1909 amounting to about 40 million tons, which is about one sixth of the estimated * amount re- maining in the Chilian and Peruvian beds. The export duty 1 The estimates of ten years ago placed the total supply at 81 million tons, but a so-called official report made in 1909 (American Fertilizer) estimates 246 million tons. 526 VARIOUS FERTILITY FACTORS yields an annual revenue of more than $20,000,000, or about three fourths of the total income of the Chilian government. Most of the sodium nitrate imported into the United States is used for the manufacture of explosives. Calcium nitrate. The artificial fixation of atmospheric nitrogen by an economic and practical method is a problem whose solution has been given much attention for many years; in fact, it has been the dream of many a chemist and a dream which has only recently been realized. Calcium nitrate is now produced to a limited extent (chiefly at Notodden, Norway, by the aid of cheap water power) by the Birkeland-Eyde process, in which a current of air is subjected to powerful electric action, which results in the formation of nitrogen tetroxid (as observed by Priestly as early as 1775), which by somewhat complex reaction with water and oxygen yields nitric acid. This is treated with lime to form calcium nitrate, which is obtained in crystallized form, Ca(NO3)2 4 H2O, containing about 12 per cent of nitrogen. Calcium nitrate is a highly deliquescent substance, and must be shipped in air-tight containers. In an experiment at Rothamsted 10 grams of calcium nitrate (produced at Notodden in 1906) absorbed 20 per cent of water and became liquid in 3 days, and in 10 days about 50 per cent of water had been absorbed. Calcium cyanamid. Calcium cyanamid is also a product result- ing from the artificial fixation of atmospheric nitrogen, by a pro- cess recently developed by Frank and Caro of Germany. The primary materials used in the process are limestone, coke, and nitrogen gas. Calcium carbid is first produced by heating a mixture of burned lime and coke to a very high temperature pro- duced by an electric furnace: CaO + 3 C = CaC2 + CO. The calcium carbid is finely ground and then placed in closed retorts and heated to the requisite temperature in an atmosphere of nitrogen, which reacts with the calcium carbid with the forma- tion of calcium cyanamid and separation of carbon : CaC2 + 2 N = CaCN2 + C. MANUFACTURED COMMERCIAL FERTILIZERS 527 The nitrogen gas is obtained from the air, either by passing air through a hot tube containing copper turnings, which remove the oxygen by forming copper oxid (the oxid being again reduced by substituting coal gas for air), or by the Linde liquid-air process, in which advantage is taken of the difference between the boiling points of nitrogen (— 194° C.) and oxygen (— 184° C.), the nitro- gen being evaporated at the lower temperature. Potassium cyanid is a well-known substance with the formula KCN, or N=C — K, and the group or radicle, N = C — , is called cyanogen, somewhat as the group — NH4 is called ammonium, and the group — NH2 is called the amido group. Cyanamid contains the two groups, thus N = C — N=H2, and by replacing the two hydrogen atoms by one bivalent calcium atom, calcium cyanamid (N=C — N=Ca) is produced. Calcium cyanamid itself con- tains 35 per cent of nitrogen; and, if the product could be made with the one atom of free carbon as the only impurity, the nitro- gen would still reach 30 per cent, but about one third of the commercial article consists of other impurities (coal ash, lime, cal- cium carbid, sulfid, phosphid, etc.), the nitrogen being thus re- duced to about 20 per cent. An analysis of a commercial sample gave the following results: Calcium cyanamid (CaCN2) 57.0 per cent Carbon 14.0 per cent Lime (CaO) 21.0 per cent Silicon dioxid 2.5 per cent Iron oxid 4.0 per cent Calcium sulfid, phosphid, and carbonate . . . 1.5 per cent When first added to the soil the commercial calcium cyanamid with its impurities produces an injurious effect upon young plants, and to avoid this it is applied a week or two before seeding. For the same reason it cannot safely be used as a top-dressing. It has a tendency, because of its lime content, to absorb moisture and carbon dioxid from the air, and for protection is usually treated with a small amount of heavy petroleum. A ten-gram sample of calcium cyanamid exposed for 12 days at Rothamsted increased 30 per cent in weight, and some loss of ammonia occurred. 528 VARIOUS FERTILITY FACTORS Many pot-culture and field experiments have been made with calcium cyanamid which show that when properly used the nitrogen in this form has about the same value as in ammonium sulfate. When heated with water under pressure, calcium cyanamid decomposes with the formation of calcium carbonate and ammonia, as indicated by the following equation: CaCN2 + 3 H20 = CaCO3 + 2 NH3. For long-distance shipping the final product from the artificial fixation of atmospheric nitrogen is, in the opinion of the author, to be ammonium nitrate, made by using the ammonia thus pro- duced as a base for neutralizing the nitric acid obtained in the Birkeland-Eyde process. NH3+ HNO3 = NH4NO3. The ammonium nitrate thus formed, free from mineral im- purities, would contain 35 per cent of nitrogen. To produce this compound would require five plants: (i) for nitrogen gas, (2) for calcium carbid, (3) for calcium cyanamid, (4) for ammonia, and (5) for nitric acid. A publication issued June i, 1907, by the American Cyanamid Company, estimates that the original cost of a complete plant for the production of 10,000 tons per annum of calcium cyanamid would amount to $444,000, including : $155,000 for the calcium carbid plant, 70,000 for the Linde nitrogen plant, 145,000 for the calcium cyanamid plant, 74,000 for expense not itemized. This estimate of $444,000 does not include the cost of the power plant, it being assumed that a separate company will be organized to furnish the power by utilizing a natural waterfall. In connection with elaborated estimates as to the actual cost of producing cyanamid nitrogen in such a plant, the following state- ments are made in this publication: " Throughout all estimates of costs of operating it is assumed that power costs the Cyanamid Company $15 per 24-hour horse power per annum, meas- ured on the switchboard of the power company supplying the power. It is MANUFACTURED COMMERCIAL FERTILIZERS 529 assumed also that the power company and the works of the Cyanamid Com- pany are so close together that there will be no appreciable loss in electric transmission." " Thus the total estimated cost, including interest upon the estimated in- vestment of $444,000, is $45 per metric ton, equivalent to 22\ cents per kilogram of nitrogen, or 10 cents per pound." It should be noted that this 10 cents per pound for cyanamid nitrogen is the estimated cost to the manufacturers, and that it includes no allowance for transportation of the finished product from the factory to the farmer, no allowance for the cost of adver- tising and selling, and no allowance for any profit to anybody. It should be noted, too, that the 4 million pounds of combined nitrogen which such a plant could produce in one year would be sufficient, if none were lost in drainage waters, to meet the " grow- ing demands " of the average corn crop of the United States for less than 200 minutes. A calcium cyanamid factory is located at Niagara Falls. SOURCES OF COMMERCIAL POTASSIUM There are three important sources of commercial potassium: (i) the German mines, (2) the salts recovered from the evapora- tion of sea water, and (3) wood ashes. Potassium salts of Germany. The very extensive salt deposits in the region of the Harz Mountains in northern Germany con- stitute at present by far the most important source of commercial potassium. These deposits were discovered by borings made near Stassfurt in 1857, and the potassium salts are found chiefly in strata overlying the much thicker stratum of common rock salt. It is estimated that these German salt deposits cover an area of a million acres, and that the supply of potassium which they con- tain is sufficient to supply the present rate of mining for 190,000 years. It is thought that these salt and potash beds were formed in ancient geologic time by the evaporation of sea water confined in lakes somewhat like the Dead Sea, or Great Salt Lake, except that there was at times connection with the ocean which supplied the salt water. Evaporation carries off water vapor ami leaves the salts in solution, but if the evaporation proceeds far enough, the 53o VARIOUS FERTILITY FACTORS less soluble salts, such as (i) calcium sulfate (gypsum) and (2) sodium chlorid (common salt), begin to separate in crystals which settle to the bottom; and with further evaporation of water the more soluble salts of potassium and magnesium finally separate in crystals which are deposited in strata above the principal salt deposits. After vast amounts of water had been evaporated and immense quantities of salts deposited, these accumulations sometimes be- came covered with drift material (clay etc.) several feet in thick- ness, and at a later period the sea water again came in and by evaporation left a second, stratum of calcium sulfate, and above it another immense salt deposit. The total thickness of these various strata is about 5000 feet at Stassfurt. There are many variations and irregularities, but in the main the lower stratum consists largely of calcium sulfate; next above is the sodium chlorid deposit of great depth; then a layer of the mineral polyhalite, composed of the sulfates of po- tassium, calcium, and magnesium, kieserite (magnesium sulfate), and finally a stratum varying from 50 to 130 feet in thickness, which consists largely of carnallite, a double salt of potassium and magnesium chlorid. In some places the overlying clay or earth became cracked, and water entered from the surface, so that more or less of the various salts were dissolved and redeposited in veins or pockets in com- pounds or forms not commonly found in the more general strata. Thus were formed comparatively small beds of kainit, sylvanite, and hartsalz. More than thirty different compounds or minerals are found in these Stassfurt deposits, and at least a dozen of these contain more or less potassium. By far the most abundant source of potassium is the carnallite stratum, but even the pockets or beds of kainit, sylvanite, and hartsalz are of great importance. The following are commonly accepted as the formulas which represent these minerals: Carnallite, KC1 MgCl2 6 H2O. Kainit, K2SO4 MgSO4 MgCl2 6 H2O. Sylvanite, K2SO4 MgSO4 KC1 MgCl2 NaCl 6H2O. Hartsalz, KC1 MgSO4 NaCl H2O. MANUFACTURED COMMERCIAL FERTILIZERS 531 There are three principal potassium fertilizers brought to America from Germany: potassium chlorid, kainit, and potassium sulfate. The commerical kainit usually consists of two thirds of the mineral and one third sodium chlorid, and contains about 10 per cent of potassium. It is ground and used very generally for direct application. Potassium chlorid is obtained from carnallite, and potassium sulfate from kainit, by dissolving the minerals and allowing these salts to crystallize out at suitable temperatures. Commercial potassium chlorid is usually at least 80 per cent pure, while the sulfate has a purity of nearly 95 per cent. Each contains about 42 to 43 per cent of potassium. Potassium sulfate is also produced from potassium chlorid and sulfuric acid in the manufacture of hydrochloric acid, for which sodium chlorid was formerly used. Sylvanite and hartsalz (hard salt) are sometimes ground and applied in the crude state, but the concentrated salts may also be derived from them by solution and recrystallization. Potassium-magnesium sulfate, or " double manure salt," is another Stassfurt preparation which is used to some extent. It contains, as found in the market, about 20 per cent of potassium. Its special value, like that of potassium sulfate, is for use in fer- tilizing those crops whose " quality " is injured by salts containing chlorin, particularly the tobacco crop. Wood ashes. Unleached wood ashes commonly contain 5 per . cent of the element potassium (as carbonate), 50 per cent of cal- cium carbonate, and .5 per cent of phosphorus. On most soils they are likely to be more valuable for the lime than for their potassium content; and, when applied at the rate of a ton or more per acre, even the phosphorus added is more than that con- tained in 200 or 300 pounds of the common " complete " commer- cial fertilizer. Potassium from sea water. Where common salt is obtained from the evaporation of sea water, as has been done to some extent on the southern coast of France, potassium is secured as a by- product from the concentration of the " mother liquor," and one may conceive of unlimited supplies being produced in this manner where the climatic conditions and other natural advantages can be utilized, as on an arid coast and under a tropical sun, especially 532 VARIOUS FERTILITY FACTORS where tidal power could fill extensive reservoirs and where a mountain stream could serve to dissolve and remove the salt deposit after the " mother liquor " is drawn off for further con- centration. It is sometimes claimed that potassium salts, especially kainit, have some power to prevent damage from fungous diseases and injurious insects, but it may be questioned whether the effect is direct or indirect; data already given show very conclusively that not only potassium salts, but also the salts of magnesium and sodium (usually to a smaller extent) , produce marked benefit in many instances. In most cases, however, any influence which aids directly or indirectly in the proper nourishment of the plant will thus enable the plant itself better to resist attacks of insects or disease. The author has frequently observed that insect in- juries are much more apparent on corn grown on poor land than on adjoining plots treated with phosphorus in connection with farm manure or crop residues. Where the soil is markedly deficient in potassium, as compared with normal soils, as is the case with certain peaty swamp soils, and where the application of potassium salts on such soils produces marked benefit, while sodium salts produce practically no benefit (see Table 91), there can be no question regarding the need and value of potassium for its own sake; and even where the soil con- tains normal amounts of potassium, if enormous crops are to be grown that draw very heavily on potassium, as the 40 to 50 tons per acre of mangels on Barn field at Rothamsted (see Table 716), the time will come when potassium must be returned. (This, how- ever, is also the case with magnesium and calcium.) On the other hand, where the plowed soil contains sufficient total potassium to meet the draft upon it for, say, two thousand years, and where sodium or magnesium salts produce about the same effect as potassium salts, and where potassium produces little or no effect if applied in connection with liberal amounts of decaying organic matter, the conclusion may be safely drawn that the addition of commercial potassium is not essential in adopting systems of permanent agriculture, for even the slight erosion that occurs on nearly level lands will possibly provide an absolutely permanent supply. CHAPTER XXVIII CROP STIMULANTS AND PROTECTIVE AGENTS A CLEAR distinction should be made betweeen the use of plant food in systems of permanent agriculture and the use. of crop stimulants or crop " protectors." Unquestionably there are con- ditions under which the use of some particular substance, other than plant food, will produce a sufficient increase in the yield of the crop for which it is applied to more than pay the cost; and, further- more, the use of such material may in some cases be advisable, but it should be used with intelligence and full understanding of its effect. Land-plaster. Land-plaster (native calcium sulfate) is a well- known crop stimulant, but it contains neither nitrogen, phos- phorus, potassium, nor lime. Thus it supplies no plant food of value and has no power to correct soil acidity, and its physical effect on the soil is probably injurious rather than beneficial. In fact, it is the common report that the soil tends to become hard and more compact with the long-continued use of land-plaster; but whether this effect is wholly due to