Si- 203 .N38 1929 Bulletin OF THE National Research Council NUMBER 67 THE MINIMUM PROTEIN REQUIREMENTS OF CATTLE Report of Committee on Animal Nutrition H. H. Mitchell N8I3 Published by The National Reseaech Council OF The National Academy of Sciences Washington, D. C. 1929 PI handle t witi The Universi Librai :.^uA>^ C) I jrning Publications e rch Council )nal Academy of Sciences fcearch Council which is represented scecutive Committee. It is open for f the National Research Council on National Academy of Sciences. dings " is $5 per year. Business Academy of Sciences, Washington, lonal Research Council lal Research Council, other than pro- opriate agencies of publication have pegular intervals. A discount of 20% '[ill be given to all who subscribe for Publication Office. Numbers of the es based upon the cost of manufac- ts of individual numbers, see end s of the National Research il ces dependent upon the cost of d by or for the National Research list price will be given to all who from the Publication Office. (For s, see end of this report.) Reprints and Circulars "of the ed by remittance, should be ad- search Council, Washington. D. C. BULLETIN ,(j^^^ OF THE NATIONAL RESEARCH COUNCIL February. 1929 Number 67 THE MINIMUM PROTEIN REQUIREMENTS — OF CATTLE 5Vai\ ; (ry^, r- ^ ^- ^ '- - '^-^"^'^ ^ ^' i>^.vo-X ' Repoffof Committee on Animal Nutrition^ H. H. Mitchell *The members of this committee of the Division of Biology and Agriculture are: Paul E. Howe, U. S. Bureau of Animal Industry, Washington, D. C, Chairman; F. W. Christensen, State College Station, Fargo, N. Dak.; E. B. Forbes, State College, Pa.; F. G. King, Lafayette, Ind.; L. A. Maynard, Ithaca, N. Y.; H. H. Mitchell, Urbana, 111.; H. J. Gramlich, Lincoln, Nebr.; A. G. Hogan, Columbia, Mo.; F. B. Morrison, Ithaca, N. Y.; C. R. Moulton, Chicago, 111.; E. G. Ritzman, Durham, N. H. (,H.^o| IV ^ THE MINIMUM PROTEIN REQUIREMENTS OF CATTLE H. H. Mitchell PRELIMINARY CONSIDERATIONS The question of protein requirements of farm animals possesses both a scientific and a practical significance. It is obvious that the food require- ments of animals and the factors modifying these requirements are legitimate subjects for scientific research whether they relate to farm animals of great economic importance to man, or to laboratory animals of no economic importance, or to man himself. In each case the problems involved are those of experimental biology, and the results obtained will aid in a better understanding of the science of nutritional physiology. "\K Experimental investigations with farm animals which deal with the ■s^ entirely practical questions of feeding and management are more under- ^ standable, more readily explainable, and hence more intelligently appli- ■v^: cable the more exact and extensive is the available knowledge of the food <, requirements of such animals. For example, the favorable effect of the addition of a nitrogenous concentrate to the ration of fattening cattle may be due to its content of protein or of phosphorus, or it may be due to some unknown specific effect of the concentrate on the digestive tract resulting in a greater desire for food and a greater consumption of it. 0 Obviously, the decision as to which of these possible explanations is cor- ^- rect will have much to do with the practical application of the experi- ^ mental finding to other rations than those used, and even to other '>- nitrogenous concentrates. The greatest return from even these wholly ■^ practical investigations can be realized only when their results are as completely explainable as the control and the measurement of experi- mental conditions will permit. Complete and accurate information regarding the nutrient requirements of farm animals and the nutritive f value of farm feeds will obviously aid in interpreting intelligently the , outcome of such practical studies. "^ What is true of practical experimentation with farm animals is r^ equally true of scientific experimentation. An accurate knowledge of the food requirements of the experimental subjects is not infrequently a vital prerequisite to the correct interpretation of the results secured. Much of the German experimentation on the capacity of ruminants to 3 4 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL synthesize protein in the paunch from urea and other simple nitrogenous compounds/ has been rendered largely unintelligible through lack of an accurate knowledge of protein requirements. One of the most practical contributions of the science of nutrition to the feeding of farm animals is the formulation of feeding standards applying to different classes of animals and to a variety of conditions. How practical these standards have proved to be and how greatly they have modified feeding practices cannot be told. Whether or not their value in increasing the financial profits of feeding operations is con- siderable, undoubtedly knowledge of feeding standards and their limita- tions will aid materially in the intelligent appreciation of the live stock business, particularly in the ability to cope successfully with changing conditions of feed supply and to avoid exploitation by manufacturers of commercial feeds and other products for live stock. Feeding standards should promote maximal production with a mini- mum of overfeeding. They should include a factor of safety, so that a normal variation in the composition and nutritive value of feeds and in the functional capacities of animals will rarely if ever result in under- feeding. But obviously, a factor of safety cannot be scientifically included in a feeding standard until the actual minimum requirements of animals for the different nutrients have been determined. For the same reason an engineer cannot intelligently impose a factor of safety in the construction of a bridge unless a satisfactory estimate of the maxi- mum load that the bridge will have to bear can be made. Hence, feeding standards for farm animals must ultimately be based upon satisfactory determinations of minimum animal requirements. It may never be necessary or advisable to feed a farm animal in exact accord with its protein requirements, but when an animal is non- producing at certain seasons of the year, or when protein concentrates become relatively high in price, it may become expedient to approximate these requirements. In such cases it is clear that an exact knowledge of protein requirements, as well as of the protein values of farm feeds, becomes of immediate practical significance. For these reasons, a study of the protein requirements of farm animals is well worth while. In a recent report of the Subcommittee on Animal Nutrition (^) a method of measuring such requirements was proposed, differing in all essentials from methods commonly used in previous ^In most of these experiments, the question of the nutritive value of the urea addition depends for its answer upon the question whether the digestible crude protein of the basal ration is adequate in itself to cover the protein requirements of the animal. