ABSTRACT CORRELATION STUDIES WITH CORN USING SEASONAL SOIL AND TISSUE TESTS by Hari Mohan Singh The objective of the study was to compare the usefulness of nu- tritional indices derived for corn from seasonal soil tests for nitrate, phosphate, and potassium, with those from seasonal tissue tests for the same nutrients, using recognized concepts for the interpretation of foliar analysis data. Samples of soil and corn midribs were collected periodically during the growing season from several field eXperiments. The time of appearance of visible nitrogen deficiency symptoms was noted, and final yields of corn were taken. Multiple regression analysis, employing two types of curvilinear equation, was used to correlate yields with soil tests or tissue tests for each sampling date. Soil tests and tissue tests for all three nutrients were found, by analysis of variance, to be influenced by numerous management and soil factors, as well as by climatic conditions and time. Specific relationships, however, were frequently very different for soil tests than for tissue tests. Soil test variations during the growing season appeared to reflect changes in rates of release of nutrients from soil sources and removal by corn. Tissue test variations appeared to re- flect variations in rate of uptake and rate of assimilation of nu- trients by the corn. Soil nitrate levels of 20 pounds per acre or less anticipated the development of visible nitrogen deficiency symptoms by one to three weeks. The corresponding "threshold" level for tissue nitrate was Bari Hohan Singh 200 ppm. N, on a green tissue basis. Multiple regression analysis of unit observations (rather than treatment means) revealed distinct optimum levels of soil and tissue nitrate which were higher than the threshold levels (40 pounds per acre and 400 ppm., respectively for soil and tissue). ' The "threshold” levels for development of visible nitrogen de- ficiency symptoms appear to correspond to "critical" levels in the zone of poverty adjustment, as defined in established schemes for interpretation of foliar analysis data. The I'optimum" levels found appear to represent an upper limdt of balanced nutrition, beyond which accumulations of soil or tissue nitrate reflect critical de- ficiencies of some other nutrient or of some other factor of growth. The range between threshold and optimum.levels may represent a zone where response is variable due to varying degrees of “nutrient substitution" -- in the sense used by production economists. Similar relationships were observed for soil and tissue phos- phate in one experiment. Soil tests for nitrate were better correlated with yield, par- ticularly during periods of rapid corn development, than were tissue tests for nitrate. The reverse was true for phosphorus tests. Es- sentially no significant correlations were found for soil or tissue tests for potassium.in the'experhments studied. In general, the best multiple correlations with both soil and tissue tests were obtained at about silking time -- or, in other words, near the end of the grand period of growth and physiological development in corn. Bari Hohan Singh Useful soil test correlations were obtained where wide differ- ences in nitrogen fertility had been established by previous treat- ments imposed over a period of years. CORRELATION STUDIES WITH CORN USING SEASONAL SOIL AND TISSUE TESTS by Kari than Singh A THESIS Submitted to Hichigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Soil Science 1961 To my Parents This thesis is dedicated to my parents, who encouraged me to come to the United States for advanced studies and who cared for my family during my absence from India. 11 ACKNOWLEDGMENTS The author wishes to express his sincere gratitude to Dr. A. R. Holcott, under whose guidance this investigation was carried out. He is also greatly indebted to Dr. R. L. Cook for his assistance in obtaining financial aid and for the facilities made available for conducting the research. Sincere thanks are also extended to graduate students and other staff members of the Soil Science Dapartment, including secretarial staff, for their assistance and help in the completion of this study. Acknowledgment is also extended to‘Hr. B. A. waldecker, Data Processing Research Division, for assistance in programming data for analysis of variance, and to Mr. B. R. Hoffnar and members of the Statistical Pool, Department of Agricultural Economics, for com- putations involved in the multiple regression analyses. The financial aid of the Nitrogen Division, Allied Chemical Corporation, is gratefully acknowledged. iii TABLE OF CONTENTS PAGE mnuulou00000000.0..0000000000.00.00000000000000000000.00000 l lem “VIE"..0.000000000000000 0000000000000000000000000000 J-‘b rm.hrissue Test.0 ....... 0 ..... 0.0.0.0000000 0000000000 0.00000 Objective of Tests... ....... .... ...... . ............ . ........ Theory behind Fresh Tissue Tests.................... ........ Role of Nitrate in the Plant................................ Roles of P and K.in the P1ant........... ........ ............ Correlation of Tissue Tests with Crop Response...... ........ Parts and Number of Plants to be Tested ..................... Time of Day for Tissue Tests.......... ...................... Stage of Growth for Tissue Tests ........................... . OQQNO‘O‘UUI H Soil Tests........ ............................................ 10 Soil Tests in General Use ................................... 11 Soil Test Correlation...... ......... ....... .............. ... l3 HATERIALSANDIETRODS ......... . ................ 16 Field Rxperiments............................................. l6 Rotation Experiment...................... ....... ............ 16 Residue Rxperiment.......................................... l7 Nitrogen Source, Rate and Time of Application Experiment.... 18 field Procedures.... ........................... ...... ......... 19 Sampling................ ........................... . ........ 19, Barvesting........................................ .......... 19 Laboratory Procedures..................... ...... .............. 20 frozen Tissue Tests......................................... 20 Soil Analysis ............ . .................... .. ............ 20 iv TABLE OF CONTENTS (Continued) PAGE RESULTS000000.0.0..00000000000000000000000000000000000000000000. 21 Effects of Treatment on Soil and Tissue Tests and Yields Of corn. 0 000000 0 0 0 0 0 0 0 O 0 0 0 0 0 0 000000000 0 0 0 0 0 0 0 0 O O 0 0 0 0 0 . 0 0 0 0 0 0 21 Rotation Experiment.... ............... . .................. ... 21 Tissue nitrate....... ............... . .......... ........... 21 ri’me phosphor“, 0 0 0 00000 0 0 0 0 0 0 0 0 0 000000000 0 0 0 0 O 0 0 0 0 000000 21 r1.sue pot“.1u0 0 0 0 0 0 O 0 0 0 0 0 000000 , 0 0 0 0 0 O 0 0 O 0 0 0 0 0 0000000000 26 $011 nitrate. 0 0 0 0 0 0 0 0 0 0 I 0 00000 0 0 0 0 0 0 . 0 0 00000 0 0 0 0 0 0 000000 0 0 29 Soil phosphorus.... ...... . ................. . .............. 32 Soil potassium”... ...... ............. .................. ... 35 $011 reaction0 0 0 0 0 0 0 000000 0 0 0 0 0 0 0 0 0 0 O 0 0 0 0 0 0 0 000000000 0 0 0 0 O 38 Corn yields.... ..... . ....... ... ..... ... ................... 38 'com populations 0 0 0 0 0 0 O 0 0 0 0 0 00000 0 0 0 0 0 O 0 0 0 0 0 0 0 0 O 0 0 0 0 0 00000 38 . Re. 1due Experimnt 0 0 0 0 0 0 0 0 0 0 0 0 O 0000000 0 0 0 00000 0 O 0 0 0 0 0 O 0 0 0 0 0 0 43 Tia-“e nitrate. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O 0 0 0 0 0 0 0 0 0 0 0 0 0 O 0 O 0 0 00000 0 0 43 “S'ue phO’phorus 0 0 . 0 0 0 000000000 0 0 0 0 O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 45 Tis‘ue potaaiun. 0 0 0 0 O 0 0 0 0 . 0 0 0 0 0 0 0 . 0 I 0 00000 0 0 0 0 0 0 0 O 0 000000 45 8°11 nitrate. 0 0 0 0 0000000 0 0 0 0 0 0 0 0 0 0 O 0 0 0 0 0 0 0 000000 0 0 0 0 0 00000 45 Soil phosphorus....... ...... ........ ................... ... 49 Soil potassium».............. ......................... .... 51 Soil reaction......... ....... ..... ......... . .......... .... 51 com y161d8000000 0 00000 0 0 000. 000000 0 0 00000 0 0 0 0 00000000 0O 0 O 51 Corn populations...... ............................ .. ...... SS Nitrogen Sources Experiment ................................. 55 Tissue nitrate... ......................................... - 56 Tissue phosphorus ..... ...... ...... ...... ..... ..... ........ ' 59 Tissue potassiuma.. ................................. . ..... 62 Soil nitrate .............................................. 65 Soil phosphorus............ ............ . .................. 68 8°11 potasaium0 0 0 . 0 0 0 0 0 0 . 0 0 0 0 0 0 0 0 0 0 0 0000000000000000000 0 0 0 71 $011 reactionssssssssssesssssssssassasses ooooo s sssssssssss 74 corn Yields 0 0 . 0 00000000000 0 0 0 0 0 0 0 0 O 0 0000000000000000000000 77 com populations 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0000000 80 Seasonal Distribution of Nutrients in Corn Tissue and 1ns°1100000000000.0000 00000 000000 000000000 000000000000 00000 80 General Relationships Between Nitrogen Deficiency Symptoms, Soil and Tissue Tests, and Corn Yields ....... ... ...... . ..... 91 TABLE OF CONTENTS (Continued) I ' DATE Correlation Studies... ......... . ......................... ..... 98 Rotation Experiment, Tissue Tests.... ........... ....... ..... 99 Rotation Experiment, Soil Tests.... .......... ............... 103 Nitrogen Sources Experiment.............. ..... .............. 107 DISCUSSION...................................................... 119 SUNNARY......................................................... 124 CONCLUSIONS..................................................... 126 LIST OF mEnNCESO000000O000000000000.00.00.0000000000000000000 127 vi 10. 11. 12. 13. 14. LIST OF TABLES PAGE PPN.N03-N in tissue by dates of sampling. Ferden Farm rotations l, 6 and 7. Sims clay loam" 1960............... 22 PPNIP in tissue by dates of sampling. Ferden Farm rotations 1, 6 and 7. Sims clay loam. l960............... 24 PPM K in tissue by dates of sampling. Ferden Farm rotations l, 6 and 7. Sims clay loam. l960............... 27 Pounds per acre N03-N in soil by dates of sampling. Perden Farm rotations 1, 6 and 7. Sims clay loam, 1960... 30 Pounds per acre P in soil by dates of sampling. Perden Farm rotations 1, 6 and 7. Sims clay loam. 1960.......... 33 Pounds per acre R in soil by dates of sampling. Ferden Farm rotations l, 6 and 7. Sims clay loam. 1960.......... 36 Soil pH by dates of sampling. Perden Farm rotations 1, 6 and 7. Sims clay loam. 1960....................... ..... 39 Corn yields, fertile stalks, barren stalks and total stalks per acre. Perden Farm rotations 1, 6 and 7. Sims Cl.y1°m0 1960000000000000000000000000.000000000000000000 41 PPN.N03-N in corn tissue by dates of sampling as related to organic residues and nitrogen treatments. Sims clay 1°“. ’arden ram. 19600000000000.00000000000000000000000 44 PPN.solub1e P in corn tissue by dates of sampling as related to organic residues and nitrogen treatments. ’ Sims clay loam. Perden Fanm. 1960 ..... ............. ...... 46 PPM K in corn tissue by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam. Perden Farm.1960. ....... ..................... 47 Pounds per acre soil NO3’ by dates of sampling as related to organic residues and nitrogen treatments. Sims clay 10m. Ferden Pam. 19600000000000000000.00000000000000000 48 Pounds per acre soil P by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam. Petden Farm. 19600000000000000000000000000000000000 50 Pounds per acre soil N by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam. Perden Fam0 1960000000000000.000000000000000000000 52 vii 15. 16. 17. 18. 19. 20. 210 22. 23. 24. LIST OF TABLES (Continued) PAGE Soil pH by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam. Ferden Perm. 1960........ ......................... . ....... 53 Corn yields, fertile stalks, barren stalks and total stalks per acre as related to organic residues and nitrogen treatments. Sims clay loam. Perden Perm. 1960.. 54 PPNLN03-N in corn tissue by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annua1 treatment. Hillsdale sandy loam.. East Lansing. 1960............................ ..... 57 PH! soluble P in corn tissue by dates of sampling as re- lated to nitrogen materials, rates and times of applica- tion in the fourth year of annual treatment. Rillsdale sandy loam. East Lansing. 1960........................... 60 PP! K in corn tissue by dates of sampling as relatedato nitrogen materials, rates and times of application in the fourth year of annual treatment. Nillsdale sandy loam. East Lansing. 1960................................. 63 Pounds per acre N03-N in soil by dates of sampling as related to nitrogen materials, rates and times of applica- tion in the fourth year of annual treatment. Nillsdale sandy loam. East Lansing. 1960........................... 66 Pounds per acre P in soil by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960.....................;........... 69 Peunds per acre R,in soil by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Rillsdale sandy loam. East Lansing. 1960................................. 72 Soil pN.by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Nillsdale sandy loam. East Lansing. 1960............................................. 75 Corn yields, fertile stalks, barren stalks and total stalks as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Nillsdale sandy loam. East Lansing. 1960..... ..... ....... 78 viii TABLE 25. 26. 27. 28. 29. 30. 31. 32. LIST OF TABLES (Continued) PAGE Coefficients of linear correlation among tissue tests and yield. Perden Farm rotations l, 6 and 7 ..... . ......... 100 Regression statistics for the multiple regressions of yield (Y) on tissue tests for N, P and R,on four dates. ’erdn "n rot‘fim 1’ 6nd7000000000000000.00000000000 101 Coefficients of linear correlation among soil tests and yield. Perden Parm.rotations 1, 6 and 7............ ....... 104 Regress on statistics for the multiple regressions of yield on soil tests for N, P and R on six dates. lard“ tamrotations 1, 6‘nd 7000000000000000000000 000000 105 Coefficients of linear correlation among tissue tests and yield. Nitrogen sources, rates and times of application. East Lansing......................... ........ 110 Regressi statistics for the multiple regression of yields ( ) on tissue tests for N, P and E.on three dates. Nitrogen sources, times and rates of application experi- ment. East Lansing..... ......... .......... .......... ...... lll Coefficients of linear correlation among soil tests and yield. Nitrogen sources, times and rates of application experiment. East ms1m000000000000000000 00000 00000 000000 112 Regression statistics for the multiple regressions of yields (T) on soil tests for N, P and K,on five dates. Nitrogen sources, rates and times of application experi- ment. East Lansing................................... ..... 113 ix FIGURE 10. 110 12. 13. LIST OF FIGURES Soluble nutrients in midribs of corn in three rotations on Sims clay loam................... ........ .. Available nutrients in Sims clay loam.planted to corn in three rotations ..................... ... ............ .. Soluble nutrients in midribs of corn following residue treatmntsonSImclay1003............................ Available nutrients in Sims clay loam planted to corn fall-Wing re.1due treatunts0000000000000000000000000000 Soluble nutrients in midribs of corn on Billsdale sandy loam as related to time of application of 80 pounds N from ammonium sulfate..... ........ ............. Available nutrients in Hillsdale sandy loam.p1ented to corn as related to time of application of 80 pounds N frmm1‘m8u1f‘teeseassesseseeeeseeessseoeeeesaeeeee Corn yields vs. tissue nitrate in relation to sampling d‘te’ rotation m nitrogen treamt0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Corn yields vs. soil nitrate in relation to sampling date, rotation and nitrogen treatment...... ............. Corn yields vs. tissue nitrate in relation to sampling date and time of application of two nitrogen sources.... Corn yields vs. soil nitrate in relation to sampling date and time of application of two nitrogen sources.... Corn yields calculated as a function of tissue P on four dates. Rotation experiment, Perden Farm. 1960.... Corn yields calculated as a function of soil tests for N, P and R 17 days after tasseling. Rotation experiment, Perden Farm, 1960............. .............. Corn yields calculated as a function of soil tests for N, P, and R 17 days after tasseling. Rotation uperimnt, Ferden ram, 19600 0 0 00000 0 0 0 0 0 I 0 0 0 0 0 0 0 0 0 0 0 0 0 PAGE 81 82 83 84 85 86 92 93 95 96 FIGURE 14. 15. LIST OF FIGURES (Continued) Corn yields as calculated functions of tissue tests and soil tests (variable N, average P and E), at tasseling time. Nitrogen sources experiment, East Lansing, 1960...... ...... ...... ............. ....... Corn yields as calculated functions of soil tests (variable N, average P and R), 11 days after tassel- ing. Nitrogen sources experiment, East Lansing, 19600000000000000000.000.00.0000.00000000000000000000000 xi PAGE 117 118 Imowc'non Chemical methods for estimating fertilizer requirements of soil date from Liebig's discovery of the principles of plant nutrition. It has been established by many workers in the past that plants obtain nutrients from soils. It was found that only a fraction of the total nutrients in soil is available to plants. So attempts were made to get some measure of the rate at which these nutrients moved into the soil solution. Different methods weredevised by different workers to measure the nutrients in solution. This led to the coining of such words as ‘active,“ "dormant," ”available,” "non-available," "exchangeable,” ”fixed," etc. All these attempts were based on the idea that certain specific extractants would remove from soils these nutrients which are available to plants. Employing this concept, some success was had in predicting crap response to applied fertilizer from chemical tests of the soil. During the last half century, rapid chemical tests of soils and plant tissue have been brought into practice. In the hands of experienced workers such tests are proving more valuable every day. Their use has been extended from laboratory and greenhouse to the field. Rapid tests are simple and inexpensive and afford a means of on-the-spot diagnosis. This has made possible nutritional surveys of relatively large areas in a short time at low cost. It has been possible to establish critical tissue concentrations for each nutrient for individual crops by making observations in which yields are compared to analytical values. Sim- ilarly, critical soil test levels for P, R and some of the secondary and minor nutrients have been established for specific soils and specific crops. Under intensive greenhouse culture, critical soil test levels 1 2 have been determined with a high degree of precision. Under field con- ditions, correlations between crap yields and soil tests are much less exact; the soil tests provide general guides which must be augmented by specific information regarding soil type, cropping and management history and by skill and experience on the part of the prognosticator. Soil tests for P and R.provide a basis for arriving at reasonable rates of phosphate and potash to use on specific soils. No soil test for nitrogen has been found with any widely accepted usefulness in pre- dicting in advance how such nitrogen might be released from the soil dur- ing a given growing season. Rough estimates can be made from a knowledge of soil type, organic matter content and soil management history. Release of nitrate during incubation has been used successfully for predicting fertiliser nitrogen needs of non-leguminous crops, except where these follow forage legumes. The rate of release of soil nitrogen varies during the season and from one season to another. There is need for methods of estimating the current nitrogen status of corn during the growing season. To be most useful, such information should be timely and allow for correction of deficiencies during the current season and before potential yields have been irreversibly reduced. In Japan, excessive fertilization is a problem, whereas in newly developed countries under-fertilization is the rule. In the United. States, soil management practices present the whole spectrum.between these extremes. The need for more accurate interpretation of soil and tissue tests is, therefore, of significant world-wide concern. There is an increasing emphasis on the use of mathematical correla- tion, using functional models which can be readily transformed into 3 economic terms. A large body of economic theory has evolved, based on such functional relationships. The concepts of production economics, however, have been refined far beyond the level of reliability yet re- alised by supporting agronomic data. The objectives of the present study were the following: (1) To investigate the extent to which tissue tests for nitrate used alone or in conjunction with tests for P and K may be employed to determine supplemental nitrogen needs of corn during the growing season. (2) To investigate the relationships between soil levels of nitrate, P and R.during the growing season and response of corn to supplemental nitrogen. (3) To determine 'threshold"’va1ues for nitrates in soil and tissue where actual nitrogen deficits develop prior to the appearance of visible deficiency symptoms in the corn plant. (4) In approaching the above objectives, to use mathematical models of interest to production economists. LITERATURE REVIEW Fresh Tissue Tests The correlation of mineral elements in fresh tissue with crop be- havior was first undertaken in the U. S. A. Boffer (48) made qualita- tive tests for Fe and E.in fresh corn stalks and showed accunulation of Fe at the nodes when R.was limdting. In the beginning, sap was extracted either by crushing green tis- sues or extracting under heavy pressures. Later, simple extracting solutions were used for extracting soluble nutrients. water was used’ by Nightingale (70, 71) for detection of calcium deficiency in sugar- cane and pineapple, and by Harsh (56) for calcium and boron in tobacco and corn. Page and Burkhart (74) used boiling water for extraction of nutrients from fresh material for diagnosis of nitrogen, phosphorus and potassium deficiencies in peanuts and cotton. Emmert (30, 31, 32, 33, 34, 35) used two percent acetic acid for extracting nitrate, phos- phate and potassium. Later Carolus (17, 18)extended the use of the same extractant to include calcium and magnesium in vegetable crops. Thornton (93) et a1., at Purdue, suspended finely chopped tissue directly in the chemical reagents used for detecting the nutrient; e.g. cobaltinitrite for R, ammonium molybdate for P and diphenylamine in sulfuric acid for NO3'N. This method has been used by Scarseth (85), .Atkinson (4) and Hark (98) for the diagnosis of mineral deficiencies in small grains, tomatoes, potatoes and corn. Carolus (17, 18), Hester (47), Peech and English (77) have used a Haring blendor for extraction of nutrients from fresh tissues. Carolus Iased two percent acetic acid, but Hester (47) adopted the acetate buffer 5 solution (pH 4.8) developed by Norgan (66) for the 1atter"s soil test- ing system. Fresh tissue testing under field“conditions has become prominent in all states of the U. S. A. and at many agricultural experiment sta- tions of other countries. Objectives of Tests The immediate aim of fresh tissue testing is to assay the nutrient status of the plant sap. The principal objectives of such an assay have been cited to be: (1) To aid in determining the nutrient supplying power of the soil. (2) To aid in determining the effect of treatment on the nutrient supply in the plant. (3) To study relationships between the nutrient status of the plant sap and crop performance as an aid in predicting fer= tiliser requirements. Theory behind Fresh Tissue Tests In tissue tests soluble tissue constituents are assayed to deter- mine unassimilated N, P, R and other elements. Essentially all of the potassium in plant tissue is present in solution as the cation. The actual level of soluble I or P in the plant at a given time represents an equilibrium.between rate of uptake from the soil and rate of metabolic assimilation within the plant. It has been assumed that imbalanced ‘nutrition is reflected by low levels of one or more nutrients accompanied ‘by abnormally high accumulations of one or more of the other nutrients. Thus, one obtains information regarding the factors of nutrition which 6 may be limiting at any particular period. Since the nutritional status of the plant is not static, the demands of the plant change as it proceeds through its life cycle. Role of Nitrate in the Plant Nightingale and Eckerson (25, 70) have reported that N03“ does not directly affect the growth responses of plants. Environmental factors such as length of the day, light intensity, temperature or other external factors may affect the rate of assimilation of nitrate nitrogen. There occurs reduction of N03' by a nitrate reducing enzyme system (reductase). The NB4+ produced combines with organic acids formed by oxidation off sugars, giving rise to amino acids which are then converted to complex proteins and other organic nitrogenous materials. Active synthesis of amino acids and proteins in many plants does not occur unless the tissues concerned contain a liberal reserve of N03‘. I According to Nehlich (61), N03’ ion is rapidly reduced in plants to NN3, probably in the presence ofla molybdenumecontaining enzyme. In corn, the highest concentration of N03“ is found at the base of the stem. The concentration becomes progressively lower toward the top of the plant. Thus it seems reasonable to suppose that the reduction of nitrate and the assimilation of amnium nitrogen occurs at least to some extent in all parts of the plant. Nightingale (70) found in the pineapple, that soil temperatures of 68° F. or lower resulted in limitation of absorption of nitrate by roOts. Roles of P and R in the Plant There are important interdependent relationships between nitrate, 7 P and E.in the nutrition of plants. Phosphorus functions as a carrier of energy, is required for respiration, and influences the synthesis of proteins. It plays an important role in plant metabolism as an ac- celerator of oxidative enzymes, as a promoter of root development and a regulator of maturity, and as a component of many vital compounds. Potassium serves as a catalyst, condensing agent and translocation regulator, is necessary for the formation of carbohydrates, oils and proteins, stimulates enzyme activity, and is required for cell division, reduction of nitrates, and chlorophyll formation. 