the wearing out of the organic matter, or in some part due to the cementing properties of the calcium sulfate, cannot be stated with certainty. When de- hydrated, calcium sulfate becomes plaster of Paris, and it is a constituent of different cementing materials. The temporary beneficial effect of land-plaster is probably due to its chemical action in the soil. It may convert more or less of the supposedly difficultly available iron phosphate into the more readily available tricalcium phosphate, as indicated by the follow- ing equation : 2 FeP04 + 3 CaS04 = Ca3 (PO4)2 + Fe2 (SO4)3. Very possibly this or some similar reaction occurs to a limited extent when calcium sulfate is applied to a soil containing iron 533 534 VARIOUS FERTILITY FACTORS phosphate. Another possible reaction may result in the liberation of potassium or magnesium from polysilicates, as roughly indicated by the following equation: AlFeMgNaKI+2 (SiO3)y (H2O)Z + CaSO4 = AlFeMgNaKxCa (SiO3)y(H2O)z+K2SO4. With large supplies of potassium present in polysilicates, heavy applications of calcium sulfate would doubtless liberate some potassium sulfate, although under the opposite conditions mass action would force the reverse reaction; that is, heavy appli- cations of potassium sulfate to a soil containing much calcium in polysilicates would liberate some calcium sulfate. The potassium sulfate liberated from the insoluble silicate, as indicated above, may serve directly as plant food, or it may react to increase the availability of phosphorus, thus: Ca8 (PO4)2 + 2 K2SO4 = CaK4 (PO4)2 + i CaSO4. These equations are given to show some of the possible reactions that may occur when a soluble salt is added to the soil. A dozen different reactions may be taking place at the same time within the same cubic inch of soil, and it is easily possible that, while one reaction is taking place in one part of the cubic inch, the reaction is running in reverse order in another part, depending upon the mass, composition, and concentration of the insoluble and soluble salts. If a solution of ammonium sulfate or potassium chlorid is per- colated through a stratum of soil, an examination will usually show that, while ammonium or potassium passed into the soil, calcium and magnesium have passed out in the percolate. If tricalcium phosphate be shaken with pure water, practically no phosphorus will be found in the nitrate; but if a solution of some neutral salt, such as sodium chlorid or potassium nitrate, be sub- stituted for the pure water, very appreciable amounts of phosphorus are dissolved. Land-plaster has been much used in some parts of the North Central and Eastern states, and for a time it usually gives quite profitable results; but finally, the element of real value that has been liberated by the action of the land-plaster becomes so depleted CROP STIMULANTS AND PROTECTIVE AGENTS 535 that even heavy applications of plaster fail to liberate sufficient for profitable crops, and thus the plastered land is made poorer than the untreated land. A common method of advertising has been to write the word PLASTER with a heavy application of the material in large letters on a cultivated field in vi.ew of the public road. In the larger, greener growth of grain crops or grass, the word PLASTER can be read by the passers-by; and thus the landowner is induced to plaster his whole field. If, however, he would only apply plaster year after year where the word was first written, the time would come when the word could not be read; and, if he still continued the applica- tions, ultimately he would again be able to read PLASTER, if we may judge from the testimony of common experience. Common salt. Common salt (sodium chlorid) is sometimes used as a crop stimulant, but its beneficial effect is likely to be even less durable than that of land-plaster. However, where common salt or other soluble salts, such as sodium sulfate or magnesium sulfate, or mixtures like kainit, are applied in connection with sufficient supplies of phosphorus, nitrogen, and lime, the effect of the stimu- lant must be confined chiefly to holding the phosphorus or other necessary elements in available form and to liberating potassium from the soil; and where the natural supply of potassium is ex- tremely large, the beneficial effect of the applied salt may continue for many years, as is well shown in the results from Rothamsted. Kainit. Kainit, of course, also supplies some potassium, and is thus in some part a fertilizer, though in large part a stimulant. To some extent this is also true of the common acid phosphate, which contains phosphorus mainly in the form of a soluble acid salt, and twice as many molecules of manufactured land-plaster : Ca3 (PO4)2 + 2 H2SO4 = CaH4(PO4)2 + 2 CaSO4. It is difficult to conceive of a more effective combination than about 200 pounds per acre of a mixture of acid phosphate and kainit applied twice for each five-year rotation of corn, oats, wheat, clover, and timothy. With all crops removed, such a system would doubt- less as thoroughly deplete the soil as any that could be devised, clover itself, used in this way, being a very powerful soil stimulant. If anything could be added to hasten the action, an application 536 VARIOUS FERTILITY FACTORS of five tons of farm manure about once in ten years (spread very uniformly, and an occasional dressing of burned lime, would make this system of ultimate land ruin very complete. A mixture of 300 pounds of acid phosphate and 100 pounds of kainit in five years would, with the manure system, furnish about 5 pounds of nitrogen, 5 pounds of phosphorus, and 6 pounds of potassium per acre per annum; whereas, crops as large as we ought to try to produce would remove from the soil as a yearly average about 100 pounds of nitrogen, 20 pounds of phosphorus, and 80 pounds of potassium (see Table 13). When crops one half as large are produced under such a system of fertilization, the soil under the action of these stimulants must furnish about nine tenths of the nitrogen, half of the phosphorus, and six sevenths of the potassium required for the crops. An invoice of his stock of fertility will help the landowner to plan wisely for the future, because he can thus know in advance what the ultimate effect must be of such systems. Protective agents. As protective agents we may include materials which tend to ward off disease or insect enemies; and the effect may be produced by substances destructive to fungi or repellent to insects. Kainit is thought to act sometimes as a fungicide, and tankage is held by some to prevent attack from certain insects. Any treatment which hastens the normal growth of the plant usually helps the plant to resist or overcome the attack of insects or disease; and it is apparently true that imperfect or abnormal plants are more likely to suffer from such attacks than normal, healthy plants. It has been suggested that sucking insects prefer the concentrated sap of weak or somewhat withered plants to that of vigorous succulent plants. Doctor Forbes has suggested that the very dilute juice of a rapidly growing plant may constitute a starvation diet for healthy insects; in other words, that the capacity of the insect for such juice is not sufficient to furnish it with the amount of nutrition necessary for maintenance and reproduction. It is common observation that chinch bugs may attack and destroy wheat that would otherwise yield 10 or 15 bushels per acre, while wheat growing in the same field on land capable of producing 30 bushels or more per acre is not attacked. The author has noted in several different seasons that corn growing on land that will CROP STIMULANTS AND PROTECTIVE AGENTS 537 yield 40 to 60 bushels per acre has been very severely injured by the colaspis root worm, while no apparent damage was done on adjoining well-fertilized plots which produced 80 to 100 bushels per acre, although the insects were found in both parts of the field. In any or all of these ways small applications of fertilizers or stimu- lants may produce results in crop yields far beyond the direct nutrient value of the plant food applied. The practice is somewhat common in places of coating seed corn by stirring with a paddle dipped in warm tar, and to some extent castor oil has been used for the same purpose, the appli- cation being made two or three weeks before planting so that the oil coating may have time to " dry on." Turpentine has also been used, and some have advised putting powdered sulfur with the seed in the planter boxes. These of course are solely protective agents, if they have any value. Their use is based upon experi- ence, however, and not upon experiment, and thus far the practice seems to rest upon no better foundation than that of planting potatoes " by the moon," or " witching " for water, an " art " which fails to find water twice in the same place if the operator is blindfolded. CHAPTER XXIX CRITICAL PERIODS IN PLANT LIFE IN this connection we may well consider another cause of differences in crop yields quite out of proportion to the difference in soil treatment. There may be, and often are, critical periods in the life of plants, when some small measure of assistance may change prospective failure into marked success. Thus, it is not infrequently a question of life or death with the clover plant when the nurse crop is removed; and, while most of the plants may die at that time on untreated land, a good stand of clover may be saved where a very light application had been made of manure, fertilizer, or soil stimulant. The nutrient value of the application may not be sufficient for half a ton of clover, but the difference in yield may amount to one or two tons; and from the larger crop larger resi- dues are left on and in the soil, resulting in a larger crop of grain the following year. Thus, enormous credit may be given, which is not at all deserved, on the basis of total plant food concerned. One of the most critical periods in the life of the corn plant is at the time the ears are forming, and an ample supply of moisture appears to be especially necessary at that time. If a severe sum- mer drouth is coincident with this critical period, the yield of grain is likely to be small; and any soil treatment which has the effect either of hastening or retarding the development of the plant, and thus of bringing the earing time either before or after the drouth, may very markedly affect the crop yield. The farmer is usually most anxious for conditions under which his wheat will " fill " well, and since this is influenced very appre- ciably by the temperature during this critical period, it follows that very marked effects upon both yield and quality may sometimes result from any soil treatment that causes the wheat to " fill " either a few days earlier or a few days later than on the untreated land. On the other hand, the treatment, whether applied as a crop 538 CRITICAL PERIODS IN PLANT LIFE 539 stimulant or in a system of permanent soil improvement, may some- times be the means of bringing this critical period at the time when the weather conditions are most unfavorable, while the untreated land may mature a larger crop at the more favorable time. An instance has been reported of a field treated with half a ton per acre of raw phosphate having produced a crop of 45 bushels of oats free of rust, whereas only 20 bushels of badly rusted oats were produced from similar seed on adjoining untreated land. Two influences may help to produce this difference: the added phos- phorus tends to balance the -food ration and thus to strengthen the oats against the fungous disease (and against lodging, as well), and also to hasten the maturity by which the crop escapes the rust which might attack the plants maturing later and perhaps under weather conditions more favorable for the development of the disease. As was stated by the author to the farmer who reported this experience, the marked difference in yield is not to be credited even largely to the phosphorus because of the plant food for its own sake, but rather to a combination of influences to which the added phosphorus proved to be the key. While such examples may serve temporarily as good advertise- ments for the treatment applied, they are just as misleading for wide application as are the occasional reports of damage to crops produced by applying manure. All of this serves to emphasize the importance of having some fundamental knowledge upon which to base definite systems of permanent agriculture. For this pur- pose we must rely primarily upon the absolute facts furnished by chemistry and mathematics and be guided only by the results of carefully conducted and long-continued experiments. Single examples can be found in support of almost any practice or theory that can be advanced; but a mere experience, though it be repeated, invariably with the same result, for fourscore times, furnishes no proof whatever that the octogenarian will live to celebrate another birthday. A small amount of readily available plant food, such as 50 pounds of sodium nitrate per acre as a top dressing for wheat on poor land in a cold spring, may produce a sufficient increase in yield to more than pay the cost of the nitrate. Likewise, 100 pounds of " am- moniated bone and potash," carrying perhaps 2 pounds of nitro- 540 VARIOUS FERTILITY FACTORS gen, 4 pounds of phosphorus, and 2 of potassium may be dropped in or near the hill of corn with a " fertilizer attachment " to the planter, and, under adverse conditions of soil and season, the crop increase may show some profit. It should be clearly understood, however, that all such systems of fertilizing are of themselves only an aid to soil depletion, because the " good start " thus given to the crop enables it to draw upon the soil itself for larger supplies of one or more elements of plant food than would be furnished by the untreated soil and the fertilizer applied. Quite independent of any such practices, the landowner should make ample provision for maintaining the fertility of the soil, on normal soil, by large use of phosphorus and farm manure or legume crops and crop residues, sufficient limestone being applied when necessary to prevent or correct soil acidity. Where this is done, however, the use of " starters " is usually unnecessary and unprofitable. Indeed, the dropping of a small quantity of fertilizer in the hill of corn (or near it) is sometimes a source of damage, not so much because it may injure the seed or young plant, but because it does not encourage the normal development of the root system in proportion to the early growth of the plant, and as a consequence the crop may suffer from drouth later in the season much more than the unfertilized corn. CHAPTER XXX FARM MANURE THE value of farm manure is governed largely by four modify- ing and somewhat related factors. First, the composition of the materials used for feed and bedding. Second, the dryness, or dry-matter content, of the manure. Third, the preservation or stage of decomposition or waste of the manure. Fourth, the kind of animals producing the manure. As a general average a ton of fresh-mixed cattle and horse manure contains about 500 pounds of dry matter, 10 pounds of nitrogen, 2 pounds of phosphorus, and 8 pounds of potassium. It would be produced from about 810 pounds of air-dry feed (yielding 270 pounds of dry excrement) and 270 pounds of air-dry bedding (containing 230 pounds of dry matter). On this basis, four tons of air-dry feed and bedding (used in the proportion of 3 to i) would produce about 7^ tons of average fresh manure containing 25 per cent of dry matter and 75 per cent of water. Roughly, this represents the theoretically possible production of manure on the farm, if all crops grown are used for feed and bedding. If the crops sold from the farm amount to one third of the total produced, and if one fifth of the manure made is lost be- fore it is applied to the land, then for every ton of air-dry produce harvested and removed from the land one ton of manure could be returned. If we count 85 per cent of dry matter in the air-dry feed and bedding, and 66f as the average digestion sufficient for the dry matter in the food consumed (see Table 29), and 75 per cent of the nitrogen and phosphorus and 90 per cent of the potassium re- turned in the manurial excrements, then a ration of 500 pounds of clover hay and 310 pounds of corn, with 270 pounds of