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 5 studies. The present report is an attempt to apply this method to a determination of the protein requirements of beef and dairy cattle by the use of experimental data already available. The essential features of the method of measuring protein require- ments that will be exemplified in this report are as follows : 1. The protein requirements of animals, as well as the protein values of feeds, are best expressed in terms of nitrogen, or of a conven- tional protein, such as the ordinary N x 6.25, rather than in terms of true protein. Throughout this report, the term " protein " is used in this conventional sense. 2. The problem of protein requirements must be considered as separate and distinct from the problem of the protein values of feeds and rations, if permanent progress is to be made in respect to either. Hence the attempt to measure protein requirements in terms of digestible feed protein should be abandoned. Such expressions of protein requirements have served a useful purpose in the past, and may necessarily serve for several years to come until the information required for a more rational expression becomes available; but future investigational work may well be planned along other lines. 3. Protein requirements may be conveniently and rationally ex- pressed in terms referable to the animal rather than to its feed. Although the percentage of nitrogen in the different nitrogenous compounds of the animal body varies greatly, the need for them, or for their precursors, the dietary amino acids, may be satisfactorily measured by the total nitrogen content of the tissue constituents catabolized endogenously, in the case of maintenance, or by the total nitrogen content of the new tissues formed in gro\vth, fattening and reproduction, or by the total nitrogen content of the milk produced in lactation. These nitrogen values may for convenience be converted into conventional protein. 4. The use of such measures of protein requirements can be made to the best advantage only in conjunction with measures of the protein values of feeds based upon (a) the total nitrogen content of the feed, (b) the wastage of nitrogen in digestion, (c) the necessary wastage of absorbed nitrogen in the process -of its conversion into tissue constituents or the constituents of body secretions, and (d) the supplementary relations between the available nitrogenous constituents of the feed and those of the other feeds with which it is commonly fed. It follows, therefore, that the results obtained in the folloivmg study relative to protein requirements are not to he compared with values already current in the literature. Heretofore, protein requirements have been expressed in two ways, either in terms of digestible crude protein or 6 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL in terms of digestible true protein. Statements thus expressed are ambig- uous, since they are not related to protein of any definite or known biological value. In this report, protein requirements are expressed in terms of animal expenditures or storages of nitrogen, converted for convenience into conventional protein (N x 6.25). They may be considered as representing the requirements for digestible crude protein possessing a biological value of 100. They are thus definite in their significance, minimal in the truest sense of the tvord, and adaptable to any protein mixture the biological value of which may be satisfactorily estimated. THE PROTEIN REQUIREMENTS FOR MAINTENANCE Existing protein standards for maintenance. — From a study of Ameri- can and foreign investigations on farm animals in which low-protein rations were used, Armsby(2) has concluded that the protein require- ments of swine, cattle, sheep and horses are very closely the same per unit of weight. This evident similarity between animals differing so widely in size and dietary habits is probably more significant than the actual numerical requirement resulting from this study, i. e., 0.6 pound digestible crude protein per 1000 pounds live weight. It is of some sig- nificance also that Sherman (^), in a summary of similar, though more extensive and exact, investigations on adult human subjects, also arrived at approximately the same relation of protein requirement to body weight, i. e., 0.6 gram per 1000 grams. But there is abundant evidence that these requirements are too high, even in terms of digestible protein. Armsby himself has noted certain Danish experiments on two dry cows in which nitrogen equilibrium was attained on intakes of 0.21 pound and 0.25 pound of digestible protein per 1000 pounds live weight, and another experiment on steers in which similar quantities of digestible protein sufficed for maintenance. The long-continued feeding experiments of Perkins (*) on dairy cattle may also be cited to the same effect. Subtracting the protein secreted in the milk of these cows from the digestible crude protein intake, left balances of crude protein varying from 0.43 to 0.68 pound per 1000 pounds live weight, averaging 0.52 pound. Since ■ a 100 per cent conversion of digestible crude protein into milk protein would probably never be realized, the actual amounts of protein used for maintenance by these cows were probably distinctly less than the values above cited.* The recent results of Buschmann(^) possess much the same significance. * Much the same experience has been more recently reported by Maynard, Miller, and Krauss in Memoir 113 of the Cornell University Agricultural Experi- ment Station (p. 17). PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 7 Sherman also has noted many experiments on men with a much lower indicated protein requirement for maintenance than the average cited above, and has himself, in conjunction with his associates, reported experimental observations of this description. Hindhede's work has afforded further striking confirmation (^) of the adequacy throughout prolonged feeding periods of amounts of protein only one-half the Sher- man average. It is, therefore, clearly evident that the standard require- ment of 0.6 part of digestible protein per 1000 of body weight includes a large margin of safety. The excretion of urinary nitrogen hy fasting cattle or cattle on low- protein rations. — However, if the maintenance requirement for protein is to be measured in terms referable to the animal rather than to its food, the values just considered are not directly pertinent to the problem. Attention must rather be directed to the daily excretion of urinary nitrogen of cattle on low-nitrogen or nitrogen-free rations. It is also important to consider the nitrogen excretion of fasting cattle. Forbes, Fries and KrissC) have found the average daily