BIOLOGICAL METHODS FOR DETERMINING FERTILIZER REQUIREEEETTS OF CROPS by Harold Blair Heeney x“ A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of - MASTER OF SCIENCE Department of Botany and Plant Pathology 1951 THESIS é/> 7/7 2 7, w/ _5 Acknowledgement The writer wishes to express his appreciation for the stimulating suggestions and valuable criticisms given during the course of this study by Dr. I. Lyle wynd, under whose direction this work'was done. The financial assistance of the Agricultural Institute of Canada which has made this study possible is gratefully acknowledged. 253H12 PEOPLE-PACKED PLANET ”If the United States seems to be filling up-«andfit's certain that Americans have less elbow room than they once did-~the phenomenon is not exactly unique. So is the world. New statistics gathered by the United Nations say that with emphasis. During the last three decades, according to U. N. 'pOpulatian experts,‘ the number of human beings on this planet has risen from an estimated 1,83h,000,000 to an estimated 2, 378,000,000. The total population of the globe rose by an.average of approximately 1 percent a year during that period. If this rate of increase is maintained there will be twice as many people a century from now as there are today. In the year 2051, then, there may be around h,750,000,000 individ- uals on this terrestrial ball-to use what sounds like a most appro-. priate phrase from an old hymn. That's a lot of humankind. It's fairly obvious that if the earth cannot support its present population in reasonable comfort, tremendous advance will be necessary if it is to feed, house and clothe twice as many people. Is such an advance possible? Merely to live in the United States is to realize that the world can be made to produce more abundantly than some pessimists can imagine. Americans have made their share of this continent support 150,000,000 people at an unprecedented standard of comfort. Perhaps the miracle can be accomplished. But future generations will inhabit a more crowded sphere. Those unfettered souls that must have the great Open spaces will have only the deserts or the oceans.'-wEditorial from the State Journal, Lansing. Michigan BIOLOGICAL mesons FOR DETERMINING FERTILIZER REQUIRmmITs or CROPS Table of Contents I. HISTORICAL BACKGROUND OF THE MODERN CONCEPTS OF FERTILIZER'USE...1 A. Early theories of the nutritional requirements of plants....1 1. 2. Introduction...........................................1 The Theory of the Four Elements........................2 The Theory of Water as the Nutrient of Plants..........3 The Theory of Earth as the Nutrient of Plants..........h The Theory of Humus as the Nutrient of Plants..........5 The Beginning of the Biochemical Approach to the Problem of Plant Nutrition...........................9 B. DeveIOpment of Biological Methods for the Determination Of Fertilizer RequirmentSOOOOOOOOOQO0.000.000.000000000012 1. 2. 3. Introduction..........................................12 The Law of the Minimum................................12 The Laws of the Minimum and Maximum Nutrient Content of Plants...................................20 Summary of the Work during the Last Half of the Nineteenth Century which was Instrumental in the Development of Modern Procedures....................2# a. Field Trials.....................................24 b. Pot and Solution Cultures........................25 c. Plant Analysis z§g§2g_Soil Analysis..............27 d. Early Methods Of Chemical Analysis...............29 II. FACTORS AFFECTING TEE ABSORPTION AND CONCEITRATION OF MIENTS IN ms..............................................32 A. Introduction..................................................32 B. Mechanism of Nutrient Absorption..............................32 C. Internal Factors Affecting the Concentration of Nutrients in Plants...................................................37 1. Species and variety......................................37 2. Tissues..................................................#2 3. Diurnal Rhythms..........................................h9 a. Nitrogen............................................h9 b. Phosphorus..........................................u9 c. Potassium...........................................54 d..Ash.................................................54 h. Season and Stage of DeveIOpment..........................55 a. Introduction........................................55 b. Nitrogen............................................56 c. Phosphorus..........................................65 d. Potassium...........................................68 D. External Factors Affecting the Concentration of Nutrients in Plants...................................................69 l. Climate..................................................69 a. Illumination........................................69 b. water...............................................7O c. Temperature.........................................73 d. weatherooooOOOOQoonQOOec9000000000000000000000000007“ 2. Pests and Diseases.......................................82 3. Management...............................................83 h. Soil.....................................................86 III. FIEED TRIALS............................. ....... .................88 A. Introduction.......................................:..........88 3. Statistical Analysis of Data..................................91 1. Introduction.............................................91 1 2. Experimental Area........................................93 3. Replication..............................................9# h. Randomization............................................95 5. Local Control............................................95 C. Applications..................................................96 1. Introduction.............................................96 2. Correlation of Crop Yields with Fertilizer Applications...........................................98 3. Correlation of Tissue Analyses with Fertilizer Applications..........................................100 h. Discussion..............................................10h IV. LAW'OF DIMINISHING-RETURNS......................................106 A. Introduction.................................................106 B. Development of the Law of Diminishing Returns................107 1. Basic Assumptions and the First ApproximatiOn...........107 2. Criticisms of Mitscherlich's Equation...................114 a. Major Criticisms...................................11u iv. b. Criticisms by Pfeiffer and his Co-workers..........1lh c. Criticisms by Briggs...............................119 d. Other Criticisms of the ”C" Factor.................123 3. Application to Multiple Factors.........................12h a. Work of Baule......................................121+ b. Wbrk of Balmukand..................................131 c. WOrk of Gregory....................................133 b. Second Approximation....................................135 5. Status at the Present Time..............................1h0 V. APPLICATIONS OF THE LAW OF DIMINISHING RETURNS...................148 A. Pot Cultures.................................................148 1. Mitscherlich's Technique................................148 2. Stephenson and Schuster Technique.......................156 3. Jenny Technique.........................................157 h. value of Pot Cultures...................................159 B. Field Techniques.............................................162 1. Introduction............................................162 2. Procedure of Magistad and Co-workers....................162 a. Introduction.......................................162 b. Calculation of Constants by the Method of Least Squares....................................165 c. Calculation of the Constants by the Simplified Procedure........................................l66 d. Determination of the Available Nutrient in the $011.0...0000OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO....168 V. e. Determination of the Profitable Amount of Fertilizer.......................................168 f. Discussion.........................................171 3. Procedure of Roger Bray.................................173 a. Introduction.......................................173 b. Correlation of the Potassium Test with Percentage Yield.................................175 c. Application Of the Baule Percentage Yield Concept to Several Nutrients at the Same Time............178 d. Correlation of Soil Analyses with Fertilizer Requirements.....................................180 e. Discussion.........................................184 4. Procedure Of Willcox....................................185 a. Introduction.......................................185 b. Procedure for Research Workers.....................186 (1) Preparation of the Standard or Universal Yield Diagram...............................186 (2) Interpretation of Field Plot Data............188 c. Procedure for Farmers..............................198 d. Discussion.........................................201 VI. USE OF SEEDLINGS FOR DMI’BMINING NUTRIENT DEFICIENCIES OF caors.............................................. ........ .....205 A. Introduction.................................................205 B. Neubauer Procedure...........................................205 10 TGChnique......................o......................00205 2. Discussion..............................................208 C. Thornton Procedure...........................................209 D. McGeorge Procedure...........................................210 E. Discussion.......................................... ....... ..210 VII. USE OF MICRO-ORGANISMS FOR THE DETERMINATION OF FERTILIZER REQUIREMENTS............................................ ..... ...21b A. Introduction.................................................21b B. Azotobacter...................... ...... ......................214 1. Technique...............................................21# 2. Discussion..............................................217 C. Aspergillus niggg............................ ..... ...........219 1. Introduction............................................Bl9 2. Niklas Procedure........................................220 3. Mehlich Procedure......................... ....... .......220 h. Mulder Procedure........................................2i21 5. Discussion..............................................222 D. Cunninghaggllgp........................... ..... ..............223 1. Technique...............................................223 2. Interpretation..........................................227 3. Discussion..............................................229 I. value of the Microbiological Methods.........................230 VIII. FOLIAR DIAGNOSIS...............................................231 A. Introduction.................................................231 B. Theory of Foliar Diagnosis...................................233 1. Experimental Basis......................................233 2. Definition of Terms.....................................235 3. NPR Unit....................................... ..... ....237 C. Technique of Application.....................................241 D. Applications...................................... ......... ..25h 1. Effects of Disease......................................254 2. Effects of Cultural Methods.............................25b 3. Effects of Nutrient Carriers............................255 4. Effects of Climate......................................260 5. Fertilizer Requirements.................................260 a. Introduction.......................................26O b. Montpellier Technique..............................261 c. Pennsylvania Technique.............................266 E. Discussion...................................................269 Ix. TRIPLE ANALYSIS.................................................273 A. Introduction.................................................273 B. Technique........................ ..... .......................274 C. Discussion...................................................277 X. CRITICAL CONCENTRATION..........................................281 A. Introduction.................................................281 B. Correlation of Critical Concentration.with Yields............283 1. Introduction............................................283 2. Determination of Critical Concentrations................289 3. Discussion..............................................30U 0. Nutrient Batios..............................................306 D. Presentation Of mta..0...l....0.........................O...310 E. Discussion....... ....... ................ ................. ....31h XI. LEAF PAINTING Aim INJECTION. ................. 316 A. Introduction......................... ......... ' ......... ......316 B. Technique.................... ..... ........ .................. .316 C. Applications. ...... .......................... ............. ...319 D. Discussion............. ......... ....... ...... , .......... ......320 III. VISUAL SYL?IUI~IS......... ...... ......... ..... ...................321 A. Introduction.................................................321 3. Application.......................... ....... ..... ....... .....321 C. Discussion...................................................323 XIII. WEEDS............................. ...... .... ................ ...327 XIV. CHLOROPEYLL....................................................330 XV. SUMMARY.......... ..... ........................ ........... ......335 m0 LITMTURE CITmecooooo oooooooo ooooooeoeoooc-ooaeoecoo oooooo 000338 BIOLOGICAL METHODS FOR DETERMINING FERTILIZER REQUIREMENTS OF CROPS I. HISTORICAL BACKGROUND OF THE MODERN CONCEPTS OF FERTILIZER USE A. Early theories of the nutritional requirements of plants I. Introduction Since man first planted a seed and produced a crOp, the problem of his conscious control of the conditions affecting plant growth has occupied his attention. The mysterious development of a vigorous plant from an apparently lifeless seed has excited the curiosity of mankind for hundreds of years. In recent time, the problem of assist- ing nature to provide for the plant's needs has become vitally im— portant because of the increase of pOpulation and the great complexity of its requirements. It is obvious to agricultural workers that the world acreage of usable agricultural soils is decreasing at an aston- ishing rate. Inquiry into the methods of determining the proper use of fertilizer for crOp production is one of the most important problems of modern agriculture. The current scientific approach to this problem was not deveIOped in recent time, but has evolved slowly over a period of at least two thousand years. while a narration of the errors of early scholars should not be given undue prominence, a knowledge of some of their efforts to understand plant growth is necessary if we are to fully comprehend the problems confronting us at the present time. Erroneous theories often have been the stepping stones to the discovery of scientific truth. The purpose of the present study is to inquire into the various biological methods for the determination of the fertilizer needs of crops. This study involves careful inquiry into how early work was in error, an evaluation of the current procedures, and the development of necessary improvements of these procedures. It is erroneous to assume that the present methods are necessarily correct because they are modern and that they satisfactorily serve the needs of agriculture. A re-investigation of the major aspects of the problem based on ori- ginal literature is essential if we are to avoid the prOpagation of faulty theories and inadequate technical procedures. 2. The Theory of the Four Elements Many of the fundamental aspects of the nutritional requirements of plants are now so well known that they are frequently taken for granted, without any realization that their historical development . involved the efforts of many pe0ple over a very long period of years. There is no precisely defined beginning of man's effort to understand the nature of plant growth, but certainly Aristotle's philosOphy that the four 'elements', fire, air, earth, and water, were the units comprising all matter might be considered the point of departure which led ultimately to our modern concepts. Aristotle believed that the union, or mixis, of the four “elements" in the soil generated the nutritive organic matter of plants in the form of exceedingly minute particles, or homoiomeria. These particles were supposedly assimilated through the roots of plants and deposited unchanged in tissues and organs. Aristotle thus laid the foundation for the work of many of his successors for his theory was the earliest expression of what is now known as the humus theory. The humus theory, or belief that plants were nourished by preformed organic matter in the soil, dominated the study of plant nutrition for over 2000 years. Dven at the present time, this concept, in modified form, is the basis of the so-called “organic gardening" theory of plant nutrition. 3. The Theory of Water as the Nutrient of Plants About 300 years ago, a Flemish physician and chemist named John Baptiste. van Belmont (1652) used quantitative chemical experiments for the first time in the study of the nutrition of plants. He ridiculed the ancient philosOphers for their failure to utilize experimental data. An early experiment of van Belmont, which was fundamentally very sound, convinced him that water was the primary substance from which plants were formed. This conclusion was the result of a now famous experiment which has been described many times as an example of the improper interpretation of essentially accurate observations. Van Belmont stated, 'I was able to show by the following experi- ment that all vegetables are produced immediately and materially from the single element water. I took an earthen vessel in which I placed 200 pounds of soil previously dried in an oven. I then watered it with rain water and planted therein a willow branch weighing 5 pounds. After an interval of five‘years the tree which had sprung up weighed 169 pounds and. some 3 ounces. The earthen vessel which was always watered when necessary with rain water was large and embedded in the ground: lest any flying dust should get mixed with the soil, an iron cover plated with tin and provided with a large opening closed the mouth of the vessel. I did not determine the weight of leaves which fell during the four months. At the end‘of the experiment I dried the soil again in the vessel and obtained the some weight of 200 pounds lacking about 2 ounces. The 1614 pounds of wood, bark, and roots were therefore derived from water alone.“ In as much as 98 percent of the fresh weight of many plants. lettuce for example, consists of water, it is not surprising that the comparatively guide quantitative data of van Helmont led him to an incorrect conclusion. ’4. The Theory of Ihrth as the Nutrient of Plants Jethro full (1733), commenting some 150 years later on the willow branch experiment conducted by van Belmont, said that those who agreed with the conclusion drawn by van Helmont 'were deceived in not observ- ing that water'has always, in its intervals, a charge of earth from which no art can free it.“ In full's Opinion, it was not the liquids 5. in the soil, but the very minute particles of earth suspended in it that constituted the "preper pablum' of plants. Tull believed that the pressure exerted by the growing roots forced these particles into the 'lacteal mouths of the roots” and into the translocation system of the plant. He summed up the opinions of the agricultural workers of his period when he stated, “It is agreed that all the following mater- ials contribute in some way to the increase in plants, but it is dis- puted*which of them is that very increase or food: nitre, water, air, fire, earth.' The great contribution to the knowledge of plant nutri- tion made by Tull was his emphasis on the soil itself as the source of plant nutrients. That he was wrong in concluding that the earth particles were the nutrients, rather than the minerals which they contained, does not detract from the value of his contribution, for from his time onward, the study of the nutrient requirements of plants was centered on the soil. 5. The Theory of maus as the nutrient of Plants The humus theory had its origins in the homoiomeria of Aristotle, as noted earlier in our discussion, and it was gradually expanded and developed for a period of 2000 years. The theory that human was the nutrient of plants reached its greatest importance during the time of Thaer (18hh), Berzelius (18h2), and Humphrey Davy (1813). According to this concept, plants feed on substances similar to their own.or- ganic components. The organic matter of the soil, or humus, was re- garded as the chief nutrient of plants and the major source of soil fertility. It is interesting to note that the proponents of this theory failed to concern themselves with the origin of the humus, and like many farmers today, assumed that there was an inexhaustible native supply of this substance in the soil. The roots of plants were be- lieved to extract the humus from the soil and to transform it into plant structures by combining it with water. Some of the followers of the humus theory believed that minerals were not essential for plant growth, and others believed that they acted as stimulants to growth rather than as essential nutrients. The application of more accurate chemical methods made it obvious that the large amounts of carbon in plants had to be explained. Noth- ing was known of the role of carbon dioxide in the atmosphere, and as a result the preponents of the humus theory firmly believed that the carbon component of plant structures originated from the humus in the soil. Many workers opposed the idea that the humus in the soil provided the carbon in the plant, and they began to accumulate information pointing towards the true importance of the leaves as the organs in- volved in carbon assimilation and to the roots as the organs involved in nutrient absorption. Ingen-Housz (1779) and Senebier (1782) came to the conclusion that 'purification' of the air occurred during periods of light, while 'vitiation' took place in darkness. Senebier argued that the increased weight of the willow tree in van Helment's experiment came from the 7. 'fixed' air. He said, ”If then the fixed air dissolves in the water of the atmosphere, combines itself in the parenchyma with the light and all other elements of the plant: if the phlogistine of that fixed air is precipitated in the organs of the plant, if this precipitate remains, as we think, since this fixed air goes out of the plant in the form of dephlogistine, it is clear that the fixed air, combined in the plant with the light, leaves there a material which was not there previously, and m experiments on etiolation demonstrate this.‘ Lavoiser (see Browns 19%) concluded in his discussion of the origin of carbonaceous matter in craps that it might seem logical to suppose, superficially, that plants obtain their carbon from the vege- table earth, the hunms, soil and manure. But because some plants live in apparently pure .water and air, Lavoiser considered it more reason- able to assume that this carbonaceous material is derived from the decomposition of carbon dioxide obtained from the air. It remained, however, for de Saussure (1801+) to establish beyond a doubt that green plants assimilate the carbon from the carbon dioxide in the atmosphere. He was the first to state that the soil furnished a small but essential part of the essential nutrients of the plants, and that the fertilizing value of humus was due chiefly to its mineral content. Viemnn and Polstorff (1&2) conducted a series of precise experi- ments on the growth of plants in extracted humus, and they found that the residue in 100 grams of the extract exposed to the air for a month was 136 milligrams, while the residue in 100 grams of extract in which plants had been grown for a month was 132 milligrams. Because of this slight difference of only 1+ milligrams, these workers concluded that humus plays an insignificant role in plant nutrition. The theory of the humus nutrition of plants received serious blows from new investigators, but it remained for Liebig (18m, 1852) to complete its overthrow and to clarify the problem of plant nutrition in the minds of his contemporaries. Liebig stated in a series of lec- tures before the British Association for the Advancement of Science that 'the primary source whence man and animals derive the means of their growth and support is the vegetable kingdom. Plants, on the other hand, find new nutritive material only in inorganic substances.” Liebig so effectively disposed of the humus theory that from his day forward primary consideration has been given to the study of the ash of plants in relation to the mineral content of the soils on which they were grown. Liebig's successful attack on the humus. theory of plant nutrition ended the profitless philosophical concepts originating from Aristotle's homoiomeria and made possible the scientific approach to the problem of soil fertility. The foregoing discussion indicates that the efforts of the most brilliant men of.a period of two thousand years were necessary to deter- mine the materials serving as nutrients for plants. During Liebig's time, agricultural science attained a fairly sound scientific basis. Agricultural chemists, having attained a rudimentary knowledge of the 9. chemical components utilized by plants, then turned to the study of the quantities of these nutrients required for the optimum production of crops. 6. The Beginning of the Biochemical Approach to the Problem of Plant Nutrition At the same time that the various theories of plant nutrition were being proposed and discarded, the first true approach to the deter- mination of the nutrient needs of the plant already was under way. Tully 150 years ago, Chaptal (1801) described the properties, extrac- tion, and utilization of various mineral constituents of vegetables. At about the same time, Hermbstadt (1801+) conducted what were probably the earliest large scale field emeriments. Hermbstadt attempted to answer a number of pertinent questions. In his own words, '(1) Do the nutrient elements of the fertilizer reappear in the organs of plants? (2) Do the elements participate in the formation of such organs and of the plant constituents deposited therein? (3) Can the quantity of each plant constituent be increased by increasing the amount of the element adapted for its formation? (’4) Can we deter- mine whether the same field, without addition of new fertilizer, suffers a decrease in yield with each year and whether, by a rotation with other cereals or with seed crops etc., a better yield of grain can be obtained without adding new fertilizer?" Even the thought behind these questions was far ahead of the current thinking of his the. 10. In the first tabulation of its kind, De Saussure (1801+), using what are regarded now as very inaccurate chemical procedures, gave percentages of water soluble salts, insoluble phosphates and carbon- ates. silica, alumina, and metallic oxides in the ash of leaves, branches, wood, bark, flowers and fruit of seven different trees and in the ash of the straw and grain of wheat, oats and barley. He came to the conclusion that the nature of the soil exerted a pronounced influence upon the percentages of the mineral constituents of plants. Carl Sprengel (1837), one of the most meticulous chemists and experimenters of his time, established the foundations for the thought of many later workers in the field of fertilizer use when he expressed his belief that the chemical analysis of crops might indicate their nutritional deficiencies: He also believed that the quantities of the various nutritional ions in the plant are dependent on the chemi- cal composition of the soil in which the plant was grown. As early as 1842, Viegman and Polstorff conducted pot culture experiments using sand and nutrient solution designed to attack the validity of some of the contemporary theories of their time, although these workers were not specifically concerned with the humus theory. Liebig (18’40, 1852, 1863) gavue a great impetus to the study of plant nutrition and the use of fertilizers when be formulated his Law of Minimum and applied the law to the response of crops to fertilizers. Prom his time on, experimenters in the field of plant nutrition have continued the never ending search for the proper nutritional require- ments of agricultural plants. 11. During the latter part of the nineteenth century, various workers utilised the chemical analyses of plants in their studies of the responses of crops to fertilizers. For example, Hellriegel (1869) studied the optimum concentration of nutrients in barley in this manner. Petersen (1875) and Volffe (I872) studied oats, Hanamann (1878) studied various cereals. Many students during this period, in addition to those mentioned, also used chemical analyses of crops in their efforts to determine the proper concentration of nutritional ions in plants. Some agricultural chemists of the period attempted to correlate the percentages of nutrients in plants with their quantitative res- ponses to fertilizers. Liebscher (1893) compiled data from field studies: Hellriegel (1893) used nutrient solutions in quartz or sand: and Joulie (1891!) carried out a systematic survey of farmers' fields. Ihe above brief description of the work of a few investigators is sufficient to show that practices in use today had their origin in the agricultural chemical revolution of- the latter half of the nine- teenth century. In many cases, the techniques used during the period were good, but the inadequate status of the current knowledge of plant physiology frequently permitted faulty conclusions to be drawn» 3. Development of bioIOgical methods for the determination of fertilizer requirements 1. Introduction Since the work of Liebig laid the foundations upon.which developed the later biological approach to the problem of fertilizer use, it is natural to regard this early work important, not only historically, but also as the stepping stone which led to the present day techniques. It will be recalled that Do Saussure (180h) attempted to use the chemi- cal compositions of plants as indications of their nutritional require- ments, and he observed that these compositions varied with the soil, with the part of the plant, and with the age of the plant. Liebig (1852, 1863) probably was the first of the many'investigap tors who tried to give a mathematical expression to the magnitude of crop response to fertilizer. Although these expressions appear as a mass of formulae and geometric curves, one must remember that science speaks plainly only through.mathematics. with this thought in mind, the development of Liebig's laws appears as an essential part of the historical background of the currently used bdological methods for determining the fertilizer requirements of crops. Consequently, these early mathematical expressions of the responses of crops to fertili- zers will be discumsed briefly as a foundation for our understanding of the more recent advances of the twentieth century. 2. The Law of the Minimum It is interesting to note that the Law of the Minimum was the earliest attempt to relate by precise mathematics the composition of the soil and the plant to the fertilizer requirements for maximum crop yields. While it is commonly assumed that this law was first expressed by Liebig (1852, 1863), a survey of the literature of his period reveals that as early as 1801+ De Saussure had a vague concept of the effect of minimum factors, and Sprengel (1837) in his 'Die Bodenkunde oder die Lehre von Boden" revealed that he probably had a clearer concept of the facts than did Liebig when he stated, ”The soil is often neither too stiff nor too porous, neither too wet nor too dry, neither too cold nor too warm, neither too high nor too low: it may be situated under very advantageous climatic conditions, it may have an abundant supply of humus, and be located on a favorably inclined slaps, and yet may often be unproductive because it is defi- cient in m 1M9. glgent that is necessary as a food for plants. Again it may also fail to bear good crops because it contains a very easily soluble plant food in too great an excess or because it con- tains substances that act as poisons to the growth of plants“ (Browns 19%). I ' 1'he concept of the effect of limiting factors, however, was not understood clearly by agricultural chemists until Liebig expressed it with clarity in his writings. In his first publication (1840) he expressed the Opinion that the percentage of the nutritional ions in plants was constant and he assumed that this chemical composition represented their inherent nutrient needs. Consequently, many workers came to the conclusion that they could maintain the fertility of soils merely by returning to the soils the total amount of nutrients re- moved by the crop. This has been commonly referred to as the books keeping method of soil fertility, a procedure which.was strongly supported by contemporaries of Liebig and which we still find at the present time featured prominently in the advertising propaganda of the fertilizer industry. The popular term now is “mining the soil”. Continuing his earlier study, Liebig (1852) arrived at the opinion that there exists a. Law of Exhaustion for each cultivated plant. He stated, 'Ihis state of exhaustion inevitably happens even when there has been*withdrawn from the soil, by a course of crops, only one of all the different mineral substances necessary for the nourishment of plants. For the one which is wanting or exists in deficient quanp tity, renders all others inefficient, or deprives them of their activity.“ Liebig expressed this concept mathematically by the equation, P g I - R, where P was the yield of the crop, l' the amount of soil nutrients, and R.the “resistance" exerted by other factors affecting growth. This formula gave a mathematical expression for the bookkeeping theory of plant nutrition. the most concise and clear expression of this Law of the Minimum is found in Liebig's “Natural Laws of Husbandry“ (1863). In his own ‘words, 'lvery field contains a maximum of one or several, and a mini- mum of one or several other nutrient substances. It is by the minimum that crops are governed, be it lime, potash, nitrogen, phosphoric acid, 15. magnesia, or any other mineral substituent: it regulates and deter- mines the amount or continuance of the crops.’I Lagatu and Maume (1927,d) quoted the French edition of ”this book, 'Les Lois Naturelles de l'agriculture' (1863), by stating that nutrient ions are absorbed by plants in definite proportions so that if the maximum amount of one ion is absorbed, the accumulation of other nutrients is retarded. The implications of the Law of the Minimum may be summarized as follows: 1. Plant growth increases with increases of the limiting factor until the factor ceases to be limiting. ’ 2. If two factors are limiting or nearly limiting growth, an in- crease of one will have but little effect, while increases of both will exert a very considerable effect. 3. If the absorption of one nutritional ion reaches the maximum limit, the absorption of other ions is retarded. l'rom the time that the Law of the Minimum was first stated, it became apparent that it did not follow strictly the interpretation that Liebig himself had applied to it. he first adequate attempt to apply the law to experimental data was made by Hellriegel and his co- workers. In a series of papers, Hellriegel and Wilfarth (1888), and Hellriegel, Vilfarth, Boner, and Wimmer (1898), described the growth of barley in sand cultures which were supplied with all necessary nutritional ions, with the exception that the amount of one nutrient 16. salt was varied. A.portion of these data are assembled in table 1 and are shown graphically in figure 1. The results of varying the magni- tude of the nitrogen supply indicate that the first increment of nitro- gen produces a certain increase in yield: but the yield increases due to the second and third increments are proportionately greater. The yield is greater than is expected, if as Liebig assumed, the effect of nutrient supply is proportional to the amount present. The fourth and fifth increments of nitrogen produce a lesser increase in yield and a sigmoid curve results when the data are presented graphically. This sequence of effects are not restricted to nitrogen for Hellriegel obtained a similar curve when he varied the supply of potassium. In the early years of the twentieth century, Mitscherlich (1909) subjected the relationship of nutrient supply to crap yield to a rigorous mathematical analysis and proposed a different statement of the Law of the Minimum. Mitscherlich's version states that the in- crease in yield per unit of the limiting nutrient supplied is directly proportional to the decrement of the yield from its maximum magnitude. This statement of the Law of the Minimum has been.widely accepted as a more exact indication of the relationship between.plant growth and nutrient supply than.was Liebig's earlier statement. Mitscherlich's restatement of the Law, in turn, has been severely criticized not only by his contemporaries Briggs (1925), Balmukand (1928) and Pfeiffer, Simmermacher and Rippel (1919), but also by later ‘workers such as Macy (1936). These investigators upheld the principle o.om o.HN o.mm o.mm o.on m.mH eastwaaaas ea .naenm nee sense: e.me e.mn e.me o.mm c.5m m.HH econeonn assesses ace omeeneoeom ninja mm we anemones.“ saucepan muons... omwé mmué 9.56 33.: I. you. can?» commence“ no cache ama.w~ mmm.a~ mmm.aa mom.0H mmm.e Nea.nv someones cases see no sense om: 8N wma N: mm o. @3385 semen»: no 25.83:: .Ammmfi £393.33 cad flowcuhdaomV mohflugo dado a.“ 5.0.2.3. no Autos» no emanate.“ demands Ho codenamed.“ ebummooodu no cacao ems .H capes 18. (N N p -c 0 Day ”errata fmoeucso IN 87:15. ‘ o~ 0 L A a _L o us :10 336 490 ”8"" Alarms-N cuppa“: 4a Conway». Figure 1. The effect of nitrogenous fertilizer on the growth of barley. (Hellriegel and Wilfarth, 1888) of Mitschlerich's interpretation but differed as to its proper mathemap tical statement. A.further critical consideration of the Law of the Minimum is found in the work of Lagatu and Maume (1927,d) which involved the study of the ratios of the percentages of nitrogen, phosphorus, and potassium in plants grown at different levels of fertility. These workers noted that if one element was limiting, the ratios of the concentrations in the plants of the other elements to that of the deficient one increased. In.other'words, they believed that when one nutrient‘was deficient, the proportion of the others in the plant increased,'uhile that of the deficient nutrient decreased. Such results obviously were in contra- diction to those expected from the Law of the Minimum. Wallace (1928) working with apples and Bartholomew, watts, and Janssen (1933) working ‘with potatoes obtained data consistent with those of Lagatu.and Maume, but Murneek and Gildehaus (1931) apparently obtained a reverse effect in their work on the nutrition of apples. As already pointed out, the applications of the basic principle of the Law of the Minimum resulted in the procedure commonly referred to as the bookkeeping method of plant nutrition.and which is nmw known.as "mining the soil.‘ This bookkeeping approach.was supported strongly by Liebig and Wolff (1871) who catalogued the chemical analyses of many plants in order to return to the soil the nutrients ‘which were removed by crops. This approach to the maintenance of fertility is short sighted, and it is unfortunate that at the present 20. time may agriculturists still care for our agricultural heritage on this basis. 3. The Laws of the Minimum and Maximum Nutrient Content of Plants Goodall and Gregory (19a?) credit Weinhold (1862, 1864) as the first to use the chemical analysis of- plants as an index of the amount of available nutrients in soil. At about the same time, Hellriegel (1867, 1869) studied barley in sand cultures and found that the per- contage of potassium in the grain increased as the supply of potassium was increased. He concluded that maximum yields were obtained when the potassium content of the straw was 0.5 percent of the dry matter and 0.38 percent in the grain. Wolff (1868) reported a minimum per- centage of Cam in oats, and on the basis of further work, he published (1877) data indicating the minimum percentages of the six principal nutrients in cats as well as percentages associated with satisfactory plant growth. Heinrich (1882) made a significant advance in using this proce- dure for determining fertilizer requirements when he analyzed a specific portion of a plant, namely the roots of oats, when grown at various levels of nitrogen. The roots were chosen as they were 'more depleted at maturity the poorer the soil. The Whole of the available nutrients have migrated to the aerial parts of the plant, to be used for assimilation and formation of new organs.' Heinrich proposed a ILaw of the Minimum” which stated that if a nutrient was lacking or present in a limiting amount, the concentration of that element in the root was at a minimum value. On the basis of this statement, he gave minimum concentrations of the six major nutrients below which the nutrient was deficient. Yon Dikow (1891) also reported data based on root tissue, but he included as 'roots' of barley all parts of the plant below the first node. The minimum percentages for P205 and nitrogen were reported as 0.13 and 0.63 percent respectively. These minima were based on the fact that material from three of his four plots exhibited approximately these values. Von Dikow also suggested that the range of the mineral content of the plant was limited by a maximum as well as by a minimum magnitude, and he proposed that Heinrich's Law of the Minimum be supplemented by a Law of the Maximum. This new law stated that until a certain maximum concentration of nutrients was reached, fertilizers would not produce the maximum growth. Helmkampf (1892), using oats and wheat subjected to differential fertilizer treatments, noted that the percentage of nitrogen in the roots of plants grown on all plots approximated Heinrich's minimum value and that the crop showed marked responses to the treatments. The potassium content increased with increased applications of potassium, and the phosphorus content remained constant between 0.1v0 and 0.48 percent calculated as P205, regardless of the treatments. He concluded that in this latter case, the maximal value suggested by von Dikow had been reached. Stahl- Schroder (190R), on the other hand, reported results obtained.with oats and wheat showing evidence of a luxury consumption, and conse- quently he rejected von Dikow's proposal of the existence of minimum and.maximum concentrations of nutrients in the plant. In a series of papers, Atterberg (1886, 1887, 1887a, 1887b, 1888, 1888a, 1889, 1901) published the results of numerous experiments with cats using sand cultures and field experiments. He reported in detail the results of chemical analyses of the various portions of the plants. He gave minimal and average values for the principal nutrients but did not accept the idea of a maximum concentration.as postulated by von Dikow (1891). Atterberg's results (1888) indicated that very high supplies of the phosphate ion.might continue to increase the phosphorus content of tissue even though there was a depression of the yield. the true significance of a minimum percentage of a nutrient in a plant tissue from a nutritional requirement standpoint is still being debated. the Law of the Maximum has received considerable attention as it applies to the problem of the luxury consumption of nutrients. While some‘workers before the turn of the century did not accept the concept of luxury consumption of nutrients, the Breslau.experiments of Pfeiffer, Simmermacher, and Rothmann (1915) showed that luxury con- sumption did occur. Whereas the Law of the Minimum as proposed by Liebig (1852, 1863) - and later by wolff (1871, 1877) had a practical application in the bookkeeping concept of plant nutrition, the Law of the Minimum as 23. proposed by Heinrich (1882) and the Law of the Maximum as proposed by 'von Dikow (1891) found their expression in concepts of the critical concentrations, threshold concentrations, and optimum nutrient ratios used very extensively at the present time. The greatest portion of this early'work‘was based on cereals and many workers proposed minimum levels of nutrients in the plants. Hellriegel (1867) gave minimum values of 0.5 and 0.38 percent of potash in the dry matter of the straw and grain respectively. Vblff (1876, 1877) collected all the data available at the time for plant analyses and published them in his book"Ashen Analysis“. From these data, he was able to postulate values for the ”necessary minimal content” in plants and also for 'good average development“. The work of the later workers, Heinrich (1882), Baessner (1887), von Dikow (1891), and Helmkampf (1892),. followed similar lines. Atterberg (1901) probably summed up the ideas of the period previous to 1900 in the report of his oat experi- ments already mentioned. He concluded that if the supply of a nutri- ent is increased, the nutrient is assimilated by plants in increasing quantities and its percentage in the plant is increased. He dee- cribes the use of his minimal values as follows: "The element which is present in the plant in minimum quantity may be determined as follows: The percent content as shown.by the analysis is compared with.the average and minimum content of oats. In general, the fertilizer ingredient whose percentage is farthest below or least above the average figure representing the minimum content is present in minimum quantity." Atterberg believed that this procedure made it possible to deduce the fertilizer deficiencies of the soil. During the period Just discussed, the majority of the possible approaches to the problem of estimating the fertilizer requirements of crops already had been considered and at least some experimental work had been carried out. The great value of the work of this period, however, was that it served as a fertile source of ideas on a subject which even today is obscure. Whereas no great scientific advance was made, the ideas that the work produced, although they in many ways were rather elemental, were the sources from which the in- vestigators of the twentieth century were able to build the more scientific approach we know at the present time. 1+. Summary of the Work during the Last Half of the Nineteenth Century which Was Instrumental in the Development of Modern Procedures 9., Field Trials flie majority of agricultural workers have at some time made use ‘ of field trials to determine the fertilizer requirements of crops. The best known, and one of the earliest field trials, was that de- signed by Iawes and Gilbert (1851) to test Liebig's (18110) theory of . mineral fertilization. These investigators concluded that the nutri- ent composition of the crop removed is 'no direct guide whatever. to fertilizer requirements. These trials have been continued to the ‘present day at Rothamstead in the same manner as originally planned by Lawes and Gilbert. Joulie (1882) conducted work in France using the data obtained from soil analysis, tissue analysis, and field trials to design a program to build up the soil: consequently he might be credited as the first to apply data obtained in the laboratory to the practical management of farm lands. Many other workers conducted their nutrition- al studies on the basis of field plots, including Helmkampf (1892), Liebscher (1893), and,Atterberg (1901). Some investigators, however, observed that the field trials did not yield as conclusive data as did pot experiments. For example Liebscher, Kretschmer, von Seelhorst, and 'illms (1898) and also Atterberg (1901) recognized the inherent 'weakness of field trials. Atterberg, however, pointed out that plant analysis and pot culture techniques,‘while giving a better indication of the nutritional status of the plant, did not lead directly to the determination of the amount of plant nutrients available in the soil, and consequently admitted the necessity of the supplementary use of, I field trials. b. Pot and Solution Cultures The development of the use of pot cultures followed the overthrow of the Humus Theory by Liebig (1840) and the establishment of the miner- al theory of fertilization. The difficulties experienced by the ‘workers of the time in.obtaining good results under natural conditions also encouraged the use of pot cultures under more or less controlled conditions. Any worker in the field of biological science realizes 26. only too well the difficulties of obtaining consistent research data under uncontrolled natural conditions. Sachs (1860) was one of the first to further refine experiments in plant nutrition by using water solutions of pure chemicals to grow plants. He published one of the first standard formulae for growing plants in culture solutions, and Knops (1865) soon after proposed a solution that has since become one of the most widely used in the study of plant nutrition. Other solutions prOposed during the period were those by Tollens (1882), Schimper (1890), and Pfeffer (1900). Hellriegel (1867) working with barley in sand cultures provided with nutrient solutions proposed minimum concentrations of potassium. He later (1869) obtained similar data.using water cultures. Other workers of the period.who used pot or solution cultures in their study of the fertilizer requirements of craps were Wolff (1876), Petersen (1876), Helmkampf (1892), Hellriegel and Wilfarth (1888), Hellriegel, Wilfarth, Homer, and Winner (1898), Hellriegel (1893) (1893a), Liebscher, Kretschmer, von Seelhorst, and Willms (1898), and.Atter- berg (1901). It was apparent, at least to some of the workers, that there were certain limitations in the use of pot and solution culture tech- niques. Wolff (1877) stated that 'the fact that the experiments were conducted in water culture somewhat invalidates the results and tends to prevent a valid field interpretation”. Somewhat later Liebscher and his co-workers (1898) concluded that, while the experiments appeared to be inconclusive when the data were compared to those obtained from 2?. pots, care would have to be taken in the interpretation of the results obtained under artificial conditions. These views are recognized by most of our own cantemporary workers in the field of plant nutrition, but unfortunately their implications are not always acted upon. Certainly it must be agreed that the artificial conditions im- posed by pot culture techniques are not conducive to the direct applica- tion of data obtained by their use to practical procedures in the field. Under the circumstances, it was logical to look directly to the plant, growing under various conditions, as the source of data concerning its nutritional status. c. Plant Analysis m Soil Analysis The problem of whether to analyze the soil or the plant to deter- mine fertilizer requirements still perplexes students at the present time. As yet, no definite answer can be given. As soon as the source of plant nutrients was settled by Liebig, the question of whether to analyze soil directly to determine their amounts, or to regard the plant as nature's ideal nutrient extracting agent troubled students of the fertilizer requirements of craps. The earliest concept of the use of plant analysis was based on the bookkeeping method of calculat- ing fertilizer requirements proposed by Liebig (18110) and others. As noted previously, Weinhold (1862, 1861+) probably was the first investigator to consider the chemical analysis of the plant as an index of the availability of nutrients in the soil. He believed that the 28. composition of the plants growing naturally in a given location indi- cated the amount and availability of the nutrients in the soil. He made the error, however, of considering that when more than one species 'was abundant in a given.area, all such species should have identical chemical compositions. He tested this supposition, and when he found that it was not valid, he dropped the investigation, concluding that the physical properties of soils were of greater importance than were its chemical properties. weinhold's work stimulated other workers to correlate the chemi- cal components of plants with the amount of added fertilizer. The most important contribution to this approach was that made by Holff (1871, 1880) who collected all the chemical analyses published from 1800 to 1880 and compiled them in his classic book 'Aschen.Analysis'. Investigators did not give much thought to the use of plant analysis as a diagnostic procedure to determine the nutritional status of the plant until around 1890. Helmkampf (1892) stated the problem when he said, Fplant analysis in connection with fertilizer experiments furnishes indications of‘whether or not a soil is defi- cient in one or more of the essential nutrients.......investigations concerning the proper part of the plant to analyze, the proper stage of its growth, etc., need to be conducted." Liebscher (1893) also concluded that “since the richness of the soil in phosphoric acid exerts an influence on the composition of the grain, the analysis of the grain may be taken as an indication of the richness of the soil.“ 29. Januszewski (1895) following the same line or thought as did Joulie (1894) stated that “plant analysis in fertilizer experiments can only be used when careful estimation and evaluation of the soil as to its physical pr0perties is taken into consideration". Remy (1896) believed that the amount of nutrients absorbed in a given time was in no way dependent on the amount of'plant tissue produced during that time, and he also believed that weather influenced the rate of nutri- ent absorption within certain limits as well as did the amounts of nutrients available. Towards the end of the nineteenth century, the major workers had some to the conclusion that the chemical analysis of plants was a more reliable procedure for the determination of the nutrient requirements of crops than was the analysis of the soil. This view probably was given undue emphasis because of the generally unsatisfactory analytical techniques at their command for the proper analysis of soil. d. Early Methods of Chemical Analysis As the chemical methods of analysis used during the nineteenth century are obsolete and are of historical interest only, attention ‘will be given here only to their general features. In this early‘work, the madority of the investigators based their results on the percentages of total ash in the plant. Lowry and Tabor (1931) credit Palladin (l86h) as the first to use the sap of plants for the study of their nutrition. 