ISF 99 N4 Copy 1 JH THE AMINO ACID CONTENT AND NUTRITIVE VALUE OF THE PROTEINS OF COTTONSEED MEAL BY WILLIAM BARBOUR NEVENS B.S. University of Wisconsin, 1914 THESIS Submitted in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY IN ANIMAL HUSBANDRY IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS 1921 THE AMINO ACID CONTENT AND NUTRITIVE VALUE OF THE PROTEINS OF COTTONSEED MEAL BY WILLIAM BARBOUR NEVENS B.S. University of Wisconsin, 1914 THESIS Submitted in Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY IN ANIMAL HUSBANDRY IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS 1921 ^A' LIBRARY OF CONGRESS RECEIVED MAY 261922 f)OCUM£NTvi D?VI3!Oi ■■ij TABLE OF CONTENTS I. Amino Acid Content PAGE 1. Introduction 375 2. Methods Employed 377 3. Discussion of Results 379 4. Summary 398 5. References i 399 II. Nutritive Value 1. Review of Previous Work 552 2. Methods Employed 552 3. Discussion of Results 573 4. Summary 582 5. References 588 Digitized by the Internet Archive in 2011 with funding from The Library of Congress http://www.archive.org/details/aminoacidcontentOOneve Reprinted from Journal of Dairy Science Vol. IV, No. 5, September, 1921 THE PROTEINS OF COTTONSEED MEAL 1 I. AMINO ACID CONTENT W. B. NEVENS From the Department of Animal Husbandry, University of Illinois, Urbana I. INTRODUCTION One of the earliest references to the proteins of cottonseed meal was made by Ritthausen (1), who separated the pro- teins in the form of spheroids. Osborne and Voorhees (2) isolated a protein from cottonseed meal which had the nature of a globulin, being soluble in salt solutions, and comprised 42.3 per cent of the total nitrogen of the meal. Another protein (or proteins) was found to be insoluble in salt solutions but soluble in 0.2 per cent potash solution and amounted to 44.3 per cent of the total nitrogen of the meal. Two per cent of the total nitrogen was present in the form of water soluble proteose. TABLE 1 Percentage of nitrogen in the different groups in various proteins Globulin, cottonseed Globulin, wheat Zein, maize Hordein, barley Nab AMMO- NIA 1.92 1.42 2.97 4.01 BASIC N 5.71 6.83 0.49 0.77 NON- BASIC N 11.01 9.82 12.51 12.04 NlN MgO PRECIPI- TATION 0.28 0.16 0.23 TOTAL N 18.64 18.39 16.13 17.21 The distribution of nitrogen in various protein bodies was studied exhaustively by Osborne and Harris (3), who employed the modified Hausmann method (4). The following values are typical of their results. 1 The results presented in this paper formed part of a thesis submitted to the Graduate School of the University of Illinois in partial fulfillment of the require- ments for the degree of Doctor of Philosophy in Animal Husbandry. 375 JOURNAL OF DAIRY SCIENCE, VOL. IV, NO. 5 376 - W. B. NEVENS These investigators state that "This wide variation in the proportion of basic decomposition products of the various pro- teins .... raises important questions regarding their food value." Osborne (5) further found that the basic nitrogen of the globulin of cottonseed, as determined by precipitation with phosphotungstic acid, consisted of 3.46 per cent histidine, 13.51 per cent arginine and 2.06 per cent lysine. The content of the mono-amino acids of the "edestin" of cottonseed meal was determined by Abderhalden and Rostoski (6) by the use of the Fischer ester method (7), and is as follows, calculated for dry, ash free edestin of cottonseed: per cent Glycocoll . 1.2 Alanin 4.5 Amino valerianic acid Present a-proline 2.3 Leucin 15.5 Glutamic acid 17.2 Aspartic acid 2.9 Phenyl alanin 3.9 Serin 0.4 Tyrosin 2.3 Tryptophane Present The quantitative determination of the amino acids of feeding- stuffs by means of the Van Slyke method (8) was undertaken by Grindley and his co-workers (9), and a little later by Nollau (10). In Nollau 's procedure, samples of the finely ground feeds were hydrolyzed with 20 per cent hydrochloric acid until the content of amino acid, as determined by the Van Slyke method, became constant. The material insoluble in hydrochloric acid was filtered off, the clear extract concentrated under diminished pressure and made up to a certain volume. The total nitrogen content of this extract was used as a basis for calculating the final results. In a report of the subsequent work of Grindley and co-workers (11), it is claimed that since Nollau filtered off the solid residue after hydrolysis of the feedingstuff and before making his total nitrogen determinations upon which the final calculations were based, that his results are not accurate since a part of the total nitrogen was undoubtedly discarded in the solid residue. THE PROTEINS OF COTTONSEED MEAL 377 The heats of combustion of several vegetable proteins were carefully determined by Benedict and Osborne (12). The globulin of cottonseed was found to yield 5598 calories per gram, compared to 5358 calories per gram for the globulin of wheat and 5916 calories per gram in the case of the hordein of barley. In commenting upon their determinations, the inves- tigators state that " many irregularities. . . . appear, which are doubtless due to the different proportion of the various amino acids which constitute the molecules of the different proteins." It is evident from the foregoing discussion that our knowledge of the composition of the proteins of cottonseed meal is very incomplete. The globulin is the only protein of cottonseed which has been isolated in pure form and whose composition has been determined. The globulin, however, according to Osborne and Voorhees, contains only 42.3 per cent of the total nitrogen of the cottonseed. The character, identity and chemical composition of the remaining proteins are practically unknown, and it is evident from the data given above that our knowledge of the distribution of the nitrogen in the proteins of cottonseed meal is very meager indeed. The investigation of the distribu- tion of the nitrogen in the proteins of cottonseed meal therefore constituted the object of this study. II. METHODS EMPLOYED IN CHEMICAL ANALYSIS 2 The method of analysis employed consisted of two main procedures. The first consisted of a series of extractions whereby the nonprotein together with a very small amount of protein was first removed, and following this, the proteins were extracted from the residual matter of the sample which consisted mostly of fiber. The second main procedure embraced the hydrolysis of the extracted proteins, and the analysis of the resulting solution 2 The method of procedure here outlined is one which has been developed and perfected in this laboratory by Dr. H. S. Grindley, Mr. T. S. Hamilton and asso- ciates (9, 11, 33, 36 and unpublished manuscripts). The method of extraction preliminary to hydrolysis of the proteins has been developed entirely in this laboratory, while the actual determination of the nitrogen in the different groups follows closely the method of Van Slyke, but includes modifications perfected in this laboratory. 