wheat 542 VARIOUS FERTILITY FACTORS straw for bedding, would make a ton of manure containing 500 pounds of dry matter, about 12^ pounds of nitrogen, 2 pounds of phosphorus, and 9^ pounds of potassium. Or a ration containing 500 pounds of timothy«hay and 310 pounds of oats, with 270 pounds of oat straw for bedding, would make a ton of manure containing 500 pounds of dry matter, about n pounds of nitrogen, 1.7 pounds of phosphorus, and 9 pounds of potassium. Some loss of nitrogen is likely to occur by volatilization, and both nitrogen and potassium are very likely to be lost in the liquid excrement. For the most common rations used in live-stock farming, 10, 2, 8 represent very approximately the average pounds of the three elements, nitrogen, phosphorus, and potassium, in a ton of " aver- age fresh manure." By leaching and fermentation the dry matter, nitrogen, and potassium are lost in approximately the same pro- portion, but the phosphorus is lost only about half as rapidly, so that one ton of average yard manure, resulting from perhaps two tons of fresh manure, contains about 500 pounds of dry matter, 10 pounds of nitrogen, 3 pounds of phosphorus, and 8 pounds of potassium, one half of the dry matter, nitrogen, and potassium, and one fourth of the phosphorus having been lost. Wheat bran contains about 24 pounds of phosphorus per ton, so that, for every 100 pounds of bran used in the ration, nearly one pound of additional phosphorus will be found in the manure. This illustration and reference to the average composition of food stuffs will show how important the factor of food is in affecting the quality of manure. Most analyses of manure represent the product in a more or less decomposed state, in which case the phosphorus content is likely to be appreciably higher than in strictly fresh manure, and even manure commonly called fresh is likely to have lost some nitrogen and potassium in the liquid excrement. The following analyses include some accepted averages from the best authorities. While these general averages may be satisfactorily applied to large quantities of mixed manure, or in estimating the amounts of plant food in repeated applications of fresh or yard manure, respectively, they cannot safely be used for small single lots, unless the per cent of dry matter is determined and the character of the feed and bedding used is known. FARM MANURE 543 TABLE no. COMPOSITION OF FRESH MANURE Pounds per Ton AUTHORITY KIND OF MANURE DRY MATTER NITROGEN PHOS- PHORUS POTASSIUM Wolff Cows 4 Rice, hulled .... Sweet corn .... 850 1 80 10.5 4-6 0.8 0-3 0.7 2.O SEEDS OF LEGUMES Soy bean 8 so O.2 O.I 28.1 Pea 850 n 8 3.6 8 4 I.I o 8 o.oo O.I 0.4 23. 4. Red clover .... Horse bean, Vicia . Garden bean Peanut kernels . . . 850 850 850 850 3°-s 40.8 39-0 38.0 6-3 5-3 4-2 3-3 "•3 10.7 10. 1 7-o 3-° i-3 i-3 1.8 i.i i.i 0-4 0.4 0-4 O.2 0.09 O.O9 °-3 0.2 °-3 o-S °-5 o-3 38.3 31.0 27.4 OIL SEEDS Cotton 850 36 5 4.2 8.4 3.1 1.2 0.3 O.O4 T.6 °-5 33-8 Flax 850 32.8 5-7 8.0 2-7 1.8 0.3 0.2 0.5 32.6 Hemp 850 26.1 7-i 7-6 i-5 7-5 O.O4 2-5 o-3 46.3 STRAW Corn (Maize) . . Oat Wheat 850 850 850 4.8 5-6 4.8 i-7 1.2 I.O 13-7 13-5 5.2 1.6 i-4 0.7 3-5 3-° 1.9 I.O 0.6 0.4 6.2 '3-4 14.4 0.4 i-5 0.4 0.6 2-7 0.8 45-3 61.6 46.0 Rye 850 4.0 I.I 7.1 0.7 2.2 0.6 8.8 0.5 0.8 38.2 Barley Buckwheat .... 850 850 6.4 13.0 0.8 2-7 8.8 20.4 0.7 1.2 2-3 6.9 0.7 i.i 10.90 1.4 1.2 0.8 J-5 4-i 45-9 5i-7 COBS AND CHAFF Corncobs .... Oat chaff .... Wheat chaff .... Rye chaff .... 850 850 850 850 2-3 6.4 7-2 5-8 0.09 0.6' i-7 2.4 1.9 3-7 6.9 4-3 O.I 0.9 0-7 0.7 O.I 2.8 1.2 2-5 0.04 i-4 0.04 0.6 23-5 34-8 30.9 0.07 2.1 r-3 O.2 O.2 0.8 0.4 4-5 71.2 92.0 82.7 HAY Redtop 81:0 10.5 i.e 8.0 Red clover in flower . Alsike clover . . . Alfalfa, early bloom. . Red clover, ripe Red clover, in bud Red clover, young . White clover in flower From very young grass oooooooo oooooooooooooo'oc 19.7 24.0 23.0 12.5 24-5 35-5 23.2 25-5 2-5 1.8 2-3 1.9 3-i 4-5 3-5 3-2 15-7 9-4 12.3 8-3 21.5 25-3 ii. i 26.3 3-9 3-i 1.9 4-2 4-7 4-7 3-6 2.8 14.6 9.8 18.2 "•3 iS-i 17.1 13-3 7-2 0.8 0.6 i-5 0.6 °-7 0.7 1.8 i.i 0.8 0.8 2.8 1.4 0.9 1.2 i-3 7-5 0.8 0.9 0.8 I.O I.I 1.4 3-3 I.O 2.2 2.2 1-9 1-3 2.4 3-3 2.6 8.4 57-6 40.0 62.0 44-7 68.4 82.3 61.1 82.4 GRASSES Timothy 3OO ?.4 I.O s o 1.2 O 2 7.1 O.3 i.i 20. ? Rye grass .... Orchard grass . . . Rich pasture grass 300 300 218 5-7 7.2 I.O 0.6 0.8 5-9 4-9 6.8 0.2 °-3 0.7 I.I 0.8 1.9 °-3 0.2 °-3 3-i 2.8 1.9 o-5 0.6 O.2 2.1 i-3 2.1 20.4 17.8 21. 1 604 APPENDIX TABLE 1216. COMPOSITION OF PLANTS AND PLANT PRODUCTS (CHIEFLY AFTER VON WOLFF, 1889)— Continued (Pounds in 1000 of Produce) *E § o JjD r is I M P § g z 9 H) PRODUCE Eg °l o • H £ 0 « w & ^g i! O P 2* I u Ix I 01 0 1 1 2 5 «M LEGUMES Red clover, young . . 140 6.0 0.7 4-3 0.8 2.8 O.I 0-9 O.2 0.6 14.0 Red clover in bud . . 1 80 5-3 0.7 4.6 I.O 3-2 O.2 0.2 O.2 o-5 14.7 Red clover in flower . 200 4.8 0.6 3-7 0.9 3-4 O.2 O.2 O.2 o-5 Alsike clover . . . 1 80 5-3 0.4 2.0 0.7 2.1 O.2 O.I O.2 8!6 Lucern, or alfalfa . . 260 7.2 0.7 3-8 o-S 6.1 0.4 0.8 O.2 0^6 19.2 White clover in flower i95 5-6 0.8 2.6 0.8 3-i 0.4 o-3 0.7 0.6 14-3 FRUITS Apple, entire fruit 169 0.6 O.I 0.7 O.I 0.07 0.4 0.04 0.4 2.2 Pear, entire fruit . . 169 0.6 O.2 O.I O.2 0.08 0.05 0.2 3-3 Grape, entire fruit 170 1.7 0.06 4-2 O.2 O.07 O.2 O.I O.O/ O.I 8.8 Raspberries .... 1 80 1.5 2.1 2-9 Strawberries .... IOO 1.5 o-S 2-5 ROOTS, TUBERS, BULBS Potato 250 3 A O.7 4-8 O 1 O.2 O.2 0.09 O.2 o •? 9e Sugar beets .... "T- 1.6 / 0.4 3-2 w*o 0.4 O.I 0.9 0.4 o-3 • j Turnips 80 1.8 O.3 2.4 O. I O-5 O.? 0.5 0.3 64 Rutabagas .... 130 2.1 o o-5 2-9 0.2 0.6 O °-3 "•T- 7-5 Artichoke .... 2OO 3-2 0.6 3-9 O.2 O.2 0.2 0.9 0.7 0.4 9.8 Onion . . . > . 140 2.7 0.6 2.1 O.2 I.I 0.2 o-3 O.I 0.2 7-4 Beets v . I2O 1.8 o.t 4.O O.2 O.2 O.I i . i O.9 9T ' L Carrots . . . .'" . 2 .2 o 0.5 2 ? O.2 0.6 O.2 0.09 o.oo i « ^ O.4 8.2 Parsnip . . . .v . 2O7 5-4 0.8 * J 4-5 0.4 0.8 0.2 0.09 O.I 0.4 IO.O Radish . . . .v. 67 I .Q O.2 O. I o c O. I -7 QC A O Horseradish .... 233 4-3 0.9 6.4 O.2 i-4 2.0 0.7 0.3 0-3 19.7 "VEGETABLES" Cabbage, outer leaves no 2.4 0.6 4-8 0.4 2.O I.O 0.05 I.I J-3 15.6 Cabbage, heart . . Cucumber, fruit . . IOO 44 i!6 ol 3-6 2.0 O.2 O.I 0.9 0.3 o-S 0.2 0.05 0.2 0.6 0.4 o-5 0.4 9.6 5-8 Lettuce 60 O 3 3T • A O. I 0.6 0.6 8.1 Asparagus, sprouts . 67 3-2 0.4 I.O O.I 0.4 O.I 0.2 0.7 o-3 Cauliflower, heart . . 96 4.0 o-7 3-o O.2 0.4 0.4 O.I 0.4 °-3 8.0 MISCELLANEOUS Wheat flour .... 900 22.O 2-5 4-5 Wheat bran .... 875 25.0 12.2 13.0 Gluten meal .... 900 50.0 r-3 0.4 Hominy feed . . . 900 I6.3 4-3 4-0 Peanut-kernel cake 900 76.7 8.7 12.5 Soy-bean cake . . . 900 68.0 IO.O 15.0 Rape cake .... 900 50.0 8.7 10.8 Linseed cake . . . Cotton-seed cake . . 850 850 47-2 62.1 6.8 12.7 10. 1 12.6 4.8 5-9 2.O °-7 2.9 2-5 0.6 0.4 Si-3 66.4 Tobacco leaves . Tobacco stems . . . Flax stalks .... 850 850 850 34-8 24.6 3-o A. 2 1.8 35-4 24-4 7.8 6-5 1.2 37-5 9.2 4-8 3-5 0.9 0.8 4.0 0.8 0.8 3-5 5-T 1.8 9-4 2.4 1.3 140.7 64-7 31-1 Hemp stalks . . . 850 0.9 4-4 1.2 11.4 O.2 1.4 0.4 0.6 Hops, entire plant . . 850 25.0 2-5 14-7 4.2 1.2 6.2 1.4 3-7 72.9 APPENDIX 605 SECTION IV STATISTICS OF AGRICULTURAL PRODUCTS1 CROP AREAS, YIELDS, AND VALUES, 1908 The final revised estimates of the Crop Reporting Board of the Bureau of Statistics, United States Department of Agriculture, based on the reports of the correspondents and agents of the Bureau, supplemented by information derived from other sources, indicate the acreage, produc- tion, and value, in 1908 and 1907, of important farm crops of the United States to have been as follows : CROP ACREAGE (Acres) PRODUCTION FARM VALUE, DEC. i Per Acre (Bu.) Total (Bu.) Per Bushel Total Corn, 1908 . . 1907 .... 101,788000 99,931000 26.2 25-9 2,668,651000 2,592,320000 $0.606 .516 $1,616,145000 1,336,901000 Winter wheat, 1908 1907 . 30,349000 28,132000 14.4 14.6 437,908000 409,442000 •937 .882 410,330000 361,217000 Spring wheat, 1908 1907 17,208000 17,079000 13.2 13.2 226,694000 224,645000 .911 .860 206,496000 193,220000 Oats, 1908 .... 1907 .... 32,344000 31,837000 25.0 2.3-7 807,156000 754,443°°° .472 •443 381,171000 334,568000 Barley, 1908 .... 1907 .... Rye, 1908 6,646000 6,448000 1,948000 1,926000 25-1 23.8 16.4 16.4 166,756000 153,592°°° 31,851000 31,566000 •554 .666 .736 •731 92,442000 102,290000 23.455°°° 23,068000 i9°7 Buckwheat, 1908 . . 1907 . . 803000 800000 19.8 17.9 15,874000 14,290000 •756 .698 12,004000 9,975000 Flaxseed, 1908 . . . 1907 . . . Rice, 1908 2,679000 2,864000 655000 627300 9.6 9.0 33-4 29.9 25,805000 25,851000 21,890000 18,738000 1.184 •956 .812 .858 30,577000 24,713000 17,771000 16,081000 1907 .... Potatoes, 1908 . . . 1907 . . . 3,257000 3,124000 85-7 95-4 278,985000 297,942000 .706 .617 197,039000 183,880000 Hay, 1908 .... IQO7 . 46,486000 44,028000 875000 820000 2 r-52 2 i. 45 4 820. 2 4 850-S 2 70,798000 2 63,677000 * 718,061000 4 698, 1 26000 3 8.98 3 n. 68 S.io3 S.I02 635,423ooo 743,5°7°°° 74,130000 71,411000 Tobacco, 1908 1907 . . . 1 Figures furnished by the Bureau of Statistics, United States Department of Agriculture, except where otherwise credited. 2 Tons. 3 Per ton. 4 Pounds. 6 Per pound. 6o6 APPENDIX CORN AVERAGE YIELD PER ACRE OF CORN IN THE UNITED STATES, 1898-1907, BY STATES; AND ACREAGE IN 1907 STATE OR 1898 1899 19OO 1901 1903 1903 1904 1905 1906 1907 1907 TERRITORY (Bu.) (Bu.) BU.; (Bu.) (Bu.) (Bu.) (,Bu.) (Bu.) (Bu.) (Bu.) (Acres) Maine .... 40.0 36.0 36.0 39-4 21.7 30.2 39-7 34-3 37-o 37-o I2OOO New Hampshire 41.0 39-° 37-Q 38.5 23-3 21.0 27-3 37-o 37-5 3S-o 26000 Vermont . 43 -° 36.0 40.0 40.0 21.8 23-4 35-9 34-7 35-5 36.0 55°°° Massachusetts . 40.0 36.0 38.0 40.5 3'-3 24.0 36.0 37-5 39-7 36.0 44000 Rhode Island 34-o 31.0 32.0 32.1 28.4 3O.I 34-1 32. s 33-1 31.2 IOOOO Connecticut . 37-o 39-o 38.0 39-o 3i-5 22.4 38-9 42.7 40.0 33-o 56000 New York . . 33-° 31.0 32.0 33-° 25.0 25.0 27-3 3i-5 34-9 27.0 600000 New Jersey . . 37-o 39-° 33-Q 36-9 34-5 24.0 38.0 35-8 36-3 3*-5 278000 Pennsylvania 37-o 32.0 25.0 35-° 36.1 31.2 34-o 38-9 40.2 32-5 1,413000 Delaware . . . 25.0 22. 0 24.0 30.0 28.0 27-S 3°-4 30-4 30.0 27-S 193000 Maryland . . 31.0 32.0 26.0 34-2 32.4 28.7 33-4 36-9 3S-o 34-2 649000 Virginia . . . 22. 0 20.0 16.0 22.2 22. 0 21.8 23-3 23-4 24-3 25.0 1,841000 West Virginia 29.0 26.O 27.0 23.0 26.5 22.6 25-3 29.8 3°-3 28.0 760000 North Carolina . 14.0 I3.0 12.0 12.0 13-9 14.7 'S-2 '3-9 iS-3 16.5 2,732000 South Carolina . IO.O 9-0 7.0 6.9 IO.4 10.3 12.4 10.9 12.2 iS-i 1,974000 Georgia . . . 9.0 IO.O IO.O IO.O 9.0 II-7 11.9 II.O 12. 0 13-° 4,426000 Florida . . . 9.0 IO.O 8.0 9.0 8.6 9.9 10.7 10. 1 II.O "•3 621000 Ohio .... 37-o 36.0 37-o 26.1 38.0 29.6 32.S 37-8 42.6 34-6 3,400000 Indiana . . . 36.0 38.0 38.0 19.8 37-9 33-2 3I-S 40.7 39-° 36.0 4,690000 Illinois . . . 30.0 36.0 37-o 21.4 38.7 32.2 36.5 39-8 36.1 36.0 9,521000 Michigan . . . 34-o 25.0 36.0 34-5 26.4 33-5 28.6 34-o 37-° 3°-* 1,900000 Wisconsin 3S-o 3S-o 40.0 27.4 28.2 20-3 20-7 37-6 41.2 32.0 1,459000 Minnesota 32.0 33-o 33-o 26.3 22.8 28.3 26.9 32-5 33-6 27.0 1,615000 Iowa .... 35-o 31.0 38.0 25.0 32.0 28.0 32.6 34-8 39-S 29-5 9, 160000 Missouri . . . 26.0 26.0 28.0 10. 1 39-Q 32-4 26.2 33-8 32-3 31.0 7,775000 North Dakota . 19.0 23.0- 16.0 22.6 19.4 25.2 21.2 27-5 27.8 20. o 154000 South Dakota . 28.0 26.0 27.0 2I.O 18.9 27.2 28.1 31.8 33-5 25-5 185000 Nebraska . . 2I.O 28.0 26.0 I4.I 32-3 26.0 32.8 32.8 34-i 24.0 7,472000 Kansas . . . 16.0 27.0 19.0 7.8 29.9 25.6 2O-9 27.7 28.9 22.1 7,020000 Kentucky . . 31.0 2I.O 26.0 I5.6 27.0 26.6 26.9 29.7 33-° 28.2 3,300000 Tennessee . . 26.0 2O.O 2O.O 14-2 21.9 23-5 25.0 24.6 28.1 26.0 3,014000 Alabama . . . 15.0 12. 0 II.O IO-9 8-4 14.8 IS.O 14.8 16.0 '5-5 2,961000 Mississippi . . 18.0 16.0 II.O IO-9 "•5 18.4 19.1 i4-3 18.5 1 7.0 2,500000 Louisiana . . . 18.0 18.0 17.0 13-7 12.5 20.6 19.9 i3-7 17.2 J7-5 1,600000 Texas .... 25.0 18.0 18.5 n.6 8.1 24.2 22.6 21.3 22.5 2I.O 7,409000 Oklahoma . . 19.0 26.0 7-3 25.8 23-3 28.1 25-3 32-9 24.4 4,650000 Arkansas . . 20. o 20. 0 19.0 8.1 21-3 20.9 21.6 i7-3 23.6 17.2 2,525000 Montana . . . 28.0 23.0 15.0 25.0 22. 0 24.1 22.2 19.4 23-4 22-5 4000 Wyoming . . 16.0 22. 0 34.0 39-S 19.8 19.4 32.5 26.9 27.0 25.0 3000 Colorado . . . 18.0 17.0 19.0 17.1 I6.S 19.8 20.5 23.8 27.9 23-5 IIIOOO New Mexico. 2I.O 2O.O 22. 0 31-6 22. 0 24.0 22.7 25-3 29.4 29.0 42000 Arizona . . . 18.0 20. 2 22.4 23.8 27.0 29-5 37-5 8000 Utah .... 21. 0 2O.O 2O. O 19.4 2O. I 21.4 33-2 26.2 32.0 25-5 IIOOO Idaho .... Washington . 12.0 23.0 20. 0 23.0 J7-5 24.7 23.0 34.5 23.1 29-3 24-7 27.2 24.2 28.3 25.2 30.0 27.0 5000 I200O Oregon . . . California . . 24.0 26.O 22. 0 27.O 23.0 25.0 20.8 31.0 23-4 3°-S 25.8 30-7 28.8 28.6 23.0 32.0 27.6 34-9 27-S 34-0 16000 54000 Gen'l average 24.8 2S-3 25-3 16.7 26.8 25-5 26.8 28.8 3°-3 25-9 99.931000 Total APPENDIX 607 CORN — Continued ACREAGE, PRODUCTION, VALUE, PRICE, AND EXPORTS OF CORN IN THE UNITED STATES, 1849-1909 Year Acreage Aver- age yield per acre Production Aver- age farm price per bushel, Dec. i Farm value Dec. i. Domestic exports, including corn meal, fiscal year beginning July i Per Cent of Crop ex- port- ed Acres Bu. Bushels Cents Dollars Bushels 7 632860 P.ct. 1866 867,946295 1.8 T867 23.6 768,320000 1.6 jg68 34 887246 906,527000 46.8 8,286665 1869 . . . 59.8 1870 .... 28 3 1871 99 1 ,898000 3.6 :872 30.8 35.3 j873 23.8 932,274000 44.2 1874 58.4 1871; 36.7 !876 1,283,827=100 1877 26 7 34.8 6 5 1878 .... 87 884892 6.3 1879 580,486217 6.4 1880 1,717,434543 5.5 1881 . . . 18 6 63 6 !882 48.5 783 867175 2 6 1883 68 301889 1884 69 683780 25.8 1885 . .... 32 8 1886 36.6 1887 1888 3.6 1889 ...... 1890 78,319651 27.0 2,112,892000 28.3 597,918829 103,418709 4.9 1891 1892 1893 76,204515 70,626658 27.0 23.1 2,060,154000 1,628,464000 40.6 39-4 ,6 c 836,439228 642,146630 76,602285 47,121894 3-7 2.9 1894 1895 1896 1897 62,582269 82,075830 8l,O27I56 19.4 26.2 28.2 1,212,770052 2,151,138580 2,283,875165 45-7 25-3 21-5 554,719162 544,985534 491,006067 28,585405 101,100375 178,817417 2.4 4-7 7-8 1898 24 8 28 7 82 IO8587 8 6 16 7 60 5 28 028688 i 8 1902 1903 1904 94,043613 88,091903 26.8 25-5 26 8 2,523,648312 2,244,176925 40-3 42 5 1,017,017349 952,868801 76,639261 58,222061 3-o 2.6 1905 1906 94,011369 28.8 2,707,993540 41.2 ,116,696738 119,893833 86 368228 4-4 1907 51 6 1908 60 6 55? 3 " R 11 * ' ' '»!/W\ ', 'o4' Census figures of production. 6o8 APPENDIX CORN — Continued CORN CROP OF COUNTRIES NAMED, 1902-1906 COUNTRY 1902 (Bu.) 19O3 (Bu.) 1904 (Bu.) 1905 (Bu.) 1906 (Bu.) United States . . Canada (Ontario) . Mexico .... Total No. America Argentina . . . Total So. America Austria-Hungary . France .... Italy .... 2,523,648000 21,159000 78,099000 2,244,177000 30,211000 90,879000 2,467,481000 20,880000 88,131000 2,707,994000 21,582000 85,000000 2,927,416000 24,745°°° 70,000000 2,622,906000 2,365,267000 2,576,492000 2814,576000 3,022,161000 84,018000 148,948000 175,180000 140,708000 194,912000 89,944000 i.S5.355°0° 179,701000 146,369000 198,084000 139,126000 24,928000 71,028000 16,000000 68,447000 48,419000 25,272000 183,994000 25,360000 88,990000 14,000000 80,272000 50,464000 18,759000 89,7S7ooo 19,482000 90,545000 15,000000 19,598000 25,920000 21,300000 139,307000 24,030000 97,265000 16,000000 59,275000 33,33io°o 31,880000 215,636000 14,581000 93,007000 16,000000 130,546000 70,501000 30,000000 Portugal .... Roumania . . . Russia (European) . Spain Total Europe Africa 429,716000 504,154000 303,858000 442,168000 618,057000 36,899000 7,256000 36,1 18000 4,987000 38,862000 9,972000 37,6550°° 8,374000 37,700000 8,608000 Australia .... New Zealand . . Total Australasia Grand total . . 590000 627000 547000 506000 653000 7,846000 5, 614000 10,510000 8,880000 0,261000 318,7311000 3,066,508000 3,109,432000 3,449,648000 3,886,163000 CORN, AVERAGE YIELDS PER ACRE, BUSHELS YEARS SOUTH CAROLINA GEORGIA IOWA • ILLINOIS UNITED STATES 1866-1875 (10 years) 9-7 "•3 34-3 29.9 26.1 1876-1885 (10 years) 8.8 10.3 31-8 27.2 25-5 1886-1895 (10 years) 10.2 II. 2 30.1 29.0 23-4 1896-1905 (10 years) 9-5 10-5 32-4 34-5 25.2 1866-1885 (20 years) 9-3 10.8 33-o 28.6 25.8 1886-1905 (20 years) 9-9 10.8 31.2 3i-7 24-3 1866-1905 (40 years) 9.6 10.8 32.1 30.1 25.0 CORN, SINGLE-YEAR RECORDS 1899, bu. Per acre 1909, bu. per acre 9.0 16.7 10.0 13-9 31.0 3I-S 36.0 35-9 25-3 25-5 1899, acres of corn 1909, acres of corn 1,857000 2,218000 3,249000 4,400000 7,815000 9,200000 6,865000 10,300000 82,109000 108,771000 1899, bu. °f corn 1909, bu. of corn 16,713000 37,041000 32,495000 61,160000 242/250000 289,800000 247,150000 369,770000 2,078,000000 2,772,360000 1899, price per bu. 1909, price per bu. 45** 9oj* 49^ 86^ 35^ 49^ 36? 