S 651 .W6 I Copy 1 A BIOCHEMICAL STUDY OF NITROGEN IN CERTAIN LEGUMES BY ALBERT LEMUEL WHITING B. S. Massachusetts Agricultural College, 1908 M. S. Rhode Island State College, 1910 THESIS Submitted in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY IN AGRONOMY IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS 1912 A BIOCHEMICAL STUDY OF NITROGEN IN CERTAIN LEGUMES BY ALBERT LEMUEL WHITING B. S. Massachusetts Agricultural College, 1908 M. S. Rhode Island State College, 1910 THESIS Submitted in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY IN AGRONOMY IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS 1912 6"^ UNIVERSITY OF ILLINOIS Agricultural Experiment Station BULLETIN No. 179 A BIOCHEMICAL STUDY OF NITROGEN IN CERTAIN LEGUMES By albert L. WHITING URBANA, ILLINOIS, MARCH, 1915 Contents of Bulletin No. 179 PAGii INTRODUCTION 471 HISTORICAL 472 BIOLOGICAL 475 Infection 476 Inoculation as It Occurs Under Field Conditions 480 Growth of the Nodule 481 BACTERIOLOGICAL 482 Bacillus radicicola 483 Growth and Endurance of B. radicicola 484 Identity of B. radicicola 484 Enzyme Production by B. radicicola 485 Slime Production by B. radicicola 486 Isolation of B. radicicola 486 Dissemination of B. radicicola 486 Fixation of Nitrogen Without the Legume Plant 487 Bacteroids 487 THEORIES OF ASSIMILATION, FIXATION, AND IMMUNITY 488 Theories of Assimilation by the Plant 488 Theories Regarding the Chemical Phenomena of Fixation 489 Theories of Immunity 492 PRACTICAL CONSIDERATIONS WITH REGARD TO LEGUME FIXATION 493 Mutual Symbiosis 493 Amount of Nitrogen Fixed per Acre per Year 495 Value of Legumes as Nitrogen Retainers 496 Cross Inoculation 497 Associative Growth of Legumes and Non-Legumes 498 CHEMICAL 499 EXPERIMENTAL 503 Plan of Investigations 503 Part I: Studies to Determine Thru Which Organs Legumes Obtain Atmospheric Nitrogen 503 General Plan of Experiments 505 Experiment I 507 Experiment II 507 Experiment III 509 Experiment IV 514 Experiment by Kossowitsch 519 General Consideration of Gas Experiments 521 Practical Application of Results 522 PAGE Paet II: Eelative Percentages of Nitrogenous Compounds in the Various Parts of the Soybean and Cowpea at Definite Periods OF Growth 522 Methods Employed in the Growth and Preparation of Samples. .523 Analytical Methods 525 Discussion of Some of the Methods Used 527 Qualitative Tests 527 Series 100 (Soybeans) 528 Series 500 (Soybeans) 532 Series 700 (Soybeans) 533 Series 600 (Cowpeas) 536 Dis^cussion of Tables 537 CONCLUSIONS 541 Illustrations Plate page I.— Nodules of Robina Tyjie on Roots of Soybeans 477 II. — Nodules of Lupinus Type on Roots of Lupine Seedlings 478 III.— Nodules of Robina Type on (A) Red Clover; (B) Vetch; (C) Sweet Clover 479 IV. — Cowpea Seedlings in Preparation for Gas Experiments 504 V. — Experiment II: Cowpeas at Harvest (37 Days) 506 VI. — Experiment II: Plants Grown in Air and in COj-j-0 508 VII. — Experiment III: At Beginning and 10 Days Later 510 VIII. — Experiment III: 52 Days from Beginning and at Harvest (83 Days) 511 IX. — Experiment III: Roots from Plants Grown in CO. -|- O and in Air. .513 X. — Experiment IV: At Beginning and 16 Days Later 515 XL — Experiment IV: 41 and 59 Days from Beginning 516 XII.— Experiment IV : At Harvest (95 Days) 517 XIII. — Experiment IV : Roots from Plants Grown in CO2 -j- O and in N + COo + 518 XIV. — Typical Jar of Five Cowpeas Being Grown for Samples 523 XV. — Graph Showing Soluble and Insoluble Nitrogen in Series 100 531 XVI. — Graph Showing Soluble and Insoluble Nitrogen in Series 700 and 500.535 XA-'II. — Graph Showing Soluble and Insoluble Nitrogen in Series 600 53S Figure 1. — Root hair of common pea, showing infecting strand 476 2. — Root cell, showing infecting strand passing thru it and the formation of lamellae 476 3. — Young nodule magnified, showing affected root hair and same root hair more highly magnified 480 4. — Young nodule, showing the beginning of the differentiation of its tissues. .482 5. — B. radicieola, showing shajie and flagella 483 6. — Bacteroids, showing shape, and occurrence of vacuoles 488 A BIOCHEMICAL STUDY OF NITROGEN IN CERTAIN LEGUMES^ By ALBEET L. WHITING, Associate in Soil Biology INTRODUCTION The investigations considered in this publication bear on the biochemical nature of the element nitrogen, especially as concerns its fixation and assimilation thru the symbiotic relationship of Bacillus radicicola and certain members of the botanical family known as Leguminosae. The sources of the element nitrogen available for agricultural purposes are numerous. Of these the atmosphere is by far the most important and most extensive. Above each acre of the earth's surface there are about 69 million pounds of atmospheric nitrogen, and science has shown that by thoroly scientific systems of management this nitrogen may be appropriated for soil improvement at a minimum expense. By growing legumes, atmospheric nitrogen may be obtained at a low cost, often at no net cost, for most agricultural leguminous crops are worth growing for feed or seed alone. In commercial fer- tilizing materials, nitrogen costs from fifteen to twenty cents per pound, an amount from two to five times greater than that expended for any of the other essential elements of plant food. It is of passing interest to note how greatly disproportionate the cost values of these elements are to the relative supplies, when the nitrogen in the air is considered. The United States spends annually, abroad, over 32 million dol- lars in the purchase of combined nitrogen for use in various opera- tions, agricultural and otherwise.^ Of this amount I6I/2 million dol- lars are expended for the purchase of sodium nitrate, which is the ^Submitted to the Faculty of the Graduate School of the University of Illi- nois in partial fulfilment of the requirements for the degree of doctor of phi- losophy, June, 1912. Revised to date of issuance. ^Norton : Special Agent Series, Dept. of Commerce and Labor, Bur. of Manf r., No. 52, 9-11. 471 472 Bulletin No. 179 [March, most important commercial form of inorganic nitrogen. The present world supply of this salt is estimated at 454,576,200.000 pounds.^ How insufficient this supply is, when measured by crop require- ments, may be realized from the fact that the following nine important crops of the United States, — corn, wheat, oats, barley, rye, potatoes, hay, cotton, and tobacco, in the year 1910, required for their growth 11,500,000,000 pounds of nitrogen.^ If sodium nitrate were used for growing these crops at the rate stated above, the supply would be ex- hausted in about six years. On the other hand, the nitrogen above only one square mile, weighing 20 million tons, would be sufficient to supply what the entire world, at its present rate of consumption, would require for the next fifty years.^ The nitrogen above four acres would furnish more than the actual yearly consumption of commercial nitrogen in the entire United States. The wonderful possibilities presented by such an extensive source of plant food, and the fact that over 100 million dollars are invested in commercial fertilizers each year in the United States, a large part of which is wasted or uselessly applied, together with the great natural losses of nitrogen that occur, tend to emphasize greatly the need of a proper utilization of this unlimited reserve supply. Further, it is well recognized that the maintenance of the nitrogen supply is the greatest of our soil problems. Nitrogen cannot be purchased on the market at a price that will permit its extensive application in growing the im- portant crops of the United States. There is only one logical and in- exhaustible source of nitrogen for the world to utilize in the produc- tion of crops. That source is the atmosphere, from which nitrogen is most economically and easily secured as a result of the symbiotic relationship between B. radicicola and leguminous plants. HISTORICAL For several centuries certain plants of the Leguminosae have been used as soil improvers. A few of the more important references to their uses are considered here. In Roman literature, among the works of Columella,'* mention is made of the Roman farmers regarding beans as possessing the prop- erty of enriching the soil, and attention is also called to the practice of plowing under lupines. Alfalfa and vetches were observed to pro- duce similar results to those of lupines and beans. Like notations ^Eeview of Eeviews, April, 1910. ^Yields taken from U. S. Yearbook, 1910. For calculations see Hopkins' "Soil Fertility and Permanent Agriculture" (1910), 154; also 603-604. 'Norton: Special Agent Series, Dept. of Commerce and Labor, Bur. of Manfr., No. 52, 9-11. ^Marshall: Microbiology (1911), 273. 1915] A Biochemical Study of Nitrogen in Certain Legumes 473 may be found among the writings of Thaer and Walz.^ Gasparin^ constantly calls attention to the power of leguminous plants to add nitrogen to the soil. Jethro Tull-^ wrote concerning the efficiency of legumes in restoring depleted soils, mentioning especially sanfoin and alfalfa. It may be noted here that Hellriegel, who later was most promi- nent in the discovery of the relation existing between legumes and bacteria, wrote in 1863 as follows: "Clover plants may develop nor- mally and completely in mere sand to which the necessary mineral constituents of plant food have been added in assimilable forms, even when this soil contains no trace of any compound of nitrogen or of organic matter." Schultz-Lupitz^ in 1881 reported results that were of both chemi- cal and practical significance. After growing lupines for fifteen con- secutive times on a sandy soil, without the application of nitrogenous materials, he observed that the yields did not diminish ; and when he grew cereals on the same land after the lupines, he found that the yields of the grains were two and three times the yields where no lupines had been grown. Analyses of the soils at the end of this time showed that where the lupines had been grown, the nitrogen content of the surface six inches had increased by .06 percent. Frank^ veri- fied these results with twenty years of lupine culture on the same fields. About this time a great deal of interest centered on pot-culture experiments with legumes. Many physiologists and chemists worked on the problem of nitrogen collection by legumes. Prominent among these were the scientists Boussingault,^ and Lawes, Gilbert, and Pugh,'^ who, owing to their great accuracy, sacrificed the possibility of becom- ing the discoverers of this important relationship. In their great care, they destroyed the vital agency {B. radicicola) necessary for the ac- complishment of this symbiotic fixation. Later, in 1886, Hellriegel and his co-worker Wilfarth^ made the classical discovery that legumes obtain atmospheric nitrogen thru the association of microorganisms living in the nodules. In a preliminary report read before a section of scientists assembled on September 20, 1886, at Berlin, Hellriegel announced his findings ; and in a more com- ^Storer: Agriculture (1906), 2, 97. ^Ibid. ^Lipman: Bacteria in Relation to Country Life (1908), 206. 'Schultz-Lupitz: Landw. Jahrb. (1881), 10, 777. "Frank: Landw. Jahrb. (1888), 17, 501. "Boussingault: Ann. Sci. Agron. (1909), 26, Ser. 3, 4, 102-130. 'Lawes, Gilbert, and Pugh: Eothamsted Experiments (1905), 6-7. 'Hellriegel and Wilfarth: Tagblatt d. Naturforscher Versamml. z. Berlin (1886), 290. 474 Bulletin No. 179 [March, plete account rendered two years later, he made known to the world his researches. These are summarized as follows i^ 1. The legumes differ fundamentally from the grains in their nutrition with respect to nitrogen. 2. The grains (Gramineae) can satisfy their nitrogen need only by means of assimilable combinations existing in the soil, and their development is always in direct proportion to the provision of nitrogen which the soil places at their disposal. 3. Outside the nitrogen of the soil, the legumes have at their service a second source from which they can draw in most abundant manner all the nitrogen which their nutrition demands to complete that lack when the first source is in- sufficient. 4. That second source is the free nitrogen — the elemen- tary nitrogen of the atmosphere which is furnished to them. 5. The legumes do not possess by themselves the faculty of assimilating the free nitrogen from the air; it is ab- solutely necessary that the vital action of microorganisms of the soil come to their aid in order to attain this result. 6. In order that the nitrogen of the air can be made to serve the nutrition of the legumes, the sole presence of lower organisms in the soil is not sufficient; it is still necessary that certain among them enter into a symbiotic relationship with the plants. 7. The nodules^ of the roots must not be considered as simple reservoirs of albuminoid substances; their relation to the assimilation of free nitrogen is that of cause to effect. Schloesing and Laurent^ after growing legumes in a confined at- mosphere, gave out the following direct evidence of the fixation of at- mospheric nitrogen. Atmospheric nitrogen introduced into culture vessel 2681.2 ccm. Atmospheric nitrogen withdrawn 2653.1 ccm. Amount of nitrogen assimilated 28.1 ccm. (=36.5 mg.) Nitrogen in the soil and crop 73.2 mg. Nitrogen in the soil and seed 32.6 mg. Nitrogen assimilated 40.6 mg. *Hellriegel and Wilfarth: Beil. Zert. d. Verins. fiir die Eubenzucker In- dustrie, Berlin, Nov., 1888; or Lafar Handbuch der technischen Mykologie (1904- 06), 3, 31. "Nodules substituted for tubercles. 'Schloesing and Laurent: Compt. Rend. Acad. Sci. (1890), 111, 750: (1892) 115, 659, 732. 1915] A Biochemical Study of Nitrogen in Certain Legumes 475 In addition to the scientists mentioned above, Atwater and Woods, Berthelot, Miintz, Ville, Maze, Deherain, Frank, Hartig, Nobbe, Hilt- ner, Warrington, Hopkins, and many others have done much careful work in solving the problem and applying the truths discovered. BIOLOGICAL Nodules,^ which are the visible manifestations of infection, were observed upon the roots of legumes by Malphighi^ as early as 1687. The investigators of those times believed that the nodules were the re- sult of pathological processes, — that they were lumps, knobs, warts, and even galls. In 1853 the modern conception of the nodule as a nor- mal growth on the legume plant was established by Ij. C. Treviranus.