. DETERMINATION OE TOTAL CARBON gs; , . OY DRY COMBUSTION AND ITS . RELATION ,TO FORMS OF SOIL NIIIIOOEIITI’:~ ' As MEASURED III THE LABORATORY“? -~ AND IN THE GREENHOUSE ,, TheSIS for the Degree Of; Ph D MICHIGAN STATE UNIVERSITY JULIETA A. OVEIERA BELO * , A 1970 0-169 This is to certify that the thesis entitled DETERMINATION OF TOTAL CARBON BY DRY COMBINATION AND IT'S RELATION TO FORMS OF SOIL NITROGEN AS MEASURED IN THE LABORATORY AND IN THE GREENHOUSE. presented by Julieta Belo has been accepted towards fulfillment of the requirements for Ph.D. degree in Soil Science «J I Major professor Date , T q a! LIBRARY Michigan 5qu University AUG 1 32004 MAY I 2004 ABSTRACT DETERMINATION OF TOTAL CARBON BY DRY COMBUSTION AND ITS RELATION TO FORMS OF SOIL NITROGEN AS MEASURED IN THE LABORATORY AND IN THE GREENHOUSE BY Julieta A. Ovejera Belo A dry combustion method of determining total C In soils was developed. The method involved in the use of a high fre- quency induction furnace equipped with a direct reading carbon analyzer. In this procedure, IOO mg of mineral soil or 20 mg of organic soil, ground to pass an 80 mesh screen, was found to be the best sample size to use. The dry combustion method was compared with a wet combustion method and chromic acid reduction method. There were no significant differences between the values obtained by the three methods, although higher values were obtained by dry combustion than by either wet combustion or chromic acid reduction. Since dry combustion oxidizes all forms of C, the carbonates in calcareous soils need to be either removed or determined separately if organic C need to be determined. Organic C determined by dry combustion was correlated highly with total soil N, and accounting for 78% of the variation in total soil N in the 59 soils studied. The correlation of organic C with mineralizable N (released in a Julieta A. Ovejera Belo lA-day aerobic incubation at 30 C) was significant, but the correlation coefficient was low; organic C accounted for less than 20% of the variation in mineralizable N. Various laboratory measurements related to N availability, including organic C, total N, initial mineral N (NH4+ + NOS'), and mineralizable N, were correlated with dry matter yield and N uptake by oats and corn grown successively in the same pots in the greenhouse. Organic C was significantly correlated with yield of corn, with N uptake by oats and corn, and with total N uptake by the two crops, but was not correlated with yield of oats. Total soil N was significantly correlated with yield and N uptake for each crop and with the total uptake by both crops. The correlation coefficients obtained when organic C and total N were correlated with total N uptake by oats and corn, however, were low; organic C accounted for l7% and total N, 35%, of the variation in total N uptake by the two crops. Total mineral N at the end of the IA—day aerobic incubation period (initial + mineralizable) was more highly correlated with yield and N uptake by oats and corn than was mineralizable N alone. This may be attributed to high initial accumulation of mineral N prior to cropping, hence, a great part of the N taken up by the crop came from mineral N initially present in the soil. Julieta A. Ovejera Belo In a multiple regression analysis in which organic C, total N, initial mineral N, and mineralizable N were the independent variables and in which the yield and N uptake were the dependent variables, partial correlations were significant only when initial and mineralizable N were compared with yield and N uptakeof the first crop (oats) and with total N uptake of both crops. The simple and partial correlations obtained between the various laboratory measurements of N availability and yield and N uptake of oats and corn suggest that the mineral N present in the soil before cropping should be taken into consideration when estimating the fertilizer needs of the crop to be grown. DETERMINATION OF TOTAL CARBON BY DRY COMBUSTION AND ITS RELATION TO FORMS OF SOIL NITROGEN AS MEASURED IN THE LABORATORY AND IN THE GREENHOUSE BY Julieta A. Ovejera Belo A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Crop and Soil Sciences Department I970 .35) Q ~ Os .7 /»/A II ACKNOWLEDGMENTS The author expresses her sincere thanks and appreciation to Dr. E. C. Doll for his patient guidance and helpful suggestions throughout the course of this study and for his invaluable help during the preparation of the manuscript. The author is grateful to Drs. R. L. Cook, B. G. Ellis, B. D. Knezek, A. R. WoIcott, and R. S. Bandurski for serving on her guidance committee. In particular, appreciation is extended to Dr. A. R. Wolcott for his help and advice during the preparation of the manuscript. The writer is grateful to the personnel of the soil testing laboratory for their assistance in the laboratory work. Special acknowledgment is extended to Grace Woodman for the nitrogen analyses and to Sherry Sawyer for keypunching the data for computer analyses. Appreciation is extended to Mr. Burton Cedarstaff for doing the sketches of the apparatus used. The writer would also like to thank Max McKenzie and Charles Bethke for their help in the greenhouse work. A special thank you is extended to the author's husband, Panfilo S. Belo, Jr. for all the help he had unselfishly given and most of all for his understanding. The financial assistance from Tennessee Valley Authority is gratefully acknowledged. TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . Ii INTRODUCTION. . . . . . . . . . . . . . . . . . . . . l REVIEW OF LITERATURE. 3 Determining Carbon in Soils. 3 Dry combustion. 3 Wet combustion. 5 Chromic acid reduction methods. 8 Relation of organic matter to organic carbon. . . . . . . . . . . . . . . . . . . . l0 Forms of Nitrogen in Soils . . . . . . . . . . . ll Organic combinations. . . . . . . . . . . . . lI Inorganic forms of soil N . . . . . . . . . . l4 Losses of Nitrogen from the Soil . . . . . . . . l5 Measurements of Available Soil Nitrogen. . . . . l7 MATERIALS AND METHODS . . . . . . . . . . . . . . . . 24 Soil . . . . . . . . . . . . . . . . . . . . . . 24 Laboratory Procedures. . . . . . . . . . . . . . 24 Total carbon in soils . . . . . . . . . . . . 24 Total nitrogen of soils and plant tissue. . . 35 Measurement of available nitrogen by aerobic incubation. . . . . . . . . . . . . . 36 Greenhouse Experiments Statistical Analysis RESULTS AND DISCUSSION. Carbon Determinations. Development of a procedure for determining carbon in soils using an induction furnace carbon analyzer . Size of sample. Fineness of grinding. Comparison of soil carbon determinations made using the dry combustion, the wet combustion, and the chromic acid reduction methods Correlation of organic carbon with total, mineral, and mineralizable nitrogen in soils. Greenhouse Experiment. Correlation of Various Laboratory Measurements of Soil Nitrogen with Dry Matter Yield and Nitrogen Uptake in the Greenhouse. Relation between organic carbon and yields and nitrogen uptake . Relation between total nitrogen and yields and nitrogen uptake . Mineral and mineralizable nitrogen. Multiple correlation of soil tests for available N with greenhouse yield and N uptake. SUMMARY AND CONCLUSION. BIBLIOGRAPHY. 37 40 4I 4l 62 61+ 64 7O Table LIST OF TABLES Description of selected Michigan soils used in the evaluation of carbon and nitrogen determinations and water and soil amendments added in the greenhouse experiment. . . . . . . . . . . . . Effect of size of sample on the determina- tion of total C in mineral soils by dry combustion using the induction furnace carbon analyzer . Effect of size of sample on the determina- tion of total C in organic soils by dry combustion using the induction furnace carbon analyzer Effect of fineness of grinding on the determination of total C in mineral soils by dry combustion using the induction furnace carbon analyzer Comparison of carbon determination by means of dry combustion, wet combustion, and chromic acid reduction. Soil pH, and levels of total carbon and total mineral, and mineralizable nitrogen in selected Michigan soils. Linear correlation coefficients among soil carbon and various nitrogen determinations in selected Michigan soils. Dry matter produced and nitrogen removed from soil by oats and corn grown in the greenhouse on selected Michigan soils Simple correlation coefficients between yields and N uptake by oats and corn and measurements of soil N in selected Michigan soils 25 44 46 52 55 Table l0 Relation between yield and N uptake by oats and corn and various determinations for estimating availability of soil nitrogen as revealed by partial correlation and multiple regression analyses. 77 Figure LIST OF FIGURES Diagram of apparatus for determination of carbon in soil by wet combustion Diagram of the induction furnace carbon analyzer Relationship between organic C and total N in selected Michigan soils. Relationship between soil organic C and total N uptake by oat and corn in the greenhouse. Relationship between soil total N and total N uptake of oat and corn in the greenhouse. Relationship between mineral N (NH4+ + N03-) released during l4 days' aerobic incubation and total N uptake of oat and corn in the greenhouse . Relationship between total (NH4+ + N03“) N (initial + released) after l4 days' aerobic incubation and total N uptake of oat and corn in the greenhouse. . vii 30 33 57 66 73 75 INTRODUCTION The ability of a soil to supply nitrogen (N) to plants is largely dependent upon the amount of organic matter in the soil. Soil organic matter is a natural storehouse of N, since from 92 to 96% of the N in surface soils is found in organic forms (Mortensen and Himes, I964) and usually less than O.l percent is in available mineral forms (Stevenson, I965). Nitrogen in organic form is not usually directly utilized by plants, but is first converted to ammonium or nitrate by a process called mineralization. Usually, only I to 3 percent of the total soil N is mineralized by microbial processes during one growing season (Keeney and Bremner, l96éfi. The striking increase in crop yields due to increased use of N fertilizers clearly show that the amount of N mineralized during one growing season is seldom adequate for optimum crop yields (Bremner, l965d). The simple and effective laboratory test is needed to measure available soil N and to predict crop response to N fertilizers. Many chemical and biological methods have been proposed, but Bremner (l965c)states that most of these methods have very limited value and may be criticized because they are completely empirical and make no allowance for the fact that the N-immobilization cycles in soils are controlled by the supply of energy for microbial processes. For example, the limiting factors in the aerobic incubation procedure are (l) the accumulation of ammonium-N in addition to nitrate-N, (2) the difficulty of maintaining a constant moisture content, and (3) the need to avoid conditions conducive to denitrification. The organic matter content provides an index of the total substrate which is the source of mineral N. The content of organic matter therefore may give an estimate of the amount of potentially available N in a soil. Organic matter content is generally calculated from carbon (C) analyses. The objectives of this study were: (I) to develop and evaluate a method of determining C in soils using an induction furnace carbon analyzer, (2) to relate C to total N and mineralizable N in selected Michigan soils, and (3) to relate C, total N and mineralizable N to plant yield and total uptake of N in greenhouse studies. REVIEW OF LITERATURE Determining Carbon in Soils Estimates of soil organic matter are frequently based on organic carbon (C) measurements since C is readily measured quantitatively (Allison, I965). Organic C is usually determined either by (I) quantitative combustion procedures, wherein C is determined as C02, or by (2) reduction of the dichromate ion by organic matter. Dry combustion The dry combustion method for determining total C in soil and plant material has been used as the standard procedure for comparing the accuracy of other methods. In this method, the soil or organic material is heated to a high temperature in a stream of COZ—free air or 02 and the C02 produced by combustion is absorbed in alkali and determined titrimetrically or by the use of an absorbent which is suitable for gravi- metric determinations. Two types of furnace are used for dry combustion: (I) resistance, either medium or high temperature, and (2) induction (Allison, Bollen, and Moodie, I965). In a resistance furnace, the sample is heated by radkiion, conduction and convection in a tube surrounded by heating elements made of high-resistance materials (usually nichrome for medium—temperature or silicon carbide for high temperature furnaces). In an induction furnace, the source of energy is high-frequency electro- magnetic radEtion. Materials such as soil that do not heat by induction can be heated indirectly by mixing the soil with materials that do heat by induction, such