urinary nitrogen of two fasting cows per 1000 pounds live weight to be 46.5 gms. and 43.6 gms., for the sixth to the ninth day of fast, inclusive. This is equivalent approximately to 0.1 gm. of urinary nitrogen per kilogram of body weight, or to 0.6 pound of crude protein per 1000 pounds of body weight. For two steers weighing approximately 600 kgms., taken from a sub- maintenance ration and fasted for 10 days. Carpenter (8) found the nitrogen excretion in the urine for the last 3 or 4 days to average .059 gm. and .067 gm. per kilogram per day, equivalent to 0.37 and 0.42 part of protein per 1000 of weight. It is, however, well known that the nitrogen excretion of fasting animals includes a considerable fraction of nitrogen representing the catabolism of tissue constituents serving as a source of energy. Since this function may be served by the non-nitrogenous nutrients, it does not represent a function of protein that should be considered in the deter- mination of protein requirements. The protein requirement for main- tenance, therefore, does not bear any constant relation to the excretion of urinary nitrogen during fasting, but it is directly measured by the excretion of urinary nitrogen during the adequate feeding of a nitrogen- free ration, i. e., what might be called specific nitrogen starvation. Unfortunately no experimental data appear to be available on the nitrogen output of cattle under conditions of specific nitrogen starvation. It must, therefore, suffice to cite experiments in which this condition is more or less closely approximated. Thus, reference may be made to the experiments of Bull and Grindley(^) on the nitrogen metabolism of steers on various rations. In these experiments, 2 steers were brought 8 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL to nitrogen equilibrium on a maintenance ration of 1 part of clover hay to 5 parts of corn. During this period, the nitrogen output of these steers averaged daily .046 gm. and .048 gm. per kilogram of weight, equivalent to 0.29 and .30 part of protein per 1000 of weight. Titus (^°) has determined the daily output of urinary nitrogen of three steers on a ration containing 40 per cent of alfalfa hay and 60 per cent of paper pulp. These values per kilogram of live weight are .042, .052, and .045 gm., or, in terms of protein per 1000 of weight, 0.26, 0.32 and 0.28. The Laboratory for Agricultural Eesearch in Copenhagen has pub- lished (^i) the metabolism data from 2 dry cows subsisting upon a low- nitrogen ration supplying 46 to 47 g-ms. daily. Nitrogen equilibrium was established on this ration with average daily excretions in the urine of 14 gms. and 15 gms. of nitrogen, equivalent to .029 gm. and .034 gm. per kilogram of body weight, or 0.18 and .21 part of protein per 1000 of body weight, Honcamp, Koudela, and Muller(^2) \^^yQ investigated the nitrogen metabolism of 2 cows in milk on a low-nitrogen ration of oat straw, corn, potato flakes, molasses, dried beet pulp, and salts. The nitrogen intake was 85 and 89 gms. daily, and the urinary nitrogen output was 0.042 gm. and 0.035 gm. per kilogram of body weight, or .26 and .22 part of protein per 1000 of weight. If the disposal of endogenous metab- olites is not influenced by the function of milk secretion, these values are comparable to similar values secured on dry cows. The nitrogen output of 2 young growing heifers subsisting upon a low- nitrogen ration was determined by Hart, Humphrey and Morrison (^^). The calves weighed from 300 to 400 pounds and the ration used con- sisted of wheat straw, corn starch, cane sugar, calcium phosphate and salt. The average daily excretions of urinary nitrogen were equivalent to 0.036 gm, and 0,030 gm, per kilogram of weight, or 0,20 and 0,17 part of protein per 1000 of weight. In a later somewhat similar experiment upon a younger calf, Steenbock, Nelson and Hart(^*) obtained a mini- mum excretion of urinary nitrogen of 0,045 gm, per kilogram of weight. The minimum excretion of urinary nitrogen per kilogram of body weight, for various farm animals and man. — As Armsby's study of the amounts of digestible feed protein per unit of body weight required for the maintenance of weight or of nitrogen equilibrium has indicated a remarkable similarity among different species of farm animals in this respect, it is of interest to compare different animals also with respect to the excretion of urinary nitrogen per unit of weight on nitrogen-free or low-nitrogen rations. Such a comparison for the larger farm animals from such available data as have been located has been made in Table 1. The data in this table for each species are arranged in the order of decreasing ratios of urinary nitrogen to body weight. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL TABLE 1 Recorded Determinations of the Daily Excretion of Urinary Nitrogen by Farm Animals on Low-Nitrogen or Nitrogen-Free Rations Pig Species Body of weight animal kgms. Daily urinary nitrogen gms. Urinary nitrogen per kgm. body wt. gms. ; 22.2 1.60 0.072 14.3 0.96 0.067 17.7 1.09 0.062 19.5 1.09 0.056 16.8 0.90 0.054 25.0 1.32 0.053 38.1 2.00 0.052 10.9 0.54 0.050 38.1 1.88 0.049 40.0 1.95 0.049 46.3 253 0.048 38.5 1.83 0.047 26.3 1.19 0.045 37.2 1.61 0.043 68.1 2.65 0.039 41.0 1.54 0.038 74.9 2.61 0.035 Lttle 145 6.48 0.045 385 16.32 0.042 177 6.33 0.036 440 16.40 0.035 443 15 0.034 168 5.03 0.030 485 14 0.029 eep 47 3.39 0.072 38 2.59 0.068 35 2.37 0.068 45 2.63 0.058 43.5 2.39 0.055 45 2.41 0.054 35 1.81 0.052 40 1.91 0.048 42 1.84 0.044 40 1.71 0.043 54 2.02 0.038 31.9 0.99 0.031 33.1 1.03 0.031 43.5 1.16 0.027 44.1 1.05 0.024 Investigator and reference McCollum and Hoagland(l5) McCoIlum and Steenbock(l6) Pfeiffer(n) Mitchell and Kick(l8) McCollum and Steenbock(16) MitcheU and Kick(l8) McCollum and Hoagland(15) McCollum and Steenbock(l6) PfeifferC") McCollum and Steenbock(i6) U U li Morgen et al (19) McCollum and Steenbock(l6) Steenbock, Nelson and Hart(14) Honcamp, Koudela and Miiller(l2) Hart, Humphrey and Morrison (13) Honcamp, Koudela and Miiller(l2) Copenhagen investigators (H) Hart, Humphrey and Morrison(13) Copenhagen investigators (H) Morgen, Beger and Westhauser(19) a (( (< u « " « « (20) a it (t .0(Me«5 '-I T-^ (M (M ■* ■^ bi CO lo O be r-oiOi— ico-^ioco ^ --< (M (M (N (N C^ O CO 05 CO (M «0 Q O -^ O Tt< -!ti >o O ^ --H (M (M (M (N IM "^ffl «P'-|0qO5i-ipiqp-«i^ ■*'■<* -H CD oi CO 5Siac00'-iT}--Hco03 r^tj)CDOOCOiOCOC005iMC^iM *-'^ rt<-irt^(MC.0-^05t^C5IMI>CO m'5fcI)r-ic5c5cOCOidtOI^«OOSo Ji d CO l^ t^ --I ^^2- lO UO ■<* Tjl (/j*^ O »0 lO "5 (M ^ lO t^ CO '- CO O >— 1 (M ^H o lO "5 »0 lO "O lO PROTEIN REQUIREMENTS OF CATTLE: MITCHELL OS "-JiOOQi-HOiOOicOO M csi . QOitaD OOOOiO^H^HCOiOOO ?^ ^^'^'-<(MCOCO'3<'• Oi lO OS M <3i Tj; CO C^ (m' CO Tf< ^ S •^ o lo r>. ■O^ in' Co' CO Tti CO >-• CO T^ »0 ■<*< (M p ^ CO (N (M O M' in Tf' Tji p'--COO(MO'. — — -■■ lO lO oa 1^ ^ _ ^ (^ f^ •^(NOQOfOeOt>.cOlN(M o ''■i '^C. feS^5S£5J?