378 W. B. NEVENS for certain amino acid and other groups according to the general method of Van Slyke (8). The sample of cottonseed meal was prepared from good quality commercial meal, finely ground and passed through a 40- mesh sieve. Each sample taken for analysis weighed 15 grams and contained 1.0194 grams of nitrogen, or about 6 grams of protein. In the first three extractions, which were carried out con- secutively, cold anhydrous ether, cold absolute alcohol and cold 1.0 per cent trichloracetic acid were used. The samples of feeding stuff were placed in 500 cc. centrifuge bottles and 100 to 200 cc. of the reagents added. The bottles were placed on a shaking machine which rolled them back and forth continually. Usually two extractions were made each twenty-four hours, one extraction period being seven to eight hours and the other 15 to 16 hours in length. At the end of the extraction period, the sides of the bottles were washed down with the reagent, the bottles centrifuged and the clear supernatant liquid decanted. Usually six or seven extractions with each reagent were necessary. The ether and alcohol extracts were filtered and any residues returned to the centrifuge bottles. After slight acidification with sulphuric acid, the ether and alcohol were evaporated and recovered and total nitrogen determinations made on the residue. The small amount of protein removed in the trichloracetic acid extracts was recovered by precipitation with colloidal ferric hydroxide (containing 5 per cent Fe 2 3 ) in boiling solution. The precipitate was transferred to a digestion flask with 20 per cent hydrochloric acid, and total nitrogen determined in the filtrate. The bulk of the proteins was removed from the residue re- maining after extraction with 1.0 per cent trichloracetic acid by extraction, first, with dilute sodium hydroxide solution, then with 20 per cent hydrochloric acid followed by treatment with 5 per cent sodium hydroxide solution. The dilute sodium hydroxide solution used during the shorter period was a 0.2 per cent solution and that during the longer period a 0.1 per cent solution. These extracts were neutralized, acidified with hydro- THE PROTEINS OF COTTONSEED MEAL 379 chloric acid and concentrated in vacuo to a small volume. An equal volume of concentrated hydrochloric acid was then added. The residues remaining after treatment with dilute sodium hydroxide solution were boiled for three minutes with 20 per cent hydrochloric acid. After cooling, the solution was filtered off, the residue washed and the procedure repeated once. The washings were evaporated to a small volume, an equal volume of concentrated hydrochloric acid added and the washings then combined with the main hydrochloric acid extract. The residues insoluble in hydrochloric acid were treated three times with 5 per cent sodium hydroxide solution using centrifuge bottles as in former extractions. After washing the residues nearly free from alkali, they were submitted to Kjeldahl analysis. The extracts and washings were acidified with hydrochloric acid, concentrated in vacuo and transferred to digestion flasks with an equal volume of concentrated hydrochloric acid. The proteins precipitated by colloidal iron and the proteins removed by extraction with dilute sodium hydroxide, 20 per cent hydrochloric acid and 5 per cent sodium hydroxide were completely hydrolyzed by boiling for fifteen hours upon a combined electric plate and sand bath under reflux condensers. The resulting solutions were combined and analysis for the chemical groups characteristic of certain amino acids executed essentially as directed by Van Slyke (8), but with the use of minor improvements perfected in this laboratory. III. DISCUSSION OF RESULTS The results obtained by application of the method of chemical analysis outlined in the preceding section to eight portions of the same original sample of cottonseed meal are shown in the accom- panying tables 2 and 3. Table 2 shows the values expressed in percentage of the total nitrogen present in the sample of feeding stuff when it was taken for analysis, while table 3 shows the same values expressed in percentage of the feeding stuff itself. Two averages are included in the tables. The first is com- piled by taking the average of all values obtained by analysis of the entire eight samples. The second average is obtained by § P. s .3 .S? A », -^ m -2 s to H O aoj pa^tmoooB naSoa^ra |i3^ox CM OJ OJ l*- OJ 00 CO b- OJ OJ —1 OJ CO O r~ oo 00 CJ OJ OJ tP 00 CO W5 i-H CO OJ t^ OJ OJ oj co O OJ O rh oo oo OJ OJ 99.478 99.765 g B O ►4 H ", O fx O pj « ■ CO co co CO CO CO OJ T-H I> CO CM 1^ o !>• O IN CO sasisq aq:* raoij a^BJ^ -jg aq^ jo noi^njos uiojj anpisaj nj o o o o o © CJ CO !>• CO O i-i o o C3 o o co co i— ( i-H o o t> CM >-H 00 CM (M o o ljQ i-H lO o CO CO o o -xrni jaq^a-loqooi^ ptare ui ajqrqog CD CO o -* OS O O H ■<* oo co t^ J> o tH i-i i—l CO CD OJ o o i-H 00 oo -* O T-H lO CO i< co CO tV o o (a^idioaid a^sStirnoqdsoqd jo uonisbdraooap Suunp noi^njos raojj'paja^ig) nimn'q paqjospranft co »o o ■* -Cfl l-» ^ oo o q i^ oo CO ■* oo ■* i-H T-H T-H CO i-H r^ o o co co O CO CO "3 o o OJ OJ CM ■># CO t-H O CO CO Tti OJ OJ i-H OJ lO CO co co OJ OJ i-H CO O CO CO CO OJ OJ sasrcq aq^ rao-rj a^Bi^g ui £j oniun3-uo£[ OJ CO t^- CO o ■* Tji cm oo io O »0 OJ OJ CO (M oo ■* ■* id CO -* CM t-H i-H T-H CO CO 1-H 1— 1 o o oo i> CM o 00 CD co tj5 CM OJ CO >CH CO CM i-H 00 l> CD ■* CD CO CO o o CO t-H Tp HO * CM co CO CM T-H oo OJ cm' 15 ts O a g S o M ft. O 8 naSoj^ni uia'jojduon j^o j, 7* CO «3 lO d — 1 Ci Ci o CD O Ci 00 Ci Ci Cl Ci oo co oo o co ID CO i-H o CO co CO i-H CO t~ CO 00 CO CO Ci CO CO co CO CO co CO CO CO CD Ci Oi o o •<* o CO o o CO 00 o CO CD 00 t~- o o O O o o o o d © ID iO ^ CO oo CO o SI 1—1 i-H ID i—l i— 1 © 00 ^P CO CO CO o o o o o o d d CO J>- ■* l-H co ci Ci Ci ID o Ci o oo o H ID CO I> Ci 00 © o o 1—4 i-H 1— t d d i— 1 CO CO r-l O O 00 T— 1 oo *o CO 00 CO CO CO © CO I-H ID CO CO 1—1 i— I o o i-H i-H 1-5 l-H ID oo CO Ci 1—1 1> Ci 00 ID O CO co Ci o o o d d CO CO Ci I>- CO O co o »d o CO CD o CO CO Ci CO i-i 00 Ci Ci CI Ci o i-H CO o CO ID Ci Ci *# CO O I"- »o o 1—1 CO Ci CO ID CO Ci O CO 1—1 oo lO t- co t^ ID CO - CO CO o Ci CD ID 00 CO o Ci tv- o o co oo -* ID CO CO ID "tf Tfi CO oo co CO CO ID i—l CO o lO CO CO O 1^ t^ Ci i-H CO CD CO Ci Ci ID 1> t^ t^ co co co co Ci Ci CO Ci CO CO CO o oo 1—1 ID 1> CO CD •* o Ci Ci o o o o o o d d o »o 00 o ID CO CO CO Ci o Ci Ci o ID CO o co Ci Ci t— 1 1— < 1>- i—( oo Ci o oo oo l-H l-H oo T— 1 CO co o o CO CO ID ^H 00 Tin ->* Ci Ci Ci Ci Ci o cS CO CO CO CO o 00 Ci >D 00 CO CO CO CO CO o CO Ci CO CO CO CO Ci i^ Ci I> CO CO CO CO CO C-l CO oo Ci >D CO ID i-H CD O O CO l-H CD CO !>. CO CD co Ci O Si i— i o Ci CO ID CO ID r}H »d CO t^ ID ID Ci 9 8 CD O Jr^ id »D ID o O O O O § i—i oo o o ID »D Ci CO O i-H o o o o o CO o 00 o CO CD bO bfl I-H JH > > < < a s -4J O o o o «•-• bfl a m CD =3 O r3 CD OS > o 2 .■s S ft a X a o 01 CO CD 01 a.2 £ £ & O CD CD *d ft tl o o O O o CD T3 o3 03 o3 a a "ci'a CO :-i a <4-i o o S CD co bfl o CD bfl U Sh 53 2 oi u S CD CD CD 0) £ > > A (A < I co co sis^bub a^t^is UB A J° s^insajt + "i^j uia^oiduo^i CO CM CM © co tP CO CO CO CO CO CM OS a> CO CO CO tH co Ci (M Q CO CO C5 CM !>. CO iO -* CO CO co CO co oo i-H IO »o IO CD CD g i o ij « to a r-i O J K 5 F< 8? e§ gg "a « W P S38Bq p a^Bi^t; jo uoivqos uiojj anpisay 1-H CO o o o o o © C3 IO i-H CO o o o o o o T« i-H o o o o o o co oo o o o o o o JV CD Tf IO i-H O © © © © sasBq jo uoi^njos tnojj anpisa'jj 33 i—i i— i o o o o O CM t^ (M O CM o o o o CO CO CO CO I-H i-H o o o o io co CO CO i-H i-H o o o o co © •> Q 1-H ■* © © o o jaqia-loqocqB \Rhvz ajqnjog CO © . i-i © CO t- o o o © o> co 03 CO o o o o co co i-H IO <* CO o o o o IO OS o o o o © o !§ © o "3 § to (a^BjidioaJd a^sSurnoqdsoqd jo uot^isod'raooap 3uunp uoi^rqos uiojj ' paaa^m) uiuiriq paqiospsuQ co co CO IO CO ■* 1—1 I—I o o co lo co co co t^ 1-H O o o 05 IO • co oo CM OS i-H O o o CO i-H Tfl C2 o o CD ■* co co 00 t~ © o © © HBJJiB Suoi^s ui a[qiqosaj o »o I—I o d o d IO o IM o d oo IO 1-H o d i on s < S3 -=! H M S- J (0 > w o to a p OQ w K sasBq uiojj a^Bujg ni uaSoi^tu ouiuib-uo^i CO Tj< t- © cm i-i o o CO C5 to o CO o saSBq uiojj a^Bii'g ui naSoatjta ouiuiy t- o ic oo I—I 1—1 CO CO CM CM o co OS i-t CO CO CM CM O CO i— 1 OS CO I-H CO CO CM CM S3 !