^ Various theories have been proposed as to the function of these peculiar outgrowths, some advancing the idea that they were storage reservoirs or stimuli whereby the plants obtained nitrogen from the atmosphere thru their leaves. Recently Jost'* has called them "bac- terium galls," local hypertrophies not dissimilar to those sometimes caused by animal life. An astonishing conception has crept into the minds of the authors of certain general textbooks on bacteriology and plant physiology that nodules are abnormal growths and that their re- lationship to plants is either wholly or partially parasitic. It seems preferable, even to those familiar with the limitations of the theory, to describe this relationship as a normal condition and a true mutual symbiosis. That the formation of these nodules is due to external infection was definitely shown in 1887 by Marshall Ward,^ who was able to inocu- late the roots of young legumes by placing them in contact with old nodules. In Germany the first attempts to grow soybeans (Glycine Jmpida) in the botanical gardens resulted in failure, and it was not until soil from the natural habitat of that plant was imported for inoculation that soybeans were grown successfully.^ The history of the introduction of alfalfa culture in the states of Kansas and Illinois exemplifies in a large way this need of inoculation. From this experi- ence developed fhe soil-transfer method and tJie glue metliod'^ of inocu- lation, hotJi of ivliicJi are recognized today as superior to tlie use of so- colled commercial cultures. 'Nodules are recognized on the following non-leguminous plants: alders (Alnus glutinosa), silverberry (Eleagnus), sweet gale {Myrica Gale), sago palm, an evergreen, (Podocarpineae) , cycads (Cycacadeae), birthwort (Arisiolociaceae) . Nitrogen-fixing bacteria resembling B. radicicola have been found in the alder, silverberry, sweet gale, and. five varieties of podocarpus. =Malphighi: Op. (1687), 2, 126, Leiden. 'Treviranus: Bot. Ztg. (1853), 11, 393. .*Jost: Plant Physiology (Gibson 1907), 237. "Ward: Phil. Trans. Roy. Soc. London (1S87), 178, 139. 'Soil inoculation experiments were instituted as early as 1887 at the Moor Culture Experiment Station, Bremen, Germany. '111. Agr. Exp. Sta. Buls. 76, 94. 476 Bulletin No. 179 [March, Two types of nodules have been recognized by Tschirch;i Lupinus (lupine) represents one type and Robina (locust) the other. As may be seen by reference to Plates I and II, they differ in morphological appearance. The Lupinus tj^pe involves a swelling of the central root cylinders themselves, while in the Robina type only the epidermal and the endodermal tissues seem to enlarge. According to Tschirch, the nodules of lupines alone are of the first type, while those of all other legumes belong to the second. The figures in Plate III are sufficient to illustrate the most com- mon shapes of the Robina type. The shape varies with the different species of legumes, and to a certain extent with the individuals on the same legume plant. In the experimental work reported in this publi- cation, over twenty thousand nodules were examined closely, and it was not uncommon to find on the same plant notable variations due to external obstructions to growth. Infection The artificial inoculation of a plant is easily accomplished by con- tact. If the epidermis of the root is wounded and the infecting or- ganism {B. radicicola) brought into contact with the wound, nodules Fig. 1. — Boot hair of com- mon pea (Pisum sati- vum), showing infecting strand (x300) (After Prazmowski) Fig. 2. — Eoot cell, showing in- fecting strand passing thru it and the formation of lamellae (x650) (After Prazmowski) "Tschirch: Ber. deut. Bot. Gesell. (1887), 5, 58. 1915] A Biochemical Study of Nitrogen in Certain Legumes 477 Plate I,— Nodules op Eobina Type on Roots of Soybeans (Enlarged) 478 Bulletin No. 179 [March, Plate II.— Nodules of Lupinus Type on Boots of Lupine Seedlings (After A. Meyer) 1915] A Biochemical Study op Nitrogen in Certain Legumes 479 480 Bulletin No. 179 [March, result. Inoculation in pot cultures is attained by placing an infusion on the seed or in tlie medium. A similar method is successful with water cultures. Inoculation as It Occurs Under Field Conditions Studies of inoculation as it occurs in the field show the following generally accepted phenomena : As the tip of the root hair of the legume pushes itself out into the soil, it chances to come into intimate contact with the organism B. radicicola. Some scientists have exploited the view that the organ- ism is attracted to the plant by chemotaxis, believing that the plant excretes a substance, probably a carbohydrate, which diffuses into the soil solution and attracts the motile organism. While it has been rather definitely shown that this organism progresses in the soil at a rapid rate, nevertheless the number of root hairs in- fected^ is too small to lend support to a chemotactic theory. However the case may be, the organisms clus- ter at the tip of the hair and by means of an enzyme (or otherwise) rapidly dis- solve the cellulose of the cell wall, thus enabling the organism to enter the root hair. As a result, there is a decided bending of the tip, causing it to resemble a shepherd's crook. This was early observed as a sign of complete infection. It is claimed that other root hairs which form after infection are immune to the attack of other legumi- nous bacteria.2 The organisms, by rapid division and growth, ad- vance thru the center of Fig. 3. — Young nodule magnified, showing af- fected root hair and same root hair more highly magnified (After Atkinson) the infected root hair. Prazmowski^ found organisms in the cell ^Pierce, G. J.: Proc. Cal. Acad. Sci. II, No. 10 (1902), 295-328. Pierce found the proportion with bur clover to be 1 : 1000. 'Fred: Vir. Agr. Exp. Sta. Ann. Ept. 1909-10, 123-125. ^Prazmowski: Landw. Vers. Stat. (1890), 37, 160-238. 1915] A Biochemical Study of Nitrogen in Certain Legumes 481 sap and even in the epidermis only two days after inoculation. In this advance an infection strand (Infektion-schlauche) is formed, which consists of gelatinous material, and in the earlier stages of de- velopment this strand may be traced from the root hair into the inner tissue of the root and from cell to cell thruout the nodule. This infect- ing strand is not supposed to constitute a portion of the living tissue, nor is it a well-defined tube; but, as Fred has recently shown, it con- sists of a large number of zoogloea occurring adjacent to one another, in which separate bacteria can be distinguished. The infecting strand branches profusely, and it was this habit of growth which caused the early investigators to consider it the mycelium of a fungous growth. Growth of the Nodule The presence of B. radicicola in the tissues of the root causes a rapid cell division in the pericycle. These cells become larger and contain more protoplasm than the surrounding cells, and as growth takes place, the cortical parenchyma and epidermis are forced out- ward, thus forming a nodule. The growth of the nodule is apical. The various tissues common to the plant are present (see Fig. 4). In the central portion of the nodule is the so-called bacteroidal tissue, which is ochre, flesh, or gray in color, according to the age of the nodule, and in this portion the infecting strand (Infektion-schlauche) is distin- guished in the young nodule. It ramifies thruout the cells, causing those which it enters to lose their power of cell division but not of growth. Later, or in older nodules, the infecting strand is not visible, and the bacteroidal tissue loses its firmness. At the period when seed formation is at its height, most of the nodules are soft, and the inter- nal tissues slough off, leaving the more resistant epidermal tissue as a mere shell, which later decays. The endurance of the nodule depends upon several factors, — chiefly, however, upon the kind of legume plant on which it is produced and the need of nitrogen by that plant. Pierce^ considers the nodules as originating endogcnously from the same layer of cells as the lateral roots, and as being morphologi- cally similar to them ; however, as the lateral roots rupture the epider- mis, the above statement is not entirely in accord with what actually takes place. The nodules are largest and most numerous where aeration is best in the soil. In saturated soils they occur at the surface and are often found colored green, very similar to sunburned potatoes. Nodules form in solutions, and exceptionally well in certain nutrient solutions. Several interesting instances have been brought to the attention of the Experiment Station, in which the observers believed that the nodules ^Pierce, G. J.: Proc. Cal. Acad. Sci. 11, No. 10 (1902), 295-328. 482 Bulletin No. 179 [March, had grown above the ground. These peculiarities were undoubtedly caused by unobserved physical conditions occurring at the time of in- fection or afterward. Jr^iicfir^J sTr o-r<^' auTt ,0T rf-^ Tner'5' L^7-)clocJerrr)t<^ Fig. 4. — Young Nodule, Showing the Beginning of the Differentiation OF ITS Tissues (After Prazmowski) BACTERIOLOGICAL Minute bodies were first detected in nodules in 1866 by Woronin/ a Russian botanist. At that time bacteria were not recognized, and it was not until 1887 that they were demonstrated to be true bacteria by Wigand.^ 'Woronin: Bot. Ztg. (1866), 24, 329. 'Wigand: Bot. Heft. Forsch, a.d.Bot. Bart. 3 Marburg (1887), 288. 1915] A Biochemical Study of Nitrogen in Certain Legumes 483 Immediately afterward, Beyerinck^ isolated the organism on an artificial medium composed of a decoction of pea leaves, gelatine (7 percent), asparagine (.25 percent), and saccharose (.5 percent). He named the organism Bacillus radicicola, altho he described an organ- ism bearing a single polar flagellum. This organism became generally described as Pseudomonas radicicola, and some writers still prefer this designation. The organism has been known under a variety of terms, as ScJiinzia leguminosarum, CladocJiitrium tuberculorum, RJiizohium radicicola, RJiizohium leguminosarum. Bacterium radicicola. Micro- coccus tuherigenus, Myxobacteriaceae, Actinomyces, and Phytomoyxa. Such inappropriate names as RMzohacterium japonicum and Rliizo- hium sphaeroides are applied to certain special races. In commercial and general use, the organism is labeled as pea bacteria, bean bacteria, alfalfa bacteria, et cetera. Recent studies on the number of flagella possessed by this organ- ism have indicated that the organism is a bacillus, and it is there- fore desirable to adopt the original name B. radicicola, as proposed by Beyerinck. Bacillus radicicola These bacilli are rod-shaped organisms possessing numerous fla- gella (6 to 20), wldch are peritrichous. When full grown they vary in length from 1 to 4 or 5/*.^ It is not uncommon to find them from .5 to .6/A wide and from 2 to 3ju, long, and some have been found to measure only .18jLi wide and .9/^ long. The organism is actively motile. It is strongly aerobic, and in this connection Pierce calls attention to the intracellu- lar spaces in the root, which make it un- necessary to assume, as has been done, that it must live anaerobically. It is known that this organism does not form spores, but its means of enduring in the soil has not yet been determined. The bacilli prevail in the young nodule, while the branched forais, or bacteroids (see page 487), predominate in the older structure. B. radicicola grows well on a great va- riety of culture media, perhaps best on a medium of ash-maltose-agar or one of legume extract plus a sugar and dipotassium phosphate. Dextrose, saccharose, and maltose are suitable carbohydrates. The cultural characteristics of the colonies and the morphology of the organism will not be considered at this time, but it might be stated that many modifications occur on various media. 'Beyerinck: Bot. Ztg. (1888), 46, 725; also 741, 757, 780, 797. *M=Diieron, or 1/25,000 of an inch. Fig. 5. — B. radicicola, show- ing shape and flagella 484 Bulletin No. 179 [March, Growth and Endurance of B. radicicola The optimum temperature for B. radicicola varies between 18° and 26 °C. The thermal death point, according to Zipfel/ is 60° to 62° C. Growth is perceptible between 3° and 46° C. B. radicicola is not very sensitive to the reaction of the medium, which may be either acid or alkaline. Under field conditions, the organism exists in extremely acid soils, especially the race peculiar to legumes which thrive well on very acid soils. Experiments have dem- onstrated that the bacteria can withstand any degree of acidity or of alkalinity in the soil, that the particular legume itself can endure. That the organism endures at least two years in dry soil was de- termined by Ball.2 Harrison and Barlow^ found that the limit of viability on ash-maltose-agar varied somewhat, but that in the ma- jority of cases it was about two years. No doubt the organism will live much longer than this on artificial media when suitable conditions of growth are maintained. How long the clover or alfalfa organism will exist in a soil under field conditions is not yet known, but prac- tical observations indicate that it must be many years. The statement is quite generally made that B. radicicola becomes ' ' nitrogen hungry ' ' when cultivated thru several generations on nitro- gen-free media. This fact has not been sufficiently demonstrated to be accepted, for while Siichtung, Hiltner, and others have found that these organisms survive cultivation on nitrogen-free media for a year and at the end of that time possess the same ability to effect inocula- tion and nitrogen fixation in the legume as organisms obtained from fresh nodules, yet the bacteria had apparently made no appreciable gain in ability to effect inoculation. Garman and Didlake^ failed to find that nitrogen-free medium possessed any particular advantage over a legume-extract medium in causing the organism to become ' ' nitrogen hungry. ' ' Indentity op B. radicicola It is believed by some that the various legumes have different species of bacteria. Evidence has been produced which indicates that nodules do not contain but a single race of infecting organisms. Gnio de Eossi^ reported the finding, in artificial cultures, of two organisms which differed in that one formed a large hyaline colony, not developing well in beef and peptone gelatine, while the other ^Zipfel: Centbl. f. Bakt. 2 Abt. (1912), 32, 97-137. Five minutes taken as the time of exposure instead of ten minutes. "Ball: Centbl. f. Bakt. 2 Abt. (1909), 23, 50. ^Harrison and Barlow: Centbl. f. Bakt. 2 Abt. (1907), 19, 429. 'Garman and Didlake: Ky. Agr. Exp. Sta. Bui. 184, 352. "^Gino de Eossi: Centbl. f. Bakt. (1907), 18, 289-314, 418-489. 1915] A Biochemical Study op Nitrogen in Certain Legumes 485 formed white non-transparent colonies in beef gelatine. He believed that he had found another organism associated with B. radicicola. This work has not been sufficiently substantiated to be accepted as final. Greig-Smith^ reported having found three races of this or- ganism in the same nodule. Hiltner and Stormer^ classify nodule bacteria into two groups, RJiizohium radicicola and RJiizohium heyer- inckii. The former they associate with lupines, serradella, and soy- beans ; the latter with all other legumes. On the other hand, the results of many investigators^ (especially Laurent,^ who obtained nodules on the pea with organisms from thirty-six different legumes, and Nobbe et al.,^ who worked on the adaptability of nodule bacteria of unlike origin in different genera of Leguminosae), seem to support the theories of the identity of nodule organisms and the presence of only one race in the nodule. On the whole, present experimental evidence is slightly in favor of the view that there is only one species of this organism thruout the entire family of legumes.