^«=>O'='00c0 ■^E fOQqO'-!^-ooooo^^t>■03 '^' S '"' r^ 52 c*^ «ti CO CO CO CO c^j o «=c '^i^O'— iTri(M(Mio»oooo LO^corti(N(>iTj;ididcd < s C0tqt>.pTt(O5t--.O5(Nt>.p £2 2 ^ ri P "^ <=^ <^i f^' r**' 00 2'Zht '~lS"^^'-''-'fOM:^ ^(N(NCOCOiOiOl^000000 M CO t>. I »0 (N Tt* lO t>. CO -H P CO O T-< Cfl r^ ~ 26 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL ooi— 100.— it^,— I,— irfco US ^H >-H^-.cpo5»-^^-H0^5C>^» '^ 1-1 1-1 CO CO ;0 CD t>;OqTti'-;coot>: cShyoO'-HiOO'^COC^.-iod'O ^■^ r-i^c^coiot^ooooo 2 cpcoooo;cp«ococDO P_ »/^ /■~^ ^/-s I r^ — ^s*C ^'^^^ J jm b^ * .- »- GO 00 ^- w^ (M CO Tj( t- C30 r- t^ £ cp'-; -^^ TjH .-< o oi 00 (M CO ■*■«*< T}< lO OJ o fcue (M -^ >0 ^ H ^ rt ^ ra a;_a _ ^ ^ c-q f^ ^ So TtHOJooococot^ w^ >o'OiracoO(MO ^■^ lO "3 lO lO »0 O lO ■»ti o t^ 05 r^ CD Tj( lO »0 CD (M IM (M O .-I lO iO UO o _o g 0> CO 00 05 (N a he 00 O -H lO r-< 05 "5 0-« CO CO CD CD . CO Tt< •<# IM CO CD (M CO CO CO QboOlCDl^Cll^Olt^CiO'M'O - .a COiMiCCO^COCOiOOliO r-H .-I (M CO C'TI CO " ^ oqiot>:prt!osi--:05Cii>:cp S)E cocdcd'-ioooCJcico'^co ^K T-ir-ii— ((M(MiM(Mrfi>0»OiO >1X l» TT .tc a (M n >!S '-i(N(MCOCOUO>Ot>.OOQOOO So CDt^t>..— iLOCra»0l0>0'0i0»0i0i0 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 27 O tc CO CO Tj< ;0 oJ •«* o 05 O ^ i-H T-H .-H CM • t^ >ocorocOT}<,— ir},J- to O 1^ -^-^-HeMCOCOrt^.^^ J ^ lOOOOOO— itO^CJCS g5 »f3'S<»OTj. -^' 4,g p-^ '-Hi— I.— iCMCO-^'^lfJiO u o '*CMOOOCOCOt^tDCMCM iS UJIOOCOOCMOC^O^ -g pQOpcOt^OTj<00^-^ I oiP f001^T-iOt^x>;CM»OiO ppt^t^oqpopcooo ■^ocooocdod'^'coTtHco i-HT-H^rtCMCMOJCM Oqiqt>;prt.p CO CO ;o '-< o -a;H '-I'-Hi-HCMCMCMCM •a ME CM»:H5iooco<*.-HooO'-<-H Ooitl'-'OpCM'— i»-cC0TJ— iCMCMcoco»o»or>.oooooo ^^ iC'0"<#Tt* t-^ r-< '^ -« r-H^Ht— i^(MiMMCCirofOCOrt<'*TtiTti 0!6J)O00O(MirC-^l0«0t>.G00iOO^H^H»-l eQ'*^.-i,-^rtC^c;'#p'#t>.p'-<-HpO5t^'«*!r-; Io5i-^050r--Icsic^'co-^>o f< £ S •* P ■*. •ti iD M CO lO o oot>^>ocip'*pcqt~-:pcqio o6o50^'r-icicsic6fOTi.U5^tN.i-HiqoOppi-;PoOp T)H -H 00 o d' 00 »o ^ t^ ■*' o CD oi i>^ CO ^H(MC<|C0Tt<'*»OcOOt^0000Ci05O r>.t^r-lr^t>.r-<(M^H05lO'— llOOO'-l OCOOoOOJ'-HiMCOfOrtiiOiOiocO o^^ooooi«ot^eOcooO'-i^HOi>.coi>' o">c rt^cicooo^Hirsoo.— i-»* M ,-(■— c(M(NcOfOfOTtiTfiTfi'*iiOiOio»0'C CO -*i CD t>- o pppp^pTtii>oot>tpcqp'*co O 00 -4 ^' CO I>^ Ci O ^' (N CO ■«^< rtH >o »o r-Hrtrtr-lrt(M(MC^CSl(N(M(MCN JjSE p'rtjri*. pcqcq^Hoqcot^— icoc6-^''^idiOio'Oic ^ii P CO ^ 05 S? ^idc5cico^TttooiM«oocoi>.T-Hio feF-05TtiTj<050oocoio-*^Hoocot^ ^ rt'-ir-iCvacicvjc^tNc^cococococo .^jj c CO t>. O ic CO ^^yr-HC^COr*-'!' •<*<0505C0C0O(M--i00(M Ttl05'^0iC0l>-OC0«O00 Oi0CDCOt>-l>00000000 d2 COOJCIICOO'— iTt't^O'.-J^i-'WJ;.-^ IB T-irti-((Mcjc«cococoeO'>* PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 29 t^(N-Hr>.oix-HicoOTj.'— icooo f QpOC3C5^»CCOCOCOiO'*'rO<-HOOOt>- 2 "^ C^ (N (N (m' C.o--J C^' ^ ci (>J c^" c^ (N (m' oocor-^icoqcpt>.i>.cq05'<#05T^(>)cp * CO (N C^' (>i C^ (N (m' Cvi T-H ,-i O O Oi 00 ^TficoooeooocooiTtiococ^Osoco c^T*HTii ei i-H r-( -H ,-(■ *j iCpc<5Tj-0^t^ e: idio-^T}.OfO«DC5(N»COO-^ ,-HT^,-ic^c^esfQcococo'*-^'a"»o 30 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL Since the weight of a steer is a better indication of its stage of growth than is the age, the same sort of data may be obtained for equal intervals of weight rather than of time. From Equations (3), (3), and (4) the average age of each group of steers at weights of 100, 200, 300, etc., pounds in weight may be computed, and the composition of the steers at these ages determined as were the values in Table 13. The percentage composition, derived directly from such estimates, for the steers at equal intervals of weight has been summarized in Table 15. The rate of deposition of nutrients and the percentage composition of gains at different ages and iveiglits. — The ultimate purpose of this math- ematical analysis of the Missouri data on steers is to secure satisfactory estimates of the rate of deposition of each of the body constituents at different times and at different weights. This information may be readily obtained from equations obtained by differentiating Equations (9) to (20) inclusive, in accordance with the general Equation (5). The linear equation used in fitting the age-fat content data of Group I (see foot- note to page 19) gives a constant, upon differentiation, of 7.932 kgms. of fat per month. For this group, therefore, the rate of deposition of fat was constant.* Obtained in this manner, the rate of gain per day by the steers of each of the three groups in each of the constituents will be found, for equal intervals of time in Table 16, and for equal intervals of weight in Table 17. The energy increments in these tables are obtained by assuming a gross energy value for protein (Nx6.25) of 5.7 cals. per gram, and for fat a value of 9.5 cals. per gram. ^The differential equations obtained, from which the rate of deposition of the constituent in question per month ( , j at any time t may be calculated are as follows: Water content: Fat content: Group I dW Group II dW dt Group III dW dt Nitrogen content : Group I dW dt Group II dW dt nrnnn TTT dW dt = sr-see-""'" = 35.006-""" — is.oee-"''" = 2.8096-'"'" = 1.1496-"*^" = 0.4486-"="" Ash content: dW dt dW dt dW dt dW dt dW dt dW dt = 7.932 = 2.5116-'""*"' = 1.0136-""'"' z= 2.5686-"="'' = 1.1966-"='"' = 0.8406-""^' \ PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 31 5 00 o o P p CO < ^ •^ CO ^ ■^ UOO-HCOCOCOiM (N I>; CO p p CO 00 CO CJ 05 O i-< 00 (M kO -H CO 00 I-- CO « O B3 J 05 ". pTf:P !H ci (m'ojoooocjo^co j= 1^00(Nl>.OTtit^00i-iTti ^ PO^. T-HIM. C^(M(NCOCO "'^ CO Tji Tji TjJ TJH Tji TjJ •«}; T)< Tj( P e ri C>-J o _J2 § '^. CO t^ ^ be (N M (N c^ »opp (2 o^-Hc^cocdT)<"30t>^ fc i>toocot>^pTf;t>.oqt>.'^j< 3 oidt>^t^cOidco'— 'cit^ o O 'ooqpooP'^t^pppooco'*'— iTTcDio O5o6c6<^^cs(^^c^^c^<^i(^il^^<^i(^«(^^-Hooj ^ O00S;i0f0'-io»0(M00rti05C0i-<«0T-<»0 _^ p(N'^. ■<#T}-;io<-ioqppi^ fe ^»'cocoooaj— 'c^co-^'ior^odot-icocooo r-i^i-Hi-H(Mc^(M(M(NIM(MCOC0COCOCO 00 CO poooO'-H'-jc^coco ' CD «0 l6 T*< CO im' — < ^ Tj*lO"'lO'*<'^T}<-^'<*l'«*lTtlT*; CO CO CO ^£ O O P o o o o o o S o p o o CO Tj< "0 (X5 ^- 00 05 oooooooooo oooooooooo O>-h(MC0t}. ■«# .— I 05 CO -^ (M 50 CO CO »0 «0 lO lO r>. 03 (M CO o 00 t^ lo ic ■* ■* ■* O CO »0 (M -H Tti (M T-i o ai ^ ^ "^ ^ CO .S c/5 '^' O »0 (M. 05 CO !>; t>; 00 P CO C. _, ^. w. -.; w- a-SgS cocb'ococ^^'oo5CJo6i>i>cdidib'!05»oc^cd cd»cio-;i>:cqpcop .Sc-Ss ooooc3coooif5cococoicooc.coioiOTt<'!j( ,2 Q.i2 c lo n -H o3 r>. uj t'j CM t_) O "^X! bDfoCOCO(M(MlMiM(M(M COC5(M>OCO^-^t^OCOCDOiiMiOOO .-i^,-i(M(M(NCOCOCOCOTt.i-HTticot>.c^»oocot>-t>-t>»(Mooco C^TtlC005COOOc005»C'-iOO>OCOOOO coc^^ooo50500oooot>.i~-t>.i>.co t^ o CO o '?