>- b- co CO © o CM CM o «o i-H i-H CO t- co oo CM CM tg bfl uaSoi^iu auisXrj i-H CO CO CO CO i-H oa co o o CO OJ "* CO CM i-H o o O CO CO C5 8c3 o o i-H IO 1-H CO o o CO CO © © t- oo C^l CO f CO CO CM © © 5 « uaSoj^iu anipi^eijj CM ■* ■* —I i-l CO Tfi co o o T-H CO o -* o o t^ CM I> CD co ■<*< Tp IO © © © © TM i-H t^ © tH CO o © wary of the results of (Results expressed ] uaSoijra aui^sjtQ CO CO CO CO o o d d co co iO CO IO CO o o o o co co O "* o o o o eo co t^ o CO © lOqoajB a^njosqB ni ajqnjog CO CO o d o CM o d 3 •"CH o d i-H © © «o eo © © jaq^a ut aiqrqog I-H o o d 1—1 CO o o d co i-H o d © © © »o »o 8 d aaawnN cnaiMvs i-H o CM o CO o IO u 382 o OO CM CM o 00 oo CM o CO IO CO CM CO CO oo or O 00 1^ CO CM C5 o CC © t-H t— 1 CO © CO CO t> t- CO CO O CO CN CO T-H Oi CO io CM IO O r-i CO CM o o CO T-H CO t^ o o o o CO © T-H CO o CO CO m © ID tH o o t-H CO -* r-l © © o o o o u u io io t^ CM t> CO OS (N CO CO CO "O CO »o •o IO c-> o © o o o o c 1 o o o o o o o O o o © o o o o o o o _o_p_ CO IO CM Oi CO io o o o o CM -* © CO CM CO O O CM t-H O t-H TH t-H CO 00 © T-H T-H CM CM O CM T-H Oi IO CM CO o CM CM t- 00 T-H T-H o o o O O o ©' © o o T-H CO CO CO oo oo o IO CO CO CO IO CO CM Oi o CO oo Oi 35 IO CO CD © CM CM CM CM CM CM CM CM CM* © IO CM IO CO CO T-H oo co CM CO T-H IO CM CM r^ Oi CO CO Oi T-H CO t-H 00 © 00 oo io CM CM O O o O o o © © -$ oo CD 00 CO ■>* IO CO 1^ CO T-H co IO oo CO CM Oi CO l> i> CM ^8 O O o o o O d © IO io ID CO CO CO O O T-H o Oi o T-H IO IO o T— 1 T-H IO O T-H © ■HH T-H © © © © o o o o o o © © CO o O CM CO t-H CO CO CD 00 T-H CM CM CM CD IO CO CD CO oo CO CM CO T-H CO CM CM T-H T-H CO "* Tp © CM CM Oi Th ■Hr CO oo oo Tjr © T-H IO CM T-H > > HJ T3 383 384 W. B. NEVENS averaging the results secured in the analysis of the complete samples C2, C3, C6 and C7. It is believed that the latter average more nearly expresses the actual composition of the commercial cottonseed meal used, for the following reasons: (a) These samples, i.e., C2, C3, C6 and C7 show the best agreeing results throughout. The two parts of sample CI agree well in the amount of arginine nitrogen, but show a considerable differ- ence in the amounts of amino nitrogen, non-amino nitrogen and histidine nitrogen. The totals of the nonprotein nitrogen plus the protein nitrogen are considerably below the average of all the samples. In sample C4 the non-amino nitrogen is particularly low, this value being one of the principal factors contributing to the noticeably low nonprotein plus protein of this sample. Sample C5 is omitted from the average partly on account of the non-agreement of its arginine nitrogen and histidine nitrogen values. Of the latter values, one is 3 per cent above the average of all samples, (b) The second average, i.e., of samples C2, C3, C6 and C7, includes values which are most free from obvious errors. In making the determinations in the case of sample C4, two determinations were lost, and in the case of sample C5, one determination was lost. While the results obtained for sam- ple C8 agree fairly well throughout with themselves and with the average, the results are consistently high, and it is excluded from this average on the grounds of the totals obtained, which are obviously too high. Nonprotein nitrogen. The first section of tables 2 and 3 shows the amount of nitrogen removed in the preliminary extractions with absolute ether, absolute alcohol and trichloracetic acid. While the absolute ether in the cold is used primarily to remove the lipins, such as the oils, waxes, etc., it also dissolves various amounts of other substances, such as coloring matters, and at the same time a small amount of nitrogen. The thorough extrac- tion with absolute alcohol following the treatment with absolute ether presumably completes the extraction initiated by ether. The alcohol removes somewhat more nitrogen than the extrac- tion with ether. The bulk of the nonprotein nitrogen accounted for, about 89 per cent of the total, remains, however, in the tri- THE PROTEINS OF COTTONSEED MEAL 385 chloracetic acid extracts after precipitation of the proteins by colloidal ferric hydrate and the removal of the precipitate by filtration. That the nitrogen determined in these latter extracts is not protein nitrogen is apparent from the work of Van Slyke, Vinograd, Vilchur and Losee (13), Hill (14), Wolff (15), and others. It seems from the study of the character of the nonprotein nitrogenous constituents of feedingstuffs by Grindley and Eckstein (16) that the forms of nitrogen represented in this classification consist principally of those forms naturally re- sulting from the cleavage of the proteins upon hydrolysis and therefore could not interfere in the determination of the charac- teristic chemical groups of the proteins were they not removed in the preliminary extractions. Neidig and Snyder (17), who recently determined the proportion of nitrogen in the form of ammonia in the ether extracts and alcohol extracts of different kinds of silage, found that from 28.1 per cent to 100 per cent of the ether extract nitrogen consists of ammonia nitrogen, while from 14.2 per cent to 23.4 per cent of the alcohol extract nitrogen is yielded as ammonia nitrogen. This indicates that only a part of the nitrogen soluble in ether and alcohol would appear in the ammonia fraction were it not removed previous to hydrol- ysis. Therefore, the removal of the nonprotein nitrogen at this point avoids possible complications in the further prose- cution of the analytical procedure, and it is believed that the accuracy of the further determinations has been increased over that of previous methods by the removal of the nonprotein nitrogen before hydrolysis of the proteins. The total amount of the nonprotein nitrogen present in cottonseed meal found by the method used amounted to 6.106 per cent of the total nitrogen contained in the feedingstuff. Results of the Van Slyke analysis. It is a matter of common knowledge that one of the important sources of loss in the analy- sis of proteins by methods involving the employment of acid hydrolysis is the formation of an insoluble black substance called humin. The term melanin is also applied to this substance, on account of its supposed relationship or similarity to the 386 W. B. NEVENS naturally occurring body pigments. The amount of humin formed in acid hydrolysis of the proteins is greatly increased by the presence of carbohydrates, as shown by Gortner and associates (18, 19), Hart and Sure (20), and Osborne, Van Slyke, Leaven- worth and Vinograd (21). A part of the humin formed, how- ever, remains in solution in the hydrochloric acid, and is termed soluble humin. In these experiments the soluble humin which is adsorbed by the lime used in neutralizing the hydrolysate when determining ammonia nitrogen, carried with it a larger amount of nitrogen than the insoluble humin. The sum of the insoluble plus the soluble humin nitrogen found amounts to 6.589 per cent, which constitutes no inconsiderable error, since at present it is im- possible to determine the character of the nitrogen discarded in this form. It may be noted by referring to table 1 that the amount of soluble humin nitrogen in the first four samples is considerably greater than in the succeeding four. Possibly this is due to a slight variation in the analytical procedure. In the case of samples CI to C4, inclusive, it was necessary to add 75 to 90 cc. of calcium hydroxide in order to neutralize the hydrochloric acid before distillation of ammonia. About 20 cc. in excess were then added. With the next four samples evaporation of the acid hydrolysate in vacuo was continued longer in order to drive off a greater proportion of the hydrochloric acid. In conse- quence, only 30 to 40 cc. of calcium hydroxide were necessary to effect neutralization, and in these cases only a small excess, about 10 cc, of calcium hydroxide was added. The hypothesis is put forward that the presence of a large excess of calcium hydroxide during the distillation of ammonia may result in the adsorption of some amino acid nitrogen which is incompletely removed in the subsequent washing of the sticky mass. When it is recalled that the proteins of cottonseed meal form approximately 43 per cent of the feedingstuff, the amount of nitrogen in the humin resulting from the hydrolysis of these proteins, as determined in these experiments, is not excessive when compared to the amounts obtained in the hydrolysis of THE PROTEINS OF COTTONSEED MEAL 387 pure proteins by Van Slyke (8) , some of whose results are shown in the accompanying table. The amount of nitrogen recovered as ammonia was quite constant in all the samples. Little can be said in regard to the significance of this fraction, aside from the fact that the propor- tion of the total nitrogen of cottonseed meal which appears as ammonia is quite in harmony with that of other