*^ This conception is not easily reconciled with field observations, for under natural conditions this organism has become so modified as to make it appear that there are many species. Contamination of the nodule has undoubtedly been responsi- ble for varying conclusions in this connection. Enzyme Production by B. radicicola Hiltner'^ reported the finding, by filtration thru porcelain, of a substance produced by B. radicicola which can dissolve the cell wall of root hairs. No proteolytic enzyme has as yet been reported. Further, Beyerinck claims that no enzyme has been found which at- tacks lime, starch, or cellulose, or which is capable of inverting sac- charose. In recent studies, Fred,^ altho unable to detect a proteolytic enzyme, obtained slight evidence of the presence of oxidases in the slime of various legume bacteria. These results suggest the need of further studies on enzj^me production by this organism. Similar to all microorganisms, it has the ability to reduce methylene blue to the colorless leuco-compound. ^Greig-Smith: Jour. Soc. Chem. Indus. (1902), 26, 304-306. ^Hiltner and Stormer: Arb. K. Gsndhtsamt, Biol. Abt. (1903), 3H. 3, 151. 'Harrison and Barlow: Centbl. 2 Abt. (1907), 19, 429. Kellerman: Centbl. f. Bakt. 2 Abt. (1912), 34, 45. Buchanan: Centbl. f. Bakt. 2 Abt. (1909), 22, 371. ^Laurent: Exp. Sta. Rec. (1890), 2, 186. "Nobbe et al.: Centbl. f. Bakt. 2 Abt. (1895), 1, 199. " " '' : Ibid (1900), 6, 449-457. 'Kellerman (see note 3) reported inoculation of soybean, lupine, and also of alfalfa from a culture originally isolated from alfalfa and kept on artificial media in the laboratory for six years. 'Hiltner: Centbl. f. Bakt. 2 Abt. (1900), 6, 273. •Fred: Vir. Agr. Exp. Sta. Ann. Rpt. 1910-11, 123. 486 Bulletin No. 179 [March, Slime Production by B. radicicola B. radicicola produces, on artificial media, a gum, or slime, which is partly soluble and party exists as a zoogloeal mass. The organisms, on suitable media, have been observed to surround themselves with definite capsules several times thicker than themselves. These cap- sules are rather distinct at first b\it later form a gelatinous mass. Greig-Smith^ and Maze,^ who studied this slime, claimed for it a nitro- genous substance. The results obtained by Buchanan, Gage, and Fred agree and are in direct refutation of the above. This is impor- tant to bear in mind in connection with the theories later to be dis- cussed. Gum is not formed from a carbohydrate containing less than five carbon atoms. Isolation of B, radicicola from Soil Until recently B. radicicola had never been very successfully isolated from a normal soil except by means of a legume plant. Gage,^ after a long, tedious process, obtained from soil an organism which was capable of producing nodules on red clover and which appeared to be identical with B. radicicola. Still more recently Greig-Smith^ has reported the isolation of this organism from soil, but his work has not been substantiated by others. Maze early attempted the isolation of B. radicicola from sterilized and non-sterilized soils to which pure cultures had been added. He isolated the organism from the soil which had been sterilized before the addition of the culture, but he was unable to recover it from the unsterilized soil. Kellerman and Leonard^ isolated (on the agar rec- ommended by Greig-Smith) an organism which inoculated alfalfa from soil that had been sterilized and subsequently inoculated with living organisms of B. radicicola. Lipman and Fowler*^ were able to isolate the organism peculiar to vetch {Vicia sicida) on soil-ex- tract agar and proved out the organism. They attained success in about 40 percent of the cases, judging from the condensed report recently published. Dissemination op B. radicicola Those familiar with pot-culture experiments and inoculation ex- periments with legumes easily understand that legume bacteria are disseminated in many ways. In fact, sterile conditions are difficult to maintain. The layman, however, may wonder how legume bacteria 'Greig-Smith: Centbl. f. Bakt. 2 Abt. (1911), 30, 552-556. =Maze: Ann. Inst. Pasteur (1898), 12, 128. 'Gage: Centbl. f. Bakt. 2 Abt. (1910), 27, 7-48. *Greig-Smith: Centbl. f. Bakt. 2 Abt. (1912), 34, 227-229. 'Kellerman and Leonard: Science (1913), 38, 95-98. 'Lipman and Fowler: Science (1915), 41, No. 1050, 256-258. 1915] A Biochemical Study of Nitrogen in Certain Legumes 487 not common to a certain locality creep in. A few agents concerned in the transfer of the organisms are here cited. The seeds themselves are a common means of distribution. The ruptured seed coats offer an opportunity for the various kinds of bac- teria to accompany the seeds in a most persistent manner. Sometimes wind is responsible for the dissemination of these bacteria. An inter- esting instance is cited by Ball of Texas: a wind storm blew off the roof of the culture-house in which legumes were being grown under sterile conditions, and as a consequence the various plants under ob- servation became inoculated. Water has been known to aid in inocu- lating large areas during washing and floods. The transfer of un- cleaned seed is sure to result in the conveyance of some inoculating or- ganisms in the impurities accompanying the seed. Cultivation, espe- cially harrowing, is also responsible for the spread of the organisms. The addition to soils of legume residues from either fields or stables is still another common means of dissemination. Fixation of Nitrogen Without the Legume Plant Maze,i in 1897, demonstrated that nodule bacteria have the power to assimilate atmospheric nitrogen in the absence of a legume. His researches have been verified by others, altho the amounts of nitrogen obtained in his experiments have never been equaled. Recently con- ducted experiments^ on this question show that as an average, in liquid and in solid media, about 1.2 milligrams of nitrogen are fixed per 100 cc. of medium. A fixation in the absence of the legume plant has been found in sterile sand and in soil. How important this kind of non-symbiotic fixation is, has yet to be more fully determined. At the present time it is generally recognized as insignificant compared with symbiotic fixation in the nodules of legumes. Bacteroids Bacteroids are believed to be a form which appears in the devel- opment of B. radicicola. The cell activities of this form are main- tained in a way similar to that in which the cell activities of the rod- shaped form are maintained. Stefan^ states that these bacteroids are thin-walled and capable of division when young, but that when older they become swollen and finally degenerate. Fred was able to observe the changes which occur in the organism in passing from the bacillus to the extreme vacuolized bacteroidal form. The organism at first apparently thickens at one end and then branches into the bacteroid, which is characterized by rounded outgrowths, kno^vn as vacuoles. These vacuoles appear at a definite period of growth and evidently 'Maze: Ann. Inst. Pasteur (1897), 11, 44. 'Fred: Vir. Agr. Exp. Sta. Ann. Ept. 1909-10, 138-142. 'Stefan: Centbl. f. Bakt. 2 Abt. (1906), 16, 131-149. 488 Bulletin No. 179 [March, are not a sign of polymorphism, but are a further development of the bacteroid. (They require special staining to be made visible.) ^ 0^ Bacteroids occur in the nodule as <^ ,g^ esM <^ ^r" ^^ ^^ ^^ culture media. Their " morphology varies according to the ^ ^ rm^ ^^ constituents of the culture media. Aj Some writers prefer to call these ir- tjj % <^ P P ^ Cl regular organisms degenerate or in- ^^^ ^ volution forms. They were first ob- Fig 6.— Bacteroids, showing ^^^^^^ -^^ artificial media in 1888 by shape, and occurrence or • ^ . ^ ^ ^ t-,i vacuoles Eeyermck,! and have been studied by Hiltner,2 Stutzer,^ Buchanan,'* Fred,^"" and others. A medium rich 'in carbohydrates or the glucosides of amygdalin or salicin offers very favorable conditions for bacteroid formation. Of fifteen carbohydrates tested, mannite has proved par- .ticularly suited to their development. Glycerine is better than most nutrients, while the salts of organic acids have been found unsuitable. On careful observation the following factors have been found to exercise no influence upon bacteroid formation: temperature, light, osmotic pressure, decreased oxygen pressure, reaction of medium, ni- trogen-hunger, specific formative materials in the legumes, and the ac- cumulation of metabolic products. From the above observation it is evident that nutrition is a strong factor in bacteroid formation. It is claimed that each of the various legumes exhibits a different shaped bacteroid which is characteristic of that legame. In studies conducted at the Virginia Experiment Station the bacteroids pos- sessed by the Egyptian, the crimson, and the red clover were found to be very similar, while those of the vetch differed somewhat. More extended research regarding the appearance of bacteroids as connected with the beginning of nitrogen assimilation, is sorely needed. THEORIES OF ASSIMILATION, FIXATION, AND IMMUNITY Theories of Assimilation by the Plant In brief it may be said that the two main suppositions regarding assimilation are as follows: (1) that the bacteroids are bodily absorbed by the plant fluids; and (2) that the bacteroids, by some sort of change, produce the substance containing the assimilable nitrogen which the plant utilizes. The theory that the plant absorbs these bacteroids has been chal- lenged by some on the evidence which Nobbe and Hiltner*' produced ^Beyerinck: Bot. Ztg. (1888), 46, 725. ^'Hiltner: Centbl. f. Bakt. 2 Abt. (1900), 6, 273. 'Stutzer: Ibid (1901), 7, 897. 'Buchanan: Ibid. (1909), 23, 59-91. "Fred: Vir. Exp. Sta. Ann. Rpt. 1909-10, 128-198. "Nobbe and Hiltner: Centbl. f. Bakt. 2 Abt. (1900), 6, 449. 1915] A Biochemical Study of Nitrogen in Certain Legumes 489 to show that a plant had fixed 1 gram of nitrogen while its nodules weighed only .3 gram. The relationship between the amount of nitro- gen fixed and the weight of the nodules is no criterion, however, for criticism of such a theory, inasmuch as this relationship is not at all definite but varies according to the development and needs of the plant. The failure to establish the presence of a proteolytic enzyme has also been responsible for no little criticism of the first supposition. While the second supposition seems more plausible, it must be ad- mitted that it, too, is only a theory which should be studied with the hope of isolating and identifying the diffusible substance. In connec- tion with this theory Golding^ conducted some very interesting experi- ments on the removal of the products of growth in the assimilation of nitrogen by legume bacteria. He reasoned that the plant played an important role in the removal of the products produced by bacteria in the nodule aside from the mere furnishing of suitable food. In his experiments he used a porous Chamberland filter-candle placed in a culture vessel to serve to imitate natural conditions. Aerobic condi- tions were obtained by passing purified air thru the cultures. The parts of the plants used in some of his experiments were sterilized in order to avoid the possibility of plant enzyme action. As a result of his method of experimentation, he obtained a much greater fixation of nitrogen than other experimenters had found, and the logical conclu- sion arose that the plant performs a function in the assimilation of nitrogen which is construed to be the removal of soluble products of growth. The results of Golding's most extensive experiment are embodied in the table below : N. in grams 500.0 grams of Stems and Leaves 2.86.3 20.2 grams of Eoots and Nodules (quite frei h) 094 3000.0 cc. Ammonia-free Distilled water 000 Total Nitrogen to start with 2.959 2870.0 cc. Filtrates and Drainings 731 566.2 grams of Wet Res^idue 2.570 To'al Nitrogen after experiment 3.301 Total Gain of Nitrogen during experiment 342 Theories Regarding the Chemical Phenomena of Fixation As yet, purely chemical theories of fixation arc entirely hypo- thetical ; however, they deserve consideration, for even theories un- supported by facts may have a value in stimulating thought and as- sisting in the development of more rational views. "Frank,2 a prominent Frenchman, was among the first to attempt •Gelding: Jour. Agr. Sci. (1905), 1, 59-04. Trank: l>andw. Jalirl). (1888), 17, 504-518; 19, 564. 490 Bulletin No. 179 [March, an explanation of how the plant actually obtains nitrogen. He be- lieved that it came in thru the leaves, and even recently some isolated statements hold to this idea. His view was allied with the conception of stimulation which many held ; namely, that the organisms on the root stimulated the plant to fix nitrogen in its leaves. Stocklasa,^ as a result of his chemical investigations, also believed that assimilation took place thru the leaves, — that amides were first formed, and that these, migrating to the nodules, reacted with glucose and produced protein, which served as the nutrient medium for the bacteria. In this connection he advanced the idea that the bacteria produced an enzyme which enabled the plant to effect this fixation. Loew and Aso^ in 1908 suggested that ammonium nitrite was the first compound produced, the nitrous acid being readily reduced to ammonia. Little evidence has been ol)tained to suj)port this theory ; no evidence whatever has been found under controlled conditions. Gautier and Drouin^ suggested that the nitrogen is oxidized to nitric and nitrous acid. Winogradsky^ has advanced the idea that the free nitrogen in the plasma of the organism may unite with nascent hydrogen and form ammonia, which by oxidation would become assimi- lable. In connection with the two latter theories, it should be empha- sized that the presence of nitrites, nitrates, or ammonia in the nodules, roots, or tops of legumes, inoculated or uninoculated, when grown in the entire absence of combined nitrogen, has not been established. Gerlach and VogeP have investigated non-symbiotic nitrogen fixation and have arrived at the conclusion that there is a direct union of free nitrogen with some organic compound inside the bacterial cell. Heinze*' thinks it probable that nitrogen is at once brought into com- bination with a hydrocarbon (glycogen), and suggests that a salt of carbamic acid maj^ be formed first or that carbamic acid may be pro- duced from cyanamid. There is yet another most extraordinary theory which, owing to its somewhat recent notoriety, it seems appropriate to consider. This theory is most properly called the Jamicson theory. The rather pecu- liar views embodied in it are perhaps quite well explained in an article published in TJie Spokesman Review, Spokane, Washington, March 28, 1913. Tlie Review is a bi-weekly paper devoted to agricultural inter- ests. ^Stocklasa: Laiidw. Jalirb. (1895), 24, 827-863. =Loew and Aso: Bui. Col. Agr. Tokyo Imp. Univ. (1908), 7, 507. ^Gautier and Drouin: Bui. See. Chim. Paris 78, 84-97. 