•"«? »cco^<35o6coid->9ic^r^ S. OI^OCOCOOt^COCOlMiO c^bcg csjoqt>.oqc^pcooqcopin ■3-s2S(3icoT*Hc^i— iaJodt^c6»oidTiJcococ^c^ o 5 5) ii; S ;2; i:j ;_3 t^puOTj*-*. cqcocococor).t>.t^r-t-i^i^cocococococococo S ^a =« lOCScO'^'-HCOt^cOCO M.S1aE coioo5'>i-< ^ ^ CO CO CO t>^ T)< CO CO c,m OO'^O^CDCOCOCOCOOJCOOJMCDCQ ■3 2«'S o5-*05>o--i^-*^cococo-HascocO"0 O ^"^ S) »« »o — — T}.COt^O-^05iOCOC---- — - JoCO(M005COCOt^COCOcOCOlO»OlOif5 <^ COCOfOCOC^C^C^CaC^C. o c ajcoo6TJ^1-Jo6uocO'-'05o6^^cb^OTJ^c6 lJ£COCOCSC<>C. p 00 CO (M. p CO CI T}< t^ (M. t>: -^ >-< P 0'^oiidciooocO'0'!j-CDlOkO'*'^COCOCOCld ^i CD05e^lOOO'-<'tit>.OCOCD03d"300'-< ^ ,Hi-ii-(C«dCSIC0C000C0-» c^ CO •^ o ^•^ ^ !r cj p 05 00 00 ^ t>^ rf< (N O ci S-SShS O S,^ 05 00 t>. IC lO T}< CO 5E c^'-^-Hoo05o6 •gc-gg C^00iO(MO5«C>Tti 0"'j£5X C^ ^^ '"' ^^ ^=2 ft-o r^ CO CO c^ CI (M ^ r-i =g t- CO 00 i>- 'o "O Tt< (N l>i CO O 00 00 F-l ,— I d CO CO T}< o w m ►J CQ Ui > •^ a o PQ •5 e A >i,£ *i»00005!M'«til^OC0 -"S-Sg^w '<*ioo6tdriHc.T}<.-l. CO CO C Sr/;^(MOCO(N0005"5o6«0 'S^tiS Ttir-— (lOOio^Hooicco ca s; rt P wj '^j^ v^ *4j ^^ ^^^ tN 1^ (-^ ^-^^ O ^"^ EoiOiC-^TtlCOCOfNIM'-i -I ^' <£> ^ !>.; t-i ^ "-H (M (M CO CO CQ 2 -i o - ^.oo(Moooocoi^>ncD05'^coooor>-05t^o .5 _. i rjis ot>.'^Tt-^io-Hi:oc^a5t>- C8.5 c M-^ t^ioTricofMr^oocjoooot^i^ocoioio O *'4. ■S.Siy.g cooJc6cd^'o6oTj'-^COCOlOOOCOOOOC5^ ■S.Srac 'OOkC.-(tv.rtiOI--Tji-HOO>OCO'— c00i;DiO O ^hE"5'0"*'*COCOCO(MiMC^'-i^^'-i Ht.;^, "-: oO'-i(Nt^i^cooooo-*-*oo>ococooDio 5J;«E Ttit^rtiooiooi'^CTi'^oiTfioocot^csi^ o ^"^ 5jOoa)oc5050oi>.t>.!OcoioiOTti'«*icoco(M g p CO p »o Pg Tti O 00 05 '-^OOp'C(N'^'#'-;Tt<'^. --H C^ -^ ' ' ' im POOOOOOOOOO ..Q OOOOOOOOOOO '" CO'^lO«3t>.00050^(MCO o o o o o o o o rt< »0 CO t^ 34 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL The percentage composition of the gains is given for equal intervals of age, in Table 18, and for equal intervals of body weight in Table 19. The per cent " difference " in these tables and the absolute " difference " in Tables 14 and 15 are accounted for by the " fill " of the cattle except for errors in analysis or in mathematical estimation. Negative " differences " in relation to gains may be interpreted as due to a decreasing fill. The significance of group differences in terms of growth and fatten- ing.— The essential results of this investigation in so far as the question of food requirements is concerned are contained in Tables 16 and 17, i. e., the daily increments in nitrogen, ash and energy. These increments must come from the day's food and represent those portions of the food that are being used productively. For all three groups of steers, the daily increments in these constituents decrease progressively after six months of age, except the fat increment for Group I, which remains constant. Striking differences exist among the three groups of cattle in respect to the amount of nutrients added to the body daily during growth and fattening. These differences are greatest with respect to fat and energy, and least with respect to water, nitrogen, and ash. It is important to evaluate more precisely the significance of these group differences in daily tissue increment occasioned by the imposed differences in the plane of nutrition. According to the plan of the experiment. Group II represents more nearly a normal growth of Hereford-Shorthorn steers, while Group I represents an additional and considerable fattening and Group III a retarded growth. In accordance with this plan. Group II eventually attained a fat content of about 20 per cent. Group I a fat content of about 40 per cent, and Group III a fat content of about 13 per cent. It would be interesting to compare the composition changes of these steers with those of dairy heifers during normal growth, but this is impossible to do directly in the absence of chemical analyses of the carcasses of dairy heifers of different ages. In the absence of a direct method of comparison, an indirect one must be resorted to. Brody and Eagsdale(**) have suggested a height- weight relation as indicative of the state of nutrition of dairy heifers differing in age. They point out that although the weight of growing cattle is primarily dependent upon the plane of nutrition, the height at withers is a remarkably good index of the growth attained and is not readily affected by fattening. If weight is plotted against height, the curve obtained for Jersey heifers follows very closely the curve obtained for Holstein heifers, although a considerable age difference exists with PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 35 05 "^ 05 t--; 05 t^ >-; »-J ■^ O OJ >— ' fO CD ita 00 O CO «0 •^ (N •^ t^ 05 ^ -^ t^ 00 00 CO C: 03 05 O fO lO o ocooo.. — -- — ,_iCOCO'^OJ(3500COrf?OOS *^ lO ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ th -"ti CO CO CO iS oQoooooot-^t^t>.. t^i>:t>:«p«p«pep«P«iP *j p«0(NCOTt<(NOOt--;incocqc^'*icoO'-< r^ OS ci O O >-< (m' C^' CO -^ »0 O l^ CO oi o p^ " ,_,,_|,— I,— (T-(i— (>— 11— IrHr-l'— Ir-HC^CS g cqco'*eo«D(MTtt-HO3cpkO'^i>;'-;«O0Q t; «De^ot>;-^c^'o5t^-^'c3goo«D»oco;-; I? coocoiO'0»0'*Tti'<3-OC0CD<3SIMiO00--< ijig stjw i-3_;,-iiMc^cMcocococo'*T}iCC0Ot^0iC>a05T»< o 'j; t^ o CO o _ t^ CO CO cJ 00 t^ i-Hr-l(MC^lC^(M(M(MIMIM(M I I ^,H-H-H(N(N(MC0't<-^>OOt>-Q003O ,-, CO 1-- l-"- O CO ■^ CO CO CO Tf 'S' '*. H05Q--i(M005^(MOt2 HTjo»oiocpcpcpcO TtlTt!-*'*'^'*'''*!'^''*!'^'*'*"*'^ ^ ^ {M_(MCOtDO5Cq00iO(NO5C^. "503Tt;COC>» ,=! con^idcbt-^oiofN-M^'CopOc^'OOOr-. fe ,_,^^,_i,-(,-i(MC^(M{M(MfOCOCOCOTt< ^ i>.»005C^a)— 105»-;i;pcoioooqtpt^o tS cooaJ'^oO'*05CDcOOO^HTfit>-OCO bc2 ^rt^c^{Mt-icococ^(M'oi>^Ttioiocaoi I I I b£ ^ 01 -^ t^ 00 en gr*^ -H t>. lO US 00 CO Cp^ ^ ^ ^ ^ CD •^ Tf< •<* iCi (M O uui.^jiJuO0O5Tfl(MCO00CD 00O'^C0O00^'*l>!-1 •-HIMiMfNC^C^CCCOCOTjt ■*'cococ6c^c.cp CO (M M C* o o o o t>^i— ipC^Ji— ilM. TPCD05C0'-;t^01C0 t>Jt>IocOI>^'— I'CO'OMOSCDiO'^ (N(MCOCOCO'i''1-OOCV r);cqoqi--:t^pcoQqcooi oocooodiocoooocDioco-Hoco »OM*rtiCOCOCOCO.OC0CD0i(N»0 r-ir-i^(M(M(NCOCOCOCO-<*'-* fao ;3G PROTEIN REQUIREMENTS OF CATTLE: MITCHELL O Tti 00 t- iO Tfi CD CO Tf t^ IM 00 O >C CO CD 05 •* Ol 1^ t^ Cq 00 00 CI as o »-< rci •^ ^ ^ ^ ^ ^ ^ CO Pc "titNOcDfMasCO i3a; a3oqoot>.t^cDcD rt< CO -^ 00 t^ CO O TJH kO I~^ O Tt< oi ■^' O CO o "5 lo lO (M Ttl Tf CO CO CO pq < O osh^c^i-tr-^oococ^iosi-^ • r- T-iiOo^CJ^'cD-^'rtl-^OO "SIS ^ i-ir-((M(N(M(M hji O5ocooooi>o-^g:iqTj; O -H_ (M_ CO ^ co-^'ocbt^oJc^iooi'o >ii ,-H,-.^^^.-lM(M(MCO (M. oqpc^po^. c^kooo rt^»-iiOCJ t>-CDCOiOTti'**i CO>COOT-<.-H^COCOOi pp'S;T)ico(N'-:iq»rtcqc^ipoq ■>^cococococococ6caiM'c^c^i(NC^(N..CO ^^ ppp^jHco^. p -; be CO CO (N M (M' im' C^" r^Ttiooioot>-coocDco 00l>.>O'^CO'— iOOI>-CD pr^coppr-^p'^pc^'o ' CO id 00 ^ ■*' 00 •<# ■-< o o ^ --H M T-; .