feedingstuffs. A particularly characteristic feature of the amino acid con- tent of cottonseed meal is the remarkably high content of argi- nine. This is much higher than that found in any other feeding stuff so far examined, with the exception of peanuts, although it is not so high as that found in some other vegetable proteins. Van Slyke (8) found 27.05 per cent arginine nitrogen in edestin TABLE 4 Amounts of humin nitrogen in pure proteins expressed in percentage of the total nitrogen of the protein PROTEIN AND DESCRIPTION Gliadin from wheat Edestin Fibrin (Merck's) Oxyhemoglobin ("pure, crystalized") . Dog's hair HUMIN NITROGEN per cent 0.86 1.83 3.43 3.60 7.35 while Nollau reported that hemp seed, peanuts, black walnuts, and hickory nuts have an arginine nitrogen content of more than 20 per cent. In the Van Slyke procedure the only amino acids deter- mined by direct analysis are arginine and cystine. Just how much importance may be attached to the results obtained for the latter is questionable, even though the values found in the different samples do not vary widely. These values, however, probably fall short of the true value, due to losses in the deter- mination of cystine. Van Slyke (8) has shown that boiling cystine for sixteen hours with hydrochloric acid resulted in the conversion of one-half of its nitrogen into forms not precipitable by phosphotungstic acid. Since in the analytical procedure de- scribed above, hydrolysis of the proteins was carried out by 388 W. B. NEVENS boiling them with 20 per cent hydrochloric acid for fifteen hours, it is probable that much of the cystine was destroyed during that reaction. If it is assumed that the correct value for cys- tine should be double that actually obtained, then the total nitrogen of the bases would amount to 31.75 per cent of the total nitrogen of the feedingstuff. The values given for histidine and lysine are somewhat vari- able among the different samples. These variations are likely due in large measure to the indirect method used in their deter- mination, since slight errors in any or all of the three direct determinations of arginine nitrogen, cystine nitrogen and the total nitrogen of the bases are doubtless all reflected at these points. The content of mono-amino acid nitrogen of cottonseed meal is considerably less than that found in other feedingstuffs, possibly due to the greater proportion of the total nitrogen which is formed by the basic amino acids. One of the interest- ing features of the results of the Van Slyke analysis of the pro- teins of cottonseed meal is brought out in the summation of the ammonia nitrogen, the nitrogen of the bases, mono-amino acid nitrogen, and non-amino acid nitrogen, the four groups which represent the total content of strictly amino acid nitro- gen as determined by this method. The sum of these is 83.132 per cent. While this sum is not so great as that in the case of some other feedingstuffs or of animal proteins, as determined by previous investigators employing the Van Slyke method of analysis, it is a much larger amount than it was possible to secure in most cases from comparable sources by the methods of isolation and purification employed by the earlier investigators. Thus the tabulations of Lusk (22), combining the results of Osborne and associates in this field, show the maximum amino acid content of zein of maize, to be 88.87 per cent and that of gliadin of wheat as 85.68 per cent, but in the majority of cases the sum of the nitrogen content of the amino acids actually isolated from vegetable proteins ranges from 50 to 65 per cent of the total nitrogen of the protein. THE PROTEINS OF COTTONSEED MEAL 389 Uncharacterized nitrogen lost in analysis. In the various steps of the analytical procedure small amounts of nitrogen of unknown character are included in residues and solutions which are discarded. In general, these have been disregarded by workers in other laboratories, especially those losses occurring at points indicated in table 2 by the last four of the subheadings included under the heading " Uncharacterized nitrogen lost in analysis," but in this laboratory the nitrogen discarded at each of these steps has been determined. Under the above mentioned headings, it is shown that, on the average, only 0.492 per cent of the total nitrogen remains in the residues insoluble in strong alkali, or in other words 99.508 per cent of the total nitrogen of the f eedingstuff is extracted as a result of the method employed, and that in individual cases as much as 99.78 per cent of the total nitrogen present was removed. As previous workers failed to isolate the proteins from the feedingstuff before hy- drolysis, it is not established that part of the insoluble residue discarded in their methods did not include some nitrogen in the form of non-hydrolyzed protein although this does not seem highly probable. It is believed, however, that the nearly complete extraction of the proteins before hydrolysis lends to the accuracy of the method by facilitating hydrolysis and in reducing the amount of humin. The largest item of loss occurs in the residue which remains after dissolving the precipitate of the bases in the amyl alcohol- ether mixture. This, presumably, is soluble humin which has not been adsorbed by the lime in the determination of ammonia, and fouls the solution at this point. This difficulty was also encountered by Menaul (23), who employed a preliminary precipitation with phosphotungstic acid in boiling solution for the separation of the humin and ammonia before the precipita- tion of the bases. In the present investigation, very little of the soluble humin appeared when the bases were precipitated, in most cases the precipitates being free from black particles. Washing with alternate portions of amyl alcohol-ether and water and then taking up the residue and washing thoroughly with water seemed to have little effect in reducing this source 390 W. B. NEVENS of loss. A considerable portion of the nitrogen lost is soluble in the amyl alcohol-ether mixture, while smaller losses occur in the residues resulting from concentration of the solutions of the bases and filtered from the bases. Presumably, the second, third and fourth items of loss include some nitrogen which should be credited to the bases, but the character of this nitrogen was not determined. If these losses can be reduced, the total nitrogen of the bases of cottonseed meal may be found to be somewhat greater than the amount here reported. Total nitrogen accounted for. Summation of the nitrogen found in the various fractions of the protein molecule together with that in the unavoidable losses in the procedure gives totals which average 98.75 per cent. While the use of the Van Slyke method of analysis has enabled others to account for as great a proportion of the nitrogen of feedingstuffs, it is doubtful, for reasons pointed out below, if their results give as accurate a picture of the distribution of nitrogen in feedingstuffs as is obtained by the procedure employed in the present investiga- tion. Physiological significance of the basic amino acids. Our knowl- edge of the physiological role of arginine and histidine has been enhanced by the studies of Ackroyd and Hopkins (24). Em- ploying rations in which the nitrogen was provided in the form of hydrolyzed casein from which these two amino acids had been removed by precipitation according to the method of Kossel and Kutcher, it was found that rats receiving these rations declined rapidly in weight, but that when either amino acid was returned to the ration, loss in weight was prevented and some growth ensued. The investigators suggest that possibly either of these amino acids may be converted into the other by the animal body. It was further observed that when argi- nine and histidine are removed from the ration, the excretion of allantoin, which is the main end product of purine metabolism in the rat, was lowered. Subsequent experiments proved that the falling off of allantoin excretion was not due to lowered