as tin or iron chips. A detailed procedure using a medium-temperature resistance furnace is given by Allison, Bollen, and Moodie (I965). Jackson (I952) studied the use of a Fisher induction carbon apparatus for determining C in soils. His apparatus was designed for the determination of C in steel. Results of four soil samples of widely different C contents showed a satisfactory agreement with C determined by standard dry combustion method described by Robinson (I939). The induction furnace method is described in detail by Jackson (I958). Young and Lindbeck (I964) found that the technique of Jackson (I958) often gave inconsistent and generally low results, and developed a procedure for analysis of total C in residues from soil extract or other solution and for inorganic and organic C in soils. Stewart, Porter, and Beard (I964) used a commercial Dumas apparatus which is used both for N and C determinations (Coleman Nitrogen Analyzer Model 29). The sample is combusted at a maximum temperature of about IOOOOC and about 2 minutes is required to complete the cycle. Their results agreed with those obtained using the Fisher high frequency induction furnace. Morris and Schnitzer (I967) recently reported satisfactory C analyses of dried soil organic matter extracts using an Aminco C and H analyzer with automatic read-out for C02. They used a 2-minute combustion period at IOOOOC in pure oxygen. Wet combustion The wet combustion method of determining C in soils has been more widely used than the dry combustion technique because it does not require expensive equipment. Several wet combustion methods have been described which liberate and measure C as C02. The earlier methods (Heck, I929; Friedman and Kendall, I929) required macroequipment which is tedious to assemble and disassemble and which requires considerable bench space. Clark and 099 (I942) found that a digestion mixture of 60 parts H2804 and 40 parts H3P04 prevented the formation of $02 and kept the boiling point high enough for rapid oxidation. Earlier, Van Slyke and Folch (I940) pointed out that the oxidizing mixtures of all the wet combustion methodshitherto employed did not give accu- rate results for difficultIy-combustible organic compounds. They used an oxidizing mixture of fuming sulfuric, phosphoric, chromic and iodic acids, and obtained excellent results with all compounds including such difficultIy—oxidizable substances as cholesterol and palmitic acid. They stated that this oxidation mixture (l) vigorously attacks resistant forms of C thereby requiring a minimum boiling time for complete oxidation and that (2) H|O3 facilitates conversion of CO to C02 in the presence of readily-oxidizable carbohydrate. McCready and Hassid (I942) developed a semimicro apparatus consisting of a simple purification and absorption train assembled on a small panel and using the Van Slyke—Folch oxidation mixture. The successful use of this method for C determinations in organic compounds prompted Bremner (I949) to investigate its possible use for soils and soil extracts. Using the Van Slyke-Neil manometric apparatus he obtained results as precise as those obtained using established procedures for microquantities of material. However, this method requires the use of only very small amount of sample (containing 2 to 35 mg of carbon) which may be a serious disadvantage when analyzing soil and plant materials. Cornfield (I952) modified the apparatus of McCready and Hassid (I942) by introducing a column of heated silver to remove halogens, and used Van Slyke-Folch digestion mixture. The values he obtained for soils and plant materials are very similar to those obtained with dry combustion methods. He also noted that when soils and plant materials are dried at IOOOC, a loss of from I to 4% of the organic C occurs. He suggested that when samples are dried by heat, a temperature of not more than 600C be used. Shaw (I959) and Allison (I960) developed a method of wet combustion which is as accurate as the dry combustion technique but is simpler, equally efficient, and less time consuming. They developed a simpler apparatus than those used in studies cited previously (Heck, I929; Friedman and Kendall, I929; Clark and Ogg, I942; Van Slyke and Folch, I940; Cornfield, I952; McCready and Hassid, I942). The digestion mixture was essentially that of Clark and Ogg (I942). Extensive comparison was made on many soils to compare this digestion mixture with the Van Slyke-Folch di- gesfion mixture; their results indicate that both are equally effective in determining soil organic C. However, the shorter time (3 or 4 minutes per determination) needed when using the Van Clyke-Folch acid mixture does not compensate for the increased care needed when preparing and handling an acid containing fuming H2804. Further tests showed no need for including KIO3 in the acid mixture. The equal effective- ness of the HZSOA-H3P04-K2Cr207 oxidation mixture is apparently due to the fact that residual soil organic matter is a product of microbial action and thus contains relatively little or no active carbohydrates capable of producing CO during digestion (Allison, I960). Shaw (I959) found that this procedure may also be used to determine organic C in aqueous solutions without evapora— ting the solution to dryness when the solution contains at least 575 mg C per milliliter. To remove carbonates without loss of organic carbon, he treated a weighed quantity of soil with 5 percent H2503 and evaporated the acid in a vacuum desiccator overnight. The whole process was repeated until the carbonate was completely removed as indicated by obtaining a constant weight. He also noted that heating a wet soil to I050C results in appreciable loss of organic C, thus confirm- ing the earlier findings of Bremner (I949) and Cornfield (I952). Allison (I964) added KI and AgZSOA traps to the purifying train to precipitate chloride from saline soils which are high in this anion. Anderson and Harris (I967) further modified the digestion mixture by eliminating H2304 which they said caused the formation of insoluble films of CaSOu on carbonate particles resulting in low and erratic values for inorganic carbon. They showed that their method is effective in oxidizing benzoic acid, which is very difficult to oxidize. The apparatus they developed is also more compact than those used previously. Chromic acid reduction methods Various titration methods have been developed for determining soil C which give satisfactory results when corrected by an appropriate factor. Unlike the combustion procedures, the titration methods need no modifications for soils which contain carbonates and, they are somewhat selective with respect to relatively inert C in materials such as coal, charcoal, or graphite and C in soil organic matter. The original titration method of Schollenberger (I927) consists of adding a KZCr207—H2804 solution to a sample of soil and heating the mixture to I75 C over a low flame for 90 seconds. The cooled solution is then titrated with ferrous ammonium sulfate solution using diphenylamine as an indicator. Several modifications of this method have been developed (Degtjareff, I930; Walkley and Black, I934; Walkley, I935; Allison, I935; Purvis and Higson, I939; Smith and Weldon, I940; Chesnin, I950). The most widely used procedure is that of Walkley and Black (I934) as later modified by Walkley (I935). An appropriate sample of soil is mixed with l0 ml of I.0 N potassium dichromate solution and 20 ml of concentrated sulfuric acid is added to the mixture. The heat of reaction raisesthe temperature to about l20 C and, the mean recovery of C is reported by Walkley (I935) as 77 percent. In a critical examination of the rapid titration method, Walkley (I947) discussed the factors affecting the oxidation IO of soil organic C by dichromate-sulfuric acid mixtures. He stated that the thermal decomposition of dichromate increases with the temperature of the reaction mixture and with the time of heating, and may also be catalyzed by certain soil inorganic constituents. Tinsley (I950) studied thermal decomposition in three different dichromate mixtures and concluded that soil organic matter is nearly completely oxidized in 2 to 3 hours by boiling under a condense-with a sodium dichromate-sulfuric-perchloric acid mixture, without thermal decomposition of the dichromate. Bremner and Jenkinson (I960) later compared the Tinsley and the Walkley and Black methods with the standard wet combustion method of Shaw (I959). They reported that the Tinsley method gave more reliable results with soils than the Walkley and Black method, but concluded that neither was satisfactory for precise work. Quantitative results were obtained on a wide range of plant materials with both methods. Colorimetric procedures in which either the green color of the reduced dichromate or the yellow color of the unreduced reagent is measured in a photoelectric colorimeter or with a color chart have been reported (Wilde, I943; Graham, I948; Carolan, I948). Relation of organic matter to organic carbon The organic matter content of soils may be reported as ll either C or percent organic matter. The percent organic matter is often obtained by multiplying the percent organic C by I.724, thus assuming that the organic matter in the soil contains 58% C. Lunt (I93I) reported that I.724 is too low to be used for the organic horizons of forest soils, and suggested a factor of I.86 for forest soils. Wilson and Staker (I932), concluded that no one factor could be used for all soils if precise values for organic matter content are desired. Broadbent (I953) reported that the use of the common conversion factor of I.724 may lead to an error as large as 50% in estimating sub- soil organic matter. He suggested values of l.9 for surface soils and 2.5 for subsoils, but indicated that it would be preferable to report C rather than organic matter when C is determined. Ponomareva and Platnikova (I967) suggested the factor of I.724 be increased to 20.0. Ranney (I969) developed the following regression equation to convert organic C to organic matter in Pennsylvania surface soils: Percent organic matter = 0.35 + percent carbon x I.80 Forms of Nitrogen in Soils Organic combinations The plow layer of most cultivated mineral soils contains between 0.02 and 0.4% N, most of which is in organic form. l2 The nitrogenous organic components of soils are derived from plant and animal residues, root exudates, and microbial cells, and may be largely proteinaceous in nature (Black, I958, p. 4I2). Bremner (I949, l965d, I967) reported that about one-third of the organic N in soils is proteinaceous. Numerous workers (see literature reviews by Martin, I962, and Mortensen and Himes, I964) have demonstrated the presence of amino acids and amino sugars in soil hydrolyzates. Hydrolysis studies have shown that from 20 to 40% of the total N in most surface soils is present as bound amino acids (Bremner, I949; Stevenson, I954; Keeney and Bremner, I964), and from 5 to lo percent as combined hexosamines (Bremner and Shaw, I954; Stevenson, I957; Keeney and Bremner, I964). The amount of free amino acids in soils is negligible, but it is possibly significant since plants can absorb at least some amino acids (Black, p. 4l2, I968). Although the identifi- cation of amino acids in soil hydrolyzates suggests the presence of proteins, proteins as such have not been isolated from soils (Martin, I962). The resistance of soil organic N to biological decom- position is illustrated by the fact that less than 4% of the total soil N is mineralized in any one season (Ensminger and Pearson, I950). Although soil organic N is considered I3 to be proteinaceous in character, it is much less available to plants than is the N in proteins from plants, animals or microorganisms. Waksman and lyer (I933) state that the stability of soil organic N complexes is due to the formation of Iignin-protein complexes by a gradual process involving the reaction of carbonyl groups of lignin with amino groups of proteins. Their results concerning the stabilizing effect of Iignin on protein have been confirmed by other workers (Bennet, I949; Bremner and Shaw, I957). Ensminger and Gieseking (I942) attributed the resistance of organic N compounds to microbial attack, to adsorption of N compounds by clay minerals. They demonstrated that mix- tures of proteins with certain clays are more resistant to hydrolysis by proteolytic enzymes than protein alone. Their previous work (I939) had shown that proteins are adsorbed within the llOOI“ lattice spacing of montmorillonite clays. The adsorption of protein was found to increase with decreasing pH, indicating that proteins react as bases. Pinck, Allison, and Goddy (I954) studied microbial decom— position of proteins in the presence and absence of mont- morillonite, and found that the stability of proteins was increased by the presence of the clay mineral. l4 Broadbent and Norman (I947) and Broadbent (I948) reported that the relative unavailability of organic soil N was due to the lack of enough energy material to support a vigorous