'Correlation of Tissue Tests with Crop Response Pettinger (82) extracted corn sap with a hydraulic press at a pressure of 6500 pounds per square inch from 15-inch sections of stalk immediately above ground and found that the concentration of nitrate nitrogen showed only a fair degree of correlation with nitrogen fertili- zation. It was well correlated, however, with the soil supply of NO3’N. He suggested the following nutrient concentrations in corn sap as standards in diagnosing soils: Very deficient NO3'N: Less than 100 ppm. Noderately deficient N03'N: 200 ppm. Ample NO3'N: More than 300 ppm. Lynd (53) found that tissue tests for N, P and R reflected soil fertilizer treatments on soil types varying widely in physical and chemical dharacteristits. Tissue tests throughout two growing seasons on rotation fertilizer experiments indicated that nitrogen was the limiting factor 1n.plant growth and that a definite change took place in the nitrogen status of plants with the initiation of the flowering period. 8 Brock (16) tested soil and corn tissue at frequent intervals dur- ing the growing season in 1958 and 1959 in two of the rotations studied by Lynd. Re found that levels of NO3' in both soil and tissue dropped sharply during a critical period beginning a week or two before tassel- ing and continuing through the pollination period. In Sims clay loam soil, corn yields were limited by nitrogen availability at levels of soil nitrate below 50 to 60 pounds I per acre during this period. In productive soils or where legumes or manure are plowed down, soil or tissue tests for nitrate may not reveal shortages until this critical period of nitrogen demand is reached. Usually this is too late to add supplemental nitrogen with conventional equipment. However, deficient tests during this period may be used in planning fertilizer and management programs for subsequent crops. According to'Nagnitski (55), it is possible to outline rough indices of composition of corn midribs for obtaining high yields of green matter. The content of nitrate N in the earlier phases of corn development be- fore appearance of bloom.must be in the range of 800 to 1500 ppm., phos- phorus content must be 40 to 100 ppm. and potassium content, 3000 to 4000 ppm. of fresh matter. In young plants showing deficiency symptoms, he found the content of various elements in fresh matter of midribs to be as follows: nitrate nitrogen, traces to none; phosphorus, 20 ppm. to traces; potassium, 600 to 1500 ppm. Parts and Number of Plants to be Tested Selection of the tissue to be tested in various crops and inter- pretation of the analyses under varying environmental conditions for thy given crop require attention to a number of factors. It is essen- 9 tial to test that part of the plant which will give the best indication of the nutritional status. Certain principles have been fairly well established by many workers. lany workers (21, 72) have suggested the following plant parts to use for a tissue test on a fully developed plant: crop Nitrogen Phosphorus Potassium Corn Kain stem or Leaf midrib Blade tissue leaf midribs near ear near ear Thornton et a1., (93) suggested sampling eight to ten plants from a single plot. In later stages of growth, leaves for analysis are selected which have finished their growth and are physiologically active. These are located somewhat higher on the stem than in earlier periods. Lynd (53) found that the third functioning basal leaf of corn plants is quite reliable as an indicator of plant nutrient status. Time of Day for Tissue Tests Helch (99) made hourly determinations from early morning until sun- set and found that the nitrate status of corn plants was lowest in early morning. As the day advanced, the nitrate level increased, reaching its peak from about 11:00 a.m. to 3:00 p.m. after which it decreased to a level somewhat above that of the early morning. Fallon (37) suggested that corn and small grain plants should not be tested in early morning hours since nitrate tends to accumulate over night. Testing during later forenoon and early afternoon was found to give accurate readings. According to him, nitrate tests should be avoided within 24 hours after a rain, since results may then be lower than normal. Further, he pointed out that lower leaves at the base and younger, im- 10 mature leaves at the top should be avoided. Plants in the same field may vary in nutrient content, so several samples should be taken from the same field for average reading. Stage of Growth for Tissue Tests The most critical stage of growth for tissue tests of corn is from silking through early ear formation. To follow the nutritional trends of the plants more closely, Krantz et a1., (50) suggested that tests be made at the following five stages of growth: (1) Corn two to three feet high, about five weeks after planting. (2) Tassels just emerging, about nine weeks after planting. (3) Silking stage, about 11 weeks after planting. (4) Late milk stage, about 13 weeks after planting. (5) Husks beginning to dry, about 15 weeks after planting. Soil Tests Soil testing is currently the most popular diagnostic technique used in.making fertilizer and lime recommendations. Routine soil test- ing services are offered by both public and private agencies in most states. The interpretation of soil tests, however, is still a matter for controversy. The precision with which fertilizer requirements can be predicted from soil tests falls far short of the precision possible in the tests themselves. This is because mathematical correlations wdth experimental data are frequently not good enough to justify using ‘them except as a guide. Or, if good mathematical correlations are ob- tained, other soil or crop factors, or climatic factors, or economic considerations frequently make it advisable to use different fertilizer applications for a given soil test (10). A great deal of research 11 activity in the area of soil fertility is directed towards the problem of correlating soil test results with crop response and economic fer- tilizer use. Soil Tests in General Use The essential difference between the various soil testing pro- cedures currently in use is in the extracting solution which is employed. The actual chemical determinations for extracted nutrients range from qualitative “quick tests" suitable for use in the field to quantative. laboratory determinations. An early objective in soil testing was to find a selective solvent or extractant which would simulate the seasonal capacity of a crop to remove nutrients from.the soil (24). This objective has been shown to be invalid, by and large. However, a high degree of selectivity is still recognized as an essential property of a useful soil extractant (13). The extractant must remove nutrient forms which are significant in crop nutrition. Exchangeable potassium.is considered to represent the major avail- able form.of this nutrient in soils. It is the form measured in a number of soil testing systems (12, 78). In general, however, potassium.ex- tractants have been selected with ionic strengths such as to give more or less relative estimates of exchangeable potassium in the major agricultural soils of a given region (52, 62, 66, 72, 91, 102). Phosphorus occurs in soils in numerous forms (19). The availability to plants of various soil phosphorus fractions has been investigated by numerous workers (9, 13, 20, 29, 76). Phosphorus uptake has been principally associated with acid-soluble. 12 and adsorbed phosphate_in the range from pH 5.0 to 8.0. The availabil- ity of acid-soluble forms is sharply suppressed at soil pH's above 8.0. The availability of the phosphate ion decreases with decreasing pH below pH 6.0 -- due to the formation of iron and aluminum.phosphates, which are only slightly soluble in acid extractants but are increasingly soluble as the pH of the extractant increases. Organic phosphorus fractions appear to contribute only indirectly to the available supply (27). ' ' ‘ Accordingly, phosphorus extractants have been selected to give relative measures, (a) principally of acid-soluble phosphorus (8, 36, 47, 52, 62, 66, 72, 91, 93, 94, 102), (b) principally of adsorbed phosphorus (73), or (c) of acid-soluble plus adsorbed phosphorus (15). Biological assays have been widely used to assess phosphorus- ' availability in soils. These include vegetative tests such as the leubauer tests (60) and the determination of TA" values, using P32 (42). Microbiological assays have been used for phosphorus, as well as for potassiumhand nitrogen (l, 81). Nitrate, or forms of nitrogen which can be readily converted to nitrate by soil microbial processes, are generally accepted to be the available forms of this nutrient. litrate levels fluctuate rapidly with crop removal, organic amendments, leaching movement and with microbial release from organic matter. In humid regions, soil nitrate levels in the field frequently show little relationship to crop growth because the quantities of nitrate found usually are small relative to the potential for release of nitrate from organic forms during the growing season. Also, leaching results in movement of nitrate to hori- zons which are usually not sampled but which are reached by plant roots 13 during the growing season. When nitrate in these deeper horizons has been taken into account, good correlations with crop yields have been obtained (100), particularly under send-arid conditions (57). 'Under intensive greenhouse management, much higher levels of nitrate are main- tained, and these are realistically related to crop response (58, 90). Because early attempts to relate tests for soil nitrate with crop response were generally unsuccessful, investigators turned to methods for estimating the potential availability of soil nitrogen (2). No single test for soil nitrogen availability has been generally accepted. In some advisory programs, total nitrogen is determined chemically or estimated from organic matter determinations. Empirical availability factors are employed to calculate the soil nitrogen contribution to specific crops (87, 103). Hivailablei nitrogen has been estimated chemically using mildly oxidative extractants (51, 81). Or available nitrogen.may be estimated by a vegetative test, such as the Ueubauer test (60). Production of nitrate during laboratory incubation of soil has been used to estimate soil delivery during the growing season, successfully in some areas (41, 67); unsuccessfully in others (3, 51). In some states, Michigan for one, no test for soil nitrogen is used. Nitrogen fertilizer recom- mendations are based on previous management and the needs of specific crops as determined by field experiments on groups of similar soils (lichigan Agr. Bxpt. Sta. Bul. 3-159). Soil Test Correlation The basis for interpretation of soil tests in terms of fertilizer recommendations varies considerably from state to state. With vegetable crops on coastal plains soils in the eastern United States, soil tests 14 are subtracted directly from known requirements for a desired produc- tion level, the balance being supplied as fertilizer, taking into ac- count average availability coefficients for specific fertilizer materials (47). More frequently, soil tests are categorized as high, madium.or low, or into responding and non-responding ranges. Fertilizer rates or ratios are then adjusted in accordance with experience gained through field fer- tilizer trials on the same or similar soils (66, 93, 94). In essence, these correlations between yield and soil test are based upon Liebig's law of the first limiting nutrient. Critical levels of specific nutrients and balance among the several nutrients are essential concepts employed. 'Hathematical correlations of crop response with soil tests have been developed by a number of investigators. In general these have been based upon the liitscherlich-Baule equation (7, 64) or some modification of 1: (10, 12, 13, 14, 44, 90). Son. limitations‘on the application of these functions to prediction of fertilizer needs have been pointed out by Black (10). In recent years, agronomists and economists have concerned theme selves primarily with the functional analysis of crop response to ap- plied nutrients, with little regard for soil tests (68). The percentage sufficiency concepts of Hitscherlich (64), Baule (7) and Spillman (88, 89) have been exploited by some investigators (28, 38, 101). Polynomial and square root functions have been employed by others because they per- mit simpler calculation of coefficients by least squares methods. These fertilizer response functions allow for diminishing returns but are not ‘baaed on any logical concept of the nutritional behavior of the plant. Empirically, they have been found to fit experimental data rather well (22, 46, 92). 15 Some acceptable agronomic principles have been demonstrated by these procedures, such as, that the profitability of nitrogen fertilizer response is dependent upon the stand of corn (80). However, the concept of nutrient substitution, which has developed from such studies, has been difficult for agronomists to reconcile with long accepted principles of plant nutrition. According to this concept increases of one nutrient can substitute for decreasing supplies of another to maintain a constant yield. This would place qualifying restrictions on such long-accepted agronomic concepts as critical levels, nutrient balance and luxury con- sumption. The physiological basis for nutrient substitution among the major elements has not been explained. However, recent studies involving correlation of corn yields with leaf composition have shown that wide fluctuations in the ratio of I to P in corn tissue can occur without affecting yield. Isoquant yield diagrams fitted to the leaf composition data were similar in form to those usually obtained for fertilizer in- put data (23). Potentially, the functional analysis of fertilizer response data can be very.useful. well-developed concepts and computational procedures are available for incorporating cost-price variables and calculating optimum.fertilizer rates and ratios for specific economic and manage- 'ment situations. However, practical application is hindered by the variability of supporting field experimental data. variations in fer- tilizer response from field to field and from year to year are great. Yield equations need to be generalized by including additional variables, such as climatic factors and soil properties, including soil tests. Much.current research is being directed toward this end (75, 79). MATERIALS AID METHODS Field Experiments Experiments were carried out on three soil types. 811: different field experiments were used for the studies on corn in 1960. Three of these were parts of previously established experiments involving sys- tematic fertilization and management of four to twenty years' duration. One involved a comparison of current response with residual response to fertilizer applied one year previously. Two experiments were con- cerned only with response to the current year's fertilizer treatment. Results from the last three experiments were enigmatic and have not been fully evaluated. The present report deals only with data from the three long term experiments . Rotation Experiment This rotation experiment was established in 1941 by the Soil Science Department of Michigan State University on the Eerden Farm in Saginaw County. The soil at this location is Sims clay loam. The purpose of the experiment was to compare seven systems of farming at two fertility levels and two levels of supplemental nitrogen fertilization. Out of seven rotations, three five-year rotations were selected for the present study: ’ (a) A rotation having two years of alfalfa-brome meadow (Rotation number 1) (b) A cash crop rotation without green manure or forage legumes (Rotation number 6) (c) A rotation having two leguminous green manure crops (Rotation number 7) 16 l7 Rotation number 1 was comprised of alfalfa-brome, alfalfa-brome, corn, beets, and barley. Rotation number 6 had beans, wheat, corn, beets and barley, and rotation number 7 had beans, wheat (gm)S, soy- beans, beets and corn (gmo. " These rotations were replicated four times. A split-split plot field design was used.“ Rotations comprised the main plots, with fer- tility levels as sub-plots and supplemental nitrogen treatments as sub- sub-plots, gain plots were 28 feet by 90 feet and were subdivided longitudinally. High fertility-sub-plots received 1600 pounds per acre of 5-20-10 over the five-year rotation period (400 pounds on corn); low fertility sub-plots received 800 pounds over five years (200 pounds on corn). One half of each fertility level sub-plot received no supplemental nitrogen, the other half received 100 pounds per acre of l sidedressed on corn and 50 pounds sidedressed on beets, beans and wheat. The present study was concerned only with corn grown in 1960 -- during the fourth cycle of each rotation. Residue Experiment This rotation was begun in 1951 on the Ferden Farm on Sims clay loam. Its purpose was to determine if the alfalfa-brome hay that is normally removed could be replaced by sawdust or straw to maintain yields and soil building qualities in the rotation. The present study deals with the comparison of residual effects of the residue treatments and supplemental nitrogen on levels of R03", P and R in soil and corn tissue and on yield of corn in 1960. The corn in 1960 was planted at the beginning of the second cycle of a five-year rotation (corn, beans, *gm.- green manure (sweet clover planted as catch crop in wheat and corn). 18 barley, and two years of alfalfa-brome). This was six years after the initial residue treatments on a block of plots established in 1954. Four residue treatments were involved as follows: (1) Two years of alfalfa-brome hay cut and removed. This treat- ment repeated each cycle of the rotation. (2) One year of alfalfa-brome hay cut and removed and the second year alfalfa cut, weighed and left on the plot. This treat- ment repeated each cycle of the rotation. (3) Thirty-five tons per acre of sawdust was added after removing alfalfa-brome at the beginning of the experiment (six years before the 1960 corn crop which was used in this study). (4) Three to four tons of wheat straw per acre applied after re- ‘moval of the second year of alfalfa-brome. This treatment repeated each cycle of the rotation. All treatments were replicated five times. A split-plot design was used. Residues comprised the main plots, with supplemental nitrogen treatments as sub-plots. llain plots were 14 feet by 90 feet and were subdivided longitudinally to give 7 feet by 90 feet sub-plots. Fertilizer had been applied at rates of 100 pounds per acre of 5-20-10 for corn, 200 pounds 0-20-10 for beans and 240 pounds 5-20-10 for barley. lo fertilizer had been used on the hay meadow. Supplemental nitrogen had been applied on one half of each residue plot at the rate of 100 pounds of nitrogen for corn and 40 pounds on beans and barley. Nitrogen Source, Rate and Time of Application Experiment This experiment was located on Hillsdale sandy loam at Michigan State University, Wilcox field. Corn had been grown continuously for ‘the four years of the experiment. 19 Two nitrogen carriers were used -- calcium nitrate and anonium sulfate. They were compared in fall, spring and summer applications at two rates, 40 and 80 pounds I per acre, the same treatments having been applied on the same plots over the four year period. Plot size was 14 feet by 50 feet with 42-inch row space between corn rows. Three hundred pounds 0-20-20 was plowed under and 200 pounds 0-20-20 was applied with the seed. Field Procedures Sampling At each location, periodic soil and tissue samplings were made during the growing season. Twenty soil cores to an eight-inch depth were composited per plot. These were passed through a four-mesh screen, thoroughly mixed and extracted for the nitrate determination within 24 hours of sampling. The balance of the sample was rapidly air-dried and held for determination of available P and R at a later date. Leaf mid- ribs opposite the basal ear were collected from eight plants per plot. Hidrib samples were taken in the late forenoon or early afternoon. These were placed in plastic bags and quick-frozen for later determina- tion of l03'-N and soluble P and K. At the time of removing the leaves, a quick test for nitrate was also made on the severed stub of each mid- rib. This was done to permit calibration of the quick test against the quantitative tissue nitrate determination. This calibration has not been completed. Harvesting Two rowa of corn per plot were harvested mechanically. Moisture lamplea were taken, and final yields were calculated to bushels per 20 acre at 1535 percent moisture content. Laboratory Procedures Frozen Tissue Tests Five gms. of finely cut frozen tissue were placed with 50 ml. of two percent acetic acid in a flaring blendor, and a small amount of char- coal (Darco G 60) was added. This was macerated for five minutes and filtered to give a clear filtrate. litrate in the filtrate was determined by the phenoldisulfonic acid method, adapted for photometric determination from Harper (45) and Prince (83). ' ’ E ' Phosphorus was determined by the colorimetric method described by Fiske and Subbarow (40). Potassium was determined with a flame photometer. Soil Analysis Field fresh soil was extracted for the nitrate determination with .0211 013804. Calcium hydroxide was used to clarify the extracts (45, 49). litrate I was determined colorimetrically after reaction with brucine (86). .Soil phosphorus was determined colorimetrically in 0.025 I hydro- chloric acid and 0.03 I amonium fluoride (Bray P1, adsorbed) (15). Soil potassium‘was determined by flame photometer in Spurway.re- serve extracts (0.13 l hydrochloric acid) (91). Soil pH was determined by glass electrode. RESUETS Effects of Treatment on Soil and Tissue Tests and Yields of Corn Rotation Experiment Tissue nitrate Data presented in tables l-a and l-b show l03' in tissue of corn grown on Sims clay 1oam.at the Ferden Farm. Rotation 1, having two years of alfalfa-brome plus manure preceding corn, reflected the highest l03' accumulation in tissue throughout the season. The lowest accumula- tion of l03‘ early in the season was found in rotation 6, where corn was grown in rotation with other cash crops. Rotation 7, where two sweet clover catch crops were grown for green manure, occupied an in- termediate position early in the season. There was a tendency for the relative rank of rotations 6 and 7 to change during the season. How- ever, later differences were not significant and there was no signifi- cant date x rotation interaction. The rate of application of fertilizers did not have any effect on '03' accumulation. Supplemental nitrogen favored increase in tissue l03' in all rotations and also at both high and low levels of fertilizer application. There were no significant interactions between the treatment variables. Individual factors af- fected tissue nitrate in essentially the same way at each level or combination of the other factors (table l-b). Tissue phosphorus Soluble phosphorus accumulated to higher levels in corn midribs in Rotation 6 than in Rotations 1 and 7 (tables 2-a and 2-b)l High accumulations of phosphorus in Rotation 6 were associated with low 21 22 Table l-a. -- PPH.H03-H in tissue by dates of sampling. Ferden Farm rotations 1,6 and 7. Sims clay loam..1960.