52? 37-2^ 59-6?1 1899, value of crop 1909, value of crop $ 8,357000 33,337ooo $16,247000 52,598000 $55,7I7°oo 142,062000 $64,259000 192,280000 $ 629,210000 1,652,822000 APPENDIX 609 INTERNATIONAL TRADE IN CORN, INCLUDING CORN MEAL, 1902-1906 GENERAL NOTE. Substantially the international trade of the world. The exports given are domestic exports and the imports given are imports for consumption, as far as it is feasible and consistent so to express the facts. While there are some inevitable omissions from such a table as this, on the other hand, there are some duplications because of reshipments that do not appear as such in official reports. For the United Kingdom import figures refer to imports for con- sumption. EXPORTS COUNTRY YEAR BEGIN- NING 19O3 (Bo.) 1903 (Bu.) 1904 (Bu.) 1905 (Bu.) 1906 (Bu.) Tan 46,959590 3,010624 4,346609 7,883279 4,726324 43,013192 44,148590 1,091588 76,63926 i 703770 1,528000 82,845915 310804 6,579655 5, 089114 5.373J94 31,080198 25,349683 171767 58,222061 1,004063 1,086000 97,221783 174342 6,287688 9,762657 4,449009 18,042377 18,633663 130225 90,293483 2,002431 1,009000 87,487629 63218 8,078215 3,870090 4,278515 i,44i437 7,372386 806115 "9,893833 28519 4,100325 106,047790 22361 6,588557 5,658500 6,010176 1 23,394301 29,878141 1,755446 86,367988 2 034696 23,547299 Austria-Hungary . Belgium .... j Mi- Ian. Jan. Tan Netherlands . . . Roumania . . . Russia .... Servia .... United States . . Uruguay .... Other countries Total .... Jan. Jan. Jan. Jan. July July 234,050827 2I7,II2454 248,006658 237,420282 250,205255 IMPORTS Austria-Hungary . Jan. 5,87497i 11,130274 14,090377 18,511368 7,118221 Canada .... Cape of Good Hope Cuba July Jan. Jan. 7,i54522 1,943896 1,150176 11,333530 3,471281 619326 12,003574 1,236927 696517 11,779679 2,171601 I.84374.8 2 15,233894 215007 2,480087 Denmark .... EorvDt . Jan. Ian. 12,355050 SS266 8,772022 14.21; T.1 9,284777 StOn 10,859257 18,855752 France Germany 3 . . . Italy j Jan. Jan. (an. 8,674931 35,454243 8,216902 n,347II4 37,527343 15,092527 10,124353 30,450853 &,^6^I2T, 11,122512 36,538366 ^.00287^ 14,509103 44,883053 8 666763 Mexico .... Netherlands . . . Norway .... Portugal .... Russia Jan. Jan. Jan. Jan. Jan. 142102 15,817237 637387 759967 !•? S822 496028 20,160078 765246 366605 41:771 c 476182 16,547198 555991 531889 62^^26 1,454327 16,234785 544596 2,724050 163070 2 2,079553 25,305233 718277 4 2,724050 2 437868 Spain Jan. OO7272 1,484490 2 761426 2 637I445 189357 2,6l I2O2 2,197476 101,284919 234986 2,704457 1,422985 86,076697 49I035 2,498380 1,277353 84,156490 564946 2,887291 * i, 277353 97,736852 3,260478 7>3I847° 3,309436 7,429351 2 7, 090991 Total. . . . 210,483315 257,091403 221,026621 243,057067 277,005211 1 Average, 1902-1905. 2 Preliminary. 8 Not including free ports prior to March i, 1906. * Year preceding. 6io APPENDIX WHEAT WHEAT CROP OF COUNTRIES NAMED, 1903-1907 COUNTRY 1903 (Bu.) 1904 (Bu.) 19O5 (Bu.) 19O6 (Bu.) 19O7 (Bu.) United States . . Canada .... Mexico .... Total No. America Argentina . . . Total So. America Austria-Hungary . Belgium .... Bulgaria .... Denmark .... France .... Germany .... Greece .... Italy 637,822000 85,271000 10,493000 552,400000 75,213000 9,393000 692,979000 113,441000 7,000000 735,261000 132,705000 7,000000 634,087000 96,606000 10,000000 733,586000 637,006000 813,420000 874,066000 740,603000 103,759000 129,672000 150,745000 134,031000 155,093°°° 110,113000 155.185000 160,834000 151,604000 178,636000 226,721000 12,350000 35.55 100° 4,461000 364,420000 130,626000 8,000000 184,451000 4,258000 307000 8,000000 73,700000 551,728000 10,885000 128,979000 5,538000 26,000000 46,524000 204,406000 13,817000 42,242000 4,302000 298,826000 139,803000 8,000000 167,635000 4,423000 2I2OOO 9,OOOOOO 53,738000 622,255000 11,676000 95,377000 5,135000 23,000000 35,624000 228,138000 12,401000 40,736000 4,083000 335-453°°° 135,947°°° 8,000000 160,504000 5,109000 329000 5,000000 103,328000 568,274000 11,280000 92,504000 5,529000 20,000000 57,422000 268,675000 12,964000 55,076000 4,161000 324,919000 144,754°°° 8,000000 176,464000 4,978000 303000 9,000000 113,867000 45°>963°°° 13,21 1000 140,656000 6,656000 25,000000 57,583°°° 185,059000 12,000000 30,000000 4,000000 369,970000 127,843000 8,000000 177,543°°° 5,000000 20000O 6,000000 42,237000 455,OOOOOO 8,375000 100,331000 5,953000 16,000000 53,860000 Netherlands . . . Norway .... Portugal .... Roumania . . . Russia (European) Servia Spain Sweden .... Turkey (European) England .... Total United Kingdom . . Total Europe . British India, includ- ing such native states as report . Cyprus .... Japan Russia (Asiatic) Turkey (Asiatic) . Total Asia . . Algeria .... Cape of Good Hope Egypt Natal 50,321000 39,o82OOO 62,188000 62,481000 58,275000 1,830,526000 1,747,262000 1,803,132000 1,826,422000 I,6l6,o86oOO 297,601000 2,477000 9,600000 69,659000 35,000000 359,936oo° 2,176000 19,754000 44,494000 35,000000 283,063000 2,441000 18,437000 68,01 1000 35,000000 320,288000 2,410000 20,283000 57,427000 35,000000 315,386000 2,OOOOOO 22,932OOO 56,OOOOOO 35,OOOOOO 430,516000 477-55°°°° 423,152000 451,586000 447,5I8OOO 34,035000 i,755°00 12,000000 4000 294000 7,523000 25,484000 2,000000 12,000000 7000 486000 10,519000 25,579000 2,000000 12,000000 4000 483000 5,720000 34,080000 2,000000 12,000000 8000 542000 4,409000 3I(I2OOOO 2,OOOOOO 12,000000 6OOO 50000O 6,OOOOOO Sudan (Anglo- Egyp- tian) .... Tunis .... Total Africa . . Australia .... New Zealand . . Total Australasia Grand total . . 55,61 1000 50,406000 45, 705000 53,030000 51,626000 12,768000 7,693000 76,488000 8,140000 56,215000 9,41 1000 70,681000 7,013000 68,l85COO 5,782000 20,461000 84,628000 65,626000 77,694000 73,06~OOO 3,189,813000 3,152,127000 3,32°,959°°° 3,435,401000 3,108,526000 APPENDIX 611 WHEAT — Continued ACREAGE, PRODUCTION, VALUE, PRICE, AND EXPORTS or WHEAT IN THE UNITED STATES, 1849-1909 Year Acreage Aver- age yield per acre Production Aver- age farm price per bushel, Dec. i Farm value Dec. i Domestic exports, including flour, fiscal year beginning July i Per cent of crop ex- ported Acres Bu. Bushels 100,485944 173,104924 151,999906 212,441400 224,036600 260,146900 235,884700 230,722400 249,997100 281,254700 308,102700 292,136000 289,356500 364,194146 420,212400 448,756630 498,549868 383,280090 504,185470 421,086160 512,765000 357,112000 457,218000 456,329000 415,868000 490,560000 399,262000 611,780000 515,949000 396,131725 460,267416 467,102947 427,684346 530,149168 675,148705 547,303846 522,229505 748,460218 670,063008 637,821835 552,399517 692,979489 735,260970 634,087000 664,602000 737,189000 Cents Dollars Bushels 7,535901 17,213133 12,646941 26,323014 29,717201 53,900780 52,574111 38,995755 52,014715 91,510398 72,912817 74,750682 57,043936 92,141626 150,502506 180,304180 186,321514 121,892389 147,811316 111,534182 132,570366 94,565793 153,804969 119,625344 88,600743 109,430467 106,181316 225,665811 191,912635 164,283129 144,812718 126,443968 145,124972 217,306005 222,618420 186,096762 215,990073 234,772516 202,905598 120,727613 44,112910 97,609007 146,700425 163,043569 123,000000 P.ct. 7-5 9-9 8-3 12.4 13-3 20.7 22.3 16.9 20.8 32.5 23.7 25.6 19.7 25.3 35-8 40.2 37-4 31-8 29-3 26.5 25-9 26.5 33-6 26.2 21.3 22.3 26.6 36.9 37-2 41-5 31-5 27.1 33-9 41.0 33-o 34-0 41-4 31-4 30.3 18.9 8.0 14.1 20. o 25-7 18.5 £866 15,424496 18,321561 18,460132 19,181004 18,992591 19,943893 9-9 1.6 2.1 3-6 2-4 1.6 1.9 2-7 2-3 i.i 0.5 3-9 3-1 3-8 3-1 O.2 3-6 1.6 3-0 0.4 2.4 2.1 I.I 2-9 I.I 5-3 3-4 1.4 3-2 3-7 2-4 3-4 5-3 2-3 2-3 5-o 4-5 2.9 2-5 4-5 5-5 4.0 4.0 5-8 152-7 145.2 108.5 76.5 94-4 II4-5 111.4 106.9 86.3 89-5 96.3 105.7 77-6 110.8 95-1 119.2 88.4 91.1 64-5 77.1 68.7 68.1 92.6 69.8 83.8 83-9 62.4 53-8 49.1 50.9 72.6 80.8 58.2 58.4 61.9 62.4 63.0 69-5 92-4 74-8 66.7 87-4 92.8 99-0 232,109630 308,387146 243,032746 199,024996 222,766969 264,075851 278,522068 300,669528 265,881167 261,396926 278,697238 385,094844 325,814119 497,030142 474,201850 456,880427 445,602125 383,649272 330,862260 275,320390 314,226020 310,612960 385,248030 342,491707 334,773678 513,472711 322,111881 213,171381 225,902025 237,938998 310,602539 428,547121 392,770320 319,545259 323,515177 467,350156 422,224117 443,024826 510,489874 518,372727 490,332760 554,437000 616,826000 730,046000 1867 1868 1869 1870 1871 1872 . . .... 20,858359 22,171676 24,967027 26,381512 27,627021 26,277546 32,108560 32,54595° 37,986717 37,709020 37,067194 36,455593 39,475885 34,189246 36,806184 37,641783 37.336138 38,123859 36,087154 39,916897 38,554430 34,629418 34,882436 34,047332 34,618646 39,465066 44.055278 44,592516 42,405385 49,895514 46,202424 49,464967 44,074875 47,854079 47,305829 45,211000 47,557000 46,723000 1873 . 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 . .• 1898 1899 1908 Census figures of production. 6l2 APPENDIX WHEAT — Continued AVERAGE YIELD PER ACRE OF WHEAT IN THE UNITED STATES, AND ACREAGE IN 1907 8-1907, BY STATES, STATE OR 1898 1899 1900 1901 1902 1903 1904 1905 19O6 1907 1907 TERRITORY (Bu.) (Bu.) (Bu.) (Bu.) (Bu.) (Bu.) (Bu.) (Bu.) (Bu.) (Bu.) (Acres) Maine .... 19-5 22.5 19-5 23-9 25-3 2S-S 23-3 23.0 24.8 26.2 8000 Vermont . . . 22.5 22.0 23-5 18.7 18.8 20.9 25-1 18.8 22.3 23.0 1000 New York . . 21.2 l8.S 17.7 I3-1 16.8 17.8 JI-3 21.0 20.0 17-3 416000 New Jersey . . Pennsylvania 17.4 17-5 I4-S I3.6 19.1 J3-S 16.8 17.1 16.0 iS-8 14.0 IS -6 '3-3 14.1 16.4 I7.I 18-3 17.7 18.5 18.6 108000 1,618000 Delaware . . . *3-3 12.8 20.3 18.5 16.5 IO.2 14.9 I3.8 16.0 20.5 I2OOOO Maryland . . . Virginia . . . 15-3 I4.I 14.1 8.4 19-5 11.9 17.2 10.9 14.7 5-7 12-5 8.7 13-4 IO.2 I6.3 11.4 16.0 12.5 19.0 '2-5 777000 655000 West Virginia I3.8 9-3 9.8 10.9 7-7 IO.2 IO.I 12.3 12.7 12.2 367000 North Carolina . 9.2 6.7 9.6 8.7 5-3 5-1 8.6 6.7 9.1 9-5 560000 South Carolina . 10.6 6-5 9.0 8.8 5-6 6.5 8.1 6.1 9-3 8.5 314000 Georgia . . . IO.O 6.8 9.1 8.2 6.0 6.2 8.8 6.9 IO.O 9.0 297000 Ohio .... 16.9 14.2 6.0 iS-3 17.1 13-7 "•5 17.1 20.4 16.3 1,882000 Indiana . . . 15.6 9.8 5-3 15.8 16.0 IO.O 9.2 18.3 20.7 14.4 2,362000 Illinois . . . II.O IO.O 13.0 17.6 17.9 8.4 13-8 16.0 19-5 18.0 2,228000 Michigan . . . 20.8 8.4 7.6 ii. i 17.7 15-5 9.8 18.5 I3-1 14-5 878000 Wisconsin . . 18.0 J5-5 iS-5 16.1 18.1 15.6 iS-S 16.6 16.3 14.1 2IOOOO Minnesota . . 15.8 i3-4 10.5 12.9 !3-9 I3-1 12.8 i3-3 10.9 13.0 5,200000 Iowa .... 16.7 13.0 15.6 16.2 12.7 12.4 n.6 14.2 iS-7 i3-4 569000 Missouri . . . 9.8 9-9 12.5 !S-9 19.9 8.7 17.7 12.4 14.8 13.2 2,213000 North Dakota . 14.4 12.8 4-9 I3-1 JS-9 12.7 n.8 14,0 13.0 IO.O 5,513000 South Dakota . 12.4 10.7 6.9 12.9 12.2 13-8 9.6 13-7 13-4 II. 2 2,900000 Nebraska . . . 16.4 10.3 12. 0 17.1 2O.9 !5-7 13.6 19.4 22. 0 18.1 2,535°00 Kansas . . . 14.2 9.8 17.7 18.5 IO.4 14.1 12.4 13-9 I5-I II.O 5,959000 Kentucky . . . iS-4 9.1 I3.0 12. 1 9-3 8.4 11.4 "•3 I4.I I2.O 734000 Tennessee . . 13.2 8.7 9-9 10.8 7-2 7-1 "-S 7-2 !2-5 9-5 779000 Alabama . . . I2.O 7.6 9-5 8.7 6.0 9.1 10.3 9.6 II.O IO.O 89000 Mississippi . . 13-9 7-7 9.6 8.8 8.0 8.0 8.8 10.8 IO.O II.O 2000 Texas .... 14.8 ii. i 18.4 8.9 9.0 i3-4 10.7 8.9 "•5 7-4 38OOOO Oklahoma . . 14.9 13-3 19.0 16.4 ii. i 14.9 11.7 8.2 14.0 9.0 959000 Arkansas . . . II.O 8.6 IO.I 8.8 9.1 7.0 IO.I 7-9 10.8 9-5 154000 Montana . . . 29-S 25-7 26.6 26.5 26.0 28.2 23-9 23.8 24.0 28.8 . I39OOO Wyoming . 23-7 18.8 17.6 24-5 23-5 20.9 22.1 25-4 28.7 28.5 3OOOO Colorado . . . 26.3 23-7 22.6 24.1 18.0 26.6 22.8 25.0 32-5 29.0 293OOO New Mexico . . 23.8 13.8 21. 0 21.5 17.1 18.4 12.8 22.2 25.0 24.0 46OOO Arizona . . . 3'-7 l5-3 14.6 21.8- 18.7 25-3 25-5 24.4 25.2 25-9 15000 Utah .... 28.0 20.7 2O-9 20.5 21.2 22.6 26.6 26.4 27.4 28.8 i 6 i ooo Nevada . . . 29.0 18.0 24-5 25-1 27.1 27.6 26.2 27.0 3«-S 32.0 30000 Idaho .... 31.0 24.2 20.8 21.2 22.1 21. 1 22-9 28.2 24.4 25-3 342000 Washington . . 24.2 22.7 23-5 29.1 22.2 20-3 22.2 24.6 20.8 26.0 1,349000 Oregon . . . California . . 20.5 9.1 19.2 14.1 13.8 10.3 21. 1 13.0 2O. O IO-9 18.2 II. 2 19.0 10.8 18.6 9-3 2O.O I7.I 23-4 15-0 651000 1,368000 or total !5-3 12.3 12.3 I5.0 14-5 I2.9 I2.S 14-5 IS-5 14.0 45,211000 APPENDIX 613 INTERNATIONAL TRADE IN WHEAT, INCLUDING WHEAT FLOUR, 1902-1906 EXPORTS COUNTRY YEAR BEGIN- NING 1903 (Bu.) 1903 (Bu.) 1904 (Bu.) 1905 (Bu.) 1906 (Bu.) Argentina . . . Australia . . Austria-Hungary . Belgium .... British India . . Bulgaria . . . Canada .... Chile Jan. an. Jan. Jan. Apr. Jan. Jan. an.- }an. an. San. • an. an. uly 25,672368 10,799165 5,534270 I3,89°599 21,389010 9,320644 39,820238 1,043883 4,044662 37,3498°4 34,715888 114,872260 1,875580 202,905598 14,499026 65,421537 1,489763 5,532485 13,362799 50,684276 13,185710 24,452019 2,270728 7,956750 40,218462 31,860939 158,064833 2,016358 120,727613 9,3"3°7 90,115119 38,850166 3,984789 18,217597 83,128272 20,286368 21,110205 3,146416 8,640465 41,268227 26,718698 174,334182 3,098326 44,1 12910 10,955245 112,718476 32,506453 3,630659 18,496029 37,483073 17,508259 48,566652 706932 10,512765 53,951447 65,246599 181,759796 3,520627 97,609007 I3,IJ6253 89,128802 38,878679 4,059153 18,030378 32,213417 11,037613 42,224469 1 706932 10,350641 33,626290 '65,246599 3i37,i30392 3,365644 146,700424 3 14,227728 Germany 2 . . . Netherlands . . Roumania . . . Russia .... Servia .... United States . Other countries Total . . . 537,732995 546,555579 587,966975 697,333027 646,927161 IMPORTS Belgium .... Brazil .... Denmark Finland .... France .... Germany 2 . . . Greece .... Italy an. ] an. , an. 'f an. '_ an. Jan. Jan. Jan. 57,507743 10,845841 5,865624 3,026987 10,509786 77,822604 6,396218 43,33019° 2,427147 55-75286i 336955 2,620395 7,953342 15,226501 2oo,577oo4 43,509254 59,797102 12,129189 5,467021 3,442443 18,516169 72,501263 6,207668 43,174711 9,l64759 58,552553 2,748269 3,363238 ' 8,658924 16,324627 217,100937 58,579453 64,160454 i3,74535i 5,373202 3,4i376i 8,625293 75,436433 5,207403 29,670497 6,702045 58,916277 3,282298 8,253950 8,446395 17,220343 2i9,7i3497 47,873864 64,976813 14,983303 4,691567 3,580581 7,347i85 85,136923 5,863742 43>I04i99 7,873865 70,380247 4,672573 35,502385 7,5i5498 16,158553 212,089481 49,5i8455 68,178372 1 6,303441 5,648708 3,966877 11,732007 74,873885 7,924950 50,541670 5,622967 54,678154 '4,672573 20,040928 8,216745 16,196009 208,920370 349,598374 Japan .... Netherlands . . Portugal . . . Spain .... Sweden . . . . Switzerland United Kingdom . Other countries Total . . . Jan. an. ^ an. ' an. ] an. ] an. 543,708452 595,728326 576,041073 633,395370 607,116030 1 Year preceding. 2 Not including free ports prior to March i, ' 8 Preliminary. 1906. 614 APPENDIX WHEAT — Continued QUANTITY AND PERCENTAGE OF EXPORTS OF DOMESTIC WHEAT, 1907 AND 1908, BY LEADING PORTS CUSTOMS DISTRICT YEAR ENDING JUNE 30 — 1907 1908 (Preliminary) Bushels Per Cent of Total Bushels Per Cent of Total New York 18,679225 6,012732 7,198844 &,39I45° 5,026578 14,172021 2,622505 5,496935 1,940582 7,028551 24.4 7-8 9.4 II.O 6.6 18.5 3-4 7.2 2-5 9.2 21,478019 14 699237 13,411581 12,357077 8,763989 8,112828 5,261870 5,020235 3,845004 7,262321 21.4 14-7 13-4 12.3 8.7 8.1 5-3 5-o 3-8 7-3 Puget Sound Willamette Philadelphia Baltimore Galveston Boston and Charlestown New Orleans . Duluth All other Total 76,569423 IOO.O 100,212161 IOO.O AVERAGE YIELD OF WHEAT IN COUNTRIES NAMED,' BUSHELS PER ACRE, 1898-1907 YEAR UNITED STATES * RUSSIA, EURO- PEAN* GER- MANY2 AUSTRIA 2 HUNGARY PROPER 2 FRANCE 1 UNITED KING- DOM' Average (1888 to 1897) 12.8 8.4 22-7 15-6 17.9 17.6 30.1 1808 iz.t o 6 18 o « 8 1800 8 7 28 4 17 8 T? 8 IQOO 8 -i IOOI 8 i if. » i .9 T& e 1902 IOO3 T- Q 1904 29.2 TO 7 27 8 1905 a8 - Tfi - 1906 10. 0 20-5 19. I9O7 .... 7-7 3°-3 22-5 29.0 I5-1 Average (1898 to 1907) '3-9 9-3 28.4 18-3 17.9 20.8 32.6 1 Winchester bushels. 2 Bushel of 60 pounds. 