microbial population rather than formation of resistant complexes with other soil compounds. They found that part of the C and N mineralized when tracer-labelled organic materials were added to the soil was derived from the original soil organic matter. Inorganic forms of soil N Inorganic N occurs in soils chiefly as ammonium (NHOT) or nitrate (N03'); however, under certain conditions nitrite (NOZ') may accumulate(AIexander, l96l, p. 285). Plants utilize N primarily as the mineral form. Nitrite and nitrate N are mainly present in the soil solution as freely diffusible ions, while a portion of the ammonium N is held in exchange- able form by soil colloids, and can be displaced Into the soil solution by the addition soluble cations to the soil. The determination of exchangeable ammonium, nitrite and nitrate in soils Is complicated by the fact that the amount of these forms of nitrogen are subject to rapid changes through ammonification, nitrification and other microbial processes. l5 In certain soils, some of the ammonium N is intimately associated with clay minerals and is not exchangeable with other cations; this has been referred to as fixed ammonium (Allison, and Roller, I955). Bremner (l965b) defined fixed ammonium as ammonium which is released by 5 N HF-l N HCI solution after the soil has been extracted with alkaline potassium hypobromite (KOBr-KOH) solution to remove exchangeable ammonium and labile organic N compounds. The proportion of total N in a soil which is fixed as ammonium generally increases with depth (Allison, Kefauver, and Roller, I953; Legg and Allison, I959). Allison, Doetsch, and Roller (l95l) found that only about l0% or less of the fixed ammonium is available to nitrifying organisms. Losses of Nitrogen from the Soil Inorganic N can be lost from the soil by volatilization of ammonia or by leaching or denitrification of nitrates. The possibility of such losses must be taken into considera- tion in any study of soil N. The process of denitrification in soils has been reviewed by Martin (I962) and Pesek (I964). Bremner and Shaw (I958) described denitrification as a biological process whereby nitrates are reduced to gaseous N compounds such as nitrous l6 oxide and molecular N. Their work, together with that of several other workers (Nommik, I956; Arnold, I954; Jones, l95l) support the view that denitrification occurs only when the oxygen supply is restricted. In contrast to these observations, Broadbent (l95l) and Broadbent and Stojanovic (I952) found evidence that appreciable denitrification can occur under apparently aerobic conditions. In aerobic studies, Cady (I960) found that both free oxygen from the air and chemically-combined oxygen from added nitrate were used by the microbial population at the same time; this indicates that anaerobic conditions are not required for gaseous loss of N to occur. Kefauver and Allison (I957) and Skerman, Carey, and MacRae (I958) observed nitrite reduction under aerobic conditions. The reduction of nitrite under aerobic conditions as a factor of denitrification may be of considerable importance, since nitrite may be formed in soils either from reduction of nitrate when oxygen is lacking or from oxidation of ammonium when oxygen is adequate (Broadbent and Clark, I965). Jannson and Clark (I952) attributed aerobic denitrification to the possibility that a micromosaic of aerobic and anaerobic spots may develop even in a biologically active system surrounded by a pure oxygen atmosphere. l7 Bremner and Shaw (I958) found that loss of N occurred under water-logged conditions after complete disappearance of nitrate. This led them to investigate the possibility of volatilization of ammonia formed either by reduction of nitrate or by ammonification of organic N. Their results showed that less than I% of the N added as nitrate was volatilized to ammonia. According to Fuller (I963) volatili- zation losses from ammonium-containing fertilizers under aero- bic conditions are greatly reduced by placing the fertilizers below the surface of the soil; however, losses due to volatilization of ammonia formed by mineralization of organic compounds in the soil are rarely significant. Volatilization loss of NH3 can be expected when ammonia-containing or ammonia- forming fertilizers are applied on the surface of the soil, or are injected to inadequate depth or into soils that are too wet or too dry at the time of injection. Measurements of Available Soil Nitrogen For accurate determination of N needs of crops, field trials and pot experiments are probably the most direct l8 empirical methods, but are laborious and time and space consuming (Harmsen and Van Schreven, I955). Quick laboratory methods have been investigated, and several biological and chemical methods have been proposed. Bremner (I965) described three types of biological methods, each of which involves incubating the soil under conditions which promote microbial activity, and subsequently measuring some product of microbial synthesis or degradation. The methods involving estimation of mineral N (NHA+ + NO3') produced during incubation under conditions which promote mineralization of soil organic nitrogen have been the most widely used. Harmsen and Van Schreven (I955) reviewed research over the last 40 years and concluded that the Incubation techni- ques can effectively estimate the relative release of N to crops. This is probably because the factors involved in the release of mineral N during incubation are also involved in the release of N to crops during the growing season (Bremner, I965). Numerous reports indicate that the incu- bation technique will provide a fairly accurate index of the availability of soil N to plants (Fraps, l92l; Black, Nelson, and Pritchett, I947; Pritchett, Eno, and Malik, I948; Allison and Sterling, I949; Fitts, Bartholomew, and Heidel, I955; Hanway and Dumenil, I955; Munson and Stanford, I955; l9 Saunder, Ellis, and Hall, I957; Eagle and Mathews, I958; Synghal, Toogood, and Bentley, I959; Gasser, l96lb; Gasser and Williams, I963). If an aerobic incubation technique is used to determine the N-supplying power of a soil, satisfactory conditions conducive to biological activity must be established and maintained. Black, Nelson and Pritchett,(l947)mixed equal parts of air—dry soil and quartz sand to increase permeability of the soil t>air and water. Other workers (Stanford and Hanway, I955; Eagle and Mathews, I958; Synghal, Toogood, and Bentley, l959)preferred to use vermiculite, but Timmons (l96l) found that vermiculite can lead to serious errors, since it can absorb ammonia from the atmosphere, thus leading to high results. On the other hand, vermiculite can fix ammonium, and the use of vermiculite which has not been carefully pretreated to destroy its ammonium-fixing capacity may affect the results of the incubation. Bremner (l965c) found that the amount of water required for maximal aerobic mineralization of nitrogen is practically the same for different soils (0.6 ml per gram of soil) if the soil sample is mixed with three times its weight of 30- to 60-mesh quartz sand before incubation. The data and evaluation tests associated with the development of this method have recently been published (Keeney and Bremner, l966c). 20 In an attempt to eliminate the problems of establishing and maintaining satisfactory conditions for aerobic incuba- tions, Waring and Bremner (I964) proposed that the incubation of soils under water-logged conditions may provide an index of nitrogen availability. They found a close relationship between the amount of ammonium produced by their method and the sum of ammonium, nitrite and nitrate N produced by incubation under aerobic conditions. Waring (I967) later confirmed that ammonium production during incubation under waterlogged conditions proved to be as good an index of nitrogen availability as aerobic mineralization. However, Robinson (I967) reported that available mineral N obtained after aerobic incubation of fertilized soils was not well correlated with available ammonium N obtained after anaerobic Incubation. Available ammonium was less well correlated with yield and leaf N parameters than available mineral N. A complicating factor in the anaerobic incubation test could be the initial presence of nitrate which might be reduced to ammonium during incubation. Moreover, Waring and Bremner worked with soils of similar type, whereas Robinson used different soils and Included eight distinct types developed over several different parent materials. Preparation of the soil sample for subsequent incubation (air-dry and storage) was found to affect mineralization of N. 2l It has been found that air-drying increases mineralization (Birch, I960; Gasser, l96l; Cunningham, I962). Some workers also found that mineralizable N increased with time of storage (Acharya and Jain, I954; Birch, I960; Gasser, l96l; Cunningham, I962). From the standpoint of rapidity, convenience, and precision, the chemical approach to the problem of obtaining a laboratory index of soil nitrogen availability is attractive. The methods Involving estimation of the amount of N or ammonium released by treatment of the soil with various reagents have received considerable attention. During the past decade, a number of extraction procedures have been devised. Cornfield (I960) estimated ammonium released with normal sodium hydroxide by microdiffusion in a Conway unit. Keeney and Bremner (l966b) proposed a rapid chemical method involving extraction of the soil with boiling water with the subsequent treatment with potassium sulfate solution and determination of total N in the extract. In this method, the soil-water mixture was boiled under reflux for l hour and lo percent K2804 was added to the cooled mixture. The total N of the filtered extract was then determined by a semimicro Kjeldahl distillation procedure. The water-extracted nitrogen corre- lated as well with N uptake by ryegrass as did N released 22 during aerobic and anaerobic incubation. Methods In which soils were extracted with neutral, 0.5 M sodium pyro- phosphate or 0.0l M calcium chloride solutions at IOOOC, or autoclaved at IZIOC In 0.0l M calcium chloride were proposed by Stanford (I968, I969) and Stanford and Demar (I969). Total distillable N obtained by the above procedures correlated highly with mineralizable N. Evidently, the removal of soil organic N by successive extractions with dilute salt solutions represents a gradual and somewhat selective dissolution of N susceptible to mineralization. Stanford and Demar (I970) demonstrated that the diffusible NH3-N determined by a modified Conway method correlates as highly with mineralizable nitrogen as does distillable nitro— gen, and should serve equally as an Index of soil nitrogen availability, Jenkinson (I968) proposed that barium hydroxide—extractable ”glucose” Is an index of potentially available nitrogen In soils. The amount of glucose extracted was significantly correlated with the yields of barley In a field experiment,with uptake of N by ryegrass In a pot experiment, and with amount of N released by the soils during incubation. An Important advantage of the chemical methods over the incubation methods is that the amount of extracted N by the 23 chemical method Is unaffected by air-drying and air dry storage of soil samples (Keeney and Bremner, l966b). As already pointed out earlier In this literature review, air drying and air dry storage have marked effects on the results obtained by Incubation methods. MATERIALS AND METHODS Soil Fifty-nine soils varying In texture from sand to clay were used in this investigation (Table I). Soils l to 37, Inclusive, were collected during the summer of I969 and 38 to 59, Inclusive, during the summer of I967; all were sampled to plow depth. The bulk samples of the soils were air-dried, and passed through a l/2” sieve. Subsamples were ground to pass a 20-mesh sieve for subsequent laboratory analyses. Samples for C and total N analyses were ground again to pass an 80-mesh sieve. Laboratory Procedures Total carbon In soils Three methods of determining C in soils were compared: (I) the Walkley-Black chromic acid titration method (Walkley and Black, I934; Walkley, I935, I947), (2) the wet combustion method of Shaw (I959), and (3) a dry combustion method using an Induction furnace equipped with an automatic carbon analyzer (LECO 70-Second Carbon Analyzer manufactured by the labdauxy Equipment Corportion, St. Joseph, Michigan). 24 25 _ A NN mumoumz ummzz Emo_ >Ucmm m— N mN mumoomz cLou Emo_ >pcmm N— N RN comm: cLou EMOJ .