30- The earlier work involved the analysis of the whole plant, but later many investigators recognized that different portions of the plant differed in their value as indicators of nutrient uptake. Heinrich (1882) concluded that “the roots are the most appropriate organ for comparitive investigations involving the determination of nutrient conditions in the soil.I Hellriegel (1869) considered that the composition of the straw of cereals was the best indicator. Von Petersen (1876) found that the entire top was a better indicator than was the straw or grain taken separately. Other workers, such.as wolff (1877) and Helmkampf (1891, 1892), agreed with von Petersen. 0n the other hand, Januszewski (1895), and later Atterberg (1901), regarded the composition of the grain as the best indicator of the availability of nitrogen and phosphorus, and the straw as the best indicator of the potassium supply. It is apparent that by the end of the nineteenth century, sufficient progress had been made to indicate the necessity of using specific portions of the plant, if the chemical analysis of plants was to serve as a procedure for determining fertilizer require- ments. Our discussion indicates that nothing really new has been added in recent times. Frequently it is assumed that tissue analysis is a recently developed concept} but the advances since the end of the nineteenth century have been those of inproved techniques, particu- larly analytical techniques, rather than the development of new approaches. If one were to search for the reasons why the early work 31. failed to show good correlations between the chemical analysis of plant tissue and the nutritional status of the plant, it would be apparent that the chemical procedures used.were inadequate and that too little ‘was known about the factors affecting the absorption of nutrients. It is essential, therefore, to discuss in some detail the factors involved in the absorption and concentration of nutrient ions in the plant if a true understanding of the proper use of "tissue testing” is to be attainede II. FACTORS AFFECTING THE.ABSORPTION.AND CONCENTRATION OF NUTRIENTS IN PLANTS A. Introduction -Hany factors influence the absorption of nutrients by plants and their percentages in plant tissues. Any attempt to utilize biological methods, that is to say, either the growth of plantscnf their chemical composition, to determine the fertilizer requirements of crops depends on the assumption that only the amount or availability of a nutrient in the soil influences its effect on growth or its concentration in the tissues of the plant. Strictly speaking, this hypothetical situation is never true. However, this is not equivalent to a condemnation of biological methods. The use of biological methods theoretically is the best of all procedures for the determination of the fertilizer requirements of crops, provided of course, that all factors affecting the results other than the availability of nutrients in the soil are constant or insignificant in their magnitude. It is essential that the many factors affecting the responses of plants to nutrients be understood and constantly borne in mind if the use of biological methods is to serve as an adequate technique for the determination of the fertilizer requirements of crops. Conse- quently these factors will be discussed briefly. 3. Mechanism of Nutrient Absorption A.complete understanding of the mechanism by which plants absorb 33. their nutrients from the soil would make it possible to evaluate accurately the influence of all factors on the response of crops to fertilizers. Unfortunately, a complete comprehension of this intri- cate biological phenomenon is not possible on the basis of present knowledge. However, such evidence as is available should be utilized in.an attempt to determine how various factors affect the uptake of nutrients by plants. The preponderance of evidence leaves no doubt that the process of nutrient accumulation by plants is intimately associated‘with the metabolic activities of plant cells, and it cannot be regarded merely as a passive entry of ions into the roots. The important question is the mechanism by which nutrient ions enter the plant because the pro-- case is affected by factors other than the amount present. The meta- bolic aspects of ion accumulation have been investigated almost ex- clusively by the nutrient solution technique. The application of this experimental approach to problems of fertility depends on the assump- tion that a basic similarity exists between the absorption of ions from a nutrient solution and from the soil. Such an assumption is based on the theory that the nutrients move unhampered by the surface forces residing on colloidal particlesf but the solid phase of soil is regarded as the source of most nutritional ions, particularly cations, in the soil solution. Recent research by Comber (1922), Truog (1928), and Jenny (1950) has shown that the extracted soil solution does not provide the answer to the problem of the mineral nutrition of plants. The attempts of Camber and Truog to attribute a more important role to the solid phase of the soil were unsuccessful because no theory was advanced which provided a convincing mechanism involving the solid phase. Jenny (1950) has shown that the uptake of radioactive sodium, at higher concentrations, is decidedly greater when adsorbed on.clay. suspensions than.when.present as a chloride or bicarbonate salt in solution. In the case of an ammonium saturated clay versus a solution of ammonium chloride, the uptake of ammonia by the roots was nearly equal under both conditions. On the other hand, a solution of potassium chloride provided a better source of potassium than did a potassium saturated clay, as would be expected on the basis of the solution theory of nutrient availability. It is sufficient for the purpose of the discussion to emphasize that the absorption of cations adsorbed on clay particles bears no simple relationship to their absorption from aqueous solutions. The theory of contact exchange prOposed by Jenny and Overstreet (1938) involves a mechanism for the absorption of nutrients held colloidally on the clay particles. This mechanism can be deduced from theoretical considerations concerning the nature of colloidal surfaces. Specifically, it rests on the redistribution of ions when they find themselves in intermingling colloidally held ionic layers. Accordingly, the contact ‘f absorption theory embraces all colloidal systems and not only those in the soil and in the cells of roots. The absorption .330 of cations by roots from colloidal surfaces in the soil by contact exchange and by direct absorption.from solutions is diagramaticalky shown in figure 2. The application of the theory of contact exchange to the mineral nutrition of plants postulates that swarms of ions intermingle in the electrical fields of the colloids in the soil and those of the absorbing cells of the roots. Inasmuch as an ion is attracted by two colloidal surfaces it goes to the surface possessing the greatest attractive force. If this be true, the transfer of ions from the soil colloids to the root does not involve a true solution. The uptake of ions adsorbed on colloidal surfaces by this mechanism is largely independent of the water content of the soil and of the transpiration stream of the plant. Both solutions and contact mechanisms operate in the soil. As far as major nutrient cations are concerned, the soil solutionuwould contribute a relatively smaller amount of the total absorbed by the plant. The lower the salt content of the soil solution, the greater will be the relative importance of contact exchange‘with the soil colloids. For those micro nutrient ions, such as iron, which.are largely insoluble at higher pH values, contact exchange may well be the dominant mode of their acquisition by plants. The direct application of the theory of the metabolic absorption of ions to the problem of nutrient absorption from soil is seen to be somewhat more difficult than to the problem of absorption from solutions. 36. SOLUTION Excmmsg . CONTACT E xcuANec ‘ K K K F 93¢: 3:33... + H 0 H * .23.!“ .3321. ,+ '33.!“ K' K INTE R +K.O H + INT'EMED. H K K‘ K STATE + K0 H + H K \Y I FINA L K K K s TATE +H 0 H *1 + , K K r, K Figure 2. Diagrammatic representation of absorption by contact-exchange and from solutions. (Jenny and Overstreet, 1938) But since the metabolic activities of the absorbing cell undoubtedly are involved in the entry of ions into the plant, irrespective of external conditions, it is apparent that all factors affecting meta— boliem will also influence absorption. C. Internal Factors Affecting the Concentration of nutrients in Plants 1. Species and variety Drake and Scarseth (1940) have emphasized that plants of different genera may differ profoundly in their nutrient uptake, and Wimer (1937) pointed out that this may be true also for species of the same genus or even for varieties of the same species. Many workers have reported their observations on the different amounts of the nutritional ions contained in plants of different species grown under identical conditions. The differences in the composition of plants has been studied ever since chemical analyses of plants were first carried out. Arendt (1859) stated that “our literature is over rich in ash analyses of many kinds but they'may be briefly summarized as follows: Plants must have a certain amount of inorganic nutrients. This amount is different for different plants. The number of inorganic elements is about the same in all plants, although their relative amounts differ due to the various chemical and physical natures of the soil and due to the kind of plant and due to the age of the plant and due to the various natures of the different organs." Wolff (1871, 1877) in his book 'Aschen Analysen" compiled the results of innumerable analyses of plant materials reported by various authors from 1800 to 1880. Although 380 in some cases, the analytical methods were crude, WOlff's tabulated data indicated that the ash content varied in different kinds of plants, in different stages of their growth, and in different locali- ties. Since this early compilation, differences in the chemical com- position of different species have been noted by many workers, even when plants were grown under similar conditions. Richardson.(l920) determined the composition of the ash of certain dune plants and in the sand in.which they grow. The most striking differences observed in the proportion of elements absorbed was in the amounts of silicon, calcium, and aluminium, although differences were also noted in potassium and magnesium. Working with alfalfa and tobacco grown on soils containing both lithium and strontium, Headden (1921) found that the alfalfa had little lithium but a considerable amount of strontium in the ash, while in the case of tobacco the reverse was true. Skinner, and Reid (1921) reported differences in the mineral content of wheat and clover grown in the same nutrient solutions. Greaves and Nelson (1925) obtained similar results. Newton (1928) grew sunflowers, beans, wheat, barley, peas and corn in the same container. He harvested the wheat and barley at the heading stage, the corn.at the time of tasseling, and the peas and beans at the blooming stage, and then determined the percentage of the various nutrients present in the plants. His results are summarized in table 2. Not only did Newton detedt differences in the various species, but also found that the absorption of nutrients from the culture solution.was greater than from a loam soil. He did find, however, that there was less variability in the amounts absorbed 39. Table 2. Chemical composition of various crops grown in nutrient solution in the same container. (Newton, 1928) Composition as fiercent Crop of Dry Matter 0a 1: Mg 11 ' r sunflower 2.2 5.0 0.64 3.6 0.56 Beans 2.1 4.0 0.59 3.6 0.55 Wheat 0.8 6.7 0.41 9.5 0.09 Barley 1.9 6.9 0.5% h.7 0.52 Peas 1.6 5.3 0.50 h.5 0.19 Corn 005 3e9 0.140 209 0e39 from the solution than from the soil. Trelease and Martin (1936) reported that Western wheat grass, Agggnggpigmithii Regdb., grown on various soils accumulates from 1 to 60 parts per million of selenium, while Astragglus bisulcatpg A.GRAy grown on the same soil accumulated from 200 to #300 parts per million. Daniel (193a) found that legumes ordinarily contained 3.9 times as much nitrogen as did grasses. Plants that were low in calcium and phosphorus remained low in those elements, even.when grown in fertile soils. Those plants containing relatively high amounts of these elements always contained relatively large amounts even though grown.on a poor soil. It has been established that the variety of a plant also exerts an effect on its composition. Wolff (1877). from his study of the data in “Aachen Analysenz suggested that there were differences in composition depending upon variety. Since 1930, much information on the varietal difference in chemical composition of plants has been published. Such differences, probably governed by genetical factors, have been reported in oats, (Lundegardh 1931, 1932); apple, (Wallace 1932: Hill, Johnson, Heeney and Buckmaster 19513 Pdper, 1936): wheat, (Hanna and Dulac 1931;, 1931a, 1935a, 1936, 1937. 1937a, 1938a): and sugar cane, (Borden 1936: Beauchamp and.Alvarino 1990). Even the rootstock has been shown to have an effect upon the composition of the plant budded or grafted on it (Roach 1931: 5111 unpublished data). 41. The work of Yates and Watson (1939) showed that the differences between some varieties of a species may not be significantly great, in contrast to the rather large differences reported by most workers. These varietal differences in the chemical composition of plants make it essential to consider their magnitudes in any attempt to develop biological techniques for evaluating fertilizer requirements. The effects of variety on the percentages of nutrients in plant tissues implies the necessity of determining standard concentrations for each variety. The published literature seems to indicate that these varie— tal differences are merely expressions of the ability of plants to absorb nutrients, and do not reflect necessarily their specific physio- logical needs. As with any biological study a definite opinion cannot be ex- pressed concerning the necessity of setting up standards of composi- tion for each variety of plant. A.number of workers have considered that their data indicate that, in some cases.at least, varietal differ- ences and even species differences may be so small that general conclup sions are possible. For instance Graig (l9b1) concluded the same Optimal values for nutrient concentrations in sugar cane could be used for any of the studied varieties of this crop. The results ob- tained by Lundeggrdh (l9hl) for oats were considered by him to apply to other cereals as well. Wallace (191m) and Wallace and Osmond (19%) considered that, in general, the minimum levels for the concentrations of potassium and magnesium “,", found for apple trees would be generally applicable to other fruit trees. Crowther (1936), working with cotton, showed that while some varieties had a particularly high nitrogen con- tent, they responded to nitrogenous fertilizer to the same extent as did varieties normally containing much lower percentages of nitrogen. Even if it is admitted that there are a few instances in.which . generalities can be made, one must conclude that critical values must be related to a comparatively closely related group of plants, and must be applied to different varieties of the same species with care. 3.'Tissues There has been a considerable number of data reported that indi- cate that the various tissues within a plant differ in their chemical composition. As early as 180b, De Saussure noted that differences existed in chemical composition of the ash of leaves, branches, wood, bark, flowers, and fruit of seven different species of trees and also in the ash of straw and grain of‘wheat, corn, barley and oats. This tabula- tion.probab1y was the first of its kind and it led to the sampling techniques used at the present time. During the nineteenth century the significance of De Saussure’s data was not fully recognized and only isolated workers paid attention to them. In 1859,.Arendt studied cats, and he analyzed the ears, the two upper leaves, the three lower leaves, the highest stem internode, the two middle internodes, and the three lower internodes of the plant at 43. five different stages in the develOpment of the plants. Arendt's data not only confirmed the variations reported by De Saussure, but also indicated the striking differences between the samples taken at different periods during the growing season. Wblff (1877) also conduct— ed experiments with cats growing in.water cultures. He supplied successive increments of single nutrients under conditions of an adequacy of the others. After analyzing the grain, straw, chaff, and the roots, he came to the conclusion that minimum concentrations of nutrients could be specified for the aerial parts of the plants. ,A condensation of his data appear in table 3. The results of wolff, Arendt, and De Saussure convinced many workers that the analysis of the whole plant provided the best index to its nutritional status. Consequently, before 1900 Helmkampf (1892) and others followed this procedure because, they argued, a composite sample of the plant averaged the variations in its tissue. Other workers, for example Heinrich (1882) and Atterberg (1901), verified the work of De Saussure and believed that consideration should be given to the differences in the nutrient concentration occurring in specific portions of the plant. Atterberg concluded that the differences in the percentages of nitrogen and.phosphorus were greatest in cat grain, while analysis of the straw, or of‘whole plants, indicated the greatest differences in potassium content. Remy (1906; 1906a), working with various plants, reported that the concentrations differed in different tissues and believed that a Table 3. Necessary minimal concentrations of the sixzmajor nutrients in cats, as well as concentrations required for good average development. Based on analyses of the whole mature plants grown.in water cultures. Concentrations expressed as percentages of the dry material. (Condensed from Wolff 1876, 1877) Composition N P K Ca Mg S Necessary minimal 0.6 0.15 0A2 0.11 0.06 0.01» concentrations Necessary concentrations 1.0 0.22“ 0.66 0.18 0.12 0.08 for good average development #5. selection of tissue would indicate the nutritional status of the plant. Pfeiffer, Simmermacher, and Rippel (1919), working with oats, and Cook (1930), working with wheat and rye, again showed that the different tissues varied in their chemical composition at various stages during the growth of the plant. In a search for more accurate correlations between the nutrient contents of crops and their responses to fertilizers, workers have shown, incidentally, many examples of differences in the percentage compositions in closely related tissues. wallace (1928, 1929), works ing with gooseberries, found significant differences in the laminae and petioles of leaves. Emmert (1931), working with tomatoes and lettuce, found that the conducting tissues, petioles of tomatoes, and midribs of lettuce, all contained higher concentrations of nutrients than did the blades of these craps. Legatu and Maume (l930d) even showed differences in the concentration of nutrients in the leaves on.a single branch of the grape. The concentration.was greater in younger than in older leaves. Ulrich (l9h8) reported that the potassium content differed in the petioles and leaf blades of grapes and that the effect of fertilizer applications‘was expressed more clearly in.the petioles. Whereas there were only small differences in the nutrient concentrations in the blades of grape leaves as the result of potash fertilization as indicated in figure 3, the differences evident in the petioles were quite large as indicated in figure h. Ulrich's data also show that the leaf blades seldom contained nitrates, whereas the petioles did. 46. 8LAD€3 reacts/1' R IN DRY A66. SEPT. “32' “,9 . a 2.5 arr: as s AMPLE Figure 3. Potassium concentration in the blades of recently matured leaves of Petite Sirah grapes in Alexander Valley, California in 1940. (Ulrich, 19MB) #7. 3 Q a. ( N “I ’00 A t ‘3 s '0‘ *C 1.0 . g .05 p \ 1 \ Figure A. Potassium concentration in the petioles of recently matured leaves of Petite Sirah grapes in Alexander Valley, California in 1940, showing differences in its concentration induced by potash fertilizer. ( Ulrich, l9h2) 1+8. Thornton (1932) and Hoffer (1926) considered that their field testing methods could be pr0perly interpretated only if specific tissues were sampled. Lagatu.and Maume (1933) found differences between the leaves on fruiting and non-fruiting branches on the same grape vine. Sturgis and Reed (1937) reported that the nitrogen and phosphorus con- tent of the leaves and stems of rice‘was only one-third that of the panicle. Raleigh and Chucko (19hh), studying tomatoes, concluded that variations in the nutrient supply exerted a smaller effect on the com- position of fruits than on either the vines or the roots of the plant. A few recently reported data reveal differences in the concentra- tions of nutrients even within an organ itself. Emmert (1931) presented evidence that the composition of the edge of a leaf of lettuce more nearly indicated the nutritional status of the plant than did the central region of the lamina. Nightingale (19u2, 191+2a) founddiffer- ences between the meristematic base of the leaves of pineapple and the remainder of the leaf. Subsequently he sampled only the meri- stematic region as it appeared to reflect differences in the effects of fertilizers to the greatest extent. The differences in.percentages of nutrients in various tissues presents a serious problem to the use of tissue analysis for the deter» . mination of fertilizer requirements, and the most suitable tissue for analysis must always be selected. The significance of this situation will be discussed in a later section, and it is sufficient for our purpose to emphasize that the majority of investigations have shown #9. that the analysis of whole plants is unsatisfactory. 3. Diurnal Rhythms a. Nitrogen It was apparent to the early workers in the field of plant nutri- tion that consideration should be given to the effect of the time of day on the nutrient content of plant tissues, but in this early work, there were expressed differences of Opinion concerning the nature of these variations. Kosutany (1897) (see table 1+ and figure 5), Otto and Kboper (1910), Keoper (1910), and Susuki (1897) found increased amounts of nitrogen in.plants during the night, while Schulze and Schutz (1909), and Piagorini (1911+) found a decrease. Nightingale (1927) found that the period of daily illumination of tomato plants caused variations in their carbohydrate and nitrogen contents. Scarseth (1943) found that plants suffering from inadequate nitrogen had virtually no nitrate dur; ing the afternoon although he obtained positive tests during the early morning. Diurnal variations were shown also by Woodward, Shepherd and Tysdal (l94h) in the percentages of nitrogen in the dry matter of alfalfa. Their data are summarized in table 5 and figure 6. b. Phosphorus McCool (1926) working with the extracted saps of various plants, and Phillis and Mason (19h2) working‘with cotton leaves found that the concentrations of phosphate in the sap and the percentage in the leaves were higher during the morning than after midedayy which indicated a 50. Table h. variations in the percentage of nitrogen in the leaves of the grape during the day and night through the growing season. (Kosutany 1897) Percentage nitrogen Date in dry'material Day Night June # 4.161 #.616 June 19 b.25h “.276 July 3 '3.829 4.031 August 1h 3.h06 3.h61 August 28 3.994 n.0h8 September 25 3.158 3.119 October 9 2.680 2.258 Average 3.581 3.687 51. Farm. Nmoetu As 'I. or Day Hanan . l ‘ . . .‘to Jun JUL N“- " '0 I, 8 V ' {I ’ Figure 5. Variations in the percentage of nitrogen in the leaves of the grape during the day and night through the growing season. (Kosutany, 1897) Table 5. Percentages of nitrogen in the dry matter and of the ash in the dry matter of alfalfa har- vested at 2 hour intervals over a 2h hour period. May 30 and 31, 1941+. (Woodward, Shepherd, and Tysdal 19h4) Hour Percentage dz; Nitrogen Ash 7 AM 2.55 7.20 9 2.68 6.h8 11 2.60 6.12 1 PM 2.60 6.1a 3 2.97 6.09 5 2.5# 5.84 7 2.50 5.99 9 2.52 6.93 11 2.54 7.34 1 AM 2.60 7.92 3 2.66 8.67 5 2.62 8.08 7 2.99 7.66 53. 9“ 5 - E t z a D 1 ’- c a. O c 6+ g l ‘ l e z i < r . * 2 3 I . . ‘ ‘ A . . A g n a. L. 9 n An 3 I f D u .1" 8 I 1 - Figure 6. Variations in the percentages of nitrogen and of ash in the dry matter of alfalfa harvested at two hour intervals over a 21. hour period. May 30 and May 31, 19141.. (Woodward, Shepherd, and Tysdal, 19141.) decrease during the day followed by an increase during the night. Previously, Schulze and Schutz (1909) had reported that the percent- ages of phosphate in the leaves of the Box Elder were higher in the afternoon throughout the growing season. c. Potassium A.few data showing the diurnal variation in the potassium content of plants have been reported in the literature. Schulze and Sdhutz (1909), working with the Box Elder, noted that the potassium content in the leaves expressed as the percentage of the dry matter was higher in the afternoon than in the morning. Penston (1935, 1937. 1938) found that the amount varied from 2.5 to “.1 percent in potato leaves and from 3.7 to h.9 percent in maize over a 2h hour period. d. Ash Less attention has been paid to the diurnal rhythms of the total ash content of plants than to its individual components. Vbodward (19%) found that the total ash in alfalfa varied by as much as to percent of the mean depending upon the time of day when the material was harvested. His data, presented in table 5 and figure 6, show that the lowest percentage of total ash occurred at five P.M. and the highest at three A.M. lhile there are conflicting reports concerning the magnitude of diurnal variation of the total ash content of plants, such variations do exist. Consequently, the diurnal factor must be considered in the 55. use of any procedure for diagnosing the nutrient status of plants by the chemical analysis of their organs or tissues. 9. Season and Stage of Development a. Introduction Data showing seasonal variations in the chemical composition of plants frequently are misleading for there are two unrelated factors involved. The seasonal factor actually is a composite of such individual- factors as length of day, temperature, humidity, precipitation, etc. If a seasonal effect on the chemical composition of plants becomes evident, it is difficult indeed to generalize as to the cause. For example, the high temperatures in August, or the midsummer drowth, or some other factor, or combinations of factors may exert the dominant influence. But while the biotic factors comprising the seasonal environment of plants is passing through its normal, or abnormal phases in some cases, the plants themselves are changing as they proceed with their deve10pment. It becomes impossible, therefore, to separate the rela- tive importance of seasonal factors, pgg_gg, from the effects of the advancing maturity of the crop. Strictly speaking, seasonal effects should be regarded as separate from the developmental factor associated with the plants themselves, but this is not possible when crops are studied under field conditions. 56. Perhaps true seasonal effects on the composition of crops could be better studied by comparing crops grown in successive seasons, rather than by comparing them through a given season. But arbitrarily, the data that have been reported on the basis of the date of sampling are grouped as seasonal in our discussion. b. Nitrogen Over a period of years much information has been obtained indicat- ing that the concentration of the components of plants vary with the date of sampling throughout the season. Remy (1896) using wheat, oats, and barley, Kosutany (1897) using the grape, and Schulze and Schutz (1909) using the leaves of the Box Elder, all reported decreases in the percentage nitrogen in the dry matter of the plants as the season progressed and the age of the plant increased. Liebscher (1887), using oats, barley, and wheat, reported that in the early part of the season the accumulation of nitrogen was relatively higher than of total organic matter, whereas in the period of seed formation the accumulation of total organic material was greater. His data are presented in abbreviated form in table 6 and figure 9. Adorjan (1902) also showed that wheat, previous to the Jointing stage, contained the greater percentages of nitrogen. In fact, he reported that the uptake of nitrogen was three times greater during this stage than during the later stages of growth. This worker stated, 'In youth, 57. 00H 00H 03 00..” d: I... II II 1.. II II I... II II 1.. a.mm o.ma m.am m.am sea in in a- it in in it in u: in m.am a.om H.oa N.am. om m.ma ooa ooa. ooa aoa u. a. nu is in m.Na w.om a.am m.mm ma 00H m.ma m.mm N.am mm nu -u in nu :1 ~.n~ w.~a m.mH m.a Ne a.aa «.me m.am n.ma. mm a.aa ooa ooa ooa ma o.oa m.a H.0H H.a ma m.aa «.mm H.am a.aH an .ooa m.mm H.2m m.om am H.a a.m o.~ m.H an H.m m.m o.~a m.m mm o.na n.me m.aw a.an an m.o v.0 H.o m.o ca m.o m.o m.H n.o ma 0.3a m.me w.ms n.om ma wqaufidaa wawpqmaa m m N hounds poems N m N. hounds henna N m N unease noaumdfiahom o a o a a hen neon o a o a a sun neon o a o a a she hoped sang deco names .eoaq one aoeaomv flamed .Ho ho aoeaaoemv madman Ammma .hoflompoaflv memo no ounces aaafiuss.oma .aoaaca and ma uaouh chapsHoa on» end soapmaflauom seams when mo menses one nomsuop nofipeaom .OOH as some» mu uses “pence mono .muompsc uaoaamb hp dopaoama we macho adoaasb now own use .mo m .noMouuan .m oHpma 58. wheat takes up most of its nutrients, stores them, and then uses them for kernel formation.at a later time.” More recent work, however, has presented the opposite view. Salter and Amos (1928) concluded that in a general way, the rates at which ' nutrients are absorbed bear no constant relationship to the rate at which.p1ants increase in dry matter and that the nutrient absorption proceeds at an ever increasing rate throughout the season. The view of Salter and Amos is supported by their data which have been partially re— produced in figure 7 and table 6, and also by the data of Pfeiffer, Simmermacher and Rippel (1919), and of Pfeiffer and Rippel (1921) which are shown in table 6 and figure 8. These data present a striking contrast to those of Liebscher presented in table 6 and figure 9. This earlier work indicated that the greatest portions of dry matter, nitrogen, P205, and K20 were accumulated during the initial stages of plant deve10pment. The more recent work of Salter and.Ames (1928), of Pfeiffer and his co-workers, and of others shows evidences indicating that not only is this not true, but that the reverse situa- tion actually is a more accurate picture. The true significance of these data is that they show not only different levels of the various components in the plants through the season, but also show that these components are absorbed at varying rates through their life cycles. The majority of the work discussed was based on cereals. At the present time, more consideration is being given to the horticultural annual and perennial crops. It is apparent, that the seasonal fluc- tuations of the nutrient components in cereals also occur in other 59. i 1 N Page‘s Kto or on, ‘5 '2' '3 Runmve Y'slos 20 *0 ‘0 80 '00 0 A75 seen PLAN-ruse Figure 7. Total dry weight, nitrogen, phosphorus, and potassium in corn grown at Wooster 1927, and sampled at various times throughout the season, expressed as percentages of their maximnn values. ( Salter and Amos, 1928) 60. 0 j 3 Rem-rave YIELDS a: DN,N,P.,05-,K. .ue-loooo. . l l 20 4O 60 80 I00 Figure 8. Total dry weight, nitrogen, phosphorus and potassium in barley grown at Breslau, l9l9, expressed as percentages of their maximum.values. ( Pfeiffer and Rippel, 1921) 61. Reuflvs mm or on ~,P.o., K.O l l J 30 ‘50 00 .0 I” DA y 0 non GIRHINATION Figure 9. Total dry weight, nitrogen, phosphorus and potassium in oats, expressed as percentages of their maximum values. ( Liebscher, 1887) . 62. crops. Murneek (1930) concluded that the growth and development of the apple tree in early spring is largely supported by the nitrogen stored in the tree during the previous year. Consequently, all woody structures of the tree showed a progressive decrease in nitrogen con- tent from early spring until active growth had ceased. The nonebearing spurs had a maximum nitrogen content at the time of bud swelling and a minimum content when vegetative elongation had ceased. There was a seasonal percentage decrease in the nitrogen in the leaves. The value decreased from 4 to less than 1 percent. Murneek found that from 35 to #5 percent of the total nitrogen in the leaves was transported back into the tree prior to abscission. The soluble nitrogen in the leaves I usually increased during this period while the insoluble fraction decreased. Hulay (1931, 1932), studying a six year old Bartlett pear tree, found that the percentage of total nitrogen in the shoots began to increase at the end of October, reached a peak in December, remained constant until February, and sharply decreased in March.when new growth began. In flay it had reached a minimum. The 'foliar diagnosis” technique deve10ped by Lagatu and.Maume (1934, 1937. 1937a), Thomas (19h5), and Thomas and Mack (19h0, 19h4) has been used over a period of years to show that the concentration of nitrogen and the relative amounts of nitrogen in the group of componp ents, NPK, vary through the season. An example of this seasonal varia— tion is shown in table 7 and figure 10. These data were obtained from 63. Table 7. Concentrations of calcium, nitrogen, phosphorus, and potassium, expressed as percentages of the dry matter, in the third leaf from the base of the main stem of the potato at various dates through the growing season. (Lagatu and Maume 1930) May 2 May 22 June 2 June 11 June 19 Check plot Ga 5.65 5.28 8.40 9.40 9.40 N’ 3.45 2.56 2.22 1.98 1.92 K O 5.60 4.61 4.72 4.62 5.01 P3205 0.88 0.77 1.04 0.82 0.8? Nitrogen treated plot Ga 6.60 8.47 9.74 9.31 9.12 N 4.35 3.11 2.80 2.52 2.55 120 4.80 4.90 3.81 3.66 3.81 0.62 0.51 0.79 0.39 0.38 9205 64. .__.__.CA HPIOI II- ,____,,N o———o Koo 5 b" '— ( :74 > 1 CD 1." ° “\V E. \\_A I.~13.\.\__~ u x , m Q’PW W may JUNE May Jun: auxin“? anal/v9 CHECK NITROGEN Figure 10. Concentrations of nitrogen, calcium,potassium and phosphorus expressed as percentages of the dry matter of the third leaf from the base of the stem of the potato through the growing season. Samples taken from a plot receiving nitrogen and from one receiving no fertilizer. ( Lagatu and Maume, 1930) two plots, one unfertilized and one treated with nitrogen. The ferti- lized plot yielded 264 pounds more potato tubers than did the unfer- tilized. Table 7 and figure 11 show that the nitrogen decreased in its percentage of the dry matter as well as in its relative amounts with respect to the other elements. c. Phosphorus Analytical data based on many samples of many plants reported by various workers indicate that the concentration of phosphorus in plants exhibits less variation than do the other major nutrients. In a study of the composition of the leaves of the Box Elder, Schulze and Schutz (1909) found a gradual decrease in the percentage of phosphorus from 1.697 percent of dry matter in May to 0.374 percent in September. Since this early work, the gradual decrease in the phosphorus concentrations in tissues during the growing season.has been confirmed by many investigators. Pfeiffer et a1 (1919), for example, reported data showing that the phosphorus concentrations, expressed as percentages of P595 in the dry matter of barley leaves, decreased from 01462 percent in May to 0.178 percent on July 23. The phosphorus content of the straw showed the same relative decrease during the progress of the season. The percentage in May was 0.624 and on.Ju1y 23 it was 0.2626. In a later report covering the three-year period from 1918 to 1920, Pfeiffer et a1 (1921) confirmed the previous 66. 0‘5 e z ' \ 2 cuacn 30 o- 3 \/( o :3 5‘31 3?) P205 Figure 11. Seasonal effect of the amount of nitrogen, P20 , and K20 in the NPK-unit in the third leaf from the base of the stem of the potato through the growirg season , Montpellier 1928. The numbers 1, 2, 3, h, and 5 represent the sampling dates given in table 7. (Lagatu and Maume, 1930) 6?. conclusion that the percentage of phosphorus in the dry matter decreased through the season. Hewever, there are data conflicting with those of Pfeiffer, for AdorJan (1902) showed that, during the Jointing stage, there was a high rate of phosphorus absorption.which was related to the require- ments of the inflorescence. .After jointing, no significant amount of phosphorus was absorbed. More'work has been done on the seasonal fluctuations in the con- centration of phosphorus in plant tissue by the use of the foliar diagnosis methods than by other procedures. Data of this type have been reported by Lagatu and Maume (1934), Thomas (1945), and Themas and mask (1940). The work bf Legatu and Thomas does, in a general way, confirm the decrease in the concentration of phosphorus with the advancing age of the plant tissues; they also make it evident that this decrease may not always occur because of the influence of climate. Some of their data showing the complexity of seasonal phosphorus fluc- tuations are given in table 7 and figures 10 and 11. In 1928, the phosphorus concentration in potatoes reached a maximum value in.the tissue on about June 2 in both the treated and the untreated craps, and decreased after this data. Figure 11 illustrates this same effect is seen in the relative amounts of phosphorus in the unit, NPR, in the plants. During this work, the authors noted that the temperature rose sharply around may 24. They believed that this sharp rise in temperature caused the deviation from the normal gradual decrease in phosphorus as growth progressed. 68. d. Potassium 0f the three nutrients, nitrogen, phosphorus, and potassium, potassium exhibits the most consistent relationship between its per- centage in the plant and the advancing season. Investigators consis- tently have reported decreases in.potassium concentration.with.advanc- ing season, among them Schulze and Schutz (1909) working with the leaves of the Box Elder, Liebscher (1887) working with cats, Pfeiffer et a1 (1919, 1921) working with barley, and Salter and Ames (1928) working'with corn. Some of the data of these workers are assembled in table 6 and figure 7. Legatu.and Maume (1930, 1934) considered that, while there was a gradual decrease in the potassium concentration in potato tissue, this decrease'was not as striking as that of nitrogen. This conclusion'was based on the data in table 6 and figure 10. These workers also showed by the data presented in figure 11 that the amount of potassium relative to total nutrients of the group NPR decreased as the season progressed. Ulrich (1942) concluded from his data based on the blades and petioles of Petite Sirah grapes in 1940 that the percentage potassium in the dry matter decreased through the season even though his experimental plots had been variously fertilized. Many workers have described this effect of season, among whom are Burd (1919), Duley and Miller (1921), and Knowles and Watkins (1931). 'hile only the three major fertilizer components. nitrogen, phos- phorus and potassium, have been discussed, the data presented emphasize 69. that one of the most important factors involved in the determination of the fertilizer requirements of plants by biological responses in tissue composition is the seasonal variation of the concentrations of nutrients in the plants. The resolution of this problem is complicated further by seasonal variations in the supply of nutrients in the soil. Reali- zation that such variations exist makes it apparent that the date of sampling influences the data obtained. Careful standardization of the sampling date, therefore, is necessary if comparable and interpretable data are to be gathered. 1A correct sampling date must take into con~ sideration all of the factors affecting the concentration of nutrients in.plants. D. External factors Affecting the Concentration of nutrients in.Plants 1. Climate a. Illumination A.litt1e work has been reported on the effect of illumination on the composition of plants. Although the data are not conclusive, they should be taken into consideration in any attempt to relate the chemi- cal composition of plants to their fertilizer requirements. Mitchell (1936) reported that the percentages of phosphorus and potassium in White Pine seedlings were influenced by the amount of illumination the plants received. Studies carried out by the Rhode Island Experiment Station showed that the nitrate content of the best also was considered to be affected by light (3.1. Agric. Exper. Sta. 1935). wark (1938, 1939) believed that cereals were so influenced by illumination that samples for analysis should be taken only in.bright weather, if they were to be used for nutritional diagnosis. Pfeiffer, Blanck, and Iriske (1913) and Pfeiffer, Blanck, and Flungel (1912) showed that shading of cats growing in pot culture induced an in- creased percentage of nitrogen in the straw. These data are pre- sented in condensed form in table 8. b. Water Tollens (1902) reported the results of his work on potatoes, which had been published previously by Drszewski and Tollens (1900), ‘which indicated that the percentages of total ash increased in.potato tubers as the available water in the soil increased: and he concluded that the percentages of nitrogen, potassium, and phosphorus in the ash, as well as the amount of dry matter, are all influenced by plant food in the soil and also by its water content. Pfeiffer, Blanch and Ilungel (1912) grew cats in pot cultures receiving different amounts of nitrogen, and he observed that excess water appreciably decreased the percentage of nitrogen in the dry matter of the straw. The data are condensed in table 8. Pfeiffer, Blanck, and rrloho (1913) carried out an extensive investigation covering the growth of cats in seven different soils, each.at four levels of moisture. The data summarized in table 9 show that the percentage nitrogen is decreased while that of phosphorus is increased by additional water. Potassium appears to 71. Table 8. Percentages of nitrogen in the straw of barley, grown in pots, as influenced by increasing amounts of water. (Pfeiffer, Planck, and F1unge1 1912) ' Nitrogen Percentage nitrogen in dry straw added (grams) Intermittent Moderate Excess Excess water water water water Plants shaded 0 0.58 0.50 0.11 0.5? 0.355 0.63 0.57 0.108 0.57 0.710 0.82 0.55 0.308 0.62 1.065 0.76 0.79 0.506 0.76 1.b20 1.01 0.87 0.659 0.91 1.775 1.71 1.09 0.885 1.01 2.110 2.00 0.98 0.835 0.87 72. Table 9. Effects of different moisture levels on the relative concentrations of nitrogen, phosphorus and potassium in cats. The concentrations obtained at the lower amount of water are taken as 100. (Pfeiffer et a1 1913) Year Water level N P205 K20 1911 Low 100 100 100 High 7 7 126 109 1912 Low 100 100 100 High 81 131+ 112 be only slightly increased. These data confirm those of a similar nature obtained by Seelhorse and Tollens (1901), and.later also those obtained by Pfeiffer et a1 (1919), Richards (1932). Yates and Watson (1939), Lilleland and Brown (19%), and by Greaves and Carter (1923). It is not possible, however, to know if this effect is due to the water alone or to the greater amounts of nutrients made available to the plant by the water. c. Temperature The literature does not reveal any information on the direct effect of temperature on the composition of plants. However, the fact that temperature might influence the percentages of the elements in the plant ash has been considered by'a number of workers. For ex» ample, Tbllens (1902) believed that the percentage of potassium, phosphorus, etc. in plant ash as well as the amount of dry matter are influenced by variations in temperature, perhaps as they affect the soil. Traps (1931) also noted that ”the interpretation of chemical analysis as related to the capacity of the soil to supply plant food is affected by various factors including the temperature.“ In.w0rking with potatoes in 1928, Lagatu.and Maume (1930a) obtained the followb ing experimental data from which an effect of temperature may be de— duced. The phosphorus content tended to reach a maximum value in all curves at a period between June 2 and June 19. A.burst of growth of the potato plants also was apparent during the same period. The average temperature until May 22 was 10° c but betwaen May 22 and June 2 the average temperature increased rapidly to 210 C and re- mained approximately constant at 210 C after June 2. These workers considered that the deviations from the normal concentrations of phosphorus in the plantS'which occurred could be attributed to the influence of temperature alone. d. Weather Weather represents the combined effect of moisture, temperature, illumination, and all the other physical factors comprising climate. Certainly, this complex of factors induce important changes in the composition of plants and unfortunately its influence cannot be con- trolled in field experiments. Lagatu and Maume (1930) aptly stated the influence of the physi- cal factors constituting weather on the accumulation of nutrients in leaves as follows: “There exists what may be designated a 'chemism' due to the influence imparted by the fertilizer itself and also a 'chemism' due to the effects of meterological conditions.I Workers in the field of plant nutrition are well aware that the same fertilizer treatment may produce very different crop yields in meterologically different years. The 'foliar diagnosis“ technique of Lagatu.and Maume has shown different percentages of nutrients in.plants grown.in different seasons varying in weather. They detected these differences in potatoes (1930a), and in the grape vine (1928). 75. Their data based on grapes are condensed in table 10 and figure 12, and they clearly show the effect of the climatic factors on the con- centrations of nutrients in the leaves, even though the same plants were sampled during successive seasons. In a season of adequate rain- fall and optimum temperatures (19210 the yield and the levels of nutrients in morphologically similar leaves were greatly increased (1925). These workers placed great stress on the necessity of evaluat- ing climatic factors in any procedure involving the determination of nutrient requirements of plants by foliar diagnosis techniques. They stated (19310 that, “in an area which can give, for the same season, meteorological conditions that are very different from year to year, the problem of fertilizing a cultivated plant can not be definitely solved. haployment of fertilizers is speculative only, depending for its value on many factors." Thomas and Mack (1941) also demonstrated the influence of the climate on the nutritional status of tomatoes grown in the greenhouse in the spring and fall, and reported striking differences beWeen the percentage of nitrogen, phosphorus and potassium in the dried foliage. This work was conducted under the differences in such climatic condi- tions as light intensity and length of day occurring in the spring and fall in contrast to the previous work of Lagatu and Maume (1928) which was conducted under different conditions of moisture and tem- perature. The climatic effects are seen in table 11 and figures 13, 11b, and 15. The yield was always higher in the spring than in the Table 10. 76. The effect of two meterological different seasons (1921+ and 1925) on the nitrogen, phosphorus and potassium in the third leaf from the base of the main stem of potato plants grown on a check plot and a plot receiving a complete fertilizer. The data are ex- pressed as percentages of dry material. (Lagatu and Maume, 1930)) Season 1921+ Season 1925 Annual rainfall 725.1 mls. Annual rainfall 529.1 mls. Date of Date of sampling 1! X20 P205 sampling N K20 P205 Unfertilized plot ' “May 10 ltJtO 2.05 0.89 May 11 1.98 1.66 0.50 gJune 11+ 2.64 . 1.88 0.59 June 14 1.65 1.39 0.32 9 EJuly 12 2.55 1.65 0.51 July 12 1.50 1.50 0.26 ELVIS.“ 9 2.00 1.142 0.144 Aug. 9 1.20 1.145 0.11-t asept. 13 2.20 1.14 0.1m Sept. 13 1.35 0.83 0.13 - Fertilized plot (Annual application since 1919 -------- 80 Kg. N, mo consisting of 80 Kg. nitrogen 90 Kg. K, 1233“ 75 Kg. P, Superphosphate May 10 h.5 n.37 0.74 May 16 2.59 1.98 ‘ 0.80 gJunc 11+ 3.1a £7.12 0.58 June 1'4 2.15 1.71 0.70 OH Eduly 12 3.0a n.05 0.5!; July 12 1.95 1.66 0.50 EAug. 9 2.78 3.15 0.51 Aug. 9 1.30 1.73 0.36 asept. 13 1.62 2.32 0.32 Sept. 13 1.75 1.80 0.25 a Unfertilized 560 Kg. #40 Kg ; Fertilizer]. 795 Kg. 540 Kg 77. ‘ Tntxrco , o——-o N s .-—--O “‘0 E ”1-.....“ P303 1: y. K O i W E Q. JunduyMM-On M0 J A..S‘. “ I929' '“Uggug' ”to" Figure 12. The effect of different climatic seasons on the yield and chemical composition of the leaves of the grape during the seasons 1921. and 1925. (Lagatu and Name , 1928) 78. m.m H.NN m.~a m.emm a.aa m.ma .m.~nm o.w nm.o a.a n.n Hm .aoo H.a a.aa s.ms a.swm m.a~ a.sa a.mmm o.c m.H m.H a.m ma .aaom H.a m.ma m.ow n.Nna o.mn a.mm a.~an H.m a.” m.a N.m m .aoom m.m m.a~ a.mm a.nom a.aH H.aa N.naH m.m ma.o N.N a.~ am as: n.m m.o~ m.~a c.mmn m.om a.ac o.amm a.m ma.o m.a n.m am ”Haas m.a ~.Na n.ma m.aan m.am n.Hm n.mmm a.a a.a ~.H w.: m Hausa aoHa eosaaaaaoa.uannmv m.n a.sm a.mw H.nmm n.a~ m.ow ~.nmm o.c o.a m.~ H.m Hm .moo v.5 a.ma a.ma e.amm w.am a.ma 0.5mm H.a e.H a.H 0.: mm .aoom o.w m.sH o.aa H.nma «.mm a.aw m.anm N.m a.” m.H a.e n .apnm ~.m «.ma m.ae a.snm m.ma m.aaa s.HmH n.n mc.o m.~ m.H am as: a.a n.mm m.am H.mmm ~.sH a.ooe a.awa s.m ac.o n.~ a.~ am Hates m.a ~.H~ o.na e.wam H.mH m.aa m.ma~3 m.m mm.o m.H m.m m Hates upoaa wouaaupnom nahomamogm m.e H.0m a.mm «.mam n.nH s.aa H.ao~ as.m «v.0 on.a oa.~ HN .eoo s.m m.HN m.~a a.mmm «.mfi m.wa m.Hmm Ha.m om.o mw.H cc.m mm .acom m.n a.aH c.aa a.ams m.a~ m.sc n.mnm mp.s am.a mm.H am.a m .aoem c.a m.ma m.am m.cam m.ma a.amfl m.NmH mw.m no.0 ao.m aa.~ am as: m.a n.~m ~.mw m.amm H.¢H a.mo~ H.aom ao.m aw.o on.a oa.~ am Hades H.m H.om m.Ha a.man m.Hm H.aa a.~mm mm.a an.” am.a ma.n m Hates epoHa nonaawaueu collaoam momma ea ml 03 on a & R & can mama n a cum “can u a a a can Was." u wanna: so once caaqslxmz no soaaanoesoo reassurance seawsflaax. octagon no accuses Heston: Addmd .300: one eds—on.a .anne 23253.3 on» Mdubwooeh nun—dam]! oudfioa Ho HHGM 05 dd find Uflfihmu 05w 9.“ Hflod fiahuh 0:» No a «gig on» No fibuauuofisoo on.» «and .uofioflgufioo EthHHHuE .GMNdHON .00th one an CNN Udd . CNN .2 ho ctMflafioohonm .HH OHnflR 79. K 5 . c ’ ' u 8: d I: D 3 ' a O ‘ y . I r o. I 2 -._ / m f‘“ s 1‘ 7%" . I ...O I. s P ‘h... .5 L I s o s o .0 C O O I C. C C l O . . DA". on Banmuo Yuma In... M M “oh. (01.31:. 623m. —— N h———‘ “0.. u ............ . “‘0 Figure 13. Changes during growth in the percentages of nitrogen, P 20 am K 20 in the fifth leaf of Spring and Fall crops 20? tomatoes receiving the treatments indicated. Yields expressed in pounds per plot are indicated. ( Thomas and Mack, 1941) 80. ¢ 8 r u I: < Z 1‘ s a E O a; .f , .__————a(2.~)PK on m" _____ - p i t ................ 0 CHECK o 9. ’ p. a U 0 C e l l l l l l . . OCT. Avg“. saga :29"; 51;! air; a! Figure 1L. Percentages of N +P205+K20 in the fifth leaf of tomatoes in the Spring and Fall, fertilized as indicated. (Thomas and Mack, 191.1) 81. Spam. ,___._.mn FAU- o———OCN ‘CK o-----C(IN)'“ Figure 15. Changes during the groch of tomatoes in the N'PZOS'KZO equilibrium in the fifth leaf. The plants were grown in Spring and Fall with fertilizer treatmnets as indicated. (Thomas and Mack, 191.1) 82. fall except in the phosphorus series which suffered from the Egsarium wilt (figure 13). The concentrations of nutrients in the fifth leaf from the base was always higher in the fall than in the spring although its yield of dry matter is lower (figure 1h). The relative percentage of nitrogen is higher and that of potassium is lower in the NPK unit during the fall (figure 15). The percentage of nitrogen in the dry matter of the leaf tends to be higher in all treatments in the fall. while that of potassium tends to be variable (figure 13). 2. Pests and Diseases The effects of pests and diseases on the chemical composition of plants probably are those most frequently overlooked by students of the nutrient concentrations in plants. Eaton (l9h#) has shown that nutrient deficiency may sometimes result in an increased susceptibility to disease and this situation makes it difficult to separate the effects of nutrition from those of disease. It is, of course, true that dis- eases and pests may upset the nutritional status of the plant. Ihe more apparent diseases and pests are easily recognized and may be avoided in the collection of samples, but real problems may arise‘when pathological conditions occur which are not evident to the person collecting the samples, and which may lead him to erroneously interpret nutritional data based on chemical analysis. l. 1.1? 83. 