me- tabolism. From these observations and from the fact that the arginine, histidine and guanine molecules have similar structural THE PROTEINS OF COTTONSEED MEAL 391 relationships, it was concluded that possibly one of the functions of arginine and histidine is to furnish the raw material for the purine metabolism of the animal organism. The above conclusion regarding the importance of arginine in purine metabolism is given added weight by the findings of Myer and Fine (25) regarding the creatine content of muscle. Differences of as much as 2.5 per cent in the creatine content of muscle were noted as a result of feeding rations high and low in arginine. That cystine plays an important part in nutrition has been brought out by several investigators, among them Osborne and Mendel (26). The latter obtained adequate growth by the addition of cystine to rations containing 9 per cent of casein, on which growth had been limited. Geiling (27), working in this laboratory, concluded that cystine seems to be necessary for the maintenance of adult mice. The importance of cystine to the animal organism is admirably set forth by Matthews (28). In the intermediary metabolism of the body, that is, the metabolism of the tissue, sulphur probably plays a very important role. This is shown not only by the fact that it is absolutely necessary for the continued existence of the body, as necessary as nitrogen or any of the other elements, but also by the fact that it is one of the most labile elements of the protein molecule. No other element is split off from the proteins with greater ease than this. It is, indeed, the labile element par excellence. Moreover, cysteine, which is one of the amino acids, readily oxidizes itself. It is a reducing body. It oxidizes spontaneously and there are many points in its oxidation which strongly resemble the process of respiration. Thus the most favorable concen- tration of hydrogen ions for the oxidation of cysteine is the same as that in protoplasm; both cysteine and protoplasm are poisoned by many of the same substances, such as the nitriles, the cyanides, acids, and the heavy metals; their oxidations are catalyzed or hastened in the same manner by iron, arsenic and some other agents. For these reasons it has been suggested by Hefter and the author that there is more than a superficial connection between the oxidation of cysteine and the respiration of the cell. 392 W. B. NEVENS The necessity of lysine for growth has been conclusively dem- onstrated by Osborne and Mendel (29). When gliadin of wheat, which contains only a minute amount of lysine, formed the sole source of protein in the rations of rats, the live weight of the animals was maintained over long periods, but normal growth could not be secured. When lysine was added to the rations, normal growth occurred. In other investigations (30), in which zein of maize was used as the source of the protein, it was found that a rat could be maintained at an almost constant weight of 50 grams for a period of one hundred and eighty-two days when tryptophane was added to the extent of 3 per cent of the zein. The further addition of lysine induced normal growth. Further study (31) of the necessity of lysine in the ration convinced these investigators that about 2 per cent of the protein of the ration must consist of lysine in order to pro- mote normal growth in the rat. Osborne and Mendel (32) also demonstrated the necessity of lysine for the growth of chickens. In view of the essential role which the basic amino acids play in nutrition as brought out above, it is reasonable to assume from a survey of the analytical results of cottonseed meal secured in this investigation that the proteins of this feedingstuff have a high nutritive value. The combined arginine and histidine content of cottonseed meal is greater than that of any other feedingstuff so far analyzed with the exception of the peanut. This feature alone is of great importance in view of the fact that arginine and histidine seem to be interchangeable in nutri- tion. While the lysine content cannot be said to be exceptional in any particular it seems apparent from the above discussion, that the combined proteins of cottonseed meal contain sufficient amounts of both cystine and lysine to render them adequate for nutrition. Comparison with previous analyses of cottonseed meal. As shown in the introduction, there are but few determinations of the chemical composition of the proteins of cottonseed meal. The earliest studies were made upon one of the isolated pro- teins, the globulin or "edestin" of cottonseed, the results of which can not well be compared to analyses of the combined THE PROTEINS OF COTTONSEED MEAL 393 proteins, since, as previously mentioned, the globulin contains but 42.3 per cent of the total nitrogen of cottonseed meal. In the accompanying table 5, the values secured by two previous investigators who made analyses of the combined proteins of cottonseed meal are brought together for comparison with those obtained in this investigation. It is evident from the results presented that there is a general agreement between the three sets of values, but that there are considerable differences in several important particulars. As pointed out in the introduction, Nollau (10) calculated his re- sults upon the total nitrogen content of the hydrolyzed solution after filtering off the solid residue insoluble in hydrochloric TABLE 5 Distribution of nitrogen in cottonseed meal as determined by different investigators (Results expressed in percentage of the total nitrogen of the feeding stuff) INVESTIGATOR HUMIN N AMMO- NIA N ARGI- NINE N CYSTINE N HISTI- DINE N LY- SINE N AMINO ' N IN FIL- TRATE FROM BASES NON- AMINO N IN FIL- TRATE FROM BASES TOTAL N ACCOUNT- ED FOR Nollau 6.27 7.78 6.58 14.06 10.45 9.49 12.77 19. '52 18.74 2.74 0.65 0.91 7.57 5.47 7.40 1.94 4.78 3.81 45.02 42.82 40.12 7.49 5.43 2.68 97.48 Grindley 96.90 Nevens 98. 75 1 1 Includes 9.03 per cent N removed in preliminary extractions plus unchar- acterized nitrogen lost in method of analysis. acid. This means that all of his calculations are too high, since a part of the nitrogen of the sample was undoubtedly dis- carded in the solid residue. The value of 6.27 per cent humin nitrogen reported by Nollau must, therefore, represent the soluble humin nitrogen, which is nearly as large a value as that obtained by the writer for the sum of the insoluble humin nitrogen plus the soluble humin nitrogen. The amount of soluble humin .nitrogen found by the writer was but 3.89 per cent. Compared to the total humin nitrogen found by Grindley, et al. (9), the amount of total humin nitrogen as determined by the writer was 1.19 per cent less. The reduction of the humin nitrogen has no doubt been an important contributing 394 W. B. NEVENS factor in the present investigation in securing somewhat higher values of the basic amino aids. In view of the known effects of acid hydrolysis of the proteins in the presence of carbohy- drates, as already pointed out, it is reasonable to assume that the smaller amount of nitrogen discarded in the form of humin in these experiments than in those of Grindley may be attributed to the more complete separation of the proteins from the carbo- hydrates before hydrolysis. The method of analysis of the proteins after hydrolysis by hydrochloric acid, as employed by Grindley et al., was similar to that employed by the writer, the main point of difference between the complete procedures being in the omission by the former workers of the extractions previous to hydrolysis. At just what point the 6.106 per cent of nonprotein nitrogen re- moved by the writer in the preliminary extractions might appear were it not so removed, is not clear. However, the sum of the ammonia nitrogen, amino nitrogen and non-amino nitro- gen in the nitrate from the bases, obtained by Grindley et al, is 6.414 per cent greater than the sum of the corresponding values obtained by the writer, so it is possible that these three forms of nitrogen as reported by the former