— N 0N comm: CLOU EMOJ o— N _N comm: cLou EmO. >pcmm m N wN comm: cLoQ Emo_ >Ucmm m N :N EmOu N N :N mumoomz cLoQ >m_u m N mm cOmwz CLOU pcmm m N __ comm: cLoQ Emop >Ucmm J N NN mumoomz cLoU Emo_ >Ucmm m N NN cOmmz CLOU bcmm >Em04 N N 0N comm: cLoQ Emo_ >pcmm _ mcsuxou mucmEpcmE< ANV A>ucsoUV >L0um_£ aoLo L0\Ucm .oz __0m pmbpm Loam: co_um004 wum_meE_ ma>u __0m __0m .ucoE_Lwaxo mmsoscoocm msu c_ bobbm mucmEUcmEm __0m pcm Lmumz 6cm mco_umc_EL®u®p cmmocu_c pcm concmo mo co_um:_m>o esp cw paw: m__0m cmm_cu_z neuum_om mo comua_commo ._ m_nmh 26 _ N m_ 00mo_ CLOQ Em04 wN N _m 00mo_ cLoQ Emo_ >m_u NN N _N 00mo_ cLoQ EmOJ oN N ON 00mo_ CLOQ EmOJ mN N mN oomo_ cLoQ Emo_ >m_u :N N MN 00mo_ cLoU Emo_ >Ucmm MN N 0N 00mo_ cLoQ Emo_ c__3mx3mx NN N :N 00mo_ cLoU Emo. >m_o _N N NN 00mo_ cLoQ Emo_ >m_o ON N wN cOmmz CLOQ EMOJ @— N :N cOmmZ CLOQ Emo_ >bcmm @— Nn mN cOmmZ cLoQ Emon N— N mN mpmoooz cLoo Ewen @— Nn ON mbmoomz cLoQ Emo_ >pcmm m_ N KN mumoomz cLou Emo_ >m_o :— eczuxmp mucmEUcmE< ANV A>ucson >Lobm_; aoLu Lo\pcm .oz __0m pmppm Leumz co_umoon oum_boEE_ mq>u __0m __0m % .e_ucou ._ m_nmh 27 n _: cmmmc0uco mmmLu >m_o cmmchuco m: N N_ E_Luc< mm0pmu0m pcmw >Emo_ c__cm¥ N: N4 N_ E_Luc< mOODmDOQ pcmm >Emo_ c__cm¥ _: n N_ E_Luc< mmOOmDOA bcmm >Emo_ c__Lm¥ 0: N4 N_ E_Luc< moODmHOA pcmm >Emo_ c__cm¥ mm N4 N. E_Luc< mmODmDOA Ucmm >Emo_ c__me mm 4 J: comm: CLOU Emo_ >pcmm mm N _N cOmmz cLoU Emo_ >pcmm mm N mN comm: CLOQ pcmm >Em04 mm N w_ mumoumz mt_mt_< Emo_ >ecmm :m N NN oomo_ :LOU bcmm mm N mN 00mo_ Ccoo EmO. >Ucmm Nm N NN 00mo_ cLoU pcmm >Em04 _m 4 mN 00mo_ cLoU Emo_ >m_o om N RN 00mo_ cLoU pcmm >Em04 mN mcszmu 3.1%.: OOEOE “wwmwk .mmfiefl: 3%“qu am“ .a.ucoo ._ O_nmc 28 0c_NuN ”me_nun_ N4 N. E_m0ucoz mOOOMDOQ Emo_ >Ucmm E_moucoz mm N :_ mxmmx_mx mOODmHOA Ucmm >Emo_ memx_m¥ mm N m_ m_oom:H note 02 Emo_ >m_o cmcm_3 mm N _N o_m_ madmmcm QoLo oz Emo_ >m3mco mm N NN m_oomnp note 02 Emo_ >Ucmm ___;xcma mm N4 mm 3mc_mmm doco oz Emo_ >m_o >u__m mE_m :m 4 m_ couchOI m00pMu0¢ Ewo_ >Ucmm mc_m_csz mm Nu m_ couzmsoz mo0ump0m Emo_ >pcmm mc_m_c:z Nm 4 w_ couchOI mOOuMuOm Emo_ >Ucmm mc_m_c:z _m N NN m_m_ msdmmcm moODmOOQ u__m cm_EOLOm om N om m_m_ madmmcm mm0umu0m u__m cm_Eo50m m: N4 wN m_m_ madman; mOOOmDOQ u__m cm_EmLOm w: 4 mm cmmchuco mmmcu >m_o cmmmc0uco N: 4 mm commcouco mmme >m_o cmmchaco o: 4 mm cmmmc0uco mmme >m_o cmmmc0uco m: 4 mm cmmchBCO mmmco >m_o cmmmc0uco J: mcsuxmu mucmEpcmE< A&V A>ucson >Lobm_s aoLu Lo\pcm .oz m__0m pmpbm Leumz co_um004 mpm_meE_ ma>u __0m __0m .e_ucoo ._ O_Qme 29 In the Walkley-Black method, I 9 soil (80 mesh) was weighted Into a 500-ml erlenmeyer flask. Ten ml of l N KZCr207 was added, followed by 20 ml of concentrated H2504. The mixture was allowed to stand for 30 minutes to cool. Then 200 ml of distilled water, I0 ml of H3P04, 0.l g of NaF powder and 2 ml of diphenylamine Indicator were added and the mixture was titrated with 0.5 N Fe(NH)2(SOA)2.6H20 to a green end point. In the wet combustion method (Shaw, I959), the organic matter Is oxidized by heating the soil with a mixture of KzCr207, H2804, and H3P0h, and C evolved as C02. Impurities such as oxides of S and N are removed from the evolved C02, and water Is removed prior to absorbing the evoled CO2 with ascarite. The apparatus used is represented diagrama- tically In Figure l. A sample of soil (80 mesh) Is weighed into a dry digestion flask and mixed with 3 g of KZCr207 and 3 ml of water. The flask (B) is connected Into a condenser (C) through a ground glass joint. Attached to the top of the condenser Is a separatory funnel (D) with a long tube extended into the sample In the digestion flask; 25 ml of HZSOA-H3P04 mixture (l.5:l v/v) is added through the separating funnel. The flask Is heated with a micro- bunsen burner for IS minutes. The evolved C02, together with air admitted Into the digestion flask through A (Figure l), 30 pswnmwz 2.: .coepusm .4 .mnzp Npomo .x .mpvemomm mo ma:# 3 .w .wpwgmomm $0 n_:n .H .Lm30p .m .chc:$ womNI .I .mwnzp mcogvzgc< .6 can .A .mwascmcm xLOBLmawm .o .mecmecou .o .xmm_w cowpmmmwo .m .LTw A+wcza ow wasp mpvgmum< .< .AmmmF .zmsmvcovpmzneoo um; congao we :oTpmcmEmemu Low mspwemaam $0 Emgum_o NE _L0m CL mgzmvm ”mum «.5 I I J”; I 3l Is drawn through the apparatus by water suction (L) at the rate of l20 bubbles per minute (30-35 ml per minute) as measured In the H2804 in the Dreschel tube (H). The gas first passes through a tube of zinc granules (E) which absorbs oxides of N and S, Is dried by anhydrous magnesium perchlorate (F and G), bubbled Into H2504 (H) and finally absorbed in ascarite In a previously weighed Nesbitt absorption bulb U tube. Air is passed through the apparatus for at least 45 minutes after heating, and C02 Is measured by weighing the absorption bulb. The ”LECO 70 Second Carbon Analyzer“ consist of two units: (l) a high frequency Induction furnace, and (2) a direct reading carbon analyzer (Figure 2). The Instrument is equipped with a purifying train to remove moisture, C02, and acid gases from the oxygen supply. This train consIsB of an acid tower, a drying reagent tower, and a rotometer tower for measuring the flow of oxygen (Fig.2). Oxygen is used as a carrier gas at a pressure of three pounds per square Inch and a flow rate of l.5 liters per minute. The direct reading carbon analyzer utilizes the difference in thermal conduc- tivity between 02 and C02 to measure the C content of the gas. One hundred milligrams of finely ground soil (80 mesh) was weighed into a special ceramic crucible and one scoop (approximately lg) each of iron chip and tin accelerators was added. The crucible was then placed in the combustion tube 32 amcu ococp>gc< .I LemEmco co_um3ono um>_mumo pmpmmI .0 L030“ domNI nomcu mmp_xo m bcm 2V Nocz .L LOZOu ucmmmmc >Lo ame umso .m LOZOD LmuoE0uom .LmN>_mcm concmo mumcLzm co_uuzbc_ exp #0 Emcmm_e .N ®L3m_m PURIFYING 33 CARBON ANALYZER IN DUCTION FURNACE of the Induction furnace through which 02 was being passed. As the cycle was started, the sample was raised to a temperature of over I67OOC, at which It oxidized, or burned, readily. The C in the sample was oxidized to C02. After leaving the induction furnace, the gas mixture was passed through a dust trap, a sulfur trap containing MnOZ and a heated catalyst In that order, to filter out the solid tin and Iron oxides, to absorb sulfur gases which might have been oxidized during the combustion of the sample and to oxidize any CD formed to C02, respectively. Moisture was removed from the gas mixture before It entered the analyzer by an anhydrone trap. After combustion and passing through the purification train, the gas mixture (02 and C02) was passed Into a cylinder housed In a temperature-controlled oven (45°C) In the analyzer. The thermal conductivity of the gas mixture In the cylinder was then measured by a thermistor-type thermal conductivity cell. The output of the thermal conductivity cell was read on a special DC digital voltmeter as percent carbon (% C). The instrument Is calibrated to read a one-gram sample so that all the readings for mineral soils were multiplied by l0 to get % C In the soil. 35 Preliminary experiments to determine the fineness of grinding and the precise sample weight to use In the carbon analyzer were conducted. The C contents of soils ground to pass a 20 mesh sieve normally used for routine soil testing were compared with those ground to pass 80 mesh. The latter was ground for 3 minutes in a high speed impact shaker (Spex Mixer/Mill, Spex Industries, Inc., Scotch Plains, N.J.). The sample weights compared were 200, I00, and 50 mg for mineral soils and 50, 40, 30, 20, and lo mg for organic soils. Total nitrogen of soils and plant tissue The semimicro-Kjeldahl method of Bremner (I965a) was used to determine total N In soils and plant tissues except that in the KZSOA-catalyst mixture, HgO was used as a catalyst instead of CuSOO. This necessitates the addition of Na28203 to the alkali solution. Half gram of finely ground soil (80 mesh) or 50 mg of plant tissue (dried at 650C and ground to pass a 40—mesh sieve) was wrapped in a cigarette paper and added as a package Into a l00-ml Kjeldahl flask. After adding 2 ml of distilled water, the flask was allowed to stand for 30 minutes. Then l.l g of KZSOA—catalyst mixture and con— centrated H2804 (3 mlibr soil and 2 ml for plant tissue) were added. The mixture was digested until clear (2-I/2 hours for soil and l hour for plant tissue). The digested mixture 36 was made alkaline with 40% NaOH-Na23203 solution and dis- tilled for 5 minutes Into 2% borIc acid-methyl purple Indicator solution. The NH3 liberated was estimated by titration with .OI N HCI as total N. Measurement of available nitrogen by aerobic incubation The method of Incubation was essentially that of Bremner (l965c). Ten grams of soil was mixed with 30 grams of acid-washed, 40-mesh quartz sand and transferred to 8-02 wide-mouthed bottle containing 6 ml of distilled water, distributing the mixture evenly over the bottom of the bottle during the transfer. The bottle was covered with a small sheet of polyethylene held In place with a rubber band. Two small holes were punched on the polyethylene cover to facilitate aeration, while minimizing moisture loss. The bottles were placed in a constant-temperature cabinet at 300C for I4 days. At the end of the l4-day Incubation period, l00 ml of 2 N KCI was added to the soil-sand mixture and the bottle was shaken for one hour on a wrist action mechanical shaker. The suspension was allowed to stand for 30 minutes; at this time the soil-sand mixture had settled and the supernatant was clear. A 5-ml aliquot from a total volume of l06 ml of the supernatant was pipetted Into a IOO-ml Kjeldahl flask 37 and the (ammonium + nitrate)-N were released by alkaline steam distillation after the addition of 0.2 g carbonate-free MgO powder and 0.2 g of finely ground Devarda's alloy. The distillation time was 4 minutes. The NH3 liberated by steam distillation was collected Into l0 ml of 2% boric acid containing a mixture of bromcresol green and methyl red Indicators and estimated by titration with 0.005 N HCI. The amount of (ammonium + nItrate)-N initially present in the samples were determined by the same procedure. The amount of mineralizable—N reported in Table 6 was the difference between the (ammonium + nitrate)—N before and after incubation. A 5-ml aliquot of the extract of the unincubated sample was distilled without the addition of Devarda's alloy; the NH3 liberated represents the ammonium-N. This ammonium—N subtracted from the (ammonium + nItrate)-N estimated by the distillation with Devarda's alloy gave the amount of nitrate—N in the sample (Table 6). Greenhouse Experiments These experiments were conducted to correlate greenhouse yields and N uptake by oats and corn with different laboratory tests for C and N. One crop of oats and one crop of corn 38 were grown successively in the same pots. Two and one-half kilograms of each of the 59 soils were placed In polyethylene-lined No. l0 cans; the experiment was laid out In a completely randomized design with each soil replicated four times. To each pot, IOO ppm of P and IOO ppm of K were added as monocalcium phosphate (Ca(HP04)2) and potassium chloride (KCI), respectively. Those soils which tested below pH 6.2 (Table l) were lImed with a 2:l mixture of CaC03 and MgC03; the rate of liming was determined using the SMP buffer procedure (Shoemaker, McLean, and Pratt, l96l). Soils 36 and 58 did not need lime but contained less than l00 pounds per acre of Mg; 25 ppm of Mg as MgSOA.7H20 were added to these two soils. Zinc as Zn804.7H20 at the rate of 2.5 ppm for soils testing pH 5.7 to 6.5 and 5.0 ppm for those testing pH 6.6 to 8.0 was applied (Table I). Three ppm of Mn as MnSOA.H20 was added to all the soils. The lime, fertilizer, and micronutrients were thoroughly mixed with the soil by rolling on a heavy sheet of paper and placed in the containers. The pots were then arranged randomly on four adjacent benches. Thirty oat seeds were sown In each pot on December l9, I969. During planting, enough soil was removed from the pot to be used for covering the seeds to the proper depth after planting. Enough distilled water was added to wet 39 the soil approximately to field capacity. The amount of water to be added to each soil to bring it to approximate field capacity was previously determined by adding an inch of water to a IO-inch column of soil In a test tube and determining the moisture contents after 24 hours (See Table I). The seeds were spread evenly over the wet surface of the soil and covered with the dry soil. To keep the soils from drying up while the seeds were germinating, the soil surface was covered with the loose end of the polyethylene lining. Each pot was weighed so that the moisture content could be precisely controlled. The covers were removed 4 days after planting after the plants had emerged. After one more week, the plants were thinned to twenty per pot. Each pot was brought to the original weight by adding distilled water daily or every other day as needed. The oats were harvested on January l8, I970 after soils 3, I4, 25, 33, 56 and 58 showed symptoms of atrazine Injury. The plants were cut near the soil surface, placed In paper bags and dried at 650C. The dried samples were weighed and ground In a Wiley mill to pass through a 40-mesh screen. The soils In the pots were loosened and the large roots and stubble removed. On January 23, I970, 7 corn seeds were planted In each pot following the same procedure described 40 above. The plants were thinned to 5 per pot after one week. The same watering procedures as in the oat crop were