3. Management The management, or “farm practice", frequently is overlooked as tan important factor influencing all biological tests of soil fertility. The quantitative effect of this indefinite but important aspect of fertility studies is difficult to determine. Only a few data will be presented to indicate the subtle influence of the mechanical routines involved in field management. Thomas, Mack and Cotton (19142) studied the effects of irrigation practices on tomatoes, and Thomas, Mack and Rahn (19%) reported differ- ences in crop response due to the placement of fertilizer. A number of Workers have reported differences in crop response and in the compo- ssition of plants due to the type of crop rotation and cover crap: eand it is of course obvious that legumes will fix nitrogen which can 1be used by succeeding crops. Some crops have a considerable power of extracting nutrients from compounds in the soil that are relatively unavailable to other craps. Russel and Russel (1950) have shown that Ilupins are good extractors of phosphate. Rogers, Gall and Barnett (1939) suggested that the best way of preparing land for maize was to aallow the native weeds to cover the land and then.to plow them in. In ‘this way the weeds were shown to accumulate and then to supply avail- aable zinc to the succeeding crop. They suggest also that alfalfa, 1which is a strong extractor of zinc. would be a valuable cover crap in orchards. Ii'he effect of irrigation has been noted previously and emmales have been given from the work of Pfeiffer et a1 (1912, 1913). This work, however, was carried out in pot cultures. Data based on field irrigation studies are presented in table 12, and they may serve as examples of the type of effects that might be attributed to this phase of field management. These data, taken from the publications of Greaves and Carter (1923) and Greaves and Nelson (1925), were based on the growth of wheat under irrigation at Logan, Utah. The results show that increasing amounts of irrigation water decreased the per- centages of calcium, magnesium, phosphorus, potassium and total ash in the dry matter of both the straw and grain. Lagatu and Maume have conducted experiments showing the effect of different sources of the three major nutrients on their concentration in the dry material of various crops: for example, the effect of var- ious sources of nitrogen on the composition of grape leaves (1930b), various sources of phosphorus on tobacco (1935), and various sources of phosphorus on the grape (1936, 1936a) all were investigated. Their work shows that equivalent amounts of phosphorus or of nitrogen exerted a different effect on the concentration of these elements in the plants. if they are supplied to the plants in diffbrent compounds. 'Lagatu and Maume (1934b) observed an increase in the percentage of nitrogen from 0.3 to 0.5 percent of the dry matter and an increase in the phosphorus from 0.1 to 0.3 percent due to tillage practices alone. They also observed that tillage practice increased the rela- tive amounts of nitrogen and phosphorus in the NP! unit in the plants, while it decreased the relative amount of potassium. 85. Table 12. Percentages of nitrogen, phosphorus, potassium, calcium, magnesium, and total ash in wheat grain under irrigation at Legan, Utah. (Greaves and Carter, 1923) Inches of Percentages in dry material irrigation water applied Ca Mg Rf P K Ash 0 0.103 0.170 2.39 0.295 0.396 1.56 5 0.107 0.171 2.16 0.301 0.1m 1.56 10 0.122 0.172 2.18 0.306 0.h39 1.57 15 0.165 0.172 1.99 0.323 0.b91 1.71 20 0.195 0.198 1.98 0.371 0.h90 2.01 35 0.211 0.207 2.01 0.h58 0.534 2.28 67.5 0.262 0.22» 2.06 0.421» 0.535 2.19 86. The time at which fertilizer is applied also is a phase of field management studied by Legatu.and Maume (19300). Using the grape, . variety Rupestris, they established five plots which were cultivated on Nov. 21, Dec. 21, Jan.\21, Feb. 22 and March 28 respectively. One— half of each plot received a complete fertilizer while the remaining half served as a control plot. The data show a significant increase in the amount of potassium absorbed from the fertilized plots when the fertilizer was applied Nov. 21, but this effect decreased gradually and steadily as the time of application was delayed from November to March. The subtle effects of management factors often are overlooked in experiments with fertilizers. Certainly it must be borne in mind that the mere addition of fertilizers to a soil does not ensure a predictable rate of absorption of the nutrients by the plants. h. $011 The theory of the use of biological techniques to determine the amounts of plant nutrients in the soil, and to determine thereby the fertilizer requirements of creps depends on the ability of plants to truly reflect the nutrient content of the soil. Even if all factors were constant except the amount of’available nutrients, the develop- ment of the root system would still influence the results. It is obvious that, if the deve10pment of the root system itself is a reflection of the availability of nutrients, then no real problem is 87. involved, but if factors other than nutrients are acting upon the root, then a very real problem of interpretation may appear. Such non-nutritional factors as the presence of a hard-r , a high.water table, or a zone of toxic salt accumulation have been investigated by Lillel and Brown (19111). Soils with adequate quantities of a given nutrient may be unable to supply an adequate amount for plant growth because of interference of other ions. For example, Johnson (192k) reported that the absorption- of iron.was prevented by high concentrations of manganese in the soil, and Boynton and Burrell (l9hh) found that magnesium deficiency was in? duced in.app1e trees on soils repeatedly fertilized with potassium salts. Pineapple plants have been shown to be unable to absorb nitrates from soils which.were low in potassium. In this case, Nightingale (19u2) considered that an adequate supply of potassium was necessary for normal nitrate absorption. Ulrich (19h2) found that the addition of nitrates to the soil stimulated the absorption of potassium by the roots of grapes. It is probable that this was due to the greater deve10pment of the root system induced by the nitrogen. These data emphasize that a very complex set of conditions controls the type and quantity of the various nutrients absorbed from the soil by the roots of plants. The ability of the plant to absorb nutrients in such.a manner as to permit the pr0per interpretation of the complexb ity of the response of craps to fertilizers will be discussed in the succeeding sections of the present discussion. .m‘, UU. III. FIELD TRIALS A. Introduction The basis of all biological methods for the determination of fer- tilizer requirements of crOps is the use of plants as extracting agents. These procedures are in contrast to chemical methods of analysis which utilize various chemical solutions to simulate the extraction.power of the plant. Both of these methods are attempts to analyze the soil so that estimates of the fertilizer requirements of crops can be made. The field plot experimental procedure is the oldest, as well as the best known, of the biological methods of determining fertilizer require- ments. lundamentally, the method involves the application of ferti- lizers to the soil in various combinations and in various amounts, and then.using the responses of the crops as indications of the proper kinds and amounts which.may be profitably added to the soil. It has already been.pointed out that the historical experiments of Boussingault (18h1) were carried out either in the laboratory or in small pots. About 1830, however, Boussingault began a series of field experiments in.France. He weighed and analyzed the fertilizers used and the crops he obtained. At the end of each crap rotation, he prepared a balance sheet to show to what extent the fertilizer had satisfied the requirements of the crop. The results of one such experiment are given in table 13, and they illustrate his application of the bookkeeping method to the maintenance of soil fertility. This 89. Table 13. Statistics of a rotation. (Boussingault, 1841) Crop Yield in kilograms per hectar Dry Carbon Hydrogen Oxygen Nitrogen Mineral matter matter Beets 3172 1357.7 18h.0 1376.7 53.9 199.8 Wheat 3006 1031.6 160.0, 1210.8 31.3 163.8 Clover Hay 0029 1909.7 201.5 1523.0 84.6 310.2 Wheat #208 2004.2 230.0 1700 .7 1+3. 8 229.3 Turnips 716 307.2 39.3 302.9 12.2 50.4 Gate 2347 1182.3 137.3 890.9 28.0 108.0 Total during 17078 8192.7 956.5 7009.0 250.2 1065 .5 rotation Added in manure 10161 3637.6 026.8 2621.5 203.2 3271.9 Difference not +7317 44655.1 +529.7 H3876 +51.0 ~2206J+ accounted for taken.air, rain, or soil. 7 .. L. 90. bookkeeping method.was used also by Liebig who strongly advocated it as a procedure to study mineral fertilizers some 20 years later. Field plot techniques obtained their greatest impetus from the difference of Opinion between Lawes (1847) in England and Liebig in Germany concerning the importance of ammoniaoal fertilizers. Liebig had deve10ped a patent fertilizer containing phosphtes and alkalis but not nitrogen, for he believed that the nitrogen used by the plant came from the atmosphere. Lewes did not believe that the atmosphere was the true source of the nitrogen, and to prove his point established field experiments at Rothamstead. He obtained greater crop fields when ammoniacal fertilizers were used in addition to Liebig's patented fertilizer. The Rothamstead experiments were essentially the same as those performed by Boussingault but they had the added advantage of continuity. The meticulous methods used by Lewes and Gilbert in re- cording the results obtained from their field trials have given a cone stant source of valuable information to statisticians studying ferti- lizer use and crop production. The success obtained by these early workers with their field ex- periments led many investigators to question the value of pot and ' solution cultures for the estimation of the fertilizer needs of the crop. we1£r (1877) concluded, after a series of water culture experi- ments on oats, that 'the fact that they were conducted in water cul- ture somewhat invalidates the results and tends to prevent a valid field' interpretation.’l 91. A report published in the ”Journal de Agriculture de Paris” by Joulie (1889) presented the details of the field techniques he used in the improvement of the soils on his farm. Using diagnostic soil and plant analysis, he made recommendations concerning the fertilizers to be applied to the soil. He then grew the cr0ps, analyzed them, and compared the data with 'composition.types" based on averages of numerb one data from plants of good growth. Any nutritional deficiencies becoming apparent were remedied. Joulie was able to reclaim soils which previously had been unable to support good crops by following this procedure. During the nineteenth century, the deve10pment of field trials progressed rapidly.‘ In spite of the additional work and expense in? volved and the uncertainty of the generalizations drawn from them, - field trials have been used on an ever increasing scale during the past one hundred years. When used properly, their value is unques- tioned, and when data obtained by them are subjected to statistical analysis, their usefulness is very great. B. Statistical Analysis of Data 1. Introduction The application of field trial techniques to crop responses was very haphazard until the relatively recent advances in plot design and the statistical methods of analysis developed by Fisher (1935). These newer techniques were applied by such workers as Snedecor (1937), 92. Goulden (1939), and Paterson (1939). Fisher's techniques have so revolutionized the use of field trials that they now represent an exact science, at least in comparison to their previous empirical status. Field trials may not be regarded as a fundamentally sound biologi- cal method, but more accurately, they represent an overall attempt to improve cultural practices in which the ultimate measure of improve- ment is profit per acre. The general practices of field cultivation us ed today are of a relatively high level of efficiency, and any im— provements in crop yields cannot be as great as those obtained by early workers. Small differences in yield are difficult to evaluate without Draper statistical study which takes into consideration all possible measureable causes of variation in the data. Since it is apparent that there are factors over which man has no control, involved in the use of field trials, it is essential that conditions be controlled as uniformly as possible during the progress of the field trials. The results of any one field trial have only a limited application, for they theoretically are true only for the particular soil and season 1ll‘volved. To establish a general law, a number of separate experiments 0:1 various soil types and during several seasons must be conducted, 94116. this mass of data used to prove the authenticity of the conclusions. The variable nature of both soil and season poses an unavoidable Obstacle to the easy solution of fertilizer problems by field trials. 80 ils vary in fertility, not only from acre to acre, but even from root to foot. This makes it impossible to use identical soil conditions 93. for various fertilizer treatments. It is quite.conceivable that an apparent difference may be due to the difference in the original fer- tility of the plots. Similarly the prevailing climatic conditions in any season may favor certain of the treatments. Even slight changes in climate might be sufficient to cause significant changes in the relative merits of the experimental treatments. The theory of statis- tics and experimental design permit the development of techniques which, although they cannot completely eliminate extraneous effects, can minimize them to such an extent an approximate appreciation of the relative merits of the experimental treatments can be realized. Since the application of statistical methods to field trial data is so essential a general discussion of the methods of minimizing error will be presented. 2. Experimental Area The importance of the careful selection of the site for experi- mental field plots cannot be overemphasized. The area should conform to the general soil and environmental conditions under which it is intended to apply the results of the experiment. .At the same time, it is essential to select the most uniform area available in order to lessen the effect of the variable soil fertility of plots which are to be compared. Recently, an erroneous concept has developed that this is not of vital importance because of current improved techniques. While it is true that the currently used plot arrangements considerably 9#. reduce the effect of soil heterogeneity, they cannot eradicate its effect entirely. The more uniform is the site, the greater will be the probability of obtaining a true evaluation of the fertilizer treat- ments. On the other hand, it must be borne in mind that a field trial is conducted under artificial conditions, even though the site is uni- form. The results must be interpreted as to permit their application to-areas not so uniform as the plots themselves, for uniform agricul- tural areas are the exception rather than the rule. 3. Replication Statistical theory states that a single large plot cannot give data of any value for comparative purposes, and consequently the experi- mental treatment should be divided into a number of smaller but similar plots. with replication, the increase of plot size is effective, to a certain extent, in the reduction of error, but replication reduces the standard error of the mean of one treatment at a rate preportional to the square root of the number of its replications. while the value of the replication of experimental fertilizer applications is realized by the majority of agricultural workers, there are cases'where it is not only not necessary but may even be undesirable. For example, the techniques developed by Legatu and Maume (193b) involve such masses of similar data accumulated over many years that these workers now consider further replication to be unnecessary. 95. h. Randomization Statistical significance is based on the assumption that the estimates of the means and standard deviations obtained from the data approximate the true values that would be obtained from an infinitely large number of replicated plots. Such a situation requires that plots duplicating any one treatment be located at random. Fisher (l93b) states that “an estimate of error will only'be valid for its purpose if we make sure that in the plot arrangement, pairs of plots treated alike are not nearer together, or further apart than, or in any other relevant way, distinguished from pairs of plots treated differently.“ On the other hand, there tends to be a close similarity between the fertility of adjacent plots, and in consequence there is a greater probability of demonstrating real differences between treatments if they are carried out on adjacent plots. Bray (1948) in his work:vith corn in Illinois, considered that his techniques require that the plots are not randomized. one must conclude the necessity of each experiment being considered on its own merits, and the fertilizer treatments or the collection{ of be samplee’randomized or not as seems best under the circumstances. 5. Local Control A.great many types of field plan have been designed and used at various times in an attempt to obtain a more accurate control over the error and to permit more accurate interpretations. There are three basic plans for field plot experiments, the complete randomization which has no local control and consequently is of limited value, the randomized block which is the simplest type having a local control and which.permits the removal of error in one direction, and the latin square which permits the removal of error over two directions. The latin square is the most efficient design for small numbers of treat- ments. In addition to these three basic types, more complex.plot designs such as factorial, split-plot, and confounded designs all permit a more accurate elimination of error and permit a correspondingly more accurate interpretation of the data. The difficulty of the.more intricate plot designs is that the mathematical complexity of their interpretation is so great that inexperienced workers become confused in their final interpretations and may be led into errors by the complexity of the calculations designed to help them. Fisher (193“) has summarized the principles of plot design in the diagram reproduced in figure 16 which shows the interrelationship of methods for minimizing the effects of factors other than that being studied. 0. Applications 1. Introduction Field trials have been conducted for more than one hundred years. Every'worker in the field of plant nutrition has resorted on occasion to their use for the purposes of his investigations. The majority of the chemical methods of analyzing soils or plants to determine fertilizer 97. I REPLICATION I II RANDOM Loc AL. DISTRIBUTION CONTROL VALID 11')! OF DIM/NUTION Esnmm or [anon o r ERROR Figure 16. Diagram summarizing the principals from which statistical methods have evolved. (Fisher, 1931.) 98. requirements, and other biological methods have been develOped in order to avoid the necessity of field plot trials. The value of these addi- tional procedures has been subjected to criticism. Alway, Shaw, and Bethley (1926) state that “crop analysis does not appear promising as a practical method of detecting the phosphate hungry fields on prairie soils and is not to be compared in desirability with the use of small plot trials." On the other hand, Hardy, McDonald, and Rodriquez (1935) consider that "greatest reliance should be placed on the results of a statistical analysis of yield and chemical analytical data derived from a properly planned manurial experiment.” Field trials are being used currently either to correlate yields directly with fertilizers applied as Alway thought best, or to corre- late the concentrations of nutrients in the tissue with.yields from properly planned fertilizer experiments carried out on field plots, as prOposed by Hardy. 2. Correlation of crop yields with fertilizer applications Parker's work (1933) at the Virginia Truck Garden Station on tomatoes illustrates the procedure used and the inferences which may be drawn from the field plot trials of fertilizer applications. ‘Rarker conducted experiments from.l928 to 1932 in.order to find the correct fertilizer applications for the maintenance of Optimum yields 0f tomatoes under the local conditions. He applied eleven treatments on the variety Marglobe. The average yields for the five year period are shown in figure 17. It was concluded that under the local conditions 99. H t u I u '0 ' <1 tn 9 ’ H O +- e k ( Z +3 7 ‘ o 6 r / tn / °'_--" P SERIES 2 5 r F3 J koo—ood N SER'ES If i L I 1 L l N SERIES 040-6 2 IO-6 trio-6 610—6 —- P SERIES 4-0-6 lt-‘r-G Ip-e-b quay-6 413-6 K SERIES 4-10-0 If-IO-lr ‘r-IO$ 4-10’6 It'll-I2. Figure 17. The effect of increasing amounts of nitrogen, phosphorus and potassium. in a fertilizer mixture on yields of tomatoes in Westmoreland county, 1928-1932. (Parker, 1933) .vao of the experiments, the h—lO-6 fertilizer produced Optimum yields when it was used at the rate of 1000 pounds per acre. The weakness of the technique used by Parker is the same as in the majority of field trials, namely the use of treatments consisting of definite fertilizer ratios. It has already been noted that the absorption of each plant nutrient and its concentration in the plant is affected by a complex set of conditions. Surely, the arbitrarily selected combinations of nitrogen, phosphate, and potassium fertilizer units will be affected by these conditioning factors, and interpreta- tion of the results will be difficult. If one considers the possibili- ties of combining three factors at various levels, it is apparent that ghis procedure cannot determine the required amount of fertilizer for optimum yields. There can be only a suggestion of the amounts and combinations giving the best results under the conditions of the test. Unfortunately, although this weakness is apparent to most agricultural workers, little has been done to apply better techniques which would overcome the difficulty. It cannot be said that this is due to the lack of proper available procedures when the Law of Diminishing Re- turns as put forth by Bray (l9h8) and by Willcox.(1930, 1937) are re- called. Also the use of chemical analysis of plant tissue represents another more adequate approach. 3. Correlation of Tissue Analyses with Fertilizer Applications Chemical analyses of plant tissues has attained wide usage during 'tlle past few years, especially since simplified chemical procedures cbV-b. permit approximations of the compositions of large numbers of samples in a relatively short time. This technique was used by Carolus (1933) to study the same plots reported on by Parker (1933). In 1932, Carolus analyzed the plant material obtained from the plots for nitrogen, phosphorus, potassium, magnesium, calcium and total ash. The samples were taken at the beginning of fruit setting and the results were ex- pressed as milligrams per plant as suggested by Bartholomew, Watts, and Janssen (1933). These analytical and yield data are partially summarized in table 1h. These data show that there was a great rela- tive increase in both the fresh and dry weights when phosphate ferti- lizer was used. The effect was smaller in the case of nitrogen.and there was some depression in the yield when potassium was used., Carolus believed that more phosphate could have been utilized by the crap since values in the BB and CC lines (table 1h) were negative. Since the method of calculation yielded positive values in BB and CC, there is some evidence of luxury consumption of nitrogen and potassium. The final conclusion drawn from these data is that the Optimum fertilizer should contain at least 12 percent P20 , less than 6 percent nitrogen, and less than.12 percent 320. These results confirm in a general way those obtained by Parker but are of added value in that they indicate how the fertilizer ratio recommended by Parker could be improved. 9% + II .H+ I S - -I I -I I- I 8 amp? I- .m + -I om: - I I aaaaa I- I- 0 do... + I .2- I min - II 99- -I II I mm 83+ I- .am- I- 83- II HAN- -I I I- m mdm + I a - -I mi - -I- Tm .. -I 3 + II 5. a??? I R- I m8 - II TNT -I N? I 4 mmam noun .3: «3 3mm NS: :9 moi .83 $3 e23» N707: o-S-n «TS-s o-S-n mus-SJ 3:1: «707: o-S-n «73-: 9.2-: 33.2 tenants moauom ssfimmcpom find - I- 93+ I Tm - II II II II I 8 8m. - I- 9%... -I won - II II II II I o 3.3 - II 9.3 - I- R - II mom- II -I I- mm a 83- I- one - I- 33- II N.mm- II II -I m m emu + II we? I- New + I- new... II S: + I- .3. 53? II 33+ I- mmon... I 3: -I- 3.3+ I . 4 anon on? mmm 0.2 mm? news «43 was 59 New. oefiou. m-NT: m-o-n m-NH-n m-o-s whoa-n m-o-n m-NH-a m-o-e m-mH-e w-o-n 332 nonadenom nuance mahonamonm s4: - II .m - I- mag + II I- II- II II 8 82- II .mm- I- 8m + II II -II II II o 93 - II are - I- 12 + II 3d + ----- II II mm 33- II NN- I- a? + II is + ----- II II n m.~ + -I- i}: I- mxm + II 93+ -II are. + II <4 um: I- may, I- new? -I- met. ---I man... -I 4 3mm 2mm 8: n3 moan «Nam «:23 in? 83 new: 33» w-S-m w-oH-o w-S-m w-S o m-Sim w-oH-o m-S-w w-S-o w-oTw boon-o 83am tenants no when swan» .mh .uama .nama .uswa .uaw .03» M :5 sun .3. anon.— I Ifim-m-NN Imago-«n64 .NMHNWJI-mIem-Wm-I _ boon UHOdh 04.3 do 6.90 mung pfldHQ Moon an» do geochnwno unmana- oudsoa on» an - on» no UQDQOHUGH QOHSuNaE houdfiapheH 020 HO maneu60h¥:« OSu ho moObhkt van M0 GomthCEOO fi mpfiofihufifi QH behoo ho nowanhound .JVH. Danna—won _ 103. .mfimmp unwfimz hum m mo.paooHo mm was m you me 08mm on end 0 he nonneHeoHeo .mo.ma nflnseo ooahm.aam\m.oav onomonone ea mm .Heoneonoonn one» noes .news m.ma ma nofipaaompm om» onomoaoms .mama mmm ma masonamomm Mo unounoo adapo< .Akuomauanwmnummmv .mama m.HHw mp oxen casomm one» each on» as mahomamoma mnwpaompm pon deacon» on» no Hmuaopca .maw mama .mmnonamonm .ama ONH and pnmfioz macaw madam mum vodaoah poaa Ucpmouunp .mofiaom manonamoma a“ «menace menoaamoga oaaanxm mm was m mo moach50Hmo swamp an no pqcoaom m we commoumko 0 mo mofipmfibmn oo nofiumhompw acofipoaoomp om» Roam nofipmabon o noapaaompm pmooaoa c we commopako m mo mofipmfibon mm and museum Hmofivonoon» on» Beam nofiumabon m ooam condone-don on» no women owqdno a _44 . .M no .m an we doapaved on“ on mac ownmno .4 I] «consaonog .3 canoe 104, h. Discussion It has been stated previously that the field trial technique is the oldest, as well as the best known, biological method used for the determination of the fertilizer requirements of crOps. Field trials are not, however, without their limitations and disadvantages. Probably the most important of these is the heavy expenditure in labor and money required to conduct adequately such.a trial. Assuming that the design of an experiment is technically sound, the accuracy of execution of such details as cultivation, harvesting, units of measurement, deve10p- mental studies, etc., is essential for its success. Such accuracy can be guaranteed.only where skilled supervision and labor are available, and consequently field trials are too eXpensive to use on a scale suitable for the needs of individual farmers. Field trial experiments also are subject to substantial error if an improper plot design is used, and in some cases even.when the correct design is used authentic data may not be obtained. .Another disadvantage is that the results cannot be used during the year the field trial is conducted. Some 'workers have attempted to overcome this difficulty by making observa— tions at an early stage of crop development on such characters as leaf (color or rate of stem growth (Wallace l9h3). There is no doubt that field trials are the most direct way of éiiagnosing the fertilizer requirements of plants and it is equally 105. without doubt that they are the ultimate test to which the findings of any other method must be submitted. It is recognized by all workers that if predictions founded on the results of any other method fail to be confirmed by the results of field trials, such a method is unsatis- factory. While an extensive set of field trials is essential for the standardization of other techniques, these other techniques frequently will be found to be of greater practical value. 106. IV. LAW OF DIMINISHING RETURNS A. Introduction It was pointed out in the discussion on the use of field trials that many of the inherent difficulties of that approach to the deter- mination of the fertilizer requirements of crops could be overcome by the application of the Law of Diminishing Returns. An understanding of this law in principle as well as in application is fundamental to the study of the fertilizer requirements of crops. Von Liebig first emphasized the need of the various mineral nutri- ents for plant growth, and immediately thereafter attempts were made to formulate mathematically a law expressing the relationship between increments of plant nutrients and the yields resulting from them. The first of these attempts was the Law of the Minimum, already suggested by Sprengel (1837), but emphasized by Liebig (1863). This law recog- nized the importance of’a limiting factor governing plant growth. Uh; fortunately, the Law of the Minimum as prOposed by Liebig was poorly stated and numerous workers became involved in the effort to develop a sufficient body of data under controlled conditions to show that it was incorrect, and later to formulate a more correct statement. Mayer as early as 1869 realized that there were limitations to Liebig's statement of the Law of the Minimum, for he said that 'it is not possible to increase the production of plant substance at will, since a further application of labor and capital gives rise to a smaller and smaller increase in production." This probably was the 107. earliest disagreement with the straight lime relationship between nutrient supply and grmwth expressed by Liebig's law. Somewhat later, Heinrich (1882) noted that increased amounts of fertilizer not only failed to produce correspondingly great increases in yield but actually lessened the crop yield. Wagner (1883) at Darmstadt confirmed Heine rich's observations experimentally and reported his own.data to show that “the first 25 parts of nitrogen added to soil produced 127? grains of spring‘wheat: the second 25 parts, 563 grains: the third, 318 grains: the fourth, 39 grains.“ At the end of the nineteenth century, Whllny (1897-1898) laid the foundation for an interpretation of the Law of Diminishing Returns which was later proposed by Mitscherlich. 'bllny concluded from his investigations that “with addition of nutrients a rise occurs in the productive power of plants which is at first progressive and then becomes gradually smaller up to a certain limit, beyond which a further increase in the supply of nutrients causes a corresponding reduction in yield." This early work inter- ested Hitscherlich.who soon became interested in.the relationship of fertilizers to the yield of crepe, and he carried out a series of precise experiments from which he formulated his own statement of the Law of Diminishing Returns. 3. Deve10pment of the Law of Diminishing Returns 1. Basic Assumptions and the First Approximation Mitscherlich (1909) considered it necessary in order to obtain the most accurate expression for the relationship between increases in 108. yield and.amounts of fertilizer used to determine to what degree the crop yield was dependent on the vegetative factor operating at the minimum intensity. Mitscherlich's original prOposal (1909) of the Law of Diminishing Returns was based on.pot culture experiments with oats. Six kilograms of sand were fertilized with the following nutrient salts at time of planting: Salt Grams per pot 08(N03)2 3.76 NH .0 um; 2 9 CaC'I‘z 1.27 CaCO3 1.28 Mason 3. 66 In addition to these salts, each pot received a top dressing of 376 grams of Ca(N03)2 and 2.09 grams of NRhNO on June h and.August 10, or 3 approximately one and three months after planting. In this experiment, phosphorus was the limiting nutrient and was supplied as mono-, di-, and triphosphate, each at four levels. The averages of the data from four experiments, expressed as grams of dry matter in the straw and roots, are reproduced in table 15 and in figure 18. - The smooth curves found experimentally suggested to Mitscherlich that they could be expressed by a mathematical equation. He assumed 109. Table 15. Yields of oats fertilized with different amounts of phosphate in each of three forms, together with the probable errors of the yields. (Mitscherlich, 1909) Yields of dry material, expressed in grams Fertilizer expressed With With With as grams monobasic dibasic tribasic of P295 phosphate phosphate phosphate 0.10 ~ 144331.14 £6.11- : 0.1; 8.1; I 0.9 0.25 60.5 t 0.8 62.7 i 1.0 11+.o 1- 3.5 0.50 78.5 1: 3.6 78.0 5; 1.3 25.0 t 1.7 1.00 88.5 t 3.1- 91.L- 53.9 114.1 11.2 110. [00' X D 80" ' TRIBASIC O O 0 E. E 60L . . DlBASiC a x x Monoaasuc y. ( O ’10 ‘ / 9 3 2 20 >.. IO 5' [-0 GRAHS OF ngS Figure 18. Yield of dry matter of oats fertilized with different amounts of phosphate supplied as the monobasic , dibasic, and tribasic salts. The points indicated were obtained experimentally , but the curves were drawn according to the equation yaA(l-e"§x)'.(Mitscherlich, 1909) 111. that a plant or crop should produce a certain maximum yield if all. conditions were ideal and that when an essential factor is deficient, there is a corresponding reduction in the yield. He further assumed that the increased yield produced by one unit of the deficient factor is pr0portional to the decrement of the yield from its maximum. That is to say: as: “(A-‘5) where y is the yield obtained when x is the amount of the limiting factor present, A is the maximum yield obtainable if the factor were present in excess, and Cl is a constant. On integration, and assuming that i s 0 when x = o, it follows that /09re (A‘j)= /08reA-‘C,)(‘ I OY‘ /03(A~3)=/03A-cx wAere c=o.4343c, , or ‘1 :. A(I"/0—cz) Mitscherlich claimed that the proportionality factor was constant for each plant nutrient, independent of the crop, the soil, or other conditions. This means that if such were truly the case, the yields obtainable from any given quantity of fertilizer could be predicted from a single field trial. It would further be possible to estimate by pot experiment the amount of available plant food in soil. Mitscher- lich (1923) did use this formula for these purposes. In 1911. Mitscher— lich published further results obtained when phosphorus was the limiting 112. nutritional factor and confirmed the conclusions based on his earlier work reported in 1909. Mitscherlich considered that his statement of the Law of Diminish- ing Returns applied also to factors other than nutritional which affected cr0p yield. In 1912, he reported data, condensed in table 35, showing the quantitative relationship between yield and water supply. These data are significantly close to the theoretical values calculated from his equation. Spillman (1924) working independently arrived at the same basic conclusion as did Mitscherlich, and he expressed the Law of Diminishp ing Returns in the form x X Y=A-—-R, or Y=M—AR where Y is the actual yield obtained with x units of fertilizer, M is the theoretical maximum yield that can be attained with the ferti- lizer in question, and R is a constant. The fundamental difference in these statements was that Mitscherlich arrived at his formula by assuming that the slope of the curve of yield increase is proportional at all points to the amount of increase theoretically yet possible from the use of the fertilizer in question, while Spillman (l92h) arrived at the same expression by assuming that the successive incre- ments in yield due to successive increments in fertilizer applied constituted the terms of a decreasing geometric series. spillman (1921») has shown mathematically that there is no difference between 113. Table 16. Yield of tops of cats and peas grown in pots containing various percentages of soil mixed with sand, maintained at the water holding capacity. Experimental results and those calcup lated from Mitschérlich's formula are arranged in adjacent colums. (Mitscherlich, 1912) Percent soil water Yield of oats Yield of peas in the sand added mls. Obser. Cal. Obser. Cal. 20 #70 #3.? #3.4 22.h 22.8 30 690 61.1 60.2 32.h 31.7 #5 1035 81.0 80.7 33.3 42.5 70 1610 10a.0 105.1 55.3 55.3 100 2300 123.0 123.7 71.0 65.1 114. the expression l-c in Mitscherlich's equation and B in the equation he himself pr0poses. 2. Criticisms of Mitscherlich's Equation a. Major Criticisms MitscheriiCh's work stimulated a veritable flood of Controversy and it was subjected to many criticisms, both as to the way in which he applied it, and as to the way he mathematically expressed it. One severe criticism was that his maximum value, A, was determined by mathematical means and not experimentally. The question of whether the constant, C, for a particular nutrient is independent of other conditions of growth is still a matter of controversy. Finally, the logarthmic equation itself was criticized because it was confined to the expression of the effect of a single factor on crop yield, and also because it did not account for depressions in yield caused by large applications of fertilizers. Critics also pointed out that other, and simpler, equations could be used. b. Criticisms by Pfeiffer and his Co—workers Mitscherlich's theory, that the increase in yield with increasing applications of fertilizer gradually diminishes, was subjected to experimental test by Pfeiffer, Blanch and Flugel (1912). They investi- gated Ehe effect of increasing amounts of nitrogen on the growth of n” {M . oats in.pots of soil containing excess water, containing intermittent periods of adequate and inadequate water, and containing excess water 115. with the plants self-shaded. The results are shown in table 17, and it was maintained that the yield followed a parabolic curve expressed by the equation 2. §1° a-+l)>(-+ C)( The use of the equation of a parabolic curve accounted for depressions in yield resulting from the use of excessive amounts of fertilizers. The ”comparison of fit" of Pfeiffer's data to the two curves ‘38 a +bx+ cx" and /03[A-WJ) = /03 A “Cx is shown in figures 19 and 20. The parabolic function appears to show a closer fit to the experimental data. Mitscherlich (1912a) in reply to his critics, pointed out that the validity of any law of the minimum ceases‘when.no further increase in yield occurs, because the injurious effects occurring when excess- ive amounts of fertilizer are used bear no causal relation to limiting factors. He also pointed out that his formula, which contains constants of physiological significance, theoretically is better founded physio- logically than that of Pfeiffer which contains constants of only mathematical significance. Finally, Mitscherlich pointed out that all studies of crop growth from the time of Liebig had shown an ever diminishing rate of increase in yield.with successive increments of II N 5am 8w; 3 u 8a.: .xmm 3 a $932 .msm mmmé... untamm $53.9. u a .m $0.? £86m +Nuaomau u a .a II N .mam onw.a m“ N dogs .HmE ma h mmm.oxz .mam II N .er Ream 3 n noes 38: 3 a mama? .QO 30.3... 53.8 $33.? u a .n mmm.w.rnoma.mm+ RNA: as .N z .msm mafia 3 N some .3; 3 a was? .3». u , w ”memos .. 8%.: u a .. mocmoq .H .mnofiedsfleo 98.32 Jam m.mm a.aHH m.Ho m.mo o.aoa n.0HH m.mma o.mm n.ama omH.N m.mm H.HOH a.mo H.mo N.H0H m.mm m.omfl m.mm «.mma maa.a . m.mm o.mm m.oa n.0a m.mm o.nw H.HH~ m.aw m.uoa oma.~ 6 n new 0.8 we as 0.5 9Q. new as 3.8 m8; m.mo p.00 «.mm m.mm «.mo m.mm m.ma m.mo N.sa oaa.o m.a: o.aa 3.3m n.0m m.ea a.ma H.ae ~.ea o.ea _ mmm.o m.HH H.a . H.m o.m 0.0m m.:H a.m e.m m.m 00.0 m a m N H empaapoamo €538 eoaegoaeo ectompo empegoaeo ectompo $3338 3238 “some: 3 mo mamamv ovmnm mafia nope; mmooxm have: escapaapmuqH nous; onenodoz nonm3 mmooxm coeds.“ nouaaapeoh madam nfi dommoumwo .mmnpm dam madam mo meaowh vmbpmmno dad dopdddoaso ANHmH .Howsdm was .Moumam .pommHommv .masaom pqeowfidd ma downsnpm one aaofiamsvo schnammm and aoHHnvompfi: 0:» Beam dopdflsoaeo omen» was sped asapoq .uqofiafidnoo endomwoa escapee nouns use nouaaflpuom caosoMohaws mo madness muoahw>_npd3 Azmupu was namamv undo mo upoa anon mo meflogh ommaohq .mH capes 117 . View on- eum +trnaw A: 0.1-Mme.) A A A A A A A ' m a. been me out we re“ we alums N/m «mm N/PO‘I’ savanna»? wmn emu acne wann— me some» Figure 19.. Yield of oats plotted against grams of nitrogen fertilizer, showing fit of the Mitecherlich curve with the type curve y=a+bx+cx2 proposed by Pfeiffer. Plots with intermittent water and with excess water and the plants shaded. (Pfeiffer, Blanch, and Flugel, 1912) 118 . \ 8 x AcruAL VALUE. Y- arbx NHL" _____ Lug-3) - LooA - ex 8 Yleio as GRAIN +eTnAw A5 M(cns.) \ 0 A I L 4 A A A l I .A .A ‘65 ‘.I3 655 °1I0 0.065 use 385 m to calm: N/rov 6mm: N/POT excu- wM-n semen Mount: wmn segues Figure 20. Yield of oats plotted against grams of nitrogen fertilizer showing the fit of the Mitscherlich curve and with the type curve y: a+bx+cx2 proposed by Pfeiffer. Plots with excess water and moderate water. ( Pfeiffer, Blanck, and Flugel, 1912) 119. fertilizer. This had been shown by Mayer (1869), Wagner (1883), and Wollny (1897, 1898) as well as by his own work (1909, 1911). Mitscherlich‘s defense of his theory was investigated by Pfeiffer, Blanck, and Simmermacher (1915). These workers varied not only the rate of fertilizer application but also the amount of illumina- tion in order to test the combined effects of two factors on the growth of plants. Their results indicated that the Mitscherlich equation.was "most probably an accurate evaluation of the effect of growth factors on yield.” c. Criticisms by Briggs A critical study of Mitscherlich's equation.was carried out by Briggs in 1925. The basis of this study was stated in the form of two questions as follows: 'Is the relation between y and x of the form 3 = A044“) That is, is the value of c a constant when calculated from this equa- tion? Is the value of c the same for the same variable, irrespective of the plant, soil weather, etc?" I Mitscherlich maintained that these assumptions are valid because the values of y calculated on the basis of his equation, using a definite value for c, for each external factor agreed closely with the values found experimentally for different kinds of plants under various conditions. Briggs claimed that this type of data presented by Mitscherlich is not conclusive enough and to support this uses Mitscherlich's work (1912b) in.which he attempted to find the amount 120. of potassium available in a sample of soil. Using this data, shown in table 18, Briggs demonstrated that the equation .3.31 ‘3': 5I)(l“e a) more closely agrees with experimental data than the equation y = 50(’_e-3X) proposed by Mitscherlich. In a still more striking set of comparisons, Briggs showed that the closer fit to experimental data is given by the equation ‘3‘31A'1x4'b ’ ‘i’TFJ—c where b g 0.03 and c = 0.0622. While Briggs did not state that the Mitscherlich equation.was not an accurate indication of the relationship between yield and applied fertilizers, he did state that Mitscherlich‘s work (1923:) indicated that the values, A.= 50 and c = 3, were selected.without sound mathematical Justification. In this connection, he stated that "it seems that the guiding principle of assigning values to A.was that 0 should be 3 and b such that the amount of potash in sand and soil deduced from any pair of equations should agree'with that from any other pair. Not having the bias as to the immutability of c in the presence of limestone soil, but accepting the restriction as to b, we have applied other equations with the result that the potash in the sand and soil becomes 0.0088 gm. and 0.116 gmfkg respectively, as compared with Mitscherlich's figures of 0.01 and 0.08. It cannot be 121. .ma.o .om.o .mmo.o .mo.o Ammdam> 0v NNQOHo AmoeHE 8 N.oo 6.8 .83 .o.Nm 33% «K 8520 .moNa 63.0 52.3 885+“:me moH u AaummvmoH En H .36 .96 .96 AeoeHee 8 896.3an? moH u Annamvmon “so H .o.N .m.m Ammoétsném on .- QTHSmoH 3m H .m.N .m.m .m.m .m.m A839,. 3 “80.0.7598 moH n 3.333 34 H 5.3 6.1% 6.3 . 3oo.o+uvmuHm moH u AasHSmoH 3:. H .o.me .o.Hm .m.mH AmoeHee AavN eeeeHoo ANmo.o.+vaumH moH u Aaanvmoa Aeva H enhance one: sunfish coandOHmo on» noHna scum msoHpmswm m.om 0.0m 0.0m m.om o.Hm o.Hm m.Hm “.mH o.mH e.mH om.H mi: .33 was was man anon o.aN +2: de N.mH 8.0 Has H33 H43 eNe méN méN mKN ooH o.oH o.oH mN.o 0mm ~38 inn era m.HN N.HN N.NN ENH N.NH QNH H.o H.N.H H.oH oKH noH n.0H m.HH QOH Ho Nam fin o6 m N H N H N H vopdeoano sophomno dopsddoaeo mohhompo doomHsoamo doshmmpo H “0v pssHm oHoss any smupm Adv nHmnw HHom od..dqsm masHMoHHM o .4 .mmma ummHnm hp copmflsOHmoon use sNHmH .noHHaonomsz Bonn even .mnsaoa Mo mangoes mthAmh wsHbHoeon .HHon was case no monspHHa AH muse he vHon .mH oHnma .a 0.00 0.mm 0.mm m.mm m.mm 0.:0 0.am N.em m.0m 0.0m 0.Hm n.0e 0m.H 2 11 :.mm m.dm m.dm 3.3m 0.0m n.mm 0.0m n.3m m.md H.Nm 0.0m m.m: 00.0 a.Hm 0.0a m.Nm 0.mm m.Hm 0.Hm 0.Nm e.Hm 5.0: N.oe .m.0e 0.0m mN.0 e.ae 0.:0 H.me m.me 0.00 0.00 e.me N.ae 0.50 0.00 a.am m.am H.0 5.00 0.0m N.He 0.He a.em H.0m 0.mn 0.0m 5.0N N.eN m.eN n.0N 0.0 m N H m N H m N H doeeHeoHeo emeeoopo eoeeHeoHeo eoeeomeo 0oeeHeoHe0 eoeeoee0 on eeon oHoea some .ms m .HHou .ma m.m m on pamam oHom: Haoe .wx 0.N anon .ma 0.: n on aeeHm oHoea see. .ma m. .Haoe .ma 0.m 0 AnodafionOOV .md aflnma 123. claimed that the calculated values of y show a worse agreement with the observed than those calculated by Mitscherlich.’ Briggs' work brings forth only one fact, namely that it is extremely difficult to apply any single equation to the relationship between the magnitudes of crop yields and the amounts of fertilizers used. d. Other Criticisms of the '6' Factor Various workers, at different times, have reported data indicating that the e value as applied by Mitscherlich is not constant in magni- tude. Ma-gistad (1938) conducted a critical comparison of the e value by growing oats in seven different soils. These tests were conducted simultaneously in Germany and also in the Hawaiian Islands. he re- sults show that, 1: the c values are taken as 0.122 dz/hectar (153 lbs/acre) for nitrogen, and as 0.6 dz/hectsr (753 lbs/acre) for r205, and as 0.93 dz/hectar (1170 lbs/acre) for :20, then the results obtained on the same soil and using the same crap are different in the two test areas and consequently the value of e is neither constant nor independent of soil, weather, crop etc. Magistad, rat-den, and Lambert (1932) earlier had shown that the value of c prOposed by Mitscherlich was not constant under field conditions in Hawaii. his work seems to indicate that there are restrictions in the extent of the area over which the value of c is constant. 121+. ‘ Oapo’ (1938) using Hegori Sorghum, fertilized with various amounts of nitrogen, phosphate, and potassium, also found that the c values as proposed by Mitscherlich were not constant, and he proposed values of 0.805 for K20, 0.5538 for P205, and 0.5968 for m3. Previously, Mitscherlich had proposed the values of 10.93 for K20, 10.6 for P20 , and 10.10 for ma. Experimean data and the calculated results are given in table 19 together with the equations used. While this data dhows close agreement between the calculated and observed results and an apparent lack of constancy of value of c for different crops and conditions, sufficient evidence is not available to permit defin- its conclusions as to the constancy of Mitscherlich's value c. While the proceeding discussion indicates that there was dis- ' agreement as to the accuracy of the equation proposed by Mitscherlich V (1909), particularly in regard to the constancy of e value. one must conclude, until it is proven otherwise, that the first approximation of the Law of Diminishing Returns is an accurate indications of the effect of a single factor on crap yields. 3. Application to Multiple Factors a. Work of am. ' Mitscherlich's first approximation of the Law of Diminishing Returns ‘1} .-. A ((—- (fear) 125. Table 19. Actual and calculated yields of Hegari Sorghum, expressed in pounds of green material per plot, when fertilized with various amounts of 1133, p205. and :20. (Cepé. 1938) Pounds of * Crap yields in pounds of green material ferziiézor 150 fertilizer 9205 fertilizer Kéo fertilizer (2) Actual Calculated Actual Calculated Actual Calculated yield yield yield yield yield yield 1 2 3 _ 0 111.1 112.2 80.h 80.6 79.1 78.3 0.75 117.0 113.u 98.7 97.9 99.1 97.0 1.50 110.3 11u.h 108.0 109.1 113.7 109.0 2.25 116.7 115.2 116.7 116.2 116.7 117.9 lquations: 1. y . 120 (1 - 0.23051s31’Z'61+ ‘) 2. y . 129 (1 - 0.55385”6+ ‘) 3.‘ y = 136.6 (1 - 0.59681°65+‘) 126. or as it is more commonly stated /03(A‘y) = /05 A " cm cannot be used for stating the effects of two factors except by treat- ing each graduation of one factor as‘ a constant, the other factor being treated as the variable, and then reversing the procedure. In order to express the dependence of crap yield on the many variable factors aper- ating under the actual conditions of crop production, a transformation and extension of this basic formula is necessary. Baule (1918) recognized limitation of the Law of Diminishing Returns to only single factors affecting yield as it was proposed by Mitscher- lich (1909). Baule consequently modified the expression while retain- ing its fundamental assumptions. He supposed that each of the factors influencing plant growth acted in accordance with Mitscherlich's assumption, and that the final yield was given by the expression, "1 = A («-e"°"‘)(l- 5m“) ........ (Fe—‘0'“) He states that the percentage increase in the crop yield was a function of any growth factor. For instance, if one unit of a given fertilizer increased the yield 50 percent of the amount by which the attainable maximum exceeds the yield without fertilizer, then a second unit of fertilizer will increase the yield 50 percent of the renaming distance to the maximum, that is to say, two units of fertilizer will relieve alt of the deficiency. A third unit raises the yield to 7/8 01‘ its untimum, etc. Ihe efficiency of the nutrient, or other factor, 127. called by Baule the 'Wirkungsmenge' and now known as the 'Baule unit”, is expressed by the relation 1. ‘5 [og»e a c»? where c is Mitscherlich's constant and 0.? divided by c is the 'Baule unit“. The Baule unit is defined as the amount- of a factor needed to give half of maximum yield obtainable by increasing that factor while all others remain constant. Baule'shows that there is a definite mathe- matical relationship between Mitscherlich's c value and the hule unit. If the fertilizer increment added, (2), is such that jj== 21/4 then - h /—e“ #1 2. _—fl——-/° e _‘ a. ‘27 C and h is equal to mule was able to deduce mathematically that if two factors vary simultaneously, each produces its own effect on yield independently of the other. When cats were grown in sand and both the water and phos- phate supply were varied, the relative effects of the water remained the suns no matter how much phosphate was given. Ihe original data are presented in table 20. file theoretical curves expressing the yields are sigmoid if the Baule units of the two variable factors are approximately of the same order of magnitude, but if one is significantly greater, the curve 128. fable 20. Yield of'oats in pot experiments fertilized.with different amounts of phosphate and receiving different amounts of water. (Mitscherlich, 1918) units of Yield of oats 0‘3 (P0192 Water Water Ratio 1 unit 2 units 13/; a b O 6.1} 11.0 1.72 1 11".6 25.6 1.75 2 22-5 36.6 1.62 1" 29o? 53.1 I.” 8 “1.3 70.6 1.71 16 5°~3 77.5 1.53 32 55s? . 88.5 1.59 129. rulains logarithmic as is seen in figure 21. Mitscherlich, working with cats (1918), sugar cane (1919), and mustard (1919a), confirmed the validity of Baule's procedure for cal- culating the effects of multiple factors of crop yields. Spillman (192a) carried out several experiments in which more than one plant nutrient was varied, and found no indication of a sigmoid yield curve. .He noted, however, that on the portion of the sigmoid curve beyond the point of inflection an excellent fit with theoretical calculations was obtained. This showed that for the greater applica- tions of fertilizer, even in cases which involve the sigmoid curve, the simple formula is applicable. Spillman suggested that this may be due to the fact that hule's interpretation may not apply to cases in which the two variables are both nutritional in nature, but will apply when one factor is a nutrient and the other is some such factor as soil moisture. That this interpretation is incorrect has been shown clearly by Bray ‘(19h8) working with corn, ‘ fertilized with various amounts of phosphate and potassium. In conclusion, we may say. that the Mitscherlich equation in its simplest form, presupposes the presence in excess of all nutrients emcept the one limiting growth. If two nutrients are simultaneously increased in constant proportions, the combined effect’will be the product of their individual effects, as proposed by Baule (1918), and the curve relating the magnitude of the yields to amounts of fertilizer becomes signoid. The specific form of the curve varies according to the relative proportions of the factors acting on the yield. 130. I _ 7 -. _ no.0 T on: 8P5 Tf-o >22 Figure 21. Baule's curves showing the yields (y) as percentages of the maximum possible yield when two factors 3: and z vary. When x-ez the curve becomes sigmoid. (Baule, 1918) 1.)..L. b. Work of Balmulmnd The effects on crop yields of the simultaneous variation in the supply of more than one nutrient have been represented mathematically by Balmulcand (1928) in a manner somewhat different than did Baule. Balmukand considered that Mitscherlich's equation expressing yield- factor relationships, while satisfactorily expressing the response of a crop to variations of a single factor, does not adequately repre- sent the effects of simultaneous variations of two or more factors. He points out that in Hitscherlich's equation, y = i (l-e"°’), that if x is changed to x', that is to say x amount of fertilizer is increased 1' amount, then y will be changed to y' independent of the value A. , Balmukand proposed the use of the ”resistance formula“ which requires that, instead of the ratio y'/y being independent of the influence of other factors when the value of one factor changes in intensity, the expression 1 - 1', or the difference of the reciproca lse of the yields. should be iidepindent of other factors. The validity of this assumption is supported by his data, presented in table 21, taken from field plots and pot-cultures with wheat. They show that, while the ratio y'ly in the case of the lpot culture work varies from “.9 to 214.0, the reciprocal difference 1 - l varies only from 0.26 to 0.29. y y' Wad therefore preposed the use of a formula of the resis- tance type -§-=F(N)+F'(K)+F"(P) ...... c where , “NF-fife . F (W 21‘“, etc., «one... sense 36m 86 mos enema $0.8 3.8 on.a and nacho mesa 63%. 