(Tassel- ing date: July 25.) Date Seasonal Treatment AQQde_ 7/18 7(22 8(11' 9114 Average Rotation 1 31 347 171 ,149 241 227 Rotation 6 R6 146 85 137 206 144 Rotation 7 R7 267 110 118 187 . 171 15005 89 as as as 37 800 lbs. 5-20-1o** pl 261 127 146 207 185 1000 lbs. 5-20-10 32 245 118 125 215 176 LSDos . as as us as as so Suppl. 3 I1 228 116 121 215 170 Sidedressed u** Hz 278 129 149 .207 191 LSDos NS NS 27 NS NS 11:1 335 162 157 216 218 2132 359 181 142 265 237 2531 164 102 139 210 153 3592 128 69 136 203 134 3731 285 116 141 196 185 R7F2 249 105 96 177 157 13005 (3 within p) 123 as . as as 50 18005 (3 within a) as us as us as nlnl 355 154 138 238 221 aluz 339 189 160 244 233 2631 112 81 130 204 — 132 2682 180 89 145 - 208 155 nynl 218 112 93 205 157 R7H2 316 109 143 169 184 13005 (2 within a) 111 as as as 47 LSDos (H within R) H8 H8 ' 47 NS NS rlul 227 111 122 221 . 170 Fllz 296 142 _169 .194 , 200 rzul 230 120 120 210 170 rznz 261 116 130 220 181 .. {- i‘ r Lsnos (3 within a) as as as _ns 'us 13005 . (3 within 3) us as 39 as as *Total over 5-year rotation. **100 pounds H per acre on corn; 40 pounds on other row crops and wheat. 1960 l applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. 23 Table l-b. -- PPM H03-N in tissue by dates of sampling. Ferden Farm rotations l, 6 and 7. Sims clay loam. 1960. '(Tassel- ing date: July 25.) fiyggg: Seasonal Beet—met 11.1.8 14.2.9 BILL— filLJme RIFINI 334 127 138 222 205 ab abc ab a sh RIFI'Z 336 1’8 177 210 230 ab a a a a R1F2l1 375 181 139 253 237 a ab ab a a 31"282 343 180 144 278 236 ab ab ab a a 363131 125 79 120 207 133 d c .ab a c 363132 203 125 158 212 174 bed abc a a abc R6F2H1 99 84 141 202 132 d be ab a c R6F2H2 157 54 131 205 137 cd c ab 8 c R7F1H1 221 129 108 232 172 abcd abc ab a abc 117111112 349 104 173 161 197 ab abc .3 a abc 37F2N1 215 96 78 177 141 bed be h a be R7F282 283 114 114 177 172 she abc ab a abc a, b, c, d - Ranges of equivalence (Duncan, 1955). Within columns, numerical values with the same literal subscripts are not different at 5 percent. 24 Table 2-a. -- Pm P in tissue by dates of sampling. Ferden Farm rotations 1, 6 and 7. Sims clay loam. 1960. (Tassel- ing date: July 25.) 5' Date ' Seasonal treatment Code 7118 7 29 8 11 9 14 Ayers Rotation 1 11 106 106 103 73 97 negation 6 36 148 135 152 143 145 Rotation 7 s7 68 73 61 59 65 15005 23 as 48 56 21 800 lbs. 5-20-10* 81 112 102 110 87 103 1600 lbs. 5-20-10* 82 103 108 101 96 102 1.8005 ' as us as as as an Suppl. 8 n1 119 109 126 114 _ 117 Sidedressed R** Hz 96 100 85 69 87 LSDos ‘ 13 ins 22 28 14 2121 107 114 102 58 95 2132 105 98 105 88 99 2521 158 ‘121 160 144 146 25P2 138 148 144 143 143 27P1 71 70 ~ 66 60 ‘67 3732 65 77 55 57 64 LSD05 (R within F) 30 RS 51 63 23 LSDos' (8 within a) as as us as as 2181 111 111 111 66 100 nlnz 101 100 96 80 . 94 8551 186 151 205 214 189 3582 111 118 99 73 100 8731 61 * 65 63 63 63 n7u2 76 ‘ 82 60 54 68 18005 (R.within s) 28 as 54 66 23 28005 (9 within a) 23 as 38 48 24 rlsl 129 96 132 110 117 3132 95 107 87 65 88 r231 109 122 120 118 117 3232 96 93 83 73 -~ 86 18005 (9 within a) as as as as as LSDos (3 within 3) 19 as 31 40 20 *Total over 5-year rotation. **100 pounds I per acre on corn; 40 pounds on other row crops and wheat. 1960 l applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. 25 Table 2-b. -- PPHLP in tissue by dates of sampling. Ferden Farm rotations 1, 6 and 7. Sims clay loam. 1960. (Tassel- ing date: July 25.) Date if Seasonal Treatment " 7118 7129 8(11 9/14 Average RIFINI 123 123 ‘121 50 104 b be b b b klrllz 91 105 83 67 86 beds bed be b bed 313231 98 100 101 82 95 bed ede be b be 313232 111 96 108 93 102 be cdef be b b 263131 203 102 210 211 181 a ede a a a earls; 113 140 111 77 110 be b be b b 268281 168 201 201 217 197 a a a a a R6P2N2 108 96 87‘ 68 90 be cdef be b bed Ryllfll 61 _ 62 - 67 71 65 e f be b de agrlsz 82 77 66 50 69 ede def be b ede. R7FZN1 61 67 58 56 60 e ef c b e 273282 70 87 53 58 67 de def c b de a, b, ... f -- Ranges of equivalence (Duncan, 1955). Within columns, numerical values with the same literal subseripts'are not different at 5 percent. level: Suppl part1 of fl inc: m 1111 the oi 3P1 th. 26 levels of nitrate prior to and at tasseling time (tables l-a and l-b). Rate of fertilizer application had no effect on tissue phosphorus. Supplemental nitrogen reduced the concentration of tissue phosphorus, particularly in the cash crop rotation (Rotation 6) and at the low level of fertility. Growth and final yields of corn in this experiment increased with increasing nitrogen supply. ~ Tissue phosphorus accumlated where growth was restricted by inadequate nitrogen in Rotation 6. Thus, inadequate nitrogen nutrition, reflected in low tissue nitrate tests, restricted the assimilation of phosphorus in the plant, allowing equilibrium levels of soluble, unasshmilated phosphate to increase. where no nitrogen was applied on Rotation 6, tissue phosphorus continued to increase through the last two samplings. Where supplemental nitrogen had been used on Rotation 6, tissue phosphorus declined during the last two samplings to levels similar to Rotations 1 and 7. These seasonal changes were associated with a statistical date x rotation x nitrogen interaction significant at five percent. Tissue potassium Potassium accumls'tion in the green tissue (table 3-a) was found to be greater in Rotation 7 than in Rotations 1 and 6. Low or high rates of application of fertilizer, as well as supplemental nitrogen treatment, had no consistent effect on potassium accumulation. However, about tasseling time, there was a highly significant rotation x fertili- zer x nitrogen interaction. The data for the July 29 sampling in table 3-b show that the relative ranks of the three rotations were significantly altered by both fertility level and supplemental nitrogen treatments during this period of high nutrient requirement. As in the case of 27 Table 3-a. -- PPHLR.in tissue by dates of sampling. Ferden Farm“ rotations 1, 6-and 7. Sims clay loam. 1960. (Tassel- ing date: July 25.) Date Seasonal Trea‘ nt gods ~7 18 7 29 8 ll 9 14 ' e e Rotation 1 21 337 268 277 , 310 298 Rotation 6 R5 383 297 274 320 318 Rotation 7 27 403 348 294 380 ‘ 356 13005 33 45 as 21 17 800 lbs. 5-20-10* :1 381 296 276 342 324 1600 lbs. 5-20-10* 32 367 ‘313 288 332 325 LSDos as us as as us no Suppl. 8 N1 393 303 285 326 327 Sidedressed n** s; 356 305 278 348 322 LSDOS as as as as as 2121 346 255 269 321 298 3122 327 281 286 299 298 2521 391 283 280 318 318 R632 374 311 267 323 319 nyrr 406 350 278_ 386 .355 2722 401 346 310 375 358 13005 (R.w1th1n r) 62 55 as 49 24 18005 (9 within a) as as us as us 21:1 348 275 280 299 300 2132 326 261 275 321 296 R631 405 298 279 316 324 2682 361 295 268 325 312 2791 426 338 296 364 356 2732 380 358 291 396 357 LSDOS (R within N) 59 56 RS 42 25 18005 (n within a) as as as us _ us pill 408 292 291 333 331 r182 355 300 260 350 316 9281 f 378 315 280 319 323 3292 357 310 296 345 327 13005 (F‘within s) as us as as as LSDos (a within 3) as as us as us *Total over 5-year rotation. **100 pounds K per acre on corn; 40 pounds on other row crops and wheat. 1960 N applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. 28 Table 3-b. -- m R in tissue by dates of sampling. Ferden Perm rotations l, 6 and 7. ing date: July 25.) Sims clay loam. 1960. (Tassel- - Date Seasonal Treaggggt 7118 7129, 8111 9114 Avergge erlll 363 288 288 301 310 ab abcd a cd bed erllz 330 222 250 342 286 b d a abcd d 313231 332 261 272 297 290 b ed a d ed R F 2 322 301 300 301 306 1 2. b abc a cd bed Rgtlll 415 250 295 316 319 ab ed a bed bed aarlnz 368 316 266 321 318 ab abc a shed bcd asrzal 395 347 263 316 330 ab ab a bed abc Rgrzlz 353 275 271 330 307 ab bed a abcd bed R7F1fl1 446 338 290 383 364 a ab a abc a 375132 366 361 266 288 345 ab a a ab ab = RyTzll 407 337 303 _345 348 ab ab a abcd ab 273232 395 356 317 405 368 ab a a a a a, b, e, d - Ranges of equivalence (Duncan, 1955). Within columns, numerical values with the same literal subscripts are not different at 5 percent. 29 phosphorus, potassium levels tended to be lowest where corn was most vigorous (Rotation 1) and tissue nitrate levels were highest (table l-b). —Haximum potassium levels in Rotation 7 were associated with low tests for phosphorus (table 2-b). Thus potassium levels tended to be related inversely to both nitrogen and phosphorus. Soil nitrate Ritrate accumulations in soil were affected significantly by rota- tions and supplemental nitrogen treatments through most of the season (tables 4-a.and.4-b). The.high.level of organic additions in the form of alfalfa residues and livestock manure in Rotation 1 resulted in the maintenance of relatively high nitrate concentrations through the last sampling date. The two sweet clover green manure crops in‘Rotation 7_ ‘ contributed to significantly higher soil nitrate levels up until tassel- ing time in this rotation than in the cash crop rotation (Rotation 6). There was no evidence on June 21 of residual carryovers from previous years' nitrogen sidedressings, nor of any response to 50 pounds of nitro- gen sidedressed on June 9 of the current season. A second 50-pound . application of nitrogen was made with the last cultivation on June 30. Through the balance of the season, soil nitrate in sidedressed plots remained 6 to 13 pounds higher than in those which received no supple- *mental nitrogen. This compares with an 18 to 24-pound differential main- tained by Rotation 1 over Rotation 6. The level of fertility had no effect on soil nitrate levels in general. However, on July 29, a rotation at fertilizer interaction was encountered, such that nitrate levels were significantly reduced in Rotation 1 by the higher rate of fertilization. A.similar relationship 30 Table 4-a. -- Pounds per acre 803-N in soil by dates of sampling. Ferden Farm rotations 1, 6 and 7. Sims clay loam. 1960. (Tasseling date: July 25.) Date Seasonal Treatment Code 6/21 712 7118 7129 8111 9114 Average Rotation 1 R1 42 37 37 35 38 25 36 Rotation 6 R6 19 13 14 16 16 7 14 Rotation 7 R7 29 18 19 19 19 7 18 LSDOS 7 4 5 6 10 11 3 800 lbs. 5-20-10* F1 31 24 22 25 24 14 23 1600 lbs. 5-20-10? F2 29 21 25 22 25 12 22 1.8005 NS NS NS NS NS NS NS No Suppl. 8 N1 31 20 19 19 18 8 19 Sidedressed N** N2 29 26 28 28 31 18 27 ‘ LSD05 NS 4 2 4 7 6 2 R131 44 '39 38 40 40 27 38 Rle 40 35 36 30 37 23 34 R5F1 20, 13 12 16 14 8 14 Rng 19 13 16 16 19 6 15 R7F1 30 20 15 18 18 7 18 R7F2 27 16 23 21 19 7 19 LSD05 (R within F) 7 8 7 8 11 12 3 LSDOS (F within R) NS NS NS 7 NS NS NS R181 44 35 32 28 32 14 31 R182 40 39 42 43 45 35 41 R681 20 10 10 12 9 5 11 R6N2 19 16 18 20 23 9 18 R.ll 29 15 14 16 14 5 15 R7N2 28 22 24 22 24 10 22 LSDos (R within I) 8 7 6 7 13 13 4 LSDOS (I within 3) NS NS 4 7 12 10 3 F131 31 22 l9 18 17 7 19 F132 31 27 25 31 30 21 27 F281 30 18 19 19 19 9 19 Fzflz 27 24 31 26 31 16 26 LSDOS (F within N) NS NS 4 NS NS NS NS LSDos (8 within 1?) us 6 3 6 10 8 1 *Total over S-year rotation. **100 pounds N per acre on corn; 40 pounds on other row crops and wheat. 1960 N applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. 31 Table 4-b. -- Pounds per acre 803-! in soil by dates of sampling. Ferden Farm rotations l, 6 and 7. Sims clay lomn. 1960. (Tasseling date: July 25.) Date ;1 " Seasonal eatment 6 21 7 5 7/18 7129 8 ll 9 14 Avera e thlfll 46 38 33 27 33 l 13 32 a a b bed she c 'be thllz 42 40 42 53 46 40 44 a a a a a a a Rllzll 43 31 31 28 31 16 30 a ab b be abc be R P 37 38 41- 33 44 30 37 1 2‘2 ab a ‘a b ab ab b R‘PIRI 19 10 10 11 8 4 10 ed d e e d c h 269132 20 17 15 21 20 12 17 ed ed ede ede ed c defg R6F281 21 ll 10 13 11 5 12 ed d e e d e gh ‘6’2'2 18 15 21 20 27 7 18 d cd c ede abcd c def 37'131 30 17 l3 16 11 4 15 be ed de de d c fg R7P1R2 30 24 18 20 25 ll 21 be be ed ede bed e de 3715111 28 ‘ 13 16 17 16 5 16 bed ed ede ede cd c efg R7F2N2 27 19 30 24 23 10 22 bed bed b bed ed c d a, b, ... h - Ranges of equivalence (Duncan, 1955). numerical values with the same literal subscript are not different at 5 percent. Within a column, 32 with fertilizer rate was maintained in all samplings, although the diff ferenees were smaller and not statistically significant. This suggests that more rapid growth of corn on the high fertility plots resulted in more rapid uptake of nitrogen, thereby lowering the equilibrium level of nitrate in the soil. A.simd1ar effect was observed during the first two samplings in Rotation 7. In both of these rotations there "“6‘ small yield response to fertility level (table 8-a). It is of interest to note that relationships between soil nitrate and.treatment were more distinct and differences attained a much higher order of_statistical significance than was true for the tissue nitrate determinations. Soil phosphorus Available soil phosphorus values (tables 5-a and S-b) reflect the heavy demand for phosphorus of the vigorously growing corn in Rotation 1. These low soil phosphorus values were associated with high soil and tissue [03' and with moderately high tests for tissue phosphorus in this rotation. Final yields were also higher than in Rotations 6 and 7. In the ease of Rotation 7, relatively high soil tests for_phosphorus were accompanied by the lowest levels of phosphorus in the tissue (table 2-a). Some unknown factors were obviously acting in Rotation 7 to en- hance asshmilation of phosphate in the crop or to retard its uptake from the soil. ‘Soil phosphorus was consistently four or five pounds higher at the high rate of fertilization than at the low, except right at tasseling time, when no significant effect was observed. The lack of significant differences at tasseling time may have been due to reduced rate of uptake 33 Table S-a. -- Pounds per acre P in soil by dates of sampling. Ferden Parm.rotations l, 6 and 7. Sims clay lean. 1960. (Tasseling date: July 25.) . Date Seasonal Treatment Aggde 6.21 7 7 18 7 29 8 11 9 14 Aver e Rotation 1 21 14 22 24 25 14 14 19 Rotation 6 35 22 28 30 28 21 21 25 Rotation 7 27 24 32 31 29 23 23 27 LSDOS 7 8 NS NS 4 7 3 800 lbs. 5-20-10* :1 17 f25 26 26 18 17 22 1600 lbl. 5-20-10* 92 22 1 29 31 28 21 22 25 18005 ’ 3 3 4 as 2 2 1 an Suppl. l N1 20 27 30 26 20 21 24 Sidedressed N** 82 19 27 26 28 19 18 23 13005 as as 4 as as 2 1 21:1 12 22 21 24 11 11 17 2132 15 22 .27 25 16 16 20 2621 18 27 29 26 20 20 23 2512 25 29 31 30 23 23 27 2731 22 28 27 29 22 ~ 21 25 2712 26 36 35 29 25 25 29 18005 (3 within 9) 7 8 as as 5 8 3 13005 (3 within a) 5 5 7 s8 3 4 2 ' ' alnl ‘ 14 23 27 . 23 14 14 .19. 21s2* 14 21 21 27 13 13 . 18 35:1 23 29 31 29 22 22 26 2532 20 27 28 ~27 21 21 24 27:1 24 30 32 28 24 25 27 a7n2 23 34 30 31» 23 21 27 1.8005 (11 within a) 7 8 as as 5 8 4 LSD05 (R within R) NS 4 6 RS RS 3 2 5 ' rlsl 17 25 26 25 18 18 21 F1R2 18 26 25 28 18 17 22 2231 23 29 34 28 22 23 27 szsz 20 29 27 29 20 20 24 15005 (2 within a) 3 4 5 as 2 3 1 18005 (n within 3) as as 5 as us 3 1 *Total over 5-year rotation. **100 pounds I per acre on corn; 40 pounds on other row crops and wheat. . 1960 N applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. Table 5-b. -- Pounds per acre P in soil by dates of sampling. Ferden Farm rotations 1, 6 and 7. Sims clay loam. 1960. (Tasseling date: July 25.) Seasonal Treatment 6 21 7 . 6 l8 7 29 8 11 9 l4 Avera e erlll 12 24 23 22 12 12 17 h de be b d ef e erllz 13 20 19 26 11 ll 16 gh e c ab d f e erzll 16 22 31 23 16 17 21 : efgh de ab ab e de ed Ritz!2 15 22 23 28 16 16 20 fgh de bc ab c def d R‘PIRI 18 26 28 25 19 19 23 defg ede abc ab be ed bc R‘Fllz 19 27 29 27 22 20 24 def ed abc ab ab bcd b R P l 28 ’32 35 32 25 25 29 6 2 1 8 abc a . a a ab 8 R‘PZR2 22 27 27 28 21 21 24 bed ed abc ab ab bcd b a7rla1 21 26 28 28 23 23 25 cde ede abc ah ah "abc b 27111162 22 31 27 31 20 19 25 ' bed bc abc ab b ed b R7P2R1 27 34 37 28 25 27 30 ab ab a ab a a a 117112112 ’ 25 38 32 31 25 23 29 abc a ab ab a - abc a a, b, e....h - Ranges of equivalence (Duncan, 1955). Within a column, numerical values with the same literal subscript are not different at 5 percent. 35 or to more rapid mineralization of organic soil phosphorus, since soil phosphorus levels were at or near their seasonal high points. Soil potassium Soil tests for potassium.(tablea 6-a and 6-b) reflected the influ- ence of treatment factors to a lesser degree than did the tissue tests for potassium (table 3-a). Rein effects of rotation, fertility level and supplemental nitrogen treatment were generally non-significant. On July 18, just before tasseling, there was a significant rotation x fertilizer interaction due to the fact that available soil potassium was augmented by the double rate of fertilization in Rotations 6 and 7 but not in Rotation 1. A.direct relationship between soil potassium and rate of potash addition was maintained with moderate consistency only in Rotation 7, giving rise to a significant date x rotation x fertilizer in- teraction. The seasonal averages showed a significant difference for fertilizer only in Rotation 7. Low tissue phosphorus tests (table 2-a) suggest that low availability of phosphorus may have restricted growth in Rotation 7, thereby reducing crap demand for soil potassium and allow- ing it to increase. (A.similar fertilizer x nitrogen interaction operated through most of the season to give a significantly higher seasonal average soil test for potassium at higher rates of fertilization only in plots which received no supplemental nitrogen. These results are consistent with the concept that seasonal levels of available potassium reflect the intensity of crop demand on the one hand and rate of release from un- available soil sources on the other. 'The fact that treatment-associated variation in potassium soil tests at this location was generally much less than for soil I03' and phosphorus indicates that potassium.was not 36 Table 6-a. -- Pounds per acre R.in soil by dates of sampling. Ferden Farm.rotations 1, 6 and 7. Sims clay loam. 1960. (Tasseling date: July 25.) Date _ Seasonal‘ Treatment Code 6131, 7/5 7118 7129 8111 9114 Average Rotation 1 R1 174 180 175 174 167 159 171 Rotation 6 a, 176 189 197 166 161 166 176 Rotation 7 '27 182 187 188 182 172 166 179 15505 I as as as as as as as 800 166. 5-20-10* Pl 177 188 180 175 166 161 ‘ 174 1600 166. 5-20-10* :2 ‘178 183 193 173 167 166 177 13005 as as 10 as as as as so Suppl. a 1:1 178 185 184 174 165 167 175 Sidedressed a** a2 177 185 189 174 168 160 175 LSDOS as as as as as as as alrl 177 176 180 171 172 164 173 2132 171_ 184 170 178 162 154 170 2‘11 177 198 .189 170 160 160 175 2‘92 176 181 204 162 163 172 176 2751 177 190 170 184 166 159 174 2722 187 184 206 180 178 174 184 13005 (3 within a) as as as as as as 13 18005 (t within.a) as as 17 as as as 8 R131 171 182 175 171 168 156 170 alaz 177 178 176 178 166 162 173 n‘al 179 188 191 162 161 170 175 a5a2 174 191 202 170 162 162 177 Rle 184 187 187 190 166 176 181 ayaz 180 187 189 174 178 157 177 13005 (3 within a) as as as as as as as 18005 (a within a) as as as as as 12 as ' alal 176 185 173 175 162 162 172 alaz 178 190 - 187 175 169 160 176 azal 180 186 196 174 167 172 179 azaz 176 180 191 173 168 160 174 1.81105 (a within a) as as 16 as as as 6 5 LSDOS (3 within F) NS NS NS NS NS NS *Total over 5 year rotation. **100 pounds N per acre on corn; 40 pounds on other row crops and wheat. 1960 8 applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. 37 Table 6-b. -- Pounds per acre I in soil by dates of sampling. Ferden Farm rotations 1, 6 and 7. Sims clay loam. 1960. (Tasseling date: July 25.) ¥Qg£g ~ Seasonal Treatment 6 21 7 5 7 18_ 7 29 8 11 < 14 Avera e altlfli 172 178 182 168 170 156 171 a ab bed a sh cd bc thlnz 182 174 178 174 174 172 175 a b bed a ah abc bc ‘1’2'1 170 186 167 173 ‘ 166 156 169 a sh cd a b cd c erzlz 172 182 173 183 158 152 170 a ab bed a b ' cd bc R‘Flll 178 190 182 163 154 160 171 a ab bed a b bcd ‘bc ‘6'1‘2 176 206 197 177 166 160 180 a a abc a b bcd ab RGIZII 180 186 200 162 168 180 179 a ab abc a ah ah abc Rsrzflz 172 176 208 163 158 164 173 a ah ah a b abcd bc nyrlal 178 188 154 195 164 170 174 a ab d a b abc bc a7r1a2 176 192 186 174 168 148 ’ 174 a sh abcd a ab d be R7F2l1 190 186 220 186 168 182 , 188 a sh a a ab a r a 117112112 184 182 193 174 188 166 181 a sh abc '7 a a shed, ab a, b, c, d - Ranges of equivalence (Duncan, 1955). Within a column, numerical values with the same literal subscripts are not different at 5 percent. 38 as critically limiting as the other two. This conclusion is supported by the extent to which variation in tissue tests for each of these three nutrients could be associated with treatment. Soil reaction Soil pH (tables 7-a and 7-b) did not show any remarkable variation during the growing season, although high fertiliser level and supple- asntal nitrogen occasionally lowered pR. Corn yields Data.in table 8-a show that the principal yield responses were associated with rotations and supplemental nitrogen treatment. Yields were somewhat higher at the high level of fertilization than at the low, but differences were not significant.‘ This would indicate that nitrogen was the nutrient which principally controlled yield. Responses ranged from a 41-bushe1 increase for supplemental nitrogen applied on Rotation 6 at the high level of fertility to a one-bushel decrease where nitrogen was applied on Rotation 1 at the low fertilizer rate (table 8-b). It is apparent that the response to nitrogen in this experiment followed a diminishing returns function of some sort. Corn populations Germination and survival of corn were influenced by fertility in this experiment. Total stalks at harvest time were increased signifi- cantly by the high level of fertilizer addition, particularly where supplemental nitrogen was also used (table 8-a). Barrenness, on the other hand, was unaffected by rate of fertiliza- tion, but was strongly influenced by rotations and by supplemental nitro- gen treatments. Comparison with table l-a shows that barrenness was 39 Table 7-a. -- Soil pH by dates of sampling. Ferden Farm.rotations 1, 6 and 7. Sims clay loam. 1960. (Tasseling date: July 25.) . Date Seasonal Treatment Code 6/21 7(5 7118 7(29 8(11 9Z14 Average Rotation 1 RI 6.5 6.5 6.6 6.2 6.5 6.6 6.5 Rotation 6 a, 6.7 6.6 6.7 6.4 6.6 6.8 6.6 Rotation 7 Ry 6.5 6.4 6.6 6.1 6.5 6.6 6.4 LSDOS NS US ’5 NS NS NS NS 800 1b.. 5-20-10* F1 6.6 6.5 6.6 6.2 6.6 6.7 6.5 1‘00 1b.e 5-20-10* F2 ‘e5 ‘65 ‘0‘ 6.2 ‘e4 ‘0‘ 6.5 LSDos ‘ as as as as .04 as as lo Suppl. I N1 6.6 6.5 6.7 6.3 6.6 6.8 6.6 Sidedressed R** Hz 6.6 6.5 16.6 6.2 6.4 6.6 6.5 LSDos ~' as as 0.06 as as 0.09 0.03 3131 6.5 6.5 6.5 6.2 6.5 6.6 6.5 3132 6.5 6.5 6.6 6.2 6.4 6.7 6.5 n‘al 6.7 6.6 6.8 6.4 6.7 6.8 6.7 n‘rz 6.7 6.6 6.7 ,6.3 6.5 6.8 6.6 n7r1 6.5 6.4 6.7 6.1 6.5 6.6 6.5 27:2 6.5 6.4 6.6 ‘6.2 6.4 6.5 6.4 LSDOS (R within F) NS NS NS NS NS NS NS LSDOS (F within 3) NS NS NS NS 0.08 NS NS ‘ 31'1 6.5 6.5 6.6 6.2 6.6 6.8 6.5 [1H2 6.5 6.5 6.5 6.2 6.4 6.5 6.4 n‘al 6.7 6.6 6.8 6.5 6.7 6.9 6.7 asaz 6.7 6.6 6.7 6.3 6.5 6.7 6.6 R7l1 6.5 6.5 6.7 6.1 6.5 6.6 6.5 RyNz 6.5 6.4 6.6 6.1 6.4 6.5 6.4 LSDOS (R within N) NS NS NS NS NS NS NS LSD05 (I within R) NS NS 0.11 NS 0 14 0.16 0.06 A 3131 6.6 6.6 6.7 6.3 6.6 6.8 6.6 Fllz 6.6 6.5 6.6 6.2 6.5 6.6 6.5 P231 6.6 6.5 6.7 6.3 6.6 6.7 6.6 azaz 6.5 6.5 6.6 6.2 . 