^ 1— (l>.cDr>»03(^(7^)ujuui-H-':riJU-'T'»-^'— '^jvi^ >* CO(M(M(M(MCOCOCOCOTj cDOT}-00050 oooooooooo oooooooooo C^ CO TJH lO CD 1-- oo PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 37 respect to the weight-height relation. Thus, a lO-month-old Holstein heifer possesses the same weight-height relation as a 15-month-old Jersey, The steers of Groups I, II and III, although differing markedly in their age-weight relationships, were strikingly similar in their age- height relationships, indicating that skeletal growth, at least, was ver}* little affected by the plane of nutrition. Thus, at 4 years of age, the average height at withers for all groups was very close to 56 inches, although the average weights were, in order, 1945, 1238, and 897 pounds. If the height-weight relation of these beef steers is compared with that of the dairy heifers cited by Brody and Eagsdale, it becomes evident that a better agreement exists between the Group II steers and the dairy heifers than between either the Group I or the Group III steers and the dairy heifers. Thus, at a height of 40 inches, the dairy cattle weighed about 365 pounds, while the steers of Group I weighed about 415 pounds, those of Group II about 415 pounds, and those of Group III about 320 pounds. At a height of 45 inches, the dairy cattle averaged about 550 pounds in weight, while the steers of Group I averaged about 710 pounds, those of Group II about 595 pounds, and those of Group III about 450 pounds. For a height of 50 inches, the dairy cattle weighed 855 pounds, the Group I steers 1070 pounds, the Group II steers 810 pounds, and the Group III steers 605 pounds. For greater heights, the Holstein heifers weighed increasingly more than the Group II steers, though far less than the Group I steers. From this comparison, it appears evident that the nutritive condition of the steers of Group II was kept approxi- mately the same as that of growing dairy heifers of equal height, except at the greater heights (above 50 inches) where the condition of the dairy cattle (Holstein only) was superior to that of these steers. This constitutes confirmatory evidence that the body changes occurring in the steers of Group II, approximate what might be considered normal growth changes. If this is true, then the changes occurring in Group III represent subnormal growth changes and need not be considered further in this discussion of the protein requirements of growth and fattening. It is a matter of interest, however, to investigate the difference in nitro- gen content and in the rate of nitrogen deposition between Group II and Group I, as revealed in Tables 13 and 16. The steers of Group I con- tained on an average a larger amount of nitrogen at all ages, and for the first two years averaged a greater rate of nitrogen deposition. Are these differences due only to a greater rate of gTowth in Group I than in Group II, or are they due in part at least to the greater rate of fatten- 38 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL ing? If the latter is true, then an examination of the nitrogen content of the fatty tissues in the two groups of steers should reveal a corre- sponding difference. Does fattening involve nitrogen deposition? — Although all of the fatty tissues from the carcasses of these steers were not analyzed sep- arately, separate analyses are reported for the offal fat, the kidney fat and the fat from the round, loin and rib, except, apparently, when there was no considerable amount of fat on the rib cut. In Table 20, the weights of nitrogen in the analyzed fat samples are summarized and totaled separately for the steers of each group. The steers whose results are collected on the same line of this table were of approximately the same age when slaughtered. The total nitrogen content of the adipose tissues of the steers of Group I was always considerably greater than that of the steers of Group II of approximately the same age, the same relation existing be- tween the steers of Group II and those of Group III. Evidently the deposi- tion of fat in adipose tissue involves cellular proliferation (or cellular protoplasmic enlargement), and therefore fattening possesses its own nitrogen requirement, distinct from that of growth. This difference in adipose tissue nitrogen may account largely for the difference in nitrogen content of Group I and Group II steers of like age, and for the more rapid deposition of nitrogen in the Group I steers during the first two years, although the differences in Table 20 are far too small to account for the differences in Table 13. However, a greater nitrogen content of the other adipose tissues, such as the subcutaneous, intermuscular, and marrow connective tissues, in the steers of Group I over that of the steers of Group II also must have occuiTed. A comparison of the nitrogen content of the lean samples for the steers of Groups I and II slaughtered at ages of two years or more is of interest in this connection. The nitrogen in the lean of the round for the five oldest steers in Group II averaged 94 per cent of the nitrogen in the same samples for the five oldest steers of Group I. For the lean of the loin and the rib, however, this percentage was 78 and 83, respectively. But the lean of the latter cuts was considerably fatter than the lean of the round, and hence the greater nitrogen content for the Group I steers over that for the Group II steers in these cases may have been due to a considerable extent to the increased content of adipose tissue cells. Further light may be thrown upon this question by computing the percentage composition of the added empty weight of Group I over PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 39 O CD CO ■-< t-^ OJ i S "5 00 00 o a; lo •* lO ■^ T»< fc 00 o t^ 00 lO cs ■^ o .-1 -H (M M ■-I (M CM 'i; CO CM CM 00 — < -h' o c»6 o CO •«*< CO I-- ■^ t^ Tf uo 00 lO o c; Tf< p ^ id CO lo O 00 '-- p CO 00 CO 00 CO ■^' i-( CD CO lO lO — lO Tt< 05 o CO CM C^l o o IQ lO iC »0 lO C3i »-< CM C5 05 1> O CM CM CM CM CM 00 CM CM CM ■^ O O CM kO CO CD CM CO CM ?5 co' 05 W5 •g P CO p .-H t>. J CO CM oi CO •^' —I ^ CM Tfi p CM p i-H CO OJ CO oi P CM 05 t^ lO p 00 00 CO p c^i GO CO rt' cd' t^ i-o 00 t^ i^ LO Tj< o CO r- 5 ^ CM CO CM t^ ■^ CM :=3 CM 00 "« t^ »o CO 00 ifi CM p ■* CM ■^ to o i>^ CO »C 00 00 CO CO CO CO CD lO lO CM UO o lO lO 00 CO CO CM 1^ o lO CD C^l i-O .« COiCCO^CM^^^iOCOcOOO O Tjlr-l,— (Ttll>.03lOt^t^05»0 ^ 1— I.— I,— (^CMCOTtiiOlOTti 'OppCOQOTjHCMOOcOp ^oiot^^coc^icoTf'oococo '-''— i^HCM'OiocDOlOS'— 1 C >OpeMTj- C 1— (OOppO^OprJipoCO o lO»oo6^^o6^^CTjoj-^cdo6 pj »-iCMCMCMCOCOOOT}.Tt< C 'OP'-Hpe-JpiOt-.lOp.— I S CM'-Iu5cocMJ~-^Tf-t>.iO'— iC^1'*i»Ot>-CO>— I iOiOt»). However, in this work the maternal tissues and fluids appar- ently^ were not included in the chemical samples. Eckles has shown (^^) that at full term the placenta and amniotic fluid contain 18 per cent as much protein as the fetus (Holstein) ; at younger ages the pro- portion may be much greater. The neglect of these tissues would thus introduce a considerable error in estimating the daily protein require- ments of pregnancy. Illinois investigations on the bovine fetus. — During the last year. Professor W. W. Yapp of the Department of Dairy Husbandry, Uni- versity of Illinois, has secured some valuable data on the composition ' The description of the preparation of the fetuses for analysis given in the bulletin is not sufficiently complete to warrant a definite statement on this point. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 49 of \) fetuses, of dillorciit ages, with their meinhraiies. The amniotic fluid was collected, but its nitrogen content was not determined. Through the courtesy of Professor Yapp, these unpuhlishcd data have been made available for this stud}' of the protein requirements for intra-uterine growth. TABLE 24 Age and Breeding of Bovine Fetuses Analyzed bv W. W. Yai'p Fetus Age from date of Xo. breeding days nam Sire 718 91 F2 Holstein-Guernsey cross Pure-bred Guernsey bull 2 95 Grade Holstein " " Holstein " 26 95 Holstein-Guernsey back-cross to Hoi- '' " Guernsey " stein sire. 21 104 Holstein-Guernsey back-cross to Hoi- " " '' " stein sire. 27 123 Holstein-Guernsey back-cro.ss to Hoi- '• " " " stein sire. 1 162 Holstein-Guernsey back-cross to Hoi- " '' " " stein sire. 644 162 F, Holstein-Guernsey cross " " " " 11 167 Grade Holstein " " Holstein " 19 186 Grade Holstein " " " " TABLE 25 The Protein Content of Bovine Fetuses of Different Ages, FROM Data Secured by W. W. Yapp Fetus No. Age days Weight of fetus and membrane gms. Protein content pet. Protein content gms. 718 91 363 5.85 21 2 95 489 5.25 26 26 95 666 6.16 41 21 104 513 5.59 29 27 123 1122 6.84 77 1 162 6179 9.20 568 644 162 4838 9.22 446 11 167 5531 9.49 525 19 186 13002 11.08 1441 The fetuses were obtained from grade Holstein and Holstein-Guernsey cows slaughtered at varying periods after being bred either to a pure- bred Holstein bull, in the case of the Holstein cows, or to a pure-bred Guernsey bull, in the case of the cross-bred cows. The description of the fetuses is contained in Table 24 and data relating to their nitrogen content will be found in Table 25. Fetuses Nos. 2, 11 and 19 were sired by the same bull, as were the remaining six. 4 50 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL At these early ages, it seems justifiable to treat these data as homo- geneous, in spite of the fact that some of the fetuses were largely of Holstein blood, while others were largely of Guernsey blood. In the same series may be included Eckles' analysis of a full-term Holstein fetus, weighing 95 pounds. With its placenta tbis fetus contained 9162 gms. of protein. Since the average weight of Holstein calves at birth is 89 pounds, instead of 95 pounds, the protein content of the full-term Holstein fetus with its membranes may be roughly estimated at 89 -— X 91(52 = 8585 orams. 95 A mathematical analysis of the data. — The data at hand evidently give only a fragmentary picture of the intra-uterine growth in protein of the bovine fetus. A mathematical description of the data by a con- tinuous function is essential to any effective use of them. The growth in weight of the chick embryo has been closely described by Murray (^*) by means of the simple exponential (or parabolic) equation. W = kt" (25) in wliich W is the weight of embryo, and t is the incubation age. In adapting this equation to the intra-uterine growth of the mouse embryo, MacDowell and Allen(^^) found it necessary to change the significance of t from conception age to something less, specifically 7 days less. This reduced age they call the " embryo age " on the following basis : " Since the first stages of development of a mammal consist of the formation of the pro-embryo, a consideral)le period elapses before the first organization of the embryo proper. This is the justification for assuming embryo age to be less than conception age. In the mouse the first differentiation of the embryo proper (primitive streak) is not found before the end of the first week. Thus the embryological evidence bears out the purely graphical resiilt obtained by shifting the age (t — 7) until the embryo weights fit a logarithmic straight line." Thus, the mouse embryo at seven days after conception and the chick embryo at the beginning of incubation are in practically the same stage of development. Evidently the shifting of the embryo age seven days ahead of the conception age will also include whatever error is made in measuring conception age from the time of copulation with the male. In accordance with the experience of MacDowell and Allen, the nitro- gen-growth data of bovine fetuses were not very closely described by an equation of type (25), but if t is taken as one month less than the age counted from the date of breeding, a fairly satisfactory fit is obtained, the equation becoming W = 0.143 (t-1)*-"' (26) PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 51 W being expressed in grams, and t, the age from breeding, in months. The agreement between observations and corresponding values derived from this equation is given in Table 26. Having obtained a continuous mathematical function descriptive of the relation between the age and the nitrogen content of the fetus, in so far as the limited amount of available data will permit, it is possible to derive a second mathematical function or equation from which the rate dW of nitrogen deposition, — j— ? may be computed for any age The dif- ferentiation of Equation (26) will give dW ■!^ =0.611 (t- 1)3-29 (27) from which, by substituting any desired value for t, the monthly rate of nitrogen retention in pregnancy at the particular time chosen may be TABLE 26 The Nitrogen-Age Relation of Bovine Fetuses , BY Observation and by Calculation from Equation (26) Age in Nitrogen content Fetus Age in Nitrogi en content Fetus Observed Calculated Observed Calculated No. mos. gms. gms. No. mos. gms. gms. 718 3.0 3.4 2.8 644 5.4 71 82 2 3.2 4.2 4.2 1 5.4 91 82 26 32 6.6 i2 11 5.6 84 99 21 3.5 4.6 7.2 19 6.2 231 167 27 4.1 12.3 18.2 Mo. 9.4 1373 1305 obtained. In Table 27, this has been done for successive months from 1 to 9, the results being divided by 30 to give the daily retentions of nitrogen in grams, A tentative estimate of the protein requirements of pregnancy. — In this table an attempt has been made to adapt these values to difEerent breeds of cattle. The values themselves are considered as applying to the Holstein breed, since the fetuses themselves were largely of Holstein breeding and since the value at full-term was taken from a Holstein fetus. The gestation periods of different breeds of cattle are the same, but their birth weights vary considerably. It seems reasonable to assume tentatively, therefore, that the rate of nitrogen retention at any age will vary among the different breeds of cattle in proportion to their birth weights. The values given in Table 27 for the Jersey, Ayrshire, Guern- sey and Dairy Shorthorn have been computed on this basis from the average birth weights given by Eckles(5*), i. e., 89 pounds for Holstein 52 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL calves, 55 pounds