9a m: Sam... Hmmfi $32 wnfi ma momma we? 8.? sea m h .5 arm uh. comes»? H t H no mashed can mama ma eocouommdc on» ma dosacppo non non ea sea Hmooaafioom munouh mo canon cowoapun masawfiaaa: owm no . non non macaw a“ mdaofih mono madhmaannz sped cascade com a a; 1* . . SQR .a com 08 «an 2: 2: 08 means 0 was. m 83: a a... a... «on 8a 2: _ 8N . . 392 .a com com . i... . II II ..I. once hon .mpH ease non .mpfi once you .mpa .AOIomlov once non .mpH shoe non .epH ease sea .epH .m m .m one. see norm: .nomunsma enemasoneheas cushions: Seneca semen H I H .h nHoAmsm no acaoaq no anaoa< mo anpoa< Mo «speed we essoafl No onuoafl O 33» seer nannies seen uHo«h_MHeh6eoam. till though own» He announce? MflH9d0dUfld lOHfivflfio “OR Ebb“ find UHOwMJMMGHWdW” .«mmmn .pacmsammm» h\.a oxen one some 0"")! "[||» I \H I 5}” mafia) 1 do has one 3. OHM“ 508.“ page! H) “Riemann” DWI DH l2). in.which N, K, and P are the nutrients added to the soil, n, k, and p are the nutrients originally present in the soil, and an, ak, and ap are constants for the type of fertilizer used. Balmukand reported the actual and calculated values obtained from an experiment with potatoes in support of the accuracy of his mathemati- cal theory. His data given in table 22 show the yields of potatoes expected from theory and those actually obtained on plots receiving Jfour different amounts of nitrogen, and four different amounts of potash, in all combinations. He concluded that the resistance formula fits the observed data closer than does the Mitscherlich formula. and that the parameters appropriate to each nutrient, being independent of the abundance of other nutrients, are capable of direct physical inter- pretation. While this approach by Balmukand is rather unique,‘we must conclude that it does not detract from the value of Mitscherlich‘s equation. The mathematical statement of Balmukand is sound, and probably states the effects of multiple factors on crop yields as well as does Baule‘s equation (1918), but the complexity of the mathematics used by Balmukand renders his resistance formula imprac- tical for routine use. c. Work of Gregory The'work of Gregory (1937) indicated that under certain circum- stances the relationship of crop yields to the simultaneous increases in two plant nutrients, added in constant proportions, is represented 33.0 .1... $86 a no 886 M $86 a as 36 H no; u e 8.0 a 36 u a 8.: 3.2 3.3 No.3 no.2 3.2 «are «fie son $.S 3.3 ”man 8.3 mug: 3.3 mo.m on.a com on.a Sm 1mm Sm mum mad man E2. 8” mod 2k. max 86 8.x. _ 8.... e: 8;. c possesses". esters confiscate science oeesasoaeo uptempo messages entrees moses so seeps admwu nommu admfiu camAAHpu HO who.“ . MO 0H0.“ NO on: gfiflogfl mom .mpa co: non .mpH com hem .mpH 00H vouaaapuowfip no nausea shoe sea was» a“ mooaeuoa no caeun Amman .uneasaammv .mmma .eaoam numbness» ca macaw .eaos you one» no consensus mooaeaod mafia s.uueu no cavemen s.ucmxpaden no semen space» dooedsoaco use Happo< .NN sepia figures...» h’zj‘o \f in a. . Jun. .3. by a straight line. These results were obtained by adding nitrogen, potassium and phosphorus fertilizers to barley. When the yields were plotted against the amounts of the nu rients taken up by the plant, three types of relationships were established: 1) with an excess of all nutrients excepting the one being added in increasing amounts, the relationship of yield to the uptake of the nutrient is given by a curve of the typical Mitscherlich type: 2) when the yields are plotted against the uptake of one of the nutrients available in adequate amount, a curve showing an ever increasing yield is obtained: 3') when the supply of the two nutrients were increased in a certain fixed proportion, the curve relating the yield to the uptake of either nutrient becomes linear. These relationships are shown in figure 22. It was further shown in sand culture that the uptake by the plant of the nutrient in minimum was proportional to the amount of nutrient applied to the sand. These data appear in figure 23. Consequently, a rough approxi- mation of the yield can be derived from the amounts of a nutrient absorbed by the crap. ’4. Second Approximation Experimental work in the field frequently shows that increasing incranents of a fertilizer applied to the soil give regular and profit- able increases in crop yield for the first few increments, but with further increments, the yield not only ceases to increase but may be depressede 136. }. c Q. s ' us ‘ 5 to E c J a ; so .1 E l- o l s A a .1 .1 .a 1 . o .I .3 .3 .* .3 . 'I '3 '3 "f 5 ‘ 7 'U ’ 0mg Novasosu Umad‘us) TOTAL “.0 Una“. ants.) Figure 22. Relationships between total crop yields and amounts of nitrogenrand potassium absorbed from.the soil. One nutrient in.mdnimmmimmsns that the others are present in excess. (Gregory, 1937) w' e‘_ 137. mcswmo N (mu, -N.n.) A mcnsmmo NwK(N5K5 - um.) x mcsiumo 9 (mp, -~.p.) o RAT: or UPTAKE m M0-/Pd'r m 2. Watts !_1 . . . ,- 761551 V9 V3 0 Figure 23. Relationships between the rate of uptake of a nutrient and its concentration in the soil. Experiments carried out with barley and harvested six weeks after germination. (Gregory, 1937) 138. Mitscherlich's (1909) first approximation of the Law of Diminish- ing Returns was based on experiments under conditions which pertained only to the first part of the yield curve. As long as these condi- tions existed, the law apparently eacpressed a true relationship between the amounts of fertilizer applied and the crop yields obtained. Other experimenters, notably Pfeiffer, Blanck and Flungel (1912), and Briggs (1925), operating under conditions which produced a depression of the yield, where c was not constant, and the logarithmic equation did not apply, assumed that Mitscherlich's concept was inherently erroneous. Hitscherlich (1928) met the criticisms with an extensive new series of experiments in which the three principal plant nutrients, nitrogen, phosphate, and potassium; were employed in different com- binations, and them proposed his second approximation of the law. In the development of the second approximation, Mitscherlich (1928) noted that the yield curves from all the experiments in which depressions of yield occurred because of an unbalanced nutrition of the plants, exhibited two portions: an initial portion along which the increments of yield exhibited the normal course in accordance with the normal yield equation, and a subsequent portion which followed a different mathematical law. the initial portion, of course, follows the curve given by -C‘x —-C.'X 3 new )0, 3 A(I— e ) which may be expressed as d 7?}: C(A-lj) If both sides of this expression are divided by the factor y, then 4 _ . H-‘j c410”? “L 3%“: ‘72: Iw/o' 7‘ ‘9 where A-‘y: A(l0-cx)l and my ‘3: A(l“'/0 ) But y is preportional to the yield depression factor -2k11, and by adding this factor to the equation the second approximation of the Law of Diminishing Returns is obtained as follows: .L J I ’c" v 2%: (c'fifxflm 3% “(“5 {Km £10-")d" ‘IQ k'x) 17‘ .: /o 9((‘10- CX)‘(- ___I 0 - 0 1.5 O-‘l :1" '0 - Guns N 9n POT GRAMS P20, rm ror Figure 26. Continued. B. Soils 85 and LA. (Magistad, 1938) 15 L #4 u"‘""" """ - GERHINY ._ _..__ -J‘AWQH A l ‘ A_—‘ 1} .I .1, a. It: Gums K.O nu ror 156. Better results were obtained by recalculating the c values. Even as late as 1943 there was discussion of the possible value of these pot tests. Olsen and Shaw (1943) reported that the Mitscher— lich pot tests gave a somewhat better differentiation of the available potassium than did the chemical tests of soils that were known to res- pond to added potassium. The complex and expensive equipment required to conduct the Mitscherlich pot tests preperly has discouraged its use and has en— couraged the development of a number of simpler and less expensive techniques, even though they may not be as accurate. 2. Stephenson and Schuster Technique The technique of conducting the pot test proposed by Stephenson and Schuster (1941) is a modification of the Mitscherlich method so that the availability of crap nutrients other than nitrogen, phosphate and potassium might be determined. The method is based on the vigor- ous extraction of nutrients by rapidly growing sunflower-plants on.a small quantity of soil. The plants are grown in the greenhouse in N0. 1 tall, lacquered fruit cans containing 400 grams of air dried soil. The quantity of soil required depends upon the number of plant nutrients being tested. The bulk sample should be composed of a sufficient number of subsamples to represent the field as a whole. When sieved, each can should receive a maximum of’400 grams of the thoroughly*mixed, air dried soil. Nutrients are supplied from molar 157. stock solutions of HHéPQh, HéHPCn, K01, CaHé(POh)2, Ca(H03)2, Hgsou. 7320, and CaSOh, and from a 0.00567 molar solution of 3203.320. For the complete fertilizer treatment, all nutrients. properly diluted, are added and for other treatments, all nutrients excepting the one being tested are added. The nutrient ions are supplied at the follow» ing rates expressed as parts per million: N'; 200, P = 217, HI: 350. Hg : 168, S = 224, and IB 3 2.5. Lime, when used, is applied at the rate of 0.8 gram of calcium carbonate per can. The sunflower seeds are planted, and after about nine days, the seedlings are thinned to five per can. The soils are kept at an optimum moisture content by adding distilled water at frequent inter- vals during the growth period of 6 to 10 weeks. .At the end of the growing period, the plants are cut off at the soil level and dried at 700 C. The average dry weight, obtained by acw'treatment is compared.with that of the plants receiving the com- plete fertilizer treatment. Ihile Stephenson and Schuster (1941) have used their technique with.good results on more than 20,000 soil samples, it would appear that it lacks the quantitative mathematical interpretation possible by'the Mitscherlich method for there is no direct way of using the data to determine the actual fertilizer requirements of a crop. 3. Jenny Technique vandecaveye (1948) described a pot culture technique of Jenny using lettuce as the indicator crap. This crap is grown in pots in the greenhouse in the‘winter or outside in the summer. Lettuce is grown in sixfinch flower pots coated inside and out with two layers of asphaltum paint and*with an additional layer of aluminum paint on the outside. Drainage water was collected in saucers treated in the same manner. Each.pot requires approximately 4 pounds of soil. and depending on the number of nutrients being tested, and the number of replicates used, the total amount of soil required is between 100 and 200 pounds. The sample being tested should be composited from a number of samples representing the upper 8 inches of the profile. Jenny'bases his fertilizer treatments on the following units, indicated as subscripts to the nutrient in question. Nutrient Milligrams Pounds per pot per acre Pi P205 ; 80 P205 : 100 x1 x20 :80 1:20:100 He suggests that the most useful combinations of units are as follows: NBPBIS léPbHé HéPblé HéPéKs NbPéIé Before the lettuce seedlings are transplanted, apprOpriate amounts of stock solutions are added to the pots and thoroughly mixed with the soil. Because in combination of unitsIHjPBHé’has been shown over a period of years to give the highest yield and best growth of lettuce on all soils, Jenny considered that this rate of fertilizer application be considered optimum. One lettuce plant is transplanted from a flat to the moist soil when about 4 weeks old. one plant to each pot. Distilled water is added regularly and any drainage water accumulating in the saucer is returned to the pot. After a six weeks growth period in the soil, the plants are harvested, dried at 700 0.. and weighed. The yields are recorded as percentages of the theoretical maximum. Jenny used these percentage values because they are less influenced by the environment than are the actual yields. Table 26 represents a tentative scale of nutrient deficiencies based on the percentages obtained. This technique requires much less time and equipment than does Hitscherlich's procedure. 4. value of Pot Cultures Pot cultures have been used in every phase of nutritional research and they still represent a very important technique. All such methods when used for the diagnosis of fertilizer requirements exhibit important weaknesses. The volume of soil accessible to the roots of the plants in the pot is much less than in the field, the mechanical disturbance of the soil in the preparation of the cultures changes the physical composition of the sample. Both of these factors affect the avail- ability of the nutrients to the plant. Consequently, results obtained with.pot cultures can be compared to those obtained under field condi- tions only with reservations. It has been reported by Stephenson and Table 26. lettuce in.pot culture by the method of Jenny. Tentative scale for estimating nutrient deficiencies by growing (see vandecaveye, 1948) Perti1izer added Yields as percentages of maximum Nutrient Bbr standards For being Definite Probable Uncertain (100 percent comparison tested deficiency deficiency dlficiency yield) NP! PE FPKL PKL 11 Less than 20 20 - 50 51 - 70 NPKLS PKLS NPKLS MI PKLS M3 NP! NK NPKL HXL ' P Less than 20 20 - 50 51 - 65 NPKLS NKLS EPILS HE NKLS HE NP! NP NPIL NFL K Less than 70 70- 75 76 - 80 NPKLS NPLS ms 1m ms 1m , HPKS NP! NPKLS NPKL ‘ 3 Less than 66 66 - 76 77 - 83 NPKLS H3 NPKL M3 IPII NP! HPHLS NPKS Lime Less than 55 55 x 73 74 - 80 EPILS "I NPKS ME 9 00 161. Schuster (1940) that in some cases the responses of plants observed in 'pots may be the reverse of those observed in the field. Another questionable'aspect of most pot culture techniques is the use of specific indicator plants. It was pointed out earlier in the present discussion that different species differed in their ability to absorb nutrients from the soil, and these differences certainly would limit the usability of data obtained from an arbitrarily selected species. Another limitation to the use of pot culture techniques is the expense involved in both equipment and labor. While the experiments can be conducted with less expense than field trials, they are more costly than either chemical analysis of soil or of plant tissues. The most accurate pot culture technique is that of Mitscherlich. Whereas other workers, such as Jenny (see vandecaveye, 1948), and Stephenson and Schuster (1940), have attempted to simplify Mitscher- lich's rather expensive and complicated procedure, their modifications have lessened the accuracy of the method. While it is agreed that pot- culture techniques may yield valuable information as to the relative amounts of available plant nutrients in a soil, it should be recalled that they cannot be used for a quantitative approach to the fertilizer requirements of a crop. Mitscherlich's procedure, however, appears as an exception to this generality. 3. Field Techniques 1. Introduction The Law of Diminishing Returns has been used most frequently for the interpretation of crop yields obtained under field conditions. that of these applications were made during the first quarter of the present century in attempts to either prove or disprove the validity of applying the mathematical expression of the Law to the study of single factor relationships. iithin the last ten to fifteen years, however, some interesting work has been conducted with the Law of Diminishing Returns as the basis of interpretation by such workers as Magistad, larden, and Lambert (1932), J'arden and Magistad (1932), Capo (1938), Villcox (1941), Olsen and Shaw (1943), and Bray (1948). The work of Uillcex (19M) and Hagistad, rarden and Lambert (1932) was a direct applica- tion of the principle as laid down by Hitscherlich and Spillman, while other work, such as that of Bray (1948), is based on modifications of the concept. The discussion below is based primarily on the field techniques of these last three workers because it seems probable that their work will furnish the answer to the problem of the quantitative determination of fertilizer requirements. 2. Procedure of Magistad and Oo-workers a. Introduction Since Magistad and his coworkers used the type of analysis put forth by Spillman (1922;) in preference to that of nitecherlich (1909), 163. it is necessary to emmine the details of the Spillman technique as it was mentioned only briefly in the general discussion on the Law of Diminishing Returns. Spillman's equation expressing the Law of Diminishing Returns is Y = M - m’ where Y is the yield of the crOp in tons per acre, 14 is the theoreti- cally maximum crap yield, A is the yield increase due to the applica- tions of large amounts of fertilizer, R is the ratio between successive increases in yield resulting from equal additions of fertilizer, and x is the number of units of fertilizer applied. Spillman (1921+) showed that this equation was similar mathematically to that originally pro- posed by Mitscherlich (1909) and Baule (1918). As with the Mitscher- lich equation, the Spillman equation does not apply when the nutrients are high enough to cause a depression in crop yield. An experiment- conducted on pineapples by Magistad, Farden and Lambert (1932) will be taken as an example of this procedure. the fertilizer was worked thoroughly into the upper three inches of soil. In December, 1927, 60 percent of the nitrogen. 57 Percent of the P20 , and all but fifty pounds of the X20 were applied while the rest of the 320 was applied in May and November, 1928. The ferti- liser treatments and the crop yield data are given in table 27. and their mathematical analysis will be briefly described. i'he major differences between the equation Log (1 - y) : Log 1. - bc mm.wa me.s flamed «men eon onH m.amm n 2.? 3.: mmmfl RR eon ems 3.8 o. c~.ca NH.e mmoma amen com omH m.amm m Hm.ma mm.n amass ween «NH omH m.am~ q . . e m e _ cued sea A epHv A spay ceases we 0 u. o m a pass» upmcam Mo .adshm mo assess choc mom common peak “mama .eeepsdn sec deepen .eceeamezv .oeu we neurone mummsoauqd use .momm no end unwoupws mo nuances shamans no dopmmmmoo munmspmoaa .ommn a“ vopmo>amm macho oHaaeodua on» you some uHlo use nouafiapaoh .nw capes and that proposed by Spillman, Y : M - ARx, are the constants. Mitscherlich obtains his A constant directly from the yield of a plot receiving fertilizers for maximum yield and maintained that his e value was constant. Spillman, on the other hand, has three distinct constants M, A, and R, each of which must be determined for the par- ticular experiment. There are two methods by which these constants can be calculated, both of which will now be explained. b. Calculation of Constants by the Method of Least Squares ll'he most probable values of M, A, and R are those which reduce to a minimum value the sum of the squares of the differences between the observed and the computed yields, or Y. That is 2 2 (I . (M - Ann) + (I = (M - A1112) ...........e minimum where x1e x29~zje ......... o. xfi,= unitSOf fertilizer V - corresponding yields ‘1’ Yé, onoooaeoooee. Yfi _ 2 The values of M, A, and n that reduce the equation, I g (M - A1111) to a minimum are obtained by differentiating the equation first with 1‘esjpect to M, then with respect to A, and finally with respect to B. when these differentiations are set to equal zero and then solved for M and A we have: M= the AER”) 166. A : «(TR:—éYR‘ ' (éR) -«€R“‘ A : mgéYxRx-éYfl‘xRx) * Warm—meta” Having obtained these equations, the next step is to try a series of values of R and solve for A1 and A2. When these values of A1 andA2 are plotted against the R values and the curves drawn, the intersec- tion of the curves will be the actual value of R and A. c. Calculation of the Constants by the Simplified Procedure A simpler procedure that may be followed to obtain the values of 14,.A, and 3.19 to find the greatest common divisor of the amount of ‘fertilizer applied per acre for each treatment and indicate this as "units" of fertilizer as has been done as shown below: Lbs. x20 used 0 in units of Yield in tea; per acre 2 lbs. used per of fruit'per acre acre (1) (Y) 125 5 16.21 200 8 17.27 300 12 18.17 400 16 18.58 Then by assuming a value for R and solving the equations for A1 and 167. A2, Having obtained these values, the values of A1 and A2 are obtained it is found that for this data A1 = 6.4599 and A2 = 6.491”. for values of R above and below the original estimated R value. The results will be of the nature of that shown below. R A1 A2 Difference .8ti 6.682 6.757 -.o752 . 85 6.1699 6.1%? -.o348 .86 ' 6.2813 6.21:3? .0376 From these figures, it is apparent that the true value of R lies between 0.85 and 0.86. To find the actual value of R, the A1 and A2 values are plotted against R as has been done in Figure 27. The intersectien of the two curves gives the true value of R, in this case 0.85h, and the true value of A, in this case 6.38. Then from the equation M = %(< H6867 it is possible to find M which in this case is 19.0 tons per acre. Using the three constants, the yield curve is given by the equation I : 19.1 - 6.38(0.85hx) where x is the number of units of 25 pounds each of K20 used per acre, or Y s 19.1 - 6.38(o.538x). where x is the number of units of 100 pounds of Kéo per acre. The 168. curve shown in figure 28 can now be used to indicate the units of x required to give Optimum yield. d. Determination of the Available Nutrient in the Soil As with Mitscherlich's equation, the Spillman equation can be used after the constants have been calculated to determine the amount of a plant nutrient in the soil, that is, we can obtain a biological soil analysis. An examination of figure 28 shows that when no potash is applied, a yield of 12.72 tone is the equivalent of Y, and if the curve is projected to the abscissa we theoretically have no yield at all. This occurs when 1:: 1.78 which is equivalent to Mitscherlich's b value. Thus in this case, when Y': 0, then x:: 178 pounds per acre. or the theoretical amount of available K20 in the soil. 9. Determination of the Profitable Amount of Fertilizer Farden and Magistad (1932) used the equation described above to determine the economic limit of fertilization. They considered both the cost of the fertilizers and the economic value of the craps they produced. They reasoned that if one ton- of pineapples sold for V dollars and if x units of K produced Y tona ofjpineapples per acre, then the total value per acre would be VY dollars. If one unit of K costs I dollars and the N'and P used cost L dollars, then the total cost of the fertilizer would be (Misc) dollars. The profit from the crOp would be 169 . Figure 27. Graphic method for the determination of the true values of A and B in the equation Y==I~I-ARx. (Magistad, Farden, and Lambert, 1932) 170 . 30 ‘ Yo'cuo '0» Tons V V -2 .’ o I z 3 , Umva of K,O mt acne Figure 28. Fit of the observed values of crop yield in tons per acre to the yields calculated from the equation, Y=19.1-6.38(o.538)x. (Magistad, Farden, and Lambert, 1932) 171. P : VY - (1.1-Rx) but Y :- (M - Alix), hence p = vm - 1m") - (In-Ex) If one ton of pineapples sold for 25 dollars, and if 287.5 pounds of nitrOgen and 150 pounds of P205 cost #5 dollars, one unit of %0 cost 6.30 dollars, then P = 25 (19.1 - 6.38 (.5395)? -- (45+6.3x) If various values of x are substituted, the curve reproduced in figure 29 is obtained. The most economical rate of fertilizer applica- tion is represented by the peak of the curve. This point may also be found by the following mathematical manipulation: in; ....23a3VA loTR'R‘- K d1 “/8411 é£=o R‘: K -4303 VA A] R X : [,;R(/OJK -—/DJ(—.1303VA /a? R) The value of x is found to equal u.u5 units of K20 and from the curve in figure 29 the maximum profit is found to be 39h.76 dollars. 1'. Discussion There is a serious disadvantage to the Magistad technique which was frequently found in the earlier attempts to apply the Law of Dimin- ishing Returns. His technique is based on the actual crop yields per 172. too in 3 x «c a 2 1 ¢ 3 = s. a; 10 :E :310 iii: l .. ti 2 .. 2, 8 1:5: son :: loo .AE . . 6 1! UNIT. no MACK Figure 29. Graphical method for determining the rate of fertilizer application which wi11 give maximmn financial profit. (Farden and Magistad, 1932) 173. acre and consequently the result is subject to serious meterological effects. If this source of error was corrected using the percentage of the maximum yield rather than the absolute yield, the disadvantage of the technique would be greatly diminished. However, the procedure possesses a valuable feature not often considered because it takes into consideration the economics of the fertilizer problem. It per» I mits the calculation not only of the amount of fertilizer that will give the greatest yield but also of the amount which produces the greatest financial profit. i As the complexity of the mathematics involved in the Magistad technique does not allow easy interpretations of field trials, the practical value of the technique for general fertilizer investigations is small. Consequently, other techniques for applying the Law of Diminishing Returns are more useful. 3. Procedure of Roger Bray a. Introduction The bases of Roger Bray's technique of applying the Law of Diminp ishing Returns is his own (19h8) mobility or elasticity concept of plant nutrients, and the percentage yield concept of Baule (1918). Bray's mobility concept assumes that "the available soil nutrients have a variable availability which depends upon the mobility of the nutrients in the soil and on the nature of the plant. Those with little mobility like potassium, phosphorus, calcium and magnesium tend 174. to follow the Baule (1918) percentage yield concept. The mobile nutrients, like nitrate nitrogen and water, tend to follow the Liebig idea of a limiting nutrient.” The Baule percentage yield concept stated that the final yield is the product of all the factors involved in yield. If each nutrient level is expressedin terms of its ability to produce a certain percentage of the maximum yield attainable with a sufficiency of the nutrient the final yield is the product of the various percentage yields. Bray studied the problem of the determination of fertilizer re- quirements of crops and stated the problem in the form of two questions: ”1. How deficient is the soil to a given nutrient: that is, how much will it respond to addition of adequate amounts of the nutrient? 2. How much fertilizer is needed to give this response: that is, what is an adequate amount?" While any well designed field plot experiment answers both of these questions, the application of the results obtained by a field plot experiment to soils and conditions other than those under which the trial was conducted is the real problem which must be solved. Bray (19h8) studied problems of potassium and phosphorus avail- ability in the soils of Illinois and attempted to answer the questions he posed. The first question was answered simply by comparing the crop yield from a plot which had received adequate amounts of all plant nutrients with the yield from a plot treated similarly but with one nutrient omitted. The second question was answered by the -'!:‘-‘ inmaw' ' a t. 5.3.1., grill. ‘45 In , [Hart oi 175. application of a series of increments of one nutrient across plots fully fertilized with all other nutrients. Bray pointed out that the fault of the majority of field experi- ments based on single-increment studies is that the increment applied represents a practical amount. This of course will neither indicate the extent of the deficiency nor the amount of fertilizer needed to overcome the deficiency. Bray avoided this fault by a modification of the MitscherlichrBaule theorem so that the constancy of c value need not be assumed under all conditions. and he succeeded in develop- ing a technique by which the quantity of fertilizer required for a given crop could be determined by a simple chemical analysis of the soil. His analytical methods have been standardized against field trials over a wide area and over‘a number of years by using the Mitscherlich logarithmic equation as the basis of comparison. b. Correlation of the Potassium Test with Percentage Yield The correlation of the chemical test for available potassium with the percentage yield illustrates the general procedure used by Dray. It is important to emphasize that the chemical methods used by Bray were precise laboratory techniques for determining the total exchangeable and water soluble forms of potassium and the chemosorbed and easily acid-soluble forms of phosphorus. Bray established a series of 23 test plots on soils which were fully fertilized except for potassium. One plot of each set was supplied with adequate amounts of lime, phosphorus, and potassium (LPK) 176. while the second plot received lime and phosphorus only (LP). Green manure legume crops were grown in rotation to supply nitrogen. Before fertilization, each of the plots were thoroughly sampled to a depth of 6.6 inches. The yields of corn obtained on the LP plot were expressed as percentages of the yield from the LPK plot which was considered to have given a maximum, or 100 percent, yield. For example, a soil which.contained 158 pounds of available potassium per acre of 2,000,000 pounds of soil, the crop yields were 79.3 bushels of corn on the LP plots and 89 bushels on the LPK plots. Thus the yield from the LP yield is 89 percent of that from the LPK plot. The percentages of the maximum yield obtained from the 23 plots are plotted against the amounts of available potassium expressed in pounds per acre in figure 30 together with the yield curve obtained by the equationfififilry) : LogA - clbl' where A.is the 100 percent yield from the LPK plot, y is the yield from the LP plot, c1 is the propor- tionality constant, and b1 is the amount of potassium found by chemi- cal test on the LP plot. . While Bray (19h8) used the logarithmic equation developed by Mitscherlich, he did not agree with Mitscherlich(1909), Baule (1918), and hilcox (1937) that the c value is constant under widely different conditions. He therefore approached the problem in a slightly differ- ent way. He averaged the proportionality constants for a large number of field tests carriedout over an area. If the individual values of the proportionality constants were not significantly different, he 177. guns or co»: In Mr nor Logan-15) :- Lo, IOO‘ «>955 (noun-0.8““ } co so too too loo acumen-Ls K m Ibs. by Text Figure 30. Relationships between exchangeable potassium and corn yields on untreated plots receiving no K20. (Bray: 19148) ' 178. considered that the average value would serve as a constant over the particular area. In his study of available potassium, Bray averaged the values of c1 and by taking this averaged value he constructed the curve shown in figure 28. 'As the standard error of estimate for this curve was 5 percent, Bray considered that it was possible to use the same inter- pretation of his chemical test for available potassium for all soils and conditions represented in his study. ‘Using this curve, Bray pre- pared table 28 to predict the percentage increase in yield obtainable when adequate potash fertilizer was used for the different crops. Bray realized that in nature there is seldom a single factor governing plant growth and he therefore assumed that any practical procedure must be capable of determining the requirements for more than one fertilizer at the same time. He accomplished this by applying Baule's percentage yield concept as described in the succeeding section. d. Application of the Baule Percentage Yield Concept to Several Nutrients at the Same Time The Baule percentage yield concept states that if there is enough available potassium in the soil to give 80 percent of the maximum yield obtainable with adequate potassium, and enough phosphorus to give 70 percent of the maximum yield obtainable with adequate phosphorus, then the yield obtained.will be Yield: 99... 2.9... :56 percent 100 100 179. Table 28. Relationship between the percentage of maximum yield without added potash and chemical test for available potassium. (Bray. 19h8) Yields without added potash Available potassium Corn Soybeans Wheat or oats (b1) (e1 0.0065) (c1 0.0077) (c1 0.012) Lbs. per Percent Percent V Percent acre no #5 51 66 60 60 65 80 8o 70 76 90 100 78 83 9“ 130 85 90 97 150 90 93 98 200 95 97 - 300 98 - -~ 180. This means that if all factors affecting growth.were optimum excepting potassium and phosphorus, and if these nutrients are present in the percentages of the Optimum concentration as indicated, the crap yield would be 56 percent of that obtainable if adequate potassium and phos- phorus were present. Bray (l9hh, l9hba) used the data from the potassium experiments shown in table 28, and a further set of data for experiments with phosphorus in his application of this concept to crOp yields in Illinois. The agreement which he found between the calculated and the observed yields are shown in figure 31. .Uhlike Baule (1918) and Mitscherlich (1909), Bray does not consider that the percentage yield concept can be applied to such mobile nutrients as nitrogen and water and therefore restricts the application of this concept to phosphorus and potassium. d. Correlation of Soil Analyses with Fertilizer Requirements Bray defines fertilizer requirement as the amount necessary to obtain 98 percent of the maximum yield obtained if the nutrient under‘ consideration.were present in an Optimum amount. The value 98 percent is taken rather than 100 because of the increasing requirement as the theoretical maximum is approached. As the fertilizer requirement changes from year to year, with the kind of fertilizer used and with the method of application, it is necessary that new field trials be (conducted for each fertilizer and for each method of its application. It is assumed, of course, that the fertilizer requirement follows the 181. was mean” 1.. VI‘L’ “M“ 3““ “a” 11""; 110 I snow“ 5,;y"“°'c 0.. a. PLOT . FROM P ¢ K 1:5;- p 4 K Tu?! ] loo- '. u c , 0 z ( i E 00- . .. . g L; I P”, ‘. . . ,_ r‘ :_ [/4 z {3 3335/, M no» [253:3 A [/4 f-l-i‘ ""3 / 2 [5.352% ?I 1 - I _ z u Flt-£3 - F's": , . r E j 1 .. ‘ ’ L“ 20 - 5 i ,<_ I ' ‘ ‘23 9. E J ¢ 2 O > 0 ¢ ‘ u a 2 3 < a s 1 0° 91; ca 0 m [a :3 __ Figure 31. Comparison of actual and calculated increases in com yields from field plots. Calculated yields are derived from Baule's percentage yield concept, and are based on chemical analysis of the soil. (Bray, 194M.) 182, Law of Diminishing Returns. If it is further assumed that the amount of a nutrient found by field test required to produce the maximum crop yield on a particular plot is accurate, then the application of the Law of Diminishing Returns permits the determination of the complete range of fertilizer requirements, as exemplified for the potassium tests presented in table 29. Bray conducted his experiments on a series of fields over the State of Illinois, but he pointed out that the same information could be obtained by an adequate field plot design. For example, if for a given crop, an average c value in the equation Log (Apy) ; Log Apclbl is found to apply over a wide range of soil conditions, then a chemi- cal analysis of the soil can be used to estimate the fertilizer re- quirements of the crop throughout the area. Experimental procedure requires that one plot be adequately fer- tilized with all nutrients except one, and an adjacent plot be treated ‘with the optimum amount of the fertilizer being studied. The experi- znent should be repeated over as wide a range of soil types and condi- tions as practical. If it is found that the c1 value is constant, then the method is applicable to the area in question. To establish the more exact fertilizer requirement for each , chendcal analysis of the soil, a study of the rates of fertilizer emailication is required. Such a study may be conducted as a separate experiment or it may be conducted simultaneously by applying a further modification of the Mitscherlich formula: L05 («Ar-3') : Log A - (clbl Q-cx) 183. Table 29. Approximate potassium requirements, calculated as potassium chloride, for individual crops based on chemical analysis of the soil. (Bray. 19h8) Potash requirements for different crops in terms of muriate of potash (0-0-50) Available potassium Corn and clovers Soybeans Wheat and oats by test (c 0.0094) (c 0.015) (c 0.020) lbs. per lbs. per lbs. per lbs. per acre acre ' acre acre 110 153 94 62 60 136 8# 50 80 12h 72 36 100 110 62 24 130 92 1+6 -_ 150 7n 36 ~~ 200 ’40 - -.. where c is the constant for the fertilizer used and x the increment or amount of fertilizer added. It is advisable to establish a central plot deficient in the fertilizer being studied and four surrounding plots receiving increas- ing increments of the nutrient under investigation. The greatest fer- tilizer application should be adequate to support the maximum yield in so far as this nutrient is concerned. It is important to point out that it is essential to have as little soil variation as possible between the deficient plot and the various increments and consequently- the plots should not be randomized. 9. Discussion The contributions of Roger Bray constitute the most interesting advances of the present day in the study of the fertilizer requirements of crepe. His studies have been conducted.with corn, wheat, oats, and soybeans in Illinois, and other workers are applying his techniques to other creps in many areas. The writer is applying the technique to the tomato fields of Prince Edward County in Ontario, and to the potato fields near Guelph, Ontario. While there is no doubt that Bray's procedure may be applied in areas of relatively uniform soil conditions such as are found in Illinois, there is the possibility that difficulty will be encountered in areas where wider variation of soil types are found. If this possibility proves to be unlikely, or of small significance, Bray's contribution to agricultural science is of a high order. One point that might be questioned in Bray's concepts is his ‘belief that those nutrients regarded as "mobile" cannot be investi- gated by Baule's percentage yield concept. As was pointed out earlier, the amounts of nitrogen and water applied by numerous workers was part of the basis upon which the logarithmic yield equation was developed. Appleton and Wynd (1951) found that in 19b8, the relation “between yield and applied irrigation water followed the relationship expected from Bray's mobility concept, while in the following year this relationship followed Baule's percentage yield concept. Whether or not Bray's concept of the special problem of the "mobile" nutri- ents is correct, his contribution to the problems of potassium and phosphate fertilizer use is of unquestionable value. 1+. Procedure of Willcox a. Introduction One of the less generally understood techniques of interpreting :field experiments in order to show the maximum crop yield obtainable with the use of fertilizer, is that proposed by Willcox (1930-19n9). This is probably due to the general lack of understanding of the Law of Diminishing Returns. Perhaps the application of this law has been hindered by the very mass of evidence which Mitscherlich (191w), Gerlich (191+?) and Willcox have presented to support it. Only Bray and Villcox have made significant attempts to apply the Law of Dimin- 1Bhing Returns in recent years. 186. The greatest difference between the work of Bray and Willcox is the adherence of Willcox to the idea of the constancy of the effect factor as expressed in Baule units and as used in the equation Log (Apy) ;_-_ Log A - .301 x where x is in Baule units. Willcox has prOposed techniques which may 'be used by the research workers for their interpretation of the data obtained from field trials (1915/0, 19M, 1941a, 1941b, 19%, 1916, 19h5a, 19b7, 191+7a) and also a simpler "farmers" technique (undated circulars a and b). b. Procedure for Research Workers (1.) Preparation of the Standard or Universal Yield Diagram Willcox (191+7a) described the construction of a universal yield diagram which is the basis of the interpretation of his yield data. He plotted on a standard sheet of graph paper the units of yield per acre as the ordinates and the Baule units of fertilizer as the abscissa. He then drew the curves from A _-_ 10 to A :- 2’4 inclusive, using his Published co-ordinates (l9h7a), and obtained a diagram similar in form to that shown in figure 32. After these whole number curves have been drawn the intermediate curves may be drawn by interpolation. Willcox I‘efers to this diagram as the universal yield diagram because it pre- sents the crop yields obtained up to 8 Baule units of a plant nutrient, beyond which no increase in yield could be expected. As field plot experiments frequently use less than 1 Baule unit 187. B UNITS or YIELDS . 1 Figure 32. General agrobiologic yield diagram from.O to 7 Baule'units. (Willcox, l9h7a) per acre of each of the major plant nutrients, it is inconvenient to plot yields obtained by these small additions on this general diagram. Willcox suggested that the investigator provide himself with two standard yield diagrams, one for very small increments of fertilizer (from O to 2 Baule units), and one for intermediate quantities (from O to 4 Baule units). This permits a spreading of the curves which facilitates the convenient use of the diagram. Table 30 permits the determination of the co-ordinates for a diagram of any scale. This table shows the percentage of the maximum yield obtainable from a given number of Baule units of any nutritional factor. For example, if the coordinates for A.= 25 are being computed, the head of the first column in table 30 indicates that 0.1 Baule will produce 6.? percent of the maximum yield. For curve A.= 25, 1.65 or 0.067 x 25 is the point where this curve crosses the ordinate at 0.1. This pro- cedure repeated gives the complete series of curves necessary for the universal yield diagram. It should be pointed out that the magnitudes of the Baule units are as follows: Nitrogen - 223 pounds per acre P205 _ #5 pounds per acre KéO 82 pounds per acre (2.)Interpretation of Field Plot Data The use of the general or universal yield diagram described above has been described by Willcox in a series of papers beginning in 1941 189. om.mm o.m mn.He m.m He.em a.m mN.Ha m.a He.ee m.o Ho.ea o.m oH.He m.m Hm.mm o.~ mm.eo a.H am.me m.e mm.mm o.e mm.om e.m mm.mm m.m Ho.aw m.H ee.mm a.o 36m 06 3.8 Tm no.8 sum no.3 min Norma m6 mm.wm o.m ma.mm ~.m mm.ma m.m HH.Nm e.H mm.mm m.o ma.mm 0.: em.mm H.n em.ma N.m mm.mm m.a Hm.em e.o on.mm m.m on.am o.n aw.ma H.N ee.mm ~.H aa.mfi m.o mm.mm m.m om.mw m.~ oo.ma o.m mm.mm H.H em.~a m.o Hm.mm a.n em.mm m.N Hm.ma m.H 0.0m o.H oa.m H.o memes heads mead: means means < so a ofleem q no a oases «_eo a oases «_eo a oHeem «see a ofleem Asthma 683.35 .033 has so aeawma .30; Hemaoefiab e msfipamaoo no.“ spam .om 0..“me 190. and continuing to the present day. The majority of these papers are examples of the applications to the data published by other workers. Willcox himself hasnot conducted experimental work under field condi- tions. It will serve our purpose to explain his technique of inter- pretation by referring to one of his applications (1943) based on data published by Olsen and Shaw (me). In this study, as seen in figure 33, the universal yield diagram consists of a series of Mitscherlich yield curves numbered from 6 to 1?, calculated from corresponding values of A by the Mitscherlich— Baule equation, log (by) ,-_- log A — 0.301 x The first step in the procedure was to obtain the total yield in pounds of dry matter. Since this gives large values, it is convenient to convert these into units which will fit on the diagram. The most convenient unit is 800 pounds. The yield data from the Olsen and Shaw experiment, when so converted, are shown in table 31. It was necessary to convert the amount of fertilizer added to'Baule units. Olsen and Shaw (l9h3) used O, 10, 20 and 110 pounds of K20 per Etcre and as l Baule unit is equivalent to 82 poundsof 320 per acre, ishe number of Baule units added were 0.122, 0.2h4, and 0.488, as indicated in table 31. The interpretation is given below, using the experiment on Miami Silt loam in 1939, and represented by B in figure 33. A sheet of transparent paper was placed over the universal yield diagram. A line was drawn over the horizontal line corresponding to the ordinate 9 on 191. 9 1 “mu 0' goo roams) YIELD(OF CORN AND STOVE“ q l I \ anuu UNIT: w “,0 Figure 33. Four of Olsen and Shaw's (191.3) field experiments plotted on the universal yield diagram, A. Clemont 5111-. loam (191.0) E. Miami silt loam (1939) C. Muskingum silty clay (19w) D. Wooster silt loam (191.0) (Willcox, 1943) seao pflam senses: a seam pflnm sees; m has 33m gyms: o . econ 3.8 Scenes 4 $4.0 $06 on? mwdé mofi. wnwm esm.o 00.0 cow: esm.o mo.a mmom NNH.0 mm.m new: NNH.o am.m mamm o mmfi .33 n 0 $6 8% 0 mm:.o «H.HH mmmm mms.o sm.a samm ssm.o mm.oa mmew ssm.o mm.m was: mea.o sw.m saws mma.o aa.m ans: 0 on.a uni. m o #0.: mmmm 4 emu one 323 3.29 3500.. 30m mean: and? wagon 20m 0.33“ ago; madam ago; “new.” .Hooaagv 330.3,; .3 gamma .30?» «.3355, on» no soapsposfimufi no.“ . gadgoaso ms A963 3% was «830 Son.“ 500 Mo mwamfn 23 no.“ 33 Jun 3”pr __.-— 193. the diagram, and on a vertical line above this, a point is indicated at the value of 9.30. A second point is drawn at 0.122 Baule units to the right corresponding to ordinate 9.84, a third at 0.2hh Baule units to the right of the first dot corresponding to ordinate 10.56, and a fourth point at 0.b88 Baule units to the right corresponding to the ordinate 11.17. The transparent paper is then removed to the right or left until the four dots aline themselves, as close as possible to one of the curves. The value of A is given by the number of the curve which shows the closest fit. In this case, the curve is number 16 and A is equivalent to 12,800 or 16 x 800 pounds per acre yield of corn plus stover. This is the maximum yield that could be obtained on this field by using potash in such a year as 1939. The amount of potash in the untreated soil is determined by finding the Baule units of Kéo in the check plot, which produced 9.30 units of yield, and then multiplying by 82 to convert the value to pounds per acre of Kéo. The same procedure will give curves of similar type for the other data in table 31. Besides the maximum yield (A) and the amount of a nutrient in the soil (b), there are other important conclusions that can be drawn :from the use of this universal diagram. For example, a curve located iis is the one for Miami silt loam indicates that the soil contained a éaood supply of nutrients other than potassium, but that potassium was iihe limiting factor. From the slope of the curve, one could expect iihat larger additions of potassium fertilizer would further increase the yield in this case. The equation, log (lS—Y) : log 16 - 0.301 x can be used to determine what increase in yield could be expected.with the addition of the optimum amount of potash. For example, if Olsen and Shaw had added #0 pounds more of K20, they would have added a total of 2.251 Baule units. The equation above shows that this further addition would have resulted in an increase of 1176 pounds of corn plus stover. The same situation is apparent from the data obtained on Cler- mont silt loam (A). On the other hand, the Muskingum silty clay (C), and the Wooster silt loam (D) contain identical amounts of potash, and when yield data from these two soils are plotted, the curves in the diagram indicate that the factors causing the different yields are other than the amount of available potassium. This conclusion follows because the yield curves fall on the flat portion of the Mitscherlich curve. These examples cited by Willcox in.which all points fall on the curve are, of course, exceptions and Willcox has found it necessary to study the instances in which this is not the case. In the ”scatter“ type of distribution discussed by Willcox (194fl, 194ha) the points cannot be made to fit smoothly on any of the normal yield curves, lyut arrange themselves in the form of an inverted crescent in some (cases. as illustrated in figure 32, and in other cases they oscillate 1?rom one side of the standard curve to the other. Willcox maintained that, except in the most extreme cases, the aberrant points of this type line up along the curve with sufficient closeness to make iden- tification of the most probable value of A fairly certain. In the case of the experiment with sugar cane, the results shown in figure 3b indicate that the aberrancy probably was due to small differences in the individual plots and represented experimental error over which the investigator had little or no control. A second type of aberrancy is shown in figure 35 and is referred to as the "near-end” depression, and was discussed by Willcox (l9bh, l9bba, undated c). This particular type of aberrancy is marked by an abnormal spread between the yields obtained from the untreated plots and those obtained by the first increment of fertilizer, even though the yields obtained by the addition of the higher increments conform to the normal yield curve. Willcox does not consider that he had a certain explanation of this effect, but he suggested that it may have been due to one of two causes: either from the depression of the third and fourth points due to the heavy application of potash, or to the influence of the potash on some unknown soil condition which established a level of fertility qualitatively different from that existing in the untreated plot. A third type of aberrancy is referred to by Tillcox as a "far-end" depression.which he discussed in a series of papers (l9hla, 19kt, 19h5, 19h5a, 19h7). This effect, due to the depressive action of some factor on growth, was expressed by Mitscherlich (1928) mathematically in the equation y = A (1 - 10"”) 10"”2 1% . 3 Umu pun: op Susan' 3 V A L k A or 5 Damn or K.O Figure 34. The scatter type of aberration in a sugar cane-potash experiment conducted in a good(G) season and in a poor (P) season. (Willcox, 19m) 1‘27 . ~ 0 r 3 or wean 5 UNIT YItLDO Figure 35. The "near-end" type of aberration in sugar cane treated with potash in a good (G) season and in a poor (P) season. (Willcox, 19W) 198. An example of this effect is given in figure 36. This curve was pre- pared by Willow: (19%) from Prescott‘s data on corn yields and it shows that the yield is depressed by the higher applications of nitro- gen. Willcox maintained that this type of aberration generally is due to the lack of "agrobiological balance" between the increased amount of the nutrient being studied and the amounts of-other factors which.are being kept at a constant Optimum. It would seem that in such.instances, some other factor must be corrected before the maximum benefit could be obtained from the fertilizer. c. Procedure for Farmers Willcox (undated a, b) recognizes the necessity of a simple method whereby a farmer could determine how much plant nutrient is contained in an untreated soil and how much fertilizer could be profitably used. He proposed that the farmer select a strip b8h by 105 feet on the field to be tested, and plant a crOp such as corn. The test strip will therefore be comprised of 35 rows wide. If all three major fertilizer components are to be investigated, alternate groups of five rows should be fertilized with Kéo, P205, and nitrOgen at a given rate. as indi- cated in figure 37. The crop should be given the normal care and at harvest the farmer would measure or weigh the crOp of grain from the middle strip and from the two adjacent untreated strips. The yields from the untreated strips would be averaged for comparison with the yield from the treated strip. 199. '0 ,"5 ‘. sn- 0 3 Ila" 10 n t . .3 igbb ‘u ‘5'- D g O'H- P I A l l ‘ ' ‘ .2 .h 06 '8 Figure 36. "Farbend" depression found in Prescott's field test with corn. ('Willcox, l9hh) 200, ‘fi we 4’ a U N TREAT ED Then to wn N no UNTEEATED I TREATED wrrn 9‘05 '05 UNTKIATCD TREATID WH’I NITROOCN UNTRCATIO i Figure 37. Plot design for three one-plot fertilizer tests with corn. (Willcox, a) 201. The data the farmer will have at his disposal in the case of a ! study of the effects of potash on corn would be as follows: Amount of Rio used 38 pounds per acre Yield from treated strip 72.3 bushels per acre Yield from untreated strip I 51.0 bushels per acre Percent increase in yield al.76 percent due to the added potash By referring to figure 38 or to other appropriate curves, (Willcox b), and by locating the percentage increase in yield on this graph, the farmer would find that there were 58.5 pounds of K20 in the untreated soil. Then by referring to figure 39, or to other appropriate curves, he would detect what percent of the maximum yield this value repre- sented. In the present instance, the value is no percent which means that the 51 bushels obtained are only to percent of the yield obtainp able by adding an adequate amount of potash. Finally, by calculating the cost of the fertilizer, and the value of the increased yield, the farmer could determine from figure 39 how much fertilizer he could apply economically. (1. Discussion Unfortunately the techniques preposed by Willcox enjoy no general recognition as to their value and importance. The greatest disagree- ment concerning their value has centered around the question of‘whether or not the effect factor c in the Mitscherlich equation is a constant 202 . J23 YIELD PERCENT INCREASE OF w—i I | V J O n to 70 D ’0 Pouuos or «.0 muummto sou. —--b-——dD——-1 Figure 38. Diagram for use with the one-plot technique to show the amounts of K20 in the soil. (Willcox, a) 203. a / g z E E i 60:- < E ( ,L a an _ 0 j// *- J 2 u: Q 29 _—+- —<—-————4 a: K K o no zoo Joe we 5'00 600 pouuos K,o PER ACRE Figure 39. Diagram for use with the one-plot technique to show the amount of K20 fertilizer which may be profitably applied to a crop. (Willcox, a) 20h. under the widely different conditions under which Willcox used it. He quoted one worker's comments on this point as follows: ”It is unthinkb able that one equation'will be equally applicable to all growth factors unless that equation is so general or lacking in precision as to be of very little value..........Surely, not all growth factors work:in the same way." (Willcoz c). The criticism of "unthinkability" is not evidence and in no way could such a subjective reaction discredit the work of Willcox. It is by demonstration, not by "unthinkability,' that Willcox's prOposals should be investigated. The mass of information on a large number of crOps growing under widely different conditions accumulated by Willcox does not necessarily prove that his concepts are correct, but it does Justify careful con- sideration of his work. Until actual eXperiment adequately disproves the work of Willcox, his concepts must be regarded as a significant contribution to the science of fertilizer use. It is the ouinion of the writer that the techniques described above should not be delegated to the minor role they now possess, but should be applied more generally in an effort to show by actual data if they can replace current methods based on statistical treatment. If these techniques can be shown to apply. they will be of extreme importance for they put the interpretation of chemical analyses of soil on a firm foundation and permit individual farmers to conduct and evalur ate their own field trials. 