'Winogradsky: Centbl. f. Bakt. 2 Abt. (1901), 7, 842. 'Gerlach and Vogel: Centbl. f. Bakt. 2 Abt. (1902), 9, 817-821, 881-892; (1903) 10, 636-644. ''Heinze: Landw. Jahrb. (1906), 35, 907. 1!)15] A Biochemical Study op Nitrogen in Certain Legumes 491 "Do Plants Directly Absorb Free Nitrogen from the Air? ' ' Scotch Scientist is at Odds with the Common Belief "The doctrine that plants directly absorb free nitrogen from the air con- flicts with the earlier beliefs. It is held that plants, with the exception of legumes, cannot utilize the nitrogen of the air, the explanation in the eaee of legumes being that by the aid of nitrogen in organisms on their roots the?e plants utilize atmospheric nitrogen. ' ' Thomas Jamieson, Director of the Agi-icultural Eesearch Association of Scotland, takes issue with this belief. Mr. Jamieson has made a life study of the problems of plant nutrition. An abstract of his views is given in the New Zealand Journal of Agriculture. Mr. Jamieson proceeds to show: ' ' 1. That the legume-tubercular theory is untenable. ' ' 2. That the nitrogen of the air is directly used. "3. That the application of this knowledge is valuable to the agricul- turalist **********. ' ' Mr. Jamieson disagrees with the theory that the tubercle formation on leguminous plants is a normal growth containing a net structure of plant food thru the union of fungus and a legume. He says : " '1 regard the tubercles as abnormal growths. I hold that no "symbiotic" action takes place ; that the fungus is not a fixer of the nitrogen ; that the legume plant is itself the fixer, and that it fends its manufactured albuminous products to heal up the wound or to counteract the drain of the parasitic fungus ; and that the tubercle has nothing to do with the fixation of the nitrogen of the air ****************' "As to~ the explanation of tubercles of leguminous plants Mr. Jamieson says: " 'The plant being attacked by the fungus, a wound is made, the fluid of the plant courses to repair it, and not only is the leguminous fluid of the plant rich in nitrogen, but its most nitrogenous fluid, albumen, is just a plastic material like the white of an egg, especially suited to heal the wound and to form a sac round the invader. " 'There is nothing exceptional in the bearing of tubercles by the legume. The nodules, or tubercles, are well displayed. The legume is a plant specially sought by fungus demanding nitrogen. It is provided with a means of supplying the element, hence it is specially attacked. ' "Further investigations lead Mr. Jamieson to the conclusion that nature pro- vides special means for all plants to absorb nitrogen. Even the hardest leaves are soft in the earlier stage. The cultivated members of the legume family have broad, soft leaves studded with apertures, supposed to serve for exhalation. It is accepted that the green cells or the chlorophyl contained by these cells de- compose the carbonic-acid gas. Cannot a similar action extend to nitrogen? **#»****«»»****^ "The effect on the soil of producing certain crops, as cereals and grasses, is to reduce the available plant foods. Of these, in their simple forms, the most important supplied by fertilizers are nitrogen, phosphorus, and potash, and of these the farmer can avoid the expense of the purchase of nitrogenous manures by the adoption of a rotation to include those plants that are rich in nitrogen. This is not new. Legumes have been availed of from the earliest recorded time as preparatory to cereals. What is new is that there is a wider field of plants for selection and the farmer knows why these jdants enrich the soil for the ni- trogen-demanding cereals. The plants, among others mentioned by Mr. Jamieson, are rape, mustard, and turnips." The above theory has received more contradiction and less sup- port than the others reviewed. ^ 'Henry: Ann. Sci. Agron. (1909), 26, 102-130. Vageler: Centbl. f. Bakt. 2 Abt. <1909), 22, 452. Kovessi: Compt. Eend. Acad. Sci. (1909), 149, 56. Kny: Ber. deut. Bot. Gesell. (1909), 27, 532. Mameli and Pollacci: Ann. Sci. Agron. (1914), 31, 141. 492 Bulletin No. 179 [March, Three possible chemical processes in fixation have been consid- ered by scientists: 1. Keduction 2. Oxidation 3. Direct union into an organic compound The first two possible processes have received no chemical verification, while the third is supported only by data which are of an eliminative character. More data of a similar nature will be found in Part II of the experimental section of this bulletin. It is interesting to note that opinion seems to be strengthening in support of the theory of the direct union of nitrogen gas into organic combination, in spite of the fact that such a combination is unknown in chemistry to take place at ordinary temperatures. It is possible, however, to unite nitrogen into organic combination at at- mospheric pressure, altho a high temperature is required. Theories op Immunity Reasoning from animal life, it seems logical for one to believe that the relative strength of a legume plant or of B. radicicola may vary under certain conditions so that the plant will resist the entrance of the organism. Inoculation experiments have produced data show- ing that B. radicicola causes a certain resistance on the part of the plant, making it necessary in some cases to employ organisms of greater efficiency in order to produce inoculation. Hiltner^ has given the six following conditions as instances in which immunity demonstrates itself. 1. The organisms cannot get into the plant. 2. The organisms gain admission into the plant but do not pro- duce nodules because the plant, by its greater resistance, absorbs the bacteria. 3. The organisms enter the plant and produce nodules, but no fixation of nitrogen occurs. 4. The organisms enter, produce nodules, and nitrogen is fixed and assimilated by the plant. 5. The organisms are so efficient in comparison with the plant that the latter is injured. 6. The organisms are parasitic and the plant is actually killed. In the pursuance of the investigations reported in this bulletin no indication of the existence of any of these conditions, except No. 4, was observed, and in no instance under normal conditions did in- oculation fail to produce nodules and cause a fixation and an assimi- lation of nitrogen. ^Lafar: Handbuch der technischen Mykologie (1904-6), 3, 45. 1915] A Biochemical Study of Nitrogen in Certain Legumes 493 Various other similar theories of resistance have been proposed, all of which are permeated with the idea of natural or acquired immunity. Prominent among these might be mentioned Siichtung's theory of equilibrium.! Siichtung assumed that the bacteria produced a toxin and the plant an antitoxin, and that the degree of equilibrium deter- mined the extent of nodule formation, the plant becoming immunized by an antibody and not by a substance produced by the bacteria; further, that the nitrogen supply in the plant was regulated by the production of this antibody. It is conceivable that in some of the cases observed the apparent immunity may have been due to a weakness on the part of the bacteria rather than to resistance by the plant. Siich- tung's equilibrium theory considers varying conditions of virulence on the part of the bacteria and varying degrees of resisting ability by the plants. His theory was advanced as the result of carefully conducted experiments which showed that there were variations in virulence between organisms of the same kind when grown upon artificial media and when obtained from fresh nodules. The inoculation of legumes in solution is inhibited by potassium nitrate, tho a convincing explanation of this inhibition has not yet been offered. Some experimenters believe that the immunity of a plant is strengthened by its nitrogen nutrition ; others hold that bacteria find another source of nitrogen nutrition in the nitrate and hence do not seek the plant. It has been recently shown, however, that the organism will produce nodules after the concentration of the nitrate has been reduced by the plant, which would tend to show that the immunity of a plant is not strengthened and that the organism is not permanently injured by a solution of potassium nitrate suit- able for the plant. Inoculation under field conditions is no doubt inhibited by physi- cal and antagonistic biological factors, which have been considered only briefly by most investigators. The subject of immunity in plants has been given little attention thus far, but the increasing number of bacterial diseases in the plant kingdom will undoubtedly lead to research in this direction. It is well known that bacterial diseases of plants are the most difficult to control ; the need of investigation in this unexplored field of immunity as an aid in their control is imperative. PRACTICAL CONSIDERATIONS WITH REGARD TO LEGUME FIXATION Mutual Symbiosis -Mutual symbiosis may be defined as the contiguous association of two or more morphologically distinct organisms not of the same kind, 'Suchtung: Centbl. f. Bakt. 2 Abt. (1904), 11, 377. 494 Bulletin No. 179 [March, resulting in an acquisition of assimilated food substances. It implies that the organisms concerned have the power of independent exist- ence, but that both are benefited by the close association. The relationship existing between B. radicicola and legumes is one of mutual sjonbiosis. The facts which bear out this belief are too convincing to need explanation. However, some prefer to call the re- lationship a truly parasitic condition, while others consider it to be parasitic in the beginning and later a true mutual symbiosis. This latter conception would seem to be plausible, yet no exact data have been produced to show that a parasitic condition exists at any stage. ^ The result of this mutual symbiosis is wonderfully characteristic of nature as well as astounding when one considers the corresponding chemical process, in which the energy expended is so apparent and the temperature required so high.^ The energy values in the symbiotic fixation of nitrogen by B. radicicola and legumes have never been de- termined. "When B. radicicola and Azotohacter are grown under simi- lar conditions, apart from their respective hosts,^ less organic carbon per unit of nitrogen fixed is oxidized by B. radicicola than by Azoto- hacter. Present knowledge indicates that a very great amount of energy is necessary for the fixation of atmospheric nitrogen by Azoto- hacter. Table 1 presents the amounts of some of the common materials that must undergo rather complete oxidation in order to furnish suf- ficient energy for the addition of fifteen pounds of atmospheric nitro- gen to the surface soil of an acre by Azotohacier. The figures show that in the case of dextrose 66% times as much organic matter is re- Table 1. — Amounts of Materials Necessary for the Fixation of Fifteen Pounds of Atmospheric Nitrogen per Acre by Azotobacter Kind of material Dextrose (sugar) . . Fresh clover tops. Fresli lupine tops. Wheat straw Corn stover Oak leaves Pounds required 1 000^ 1 212 2 000 4 300 5 500 11 500 'This figure represents the minimum amount of dextrose consumed per unit of nitrogen fixed; in other words, 1 gram of dextrose (yielding 3,750 calories) is necessary for the fixation of 15 milligrams of atmospheric nitrogen by Azoto- hacter. ^Experiments are now in progress at this station with the view of obtaining data on this question. "The most recent figures show that 1 kilowatt hour yields 70 gi-ams of nitrogen in the form of cyanamid ; in other words, 1.35 horse-power yields 70 grams, or 8.74 horse-power per hour yields 1 pound of nitrogen. *Algae are understood as the host for Azotohacter. The word host as used in this publication is not intended to convey the idea of parasitism. 1915] A Biochemical Study of Nitrogen in Certain Legumes 495 quired as there is nitrogen fixed, and in the case of oak leaves, 766 times as much. In the oxidation of such large amounts of organic carbon it is easily seen that the volume of organic matter in the soil is greatly reduced. It has been definitely shown that Azotohacter lives in a symbiotic relationship with algae. It is also well known that our normal soils possess an abundant algal flora. In view of these two facts it may possibly be found that nitrogen is accumulated by Azotohacter with- out the above reduction in the volume of the organic matter of the soil. Whatever the future may disclose, the only fact that now re- mains to be pointed out is that in the symbiosis between legumes and B. radicicola, instead of there being a decrease in the organic matter of a soil, a material increase is bound to result. While fixation pro- gresses, organic matter is being manufactured by the plant, which is later returned to the soil. This process is, then, a constructive one, as compared with the destructive non-symbiotic fixation. Amount op Nitrogen Fixed per Acre per Year The amount of nitrogen added to a soil depends in part upon the relative supply of that element in the soluble and decomposable forms, organic as well as inorganic. The poorer the soil the greater the amount of nitrogen that will be fixed, tho in a rich soil in which the nitrogen is not in an available form, large amounts may be fixed. For instance, altho a peat soil contains in one acre some thirty thousand pounds of nitrogen in the surf",ce million pounds (0 to 6% inches), yet because of a lack of proper organisms in the soil to decompose the organic matter, which resists natural decay and there- fore does- not readily furnish nitrogen to plants, a legume crop may add large amounts of that element. Where nitrates are present in large amounts, they are taken up by legumes; but where they are present in only small amounts, as is the case during dry seasons even on the common prairie corn-belt land, atmospheric nitrogen is fixed by legumes. The fixation varies with the seasonal conditions, a hot, moist season being best suited to the summer legumes. Among other factors the kind of legume and the duration of its growing period affect the amount of nitrogen added. The annual legumes must necessarily fix nitrogen much faster than the biennial or perennial legumes. The yield does not necessarily in- dicate the amount of fixation, as some legumes which yield much less hay and seed than others may have a greater total nitrogen content. The most reliable data which now exist indicate that two-thirds of the nitrogen in legumes grown on soils of normal productive power is obtained from the air.^ These figures, contributed by the Illinois 'Hopkins: 111. Agr. Exp. Sta. Buls. 76 and 94. 