6.3 6.6 6.4 LSDOS (F within N) NS NS NS NS 0.09 US as LSDOS (R within P) NS NS 0.09 NS 0.11 0.13 0.05 g *Total over 5 year rotation. **100 pounds N per acre on corn; 40 pounds on other row crops and wheat. ..1960 l applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. 40 Table 7-b.- Soil pH by dates of sampling. “Ferden Farm rotations 1, 6 and 7. Sims clay loam? j1960. (Tasseling date: July 25.) —1 , . Date Seasonal Treatment 6/21 ‘715‘ > 7118 7/29 8/1_7 9/14 Averagg 111111111 , 6.5 M 6.6 6.6 6.2 6.5 6.8 6.5 g a ab _ abc ab abc a _ bc ‘1'1'2 6.5 6.4 6.4 6.2 6.4 6.5 6.4 a b c ab bc b e 111172111 6.6 6.5 6.7 6.3 ' 6.6 6.8 6.6 _ ~a -ab ab ab ab a ab ‘1’2‘2 6.5 6.5 6.6 6.2 6.3 6.5 6.4 a ab abc ab ,c b_ c Rgtifll 6.7 6.7 6.7 6.5 6.7 6.9 6.7 a a ab a a a‘ a a a 6.7 6.6 6.8 6.3 6.6 A 6.7 6.6 ‘6 1 2 a sh a ab ab ab ab a ab 6 a a a a “'2': 6.7 606 666 662 6.4 607 665 a ab a c ab be 'ab bc 37’1'1 6.5 665 ‘67 600 666 667 665 a ab ab b ab ab bc 37,1!2 6.5 6.4 6.6 6.1 6.4 6.5 6.4 a b abc b be h c s7rza1 6.5 6.4 6.6 6.2 6.4 6.5 6.4 a b ,abc ab bc b c 37'2'2 605 604 665 661 ‘ 604 665 604 a b be b be b c a, b, c - Ranges of equivalence (Duncan, 1955).‘ Within a column, nuacrical values with the same literal subscript are not different at 5 percent. 41 Table 8-a. -- Corn yields, fertile stalks, barren stalks and total stalks per acre. Ferden Parm.rotations l, 6 and 7. Sims clay loam. 1960. (Tasseling date: July 25.) . can yield W Treatment Code per acre Fertile Barren Total Rotation 1 R1 98.4 14352 858 15210 Rotation 6 R5 67.4 13869 1858 15727 Rotation 7 R7 79.8 14748 1229 15977 LSDOS 16.3 RS 383 NS 800 lbs. 5-20-10* F1 80.2 13958 1262 15220 1600 lbs. 5-20-10* P2 83.5 14688 1368 16056 1.3005 as 699 as 711 No Suppl. R R1 71.4 13808 1762 15571 Sidedressed l** R2 92.4 14837 868 -15706 was .- 5.0 535 283 as R131 96.2 13851 793 14645 3132 100.7 ‘14852 922 15775 “6’1 67.8 13722 1750 15473 2522 67.0 14015 1966 15982 R7F1, 76.6 14300 1242 15542 3722 83.0 15197 1216 16413 LSDos (R within P) 17.9 NS 564 NS 1.8005 (F within a) as as as as Rlll . 96.1 14265 897 15162 ,alaz 100.8 14438 819 15257 R6N1 50.3 12816 2785 15602 R6N2 84.5 14921 931 15852 R731 67.8 14343 1604 15947 Rynz 91.8 15154 853 16008 1.3005 (8. 61:111.: a) 17.4 1315 516 as LSDO5 (R within R) 8.6 927 491 RS Flll 71.3 13777 1730 15507 Fllz 89.2 14139 793 14932 lel 71.5 13840 1794 15634 32’2 95.6 15536 943 16479 LSD05 (F within R) NS 880 RS 857 LSD05 (R within P) 7.0 757 401 679 *Total over 5 year rotation. **100 pounds K per acre on corn; 40 pounds on other row craps and wheat. 1960 8 applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. 42 Table 8-b. -- Corn yields, fertile stalks, barren stalks and total stalks per acre. Ferden Farm rotations 1,,6 and 7. Sims clay loam. 1960. (Tasseling date: July 25.) Corn Yieldgflrertile'Stalks - Barren Stalks ‘Tota1.Stalks Treatment Per Acre Per Acre Per Acre Per Acre thlll 96.7 14127 845 14973 ab be be ed 313132 95.7 13575 741 14317 ab ed c d’ erzll 95.4 14403 948 15352 abc -abc be bed 31’2‘2 105.9 15300 897 16197 a ab bc abc sealal 54.1 13023 2725 15749 fg ed a abc asrlaz 81.6 14421 776 15197 cd abc e bed Rglzll 46.5 12609 2846 15456 g d a abcd R6P2R2 87.5 15421 1086 16508 be ab bc ab R7F1N1 63.0 14179 1621 15801 ef be b abc R7P1N2 90.2 14421 862 15283 bc abc be bcd R7P2N1 72.5 14507 1587 16094 de abc b abc R.P N 93.5 15887 845 . 16732 7 2 2 she a be a 1a, b, c....g - Ranges of equivalence (Duncan, 1955). Within a column, numerical values with the same literal subscript are not different at 5 percent. 43 inversely related to tissue nitrate prior to and at tasseling time. The number of fertile stalks (stalks with ears) at harvest time was the net result of variations in survival (total stalks) and barren- ness. As a result, fertile stalks were significantly influenced by rate of fertilization and supplemental nitrogen treatments, and, where no supplemental nitrogen was used, by rotations as well. Residue Experiment Tissue nitrate Data presented in table 9 show the accumulation of nitrate in corn tissue on Sims clay loamLas influenced by organic residues and supple- :mental nitrogen treatments. Leaf midrib samples were taken from only two of the five replications in the experiment. As a result, the level of statistical significance attained was low. Effects of residue treat- ments were not significant and no consistent relationships were main- tained from.onesampling to another. Tissue nitrate was consistently higher with supplemental nitrogen treatment than without. This effect was most pronounced just before tasseling (July 20 sampling) on the check and the sawdust plots. High accumulations of tissue nitrate were found at early stages of growth. litrate was found to increase in the second sampling and then it dropped suddenly at about tasseling time. On July 20, the con- centration of nitrate in corn tissue was still two to three times higher than for the corresponding July 18 sampling in the previous experiment (table l-a). Nitrate levels in the last two samplings were similar in the two experiments. Changes from one sampling date to another were significant at the one percent level of probability. 44 Table 9. -- PRH.IO39N in corn tissue by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam. Ferden Farm. 1960. (Tasseling date: July 25.) ’ Date Seasonal Treatment . Code 6728 717:. 7/20 8711 9112T_ (Ayggggg Check** al 1062 1333 636 94 225 670 Alfalfa-brome+’ R2 1207 2029 599 221 235 858 Sawdust++ 23 1060 1755 643 107 275 768 Straw: 34 1119 1600 545 155 204 724 LSDOS NS IS IS NS NS NS No Suppl. I RI 970 1663 513 109 231 697 Ave. N Treatment* 112‘ 1254 1695 698 180 238 813 I LSD05 as as 128 as as 117 alal 1025 1415 483 84 223 646 alaz 1100 1251 790 104 227 694 R281 1031 1760 515 146 227 736 ‘2N2 1382 2298 683 296 244 980 naal 1001 1760 479 65 300 721 aaaz 1119 1751 806 150 250 815 2.4a1 823 1719 575 141 175 686 Rglz 1415 1481 515 169 233 762 1.3005 (n within a) as as _ as as as as I (a within a) as as 255 as as 235 1.3005 *100 pounds 1' per acre on corn, 40 pounds on beans and barley in a 5 1960 1! applications , year rotation. "All second-year hay removed. +Second-year hay left and plated down for corn. ‘H'35 tons sawdust applied in 1954. *4 tons wheat straw plowed down for corn. a6 l on alfalfa-brome (2 years). on corn: 50 pounds on June 9 plus 50 pounds on.June 30. 45 Tissue phosphorus Tissue tests for phosphorus (table 10) were essentially unrelated to treatment. The actual test values for the period from.just before tasseling to the end of the season were about the same as for Rotation l in the previous experiment (tables 2-a and b). Tissue potassium Potassium accumlation in plant sap (table 11) was found to behave seasonally like tissue R03'. Potassium.accumulation was high in the beginning, showed an increase on the second sampling date and then it dropped abruptly about tasseling time. Thereafter it showed a decline, except on the last sampling date. 0n the last sampling date there was a tendency for tissue potassium to increase again. The lowest levels of potassium1were invariably found where supple- mental nitrogen was used without residue addition. Where alfalfa-brome .hay or wheat straw were returned, tissue potassium was consistently higher than on the check, particularly where supplemental nitrogen was used. This residue x nitrogen interaction was significant at the five percent level on August 11 and in the seasonal averages. Potassium levels where sawdust had been applied were intermediate between those for the check plots and those for alfalfa-brome or wheat straw. Soil nitrate Nitrate accumulation in soil (table 12) was strongly influenced by residue treatment up until tasseling time. On plots which had re- ceived 35 tons of sawdust six years earlier, nitrate levels were signifi- cantly higher than the unamended check during the first two samplings. Nitrate was lower than the check through the third sampling on plots where four tons of wheat straw was plowed down for the 1960 corn crOp. 46 Table 10. -- PPM soluble P in corn tissue by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam. Ferden Farm. 1960. (Tasseling date: July 25.) Aggge Seasonal Treatment Code 6728 7/7 7/20 8711 9716 Average Check** R1 113 56 108 70 83 86 A1fa1fa-brome+ R2 106 50 97 97 73 85 Sawdust++ R3 133 47 105 76 65 - 85 Strawi a, 108 45 75 82 76 77 LSD05 17 NS NS NS NS NS No Suppl. N N1 118 43 98 80 7O 82 Ave. N Treatment* N2 113 55 94 82 79 85 15005 as as as as as as RlNl 115 55 107 80 77 87 R1N2 112 57 110 60 90 86 R2N1 102 50 107 110 80 90 RzNz 110 50 87 85 67 80 R3N1 145 42 105 65 62 84 R3N2 122 52 105 87 67 87 R4N1 110 27 75 67 60 68 RaNz 107 62 75 97 92 87 LSDOS (R within N) NS NS NS NS NS NS LSDos (N within R) NS NS NS NS NS NS ‘*100 pounds N per acre on corn, 40 pounds on beans and barley in a 5- year rotation. No N on alfalfa-brome (2 years). 1960 N applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. ifihAll second-year hay removed. 'fSecond-year hay left and plowed down on corn. 'F+35 tons sawdust applied in 1954. 3F4 tons wheat straw plowed down for corn. Table 11. -- PPM 47 K in corn tissue by dates of sampling as related to organic residues and nitrogen treatments. Sine clay loam. Ferden Farm. 1960. (Tasseling date: July 25.) Date Seasonal Treatment Code 6(28 7/7 7L§Q—» 81;;w 9/16 Average Check** R1 270 315 238 191 322 267 Alfalfa-brome+ 112 337 451 276 253 318 327 Sawdust++ R3 330 401 281 237 322 314 sci-wt 114 353 452 267 262 351 337 151105 as 31 as 38 as 24 an Suppl. N N1 310 398 265 228 316 304 Ave. N Treatment* N2 335 411 266 243 340 319 LSD05 as as as as as as RINI 275 332 240 207 332 277 RlNz 265 297 237 175 312 257 RZN1 350 420 270 235 317 314 R2N2 345 482 282 272 320 340 R3N1 315 385 275 192 277 289 R3N2 345 417 287 282 367 340 RgNl 322 457 275 280 340 335 RgNz 385 447 260 245 362 340 LSD05 (R within N) NS ' 62 NS 54 NS 37 1.8005 (a within a) as as as 54 as 40 *100 pounds) N per acre on corn, 40 pounds on beans and barley in a 5- 1960 N applications year rotation. No N on alfalfa-brome (2 years). cu: corn: 50 pounds on June 9 plus 50 pounds on June 30. **A11 second-year hay removed. +Second-year hay left and plowed down on corn. H35 tons sawdust applied in 1954. 4'4 tons wheat straw plowed down for corn. 48 Table 12. -- Pounds per acre soil N05 by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam, Ferden Farm. 1960. (Tasseling date: July'25.) ggggge Seasonal Treatment Code 6728 777 7/20 7727 8711 9716 Average Check** 31 47 39 39 48 39 21 39 Alfalfa-brome+ R2 43 44 36 48 36 21 38 Sawdust++ 33 54 49 ' 36 43 40, 20 40 Straw: a“ 39 34 25 42 32' 16 31 18005 6 6 9 as as as 3 No Suppl. N ‘ N1 41 4O 31 38 28 12 32 Ave. a Treatment* a2 50 43 37 52 45 27 42 13005 5 as as 8 9 5 3 alal 40 38 31 40 30 13 32 slag 54 40 46 56 49 30 46 azal 39 40 37 40 33 ' 13 34 azaz 47 47 36 56 39 29 42 8.3a1 48 49 36 39 30 13 36 RgNz 60 50 36 46 50 27 45 117a1 37 33 22 35 22 8 26 R4N2 40 34 29 49 41 24 36 13005 (8 within a) 9 8 12 as as as 5 13005 (a within a) 10 as as as 19 10 6 *100 pounds N per acre on corn, 40 pounds on beans and barley in a 5- year rotation. No N on alfalfa-brome (2 years). 1960 N applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. ' *tAll second-year hay removed. +Second-year hay left and plowed down on corn. ++35 tons sawdust applied in 1954. #14 tons wheat straw plowed down for corn. 49 This reflects the imobilizing effect of recently applied carbonaceous residues and is apparent in the seasonal averages at both levels of nitrogen addition. Incubation data for this same experiment, reported by Nora (65) and Au (6), indicate that the immobilizing effect of the heavy sawdust application had been largely dissipated by the fourth year after incor- poration of the sawdust. The addition of supplemental nitrogen resulted in significant in- creases in the level of soil nitrate, except in the second and third samplings. There was a drop in level of nitrate during this period with all treatments, followed by a very temporary rise just at tassel- ing time. This would suggest that corn growth was such as to require large amounts of nitrogen during the two weeks prior to tasseling. The apparently reduced demand for nitrogen at tasseling time may have been related to climatic conditions. During the balance of the season, significantly higher levels of nitrate were maintained where supplemental nitrogen was used with all residue treatments. The observed seasonal fluctuations in soil nitrate gave rise to interaction mean squares significant at five percent for the date x residue interaction and at one percent for date x nitrogen treatment. Soil nitrate levels in this experiment were similar to those for Rotation l in the previous experiment (tables 4-a and 4-b). Soil phosphorus Available soil phosphorus (table 13) was unaffected by either residue or nitrogen treatments. There was a progressive decline in soil phosphorus through the season, giving rise to a highly significant mean square for date averages. 50 irable 13. -- Pounds per acre soil P by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam. Ferden Farm. 1960. (Tasseling date: July 25.) 72323, Seasonal ‘Treatment eggs. 6128i, 7/7 7/20 7727 8711 2716 Average Check** 21 21 16 13 17 12 10 15 'Alfalfa-brome+ 32 26 15 14 17 13 11 16 Sawdust++ a3 23 15 15 19 11 11 16 straw¢ 24 21 16 13 16 12 10 15 LSDOS as as as as as _as as No Suppl. N N1 24 16 13 17 12 ' 10 15 Ave. N Treatment* N2 22 15 14 17 12 ll 15 18005 as as as as as as as 8.1a1 21 17. 13 17 14 1o 15 111a2 21 15 13 ' 17 11 11 15 azal 28 15 15 17 12 10 1e azaz 25 14 13 16 13 12 15 naal 23 15 15 18 11 10 15 Hafiz 24 16 15 19 11 . 11 16 84a1 24 17' 12 16 12 10 15 aaaz 19 15 14 17 13 10 14 L3005 (a within a) as as as as as as as 13005 (a within a) as as as as as as as *100 pounds N per acre on corn, 40 pounds on beans and barley in a 5- year rotation. No N on alfalfa-brome (2 years). 1960 N applications on corn: 50 pounds on June 9 plus 50 pounds on June 30. **All second-year hay removed. +Second-year hay left and plowed down on corn. ++35 tons sawdust applied in 1954. 4 tons wheat straw plowed down for corn. 51 Except for the first sampling date, soil phosphorus levels were lower than for the preceding experiment (tables 5-a and S-b). Soil potassium The amount of available potassium.in the soil (table 14) was not affected as much by the various treatments as was tissue potassflmn (table 11). In both cases, however, low levels of potassium were as- sociated with the check treatment. The seasonal average for this treat- ment was found to differ from.that for alfalfa-brome or sawdust at the one percent level of probability. Supplemental nitrogen had no effect on soil potassium. The actual quantities found were similar to those in the previous experiment (tables 6-a and 6-b). Soil reaction Data on soil pH are presented in table 15. variations in pH were not great (from 0.1 to 0.3 pH units). However, differences associated with nitrogen treatment were significant at the one percent level of probability during the last four samplings. The date x nitrogen inter- action was also highly significant. This can be attributed to the fact that pH variations from date to date were larger where no supplemental nitrogen was used than where it was. The acidifying effect of the nitro- gen fertilizer tended to counteract unknown factors which tended to raise the pH during the course of the season on plots which received no supple- mental nitrogen. This behavior was simdlar to that observed in the previous experiment (tables 7-a and 7-b). Corn yield Corn yields shown in the first column of table 16 were very uni- 52 Table 14. -- Pounds per acre soil R.by dates of sampling as related Sims clay loam. Ferden Farm. 1960. (Tasseling date: July 25.) to‘organic residues and nitrogen treatments. Seasonal‘ 7/7 7/20 .7727- 8711 .2716 .Avergge .173 172 169 155 NS 166 168 NS 168 178 166 178 178 160 152 158 NS NS " .‘ ~ Date - Treatment . Code 6738 Cheek** ' 21 171 162 218 Alfalfabbrome+ 22 ~177 184 224 Sawdust++ 23 180 175 232 strnwt' R4 181 174 222 18005 as as as an Suppl. a a1 177 174 225 Ave. N Treatment“ N2 177 173 223 LSDOS as as as ‘ 21a; 169 164 222 alaz 173 -160 214 8.2a1 177 184 226 azaz 177 184 222 23a; 184 176 226 33a; 176 174 239 R4N1 180 174 226 R4N2 183 175 219 15005 (2 within a) as as as LSDOS (a within a) as as as 168 185 174 178 NS 180 in NS 174 163 190 180 176 171 180 176 NS NS 147 161' 155 157 9 155 155 NS 147 147 156 166 156 153 161 153 12 NS 173 184 181 178 180 178‘ as 174 172 183 184 183 179 179 177 8 as *100 pounds N per acre on corn, 40 pounds on beans and barley in a 5- year rotation. No N on alfalfa-brome (2 years). on corn: 50 pounds on June 9 plus 50 pounds on June 30. **Ail second-year hay removed. '*Second-year hay left and plowed down on corn. 4*E35 tons sawdust applied in 1954. tons wheat straw plowed down for corn. 1960 N applications 53 Table 15. -- Soil p11 by dates of sampling as related to organic residues and nitrogen treatments. Sims clay loam. Ferden Perm. 1960. (Tasseling date: July 25.) Seasonal Treatment Code 61:28 7/7 172 O 7727 8711 9716 Average Cheek“ R1 6.2 6.3 6.1 6.1 6.3 6.2 6.3 Alfalfa-brace" 112 6.2 6.3 6.2 6.1 6.3 6.3 6.3 Sawdust” 113 6.1 6.3 6.1 6.0 6.2 6.3 6.2 sci-mi 11,, 6.2 6.3 6.2 6.1 6.3 6.3 6.3 1.31105 .05 as as as as as .03 No Suppl. N N1 6.2 6.4 6.2 6.1 6.4 6.4 6.3 Ave. N Treatment* N2 6.2 6.3 6.1 6.0 6.2 6.1 6.1 LSDOS as .06 .05 .08 .07. .08 .03 111a; 6.3 6.3 6.2 6.1 . 6.4 6.4- 6.3 .11., 6.2 6.3 6.1 6.0 6.2 6.1 6.1 azal 6.2 6.4 6.2 6.2 6.4 6.4 6.3 112a2 6.2 6.3 6.2 6.0 6.3 6.2 6.2 8.3111 6.2 6.3 6.2 6.1 6.3 > 6.4 6.3 a3a2 6.1 6.2 6.0 6.0 6.2 6.1 6.1 24111 6.1 6.4 6.3 6.2 6.3 6.4 6.3 11,,a2 6.2 6.3 6.2 6.0 6.2 6.1 6.2 1.81705 (R within N) .12 NS NS NS NS NS .05 151305 (N within R) NS NS .11 .17. .15 .16 ‘.07 *100 pounds 1N per acre on corn, 40 pounds on beans and barley in a 5- year rotation. No N on alfalfa-brome (2 years). 1960 N applications on corn: 50 pounds on June ‘9 plus 50 pounds .on June 30. **All second-year hay removed. +Seeond-year hay left and plowed down on corn. ”35 tons sawdust applied in 1954. 31: 4 tons wheat straw plowed for corn. 54 Table 16. -- Corn yields, fertile stalks, barren stalks and total stalks per acre as related to organic residues and nitrogen treatments. Sims clay loam. ‘lerden Farm. 1960. (Tasseling date: July 25.) Yield Stalks per acre Treatment Code . A Per 1e Barren Total Check** R1 104 13737 532 14289 Alfalfa-brome+‘ R2 108 14007 655 14662 Sawdust++ R3 108 14634 669 15304 strut 114 105 13613 538 14151 1.5005 3 as as as No Suppl. N N1 105 13710 658 14369 Ave. N treatment* N2 107 14286 548 14835 LSDOS as as as as R1N1 104 13744 593 14338 R1N2 104 13731 510 14241 RQNI 106 13386 593 13979 112a: 109 14628 717 15345 R3N1 108 14710 745 15456 R3N2 108 14559 593 15152 RgNl 103 12999 703 13703 R4N2 108 14227 372 14600 LSD05 (R within N) NS NS NS NS LSDos (N‘within R) NS NS NS NS *100 pounds N per acre on corn, 40 pounds on beans and barley in a .5-year rotation. an N on alfalfa-brome (2 years). 1960 N applica- tions on corn: 50 pounds on June 9 plus 50 pounds on June 30. **A11 second-year hay removed. '+Second-year hay left and plowed down for corn. ‘*+35 tons sawdust applied in 1954. *=4 tons wheat straw plowed down for corn. 55 form on these plots. Yields varied from 103 to 109 bushels per acre. The only significant differences were related to residue treatment. The uniformly low phosphorus soil tests (table 13) suggest that phosphate availability was the factor which primarily controlled yields on these plots. There may have been some response to supplemental nitrogen in the case of the straw residue treatment. The five-bushel yield increase for nitrogen with straw was not significant, but it was associated with relatively low soil nitrate levels (table 12). The significantly higher yields for the alfalfa-brome and sawdust treatments were very likely in response to phosphate released from organic sources. The rather high tissue test for phosphorus on the sawdust plots early in the season is consistent with this view (table 10). Corn populations Stand counts, shown in the last three columns of table 16 were not influenced significantly by treatment. Total stands were somewhat higher on alfalfa-brome and sawdust plots than on the other residue plots. There was also a tendency for supplemental nitrogen to reduce the propor- tion of barren stalks, particularly in combination with straw treatment. Nitroggn Sources Experiment This experiment involved a comparison of calcium nitrate and am- monium sulfate at two rates in fall, spring and sun-liar applications for corn. The same treatments have been applied on the same plots over a four-year period. Treatments were replicated four times and all plots were sampled for soil and tissue tests. 56 Tissue nitrate Data presented in tables 17-a and 17-b show 1103' in the tissue of corn grown on Hillsdale sandy loam. A statistical analysis combining all of the data showed significance at the one percent level for dif- ferences between means for sampling dates and for all main factor effects on seasonal averages. The interaction mean square for date x materials was significant at five percent. Nean squares significant at one per- cent were obtained for interactions between sampling date, on the one hand, and effects of rate and time of application and the average effect of nitrogen, on the other. The major changes associated with sampling date were a sharp drop in nitrate at tasseling time, followed by a further, less drastic decline during the last sampling interval. The highly significant interactions between treatment factors and date of sampling were due to the fact that rates of seasonal decline were different for different treatments. The most rapid seasonal declines were associated with high early levels of nitrate. Nain effects of treatment are strikingly apparent in the first two samplings and in the seasonal averages. The use of supplemental nitro- gen enhanced nitrate ac’cumlation in (the first two tissue samplings, but this effect was significantly reversed in the last sampling. The en- hancement in the first two samplings was greater at the 80-pound rate of nitrogen fertilization than at 40 pounds, and again there was a tend- ency towards reversal of the effect in the last sampling. The fertilizer source of nitrogen was found to affect nitrate ac- cumlation in the tissue. Amonium sulfate maintained higher levels of tissue nitrate in the first two samplings at all rates and times of application than did calcium nitrate. Differences in the last sampling 57 Table l7-a. -- PPN N03-N in corn tissue by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1) Da e ' Seasonal M *ng: 7/14 8 4 12 Aver e No Suppl. N N1 361 107 92 177 Ave. N Treatment N2 725 259 57 309 LSDOS 188 103 30 84 Ann. Sulf. N1 810 288 56 344 Calcium Nitr. N2. 640 230 57 273 LSDOS ' 105 57 NS 46 40 lbs. N per A. R1 655 183 60 261 80 lbs. N per A. 12 795 335 54 ’356 151105 105 57 as 46 Late Pall* T1 608 214 62 247 Spring“ . T2 876 300 47 '388 Sumner“ T3 691 . 