comprise some nitrogen not derived from the proteins as such. It is evident from the table that the nitrogen of the bases as found by Grindley and his coworkers are in much closer agree- ment with those obtained by the writer than those reported by Nollau. The latter 's figures for arginine are obviously too low, while his cystine values are more than four times as great as those of Grindley et al. and three times as great as those of the Writer. Accordingly, the lysine nitrogen values as calculated by Nollau are correspondingly too low. The values for the total nitrogen of the bases as found by the three investigators in the order given in the table are as follows: 25.02 per cent, 30.42 per cent and 30.84 per cent respectively, the last being nearly 0.5 per cent higher than previous determinations. The total nitrogen accounted for in the three reports is like- wise shown to be 97.86 per cent, 96.90 per cent and 98.75 per cent. The greater amount in the last case is evidently due in part THE PROTEINS OF COTTONSEED MEAL 395 at least, to the inclusion of the determinations of the uncharac- terized nitrogen lost at points were unavoidable losses occur in the method of analysis. These losses were not determined by the first two investigators. Comparison of the distribution of nitrogen in cottonseed meal with that in other feedingstuffs. A comparison of the results of analysis of the proteins of cottonseed meal, as discussed above, with those obtained by Hamilton, Grindley and Nevens (33) for alfalfa hay, oats and corn is of value in studying the relative nutritive value of the proteins of these feedingstuffs, as well as the applicability of the general method of analysis to feeding- stuffs which vary widely in composition. In the analysis of oats and corn an additional preliminary extraction, which involves the use of hot trichloracetic acid, is employed to remove the starch. This extraction is not necessary in the case of cottonseed meal and alfalfa hay on account of the absence of starch in the former, as stated by Withers and Fraps (34), and the relatively small amount of starch in the latter. The first point of interest in contrasting these feedingstuffs, as may be noted by reference to table 6, is their content of non- protein nitrogen. Oats contain more than twice as much non- protein nitrogen as cottonseed meal, while alfalfa hay contains more than three times as much. Hart and Bentley (35) found that 23.5 per cent of the nitrogen of alfalfa hay is present in a water soluble form, while Grindley and Eckstein (16) found a value of 28.4 per cent for the same feedingstuff. The amount of total humin is greatest in the case of alfalfa, a natural result, since the proteins are more difficultly extracted from those feedingstuffs containing large amounts of crude fiber. The amount of humin in the case of corn is very small indeed, considering the high percentage of carbohydrates in this cereal, and compares very favorably with the amounts of humin re- sulting from the hydrolysis of pure proteins as shown in table 3. Cottonseed meal occupies a medium position in respect to the proportion of humin nitrogen. The most striking difference between these four feeding- stuffs is in their basic amino nitrogen content. Cottonseed § e t3 *a _-»j s es- se o -+j g ° ° l^ 1 r* M « < Eh O Eh joj pa^tmooora naSoj^ra jb^ox CM »o CM OS O o OS OS OS OS OO OS OS o !>• 00 OS DC s I" > i-3 -1 fc 4 CQ 55 -*! S* m w Eh Eh O 1 P m a M STS^p3UB ai[Aig nB A J° ajpsaj -f- uia^ojduon jtf:jox o CM C3 © o 1— 1 OS CM © CO OS CM 00 OS sasuq toojj a^tu^jg ut jsj piae omrara-uo^q; 1—1 1—1 © CO 00 CO CO i—i CM CM* sassq uiojj a^BJiqg ui ^ P! 01 * ouiray CO o CO co i— i CO CM l-H naSoj^ic anisi£ r j CO T-H 00 cnJ O © CM oi © oo CO uaSoi^iu auipt^sijj r-H co o> CO CO Os U5 iM CO oo OS CO naSoj^iu aui^sXQ os OS o OS © CM O i— i CD O os d uaSoj^iu auraigjv CO OS i— I i— I CM t- oo CO CO oo i-t naSoj^ra mnommy CO co 03 i— 1 i— 1 CD CO Os T-H d naSoj^ra uiranq ajqnjog 1—1 oo CD i— I CO (M* o OS oo CO uaSoj^iu uininq ajqTqosuj o OS CO CO CO © co co OS OS CO CM S5 H o « Eh g w O « g o S3 uaSoj^iu niaiojdnott iu}\ox o OS o Os CO OS j £ o K K 63 a 63 £ t> 3 fc; 01 P < P- o ►3 ►3 fk ►J 63 63 Ph pj m < 5« 85 63° O o O 63 111 s *a O O K o RAT NUMBER H 63 O Hi P o P2 Eh 63 2 Pi Z 3-1 4 H 2 P Z O *H 3 H a go O M n H < h. k gig < m O H z o H O O K O En ° 5 H °S fc o m ■ c n & n O O 8 aj O d 5 « a ag oh P5 K grams grams mgm. mgm. mgm. per cent per cent 1 2 + 3 Cottonseed meal 98 9.50 124.8 54.5 25.3 64 30 2 2 + 3 109 8.62 107.1 38.2 30.4 64 19 3 2+3 127 11.99 169.6 91.3 28.0 71 44 Average 111 10.04 133.8 61.3 27.9 66 31 1 4 + 5 Cottonseed meal + 114 10.83 125.9 52.9 30.6 67 26 2 4 + 5 alfalfa hay 126 11.06 131.7 58.1 35.7 71 29 3 4 + 5 159 15.68 185.0 89.5 33.7 67 34 Average 133 12.52 147.5 66.8 33.3 68 29 2 6 + 7 Cottonseed meal + 136 9.82 124.3 45.6 38.2 63 21 3 6 + 7 alfalfa hay + corn 186 14.01 173.0 72.5 39.4 65 27 Average 161 11.92 148.7 59.0 38.8 64 24 4 2 + 3 Corn 118 7.75 118.9 34.6 22.8 48 15 5 2 + 3 108 6.98 105.9 24.7 27.1 49 8 5 2 + 3 115 6.92 104.5 34.1 20.3 52 16 Average 114 7.22 109.8 31.1 23.4 49 13 4 4 + 5 Cottonseed meal + 129 9.81 141.4 59.5 24.9 60 30 5 4 + 5 corn 118 8.30 127.8 43.6 29.2 59 21 6 4 + 5 124 8.11 120.2 48.9 22.3 60 28 Average 124 8.74 129.8 50.7 25.5 60 26 5 6 + 7 Cottonseed meal + 139 10.90 136.9 49.3 34.9 62 18 6 6 + 7 alfalfa hay + corn 141 11.20 146.8 64.8 25.1 61 28 Average 140 11.05 141.9 57.1 30.0 61 23 7 2 + 3 Alfalfa hay 104 9.42 99.4 37.2 21.4 60 16 8 2 + 3 103 8.69 92.9 33.7 20.0 58 11 9 2 + 3 121 12.55 128.1 61.2 29.4 70 30 Average 109 10.22 106.8 44.0 23.6 62 19 7 4 + 5 Alfalfa hay + corn 115 12.64 149.1 56.8 23.3 54 20 8 4 + 5 118 14.08 175.8 78.9 23.3 59 28 9 4 + 5 134 12.51 166.6 70.6 32.7 62 29 Average 122 13.08 163.8 68.8 26.4 58 26 8 6 + 7 Alfalfa hay + corn 142 11.81 152.6 65.0 27.0 61 26 9 6 + 7 + cottonseed meal 154 13.06 184.0 76.1 38.0 62 29 Average 148 12.83 168.3 70.6 33.0 61 27 577 578 W. B. NEVENS Utilization of proteins. In the summary of results shown in table 8 two methods of calculating the utilization of the proteins fed are included for comparison, although, for reasons discussed above, the second method, namely, the utilization of the absorbed nitrogen for both maintenance and growth, is employed in the discussion here. The utilization of the nitrogen absorbed from a cottonseed meal ration containing 10 per cent of crude protein was found to be 66 per cent, using the average results of six metabolism periods with three rats. The utilization of the proteins of alfalfa hay was found to be only slightly less than that of cottonseed meal, namely, 62 per cent. When these two feeds were combined in such proportion that each furnished about an equal amount of digestible protein to the ration, very interesting results were secured, indicating a slight supplementary effect of the proteins from these two sources. This effect was not pronounced, the utilization percentage being 2 per cent above that of cottonseed meal alone and 6 per cent above that of alfalfa hay alone. It is noted that during the periods when the cottonseed meal alfalfa hay ration was fed, greater quantities of feed were consumed and larger amounts of nitrogen were absorbed than with either the cottonseed or alfalfa hay rations alone, which may in some unknown way have operated in effecting a more efficient utilization of the nitrogen, although the same conditions hold true in the case of both groups of rats which received either the corn or the alfalfa ration during two periods and were then changed to rations containing proteins from both sources. It was found that the proteins of corn were utilized the least efficiently of those of the three feedingstuffs compared. When corn was combined with cottonseed meal or with alfalfa hay the resulting utilization coefficients tended toward a mean of the utilization coefficients secured with these feedingstuffs when fed alone, but were nearer that of the feed other than corn. For example, the utilization coefficients found