3 For the ten years, 1886 to 1895, the average yield of wheat was 27.7 bushels per acre in Holland, and 37.3 in Denmark ; and for the succeeding ten-year period, 1896 to 1905, the average yield in Holland was 31.2 bushels, and in Denmark 40.6 bushels per acre, with only two years below a 4o-bushel average (35.2 bushels in 1898, and 29.2 in 1901). — C.G.H. APPENDIX OATS ' OAT CROP OF COUNTRIES NAMED, 1903-1907 COUNTRY 1903 (Bu.) 1904 (Bu.) 1905 (Bu.) 1906 (Bu.) 1907 (Bu.) United States . . Canada .... Mexico .... Total North America . . Austria .... Hungary proper . France .... Germany . . Russia (European) Great Britain — England . . . Scotland . . . Wales .... Ireland .... Total United Kingdom . . Total Europe . Russia (Asiatic) Total Asia . -. Total Africa . . Australia .... New Zealand . Total Australasia Grand total . • 784,094000 211,192000 13000 894,596000 208,024000 18000 953,216000 234,099000 17000 964,905000 251,194000 17000 754,443°°° 210,869000 17000 995,299000 1,102,638000 1,187,332000 i, 216, 1 16000 965,329000 128,330000 87,3340oo 300,366000 542,432000 728,049000 85,400000 36,379000 6,832000 58,816000 109,611000 62,775000 257,811000 477,852000 1,065,068000 86,728000 37,034000 7,661000 60,142000 123,880000 78,009000 269,581000 451,017000 851,667000 76,453000 36,390000 7,264000 60,754000 154,551000 87,7330°° 256,943000 580,875000 633,291000 84,102000 35,108000 8,063000 62,751000 170,657000 79,484000 314,132000 630,324000 820,621000 94,707000 36,056000 7,875000 60,080000 187,427000 191,565000 180,861000 190,024000 198,718000 2,268,425000 2,402,641000 2,203,967000 2,222,575000 2,493,532°°° 71,734000 SQ.^SOQO 84,995000 79,713000 85,176000 72,215000 12,116000 7,527000 22,452000 59,552000 14,309000 85,397000 12,077000 80,072000 12,418000 85,576000 12,008000 18,094000 i5,583oo° 9,064000 15,012000 10,805000 13,108000 14,041000 11,555000 29,979000 33,677000 24,076000 23,913000 25,596000 3,378,034000 3,612,817000 3,512,849000 3,555,094000 3,582,041000 AVERAGE YIELD OF OATS IN COUNTRIES NAMED, 1898-1907 Bushels per Acre YEAR UNITED STATES J RUSSIA, EURO- PEAN2 GER- MANY 2 AUSTRIA z HUNGARY PROPER 2 FRANCE 1 UNITED KING- DOM1 Average (1888 to 1897) 25-7 16.8 36-9 23-9 25-3 29.2 43-r 1898 28 4 4.6.1 1800 . 48.0 3O.2 33.3 27.8 44.2 1900 3 * 48 o 28.1 25.7 43. C IQOI . 2C.8 44.6 2^.6 28.1 23. <; 42. 0 1902 1903 IQO4 . 34-5 28.4 32.1 21.8 17.7 2<;.7 50.1 51.2 46.2 27.7 28.3 24.3 34-o 34-5 21;. 6 29.2 31.6 27.2 48.3 44.2 44-2 loot; . 20. 2 43.6 27.7 31. 1 28.6 43. Q 1906 31.2 23-7 19.7 55-7 58.2 34-1 35-7 34-3 29.7 27.0 31-8 46.1 45-1 Average (1898 to 1907) 29.8 19.4 49-3 28.6 3°-9 28.1 44-7 1 Winchester bushels. Bushels of 32 pounds. 6i6 APPENDIX BARLEY BARLEY CROP OF COUNTRIES NAMED, 1903-1907 COUNTRY 1903 (Bu.) 1904 (Bu.) 1905 (Bu.) 19O6 (Bu.) 1907 (Bu.) United States . . Canada .... Mexico .... Total North 131,861000 39.°35°00 9,061000 139,749°°° 42,244000 7,355°°° 136,651000 45,389000 7,000000 178,916000 50,820000 7,000000 153,597°°° 45,235°°° 7,000000 America . . 179,957°°° 189,348000 189,040000 236,736000 205,832000 Austria .... Hungary proper . France .... Germany . . . Russia (European) Great Britain — England . . . Scotland . . . Wales .... Ireland .... 73,873000 64,577000 43.345°°° 152,653°°° 350,486000 50,628000 7,739000 2,981000 6,076000 66,815000 49,915000 38,338000 135,409000 339,717000 48,511000 7,408000 3,077000 5,478000 70,469000 62,453000 40,841000 134,204000 338,836000 48,778000 8,257000 2,906000 7,181000 76,024000 69,747000 36,538000 142,901000 304,276000 51,543000 7,803000 3,1 16000 7,21 1000 78,548000 63,078000 45,°95°°° 160,650000 344,104000 51,912000 7,466000 2,885000 6,995000 Total United Kingdom . . 67,424000 64,474000 67,122000 69,673000 69,258000 Total Europe . 93I,758°°° 841,070000 867,392000 907,895000 911,451000 Cyprus .... Japan .... 3,969000 3,122000 2,980000 2,778000 3,000000 90 544000 Russia (Asiatic) . 6,984000 6,538000 8,130000 7,763000 9,345000 Total Asia . . 70,728000 90,512000 88,596000 94, 55 8000 102,030000 Africa .... 50,987000 52,097000 35,7°3°°° 44,102000 44,205000 Australia .... New Zealand . . 1,184000 1,172000 2,740000 1,197000 2,084000 1,164000 1,945000 1,056000 2,319000 1,068000 Total Australasia 2,356000 3,937000 3,248000 3,001000 3,387000 Grand total . . 1,235,786000 1,176,964000 1,183,979000 1,286,292000 1,267,814000 AVERAGE YIELD OF BARLEY IN COUNTRIES NAMED, BUSHELS PER ACRE, 1898-1907 YEAR UNITED STATES l RUSSIA, EURO- PEAN * GER- MANY1 AUSTRIA ! HUNGARY PROPER 2 FRANCE * UNITED KING- DOM l Average (1888 to 1897) 23.2 12.6 27.6 20. 2 20.3 21-5 34-4 1808 1800 11 8 « 8 1900 21 8 IQOI 12 "7 IOO2 ic 6 IOOT ?f> i 24 8 IQO4. 22 8 IQ 8 1905 26 8 1906 28 i 26 8 20 8 l6 2 1907 .... 23 8 ,Q , ' if, 8 *4"4 Average (1898 to 1907) 25-5 *3-7 34-4 23-9 23.8 22.9 34-9 1 Winchester bushels. 2 Bushels of 48 pounds. APPENDIX 617 RYE RYE CROP OF COUNTRIES NAMED, 1903-1907 COUNTRY 1903 (Bu.) 1904 (Bu.) 1905 (Bu.) 1906 (Bu.) 1907 (Bu.) United States . . Canada .... Mexico .... Total No. America Austria-Hungary * . France .... Germany .... Russia (European) United Kingdom . Total Europe . . Russia (Asiatic) Total Asia . . Australia .... New Zealand . . Total Australasia Grand total . . 29,363000 3,915000 136000 27,242000 2,995000 67000 28,486000 2,748000 70000 33,375°°° 2,273000 70000 31,566000 2,002000 70000 33,414000 30,304000 31,304000 35,718000 33,638000 132,267000 57,951000 389,923000 879,883000 2,000000 i37»963oo° 52,141000 396,075000 977,981000 2,000000 151,641000 58,116000 378,204000 708,692000 2,000000 IS3,SIS°°° 50,429000 378,948000 638,675000 2,000000 129,234000 58,578000 384,150000 776,000000 2,000000 I,S94,37°000 1,681,280000 1,436,406000 1,371,881000 1,479,851000 32,059000 30,457000 28,750000 28,169000 32,000000 32,059000 30,457000 28,750000 28,169000 32,000000 78000 40000 131000 2IOOO 85000 33000 94000 65000 89000 43000 i 18000 I52OOO 118000 159000 132000 1,659,961000 I,742,I93OOO 1,496,578000 1,435,927000 1,545,621000 AVERAGE YIELD OF RYE IN COUNTRIES NAMED, 1898-1907 (Bushels per Acre) YEAR UNITED STATES RUSSIA, EURO- PEAN GER- MANY AUS- TRIA HUN- GARY PROPER FRANCE IRE- LAND Average (1888 to 1897) . 13-5 10. 0 19.0 iS-S 16.3 17.1 25-4 1898 n.6 18 i 1800 14.4 12.8 21 t 18 7 17 7 18 2 2S 8 IQOO ic. i 12.7 I C I 2^.7 IQOI IST IO.3 i«.8 16 7 IQO2 .... 17.0 12. <; 18.2 28.1 IOO1 18 2 18 2 18 i IQO4. 16 6 1905 16 e 18 s IOO6 16 7 8 8 19 8 16 ? 27 6 !9°7 16.4 10.8 25-7 18.8 16.2 18.2 27.0 Average (1898 to 1907) . 15.8 "•5 24.6 18.1 17-5 17.2 26.7 NOTE. The student may well remember that the reported crop yields, acreage, and production are based upon estimates, while the statistics for im- ports and exports are based upon definite data. In census years the estimates are made with greater care and detail, but even these are largely estimates. In some cases the annual estimates are undoubtedly far from the facts, as is strongly suggested by comparing the federal "statistics" with the "statistics" 6i8 APPENDIX POTATOES POTATO CROP OF COUNTRIES NAMED, 1902-1906 COUNTRY 1902 (Bu.) 1903 (Bu.) 1904 (Bu.) 1905 (Bu.) 1906 (Bu.) United States . . Canada .... Mexico .... Newfoundland 284,633000 51,206000 347000 1350000 247,128000 56,944000 539000 1,350000 332,830000 55,436000 527000 1,350000 260,741000 55,257000 400000 1,350000 308,038000 59,804000 400000 1,350000 Total North America . . 337,S36oo° 305,961000 390,143000 317,748000 369,5Q2ooo Chile .... 11,616000 10,349000 6,131000 6,532000 6,532000 Austria-Hungary . France .... Germany . . . Russia (European) 584,619000 441,055000 1,596,969000 1,028,036000 544,166000 426,422000 1,576,361000 887,600000 520,461000 451,039000 I,333,326o°° 893,908000 765,117000 523,876000 i,77S,S79°°° 1,032,888000 709,237000 372,076000 i.577,653°°° 939,717000 United Kingdom: Great Britain . Ireland . . . 119,250000 101,761000 108,779000 88,227000 133,961000 98,635000 140,474000 127,793000 128,005000 99,328000 Total United Kingdom 221,01 IOOO 197,006000 232,596000 268,267000 227, 333°°° Total Europe . 4,280,644000 4,038,566000 3,843,081000 4,779,59°°°° 4,3°5-3I3°°° Japan .... Russia (Asiatic) . 7,4l8oOO I3,I42OOO 9,824000 19,364000 11,274000 18,800000 16,255000 18,865000 16,255000 16,481000 Total Asia . . 20,560000 29,188000 30,074000 35,120000 32,736000 Total Africa . 3,884OOO 3,541000 4,048000 4,071000 4,138000 Australia . . . New Zealand . . I2,039OOO 7,72IOOO 14,973°°° 7,215000 16,777000 7,795000 11,071000 5,025000 10,016000 4,607000 Total Australasia I9,76OOOO 22,188000 24,572000 16,096000 14,623000 Grand total . . 4,674,OOOOOO 4,409,793000 4,298,049000 5,159,157°°° 4,732>934°°° from some of the states which have crop-reporting boards, such as the Illinois State Board of Agriculture and the Iowa Weather and Crop Service. Thus in 1907, Illinois produced, according to the federal report, more than 40 million bushels of wheat, but less than 25 million bushels by the state report. In 1908 the federal estimate gave Illinois 9,450,000 acres of corn, while the state report was 6,780,000 acres. The same year the state of Iowa claimed to produce only 4,968,250 bushels of wheat, but in the federal report Iowa receives credit for 8,068,000 bushels. These are extreme variations, but they should serve to emphasize the fact that crop "statistics" should not weigh heavily as against actual data, such, for example, as the records of the long-continued field experiments of Rothamsted, Pennsylvania, etc. On the other hand, the United States crop statistics are probably as good as those from any nation; and, while gross errors may appear in specific instances, APPENDIX 619 RICE RICE CROP OF COUNTRIES NAMED, 1902-1906 [Mostly cleaned rice. China, which is omitted, has a roughly estimated crop of 50,000,000,000 to 60,000,000,000 pounds. Other omitted countries are Afghanistan, Algeria, Brazil, Colombia, Federated Malay States, Madagascar, Persia, Russia (Asiatic), Trinidad and Tobago, Turkey (Asiatic and European), Venezuela, and a few other countries of small production.] COUNTRY 1903 (Lb.) 1903 (Lb.) 1904 (Lb.) 19O5 (Lb.) 1906 (Lb.) United States: Contiguous . . Hawaii .... Mexico 319,400000 33,400000 40,000000 560,800000 33,400000 48,700000 586,000000 33,400000 62,000000 378,000000 33,400000 . 62,000000 496,000000 33,400000 62,000000 Total No. America 401,600000 652,000000 690,800000 482,800000 600,800000 TotalSouth America 85,600000 87,500000 95,100000 97,300000 120,500000 Italy Spain 3 5 q. 800000 417,100000 394,600000 478,800000 475,400000 Total Europe . . 1,038,300000 1,188,500000 1,167,500000 1,166,500000 1,215,000000 British India: British Provinces Native States . Japanese Empire Total Asia . . Total Africa . 72,688,000000 799,000000 13.295,300000 105,075,200000 22,200000 68,580,000000 838,000000 16,809,200000 104,887,800000 22,200000 71,561,000000 764,000000 18,658,700000 110,212,200000 22,200000 67,916,000000 640,000000 14,639,200000 102,147,900000 21,800000 67,464,000000 640,000000 17,185,900000 104,974,000000 21,800000 Grand total . . 106,626,400000 106,841,000000 II2,IOO,80000O 103,919,100000 106,943,900000 they doubtless approximate the truth as a general rule, and are more trust- worthy, especially for purposes of comparison, than the state estimates. It should be kept in mind that certain crops, such as wheat, are now quite regularly fertilized in some of the Eastern and East Central states. Comparison of crop yields in different states is most significant when the acreage is also comparable. In 1907 the average yield of wheat per acre was 23 bushels in Vermont and only 1 1 bushels in Kansas ; but Vermont raised one thousand acres and Kansas raised six million acres, and on many thousand acres the Kansas yield may have exceeded 30 bushels per acre. In 1905 the average yield of corn per acre was 42.7 bushels in Connecticut and only 39.8 bushels in Illinois; but Connecticut raised only 55,595 acres of corn, while Champaign County, Illinois, raised some 200,000 acres of corn, which made an average yield of 65 bushels per acre. South Carolina is authentically credited with having produced 239 bushels of air-dry corn on one acre of land in one season, but the average yield for the state for the 44 years, 1866 to 1909, is 10 bushels per acre. Even the average yield of corn for the United States has varied from 30.8 bushels per acre, in 1872, to 16.7 bushels, in 1901 ; and if these records were interchanged the average yield would become the same for two successive 2o-year periods, 25.0 bushels per acre. 620 APPENDIX SUGAR SUGAR PRODUCTION OF COUNTRIES NAMED, 1903-1904 TO 1907-1908 [European beet sugar, as estimated by Licht; United States beet sugar, from reports of Department of Agriculture on the Progress of the Beet-sugar Industry in the United States; production of British India (except 1907-1908) and of Formosa and Natal prior to 1905-1906 from official statistics; other data, from Willett & Gray. The estimates of Willett & Gray do not include the production of China, and some other less important sugar-producing countries.] COUNTRY 1903-4 (Tons1) 1904-5 (Tons1) 19O5-6 (Tons1) 1906-7 (Tons1) 1907-8 (Tons1) CANE SUGAR United States : Contiguous — Louisiana 228477 19800 328103 130000 335000 15000 380576 145000 330000 I2OOO 383225 213000 230000 13000 392871 2IOOOO 335ooo I20OO 42OOOO 217000 Texas Noncontiguous — Hawaii Porto Rico Total United States (except Philippine Islands) . . . Mexico 706380 875576 938225 845871 984000 107547 1,040228 2,143911 107038 1,163258 2,410477 107529 1,178749 2,545l82 JOSOOO 1,427673 2,680175 II5OOO 1,200000 2,6090OO Cuba Total North America . . . Total South America . . . Total Asia 601134 590382 700001 6lOI5I 58600O 2,876671 3,333672 2,926209 3,455446 3,48l477 Total Africa Total Oceania 355747 25J340 317967 349000 270000 163328 216213 230000 249000 27600O Total cane-sugar production BEET SUGAR United States 6,168791 6,820676 6,735°81 7,360172 7,233477 214825 6710 216173 8034 279393 11419 43J796 11367 413954 7943 Canada Total North America . . . Austria-Hungary .... 22IS35 224207 290812 443l63 421897 I,l67959 209811 804308 1,927681 I2355i 1,206907 441116 889373 176466 622422 1,598164 i3655i 953626 332098 1,509870 328770 1,089684 2,4i5r36 207189 968000 415000 1,344000 283000 756000 2,238000 181000 1,470000 445000 1,460000 235000 725000 2,135000 175000 1,410000 435000 Belgium France Germany Netherlands Russia Other countries Total Europe . . . 5,881333 4,708700 6,933649 6,717000 6,575000 Total beet-sugar production Total cane and beet sugar . 6,102868 4,932907 7,224461 7,160163 6,qg68g7 12,271659 ", 753583 13,959542 i4,52°335 !4,230374 1 Tons of 2240 pounds, except beet sugar in Europe, which is shown in metric tons of 2204.622 pounds. APPENDIX 621 SUGAR — Continued PRODUCTION OF SUGAR IN THE UNITED STATES AND ITS POSSESSIONS, 1874-75 TO 1907-8 T^TTTTT ( >ANE SUGA R YEAR OAAl SUGAR (Long Tons) Louisiana (Long Tons) Other Southern States (Long Tons) Porto Rico (Long Tons) Hawaii (Long Tons) Philippine Islands (Long Tons) TOTAL (Long Tons) 1874-1875. . . ] (60047 3454 72128 11197 126089 273015 1875-1876. . . _ _ 1 72954 4046 70016 11639 128485 287240 1876-1877 . . . IOO ' 85122 3879 62340 11418 121052 283911 1877-1878 . . . J 65671 5330 84347 I7I57 120096 292701 1878-1879 . . . 2OO 106908 5090 76411 21884 129777 340270 1879-1880 . . . I2OO 88822 3980 57057 28386 178329 357774 1880-1881 . . . 500 121867 5500 61715 41870 205508 436960 1881-1882 . . . \ , / 71373 5000 80066 50972 148047 355958 1882-1883 . . . j 5°° 1 I35297 7000 77632 51705 193726 465860 1883-1884. . . 535 128443 6800 98665 63948 120199 418590 1884-1885 . . . 953 94376 6500 70000 76496 200997 449322 1885-1886 . . . 600 127958 7200 64000 96500 182019 478277 1886-1887 • • • 800 80859 4535 86000 95000 169040 436234 1887-1888 . . . 255 I5797I 9843 60000 I 00000 158445 486514 1888-1889 . . . 1861 144878 9031 62000 I2OOOO 224861 562631 1889-1890 . . . 2203 124772 8i59 55000 I200OO 142554 452688 1890-1891 . . . 3459 215844 6107 50000 I25OOO I36035 536445 1891-1892 . . . 5356 160937 4500 70000 "5598 248806 605197 1892-1893 . . . 12018 217525 5000 50000 I4OOOO 257392 681935 1893-1894. . . J995° 265836 6854 60000 136689 207319 696648 1894-1895 . . . 20092 3*7334 8288 52500 131698 336076 865988 1895-1896 . . . 29220 237721 4973 50000 201632 230000 753546 1896-1897 . . . 37536 282009 5570 58000 224218 2O2OOO 809333 1897-1898 . . . 40398 3I0447 5737 54000 204833 I78OOO 793415 1898-1899. . . 32471 245512 3442 53826 252507 93000 680758 1899-1900 . 72944 147164 2027 35000 258521 62785 578441 1900-1901 . . 76859 270338 2891 80000 321461 55400 806949 1901-1902 . 164827 321676 3614 85000 3!75°9 78637 971263 1902-1903 . . . 194782 329226 3722 85000 391062 9OOOO ,093792 1903-1904 . . . 214825 228477 2 i 9800 130000 328103 84OOO ,005205 1904—1905 . 216173 335000 2 15000 145000 380576 106875 ,198624 1905-1906 . . . 279393 330000 I2OOO 213000 383225 M5525 ,363M3 1906-1907 . . . 431796 230000 2 13000 2IOOOO 392871 I45OOO ,423167 1907-1908 . . . 413954 335ooo 2 I2OOO 2I7OOO 420000 135000 ,532954 1 Production uncertain; not exceeding quantity stated. 2 Texas. 