496 Bulletin No. 179 [March, Experiment Station, were obtained by analyzing inoculated and unin- oculated legumes from like areas of normal soils, and as a result of pot experiments. Computed by these data, a 3-ton crop of cowpea hay adds 86 pounds of nitrogen per acre, a 25-bushel crop of soybeans with 21/4 tons of straw adds 106 pounds, a 4-ton clover crop adds 106 pounds, and a 4-ton alfalfa crop adds 132 pounds. A nitrogen gain of 200 pounds per acre has been reported by the New Jersey Experiment Station^ with crimson clover. At the Rhode Island Experiment Station," as a result of a pot-culture experiment, it was found that nitrogen had been added at the rate of 400 pounds per acre per year. This experiment extended over five years, and legumes were grown both in the summer and in the winter. The tops of the summer legumes (cowpeas and soybeans) were removed from the soil, while the winter legume (vetch) v»^as turned back into the soil. It should be noted, however, that an acre of this soil to a depth of 6% inches contained only a little over three thousand pounds of nitrogen. Moreover, in this experiment optimum conditions were established, and no losses were possible from drainage, — which factors would tend to make these results much higher than would be obtained under field conditions. In considering soil enrichment by clover, ten years' results of a field experiment at the Experimental Farms, Ottawa, Canada,^ are important. In this experiment a light, sandy loam with a sandy sub- soil was planted to clover continuously, being reseeded every two years. The clover was cut and left to decay on the land. In ten years the nitrogen content of this soil was doubled. The yearly gain of nitrogen was fifty pounds per acre. It was found that from two to three times that amount was added, but that all but fifty pounds was dissipated by bacterial activities and in other natural ways. Analyses of the clover crop also brought out the fact previously men- tioned that the amount of nitrogen fixed is influenced in part by the season. Value of Legumes as Nitrogen Retainers Legumes have a very great value aside from their role in the nitrogen-fixing process. It is well known that they require more nitrogen for their growth than other ordinary farm crops, and that they therefore contain more of it per ton. It seems very appropriate, therefore, to select legumes for such purposes as holding soil from 'N. J. Agr. Exp. Sta. Ept. 1894, 158. -R. I. Agr. Exp. Sta. Bui. 152. ^Experiiueutal Farms, Ottawa, Ept. 1912, 145. 1015] A Biochemical Study of Nitrogen in Certain Legumes 497 washing and preventing sands from shifting, for not only do they serve these purposes well, but at the same time they conserve rela- tively more nitrates from loss than non-legumes. Of course a con- dition might occur in which a legume would draw all its nitrogen from the soil, but even in such a case a legume would be preferable to a non-legume, as by its use relatively more nitrogen would be kept from leaching and so saved for future crops. Cross-Inoculation There are relatively few cases of cross-inoculation that have been definitely determined as occurring under natural conditions. The most important example is the cross-inoculation that takes place be- tween the sweet clovers and alfalfa. Bur clover (Medico go lupu- lina) is another source of inoculation for alfalfa. The wild vetches serve for inoculation of the cultivated vetches. It w^ould seem that many such cases may exist in which the wild specie of a legume con- tains the organism for the inoculation of the cultivated legume of the same specie or even of an entirely different specie or genera. Inves- tigations along this line have not been carefully undertaken as yet. It has been possible under laboratory conditions to cross-inocu- late in many different ways. The data furnished by Laurent, re- ferred to in an earlier part of this publication (page 485), together with that furnished l)y Moore, Kellerman and Leonard, and others, is of interest. Laurent produced nodules on the pea with the organ- isms from thirty-six different species of legumes. Moore^ produced nodules on many legumes with the pea organism, among which were crimson clover (Trifolium incarnatinn), white clover {Trifolium re- pens), red clover {Trifolium. pratense), berseem {Trifolium Alexan- drinum), alsike {Trifolium. lijihridium) , sweet clover {Melilotus alba), cowpea {Vigna catjang), alfalfa {Medicago .sofwa), broad bean {Vicia faba), common bean {PJiaseolus vidgnris), fenugreek {Trifolium foe- num graecum), hairy vetch {Vicia villosa) , scarlet vetch {Vicia ful- geus), and yellow vetch {Vicia lutea). The results published by Kellerman and Leonard represent the extreme in cross-inoculation at the present time. It will be recalled that they have reported the inoculation of the soybean, the lupine, and alfalfa with an organism originally obtained from alfalfa nodules,^ altho it has been quite generally believed that the soybean was representative of a special class as regards its inoculation, and the same can be said regarding the lupine. Legumes may be grouped as follows according as their ^Moore: U. S. Dept. Agr. Bur. Plant Indus. Bui. 71. ^See page 485, note 6. 498 Bulletin No. 179 [March, bacteria are interchangeable for the purposes of inoculation: Group 1 Alfalfa, sweet clovers, bur clover, black niedick ' ' 2 All true clovers ' ' 3 Cowpea, partridge pea ' ' 4 Soybean " 5 Bean " 6 Peas (garden and field), vetches (cultivated and wild), sweet peas, lentils ' ' 7 Lupines ' ' 8 Sanfoin ' ' 9 Locust Nobbe, Hiltner, and SchmicP obtained inoculation and nitrogen fixation with the locust and vetch cross-inoculated with each other and with pea bacteria, as shown below : Inoculated with Locust Vetch Pea-bacteria Nitrogen Locust (Eobina) 232.1 13.5 21.2 assimilated Vetch (Vicia) 12.9 264.0 22.6 Some prefer to divide the legume bacteria into two classes accord- ing to the beneficial or detrimental effect produced by lime upon the legume. In this classification, alfalfa would represent one type and serradella the other. The question of cross-inoculation is far from settled. It is easily seen that a great many interesting problems, aside from the purely scientific studies of the laboratory, are presented for the soil biologist in the pursuance of this field of research. Associative Growth of Legumes and Non-Legumes The associative growth of legumes and non-legumes has been given renewed notoriety in recent publications. Practical observa- tions of long standing have indicated that a non-legume benefits by the presence of a legume during the second year of its growth — as might reasonably be anticipated. The proof of a benefit by association dur- ing the first season is not sufficiently established for a generalization, for errors in sampling, in methods of experimentation, and other un- favorable conditions have crept in and overshadowed the full value of the data reported. The problem of associative growth involves many details that must be further studied. The stimulation caused by the struggle for existence in association may increase the height of the crops or the amount of the organic matter produced, yet not necessarily the nitro- gen content. In the work under observation at this station, it appears that the nitrogen Avhich is returned to the soil as the nodule sloughs off: could hardly be utilized by an ordinary annual non-legume crop. It is yet to be determined whether either the legume itself or its ^Nobbe, Hiltner, and Schmid: Landw. Vers. Stat. (1894), 45, 12. 1915] A Biochemical Study of Nitrogen in Certain Legumes 499 nodules exude nitrogenous compounds during their active period of growth. CHEMICAL The chemical composition of legumes from the standpoint of their nitrogenous constituents has been investigated to some extent, but the studies closely related to this point are relatively few. The follow- ing data are very general in character and relate to studies concern- ing the total nitrogen content of the different parts of legumes at dif- ferent periods of growth. Studies upon some of the various nitroge- nous compounds are also included. In 1895 Stocklasa/ working with lupines {Lupinus luteus and Lupinus augusfifolius), found that the nodules were richest in the element nitrogen at the time of blooming, while the roots appeared to be richest in that element at the fruiting period. His results are given in Table 2. The figures for the nodules indicate that the nitro- gen is either taken up by the plant for seed production or diffused into the soil. Table 2. — Total Nitrogen in Lupinus Luteus: BY Stocklasa Results Obtained (Percentage on dry basis) Period Roots Nodules Blooming Fruiting Maturity 1.64 1.84 1.42 5.22 2.61 1.73 Stocklasa also determined protein, amides, and aspai'agine in lu- pine nodules. The protein was obtained by the Statzer method, the amides by the Kjeldahl method, and the asparagine by calculation from the ammonia obtained by distillation with magnesium oxid. Table 3 shows his results. Table 3. — Nitrogen Compounds in Lupine Nodules: BY Stocklasa Results Obtained (Percentage on dry basis) Period Protein Amides Asparagine Blossoming 3.99 1.54 .35 .15 .34 Maturity Trace The presence of asparagine in the nodule is important, as it is thought to be intimately related with the formation of protein. In 1901 Wassilieff- studied the nitrogen compounds in white lupine {Lupinus alba) seeds and seedlings. He found that the seeds contained 7.G8 percent of total nitrogen; and that of this, 6.89 percent was in the form of protein and .53 percent was precipitated by phos- •Stocklasa: Landw. Jahrb. (1895), 24, 827-863. ^'Wassilieff: Landw. Vers. Stat. (1901), 55, 45-77. 500 Bulletin No. 179 [March, photungstic acid, leaving a difference of .26 percent, asparagine. The occurrence of asparagine in large amounts in the seedlings is shown b}' the data given in Table 4. Table 4. — Nitrogen Compounds in Pourteen-Day-Old Green Seedlings of White Lupines: Kesults Obtained by Wassiliefp (Expressed in percentage on dry basis) Parts P. T. A.^ nitrogen Asparagine Protein Total nitrogen Leaves .53 .63 .42 .46 1.45 3.83 4.57 2.20 4.11 2.44 1.56 1.87 6.57 Cotyledons Stems 7.83 6.77 Roots 5.40 ^P.T.A. : This abbreviation for phosphotimgstic acid will be used thrnout this publication. Wassilieff also demonstrated the presence of leucine and tyrosine in the cotyledons of one-week-old seedlings of white lupines. These and other amino acids would be expected to be present when the pro- tein of the seed is breaking down for the nutrition of the seedling. Knisely^ analyzed the leaves, pods, stems, roots, and nodules of lupine plants for total nitrogen at three distinct periods of develop- ment. His results show better than the others presented where the nitrogen accumulates as the plant matures. Table 5. — Total Nitrogen in Lupines: Results Obtained by Knisely (Expressed in percentage on dry basis) Period Leaves Pods 3.07 3.38 3.68 Stems Roots Nodules Full bloom Pods well formed Pods very large 4.02 3.70 3.41 1.15 .88 .90 .92 .83 .66 5.17 4.29 3 70 Schulze and Barbieri- examined lupine and soybeans seeds and seedlings for nitrogen and obtained the results shown ' in Table 6. Table 6. — Nitrogen in Lupine and Soybean Seeds and Seedlings: Obtained by Schulze and Barbieri Results (Express ed in percentage on dry basis) Material Total nitrogen Protein P.T.A. nitrogen .24 .13 1.60 2.17 .56 Filtrates from P.T.A. Lupine seeds 8.63 6.73 10.64 10.51 7.42 8.17 0.32 3.40 2.33 3.86 2^ Soybeans 28 Lupine dark seedlings 11 to 12 days old Lupine dark seedlings 12 days old 5.64 6 01 Soybean seedlings 15 days old 3.00 'Knisely: Ore. Agr. Exp. Sta. Rpt. 1909, 30-31. ^Schulze and Barbieri: Landw. Vers. Stat. (1881), 26, 241. 1915] A Biochemical Study of Nitrogen in Certain Legumes 501 They also found a large amount of asparagine in both the lupine and the soybean seedlings. Schulze^ has made a careful study of the compounds in plants, and has formulated the hypothesis that the same decomposition products arise from protein in the plant as outside it, but that in the plant the compounds are further altered, thereby affecting in varying degree the individual products of the hydrolytic decomposi- tion. A comparison of the analyses of pea seedlings one week old and those three weeks old showed the following dilt'erences : 1 week. 3 weeks. Leucine Tyrosine Arginine Asparagine abundant little present absent much less absent almost absent very abundant Arginine and amido acids w^ere shown to be present in the lupine cotyledons, but asparagine was absent, altho the latter substance was found in the stem of the seedling. It has been suggested that the oc- currence of asparagine is associated with the disappearance of amido acids and not of protein. Phenyl alanine, tyrosine, and tryptophane have been reported in the white lupine {Lupinus alha), tyrosine and tryptophane in vetch {Vicia sativa), and tryptophane in the garden pea {Pisuni sativum).- Smith and Robinson^ found 4.19 percent of nitrogen in soybean nodules and 3.90 percent in cowpea nodules. They observed that inocu- hition increased the protein content of soybean plants without in- creasing the yield of beans. This has been noted by other experi- menters. Hopkins^ has reported the analyses of cowpea plants for total nitrogen with and without inoculation. The nodules, roots, and tops were analyzed separately, as will be seen by reference to Table 7. Table 7.- — Nitrogen Fixation by Cowpeas: Kesults Obtained by Hopkins (Expressed in egs.) Treatment Tops Eoots Nodules Nitrogen fixed 146 38 171 55 143 40 9 3 10 4 8 4 11 18 17 125 Ten plants without bacteria 140 124 ^Schulze: Zeits. f. Physiol. Chem. (1895), 24, 18; 30, 241. =Schulze et al: Zeits. f. Physiol. Chem. (1887), 11, 43; (1906), 48, 387, 396; (1910), 65, 431. Gorup Besamez: Ber. deut. Chem. Gesell. (1887), 10, 781. 'Smith and Robinson: Mich. Agr. Exp. Sta. Bui. 224, 125-132. -Hopkins: 111. Agr. Exp. Sta. Bui. 94, 319. 502 Bulletin No. 179 [March, The inoculated plants contained a much greater percentage of nitrogen than the uninoculated, the average content of the inoculated being 4.24 percent in the tops, 1.48 percent in the roots, and 5.92 per- cent in the nodules, while the average content of the uninoculated was 2.48 percent in the tops and .88 percent in the roots. The ash and the ash constituents of the nodules and the roots of lupines have been determined by Stocklasa,^ as presented in Table 8. The total ash of the nodules was found to be 6.32 percent, while that of the roots was found to be 4.55 percent. Table 8. — Ash Constituents in Lupine Nodules and Roots; Results Obtained by Stocklasa (Expressed in percentage) Constituents Nodules Roots Si 1.59 4.90 (5.51 17.31 16.94 7.41 7.64 .83 1.90 S P K Na Me 0.38 4.28 12.05 19.94 7.05 Ca Fe 12.04 .75 The analyses of red-clover nodules show a potassium content of 2.63 percent in the dry matter.