followed. The corn plants were harvested on February 22, I970. The plant materials were dried, weighed and ground for total N analysis as described previously. Statistical Analysis The data were statistically analyzed and graphs were drawn using a Control Data Corportation (CDC) 3600 digital computer. The yield of dry matter and N uptake by oats and corn weresubjected to analyses of variance. A least squares routine was used to correlate C and N analyses of the soils with yields and N uptake of the oat and corn crops. This routine calculated simple correlations, multiple and partial correlation coefficients, and analyses of variance for multiple regression. Graphs were drawn showing the rela— tionships of total N with C, and of total N, C and mineral N production with the total N uptake of oats and corn in the greenhouse. RESULTS AND DISCUSSION Carbon Determinations Development of a procedure for determining carbon in soils using an Induction furnace carbon analyzer A rapid and precise method of determining C In soils was investigated for use In routine soil testing. The method involved the use of a high frequency Induction furnace equipped with a direct reading carbon analyzer. The first step In the study involved the determination of the optimum sample size and the fineness of grinding for precise analyses. Size of sample: Preliminary tests indicated that l.0 and 0.5 9 samples of finely ground mineral soils were in- completely combusted during the burn cycle. One—gram samples were apparently too large since a portion of the sample remained unoxidized after completion of the cycle. When l.0 and 0.5 9 samples were used, values obtained for % C Indicated that only about half of the C In the sample was measured when the results were compared with those obtained using 0.l 9 samples. With organic soils, the use of as small as 0.l 9 resulted in explosive flashes as soon as combustion started, causing the rubber tubing connection leading from the combustion chamber to the dust trap to break so that no C could be determined. 4l 42 Therefore, further tests were made to determine the optimum sample size to use for both mineral and organic soils. On mineral soils, 50, IOO, and 200 mg samples were used (Table 2) and on organic soils, IO, 20, 30, 40, and 50 mg samples were used (Table 3). The values for % C obtained using 200 mg samples werelower than those obtained using 50 and I00 mg samples for all the mineral soils (Table 2). No differences were noted between the results obtained using 50 and l00 mg samples. Consequently, l00 mg samples of mineral soils were used for all subsequent analyses, since the use of a larger sample minimizes sampling error. 0n organic soils (Table 3) no difference In the C contents were noted between I0 and 20 mg samples. Assuming that C determinations in the IQ and 20 mg samples represented all (l00%)of the C In the soil, recovery of C varied from 30 to 70% when 50 mg samples were used, from 74 to 93% with 40 mg samples, and from 88 to 95% with 30 mg samples. Consequently, 20 mg samples of organic soils were used In all subsequent analyses. Fineness of grinding: In the above study, the soils were first ground in a ball mill to pass an 80—mesh sieve. If this additional grinding were not needed however, con- siderable time could be saved since soils are routinely ground to pass a 20-mesh sieve for soil testing. Subsequent experiments comparing the two mesh sizes (20 and 80) Indicate 43 Table 2. Effect of size of sample on the determination of total C in mineral soils by dry combustion using the Induction furnace carbon analyzer Sample Size (Mg)l Sample No. 50 l00 200 TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTRCTFTTTTTTTTTTTTTTTTTT l 0.9l 0.92 ‘ 0.83 2 l.l7 l.l8 l.05 3 1.38 l.40 I.34 4 1.53 1.53 1.46 5 1.53 1.58 1.39 6 1.60 1.60 1.50 7 1.62 1.52 1.45 8 1.69 1.75 1.66 9 l.92 l.94 l.9l IO l.96 l.96 I.86 II 2.28 2.20 2.l6 12 2.57 2.58 2.50 l3 2.98 2.92 2.34 14 3.62 3.64 3.44 15 4.48 4.48 4.39 Mean 2.08 2.08 l.95 1These values are averages of duplicate analyses. 44 Table 3. Effect of size of sample on the determination of total C in organic soils by dry combustion using the Induction furnace carbon analyser Sample No. Sample Size (Mg)l 0 20 30 0 50 I0 I 24.60 24.40 23.30 22.70 17.20 2 31.40 31.20 29.80 27.90 9.70 3 42.70 41.20 38.03 30.90 17.00 4 42.65 41.10 37.07 31.06 17.40 Mean 35.34 34.48 32.05 28.l4 l5.33 lThese values are averages of duplicate analyses. 45 no appreciable differences In values obtained for C due to fineness of grinding of the samples (Table 4). However, using the coarser samples (20 mesh) frequently caused explosive flashes which resulted In sample loss and which necessitated cleaning the instrument and rerunning the sample. Moreover, duplication of results obtained using the coarser samples was not as good as those obtained using the finer samples, because of the difficulty of obtaining a small representative subsample from coarsely-ground samples. Therefore, finely-ground samples (80 mesh) were used in all subsequent analyses. These samples were prepared by grinding approximately l0 9 of mineral soil or 5 g of organic soil in a ball mill for at least 3 minutes. Com arison of soil carbon determinations made usIn the dr combustion the wet combustion and the chromic acid —_——.1__—_______2—_—_—__ reduction methods The dry combustion method using a high frequency Induc- tion furnace with a direct reading carbon analyzer was compared with the Walkley and Black (I934) chromic acid reduction method and the Shaw (I959) wet combustion method in which C02 evolved during the acid dichromate oxidation of organic matter with the application of heat is determined gravi- metrically (Table 5). A high degree of correlation was observed between the three methods, as indicated by signi- ficant correlation coefficients shown at the bottom of Table 5. 46 Table 4. Effect of fineness of grinding on the determination of total C in mineral soils by dry combustion using the induction furnace carbon analyzer Sample Fineness of Grinding NO- 20 Mesh 80 Mesh l 2 Mean I 2 3 Mean % C l l.08 l.02 l.04 l.05 l.02 l.03 l.03 l.03 2 0.92 0.97 l.04 0.98 l.05 l.03 l.04 l.04 3 0.95 0.90 0.94 0.93 0.95 .97 .97 .96 4 l.l4 l.0l l.02 l.06 l.92 l.94 l.94 1.93 5 0.9l 0.90 0.85 0.89 0.95 0.98 0.98 0.97 6 0.98 0 93 0.98 0.96 0.92 0.95 0.95 0.94 7 0.87 0.94 0.97 0.93 0.93 0.93 0.93 0.93 8 l.l8 l.l6 I.32 l.22 I.33 I.36 I.33 I.34 0.99 0.99 I.l0 l.03 l.08 l.I0 l.06 l.08 10 0.92 0.94 0.94 0.93 l.00 0.98 0.99 0.99 ll I.34 l.23 l.24 l.27 I.34 1.30 I.32 I.32 l2 I.87 l.96 I.8l I.88 I.85 l.92 1.93 1.90 l3 I.89 I.8l l.95 I.88 2.00 2.04 2.08 2.04 l4 I.30 l.20 l.24 l.25 l.24 l.20 l.22 l.22 l5 I.72 l.68 I.89 I.76 I.78 1.73 I.74 1.75 47 Table 5. Comparison of carbon determinations by means of dry combustion, wet combustion, and chromic acid reduction Method of Carbon Determination Soil —DW—_—_—Wet—_———_—ChrcmiicAcid No. Combustion Combustion Reduction __———%W———— 1 1.45 1.31 1.21 2 1.24 1.10 1.09 3 2.72 2.70 2.09 4 1.82 1.79 1.65 5 2.18 2.17 0.58 6 2.26 2.13 1.83 7 1.70 1.67 1.39 8 1.54 1.42 1.20 1.42 1.31 1.07 10 1.51 1.41 1.12 11 1.90 1.52 1.27 12 1.66 1.45 1.12 13 2.26 1.88 1.63 14 2.02 1.82 1.43 15 1.66 1.29 1.14 16 1.15 0.86 0.53 17 2.74 2.59 2.14 18 1.92 1.51 1.35 19 2.36 1.75 2.17 20 1.18 1.13 0.96 48 Table 5, Cont'd. Method of Carbon Determination Soil _D—w——_—TIEt_—_——T—W No. Combustion Combustion Reduction ——__——_ch—_——— 21 1.17 1.03 0.90 22 1.64 1.47 1.38 23 1.42 1.19 1.08 24 1.46 1.20 1.15 25 1.03 0.86 0.78 26 1.21 1.11 0.92 27 1.52 1.34 1.62 28 1.21 0.97 0.87 29 1.76 1.44 1.43 30 1.79 1.50 1.43 31 1.29 1.24 1.20 32 1.35 0.92 1.24 33 0.84 1.36 1.10 34 0.89 0.66 0.60 35 3.52 3.23 3.66 36 0.79 0.80 0.62 37 11.20 10.30 8.48 38 0.69 0 67 0.55 39 0.66 0.56 0.54 40 0.69 0.63 0.60 41 0.66 0.75 0.61 42 0.88 0.76 0.83 Table 5, Cont'd. et 0 0 Car on etermination Soil TTTDT7TTTTTTTTTTTTTTTTTTTEfiTTTTTTTTTTTTCETBETETAETH No. Combustion Combustion Reduction TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTT3ITEiTEET‘TTTTTTTT—TTTTTTTTTT 43 1.95 2.10 1.94 44 1.85 2.03 1.81 45 1.42 1.96 1.61 46 1.33 1.74 1.49 47 1.46 1.69 1.53 48 1.71 1.71 1.37 49 1.57 1.54 1.35 50 1.66 1.88 1.42 51 1.72 1.43 1.26 52 1.78 1.43 1.35 53 1.74 1.65 1.22 54 4.64 4.00 4.43 55 2.33 2.02 1.69 56 2.44 2.26 1.84 57 1.39 1.10 0.88 58 1.02 0.84 0.63 59 1.05 0.75 0.50 Mean 1.79 1.64 1.44 LSD (0.05) for Methods: N.S. Correlation coefficient (r) between the dry combustion method and: Wet Combustion 0.9897 mmMCAcM Reduction 0.9687 50 The wet combustion method was slightly more closely corre- lated (r=0.9897) with the dry combustion method than was the Walkley-Black method (r=0.9687). The dry combustion procedure resulted In generally higher C values than the wet combustion or the chromic acid procedure. The average C content of all the soils found by the dry combustion, the wet combustion and the chromic acid titration were 1.79, 1.64, and 1.44%, respectively. However, an analysis of variance (Steel and Torrie, 1960) did not indicate significant differences between the different methods except for soils 5 and 37. In these two soils, the chromic acid reduction procedure resulted In significantly lower values than either the dry or the wet combustion procedures. The low results with chromic acid reductions on these two soils probably Indicate that they contain a type of organic matter which Is less susceptible to the oxidizing conditions of the dichromate-sulfuric acid mixture than Is the organic matter in the other soils. According to Walkley (1947), one of the reasons for variation In the titration procedure is variation in the composition of soil organic matter; certain types of organic matter are more susceptible than others to various oxidizing conditions. The Induction furnace carbon analyzer is suitable for routine analyses because a large number of samples can be 51 rapidly analyzed at one time. Moreover, the apparatus Is simple to operate. The results discussed above Indicate that the carbon analyzer is suitable for ugawith both mineral and organic soils which vary widely In C contents. As In other dry combustion methods, however, all forms of C in soils are completely oxidized. It Is very doubtful, however, If an oxidizing agent exists which will specifically oxidize the carbon of soil organic matter and not decompose some inert carbonaceous materials such as coal and charcoal In the process (Shaw, 1959). For calcareous soils, carbonates must either be removed before C Is determined, or analyzed separately and subtracted from total C values obtained if organic C is to be calculated. None of the soils used In this study contained an appreciable amount of free carbonates, so that the C values reported herein represent mainly organic C. Correlation of organic carbon with totalz mineral, and mineralizable nitrogen In soils The relationships between organic C, as estimated by the dry combustion procedure developed In this study, to total N, mineral N present initially in the soil, and mineral N released In a 14-day Incubation period (Table 6) were evaluated using simple linear correlations. Organic C was 52 Table 6. Soil pH, and levels of total carbon and total, mineral, and mineralizable nitrogen In selected Michigan soils Soi otal Total Initial Mineral N Mineral- C N No. pH Carbon N ———————————————————— Izable2 Ratio NHQ-N N03-N Total 0 ppm 1 5.9 1.45 0.133 5.60 74.20 79.80 32.37 10.90 2 6.8 1.24 0.102 9.55 14.20 23.75 34.34 12.16 3 7.0 2.72 0.231 11.04 14.92 25.96 76.60 11.77 4 6.5 1.82 0.137 14.20 20.52 34.72 54.03 13.28 5 6.1 2.18 0.163 19.40 8.06 27.46 43.54 13.37 6 6.5 2.26 0.215 18.14 7.74 25.88 87.72 10.51 7 6.5 1.70 0.150 27.61 21.63 49.24 88.03 11.33 8 6.3 1.54 0.124 1.58 21.30 22.88 108.87 12.42 9 6.5 1.42 0.104 9.86 27.45 37.31 56.57 13.65 10 6.7 1.51 0.126 19.72 13.41 33.13 96.65 11.98 11 6.6 1.90 0.137 10.73 24.61 35.34 54.60 13.86 12 6.5 1.66 0.123 8.91 5.29 14.20 104.53 13.50 13 5.6 2.26 0.160 5.52 27.61 33.13 50.50 14.13 14 6.2 2.02 0.159 19.72 53.33 73.05 93.65 12.70 15 6.1 1.66 0.111 17.04 1.11 18.15 78.49 14.95 16 6.7 1.15 0.072 14.83 6.71 21.54 128.75 15.97 17 6.0 2.74 0.204 26.35 26.90 53.25 42.21 13.43 18 7.0 1.92 0.149 39.13 7.42 46.55 50.09 12.89 19 6.3 2.36 0.183 39.45 20.51 59.96 56.41 12.90 20 6.5 1.18 0.122 20.67 2.87 23.54 92.04 9.67 53 Table 6, Cont'd. _.——1——————__—'.—T—._—_—_—____——_—_—.