264 61 291 LSDOS 128 NS NS 57 “131 754 ‘ 195 62 “ 290 11le 866 381 50 398 11le 556 171 57 231 N2R2 725 289 57 315 LSDOS 148 81 NS 66 “171 721 261 49 278 11sz 996 312 60 433 N1T3 713 291 60 320 11le 495 167 75 216 11sz 756 287 34 342 . N2T3. 670 237 62 261 LSD05 181 99 29 80 11121 571 . 159 '63 , 223 R1T2 821 177 59 323 R1T3 572 . 213 57 236 R2T1 645 268 61 271 R2T2 931 422 35 452 . R2T3 811 315 66 346 LSDOS . 181 99 NS 80 *Broadcast on surface (soil temperature near‘freezing). 1It‘lrl'lowed down just before planting. ***Sidedressed in hands when corn’was knee high. (1960 application made on June 15.) ' 58 Table l7-b. -- PPN.NO3-N in corn tissue by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treamment. Nillsdale sandy looms East Lansing. 1960. (Tasseling date: August 1.) Treatment V Dage . Seasonal Code 7114_ .814 9112 Ayegage “1‘111 721 188 55 258 bed de abc bcd 111111-12 908 185 71 342 ab de ab b N1R1T3. 632 214 60 271 beds ede abc bed T 721 333 42 ‘299 “1‘2 1 bed abcd be bed 1112.222 1084 ' 440 49 524 a a she a N1R2T3 793 ~ 369 50 370 abcd abc abc b NQRITI 422 131 71 189 e e ab cd R T 734 170 47 305 N2 1 2 bed ' de abc bcd 11211123 512 213 53 201 de cde abc cd N2R2T1 568 203 79 243 cde de ab bcd “2321.2 778 403 20 380 bed ab c b 1121213 829 261 - 71 322 she bcde ab be No N 361 107 92 177 e e a d a, b, c, d, e -- Ranges of equivalence (Duncan, 1955). Within columns, ' numerical values with the same literal subscripts are not different.at 5 percent. 59 were not significant. Spring application (at planting time) induced more NO3' accumula- tion in.the tissue prior to tasseling than did fall or summer applica- tions. This effect of spring application was most pronounced where the ammonium.form of fertilizer nitrogen was used and at the higher rate of application (see also table l7-b). Differences between fall and summer applications were not statis- tically significant. However, there was a.nerked tendency with calcium nitrate for summer sidedressing to promote higher tissue tests for nitrate prior to tasseling than fall application (table l7-b). . Except for the check treatment (no supplemental N), nitrate levels on July 14 were very similar to those for the comparable date in the residue experiment and two to three times higher than for Rotation 1 in the first experiment. Pinal levels in September were much lower than for either experiment conducted on heavier soil (cf. tables l-a and 9). Tissue phosphorus There was a continuing increase in soluble tissue phosphorus over the sampled portion of the growing season (tables 18-a and l8-b). The ‘mean square for sampling date was significant at the one percent level of probability. The seasonal increase in tissue phosphorus was much less marked where supplemental nitrogen was used than where it was not. Ammonium. sulfate was particularly effective in stabilizing phosphorus levels. Phosphorus tests were significantly lower with ammoniumisulfate than with calcium nitrate . High phosphorus levels late in the season were associated with treatments which had promoted relatively low nitrate tests in the first 60 Table 18-a.--, PP! soluble P in corn tissue by dates of sampling as related to nitrogen materials, rates and times of ap- plication in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tassel- ing date: August 1.) . ___2£3____§ Seasonal Treatment Code 1113 814 9112 Average No Suppl. II II 64 109 157 111 Ave. 1! Treatment N2 66 91 111 89 LSDOS IS NS NS 21 Anon. Sulf. 111 61 85 92 79 Calcium litr. H2 72 97 130 99 1.31105 10 as 30 1o 40 lbs. I] per A. R1 68 90 1.17 90 80 lbs. 11 per A. 32 65 93 104 88 1.3905 NS NS '3 , NS Late fall* '11 65 87 128 94 SpringMr A T2 68 92 90 81 Sumerm T3 67 94 114 91 1.3905 as as as as 11131 63 85 83 76 HIRZ 58 85 101 83 14221 73 95 151 - 105 1129.2 72 100 108 93 1.81305 14 IS 42 16 “11.1 63 72 98 79 11sz 57 88 74 7O H1T3 63 96 103 88 ille 67 103 158 109 1(sz 78 97 107 93 14213 72 93 125 95 1.81705 17 HS 51 '_ 20 R111 70 84 158 104 RITZ 67 88 73 74 RlT3 ' 67 98 120 93 Rle 59 90 98 84 3212 58 97 108 89 821‘3 68 91 108 90 1.81305 NS 118 NS NS *Broadcast on surface (soil temperature near freezing). **Plowed down just before planting. ”Sidedressed in bands when corn was knee high. (1960 application made on June 15'.) ‘ 61 Table l8-b. -- Pal Soluble P in corn tissue by dates of sampling as re- lated to nitrogen materials, rates and times of applica- tion in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1.) Treatment Date Seasonal Code 7 L14 8/ 4 9112 Average “1311.1 66 69 95 79 a a be bcd ”1311.2 58 88 68 67 a . a c d “1311.3 65 100 85 81 a a bc bcd 1113211 59 75 100 80 a a be bcd “1321.2 56 88 80 73 a a be cd H182T3 60 93 122 95 a a be abcd 1122111 * 74 100 220 129 a a a a 1123sz 76 88 78 ._ 81 ' a a be - bcd 323.1133 69 96 155 104 a a ab abc 1123le 60 105 95 88 a a bc bcd 1123sz 80 106 135 104 a a bc abc 1128213 75 89 95 86 p a a be bcd No R 64 109 157 111 a a ab ab a, b, c, d -- Ranges of equivalence (Duncan, 1955). Within columns, numerical values with the same literal subscripts are not different at 5 percent. 62 sampling (cf. tables l7-a and l7-b). A similar inverse relationship be- tween early season nitrate and late season phosphate was observed in the rotation experiment (of. tables l-a and 2-a). This suggests that the capacity for assimilating phosphate is influenced in the corn plant by nitrogen nutrition during early stages of growth. This hypothesized relationship is most evident in the data which relate to the two nitrogen materials. Ammonium.sulfate promoted much higher levels of tissue nitrate in the July 14 sampling than did calcium nitrate. This effect was most pronounced at the low rate of applica- tion and in the fall and spring applications. These are the treatments where the inverse relationship to late season phosphate was also most pronounced. The low rate of nitrogen fertilization was inadequate for maximum yields (table 24-a). The fall and spring applications would have influenced early stages of growth which would not have been influ- enced by the summer application. Therefore, it appears that high levels of unassimilated phosphate in the sap late in the season reflected re- stricted physiological development imposed by inadequate nitrogen nutri- tion at earlier stages of growth. Tissue potassium Changes in the level of tissue potassium.from.one sampling date to another (tables l9-a and l9-b) were highly significant. There was a sharp drop at tasseling time followed by a moderate rise at the end of the season. There was no significant relationship with treatment, except in the September sampling. Here, tissue potassium was reduced by applica- tion of fertilizer nitrogen. The reduction was greater at the 80-pound 1: {I’m IH 63 Table l9-a. -- PPM K in corn tissue by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1) Date Seasonal Treatment Code 7]}: 8/4 fllz Average No Suppl. N N1 541 421 465 480 Ave. ll Treatment R2 549 394 418 457 1.3905 as us 42 us Amman. Sulf. H1 563 393 415 464 Calcium litr. H2 535 395 420 450 1.51105 us as as ‘ us 40 lbs. 1! per A. El 547 390 431 460 80 lbs. N per A. R2 551 398 404 454 LSDOS as as - 23 as Late Pall* r, 535 394 424 450 Spring” . T2 561 385 ‘ 406 455 Sumer*** T3 551 403 422 466 LSDos as as . us as 311:1 561 383 422 460 14132 , 565 402 408 467 ”231 533 398 . 440 ‘ 460 11232 538 393 ‘ 401 440 LSDOS NS NS .33 KS 11111 549 388 429 454 11sz 574 394 393 459 u1r3 566 396 428 478 11le 521 400 426 446 11sz 549 376 419 451 321'3 536 410 416 453 LSD05 as as as us RlTl 536 386 448 453 R1T2 552 381 414 455 RlT3 552 404 430 473 Rle 533 402 400 447 RZTZ 571 389 398 455 9.213 550 402 414 459 1.51305 as us 40 as *Broadcast on surface (soil temperature near freezing). **Plowed down just before planting. mSidedressed in bands when corn was knee high. (1960 application made on June 15.) 64 Table l9-b. -- PPH - K in corn tissue‘by dates of sampling as related-to nitrogen materials, rates and tines of application in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1.) Treatment Date Seasonal Code 7/14. 814» 9112 Average HlRlTl 560 366 445 . 455 a a abc . a 1112.sz 547 401 385 445 a a c a H1R1T3 575 383 435 480 a a abc a HIRZTI 538 410 400 453 a a abc a HIRZTZ 600 388 402 473 a a abc a H1R2T3 557 409 422 476 a a abc a fl HleTl :13 :06 4‘52 :52 32311.2 557 361 443 464 a a abc a H2R1T3 529 425 425 465 a a abc a 1123le 529 394 400 440 a a abc a ”2:212 541 391 3:5 438 a a c a H2R2T3 544 395 407 442 a a abc a No R 541 421 465 480 a a a a a, b, c -- Ranges of equivalence (Duncan, 1955). within columns, numer- ical values with the same literal subscripts are not different at 5 percent. ' 65 rate than at 40 pounds. This rate effect was greater with calcium.nitrate than with ammonium.sulfate, and greater with fall application than with applications made in spring or summer. In other words, potassium behaved with respect to nitrogen fertilization in a manner similar to phosphorus in this last sampling. Soil nitrate litrate accumulations in soil were increased significantly through- out the season by additions of nitrogen fertilizer (tables 20-a and 20-b). The higher rate of nitrogen application maintained higher l03' levels in the soil. These main effects of nitrogen fertilizer were the same as their effects on tissue nitrate prior to and at tasseling time (of. tables l7-a and l7-b). In September, tissue nitrate was inversely re- lated to soil nitrate for these same treatment combinations. Except for the July 14 sampling, the soil nitrate mean squares for time of application were highly significant. The highest levels of soil nitrate throughout the season were associated with the summer sidedressing. The spring application maintained higher soil nitrate tests than the fall application through the August 12 sampling. In contrast to this, tissue nitrate prior to tasseling was such higher where nitrogen was applied at planting time than where it was applied either in the fall or as a sidedressing in mid June. There was no significant main effect of nitrogen source on soil nitrate levels. However, there was a significant to highly significant interaction between materials and time of application on most sampling dates and in the seasonal averages. This was due to the fact, with fall application, higher levels of nitrate were maintained in the soil by ammonium.sulfate than by calcium nitrate, whereas the reverse was 66 Table 20-a. -- Pounds per acre 803-! in soil by dates of sampling as re- lated to nitrogen materials, rates and times of applica- tion in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: .August 1.) Date . Seasonal Treatment Code 6 0 -7 l4 8 8 12 9 l2 Avera e No Suppl. I ll 20 43 14 7 ‘ 7 “ 18 Ave. N Treatment H2 41 57 36 l7 13 33 ‘ LSD05 12 HS 13 8 HS 7 A-aon. Sulf. M1 38 58 36 l8 14 33 Calcium litr. H2 43 57 36 17 12 33 LSDos HS HS HS HS HS HS 40 lbs. I per A. R1 36 48 31 9 ' 6 26 80 lbs. I per A. R2 46 66 41 25 19 39 15905 6 14 7 4 6 4 Late fall* T1 32 47 27 10 8 25 Spring** T2 39 60 37 18 9 33 Summer*** T3 51 64 44 24 21 41 1.3005 8 NS 9 5 7 4 ”1‘1 34 45 34 10 6 26 11le 43 70 q 37 26 21 39 11le 38 51 28 9 7 26 1122.2 49 62 44 25 18 39 LSD05 9 20 10 6 8 5 HITI 41 54 32 13 9 30 11sz 35 68 38 20 13 35 H1T3 39 51 m 37 21 19 33 lile 23 41 21 6 7 20 HQTZ 43 52 36 16 6 30 H2T3 64 77 51 27 24 49 LSDOS ll 25 13 7 10 6 RlTl 29 47 22 7 5 22 R1T2 32 49 32 8 6 26 R1T3 46 49 39 13 8 31 R2T1 35 48 31 13 12 28 R2T2 46 70 42 28 12 40 R2T3 57 80 49 36 34 51 LSDOS ll 25 13 7 10 6 *Hroadcast on surface (soil temperature near freezing). **Plowed down just before planting. ***Sidedressed in hands when corn was knee high. (1960 application made on June 15.) 67 Table 20-b. -- Pounds per acre R03-H in soil by dates of sampling as related to nitrogen materials, rates and times of applica- tion in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1 .) Treatment Date Seasonal Code 6730 7114 8/4 8/12 9 12 Avera e “13111 34 46 29 9 4 24 cd bc bcd d c e 313112 34 51 38 9 8 28 cd bc b d c cde H1R1T3 33 40 35 13 6 26 cd c be d c de ”11211 - 48 63 36 18 15 36 be abc b cd bc bcd 313212 35 85 39 31 18 42 cd ab ab ab bc b H1R2T3 44 62 38 30 31 41 bc abc b abc ab b 11231131 24 48 15 6 - 6 20 d bc cd d c e 11211112 31 4s 27 7 5‘ 23 cd bc bcd d c e HhR1T3 59 57 42 13 9 36 ab abc ab d c bcd HQRle 22 34 27 7 9 20 d c bcd d c e 142R2T2 56 56 45 25 7 38 ab bc ab be c be 1423213 69 98 6O 42 38 61 a a a a a a No N’ 20 43 14 7 7 18 d c d d c e a, b, c, d, e -- Ranges of equivalence (Duncan, 1955). Within columns, numerical values with the same literal subscripts are not different at 5 percent. 68 true with summer sidedressing. There was no difference between the two materials when they were applied at the time of spring planting. In the case of tissue nitrate (table l7-a), on the other hand, higher levels were maintained through tasseling time by ammonium sulfate than by calcium nitrate, particularly with fall and spring applications. Since soil nitrate levels were unaffected, this suggests that the young corn crop was taking up nitrogen in both the ammonium.and nitrate forms, but was preferentially assimilating the ammonium.form. As a result, nitrate accumulated in the sap where the ammonium.form.was made avail- able in fertilizer. This is consistent with accepted concepts of nitro- gen nutrition (63, 69).1 Mean squares associated with sampling date were significant at the one percent level of probability. The seasonal decline in soil nitrate after tasseling in this sandy loam soil was, in general, more rapid than in the two earlier experiments conducted on Sims clay loam” Levels prior to tasseling tended to be higher, those at the end of the season lower. However, the 80-pound summer sidedressing with continuous corn on the sandy loam was equally as effective in maintaining soil ni- trate through the period after tasseling as was 100 pounds sidedressed on corn in rotations including two years of alfalfa-brome on the heavier soil (cf. tables 4-a and 4-b). Soil phosphorus Addition of nitrogen tended to depress available soil phosphorus (tables 21-a and 21-b). This effect was rather consistent throughout 1However, see also relation to soil pH, pp. 77. Table Zl-a. -- 69 Pounds per acre P in soil by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1.) Date Seasonal Treatment Code 6130 2114 £14 8112 2113 Average so Suppl. R R1 127 126 123 121 123 124 Ave. N Treatment 82 115 116 117 128 112 118 LSD05 12 IS IS NS NS 6 Mn. 3.312. 111 115 115 117 129 112 118 Calcium Hitr. 112 115 118 117 128 112 118 LSD05 IS IS IS as IS IS 40 lbs. I per A. R1 114 117 118 129 115 119 80 lbs. I per A. R2 116 116 117 128 109 117 LSDos IS IS . NS NS IS IS Late my: '11 116 116 121 135 116 121 Spring“ T; 118 121 121 132 111 121 Sumer*** ’1'3 111 112- 110 118 108 112 LSD05 NS NS 7 13 HS 4 an 114 117 119 132, 119 120 11132 116 113 115 126 105 115 HQRI 114. 117 117 127 111 117 HQRZ 115;. 118 118' 129 112 119 LSDOS NS NS NS _RS HS 5 11111 115 113 119 134 113 119 11sz 123 124 123 136 112 124 111133 108 108 110 117 111 111 “21.1 118 120 122 137 119 - 123 1! T 113 117 119 127 111 118 H2T3 113 116 110 120 105 113 LSD05 12 NS 10 18 IS 6 RlTl 113' 121 120 142 121 123 RITZ 116 116 119 129 112 118 R1T3 114 115 115 118 112 115 Rle 119 112 121 129 112 119 R2T2 121 126 124 135 111 123 R2T3 107 110 105 119 104 . 109 1.31705 12 IS 10 18 NS 6 *Hroadcast on surface (soil temperature near freezing). **Plowed down just before planting. mSidedressed in bands when corn was knee high. (1960 application made on June 15.) 70 Table 21-b.-- Pounds per acre P in soil by dates of sampling as related to nitrogen materials, rates and times of application in the fourth yaarof annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1.) Treatment A Date ' Seasonal Code 6130 7/14 814 8/12 9 12 A er s 111111-11 108 11s 11s ‘ 140 122 121 ab a a ab a ab 'ulRsz 127 122 121 135 114 123 ' a a a ’ ab a a HIR1T3 109 112 _ 119 122 121 116 ab a a ab a ab 313211 122 107 120 128 104 116 ab a a ab a ab 11111222 120 127 125 138 111 124 ' ab a a ab a a 31R2T3 108 105 101‘ 112 101 105 ab a b b a c H2R1T1 118 124 122 143 119 125 ' ab a a a a a HQRsz 105 111 117 123 111 113 b a ab ab a be HQR1T3 119 117 111 114 103 113 ab a ab ab a be H2R2T1 117 117 123 130 119 121 ab a a ab a ab HszTz 122 124 122 132 111 122 ab a a ab a ab 11211213 107 115 109 126 107 113 b a ab ab a ho No R 127 126 123 121 123 124 a a a ab a a a, b, c -- Ranges of equivalence (Duncan, 1955). Within columns, mwmeri- cal values with the same literal subscripts are not different at 5 percent. 71 the season. It was most evident in connection with summer sidedressing at the 80-pound rate of ammonium.sulfate. As a result, highly signifi- cant mean squares were obtained in the seasonal averages for time of application; mean squares significant at five percent were obtained for the interactions between materials and rates, materials and time of application, and rates and time of application. Comparison with tables 20-a and 20-b shows that soil phosphorus tests tended to be inversely related to soil tests for nitrate. Mean yields for various treatments (table 24-a) were also inversely related to mean soil phosphorus tests. Thus it appears that the lower soil phosphorus levels reflected higher rates of removal associated with more vigorous growth of corn. Absolute levels of soil phosphorus in this experiment were of the order of five to ten.times higher than in. either of the two previous experiments (tables 5-a and 13). Soil potassium Soil tests for potassium.(tables 22-a and 22-b) reflected the in- fluence of treatment factors to a greater extent than tissue tests for potassium.(table l9-a). However, there was no correspondence between soil and tissue tests. A combined analysis of variance of all the soil potassium data showed that seasonal trends were highly significant. These included a gradual rise to main values at tasseling time, followed by a more rapid'decline to seasonally low values in the last sampling. The seasonal peak and levels after tasseling were lower with calcium nitrate than with amonium sulfate. Highly significant mate- rials x rates and materials x time interactions were obtained. High 72 Table 22-a. -- Pounds per acre K in soil by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1.) . , Date - Seasonal Treatment Code 6130 7114 '8/4 8112 9112 Average lo Suppl. 1! -. 81 196 204 224 159 198 196 Ave. N Treatment Hz 202 208 229 ‘ 183‘ 154 195 1.s1>05 . as as as as 27 - .NS .. Anon. Sulf. 111 202" 212 241 - 198 159 - 202 Calcium litr. H2 203 203 217 ' 168’ 148 188 1.311053 as as 21 18 as 8 40 lbs. 11 per A. 111 196 217 230 179 158' 196 so lbs. 11 per A. 112 208 198 229 187 _ 149 194 Lsnos as as as as as '8 Late 1.11s 11 200 209 226 179 1:61 195 Spring" 12 204 203 224 185 153 194 813-arm '13 203 212 238 184 142 196 1.51105 us as us as as as 111111 210 226 250 199 169 211 111112 193 197 232 197 148 194 112111 183 207 209 159 146 181 112112 223 199 225 _ 177 151 19s 1.31105 27 as 29 29 ,, as 11 11111 194 203 225 199 166 197 11112 203 197 229 191 151 . 194 1111-3 208 235 271 204 - 159 215 11211 206 214 227 159 156 ~ 193 11sz 204 208 ‘ 219 ' 180 164 195 11213 198 188 205 164 125 176 1.31105 us 33 as 31 26 14 21:1 186 223 231 177 166 197 3112 189 202 210 182 155 188 1111-3 214 225 243 177 152 203 1121-1 214 194 221 182 156 193 11212 218 203 238 188 160 201 112-13 192 198 227 191 132 188 LSDOS as us as as us 14 k *Broadc ast on surface (soil temperature near freezing). “Flawed down just before planting. , ***Sidedressed in bands when .corn was knee high. (1960 application made on June 15.) 73 Table 22-b.-- Pounds per acre K.in soils by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam" East Lansing. 1960. (Tasseling date: August 1.) Treatment Date ‘ Seasonal Code 61§0 7114 8 4 8 12 9 12 Aver e H1R1T} 190 222 240 200 176 206' ' abc ab ab . ab abc b ”1311.2 210 192 216 187 148 191 abc b . b abc bode bc 1118113 230 264 295 209 184 236 ab a a a. ab a 1119.211 198 184 210 198 156 189 abc b b ab abcde bc MleTz 196 _202 241 194 154 197 abc b ab abc abcde b 111R2T3 186 206 246 200 134 194 abc b ab ab cde b 1429.111 182 224 223 154 .156 188 be ab b be abcde be ("23112 168 212 203 178 162 185 c ab b abc abcde bc H2R1T3 198 186 201 145 120 170 abc b b c e c HQRle 230 204 232 '165 156 197 ab ‘b b abc abcde b HpRgTz 240 204 234 183 166 205 a b b abc abcd b H2R2T3 198 190 209 182 130 182 abc b b abc de bc No R 196 204 224 159 198 196 abc b b abc a b fa, b, c, d, e -- Ranges of equivalence (Duncan, 1955). Within columns numerical values with the same literal subscripts are not different at 5 percent. 74 soil tests for potassium.were associated with the low rate of ammonium sulfate addition and with the high rate of calcium nitrate. The highest levels of potassium were generally associated with the sumner sidedress- ing of ammonium sulfate, whereas calcium.nitrate applied at this same time usually favored the lowest potassium tests. A rate x.time inter- action, significant at five percent, was also involved, such that low soil potassiumumost frequently followed spring application at the 40- pound rate, whereas, at the 80-pound rate, the lowest tests were found following the summer sidedressing. This complicated response pattern in the soil potassium.tests could not be related in any straightforward manner to soil or tissue tests for nitrate or phosphorus, nor to vigor of growth as reflected in final yields. Low tests were associated to some extent with high yields. However, the strong influence of nitrogen source on soil potassium.1evels suggests that cationic interactions involving ammonium.and potassium or calcium.and potassium.were also involved. Soil reaction The range of soil reactions encountered during the season in this sandy loam soil (tables 23-a and 23-b) was greater than in the two ex- periments on Sims clay loam (pH 4.8 to 5.9 as against 6.0 to 6.9 in the rotation experiment and 6.0 to 6.4 in the residue experiment). Soil re- action was influenced to a greater degree by seasonal and treatment factors, reflecting the lower buffer capacity of the lighter soil. Where no supplemental nitrogen was used or where calcium nitrate 'was used, pH rose sharply after tasseling. These late season increases *were associated with depletion of soil nitrate and were highly signifi- cantu Where ammonium.sulfate was used, pH was significantly lower all 75 Table 23-a.-- Soil pH by dates of sampling as related to nitrogen materials, "rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1) ' PEEL Seasonal W Code 6 30 7 14 8/4 8/12 9 12 Avera e lo Suppl. l 111 5.5 5.6 5.5 5.7 5.8 5.6 Ave. H Treatment 112 5.3 5.2 5.2 5.4 15.5 5.3 LSDos llS . 20 . 21 . 16 . 22 . 10 Anon. Sulf. 111 5.1 5.1 4.9 5.1 5.3 5.1 Calcium litr. 142 5.4 5.4 5.4 5.6 5.7 5.5 LSDOS .17 ll .12 .09 .12 .05 40 lbs. u pct A. RI 503 593 502 5.5 506 504 80 1b.. ' per A. ‘2 Sea 502 501 503 5e5 503 LSD05 HS .11 IS .09 IS ' ,.05 Late Fall* T1 5.2 5.2 5.2 5.4 5.5 5.3 Spring“ _ T2 5.3 5.3 5.1 5.4 5.5 5.3 Sumner“ T3 5.4 5.3 5.2 5.4 5.5 5.3 LSD05 IS IS IS IS '8 HS HIRI 5.2 5.2 5.0 5.3. 5.4 5.2 111R2 5.1 5.0 4.9 5.0 5.2 5.0 H231 504 504 5.4 506 5.7 505 ”232 504 503 504 506 508 5e5 LSDOS . 24 . 16 17 12 . 17 08 311:1 5.0 5.0 4.9 5.2 5.4 5.1 11sz 5.1 5.1 4.9 5.1 5.3 5.1 111T3 5.3 5.2 5.0 5.1 5.2 5.2 11le 5.4 5.4 5.4 5.6 5.7 5.5 11sz 5.5 5.5 5.4 5.6 5.7. 5.5 u2T3 504 503 503 506 508 505 LSDos 29 19 .20 .15 21 .09 R1T1 5.1 5.2 5.2 5.5 5.6 5.3 3112 503 504 5.1 504 505 503 R1T3 5.5 5.3 503 505 506 504 R2T1 5.3 5.2 5.1 5.3 5.5 5.3 R2T2 5.3 5.2 5.2 5.3 5.5 5.3 8213 5.3 5.2 5.1 5.2 5.4 5.2 1.51105 as . 19 as . 15 115 09 *Broadcast on surface (soil temperature near freezing). “Plowed down just before planting. mSidedressed in bands when corn was knee high. (1960 application made on June 15.) n 76 Table 23-b. -- Soil pH by dates of sampling as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam. East Lansing. 1960. (Tasseling date: August 1.) Treatment Date » - Seasonal Code 61;!) 7&4 814 8112 . 