205. VI. USE OF SEEDLINGS FOR DETERMINING NUTRIENT DEFICIEECIES 0F CROPS A. Introduction The various methods for using seedlings for the determination of plant nutrients in soils deve10ped during the last twenty-five years, were attempts to replace chemical analyses and pot culture procedures. The theory of the seedling techniques was that the plant itself was the ideal extracting reagent. The theory implies that chemical methods were restricted in their value because no extracting agent had been found which would give a true indication of the amounts of available nutrients under all conditions. The seedling techniques were develOped as a supplement to pot studies and field plot tests because of the need for a simpler and more rapid laboratory method. thermore, these procedures were thought necessary to supply more detailed information concerning the fertility of the individual fields, for of course, it is economically impractical to use field plots or not experiments for this purpose. The most acceptable procedure utilizing seedlings was that developed by Neubauer and Schneider (1923) and Neubauer (1923, 1925, 1929) and later modified by Thornton (1931, 1931a, 1932, 1932a, 1935, 1935a), Pettinger and Thornton (1931;), and McGeorge (1939. 19ers. 1946, 19u6a). B. Neubauer Procedure 1 o T 8011111 qua The principle of the Neubauer method is the intensive absorption of nutrients by a large number of rye seedlings grown on a small quantity of soil. These plants, growing under optimum conditions, form exten- sive root systems and exhaust the soil of the available plant nutrients in a comparatively short time. Chemical analysis of the tops and roots of the seedlings discloses the amounts of nutrients removed from the soil. The conduction of Neubauer's test requires meticulous care, and the details of manipulation must be standardized so that comparative data may be obtained. Failure to adequately standardize the conditions probably is the major cause of the failures to obtain satisfactory data that have been reported in the literature. It is necessary to obtain a representative sample of soil at the depth desired, to air dry it under uniform conditions, and to mix it thoroughly by passing it through a 2-millimeter sieve. A moisture content must be determined so that samples equivalent to 100 grams of air dry soil may be taken from the bulk soil sample. One hundred grams of the soil are mixed.with.50 grams of quartz sand which has been.washed with hydrochloric acid and distilled water and then dried at 1500 C. The mixture is placed in a glass dish 11 centimeters in diameter and 7 centimeters deep and moistened with 20 milliliters of distilled water. A further 150 grams of dry quartz sand are distributed evenly over the surface and moistened with an additional 20 milliliters of distilled water. Rye seeds exhibiting a high percentage germination, selected for size, and.weighing approximately h grams per 100 seeds, should be used. 207. Neubauer used the Petkuser variety of rye furnished by the Stastliche, Landwirtschaftliche, Versuchsanstalt, Dresden—A, Stubel Allee 2, Germany. The rye seeds are disinfected by soaking 1.5 hours in a solu- tion of Uspulum. Using a standard pegboard, Neubauer punched 100 uni- formly spaced, concentric holes in the soil in each container and planted one seed in each hole with the germ and placed at the bottom. To facilitate subsequent watering, a glass tube is forced into the soil in the center of the container. A further 100 grams of quartz sand, moistened.with 20 milliliters of water, are spread over the surface and the weight of each dish then recorded. Three dishes are prepared as above for each soil being studied and three blanks are used containing only quartz sand. The cultures are maintained at constant temperature, preferably 200 C. i110 C, and protected from direct sunlight. Each day, the cul- tures are brought to their original weights by adding distilled.water. The seedlings are harvested at the end of 17 days and dried. The roots and tops of the dried seedlings are ashed. and the total amounts of K20 and P295 which they contain are determined. The amounts of 2205 and x20 found in the seedlings grown in the blank cultures are subtracted from these amounts in order to obtain the amounts of the nutrients removed from the soil. These corrected values are re— corded as the number of milligrams of P205 and K20 removed from 100 grams of dry soil, and they are assumed to represent the total available amounts of the nutrients in the soil. 208. Neubauer used the data from a large number of tests, and deter- mined the limits below which deficiencies occurred. The "limit values" given by him in three publications exhibit a certain variability as indicated below. Heubauer Limit values, below which sub-maximum yields occur, eXpressed as milligrams per 130 grams of dry soil Nutrient Year Barley Oats Rye Wheat Turnips Potatoes Sugar Beets P205 1923 6 6 8 8 .3. 9 10 1925 7 7 8 8 lb 10 12 1929 6 6 5 5 7 6 6 K20 1923 11+ 17 17 15 -- 37 33 1925, 18 18 it» it» us 39 37 1929 21+ 21 17 20 39 37 25 2. Discussion The seedling technique using rye described by Neubauer and Schnei- der (1923) has been used extensively, and has been critically investi- gated, particularly by the German investigators. There is a vast number of papers in the German literature concerning the application of the method, particularly in relation to its correlation with re- sults obtained in the field. A.number of aspects of Neubauer‘s procedure have been thoroughly reviewed by Stewart (1932). Neubauer and Schneider (1923) maintained that the seedlings grown in the dark absorbed the same amount of nutri- ents as did those growing in the light. Gunther (1926) and Thornton (1935) agreed with Neubauer, but other workers, especially Wiessman (1925) did not agree, for they believed that the stronger the illumina- tion, the greater its influence on nutrient absorption. and seedling growth. Thornton (1935) reported that his data indicated that "no significant differences could be observed for plants growing simultanr eously in different light intensities or for plants growing at differ- ent seasons of the year, or under different weather conditions." Thornton states, however, that a strong, evenly diffused light is best and direct sunlight is undesirable. Probably the most important single factor influencing the results is the temperature, and this has been studied both by Neubauer (1923) and by Thornton (1935). Both workers showed that a lower temperature retards the development of the seedlings while a higher temperature materially shortens the growth period. In either case, nutrient ab- sorption is hindered. Thornton reported that differences as great as 50 percent are produced when the temperature varies from 120 C. to 23° c. As mentioned above, accurate data are obtained only if strict attention is paid to the standardization of the growing conditions. C. Thornton Procedure The seedling technique followed by Thornton (1931, 1931a, 1932, 1932a, 1935, 1935a) is essentially the same as that proposed by Neubauer (1923) except for some minor modifications necessary because of lack of laboratory space and equipment. Further modifications have been made in 210 o the choice of seed and the disinfectant used to sterilize them. Thornton used Rosen rye supplied by the Farm Bureau Services, Inc. at Lansing, Michigan, and he treated the seed with Semesan Jr. instead of with Upsulum. Thornton apparently rigidly followed the essential de- tails of the original Neubauer technique and merely altered those as- pects which rendered it more readily followed under the conditions under which he was working. D. McGeorge Procedure The McGeorge technique of conducting seedling tests is another minor modification of Neubauer's original procedure. McGeorge (1942, .19h6, 19h6a) modified the Neubauer technique to include the determina- tion of the minor nutritional elements. The details of the procedure and the calculation of the data for the minor elements are presented in the cited publication (l9#6). E. Discussion A number of factors somewhat decrease the value of the seedling methods for the estimation of the amount of available nutrients in the soil. One of these is the need for special equipment that is seldom found in plant nutritional laboratories. It is apparent that even slight differences in technique bring about large differences in results. Ames and Gerdel (1927, 1927a, 1928), for instance, used larger containers and a larger amount of soil for the growth of their wheat seedlings. When they failed to find a correlation with other methods, they concluded that the seedling technique did not indicate the amount of potassium in the soils they studied. It would appear, however, that the failure to obtain agree- ment with other procedures was probably due to the larger amount of soil which Gerdel used. If correlation is obtained with 100 rye seed- lings in 100 grams of soil and not with 100 wheat seedlings in 200 grams of soil, the lack of correlation was probably due to the failure of the wheat seedlings to exhaust the available nutrients from the soil. 0n the other hand, workers who have followed carefully the details of the technique and have adequately standardized it, have found close agreement between the results obtained by the seedling method and other biological and chemical methods. Stewart, Sackett, Robertson, and Kezer (1932), Thornton (1935), McGeorge (l9fi6), and Lea and Midgley (1934) all have reported satisfactory agreement of their data with those obtained by other methods. Thornton (1935) probably has done more than any other worker to broaden the application of the seedling method to the study of soil fertility. He has studied such problems as fertilizer fixation by the soils, fertilizer availability, and the influence of one nutrient on the absorption of another. Lea and Midgley (193h), using a combination of the Reubauer method and chemical procedures, were able to prepare a soil map of the Champlain Valley in the State of Vermont. Vblk and Truog (193G), Stewart, Sackett, Robertson and Kezer (1932), and others used the seedling method to standardize other biological techniques and also chemical procedures. It is questionable whether the seedling technique can be used for the determination of the actual amounts of fertilizer required by crops for their Optimum growth. Neubauer and Scheider (1923) have attempted to do so, but they had to make a number of broad assumptions. For instance, they assumed that the seedlings removed all of the avail- able nutrients from the soil. However, plants in the field can remove only a fraction of this total supply. These authors referred to this fraction as the ”utilization coefficient", and published the following statement of their probable magnitudes: Neubauer Utilization Coefficients expressed as percentages nutrient Barley Wheat Potatoes . Roots Legumes KEG 12 15 25 33 20 P205 20 33 . 33 33 33 Using these utilization coefficients, Neubauer calculated the "limit values" already noted. He assumed that only a fraction of the applied fertilizers could be used by the plant during a single season. Fbr example, only 60 percent of any potash fertilizer was supposed to be effective or available, and the phosphate fertilizers were considered to be utilized to the extent of 20 percent by barley and 33 percent by other crops. USing these assumptions, he calculated the required quantities of P205 and K20 needed by the crops under field conditions to counteract the deficiencies shown by the seedling test. It is apparent that these estimated quantities of phosphate and potash fertilizers are based on two assumptions: the so-called "limit values" and Neubauer's assumption that 1 milligram Of a nutrient per 100 grams of dry soil is equivalent to 23 pounds per acre. It is further apparent that since these are subjedt to error, accurate estimations from seedling tests of crop requirements are difficult. Any comparison of the results obtained by seedling tests with those obtained from field experiments must be done on the basis of the “limit values." In fact, every field experiment is, in a way, a deter- mination of the limit value for a particular crOp and soil, and it does not appear that enough attention has been paid to the relationship between limit value and soil type. The seedling methods for the determination of the amounts of available nutrients in soil are more closely related to pot cultures than to other methods. In conclusion, it can be stated that both methods serve only to differentiate the more extreme limits of the nutrient content of soils or the fertilizer requirements of crops. 211+. VII. USE OF MICRO—ORGANISMS FOR THE Dm'iINING OF FERTILIZER REQUIRE-{ENTS A. Introduction The use of micro-organisms ‘to determine the amounts of plant nutri- ents in soil, like the use of seedlings, is another method designed to replace chemical. analyses of the soil and to overcome the expenditure of time and labor involved in field plot trials and pot-culture tech- niques. The principal methods which utilize bacteria or fungi as the test organisms are the gotobagteg method (Rifles and Hirschberger, 192a: Truffaut and Bezssonoff, 1927; Winogradsky, 1927, 1928), the W m method (Benecke and sb'ding, 1927; Niklas, Poschen- rieder, and Trischler, 1930, 1930a: Mehlich, Truog, and Fred. 1933), and the Qmigghamella method (Mehlich, Fred, and Truog, 193A, 1935). These methods have been modified by many workers, but as the methods 1 are of doubtful value as practical procedures, the following discussion is limited to the presentation of their principal aspects. 3. ggtobagte: 1. Technique As early as 1921+, Niklas and Hirschberger described a method for rapidly determining available phosphorus in soils by using gotobactez, and in 1927 Truffaut and Bezssonoff also reported their use of Agotobacter as a measure of the availability of various phosphates in the soil. Vinogradsiq (1925) demonstrated that colonies of Azotobagte; grew 215. luxuriantly on soil plaques if they contain suitable carbohydrate as an energy source, and the necessary mineral nutrients. Two years later, Winogradsky and Ziemiecka (1927) reported a close similarity between the nutritional factors limiting the growth of Azotobactgr and those limiting the growth.of plants although he did not at that time use this similarity for determining plant nutrients in soils. In this connec- tion he stated: “The method is intended for the study of nutrient fixation in soil, a study which is scarcely yet underway. It is clear, however, that the reactions of microbes which are so sensitive to mine eral nutritional factors can serve to indicate the latter in the soil and to do so with a sensitivity greater than possible with chemical methods. iAzotobacter has already played this role as an indicator of the lime requirements, but the procedure used by the earlier investi- gators could not give results as precise as those obtained by the method of spontaneous cultures.“ This latter method has been described in detail by Winogradsky (1927, 1928), and has been.modified and placed on a more practical basis by Sackett and Stewart (1931). The details of the technique used by Sackett and Stewart will be described. A.sample of the soil representing the upper 6 to 8 inches of the profile is air-dried and passed through a 20-mesh sieve to remove stones and to ensure thorough mixing of the sample. If the pH value of the soil is less than 6.8, calcium carbonate should be mixedwwith the soil in an amount from 8 to 10 percent of the weight of the soil. Four SO-gram samples of the soil are weighed out and 5 percent of its 216. weight of cornstarch is added. Sandy soils, being low in the anaerobic organisms necessary for conversion of starch into forms available to Azpjobeger, must receive 1 milliliter of 50 percent sucrose solution. Nutrient solutions are added as follows: 1. No additions, used as the control. 2. 5 milliliters of 3 percent 1280 , used for potassium deficiency. 3. 5 milliliters of 6 percent NaZEPOuJZEZO, used for phosphorus deficiency. l4. 5 milliliters of 3 percent £23130“, used for potassium and phosphorus deficiencies. The suggestion is made that from 5 to 10 grams of Kaolin be added to improve the texture of sandy soils and 10 grams of coarse silica be added to improve the aeration of clay soils. Enough distilled water is added to give the soil the consistency of moulding clay and this material is pressed into a Petri dish and the surface is smoothed with a microscope slide. Since Azotobacter is affected greatly by moisture variations, care should be taken to make “certain that equivalent amounts of moisture are present in each culture. One milliliter of a suspension of W grown for 72 hours on mannite-agar is added to the center of each culture. This suspension is prepared by washing the growth from the agar with 0.85 percent sodium chloride solution and diluting to 100 milliliters. The inocu- lated soil cultures areallowed to stand in a moist atmosphere for 72 hours at 30° C. and the bacterial growth observed. 217. Sackett and Stewart (1931) base their interpretation of the experimental data on the number, size and pigmentation of the colonies appearing on the soil plaques. They recommend that the growth of the Azotobacter colonies be compared and the cultures classified as follows: Class 1. Very deficient soil. No, or few to numerous, extremely small pinpoint colonies on unfertilized plaques. Few to numerous medium to large, distinct and vigorous colonies on fertilized plaques. Class 2. Moderately deficient soils. 'Few to numerous, but small and weak, colonies with little or no pigment on unfertilized plaques.’ Few to numerous, distinct and vigorous colonies on the fertilized plaque. ' Class 3. Slightly deficient soil. Number of colonies on unfer- tilized plaque as numerous as on the fertilized plaques, but smaller and less luxuriant. Few to numerous, distinct and vigorous colonies on the fertilized plaque. 01838 b. Not deficient soils. Colonies on both fertilized and unfertilized plaques equal in number and development. 2. Discussion fhere is no doubt that the Azotobactez technique is a useful tool for the diagnosis of soil deficiencies although some difference of opinion is evident as shown by table 31 prepared by Halverson.and Hoge (1942). the technique is qualitative only, and its quantitative application nacho doom eoooeooa Se page e8.“ 9: ofiaooe £8333 218. opsaamoma oa03_copmoa mafia» Had dopmofivad puma mupomdnosmmmz mmmH .dmoa uoHHmh mofiosofiofimee opmnamona you mean» capped no ”way stoH mama maze» mouse on» no uaoonom NH hams a“ anowpfieaoo eaoam mod: encapsaouaoo maoufiad Nmma seesaw unmade you and» seem“: hopedpoponq non apnea Ioawsdmu aaafiqfla on» dopuoamm .umhaoou gone up“: unofipmaoauoo doom thmnommou obsm pwma cayenne Nmmn meson madamou sameness chem «mos mouepm dynamo; Nmmfi steam one mampflgsn pump henchmen noqu dfid now one 3 8233 no: pee» on» use; eofioaom 8238 ~an upeopom Jeosoem 338$ bases see» one eoeeoEooom oefioaoo HRH totem ego £833 seamen coda op memoHamas nos memes smash omma «humans was omega madamou wsdwmasooco seem mamma «30H mmmfl ambaflnam use aoxamz .meompma Hmofisomc was memo» eaofim hp vosuwmnoc apnea museum mmma _amoqnoanfisa 0.3.2? no we: who» on» use» doauoaom deoausmfiom mmmH newsman—wan oumnamonm non owe mud obouaas pom ewe pap was» one doeqoeaoomm hudauoo deH homaopnomham use seamen 3328... 233a do. 9: sonata .33. $3 .32 banishment. soaemfimobsu on» up dosages? 30m no 0938 use» «Javanese-an newado«apum O‘QL‘N‘J 0.031s .344...» HHOWHQS! 219. would be difficult. No conclusions could be drawn other than general opinions concerning what fertilizer treatments might improve crop yields. Stewart, Sackett, Robertson, and Kezer'('l932) claimed that it is sufficiently quantitative for practical purposes and that it could be used to indicate the amounts of fertilizer necessary to ob- tain maximum crop yields, especially in the case of phosphorus. These workers, as well as Dahlberg and Brown (1932), have used the method in extensive surveys of different soils and concluded that the proce— dure was as accurate as either chemical analyses or seedling methods for determining nutrient deficiencies. The failure of the method, however, to give truly quantitative data has delegated it to the role of a general survey method, as also are the seedling techniques. 6. £2 ergillug giger 1. Introduction vendeceveyo (1948) credits Butkewitsch, in.Russia, as the first to suggest the use of gpergillus gigs; as a biological indicator of the amounts of available plant nutrients in soils. Benecke and Sailing (1928), however, first deve10ped the method, andused it to determine the available phorphorus and potassium in soils. They compared growth of the mycelium in liquid cultures to which a definite amount of phos- phorus and lpotassium and a small amount of soil had been added. The first extensive use of the method was by Niklas, Poschenrieder and Trischler (1930, 1930s), Niklas and Poschenrieder (1932, 1936). The 220. procedure was modified later by Mehlich, Truog and Fred (1933) and by Mulder (1939-i9no). 2. Ni klas Procedure The original method devised by Niklas and his co-workers was an adaptation of the earlier methods which permitted the quantitative determination of available phosphorus, potassium and to some extmt also the magnesium in the soils. They added 2.5 grams of air dry soil to 30 milliliters of a nutrient solution inoculated with a suspension of Agpgrgillus pigs}; spores. The culture flask was incubated at 35°. C. for four days, and the mycelium was then washed, and dried at 50° to 600 T. for 12 hours and then at 60° to 90° 0. for 1 hour. They cited the minimum weights for the dried mycelial pads below which a'deficiency of the particular nutrient being studied occurred. Niklas used the method to test a large number of soils in Germany and found a close agreanent between the results of the Aspergillgg technique and those of Neubauer obtained from the growth of rye seedlings. 3. Mehlich Procedure A critical study of the Niklas procedure induced Mehlich, Tnzog and Jrcd (1933) to add omsrou.2séo to the medium as a buffer. when this was done, they obtained a better growth of the mycelium, particularly if the soil was strongly acid. They further obtained more consistent data if the mycelial mats were dried at 70° to 90° c. for 12 to 1b 221. hours and at 105° C. for 2 hours, and then analysed the dried material for the nutrients being studied. In order to permit quantitative determinations and accurate com- parisons, Mehlich, Truog and Fred (1933) calculated the data on the basis of the amount of x20 taken up by the mycelium from 100 grams of dry soil. Their results indicated that Aspergillus Ligg; removed considerably more of the nutrient from the soil than did Neubauer's rye seedlings. As a result of their comparisons of their data with those obtained by the Neubauer method, and with those obtained by chemical analysis, they proposed general standards for the estimtion of X20 deficiencies in soil, either by the weight of the dry mycelial pads or by the amount of K20 removed by the mycelium from 100 grams of dry soil. These standards were as follows: Weight of K20 absorbed by Degree of potassium four pads A. giggg from 100 deficiency in the grams of soil soil Less than.1.140"grams Less than 15 mgs. Very deficient, potash required 1.140 to 2.00 grams 15 to 20 mgs. Moderately to slightly deficient, small applica- tions of I might be beneficial 200 grams or more 20 mgs. or more Not deficient. No need for potassium 1+, Mulder Procedure The technique developed by Mulder (1939, 19110) has a limited application as it is adapted to the determination only of copper or 222. or magnesium deficiencies in soil. It differs from the procedures of Niklas and of Mehlich in that the color of the mycelium and spores serves as the quantitative index, rather than the weight or chemical analysis of the mycelium. The method is based on the addition of a given quantity of soil to the nutrient solution and the comparison of the color of the mycelium with that evidenced by mycelia grown in the same manner, but without the addition of soil, and with the addition of definite increments of copper or magnesium. Gradation from the colorless mycelium grown in the deficient cultures to a dense black when adequate amounts of the two elements are present is considered to be sufficiently predictable to permit interpretation. The method apparently is reliable as the results are easily reproducible. 5. Discussion The W methods have the same faults as those inherent in use of gotgbagte; in that they are mainly only qualitative, although, as has been noted, some attempt has been made to place them on a quan- titative basis. The procedure has been used on various occasions by Truog and Fred (1933) to check the results of chemical analyses and those obtained by the Neubauer, Niklas, and Mehlich procedures. It seems apparent that the data should correlate with those obtained by chemical methods. Smith, Brown and Schlots (1932) report a comparison of the Niklas method with the Truog method in which 0.002 N sulphuric acid was used as the extracting solution. The results of this comparative 223. study are shown in figure 1K). It may be seen that the two methods are qualitatively, but only roughly quantitatively, similar, and consequently the Aspergillus technique may not be used to indicate the amounts of fertilizer required, although it will show if the soil is low enough in phosphorus or potassium to warrant further investigation. D. Cunninghamella 1. Technique The Cunnigghamella plaque method was developed by Mehlich, Fred. and Truog (1931;. 1935) for the quantitative determination of available phosphorus in soils. The procedure depends on the growth of a species of W on the surface of moist soil supplied with nutrient solutions. It is the only microbiological technique having a mathe- matical foundation. The culture dishes used were devised by Mehlich, Fred, and Truog (1935) and shown diagramatically in figure ’41. This dish, replacing the Petri dish used in earlier work (1934) permitted a more uniform growth of the fungus and has given good results with a large number of tests. The soil was screened through a 20-mesh sieve. A ten-gram sample was mixed with Just enough nutrient solution of the composition given below to saturate the soil with moisture. 22h. «S S , g NmLAI Pia-rues ""1 ’00 z Tau“ Mun-o / , E E Q. 2 . to Ill ‘3, a 2 £3 a 9: a‘. a s 4 *0: I: o. A A 1 l l 1 I l l I o L a JP L‘uPBL-PPcLPecu. 1P mam Figure 1+0. Available phosphorus six weeks after the soil ‘was placed in the greenhouse. Soil was treated and kept at 2.2-21+ percent moisture throughout the six week period. L-3 tons lime per acre. sP—lZO lbs. 20 percent superphosphate 2L-6 tons lime per acre. per acre. rP- 1000 lbs. 300 mesh Pe- 780 lbs, of 300 mesh rock phosphate rock phosphate per acre. per acre. (Smith, Brown, and Schlots, 1932) mlllllllllllllllll . ‘ mm GLAZED CAVITY g CLAY soov 877m ' Figure Ll. Diagram of the plague used in the Cunninghamella method of soil analysis. (Mehlich, Fred, and Truog, 1935) 226. Stock solution of salts Nutrient solution X Cl 20.0 gms. Glucose 25.0 gms. FeSOn.7HZO 0.2 gms. Peptone 0.5 gms. ZnSO .73 0 0.2 gms. Stock solution 5.0 mls. h 2 of salts. Héo 1000 mls. Eéo 1000 mls. * The nutrient solution must be prepared fresh every-48 hours in order to avoid.contamination.with.micro—organisms. The saturated soil is packed into the plaque and the plaque placed in a pan of distilled.water”which is loosely covered to allow aeration. The stock fungus culture was grown on malt extract agar containing 2.5 percent malt extract and 2 percent agar in.water. A.transfer was made from the stock culture at least once every four months to prevent deterioration. The soil was inoculated with a rich spore suspension prepared by adding 2 milliliters of nutrient solution to a test tube culture of the fungus which was from one to four weeks old. One drop of this suspension.was placed on the soil plaque with a wire loop. Two species of the fungus have been recommended. Cannigghamella W has been used most frequently, but for adequate growth on calcareous soils it is necessary that they be neutralized. Egggiggr hamella blggpglggag§,is recommended when unneutralized calcareous soils are used. The inoculated soil is incubated at 20 to 29° C. for a period of #8 hours when testing mineral soils, and for 72 hours when testing organic mucks or peats. 2. Interpretation I-iehlich, Fred, and Truog (193b,) carried out tests using soils of known crop response to phosphate fertilizers and found that the dia- meter of the Cunninghamella colony was a reliable index of the relation- ship between the phosphorus in the soil and crop yields. They used a modification of the Mitscherlich (1909) equation and calculated curves showing crop yields on the basis of increasing increments of phosphate fertilizer. The constant c factor was calculated from the equation c = 108(531) - log(A—y2) ’2 " ‘1 where A is the maximum growth of the fungus, y1, y2, etc. are the yields obtained from :1 and :2 amounts of added phosphate fertilizer. The constant was then used to calculate the amount of available phosphorus in the soil from the relationship, .0 _ logA - loan-3') O where y is the growth obtained without the addition of phosphate. Ten soils were studied in order to calculate the values used in the construction of the curve shown in figure ’42. Mehlich, Fred, and Truog believed that the curve obtained in this manner eliminated the unsatisfactory relationship between the diameter of fungal colony and the phosphorus present in the soil. 228. m MM. .9 DIAMETER 0F LATERAL '5' GROWH n A A a l A A L A I 00 so ‘0 7" ’° no AVAILABLE Puonuonus m sous 0‘er Figure #2. Relationship between the lateral growth of Cunninghamella and the amounts of available phosphorus in soils. ( I-Iehlich, Fred, and Truog, 193A.) The standards used for the interpretation of the data were based on a comparative study of the results with those from field plot tests. Standards must be prepared for both 9. eleg_a_n§ and Q. blakesleeana as tests have shown that in the case of the former, sufficiency of the fertilizer is indicated when growth of the colony reaches a diameter of 22 millimeters, while in the latter, sufficiency is indicated by a growth of 16 millimeters in diameter. The standards for Q. elm are as follows: Diameter of Degree of Need for phosphate colonies phosphorus deficiency fertilizer Less than 10 mm. Very deficient Great 11-15 mm. Moderately deficient Moderate 16-21 mm. Slightly deficient Slight More than 22 mm. Not deficient Done '3. Discussion .The (hmnipghamella plaque technique has been checked against the Energillug method, the Mitscherlich method, a chemical method, and field plot tests by Mehlich, Fred, and Truog (1934) using a large number of soils from widely separated areas and good correlations were obtained. Mooers (1938) studied 129 soils from nine soil types and concluded that the W and the W gigg; tests were more reliable for the determination of available potassium than was the 230. Neubauer seedling test. The Cunnigghamella technique. like other methods using micro-organisms, will not permit the determination of the actual amounts of fertilizer needed for Optimum crap yields. 3. Value of the Microbiological Methods The microbiological methods are in the same class as the seedling methods as they yield data which are primarily qualitative rather than quantitative in so far as the determination of the fertilizer require- inents of soils are concerned. Like the seedling tests, they can be 'used to determine the relative amounts of the various nutrients in the soil, and for this purpose they are pnbbably reliable. Unfortunately they are not adaptable to the determination of nitrogen, and in order to get accurate tests for potassium and phosphorus it may be necessary to use two separate techniques. As far as quantitative evaluations of soil fertility are concerned, the microbiological procedures contribute but little to the determina~ tion of soil fertility and fertilizer requirements because more reliable and more mathematically and chemically sound procedures are available. The use of micro-organisms to estimate how much nutrient a crop will remove from the soil has the same sources of error as many types of chemical analyses because there is no inherent relationship to craps ‘which.might serve as bases for correlations. Tb base an estimation of the fertilizer requirements for maximum crop yields on the amounts of nutrients available to micro-organisms is a rather hazardous procedure. 23].. VIII. FOLIAR.DIAGNOSIS 1. Introduction An increasingly pOpular procedure for determining the nutritional status of crops is based on simplified chemical analysis of the leaves. This foliar diagnosis method was deve10ped through the efforts of Lagatu and Maume at Montpellier, and of Thomas and Mack in Pennsylvania. The percentage composition of the dried plant material, or the number of milligram equivalents, was regarded by these workers as indicative of the nutritional status of crops grown on experimental field plots. The procedure implies that such data are adequate for the diagnosis of the nutritional requirements of plants of unknown history and therefore may be used to determine fertilizer requirements. Beginning in 1921+, Lagatu and Maume have published a series of papers describing the techniques and applications of foliar diagnosis. General accounts of the work have been published by Lagatu and Maume (1931+, 1937, and 1937a). The original work was carried out on the grape vine by Legatu.and Maume (192a, 192th, 1925, 1926, 1927, 1927a, 1927b. 1927c, 1927a. 19276. 1928. 1928a. 1929. 1930b. 1930c, 1937c. 1938, 1938a), but later they also applied the method in their studies of the potato (1929, 1930. 1930a, 1930e, 1931. 1932a, 1932c, 1933. 1933a. 193M. 193%. 193%. 1935a. 1935b. 1935c. 1937a). and later Maune and Dulac (193ta, 1935, 1935a. 1936, 1937. 1937a, 1938) and Hanna and Bouat (1937. 1937a), applied the procedure to wheat. Lagatu and Maume (1935, 1935a, 19350, l936e) also studied the nutritional status 232. of tobacco by foliar diagnosis. This impressive list of published papers proceeding from the work of Lagatu.and Hanna and their co- workers indicates the extensive amount of work that has been done'with this technique. Thomas (1929, 1930, 1932, 1934) and his associates in.Pennsylvania adopted the procedure and applied it in,a modified form to various vege- table and fruit craps. The general descriptions of the procedure, 1hr cluding the modifications they made, may be found in the papers of Thomas (1939, 1945), and of Thomas and Mack (191m, 19%). Thomas and his collaborators applied their techniques to corn (Thomas 1938, 1939, 1939a, 1939b, 1§39c, 1943; Thomas and Mack 19h3, l9hha), potatoes (Thomas 1936, 1937. 1938a, 1938b: Thomas and Mack, 1938, 1939c, 1939a), tomatoes (Thomas and Mack 19u0e, 19m, 1911-la), beans ('fliomas, Mack. and Cotton 19112), and peaches (Thomas, Mack, and Fagan 19148). Ibliar diagnosis was used net only for interpreting the results of fertilizer experiments, but also for elucidating the effects of disease, cultivation, irrigation, pruning, and varying weather condi- tions. Legatu.and Maume (l935d, 1938a) appear to have used the method on occasion to indicate the causes of deficiency disorders of unknown origin. The foliar diagnosis school has adapted the view that the 'nutrition' of the plant is governed not only by'the nutrients in the soil or even.by those absorbed by the plant, but by the whole complex of environmental factors. Therefore, the chemical compositions of the leaves and the 233. changes they undergo during the growing season provide the most direct information obtainable concerning the I'nutrition" of the plant. By definition, then, data obtained by foliar diagnosis cannot be readily used for the determination of fertilizer requirements for optimum yields. Ibliar diagnosis methods are of such great value for the study of the nutritional status of crops that their details will be discussed in detail in the succeeding paragraphs. 3. Theory of Foliar Diagnosis 1. Experimental Basis Foliar diagnosis developed from the belief expressed by Lagatu and Maume (l93h) that 'the first condition necessary to guide us to the proper nutrition of the plant is to know, at a given time, what the nutritional status is F Agricultural science has shown that the mineral composition of a particular species is not a fixed characteristic but varies over a'wide range according to the conditions influencing the plant growth (see Section II). Lagatu.and Maume realized the complexh ity of the problem of crop production and believed that the logical procedure was to divide soil fertility investigations into separate lines of attack, namely, physiological studies of the plant designed to determine the requirements for maximum growth, and chemical and physical studies of the soil designed to determine how a particular soil can.be improved so that better crop yields will be obtained. It 23“. was through the first of these avenues that Lagatu and Maume attacked the general problem of crOp production. The experimental facts upon.which foliar diagnosis are based have been set forth by Lagatu.and Maume and by Thomas and Mack in the publica~ tions already cited. It was assumed that two morphologically homologous leaves or leaves of the same physiological or metabolic age from plants of the same species and variety are the seat of identical physiological processes, at least when.the environment of the two is identical. Conversely, the physiological processes are different‘when the environment is different (James, 1931; James and Penston, 1933: Lagatu.and Maume, 1930, 1932c. 1933; Thomas, 1937: Thomas and Mack, 1939). The chemical composition of leaves of the same physiological age from plants growing on homOgeneous soil, but which.have received differb ent fertilizer treatments, differs. The addition of any'fertilizer effectively influencing the response of the plant always is associated with.an increase of the nutrient in question in the dried foliage (Lagatu and Maume, 1930, 1932c: Thomas, 1937; Thomas and Mack, 1939). Further, the change in the composition of the leaves is related to the yields of the plants (Lagatu and.Maume, 1932c; Gregory, 1938; Vinet and Lemesle, 1930; Thomas and Mack, 1937). The magnitude of the variations in the composition of the leaves is relatively large and is easily determined. The leaf is, therefore, a highly sensitive index of the factors affecting the nutrition of the plant, (Lagatu and Maume, 1928, 1930, 1930a, 193ue; Maume and Dulac, 235. 1935a; Thomas and Mack, 1939a, 1941, 19u3). The prOponents of foliar diagnosis base their procedures on the above facts. Data will be presented later to show whether or not these ”facts“ are based on sound experiment. 2. Definition of Terms Since the idea of foliar diagnosis is based on the work of two men, Lagatu and Maume, it is necessary to define the terms used by them in order that their work:may'be discussed. Lagatu.and Maume (1930) define foliar diagnosis as the determiner tion of the chemical composition of the leaf with respect to the dominp ant nutritive mineral ”elements", nitrogen, P205, and Rho, at the time of sampling. The foliar diagnosis for any given season consists _ of a sequence of chemical determinations indicating the sequence of the nutritional status during the growth period. The composition is expressed.as percentages of the components in the dry matter of the leaf without reference to the weight of the dry matter at each sampling or to the number of leaves sampled from each plant. All of the work by Legatu.and Hanna, and therefore the whole concept, is based on this method of expression of the composition of the plant. Bartholomew, hatts, and Janssen (1933) in a study of the method criticized the use of the percentage composition of the dry matter and preposed the use of the milligrams of the indicative com- ponents in the total amount of dry matter. But Thomas (193?) opposed 236. this saying that 'this is a misunderstanding for its (percent of dry matter) use is incorporated in the definition of foliar diagnosis” and in 1939 he insisted that there can be "no other basis of expressing re- sults than in the percentage of the dry weight of the leaf.“ Lagetu and Maume (1930) defined the Intensity of nutrition as the sum of the percentages of the “elements", N+P205+K20, in the dried foliage at the time of sampling. - The Quality of Nutrition is the ratio of the percentages of the three-”elementsI in the NPR unit at the time of sampling. This defini- tion of the quality of nutrition is the essential difference between the procedures deve10ped at Montpellier and in Pennsylvania. According to Lagatu.and Maume (1930) this term is calculated on the basis of the ratios of the percentages of the nutrients in the NPR unit in.the dry matter of the leaf. On the other hand, Thomas (1937) and Thomas and Mack (1939) express this magnitude in.milliequivalents because the chemical reactions taking place in the leaf are involved in this relar tionship, but in the case of the magnitude of the intensity of nutri- tion, consideration is given only to the presence of the “elements" in the leaf whatever their role may be! The Quantitative Index, as defined by Thomas and Mack (1939), expresses the effect of one fertilizer "element“ at the moment of sampling. It is the ratio of the amount of the 'element” contained in.the leaf of the plant growing on the experimentally fertilized plot to the amount in the leaf of a plant growing on an unfertilized plot. 237. 3. NPKUnit The NPR unit is the expression of the quantity or intensity of nutrition of the selected leaves and consists of the sum of the per- centages of the major nutrients, N+P205+K20, in the dried plant material. If x is the percentage of N, y the percentage of P20 and z 5’ the percentage of K20, then z+y+z :8 (1) Dividing equation 1 by s we get, GIN 4.. "I‘d +5. :1 (2) e where x/s, y/s, and z/s represent the proportion of the NPK unit contri- buted by N, P205, and £20 respectively. If we let x/s .-. :1, y/s : yl, and s/s : 21 then by substituting in equation 2 we have 11+ yl + 21 3 1 (3) This expresses the preportions of nitrogen, phosphoric acid and potash that are present in a unit quantity of the leaf, and conseciuently, it is an expression giving the physiological ratios between N, P20 , and 1:20 in the 'laboratory' of the leaf (Lagatu and Maume, 1934) indepen- dent of the total quantity of these nutrients in the dry matter. The NH: unit (8) is therefore a magnitude expressing the intensity of nutrition at the moment of sampling the leaf. On the other hand, the 238. composition of s gives the ratios of the major nutrients in the NPR unit or the "quality“ of nutrition. It is apparent that these two magnitudes quantity and quality are distinct but are related in the physiological processes of the leaf. The schematic diagram preposed by Lagatu and Maume (1931+) is an attempt to show more clearly what is implied in the NPR unit. If B in figure #3 represents a leaf into which nitrogen, P205, and K20 are migrating together with other materials (R), a portion of each will remain in an unelaborated form. The method of foliar diagnosis may be represented by a vessel, 1'1. of capacity 100 connected through the outlet tube M to the vessel 3'. If we now imagine a third vessel, 1'2, of unit capacity to be connected with F1, to be completely filled, and into which R, P20 , and 320 are entering in the same preport ion as they are present in II' and I , then if the N, P O , and 120 separate into layers as shown, 25 the sum of the quantities i1. yl, and z of nitrogen, phosphoric acid 1 and potash respectively, in F2 is always equal to l and this vessel may be regarded as the indicator of :1, yl, and 2 when :1 :- x/e. l 3'1 -.-.- y/s, and :1 = z/s. Lagatu and Maume (1930, 1931+, etc.) used the percentages of the components for calculating the NP! unit, but Thomas and Mack, on the other hand, considered that the 'quality factor“, that is the prepar- tions of the elements found in the leaf at the various sampling dates, is more suitably expressed as milliequivalents because this factor 239. l t t J N RQKQR. v '00 NsP‘Ofk‘o'I-R l ’. lewd-r 5 ‘— r, F, r Figure 1.3. Diagramatic representation of the concept of foliar diagnosis. (Lagatu and Name, 1931.) 2140. (quality factor) must be dependent and related to the chemical reactions involved. Thus if Mx is the percentage of nitrogen, ity the percentage of 2 may be expressed in milliequivalents by P205, and M2 the percentage of K O in the leaf, then these quantities Ex 8 1000 M1 3 By = 1020 My : and E2 = 1000 M2 17 P205 1 2 K20 If we divide the expression S : Ex+ By + Ez by S, then 24. £14.35: 1 S S S where Ex/ S, 337/ S and Ez/S represent fractions indicating in milligram equivalents the proportional parts of N, P205, and K20 in a unit quan- tity of the three major nutrients. If I: be denoted by x', then Ext x'S, S If g be denoted by y', then Eyzy's, and Ez If '3— be denoted by 2', then Ez az‘S, and 1'1. y'+ z' = 1. This equation is multiplied by 100 to avoid fractions and - lOOx'. + lOOy' + 1002' a 100, and I + Y + Z a 100 where 1001' n X, looy' = Y, and 1002' : Z. Finally, 1+- Y + Z is the NP! unit. Legatu and Maume presented their data in two forms which have been adopted by Thomas and his collaborators. 241. In the first of these methods, the percentages of each of the elements are plotted as ordinates with the sampling dates as the abscissa. This type of graph shows the increase or decrease of an element with the advancing age of the leaf. An example of this type of graph is reproduced in figure 1&5. The second method is the three component system based on the tri- linear diagram. The geometrical theory upon which this system is based is explained by Thomas and Mack (19240). An example of a simplified trilinear coordinate diagram showing the changes in the NPK unit through the period of sampling is shown in figure ’44. In this figure point 1 indicates that nitrOgen comprises 71.8 percent, P205 20.1 percent, and K20 8.08 percent of the NPK unit and point 2 indicates that nitrogen comprises 51.5 percent, P205 18.8 percent, and I20 #.6 percent. This method is a derivitive of the first and shows the changes in composition with respect to nitrogen, P205, and £20, re- sulting from any particular treatment. 0. Technique of Application The effects 'of applications of the major fertilizer components on the composition of the grape vine throughout the season were studied by Lagatu and Maume in a series of papers (1921+, 192%., 1925). Their work indicated that the absolute and relative amounts of the three major bases CaO, MgO and I20 as well as of nitrogen and P205 remained in the same order in leaves sampled at monthly intervals from May 18 s 4 m a A .. 1 1,. 1.;DD D D : .DDD D DD.D . D+.D DJ . D . D. 11D D DDDD DDDD. D D D D D D D D .D D .. D D D D D 4 4 4 4 4 4 4 4 4 4 4 4 4 4D4 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4 D D4 4 4 4 4 4 4 4 4 4 4 4 4D4D 4D4D4D4 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D D D D D D D D D D D D D D D , mommaa.»”a.“«a.“ana.»a..ao...messs.fia a»..awsfiwfisfi.an».Haaaaaaafifi.«memes».as.“as.N«newsman«News». ".4 4D D4D D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D404D4D4D4D4D4D4D4D4D4D D D D D D D D 0 D 4 4 4 4 4 4 4 4 4 4 4 4 4D4 D4D4 D4D4 D4D4 D4 D4D4 D D4D4 D4 D4 D4D4D4D4D4 D4D4D4D4D4 D4D4D4D4~3D4D4D4D4D4 4D4D4D4D4 . 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D D4D 9 D40 0 0 0 0 0 0 0 0 0 0 4 4 4 4 4 4 4 4 4 4 4 4 4‘4 4D4D4D4 4 . D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D D D4D D D D D D D D t D 4D4D4D4D ' D4D4D4D4D4D4D4D4D4D4D4D4D4 4D4D4D4F4D4 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D404D D4D D D D D D D D D D D D D D 4 4 4 4 4 4 4 4 4 4 4 4 4D4D4D4 4 4D4D4D4D4D4D4D4 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D D D D D D D D D D D D D D 4D4D4D4D4D4D4D4 4D4D4D4D4D4D4D4D4 D4D4D4D4D4D D404D404 4D4D4D4D4D4D4D4D4D4D4D4D4D404D40 0 4 4 4 4 4 4 4 4 4 4 4D4 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D 4D4D4D4D D4 4D4D4D4D4D4D4%D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D 0N0 D404D4D4D4D4D4D4D4D4D4D4D4D D40 0 0 0 0 0 0 0 0 0 0. 0 0 4 0D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4 4D4D4D4D4D4D4D 4 4D4D4D D4 4D4D4D4D4D4D D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D 5D D4D4D4D4D4D4D4D4D4D4D4D4D4D D D D.D D D D D D D D D D D 4 4 4 4 4 D4 4 . 4 D4D4a 4D4D4 D4D 45 g4 4 4 4D4D4 4 4 4 4 4D4 4 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D D4D D D4D D D D .. . 4D4D4D4D4D4D4D4D4D4 .. D4D4D4D4D4D4D4D4D4D4D4D4 4 4D4D4D D D D D 4 4D0D0hD0DAWWD4D0D0D4D4D4D4D4D4D4D4D4D4D4 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4m4D4D4D4DmeD4D4D4D4D4DflD4D 4 4D4D4D4D4D4D4D4D4 I D4D4D4D4D4D4D4D4D4D4D454D4D4 4D4D4D4D4 4D4D4D4D4D4D4D4D 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4D4D4D4D4D4D4D4D4D 4D4D4 D4D4D§4D4V§4D4D Do D DOD D4D4D4D4D4D4D0D4D4D4D4D4D4D4D D4D D4D D4D4D D D D D D D D D D D D 4D4uofi4 D4 4D4D4D4D§4D4 4D4D4D4D4D4D4D4 D4D4D4D4 4D4D4D4D4D4D4D4D4D4D4 D4D4QD4 . 4 4 4 4 4 4 4 4 4 4 4 4D4D4D4D4D4D4 4 4 4 4 4 4 4 4 4D4D4D4D4D4D4D4D4D4D4 4D4D4 D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4 4D4D4D4D4D4D4D D D D D D D D D D D D fisfisfihfiamaasausage”. fiaumwfifimaam«seesaw.”amass.”ems." am warms...assesses.» was”... 4D4D4D4D4D4§D4D4D D4 D4D4D4 4D D D D D D D D D 4 4 4 4 4 4 4 4 4 4 4 4 4D4D4 4g 4D4D4D4‘D4D‘4 ‘4DaD4 4D4D4D4D4D4D4D4D4! 4D4D4 4D ; D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D 4 40.4040 040 0 0 0 040 0 0 54. 0D» «Dvfi D4 D00 D4D4D4D4D4D4D4D4D4 4D§4 4D4D4QD4D4D4D4DN%D4D4D D4D4D4D4D4 . 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D D40 0 0 0 .0 D 0 0 0 0 0 0 0 0D4Dfi 4D4D4D4 Q4. ‘ 4D§4D4 D4 D4D4D§4Dfifl4D4D4 4 4D4D4D4D4D4D4 D4D4D4D4 4D4 4 . .. 54D4D4D4D4D4D4D45D4D4D4D4D4D4D4D4D4D4D404D40 D D D D D4D D D D D D D D ”“3. . 4 ‘4 .4 444444 4 44 B4D§ DD4 4 D§§§D r4D4§D4D4§D4D4DOD4DON¢D (D. D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D D4D D D D D D D D! E D 04 .. VD 4 D D0H4D % D4 4D D4 4D4§4D4D4D4D4D4§D4D4D4D4QD4M4DNa."4D4D4D4D4§D4D4D4D4D4D4D4D4D4D4D4D4D4 4D4D4D4D4D4DEP DfivfiflgDflfiflh 4 4D #4 4 4 4.... 5955.457 D4 4D 9 o4 .04 4D54 D4D4D4D4a 4“? 4 4 4 4 4 4 4 4 a; uD4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D4 D4D4D4D4D4D4 4 4D4D§D Ds hogD D D D oDfiDWDflD4DN 4&4?4D4D1D1D§4D4Da04D4D4D4D404D4D4D4~D4D4D D 4D4D4‘ D4‘D4D4D D4D4D4D4D4D , 4 4 4 4 4 4 4 4 4 4 4D4D4D4D4D D4D4D4D4 4D4D4D4D4D4D§4D4 40W4hr44 4D4DH0N0M0N D04 D0D0D04DM¢DD0DD 0D . ..rm,0D0D0D0D4D4D4D4D4D4D4D4D4D0D0D4D4D4E4D4D4D4D4 44 D§4§F§E4D§ 4D4D D4D4D4D4D4D4D4D4D4D D4D4D4 4D4D‘I4D45»4D4D4D4D4D4D4D4D4D4Q §4D4D4D4D4D4B4 {.5 .1D... D4D§4§D4D4D§4§ a .5 55:53. as .....m...ws....mséfi .. sausafifisas sgsms 53?... 4D4 D4D4D4D4D4D4D4DD D4 4 .n. % - 4 :4 44 . D4D4D4D4 D .,D4D . 4 E 5D D as, .fiawsamsé shame»... massageéww magma»? Emma s. 4D4 4D4D D4D4D454D§4§D .4D4 55% . ”D4D .. 4D 5 g 95 $4 . 4’ \, lair/h: C. J. / 4 IV... ‘ $4 D4D4D‘egg - 4 §4Dg DOD ED4’4hDdg4 4 .. , 4D ED4‘QD4Q E‘DED x .. . . fine... AW. 5.5 as 8 . v 4 4! eDEDgéi ‘4 D4, . , EfDD 4D4D4D4D4D4D§40W% . .. ._ ,. H. ahufi m... 4D4§4§D4 4 Q5453: .4an .. __ __ . 34D _. 4aD D 454; u. , .. . 21.2. shoring the changes in the NPR-unit through the period Figure 1.1.. Simplified trilinear coordinate diagram of sampling. . 4D4D4D4D4D4D€4D4D §§D4D4D4D4D .D4D4D4‘D a 4D4D4D4D4D4D4D4D4D4D4D4D4D4D4D 4D4 ¢4D4D4D4D4D4D4D4D4D4D4D4§D4D4D4 4D4D4D4D4D4D4D4D4D4D4D4D4 4D4 4D4D4D4D4D4h D4DV D4 D4D4D4D4D D4D4QD4D4D4 D4D4D4D 4D4D4D4D4D4 D4D4 4D4D4D4D 4D4D4D4~flD D4D4 4D4~D4D4D4D4D4D4D4D4D4 4D4D4D4D4 4D4D4‘D4 ‘4D4D4D . D4D4D4D4 4D4D4D4D4D4D4D4D4 gD4D4D4D4D4D4 4D4D4D4D4D4D4 4D4 ., 4D4. 4 4 4 4 243. to September 18. They concluded that such data could serve for recog- nizing the insufficiency or excess of a nutrient. In a continuation of the study they explored the possibility of selecting a specific organ which.wou1d represent the “intensity” and "quality" of nutrition. In one experiment, they studied the effect of pruning three of the nine bunches of grapes from the vines growing under identical conditions. From the data shown in table they concluded that there was remarkable similarity between similar organs, leaves in this case. On the basis of studies of this nature, the pro- ponents of foliar diagnosis have selected specific tissues for analysis. For grapes, they use the two leaves at the base of the fruiting struc- ture (Lagatu and Maume 1930; Thomas 1937): for tobacco, the second and third leaf from the base of the stalk:(Lagatu and Maume 1936s): for wheat, the tOps (Maume and Dulac 1934); for tomato, the fifth leaf from the base (Thomas and Mack 1943) or the fourth leaf from the base (Thomas, Mack and Bahn 19hh): for beans the basal leaves (Thomas, Mack:and Cotton 19b2): and for peaches, the leaves from the middle of the ter- minal growth (Thomas, Mack and lagan 19n8). The necessity of the precise definition of the sample is shown in the'work of Lagatu.and.Maume (1930d) on the grape in.a study of the effect of different fertilizations. ‘Using the variety Grandpfloir they sampled at various periods during the growing season the first two, the fourth and fifth, and the seventh and eighth leaves from the base of the fruiting stalk of plants from an.unfertilized and from a fertilized plot. The percentages of CaO and nitrogen in the dry matter are given in figure ’45. While each type of sample follows the same relative course of changes through the season, the older leaf represents the composition which will be the endpoint of the younger leaf. This also has been shown to be true for tobacco by Lagatu and Maume (1935, 19356). Experimental work by Thomas and Mac}: (1937) on potatoes will be described as an example of the details of the application of the foliar diagnosis procedure. This work was carried out on material from the vegetable Fertility plots at the Pennsylvania Agricultural Experimental Station during the year 1935. The plot treatments and yields are given in table 33 and the table 31+ gives the compositions of the fourth and fifth leaves expressed in percentage of nitrogen, P20 , and X20 'in 5 the dried foliage, as milligram equivalents, and as fractions of the NP! unit. To illustrate the mathematics involved in the calculation of the points of the NH! unit, data from the nitrogen plot sampled July 7 will be taken as an example. The chemical analyses show that N :- 5.12 percent, P205 3 0.11014 percent, and 120 g 3.97 percent of the dry leaf material. The intensity of nutrition is therefore 5.12 + 0.14014 + 3.97 : 9,181+ for the leaves sampled on July 7. To evaluate the M unit the percentage values are converted into milligram equivalents of nitrogen, P205 and K20 as follows: 21+5 . zo- .