——— SOI1 Total Total Initial Mineral N Minera - C/N No. pH Carbon N ———————-—————-————— izable Ratio NHg-N N03-N Total ° PPm 21 5.6 1.17 0.086 17.36 22.48 22.48 56.80 13.60 22 6.2 1.64 0.115 10.81 1.03 11.84 68.00 14.26 23 6.7 1.42 0.112 1.90 16.09 17.99 39.60 12.68 24 6.9 1.46 0.118 6.31 3.79 10.10 52.62 12.37 25 6.7 1.03 0.080 2.68 17.04 19.72 51.84 12.88 26 7.4 1.21 0.102 7.26 3.55 10.81 45.84 11.86 27 7.3 1.52 0.121 2.68 15.31 17.99 54.76 12.56 28 7.6 1.21 0.074 5.68 5.29 10.97 47.88 16.35 29 7.1 1.76 0.131 6.07 4.50 10.57 61.22 13.44 30 5.5 1.79 0.096 8.88 3.98 12.86 60.51 18.15 31 6.0 1.29 0.105 7.34 4.57 11.91 111.16 12.29 32 6.8 1.36 0.116 22.40 5.83 28.23 101.94 11.64 33 6.2 0.84 0.054 6.55 9.62 16.17 76.13 15.56 34 6.0 0.89 0.070 9.47 30.29 39.76 34.56 12.71 35 7.3 3.52 0.167 3.94 4.32 8.26 68.90 21.08 36 7.0 0.79 0.065 6.78 16.73 23.51 61.06 12.15 37 5.5 11.20 0.535 49.70 14.60 64.30 119.96 20.93 38 5.6 0.69 0.067 6.15 4.74 10.89 42.76 10.30 39 5.7 0.66 0.065 6.78 13.42 20.20 48.91 10.15 40 5.6 0.69 0.066 10.89 24.14 35.03 23.67 10.45 41 5.7 0.66 0.60 6.55 6.47 6.47 47.78 11.00 42 6.6 0.88 0.067 13.41 19.57 32.98 39.60 13.13 Table 6, Cont'd. ou ota Total Initial Mineral N inera - No. pH Carbon N ____________________ izable Ratio NHh-N N03-N Total 0 ppm 43 5.2 1.95 0.187 17.36 13.64 31.00 115.11 10.43 44 5.1 1.85 0.161 12.07 23.27 35.34 93.89 11.49 45 4.9 1.42 0.129 12.39 18 46 30.85 85.52 11.01 46 4.9 1.33 0.147 18.93 6.32 25.25 98.61 9.05 47 4.8 1.46 0.144 8.36 21.22 29.58 83.24 10.14 48 6.2 1.71 0.136 16.88 26.19 43.07 84.74 12.57 49 6.2 1.57 0.174 21 30 19.09 40.39 81.26 9.02 50 6.1 1.66 0.143 8.52 26.98 35.50 79.52 11.61 51 5.5 1.72 0.093 2.68 15.78 18.46 49.39 18.49 52 5.6 1.78 0.113 36.78 2.51 39.29 61.69 15.75 53 5.5 1.74 0.092 20.04 18.57 38.97 39.10 18.91 54 6.2 4.64 0.368 10.26 8.35 18.61 186.58 12.61 55 7.4 2.33 0.196 8.76 10.41 19 17 69.03 11.89 56 7.3 2.44 0.206 29.19 5.84 35.03 76.57 11 84 57 6.9 1.39 0.093 0.79 23.11 23.90 57.91 14.95 58 6.7 1.02 0 065 23.50 11.61 35.11 35.11 15.69 59 5.8 1.05 0.053 10.26 21.98 32.24 30.47 19.81 ——————_———————————————————_—— —————————‘_—_—___——___——___ 10rganic C determined by dry combustion. 2Mineral N (NHL.+ + N03') released during 14 days of aerobic incubation at 300C. 55 Table 7. Linear correlation coefficients among soil carbon and various nitrogen determinations Michigan soils Tess compared Carbon and Total N Carbon and initial Mineral N Carbon and Mineral N released Total N and initial Mineral N Total N and Mineral N Initial Mineral N and Mineral N released C/N ratio and Carbon C/N ratio and Total N released C/N ratio and initial Mineral N C/N ratio and Mineral N released **Significant at .001 level or less 1’ .883** .012 .440** .161 .57244 .183 .247 .205 .180 .210 in selected Correlation coefficients r2 .780 .003 .194 .026 .328 .033 .061 .042 .032 .044 Figure 3. Relationship between organic C and total N in selected Michigan soils % soil total N .352 .321 11 57 0.883 0.017 + 0.068X 2.05 2.85 3.65 % Organic carbon 4.45 58 Table 8. Dry matter produced and nitrogen removed from soil by oats and corn grown in the greenhouse on selected Michigan soils Soil Dry Matter Soil N removed NO' Oats Corn Oats Corn Total ————'g7_p?t—.———ppm_—— 1 2.23 4.14 33.81 25.17 58.98 2 0.68 2.02 10.69 10.23 20.92 (3) (0.42) (3.96) (6.90) (26.72) (33.62) 4 1.98 2.36 26.33 13.80 40.13 5 1.69 1.89 19.64 9.90 29.54 6 1 96 3.23 29.40 21.35 50.75 7 2.76 2.01 35.36 11.65 47.01 8 1.79 2.21 26.98 11.82 38.80 9 2.46 2.58 36.89 15.51 52.40 10 2.71 1.72 36.01 8.79 44.80 11 0.92 3.72 14.65 25.92 40.57 12 1.96 1.33 17.36 7.91 25.27 13 2.61 2.57 36.53 14.88 51.41 (14) (0.22) (5.31) (3.98) (53.28) (57.26) 15 1.59 2.03 19.41 10.58 29.99 16 1.57 1.96 21.67 10.94 32.61 17 2.41 1.82 33.61 9.55 43.16 18 2.24 1.78 24.60 8.95 33.55 19 2.88 1.88 40.31 10.48 50.79 20 1.81 1.16 22.24 9.22 31.46 Table 8, Cont'd. Soil No. 21 22 23 24 (25) 26 27 28 29 30 31 32 (33) 35 (36) 37 39 40 41 42 59 Dry Matter Oats Corn g/pot 2.36 1.27 2.26 1.55 2.04 1.52 0.84 2.08 (0.42) (2.87) 1.97 1.87 1.57 0.96 1.52 1.24 1.89 1.82 2.53 2.88 2.04 2.17 2.27 1.92 (0.33) (2.47) 2.00 2.16 1.53 1.29 (0.60) (1.21) 3.34 3.47 1.89 1.84 1.86 2.04 1.82 2.20 1.97 2.29 1.95 1.58 Soil N removed Oats 32. 21. 17. 13. (6 18 17. 12 16 36. 18 24. 51 01 89 06 .69) .68 26 .80 .37 16 .77 81 .44) Cor PPm 7. 9 8. 10. (15. ll. 8. 6. 11. 18. 11. 11. (12. ll. 7. (6 34. 9. 9. 1'1 49 .69 21 .73) 16 42 44 11.49 11.65 8.22 Total 40. 30. 26. 23. (22 85 26 28. .99 32. 27. 31 00 70 10 .35) .06 .04 .29 .01 .08 .17 .21) .97 .39 (16. .07 .00 20) 32 3O 60 Table 8, Cont'd. Soil Dry Matter Soil N removed No.1 "—7i527_'_—_——E37377 ‘03?§—_—_7i;¥T————_1SEST g/pot ppm 43 3.29 1.88 50.21 19.68 69.88 44 3.49 3.04 49.74 23.76 73.50 45 3.69 4.53 54.37 37.50 91.87 46 3.76 2.70 53.49 17.86 71.35 47 3.37 3.14 50.41 22.24 72.65 48 2.80 3.74 41.06 20.34 61.40 49 2 81 4.67 41.45 27.36 68.81 50 2.95 4.22 43.77 23.65 67.42 51 2.41 2.62 30.85 14.96 45.81 52 2.19 3.50 30.25 21.45 51.70 53 2.23 3.24 31.13 20.97 52.10 54 3.35 6.66 46.51 54.25 100.76 55 2.41 1.83 27.90 10.15 38.05 (56) (0.42) (4.71) (6.80) (44.34) (51.14) 57 1.97 1.86 26.17 10.26 36.43 (58) (0.49) (2.87) (7.88) (15.63) (23.51) 59 2.02 2.11 21.64 10.14 31.78 Coefficient of variation .096 .159 .104 .183 .099 Standard deviation .88** 1.19** 33.7344 27.1343 48.41** .—————————.———_——.—.——_—————_——.— W 4*significant at .0005 probability level. Values enclosed in parentheses were not included in the -4-Aa-2F4-2AA'I aan'l..n.,- 61 highly correlated with total N, and 78% of the variation in total N (Table 7, and Figure 3) between the various soils could be attributed to variations in organic C. Although organic C was significantly correlated with mineral N released during aerobic incubation (Table 7), less than 20% of the variation in mineralizable N could be attributed to organic C. Greenhouse Experiment In the greenhouse experiment, one crop of oats and one crop of corn were grown successively in the same pots. The yields and N uptake by oats and corn are given in Table 8. As mentioned previously, soils 3, 14, 25, 33, 36, 56 and 58 showed symptoms of atrazine injury when the oats were harvested; although the data from these soils are included, they were not used in the statistical analysis of the data and subsequent discussion of results, since N was at the first limiting factor in growth. Nitrogen uptake was signi- ficantly correlated with yield of both oats and corn (r=0.933 and 0.949, respectively). There was a wide variation among the 52 soils in their ability to provide N for the growth of oats and corn in the greenhouse. Total N removed by the two crops from different 62 soils varied from 19.28 to 100.76 ppm. Highly significant differences were noted among the soils with respect to both yields and N uptake by oats and corn. N removal by oats varied from 5.44 to 54.37 ppm, with an average for all soils of 26.17 ppm. Nitrogen removed by corn varied from 6.49 to 54.25 with an average of 16.36 ppm. The higher N removed by the first crop (oats) is probably due to a relatively high initial content of mineral N (NHu + N03) which ranged from 8.26 to 79.80 with an average of 29.44 ppm. The average total N removed by the two crops was 42.54 ppm. The average yield, however, was lower in the oats (2.03 g/pot) than in corn (2.54 g/pot). This seems to follow Willcox's (1954) llinverse yield-nitrogen law‘I which says that the higher the specific yield of dry substance of a crop the lower the N content in that dry substance. According to Viets (1965), Willcox's law seems to be an offhand reasonable relation when one compares a series of different crops grown on the same soil and fertilized alike. Correlation of Various Laboratory Measurements of Soil Nitrogen with Dry Matter Yield and Nitrogen Uptake in the Greenhouse The laboratory soil tests for availability of soil nitrogen refer to C, total N, mineral N (NH4+ + N03') initially present in the soils and mineral N released in a 14-day incubation period (Table 6). Simple linear correla— tions (r) were calculated between the soil tests and yield and N uptake by oats and corn (Table 9). Multiple regression Table 9. Simple correlation coefficients between yields and N uptake by oats and corn and measurements of soil N in selected Michigan soils Correlation coefficients (r) Soil test Yield N uptake Oat Corn Oat Corn Total Carbon (%) .246 ,386** .300* .473*** .412** Total N (%) .425** .505*** .482*** .602*** .588*** Initial NHh-N .260 .044 .281 .031 .193 Initial N03—N .185 .313* .301* .225 .296* Initial (NH4+N03)-N .321* .292* .432** .210 .373** Aerobic Incubation (NH4+N03)—N released .430** .366** .439*** .485*** .506*** (NH4+N03)-N released + Initial NO3_N .522*** .50 *%* .580*** .595*** .646*** Released + Initial (NH#+N03)_N .568*** .491*** .628*** .569*** .665*** C/N Ratio -.280* -.151 -.267 -.169 -.249 *Significant at .05 level **Significant at .01 level ***Signif1cant at .001 level 64 analyses were calculated using the soil tests as independent variables and yield and N uptake as dependent variables (Table 10). Relation between organic carbon and yields and nitrogen uptake Significant correlations were obtained between organic C and yield of corn, N removed by oats and corn, and total N removed by the two crops; no significant correlation was obtained with yield of oats (Table 9). Only 17% of the varia- tions in total N uptake by oats and corn could be attributed to organic C. Although the correlation between organic C and total N uptake by oats and corn was significant, the relationship shown in Figure 4 does not seem to indicate that organic C is a precise index of a 5011's capacity to supply N to the plants. It may be useful, however, in a very general way, as a means of separating soils on which a relatively small response to fertilizer N would be expected from those on which a larger response might be expected. Relation between total nitrogen and yields and nitrogen uptake Total soil N was more highly correlated with total N uptake by both oats and corn than was organic C (Table 9.) Note that the correlations between total soil N and organic C and yield and N uptake were higher for the second crop (corn) Figure 4. Relationship between soil organic C and total N uptake by oat and corn in the greenhouse. ppm N removed from soil 66 0.412 62.48 + 27.82 X mm JG) NN 3 25 3.65 % Organic carbon 4.05 4.45 67 than for the first crop (oats). Apparently, most of the N utilized by the first crop was from the mineral N initially present in the soil, and after the initial mineral N became exhausted, N released from the organic forms became relatively more important as a source of N for the plants. Total N in the soil accounted for about 35% of the variation in the total N uptake by oats and corn (Figure 5). Although earlier investigators (Fraps, 1921; Fraps and Sterges, 1932) reported that the higher the total soil N, the more mineral N is released in incubation experiments and under field conditions, Gainey (1936) found that N mineralization and fertility of a soil are more closely associated with the characteristic properties of only a small part of the total N. His investigation showed that different amounts of inorganic N were released from different soil types which contained nearly equal amounts of total N. Allison and Sterling (1948) reported that total N appeared to be a rough index of N-supplying power only for soils of the same type, under like climatic conditions, and when thoroughly humified soil organic matter is fairly uniform in quality regardless of the past agronomic treatments. The inconsistencies reported above would suggest that the low correlation obtained between total N and N uptake by oats and corn in this study, using soils of different types and wide variation in total N would be expected. It seems Figure 5. Relationship between soil total N and total N uptake of oat and corn in the greenhouse. ppm N removed from soil 95.58 88.44 80.30 72.16 64.02 55.88 47.74 39.60 31.46 23.32 .163 .226 % soil total N .289 0.588 42.21 + 518.73X .352 70 unlikely that the total N content could be used to precisely predict the N needs of these soils. Mineral and mineralizable nitrogen Mineral N released during 14 days aerobic incubation at 30 C (Table 6) was significantly correlated with yield and N uptake of oats and corn (Table 9). The correlation coefficient for N uptake was higher for the second crop (corn) than for the first crop (oats). This would be ex- pected since it is likely that the first crop had taken up much of the mineral N initially present in the soil and N for the second crop would then be derived largely from mineralizable sources. This may also explain the signifi- cant correlation between the initial mineral N in the soils and N uptake by oats, while no significance was shown for the same correlation for corn. Initial NHu-N was not correlated with yield or N uptake for either crop, but initial N03-N was significantly correlated with yield of corn, N uptake by oats, and total N uptake by both crops. Correlations with yield and N uptake were higher when initial NO3-N was included with mineralizable N. The highest degree of correlation, however, was obtained between total mineral N (initial + released) and N uptake by each crop separately and total N uptake by both crops. 