622 APPENDIX INTERNATIONAL TRADE IN SUGAR, 1902-1906 EXPORTS COUNTRY YEAR BEGIN- NING 1902 (Lb.) 1903 (Lb.) 1904 (Lb.) 1905 (Lb.) 1906 (Lb.) Austria-Hungary . Argentina . . . Belgium .... Brazil .... British Guiana British India . . China .... Cuba: . . . . Dutch East Indies Egypt .... Formosa . . . France .... Germany Mauritius . . . Netherlands . . Peru Jan. i Jan. i Jan. i Jan. i Apr. i Apr. i Jan. i Jan. i Jan. i Jan. i Jan. i Jan. i Jan. i Jan. i Jan. i Jan. i Jan. i Jan. i Jan. i Apr. i 1,500,882186 91,919510 296,287771 301,498062 280,284480 55,645520 89,945867 1,781,561643 1,904,371591 98,521149 100,873254 804,993320 2,367,596256 331,172713 310,694069 258,738790 217,486869 50,077330 288,610934 105,861392 617,792000 1,564,437691 66,888231 257,180695 48,256967 282,125760 57,474592 39,890000 2,118,279646 1,907,867945 86,469803 54,128545 469,129814 2,249,141034 375,505049 287,238939 281,482880 188,114307 107,862584 540,418988 90,460944 609,680000 1,125,102823 40,368833 406,944665 17,331526 239,043840 57,211504 48,787467 2,450,166945 2,318,243282 50,620531 79,518816 636,360461 1,720,574091 435,923559 403,476558 290,916853 191,917567 80,432029 398,854898 106,573936 569,646000 1,265,791878 4,847964 304,193682 83,216786 261,072000 64,546944 69,228800 2,412,915391 2,314,655085 67,821106 93,930689 658,062149 1,636,803746 361,987596 215,001603 295,935805 239,196273 220,925074 81,179056 901,932101 1,631,529629 233690 462,976753 187,278992 257,490912 58.660896 23,106000 2,643,700975 2,197,208868 10,495854 93,930689 617,793487 2,671,881051 410,917817 356,157015 295,935805 285,39474? 4i,433i35 206,854118 100,809856 952,547723 Philippine Islands Reunion .... Russia .... Trinidad and Tobago . . . Other countries . Total .... 11,854,814706 11,682,034414 11,677,016184 11,594,676863 13,506,338012 IMPORTS Australia . . . Jan. i 208,551056 205,026640 85,198624 55,923056 94,137680 British India . . Apr. i 549,868704 672,147168 724,262224 862,453200 1,090,152784 Canada .... Cape of Good Hope July i Jan. i 388,370832 120,365406 390,544660 104,629048 346,752590 101,468941 448,962523 82,805094 423,689614 87,165626 Chile .... Jan. i 97,002936 115,467959 124,139619 75,610563 75,610563 China .... Jan. i 607,880000 435,7"467 509,959200 626,433333 896,422400 Denmark . . . Jan. i 42,051621 77,3745i6 82,865127 76,080072 45,254827 Egypt .... Jan. i 22,844441 16,920099 45,843510 86,880895 76,321099 Finland .... Jan. i 61,752745 72,691465 71,263531 73,772007 83,322752 France .... Jan. i 220,187363 288,073883 179,849557 179,460755 222,562321 Italy . Tan 00 t J Japan .... Netherlands . . New Zealand . . Jan. i Jan. i Jan. i 47,3555O1 351,750533 248,799655 84,878074 *4> 477532 523,131067 203,061092 88,197686 4,928873 547,300400 208,329129 91,841944 11,251729 289,129733 167,742700 89,439230 31, 832317 104,816933 118,406076 93,329376 Norway .... Jan. i 82,791956 83,524155 76,703054 77,993596 80,364138 Persia .... Mar. 2 1 167,114080 179,412238 154,815921 154,217415 154,217415 Portugal Jan. i 63,630016 68,765610 72,490231 70,011389 70,011389 Singapore . Switzerland . . Turkey .... Jan. i Jan. i Jan. i 93,271733 180,272161 273,612826 102,369867 192,015742 273,612826 114,407600 175,444701 273,612826 117,958267 192,011994 273,612826 117,958267 187,653456 273,612826 United Kingdom . United States . . Uruguay . . . Other countries . Jan. i July i July i 3,440,232768 4,216,108106 43,235210 312,617000 3,099,985504 3,700,623613 39,934265 361,562533 3,409,501648 3,680,932998 49,814318 383,861800 3,099,597648 3,979,331430 33,838445 587,122591 3,420,616976 4,391,830975 33,838445 552,306790 Total .... 11,924,544723 11,309,260635 11,515,588366 11,711,640491 12,725,444045 APPENDIX 623 BUTTER INTERNATIONAL TRADE IN BUTTER, 1902-1906 EXPORTS COUNTRY YEAR BEGIN- NING 1902 (Lb.) 1903 (Lb.) 1904 (Lb.) 1905 (Lb.) 1906 (Lb.) Argentina . . . Australia . . . Austria-Hungary . Belgium .... Canada .... Jan. Jan. Jan. Jan. July 9,094269 7,77797i 15^589764 5,643r78 34,128944 1,175094 30,901910 13,728181 4,492080 24,568001 11,672157 64,788542 n.233431 4,340012 31, 764303 11,890040 55,90415! 8.944I51 3,800594 34,031525 9,712076 75,765536 7,740648 3,704232 18,243740 Denmark . . . Finland .... France .... Germany . . . Italy Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. 153,808614 21,315888 53,879727 4,849727 13,420636 5o,4i3634 28,447776 3,OI557° 83,463073 44,213494 176,664571 22,700563 59,7!4579 2,796343 14,176381 5i,659I35 3I,93l872 2,717219 90,863488 44,248776 179,745595 26,891790 49,842670 1,766564 12,375425 52,053041 35,208320 3,367075 87,705713 43,144662 176,081731 35,1359°! 49,781584 1,834907 I3,359?89 51,162980 34,240864 3,612714 86,966484 40,636298 175,043639 33,192114 39,307325 95I5I5 10,746430 56,404861 35,865200 3,281403 114,369238 35,7i28i7 Netherlands . . New Zealand . . Norway .... Russia .... Sweden .... United States . . Other countries Total . . . July i 8,896166 2,91 1000 10,717824 2,982000 10,071487 2,457000 27,360537 3-952034 12,544777 3,726146 540,869431 596,613867 628,427787 638,696284 636,311697 IMPORTS COUNTRY YEAR BEGIN- NING 1903 (Lb.) 1903 (Lb.) 1904 (Lb.) 1905 (Lb.) 1906 (Lb.) Australia . . . Belgium .... Brazil .... Cape of Good Hope Denmark . . . an. _ an. _ an. _ an. ' an. 6,901779 7,375362 6,270893 6,341566 15,432354 1,887148 9,788817 5,496134 6,055075 12,786808 43873 9,727714 5,642179 5,294516 13,007270 592201 10,054979 6,567718 5,25I72i 12,566345 7oi43 11,128520 5,344412 4,681766 I3>°49158 Dutch East Indies Egypt .... France .... Germany . . . Natal .... Jan. Jan. Jan. Jan. Jan. 2,788108 2,199657 12,042518 36,794039 1,662002 2,945909 2,366386 10,260344 53,558205 2,121121 3,021377 3,126945 10,067424 75,705838 3,I7i875 2,957073 3,066949 10,066650 79,524904 2,142003 3,049962 2,958784 11,402808 80,896179 2,142003 Netherlands . . Russia .... Fweden .... Switzerland . . . Transvaal . . . Jan. Jan. Jan. Jan. Jan. i,5i4533 856054 1,148959 9,7o5i87 3,269411 2,665917 838214 919839 10,970199 5,119642 5,858391 i,!5839° 1,305925 10,889289 4,514468 5,439836 1,103318 911993 n,955445 4,73M33 5,630865 577805 1,316117 7,882660 4,73J433 United Kingdom . Other .... Total .... Jan. i 440,221264 13,984000 447,684496 14,563000 465,285968 12,295000 456,662976 17,009360 477,092448 18,968003 568,507686 590,027254 630,116442 630,604904 650,863066 624 APPENDIX COTTON COTTON CROP OF COUNTRIES NAMED, 1902-1906 [No statistics for Siam and some other less important cotton-growing countries, of 500 pounds, gross weight, or 478 pounds of lint, net] Bales COUNTRY 1903 (Bales) 1903 (Bales) 1904 (Bales) 1905 (Bales) 1906 (Bales) United States 10,630945 9,851129 13,438012 10,575017 13,273809 103910 168098 253271 253271 253271 Total North America .... 10,740345 10,028813 13,701054 10,840532 i3,539663 Argentina 17 26 142 495 IOOOO •?osooo 285000 22OOOO 270000 365000 Peru 38200 43776 45672 49190 58283 Total South America .... 3493°5 335i84 271674 326269 439866 Greece 8200 8200 8200 8200 8200 Italy 27OO 27OO 2700 2700 Turkey 8000 7000 6OOO 7000 7000 Total Europe 20596 19650 18710 19705 19713 British India, including native states China S,1^10 I 2OOOOO 2,995875 I,2OOOOO 3,028000 1,200000 3,320000 1,200000 3,505000 1,200000 Dutch East Indies 8267 12632 15367 13280 15944 I7OI2 16262 12370 9239 Korea 7OOOO 7OOOO 70000 70000 Persia 56282 71509 81931 81931 sccooo 612000 553727 Turkey, Asiatic 6OOOO 6OOOO 60000 60000 60000 Total Asia 5,OO48lI 4,961604 5,038995 5,395758 5-523976 British Africa 27 751 4563 54OQ 8892 Egypt I 3487 SO 1,316212 1,234984 1,440107 206 German Africa 1371 I4.8o 1738 62 62 Portuguese Africa 61 6 179 «i8 518 Sudan (Anglo-Egyptian) . . . 6517 6517 15097 19441 17782 Total Africa Total Oceania 1,216353 93 !, 356227 312 1,337811 123 1,262110 "33 1,469306 108 Grand total I7,33I5°3 16,701790 20,368367 i7,8445I7 20,992632 NOTE. The total area of hay grown in the United States varies from 40 to 50 million acres, the production varies from 50 to 70 million tons, the average value varies from 500 to 700 million dollars, and as a ten-year average (1900 to 1909) the yield is 1.44 tons per acre and the farm price $9.59 per ton. APPENDIX 625 COTTON — Continued COTTON ACREAGE, BY STATES, 1902-1907 106 [As reported by Bureau of Statistics, Department of Agriculture] STATE OR TERRITORY 1903 (Acres) 1903 (Acres) 1904 (Acres) 1905 (Acres) 19O6 (Acres) 1907 (Acres) 39864 38664 36000 35000 North Carolina .... 1,076350 1,155028 1,306968 1,085568 1,374000 i ,408000 Florida 253288 268666 1,617678 7, 8011578 Arkansas 1,901841 754811 1,925191 783196 2,051185 881341 1,718751 2,097000 1,950000 Missouri 59786 66496 794°3 66444 418184 9IOOO 71000 I Indian Territory .... 657533 702966 813642 816638 901000 ^2,196000 United States .... 27,114103 28,014860 30,053739 26,117153 31,374000 31,311000 PRODUCTION OF COTTON, IN SCO-POUND GROSS WEIGHT BALES, BY STATES, AND TOTAL VALUE OF CROP, 1902-1903 TO 1907-1908 [As finally reported by U. S. Census Bureau] STATE OR 1902-3 1903-4 19O4-5 1905-6 1906-7 19O7-8 TERRITORY (Bales) (Bales) (Bales) (Bales) (Bales) (Bales) 13862 North Carolina 549542 528707 703760 6l9I4I 579326 605310 South Carolina . . 925400 787425 1,151170 1,078047 876181 T, II922O Georgia 1,887853 1,815834 Florida 52386 68797 986221 Mississippi .... 1,443740 1,432796 1,798017 1,198572 1,530748 1,468177 882073 Texas 2,498013 2,471081 3,145372 2,541932 4,174206 2,300179 Arkansas .... 970205 734593 930665 619117 941177 774721 Tennessee .... 3I7I49 248096 329319 278637 306037 275235 Missouri .... 42255 37813 51570 42730 54358 36243 Oklahoma .... Indian Territory . . 193784 351598 186589 278347 335064 469254 326981 350125 487306 4IOS20 [862383 All other .... 1263 772 2019 1416 2270 2734 United States . . 10,630945 9,851129 13,438012 10,575017 13,273809 11,107179 Total value of crop $421,687941 $576,499824 $561,100386 $556,833817 $640,311538 $613,630436 626 APPENDIX INTERNATIONAL TRADE IN OIL CAKE AND OIL-CAKE MEAL, 1902-1906 EXPORTS COUNTRY YEAR BEGIN- NING 1903 (Lb.) 1903 (Lb.) 1904 (Lb.) 1905 (Lb.) 1906 (Lb.) Argentina .... Austria-Hungary . Belgium .... Canada .... Jan. Jan. Jan. July 18,984000 64,246433 128,843602 28,830032 19,989308 88,614781 137,066773 29,002624 29,019439 92,352938 145,834669 10,115392 29,277380 77,134433 160,163061 26,227376 20,524298 58,999874 176,470002 44,39736o Denmark .... Jan. 4,045586 8,682295 4,417928 5,676571 3,101969 Germany .... Italy an. 328,760326 375,254222 436,964238 307,800450 36l,59262I Netherlands . . . Russia United Kingdom . United States . . Jan. Tan. Jan. July 139,814583 850,095204 53,146240 1,679,394359 136,734208 1,028,500994 53,146240 1,503-232680 154,525289 1,084,331094 48,462400 1,894,577648 143,200470 977,376790 57,830080 1,918,171084 147,620993 1,152,431674 58,524480 2,063,732272 IMPORTS Austria Hungary Jan. 7,656432 21,750580 27,340840 26,469794 24,769590 Belgium .... Jan. 353,641510 421,696899 445,202134 448,216564 510,213668 Canada .... July 3,521616 3,8o8224 3,953376 2,308432 3,656912 Denmark .... Jan. 654,111347 776,875723 757,48l664 842,875492 846,259587 Dutch East Indies . Jan. 15,691801 15,977041 31,004951 19,075498 26,850775 Finland .... Jan. 12,594155 7,2O5I92 13,948954 11,179475 14,543404 France Jan. 238,507681 279,980299 292,015079 323,719234 237,725713 Germany .... Italy Jan. Jan. 1,074,490655 1,108,355853 I, 231,409255 1,285,529859 1,325,622674 T * Jan. 7,909522 78 l828OO 8 fi 5,209963 6 Netherlands . . . Jan. 55,550267 461,479090 476,967295 495,921130 110,074533 510,951427 564,097473 Sweden .... United Kingdom . Jan. an. 142,046653 861,678720 163,633913 811,708400 219,913686 823,034720 226,374498 779,368320 264,890580 797,115200 Total .... 2 1 ,898000 25,7O2OOO 54,076000 153,688134 112,894136 3,910,777440 4,2O2,27944O 4,484,750758 4,763,041223 4,870,551653 SECTION V METHODS OF SOIL ANALYSIS Collecting soil samples. After one has become familiar with the typical boring of the soil type to be studied, the sample is collected by taking borings from 10 to 20 different places, each of which should appear to be truly representative of the soil type. These borings, thoroughly mixed, should make a trustworthy sample for analysis. An auger about i£ inches in diameter, with the screw point and the vertical lips filed off, APPENDIX 627 is the most satisfactory implement to use. The stem may be cut in two and a steel rod of good quality welded in to make the auger about 40 inches long. Ordinarily, samples may well be taken in sets of three : the surface, or average plowed soil (o to 6f inches), the subsurface, or that which can possibly be moved with a subsoil plow (6| to 20 inches), and the subsoil (20 to 40 inches), corresponding to about 2 million, 4 million, and 6 mil- lion pounds, respectively, of ordinary soil. The surface boring is made and the hole enlarged about ^ inch in diameter, the soil all being saved. The subsurface boring is then taken and the hole again enlarged, but the extra soil is not saved. Finally the subsoil boring is taken and the soil saved from only one half (one groove) of the auger. This provides about equal quantities of soil from each stratum. Preparation of sample. The sample of soil after air-drying is pul- verized to pass through a sieve with round holes i mm. in diameter. Any gravel which does not pulverize as easily as the dried lumps of clay is weighed and its percentage determined, after which it is discarded. After thorough mixing, the sample is then placed in a tight jar and labeled for analysis. Dry matter. Five grams of soil are placed in a glass weighing tube fitted with glass stopper, which, with the stopper removed, is placed in a hydrogen bath and heated for five hours at a temperature of 105° to 107° C., the stopper replaced, and the tube allowed to cool in a desiccator. On weighing, duplicate samples should check within 5 mg. The results of all analyses are calculated to the dry basis as found by this determination. Reaction. The reaction of the soil is determined by the test sug- gested by Veitch (Bulletin 73, page 136, Bureau of Chemistry, United States Department of Agriculture). About 10 g. of soil are placed in a Jena flask with 100 cc. water, thoroughly shaken, and allowed to stand over night. Fifty cubic centimeters of the supernatant liquid are carefully drawn off and boiled down with a few drops of phenolphtha- lein in a Jena beaker to 10 cc., or until the appearance of a pink color which indicates alkalinity. If no color appears, the soil is either acid or neutral. In case the soil is acid, its acidity, calculated to calcium car- bonate required to neutralize, is determined; and in case it is alkaline, the carbonate carbon present is determined and calculated to calcium carbonate. Acidity. Place 100 g. of soil in a 400 cc. wide-mouthed bottle, add 250 cc. normal potassium nitrate solution, stopper, and shake continu- ously for three hours in a shaking machine or every five minutes by hand. Let stand over night. Draw off 125 cc. of the clear supernatant liquid, 628 APPENDIX boil ten minutes to expel carbon dioxid, cool, and titrate with standard sodium hydroxid solution (of which i cc. is equivalent to 4 mg. of cal- cium carbonate), using phenolphthalein as indicator. The acids and acid salts of the soil are difficultly soluble in water, but by treating with a salt solution, as potassium nitrate, a double decomposi- tion takes place, carrying acidity into solution. An equilibrium is reached, however, before this reaction runs to an end, and if, after having drawn off 125 cc. to titrate, 125 cc. of fresh potassium nitrate are added to the bottle and the bottle again shaken for three hours, 125 cc. drawn off will give a titration which is more than one half of the first. By continuing this process until the last 125 cc. shows practically no acidity, we have a series of titrations the sum of which represents the total acidity of the 100 g. of soil. It has been found by working with a number of different soils, that as an average the sum of such a series is 2\ times the first ti- tration. Consequently, when the sodium hydroxid is made up so that i cc. is equivalent to 4 mg. of calcium carbonate, and 125 cc. (which represents 50 g. of soil) are titrated, each o.i cc. required to neutralize corresponds to i mg. of calcium carbonate required by the 100 g. of soil, or to o.ooi per cent of calcium carbonate required by the soil tested. The titrations of duplicate samples should not differ more than 0.8 cc. for soil samples requiring less than 100 cc. NaOH. Carbonate carbon. Carbonate carbon, when present, is determined volumetrically in the apparatus used for total carbon, described and illustrated in the Journal of the American Chemical Society, Vol. 26, pages 294 and 1640, by treating the air-dried soil with dilute (i : i) hydro- chloric acid and measuring the gas evolved both before and after absorp- tion of carbon dioxid in an alkali pipette containing a 33 per cent solu- tion of potassium hydroxid. The size of sample used for this test varies (according to the amount of calcium carbonate present) between two and ten grams. Duplicate tests of ordinary soils not very high in inorganic carbon should check within 0.2 to 0.4 cc. These results are calculated to and reported as calcium carbonate present. Corrections must be made for pressure and temperature, and absorp- tion of carbon dioxid should be repeated to a constant reading; also the gas should be allowed to stand for three minutes before the initial and the final readings. Organic carbon. The total carbon of the soil is determined by means of Parr's apparatus * as modified by Pettit 2 to contain an absorp- tion pipette of potassium hydroxid. 