- 1 ___.... : ‘ , ------- t."‘ E III . ' t- i g g E g 0' I"! . J j '0’ / ' I " - /" I d‘ ,l" / , u o \ \ \ \ E O 5 mi fitigtfifi :fit‘cg‘vigq? Cues TREATED Figure AS. The percentages of nitrogen, and 0&0 in the dry matter of the first and second leaf from the base of the fruiting stalk (C), the fourth and fifth leaf (B), and the seventh and eighth leaf (A) on a plot treated with nitrogen fertilizer and an untreated plot at four dates during the growing season. (1. June 13, 2. July 10, 3. July 27, and A. August 7) Plants studied were the grape. (Lagatu and Maume, 1930d) 246. Table 33. Plot numbers, fertilizer treatments, and yields of potatoes at the Pennsylvania Agriculture Experiment Station 1935. (Thomas, 1937) Plot Fertilizer Fertilizer 'N, P O and Symbol Yields number applied applied per 0 gq‘giva- used per plot plot in lbs. lents, in in lbs. _ pounds 2 RENOB 4.0 0.6 N‘ 109 3 Superphosphate 6.25 1.0 P 114 ’4 Potassium 1.666 0.8 K 155 chloride 6 name 4.0 0.6 HP 124 Supe9phosphate 6.25 1.0 7 naN03 4.0 0.6 NE 163 K01 1.666 0.8 8 Superpho sphate 6. 25 1. 0 PK 1148 KC]- 1.666 0.8 '10 None 4.0 0.6 m 162 Supegphosphate 6.25 1.0 101 1.666 0.8 247. 4N eeswsq a he seems4 n “mN sHse N “a sHse H ma.aN am.s N.aw. n.nnm m.Nm m.mH . N.eNN mm.e Nmm.o 4H.m a Ne.mN mm.m N.mh m.man o.ooH m.oN m.mmN 0a.: Nws.o Nm.n m mw.mN Hm.m m.oa m.oam e.Nm m.HN o.aaN sm.e mom.o mm.m N mmoN in: «.8 90% TSH are mfimm one same 3.: H was mm.Nm am.m m.Hm m.mNm n.50H N.mH e.H0N so.m one.o Nm.N a HH.Nm He.m m.He 5.3mm «.mHH m.oN m.eHN Nm.m ome.o mo.m m 4N.om we.m o.em m.eam «.mHH w.HN m.mmN Nn.m OHm.o mm.n N ao.eN mm.m w.mw N.mms m.maH e.mN m.eNm on.a ONw.o mm.: H mm em.Hm mo.s m.mw m.mhn N.mHH m.sH a.NmN we.m Nmn.o mN.m s Ho.mN ha.s N.aw m.Noe e.NHH N.mH m.moN mN.m Nne.e on.a n we.aN so.m n.am w.aNe m.mHH m.HN a.amN mm.m oHn.o no.: N mm.wN sa.n a.mm m.st H.HeH w.mH H.amm mm.m meme.o oH.m H as wN.m mm.m m.mm a.msN H.nN a.mH m.mmN mo.H omn.o on.a a mH.m Nw.h «.mm p.0Hm N.mN m.ON m.ehN mH.H eme.o Ha.n n aa.m on.a s.mw H.Hmm n.Nm m.NN n.maN Nm.H Nam.o am.m N mo.NH mm.m m.Nm m.mss Hm.mm m.eN o.ahm mm.N oem.o sH.m H mm NN.mn mm.e m.Hm m.on m.mOH a.mH m.mmH oo.m pan.o we.N a on.mN mm.m a.mm m.mhm n.eOH m.mH o.NeN Hm.s sme.o em.n n mo.aN mm.m m.ah N.eHm H.HHH w.NN n.maN NN.m o:m.o am.n N sm.aN ma.n «.me m.Hme h.emH m.aH H.mNm Nn.h sN.s Hm.s H m sa.m am.a s.Nm w.esN m.oN o.mH N.mmH HH.H one.o ma.N a ma.m no.5 a.Nm 5.0mm m.aN m.HN m.Hmm mN.H on.o 3N.m n m.HH mN.a n.Hm m.mom m.sm o.NN o.aeN sm.H HNm.o wa.n N mH.NH NN.m o.Hm m.Noe m.Hm o.mN n.mNm Ha.N omn.o am.a H m mo.NH hh.e n.mm N.NHm a.am m.sH «.mmN aa.H esm.o sm.n a m.HH mm.m m.Nm o.mmm m.Ne m.mH n.mmN Ho.N Nae.o mH.s m ma.mH ms.m e.ow H.Nem H.sm w.NN s.mHn sm.N emm.o Ne.s N mo.mH mw.m N.ma n.ame m.em H.5H m.mem am.n eos.o - NH.m H a m Mel Am. 08 on as ea m\en00H m\stOH m\smooH m e an a m: N .H N as an we own “we a one o .H a shoe oess_umz no aoHeHoossoo senoHsDHseoHHHH: omoHHoe soHee mo soHsHeoesoo ease lessee Can... .9269: .925 4.3 can and mpgaggoo figmHHHHE .omcHHoH ooze on» an ONH and .homm .am he mayo» HS 1 I I I ll!" 1' I? 1 'I“ i ‘ $le it] . ‘1‘ ‘ 4 \ i ‘ J1: .- rlllll‘llll .Iillullvilnli [111 i i. w i Mi} . . v o 4 - a o v o u 4 s a e . e a 4 n e e A c I 4 e 4 v e H _ D e u e 4 v o u D u . 4 . . . c a u e e v n 4 . . . v e a u . n e . 4 a o a . 4 A 4 4 a e u e 4 4 e 4 I O D D c ‘ C I D C I n . 4 e v . s e D u e u p p e 4 D D D o . n . . . e r 4 e e A . . . 4 . . e 4 I e I 4 u . . 4 s 4 9 n 4 4 A D 4 D t 4 4 I e u 4 I 4 4 D s D v D v u . . 4 . s e u 4 .. A - 4 - 4 . o 4 . v e u u 4 . a o p 4 n . 6 4 . . . 4 4 e e 4 D n 1 . 4 . . . 4 n o e e 4 I 4 I I 4 v I V e . . . . o e D e e 4 . e u D a 4 4 e v e . . . . c 4 h . v e o u . o D . v . D 4 v u . a o n e. s v e 4 I . u . v . . o . o a . 2&8. Ex . 1000 x .0711; x 5.12 = 365.568 ME of E E2 4 1000 x .0213 x 3.97’: 84.561 ME of Rio my 3 1000 x .ou23 1: mu 3 17.089 ME of P205 5 a 365. 568+17.089 + 8a.561 = #67318 Therefore x = 100 .33. _-. 78.2144 8 E r z 100 _E . 3.658 S E S S : Total : 100 These values are the three points required for plotting on the trilinear coordinate graph. The relationships between the percentage composition of the dried leaves, and the increasing age, in.plants growing on variously treated plots are shown in the data plotted in figure A6. This type of presen- tation shows. for instance, that the nitrogen fertilized plots actually contained more nitrogen than the untreated plots, and that the per- centages of nitrogen in the tissue decreased with the advancing age of tissue. It shows that both the phosphorus and potassium content was increased by the added qusand Kéo fertilizers respectively. It is apparent that the periodic analysis of the leaves from plants growb ing on plots treated.with.different fertilizers reflected the effects of these fertilizers. The positionsxand forms of the curves indicate the changes in the nutritional status of the plants with respect to 2&9. gen. g" Day ”A7113 0 v o "/ , K Fluent? T nzA'r HIM-r I Figure 1.6. Periodic analysis of the fourth. and fifth leaves of potato plants growing on differentially fertilized plots. The percentages of the major nutrients in the dried tissues are plotted as the ordinate and the dates of sampling as the abscissa. (Thames, 1937) nitrogen. phosphorus and potassium. The steepness of a curve repre- senting a particular nutrient indicates the relative demand for the 'nutrient in relation to its supply. Thomas (1937) warns that the interpretation of the urn unit can be confused by three things: an increase or decrease in the intensity of nutrition: a changetin the composition of the NPK unit; or a change in both simultaneously. The intensities of nutrition from the data of Thomas (1937) are shown in figure 4? for plants growing on.variously fertilized plots. The intensities are the ordinate and the dates of sampling are the abscissa. This presentation indicates the differenr ces due to the age of plant and the fertilizer treatments. The ouality of the nutrition, as shown in the trilinear coordinates (figure #8), indicates how the NPK unit is displaced from one sampling period to another. It is also of value to consider the relationship between the ins tensity of nutrition (n+r205-I- x20) and the mean m unit in relation to the yield. These values are presented in table 35. Data.presented in this form indicate the ratio of the elements in the leaves which.is favorable to-optimum crop yields and.which, as will be shown later, has been used by Lagatu.and Maume (1937) to estimate fertilizer re- quirements. 251. 3 s». 3' u... » If a. ’ NK z PK -7 NPK ’3 ’41“ 3.5» N + 21*. NP :5 1 £9 ii) .30 “0’ 5° DAys Figure #7. Relationships of the intensities of nutrition of plants growing on differentially fertilized plots. (Thomas, 1937) 252. Figure 1.8. A, relative position of the NPR-unit. B, course of nutrition of the nitrogen, NPK, and potassium plots. C, Course of nutrition of the nitrogen, NK, and NPK plots. D, course of nutrition of the PK, NP, NPK, and phosphorus plots. (Thomas, 1937) 0‘ .3 t.‘ 0.. n I.) 5. 4 a , . I no. L. “I: . .“. 4"“! . '1 ’ .c 1 t 4L...-. g... “Radix: use 7.. ... O 3' ..~l."n 4.x s .o4L4.n\- r... t -. o,..A.(11‘.M» cue-.0! . 0 9|! IV 0 n. . ‘ . i: I \Hfl nor,“ .4 ‘9'“ ,.tr.'. .Wflil.» 253. mm.ee mos em.e ae.na Hm.s NH.Hm z m mm.mh rHH mw.m mm.oa m~.e on.am m ao.me sea oa.m on.a m~.w mH.am as N om.om and on.a mo.Hm sw.m mm.mm ma mo.mm mmH ma.m on.mm mm.n Hm.mm u a mm.mm wed on.a mH.mN mH.m me.mw use ooH mma sm.oH Hm.mm mm.n mm.mm M2 can mama n 08 0.3% Hum 93% ado Ohom co." u N2 no.3. mmozaamn .30.“ mo 305 - woos .393 m go?» .39 .mpHA “83.392“ Amwoaapasu .30.“ Ho H305 spoons moose ooaeoaem means» we seaenoeaH eaan.umm one: Ahmed .ansogav Hpom pd “83733" mo .3352: some on» was 343 E does on» no non—Sou .339 6333.3.“ ham-90.3.9» do uooagom no dang on» on mouse wagon-on .mn edema 25#. D. Applications 1. Effects of Disease In a few instances. foliar diagnosis has been used for the study of the effects of various diseases on the nutritional state of plants. For example, Lagatu and Maume (1935b) working with tobacco showed the effect of the wild fire disease. The results indicated that the dis- eased plants contained relatively higher percentages of nitrogen.and lower percentages of potash in the NPK unit than did the healthy plants. It was further noted that apparently healthy plants growing in the diseased plantation.were intermediate between the two others. It was concluded that these plants, although not showing symptoms of the disease. were approaching the diseased condition. Foliar diagnosis, therefore, was thought to detect plant diseases at an early stage of their development. Thomas, Mack and Pagan (19h8) also used foliar diagnosis to study the effect of bacterial leaf spot on the nutrition of peach trees. 2. Effects of Cultural Methods Ibliar diagnosis has been used to study the effects of cultural practices on the nutritional status of the crop. For instance, Legatu and Maume (1935c) studied the effect of farmyard manure on the avail- ability of potash to potatoes. They were able to show that luxury consumption of potash occurred, but the cr0p was able to utilize the excess potassium without incurring decreased yield. Lagatu.and.Maume 255- (l930c) also studied the influence of the time of application of fer- tilizer, of pruning of the bushes (1933. 1933b). and of tillage (l93hb) on the nutritional status of the crop. This latter work showed that the tillage practices increased the intensity of nutrition and caused a variation in the nutritional balance, or '"quality" of nutrition, in the crop as indicated in figure #9. The effect of tillage was a rather sharp increase in the absorption of phosphorus and a subsequent increase of P205 in the NPK unit. Both of these effects are evident in figure #9. Thomas and his collaborators at Pennsylvania also used the foliar diagnosis technique to study the effect of cultural treatments on the nutritional status of crops. This work included the study of the effect of applications of manure on tomatoes (Thomas and Mack 19hda), and the effect of irrigation practices on beans (Thomas, Mack, and Cotton 19h2), and the effect of different methods of fertilizer place- ment on corn (Thomas, Mack, and Rahn l9hh). Foliar diagnosis is an effective technique for the study of the effects of cultural practices on the nutritional status of a crop, and ‘while it has been applied only to the major nutrients. theoretically it should be possible to apply the technique also to the minor elements, for certainly they influence the nutritional status of the plant. 3. Effects of Nutrient Carriers .Although field plot experiments have been used.almost since their inception to study the effect of different forms of nutrients on crOp (to U1 0 \ N 1" __ 113." -----flu.lo so V ‘ ‘. - -. 0 gr: K % "4".0.‘ f “to MONT” Figure 1.9. Effect or tillage on the intensity of nutrition and the NPR-unit in the grape at Montpellier during 1933. (Lagatu and Mame, 193th) 257. yield, it was not until Legatu.and Manna became interested in 1927 that attention was paid to the effects of the different forms of a fertilizer on the nutritional status of the plant. These authors (l927, l927d, l927e) reported the effect of the forms of the major nutrients on the grape vine. Lagatu and Maume (1936f) later reported ‘ the effect of 'scories de dephosphoration", superphosphate, and basi- phosphate on the NPK unit, that is on the ”quality" of nutrition of tobacco. Their data showed that the three forms of phosphate exerted different effects on the nutrition of the tobacco plants. There was not a very great effect on the phosphorus content itself, but there was a significant effect on the nitrogen content for the nitrogen fraction of the NPK unit was greatest with basiphosphate, lowest with 'scories” and intermediate with superphosphate. Thomas and Mack (1939c) used foliar diagnosis to study the effects of different forms of nitrOgen on the nutrient content of corn. and later they (l9hha) continued the study with potatoes. In the last study, the nitrogen carriers were Ca(N03)2. nan03, and (NHh)ZSO# or tankage, and, as with the work reported by Lagatu.and Maume on.phos- phorus carriers, the different carriers of nitrogen did not signifi- cantly affect the percentages of nitrogen in the dry matter as may be seen in figure 50. The greatest effect was on the relative amount of potash in the IPK unit as may be seen in figure 51. This figure shows that as the percentage of I20 in the NPK unit decreased, the yield of the crop increased. These results show that the field plot trials used ream Puts; PLOTS}! mus flora: M.~0“ Nn'K NC‘PK "V‘PK NI‘PK 15.3 1%.; 18.-I 68.2 58.0 Ir g] .0600... P‘O‘ o 2 ’— g cot ' ¢ o R ‘ ‘ 5 \ \ ‘ \ \ \ \ o .nO Iv ‘ \ “ \ i; b \ ‘\ n \o’ \.. 6' a: ‘.O n i: g to 3Q! 13*] $3.! ‘3 TI 33’ SAMPLING Famous Figure 50.’ Percentages of nitrogen, P205, and K20 in dried foliage of potatoes from.differentially fertilized plots on four sampling dates: l-July 6, 2-July 26, 3—August 6, and h—August 26. (Thomas and Mack, l9hha) 259. ' '— v v v v ‘.‘ I” A) 2 I & P‘O‘ “‘0 “a, Symbol Treatment Intensity Yield N PK 13.23 78.1. E18 Mggure 11.99 75.3 Na Huai’K 10. 57 7L.O Tk Nq-kPK 9.77 68.2 Figure 51. Intensities of nutrition (table) and NPK equilibrium (graph) in potato leaves during the growth period from plots receiving different fertilizer treatments. The symbols Ncaa Nna: Ntk: and Ném_designate respectively calcium nitrate, sodium nitrate, tankage, and ammonium sulphate as nitrogen carriers. (Thomas and.Mack, 19hha) 260. previously to study the effect of different carriers of a nutrient on yield may have been misleading since in both of the experiments des- cribed the carrier had little effect on the absorption of the nutri- ent itself but influenced the crop yield by affecting the absorption of other nutrients. b. Effects of Climate The effects of climate on the nutrition of plants have been studied by foliar diagnosis and data from the work of Lagatu.and Maume (1928) obtained with the grape vine, and from the work of Thomas and Mack (l9hl) obtained with tomatoes, have been presented in Section II. Maume and.Dulac (1935, 1935a) carried out an extensive study of the effects of climate on the nutritional status of wheat. Legatu.and Maume studied the effects of climate on potatoes (1930a,'193ha, 1935b) and on.grape (1936c), and Thomas and Mack (l9b3) studied its effect on the nutritional status of the corn.plant. Legatu.and maume (193ha) state that it is difficult to determine the fertilizer requirements of crops if the climate varies greatly from season to season, and any recommendations made under such.varying conditions are purely speculative. 5. Fertilizer Requirements a. Introduction While the preceeding paragraphs have described the use of foliar diagnosis to study the effects of disease, cultural practices, irrigation, nan-1+. pruning,‘weather, and nutrient carriers, the technique may be used as a tool for the interpretation of the results of fertilizer field plot trials, although but few examples of this practical application have been published. Many believe that data obtained by foliar diagnosis do not supply information concerning how the mineral composition of the crop may be altered to giye higher yields. However, in the works of Legatu and Maume and of Thomas there is evidence that foliar diag- nosis may be used for determining the amount and kinds of fertilizer required by crops, and two examples of this application.will be des- cribed. b. Montpellier Technique Lagatu and Maume (1936c) studied the foliar diagnosis of the grape over the period from 1921 to 1928 under twenty different ferti- lizer treatments and found that the Optimum composition of the RP! unit for high yield is N': #1, 2295 z 8, and Ego : 51 in all stages of growth. Similarly, for the potato they report (1937c) that these values should be N': 38, P20 = h, and 320 3 58. Using these ratios they believed that recommendations for the correction of fertilizer deficiencies could be made. Such a recommendation of course, is essentially one for correcting nutritional state of the soil through several seasons, although the authors have admitted that this was difficult. A.mathematical method for interpreting the data in order to permit 262. recommendation of fertilizer applications was described (1937). If x, y, and z represent the percentages of N, P295, and K20 in.the dry matter from the fertilized plot: and if a, b, and c represent the per- centages of N, P205, and K20 obtained from the unfertilized plot; and if a0, b0, co represent the percentages of N} P295, and £20 in the plants representing the Optimum nutritional levels, then z+y+z : s a+b+c : t ao+bo+co : t o Carrying out the calculations as described previously, X+Y+Z : 100 A+ZB +0 100 A04. 30"” co 3 100 These equations are used to show the improvement resulting from fer- tilization and also how close this improvement has come to the ideal NP! units The effect of the fertilizer on the composition of the NP! unit is measured by the ratios ' Realized difference of 19A. of YeB of G Desired difference Ab-A. 15-3 00-0 These fractions are interpreted in the following manner: If the fertilizer has improved the nutritional status of the crop, each of the three ratios will be greater than 1. A.negative ratio in- dicates that a nutritional factor has exerted a detrimental effect on the composition of the NPK unit. Positive ratios, but less than 1, indicate the fraction of the desired nutritional state attained by use of the fertilizer. Positive ratios greater than 1 indicate to what extent the fertilizer used was an overestimation of that required. The effects of the fertilizer treatments on the intensity of nutri- e - t to - t manner as the effects of the fertilizer on the NPK unit. tion are measured by the ratio, , and are interpreted in the same The data used to explain the technique of interpretation (1937c) were obtained from an experiment with the grape variety Vigne de Mauguio carried on from 1929 to 1931+. Potassium was added annually as potassium chloride at a rate to give 600 kilograms of I20 per hectar. This treatment increased the crop yields during the six years of the experiment. The data presented in tables 36 and 3? permit comparisons between the fertilized and unfertilized plots. comparing the O and Z values with the optimum unit, 41:8:51, it is evident that in 1929 the percentage of K20 in the foliage and in the NP! unit was lower than desired in both the unfertilized and fertilized crepe. The addition of 600 kilograms per hectar of I 0 only partially corrected 2 this deficiency, as all values were positive for the Z " 0 ratios, I Go " c during the first year. But the percentage of the K20 in the tissue was only 10 percent of‘what it should have been. As the treatment was continued during successive seasons, the nutritional status improved to such an extent that in 19314 (table 3h) the percentage of potassium in the m unit was approximately 75 percent of the Optimum value and the percentage of phosphorus level in the NPK Imit had decreased to a point 264. Table 36. Analyses of leaves of the grape, Vigne de Maugio, grown in 1929 on unfertilized and 0 fertilized plots, expressed as percent- ages Of dry matter and in units showing the extent to which the fertilized crop approached the optimum nutritional status for high yields, (Lagatu and MaumeP193'Zg) Date of Percent Percent Percent Intensity Composition of m N P 0 0 of unit ”Op 2 5 K2 nutrition Plot D Check plot no treatment a b c t A B 0 May 10 4.49 1.22 2.04 7.75 57.94 15.54 26.32 June 15 2.27 0.78 0.82 4.37 63.39 17.85 18.76 July 15 2.33 0.47 0.61 3.41 68.33 13.78 17.89 Sept. 9 1.86 0.33 0.36 2.55 72.94 12.94 14.12 Plot it Treated plot 600 Kgs./hectar of K20 as 1:01 I ; x y z s I I 2 May ~ 10 f 0.54 1.08 2.27 7.89 57.54 13.69 28.77 June 15 2.91 0.68 1.12 4.71 61.78 14.44 23.78 July.15 2.57 0.46 0.77 3.80 67.83 12.11 20.66 Sept. 9 1085 0.29 00% 2.60 71.15 11.16 17.69 1.1 ms z-c s-t 10 Au 30-3. 00-0 to-t Kay 10 -o.40 -2.05 +2.05 +0.14 -16.97 -7.74 +24.68 -o.12 June 15 "1.61 -30“). +5.03 +003“ ”22039 -9085 +3202“ -3026 July 15 -0.70 -1.67 +2.37 +0.39 -27.33 -5.98 +33.11 +4.22 Sept. 9 -1.79 -1.78 +3.57 +0.06 -31.94 4.95 +36.88 +5.08 X -A Y -B Z -0 s -t 10 .1 so -3 co -0 t0 -t May 10 +0.024 +0.265 +0.999 -1.17 June 15 + 0.072 +0.346 +0.156 +0.10 July 15 +0.026 +0.289 ‘ +0.072 +0.09 Sept. 9 +0.056 +0.360 +0.09? +0.01 Values required for Optimum nutritional status Aomocoozzhlzazfl t0 3 7O 63 Table 37. in 1934 on unfertilized and centages of dry matter and in NPK units showing the extent to which the treated plot approached the Optimum nutritional status for high 265. Analyses of the leaves of the grape, Vigne de Maugio, grown 0 fertilized plots, expressed as per- yields. (Lagatu and Maume, 1937c) Date of Percent Percent Percent Intensity Composition of RP! crop N P205 K20 of unit nutrition Plot D. Check plot no treatment a b c t - A D 0 May 11 4.59 1.10 1.72 7.41 61.94 14.85 23.21 June 14 2.07 0.77 0.99 3.81 54.05 20.10 25.85 July 17 1.72 0.56 0.99 3.27 52.60 17.13 30.27 Sept. 5 1.59 0.42 0.88 2.89 55.02 14.53 30.45 Plot r. Annual treatment since 1929 of 600 kgs./hectar of 120 as K01 x y z 3 II I 2 May 11 4 80 1.12 2.02 7.49 60.45 14.11 25.44 June 14 2.16 0.51 1.84 4.51 47.89 11.31 40.80 July 17 1.70 0.38 1.80 3.88 43.82 9.79 46.39 Sept. 5 1.56 0.43 1.71 3.70 42.16 11.62 ~ 46.22 LA r-s z-O s-t no-1 30-3 00-0 to-t May 11 -1.49 -0.74 1- 2.23 +0.08 -20.94 - 6.85 +27.79 +0.22 July 17 -8.78 -7.34 +16.12 +0.61 -11.60 - 9.13 +20.73 +4.36 Sept. 5 -12.86 -2.91 +15.77 +1.41 -l4.02 - 6.53 +20.55 +4.74 1 -1 Y -D z -0 e -t 10-1 30-3 00-0 to-t May 11 -+0.071 +0.108 +0.080 +0.36 July 17 +0.75? +0.804 +0.778 +0.14 Sept. 5 +0.91? +0.446 +0.76? +0.30 Ialues required for optimum nutritional status Aotaoicoz 341:8:51 t0 = 7063 266. close to the Optimum value. It appears from these experiments that foliar diagnosis may be used to determine the fertilizer requirements and that the investigator also may estimate how close he has come to producing the Optimum nutritional status, and consequently yield, of the crop. C. Pennsylvania Technique Thomas and Mack (1939) have suggested a technique of interpreting data obtained by foliar diagnosis by which fertilizer recommendations can be estimated intelligently. It is essential, of course, that the fertilizer be hmown.to influence the 'intensity" of the “quality“ of ‘ the nutrition of the crop. The technique is based on the assumption that there are no differences existing in the influence of external factors other than those related to the fertilizer treatments. The data in table 38, obtained from a fertilizer experiment'with corn. may be used to illustrate Thomas and Mach's application. The data obtained from plot 28 show how the composition of the crap can approach that found in the tissue of the highest yielding plot (22). The data show that the percentage of nitrogen in the NPR unit from plot 28 was lower than that from plot 22 throughout the season. The mean value of the percentage of P’ZO5 in the NP! unit from plot 28 was somewhat lower than from plot 22. The proportion of £20 in.the NPKiunit from plot 28 is greater than from plot 22. The intensity of nutrition Of the crop from plot 28 is always higher than from the highest yielding plot at every sampling. It is apparent from these ZOY. m.mma a.mem .u.: u..- it--- n--- oamfiaw as m H.~NN «.mma mm.m ow.e on.am HH.a osaH mm H.aHH s.mea mm.o~ ae.m ap.ma Np.e Moose H Tenn lumen. swam on.a $8 do was an m.mam a.ema aa.mm mm.e Hm.oa ma.m m.~a«mam.umfizmv mu m.moa m.eaa mm.ma m~.m mm.ma mm.m osaa one ensues we on” noun a scream 930 was chem use 0.8m use chem hp Hansen“ .3...” a.“ . 232.3 98: namesake no.3 meson» ease.mmz so soaeaeoaaoo .Aanma .aeea ens nesoaav newsman 00.3» on» 29C shoe Mo 930: can .323 E 88.: 6332.89”: 23.. .mpnoapmoap nowadays.“ we???» own OHDfiH. 268. data that the intensity must be reduced from 6.73 to 5.9, the percent- age of nitrogen in the NPK unit must be increased from 70.5 to 73.4, the percentage of P205 must be increased from 7.3 to 8.2, and the percentage of K20 reduced from 22.2 to 18.4. One must find some means of estimating the amount of the reduc- tion of KéO required to decrease the proportion of K20 in the NPK unit from 22.2 to 1844. A.preliminary estimate may be obtained by compar- ing the data from plots 26 and 5 in table 38. It is seen that a reduc- tion of 12.5 Pounds of K20 reduced the K20 in the NPK unit by 15.6 percent. Since it is necessary to reduce the Kéo in the NPK unit by 3.8 units. therefore 15.6 g 1205 3.8 x 1.: (3'815f22°5) : 3.04 pounds per plot The new fertilizer which should have been used is 9:6:9.5. that is, on the basis of the foliar'diagnosis, Optimum yields would be Ob- tained by using a fertilizer containing 9 pounds of N, 6 pounds of P205, and 9.5 pounds of KéO per plot. Comparison of the data from plots 1 and 23 suggest another possi- bility, for here it is noted that liming caused a decrease in the ins tensity of nutrition, an increase infithe N'and P205 in NPK unit, and a decrease in the K20 in the NPK unit. Therefore, it would appear that the yield from plot 28 could have been improved by the addition of lime. 269. E. Discussion The problem of develOping economical and simple methods for con- trolling the factors governing the growth and composition of crops probably is the most perplexing ever to confound the investigators of agricultural problems. One of the contributions of Lagatu.and Maume and of Thomas and their collaborators is the division of the major problem into two distinct units: one concerning the study of the effects of fertilizer on the nutritional composition Of the crOp and consequently the determination of the nutritional status of the crap conducive to maximum yield, and one concerning a study of the soil itself in order to determine the amounts and kinds of fertilizer necessary for maxi— mum yields. Whether this division is of any real value is doubtful,, for surely it is less reasonable to confine the study of fertilizer requirements to the relationship between the plant composition and yield as is done in the foliar diagnosis method, than it is to study the soil in its relationship to plant composition, or even’to the study of the soil alone. Further, it is well to recall that in Section II of this report, a great number of factors were described which affected the composition of plant tissues, but over which man had little or no control. It seems very unlikely that a precisely defined chemical composi- tion of a plant can be designated as Optimum. except under the condi- tions under which the “optimum" composition.was determined. But it is upon this rather weak foundation that Legatu and Haume have 270. attempted to apply the foliar diagnosis technique to the determination of the fertilizer requirements of a given crop. From data obtained over a number of years. Lagatu and Maume have calculated "standard” ratios of certain chemical components in the leaves of potatoes as N2P2K: :- 38:4:58 (1934c, 1935a) and for grapes as 41:8:51 (1936, 1936c, 1938). Following the same techniques, Craig (1938, 1939) pro- posed 50:12:38 as the ”standard” ratio for sugar cane. It is difficult to believe that these 'standard" ratios represent inherently Optimum chemical compositions which result in Optimum yields regardless of other factors as claimed by Lagatu.and Maume. Even if we assumed that such a biological constant could exist, it is much more likely to be in the form of Mitscherlich's c constant than in the form of a ratio between arbitrarily selected nutrients in plant tissue. It must be borne in mind that these ratios are confined to only three of the large number of equally significant nutrients found in plant tissue. 1 further point detracting from the possibility of the existence of a physiologically important NPK unit as a biolOgical constant is that the “standard” ratios proposed by Legatu.and Maume are affected by many factors, including, of course, all soil factors.. Even if it were possible to accept these ratios as constants, which is difficult. to do at the present time with the small amount of data available, they could not be used with the same case Of manipulation as is possible with the Law of Diminishing Returns. 271. The foliar diagnosis technique,‘weak as it is in its ability to forecast fertilizer requirements, does indicate changes in the nutri- tional status Of a given crop. The technique may be used with.consi- derable success to study the effects on the plant of such factors as cultural practices, irrigation, placement of fertilizers, pruning, climatic conditions, diseases, and abnormal soil conditions. This type of information may then be used as a supplement to the more appro- priate techniques permitting the quantitative determination of the fer- tilizer requirements of a crOp. The foliar diagnosis method is nothing more than another form of tissue test. Currently there are two procedures used in interpreting the data on leaf composition. One interprets separately the effects of the individual nutrients and selects minimum values below which crop yield is affected (Hoagland, 1941; Ulrich, 1942, 1942a). This assumes that all other components will be present in amounts greater than their minima, and that only one dominant limiting factor is Operating at a given moment. The second, based on the foliar diagnosis technique, converts the chemical composition of the crOp into a form which.indi- cates the equilibrium between three major nutrients. namely nitrogen, P20 , and K20. Unless it can be shown.experimentally that the RPK unit is related more closely to the growth and yield of the plant than are the percent- ages Of the individual nutrients from which it is derived, not only can the extra computional work not be Justified, but even the nutritional ZIZ. relationships may be unnecessarily obscured. Such a comparison has never been made, and until it has been.made, the foliar diagnosis technique can be considered no better than the critical concentration technique. There is only one general conclusion that can be drawn concerning the foliar diagnosis. This procedure cannot be applied in a manner which.will permit the estimation of the fertilizer required for opti- mum plant growth. Although such attempts have been.made by Legatu.andc Maume, their techniques are so weak in theory and so clumsy in.practice that they are of no practical value. 273. IX. TRIPLE ANALYSIS A. Introduction The triple analysis technique for testing soil fertility and for predicting probable crap response to fertilization.was developed by Lundegardh (1938) in Sweden. The procedure is based on the assumption that the fertilizer requirements of the crop could be determined most readily by chemical analysis of the plant, and of the citric acid exe tract of the surface soil and of the sub-soil. Previous to the work of Lundegardh, most investigators limited themselves to the use of soil analysis or of plant analysis in their quest for a method for determine ing the probable responses of crops to fertilizers. Joulie (1889) and Januszewski (1895) were probably the earliest workers to consider both plant analysis and soil analysis in the interpretation of crap response data. Januszewski (1895) analyzed two soils, set up fertilizer experi- ments with.wheat on the basis of these soil analyses, and then.analyzed the grain and straw at the time of harvesting. Januszewski altered the fertilizer application as a result of the soil analyses and attempted to correct the nutrient deficiencies in such a way as to obtain‘wheat with the desired chemical composition. He concluded that chemical analyses of crOps were of value for the estimation of fertilizer re- quirements only if the soil is carefully evaluated as to its physical and chemical characteristics. Expressing the same conclusion, Lunde- gardh (1938) stated that 'a good practical test of the fertility of the soil and its fertilizer requirement must be built up from a 274. scientific knowledge of the soil and its influence on the growth of the plant" 0 B. Technique The triple analysis technique, as originally described by'Lundegardh (1938), varied according to the conditions of the experiment. In some Iinstances, the ash analyses of the plant tissue and the analyses of the surface layer of soil were used, in other instances, analyses of the plant ash, surface soil, and sub-soil were used. While the term triple analysis was prOposed to fit the latter case, it soon became a general name for the simultaneous use of chemical and biological data to determine fertiliZer requirements. The analyses of the soil included the pH, phosphorus, potassium, calcium, magnesium, manganese, capper, and iron. Lundegardh used quantitative spectrosCOpic analysis for determining the cations. For pH and phosphorus determinations, electro- metric or colorimetric methods were used. The analyses of the leaf sample was considered by Lundegardh to indicate nutrient deficiencies in the soil, and to a certain extent whether an inJurious lack of physiological balance exists between.two or more nutrients. The analyses of the soil were considered to reveal the minimum availabilities of nutrient cations, and also inhibiting or antagonistic effects of certain ions such as the effect of lime on the uptake of manganese or the effect of potassium on the absorption of calcium. It is apparent that Lundegardh regarded the plant as an ideal 275. soil extracting agent and the analysis of the soil an aid in the in- terpretation of data obtained by analysis of the crop. Unfortunately, this fruitful approach to the problem of the deter- mination of crop fertilizer requirements was not develOped on a prac- tical level. Lundegardh subsequently abandoned the very real possi- bilities of his triple analysis procedure, and limited it to the plant only, thereby following the ample of Lagatu and Maume (1931!“ 1937, 1937a), Thomas (1916). and Thomas and Mack (19M), 19%). In other words ”triple analysis" became ”tissue diagnosis”. Lundegardh's approach to leaf analysis was smewhat different than that of Lagatu, Thomas, and their collaborators. The Lmdegardh procedure was based on leaf samples collected at the flowering stage of the plant when the vegetative parts were fully grown but still vigorous, a practice in contrast to that followed in foliar diagnosis in which samples were taken throughout the whole season, and also in contrast to the techniques of the majority of tissue analysis methods which, if a single sample is taken, used samples taken as early as possible in the growing season. Lundegardh (19%) tested the practical value of his leaf analysis in 800 field experiments with cereal craps in different parts of Sweden in 1938 and 1939. 3y plotting the percentages of potassium in the leaves from unfertilized plots against the increased crop yields obtained with potassium fertilizers, Lundegardh (1918) obtained a curve approximating a hyperbola. What he actually found was an approximate inverse rela- tionship between the concentration of potassium in the leaves and the 276. increase in yield obtained with fertilizer. Similar curves were ob- tained in respect to the nitrogen content and the effect of nitrogenous fertilizers, and in respect to the phosphorus content of the leaves and the effect of phosphatic fertilizers. The mathematical equation showing this inverse relationship for a single nutrient was expressed by Lundegardh (19h3) in the form b 333%“ x, where a, b, and c are constants, y the increment in the yield due to the addition of the fertilizer, and x the concentration of the nutri- ent in the plant at the time of flowering. The general form of the equation for more than one nutrient was given as _ l2. 51 t: C HA-x'p xxackx x3}{ S where y is the increase in yield,.A is a constant representing the individual values a1, a2, and a3; :l’ 1%, and :5 are the percentages of potassium, phosphorus, and nitrogen respectively, and 0a is a correction constant'which brings the values of b1, b2, and b3 into numerical agreement with the scale of yield increments. Whether or not these equations correctly represent the relationships they attempt to express is a matter of conjecture, for the supporting evidence is far from conclusive. Curves relating analyses of the crop and its response to added fertilizers were plotted by Lundegardh.from the mass of data he ob- tained in 1938 and l939. These curves may be used to study the rela— tionship between different amounts of two nutrients in the leaves and 277. the increased yields resulting from fertilization. For example, figure 52 shows the effect of the interaction of potassium and phos- phorus. If the soil suffers from a phosphorus deficiency, as is in- dicated by the low percentage of phosphorus in the leaves, fertiliza- tion.with a potassium fertilizer alone will produce a smaller crop yield than.if both potassium and superphosphate were added. Similarly leaf analyses may show whether superphosphate fertilization is profit- able without simultaneous fertilization.with nitrogen. The tabular method of presentation of these data used by Lundegardh (l9hl) is shown in table 39. Since the curves develOped by'Lundegardh are based on data from a large number of observations over a wide area, they are of some value in suggesting recommendations of the fertilizer requirements. The technique is the only one that truly relates plant analysis and yield for a number of different factors, and if it will do nothing else, the technique demonstrates to the farmer how he can economize with mineral fertilizers. 0. Discussion The triple analysis technique as described by Lundegardh (1938) is the earliest significant approach to the study of the nutrient levels in the crop as they are related to the response of the crap to fertilizer. Any method for the diagnosing fertilizer requirements of crops is not directed.primarily toward forecasting crop yields, but rather to fore- casting the probable response of the crop to the addition of a fertilizer 278- :: nxraLI‘LATION MM! nu AC". Ill-D A'TII Y m IILOO § Incense m 0 ‘0 W 60 8° Figure 52. Probable effect on yield of cereals as a result of fertilization with 60 kilograms of 1.0 percent potassium salt per acre. As the potassium content in the tissue increases the effect decreases. Increasing supply qf phosphorus raises the utilization of potassium. (Lundegardh, 1943) 279. uu- tun: ommnoom oomuoma owe nu: oomuoom oomnoma omaaooa ooH m.mA essence a :3: romance m.mnm.d vnmpaoo A no“). cowoaufiz 03A 8TB; 0.218 8-8 .I.. msv cheeses a near .oupoapamm he hopes: sea .ms ocmuoma me access m A mum mum. en: a 83 essence a he“: accesses a A one mA us- .. 02 v “.828 a near nanosecond .oupoapasm me peace: hoe .ma oomnoma cc eoaeaeeo on Amomv oumammonaaomsm Mo season sea .wsnomauooa me access 8A 8.8 8.3 3.6m on v To? essence a he? essence och“ .ocuom omuom omnom oNJv m.wum.e essence a see: undressed on A onus 3.8 omuoa I. ma? ececeee a one: each Hagunmon omdao>< doom doom has» .oaaompasu mo heaven nos .3 comuoma 8. schemes» 3 Awoav nuance mo endgame no peace: to .3 3702 he sects souaaheeoa so sconce rapeseed as name mo mobmoa on» no panama.hAd can a“ redneck Hecaphamq< 32$ .fieoweeqod stances apneambwsaOaafiaa anooaom ms commonawo .mpso no edefih macaw one so unusuaauaom no vacuum .mh capes 280. so that yields may be improved. This may be accomplished either by applying the Law of Diminishing Returns as was done by Bray and Willcox or by the complicated procedure using tissue analyses. The minimal requdrements for the development of such a tissue analysis method would be large numbers of data from plants growing on a series of soils containing various amounts of one nutrient while others are present in-excess. The data must include the internal concentrations of the varying nutrient at a definite stage in specific organs, and. also the increases in crop yield resulting from the application of definite amounts of nutrients. The increments in yield resulting from the addition of nutrients could then be represented as a function of the nutrient concentration within the plant at the time of fertilizer application. This procedure is essentially that used by Lundegardh (1938, l9u3), but it cannot be compared either in general applicability or in ease of execution.with the practical value of applying the Law of Diminishing Returns. (0.1.. I. CBITI CAI. CONCENTRATION A. Introduction The value and possible application of critically defined concen- trations of nutrients and of nutrient ratios in plants as a biological method for the determination of fertilizer requirements of plants has been discussed by Thomas (1945), Goodall and Gregory (l9h7), Ulrich (1948), and others. The biological methods discussed in previous sections of this report have been based primarily on accurate laboratory chemical tech- niques by which plant tissues of various ages were analyzed, and their compositions expressed on a dry weight basis. 0n the other hand, the analytical techniques used most commonly for the determination of ”critical“ concentrations of nutrients in plants have been the so-called quickbtest types. These techniques probably were given their greatest impetus as a result of Hoffer's (1926) scheme of quickhtests for nitrates and potassium in the corn plant carried out under field con? ditions. Later they were 'improved" by Thornton, Conner, and Fraser (19310, and by others, and were extended to include other nutrients. From this point onwards, the development of modified "quick? techniques proceeded at a rapid rate and contributions were made by many workers. Among these, Emmert (193a), Carolus (1938), Hence (1936, 1937, 19%), and Scarseth (19h1, 19h2) merit especial consideration. At the same time that these techniques were being developed, 282. wallace (1932), Lagatu.and Maume (l93h) and somewhat later, also Thomas (1937). Nightingale (1942), Macy (1936), Ulrich (l9h1, 19b2, 19h2a), Goodall (l9h5) and Boynton and Compton (1945), adopted the more accurate analytical procedures for ascertaining the nutrient COD? tent of plants. The deve10pment of the abbreviated chemical tedhniques permitted analysis of great numbers of samples to show the effect of single nutri- tional factors. With the universal acceptance of the statistical techs niques developed by Fisher (193h), it became relatively easy to show apparent relationships and correlations between the nutrient levels in crOps and crop yields. Consequently, at the present time, the majority of research projects involving the diagnosis of the nutri- tional needs of a crop utilize critical concentrations and ratios of nutrients in the crop as they are related to yields. Goodall and Gregory (19u2) in particular have made an exhaustive study of the various techniques used by investigators and have compiled tables showing the specific tissues recommended for sampling, the times of sampling, and the critical concentrations or the standard ratios recommended for the large number of crops. This compilation shows that the majority of workers using tissue analysis have attempted to standardize their methods to take into consideration as many as possible of the causes of variation of the nutrient content pf the plant. Regardless, however, of the type or quality of analytical proce- dures used, most of the work might be divided into one of two classes: 283. interpretation through the correlation of an optimum, critical, or threshold concentration of nutrients in the plant against yield, and interpretation through correlations of ratios of nutrients in the plant with yield. Both of these approaches will be considered in the followh ing pages. It is proposed to consider the background upon.which the two above approaches are based and the various ways in which attempts have been made to correlate them with plant yields. 3. Correlation of Critical Concentration with Yields 1. Introduction The possibility of using minimum concentrations of inorganic nutrients in plants as the bases for determining nutritional defi- ciencies has been recognized for many decades by'workers growing plants in artificial culture media and in field plots. A.few classic publications are those of Hellriegel (1867), Heinrick (1882), You Dikow (1891), Atterberg (1901), Pfeiffer, Simmermacher, and Rippel (1919), numert (1935), Macy (1936), Ulrich (19M), and Hill (1950). It was found that plant growth ceases when the concentration of any of the nutrients in the leaves falls below a certain minimum value. This minimum is different for different species, varieties, types of growth etc. Macy (1936) presented a theory that there is a critical percent- age of each nutrient in each kind of plant above which there is luxury consumption of the nutrient and below which there is a poverty adjust- ment which is almost proportional to the deficiency until a minimum per- centage is reached. Macy tested his theory by studying the nitrogen requirement of barley during 1933 and 1934. The data based on the straw are given in table #0. In this table, the response to the nitro- gen fertilizer is calculated as ”3% for each increment of nitrogen and is compared with the corresponding average percentage of nitrogen in the straw. The yield curves for the straw reproduced in figure 53 show that the treatments exerted marked effects on the yield. However, the relationship between.percentage nitrogen in the straw and the res- ponse of the yield to nitrogen shown in figure 53 is independent of these factors. This situation is in agreement with the theory advanced by Macy. In this case, the minimum percentage of nitrogen in.the straw below which growth ceased was O.h percent, and the critical percentage above which luxury consumption began was about 0.8 percent. Macy COD! cluded that the critical nutrient composition of a plant is an 'ideal but inherent characteristic of the plant,” and that the critical and minimum percentages varied only under special conditions. Further, he concluded that these percentages were intimately related to vegetative growth.and yield. Macy believed that the use of his theory to determine the ferti- lizer needs of particular crops on particular soils was as reliable as were field plot tests. This conclusion assumes that the sufficiency of a nutrient is a function of its percentage in the plant. ‘ll‘l Inner-raflervi u.:..l\ pry..\?.,...v s.- P.5- ..r,il uni-1... kart lirbl....t.nl-It.,n else...- n..-‘ .o..i|.t..r.... .ntt.....‘-£ -.. .~.- . ...... .u 285. mem_ m 8.." mm” mmwm 3.." om.m a emé 0.." nm.m n . . a no; II In: . .H «Ta 3 mm mm.» H H n n H J- m--- rm. ll. ....!. . . on who m o «d a d O O o N i .‘Il' m m.“ ME cum on e; w..- if; e: se er .3 mm was 2.0 and mm.m E R5 mm.w mm.“ mm mm a. 8m R5 lit 3.... m: 3.0 .0 $5 m3 3.0 9.... ......u 3 in- 3.... mmm inc i5 85 in... I..- mm am.e on.a 0 me o me.o ma.~ an.e ae.a e damned amine and s “00H H000 o o o o o a an er an e e a s 3 84 HA maa mm SM 3.” 3.: n on.a Hm; 2 mm mm ae.o mm.e He.m N HH.H m.o ea.e ea He.o .c we. and mm 86 .o -.m ”We mwgw n6 ems “We ammo do Rm Mad 3 0 SJ. and mete. me o .o on.a o mated.» autumn» mutumua . 34 mad . .36 o.~ . . mm.- MM.M mm.” mud dun no.m mm.m mmé a mad WM MM.“ Wmm 0H MBoO Fwoo “New AN I awn-N “moN ONoN +2“ I NOoN Bed hunch“ NMN .1 an e we a me.e m~.d one as e em.e ea.a e leb 01> GIG m a a conned Z q pdooued a 2 Q. acceded a Z V «donned a a madam \r Q omddobd ado Odom 30 “M H q owddcbd ado once. .30 C. x _V awakened ado each 30 a» 1a.: «2 Renee .eoea~-- 2.86. STRAW F h YIELD HARLEY srmw (9mm) PERCENT N IN OAKLEY Figure 53. A. Yield cirvee for barley straw under _ nine fertilizer treatments. B. Relationship between percentage nitrogen content and the yield response ; (barley straw) due to nitrogen fertilization ( Macy, 1936) (CY. While much work had been done previous to that of Macy which.was based on definite minimum concentrations, the theory by Macy and its possible application resulted in furthering interest in this type of nutritional research. The general concept can be criticized because only one nutrient in the plant is considered and no consideration is given to the effects of other nutrients on the absorption and utiliza- tion of the one being studied. EWennwith this limitation apparent, Ulrich (l9h2a, 19u3, l9h6, 1948) developed Macy's ideas further‘with a slightly different interpre- tation. Ulrich defined the critical nutrient level as “that range of concentrations at which the growth of the plant is restricted in com- parison to that of plants at a higher nutrient level.“ He considered (19148) that the evidence indicated that the critical values varied only slightly in comparison with the range of concentrations above the critical level. Thus the potassium levels of the middle leaves of low potassium tomato plants grown in culture media varied from 0.91 percent in the dry matter in the early part of the season to 0.51 percent in the latter part of the season, while in high.potassium plants the percentages varied from 4.66 to 2.29 during the same period. filrich (l9h6, l9h6a, l9b8) presented data supporting his statement that "there is no profound effect upon yield when large fluctuations in nutrient concentration well above the critical level occur within the plant.' For example, his data show that in one sample. when the percentage“of potassium in the middle blade of the tomato leaf changed from O.h7 to 288. 0.51 percent, the yield changed from 146 to 21h grams of dry tops. 0n the other hand, when the concentration of potassium changed from 2.29 to 2.78 percent, only a slight effect was evident on the yield. It would seem that Ulrich's data could be interpreted more easily on the basis of the Law of Diminishing Returns. The data. however, show the difficulty of working with critical levels of single nutri- ents for it was found that when low levels of potassium were in the plant added sodium exerted only a small effect upon the level of potass— ium in the plant but exerted a profound effect upon the yield. Nb better example could be desired to show that the addition of a second ion may effect the utilization of the first even though it is presnnt in its “critical“ concentration. It is necessary to determine the effects of other factors than the one being studied on nutrient cone centrations within the plant and upon the final yield. This means. of course, falling back to the techniques of Lundegardh (1941). In any case, Ulrich (l9h3) believed that the critical concentra- tions of the various elements in the crop did not vary enough as a result of season, of the amount of sodium, or of the potassium concene tration to prevent their use as points of reference to ascertain the potassium status of the crop, but the data he presented does not appear to Justify this conclusion. As a result of the work described above and of similar approaches, many‘workers have reported critical concentrations of an element in plants below which deficiency would be expected (see Goodall and 289. Gregory l9h7). Among these workers are Carolus (1933. 1933a, 1935, 1937. 1938), Gilbert and Hardin (1927). Hester (1935), Scarseth (19Li3', l9h3a), Emmert (1935. 1935a, 1937. l9hl, 1942, 1949), Boynton (l9h2), Hill (l9u3), Hill and Cannon (was), and many others. .A considerable amount of work has been carried out by English workers, particularly at Long Ashton, on the relationships between threshold, or minimum, values, visual symptoms of deficiency, the laboratory chemical analyses, and data obtained by tissue tests. Fbr example, Nicholas (l9hh, l9h8), Nicholas and Jones (l9h4), Jones and Nicholas (1916), Nicholas and Catlaw (l9h7, l9h7a, l9h7b) have shown that the appearance of visual symptoms of deficiencies, especially of potassium, magnesium, phosphorus, manganese and boron and of visible I symptoms of excess chlorine, zinc, and.manganese are correlated with the threshold content of these substances in the plants. The value of these data is trivial because it should be apparent that when nutrient deficiencies are acute enough to produce visual symptoms, or nearly so, there is no need for chemical analyses to show that the concentrations of the nutrients in the plants are low. 2. Determination of Critical Concentrations The methods of determining critical concentrations of nutrients in plant tissue are as variable as the number of workers conducting investi- gations of this nature. There are, however, two distinct approaches to the general problem. Ulrich (l9h8) suggested that a preliminary estimate may be obtained by growing the crop in soil or in solution cultures. The averages of one such set of data based on Ladino clover supplied with various amounts of phosphate are presented in table l‘rl. The yields and the phosphorus contents of the plants are presented graphi- cally in figure 51;. Similar data for a series of potassium fertilized series of field plots appear in figure 55. These data caused Ulrich to conclude that the critical concentration for phosphorus is approxi- mately 600 parts per million in the dry material, although the data themselves indicate that this was only a rough estimate. Under field conditions, Ulrich's (l9u8) calculation of the criti- cal concentrations is based on comparing the concentrations of the nutrients in plants known to be deficient with those in plants amply supplied with the given nutrient. When this procedure is repeated over a period of seasons, correlation between yield and the concentrate— tion of the nutrient usually will be obtained, and the nutrient levels in the plants signifying deficient yields might be established, as was done by Ulrich (19b6a) for Ladino clover. Ulrich's work (191+6a) showed that he could not always depend won the results of field plot experiments, as is shown by figure 56. In this case, there appeared to be no luxury consumption, as was found, in the pot tests, and be arbitrarily considered that the critical level again was 600 parts per million although the results of the field tests do not show that this was true. 291. Table #1. Fertilizer treatments, average yields, and average P205 as parts per million in the leaf stalks of Ladino clover. (Ulrich, l9h6a) Treatment Fertilizer P 0 Yield application (Pgms (gms) gms. per pot untreated --- 378 11.2 Po 0 3.02 gms. 389 1h.9 L me 19.2 gms. P1 0 3.02 gms. M9 3203 L me 19.2 gms. P205 0942 gfls. P 0 3.02 gme. 762 #6.3 L me 19.2 gms. P205 0.8# gms. P3 x20 3.02 gms. 797 52.3 -Lime 19.2 gms. P205 1.27 gms. Bu 0 3.02 gms. 946 52.0 L me 19.2 gms. P205 1069 ans. ' 1‘72. ‘o‘ I ‘ 6° h I...