71 The significant correlation obtained between initial mineral N and N uptake shows the importance of initial mineral N to the immediate crop. This contradicts the widely recognized view, as summed up by Harmsen and Van Schreven (1955), that the content of mineral N found in the soil at a given time does not provide helpful information aboutthe need for N fertilization. They say this is because the content of mineral N fluctuates over a wide range and is influenced by such external factors as season, weather conditions, plant growth, and N fertilization. In the widely used incubation methods for measuring available N, mineral N originally present in soil is always subtracted from post incubation results to get the mineral N released. The higher correlation coefficient obtained between total mineral N (initial + released) compared to that between mineral N released alone and N uptake (Figures 6 and 7) in this study, however, suggests that mineral N present in the soil before incubation should possibly be included in the interpretation of the results of an incubation experiment. This is in agreement with the results of Smith (1966) who showed that the usefulness of both aerobic and anaerobic incubation methods of N availability measurement is severely limited by excluding the available N present in the soil at sampling time. Munson and Stanford (1955) also found that available N initially present in the soil made a 72 Figure 6. Relationship between mineral N (NH4+ + N03') released during 14 days aerobic incubation and total N uptake of oat and corn in the greenhouse. ppm N removed from soil 96.58 88.44 80.30 73 r = 0.506 V = 55.70 + 0.77 X ppm Mineral N released in aerobic incubation Figure 7. Relationship between total (NH§++N03') N (initial + released) after 14 ays aerobic incubation and total N uptake of oat and corn in the greenhouse. ppm N removed from soil 96.58 88.44 80.30 72.16 64.02 39.60 31.46 23.32 75 0.665 A y = 13.09 + 0.98 X 1 II CD m \0 I\ O\ O O '— \O O M \O \o #4 rs rs O M \O 0\ ppm Mineral N after incubation (initial + released) 76 highly significant contribution to N uptake. Olson (1960) et a1. pointed out that one of the limitations of the popular biological mineralization test for nitrogen is that mineral N, particularly nitrate, which has accumulated during fallow or from nitrogen fertilizer is not reflected by the test. Multiple correlation of soil tests for available N with greenhouse yield and N uptake Multiple correlations and regressions were calculated to determine the combined effects of the soil tests for available N on yield and N uptake (Table 10). The soil tests for organic C, total N, initial mineral N and mineral N released during incubation were used as independent variables and the yields and N uptake were used as dependent variables. The multiple correlation coefficients (R) relating the four soil tests with yield and N uptake gave higher correlations than when the soil tests were used individually in simple correlations. While the multiple correlation coefficients were highly significant in all cases, the partial correlation coefficients showed a significant rela- tionship for yield and N uptake only when they were correlated with initial mineral N and released mineral N for the first crop and the total N uptake by both crops. This is a different trend from the simple correlations (Table 9) in which mineral N released was more highly correlated with yield 77 I! mmo_ co mmo_ Lo _o>m_ .0. pm ucmo_m_cm_mkk _o>o_ _oo. um ucmo_w_cm_mkk% mmo_ Lo _o>o_ mo. pm ucmo_m_cm_m< axomm. + mxam_._ + Nxmmw.maa + _xo_k.a_ - mam.N "oxoooa z _oooa :xmoo. + MXONO. + Nxmak.m + _xmm_. - moa.o uBo; ctoo :xmoo. + mxo_o. + Nxmmk.m + _x_mm. - __o._ "So; 660 "mco_pmzoo co_mmocmom mmm. NNJ. Nod. :mm. .mm. mm 4%4mmn. 4443mm. aka—om. *kwmm. 444m_©. m "muco_o_wmooU compo—occou o_o_u_3z A xv -4kwm. amN. .iomm. Nm_. ._am. aomao_oa 2-2moz+azzv 424mma. __N. 444mma. _mN. 4mmm. Amxv z-Amoz+aIzv _o_o_c_ JAN. amN. m_N. wm_. o_N. ANxV z _oooa _a_.- aoo.- _o_.- mmo.- kw_.- A_xv cootoo _ouOH CLOU poo CLOQ poo oxoooa z o_o_> amok __om muco_o_mmooo co_um_mccoo _m_ucmm mom>_mcm co_mmocmoc o_a_u_sE pcm co_uo_occoo _m_pcmo >3 Um_oo>oc mm comoLu_c __0m 40 >u___om__m>m mc_umE_umo coL mco_pmc_ELouop m:o_cm> bcm cLou pcm mumo >o oxmpa: z bcm o_o_> coosuon co_um_om .o_ o_omh 78 and N uptake by the second crop than by the first crop. The regression coefficients in the regression equations shown in Table 10 indicate that the effectiveness of the initial mineral N was double that of released mineral N in all cases. These results again suggest that mineral N present in the soil before cropping should be considered when estimating the nitrogen needs of the crop to be planted. SUMMARY AND CONCLUSION A dry combustion procedure of determining total C in soils was developed using a high frequency induction furnace equipped with a direct reading carbon analyzer (Laboratory Equipment Corporation, St. Joseph, Michigan). This instru- ment was designed to analyze C in steel and other metals. For soils, a procedure was developed which allowed precise determination of total C using a IOO-mg sample of mineral soil or a 20-mg sample of organic soil, ground to pass an 80 mesh screen. The C contents of a range of Michigan soils were determined using the dry combustion procedure and the results were compared with those obtained by Shaw's (1959) wet combustion method and Walkley and Black's chromic acid reduction method (1934). The three methods were highly corre- lated and no significant statistical difference was found between the methods. However, higher values were obtained with the dry combustion than with the two other methods. While as accurate as established procedures for deter- mining C in soils, the dry combustion method is quicker, and ideally suited for small samples. The method was found suitable for mineral and organic soils with widely varying 79 80 C contents. A disadvantage of the method, however, is that all forms of C in the soil are volatilized as C02. For cal- careous soils, therefore, the carbonates need to be either removed or determined separately. Organic C was highly correlated with total soil N. The linear regression of N on C accounted for 78% of the variation in total N in the soils investigated. The correla- tion between organic C and mineral N released during aerobic incubation, while significant, was low, since variations in C accounted for less than 20% of the variations in mineraliz- able N. Greenhouse experiments were conducted using 59 Michigan soils so that yields and N uptake could be related to diff- erent laboratory tests for N. A crop each of oats and corn were grown successively in the same pots. Organic C, total N, initial mineral N, and mineral N released during 14 days aerobic incubation of soil-sand mixtures were correlated with yields and N uptake. The simple linear correlation data suggest that measure— ment of mineral N (both initial and mineralizable levels) are more directly related to N uptake by plants than are measurements of total C and total N. Relatively low correlations with N uptake were obtained for organic C and total N contents of the soils. 81 Total N and organic C accounted for 35 and 17%, respec— tively, of the variations in N uptake by oats and corn. Although most of the N which becomes available to the growing crop is the result of the mineralization of a small part of soil organic N, the rate of mineralization is dependent upon many factors such as temperature aeration, moisture, and type of organic matter. It is, therefore, unlikely that determination of the amount of organic C and total N present at a given time will provide a precise index of the 5011's capacity to supply N to the crop. They may be useful, however, in a general way, as a means of separating soils on which relatively small responses to fertilizer N would be expected from those on which a large response might be expected. The carbon-nitrogen ratio was found to be of no value in predicting N availability. Total mineral N at the end of the bioassy incubation period (initial + released) was more highly correlated with yield and N uptake than was mineralizable N alone. Initial mineral N was probably important here because of its high accumulation prior to cropping, so that a large part of the N uptake by the first crop was derived from the initial mineral N. The total mineral N accounted for 44% of the variation in the total N uptake of the two crops. 82 Multiple correlations and regressions were calculated using organic C, total N, initial mineral N and mineral N released during aerobic incubation as independent variables and using yields and N uptake as dependent variables. The multiple correlation coefficients (R) showed a significantly higher correlation between the soils tests and dry matter yield and N uptake than was shown by simple linear correla- tions. The partial correlation coefficients were significant only for the initial and released mineral N in relationship with yield of the first crop and the total N uptake of both crops. The simple and multiple correlations both suggest that mineral N present in the soil before cropping should be considered when estimating N needs of the crop to be planted. The practical application of the results of the evalua- tion of the soil tests for availability of N in this study, should be determined by the ability of these 5011 tesB to predict crop response to N fertilizers under field conditions. The correlation of the laboratory tests with crop response in this study were based on greenhouse pot experiments where the effect of environmental factors on the mineralization of organic N were greatly minimized. The ultimate calibration, therefore, must be with crop responses in the field. BIBLIOGRAPHY Acharya, C. N. and S. F. Jain. 1954. Nitrifiability of soil organic matter. J. Indian Soc. Soil Sci. 2: 43-48. Alexander, Martin. 1961. Introduction to soil microbiology. John Wiley and Sons, Inc., New York, N. Y. 472 p. Allison, L. E. 1935. Organic carbon by reduction with chromic acid. Soil Sci. 40: 311-320. Allison, L. E. 1960. Wet combustion apparatus and procedure for organic and inorganic carbon in soil. Soil Sci. Soc. Amer. Proc. 24: 36-40. Allison, L. E. 1965. Organic carbon. in C. A. Black (ed.) Methods of soil analysis. 11. Chemical and micro- biological properties. Agronomy 9: 1367-1378. Allison, L. E., W. B. Bollen, and C. D. Moodie. 1965. Total carbon. in C. A. Black (ed.) Methods of soil analysis. 11. Chemical and microbiological properties. Agronomy 9: 1346—1366. Allison, F. E., J. H. Doetsch, and E. M. Roller. 1951. Ammonium fixation and availability in Harpster clay loam. Soil Sci. 72: 187-200. Allison, F. E., M. Kefauver, and E. M. Roller. 1953. Ammonium fixation in soils. Soil Sci. Soc. Amer. Proc. 17: 107-110. 84 Allison, F. E. and E. M. Roller. 1955. A comparison of leaching and distillation procedures for determining fixed ammonium in soils. Soil Sci. 80: 349-362. Allison, F. E. and L. D. Sterling. 1948. Nitrate formation from soil organic matter in relation to total nitrogen and cropping practices. Soil Sci. 67: 239-252. Anderson, J. U. and W. Harris. 1967. Determination of organic carbon and carbonated in soils. Soil Sci. Soc. Amer. Proc. 31: 341-343. Anderson, J. U. and W. Harris. 1967. Determination of organic carbon and carbonates in soils. Soil Sci. Soc. Amer. Proc. 31: 341-343. Arnold, P. W. 1954. Losses of nitrous oxide from soils. J. Soil Sci. 5: 116-128. Birch, H. F. 1960. Nitrification in soils after different periods of dryness. Plant and Soil 12: 81-86. Black, C. A. 1968. Soil—plant relationship. 2nd ed. John Wiley and Sons, Inc. New York, N.Y. Black, C. A., L. B. Nelson, and W. L. Pritchett. 1947. Nitrogen utilization by wheat as affected by rate of fertilization. Soil Sci. Soc. Amer. Proc. 11: 393-396. Bremner, J. M. 1949. Use of Van Slyke-Neil manometric apparatus for the determination of organic and inorganic carbon in soils and of organic carbon in soil extracts. Analyst 74: 492—498. 85 Bremner, J. M. 1955. Studies on soil humic acid. 1. The chemical nature of humic nitrogen. J. Agr. Sci. 46: 247-256. Bremner, J. M. 1965a. Total nitrogen. in C. A. Black (ed.) Methods of soil analysis. 11. Chemical and microbiological properties. Agronomy 9: 1149-1178. Bremner, J. M. l965b. Inorganic forms of nitrogen. lg C. A. Black (ed.) Methods of soil analysis. 11. Chemical and microbiological properties. Agronomy 9: 1179-1237. Bremner, J. M. l965c. Nitrogen availability indexes. in C. A. Black (ed.) Methods of soil analysis. 11. Chemical and microbiological properties. Agronomy 9: 1324—1325. Bremner, J. M. 1965d. Organic nitrogen in soils. 1Q W. V. Bartholomew and F. E. Clark (ed.) Soil nitrogen. Agronomy 10: 93-150. Bremner, J. M. 1967. Nitrogenous compounds. p. 19-66. lg A. D. McClaren and G. H. Peterson (ed) Soil biochemistry Marcel Dekker, Inc. New York, N.Y. Bremner, J. M. and D. S. Jenkinson. 1960. Determination of organic carbon in soil. 1. Oxidation by dichromate of organic matter in soil and plant materials. J. Soil Sci. 11: 394-402. Bremner, J. M. and K. Shaw. 1954. Studies on the estimation and decomposition of amino sugars in sofls. J. Agr. Sci. 44: 152-159. 86 Bremner, J. M. and K. Shaw. 1957. The mineralization of some nitrogenous materials in soil. J. Sci. Food Agr. 8: 341-347. Bremner, J. M. and K. Shaw. 1958. Denitrification in soils. 