9 12 Avera_e 11111111 4.8 5.0’ 5.1 5.3- ' 5.4 5.1 b cd cdef bc cd cd 11111112 5.2- 5.4 4.9 5.3 5.5' 5.3 u ab ab ef bc bcd bc 11111-13 5.5 553 5.1 5.3 5.4 5.3 - a abc ~ cdef bc cd bc 11111211 5.2 521 4.8 _ 5.1 5.4 5.1 ab bcd f cd_ ‘cd cd I 11]}sz 5.0 4.9 4.8 4.9 5.2 5.0 b d f d' de d a 'r 5.2 5.1 5.0 5.0 5.0 5.1 132 3 ab bcdj def cd e cd 11211111 5.4 5.4 5.4 5.7 5.8 5.5 a ab abc a sh ab “2‘112 504 504 503 506 506 504 ' a ab abcd ab abc ab 11211113 5.5 5.4 5.4 5.7 5.8 5.6 s eh abc a ab a ”M11 504 503 504 5e6 5e7 504 a abc abc ab abc ab liszTz 5.6 5.6 5.6 5.7 5.9 5.6 a a a a a a T 5.3 5.2 5.2 5.5 5.7 5.4 “272 3 ab bcd bcde ab abc ab lo I 5.5 5.6 5.5 5.7 5.8 5.6 a a ab ~a ab .a a, b‘-, c, d, e, f -- Ranges ofequivalence (Duncan, 1955). within cola-1s, numerical values with the same literal subscripts are not different at 5 percent. 77 through the season. The acidifying effect of anonium sulfate was greater at the 80-pound rate than at 40 pounds, and the seasonal increase in the last two samplings was not significant at the higher rate. Corn yield Data in the first column of table 24-a show that supplemental nitro- gen application increased corn yields, on an average, 45 percent over the _check. There was an average increase of nine bushels per acre for 80 pounds of nitrogen over 40 pounds, regardless of material. However, the increase for rate was associated primarily with spring and summer applica- tion; the rate response with fall application was not significant at five percent probability. Summer sidedressing gave a highly significant six to seven bushel increase over either fall or spring application. However, there was a significant materials x time interaction, such that yields were equally good with spring and summer applications of calcium.nitrate, whereas, with ammonium.sulfate, the spring application was significantly less efficient than the summer sidedressing. The relatively low yield for the spring application of ammonium sulfate was associated with pH's of 4.9 to 5.1 during the period up through tasseling. At this low pH it is quite possible that the addi- tional acidifying effect of the ammonium.sulfate may have enhanced man- ganese availability to the point of toxicity during these earlier stages of growth (55)2. Extremely high tissue nitrate tests and significantly higher equilibrium.levels of soil nitrate and phosphate during this period indicate that assimilation and uptake of these two nutrients were inter- 2See also relation to nitrogen nutrition, pp. 68. 78 Table 24-a. -- Corn yields, fertile stalks, barren stalks and total stalks as related to nitrogen materials, rates and times of application in the fourth year of annual treat- ment. Hillsdale sandy loam. Rest Lansing. 1960. (Tasseling date: August 1.) 4 Yield ' Stalks ggr acre Treatment Code a . 1A Pegtile Barren - Total No Supp1. 8 H1 47 11563 4313 15875 Ave. 11 Treatment 112 68 14091 2589 16680 1.31105 8 1384 1390 as Anon. Sulf. 111 68 14156 2484 16641 Calcium nu. 112 68 14026 2693 16719 13005 as as as as 40 lbs. 11 per A. R1 64 13536 2964 16500 80 lbs. 11 per A. 112 73 14646 2214 16859 1.8005 - 4 768 as as Late 1611* 1'1 65 13703 3070 16773 Spring“ _ '12 66 13711 2570 16281- Suner 13 73 14859 2125 16984 LSD05 5 942 as 118 1119.1 63 13448 3031 16479 111112 74 14865 1938 16802 112111 65 13625 2896 16521 112112 72 14427 2490 16917 1.81105 6 1087 as as 11111 68 13828 3094 16922 11112 62 13703 2391 16094 11113 74 14938 1969 16906 11211 63 13578 3047 16625 11212 ‘ 70 13719 2750 16469 11213 72 14781 2281 17062 1.8005 7 1331 as as 11111 62 13125 3344 16469 11112 61 13250 3234 16484 11113 68 14234 2313 N 16547 11211 68 14281 2797 17078 2212 71 14172 1906 16078 11213 79 15484 1938 17422 1.81105 7 1331 as us * *Broadcaat on surface (soil temperature near freezing). *ikPlowed down just before planting. ”*Sidedressed in bands when corn was knee high. (1960 application made on June 15.) 79 -- Corn yields, fertile stalks, barren stalks and total Table 24-b. stalks as related to nitrogen materials, rates and times of application in the fourth year of annual treatment. Hillsdale sandy loam" East Lansing. 1960. (Tasseling date: August 1.) ‘ Treatment Yield §ta1ks Per Ange Code Hu,1A Fertile ’ Barren Total H1R1T1 64 13063 3813 16875 bcd b a a 11111112 58 13094 3031 16125 d b a a H1R1T3 67 14188 2250 16438 bcd ab a a “13211 72 14594 2375 16969 abc ab a a H1R2T2 67 14313 1750 16063 bcd ab a a H1R2T3 82 15688 1688 17375 a a a a HQRITI 61 13188 2875 16063 cd b a a H281T2 65 13406 3438 16844 bcd b a a 11211113 69 14281 2375 16656' bcd ab a a HQRZTI 65 13969 3219 17188 bcd ab a a . ”23212 75 14031 2073 16094 ab ab a a H2R2T3 76 15281 2188 17469 ab ab a a no H 47 13897 2721 16618 e ab a a a, b, c, d -- Ranges of equivalence (Duncan, 1955). within columns, numerical values with the same literal subscripts are not different at 5 percent. 80 fered with by this treatment (of. tables l7-a, 20-a and 21-a). Corn populations There were no significant differences in total stalks due to kind, rate, and time of nitrogen applications (table 24-a). The proportion of fertile stalks, however, was significantly increased by the use of supplemental nitrogen and by increasing rates. This effect of supple- mental nitrogen was significantly greater with summer sidedressing than ‘with either fall or spring application. These results are similar to those observed in the rotation experiment (table 8-a). . In the cash crap rotation (Rotation 6) a 68 percent increase in yield for sidedressed nitrogen was associated with a 16 percent increase in number of fertile stalks. In the present experiment, a 55 percent yield increase for summer sidedressing over no supplemental nitrogen was associated with a 28 percent increase in fertile stalks. Since the summer sidedressing promoted a larger number of ears per acre than earlier applications while total stand was not affected, it appears that flowering and pollination processes in corn are critically influenced by nitrogen availability during the period just prior to and about tasseling time. It is known that peak seasonal uptake of nitrogen by corn occurs during this period (84). Seasonal Distribution of Nutrients in Corn Tissue and in Soil A number of relationships which were pointed out during the pre- sentation of data in tables l-a through 24-b may be seen more clearly in the.graphical presentation.of data for selected treatments in figures 1 through 6. 81 ICNDCDF g ' KEY YIELD 3 750_ ROT'N.:|.-ALF.-BROME PLUS MANURE O IOI _ ROT N. 6.-CASH CROPS ONLY C] 84 ; ROT'N."7. -CA§HOCREO:S SLUS SWEET _ 500_ L V G EEN MANURE O 92 2 I'm 0 250~ 2 fl 5 V Q I 1 I 4 1 _J 6/2I 7/5 7/l8 7729 BA! 9/I4 I22C>~ . [a . 1"; 80* a N 1: ,4()_ :2 a l 1 I 1 1 _1 6/2I 7/5 7/I8 7/29 8/II 9/I4 :3 443C1F (I) 22 t- 300- E "g: 200p :E a '00 1 l l J l ’1 6/2I 7/5 7/I8 7/29 8/Il 9/I4 SAMPLJNG DATE Figure l. -- Soluble nutrients in midribs of corn in three rotations on Sims clay loam. Nitrogen s1dedressed on June 9 and 30; total of 100 lbs. N. Tasseling Date: 7/25/60. 82 8°” KEY YIELD __1 ROT'N.’I.-ALF.-BROME PLUS MANURE O IOI 8 60, ROT'N.'6.-CASH CROPS ONLY C] 84 R07'~."7.-CASH CROPS PLUS SWEET g CLOVER GREEN MANURE O 92 z 40.- M 3: :5 {.5 9" "H.I.III us 111 _J o 1 1 1 1 1 4 6/2I 7/5 7/l8 7/29 S/ll 9/14 11' £1009 <1 5 4' ~: 4.1 g s _j 0 1 L 1 1 1 4 6/2l 7/5 7/l8 7/29 8/II 9/14 52' £200 <1 i . 150 S s _l '00 l l l l 4 -- ._1 6/2! 7/5 7/l8 7/29 8/II 9/14 SAMPLING DATE Figure 2. -- Available nutrients in Sims clay loam.planted to corn in three rotations. Nitrogen sidedressed on June 9 and 30; total of 100 lbs. N. Tasseling date: 7/25/60. 83 KEY YIELD 2000- “30011 CHECK O 104 m ALF.-BROME HAY C] IOS :3 _ H @1500 H + w EAT STRAW O 105 *- C)””(>\' E 1000—- 500. Z I '5’ 500- 250» Z :2 n. a. o l l 1 1 __ J 6/28 7/7 7/20 8/ll 9/l6 IZCDF ‘f '0? 80’— fi ‘- LL N I 40»- :E a n. c) 1 1 1 1 14 6/28 7/7 7/20 8/ll 9/l6 543C)? ‘5‘ m 400‘ 2’2 " 300% E .1. >‘ 213C>~ :E n. n. IOO J 1 L l i 6/28 7/7 7/20 8/ll 9/l6 SAMPLJNG DATE Figure 3. -- Soluble nutrients in midribs of corn following residue treatments on Sims clay loam. Nitrogen sidedressed on June 9 and 30; total of 100 lbs. N. Tasseling date: 7/25/60. KEY YIELD so CHECK o 104 _, ALF.-BROME HAY [:1 we 6 WHEAT STRAW o :05 a) 6()A .2. 7 40+ I m C) 2 .3 2C)? 05 ‘3 01‘ 1 l l l 1 J 6/28 7/7 7/20 7/27 B/II 9/l6 n: :' IOO ~ <1 >' <1 .i 5()~ :5 W .1 0L..- , ,, 1, 1 ,_-__.L_J___.-L_—A... ,_1L_.____,_.__,.-., # 6/28 7/7 7/20 7/27 8/Il 9/l6 x' _i 200~ 2 >- <1 it. ISO— 05 m _1 '00L 1 1 1 1 1 . J 6/28 7/7 7/20 7/27 8/ll 9/l6 SAMPLING DATE Figure 4. -- Available nutrients in Sims clay loam planted to corn following residue treatments. Nitrogen sidedressed on June 9 and 30; total of 100 lbs. N. Tasseling date: 7/25/60. 85 KEY YIELD IOOOP' FALL O 72 m SPRING [j 67 3 750 SUMMER O 82 g NO N V 47 ..— 5 500% Z I '6’ z 250~ z x ‘L I 0. O .....-n-_-.__ L.._..___.._- ._-l_.--_.--__ , -, _ 6/30 7/I4 IGOI- a. IZOE I '5’ N I 4oL SE 0. O. o 1 J ._ ____, J ——J 6/30 7/l4 8/4 9/12 3 600L (D ‘2 I- SOOr z ' 457 4'! 4000— M 2 E 300 4 4-«—-—--—-——-——-_1. _4 6/30 7/14 8/4 9/12 SAMPLING DATE Figure 5. -- Soluble nutrients in midribs of corn on Hillsdale sandy loam as related to time of application of 80 lbs. N from ammonium sulfate. Tasseling date: 8/1/60. 86 KEY YIELD 80L FALL O 72 SPRING [J 67 _l a <> 2': W 60*- .7; V ? 40L |K> 2 ‘0 e <> 5 20- g; v ...I o _1 L 1 1 #4 ISOI- 6/30 7/I4 8/4 8/l2 9/I2 O. W 3:00» <1 >' <1 ii 50% as an _I o _L 1 1 1 J 6/30 7/l4 8/4 8/I2 9/l2 zsoI x i 3200:— /V § i 4 i 3‘ I50;- 3; .J IOOL 1 4 4 l I 6/30 7/I4 8/4 8/I2 9/l2 SAMPLING DATE Figure 6. -- Available nutrients in Hillsdale sandy loam planted to corn as related to time of application of 80 lbs. N from ammonium sulfate. Tasseling date: 8/1/60. 87 For example, inverse relationships between tissue P and (.are readily apparent within sampling dates in the data for Rotation 7 in figure 1. In figure 3, a sharp drop in tissue P on July 7 was asso- ciated with large accumulations of both N and K. This behavior in the rotation and residue experiments was associated with soil tests for P (figures 2 and 4) which were less than 36 pounds, or "low" according to criteria used in‘lichigan for evaluating the Bray P1 soil test (Michigan Agr. Expt. Sta. Ext. Bul. 3-159). Soil P tests on the Hillsdale sandy loam were very high (figure 6). Here, tissue P (figure 5) accumulated during the season, and final levels were highest where tissue N was low during earlier periods of growth. Similar reciprocal relationships between nutrients were apparent in the soil tests, although, generally, these were less marked than they were in the tissue tests. In the rotation esperiment (figure 2), high soil nitrate levels in Rotation 1 were associated with relatively lower soil P throughout the season and lower soil l.in the two samplings prior to tasseling. In the nitrogen sources eXperiment, there was a significant seasonal increase in soil K in the last sampling for the check treatment where soil nitrate was very low during the tasseling and post-tasseling periods (figure 6). The principal reason for being concerned with seasonal changes in tissue or soil tests, however, is that these may give clues as to which of the three nutrients may have had a controlling influence on corn growth at specific stages of development. In this connection, the seasonal changes in tissue test values were strikingly similar for N and K.(figures l, 3 and 5). Both reached 88 seasonal highs two or three weeks before tasseling. Both declined sharply from this point on, but the major decline occurred prior to tasseling. The two or three weeks prior to tasseling is a period of rapid growth in plant size. Rapid assimilation of nitrate accounts for the rapid decline in tissue nitrate. The associated decline in tissue I is very likely due to dilution of previously absorbed K by the rapid increase in plant size and sap volume, since potassium is never ”assimilated” by the plant in the same sense as N or P are. Essentially all of the potassium in the plant remains in solution as the cation (11, 97). The rapid drop in tissue N and K associated with periods of rapid plant growth has pertinent significance for correlation studies. De- ficiencies of either nutrient should be particularly critical during such periods. The likelihood of finding positive correlations with yield should be greater during or just after such periods of peak de- mand. In the case of tissue P, the data for the residue and nitrogen source experiments (figures 3 and 5) show minimal levels two or three ‘weeks prior to tasseling. Data for this earlier period for the rota- tion experiment were unreliable, but there is a suggestion in figure 1 that levels of P may have been lower if July 5 data were available. Ian: levels of P early in the season were associated with high tests for II and K in all three experiments. Thus, it would appear that positive correlations between yields and tissue P would be more likely during early stages of growth. However, differences in tissue P between treatment means were not great. Significant correlations with yield, if encountered, would be more likely associated with variation between 89 replicates rather than between treatment means. Late season accumulations of tissue P apparent in the data for the experiment on Hillsdale sandy loam (figure 5) appear to have been due to restricted assimilation and growth resulting from earlier de- ficiencies of nitrogen. flare, negative correlations with yield would be expected. Such negative correlations between late season tissue P and yields must be interpreted with caution. It is not likely that levels of soluble P in the tissue were toxic. Rather, any strong negative cor- relation late in the season should be regarded as an artifact arising out of conditions of growth earlier in the season. With regard to soil tests, the seasonal pattern for soil nitrate in the nitrogen source experiment (figure 6) was very similar to the pattern for tissue nitrate (figure 5). In the other two experiments (figures 2 and 4) there was no tendency for soil nitrate to reach a seasonal peak two or three weeks prior to tasseling. Such an early season peak, however, was very pronounced in 1958 and 1959 on the rota- tion experiment at the same location (16). The failure to reach a peak prior to tasseling in 1960 appears to have been due to cold, wet weather through July. Leaching and reduced mineralization of nitrogen from soil organic matter under these conditions apparently resulted in equilibrium nitrate levels which tended to be constant over the early periods of growth through tasseling time. Similar weather conditions at the location where the nitrogen source experiment was located in 1960 would not have interfered with mineralization to as great an extent because of the coarser texture of soil and generally better drainage and aeration con- ditions. 90 The relationship observed in figure Z'between treatment means for yield and soil nitrate suggests that strongly positive correla-« tions could be expected through most of the season in this rotation experiment. The data in figure 6 leads to a similar expectation for the nitrogen sources experiment. In the residue experiment (figure 4), dif- ferences in soil nitrate between trash-eats were small and the general level was relatively high through most of the season. Yield differences were also small. The likelihood of significant correlations would be considerably less than in the other two experiments. There were no sig- nificant differences between replicates, so there would be no reason to expect better correlations associated with individual plot values than with treatment means. -Although treatments frequently resulted in significant differences in soil P and/or K, absolute differences were generally small. Strong correlations with yield among treatment means would not be expected, ex- cept for soil I, perhaps, in the last two samplings on the nitrogen sources experiment (figure 6). Here again, a strong negative component of response would be expected in the September 12 sampling. This would have to be regarded as an artifact arising out of earlier deficiencies of nitrogen. Small differences in treatment means for soil P and I leave little room for significant correlation with mean yields. "However, significant to highly significant differences between soil P and K tests for replicate blocks suggest that there might be greater likelihood of significant cor- relations if values for individual plots were used. The latter approach 'would also allow for a greater number of degrees of freedom for dilution of error in the correlation analysis. 7 91 General Relationships Between Nitrogen Deficiency Symptoms, Soil and Tissue Tests, and Corn Yields 3 Among the primary objectives of this investigation was the deter- mination of ”threshold" values for nitrates in soil or tissue where actual nitrogen deficits occur prior to the appearance of visible deficiency symptoms in corn. The experiments used were selected because previous yield data had shown a primary response to nitrogen. Nitrogen deficiency symptoms observed included, a) a light, yellowish; green color, general to the whole plant, and b) yellowing of the midrib followed by necrosis beginning at the apex of the midribs of leaves in the lower half of the plant (21). In the rotation experiment, nitrogen deficiency symptoms were observed by silking time (August 11) in all plots of the cash crop rotation (Rota- tion 6) and in those plots of the sweet clover rotation (Rotation 7) which had received no supplemental nitrogen. Yields with all these treatments ‘were significantly reduced from the maximum yields developed on the alfalfa- brome rotation (Rotation l) -- see tables 8-a and b. Nitrogen deficiency symptoms did develop later on in nitrogen sidedressed plots of Rotation 7, but yields were not significantly reduced from the corresponding plots of Rotation l. lean yields for this experiment are plotted against mean nitrate val- ues for tissue and soil in the scatter diagrams in figures 7 and 8. Sig- nificantly reduced yields and nitrogen deficiency symptoms prior to or at silking time were associated with treatments where tissue nitrate had dropped to about 200 ppm. or less and soil nitrate to 20 pounds per acre or less prior to the appearance of the first tassels on July 25. Visible nitrogen deficiency symptoms appeared prior to tasseling 92 ROT'N. *I.-ALF.-BR0ME Q PLUSN d- KEY: .. *6.—CASH CROPS o -- .. ¢~ .. *7.—sw1. CLOVER V .. .. SF 4 7/I8 7/29 gIZOP IzoI a) . + +- 2'00~ ¢vIOOP s e + 4. ‘ 0 80» % sor I. V ° V o 60- 60* -I o O E O 0 ). 40 l J I #14 1 1 i _J I00 200 300 400 I00 200 300 400 PPM Nog-N PPM Nog-N 3 Bl” 9/I4 US 3 20L I20» . + + :5 (3C)" xp'.’ I()C)* o: 3' C 8 e4, 80* 80r g V V o . V V _l so 60? E 00 8 )- 4O 1 L 1 _J40 1 . I J _J I00 200 300 400 IOO 200 300 400 PPM Nog-N PPM Nog-N Figure 7. -- Corn yields vs. tissue nitrate in relation to sampling date, rotation and nitrogen treatment. means.) Tasseling date: 7/25/60. (4-plot KEEY I 93 " I. - ALF. - BROME *6. -CASH CROPS *7. -SWT. CLOVER O O V é glow ' * Ioo t E +33% I $1? I" 0 ~ . 3 8° v 8° v o 60* v 50. V O ...l O 6/2I o a: E 40 1 Lo 1 1 4O 9 ’fi IO 20 30 4 I0 20 39 4o 5: LstA Nog-N Lbs./A NOS-N (D 3 + ¢ 0 '00“ | I— i % r '5 00 a. {- 8 ¢ 80F 30— U *5 V ' V o 60* V 50» V -' O 7/l8 0 7/29 E o lo 1 1 > 40 I0 20 30 4o 40 '0 20 30 40 3. Lbs./A Nag-N Lbs.lA Rog-N ‘=’>’ «I TIOOP .0 :Iom _¢. 5 ”7% - e 0 .. -¢- 80» 9 0 80 V V LI. 0 60" V 50'V 3 0 Bl” O 9/I4 “J O _ o 1 - J L I 4 40 L 1 L " 40 I0 20 30 40 IO 20 so 40 Lbs./A NO‘g-N Lbs.lA N03-N Figure 8. -- Corn yields vs. soil nitrate in relation to sampling date, rotation and nitrogen treatment. means.) Tasseling date: 7/25/60. (4-plot 94 time where no supplemental nitrogen was used on Rotation 6. Careful examination of the data revealed that, in all cases, soil tests of 20 pounds NO3-N per acre or less were found one to three weeks prior to the appearance of nitrogen deficiency symptoms. Where soil nitrate levels of 30 pounds per acre or more were maintained through the silking stage (August 11), no nitrogen deficiency symptoms were observed at any time. No similar generalization could be made regarding the tissue tests. In the nitrogen sources experiment, nitrogen deficiency symptoms developed during the period from tasseling (August 1) to completion of pollination (August 25) in all plots except those which had received 80 pounds of supplemental nitrogen. In general, yields for these treatments were significantly less than the maximum.obtained where 80-pound rates 'were used (tables 24-a and b). These were treatments where tissue nitrate in the early tasseling period (August 4) was less than 240 ppm. (figure 9). Soil nitrate early in the season (June 30) and at tasseling time (August 1) for these same treatments was 40 pounds or less; at silking time (August 12), it was 20 pounds or less. Where soil nitrate remained above 30 pounds per acre through the silking stage, no deficiency symp- toms developed until September and yields were not materially reduced from.the maximum for this experiment. Again, soil tests of 20 pounds or less were encountered one to two weeks before visible deficiency symptoms appeared. In the residue experiment, midrib firing was first observed the first of September and only on plots which received wheat straw without supplemental nitrogen. These were plots where tissue tests less than 200 ppm. of N03-N were found at silking time (August 11), and no subse- quent recovery in nitrate level occurred in the September sampling YIELD OF CORN -BUS.IA. YIELD OF CORN -BUS.IA. 95 7/I4 8/4 soI . '— so~ . — '63 V O 70* 70 UV- w; G F O 60*" . 60’— . 50* o 50" O 1 l J l i 1 J 480 680 880 I080 I40 240 340 440 PPM NOS-N PPM N03 -N . 91L? Amm. Ca. 80 L CI KEY Sulf. Nitr. 7o» 0 V - FALL ' V R v SPRING . 0 SQL Y SUMMER I C] CHECK 0 50'- 0 I0 50 |00 PPM NOJ-N Figure 9. -- Corn yields vs. tissue nitrate in relation to sampling date and time of application of two nitrogen sources. Tasseling date: 8/1/60. (4-plot means.) 96 <1 ‘80- so~ § C] I 70L 70* 2 g O :5” V . 60- V . o E50 f 0 6/30 50 ’ 0 TIM ; 1 1 l 1— l L I L L 20 4O 60 80 20 4O 60- 80 IOO S LBS.IA.NO§-N I LBS.IA.NO3-N ‘gso- . 80‘ . (f '0 D '0 D 27 VII Fl 7GP ”60» GOI‘ II. V C o 050” 8/l 50L 8/I2 J _— e 0 — 0 >' i L 1 L . l L 1 L L 20 4O 60 80 20 40 60 _80 IOO q LBSJA. NOS-N LBSJA. NO3-N U580" . KEY Amm. C0. 3 0 CI Sulf. Nitr. I 7o~ 7 FALL V V E 0 SPRING Q o o 060 SUMMER 5.0.. D . (‘3 CHECK o _ £1150 9/I2 ; I 4 1 I 20 4O 60 80 LBSJA. NOS-N ‘Figure 10. -- Corn yields vs. soil nitrate in relation to sampling date and time of application of two nitrogen sources. Tasseling date: 8/1/60. (4-plot‘means.) 