1Jour. Am. Chem. Soc., Vol. 26, p. 294. 2 Ibid., p. 1640. APPENDIX 629 Two grams of ordinary soil (or \ to i g. of peaty soil) are placed to- gether with 10 g. of sodium peroxid in the Parr explosion bomb, 0.7 to i g. (ordinarily about 0.8 g.) powdered magnesium added to start combustion, the whole thoroughly mixed by shaking, and the charge exploded by means of a hot iron plug or an electric current. (No magnesium is used with soils high in organic matter.) The contents of the bomb are then washed into a beaker by means of a fine stream of hot water and brought to a boil to break up the coarse particles and expel as much oxygen as possible. It is then run from a separatory funnel into a flask containing dilute sulfuric acid (i H2SO4 to 2 H2O) and the gas collected in a measuring pipette. When all of the sample has been added to the sulfuric acid and boiled until it is decomposed, the flask is filled with water through the separatory funnel to force the last of the gas into the measuring pipette. After noting the volume, the carbon dioxid is absorbed in the potassium hydroxid pipette and the volume again read. In taking the initial and the final readings, the same precaution should be taken as for carbonate carbon. Duplicate samples should check within i cc. for every 100 cc. gas obtained, and corrections must be made for pressure and temperature. A blank determination must be run on the sodium peroxid, and this is best done by using first a 2-gram, then a i-gram sample of the same soil, calculating the amount of carbon in the reagents from the difference in results, e.g. Let x = carbon in reagents ; then, if 2 g. soil + x = 45 mg. C, and i g. soil + x = 25 mg. C, we get by multiplying the last equation by 2 2 g. soil + 2 x = 50 mg. C, and, subtracting the first equation from this, we get x = 5 mg. C Much better results can be obtained by determining the blank in this way than where no soil is used. The total carbon thus found minus the carbonate carbon is reported as organic carbon and is taken as a measure of the organic matter present in the soil. Nitrogen. Nitrogen is determined by the regular Kjeldahl method. Ten grams of soil (5 g. if high in nitrogen) are weighed into a Kjeldahl flask, 20 cc. sulfuric acid (more if necessary) and approximately .65 g. 630 APPENDIX metallic mercury added and the contents of the flask digested until color- less. Oxidation is completed by adding, while still boiling hot, powdered potassium permanganate until the solution is green. It is then allowed to cool and transferred with 250 cc. of nitrogen-free water to a copper flask of about 700 cc. capacity and enough strong alkali solution 1 added to more than neutralize the acid. The flask is then immediately con- nected with a still, the ammonia distilled off and collected in a flask containing a measured amount of standard hydrochloric acid. The ex- cess of hydrochloric acid is then titrated back with standard ammonium hydroxid, using lacmoid as indicator, and the amount of nitrogen in the soil calculated. A convenient strength of ammonia solution vis one in which i cc. is equivalent to .0032 g. nitrogen. Duplicates should check within 0.2 cc. A blank determination must be run, by using approximately .5 g. pure sugar instead of the soil sample, and a correction made for the nitrogen in the reagents used. Phosphorus. For the phosphorus determination the soil is decom- posed by heating with sodium peroxid as given on page 145, Bulletin 105, Bureau of Chemistry, U. S. Department of Agriculture. Five grams of ordinary soil are thoroughly mixed with 10 g. of sodium peroxid in an iron crucible of about no cc. capacity, the flame applied directly to the surface just long enough to start the action, the crucible covered, and the heating continued over a low flame for twenty-five minutes. The tip of the flame should just touch the bottom of the crucible and the heat be kept low enough so the peroxid will not fuse. In decomposing a soil very low in organic matter, such as some subsoils, o.i to 0.5 g. of powdered sugar should be added to favor the reaction. Peat soils are usually high in phosphorus, and 2\ g. are sufficient for the determination. Such soils, high in organic matter, will not fuse slowly when heated with peroxid, but by moistening the sample with 5 cc. of calcium acetate of sufficient strength to fix the phosphorus, the organic matter can be safely burned off, and after cooling, enough starch added to effect decomposition with sodium peroxid in the usual way. After decomposition, the sample is washed into a beaker, the coarser particles broken up, then transferred to a 500 cc. flask acidified with hydrochloric acid and boiled for five minutes. A little strong nitric acid is added to insure complete oxidation of the iron to the ferric condition. It is then allowed to cool and made up to volume. There should be no undecomposed soil in the bottom of the flask. The silica is allowed to 1 Containing 60 Ib. Greenbank's alkali and 800 g. potassium sulfid for each 30 litres of water. APPENDIX 631 settle over night, 200 cc. of the clear supernatant solution drawn off, and the iron, aluminum, and phosphorus precipitated by adding ammonium hydroxid to the boiling solution. If there is not enough iron present to give a very decided brown color to the precipitate, a little ferric chlorid should be added before precipitation to insure complete removal of the phosphorus from solution. The precipitate is filtered off, washed, and then dissolved with warm dilute nitric acid. It is then evaporated on the steam bath to dehydrate the silica, taken up with hydrochloric acid, heating if necessary, and the silica filtered off. The filtrate is evaporated to about 5-10 cc., care being taken that it does not go to dryness, as alumina and some silica are almost sure to separate out and cause trouble. It is then completely neutralized with ammonia, cleared up with nitric acid, approximately i g. of crystalline ammonium nitrate added, and the phosphorus precipitated at 4o°-5o° with 15 cc. ammo- nium molybdate solution, allowing it to stand on the water bath at this temperature for one to two hours, stirring occasionally for the first 15 or 20 minutes. It is then allowed to stand at room temperature over night, the precipitate filtered off through a double filter and washed with a tenth-normal solution of ammonium nitrate until free from molybdic acid and finally twice with cold distilled water.1 It is then removed together with the filter paper to a beaker, dissolved with a measured excess of standard potassium hydroxid solution, and the excess titrated back with standard nitric acid . A very convenient strength of potassium hydroxid solution is .83236 g. KOH per 100 cc. One cubic centimeter is then equivalent to 0.2 mg. of phosphorus. The nitric acid should be made equivalent in strength to the potassium hydroxid, and with these strengths of solutions, duplicates should check within 0.2 cc. A blank determination must be run, using no soil, and a correction made for the phosphorus found in the reagents. (Ammonium molybdate solution is made by dissolving 100 g. molybdic acid in 400 cc. NH4OH (sp. gr. .96) and adding very slowly to 1250 cc. HNO3 (sp. gr. i. 20), keeping the solution cool and well stirred.) Total potassium. This test is carried out as given on page 147, Bulle- tin 105, Bureau of Chemistry, Department of Agriculture. One gram of soil, one gram of ammonium chlorid, and eight grams of calcium 1 Molybdic oxid is often precipitated if the first few washings, while iron is still present, are done with either water or ammonium nitrate solution. This may be prevented by washing two or three times, until free of iron, with ammonium nitrate containing a little of the ammonium molybdate solution. 632 APPENDIX carbonate are fused as directed in Fresenius' " Quantitative Analysis," page 426, and by Hillebrand in Bulletin 305 of the United States Geo- logical Survey, where an illustration of the apparatus is given. The fused mass is transferred to a porcelain dish, slacked with hot water, finely ground with an agate pestle and transferred to a filter. After washing with about 600 cc. hot water, the filtrate and washings are run to dryness in a Jena beaker, taken up with hot water and again filtered, acidified with hydrochloric acid, concentrated to about 10 cc., and i^- cc. of a platinum chlorid solution (10 cc. containing i g. platinum) added. This is then evaporated to a sirupy consistency, taken up and washed about fifteen times with 80 per cent alcohol, three times with ammonium chlorid solution, and again fifteen times with alcohol. The precipitate is then washed through the filter with hot water into a platinum dish^ evaporated on the steam bath to dryness and heated in an air oven at 110° C. for an hour, cooled in a desiccator, and weighed. Duplicate samples should not differ more than 1.5 mg. in the final weight. A correction must be made for the amount of potassium in the reagents which is found by making a blank determination, using no soil. (Ammonium chlorid solution is made by dissolving 200 g. NH4C1 in 1000 cc. water and saturating with K2PtCl6.) Calcium. Five grams of soil (or less if high in calcium) are decom- posed by heating with 10 g. of sodium peroxid in an iron crucible, taken up with water and hydrochloric acid and made up to 500 cc., as for phosphorus. After being allowed to settle over night, 200 cc. of the supernatant solution are heated to boiling and precipitated from the hot solution with ammonia. The precipitate is filtered out on a 15 cm. filter and washed with hot water until but" a slight test for chlorids is given with silver nitrate. The filtrate is again evaporated to dryness and heated (to dehydrate any remaining silica), taken up with water and hydrochloric acid, brought to a boil, and ammonia added to pre- cipitate any remaining aluminum. The precipitate is filtered out on a small filter and washed with hot water. It should not be washed more than necessary to remove the chlorids, as the wash water carries alumi- num through into the filtrate. On heating this filtrate and allowing it to stand over night, more aluminum may be found to precipitate out. All of the aluminum must be removed by repeated precipitations. The solution is then made slightly alkaline with ammonia, brought to a boil, and to it is added slowly, while it is being stirred, enough concentrated ammonium oxalate solution to precipitate the calcium and to change the magnesium to the oxalate. After boiling until the precipitate has a granular appearance, it is allowed to stand three hours or longer, de- APPENDIX 633 canted into a filter, and washed twice by decantation. The precipitate in the beaker is then dissolved with a few drops of hydrochloric acid, a little water added, and the calcium reprecipitated, boiling hot, by adding ammonium hydroxid to slight alkalinity. A little ammonium oxalate is added, the solution allowed to stand as before, and filtered through the same filter; washed free from chlorids with hot water, the filter burned and the precipitate ignited in a blast until it ceases to lose weight, and weighed as calcium oxid (CaO). Calcium may be determined from the filtrate of the iron and aluminum precipitates in the phosphorus determination. More care should be taken, however, to wash the iron and aluminum precipitates free from chlorids than is necessary when phosphorus alone is to be determined. This filtrate must be run to dryness, and the silica dehydrated, then the procedure continued as given above. Magnesium. The filtrate from the calcium determination is concen- trated to about 50 cc. Then i g. ammonium persulfate is added, and the solution boiled for 3 minutes to precipitate manganese. After fil- tering the filtrate and washings are evaporated to dryness on the water bath and ammonium salts expelled by careful heating. The residue is taken up with 20 to 25 cc. hot water and about 5 cc. hydrochloric acid. After it has stood until all that will has gone into solution, it is made slightly ammoniacal and heated on the steam bath (or allowed to stand over night), filtered, washed with distilled water, and the filtrate and washings evaporated on the water bath to about 50 cc. If, as is sometimes the case, a precipitate forms on standing, this must be filtered out, washed, and the filtrate again evaporated to about 50 cc. It is then made acid and enough acid sodium phosphate added to precipitate all of the magnesium. While it is vigorously stirred, care being taken not to strike or rub the sides of the beaker, enough ammonia is slowly added to make it distinctly alkaline. It is then allowed to stand half an hour, 10 cc. of strong ammonia slowly added while it is again stirred vigorously, covered closely to reduce escape of ammonia, and let stand for 12 hours. The precipitate is then filtered and washed free from chlorids, using 2^ per cent ammonia water, the filter dried, burned at first at a moderate heat, then ignited intensely in the blast, and weighed as magnesium pyrophosphate (Mg2P2O7). 634 SECTION VI COMPOSITION OF SOME EUROPEAN SOILS THE data relating to the composition of European soils are very incomplete, and the analytical methods used have been far from uniform. A good compilation of these data from Germany, France, and the United Kingdom is contained in Bulletin 57 (1909) of the United States Bureau of Soils, giving principally the results secured by digesting the soils with strong acids. This compilation includes no nitrogen determina- tions, but the phosphorus, potassium, and calcium are usually given, and sometimes the magnesium, chiefly in terms of the oxids. In the Rothamsted laboratories, after previous ignition of the soil, very strong hot hydrochloric and nitric acids are employed in soil analysis, and probably this method is used quite generally in Great Britain. If so, the data for phosphorus will closely approach the total. In Germany cold hydrochloric acid is the common solvent used, and the results thus secured are not comparable with those of England. Four analyses by Burguy (Inaug.