1. Methods of investigation. J. Agr. Sci. 51: 22-39. Broadbent, F. E. 1948. Nitrogen release and carbon loss from organic matter during decomposition of added plant residues. Soil Sci. Soc. Amer. Proc. 12: 246-249. Broadbent, F. E. 1951. Denitrification in some California soils. Soil Sci. 72: 129-137. Broadbent, F. E. 1953. The soil organic fraction. Adv. Agron. 5: 153-184. Broadbent, F. E. and Francis Clark. 1965. Denitrification. in W. V. Bartholomew and Francis E. Clark (ed.) Soil nitrogen. Agronomy 10: 357-362. Broadbent, F. E. and A. G. Norman. 1946. Some factors affecting the availability of the organic nitrogen in soil. A preliminary report. Soil Sci. Soc. Amer. Proc. 11: 264-267. Broadbent, F. E. and B. F. Stojanovic. 1952. The effect of partial pressure of oxygen on some soil nitrogen trans- formations. Soil Sci. Soc. Amer. Proc. 16: 359-363. Cady, Foster, B. 1960. Gaseous nitrogen products resulting from denitrification reactions in soils. Ph.D. Thesis. North Carolina State University, Raleigh, North Carolina. 87 Carolan, R. 1948. Modification of Graham's method for determining soil organic matter by colorimetric analysis. Soil Sci. 66: 241-247. Chesnin, Leon. 1950. Modification of the Smith-Weldon wet combustion procedure for the micro-determination of soil organic matter. Agron. J. 42: 385-386. Clark, N. A. and C. L. Ogg. 1942. A wet combustion method for determining total carbon in soils. Soil Sci. 53: 27-35. Cornfield, A. H. 1952. An apparatus for the macro-determina- tion of organic carbon in agricultural materials using the Van Slyke-Folch digestion mixtures. J. Sci. Food Agr. 3: 154-156. Cornfield, A. H. 1960. Ammonia released on treating soils with N sodium hydroxide as a possible means of predicting the N supplying power of soils. Nature 187: 260-261. Cunningham, R. K. 1962. Mineral N in tropical forest soils. J. Ag. Sci. 59: 257-262. Degtjareff, W. T. 1930. Determining soil organic matter by means of hydrogen peroxide and chromic acid. Soil Sci. 29: 239-245. Eagle, D. J. and B. C. Matthews. 1958. Measurements of nitrate-supplying power of soils by an incubation method and correlation with crop yield response. Can. J. Sci. 38: 161-170. 88 Ensminger, L. E. and J. E. Gieseking. 1939. The absorption of proteins by montmorillonitic clays. Soil Sci. 48: 467-474. Ensminger, L. E. and J. E. Gieseking. 1942. Resistance of clay-adsorbed proteins to proteolytic hydrolysis. Soil Sci. 53: 205-209. Ensminger, L. E. and R. W. Pearson. 1950. Soil nitrogen. Adv. Agron. 2: 8—111. Fitts, J. W., W. V. Bartholomew, and H. Heidel. 1955. Evaluation and control of factors in nitrate production and analysis. Soil Sci. Soc. Amer. Proc. 19: 69-73. Fraps, G. S. 1921. Relation of soil nitrogen, nitrification and ammonification to pot experiments. Texas Agr. Exp. Sta. Bul. 283: 51. Fraps, G. S. and Serges, A. J. 1932. Courses of low nitri- fication capacity of certain soils. Soil Sci. 34: 353. Friedman, Theodore E. and Arthur I. Kendall. 1929. The determination of carbon and carbon dioxide. J. Biol. Chem. 82: 45-55. Fuller, W. H. 1963. Reactions of nitrogenous fertilizers in calcareous soils. Agr. Food Chem. 11: 188—193. Gainey, P. L. 1936. Total nitrogen as a factor influencing nitrate accumulation in soils. Soil Sci. 42: 157-163. Gasser, J. K. R. 1961b. Soil nitrogen. VI. Correlation between laboratory measurement of soil mineral N and crop yields and responses in pot and field experiments. J. Sci. Food Agr. 12: 562-573. 89 Gasser, J. K. R. and R. J. B. Williams. 1963. Soil nitrogen. VII. Correlations between measurements of nitrogen status of soils and nitrogen % and nitrogen content of crop. J. Sci. Food Agr. 14: 269-277. Graham, E. R. 1948. Determination of soil organic matter by means of photoelectric colorimeter. Soil Sci. 65: 181-183. Hanway, John and Lloyd Dumenil. 1955. Predicting nitrogen fertilizer needs of Iowa soils. 111. Use of nitrate production together with other information as a basis for making nitrogen fertilizer recommendation for corn in Iowa. Soil Sci. Soc. Amer. Proc. 19: 77-80. Harmsen, G. W. and D. A. Van Schreven. 1955. Mineralization of organic matter in soil. Adv. Agron. 7: 299-398. Heck, A. F. 1929. A method for the determination of total carbon and also for estimation of carbon dioxide evolved from soils. Soil Sci. 28: 225—231. Jackson, M. L. 1952. Soil organic carbon determination with Fisher induction carbon apparatus. Soil Sci. Soc. Amer. Proc. 16: 370-371. Jackson, M. L. 1958. Soil chemical analysis. Prentice-Hall, Inc., Englewood Cliffs, N. J. 498.p. Jannson, S. L. and F. E. Clark. 1952. Losses of nitrogen during decomposition of plant material in the presence of inorganic nitrogen. Soil Sci. Soc. Amer. Proc. 16: 330-334. 9O Jenkinson, D. S. 1968. Chemical tests for potentially available nitrogen in soil. J. Sci. Food Agn 19: 160-168. Jones, E. J. 1951. Loss of elemental nitrogen from soils under anaerobic conditions. Soil Sci. 71: 193-196. Keeney, D. R. and J. M. Bremner. l966a. Chemical index of soil nitrogen availability. Nature 211: 892-893. Keeney, D. R. and J. M. Bremner. l966b. Comparison and evaluation of laboratory methods of obtaining an index of soil N availability. Agron. J. 58: 498—503. Keeney, D. R. and J. M. Bremner. 1966c. Characterization of mineralizable N in soils. Soil Sci. Soc. Amer. Proc. 30: 714-719. Kefauver, M. and F. E. Allison. 1957. Nitrite reduction by Bacterium denitrificans in relation to oxidation-reduction potential and oxygen tension. J. Bact. 73: 8-14. Legg, J. O. and F. E. Allison. 1959. Recovery of N15-tagged nitrogen from ammonium fixing soils. Soil Sci. Soc. Amer. Proc. 23: 131—134. Lunt, H. A. 1931. The carbon-organic matter factor in forest soil humus. Soil Sci. 32: 27-33. Martin, A. E. 1962. Nitrogen transformation in soil excluding denitrification. A review of nitrogen in the tropics with particular reference to pastures. A Symposium. Bul. 46: 72-82. 91 McCready, R. M. and W. T. Hassid. 1942. Semimicro-determina- tion of carbon using the Van Slyke—Folch oxidation mix- ture. Ind. Eng. Chem. Anal. Ed. 14: 525-526. Morris, G. T. and M. Schnitzer. 1967. Rapid determination of carbon in soil organic matter extract by dry combustion. Can. J. Soil Sci. 47: 143-144. Mortensen, J. L. and F. L. Himes. 1964. Soil organic matter. p. 206-291. 1_ F. E. Bear (ed.) Chemistry of the soil. 2nd ed. Reinhold Publishing Corp. New York, N.Y. Munson, R. D. and G. Stanford. 1955. Predicting nitrogen needs of Iowa soils. IV. Evaluation of nitrate production as a criterion of nitrogen availability. Soil Sci. Soc. Amer. Proc. 19: 464-468. Nommik, Hans. 1956. Investigations on denitrification in soil. Acta. Agric. Scand. 6: l95-228. Olson, R. A., M. W. Meyer, W. E. Lamke, A. D. Woltemath, and R. E. Weiss. 1960. Nitrate production as a soil test for estimating fertilizer nitrogen requirements of cereal crops. Int. Congr. Soil Sci. Trans. 7th (Madison, Wis.) 11:463-470. Pesek, J. 1964. Comparing efficiency of nitrogen fertilizer. p. 365-391. in V. Sauchelli (ed.) Fertilizer nitrogen. Reinhold Publishing Corp. New York, N.Y. 92 Pinck, L. A., F. E. Allison and V. L. Goddy. 1945. Green- house experiments on the effect of green manures upon nitrogen recovery and soil carbon content. Soil Sci. Soc. Amer. Proc. 10: 230-234. Ponomareva, V. V. and T. A. Plotnikova. 1967. Some data on the degree of intramolecular oxidation of humus of different soil type (the question of the conversion coefficient from carbon to humus). Pochvovedenie 7: 85-95 (R.e.) (Tsentral. Muz, Pochv. Dokuchaeva). (Original not available for examination. Abstracted in Soils and Fertilizer 30: 544). Pritchet, W. L., C. F. Eno and M. W. Malik. 1959. The nitrogen status of the mineral soils of Florida. Soil Sci. Soc. Amer. Proc. 23: 127-130. Purvis, E. R. and Geo. E. Higson, Jr. 1939. Determining organic carbon in soils. A modification of the chromic acid reduction method. Ind. Eng.Chem. Anal. ed. 11: 20-21. Ranney, R. W. 1969. An organic carbon-organic matter conversion equation for Pennyslvania surface soils. Soil Sci. Soc. Amer. Proc. 33: 809-811. Robinson, J. B. D. 1967. Anaerobic incubation of soil and the production of NH4. Nature 214: 534. Robinson, J. B. D. 1968. Incubation methods for determining available soil nitrogen. E. Afr. Agr. For. J. 33: 295-298. 93 Robinson, W. O. 1930. Method and procedure of soil analysis used in the division of soil chemistry and physics. U.S. Dept. Agr. Circular No. 139. Washington, D.C. Saunder, D. H., B. S. Ellis, and A. Hall. 1957. Estimation of available nitrogen for advisory purposes in Southern Rhodesia. J. Soil Sci. 8: 301-312. Schollenberger, C. J. 1927. A rapid method for determining soil organic matter. Soil Sci. 24: 65-68. Shaw, K. 1959. Determination of organic carbon in soil and plant material. J. Soil Sci. 10: 316-326. Shoemaker, H. E., E. O. McLean, and P. F. Pratt. 1961. Buffer methods for determining lime requirement of soils with appreciable amounts of extractable aluminum. Soil Sci. Soc. Amer. Proc. 25: 274-277. Skerman, V. B. D., B. J. Carey, and I. C. MacRae. 1958. The influence of oxygen on the reduction of nitrite by washed suspension of adapted cells of Achromobacter liguefacens. Can. J. Microbiol. 4: 243—256. Smith, J. A. 1966. An evaluation of nitrogen soil test for Ontario soils. Can. J. Soil Sci. 46: 185-194. Smith, H. W. and M. D. Weldon. 1940. A comparison of some methods for the determination of soil organic matter. Soil Sci. Soc. Amer. Proc. 5: 177-182. Stanford, George. 1968. Extractable organic nitrogen and nitrogen mineralization in soils. Soil Sci. 106: 345—351. 94 Stanford, George. 1969. Extraction of soil organic nitrogen by autoclaving in water. 2. A kinetic approach to estimating the NaOH-distillable fraction. Soil Sci. 107: 323-328. Stanford, George and W. H. Demar. 1969. Extraction of soil organic nitrogen by autoclaving in water. 1. The NaOH-distillable fraction as an index of nitrogen availability in soils. Soil Sci. 107: 203-205. Stanford, George and W. H. Demar. 1970. Extraction of soil organic nitrogen by autoclaving in water. 111. The diffusible component of NaOH—diffusible NH3, an index of soil nitrogen availability. In press. Stanford, George and John Hanway. 1955. Predicting nitrogen fertilizer needs of Iowa soils: II. A simplified technique for determining relative nitrate production in soils. Soil Soc. Amer. Proc. 19: 74-77. Steel, R. G. D. and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw-Hill Book Company, Inc. New York, N.Y. 481 p. Stevenson, F. J. 1954. Ion exchange chromatography of the amino acids in soil hydrolyzates. Soil Sci. Soc. Amer. Proc. 18: 373-377. Stevenson, F. J. 1957. Investigation of aminopolysaccharides in soils. 2. Distribution of hexosamine in some soil profiles. Soil Sci. 84: 99-106. 95 Stevenson, F. J. 1965. Soil nitrogen. lg V. Sanchelli (ed.) Fertilizer Nitrogen. Reinhold Publishing Corp. New York, N.Y. Stewart, B. A., L. K. Porter, and W. E. Beard. 1964. Deter- mination of total N and carbon in soils by a commercial Dumas apparatus. Soil Sci. Soc. Amer. Proc. 28: 366. Synghal, K. N., J. A. Toogood, and C. F. Bentley. 1959. Assessing nitrogen requirements of some Alberta soils. Can. J. Soil Sci. 39: 120-128. Timmons, D. R., V. J. Kilmer and V. F. Stahly. 1961. A possible source of serious error in the determination of nitrate production in soils. Soil Sci. Soc. Amer. Proc. 25: 328. Tinsley, J. 1950. The determination of organic carbon in soil by dichromate mixtures. Int. Congr. Soil Sci., Trans. 4th (Amsterdam) 1: 161—164. Van Slyke, D. D. and J. Folch. 1940. Manometric carbon determination. J. Biol. Chem. 136: 509-541. Viets, F. G., Jr. 1965. The plant's need for the use of nitrogen. in W. V. Bartholomew and Francis E. Clark (ed.). Soil Nitrogen. Agronomy 10: 503-549. Waksman, S. A. and K. R. N. lyer. 1933. Contribution to our knowledge of the chemical nature and origin of humus: III. The base-exchange capacity of ”synthesized humus“ (ligno—protein) and of ”natural humus“ complexes. Soil Sci. 36: 57-67. 96 Walkley, A. 1935. An examination of methods for determining organic carbon and nitrogen in soils. J. Agr. Sci. 25: 588-609. Walkley, A. 1947. A critical examination of a rapid method for determining organic carbon in soils-effect of variations in digestion conditions and of inorganic soil constituents. Soil Sci. 63: 251-264. Walkley, A. and I. A. Black. 1934. An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soil Sci. 37: 29-38. Waring, S. A. 1967. A rapid procedure for the estimation of nitrogen availability in soils. J. Aust. Inst. Agr. Sci. 33: 39. Waring, S. A. and J. M. Bremner. 1964. Ammonium production in soil under waterlogged conditions as an index of nitrogen availability. Nature, 201: 951-952. Wilde, S. A. 1943. Rapid colorimetric determination of soil organic matter. Soil Sci. Soc. Amer. Proc. 7: 393-394. Willcox, O. W. 1954. Quantitative agrobiology. l. The inverse-yield nitrogen law. Agron. J. 46: 315-320. Wilson, B. D. and E. V. Staker. 1932. Relation of organic matter to organic carbon in the peat soils of New York. J. Amer. Soc. Agron. 24: 477-481. 97 Young, J. L. and M. R. Lindbeck. 1964. Carbon determina- tion in soils and organic material with a high frequency induction furnace. Soil Sci. Soc. Amer. Proc. 28: 377-381.