97 (table 9). Also on these straw plots without supplemental nitrogen, soil tests approaching 20 pounds NO3'-N were encountered twice prior to silking, and the September test was additionally sharply reduced (table 12). The frequency of sampling in these-experiments was inadequate for precise definition of the chronological relationship between critically low soil or tissue tests and the appearance of visible deficiency symp- toms. It does appear that both tissue tests and soil tests revealed yield-limiting deficiencies of nitrogen prior to the development of ex- ternal plant symptoms. However, soil tests were much more reliable. As a first approximation, the “threshold" level for soil nitrate, below which deficiency symptoms may be expected in one to three weeks' time, is about 20 pounds per acre of nitrate-N. If soil tests fall be- low this threshold (and remain there) at any time prior to completion of pollination, it can be expected that yields will be adversely affected. The earlier this happens, the greater the expected yield reduction. Fewer samplings of tissue were made than of soil. The seasonal distribution of tissue nitrate was, accordingly, less precisely defined. However, it is tentatively proposed that the corresponding "threshold” for tissue nitrate lies somewhere between 200 and 250 ppm., based on green weight of tissue. It is apparent from the data that, only in cases of extreme nitro- gen deficiency, do soil or tissue tests reveal shortages prior to about tasseling time. Present cultural techniques and equipment do not lend themselves to the application of supplemental nitrogen after corn has reached a height of two and one-half to three feet. Even if supplemental ‘nitrogen were applied after this time, its availability would vary with the distribution of rainfall and soil moisture necessary to carry it down 98 to the corn roots. Thus, soil or tissue tests would provide a guide for current season nitrogen applications only under conditions where shortages develop early in the season. Deficiencies revealed by later soil or tis- sue tests would, however, provide a basis for guiding management practices on the same field in future seasons. Correlation Studies In the following correlation studies, a preliminary attempt has been made to develop functional relationships between corn yields and soil and tissue tests for I, P and K. Corn populations were not considered because, in the experiments dealt with, stands were significantly related to treat- ment. no attempt has been made at this point to incorporate measurements of soil reaction and soil moisture, which were also made. Two functional models have been used. One of these was a quadratic equation, including interaction terms. This type of function allows for diminishing returns and for declining yields at excessively high levels of a nutrient. It does not, however, allow for multiple inflections such as are found in the idealized sigmoid growth curve. The second model was an exponential-power function (Carter-Halter) which does allow for multiple inflections and also integrates interaction effects without sacrificing degrees of freedom. These functions were applied only to data from the rotation experi- ment on Sims clay loam.and the nitrogen source experiment on Hillsdale sandy loam. The available variance in the residue experiment was so small that it did not appear worthwhile to work with it at this time. Tests made at different seasons of the year were used in order to determine whether correlations made at one stage of growth might be more 99 informative than another. Individual observations rather than treatment means have been used in all calculations. Rotation Experiment ,5 Tissue Tests Coefficients of linear correlation among tissue tests and corn yields for the rotation experiment are tabulated in table 25. Multiple regres- sion statistics for the two functions are shown in table 26. A.highly significant, positive linear relationship between yield and tissue nitrogen in the July 18 sampling disappeared completely in the quadratic function and was broken down into non-significant positive and negative components in the Carter-Halter function. Significant to highly significant negative linear correlations be- tween yield and tissue P on all sampling dates were retained as non- significant negative components in the quadratic function. In the ex- ponential function, they were resolved into significant to highly sig- nificant positive and negative components. Yields were not significantly related to tissue K, except in com- bination with P in the quadratic function for July 18. The proportion of total variance accounted for (i2) was generally higher for the Carter-Halter function than for the quadratic, and it was higher before tasseling (July 18), at silking time (August 11) and later (September 14) than in the early tasseling period (July 29). V The relationships expressed in the Carter-Halter function between yield and tissue P are depicted graphically in figure 11. The dominant negative segments of the observed portions of these curves are probably a carryover from.the highly significant negative correlation between tissue N and tissue P which was established early in the season (table 100 Table 25. -- Coefficients of linear correlation among tissue tests and yield. Ferden Farm Rotations 1, 6 and 7. (Corn, 1960.) Tissue Coefficients of Correlation fr) Date Nutrient P K Yield 7/18 I -.4l8** -.291 .606** P -.129 -.510** R -.223 7/29 N .069 -.225 .027 P -.123 -.360* K -.024 8/11 N .062 -.l67 .268 P -.l74 -.543** K .192 9/14 N -.l70 -.332* .028 P -.211 -.627** K .063 *Significant at 5 percent level of probability. **Significant at 1 percent. 101 .ao«usmwauouuo oaowuana we ucowoamuooow .umoouum H as uuoowuqmwuwss .huqawonooum mo HNPoH unmouom n no unmoumuawqm o no unambs .flOflufln-UO 05.... a.“ In: a.“ Mun—=9? _. Ila] new . son. «mu. 3n. .w «w e woe Reno. - u m3 $234 I a was Sean. _+ e «3 23m; + u ooooo. I M nonoo. + M onooo. I I M «Hmoo. I 1 m3 Swan. + is «3 85. + I m3 3%». + I was «as. + sum mnuoo. I saw mnmoo. I «am ooooo. I «an «ogoo. I Apnoea: a was «Sac. I e was $38. + e was £5. + e was «:2. + 18.38 2 «coco. + z mecoc. + z nnooo. I z smooo. I k NHHQO.H + k. @QNONIM .... 5. HONMN. I h. GG@QM.N I fiOHUUfl—wm «N «a w;+ o I » wag assoc codename one we A Iaoausomoexm «on. sun. mom. can. a «m Mm «woo. + an coco. +. um «coo. + aum “goo. + Hz eooo. + a! mHoo. + M2 oHoo. I M: coco. + m2 «coo. + m2.-oo. I ma nooo. I an «coo. + an coco. + NH coco. + an nooo. I N“ Hooo. I M mafia. I M «who. I u «com. + u open. I um «coo. + am sooo.. I an oaoo. I «m «coo. I m nmo.~ I m Ammo. + a «cum. +. m cume. I «z «coo. + «I moo. I «z omoo. I «z oooo. I a «How. I I moon. I 3 comm. .+ z ounw. I c.eNn+ o.-~+ ms.nHI ~.ooo+ + e . + + «N «a W + m I N. flowuouau sequence one o« a mo ouuauoaso SR , ma 3: 22 38: among» um«HNHMu mo sous: , A.oomu .ouoov .n one o .a amouuouou luau damask .uoumo know no M one m .z you eueou assume no va macaw no enouuooumou mangoes! one now aouuuwueuu moaueouwom II .om «Hook 102 IOO- if 05 80r- D 00 I 2! o: C) o sor- 1.1. CD C) ..l \ hi I ._ ._ ; 40- I A -b.N bzP...b3K ‘\ ‘ I IogY=aN c' P, CZ'K c3 \‘ ' '\ I J I 1 l I 0 40 80 l20 I60 200 240 PPM H2 Po4‘-P IN TISSUE Figure 11. -- Corn yields calculated as a function of tissue P on four dates. Rotation experiment, Ferden Farm, 1960. Parameters b2 and c2 significant at l to 5 percent. - 103 25, July 18). However, the fact that both the positive and the negative components of yield response to tissue P were significant to highly sig- nificant leads to the inference that the peaks of these curves correspond to valid estimates of the optimmm level of tissue P necessary for maximum yields. These ranged from 50 to 85 ppm".and are consistent with levels of 40 to 100 ppm. reported by Nagnitski (55). Rotation Experiment, Soil Tests“ Linear correlation coefficients among soil tests and yields for the rotation experiment are given in table 27. The corresponding multiple regression coefficients for the two functions are shown in table 28. Yields were a highly significant, positive linear function of NO3'-N on all sampling dates. In the quadratic function, on the other hand, the positive component of response to N reached significance at five per- cent on only one sampling date (July 18), whereas a highly significant negative component was expressed on four dates. A sizable portion of the positive relationship between yields and soil N was apparently dis- tributed to the NP interaction term. The positive linear response to N was reflected in the Carter- Halter function by significant to highly significant positive components. A negative N component became highly significant in the last two samplings. There was a general tendency for yields to be negatively related to soil P, the linear correlation attaining significance in the last same pling. The relationship was resolved into positive and negative compo- nents in both functions, but significance was attained only in the Carter- Halter function for the sampling made at the time of silking (August 11). The NP cross product term in the quadratic was generally positive, and it was significant at one percent on July 5 and at five percent on September 14. 104 Table 27. -- Coefficients of linear correlation among soil tests and yield. Ferden Farm.Rotations l, 6 and 7. (Corn, 1960.) Soil Coefficients of Correlation (r) Date Nutrient P K Yield 6/21 N -.451** -.055 .460** P .514** -.295-- K .215 7/5 N -.313 -.100 .656** P .455** -.035.- R .141 7/18 N -.l90 -.034 .663** P .455** -.311 , R .165 7/29 N -.037 .129 .568** P .400* .lll-, K .197 8/11 N -.383* .099 .559** P .262 -.253_ N .265 9/14 N -.356* .131 .476** P .428** -.382*. R ‘-.043_ *Significant at 5 percent level of probability. **Significant at 1 percent. 105 Table 28. -- Regression statistics for the multiple regressions of yield Y) on soil tests for N, P and K.on six dates. Ferden Farm Rotations 1, 6 and 7. (Corn, 1960.) . .. Dates of sgggling soil fig _59921 6/21 749’— .7113 Quadratic Valuesfiof b in the equation, function Y - ‘ + 2 b1 X1 ' -71.791 T +61.635 *' -23.948’r + 1.4881 N + .8967 N + 3.2699 N - .0266 N2 - .0321 112“ - .0486 112“ +2.5711 P - 4.6427 P + .0966 P - ' .0372 P2 + .0006 P2 - .0602 P2 + .9309 K + .5044 x + .6427 11* - .0022 x? - .0027 x2 - .0022 :2 - .0142 NP + .0654 aw + .0113 NP + .0080 111: + .0024 a: + .0007 s: - .0078 P: + .0192 P: + .0115 P! i2 ’f .297 .571 .621 Exponential- 4 Values ofAb in the equation power log Y - a1+ Sibi X1 fund“ - 3.3754’r - 1.35301‘ - .1300’r (Carter - .0061 s - .0038 N ~ .0064 s Halter)~ + .7656 log N +' .5771 log N** + .6579 log N* - .0172 P - .0049 P , - .0168 P + .5310 10g P + .3841 log P +' .7235 log P - .0037 x - .0019 x + .0003 x + 2.0803 log K + 1.1186 log K + .3116 log R 1'12 * .298 .481 .528 106 Table 28. -- Continued Dates of sampling soil 7 29 8111 944.... values‘of b in the equation Y-a-I-Zb1x1 -9.3655‘r +101.o4’f 4.170.561' +2.0841 N + 2.0657 s - 3.4262 s - .0261 112a - .0297 112“ - .0229 112* -5.4525 P + 5.0826 P - 6.8881 P + .0007 P2 - .1031 P2, + .0980 P2 +1.4143 r - 1.2355 x - .3148 r - .0061 12 + .0048 x2 + .0007 x2 + .0271 NP + .0433 NP + .1593 10* - .0013 at - .0003 xx + .0189 NR. + .0287 PR - .0144 P: + .0044 P1: .453 .592 .526 values offib in the equation log Y - a +u£ b1 X1 -1.37611L + 6.8340”r + 4.3245'r - .0053 N - .0088 N** - .0074 N** + .6105 log N** +' .7938 log N** 1+ .5201 log N** 1+ .0105 P . - .0316 P** +1 .0038 P .- - .6341 log P +1 1.2252 log P** - .2657 log P - .0041 K +1 .0104 K + .0045 K +1.7490 log K, - 3.8003 log K - 1.4990 log N .463 .683 .489 1This is “a“ in the equation. R2 - coefficient of multiple determination. *Value of b significant at 5 percent level of probability. **Significant at 1 percent. 107 There appeared to be little relationship between yields and soil K, although a significant positive parameter was obtained in the quad- ratic on July 18. Neither function was consistently better than the other in explain- ing total variance. Both functions were notably uninformative in the June sampling, and both yielded more information just prior to tassel- ing and at silking time than during the early tasseling period (July 29). The relationships defined by the Carter-Halter function for August- 11 are shown graphically in figures 12 and 13. .At average levels of soil N, predicted yields reached a maximum at 39 pounds N per acre and 18 pounds P. The highly significant positive and negative parameters 4 for both N and P indicate that these values represent valid estimates of the optimum.levels of these nutrients at this stage of growth. The relationships in figure 13 suggest that there may have been some response to increasing levels of soil K. However, the K parameters were not significant. A point to be noted here is that the extrapolated nega- tive slopes of these curves were determined by the curvature in observed positive segments. They are obviously misleading. This illustrates one of the pitfalls involved in the interpretation of multiple regression in- Iformation-of this sort. Inferences outside the observed range must be ‘made with caution. Nitrogen Sources Experiment Linear correlation coefficients for the nitrogen sources experiment are given in tables 29 and 31. The corresponding multiple correlation coefficients appear in tables 30 and 32. Significant linear correlations with yield were found only for tis- sue nitrate end soil nitrate. Significant parameters for tissue nitrate 108 LBS/A N03-N IN SOIL O 20 4O 60 80 I T— T I r ‘I T f j \\ ~IOO \ \ \ _' \ \ A \ \ Y=f(N) \ r 80 when K=I66 \ . and P: I8 / _. ~ P= IO'/2 «25 I l00*- ’ -« 60 II _ l .‘ I’ ' ‘ 8(3" 9 I /////’f’-“~‘\\\\\\ -l 4‘) 4' ' ’ I ‘_ _ I I _. ‘3 I I In I 60- ” A ~ 20 :z I I A S L II I Y=fIPI - U I] / when K = 3 401. II I and 3: _J 0 0 II I N- _J - 11.1 I II A ; I log Y = 6.8340 , -.0088"N + .7938**Iog N 20" as as -.O3I6 P+ I.2252 log P +.0IO4 K- 3.8003 IogK R2=.683 L L 1 .l 1 l. 1 __1 0 IO 20 30 4O LBSJA. AVAIL. P IN SOIL ’ Figure 12. -- Corn yields calculated as a function of soil tests for N, P and K 17 days after tasseling. Rotation Experiment, Ferden Farm, 1960. Parameters for N and P significant at 1 percent. YIELD OF CORN -BUS. /A. 109 I20"\ \ Y=fIKI I / \\ \ when P=I8 ,’ [I L \ I \ \ N=39 /’ ’ \ \ ‘_/ I I I00 e ‘\ / i b1 X1 433.21 40.341 +361.61L - .0162 N + 5.0952 N - 5.9511 R. - .0000 112 - .0069 112 + .0025 112 - 1157 P + 1.1401 P - 2.1766 P + 0006 92 + .0009 92 + .0017 92 (*-) + l 2529 K +1 .3068 K . - .6710 K - .0013 x2 + .0012 x2 - .0000 x2 - .0013 HP - .0290 HP + .0397 HP +1 .0001 NI - .0191 NR, +' .0143 N!.(*-) + .0002 m: - .0063 P11 + .0048 PI . + .0000 NPR +' .0001 NPR - .0001 HP! 1'12 1 - .009 .254 . .234 Exponential- values of b in the equation . power logY- a+£b1 11, function (Carter- Halter) - 7.035'r - 1.925’r + 2.5511 + .0001 N - .0003 N* - .0004 N - .0457 log N + .2762 log N** - .0695 log N + .0010 P - .0001 P - .0006 P + .0167 log P - .1447 log P + .1221 log P - .0035 K - .0016 x. + .0003 K + 3.9052 log K +.1.5915 log K - .3377 log K i2 1 .039 .265 .018 1LThis is "a" in the equation. ‘*Value of b significant at 5 percent level of probability or closely approaching significance (*-). '**Value of b significant at 1 percent. I*Coefficient of multiple determination. Table 31. -- Coefficients_of linear correlation among.soil tests and yield. Nitrogen sources, times and rates of application experiment, East Lansing. (Corn, 1960.) Soil Coefficients of correlation {r} gate Nutrient P 1!. Yield 6/30 N -.012 +.139 +.519** P +.223 -.048 K ‘-.119 7/14 N +.llO -.l42 +.238 P +.067 9.037 K ' 4.064 8/4 N -.219 +4023 +.417** P +.158 - -.l36“. 8/12 N -.038 +.24l +.465fi* P +.OO7 +.Ol7 K +.214 9/12 a -.318* -.276* +.411** P +.288* -.102 K -.216 112 *Significant at 5 percent level of probability. **Significant at 1 percent. Table 32. -- Regression statistics for the multiple regressions of 113 yields (Y) on soil tests for N, P and K on five dates. Nitrogen sources, rates and times of application eXperi- ment . East Lansing. (Corn , 1960.) Nodal, Quadratic function -2# R + .++++ +11 6&0 Dates of sampling soil 1114 values of b for the equation A Y'I'I'Sbixi 40.341 -37.051' 5.0952 n - .1189 .0069 n - .0017 1.1401 s + 1.1772 .0009 92 - .0035 .3068 x + _.3461 .0012 x - .0002 .0290 as + .0027 .0191 as + .0010 .0063 px - .0025 .0001 up: - .0000 .285 - .052 N 32 P p2 K ‘2 HP ll PK NPR Exponential-power function (Carter-Halter) -2; + 4.108" - 1.2571 - .0019 N - .0012 + .4172 log N +» .2711 + .0006 P - .0036 - .1454 log P + .7734 + .0026 K - .0015 - 1.3792 log K + .7891 .286 - .007 Values ofhb in the equation logY-a+£b1Xi log N log P log K 114 Table 32. -- Continued Dates of sampling soil 8/4 , mg 9 12 valuesfiof b for the equation Y-a+$.bixi «1497.8’r -57.54* -36.92* - 15.5116 11 + 8.4212 11* + 1.2044 s - .0090 112 - .0257 112“ - .0079 112 - 2.2275 9 + 1.0103 2 + 1.3177 P - .0118 92 ‘ + .0008 92 - .0033 92 - 3.0997 x. + .4978 x + .2825 r + .0004 :2 + .0008 x2 - .0001 x2 + .1396 1117 - .0565 119* - .0061 up + .0827 an - .0339 1m + .0034 11: + .0246 n. -- .0065 m: - .0028 n - .0007 n: + .0003 um - .0000 an .163 ' .325 .110 values ofhb in the equation logY-a+ib1X1 - .4231“ + 3.090 T - .3517 " - .0034 N - .0028 N +1 .0006 N + .4086 log 8* + .2262 log N* + .0688 log N - .0061 P - .0002 P . - .0010 P + 1.4619 log P + .0640 log P + .3107 log P + .0008 K + .0021 K - .0030 K - .3102 log K - .8782 log K + .9343 log K .209 .251 .118 TTBis is "a" in the equation. *2.- coefficient of multiple determination. *Value of b significant at 5 percent level of probability. **Significant at 1 percent. 115 were encountered only in the Carter-Halter function early in the tassel- ing period (table 30, August 4). Significant positive parameters for soil nitrate (table 32) appeared about tasseling time (August 4) and at 61111163 time (August 12) in the Carter-Halter function. In the‘quadratic model, significant positive and negative components of response to nitrate appeared at silking time. These were distributed among the linear and squared N terms and the various interactions involving N. A significant negative parameter for the Pl.interaction was also found. .A.high degree of intercorrelation among the tissue nutrients them- .selves (table 29) may have contributed to the low level of significance along functional parameters on July 14 and September 12 (table 30). In- tercorrelation among independent variables may have been-an undesirable feature of the soil test data for September, also (table 31). Even where significant multiple regression coefficients were found, however, coef- ' ficients of multiple determination (i?) were low. Neither the tissue tests nor the soil tests contributed sufficient information to account for more than 25 to 30 percent of the total yield variance of this experiment. This indicates that factors other than N, P or x.were dominantly controlling corn yields. The soil pH on these plots ranged from.4.9 to 5.8 (tabla 23-a). In this range, it would be expected that soil reaction would have a very strong influence on nutrient uptake by corn. Notably, the availability of manganese may have been enhanced at the lower pH's to the point where toxic accumulations occurred in the tissue. No attempt was made in this preliminary study to incorporate available soil pH data into the functional analyses, but this should be done. Future studies on acid soils such as this should take into consideration the effects of line, and should also include the measurement of manganese in the 116 tissue. In spite of the low 22 values for the overall equations, significant parameters should be examined for possibly useful agronomic information. In figure 14, the relationships between yields and tissue and soil nitrate levels on August 4 are plotted as calculated from the Carter-Halter equa- tion. Both positive and negative components of response to tissue N were significant, leading to the conclusion that a concentration of 400 ppm. NO3"-N in green tissue was optimm for maximum yields. The apex of the curve was broad, however, and there was little difference in expected yield between 250 and 600 ppm. Only the positive parameter for soil N attained significance at five percent. However, the negative coefficient of N was significant at the ten percent level of probability. The point on the abscissa below the peak of the curve may be taken as a reasonable estimate of the Optimum level of soil nitrate. The absolute optimum appears to have been 50 pounds N per acre, although there was little difference between 40 and 60 pounds. The exponential relationship between yields and soil nitrate at silk- ing time (August 11) is shown in the upper half of figure 15. The Optimum level of soil N03‘-N appeared to have been about 37 pounds, although there was little difference between 25 and 50 pounds per acre. The correspond- ing quadratic function is plotted in the lower half of figure 15. A very similar optimum.is indicated -- about 39 pounds NO3’-N per acre. The predicted maximum yield for the quadratic is 20 bushels greater than for the Carter-Halter function, the peak of the curve is narrower, and the extrapolated decline at higher levels of N is essentially a mirror image of the observed positive leg of the curve. The form.of the Carter- Nalter function is more reasonable in terms of actual experience with patterns of crop response in the field. .80I- S ‘ ~ ~ .6 /_ ‘~ .2 3 A a? , Y=f(TlSSUE N) z 60” I ‘r I 8 A e I log Y=-I.925 - .0003 N 3 I +.2762“Iog N -.000I P Q40” -.|447 IogI5 --00I6 I? d ' +I.59I5 Iogk‘ _2 ; l R =.265 I 20 . l J 1 J_ 0 200 _400 600 800 PPM N03-N IN TISSUE 80*- é «0 604 A 3 Y = f (SOIL N) OD ' A i / log Y = -.423I -.0034 N 8 40-,’ + .4086‘ log N -.006I E a + I.46I9 Iogg +.0008K ° - .3l02 logK _2 D = _1 I R .209 m I ; 20-- I l I _L 0 20 40 60 80 LBS.IA Nog—N IN SOIL Figure 14. -- Corn yields as calculated functions of tissue tests andsoil tests (variable N, average P and K), at tasseling time. Nitrogen sources experiment, East Lansing, 1960. YIELD OF CORN - BUS. IA. YIELD OF CORN- BUSJA. 80 60 4O 80 60 40 118 r \.\ \ *- (CARTER-HALTER) I , log Y= 3.090- .0028 N I +2262“ log N- .0002 P L +.0640 log P+.002I K —.8782 log K 52: 25, J I 'I I 0 20 40 60 80 LBS. IA N03'-N IN SOII. "‘- \ \ e ‘\ '\ '\ (QUADRATICI ‘\ A * \ —, Y = -57.54 + 8.42I2 N \ ,l —.0257“ N2 + I.0I03 P \ ‘ +.0008 P2 + .4978K ‘\ +.0008 K2 - .0565’NP L -.0339* NK .0065" PK +0003" NPK -2 R =.325 l l I 1_ 20 40 60 . 