^ The nodules, therefore, are rela- tively rich in mineral elements as well as nitrogen compounds ; and Stocklasa 's results (see Table 8) show that the chief differences between the roots and the nodules in the composition of the ash constituents are in phosphorus, potassium, calcium, and sodium. The nodules are richer in the first two elements and the roots in the latter two. The differences in nitrogen content of the various parts of the plant have already been brought out somewhat, but they will be dealt with more fully in the results presented under the experimental portion of this bulletin. The presence of the bacteria would in itself be sufficient to account for these differences. In brief, the chemical data which have been considered, altho small in amount, show the relative richness in nitrogen of the nodule as compared with other parts of the plant. They point to the accumu- lation of nitrogen in the seeds, at the expense of the other parts, as the plant matures. That the nitrogen exists in the form of protein, asparagine, and other soluble forms, is also clear. The presence of various aliphatic and earbocyclic amino acids has been mentioned. ^Stocklasa: Landw. Jahrb. (1895), 24, 827-863. -Analyzed by Aumer, 111. Agr. Exp. Sta. (unpublished data). 1915] A Biochemical Study of Nitrogen in Certain Legimes 503 EXPERIMENTAL Plan of Investigations The experimental studies herein reported arc fur convenience di- vided into two parts. Part I consists of studies made in order to de- termine thru which oi-gans legumes obtain their nitrogen from the air. Part II is concerned with an attempt to determine more definitely the mechanism of the reactions occurring in the fixation and assimilation of atmospheric nitrogen by B. 7'adicicola and legumes, a process concern- ing which science is greatly in the dark. This phase of the problem has attracted the attention of plant physiologists, physiological chemists, and other scientists outside the field of agricultural research. No les- ser chemist than Emil Abderhalden^ has written concerning it as fol- lows: "It would be very interesting to Ivuow the compounds into which these organisms convert the nitrogen. At i^resent we have no knowledge of this. We assume that the final substance produced is protein, which is then in part assimilated by the plants with the help of fermentation." Any light which may be throwji on this question will be of great value toward its final solution. PART I STUDIES TO DETERMINE THRU WHICH ORGAN LEGUMES OBTAIN ATMOSPHERIC NITROGEN For a long time it was believed that the nitrogen fixed by legume bacteria and assimilated by the plant was obtained thru the leaves, and even now many hold to this belief. Frank and Otto- in 1890 obtained analytical results which seemed to them to be proof of this theory. They believed that the bacteria were only incidentally connected with the process, acting perhaps as stimuli. The first experiment resulting in data of a contradictory nature was made by KossoAvitsch^ in 1891, but the results of this investigation were not generally accepted. Nobbe and Hiltner^ in 1899 added fur- ther evidence to the existing knowledge, but their conclusions, drawn from physiological differences, have not been substantiated by chemi- cal data, which seem more reliable than those of a physiological nature. ^Abdeihalden : f*hysiological Chemistry, Trans, by Hall, 198. -Frank and Otto: Ber. deut. Bot. Gesell. (1890), 8, 331. •'Kossowitsf'h: Bot. Ztg. (1S92), 50, 697-702, 71.'5-72:?, 729-73S, 745-755, 771-774. 'Nobbe atul Hiltiu'r: Landw. Vers. 8tat. (1S99), 52, 455-4(i5. 504 Bulletin No. 179 [March, O a 02 1 Ph 1915] A Biochemical Study op Nitrogen in Certain Legumes 505 Experiments on this question were conducted Ijy the author in 1911-1912. The general plan Avas the same thruout each experiment; the various modifications are considered under the individual experi- ments. General Plan of Experiments The plants used were the soybean and the cowpea. Uniform seeds were carefully selected and inoculated with an infusion placed directly in contact with them. They were then planted in beakers containing nitrogen-free white sand. Mineral plant food was added in solution. When the seedlings had developed two leaves and possessed small nod- ules, they w^ere carefully washed from the sand and transferred to the apparatus. The apparatus^ was arranged as follows : Woulf e bottles, placed inside battery jars painted black in order to obviate the influence of light, were connected with drier bottles, which in turn were connected with a gasometer. An outlet tube from each bottle was provided, the external end of which was immersed in water. In the first ex- periment two Woulfe bottles were used and in the others six. Sterile nitrogen-free sand containing calcium carbonate was placed in the Woulfe bottles and the young seedlings carefully transplanted, one to each. The plants were then sealed gas-tight by means of rubber tissue placed double thick about the stem. Rubber cement Avas also used to make all joints tight. Plant food, with the exception of nitrogen and calcium, was added in solution. This solution was sterilized, boiled, and cooled just pre- vious to its being used in order to prevent the addition of absorbed gases. The plant-food solutions and sterile, distilled, nitrogen-free water were added from the outer end of the outlet tube, with the gas flowing in order to avoid the possible admittance of air. The moisture content of the sand was maintained at about 12 percent. When the apparatus had been made tight, the gas was started and allowed to flow gently for eight to ten hours per day; at night it was entirely shut off. By this method the plant roots Avere kept constantly in the same atmosphere. The gas mixture used in the first three experiments consisted of 96 to 98 percent oxygen and 2 to 4 percent carbon dioxid. For the purpose of comparison, air was passed thru part of the bottles in these experiments. The gas mixture Avas made in the laboratory, great care being exercised to eliminate nitrogen, air, and other impurities. The oxygen was made from potassium chlorate and manganese dioxid, and the carbon dioxid Avas generated from marble and hydrochloric acid. The air, Avhen used as a source of nitrogen, Avas passed thru sulfuric acid before entering the gasometer and after leaAang it. In order to 'Plate V shows the apparatus in use. 506 Bulletin No. 179 [ March, 1915] A Biochemical Study op Nitrogen in Certain Legumes 50' dispel any possible doubt as to the oxygen mixture being too strong, a fourth experiment was conducted in which the effect of a mixture made of 90 percent oxygen, 7 percent nitrogen, and 3 percent carbon dioxid was compared with that of a mixture made of 97 percent oxygen and 3 percent carbon dioxid. Experiment I In Experiment I, soybeans were used. Three plants twenty-one days old were placed in position on September 1, 1911, one in each of two Woulfe bottles and a check plant left uninclosed. Thruout the experiment these plants were kept out of doors during the day. The experiment was continued for twenty-eight days. At the end of that time the plants were analyzed for total nitrogen by the official Gun- ningi method. The average of individual analyses of twenty soybean seeds was taken as the criterion from which to calculate the amount of nitrogen fixed by the plants. The results are presented in Table 9. Table 9. — Fixation^ op Nitrogen by Soybeans: Experiment I (Eesults expressed in milligrams) Plant No. Treatment Nitrogen in plant at end, 28 days Nitrogen in check seeds Nitrogen fixed 1 2 3 GOj + CO2 + O Air 10.43 10.65 17.61 11.4 11.4 11.4 (-.97) (-.75) 7.07 ^The word fixation is used in this publication in its broader sense and should be understood as meaning the fixation of atmospheric nitrogen by bacteria and the assimilation of the nitrogenous compounds formed by the plant. The error in Plants 1 and 2 is partially accounted for by a slight injury to these plants by grasshoppers and red ants. There is, how- ever, a small experimental error which is difficult to eliminate, as will be observed in the other experiments. Experiment II The experience gained in Experiment I led to the selection of cowpeas for the later investigations, since they are less subject to injury by red ants than are soybeans. Experiment II was started on November 23, 11)11, and continued until December 29, thirty-seven days. Six two-liter Woulfe bottles were planted with seedlings twenty-four days old. Air was passed thru three of the bottles and the gas mixture thru the other three. The average of individual analyses of 'In preliminary tests the Gunning and Kjeldahl methods modified to include nitrates gave no higher results than the official Gunning or Kjeldahl methods. 508 Bulletin No. 179 [March, Plate VI. — Experiment II: Plants Above Grown in Air; Those Below Grown in COa -f 1915] A Biochemical Study of Nitrogen in Certain Legumes 509 fifteen eowpea seedlings seventeen days old was used as a basis from which to calculate the nitrogen fixed by the plants during the experi- ment. Seedlings of this age were taken for analysis in order that the results of this experiment might be comparable with those of the others, altho the seedlings of the experiment when transplanted were somewhat older. Table 10. — Fixation of Nitrogen by Cowpeas: Experiment II (Results expressed in milligrams) Plant No. Treatment Nitrogen in plant at end, 37 days Nitrogen in check seedlings at beginning Nitrogen fixed 1 2 3 4 5 CO2 + O C0„ + COo + O Air Air 9.21 13.03 9.43 24.84 23.61 7.90 7.90 7.90 7.90 7.90 1.31 5.13 1.53 16.94 15.71 Note. — Plant 6 was lost in distilling thru the breaking of the flask, caused by sand adhering to the roots. The fixation shown by Plant 2 is attributed to a leak discovered around the stem of this plant some few weeks after it had been put in place ; trouble was had thruout the experiment in keeping it gas-tight. The evident fixation in the case of Plants 1 and 3 is within experi- mental error; yet since these plants were twenty-four days old when placed in the apparatus, while the check seedlings analyzed were only seventeen days old, it is reasonable to assume, from the results obtained in the next experiment, that a part at least of the assimilation of this nitrogen had taken place before the seedlings were transferred. On the plants receiving air the nodules became well developed. The accompanying photograph (Plate VI), taken at the termination of the experiment, shows the comparative development of the roots and the tops grown in the gas mixture and those grown in the air. The most interesting part of this experiment was the very evident translocation exhibited by the plants growing in the mixture of carbon dioxid and oxygen, as shown by their color. The same phenomenon was observed in the later experiments and is discussed under the general consideration of the gas experiments. Experiment III Experiment III was conducted with cowpeas in a manner similar to that of the preceding experiments. It was started on March 5, 1912, with six seedlings seventeen days old and discontinued after eighty- three days, May 27, 1912. Air was passed thru three of the bottles and the gas mixture thru the other three. In order to obtain the best possible check on the results, fifteen additional seedlings of the same 510 Bulletin No. 179 [March, Plate VII. — Experiment III: Lower Figure Showing Experiment at Begin- ning; Upper Figure Showing Experiment 10 Days Later 1915] A Biochemical Study of Nitrogen in Certain Legumes 511 .^ Plate VIII.— Experiment III: Upper Figure Showing Experiment 52 Days FROM Beginning; Lower Figure Showing Experiment at Harvest (83 Days) 512 Bulletin No. 179 [March, lot as those transplanted to the Woulfe bottles, grown from seeds 185 milligrams in weight, were analyzed individually at the beginning of the experiment. The results showed the presence of an average of 7.90 milligrams of nitrogen, while the average nitrogen content of twenty seeds of the same weight analyzed individually equaled 6.94 milli- gi'ams, making an average fixation of .96 milligram of nitrogen by these seedlings in the first seventeen days. Table 11. — Fixation op Nitrogen by Cowpeas: Experiment III (Eesults expressed in milligrams) Plant No. Treatment Nitrogen in plant at end, 83 days Nitrogen in check seedlings at beginning Nitrogen fixed 1 CO, 4-0 9.48 7.90 1..58 C0, + 7.49 7.90 (-.41) 3 CO, + 8.49 7.90 .59 4 Air Eoots 74.27 Tops 112..59 18(3.8(5 7.90 177.96 5 Air Roots 71.99 Tops 166.03 238.02 7.90 230.12 6 Air Eoots 66.71 Tops 120.51 187.22 7.90 179.32 The figures in Table 11 show to what extent fixation took place. Plants 1 and 3 may have contained more than 7.90 milligrams of nitro- gen as seedlings, altho it cannot be proved that they did, owing to the impossibility of analyzing and growing the same seedling. There was . always another possible source of error in the dissolved nitrogen gas in the water used for pressure in the gasometers. It is well to observe that in all these experimoits the gases were passed thru sulfuric acid, which eliminated the possibility of ammonia playing any part in the fixation. This is claimed by many to occur ; yet the first experiment ever made for the purpose of showing that legumes obtain nitrogen from the air was so conducted that combined nitrogen was eliminated. The plants in the carbon dioxid and oxygen mixture were from 3 to 4 inches in height and possessed two leaves at the end of the experi- ment, while those growing in the air measured from 8 to 9 inches in height and possessed nine leaves. 1915] A Biochemical Study of Nitkogen in Certain Legumes 513 Plate IX.