for the corn and cottonseed meal rations were 49 per cent and 66 per cent respectively, the mean of these two being 57.5 per cent, but THE PROTEINS OF COTTONSEED MEAL 579 the utilization coefficient found for the cottonseed meal-corn ration was 60 per cent. Possibly this represents a slight supplementary relationship. The results obtained with the ration in which cottonseed meal, corn and alfalfa hay were combined were remarkably uniform. Of the twelve values obtained with rats receiving this ration during two metabolism periods each, the lowest value was 57 per cent and the highest 67 per cent, the average of all being 63 per cent. The combination of the proteins from three dif- ferent sources failed to indicate any farther supplementary effect of the proteins. The high nutritive value of the proteins of cottonseed meal manifested by these experiments is in substantial accord with the conclusions of Richardson and Green (1), Osborne and Mendel (3, 4, 5, 6) and McCollum and Simmonds (6). They do not seem to be in harmony with the findings of Hart and Hum- phrey (8) who studied the utilization of the proteins of cottonseed meal for milk production, but since growth and milk production are dissimilar functions an absolute comparison of the results of the two experiments is not valid. Correlation of chemical composition with nutritive value. In seeking for an explanation of the differences in the nutritive value of the proteins of these feedingstuffs based upon differences in their chemical makeup, it is evident first of all that their nutri- tive values do not vary so widely as the analytical data at hand would indicate. For example, the differences found between the utilization of the proteins of cottonseed meal and alfalfa hay was but 4 per cent, while from an examination of the data in table 5 of a preceding paper, it is apparent that cottonseed meal contains more than twice as much arginine nitrogen and nearly twice as much histidine nitrogen as alfalfa hay, while the latter contains more than three times as much nonprotein nitro- gen as the former. Several theories may be advanced in explanation of this ap- parent inconsistency. In the first place, alfalfa hay is shown to have a lower digestibility than either cottonseed meal or corn. There is no evidence to preclude the possibility that the char- 580 W. B. NEVENS acter of the nitrogen absorbed from alfalfa hay differs qualita- tively from that remaining in the undigested residues. Judging from the ease with which tyrosine is split off from proteins in tryptic digestion in vitro, it is possible that the absorbed nitro- gen contains a greater proportion of amino acids essential to the body than the unabsorbed portion. Further, the sterochemical arrangement of the amino acids in the protein molecule may affect the extent to which the digestive enzymes are able to cause hydrolysis of the different proteins. A second considera- tion is the possible interchangeability of the various forms of nitrogen in nutrition, as already pointed out in the case of arginine and histidine. To what extent the nonprotein nitrogen of alfalfa hay is utilized in maintenance is problematical, but since there is no reason to doubt that the degradation products of crude protein are able to serve in this capacity, it is possible that a large part of the absorbed nonprotein nitrogen fulfills some of the requirements of the animal body. It is reasonable to assign the higher content of the basic amino acids of cottonseed meal as the reason for its superiority over the proteins of alfalfa and corn. In the case of the last mentioned feedingstuff, there is the additional factor of a comparatively low lysine content to be considered, although from the studies of Osborne and Mendel concerning the lysine requirements for growth, a lysine content of 2.2 per cent of the protein appears to be ample for normal growth. In the absence of further information respecting the char- acter of the mono-amino acid and nonprotein nitrogen content of these feedingstuffs, a detailed picture of which the Van Slyke analysis does not include, correlations between the chemical composition and nutritive value of the proteins of feedingstuffs can proceed little beyond the realm of the functions and rela- tionships of the basic amino acids. Comparison of feed consumption with that of farm animals. It was noted during the course of this investigation that the rats consumed an enormous amount of feed in proportion to their live weights. In some few cases the amount of air dry feed eaten daily was equivalent to as much as 10 or 11 per cent of the TABLE 9 Comparison of feed consumption by albino rats with that of farm animals FEED SPECIES OR BREED LENGTH NUM- BER OF AV- AV- EATEN DAILY AND OF FEEDING FEEDS IN RATION ANI- ERAGE ERAGE PER 100 CLASS OP ANIMAL PERIOD MALS FED AGE WEIGHT POUNDS LIVE WEIGHT days days pounds pounds 28 Corn + oats + alfalfa 10 227 854 2.2 Percheron hay fillies* 28 Corn + oats + alfalfa hay 10 683 1484 1.8 35 Corn + linseed meal + 4 978 2.5 Hereford steersf clover hay 28 Corn + linseed meal + 4 1466 1.5 clover hay 30 Corn + bran + linseed 4 365 472 2.9 Jersey heifers£< 90 oil meal + alfalfa hay Corn + bran + linseed oil meal + alfalfa hay 4 730 839 1.8 30 Corn + bran + linseed 4 365 656 2.4 Holstein J oil meal + alfalfa hay heifers % 1 f 90 Corn + bran -f- linseed oil meal -f alfalfa hay 4 174 730 1112 38 1.8 6.0 495 105 128 320 3.8 2.4 Sheep** { 45 127 2.1 1.1 42 Cottonseed meal -f- corn + alfalfa hay + syn- 9 129ft 8.4§§ Albino rats . . .< thetic mixture 21 or 18 per cent protein 17 175 4.6§§ more*** 200ft * From Bui. 192, 111. Agr. Exp. Sta. f From Bui. 197, 111. Agr. Exp. Sta. Data concerning "Full-feed lot." t From Nebr. Agr. Exp. Sta. Unpublished manuscript. Data concerning "Heavy fed groups." § From Henry and Morrison, Feeds and Feeding, 15th ed. p. 569. ** Calculated from data of Weiske as quoted by Armsby, The Nutrition of Farm Animals, p. 432. ff Grams. §§ Grams per 100 grams live weight. *** From Osborne and Mendel. Protein Minima for Maintenance. Jour. Biol. Chem., 1915, xxii, 241. 581 582 W. B. NEVENS live weight. It seemed of interest to compare the feed consump- tion of albino rats with that of farm animals. Such a comparison is made in table 9. The tabulations of horses and cattle include in each case two entries of the same group of animals at dif- ferent ages and weights. It is known that the horses and cattle were restricted in the amount of concentrates consumed but were offered roughage to practically the limit of their appetites. Hence a comparison of the feed consumption of these animals with that of the first group of rats, which were the ones concerned in this investigation, is warranted, but the data are indicative only. Data for the amount of feed consumed by the swine, sheep and second group of rats is not at hand. It is evident from the data presented that the rat is a voracious eater, even when receiving rations comparable to those of farm animals. A rough approximation places the relative amounts of feed eaten by rats as about three times that of various breeds and classes of farm animals, if swine be excepted. The fact is also brought out, as has previously been noted by others, that the young aDimal consumes much more feed in proportion to live weight than when older and heavier. Cottonseed meal probably not toxic to albino rats. None of the rations containing cottonseed meal seemed to exert any harmful influence upon the rats consuming it. Three of the animals received continuously for 7 weeks rations containing from 7.7 per cent to 23.9 per cent cottonseed meal with no evidence of toxic symptoms but remained in excellent nutritive condition. This observation is in agreement with those of Richardson and Green (1) and Osborne and Mendel (5). SUMMARY OF THE DISCUSSION OF THE NUTRITIVE VALUE OF THE PROTEINS Evidence is presented to show that metabolism experiments with the rat as a subject can be carried out with a high degree of accuracy. Different methods of expressing the nutritive