80 LBSJA NOS-N IN S0IL Figure 15. -- Corn yields as calculated functions of soil tests (variable N, average P and K), 11 days after tasseling. Nitrogen sources experiment, East Lansing, 1960. DISCUSSION In the three experiments for which data have been presented here, treatments represent management inputs for which costs can be calculated. In a broad sense, several different types of ”Output”’were measured. These included soil pH, soil nutrient levels, tissue nutrient levels, corn populations and corn yields. Only the last has an economic value that can be determined, at least at the present time. Actually, economic inputs (management practices and fertilizer) and economic output (yield) are separated by a whole chain of intermediate ”input-output” relationships. Fertilizers, for example, are applied to the soil. To the extent that soil tests are affected, changes in soil nutrient levels represent the most immediate ”outputs" resulting from ap- plication of the fertilizer. Soil tests do not change linearly with the ' addition of fertilizer nutrients. Some sort of functional relationship, undoubtedly, does exist, involving numerous factors such as soil type, organic matter, rainfall, crop removal, extracting procedure, etc. Soil nutrient levels, in turn, bear some sort of relationship to plant intake. This relationship again is not simple. It is affected by factors both of the external and internal environment of the plant. Within the plant, nutrients taken up by the roots represent inputs into the internal pool of soluble rameaterials. Outputs frOm this pool represent inputs into the assimilative apparatus. There they are com- bined with inputs from the photosynthetic process to produce growth and finally storage materials in grain or other plant parts. These are out- puts to which the farmer can again assign economic value. Economists and farmers are primarily concerned with relationships between the first input and the last output in this chain. This is also 119 120 an ultimate concern of the agronomist. However, the latter is more imr mediately concerned with relationships in the intervening sequence. The agronomist's contribution must be through defining theseintermediate relationships. The ultimate Objective will be better served if special- ists in different fields can come to common agreement on conceptual schemes for formalizing their findings. The economists have taken the lead in developing mathematical frmaework for the overall problem. If the piecemeal findings of other specialists can be defined according to simm- lar functional concepts, the deveIOpment of more generalized production functions will be easier. Soil tests and tissue tests represent attempts to measure intermediate inputs or outputs in the chain. Theoretically, the closer the sequential relationship between one measured input or output and the next, the easier it should be to establish some sort of predictable relationship between them. In traditional agronomic terms, numerous workers have shown that limiting and critical levels of nutrients in plant tissue are rather con- stant for a given crop at varying stages of growth and under a wide range of soil and climatic conditions (96, 97). In the present study, data from two widely different soil types provided remarkably similar estimates of deficient and Optimum levels of nitrate in corn tissue. The accumulation of similar data from field and laboratory experiments over a period Of years should establish reliable nutritional indices for this crop (corn). These would be immediately useful to the farmer in modifying his manage- *mant practices from year to year in accordance with accepted agronomic concepts. If the Observed functional relationships can be formalized mathematically and reliable parameters found, these may eventually find 121 a place in a more generalized production equation. Soil nutrient levels are further removed in the sequence of inter- mediate inputs and outputs leading to final yield. As a result, soil tests have been found to be generally less reliable than foliar analyses in diagnosing nutritional deficiencies. Frequently soils are sampled and tested in off-seasons of the year. This introduces an additional chronological separation between the tests and the period Of crop demand. This may not be a serious factor with nutrients such as P and l.where slow release from unmeasured forms tends to maintain test values at mod- erately uniform levels. Significant seasonal fluctuations in P and K were Observed in this study, but they were small, relatively speaking, to those Observed for nitrate. Large seasonal fluctuations in soil nitrate reflected variations in rate of release from soil organic matter and rates of crap removal or leaching loss. Data from three experiments on two soil types showed that soil tests for nitrate were even.more useful than tissue tests in antici- pating deficiencies and identifying optimum.levels of nitrogen for corn. The proposed deficient level Of about 20 pounds NO3'-N per acre requires further verification. The Optimum level of about 40 pounds NO3'-N per acre was in exact agreement with that found during two previous seasons at one of the same locations. The optimm value, therefore, appears to be valid over a fairly wide range of soil and climatic conditions. Its usefulness for diagnostic purposes will depend on the development of routines which will minimize changes in nitrate content Of moist soil samples between the time Of sampling and the time Of testing. The agronomic significance of discrete limiting and optimum (or critical) levels of soil nitrate is clear. If a continuous functional 122 relationship to yield can be defined and shown to have general applica- bility, a rational basis may appear at a later date for introducing it into a generalized production function with economic significance. The mathematical functions employed were useful in verifying and providing statistical support for the optimumInitrate levels arrived at by inspection of treatment means for yields and soil or tissue tests. In addition they revealed functional relationships with soil and tissue phosphorus which were not expected and would not have been found by com- paring treatment means. There were disappointingly few significant coefficients for multiple regression. This was due, in part at least, to the relatively small numr ber of plots in individual experiments. In this regard, the Carter-Halter function had somewhat of an advantage, because fewer degrees of freedom were taken up by independent terms in the equation. Numerous coefficients significant at 10 to 20 percent probability levels were encountered. These were not examined. However, it is likely that all parameters, regardless of statistical significance, may be relevant to the observed range of measurements and should be retained. It may be possible at a later date to compare these with coefficients from other experiments covering different ranges of Observation. In this way, more generalized parameters may be derived. In the rotation experiment, up to 68 percent of total variance was accounted for by the Carter-Halter regression. Preliminary examination of the residuals indicated that these were still related to treatment. More careful study of these residual variations may make it possible to distinguish between response to the current year's fertilizer applications and response to residual effects of previous treatment on soil factors. 123 In the case of the nitrogen sources experiment, residuals appeared to be related rather strongly to soil pH. The low i? values for this experiment are very likely due to the fact that soil pH was not included in the equations for the soil tests. 'langanese toxicity was very likely a controlling physiological factor. If manganese determinations had been available for inclusion in the equations for tissue nutrients, larger i2 values might have been obtained. In spite of the low level of total information provided by these functions, parameters for nitrate appeared to be valid and led to esti- mates of optimum levels which were consistent with those for other ex- periments and from previous work. Thus, useful information can be extracted from multiple regression analyses of agronomic data, even when the overall predictability of a given group of data may be low. The ”hidden“ replication associated with regression points lends additional statistical support for inferences which are made. The conceptual format of the regression equations used here are such that they may be readily integrated into more generalized production functions as their relationship to other stages of the "input- output” sequence are better understood. SUMMARY Soil tests for nitrate and available P and K were found, by anal- ysis of variance, to be significantly influenced by numerous management and soil factors, as well as by climatic conditions and time. Soil test variations during the growing season appeared to reflect changes in rates of release from soil sources and removal by corn. Tissue tests for nitrate, phosphate and potassium in corn midribs were found to be influenced by the same factors as were soil tests. Specific effects were frequently very different, however. Tissue test variations during the season appeared to reflect variations in rate of uptahe and rate of assimilation of nutrients by the corn. Soil tests of 20 pounds per acre of nitrate nitrogen or less an- 'ticipated the development of visible nitrogen deficiency symptoms by one to three weeks. The corresponding ”threshold level" for tissue nitrate was 200 ppm. N, on a green tissue basis. ” The earlier these threshold levels were breached, the greater was the yield depression at the end of the season. Only in situations of relatively low nitrogen fertility were deficient levels observed early enough in the season to correct them by nitrogen sidedressings, using conventional equipment. Deficient levels which develOped after com! pletion of pollination did not reduce yields significantly, even when visible deficiency symptoms did develop in September prior to maturity. lultiple regression analysis of unit observations (rather than treatment means) revealed distinct optimum levels of soil and tissue nitrate which were higher than the threshold levels (40 pounds per acre and 400 ppm., respectively for soil and green tissue). The Optimum soil nitrate level coincided exactly with values obtained by inspection 124 125 of treatment means from three experiments on two different soil types in 1960 and from one of these experiments over a three-year period. Hultiple regression analysis also revealed significant to highly significant relationships between yield and both soil and tissue phos- phorus in one experiment. These were not expected and were not obvious in the treatment means. Ilultiple regression curves displayed distinct optima for soil and tissue phosphorus. These were 18 pounds per acre of P in the soil by the Bray P1 test and 50 to 85 ppm. of P extractable with acetic acid in the tissue. Soil tests for nitrate were better correlated with yield, particu- larly during periods of rapid corn development, than were tissue tests for nitrate. The reverse was true for the phosphorus tests. Essentially no significant correlation was observed between yields and soil or tis- sue tests for potassium in the experiments studied. In general, the best multiple correlations, with both soil and tissue tests, were obtained at silking time -- or, in other words, near the end of the grand period of growth and physiological development in corn . CONCLUSIGNS It appears that reliable "threshold'' values can be established- for nitrates in soil and tissue where actual R deficits develop prior to the appearance of visible deficiency symptoms in corn. These threshold values appear to correspond to the "critical” level in the zone of poverty adjustment, as defined in established schemes for in- terpretation of foliar analysis data (97). Multiple regression analysis of yields and soil or tissue tests made during the growing season revealed distinct optimum.levels of nitrate for maximum.yields of corn. These optimum.levels were higher than the threshold levels. They may represent an upper limit of balanced nutrition, beyond which accumulations of nitrate reflect critical deficiencies of some other nutrient or some other factor of growth. The range between the threshold and optimum levels may represent a zone where variable response may be expected due to varying degrees of "nutrient substitution," in the sense used by production economists. Multiple regression analysis using appropriate curvilinear models promises to be equally useful in uncovering functional relationships between crop yields and soil or tissue tests for nutrients other than 126 10. 11. 12. 13. 14. LIST OF REFERENCES Allen, 0. N. Egperiggnts in Soil Bacteriology. Burgess Publishing 60., Minneapolis. Pp. 64-73. 1957. Allison, L. E. Estimating the ability of soil to supply nitrogen. Agr. Chem. 11 (4):46-48, 139. 1956. Anderson, C. R. An economic evaluation of three soil nitrogen tests. Unpub. M.S. Thesis, Michigan State University, East Lansing, Michigan. 1958. Atkinson, H. J., Patry, L. M., and wright, L. E. Plant tissue test- ing. Sci. Agr. 24:437-442. 1944. , and Levick, R. Plant tissue testing III. Effects of fertilizer applications. Sci. Agr. 28:223-228. 1948. Au, P. Mineralization of soil organic nitrogen as influenced by organic amendments, nitrogen fertilizer, crop sequence and time. Unpub. MLS. Thesis, Michigan State University, East Lansing, Michigan. 1958. Baule, B. Landwirtsch. Jahrb. 51:363-385. 1918. Baver, L. D., and Bruner, F. E. Rapid soil tests for estimating the fertility needs of Missouri soils. ‘Missouri Agr. Exp. Sta. Bul. 404. 1939. BishOp,'w. D., and Barber, 8. A. The effect of soil phosphorus com- pounds on soil test correlation. Soil Sci. Soc. Amer. Proc. 22:435-439. 1958. Black, C. A. Evaluation of nutrient availability in soils, and pre- diction of yield response to fertilization. Iowa State College Jour. of Sci. 30:1-11. 1955. Bonner, J. Plant Biochemistgy. Academic Press Inc.,'New York. 537 pp. 1952. Bray, R. B. Soil-plant relations: I. The quantitative relation of exchangeable potassium to crop yields and to crop response to~ potash additions. Soil Sci. 58:305-324. 1944. . Soil-plant relations: II. Balanced fertilizer use through soil tests for potassium.and phosphorus. Soil Sci. 60:463-473. 1945. Correlation of soil tests with crop response to added fertilizer and with fertilizer requirement. In E. B. Kitchen, ed. Diagnostic Technigues for Soils and Crops. Amer. Potash Inst. Washington, D. C. Pp. 53-86. 1948. 127 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 128 ., and Eurtz, L. T. Determination of total, organic, and available forms of phosphorus in soils. Soil Sci. 59:39-45. 1945. Brock, W. Unpublished data. Mich. Agr. Exp. Sta. (Soil Sci.). 1958-59. Carolus, R. L. Experiences with rapid chemical tests for the deter- mination of nutrient deficiencies in vegetable craps. Proc. Amer. Soc. Hort. Sci. 33:479-583. 1936. . The use of rapid chemical plant nutrient tests in fertilizer deficiency diagnoses and vegetable crop research. Virginia Truck Exp. Sta. Bul. 98:1531-1556. 1938. Chang, S. C., and Jackson, M; L. Fractionation of soil phosphorus. Soil Sci. 84:133-144. 1957. Che, C. Mg, and Caldwell, A. C. Forms of phosphorus and fixation in soils. Soil Sci. Soc. Amer. Proc. 23:458-460. 1959. Cook, R. L., and Millar, C. E. Plant nutrient deficiencies. Michigan Agr. Exp. Sta. Spec. Bul. 353. 1953. Doll, J. P., Heady, E. 0., and Pesek, J. T. Fertilizer production functions for corn and oats; including an analysis of irrigated and residual response. Iowa. Agr. Exp. Sta. Res. Bul. 463:363- 394. 1958. Dumenil, L. Nitrogen and phosphorus composition of corn leaves and corn yields in relation to critical levels and nutrient balance. Soil Sci. Soc. Amer. Proc. 25:295-298. 1961. Dyer, B. The-analytical determination of probable available mineral plant food in soils. Jour. Chem. Soc. 65:115-167. 1894. Eckerson, S. H. Protein synthesis by plants. I. Nitrate reduction. Bot. Caz. 77:377-390. 1924. . Influence of phosphorus deficiency on the metabolism of tomato plants. Amer. Jour. Bot. 16:852. 1929. Eid, M; T., Black, C. A.,and Eempthorne, 0. Importance of soil or- ganic and inorganic phosphorus to plant growth at low and high soil temperatures. Soil Sci. 71:361-370. 1951. , and Zoellner, J. A. Sig- ———_ P s nificance of soil organic phosphorus to plant growth. Iowa Agr. Exp. Sta. Res. Bul. 406. 1954. Eik, R., “ebb, J. R., Black, C. A., Smith, C. M5, and Pesek, J. T. Evaluation of residual effects of phosphate fertilization by laboratory and plant-response methods. Soil Sci. Soc. Amer. Proc. 25:21-24. 1961. 129 30. Emmert, E. ML The determination of nitrate in green tomato and lettuce tissues. Plant Physiol. 4:519-528. 1929. 31. . The effect of soil reaction on the growth of tomatoes and lettuce and on the nitrogen, phosphorus and manganese content of the soil and plant. Kentucky Agr. Exp. Sta. Bul. 314. 1931. 32. " . Field method for estimating nitrate, phosphate and potassium.in plants. Plant Physiol. 7:315-321. 1932. 33. . Tests for phosphate, nitrate and soluble nitrogen in conducting tissue of tomato and lettuce plants as indicators of availability and yield. Kentucky Agr. Exp. Sta. Cir. 43. 1934. 34. , Plant tests as a guide to fertilizer treatment of tomatoes. Proc. Amer. Soc. Bort. Sci. 38:621-622. 1941. 35. . Plant tests as a guide to fertilizer treatment of tomatoes. Kentucky Agr. Exp. Sta. Bul. 430. 1942. 36. Emerson, P. Principle! of Soil Technolggy. The Macmillan Co., New York. Pp. 402. 1930. ‘: 37. Fallon, J. Field testing reveals the story of the plant nutrients. Crops and Soils 11 (10):9-10. 1959. 38. Ferden, C. A., and Magistad, 0. C. Use of the yield equation in pineapple culture. Jour. Amer. Soc. Agron. 24:964-975. 1932. 39. Fertilizer recommendations for Michigan crops. ‘Michigan Coop. Ext. Serv. Bul. E-159. 40. Fiske, C. E. and Subbarow, Y. 'The colorimetric determination of phosphorus. Jour. Biol. Chem. 66:375-400. 1925. 41. Fitts, J. W., Bartholomew, U. V. and Eeidel, B. Correlation between nitrifiable nitrogen and yield response of corn to nitrogen fer- tilization on Iowa soils. Soil Sci. Soc. Amer. Proc. 17:119-122. 1953. ‘ 42. Fried, M. , and Dean, L. A. A concept concerning the measurement of available soil nutrients. Soil Sci. 73:263-271. 1952. 43. Fujimoto, C. R., and Sherman, G. D. Behavior of manganese in soil and the manganese cycle. Soil Sci. 66:131-145. 1948. 44. Banway, J., and Dumenil, L. Predicting nitrogen fertilizer needs of Iowa soils: III. Use of nitrate production together with other information as a basis for making nitrogen fertilizer recom- 'mendations for corn in Iowa soils. Soil Sci. Soc. Amer. Proc. 19:77-80. 1955. 45. Harper, E. J. The accurate determination of nitrates in soils. Phenoldisulfonic acid method. Ind. Eng. Chem. 16:180-183. 1924. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 130 Beady, E. 0., Pesek, J. T. and Brown, W. C. Crop response surfaces and economic optima in fertilizer use. Iowa Agr. Exp. Sta. Res. Bul. 424. 1955. Beater, J. B. Soil and plant tests as aids in soil fertility pro- grams. Comm. Fert. 63 (5):10-16, 18, 29. 1940. Eoffer, G. N. Testing corn stalks chemically to aid in determining their plant food needs. Indiana Agr. Exp. Sta. Bul. 298. 1928. Jackson, M. L. Soil Chemical Analysis. Prentice-Hall, Inc., Engla- wood Cliffs, N. J. Pp. 197-201. 1958. Krantz, B. A., Nelson, W. L., and Burkhart, L. F. Plant-tissue tests as a tool in agronomic research. 19 N. B. Kitchen, ed. Digggostic Technigues for Soils and Crops. Amer. Potash Inst., Washington, D. C. Pp. 137-155. 1948. Kresge, C. B., and Merkle, F. C. A study of the validity of labora- tory techniques in appraising the available nitrogen producing capacity of soils. Soil Sci. Soc. Amer. Proc. 21:516-521. 1958. Lunt, E. A., Swanson, C. L. W., and Jacobson, N. G. M. The Morgan soil testing system. Connecticut Agr. Exp. Sta. Bul. 541. 1950. Lynd, J. Q. Diagnosis of plant nutrient deficiencies by means of soil tests, plant tissue tests and foliar analysis. Unpub. Ph.D. Thesis, Mich. State. Univ., East Lansing. 1948. , Turk, L. M., and Cook, R. L. Application of soil tests, tissue tests and foliar analysis to field experiments. Soil Sci. Soc. Amer. Proc. (1948) 14:236-241. 1949. Magnitski, K. P. The diagnosis of mineral nutrition of plants according to chemical composition of leaves. In W. Reuther, ed. Plant Analysis and Fertilizer Problems. American Institute of Biological Sciences, Washington, D. C. Pp. 159-179. 1961. Marsh, R. P. Comparative study of the calcium-boron metabolism of representative dicots and monocots. Soil Sci. 53:75-78. 1942. Mathers, A. C., Vietz, F. C., Jr., Jensen, M. E. and Sletten, W. B. Relationship of nitrogen and grain sorghum yield under three moisture regimes. Agron. Jour. 52:443-446. 1960. Matkin, 0. A.,and Chandler, P. A. Nitrogen in nursery soils. In K. F. Baker, ed. The U. C. system.for producing healthy container- grown plants. California Agr. Exp. Sta. Ext. Ser. Man. 23:108-122. McCool, M. M., and Weldon, M. D. The effect of soil type and fer- tilization on the composition of expressed sap of plants. Jour. Amer. Soc. Agron. 20:778-792. 1928. 60. 61. 62. 63. 65. 66. 67. 68. 69. 70. 71 72. 73. 74. 131 McGeorge, W. T. ‘Modified Neubauer method for soil cultures. Soil Sci. 62:61-70. 1946. Mehlich, A., and Drake, M. Soil chemistry and plant nutrition. 39' F. E. Bear, ed. ggtgittty of the Soil. Reinhold Publishing Corp., New York. Pp. 286-327. 1958. - Merkle, F. C. Soil testing operation, interpretation and application. Pennsylvania Agr. Exp. Sta. Bul. 398. 1940. Miller. C. E. Soil Fertility. John Wiley 6 Sons, Inc., New York. Pp. 132-134. 1955. Mitscherlich, E. A. Landwirtsch. Jahrb. 38:537-552. 1909. More R., J. A. Effects of organic amendments, fertilizer nitrogen, time and cropping sequence on the mineralization of soil organic nitrogen. Unpub. M. S. Thesis, Michigan State University, East Lansing, Mich. 1957. Morgan, M. F. Chemical soil diagnosis by the universal soil testing system. Connecticut Agr. Exp. Sta. Bul. 450. 1941. Manson, R. D., and Stanford, C. Predicting nitrogen fertilizer needs of Iowa soils: IV. Evaluation of nitrate production as a criterion of nitrogen availability. Soil Sci. Soc. Amer. Proc. 19:464-468. 1955. , and Doll, J. P. The economics of fertilizer use in crop production. L9 A. C. Norman, ed. Advances in Agrongmy XI. Academic Press, Inc., New York. Pp. 133-169. 1959. Nightingale, C. T. The nitrogen nutrition of green plants. Bot. Rev. 3:85-174. 1937. . Nitrate and carbohydrate reserves in relation to nitrogen nutrition of pineapple. Bot. Caz. 103:409-456. 1942. . Potassium and phosphate nutrition of pineapple in relation to nitrate and carbohydrate reserves. Bot. Gaz., 104:191-223. 1942. Ohlrogge, A. J. The Purdue soil and tissue tests. Indiana Agr. Exp. Sta. Bul. 584. 1952. Olsen, S. R. ,Cole, C. V., Watanabe, F. S. , and Dean, L. A. Estima- tion of available phosphorus in soils by extraction with sodium bicarbonate. U. S. D. A. Cir. 939.1954. Page, N. R., and Burkhart, L. IMineral nutrient extraction and dis- tribution in the peanut plant. Jour. Amer. Soc. Agron. 333743-755. 1941. 75. 76. 77. 78. 79. 80. 81. 82. 83. 85. 86. 87. 88. 89. 132 Parks, W. L., and Knetsch, J. L. Corn yields as influenced by' nitrogen level and drouth intensity. Agron. Jour. 51:363-364. 1959. Patel, J. M., and Mehta, B. V. Soil phosphorus fractionation studies. Soil Sci. Soc. Amer. Proc. 25:190-192. 1961. Peech, M., and English, L. Rapid microchemical soil tests. Soil Sci. 57:167-195. 1944. , Alexander, L. T., Dean, L. A., and Reed, J. F. ‘Methods of soil analysis for soil fertility investigations. U. S. D. A. Cir. 757. 1947. Pesek, J. T. Agronomic problems in securing fertilizer response data desirable for economic analysis. Lg E. L. Baum. E. D. Ready, and J. Blackmore, ed. Methodological Procedures in tte Economic Analytis of Fertilizer Use Data. Iowa State College Press, Ames. Pp. 101-112. 1956. , and Ready, E. O. Derivation and application of a method for determining minimmrecosnended rates of fertilization. Soil Sci. Soc. Amer. Proc. 22:419-423. 1958. Peterson, L. A., Attoe, O. J., and Ogden, W. B. Correlation of nitrogen soil tests with nitrogen uptake by the tobacco plant. Soil Sci. Soc. Amer. Proc. 24:205-209. 1960. ‘Pettinger, N. A. The expressed sap of corn plants as an indicator of nutrient needs. Jour. Agr. Res. 43:95-119. 1931. Prince, A. L. Determination of total nitrogen, ammonia, nitrates and nitrites in soils. Soil Sci. 59:47-52. 1945. Sayre, J. D. ‘Mineral accumulation in corn. Plant Physiol. 23: 267-281. 1948. Scarseth, G. D. Plant tissue testing in diagnosis of the nutritional status of growing plants. Soil Sci. 55:113-120. 1943. Shickluna, J. C. Chemical procedures for the analysis of soil. Michigan State University Soil Sci. Mimeograph. Pp. 7-9. 1957. Smith, G. E. Soil fertility and corn production. Missouri Agr. Exp. Sta. Bul. 583. 1952. Spillman, W. J., and Lang, E. The Law of Digtgishtgg Retutgt. World Book, New York. Pp. 1-77. 1924. . Use of the exponential yield curve in fertilizer experiments. U. S. D. A. Tech. Bul. 348. 1933. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 133 Spurway, C. N. Soil Fertility Diaggosis and Control for Field, Garden, and Greenhouse Soils. Edwards Bros., Inc., Ann Arbor, Mich. 176 pp. 1948. , and Lawton, K. Soil testing -- a practical system of soil fertility diagnosis. Mich. Agr. Exp. Sta. Tech. Bul. 132. 1949. Sundquist, W. B., and Robertson, L. 8., Jr. An economic analysis of some controlled fertilizer input-output experiments in Michigan. Mich. Agr. Exp. Sta. Tech. Bul. 269:1-57. 11959° T Thornton, S. F., Conner, S. D. and Fraser, R. R. The use of rapid chemical tests on soils and plants as side in determining fer- tilizer needs. Indiana Agr. Exp. Sta. Cir. 204. 1939. Truog, E. Present status and future of soil testing and soil analysis. Soil Sci. Soc. Amer. Proc. (1936) 2:181-183. 1937. Test for available soil nitrogen. Comm. Fert. 88:72. 1954. Ulrich, A. Plant analysis methods and interpretation of results. I; N. B. Kitchen, ed. Diagnostic Technigues for Soils and Ctops. Amer. Potash Inst., Washington, D. C. Pp. 157-198. 1948. ' , Ririe, D., Bills,wF. J., George, A. C., and Morse, M. D. (1) Plant analysis, a guide for sugar beet fertilization. California Agr. Exp. Sta. Bul. 766. 1959. Wark, D. C. Tests on plant tissue as a guide to the soil's available nutrients. Jour. Australian Inst. Agr. Sci. 5:224-227. 1939. Welch, c. n. Unpub. x. s. Thesis, North Carolina State College, Raleigh, N; C. 1947. (Cited in Krantz, et a1. , p. 146, 1948. ) White, W. C., and Pesek, J. Nature of residual nitrogen in Iowa 86113. Soil Sci. Soc. Amer. Proc. 23:39-42. 1959. Wilcox, O. W. iMeaning.of the great German soil fertility survey. Soil Sci. 79:123-132. 1955. Wolf, B., and Ichisaka, V. Rapid chemical soil and plant tests. Soil Sci. 64:227-244. 1947. ' Woodruff, C. M. Estimating the nitrogen delivery of soil from.the organic matter determination as reflected by Sanborn Field. Soil Sci. Soc. Amer. Proc. 14:208-212. 1949.