— Experiment III: On the Left, Roots from Plant Grown in CO2-I-O; On the Right, Roots from Plant Grown in Axe 514 Bulletin No. 179 [March, In order to test the viability of B. rndicicola after it had grown on the plant under extreme oxygen conditions, organisms were re- moved from the nodules of Plants 1, 2, and 3, and an infusion made in sterile water. Portions of this infusion were applied to cowpca seeds that had been sterilized and planted in sterile sand. Sterile conditions were maintained thruout this test. Profuse nodule formation resulted, demonstrating that no harmful results had been produced upon the organism by its long exposure to an atmosphere with a high content of oxygen. Experiment IV Having made certain in Experiment III that no detrimental ef- fects had been produced upon B. radicicola by long exposure to an atmosphere high in oxygen, Experiment IV was instituted in order to determine if there could have been any possibility of injury to the plants in the previous experiments from the use of gaseous mixtures high in oxygen. The plan involved a comparison of the effect of a mixture of 97 percent oxygen and 3 percent carbon dioxid, and that of a mixture of 90 percent oxygen, 7 percent nitrogen, and 3 percent carbon dioxid. The nitrogen used was obtained from the air; otherwise this experi- ment was similar to Experiment III. Each gas mixture was passed thru three of the Woulfe bottles. The experiment was begun on Sep- Table 12. — Fixation of Nitrogen by Cowpeas: Experiment IV (Kesults expressed in milligrams) Plant No. Treatment Nitrogen in plant Nitrogen in check seedlings at beginning First Harvest (26 Days) Nitrogen fixed C0, + N 4- CO, + O 10.00 14.94 7.90 7.90 2.10 7.04 Second Harvest (28 Days) 00^ + N + COo + O 8.06 33.51 7.90 7.90 .16 25.61 Third Harvest (95 Days) 1 CO,-f 13.97 7.90 6.07 5 N + CO, + Leaves 129.26 Stems 45.17 Tops 174.43 Eoots 32.93 Nodules 112.02 7.90 319.38 311.48 1915] A Biochemical Study op Nitrogen in Certain Legumes 51- \ - 3 Plate X. — Experiment IV: Upper Figure Showing Experiment at Beginning; Lower Figure Showing Experiment 16 Days Later 516 Bulletin No. 179 [March, Plate XL — Lxperiment IV: Lower Pigxire Showing Plants 41 Days from Beginning; Upper Figure Showing Plants 59 Days from Beginning 1915] A Biochemical Study of Nitrogen in Certain Legumes 517 Plate XII. — Experiment IV: At the Time of Harvest (95 Days) 518 Bulletin No. 179 [ March, Plate XIII. — Experiment IV: On the Left, Roots pro.m Plant Grown in CO2-I-O; On the Right, Roots from Plant Grown in N-j-COj-f-O 3 ^V'l 1 4 3 31/2 2 3 3 2% I 3 79J5] A Biochemical Study of Nitrogen in Certain Legumes 519 tomber 7, 1912, with six cowpea seedlings eleven days old, and con- tinued for ninety-five days. The plants were harvested two at each of three periods. The results given in Table 12 need no explanation, tho it might be well to call attention to the individual differences in the plants in the amounts of nitrogen fixed. During the ninety-five days of the ex- periment, Plant 5 fixed fifty-one times as much nitrogen as Plant 1. The following comparison between the growth of these two plants is of interest. Plant 1 attained a total height of 5 inches, possessed one leaf, and on the roots were counted 60 nodules. Plant 5 reached a height of 61 inches ; its leaves measured as follows : 2 measured 5 inches along midrib, 4 inches at base " " 3-31/^ " " " . ; ) J ^ 1 ■> 1 1 ■> 1 11 11 O]/ > > ) ) 11 I > 11 2 " " ' ' ' ' II 11 -11/ 11 11 J 1 11 ! ) 2 111111 Several pods were formed, as may be seen by reference to Plate XII, one of which measured 4I/2 inches in length and was partially filled with seeds. The roots were so large that the Woulfe bpttle had to be broken in order to obtain them. The plant possessed 32 large nodules, 46 medium to large, 66 medium, and 144 small ; 288 in all. Experiment by Kossowitsch Reference has been made to a laboratory experiment conducted by Kossowitsch in the summer of 1891 (see page 503). As his work has been accei)ted by some and ignored by others, it is of particular interest. Peas were started in a mixture of four-fifths sand and one-fifth soil in which peas had been grown the year before. When the plants had developed good nodules, they were transferred to jars containing nitrogen-free sand. In some cases the roots were enclosed and in others the tops, a similar means being used in each case to make the joints air-tight. Over the tops of the jars bell-jars were placed. These were connected thru drier bottles with a gasometer. Because of mois- ture collecting in the bell- jars, absorbents were used to keep the atmosphere normal. The gas mixtures used were hydrogen and oxygen in some cases ; hydrogen, oxygen, and carbon dioxid in some ; and air in others. Combined nitrogen was also used to check up the possible abnormal condition due to the necessary manner of experimentation. Great accuracy was displayed in arranging the ap- paratus and in analyzing the gases. 520 Bulletin No. 179 [ March, a =^" 'T'co --< C5MCilOOOMCOt~«3 gs 1 CO (M T-H CO cd b ^ jj ^ rt g s &■--- co-*-*-^ ics^oarciMtocofO S "^ fl C 2; --H (D -^ & fl -t^ o .' n .-, X > May 10 53 6 " 11 1 061 y > } ) > y May 17 60 8 " 14 987 } } ) y y y May 24 67 10-12 leaves; an average of 5.4 pods 16-17 1 154 >> y y y y May 31 74 10-12 leaves; beans formed in pods 16-17 1 354 The total nitrogen determinations in the various parts of the plants of this series at different stages of growth, together with the nitrogen fixed at each of these stages, are given in Table 15. The amount of the nitrogen fixed was determined by subtracting, from the ^Thompson: Jour. Am. Chem. See. (1915), 37, 230-235. 1915] A Biochemical Study of Nitrogen in Certain Legumes 529 total nitrogen found, the average nitrogen content of five soybean seeds as shown by the individual analyses of twenty seeds. Nearly all the figures in this table represent the average of six determinations. Table 15. — Total Nitrogen in Various Parts of Soybeans and Fixation AT Different Periods: Series 100 (Milligrams per jar of five plants) Harvest Lab. Nos. Nitrogen in tops Nitrogen in roots Nitrogen in nodules Nitrogen in whole plants Nitrogen in seeds 57.30* Nitrogen fixed 1 111-112 114-115 117-118 87.10 13.35 28.04 128.49 71.19 2 121-129 204.59 22.70 47.10 274.39 57.30 217.09 3 131-139 286.91 48.44 82.95 413.30 57.30 356.00 4 141-149 356.52 40.15 60.40 457.07 57.30 399.77 5 151-159 247.82 30.82 54.56 333.20 57.30 275.90 *It would be preferable to use the analyses of uninoculated plants as a check rather than the analyses of seeds. These figures need very little explanation. The results of the first four harvests show a gradual increase in the amount of nitrogen fixed. The low results obtained at the last harvest are in accord with the results of Wilfarth and Wimmer^ and Penny and MacDonald.^ The separation of the nitrogen compounds into the various groups was carried out in this series as follows : The whole eample was dried for four hours at 50° C. both before and after grinding. The subsam- ples were weighed out and placed in 250-cc. beakers ; 100 cc. of Avater was then added and the whole heated to boiling and filtered while hot. Suction was used in filtration, and the washing was done with 50 cc. of hot water. The residue was Kjeldahlizcd as usual. The filtrates from some of the harvests were treated with 10 cc. of sodium hydroxid and distilled. The residual liquid in the Kjeldahl flask was trans- ferred to a 350-cc. beaker and the excess alkali neutralized with sul- furic acid, after which the regular P.T.A. method was applied. The precipitates obtained by P.T.A. were characteristic in their behavior. After the reagent had been added some one or two hours, a volumi- nous grayish white precipitate appeared, sometimes colored a yellow- ish green, and very graduallj^ settled to the bottom in a very thin layer, leaving the supernatant liquid yellowish green in the case of ^Wilfarth and Winimer: Landw. Vers. Stat. (1906), 63, 1-70, ^'Penny and MacDonald: Del. Agr. Exp. Sta. Bui. 86, 35. 530 Bulletin No. 179 [March, the tops, slightly straw colored to colorless in the case of the roots, and colorless in the case of the nodules. The amount of the precipitate from the determination with the tops was much greater than that with the roots and nodules. Caution was exercised in all the determinations not to allow losses or changes due to bacterial action. A few drops of chloroform were placed on the filter and in the filtrate when the determinations were sufficiently long to be liable to bacterial action. The results of the separations are shown in Table 16. Here again, as in the case of the total nitrogen determinations, most of the figures represent the average of two determinations made upon samples from each of three jars. The total soluble nitrogen was obtained by the addition of the various determined soluble forms. The total nitrogen reported in the last column is the sum of all the separations made. These results will be discussed later with those of the other series. Table 16. — Nitrogen Separations: Series 100 (Soybeans) (Milligrams per jar of five plants) Har- vest Lab. Nos. Part Insol- uble nitro- gen Total solu- ble nitro- gen NaOH nitro- gen P.T.A. nitro- gen Other nitro- gen Total nitro- gen 1 111-112 114-115 117-118 Tops Eoots Nodules 61.52 8.90 15.72 24.39 5.00 11.61 .... 4.16 .85 3.54 20.23 4.15 8.07 85.91 13.90 27.33 2 121-123 124-126 127-129 Tops Eoots Nodules 135.15 15.49 32.83 37.99 5.67 16.03 8.11 .48 9.66 29.88 5.19 6.37 173.14 21.16 48.86 3 131-133 134-136 137-139 Tops Roots Nodules 146.79 27.03 47.95 140.12 16.42 35.00 .... 25.63 .93 18.55 114.49 15.49 16.45' 286.91* 43.45 82.95* 4 141-143 144-146 147-149 Tops Roots Nodules 183.35 26.14 31.77 134.26 12.93 27.27 17.86 2.49 2.38 25.96 .85 15.35 90.44 9.59 9.54 317.61 39.07 59.04 5 151-153 154-156 157-159 Tops Roots Nodules 151.68 21.55 29.23 95.32 14.38 27.21 12.02 1.34 2.00 29.31 1.38 12.13 53.99^ 11.66 13.08 247.00* 35.93 56.44 'Taken from Table 15. 'Obtained by difference. The accompanying graph (Plate XV) shows clearly the relation- ship in which the soluble and the insoluble nitrogen exist at the various periods of growth. In this series the first harvest was not made for thirty-eight days, and therefore the period at which fixation began is not shown. Attention has already been called to the low nitrogen fixa- 1915] A Biochemical Study of Nitrogen in Certain Legumes 531 Soluble ^/md Insoluble Nit t^ooen IN To¥^5 Tfoora /^nd Nodules ofSoybe/jns JJlFFEJ^E-NT 'PEFflOnS OF JJe VE tOrr/E NT Ser,es /OO LtOtND doLuBLE □ Insoluble ■ ToF'-b J Jj Tfoor5 isn^, "HDa. (.0 Ua.. Li n^^ 74- Lou. Nodules Plate XV 532 Bulletin No. 179 [ March, tion found at the last harvest in this series. More results are necessary to confirm the supposition of a possible loss in the total nitrogen. The importance of the amount of soluble nitrogenous compounds at the various stages of growth has not yet been emphasized. Prelimi- nary studies have shown this soluble nitrogen to be much more rap- idly converted into ammonia and nitrates than the insoluble nitrogen. This would seem to have a direct bearing upon practical methods of handling leguminous crops in rotations when the shortest time must intervene between the turning under of the legume and the planting of the next crop. It would seem desirable to choose that period when the greatest amount of soluble nitrogen exists. Series 500 (Soybeans) Soybeans were used in Series 500. The plants were placed out of doors during pleasant days in August and September. From Table 17, showing the development at the four harvests, it will be seen that these plants made a more rapid growth than those in Series 100. Table 17. — Plant Development: Series 500 (Soybeans) Planted Harvested Age, Leaves and pods Height, Nodules per days per plant inches 7 15 plants Aug. 8, 1911 Aug. 22 14 3 leaves 30^ J > } ) } y Aug. 30 22 4 " 10 278 J J J J > J Sept. 8 30 6 " 14 370 1 ) > ) ) } Sept. 19 41 7-8 " ; 6 one- inch pods 14 247 (large) Tor ten plants. The total nitrogen determinations for this series are shown in Table 18. These results agree with those shown in Table 15, altho they represent earlier stages of development. Table 18. — Total Nitrogen in Various Parts of Soybeans and Fixation at Different Periods: Series 500 (Milligrams per jar of five plants) Har- vest Lab. Nos. Nitro- gen in tops Nitro- gen in roots Nitro- gen in nodules Nitro- gen in whole plants Nitro- gen in seeds Nitro gen fixed 1 2 3 4 511-519 521-529 531-539 541-549 46.99 47.32 96.96 205.40 8.50 11.08 9.76 18.65 .35 11.05 17.81 26.98 55.84 69.45 124.53 251.03 57.30* 57.30 57.30 57.30 (-1.46) 12.15 67.23 193.73 *This figure is approximate rather than exact. See note to Table 15, page 529. The separations in this series differed from those in Series 100 in that in this case a cold-water extract was made. The sub-samples 1915] A Biochemical Study of Nitrogen in Certain Legumes 533 were placed in shaker bottles and the same amount of water added as in the former series; the bottles were then put in a mechanical shaker for three hours. This method was considered to be more in ac- cord Avith natural conditions than the one used in the former series. The nodules were filtered thru a diatomaceous earth filter. The figures in Table 19 were obtained in the same manner as those in Table 16. Table 19.- — Nitrogen Separations: Series 500 (Soybeans) (Milligrams per jar of five plants) Har- vest Lab. Nos. Part Insol- uble nitro- gen Total soluble nitro- gen 25.90 4.48 NaOH nitro- gen P.T.A. nitro- gen Other nitro- gen Total nitro- gen 1 511-512 514-515 517-518 Tops Eoots Nodules 19.48 4.42 3.16 .95 3.31 .33 19.43 3.20 45.38 8.90 2 521-523 524-526 527-529 Tops Eoots Nodules 29.60 7.97 17.34 2.14 1.18 .00 4.96 .47 11.20 L67 46.94 10.11 3 531-533 534-536 537-539 Tops Eoots Nodules 70.04 8.56 16.75 31.64 2.29 1.29 2.67 .00 .00 6.48 .70 .33 22.49 1.59 .96 101.68 10.85 18.04 4 541-543 544-546 547-549 Tops Eoots Nodules 115.39 13.16 22.55 76.11 7.35 2.46 10.68 2.06 .26 65.43 5.29 2.20 191.50 20.51 25.01 Series 700 (Soybeans) Soybean seeds Avere planted on September 6 for this scries, but owing to their damping off, the jars were replanted on September 13. The seedlings that damped off were tested for ammonia, nitrites, and nitrates, with negative results. The conditions of the plants at the various harvests are shown in Table 20. Table 20. — Plant Development: Series 700 (Soybeans) Planted Harvested Age, days Leaves per plant Height, inches Nodules per 15 plants Sept. 13, 1911 Sept. 25 12 2 leaves 5 Nodules pres- ent too small to remove J ) J ) > J Oct. 6 23 3 leaves par- tially devel- oped 8-9 254 ) 1 ) > } > Oct. 14 31 5 leaves 10-12 229 if > > ) ) Oct. 25 42 7 leaves 12 288 534 Bulletin No. 179 [March, The figures reported in Table 21 present the average of duplicates of composite samples which included the whole of the material from three jars. As will be seen, these results are concordant with those of the two series already considered. Table 21, — Total Nitrogen in Various Parts of Soybeans and Fixation AT Different Periods: Series 700 (Milligrams per jar of five plants) Har- vest Lab. Nos. Nitro- gen in tops Nitro- gen in roots Nitro- gen in nodules Nitro- gen in whole plants Nitro- gen in seeds Nitro- gen fixed 1 2 3 4 711-719 721-729 721-739 741-749 40.08 48.00 68.32 110.70 8.88 8.93 11.38 20.55 7.21 8.97 17.22 48.96 64.14 88.67 148.47 57.30^ 57.30 57.30 57.30 (-8.34) 6.84 31.37 91.17 ^See note to Table 15, page 529. The figures in Table 22 showing the nitrogen separations were ob- tained in the same manner as those reported for the nitrogen fixation in Table 21 ; that is to say, they are the averages of duplicates of com- posite samples. Table 22. — Nitrogen Separations: Series 700 (Soybeans) (Milligrams per jar of five plants) Har- vest Lab. Nos. Parts Insol- uble nitro- gen Solu- ble nitro- gen P.T.A. nitro- gen Other nitro- gen Total nitro- gen 1 711 714 717 Tops Boots Nodules 9.94 3.87 28.03 4.95 4.70 .39 23.33 4.56 37.97 8.82 2 721 724 727 Tops Boots Nodules 23.64 5.50 21.56 2.55 11.96 .77 9.60 1.78 45.20 8.05 3 731 734 737 Tops Boots Nodules 26.50 7.30 6.30 29.05 3.76 1.79 13.75 1.10 .10 15.30 2.66 1.69 55.55 11.06 8.09 4 741 744 747 Tops Boots Nodules 46.45 10.67 14.52 61.97 6.31 2.70 23.75 .61 .00 38.22 5.70 2.70 108.42 16.98 17.22 The close agreement of the results of Series 700 and 500 is very evident in the data presented. By reference to the accompanying graph (Plate XVI) it will be seen more easily than in the tabular form that the soluble nitrogen predominates in the early growth of the seedling. The amount decreases during this period, however, while 1915] A Biochemical Study of Nitrogen in Certain Legumes 535 JJiFFETfE N T 'PeT^IOVd or JJe UdLOP'^E N7 •5ol uBl E I I li^bOL UI3L C ^1 6e^'f ^ 7oo Tq-p^ ?7bo76 ^ 3 Ho. 3IJ]a,. '^BJla NODUL t 5 5f T'ES t a J 7^^ /Too 7" 3 /-? Da ?2 Ba. 3QD^ Nob oLt b Plate XVI 4/Z7j 536 Bulletin No. 179 [March, the insoluble nitrogen always increases. This is true of the roots as well as the tops. It is during the period between the twelfth and the twenty-second days that nitrogen fixation begins, according to meas- urements by the most accurate chemical methods. Detailed studies are now being made of the exact time when fixation begins. Series 600 (Cowpeas) Cowpeas were used for Series 600. Owing to the smaller nitro- gen content of cowpea seeds, the plants show a need of nitrogen much sooner than soybeans, and are therefore perhaps better suited to ex- perimentation of this sort. The plants grown in this series are com- parable with the soybeans in Series 500 as regards time and conditions of growth. The data in Table 23 show the development of the cow- peas when harvested. Table 23. — Plant Development: Series 600 (Covppeas) Planted Harvested Age, days Leaves per plant Height, inches Nodules per 15 plants Aug. 8, 1911 } y J ) } } > > ) > 7 7 7 > 7 ; 7 J Aug. 22 Aug. 30 Sept. 8 Sept. 19 Oct. 5 14 22 30 41 58 3 leaves 4 " 5 " 6-7 " 8 " 7 10 12 13-14 14 450 (very small) 892 (small) 1074 2062 1992 The results given in Table 24 show a fixation of nitrogen at the end of fourteen days from the time the seeds were placed in the sand. The increased fixation is greater with the cowpeas in this series than with the soybeans in the corresponding series (500). The other gen- eral tendencies appear to be the same as in the other series. Table 24.