“ + 3 + e a o _ t ., t- e l + 0 f #0 I- I an I . : * Pnosmoau Seance :; o l 0 l9l+l l * 20 r- ' ‘ l C3 . 8 I a l L l L ‘00 600 woo woo (poo "m PHOSPHA‘I'I' Pue JPIIORUC Figure 5a. Relationship between the.yield of Ladino clover and the phosphate concentration in the petioles when phosphorus is limiting in some of the treatments (Phosphorus series of plots). Solid line is a freehand line drawn through the average points. (Ulrich, l9h6a) 292w. POTA$CI UH Stfipgg A | a ‘ g ‘0’ i o b 0 I90! 0 Q ' A .. a V ” ' o o~ * I 00 o . d» o I #9 ' l O. 0 00 2 . 0 U " I ° I 3 t- . , >- I In _ . C3 I l t l l l l «we 600 woo woo Iloo use m» PHOC'OOATK Puosruenuo Figure 55. Relationship between the yield of Ladino clover and the phosphate phosphorus concentration of the leaf petioles when the phosphorus is not limiting growth ( potassium series of plots). (Ulrich, 19h6a) 293. ‘ use | 0 ll a zoo '3 e I“ S: a|ntlo 31‘0 enh‘ I- .10 eOI :5 M- .6’1 3 .1 3 .. l‘ 80 - 0+ I * I 2' °I ' ac Cline ' Gt. ‘ 'I :60 9‘» a.» L 19.3077:- Puosnarz Puosruonus Figure 56. Relatioship between the yield of Ladino clover and the phosphate phosphorus concentration of the leaf petioles of clover grown under field conditions under various fertilizer treatments.(Ulrich, 1946) 29h. Emmert (1935a, 1937) used such.atatistical procedures as correla- tions, partial correlations and partial linear regressions to determine the limiting values of nutrient concentrations in crops. The details of his procedure are discussed in his paper describing the correlation of yield.with the amount of soluble nitrogen and.phosphate in the sane ducting tissues of potatoes at various stages of growth. His data are given in table #2 and they show the effects of nitrogenous fertilizer on the nutrients in stem tissue and on the yields. His approach was purely statistical in that he used a linear correlation, but which, of course, was not strictly accurate for the pOpulation as a whole. Emmert, however, maintained that if the data'were limited to values below the point where the curve flattens out, the correlation.was almost linear, and while not representing the population as a.whole, this segment of the linear curve showed the degree of correlation.with this limited group of data. If the higher values of nitrogen in the tissue are _ associated.with insignificant changes in yields, it indicates that a point has been reached.where the curve begins to flatten and that the Optimum for the set of conditions being studied is near this point. If, when the values above this point are removed, the partial correlation drops or remains constant, then there is an indication that the Opti- mum value probably lies above this point. But if the partial correla- tion rises, it means that the high values are associated with.lower yields and that the Optimum was below the value selected. It was by‘the above procedure, based on the data in table #2, that 295. Table 1&2. Effects of nitrogenous fertilizer on the concentration of nitrate nitrogen and of phosphate phosphorus in the stem tissue and on the yields of a spring crop of potatoes, together with various correlations, partial correla- tions, andoptimal levels of nitrogen and phosphorus. (Emmert, 1935a) Date of sample * Treatments Yields. June 15 June 21 July 15 Average in pounds average Sol. Sol. Sol. Sol. Sol. 561. $61. $01. per acre of 4 plots N p N p ' N p N 1: (WWW!) ppm ppm ppm ppm ppm ppm ppm 1013!!! Check - 127 321 228 160 198 218 116 233 181 100 lbs. NaNO3 187 896 226 265 160 324 105 495 164 500 lbs. NaNO3 200 . 747 210 780 180 627 270 718 220 Check ' 152 353 218 210 183 262 140 275 180 100 lbs. (1138);;504 207 545 207 453 132 547 118 515 153 500 lbs. (NHQZSOL, 211 712 233 913 124 1030 100 855 152 zoo lbs, 279 980 192 1267 128 1130 153 1126 158 300 lbs. (11332804 8 tons of straw 155 342 262 568 155 613 205 508 20? check . 170 284 245 260 160 218 127 254 177 10 tons of manure 199 290 268 233 167 169 196 231 ' 210 140 tons of manure 202 239 218 1+6? 146 327 155 31414- 173 check 185 288 208 217 117 242 187 249 171 Simple correlations June 15 Jun'e 21 July 15 (significance 0.273) 151 0.395 0.556 0.424 r! 0.002 0.053 0.074 H? 0.272 0.689 0.144 Partial correlations (significance 0.281) n Y . P 0.416 0.819 0.441 p r . x 0.123 0.724 0.149 N r . I 0.287 0.866 0.128 Partial correlations with plots with nitrogen above 1000 ppm removed and phosphorus above 200 ppm removed. 19’ I . P 0.376 0.515 0.170 P Y . 3 0.14146 0.1711» 0.050 Indicated direction of optimum Nitrogen above 1000 ppm above 1000 ppm above 1000 ppm Ph orus below 200 ppm above 200 ppm above 200 ppm 296. Emmert (19353) was led to the conclusion that the Optimum concentration of soluble nitrogen in the stem should be at least 1000 parts per million throughout the season and that the soluble phosphorus concen- trations should be below 200 parts per million. When, however, his data for nitrogen are plotted as in figure 57, there does not appear tO be any Justification for the selection of the value Of 1000 parts per million of nitrogen as the Optimum concentration. Indeed, the data show a very sharp break at 400 parts per million and there is greater justification in taking this value than the higher value that Emmert selected. Similarly, the phosphorus data shown.in figure 58 do not indicate any Justification for selecting 200 parts per million.as the maximum value other than the fact that the value associated with the highest yield is below 200. It is Obvious that if this value is omitted, there is no correlation between yield and phosphate concentration. The above examples have been cited to show that in a pureLy sta- tistical study, there are serious pitfalls into which the investigator can fall and that there are occasions when unjustified conclusions are drawn from such data. Statistics are of value in such a study only if used as a supplementary aid and not as a necessarily authoritative indication. I Emmert, however, realized the errors of his earlier technique and in a later study he used different statistical methods. In 1937 he described his use of partial linear regression to analyze the curvie linear relationship between crOp yields and the percentages of nutrients £97. ”sf ° °' ” 0e Ill 0 P l- ‘ < b 1 i- ‘ O m 0 C X 0 u 00x 0 o O .l u OAVIAAOI 5: ed.“ M 1 I done! o O x e OJULY 5' I50 0 A 1 L A A too 000 coo can IOOO Jmeolvm. N moon: m Sun Figure 57. The relationship between the soluble nitrogen, expressed as ppm., in the stem tissue and the yield of potatoes at three different dates. (Plotted from data of Emnert, 1935a) 2.98. x .0 O cumu- than: 6’ “Out I! “?Ud- New I i V n u 2 E 0. x e o O k 0 3 D x . '. l o .v .. 3 o x x‘ o e O '1' >' o e o 0 x x. o '50 O 1 ~ 9': 1.3 e L '00 I50 200 ISO PHospuosus m STEM Tissue(rm.) Figure 58. The relationship between the soluble phosphorus content expressed as ppm. in the stem tissue and the yield of potatoes at three different dates. (Plotted from the data of Ekmnert, 1935a) 299. in the plants. The data given in table 143 illustrate Enmert's revised technique, and are based on the response of potatoes, grown in 1935 at Lexington, Kentucky, and fertilized with nitrogen and phosphate. The method in- volves grouping the data in groups consisting of approximately the same number of observations as indicated in table 1&3. It is obvious that the mean yields of these groups should be tested for statistical sig- nificance before proceeding further. In this case, there is no signi- ficant difference between the ranges Of compositions, and further in- terpretation of the data normally would not be continued, but the procedure will be described to illustrate its possibilities. The next step is the calculation of the partial and single correlations shown in table 1&3. Then the partial regression lines can be obtained from the formula ‘1; 1 M ‘” “1%(“757 where, y is yield, n is ppm. of soluble nitrogen in stem tissue of the plant, p is ppm. of phosphorus in the stem tissue of the plant, M is mean of group, N is number in the group. r is correlation coefficient r is standard deviation. In this instance, the three partial regression lines are as shown in figure 59 and are noted as follows: 300. Table. 143. Yields of potatoes and the concentrations of nitrogen and phosphorus in the stem tissue, expressed as parts per million at Lexington, Kentuclq, 1935. (finmert, 1937) Range 1 Range 2 Range 3 N _139 - 233 N’ 257 - 555 .11. N’ 5§§_- 1625 "'11 Yield P T Yield P H n eld P" ' "‘ (ppm) (1bs.) (ppm) (ppm) (lbs.) (ppm) (ppm) (lbs.) (ppm) 139 33 195 253 38 160 556 42 188 143 43 146 264 48 146 ' 626 45 125 147 40 120 264 39 146 700 63 146 154 48 178 269 40 160 750 52 178 162 43 266 264 38 250 834 46 188 182 27 200 265 52 175_ 834 56 178 186 55 260 286 49 114 408 65 105 200 51 240 312 51 155 1125 46 160 208 45 170 330 46 260 1200 58 122 213 41 145 380 60 134 1250 80 140 233 47 146 436 57 130 1625 70 150 --— -- --- 436 64 200 --— -- -- -- -- --- 470 55 134 --- - --- 4. -— --— 476 54 ' 160 -4. -- -__ -- - -- 555 37 155 -- - --- Sum 1967 484 2116 5259 728 2499 10408 623 1680 Number 11 15 11 Mean 178.8 44 192.4 350.6 48.5 166.6 946.2 56.6 152.7 0” 30.5 8.5 45.2 96.4 8.4 40.2 30.6 11.5 37.1 ”by + 0.453 1-o.381 —+0.638 rpy 4’ 0.112 -O.254 -O.l+llv8 rap +0 .093 ”0.095 “00275 rum +0.14“? +00371 +00599 30!. /ACR£) ? .§ Porn-eta (bus. '8’ § n.7,, . .8669 no, - ~31“ PM” - .6534 Yum or A A A A A l Navaooeu on Sun Tues: (rpm) .—L. 600 Figure 59. The relationship between the yield of potatoes and the nitrogen concentration in potato stems shown as regression lines for the three ranges indicated in table 43. (Einmert, 1937) ' Range 1 1': 21.7 t 0.125 n 2 Y': 37.5 1-0.0315 n 3 1': 36.8 1'0.021 n These data exhibit several interesting aspects when plotted in the manner described. Since the data suggest that the Optimum concenr tration of soluble nitrogen is 400 parts per million, a value suggested by the previous test, it first Of all removes the danger of selecting the wrong breaking point in the data as was done in the previous'work. Secondly, it shows that between 139 and 250 parts per million of soluble nitrogen in the tissue, the yield increases 0.5 bushels per acre for each.part per million. Between 250 and 556 parts per million, the in— crease in yield is 0.126 bushel per acre, and above these values, the increase is 0.084 bushe1_per acre. These observations approximate the Mitscherlich type of yield curve. Finally, if this type of relation- ship were expanded to include a larger number of regression lines, it could be used to determine if fertilizer applications were profitable, but this could be done only if a relationship between the soluble nitro- gen in the stem tissues and the rate of fertilizer application could be shown conclusively. ' Another approach to the estimation of critical concentrations was prOposed by Carolus (1937) and.was based on experiments with potatoes at the Virginia Truck Garden Station. The crOp was sampled every seven days during the season.and a series of ”Optimum“ weekly levels of nutri- ents was obtained as recorded in table 44. The analyses were Obtained 303- Table 44. 0ptimum.weekly nutrient concentrations in potatoes growing in Virginia conditions in 1936. (Carolus, 1937) Age Of Weight Soluble Soluble K20 MgO GaO plant per plant Nitrogen Phosphorus (ppm.) (ppm.) (ppm.) (days) (51390) (PPmO) (ppm.) 42 26 1228 216 3500 250 800 49 60 988 148 5125 650 1100 56 160 1025 215 3750 675 1000 63 340 891 150 3125 1075 1100 70 470 1295 167 8250 1050 1100 77 630 1104 140 6350 950 1250 84 743 927 90 6000 1075 1800 92 673 1204 84 5900 1225 1900 Average for season 1085 152 5250 882 1315 ‘304. by 'quicketests" but Carolus states that these tests were "shorter and give a more definite indication of the nutrient uptake than complete methods, and are more reliable in diagnosing nutrient deficiencies than soil tests made after the fertilizer materials have been applied." Carolus believed that under similar growing conditions plants that did not contain at least 75 Percent of the amounts shown in table 44 should be considered to be deficient in that nutrient. The difficulty with this type of information is that, since similar conditions are difficult to define and may never be met, the Optima are Of little practical value. 3. Discussion Whether the critical, or minimum, concentrations of nutrients in plant tissues are of any real use to the research.worker for the deter- mination of the nutrient requirements of crOps is a question surrounded by much controversy. Before considering this point, however, it should be pointed out that such critical concentrations are of undeniable value for certain uses. One usable aspect of critical concentrations is their application to plant nutritional surveys. Surveys of this nature extending over wide areas disclose local deficient soils. suggesting areas which could be used to the best advantage for fertilizer investigations to be carried out later by more reliable methods. Tissue analysis is Of some value as a technique supplementary to 305. field fertilizer experiments. There can be no doubt that it also is a valuable supplement to soil analysis for the interpretation of the effects of added nutrients on the yields of crops. But since we are considering only the critical or minimum concentrations, care should be taken in their use. The use of critical concentrations and minimum levels for the confirmation of the diagnosis of nutritional deficiencies as expressed by visual symptoms is also important. Finally, such data are of value in following the seasonal trends Of the nutritional status of the plant. In spite Of the possible uses of I'critical" concentrations of nutrients in plants, such data cannot supply the information needed to determine fertilizer requirements of craps. The data are interpreted almost entirely by statistical methods, but no matter how these methods are applied, there is no way that the results can be put on a.practica1 basis. not until the investigator has data showing the yield responses of crops with varying amounts of fertilizer is he in a position to consider establishing standard values for the concentrations of nutri— ents in crops. The basing of conclusions on data from a single element is in general unsatisfactory. It has already been pointed out that the in- crease in.yield to be expected from increase in the supply of a nutrient is related nOt only to its internal concentration, but also to that of other nutrients. A.method for interpreting the results of tissue diagnosis must take into account the concentration of single nutrients and also variations in the concentrations of other nutrients. The significance of this latter aspect of the problem will now be discussed. C. Nutrient Ratios The limitations of the use of critical nutrient concentrations in plants have been recognized by investigators for some time. In many instances, the approach has been to compute and use nutrient ratios in the plant for diagnostic purposes. According to Goodall and Gregory (l9h7), this procedure involves certain hidden assumptions concerning the curves relating crop yields to nutrient content, and they explain the problem by stating, "Let us suppose a three-dimensional figure in which the vertical axis represented.whatever feature of develOpment was being measured (say, the response to a particular fertilizer application) while the two horizontal axes represented the content of the two nutri- ents in question in the plant material, all other factors being held constant: if development of the plant depends on the ratio of two nutrients'within.it, and not on their individual values, then any vertical plane through the origin must cut the figure in a horizontal line. This has never been demonstrated.I Gregory (1937), in fact; showed with barley grown in pot cultures that as the supply of nitrOgen and phosphorus increased in constant preportion to each other, the ratio of these two elements in the plant remained constant although their concentrations individually increased and the crop yield in- creased as shown in figure 60 and table #5. Moreover, when luxury consumption occurred, an increase in the supply of the nutrient in 307. Table #5. Yield of barley plants fertilized with varying amounts of nitr0gen and phosphorus, expressed in grams per pot. (Gregory. 1937) P1 P2 33 Pu as Ni 33.0 18.9 2.70 1.32 0.70 N: 10.5u 10.86 n.94 3.59 2.u5 n.64 b.43 b.20 3.50 3.47 3 En 1.98 1.98 2.20 1.97 1.27 N 1.20 1.22 1.0# 1.20 1.00 307: :7 s < E I... z a. c 2‘ 0 o s “h E “I h 9. z.. s a as.“ g 0.3» 5 a. o \.__ ‘ \c-"——-—. I . t“ “L . . . i r. . :. e a ,3 r. We“! Faun Gunman“ Wun (non GIRHmATIou Figure 60. Drift of phosphorus (A) and nitrogen (B) expressed as percent of the dry matter of whole barley plants sampled 2, h, 6, 8, and 11 weeks after germination. Nitrogen and phosphorus varying together. Balance of N:P205 constant at 3:1. (Goodall and Gregory, 19h?) excess may increase the ratio of its internal concentration to that of the "limiting" nutrient without affecting either the internal con- centration of the limiting nutrient or the yield. It appears that simple ratios, as distinct from the ratios devised by the foliar diagnosis school, can only serve as supplementary to those indications given by the actual concentrations. Richards (l9h4) be- lieved that the use of such ratios, without consideration of the con- centration of the individual components, is unjustified in most cases. He cites as an example that the results of Hunter, Toth and Bear (l9h3) concerning the calcium and potassium nutrition of lucerne are as readily explained on the basis of the individual percentages of these two cations in the plant as by their ratios. Muster (1919) stated that wheat was deficient in nitrogen if the P205:N'ratio in the straw was greater than 0.6, but Goodall and Gregory (1947) pointed out that his figures showed that the correlation of yield increment due to nitrogenroontaining fertilizer with his ratio was only —O.2969, while that with the per- centage of nitrogen in the straw was -O.6375. The fact that ratios give no information that can not be obtained from simple concentrations does not, of course, detract from the fact that the diagnosis of the nutritional status in respect of one element on a basis of its concentration in the plant may not have to be modi- fied according to the levels of the other elements. This has been shown conclusively by Lundegardh (l9hl) whose data have been discussed in the section on Triple Analysis. 310- Since the time of Atterberg (1888a, 1889) most workers have at one time or other expressed the nutritional status of a crop on the basis of the ratios of components in the tissues. This information, although considered to be of little value, has been compiled by Goodall and Gregory'(l9h7) and arranged in tabular form for the differ- ent crops considered. D. Presentation of Data The majority of workers have merely reported or suggested certain values, or ranges, of the nutrient content of crops as representing a theoretical portion of curves of the type develOped by Macy (1936). An example is the suggestion of Boynton and Compton (l9h5) that apple trees would produce an Optimum yield if the leaves sampled in July contained between 1.85 and 2 percent nitrogen. Hill (l9h3) stated that a reduction in the yield of carrots would result if the petiole sampled in the active growth period contained less than 0.0125 perb cent P205 in the fresh.material. Beauchamp, Lazo and Bonazzi (1930) reported that good growth of sugar cane resulted.when the leaves sampled.when the plants were nine months old contained K20 and P205 in the ratio of 3:1. There are a few cases in.which the‘workers have reported.success- ive ranges rather than isolated concentrations or ratios. Heinrich (1882), for instance, presented data showing the probable yield of oats in relation to the nutrient content of the roots at maturity. His . data covered successive ranges through low, average, and high yields 311. as indicated in table 146. Borden (1936) reported four levels of the potassium and phosphorus contents of the Juice of sugar cane as indi- cative,of the various levels of deficiency tabulated in table 47. He does, however, state that the results probably are only reliable for the particular variety of sugar cane that he used and then only under similar growing conditions. This method of presenting data has not had wide usage. This probably is due to the fact that investigators either seek to determine the critical levels and to bypass the inter- mediate gradations, or because they fail to visualize plant relation- ships as continuous, rather than discontinuous, functions. Probably the best approach to the use of tissue analyses for the determination of the effects of several nutrients on yield is that suggested by Nicholas (1948). This worker shows interrelatibnships of the mineral nutrients by plotting their concentrations along the radiating spokes of polygonal figures. Theoretically, there can be as many ”spokes" as there are influential elements, but Nicholas has as yet considered only N, P, K, Ca, Mg and Mn. Using a large amount of data obtained by tissue analysis, he estimated the concen- tration of these elements corresponding to normal growth and high yields, and estimated the “threshold” values, or values below which deficiency symptoms occur. Such a diagram is presented in figure 61, in which the healthy plant is represented by isometric "spokes” for each element, which results, of course, in a symmetrical figure. Intensity factors are then represented by the length of the spokes and 12. Table #6. Probable yield of oats in relation to the nutrient content of the roots at maturity expressed as percentages of the_dgy matter. (Heinrich, 1882) Nutrient Probable yield of oats ‘N 0.5-0.6 0.7-0.9 0.9 or more P205 0.1-0.2 0.2—0.3 0.3 or more K20 0.1-0.2 0.2—0.h 0.u or more Cao ’ o.2-o.3 o.3—o.u o.u or more Table 47. Percentages of nutrients in the crusher Juice of sugar canglat different levels of nutritional deficiency. (Borden 193 Level of Percent Percent deficiency P205 K20 Low 0.025 0.15 Doubtful 0.030 0.20 Medium 0.035 0.25 High 0.045 0.35 3/3, CAULIFLOUER nt-Le-ne-ue IL‘AAAA'AI‘ 11T- vawu nuns Mull pm amt. m THRESHOLD or crummy Figure 61. Diagrams showing the relationship between soluble potassium,magnesitm, calcium, phosphorus, and nitrate nitrogen in the potato and in the cauliflower growing under various fertilizer treatments. The mean of six tissue test values from NPK, NP, and NK treatments are plotted along five radii of a circle. (Nicholas, 1948) quality factors by the outline of the figure. At some place in the periphery lies the deficiency threshold region. The unsatisfactory nutritional condition is seen by the distortion of the outline due to the unequal length of the spokes. Thus, low nitrogen is often accompanied by low magnesium. This approach of Nicholas possesses advantages over foliar diag- nosis techniques because it is not restricted to the three major ele- ments, and the relationships are not shown by a single point in,a triangular system.' It possesses the disadvantage that it is not able to show the seasonal trends as readily as does the foliar diagnosis system. E. Discussion Of all various methods for determining the fertilizer requirements of crops, the use of critical concentrations probably rests on the weakest experimental foundation. The current literature is replete ‘with ”critical" and "Optimum“ concentrations as evidenced by the monograph of Goodall and Gregory (l9h7), but they are of little real value except to the individual investigator in his own area because they differ under different conditions. This resulting confusion can be of little assistance in solving the perplexing problem of determine ing the fertilizer requirements of craps. The greatest weakness of the critical concentration concept is 315. its failure to permit the estimation of the amount and kind of ferti- lizer required to give optimum crap yields. Since the determination of critical concentrations is so rapid it can be used effectively in survey studies and it is in this category that these data must be placed in relation to the other and superior techniques. Critical concentrations may also serve a purpose in the study of seasonal trends of the nutritional needs of crops and they may suggest the advisability of applying side dressings, nutrient sprays, etc., but . in this capacity they will be only of supplementary value to other techniques which can more authoritatively estimate the major fertilizer requirements. The use of critical concentrations should be completely re—examined experimentally to determine whether the use of this technique has been a detriment or an asset to agricultural research. Research.workers in the field of plant nutrition and fertilizer use should divorce them- selves from techniques designed for rapidity rather than for accuracy and they should attempt to develop new methods, or to re-examine and improve older methods in order to find more accurate procedures. for it cannot be denied that, even.with the deve10pment of more rapid and less tiresome techniques, our agricultural soils in many areas are deteriorating more rapidly than these quick methods are restoring them. 316. II. LEAF PAINTING.AND INJECTION A. Introduction The advocates of leaf painting and injection techniques have implied that they can, in a limited way, be used to determine the fertilizer requirements of crops and consequently they merit consi- deration. The diagnostic use of paint injection has been reviewed at length by Roach (1939), and by Roach and Roberts (1916). These workers have proposed these techniques as routine diagnostic methods. particularly as supplements to the use of visual symptoms. In this connection, Roach (1918) states that for "the diagnosis of mineral deficiencies several methods are necessary to suit different circum— stances: the method of leaf injection is to be used in conjunction with visual symptoms.I This seems like a weak Justification. B. Technique Roach and Roberts (19h5) proposed a series of solutions to be used for diagnostic purposes. They state that these solutions in the concentrations recommended are applicable over a wide range of craps but that it may be necessary to modify them in rare cases. The solutions recommended for leaf injection techniques by Beach are as follows, but in some instances the addition of sulphuric acid may be necessary to prevent precipitation. 317. Element Concentration Source (75) N l Urea K 1 K01 Ca 1 CaCl2 mg 0.5 Mg504.7HQO Fe 0.025 reson+o.025% (by volume) £2304 Mn 0.025 MnSOLl-t " ' ' ' Zn 0.025 znsohfiy . u u . n Cu 0.025 Cuson—r " " " Ni 0.025 Ni804+. " “_ " B 0.1 Boric acid The injection procedures discussed briefly below are described in detail by Roach (1939, l9u3, 1943a) and by Roach and Roberts (1916). In the interveinal method, the solution containing the element to be tested is held in a thimble—shaped vessel made from a cellophane drinking straw containing a small strip of filter paper glued inside. A slit is made in the interveinal area of the leaf and the end of the filter paper pushed through it acting as a wick. If a deficiency is present, there will be a localized improvement in the color of the nearby mesophyll cells. The leaf stalk technique is the most highly recommended by Roach and Roberts (19MB). It involves cutting off the blade and inserting the petiole into a glass vial containing the testing solution. Depend- ing upon the vascular system of the plant, varying effects on the color 318. of the leaves above and below the treated petiole will occur. In apple, pear, plum, and peach, the two leaves above and below the treated petiole are infiltrated with the solution on the side nearer the cut petiole, but not on the further side. Thus in four leaves, comparisons can be made between Opposite sides of the same leaf. If on the other hand, only the tip of a leaf is removed part of the blade or all of it will be affected by the solution depending upon how much of the tip has been removed. . I I In the branch method, the branches of a tree may be injected each with a different nutrient solution. The solution is fed to the tree through a piece of rubber tubing leading from the base of the container to a hole bored in the branch. This technique permits the same tree to be used for studies of several possible deficiencies. There are two methods by which the test solutions may be applied to the entire tree. One is the liquid method which involves supplying the solution through tubing leading into holes, one—fourth inch.in diameter, bored in the tree. In the other method, the dry nutrient salt is forced into small holes at three inch intervals around the circumference of the tree and the hole plugged with a cork. The leaf painting method was studied by Roberts (l94b) in his attempt to overcome the slow uptake of nutrients by fruit trees when treated by the injection method. He simply painted the leaf with the nutrient solution. Roberts claimed that by using a series of concen- trations, he could ascertain not only the deficient nutrient but also JLJO its concentration required for eliminating the deficiency. 0. Application The injection and painting techniques may be used as diagnostic procedures or as curative procedures. As far as diagnosis is concerned, the techniques have had their greatest use in the study of leaf chloro- sis of one form or another. The nutrient that corrects the chlorosis is the nutrient that is deficient, but in some instances, the response may be a local increase in size or a general increase in growth or yj. elde Corkybcore of apples was diagnosed as boron deficiency by the use of injection techniques, by Atkinson (1935), MCLarty (1936), and Young and Bailey (1936), and the'wither-tip of apples in Western Australia was diagnosed as capper deficiency by Dunne (1938). The injection of test solutions into the plant or spraying them on the foliage is the quickest way of getting temporary recovery although.permanent recovery is not likely. Roach (l93h) reported that commercial apple trees in average health and vigor made twice as much growth after injection with KQHP04 and urea, each at the rate of about 28 pounds per acre, as did untreated trees and they produced leaves thicker and healthier in appearance. In a further test, Roach (1939a) injected peach trees with manganese valued at four cents per tree, and was able to correct the deficiency. The trees previously had failed to produce a crop but produced a good average cr0p after the treatment. 320. D. Discussion Probably the worst that could be said of leaf painting and injec- tion techniques as far as their application to the fertilization of crops is concerned is that they are aimed at saving the cost of the fertilizer required by the immediate crop. Such a practice will do more damage to agriculture than good, for it eliminates, temporariLy at least, the necessity for improving the soil. 0n the long time basis, it is more important to understand the soil and to improve it than to produce a crop. Like many other types of tissue techniques, these methods permit only very vague conclusions as to fertilizer requirements, and cannot determine quantitatively the actual amounts of fertilizer required by crops. These methods cannot be considered more valuable than visual symptoms for they are not even applied until visual symptoms have appeared. The only significant use of these procedures is for the identifi- cation of specific deficiencies, especially of the minor nutrients. XII. VISUAL SYMPTOMS A. Introduction Visual deficiency symptoms have been used for the determination of the fertilizer requirements of a crop. The nutritional deficiencies are recognized by specific symptoms, mainly of the foliage, exhibited if the supply of one or more nutrients is insufficient to promote healthy growth. Before this method may be applied, the specific symptoms of each nutrient deficiency for each kind of plant must be recognizable. These specific symptoms have been determined by growing plants in artificial media under controlled conditions. Many color plates illustrating these deficiencies have been.pub- lished by the Chilean Nitrate Educational Bureau (19u1), wallace‘(l943, 19th), Cook and Miller (19u9), Bambidge (19u9), and others. B. Application Visual symptoms are intended for use in connection with other confirmatory methods for determining the nutrient status of the crap and are of no value when used alone. The various color atlases cited previously can be used for identifying the simple deficiencies, but not even the most experienced worker can interpret the symptoms of multiple deficiencies accurately in the majority of cases without resorting to confirmatory methods. Contrary to general opinion, all plants do not exhibit recognizable specific deficiency symptoms and this seriously limits their use. Only certain plants may serve as indicators of particular deficiencies. For example, the growth of sunflower plants grown under controlled conditions has been suggested by Stephenson (l9h0) and Colwell (19u3) as a means of assessing the boron status of the soils. ' Schemes in which special indicator plants and fertilizer treat- ments are combined have been proposed by Piper (19h0) and wallace (l9h3) for use in the fields where special difficulties are encountered or where an area of unknown cropping potentialities is being tested. These schemes aiming at solving during a single season the main defi- ciency problems likely to exist may have some value but they cannot give more information than adequate soil analyses. wallace (l9h3) suggested the use of certain indicator plants for diagnosing specific deficiencies. He arranges these species in groups as follows: Group Indicator plants Mineral deficiencies indicated 1 Cauliflower, broccoli N, Ca, Mg, Fe 2 Potato K, Mg 3 Sugar beet, mangold, globe beet 3, Mn it Pape, Swede P 5 Oats Mn’ The following fertilizer treatments are applied to the plot: NPR NI NPKB NP . NPKMg NPK Mn 323. If the effects of liming are to be studied, the experimental plan will be as indicated in figure 62 which shows the arrangement of plots. The crops are sown in strips running the length of the area to facilitate cultural operations and the fertilizers are applied as indicated by the cross-hatched areas. If the effect of lime is not studied, only two blocks would be planted. Wellace believed that one might obtain a fair estimation of the deficiencies in the soil by observing the visual symptoms evident from the experimental plots. 0. Discussion Frequently the symptoms produced by a deficiency of a specific nutrient on a particular species are specific, but in many cases this is not the case. 'The symptoms may vary among different species and sometimes even among varieties as described by Wallace (l9h3) and by 11111 and Johnston (19%). There are circumstances described by Wallace (l9fi3) in which deficiency symptoms may be masked entirely by pests and disease, such as eelworms or virus diseases in potatoes, and Haas (193?), Spencer and Levin (1939), and Wallace (l9b3) have described symptoms produced by pests, diseases, mechanical injury, or weather conditions that may be indistinguishable from certain mineral deficiency symptoms. For example, leaf symptoms on young cereals and §3§§§i9g_plants may be identical for wire worm, cold.weather, root injury, or phosphorus deficiency. Symptoms of chloride injury may be, and often are, almost 324. BLOCK I BLOCK II Grow 50 mt. Group 2. Potato» £9 a: é a Grow 3. Sugar boots or Molds E a g E E E g Group lo. Cauliflower or Brocolli Group 1. hp. or and” Group 5. Grow 2. 5'? In é m 3 s e a s s s E Group ‘0 Group 1. BIDCI III BLOCK IV Figure 62. Field plan for determining soil deficiencies using indicator plants and visual symptoms. (Wallace, 191+3) 325. identical with those of phosphorus deficiency as shown by Wallace (l9b3) on red currant foliage. Symptoms of certain deficiencies are very similar and difficult to distinguish. Thus, chlorosis may be the symptom produced on apples, plums, and raspberries by deficiencies of iron, manganese, or nitrogen. The value of visual symptoms also are limited by the fact that in the majority of cases only the most severe deficiency is apparent, and the effects of inadequate supplies of other nutrients are masked as described by McMurtrey (1949). Finally, in most instances, nutrient deficiencies may cause de- creased yields without inducing visible symptoms as has been.pointed out by Hill (19h3), Davis (1950), Batjer and Digman (l9h0), and Berger and Truog (19b0). ' There seems to be no possibility of using visible nutrient defi- ciencies for indicating a quantitative requirement of fertilizer. There is some value to the technique, however, for if a nutritional disorder can be diagnosed in the field, there may yet be time to correct the situation in the same season. The conclusion that one must draw conerning the use of visual symptoms is that they may be of some value in the immediate correction Of an apparent deficiency but no reliance can be placed on them without proper confirming techniques. Thewaill point to acute nutritional deficiencies until such time as a more thorough study of the problem can be carried out by more acourate methods, but surely the farmer has 326. failed to care for his soil if he waits until his crops exhibit visible symptoms of deficiency before he uses fertilizer. 327. XIII. WEEDS The use of weeds as indicators of the fertilizer requirements of crops was suggested as early as 1882 by Heinrich.who considered that the fertility of the soil might be measured by growing plants adapted to that particular soil and analyzing them to obtain indications of the nutrient content of the soil. In concept, this proposal essentially is a type of tissue testing.~ Weeds as indicators of soil fertility were first investigated by Hall (1905) whose work involved a compari-r son of the composition of the plants with that of the citric acid ex- tracts of soil. He concluded that a universal test plant, such as a weed, was needed and that until such a plant was found, the interpre- tation of soil conditions from plant ash analysis was not practical as a substitute for chemical analysis of the soil. Ecologists from time to time have suggested that the botanical composition of the natural flora or the weed flora could serve as an index of the nutritional state of the soil. Goodall (1919) analyzed weeds and compared their potassium and manganese contents with the requirements of barley and wheat for these substances. For example, figure 63 shows the correlations observed in the case of Eolygonum gonvolyglug L. This figure represents the best correlation he obtained and even in this case there was no significant difference between site means, and with the exception of one station, the means of plots at a single station also were insignificantly different. There was, however, a highly significant correlation between the composition of the weeds and crop response to fertilizer when the data from all plots were grouped together as shown by figure 63. From these results, Goodall concluded that no response to potassium fertilizers could be expected if the percentage of potassium in the dry matter of the leaves of Polygonum convolvulpg L. was greater than 1.83, but Goodall believed that his results were so inconclusive that no value could be gained from this approach. A further example may be found in the work of Christensen and Larsen (1910) who considered that the presence of such species as Scleranthus annuu§_L. and figgg; acetosella L. indicated the necessity of lime for optimum crap production. Although the distribution of a species may be generally dependent upon the nutritional status of the soil, such data are of little prac- tical value for determining the magnitude of fertilizer requirements of crOps. 317. 1? C 2 2.; < 0 K x O a k ’ E o ‘27“ m"- x c: 3 u 1 < 3 : ... B K .. d M o *4 3 a, a f3 g o g -. r -* l A 0.! a-o 1.! go 3.: PERCENT POTASSIUM IN MY WTTEI OF LEAVES OF lflulfiflflflfil SflnflflE¥flhfli Figure 63, The relationship between the increase in grain yield of barley fertilized with potassium.fertilizer and the potassium concentration in the leaves of Po onum convolvulus L. at three different locations. (Goodall, 19h9) 330. XIV. CHLOROPEYLL Foliage of plants deficient in nitrogen has a lower intensity of green color than does the foliage of normal plants. This observa~ tion has suggested to a number of workers that the chlorophyll content might be used for assessing the nitrogen requirements of crops. The earliest attempt to do this was that of Gassner and Goeze (1936). Eye seedlings were grown in pots of soil and sand to which different amounts of nitrogen.were added. After a growth period of 25 days, the chlorophyll was determined and the nitrogen requirements of the soil were estimated. ‘ More recently workerstin the United States have adopted this pro- cedure and have attempted to apply it in practice. For example, the following workers have contributed to this work: Boynton, Compton.and Fisher (l9h8), Boynton and Compton (19h5), Compton and Boynton (1945), Compton, Granville, Boynton, and Phillips (1946), Boynton and Cain (19u2), Judkins and Wander (1950), and Shear and Horsfall (l9h8). The technique used in these studies for the determination of the chlorophyll content was devised by Comar and Zscheile (19h1, 19b2)‘who applied a colorimetric procedure using wave lengths 6600 g.and 6h25 3 and an adaptation of Beer's Law. Compton and Boynton (19h5) used a series of dilutions of pure chlorophyll for standardizing the method while Heeney (1950) set up standards with fresh material. Compton, Granville, Boynton, and Phillips (l9h6) using McIntosh apples in two different orchards in the State of New York showed that 331. there was a straight line correlation between the percentage of total nitrogen in the dry tissue and the amount of chlorophyll in the fresh tissue. Heeney (1950) carried out a survey of the Standard Orchard at Ottawa which involved three soil types, four kinds of rootstocks and four fertilizer treatments, and found a significant positive correla- tion (r = +-0.6630) between the chlorOphyll content of the leaves and the total nitrogen. In a later investigation, Heeney found a correla— tion of r--+0.9974 in a more uniform fertility block at Trenton. These data are presented in figure 6“. Good correlations also were shown using tomato leaves as shown in figure 65. Compton, Granville, Boynton, and Phillips (l9h6) set up color standards for use by the grower with.which he could compare the leaves of McIntosh trees to determine the adequacy of the nitrogen supply. These standards were shown by Shear and Horsfall (l9h8) to be inade- quate for other varieties or for different locations. The results described appear to justify the use of the chlorOphyll content of plants as indications of the adequacy of nitrOgen supplies as in all cases good correlations were shown, but it should be borne in mind that the work:was done on fertility trials in which nitrogen was generally the limiting factor. If the method is applied to general field surveys, interpretation becomes a more complicated problem. When applied in this manner, the results would doubtlessly be affected by deficiencies of other elements because nitrogen is not the only nutri- ent which affects the green coloration of the leaves. Deficiencies of 332. '9 a '3‘ '1 Y P 9s v '9 u v '9 # mg. cmonopyyu IN/JOJy cm. APPLE (EA! "350: A n 1 [-0 2.0 1.2. 1.9- rorAL NIrRosr/V £xPR£33£D A: PIRCENr DRY MArrER ~ 9 5 P . Figure 61.. The relationship between total nitrogen and chlorophyll content in McIntosh apple leaves in Hendrick'e Orchard, 191.9. (Heeney, 1950) 333. p o 3! erg. cm. agar/(ya. IN 130:, cm. TOMATO (an: A A A Q A A 1809 «so A «no «no 904.005.: mrnocew m pm 7 Figure 65. The relationship between soluble nitrogen and chlorophyll content in tomato petioles grown at Smithfield 1950. (Hesney, 1950) 33“. iron, magnesium, sulphur and various other nutrients also induce chloro- sis. The interaction of these other deficiencies will affect the prac- tical value which might be attributed to the chlorophyll techniques. A further obvious objection to the use of subtle color comparisons is that of their subjective nature, and Judgments of color differences are as variable as the number of observers. This is especially impor- tant as the color variations correlated with adequate, deficient, or excess nitrogen are very slight. Judkins (l9h9) and Judkins and Wander (1950) attempted to overcome some of this error by using a photoelec- tric reflection meter. This, of course, removes the error due to varia- ble interpretation of the color charts but does not in any other way improve the value of the technique. There is no real value in this method for it is the estimation of a factor (chlorophyll) which is used to estimate a second factor (nitrogen) which is then used to estimate the third factor of the nutrient status of the crop, so that finally an estimation can be made of the fertilizer requirements of the crOp. Any method involving so many successive estimations will be so subject to error as to be of little practical value. 335 . XV. SUMMARY Confusion in the theory of the responses of creps to fertilizer is a major factor obstructing agricultural progress. An understanding of the reasons for this confusion is necessary for the future improve- ment of the productivity of agricultural soils. Consequently, a de- tailed study of the original literature upon which present theories of fertilizer use are based is essential. The overthrow of the humus theory by Liebig, and his recognition of the importance of mineral nutrients in crop production resulted in the initiation of our current concepts and increased interest in the nature of soil fertility and of cultural practices producing maximum crop yields. Although the problem of relating the numerous factors affecting nutrient concentrations in the soil and in plant tissues is very complex, a number of mathematical empressions relating these factors have been proposed. A study of the experimental data on which these expressions were based is essential to a correct understanding of fertilizer use. The most important of these empressions, the Law of Diminishing Returns, in its implications and applications presents the most direct and.practical method of studying efficient quantitative fertilizer use. This Law, however, has been severely criticized because it implies a mathematically expressed biological constant. This criticism does not seem justified because any method in.which results of field trials, tissue tests, or other techniques are interpreted so as to permit 335. general field recommendations must imply such a constant whether or not it is mathematically expressed. The applications of the Law have been replaced largely by tissue analysis at the present time, but the Law is based on a much sounder foundation theoretically than currently accepted procedures for the determination of the fertilizer requirements of crops. The techniques of Willcox, based on the Mitscherlich constant expressed in Baule units is a direct application of the Law. Bray, on the other hand, has ine troduced modifications which permit the application of the mathematical expression of the Law but does not imply the general acceptance of the constancy of a biological factor over widely different conditions. Techniques of this type are the only methods yet devised for the queue titative estimation of the fertilizer requirements of crops. In order to preserve our agricultural soils they should be given prime considerae tion in investigations of the use of fertilizers. Various techniques have been suggested including biological methods of analyzing small samples of soil. These techniques cannot serve any useful purpose, except in special cases, until such time as it can be demonstrated that they are superior to chemical methods of soil analysis. Further, they cannot be used for the accurate quantitative determination of the fertilizer requirements of crops. The techniques involving the use of the plant, leaf painting and inJection, deficiency symptoms, and chlorOphyll determinations can be used qualitatively only, even by the most experienced investigator, because of the difficulty of interpreting differences observed under 337. normal and deficient conditions. It is also apparent that the most commonly used techniques based on the analysis of the plants, such as critical concentrations and ratios, foliar diagnosis, and triple analy- sis, cannot quantitatively estimate fertilizer requirements of present or future crops. Many of these methods are, however, of considerable value in studying the seasonal trends of crop nutrition and as rapid methods of conducting preliminary surveys of soil fertility over wide areas. 33EL iDIL LITERATURE CITED Adorjan, J. 1902. Die Nahrstoffaufnahme des Weigens. J. Landw. 50:193-230. Alway, F.J. 1928. Detection of sulphur deficiency of soils by means of plants. lst. Int. Congr. Soil. Sci. 3:590-613. , and Shaw,'w.M., and Methley, W.J. 1926. Phosphoric acid content of crops grown on peat soils as an index of the fertilizer received or required. J. Agric. Res. 33:701-7h0. Ames,J.W., and Gerdel, R.W. 1927. Potassium content of plants as an indicator of available supply in the soil. Soi1.Sci. 23:199-223. , 1927a. The seedling plant method of determining soil nutrient deficiency. Soil Sci. 23:h55-h66. Appleton, J.M., and wynd, F.L. 1951. A preliminary study of the application of the percentage yield concept to the response of forage crops to irrigation water. Sci. Agr. 31:139-147. Arendt, R. 1859. Untersuchungen uber einige Vorgange bei der Vegetation der Haferpflanze. Landw. Versuch—Sta. 1:31-66. 1859a. thiz fiber die Verschiedenheiten in der Zusammenstezung der Asche desselben Pflanzenspecies bei gleichern Alter, aber verschiedener ausserer Entwickelung. Landw. Versuch- Sta. 1:66-68. Atkinson, J.D. 1935. Progress report on the investigations of corky pit of apples. N.Z. J. Sci. Tech. 16:315-319. Atterburg, A. 1386. Die Beurthleilung der Bodenkraft nach der Analyse der Haferpflanze. Landw. Jb. 15zh15-L19. 35*9. 1887. Die Beurtheilung der Bodenkraft nach der Analyse der Haferpflanze. Landw. Jb. 16:757-761. . - 1887a. Kalmar Ken. Sta. Arsberatt. (in Atterberg, 1901) 1887b.Landbr. Akad. Tidskr. (in Atterberg, 1901) 11888. Kalmar. Kem. Sta. Arsberatt. (in Atterberg, 1901) 1888a. Tio-Arsberatt., Kalmar Kem. Sta. 2:13, 1877-1887 (in Helmkampf, 1892.) 1889. Kalmar Kem Sta. Arsberatt,(in Atterberg,l901) 1901. Die Variationen der Nahrstoffgahalte bei dem Haferpflanze. J. Landw. A9:97-172. Balmukand, B. 1928. Studies ih crop variation. V. The relation between yield and soil nutrients. J. Agr. Sci. 18:602-627. Bartholomew, R.P., watts, V.M., and Janssen, G. 1933. The effect of variations in the nutrient media upon the nitrogen, phosphorus, and potassium content of plants with special reference to the tomato. Bull. Ark. Agr. Exp. Sta. 288. Batjer, L.P., and Degman, E.S. 19h0. Effects of various amounts of nitrogen, potassium,and phosphorus on growth and assimilation in young apple trees. J. Agr. Res. 60:101-116. Baule, B. 1918. Zu.Mitscherlich's Gesetz der physiologischen Beziehungen. landw. Jb. 51:363-385. Beauchamp, C.E., and Alvarino, J.E. 19AO. The mineral composition of the leaves of different sugar cane varities and its possible relationship with the efficiency of their mineral nutrition. Proc. Asoc. Tech. Azuc., Cuba. lhzh5-55. Berger, K.C., and Truog, E. 1940. Boron deficiencies as revealed by plant and soil tests. J. Amer. Soc. Agron. 32:297-301. Berzelius, J.J. 18A2. Pflanzenchemie. Jahr. Ber. Fortschr. Chem. Min. 21: p235. (in Browne, l9hh) 34o. Borden, R.J. 1936. Diagnosing the status of available phosphorus and potash in sugar cane soils. Hawaii Plant Rec. 40:3—10. Boussingault, J.E. 1841. De la discussion de la valeur relative des ‘ assolements par 1es resultants de l'analyse elementaire. Ann. Chim. Phys. lll:208—2h6.(in Russel, 1927) Boynton, D. L942. Progress towards a more scientific basis for orchard fertilization. Proc. N.Y. St. Hort. Soc. 87:192-198. , and Burrel, A.B. l9hh. Effects of nitrogen fertilizer on leaf nitrogen, fruit color, and yield in two NeW'York McIntosh orchards. Proc. Amer. Soc. Hort. Sci. hh:25-30. , and Cain, O.C. 1942. A survey of the relationships between leaf nitrogen, fruit color, leaf color, and percent of full crop in some New York MacIntosh Orchards. Proc. Amer. Soc. Hort. Sci. h0:l9-22. , and Compton, O.C. l9h5. Leaf analysis in estimating the potassium, magnesium, and nitrogen needs of fruit trees. Soil Sci. 59:339-351. , , and Fisher, E. l9h8. Further work on leaf nitrogen and leaf color as measures of the nitrogen status of frhit trees. Proc. Amer. Soc. Hort. Sci. 52:h0—h2. Bray, R. 19th. The potash problem in Illinois. Better crops with plant food 28: 8—16,l+2-l+l;. 19hha. Soil plant relations. II. Balanced fertilizer use through soil tests for potassium and phosphorus. Soil Sci. 60:h63-473. 1948. Correlation of soil tests with crop response to 3"+/. added fertilizers and with fertilizer requirement. Diagnostic Techniques for Soils and Crops. Amer. Pot. Inst., Wash., D.C. 53-87. Briggs, G.E. 1925. Plant yield and the intensity of external factors. Ann. Bot. 155:475-502. Browne, C.A. 19th. A source book of agricultural chemistry. Chronica Botanica 8(1):l—290. Burd, J.S. 1919. Rate of absorption of soil constituents at various successive stages of plant growth. J. Agr. Sci. 18:51-72. Capo, B.G. 1938. A modification of Mitscherlich's method for the determination of the nutrient contents of a soil. J. Dept. Agr. , Puerto Rico 22:137~169. Carolus, R.L. 1933. Tomato fertilization. II. The effect of different fertilizer ratios on the chemical composition of tomatoes. Bull. Va. Truck Exp. Sta. 81:1085-1117. 