-Diss. Berlin, 1899) show as an aver- age 41,330 pounds of potassium (evidently total) in two million of the loess soil of North Germany. (The phosphorus content of this soil is not given.) About 450 analyses of German soils are reported in this compilation, but for the reason given above they signify but little to the student of permanent agriculture. Wohltmann, as Director of the Insti- tute for Soil and Crop Investigations, Bonn-Poppelsdorf, in. a report (1901) on "The Fertility-Invoice of West-German Soils" ("Das Nahr- stoff-Kapital West-Deutscher Boden"), shows that the cold acid which he used generally dissolved about one fourth as much potassium from soils as hot acid (which he also used for additional potassium deter- minations), but the proportion varied with different soils from about one seventh to one half. A trial with a single soil showed that digestion with hot acid for 12 hours, dissolved one third more phosphorus than digestion with cold acid for 48 hours ; and numerous other experiments have shown that as an average the ordinary 10 hours' digestion with hot hydrochloric acid will dissolve only 85 per cent of the total phos- phorus, and with some soils less than one half of the total is thus dis- solved. Wohltmann concludes from extended chemical and cultural investiga- tions that in West Germany soils which contain less than 1200 pounds APPENDIX 635 (in two million) of phosphorus soluble in cold hydrochloric acid are in need of phosphorus fertilizer (" ersatzbedtirftig in Phosphorsaure"). The compiled data from France include about 1550 soil analyses, but here also only the plant food dissolved by the acid used is reported, and no information is given concerning the strength of acid, time, or tem- perature. While these results may have some value for purposes of comparison among themselves, they are of little or no value for compari- son with the total amounts of plant food contained in other soils. Fur- thermore this great mass of data relates to the soil of only a few prov- inces. There are in all eighty-seven different provinces, or counties, in France, and 705 of the soil samples reported upon were collected in the one province of Aisne, while 674 others were collected in Pas-de-Calais and Loire-Inferieure, and 129 more in three other provinces. The remaining 42 samples represent six additional provinces, leaving seventy- five provinces from which no soil analyses are reported. Practically all of the 1550 soil samples were evidently collected about 1890 or before, and no information is given in the compiler's report to show whether they are supposed to represent good land or poor land, although in one case a single field is represented by analyses of 73 samples of soil. In considering the analyses of European soils, it may well be kept in mind that there are still to be found areas of "abandoned" land even in western Europe, and chemical analyses of these soils are often made before attempting to bring them back into agricultural use by means of fertilizers and manures. Thus the marked differences in the plant- food content of different soils in England may serve best as an index of the agricultural history of the farms with respect to the past use of bones, guanos, phosphates, etc., while in America such differences apply not so much to individual farms, fields, or plots (see Table 73, page 411), but rather to types of soil more or less modified in the older States by the general and almost invariable practice of gradual soil depletion. The data showing the phosphorus content of soils from Great Britain make a contribution of probable value, (i) because approximately the total amount is reported, and (2) because the soil formation is frequently recorded. In the case of Dorset County, the samples appear to have been collected in connection with some sort of systematic survey or classification, as indicated by the records and also the reference: "Fifth Annual Report on the Soils of Dorset, University College, Reading, 1903." The compiler has combined the calcium found in limestone (cal- cium carbonate) with that reported in other forms, so that the calcium data have too little value to justify their reproduction here. It may be 636 APPENDIX stated, however, that, of the 286 samples of soil reported below, 129 contained an amount of acid-soluble calcium which if present as car- bonate would represent 10 tons or more of limestone per acre in the plowed soil, and of these about 60 apparently contained more than 50 tons per acre of calcium carbonate, thus suggesting the British farmer's common appreciation of the importance of having limestone in the soil. The following table shows the phosphorus reported for each of these 286 samples of soil, and the data certainly indicate that Liebig's ac- count of the tendency (even then apparent) toward the accumulation of phosphorus in British soil was well founded. As a general average of all analyses, it will be seen that the soil of England now contains about twice as much phosphorus as the most common Illinois corn-belt land (brown silt loam), three times as much as the ordinary wheat-belt soil of southern Illinois (gray silt loam on tight clay), and from four to fifteen times as much as the depleted or abandoned lands of the Atlantic Coastal Plain (such as the Leonardtown loam and Norfolk loam, the latter belonging to a series of thirteen soil types already represented by surveyed areas aggregating about 10 million acres, of which, however, soil analyses have been reported for only two types, as shown in Table 22, page 138). The amount of phosphorus in 2 million pounds of surface soil varies in the Gault soils of Kent County from 330 to 2210 pounds; in the chalk soils from 820 (Kent County) to 6800 pounds (Dorset County) ; in the Kimmeridge clay from 810 pounds (Cambridgeshire) to 7760 pounds (Dorset County); and in the London clay from 460 pounds (Surrey County) to 4100 pounds (Dorset County). Contrasted with these vari- ations, the records 1 of analysis of 555 samples of Illinois soils, in- cluding surface, subsurface, and subsoil, show an extreme variation from 540 to 2780 pounds of phosphorus in 2 million pounds of soil, the late Wisconsin yellow-gray silt loam varying from 540 pounds in 2 million of the subsurface to 900 pounds in the surface, and the early Wisconsin black clay loam varying from 980 pounds in 2 million of the subsoil to 2780 in the surface. 1 University of Illinois Agricultural Experiment Station Bulletin 123 (1908), pp. 262-294. APPENDIX 637 PHOSPHORUS IN SOILS IN THE UNITED KINGDOM Pounds per Acre in 2 Million of Soil (about 6| Inches Deep) DESCRIPTION AND LOCALITY (Original Sample Nos. ) PHOS- PHORUS (Lb.) DESCRIPTION AND LOCALITY (Original Sample Nos.) PHOS- PHORUS (Lb.) ENGLAND Berkshire Button's seed trial grounds, Cheshire (i) 870 Reading ?2?0 (2) >_• i\j 2620 Cambridge Cumberland Hatley plot (3) Joint rotation (9) .... 3400 1920 Rose-bank plot (497) . . . (499) . . . 440 440 Burgoyn's (Univ. Farm) : TTlpl/^C T T T •} (501) . . . 440 neiub 11 13 Fields 14-15 I2IO Fields 16-17 1920 Dorset Fields 18-19 790 \lluvium (38) 3660 Bowlder clay: (62) . ^\j\j\^ 2OIO Above gault (14) . I22O (41) • 4OIO Above green sand (19) . . 990 T Above gault (20) .... 93° Gravel (36) 27QO Above gray chalk (21) . . 890 • Ijf* 244O Gault soils (3) 1 220 (60 T'T' 227O (7) IIIO (64) / I74O (8) 850 j*r* Kimmeridge clay soils (12) IT ft 1280 Sen Bagshot beds (83) .... (30) .... 1570 2l8o \ J. *) J • "ou Sin (40) .... 1740 7 Oi\J (65) .... 1050 Ampthill clay soils (6) ... 840 London clay (80) 2790 Oxford clay soils (10) . . . 1030 (12) 4IOO (n) • • • 1260 (68) 1310 (22) . . . I2OO Reading beds (81) .... 2880 Lower greensand soils (5) . 1780 (n) .... 3490 (23) • 2260 (67) .... I830 (18) . I72O (61) .... 960 (9) • 1470 [unction Reading beds and (6) I72O chalk (39) 3660 v*t j. ff,\^ jvxvy 638 APPENDIX PHOSPHORUS IN SOILS IN THE UNITED KINGDOM (Continued) (Pounds per Acre in 2 Million of Soil (about 6| Inches Deep) DESCRIPTION AND LOCALITY (Original Sample Nos.) PHOS- PHORUS (Lb.) DESCRIPTION AND LOCALITY (Original Sample Nos.) PHOS- PHORUS (Lb.) ENGLAND (Continued} Dorset (continued) Chalk (15) 2620 Dorset (continued) Kimmeridge clay (87) . . . 2960 (7i) 2ZT.O (88) . . . 4360 (99) 6800 (oO 4100 (51) 2880 (77) (%i\ 2790 (94) 4270 v«2J f»rA (35) 2090 \°9) (Q\ 5230 (50) 3T4° (o; (en\ 349° (16) 2880 (57) fc-Q.\ 3°5° (72) 21. 546, 547 preservatives and reenforcing materials, 256, 547 recovered in live-stock farming, 201 Manure in culture experiments, 256, 343 Manures (see farm manure and green manuring) Marl carbonates, 167 Marl phosphates, 241 Maryland, field experiments, 261 soils, 138, 141 subsoils, 73 Massachusetts field experiments, 278 Mellilotus for green manuring, 220 Mica, 49 Michigan, investigations with legume plants, 216, 221 soils, 97 Minnesota, soil experiments, 499 soils, 100, 138 Mississippi soils, 93 Missouri soils, 89, 138 Moisture factor in crop production, 577 Molecule, defined, 4 Montana soils, 102 Mountain soils, 122, 128 Nascent, defined, 4 National Fertilizer Association report on raw phosphate, 292 Nebraska rainfall, 331, 580, 582 soils, 89 Nevada soils, 102 INDEX 651 New Jersey, pot-culture experiments, 439 soil, 138 New York, investigation of phosphorus in wheat bran, 41 soils, 75 Nitrate of calcium, 526 Nitrate of sodium, 525 Nitrification, 195 Nitrogen, 36 Nitrogen and organic matter, 194 Nitrogen determination, 629 fixation by legumes, 207 by nonsymbioric bacteria, 224, 405, 437 from air and soil, 213 gain or loss difficult to determine, 499 in animal and plant products, 154, 602 in drainage waters, 557, 563 in fertilizers, 157, 517 in rain, 309 in roots and tops of legumes, 218 in sweet clover, 220 recovered in live-stock farming, 201 retained by animals, 201 used in different amounts, 374, 423 used to give crops a start, 218, 401, 533 Nitrogenous compounds, 38 Nomenclature, 19, 565, 599 North Carolina soils, 138, 142 North Dakota investigations of flax sick- ness, 319 North Platte, Nebraska, rainfall, 331, 580 Northern states soils, 97, 138 Nuclein, 40 Oats, composition, 75, 154, 603 statistics, 615 Ocean, composition, 13 Ohio, field experiments, 245, 441 investigations with manure, 547 soils, 88, 138 in pot-culture experiments, 513 Oil cakes, composition, 604 fertility loss, 205 statistics, 626 Oil seeds, composition, 603 Oils and fats, 34 Oregon soils, 102 Organic chemistry, 30 Organic matter, defined, 30 decomposition, 195 loss in digestion, 199 methods of supplying, 198 Organic matter and nitrogen, 194 Orthoclase, 47 Oxids, defined, 17 occurrence, 53 Oxygen, 26 Pacific coast soils, 102, 130, 138 Park field, Rothamsted, 391 "Parrot" instruction, 292 Pasturing land, 199 Peat, dried, 524 Peat soils, 75, 83, 98, 100, 470 Pennsylvania, field experiments, 263, 420 investigations with manure, 202, 203 soils, 76, 142 Pentosans, 31 Periodic law of chemical elements, 9 Permanent systems of agriculture, 159 Peter, on soil fertility theories, 339 Phosphate deposits, 597 Phosphate experiments: Canada, 505 .Illinois, 283, 504 Indiana, 296 Kentucky, 288 Maine, 275 Maryland, 261 Massachusetts, 278 Ohio, 245, 442, 448, 452 Pennsylvania, 263 Rhode Island, 266 Phosphate production, 595 raw rock and slag must be fine-ground, 239 in practical agriculture, 289, 504 Phosphate report by National Fertilizer Association, 292 spreader, 179 supply, 597 Phosphates, 52, 186 low-grade, 188, 242, 598 natural, 52, 186 used in Europe, 324 Phosphatic limestone, 52 marl, 241 slag, 192 Phosphorus, 40, 52, 183, 236 Phosphorus compounds, 189 determination, 630 in fertilizers, 157, 517 • in plant and animal products, 154, 602 in wheat bran, 41 production, 595 retained by animals, 201 supply, 597 use in different forms, 237 used in Europe, 324 Photosynthesis, 29 Physical condition of soil, 576 Piedmont soils, 121, 138 Plant and animal products, composition, 157, 602 Plant food, 26 available, 107, 314, 366 652 INDEX Plant food, essential, 12 in crops, 75, 154, 602 in culture experiments, 236, 343 in sea weed, 524 lost from manure, 200, 546, 547 from plants, 549 from soils, 411, 413, 418, 556 Plant food, recovered in live-stock farm- ing, 199 retained by animals, 201 sources and cost, commercially, 157, 517 value, 154 water-soluble, 141 Plot experiments for testing soils, 569, 570 Potassium, 42 Potassium chlorid and sulfate, 531 Potassium determination, 631 Potassium, from sea water, 531 in fertilizers, 157, 517 in plant and animal products, 154, 602 Potassium salts of Germany, 529 Potato experiments, 384, 447, 511 Potatoes, composition, 154, 604 statistics, 605, 618 Prairie soils, 78, 82, 132, 138 Prefixes in chemical names, 19 Proportions, law of constant, 8 Protective agents, 536 Protein and proteids, 37 Quartz, 49 Radicle, defined, 17 Rain, composition, 309 Rainfall and drainage records, 309, 377, 413, 491, 557, 580, 582 Rainfall averages for the United States, 582 Rainfall in dry farming sections, 580, 582 Ramsay, on composition of air, 13 Ramsay and Cameron, on transmutation of elements, n Rate of growth, 32 Residual soils, 54, 126, 146, 149 Residues of crops used in soil improve-- ment, 199 Rhode Island field experiments, 266 Rice, composition, 603 statistics, 619 Rock weathering, 49 Roman agriculture, 590 Root crops, composition, 417, 604 in Canadian experiments, 511 in Rothamsted experiments, 398 Roots and tops of legumes, composition, 218 Root tubercles, composition, 215 size, 212 Rotation crops grown in experiments: Agdell field, Rothamsted, 345 Illinois, 453 Louisiana, 495 Minnesota, 499 Ohio, 245, 256, 441 Pennsylvania, 421 Rotation crops, plant food required, 75 Rotation of crops and soil fertility, 318, 339. 362. 389. 435, 443 Rotation systems, 226, 231 Russia, agricultural conditions, 332 soils, 66 Rye, composition, 603 statistics, 617 Sachs, on availability of plant food, 109 Salt, common, 535 defined, 20 Salt deposits, 53, 529 • Sand soils, 80, 98, 100, 138, 468, 498 Science, defined, i Scotland, soils, 63, 642 Seed factor in crop production, 575 Semiarid section, rainfall records, 331, 580, 582 S6n6bier's discovery of carbon fixation, 3°7 Shale, 50 Shutt, on loss of organic matter and nitro- gen, 200, 559 Silicates in earth's crust, 47, 48 Silicon, 44, 46 Slag phosphate, 192 Snow, composition, 310 Soaps, 36 Sodium, 44 Sodium in fertilizer experiments, 364, 380, 402, 508 Soil analysis, methods, 626 classification, 54, 116 composition, 58, 138 depletion, 556 by natural agencies, 61 Soil fertility theories, 300, 362, 366, 385 formation, 54 materials, 55 provinces of the United States, 116 samples, method of collecting, 626 series, 116, 132 stimulants, 45, 158, 533 structure, 116 surveys, 57, 77, 114, 517, 555 texture, 116 types, 55 Soils of Africa, 66, 67 Canada, 103 central states, 77, 138 INDEX 653 Soils of eastern states, 72, 75, 138 Europe, 63, 634 India, 66 Northern states, 97, 138 Rothamsted, 63, 411, 416 Russia, 66 South America, 67 southern states, 92, 138 Transvaal, 67 Turkey, 67 western states, 101, 138 Solution law, 366 South American soil, 67 South Carolina, phosphates, 187, 595 record yield of corn per acre, 619 Southern states soils, 92, 138 Spencer, on farming in semiarid region, 58i Spillman, on Kansas crop yields, 330 Spreader for limestone and phosphate, 1 79 Starches, 31 Statistics of agricultural products, 605 Steatite, 48 Sterile soils, 63, 367 Stimulants, 368, 394, 402, 479, 508, 533 Straw, composition, 157, 603 Structure of soils, 116 Success in farming, 584 Sugar beets, 155, 399, 604 Sugar statistics, 620 Sugars, 31 Sulfates and sulfids, 52 Sulfur 39, 106, 158 in rain, 106 Superphosphate, 191 Supply and demand of plant food, 59 Swamp soils, 80, 583 Sweet clover, content of nitrogen and or- ganic matter, 220 Symbol, defined, 7 Systems of permanent agriculture, 159 Temperature factor in crop production, 576 Tennessee, phosphates, 187, 188, 595 soils, 64, 93, 138, 367 Terminations in chemical names, 19 Terrace soils, 124 Testing soils, 565, 626 Texas soils, 95, 138 Texture of soils, 116 Thaer's soil-fertility theories, 302 Theories concerning soil fertility, 300, 362, 366, 385 Timber soils, 79, 133 Tobacco, composition, 604 experiments, 288 statistics, 605 Transmutation of elements, u Transported soils, 54 Transvaal soils, 67 Tubercles on roots of legumes, 212, 215 Tubers, composition, 604 Tull's soil-fertility theories, 300 Turkish soil, 67 Turnips, composition, 417, 604 experiments, 346, 399 Utah, phosphates, 596 soils, 101 Valence, defined, 4 Value of land, 586 Van Helmont's soil-fertility theories, 300 Van Hise, on phosphates, 560 Vegetable fats and oils, 34 Vegetables, composition, 604 Vesuvius lava, composition, 67 Virginia, field results with raw phosphate. 289 soil, 138 Vital processes in plant growth, 33 Volcanic ash, 67, 138 Wales soil, 642 Washington soils, 102 Water (see moisture) Water-soluble plant food, 141, 513 Weathering of rocks and soils, 49, 61, 174 Webster's view of agriculture, 594 Western states soils, 101, 138 Wheat, composition, 75, 154, 417, 603 Wheat bran, composition, 41, 604 Wheat grown every year: Canada, 505 Jethro Tull, 306 Minnesota, 499 Rothamsted, 363 Whitney, on potatoes at Rothamsted, 389 Whitney and Cameron's soil-fertility theo- ries, 313, 362, 367, 385 Whitson, on loss of phosphorus from Wis- consin soils, 560 Widtsoe, on arid soils and vegetation, 101 Williams, on value of manure in China, 338 Wilson, on abandoned lands, 342 Wing, on use of limestone and raw phos- phate, 289 Wisconsin experiments, 216, 221, 289, 560 soils, 99, 138 Wood ashes, 531, 602 Wyoming phosphates, 595 soils, 102 Zein, 39 Zeolites, 50