— Total Nitrogen in Various Parts of Cowpeas and Fixation AT Different Periods: Series 600 (Milligrams per jar of five plants) Nitrogen Harvest Lab. Nos. Nitrogen in tops Nitrogen in roots Nitrogen in nodules 1.96 Nitrogen in whole plants 40.12 m umn- oculated plants Nitrogen fixed 1 611-619 29.11 9.05 36.74* 3.38 2 621-629 45.46 10.25 9.22 64.93 36.74 28.19 3 631-639 91.63 16.64 18.52 126.79 36.74 90.05 4 641-649 188.40 30.28 42.33 261.01 36.74 224.27 5 651-659 439.64 73.73 64.25 577.62 36.74 540.88 ^This figure was used as it is a little larger than the average nitrogen con- tent for the seeds, 34.70 milligrams, which would make even a greater fixation ap- pear at the harvest. The results of the separations are shown in Table 25. The figures were obtained in the same manner as those in Series 700. 1915] A Biochemical Study of Nitrogen in Certain Legumes 537 Table 25. — Nitrogen Separations: Series 600 (Cowpeas) (Milligrams per jar of five plants) Harvest Lab. Nos. Part Insoluble nitrogen Total soluble nitrogen P.T.A. nitrogen Other nitrogen Total nitrogen 1 611-613 614-616 617-619 Tops Roots Nodules 15.83 6.92 11.62 2.13 3.68 .74 7.94 1.39 27.45 9.05 2 621-623 624-626 627-629 Tops Roots Nodules 23.73 9.18 7.54 21.76 1.89 2.26 5.88 .42 .92 15.88 1.47 1.34 45.49 1L07 9.80 3 631-633 634-636 637-639 Tops Roots Nodules 58.67 11.18 13.69 32.54 5.32 4.83 10.55 1.90 1.22 21.99 3.42 3.61 91.21 16.50 18.52 4 641-643 644-646 647-649 Tops Roots Nodules 97.82 2L54 26.74 93.47 7.61 15.59 32.49 2.51 9.08 60.98 5.10 6.51 191.29 29.15 42.33 5 651-653 654-656 657-659 Tops Roots Nodules 216.90 52.35 32.36 226.50 19.87 31.89 58.50 5.17 20.10 168.00 14.70 1L79 443.40 72.22 64.25 The results of this series are also presented in a graph (Plate XVII). The curve of the soluble nitrogen does not show a decrease, possibly because the change upward had taken place before the end of the first fourteen days, when the first data were taken, as cowpeas show an early fixation of nitrogen and develop extremely rapidly un- der normal conditions. The same general tendencies hold thruout this series as in the others in respect to the increase of the soluble and insoluble nitrogen. When the soluble and the insoluble nitrogen ratios are considered, the results in general agree very closely in all the series regardless of time of growth and kind of legume. Discussion of Tables The results of the total nitrogen determinations of the four series show that, as an average of eighteen harvests, 74 percent of the nitro- gen of the cowpeas and soybeans Avas in the tops, the remaining 26 percent being divided between the roots and the nodules. The figures show that in the first periods most of the 26 percent was in the roots, while later the nodules in some cases contained 18 of the 26 percent. In nine out of seventeen harvests, the nodules contained more nitro- gen than the roots of the same plants. The data showing the average daily fixation of nitrogen for five plants in the various series during the different growing periods are presented in Table 26. 538 Bulletin No. 179 [March, Soluble ^nd Ih/bOL uBle Nij fioge n IN Tof>5,f^oor::> fjND NodulE^ of (_ o v^ -H E 'j i> J] IF FE TfE^y T f^EFfiOnS OrUEVELOfr/ENT Sfir /fcS GOO Legend Soluble, n In50luBl^^ cBM Tc 0?=5 ffoc J J J ,4D^^ 2SJ]^ 3011c. ^iH--- 5,5 i7. Nodules Plate XVII 1915] A Biochemical Study of Nitrogen in Certain Legumes 539 Table 26. — Average Daily Fixation of Nitrogen in All Series Series Periods Milligrams per in days jar of 5 plants 0-38 1.87 100 (Soybeans) 38-53 9.72 53-60 60-67 19.84 6.16 67-74 (-17.69) 0-14 .00 500 14-22 1.51 (Soybeans) 22-30 6.88 30-41 11.50 0-12 .00 700 12-23 .62 (Soybeans) 23-31 3.06 31-42 5.43 0-14 .24 600 (Cowpeas) 14-22 3.10 22-30 30-41 7.73 14.02 41-58 17.44 The results included in Table 26 seem to indicate that the period during which the greatest total accumulation of atmospheric nitrogen takes place occurs between the fortieth and sixtieth days and coincides closely with the period just previous to seed formation. The greatest rate of increase of fixation and assimilation in these .series occurred in the early periods of growth. A comparison of Series 100 with the other series indicates that the growth of the plant is closely related to the rate and the amount of nitrogen fixed. The plants of Series 100 grew much slower than the others. Those of Series 500 and 600 made the greatest growth in a given period, having had the advantage of the most favorable growing season, more especially for the cowpea, which requires a higher temperature than the soybean for optimum growth. The fixation in Series 600 for the whole period of fifty-eight days was 540.88 milligrams per five plants, or an average daily fixation of 9.32 milligrams. The greatest fixation during any one period, as is evident from Table 26, occurred with the soybeans in Series 100 be- tween the fifty-third and sixtieth days, Avhen the daily average was 19.84 milligrams per five plants, or nearly 4 milligrams per plant per day. If this figul*e is calculated to an acre basis, allowing a stand of four beans per square foot, an accumulation equivalent to one and a half pounds of nitrogen per acre per day is shown. The average percentages of soluble nitrogen in the four series in terms of total nitrogen in the particular part of the plant, may be of some interest, altho it will be seen from the accompanying graphs 540 Bulletin No. 179 [March, that the amount depends upon the stage of growth when the harvest is made. These percentages were as follows: Table 27. — Percentages of Soluble Nitrogen in Each Series AS an Average of All Harvests (On the basis of total nitrogen in the given part) Series Tops Boots Nodnles 100 (Soybeans) 700 (Soybeans) 500 (Soybeans) 600 (Cowpeas) 35.9 57.7 41.2 45.0 34.7 39.5 32.4 27.5 42.3^ 18.9 8.5 34.0 ^The nodules in this series were not filtered thru a diatomaeeous earth filter but thru an ordinary filter and are therefore not included in the average given in the conclusions. The figures for Series 100 represent soluble nitrogen obtained in a hot-water extract. Series 500 and 600 are comparable with the exception of one being soybeans and the other cowpeas. The great difference in the solubility of the nitrogen in the nodules is particu- larly noticeable. There was a gradual increase in the soluble nitrogen in the nodules of Series 600 from the first harvest to the last, the percentages on the basis of total nitrogen being 23, 26, 37, and 49. A fact not brought out in the figures showing the soluble nitrogen is that in Series 700 and 500 an extremely high soluble-nitrogen content was found in the tops and the roots at the first harvest. In Series 700 the percentage in the tops was 74, in Series 500, 57 ; while in the roots in Series 700 the percentage was 56, and in Series 500, 50. On the basis of total nitrogen, the percentage of Other nitrogen in each series, as an average of all harvests, was as shown in Table 28. Other nitrogen is the difference between the total soluble nitrogen and that precipitated by P.T.A. and NaOH. It has been shown that a part of this nitrogen consists of amino acids, but as yet the total amount is unknown. Table 28. — Percentages of Other Nitrogen in Each Series AS AN Average of All Harvests (On the basis of total nitrogen in the given part) Series Tops Eoots Nodules 100 (Soybeans) 700 (Soybeans) 500 (Soybeans) 600 (Cowpeas) 25.1 36.1 30.7 28.3 29.3 32.8 23.1 17.2 18.2 6.9 16.6 The percentages of nitrogen precipitated by P.T.A. were as shown in Table 29. The variations as regards the tops are not easily ex- plainable. Undoubtedly there is a larger error in the P.T.A. deter- minations than in the others. The nodules of Series 600 contained 1915] A Biochemical Study of Nitrogen in Certain Legumes 541 large amounts of nitrogen which were precipitated by this reagent, the percentages at the harvests, from the second to the last, on the basis of total nitrogen being 9, 7, 21, and 31. Table 29. — Percentages of P.T.A. Nitrogen in Each Series AS AN Average of All Harvests (On the basis of total nitrogen in the given part) Series Tops Boots Nodules 100 (Soybeans) 700 (Soybeans) 500 (Soybeans) 600 (Cowpeas) 7.00 21.10 6.75 13.40 3.0 6.6 6.0 7.3 l.i) 1.4 17.0 The nitrogen obtained hy distillation with sodinm hydroxid ap- parently is not precipitated by P.T.A., as the percentage of Other nitrogen decreases without exception when sodium hydroxid is used. However, no definite conclusions can be drawn regarding the use of sodium hydroxid. CONCLUSIONS PART I 1. The experiments reported show conclusively that the cowpea and the soybean utilize atmospheric nitrogen thru their roots and not thru their leaves. No combined nitrogen could have been assimi- lated in these gas experiments. PART II 2. The total nitrogen determinations show that about 74 per- cent of the nitrogen of cowpeas and soybeans at the time of harvest is in the tops, while the remainder is distributed between the roots and the nodules. In the earlier periods the roots contain the larger part, w^hile later they contain much the smaller part. 3. The percentage of soluble nitrogen in soybeans and cowpeas varies with the different parts of the plant and with the period of growth. In these experiments the soluble nitrogen, as an average, constituted in the tops about 45 percent of the total nitrogen; in the roots, 34 percent ; in the nodules of the soybeans, 14 i)ercent, and in the nodules oi the cowpeas, 34 percent. 4. Phosphotungstic acid usually precipitates some form of nitro- gen. In some cases the amounts precipitated vary widely, while in others the agreement is close. In these series the nitrogen precii)itated by phosphotungstic acid averaged in the tops of both soybeans and cowpeas about 12 percent of the total nitrogen; in the roots. 5.5 per- cent ; in the nodules of the soybeans 1 percent, and in the nodules of the cowpeas, 17 percent. 542 Bulletin No. 179 [March, 5. Other forms of soluble nitrogen than those precipitated by phosphotungstic acid and sodium hydroxid occur. In these series they constituted as an average in the tops of both soybeans and cow- peas about 68 percent of the soluble nitrogen ; in the roots, 77 per- cent; in the nodules of soybeans, 89 percent, and in the nodules of cowpeas, 53 percent. 6. Fixation takes place at a very early period in the growth of the seedling — sometimes within fourteen days. It is rapid in some cases, especially with cowpeas. 7. Plants grown under the conditions of these experiments con- tain no ammonia, nitrites, or nitrates, as measured by the most ac- curate chemical methods. It is fully recognized that this work is incomplete, yet it is hoped that the study may aid in stimulating interest in some of these funda- mental problems. The lack of development of the various chemical methods used is partially responsible for some of the difficulties ex- perienced in their application. The survey presented of the chemical literature indicates a scarcity of existing knowledge regarding the more fundamental problems concerning nitrogen fixation. The biological resume points clearly to the need of more extended research along these lines. This is strikingly noticeable in the bac- teriological studies which have been undertaken with B. radicicola. Few cases are on record in which the authors actually proved out their cultures at the termination of their investigations by inocula- tion of a legume of the kind from which the organism originally came. The author takes this opportunity to express his gratitude to Professors C. G. Hopkins and J. H. Pettit for the many valuable sug- gestions they have so kindly given him. BIOGRAPHICAL SKETCH Albert Lemuel Whiting was born in Stoughton, Massachusetts, May 12, 1885. He secured his common-school education in the public schools of that town and then spent one jeav in mechanical work be- fore entering the Massachusetts Agricultural College in the fall of 1904. From this institution he received the degree of bachelor of science in June, 1908. He immediately accepted a position as assist- ant agronomist at the Rhode Island Agricultural Experiment Sta- tion, where he also instructed in veterinary science in the College. At the same time he served as secretary of the Rhode Island Experi- mental Union. While there he also took graduate student work in agronomy and botany and in June, 1910, received the degree of master of science. In the fall of 1910 he was granted a fellowship in agronomy in the University of Illinois, which he held during 1910- 12. He is now associate in soil biology in that institution. Dr. Whiting is a member of the following societies : Q.T.V. ; Alpha Chi Sigma ; the Illinois chapter of Sigma Xi ; Society of American Bacteriologists ; the American Society of Agronomy ; the American Chemical Society; and the American Society for the Promotion of Science. LIBRARY OF CONGRESS 002 756 623 6