value of the proteins of f eedingstuffs are discussed. The plan of employing the THE PROTEINS OF COTTONSEED MEAL 583 results secured in a preliminary and final metabolism period during which the animal receives a nitrogen free ration, for the calculation of the percentage of the absorbed nitrogen utilized, is favored. In comparing the nutritive value of the combined proteins of the f eedingstuffs cottonseed meal, alfalfa hay and corn, it was found that when one of these feeds furnished the sole source of protein in rations containing 10 per cent of crude pro- tein, the utilization of the proteins for the growth of albino rats was, in the order in which the feedingstuffs are named, 66 per cent, 62 per cent and 49 per cent, respectively. When rations containing these feedingstuffs, combined in various ways, but with each feed furnishing an equal amount of digestible protein, were fed, there was evident no clear cut sup- plementary effect of the proteins of one feed upon another, except in the case of the combination cottonseed meal and alfalfa hay, which showed a slight effect. No symptoms of toxicity were noted as a result of feeding rations containing cottonseed meal over a period of seven weeks. When suitable rations are provided, the albino rat consumes an enormous amount of feed * in proportion to its live weight. The writer desires to express his appreciation of the assist- ance of Dr. H. H. Mitchell in outlining the method used in this investigation and for many helpful suggestions. He is also indebted to Dr. H. S. Grindley for his encouragement and gen- eral supervision of the thesis problem. 584 W. B. NEVENS in ' ^trt-w S ,-v g « !> tH i-H rH co o i-H ^H CN I> iO »o 00 © t^ !>• w ►J !•. t^ t» 1> t^ 1> CO 00 l> co co co © © © © © s i, y w s. ^_^ to H a a "■^NiS e CO © co co 00 ■* co os NMO o rH CN CO CO ' >-• d 8.2 © lO co co Id Ui co io HI liJIO lO lO lO US id i o S g AB- SORBS N RE TAINE 1-H 1—1 "# CM o r~ CO o ■* 00 CN i-H CO O iO N CO CN rH CO CN >H ^h CO CN CO CO CO CN i-H CO CN 3 S2 3§£ s CO rh t— i co »0 rH O CO t> CD CO © CD to N (O a CO CO i> co CO CD t^ co CO 1> CO t>- © © © © S3 i Q 1 H H H g* 00 00 lO t-H CD 1-H O 1> 1> i> IQ CO S n a « z co o 00 CN «0 ■*' 00 OS i-H l^ © t>^ Sis < g lO TH 00 W5 CO os ^H CO 00 lO Ttl © 53 EE £"* ^ ^ 'a? S z <»_ g 00 CO r-l T^ CO O CN OS t>- CO CN OS tH CN) tH I> t-h t-H CO id e g CN CO OS O C5 Tt< i-H CO CN OS T-H t> CD 03 00 © rJH l> 00 t^ CN fe- ci CN CN CN CN CN co IM N CO CN O OS CO © © ■* w g l-H tH i-H W "S © © © < s Eh © 3 gaga g 00 t» T-i © CO 00 o od t^ CN CN CO CN CO 00 N H TjH O CO I> © © s I> t> 1> CO CO oo CO CD OS 1> 00 © X o 9 > 3- co CO i— 1 i-H 00 OS IC C5 tH 00 00 OS © ffi M N © 1-H lO 00 CN "*' g CO *o o CO "HH 1-H 00 © © © © © S s Q t»0 I— 1 T-l (M i-H i-H CN i-H CN CN i-H i-H CN ggil iM CO i— ( 00 t^ l> © OS r-H CO co »o © l> ■* i »o CO lO TiH lO TjH co co © 3ho5 OS 00 i—l O 00 CN © 1-H T^' i-H O © i—l i-H 1— 1 TH 1-H i-H i-H i-H P, h <» •3 w g co co CO os oo ■* co © o i-H CO CO -C 5 ° Z »-< O i-i CO O i-H "* i-H CN © CN CO I> 5" 1— 1 1—1 1-H T-H i-H i-H i-H i-H i-H t-H i-H i-H s " "5 l o CM O 1-H co © i-H ■* t-h CN © e ^ i-H i— ( t-l i-H 1-H i-H i-H i-H T-4 1-H i— 1 + + 0} ,_, ,_, S SB O a 'e3 a c3 CD 'O 2 CD r 1 - 03 CO ■oj a) (U CD -M K CD m a o CD PI o CD j CO ^ O 03 o * O c8 o °s o o O o o O aoiaaa (M (N CM co co co ^ <# ^ IQ IQ lO 0) CD CD CD PS t>0 &c bJO CJO 5§ i-H CM CO g i-H CN CO g i-H CN CO i-H CN CO g M 9 CD E? > > > > H < < < O t> t^ *~ **■ 1> os OS OS os OS OS OS os oo co co 00 00 CO oo oo 1^ rH CO t- l-H CO 1> t^ CD rH CO CO CO CO co co CD t^ CO os l-H OS o LO LO k5 >o CO "5 LO r^ t^ t^ 1> I> !>■ t^ i> CO CD CD CO CD t^ CO t^ CM CO rH OS b- CO LO O CO CO LO 1 O CN l-H O LO L0 OS CM o t> rH CO 1> OS lO (M OS rH LO CO t^ co 00 CO o CO CD CO OS CO i-H co CD CD CO CO CO co rH LO lO L0 rH rH rH rH co >o LO lO lO CD CD CO rH l-H I>. LO 00 (N CO t^ O O >o CO CO rH CO l-H CO CO rH tH lO rH rH os o »o 00 t^ l-H rH CO o (M co CM CD lO i-H O t^ OS go as rH LO GO o O d l> l> rH co CM rH T-H d lO o CO i-H rH CO co rH rH o CM LO LO rH i—i OS CM i>^ co d i-H O CO o O d o CD CO GO d CM b- o l-H t> OS GO CO t^. CO 00 t^ 00 co t^ CO t^ CO lO OS rH CO rH CM CO CD CD CO 00 o oo OS oo rH d t^ OS i-H CO t^ rH CO rH >o d CI o ,_i CM t^ CO LO CO rH CM CM IM CO O o t^ o LO 1> rH CO T-H M l-H CM T-H l-H i-H l-H l-H T-H ^ l-H i-H T-H l-H O CO t^ i-H rH rH CD OS iO o o LO CD oo I> rH co T-l 1— 1 rH *o lo rH CD rH o CO co OS 1Q rh CO O i> co d rH d co t^ t> t- CO CD CD os oo CO d co i> i— i ^ l-H CM t- CO GO OS OS GO LO OS t^ co T-H LO co LO LO (M O CO GO CO 00 l-H o T-H i-H o l-H CO T— ( CM co 03 co i-H t-H l-H i-H T-H i-H l-H T-H i-H l-H 1—1 i-H 1—1 i-H i-H T-H T— 1 CO rH > rd >> 4) *»H CD j 2 «S o T3 O w a d cd d r-l 03 03 CD O rt O c3 c3 rt u -+J %-H .+J <*H a a d CJ d CD «4H H-= J. ° "3 O w O u o o s rH O s £o o O o O o CO CO CD l> b- CM CM CM CO co CO rH rH rH LO >o LO a> 0) CD 0) CD CD bO u faO bO bO bfl (H bO CO 2 l-H > r> > > > < < ^ < < < 586 W. B. NEVENS >< Cor- rected coeffi- cient 00 © © CO 00 CO rH CN *# CD «5 N CO N "* 3 3 CO CO CO 1> t^. t^ CO CO »o 10 KJ IO IO IO i O S _ ^Sa s co ^ ^ NMO 1> HO rH ■<* © O IO CN ^^ h4S» O go KJIOUJ IO CO CO ^1 r^l ^t ^ ■^ ^ ^ "^ fl , o e jg ■ w H H 3BKS5 10 "* © rl S5 N CO CO CN O CO © 00 l^ &. t-1 CN CN CN CO CN CN rH CN CN CO rH 05 H O. 1 fc 1 fl S 3 g o o g rt OOO CO t^ CO CN K g fa pj!5 CO CO CO CO CO CO co -# CO -* CO CN t~ "a? 2 i «>_ 1 CO ■* K3 00 CO O 00 rH © 00 TH © © rH © © CO CN CO 00 CN "# 1> lO rtl CO CO CN CN CO CN CN CN CN CN CO "a ^ >3 iH CN O tH CN Ttl rH lO ^ 1> CO 1 '3' 3 °£ i O CN 00 CN 00 rH rH CN CN 00 CD fa 1 CN © 00 00 1> © © CN © 00 CO a r-^ T-t i-t J, g d s wit. CO © IO CO 00 1^ 00 00 CO CO © »o ■* rH CN »0 TfH © ^ CO IO 00 .-1 < s 00 b- 00 00 CO CO CO CO IO © 3 SaB* ■ T)H T« X Q BO rH V- 1 rH CN rH rH CN rH rH CN M Q d§*3 CN co 00 t> CN CO © CN © f~ i O O CO 1> t^ CO rH TH IO CN © H Cm "" w P, 8 © T-I © CD rH © © i-{ © 00 CO r-( r-l 7-i rH rH T-i Ch , Eh < 1 O CD CO TjH 00 © IO © ■^ CO © CO CO M< ■HH ■* © © rH © O CN G *. 1—1 rH 1— 1 r-l rH rH rH rH T-i 1—i 1-i ** ►3 a < K >o rw CO CD CO CO CO © © IO © 52 g a M 5: § CO CO CO CO ■* O © rH © © rH 8 rH i-t tH rH i-< rH rH rH r-t r-i i-l Co + + ,_, ,_, CD 03 $ j« fc S "3 5 T3 ri T3 03 ' >> >> 05 . $> , 03 c3 K 03 * CD -4- 02 ' ^3 ^3 i s 1 ° \z S 6 ^ « oj '3 c3 H-l aoiaad © co co t> I> CN CN CN CO CO CO CD CP CO n Kg bfl bC M bJD ■* K3 (O „ IO CO g 1> 00 © g I> 00 © Sh e» CD CU CU p S3 > > > > <* < < t- 1> t> t^ CO i> t^ l> 1> t>. co I> CM CO o CO CO I> lo O CO CO O CM 00 CM o CO OS oo OS i—l (Mi CM CM CM oo co CM CM co CM CM CM CM tH >o O W3 t~ CM co i-H t^ CM o Tjl CM CO lo IQ CO "3 L0 CO O CO L0 CO CO CO CO CO ^ CO OS CO co co CO C5 oo © CO t^ i-H OS GO CM LO i>. CO CO l> i> CO t- CO t~ lo CM t^ o ■* co •* CM co oo CM CM i— r T« -+l CO CO !> t^ CO CM OS o I> i-H os oo os r- 1-1 i-f i-H rH 00 uo "* o LO 00 0-1 (N o CO lo L0 th' OS 00 1> co CO O CM CO OS o rH a> GO 00 OS CM os o 00 o t-H CO i-H o os 00 CO os lo 00 o OS CO' 00 T— 1 i> OS ■HH i-H L0 •* I-H iQ o i-H c-i i-H i-H -tft o (M CM CM CM CM 01 CM CM CM CM CM CO CI 00 I-H ~H -ti CO co LO LO 1> oo oo i-H TiH CO 00 t^ o LO LO L0 CM TfH (N CM CO CM T^" CM co i-H CM i— ( T— 1 1-1 i-H T— 1 i-H i-H i-H I-H i-H l> »0 CM CM os o LO i-H N- t^ o i— I CM CO CM CM -t* o tH L0 TH CO I-H i-H ^ i-H T~H i-H I-H t-H i-H i-H i-H t^ co o 1> LO CM oo T« 1> i-H b- o o CM i-H CM CO T— 1 CO "f TtH LO I-H rH i— I T-H I-H i-H i-H 1— 1 I-H i-H rH + + a a ,_, TH s~< ■ .-; 03 03 o o o o3 cu t(H w o a «+H a 7j + 1 "~' t+H ^ >> T3 + T3 ci o3 c3 cu as CO. CIS as CD + ri 03 tH 03 CO a o a a o pH "3 03 u ■+3 -IJ o t4-H <4H 1 +3 o O o o < t^ CD CO a> as faO bfl bfi bC l> 00 OS 03 > < !> oo os o3 U > < t~ 00 OS OS > < oo CS tH as > < •J- 5 X ' c3 | a cs 1 •? S ' + o 1 «« -° © O | 1 1 V, « c3 o3 X X TS o tH! O Ph a CD o o as as o o P a 2 q o3 &fl o hn^; -w . a m CO OS ra b(l Oh fl> T3 «? n 3 Sin « 5P a 8| -i x) cy fl-a M « a b.0 as -a P >4-l H-3 «*H O a O -O T» -r^> *H a o 'O T-* *+H O t^ tH a p a a o a rg o § XD CO o ^1 ' ' o3 to < PQ