1933a. Some significant variations in the chemical composition of the plant associated with a malnutrition trouble of potatoes. Amer. Pot. J. 10:1h7-165. 1935. Experiences with rapid chemical tests for the determination of nutrient deficiencies in vegetable crops. Proc. Amer. Soc. Hort. Sci. 33:579-583. 1937. Chemical estimations of the weekly nutrient level of a potato crop. Amer. Pot. J. 1h:1Al—153. 1938. The use of rapid chemical plant nutrient tests in fertilizer deficiency diagnosis and vegetable crop response. Bull. Va. Truck Exp. Sta. 98:1531—1556. 342. Chaptal, J.A.C. 1301. Elements of chemistry. 2nd. Amer. Ed. Philadelphia.(in Browne, 1944) Chilean Nitrate Education Bureau, Inc., New York. 1941. If they could'speak. Colwell, W.E. 1943. A biological method for determining the relative boron contents of soils. Soil Sci. 56:71-94. Comar, C.L., and Zscheile, F.P. 1941. Spectroscopic analysis of plant extracts for chlorophylls a, and b. Plant Physiol. 16:651-653. , 1942. Analysis of plant extracts for chlorophylls a and b by a photoelectric spectrophotometric method. _ Plant Physiol. 17:198-209. Comber, N.M. 1922. The availability of mineral plant food. A modification of the present hypothesis. J. Agr. Sci. 12:363-369. Compton, O.C., and Boynton, D. 1945. A rapid method for determination of chlorophyll in apple leaves. Proc. Amer. Soc. Hort. Sci. 46:45-50. , Granville, w.c., Boynton, 1)., and Phillips, E.S. 1956. Color standards for MacIntosh apple leaves. Cornell Agr. Exp. Sta. Bull. 824. Cook, R.L. 1930. Effect of spil type and fertilizer on the nitrate content of the expressed sap and the total nitrogen content of the tissue of small grains. J. Amer. Soc. Agron. 22:393-408. , and Miller, C.E. 1949. Plant nutrient deficiencies diagnosed by plant sympyoms, tissue tests, and soil tests. Mich. Agr. Exp. Sta. Bull. 353:1-80. Craig, N. 1938. Part II. Chemistry. Rep. Sugar cane Res. Sta. Mauritius 8:33-46. 3‘¥3. 1939. Part II. Chemistry, Foliar diagnosis. Rep. Sugar Cane Res. Sta. 9:30-37. 1941. Part II. Chemistry. Rep. Sugar Cane Res. Sta. 11:11-17. 9 Crowther, F. 1936. The effects of variety, Spacing, nitrogen, and water supply on the developement of the cotton plant and the ; rate of its absorption of nitrogenous fertilizer. Bull. Tech. Sect. Soc. Agr. Caire 25:1—49. (in Goodall & Gregory, 1947) Dahlberg, H.W., and Brown, R.J. 1932. A study of the Neubauer and 'Winogradsky(Azotobacter)methods as compared with a chemical method for the determination of phosphorus deficiency in ‘western soils. J. Amer. Soc. Agron. 24:460-468. Daniel,A.A. 1934. The calcium, phosphorus, and nitrogen content of grasses and legumes and the relation of these chemicals in the plant. J. Amer. Soc. Agron. 26: 496-503. Davis, M.B. 1950. Progress report 1934-1948. Division of Hort. C.E.F. Ottawa. Davy, H. 1813. Elements of agricultural chemistry in a course-of lectures ' for the board of agriculture. Longman, Hurst, Rees, Orme, and Browne, London, (in Browne, 1944) von Dickow, A. 1891. Beurthleigung des Bodens nach den Wurzeln der Gerstenpflanze. J. Landw. 39:134-147.(in Goodall & Gregory,l947) Drake, M., and Scarseth, G.D. 1940. Relative abilities of different plants to absorb potassium and the effects of different levels of potassium.on the absorption of calcium and magnesium,- Proc. Soil Sci. Soc. Amer. 4:201-204. Drszewski, __, and Tollens,B. 1900. J. Landw. 48:223.(in Tollens, 1902) Duley, F.L., and Miller, M.F. 1921. The effect of varying supply of nutrients upon the character and composition of the maize plant at different periods of growth. Mo. Agr. Exp. Sta. 3 44. Tech. Bull. 42. Dunne, T.C. 1938. "Wither-tip" or "summer-dieback". A copper deficiency disease of apple trees. J. Dept. Agr.'w. Aust. 15:120-126. Eaton, F.M. 1944. Deficiency, toxicity, and accumulation of boron in plants. J. Agr. Res. 69:237-277. Emmert, E.M. 1935. New methods for the determination of the availability of nitrogen andphosphorus to plants. J. Amer. Soc. Agron. 27:1-7. 19353. The correlation of soluble nitrogen and phOSphate phosphorus in conducting tissues of potatoes at various stages of growth and yield. Proc. Amer. Soc. Hort. Sci. 33:589-594. 1937. The use of partial linear regression to analyse the curvilinear relationship between yield of vegetable crops and the content of nutrients in the lower main stem. ' J. Amer. Soc. Agron. 29:213-219. 1941. Plant tests as a guide to fertilizer treatment of tomatoes. Proc. Amer. Soc. Hort. Sci. 38:621-622. 1943. Plant tissue tests as a guide to the fertilizer treatment of potatoes. Bull. Ky. Agr. Exp. Sta. 430:1-48. 1949. Tissue analysis in diagnosis of nutritional troubles. J. Amer. Hort. Soc..54:29l. Farden, C.A., and Magistad, O.C. 1932. Use of the yield equation in pineapple culture. J. Amer. Soc. Agron. 24:964-975. Fisher, R.A. 1934. Statistical methods for research workers. Oliver, and Boyd, Edinburgh, 5th Edition. Fraps, G.S. 1931. How reliable are the existing methods for determining soil deficiencies in ash constituents of plants. J. Amer. Soc. Agron. 23:337-351. 3%5. Fuller, J.E. 1934. Application 'of the Azotobacter soil-plaque test for determining deficiencies in Massachusetts soils. Mass. Agr. Exp. Sta. Bull. 305:15-16. 1935. The Azotobacter soil-plaque test for determining mineral deficiencies. Mass. Agr. Exp. Sta. Bull. 315. Gassner, G., and Goeze, G. 1936. Versudhe zur Bestimmfing des aufnehm- baren Bodenstickstoffs durch Bestimmung des Chlorophyll- gehaltes. Ergebn. Agrik. Chem. 4:106-122. Gericke,S. 1947. Investigations on the Law of Yield. Ztschr. Pflanzenernahr. Dfing. u. Bodenk. 38:54-65, 215-229, 245-258. Goodall, D.W. 1945. Studies in the diagnosis of mineral deficiency. II. A comparisom.of the mineral composition of schorched and healthy leaves from the same apple tree. J. Pom. and Hort. Sci. 21:90-102. , and Gregory, F.G. 1947. Chemical composition of plants as an index of their nutritional status. Imp. Bur. Hort. and Plant. crops. Tech. Comm. 17. Goulden, C.H. 1939. Methods of Statistical analysis. John Wiley and Sons, Inc. New York. Greaves, J.E., and Carter, B.G. 1923. The influence of irrigation water on the composition of grains and the relationship to nutrients. J. Biol. Chem. 58:531-541. , and Nelson, D.H. 1925. The iron, chlorine, and sulphur contents of grains and the influence of drainage water upon them. Soil SCi. 19:325-333. Greene, R.A. 1932. The applicability of the Azotobacter(p1aque) method for determining the fertility requirements of Arizona 3‘1'6. soils. Soil Sci. 34:83-93. Gregory, F.G. 1928. The differential effect of ions of three salt solutions on the growth of potato plants in sand culture. Proc. Roy. Soc. London. B102:31l-327. 1937. Mineral nutrition of plants. Ann. Rev. Biochem. 6:557—578. Guittoneau, N.G. 1929. Sur 1'application des besoins des sols en elements fertilisants par les cultures spontane’es d' azotobacter. C.R. Acad. Agr. Fr. 15:83-88. Gunther, E. 1926. Kritsche untersuchungen‘fiber die Keimpflanzenmethode von Neubauer. Ztschr. Planzenernahr. Dung. u Bodénk. 85:32-36. Haas, A.R.C. 1937. Boron deficiency effects similar in general appearance to bark symptoms of psorosis in citrus. Soil Sci. 43:317-325. Haessner.__ 1887. Untersuchungen fiber den Nahrstoffgehalt in den Nurzeln und Kornern der Gerste und Verhalten desselben zu den im Boden vorhandenen assimilierbaren Pflanzennahrstoffen. (in.Helmkampf, 1892) . Hall, A.D. 1905. The analysis of the soil by means of the plant. J. Agr. Sci. 1:65-88. ' Halverson, 1.V., and Hoge, W.G. 1942. The azotobacter plaque test as applied to the determination of phOSphate deficiency in Idaho soils. J. Amer. Soc. Agron. 34:503-512. Hambidge, G. (Edited by) 1949. Hunger Signs in Crops. Amer Soc. Agron. and Amer Fert. Assoc.,‘Wash., D.C. 2nd. Edition. Hance, F.E. 1936. Soil and plant material analysis by rapid chemical methods. Hawaii Plant Rec. 40:189-299. 1937. Soil and plant analysis by rapid chemical methods. II. Hawaii Plant Rec. 41:135-186. 1941. Soil and plant material analysis by rapid Chemical 347. methods. III. Hawaii plant Rec. 45:265-296. Iiardy, F., and McDonald, J.A., and Rodriquez, G. 1935. Leaf analysis as a means odf diagnosing nutrient requirements of tropical orchard plants. J. Agr. Sci. 25:610-627. Iieadden,'w.P. 1921. Titanium, barium, and lithium in certain plants. Col. Agr. Exp. Sta. Bull. 267. Iieeney, H.B. 1950. The utilization of colorimetric determination of chloro- phyll content of leaf tissue for nitrogen determination. C.E.F. Ottawa (unpublished data) Heinrich, R. 1882. Grundlagen zur Beurtheilnng der Ackerkrume in Beziehung auf Landwirthschaftliche Production. Wismar. (in Spillman.& Lang, 1924) Hellriegel, H. 1867. Uber das Kalibedurfnis de Gerste. Jber. Fortschr. Agrik. Chem. 10:117-119. 1869. Bedeutung der chemischen Untersuchung der Ennteproducte, namenlich der Aschenanalysen, furdie Beurtheilung der'Menge und des gegenseitigen Verhaltnisses der im Boden vorhandenen aufnehm- baren Pflanzennahrstoffe. Landw. Versuch-Sta. 11:136-140. 1893. The experimental station at Bernburg Germany and its method of sand culture. Exp. Sta. Rec. 5:749—774. . . 1893a. The manurial needs of the sugar beet, Abstract in Exp; Sta. Rec. 5:233-234. , and hilfarth, H. 1888. Untersuchungen uber die stickstoffnahrung der Gramineen und Leguminosen. Zeit. des Vereins Rubenzucker Ind. 1888. , I , Romer,__, and Wimmer,__ 1898. Vegetationsversuche fiber den Kalibedurf einger Pflanzen. H. Arb. Deut. Landw. Gesell. Vol. 34. Helmkampf, A. 1891. A new method for determining fertilizer requirements. 348. of soils. Hann. Landw-u-Forstw. Ztg. p. 683 (Abstract in 1892. Untersuchungen‘uber die Feststellung des Dungungs- 'bedfirfnisses der Ackerboden durch die Pflanzenanalyse. J. Landw. 40:85-183. van Helmont, J.B. 1652. Ottus Medicinae. Amsterodami, apud Ludovicum Elzevirium (in Browne, 1944) Hermbstadt, S.F. 1804. Archiv der Agriculturchemie. Vol 1-7. Berlin. (in Browne, 1944) Hester, J.B. 1935. Interpreting rapid chemical soil tests for phosphorus for vegetable crops. Proc. Amer. Soc. Hort. Sci. 33:584-588. Hill, H. 1950. Malnutrition symptoms and plant tissue tests of vegetable crops. Better Crops with Plant Food 27:6-10, 44-45. , and Cannon, H,B. 1948. Nutritional studies by means of tissue tests with potatoes on a muck soil. Sci. Agr. 28:185-199. 1950. Nutrient composition studies on the standard orchard C.E.F. Ottawa. (unpublished data) , and Johnston, F. 1940. Magnesium deficiencies of apple trees in sand cultures and in commercial orchands. Sci. Agr. 20:516-525. , , Heeney, H.B., and Buckmaster, R.W. 1950. The Relation of foliage analysis to keeping quality of MacIntosh and Spy- varieties of apples. Sci. Agr. 30:518-534. Hoffer, G.N. 1926. Testing corn stalks chemically to aid in determining qtheir plant food needs. Ind. Agr. Exp. Sta. Bull. 298. Hunter, A.S., Toth, S.J., and Bear, F.E. 1943. Calcium-potassium 3€P§L ratios for alfalfa. Soil Sci. 55:61-72. Ingen-Housz, J. 1779. Experiments upon vegetation discovering their great power nfpurifying the common air in the sunshine and of injuring it in the shade and at night. London Edition. (in Russel, 1927) Itano, A., and Arkawa, S. 1930. Investigation on'Winogradsky's Azotobacter test as to its applicability to some rice fields and soils in Japan. Ber. Fhara. Inst. Landw. Forsch. 4:365-369. James, w.o. 1931. Studies of the physiological importance of the mineral elements in plants. 11. Potassium: its distribution, movement, and relation to growth in the potato. Ann. Bot. 45:279-293. , and Penston, N.L. 1933. Studies of the physiological importance of the mineral elements in plants. IV. The quantitative distribution of potassium in the potato plant. Ann. Bot. 47:279-293. Januszewski, Z. 1895. Ueber die Pflanzen- und Bodenanalyse in ihrer Bedeutung fur die Bestimmhng der Bodenqualitat. Biederm. Zbl. 28:77-79. Jenny, H. 1950. Contact phenomena between absorbents and their significance in the mineral nutrition of plants in soils. unpublished work. , and Overstreet, R. 1938. Surface migration of ions and contact exchange. J. Phys. Chem. 43:1185-1196. Johnson, M. 1924. Manganese chlorosis of pineapple. Its cause and Control. Haw. Agr. Exp. Sta. Bull. 52. 1 t . i, [with]? .‘lll (All 31$O. Jones, J.O., and Nicholas, D.J.D. 1944. The technique of chemical tissue tests. Progress report I. Rep. Agr. Hort. Res. Sta. Long Ashton. 1944. 79-84. Joulie, H. 1882. On permanent and temporary meadows and pastures and their functions in the economy of agricultural practice. J. Roy. Agr. Soc. Eng. Ser.2, 18:195-222. 1889. Sur la transformation des terres de la ferme d'Arcy. J. Agr. Paris 24:89-95. 1894. Sur la composition et les exigences des cereales. Monit. Sci. Quesneville Ser. 2, 8:641-654, 731-740, 807-833. Judkins, W.P. 1949. The relationship of leaf color, nitrogen, and rainfall to the growth of young fruit trees. Proc. Amer. Soc. Hort. Sci. 53:29-36. , and Wonder, I.W. 1950. Correlation between leaf color, leaf nitrogen content, and growth of apple, peach, and grape plants. Plant physiol. 25:78-85. Kitchen, H.B. (Edited by) 1948. Diagnostic techniques of soils and crops. Amer Pot. Inst., Wash., D.C. Knop, w. 1865. Quantitative Untersuchungen fiber den Ernahrungprccess der Pflanze. Landw. Versuch-Sta. 7:93-107. Knowles, F., and Watkin, J.E. 1931. The assimilation and translocation of plant nutrients in wheat during growth. J. Agr. Sci. 21:612-637. Kosutany, T. 1897. Untersuchungen fiber die Entstehung des Pflanzen- eiweisses. Landw. Versuch—Sta. 48:13-32. Lagatu, H. , and Maume, L. 1924. Evolution remarquablement rggmibm de certains rapports physiologiques (chaux, magnesie,potasse) 3:74 dans les feuilles de la vigne bien alimentée. C.R. Acad. Sci. Paris. 179:782-785. , 1924a. Etude par 1'analyse periodique des feuilles, de 1'inf1uence des engrais de chaux, de magnesie, et de potasse sur la vigne. C.R. Acad. Sci. Paris. 179:932-934. , 1925. Relation lineaire entre les quantities successives d'acide phosphorique et d'azote contenues dans la feuille de la vigne bien alimentée. C.R. Acad. Sci. Paris. 180:1179-1181. _ , 1926. Diagnostic del'alimentation d'un vegetal par l'evolution chimique d'une feuille convenablement chosie. C.R. Acad. Sci. Paris. 182: 653-655. , 1927. Sur 1'absorption de l'azote par la vigne et sue son rble physiologique. C.R. Acad. Agr. Fr. 13:452-455. , 1927a. ContrSle du mode d'alimentation d'une plante perenne (vigne) dans la sol donné'recevant une fumure donnée. C.R. Acad. Agr. Paris. 184:229-231. , 1927b. Examen critique des conditions et des resultats du contr31e alimentaire de la vigne. C.R. Acad. Agr. Fr. 13:548-553. , 1927c. Methods de controls chimique du mode d' alimentation de la vigne. C.R. Acad. Agr. Fr. 13:439-442. , 1927d. Sur 1'absorption de la potasse par la vigne, et sur son r31e physiologique. C.R. Acad. Agr. Fr. 13:448-452. , 1927s. Sur l'absorption de l'acide phosphorique par la vigne et sur son rble physiologique. C.R. Acad. Agr. Fr. 13:443-448. I i I J.JJI' III Ill- ! lil’IIuI I I. V «Ill-Ill .‘Illlv .. lull I‘ll Ilia-III Illll. 3572. , 1928. Le diagnostic foliaire applique au contr31e de l'alimentation d'une vigne de coteau avec ou sans fumure. C.R. Acad. Agr. Fr. 14:762-776. , 1928a. Antagonisme du calcaire a l'egarde de 1' absorption de la potasse par la vigne. Progr. Agric. Vitic. 90:492-497. , 1929. Le diagnostic foliaire et son degre/de se/curite. C.R. Acad. Sci. Paris. 188:1062-1064. , 1930. Le diagnostic de la pomme de terre. Premiere me’moire. Ann. Ec. Agr. Mont. 20:219-281. , 1930a. Observation, par 1e diagnostic foliaire de 1' influence de la temperature sur la mode d'alimentation d' un ve’ge’tal. C.R. Acad. Sci. Paris. 190:1516-1518. , 1930b. Absorption compares de differents engrais azotes par une vigne au cours de la mgme annee. Progr. Vitic. Agric. 93:399-403. , 1930c. Response explicite du diagnostic foliaire alors que les autres moyens d'observation restent muets. C.R. Acad. Sci. Paris. 191:579-580. , 1' 1930a. Evolution chemique comparée des feuilles de la vigne prelevéES a des hauteurs differents sur les rameux. C.R. Acad. Sci. Paris. 190:1137-1139. 11 1930e. Observation par la diagnostic foliaire du phéhomene de remplacement physiologique mutuel de deux bases: chaux et potasse. C.R. Acad. Sci. Paris. 190:389-391. 5 1931. Variations des rapports physiologiques~'“ '11 .1 1!!! All: )‘I‘I Iii] all-I'll“! II! II III". ii i 3:73. entre les constituents mineraux d'une espece vegetal. Ann. Ec. Agric. Mont. 22:284-287. , 1932. Application du diagnostic foliaire: Il suggEre contrsle et limits 1e redressement alimentaire d'une vigne mal nourrie. C.R. Acad. Sci. Paris. 194:812-814. , 1932a. La feuille peut-elle accepter une surcharge d'aliment mineral_sans profit pour le develOppement du vegetal. C.R. Acad. Sci. Paris. 194:933-935. , 1932b. Etude, par le diagnostic foliaireides‘ effets physiologique du chaulage. C.R. Acad. Agr. Fr. 18:443-452. , 1932c. Le diagnostic foliaire de la pomme de terre. Deuxieme memoirs. Ann. Ec. Agr. Mont. 22:50-158. , ' 1933. Composition compares, chez 1a vigne, de feuilles homologues prises respivement sur des souches fructiferes et sur des souches privées de leur grappes. C.R. Acad. Sci. Paris. 196:1168-1170. , 1933a. Sur les variations alimentaires des vegetaux cultivees en dehors de toute intervention d'engrais dans les conditions de la pratique agricole. C.R. Acad. Sci. Paris. 197:1558-1560. , 1933b. Composition compares de la matiere seche des feuilles homologues des raneaux fructiferes et des rameaux naturellement steriles de la vigne. C.R. Acad. Sci. Paris. 196:1hh5‘lhh70 , 1933c. Le diagnostic foliaire de la pomme de terre. deuxiEme memoirs. Ann. Agron. Paris, N.S. 2:306—362 3:1-52. 35’4. , 1934. Examen critique du diagnostic foliaire. C.R. Acad. Agr. Fr. 20:246-257. , . A . . , 1934a. Action d'un meme engrais Simple annuel sur 1'a1imentation NPK d'une mgme espECe vegetale au cours de quatre annees successives de culture dans le meme sol. C.R. Acad. Agr. Fr. 20:549-563. , 1934b. Variations qualitatives et quantitative de l'alimentation NPK d'une mEme espece vegetale dans une meme sol en dehors de toute intervention d'engrais. C.R. Acad. Agr. Fr. 20:443-448. , 1934C. Recherche, par 1e diagnostic foliaire de 1' equilibre optimum d'alimentation NPK chez une plante cultivee. C.R. Acad. Agr. Fr. 20:631-644. , 1935. Diagnostic foliaire de tabac: influence compares des scories de’phosphoration, du superphosphate, et du basiphosphate au 1'equilibre NPK. C.R. Acad. Sci. Paris. 200:502-504. , ‘ 1935a. Sur les variations de la soxmne N+P205+K20 pour 166 de mati\ere sache de la feuille d'une plant cultivé/e. C.R. Acad. Agr. Fr. 21:85-92. , 1935b. Determination par le diagnostic foliaire d' un cas d'inhibition de la potasse. C.R. Acad. Afr. Fr. 21:232-241. , ' 1935c. Demonstration par 1e diagnostic foliaire de l'efficacite du fumier pour vaincre 1'inhibition de la potasse. C.R. Acad. Agr. Fr. 21:396-404. , l935d. Variation des rapports physiologique en / a correlation avec la maladie du feu sauvage chez 1a feuille 1 It i 11“. Gil-“ll- ‘fll' [I l.l| l l I]! ill: 'll'l‘" I I." 355'. de la tabac. C.R. Acad. Sci. Paris. 201:374-376. , l935e. Sur la cinematique de la chaux, de la magnéSie, et leur rapport physiologique dans la feuille du Tabac; mEthode des relais foliaires. C.R. Acad. Sci. Paris. 200:881-883. , 1936. Contribution a 1'etude de 1'inf1uence alimentaire du superphosphate sur la vigne. C.R. Acad. Agr. Fr. 22:1018-1031. , 1936a. Contribution a l'étude de l'influence alimentaire du basiphosphate sur la vigne. C.R. Acad. Agr. Fr. 22: 1132-1138. , 1936b. Variations qualitatives de l'alimentation NPK d'une méme espEce vegetale dans une mgme sol, en dehors de toute intervention d'engrais. Progr. Agric. Vitic. 105:353-356. , 1936c. Dans quelle mesure les variations atmospheriques peut-elle, sous 1e climat Eéditeranéan, modifier chez une vigne les rapports physiologique et les quantités absorbéES d'asote, d'acide phosphorique, et de Potasse? C.R. Acad. Agr. Fr. 22:363-382. , l936d. Vigne et apports caliques. C.R. Acad. Agr. Fr. 22:478-494. , 1936c. Sur la possibilité’de variations de sens opposes et de grands amplitude au cours de une mgme année pour 1'equi1ibre NP! chez 1a feuille d'une plants cultivée. C.R. Acad. Sci. Paris. 202:1550-1553. , . 1936f. Diagnostic foliaire du tabac. Influence comparée des scories de dephosphoration, du superphosphate et du basiphosphate sur 1'équi1ibre NPK. Ann Ec. Agr. Mont. 3176. Ann. Ec. Agr. Mont. 34:3-6. , 1937. Possibilite/de mesurer separément E tout moment de la vegetation 1'effet nutritive et l'effet améliorant d'un apport d'engrais. C.R. Acad. Sci. Paris. 204:535-538. , 1937a. Sur la determination chimique de la plants cultivee. C.R. Acad. Sci. Paris. 205:549—552. , 1837b. L'acide phosphorus dans les equilibres Physiologique NPK de la vigne. C... Acad. Agr. Fr. 23:96-102. , 1937c. Intergt que préSsnte pour 1'agriculturs 1a mesure separe’de l'effet nutritif st d'effet améliorant d'un apport d'sngrais. C.R. Acad. Sci. Paris. 204:9399941. , 1938. Mesures de biochimie agricols sur des rameux de la vigne. C.R. Acad. Sci. Paris. 206:1237-1240. , 1938a. Sur la sensibilité’du diagnostic foliaire pour decaler une absorption d'sngrais par la vigne. C.R. Acad. Agr. Fr. 24:615-624. , 1943. Determination par la plants cultivss ds 1a valeur nutritive d'un milieu cultural. C.R. Acad. Sci. Paris. 216:753-755. / , , and eros, L. 1932.Etuds des variations ds la tsnsur en azote en des points trss localises du feuillage de la vigne. C.R. Acad. Sci. Paris. 194:679-681. von Langer,__, von Seelhorst,éé, and Tollensh__ 1901. J.Landw. 94: 217-223. (in Pfeiffer, Blanck, and Friske, 1913) Lavoisier, A.L. 1865. Oeuvres de Lavoisier. Ls Ministre de 1' Instruction Publique et des cultes, Paris, 1865. Lawes, J.E., and Gilbert, J.H. 1851. On agricultural chemistry. [I ll 1"!" I11" 31771 J. Roy. Agric. Soc. xii:l-40. Lea, G.L., and Midgley, A.R. 1934. Available potassium and phosphorus content of Vermont soils. Ver. Agr. Exp. Sta. Bull. 373. Von Liebig, J. 1840. Organic chemistry and its application to agriculture and physiology. Taylor and Walton, London. 1852. Letters on Modern agriculture. Walton and maberly, London. 1863. The natural laws of husbandry. walton and Maberly, London. Liebscher, G. 1887. Der verlauf der Nahrstoffaufnahme und seine bedentung furdie dfingerlehrs. J. Landw. 35:335-518. 1893. The results of experiments with cats in the years 1889-1892. Landbote 271-272.(Abstract in Exp. Sta. Rec. 4:863-864, 1892/1893) , Kretschmer,L., von Seelhorst,C., and'WillmS, J. 1898. Versuche zur Ermittleung des Dungerbedurfnisses des Acker- bodens B. Ermittelung der Dungerbedurftigkeit des Bodens aus der Zusammensetzung der Erntetrockensubstanz. J.Landw. 46:367-412 Lilleland, 0., and Brown, J.G. 1941. The potassium nutrition of fruit trees.III. A survey of the content of potassium in peach leaves from 130 orchards in California. Proc. Amer. Soc. Hort. Sci. 38:37-48. Lowry, M.W., and Tabor, P. 1931. Sap analysis by bleeding corn plants. Science (ns) 73:453. Lundegardh, H. 1931. Studier over strasadens naringsupptagande samy dettas betydelse for tillvaxten och for uppkomsten av icke- 3578. parasitara sjukdomar. (English summary). Medd. Cent. Anst. F5r36kv. Jordbr. Stockh., 403:1-146. , 1932. Die Nahrstoffaufnahme der Pflanze. Gustav Fischer, Jena. (in Goodall and Gregory, 1947) 1935. The influence of the soil upon the growth of the plant. Soil Sci. 40:89-101. ‘. 1938. The triple analysis method of testing soil fertility and probable crop response to fertilization. Soil Sci. . 45:447-454. 1941. Die Tripelanalyse. Lantbr. Hogsk. Ann. 9:127-221. 1943. Leaf analysis as a guide to soil fertility. Nat. London 151:310-311. Macy, P. 1936. The quantitative mineral nutrient requirements of plants. Plant Physiol. 11:749-764. / Magistad, O.C. 1936. A comparison of Mitscherlbeh trials on Hawaiian soils in Germany and in the Territory of Hawaii. J. Amer. Soc. Agron. 30:692-698. , and Farden,C.A., and Lambert, 0.3. 1932. Yields of pineapples as influenced by fertilization and conformity to the Law of Diminishing Returns. J. Amer. Soc. Agron. 24:610-622. Maume,L., and Bouat, A. 1937. Dosage rapids/st precis ds N, P, K, Mg, et Ca par semimicroanalyse. Ann. Eb. Agr. Mont. 23:5-43. , 1937a. Influences de la variéte et du milieu sur l'absorption du souffrs par le ble: C.R. Acad. Agr. Fr. 23:426-430. , and Dulac, J. 1934. Differences varietales dans l'absorption 13:59. de 1'azote, de 1'acide phosphoriqus, et de la potasse par des blss ayant atteint uns mEms epoque physiologique dans une name melieu. C.R. Acad. Sci. Paris. 198:199-202. , 1934a. Absorption de N, P205, and K20, CaO, and . / . I / / A / MgO par differents varieties de ble observees 1a meme annee I . dans une meme milieu. Ann. Ec. Agr. Mont. 23:96-103. , 1935. Le rapport C/N dans la plants ble a 1' épaison et a la floraisons; ses notables variations suivant 1e milieu. C.R. Acad. Sci. Paris. 200:1245-1246. , 1935a. R’o‘le important du milieu (clima‘t st sol) dans l'absorption ds N, P205, and K20 par les ble; ayant atteint une mame époque physiologique. C.R. Acad. Agr. Fr. 21:120-223. , 1936. Degre/ ds pre’cision de ite/chantiucnagc ds 1a plants ble 3 certaines epoquss physiologiques. C.R. Acad. Agr. Fr. 22:985-990. , 1937. Echantillonage de la plants ble, en vue dss analyses chimiques comparatives. C.R. Acad. Agr. Fr. 22:906-913 , 1937a. Nutrition mineral compares de blés de printsmps et de blés d'automne. C.R. Acad. Agr. Fr. 23:90-94. , 1938. Etude complementairs au chimisme du ble, (nutrition minerals compares de la tige et de L'epi). Ann. Eb. Agr. Mont. 25:9-13. Mayer, A. 1869. Das Dungerkapital und der Raubbau. (in Spillman and Lang. 1924) McCool, M.M. 1926. Relation of soil to plant cell sap. Use of fertilizers lowers the freezing point of cell sap- phosphorus content of the plant affected. Quart. Bull. Mich. Agr. EXp. Sta.. 9:60-64. 360. Me George, W.T. 1939. Factors influencing the availability of native phosphates and phosphate fertilizers for Arizona soils. Ariz. Agr. Exp. Sta. Tech. Bull. 82:295-331. 1942. Studies on plant food availabilities on alkaline calcareous soils: seedling tests and soil analysis. Ariz. Agr. Exp. Sta. Tech. Bull. 94:375-413. 1946. Modified Neubauer method for soil cultures. Soil Sci. 62:61-70. 1946a. Soil properties contributing to citrus chlorosis as revealed by seedling tests. Ariz. Agr. Exp. Sta. Tech. Bull. 112:129-165. Mc Larty, H.R. 1936. Tree injections with boron and other materials as a control for drought spot and corky core of apples. Sci. Agr. 16: 625-633. McMurtrey, J.E. 1949. Plant nutrient deficiency in tobacco. Hunger Signs in Crops. 2nd. Edition. 19-58. Amer. Assoc. Agron. and Amer. Fert. Assoc., hash. D.C. Mehlich, A., Fred, E.B., and Truog, E. 1934. The Cunninghamella plaque method of measuring available phosphorus in soil. J. Amer. Soc. Agron. 27:826-832. , Truog, 3., and Fred, E.B. 1933. The Aspergillus niger method of measuring available potassium in soil. J. Amer. Soc. Soil Sci. 35:259-270. Mitchell, H.L. 1936. The effect of solar radiation upon growth, developement and nutrient content of White Pine seedlings grown under nursery conditions. Black Rock For. Pap. L:16—22. Mitschsrlidh, E.A. 1909. Das Gssetz des Minimums und das Gesetz des abneh- msnden Bodenertrages. Landw. Jahbr. 38:537-552. 36L 1911. fiber das Gesetz des Minimums und die sich aus diesem srgebenden Schlussfolgerungen. Landw. Versuch-Sta. 97:231-236. 1912. Das Wasser als ngetationsfaktor. Landw. Jb. 42:709-717. 1912a. Zum Gesetz vom.Minimum. Eine Antwort an Th. Pfeiffer und seine Mitarbsiter. Landw. Versuch-Sta. 77:414-415. .0 1918. Versuch uber den Einfluss zweisr verschiedener Nahrstoffs auf den Pflantzenertrag. Landw. Jahrb. 52:279-284. 1919. Das Gesetz dss Pflanzenwachstums. Landw. Jahrb. 53 :167-179 0 1919a. Ein Beitrag zum.Gssetz des Pflanzenwachstums. Fuhlings Landw. Ztg. 130-133. 1923. Bodsnkunde far Land- und - Forstwirte. 4th Edition. Berlin. 1923a. Die Pflanzenphysiologische LSSung der Chemischen Bodenanalyse. Landw. Jahrb. 58:601-617. . 1928. Die Zwsite Annaherung des Workungsgssetzes der Wackstumfaktoren. Ztschr. Pflanzenernahr. Dung. u Bodenk. A. 1.2 : 273-2810 1930. Die Bestimmung des Dungerbedurfnissss des Bodens. 2nd Edition. Verlagebuchlandlung, Paul Parey, Berlin. 1947. Results of more than 27,000 field tests with fertilizers. Ztschr. Pflanzenernahrung Dung. u Bodenk. 38:22-35. Mooers, C.A. 1938. An evaluation of the Cunninghamella and Aspergillus niger’methods for the determination of the fertilizer needs of a soil. Soil Sci. 46:211-228. Mulay, A.S. 1931. Seasonal changes in total, soluble, soluble protein, 362. non protein, and insoluble nitrogen in the current year's shoot of Bartlett pear. Plant Physiol. 6:519-529. Mulder, B.G. 1939/1940. 0n the use of microorganisms for measuring a deficiency of copper, magnesium, and molybdenum in soils. Antony and Leeuwenhoek, 6:99-109.(in Vandecaveye, 1948) Munter, F. 1919. Pflanzenanalyse und Dfingerbedfirfnis des Bodens. J. Landw. 67:229-266. Murneek, A.E. 1930. Quantitative distribution and seasonal fluctuation of nitrogen in apple trees. Proc. Amer. Soc. Hort. Sci. 27:228-231. , and Gildehaus, E.J. 1931. Leaf diagnosis and interpretation of fertilizer requirements of plants. Sci. 74:39-41. Nsubausr, H. 1929. Mitteilungen fiber die Keimpflanzen.Methode. Ztschr. Pflanzenernahr. Dung. u Bodenk. B8:219-233. 1923. Ein Laboratoriumsverfahren zur Bestimmung der von den Pflanzen aus dem Baden Aufnshmbaren.Msngen von Phosphorsaure un Kali. Landw. Versuch-Sta. 100:119-128. 1925. Die Bestimmung des Dungerbedfirfnisses des Bodens. Ztschr. Pflanzenernahr. Dung. u Bodenk. A2:329-362. , and Scheider,'fl. 1923. Die Nahrstoffaufnahme dss Keimpflanzen und ihre Awsndung auf die Bestimmung des Nahrstoffgehalts der Bbden. Ztschr. Pflanzenernahr. Dung. u Bodenk. A2:329-362. Newton, J.D. 1928. The selective absorption of inorganic elements by various crop plants. Soil Sci. 26:85-91. Nicholas, D.J.D., and Jones, J.O. 1944. The rapid chemical tests of plant tissues in the diagnosis of deficiency of mineral elements. Rep. Agr. Hort. Sta. Long Ashton, 1944. 84-97. Sou; ‘, and Catlow, E. 1947. Manurial experiments on vegetable crops. XI. Effects of Farmyard manure and various fertilizer treatments on Cauliflower. Rep. Agr. Hort. Sta. Long Ashton, 1947. 103-110. , 1947a. Manurial experiments on vegetable crops. XII. Effects of farmyard manure and various fertilizers. on three varieties of potatoes. Rep. Agr. Hort. Sta. Long Ashton, 1947. 110-118. , 1947b. Manurial experiments on vegetable crops.XIII. Effects of farmyard manure and other manurial treatments on potato and cauliflower. Rep. Agr. Hort. Sta. Long Ashton, 1947 118-126. 3 1948. Application of rapid chemical tests to the diagnosis of mineral deficiency in horticultural crops. J. Hort. Sci.— 24:106-122. Nightingale, G.T. 1942. Nitrate and carbohydrate reserves in relation to nitrogen nutrition of pineapple. Bot. Gaz. 103:409-456. 1942a. Potassium and phosphate nutrition of pineapple in relation to nitrate and carbohydrate reserves. Bot. Gaz. 104: 191-223. Niklas,H. and Hirschberger, E. 1924. Eine neue Methods zur raschen Ermittlung der Phosphorsaursbedurftigkeit unserer Boden. Zeit. angew. Chem. 37:955-957. , and Poschenrieder, H. 1932. Die Ausfuhrung der Aspergillus Methods zur Prufung aui‘ Kali. Ernair. Pflanze 28:86-88. , 7 1936. Zur Feststellung der Magnesia Dungerbedurftig- keit und Magnesia-Dunewirkung im Boden mittels Aspergillus niger. Bodenk. u Pflanzenernahr. 1:235-247. 361/. , , and Trischler, J. 1930. Die cultur des Schimmolpilzes Aspergillus niger zur biochemischen Bestimmung der Kali- und Phosphorsauredungebedurftigkeit de Boden. Ernahr. Pfl. 26:97-103. , , 1930a. Eine neue mikrobiologische Methods Zur Festellung der Dungerbedurftigkeit der Bodsn. 1, Mitteilung: Bestimmung der Kalibedurftigkeit der Baden mittels Aspergillusniger. Ztschr. Pflanzenernahr. Dung. u Bodenk. A18:129-157. Olsen, S.R., and Shaw, B.T. 1943. Chemical, kitscherlich, and Neubauer methods fro determining available potassium in relation to crop response to potassium fertilizers. J. Amer. Soc. Agron. 35:1-9. Otto, R., and Kooper, H.B. 1910. Beitrage zur Abnahme Bezw. Rickwanderung der Stickstoffverbindungen aus den Blattern wahrend der Nacht, sowie zur herbstlichen ruckwanderung von Stickstoffverbindungen aus den Blattern. Landw. Jb. 39:167-172. Parker, M.M. 1933. Tomato fertilization. II. The effect of different fertilizers in ratios on the yield of tomatoes. Bull. Va. Truck Gard. Sta. 80. Paterson, D.D. 1939. Statistical technique in agricultural research. McGraw Hill Book. Co., N.Y. Penston, N.L. 1935. Studies of the physiological importance of mineral elements in plants. VIII. The variation in the potassium content of potato leaves during the day. New. Phytol. 34:296-309. 1937. A study by microchemical methods of the distribution of potassium in the potato. Ann. Bot. 45:673-692. 1938. The variation in the potassium content of maize leaves during the day. New Phytol. 37:1-14. 5361?. Petersen, von P. 1876. Ueber das Minimum.der fur die Haferpflanze nothigen Phosphorsaure und ueber die nutzbare Verbindungsform der Phosphor- saure. Jber. Fortschr. Agrik. Chem. 18/19:251-256. 1375/1376. Psttinger, N.A., and Thornton, S.F. 1934. Acomparison of the Neubauer, plant analysis of sap, and the Hoffer stalk-test methods for determining the nutrient supplies of soils. J.Amer. Soc. Agron. 26:547-561. Pfeffer, w. 1900. Physiology of plants. Vol. I. Ewart, Clarendon Press. Pfeiffer, T., Blanck, E., and Flugel, M. 1912. Nasser und Licht als Vegetationsfaktorsn und ihre Beziehungen zum Gestze vom.Minhnwn.. Landw. Versuch-Sta. 76:169-236. , , and Friske, K. 1913. Der einfluss verschiedener Vegetationsfaktoren, namentlich, des Nassers, auf die Erzielung von Maximalertragen in Vegetationsgefassen. Landw. Versuch-Sta. 82:237-298. , , and Simmermacher, W. 1915. Beziehung Zwischen dem Einfluss von Licht und Stickstoff als Minimumfaktoren auf das Wachstum der Pflantzen. Landw. Versuch-Sta. 86:45. , , , and Rathmann, N. 1915. Pflanzenanalyse und Baden analyse zur Bestimmung des Nahrstoffgehalts der Achsrboden. Landw. Versuch-Sta. 86:339-391. , and Rippsl. 1919. Uber den Verlauf der Hahrstoffaufnahme und Stofferzeugung bei der Gerstsnbezw. Bodenpflanze. J. Landw. 69:137-162. , , and Pfotenhauer, C. 1919. Hber die Verlauf der Nahrstoffaufnahme und Stofferzeugung bei der Gerstenpflanze. Fuhling Landw. Ztg. 68:81-101. , Simmermacher, W., and Rippel, A. 1919. Der Gehalt der Haferpflanzsn Phillis, €366. an Stickstoff, Phosphorsaure, und Kali untsr verschiedsnen Bsdingungen und seine Beziehungen zu der durch sine Nahrstoff- zufuhr bedingten Ertragserhbhung. J. Landw. 67:1-57. E., and Mason, T.G. 1942. On diurnal variations in the mineral content of the leaf in the cotton plant. Ann. Bot. (London) ns. 6:437-442. Pigorini, L. 1914. Studi sulla foglia di gelsa. R.C. Acad. Lincei Ser. Piper, C. 5,23:433-437. (in Goodall, and Gregory, 1947) S. 1936. The boron status of South Australian apples. J. Coun. Sci. Industr. Res. Aust. 9:245-248. 1940. The symptoms and diagnosis of minor element deficiency in Raleigh, Remy, T. agricultural and horticultural crops. I. Diagnostic methods. Boron and Manganese. Emp. J. Exp. Agr. 8:85-96. S.M., and Chucka, J.A. 1944. The effect of nutrient ratio and concentration on the growth and composition of tomato plants and on the occurrence of blossom end rot of fruit. Plant Physiol. 19:671-678. 1896. Der Verlauf der Stoffaufnahms und das Dungerbedurfnis des Roggens. J. Landw. 44:31-103. , 1906. Die Bedeutung der Pflanzenanalyse fur die Festellung des Dungsrbedurfnisses der Boden, Jber. Agrik. Chem. Folge 3 10:175. , 1906a. Dungerbedurfnis der Obstgarten. Jber. Agrik. Chem. Folge 3, 10:191. Rhode Island Agr. Exp. Sta. 1935. Measurement of nitrogen as an index of current nitrogen supplies. Rep R.I. Agr. Exp. Sta. 47:68-69. Ridhards, F.J. 1932. Physiological studies in plant nutrition. III. Further studies on the effect of potaSh deficiency on the rate of 367. respiration in the leaves of barley. Ann. Bot. London, 46: 367‘3880 , 1944. Mineral nutrition of plants. Ann. Rev. Biochem. 13: 611—630. Richardson, W.D. 1920. The ash of dune plants. Sci. 51:546-551. Roach, W.A. 1931. The chemistry of the rootstock-scion effect. I. The elements absorbed from the soil. A progress report. Rep. E. Malling Res. Sta. for 1928-30. Al4:101-104. 1934. Tree injection. Invigoration by injection of fertilizers. Rep. E. Malling Res. Sta. for 1934. 135-138. 1939. Plant injection as a physiological method. Ann. Bot. London. 3:155-226. 1939a. Diagnosis of mineral deficiencies and excesses by systematic leaf injection and analysis. Rep. E. Malling Res. Sta. for 1939. A23:51-58. 1943. The present position regarding the diagnosis of mineral deficiencies in fruit trees by plant analysis and plant injection. Rep. E. Malling Res. Sta. A27:99-103. 1943a. The diagnosis of mineral deficiency by plant analysis and plant injection. Occas. Publ. Sci. Hort. 4:40-41. , and Roberts, W.O. 1945. Further work on plant injection for diagnostic and curative purposes. J. Pomol. 21:108-119. Roberts, W.O. 1944. Leaf painting as a method of diagnosis of mineral deficiency. Ann. Rep. E. Malling Res. Sta. for 1944. p67. Rogers, L.H., and Gall, 0.E., and Barnett, R.M. 1939. The zinc content of weeds and volunteer grasses and planted land covers. Soil Sci. 47:237-243. 368. Russel,E.J. 1927. Soil conditions and plant growth. 5th Edition. Longman's Green and Co. London. , and Russel E.H. 1950. Soil conditions and plant growth. 8th. Edition. Longman's Green and Co. London. Sachs, J. 1560. Vegetationsversuche mit Ausschluss des Bodens uber die Nahrstoffe und sonstigen Ernahrungsbedingungen von mais, Bohnen, und anderen Pflanzen. Landw. Versuch-Sta. 2:219-268. Sackett,‘N.G., and Stewart,L.C. 1931. A bacterial method for determining min- eral soil deficiencies by use of soil plaque. Colo. Agr. Sta. Bull. 375. Salter, R.M.,and Ames,J.N. 1928. Plant composition as a guide to the avail- ability of soil nutrients. J. Amer. Soc. Agron. 20:808-83 . De Saussure,N.T. l804.Recherches chimiquss sur la vegetation.5th.Ed. Paris. Scarseth, G.D.1941.Soil and plant tissue tests as diagnostic aids in determin- ing fertilizer needs.Proc. Ass. S. Agr. Workers 42:53-54. 1942. Tissue testing (Quick test) in diagnosis. Amer. Fertil. 97(7):22. 1943. Methods of diagnosing plant nutrient needs. Amer. Fertil. 98(12):5-8,22,24,26. 1943a. Plant tissue testing in the diagnosis of the nutritional status of growing plants. Soil Sci. 55:113-120. Schimper, A.F.N. 1890. Zur Frags der Assimilation der Mineralsalze durch die gruns Pflanze. Flora. 73:207-261. Schulze, B., and Schutz, J. 1909. Die Stoffwandlungen in den Laubbattern des Baumes insbesonders in ihren Beziehungen zum herbstlichen Blattfall. Landw. Vsrsuch-Sta. 71:299-352. Schuster, C.E., and Stephenson, R.E. 1940. Sunflower as an indicator plant 3o 9. of boron deficiency in soils. J.Amer. Soc. Agron. 32:607-621. von Seelhorst, C. and Wilms,__ 1396. J. Landw. 46:413. (in Pfeiffer, Blanck, and Friske, 1913) Senebier, J. 1782. Memoirs Physico-chymiques. Geneva.(in Russel, 1927) Shear, G.M., and Horsfall, J. 1948. Color sa an index of nitrogen content of leaves of york and Stayman apples. Proc. Amer. Soc. Hort. Sci. 52:57-60. Skinner,J.J., and Reid, F.R. 1921. Nutrient requirements of clover and wheat in solution cultures. Soil Sci. 12:287-301. Smith, E.B., Brown, P.F., and Schlots, F.E. 1932. A comparison of the Niklas and Truog methods for the determination of available phosphorus in soils. J. Amer. Soc. Agron. 24:452-459. Snedecor, G.N. 1937. Statistical methods. Collegiate Press, Iowa. Spencer, E.L., and Levin, G.I. 1939. Frenching of tobacco. Phytopath. 29:502-503. Spillman, w.J. 1933. The use of the expodential yield curve in fertilizer experiments. U.S.D.A. Tech. Bull. 348. , and Lang, E. 1924. The law of diminishing returns. World Book Co. N.Y. Sprengsl, C.S. 1837. Die Bodenkunde oder die Lehre vom Boden nebst einer vollstandigen Anleitung zur Chemischen Analyse der Ackererden. Leipzig. (in Browne, 1944.) Stahl-Schroder, M. 1904. Kann dis Bflanzenanalyse uns Aufschluss fiber den Gehalt an assimilierbaren Nahrstoffen im Boden geben? J. Landw. 52:31-92, 193-268. Stephenson, H.B., and Schhster, C.E. 1941. Laboratory, greenhouse, and field methods of studying fertilizer needs. Soil Sci. 52:137-153. 370 Stewart, R. 1932. The Mitschsrlich, Niessmann, and Neubauer methods of determining the nutrient content of soils. Imp. Bur. Soil Sci. Tech . Conn. 25 . Stewart, L.C., Sackett, W.G., Robertson, D.N., and Kszsr, A. 1932. A comparison of the soil plaque method with the Neubauer and Hoffer corn stalk methods for determining soil mineral deficiencies. Bull. Colo. Agr. Exp. Sta. 390. Sturgis, M.B., and Reid, J.F. 1937. The relation of organic matter and fertilizer to growth and composition of rice. J. Amer. Soc. Agron. 29:360-366. Susuki, U. 1897. On an important function of leaves. Bull. College Agr. Tokyo, 3:241-252. Thaer, A.D. 1810. Grundsatze der rationellen Landwirtchschaft.(Translated by Shaw and Johnston, The principals of Agriculture. 2nd. edition. London, 1844,) (in Browne, 1944) Thomas, W. 1929. Balanced fertilizers and Lisbig's law of the minimum. Sci. 70:382-384. 1930. The conception of balance with respect to the absorption of N, P, and K by plants and the influence of the level of nutrition. Sci. 72:426-427. 1932. The reciprocal effects of nitrogen, phOSphorus, and potassium as revealed to the absorption of these elements. Soil Sci. 33:120. 1934. Misconceptions relative to the mineral composition of plants. Sci. 80:587. 1936. Mathematical expression of the equilibrium between N, 37!. A and P in plants. Sci. 84:422-423. 1937. Foliar diagnosis: Principals and practice. Plant Physiol. 12:571-599. 1938. Mathematical expression of the equilibrium between lime, magnesia, and potash in plants. Sci. 88:222-223. 1938a. Foliar diagnosis: Application of the concepts of quantity and quality in determining response to fertilizers. Proc. Amer. Soc. Hort. Sci. 35:269-272. 1938b. Foliar diagnosis: Its relation to the optimum.nutrition of the potato. Plant Physiol. 13:677-694. 1945. Present status of diagnosis of mineral requirements of plants by means of leaf analysis. Soil Sci. 59:353-374. , and Mack, H.B. 1938. Foliar diagnosis in relation to the developement and fertilizer treatment of the potato. J. igr. Res. 57:397-414. , 1939. Control of crop nutrition by the method of foliar diagnosis. Bull. Penn. Agr. Exp. Sta. 378. ’, 1939a. The foliar diagnosis of Zea Mays subjected to differential fertilizer treatment. J. Agr. Res. 58:477-491. , - 1939b. Foliar diagnosis: Physiological balance between the bases lime, magnesia, and potash. J. Agr. Res. 59:303-313. , 1939c. A foliar diagnosis of the effect of three nitrogen carriers on the nutrition of Zea Mays. J. Agr. Res. 59:303-313. , l939d. Foliar diagnosis: Nutritional differences affecting the yields of potatoes from similarly treated plots. 372. Proc. Amer. Soc. Hort. Sci. 36:573-589. , l939e. Foliar diagnosis: The influence of soil on the action of fertilizers. Plant Physiol. 14:75-92. , ‘4_ 1940. Salient features of the method of foliar diagnosis. Proc. Amer. Soc. Hort. oci. 37:253-260. , 1940a. Foliar diagnosis of differentially fertilized greenhouse tomatoes with and without manure. J. Agr. Res. 60: 811-832. , 1941. Poliar diagnosis study of the climatic effects influencing the nutrition of Spring and Fall Greenhouse tomatoes. Plant Physiol. 16:117-144. ,_ 1941a. Foliar diagnosis in relation to soil heterogeneity. Soil Sci. 52: 455-468. , 1943. Foliar diagnosis in relation to plant nutrition under different conditions of weather and soil reaction. Soil Sci. 56:197-212. , 1944. Misconceptions relative to the method of foliar diagnosis. Proc. Amer. Soc. Hort. 501. 44:355-361. , 1944a. The effect of different carriers of nitrngen on the nutrition of the potato. Proc. Amer. Soc. Hort, Sci. 44:346-354. , , and Cotton, R.H. 1942. Foliar diagnosis in relation to irrigation. Proc. Amer. Soc. Hort. Sci. 44-355-361. , , 1943. Leaf analysis as a means of determining the fertilizer requirements of crops. Amer. Fert. 98(4): 5-7,26,28. , , and Fagan,F.N. 1943. Follar diagnosis: The mineral nutrition of the peach tree with particular reference to Bacterial 3 73. Leaf Spot. Proc. Amer. Soc. Hort. Sci. 52:47. , , and Rahn, Edi. 1944. Foliar diagnosis and plant nutrition in fertilizer placement experiments. J. Amer. Soc. Agron. 36:889-901. Thornton, S.F. 1931. Experiences with the Neubauer method for determining mineral nutrient deficiencies in soils. J. Amer. Soc. Agron. 23:195-208. 1931a. The Neubauer method as applied to the determination of the availability of phosphate fertilizers. J. A.D.A.C. 14: 292-2950 1932. A field and laboratory test on plant material for diagnosing phos‘horus deficiencies. Ind. Agr. Exp. Sta. Bull. 355. 1932a. Factors affecting the availability of phosphate fert- ilizers as shown by the Neubauer method. J. A.D.A.C. 15:163-166. 1935. Soi1.and fertilizer studies by means of the Neubauer method. Ind. Agr. Exp. Sta.‘Bull. 399. 1935a. The available phosphorus and potassium contents of surface soils and subsoils as shown by the Neubauer method and by chemical tests. J. Amer. Soc. Agron. 27:46-51. , Conner, S.D., and Fraser, R.R. 1934. The use of rapid chemical Tollens, tests on soils and plants as aids in determining fertilizer needs. Circ. Ind. Agr. Exp. Sta. 204. B. 1882. Ueber einige Erlichterungen bei der cultur von Pflanzen in wasserigen Losungen. J. Landw. 30:537-540. 1901. The ash constituents of plants; their estimation and their importance to agricultural chemistry and agriculture. 374-. Part I. The ash of plants, its preparation and analysis. Exp. Sta. Rec. 13:207-220. 1902. Part II. Importance of ash analysis to plant physiology and agricultural chemistry. Exp. Sta. Rec. 13:305-317. Trealease, S.F., and hartin, A.L. 1936. Plants made poisonous by selenium absorbed from the soil. Bot. Rev. 2:373-396. Truffaut, G., and Bezssonoff, N. 1927. Hesure de l'assimilabilite de divers phosphates par leur action sur la fixation bacterienne de 1' azote. C.R. Acad. Sci. Paris. 185:85-96. Truog, E. 1928. How plants feed. Proc. lst. Inter. Congress. Soil Sci. 3:628-636. Tull, J. 1733. Horse hoeing husbandry. Previlegio Regiae Hajestatis, London. (in Browne, 1944) Ulrich, A. 1941. Nitrate test. A guide to the nitrate fertilization of sugar beets. Sugar Beet Bull. 5:77-78. 1942. Potassium content of grape leaf petioles and blades contrasted with soil analysis as an indicator of the potassium status of the plant. Proc. Amer. Soc. Hort. Sci. 41:204—212. 1942a. Nitrate content of grape leaf petioles as an indicator of the nitrogen status of the plant. Proc. Amer. Soc. Hort. Sci. 41:213-218. 1943. Plant analysis as a diagnostic procedure. Soil Sci. 55: 101-112. 1946. Plant analysis as a guide to the fertilization of sugar beets. Proc. Amer. Soc. Hort, Sci. 4:88-95. 1946a. Critical phosphorus and potassium levels in Ladino clover. Soil Sci. Soc. Amer. Proc. 10(1945):150-161. .D II 1" ill llll’ {I ll'lll. {151‘ Iii ,. all 3753 1948. Plant analysis fifi methods and interpretation of results. (in Diagnostic Techniques for Soils and Crops. Kitchen, 1948). Vandecaveye, S.C. 1948. Biological methods of determining nutrients in the soil. (in Kitchen, 1948). Vinet, M.E., and Lemesle, __ 1930. Essais sur les engrais appliques a la vigne. Bull. Mans. Soc. Ind. et Agr. Angers 101:81-97. Volk, N.J., and Truog, E. 1934. A rapid chemical method for determining the readily soluble potassium of soils. J. Amer. Soc. Agron. 26: 537-546. Wagner, P. 1883. Beitrage zur Ausbidung der Dungungslehre. Landw. Jb. 12:693. Wallace, T, 1928. The composition of leaves bark and wood of current season's shoots in cases of lime-induced chlorosis. J. Pomol. and Hort, Scl. 7:172-183. 1929. The effects of manurial treatments on the chemical composition of gooseberry bushes. I. Effects on dry matter, and ash constituents of leaves and stems of terminal shoots and of fruits, and on the total nitrogen of fruits. J. Pomol. and Hort. Sci. 7:130-145. 1932. Chemical investigations relating to potassium deficiency of fruit trees. J. Pomol. .9:111-121. 1940. Chemical investigations relating to the magnesium deficiency of fruit trees. J. Pomol. 18:145-160. 1941. Magnesium deficiency of fruit and vegetable crops. Rep. Agr. Hort. Res. Sta. Long Ashton, 1940:24-28. 1943. The diagnosis of mineral deficiencies of plants. A £376. colour Atlas and guide. Iis Majesty's Stationery Office, London. 1944. The diagnosis of mineral deficiencies of plants. A color atlas and quide. Supplement. His Majesty's Stationery Office, L0 I'ld 01'". o , Morley Davies, W., Nicholas, D.J.D., and Hewitt, E.J. 1946. Some effects of Aims and fertilizers on potatoes on a strongly acid soil as determined by visuil symptoms and chemical tests. Ann. Rep. Agr. Hort. Res. Sta. Long Ashton: 61-66 , Osmond, D.A. 1940. A preliminary note on the comparative effects of sulphate of potash, muriate of potash, and Kainit on various fruit trees. Rep. Agr. Hort. Sta. Long Ashton for ' 1940:13-18. Walker, H.B., and Sullivan, J.L. 1929. The spontaneous culture method for studying the non-symbiotic nitrogen fixing bacteria in the soil. Proc. Iowa Acad. Sci. 36:53-61. Wark, D.C. 1938. A method for testing for nitrogen deficiency. J. Aust. Inst. Agr. Sci. 4:208-210. 1939. Tests on plant tissue as a guide to the soils' available nutrients. J. Aust. Inst. Agr. Sci. 5:224-227. Weinhold, A. 1862. Analyse von Unkrautern des Bodens der Versuchstation Chemnitz. Landw. Vesuch-Sta. 4:188-193. 1864. Ueber die Uebereinstimmung der Zusammensetzung von Pflanzenaschen und derjenigen des Bodens. Landw. Versuch-Sta. 6:50-570 Wiegmann, A.F., and Polstorff, 8.6. 1842. Bestandtheile der Pflanzen Braunschweig, F. Viewig und Sohn. (in Browne, 1944) 3772 Niessmann, H. 1925. Ueber den Einfluss des Lichtes auf die Nahrstoff- Willcox, aufnahme der Pflanzen in Jugenstadium. Ztschr. Pflanzenernahr. Dung u Bodenk. B4:153-168. 0.W. 1930. Principals of Agrobiology. Palmer Pub. Co. N.Y. 1937. A B C of agrobiology. Norton and Co. R.I. 1940.A Simple graphical method of evaluating tests with fertilizers. Facts about Sugar 35(12):33-37. 1941. The fertilization of the sugar cane. II. The agrobiologic evaluation of some potash tests. Facts about Sugar. 36(6):26-29. 1941a. The fertilization of the sugar cane. III. The agrobiologic evaluation of some nitrogen tests. Facts about 1941b. A critique of field experiments with plant nutrients. Amer. Fert. 95:5—7. 1943. Interpretation of Olsen and Shaw's field tests by Mitscherlich—Baule theorem and the Universal yield diagram. J. Amer. Soc. Agron. 35:454-459. 1944.. Yield- depression effect of fertilizers and its measurement by the universal yield diagram. J. Amer. Soc. Agron. 36:20-310’0 1944a. Absolute values in fertilizer experiments. J. Amer. Soc. Agron. 36:480-486. 1945. Yield-depression effect of fertilizers and its measurement by the universal yield diagram. III. J. Amer. Soc. Agron. 37:622-628. 37V8. 1945a. Yield-depression effect of fertilizers and its measurement by the universal yield diagram. II. Report on nutritional unbalance disclosed by field tests. J. Amer. Soc. Agron. 37:9—20. 1947. Use of the standard yield diagram. Facts about Sugar 42(5):28-32. 1949. Verification of Mitscherlich's effect law. J. Amer. Soc. Agron. 41:226-229. ( undated a ) Diagrams for evaluating a farmer's one plot field test with a fertilizer. mimeographed circular. ( Undated b ) A farmer's one plot field test with a fertilizer. Diagrams for phOSphorus(P205) and nitrogen (N). Himeographed. ( undated c ) Quantitative agrobiologic evaluation of some phosphate fertilizer tests with vegetable crops in Alabama. Mimeographed ciccular. Wimer, D.C. 1927. Composition of mature corn stover as affected by variety, soil type, and fertilizer treatment. Ill. Agr. Exp. Sta. Bull. 437:174-272. Winogradsky, S. 1925. Etudes sur la microbiologie du sol. II. Sur les microbes fixateurs d'azote. Ann Inst. Past. 40:455-520. , and Ziemiecka, J. 1927. Etudes sur la microbiologie du sol. III. Sur 1e pouvoir fixateur des terres. Ann. Inst. Pasteur. 42:36-62. , 1928. Bur 1'application agronomique d'une epreuve micro- biologique. C.R. Acad. Sci. Paris. 187:161-165. 3761 Wolff, E. 1868. Bericht uber die in der Jahren 1866 und 1867 ausgefuhr- ten Vegetationsversuche in wassriger Losung der Nahrealze. Landw. Vensuch-Sta. 10:349-379. 1871..Aschen-Ana1ysen von landwirthschaftlichen Produeten. Fabrik Abfallen und wildwacksenden Pflanzen. Part I. Wiegandt and Hempel, Berlin. 1876. Ueber dae minimum der Nahrsalze. J. Ber. Forschr. Agrik. Chem. 18/19:250-251. 1877. Versuche in Wessercultur uber den Bedarf der Haferpflanzen an Stickstoff Nehrung und an fixen Nahrstoffen. Landw. Versuch- Sta. 20:395-398. 1880. Aachen-Analysen. Part II 1870-1880. Wiegandt and Hempel. Berlin. Wollny, E. 1897/1898. Untersuchungen.uber den Einfluss der Wachstums— faktoren auf das Produktionsvermogen der Kulturpflantzen. Forschungen,auf dem Gebiete der Agrikulturphyeik 20:105-123. Neodward, T. E., Shepherd. J. 3.. and Tysdal. H. M. 1944. Yield and chemical content of Alfalfa cut at different times of the day and night. J. Amer. Soc. Agron. 36:940-943. Yates, F., and Wetson. D. J. 1939. Factors influencing the percentage of nitrogen in the barley grain of Hoesfield. J. Agr. Sci. 29:452-458. Young, A. W. 1933. The Winogradsky spontaneous culture method for deter- mining certain soil deficiencies. Iowe.Agr. Exp. Sta. Bull. 157. 3130. Young, L. C.. and Bailey, C. F. 1936. Progress report on the investiga- tion of corky core of apples. Sci. Agr. 17:115-127. Ziemiecka, J. 1932. The Azotobacter test of soil fertility applied to the classical fields at Rothamstead. J. Agr. Sci. 22: 797‘8100 W.’>“\|.. not II). ; ... . . . .l . . .n. .21.).III’Afi ' I |l t u . ‘ _‘l — o . . .. . . .aa. .. .. XII. A .Aarfl ..rva. . .cr’ .0 n . u n V . u. . . _ ’3 .{r . .I t . V c u A ”cu In Q': . . . v . . .._ _. . . J . u. .1 n .. x . . . n . . , . 1 4 . I ‘. . A . . . I. . — . x _ a . Q \. . . . . .3 .1 . .3 . I K .- u . N ' , . \ v > v. . . . .m , . v y . 5‘ U . . . . . A . 0 . . . . , . .. . . s .. . _ t. v . . . l 0 ‘1 n I