for Jersey calves, 71 pounds for Guernsey calves, 72 pounds for Ayrshire calves, and 73 pounds for Dairy Shorthorn calves/ Sex differences have not been considered here, since the percentage error in the estimate is probably greater than the percentage difference in growth rate between the sexes. It is evident from these estimates that the nitrogen requirement of pregnancy is inconsiderable for the first four or five months, but that it rapidly increases from then on, until in the last two months it is equiva- lent to a deposition of 0.17 to 0.26 pound of protein daily for the Hoi- stein cow. TABLE 27 The Estimated Daily Retention of Nitrogen ix the Growth of the Bovine Fetus and Membranes * Daily retention of nitrogen Age in moa. r- ■ Holstein breed gnis. Jersey breed gnis. \ Ouernsey, Ayrshire and Dairy Shorthorn breeds 2 0.02 0.01 0.02 3 0.20 0.12 0.16 4 0.75 0.47 0.61 5 1.9 1.2 1.6 6 4.0 2.5 3.3 7 7.3 4.5 5.9 8 12.2 7.5 9.9 9 19 12 16 * Exclusive of the nitroRen retained in the amniotic fluid. At full term, Eckles estimates ( ^^ ) that the amniotic fluid contains 1.07 pounds of protein, or 77.7 gms. of nitrogen. In so far as this is of fetal urinary origin, it pos.'^esses no significance with reference to nitrogen requirements during intra-uterine growth. In a calf weighing 75 pounds at birth, Popov (^^) found 6 kgms. of pro- tein. Assuming that this was all laid down in the last 100 days of ges- tation and that the digestible protein possessed a biological value of 70, Popov computes that the pregnant cow would require an average of 85 gms. of digestible protein daily for the formation of her fetus. This is equivalent to a daily requirement of 9.6 gms. of digestible nitrogen with a biological value of 100. This estimate is comparable with the values of Table 27 for the 7th, 8th and 9th months for the Guernsey, Ayrshire and Dairy Sliorthorn breeds. THE PROTEIN REQUIREMENT FOR MILK PRODUCTION According to the general scheme of expressing the protein requirements of animals pursued in this paper, the protein requirement for milk pro- ^ The latter three breeds have been grouped together for an average birth weight of 72 pounds. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 53 duction is nioasured by the protein content of the milk produced. This fact was first recooiiized by Haecker, who says in the bulletin describ- ing his fuiulaniental investigations in milk production (^^) : During the time when the feeding experiments in milk-production in review were in progress, it occun-ed to the writer that in order to determine the actual net nutrients required to produce a given animal product, the composition of the product should be known, as well as the composition and the available nutrients in the food which is to be fed for its production, so that the nutrients in the ration might be provided in the proportions needed by the animal. Before a builder bids on a contract, he determines the quantity needed of each of the materials that are to appear in the structure. Without such specifications he would not know how much of each of the different materials would have to be provided. The fat-to-protein ratio in mill-. — It is well known that cows' milk is subject to wide variations in composition, depending upon the breed and individuality of the animal, the stage of lactation, the plane of nutrition, the character of the feed consumed, and possibly other factors. The fat content of milk appears to be the most readily affected by these factors, the sugar and ash content the least. The commercial importance of the fatty constituents of milk and the relative ease of their quantitative determination in toto are responsible for the fact that different milk samples are distinguished and graded in accordance with their content of fat. Therefore, the most profitable and practically significant method of studying the protein content of milk is to consider the fat-to-protein ratio, with the hope of finding some formula from which the protein content of milk may bo satisfactorily predicted if the fat content is known. The protein of milk appears to be more closely correlated with the fat content than any other constituent. Gaines (^') lias determined the correlation existing between the fat content and the content of protein, sugar and ash from Haecker's(^®) published analyses of mixed milk. The correlation coefficient for fat and protein was +0.812 ±0.010, for fat and sugar, +0.263 ±0.027, and for fat and ash, +0.232 ±0.027. Overman and Sanmann(^®), working with a smaller set of data, report a correlation coefficient of +0.729 ±0.017 for fat and protein, and one of + 0.184 ±0.045 for fat and lactose. The prospect would seem favorable, therefore, for devising a predic- tion formula for protein from fat. In 1899, Timpe(^^) derived the rela- tion p = 2 + 0.35f between the protein and fat contents of milk from analyses of milk from 21 cows of various breeds. Van Slyke(^®) in 1908 proposed the relation p=1.6 + 0.4f from a large, but unspecified, num- ber of analyses of herd milk and milk from individual cows of dif- 54 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL ferent breeds. Again, Andersen and Langmack(^^) in 1923 published the results of a study of 1080 analyses of the milk of Eed Danish cows, from which they derived the equation p = 1.597 + 0.446 f. From his study of the 543 analyses of milk published by Haecker, Gaines (^") arrived at the equation p = 1.46 + 0.40 f. These proposed formulas are not satisfactorily concordant. It is par- ticularly significant that Haecker's data on the composition of milk, upon which the Haecker, Savage, Eckles, and Armsby feeding stand- ards for milk production are directly or indirectly based, indicate a much lower protein content for milk of any grade than any of the formulae derived from other sets of analyses. This discrepancy is too large to be accounted for by the use of a different protein factor applied to the total nitrogen analyses. The factor used by Haecker is not reported in his publication.^ Illinois investigations on the composition of cows' milk. — A splendid opportunity to throw further much needed light on the question of the relation of the protein to the fat content of cows' milk was afforded by the courtesy of Dr. 0. E. Overman, of the Department of Dairy Hus- bandry of the University of Illinois in permitting the use for this purpose of about 2000 analyses (largely unpublished)* performed in his labora- tory during the last six years upon samples of milk of known antecedents. In this extensive work, 3-day composite samples of milk were analyzed every 5 weeks from each of 133 cows in the University herds, through 1 to 3 complete lactation periods. Analyses were thus secured upon 1002 samples of milk from 67 Guernsey-Holstein cross-bred cows, 208 samples from 14 pure bred Ayrshire cows, 268 samples from 19 pure bred Hol- stein cows, 200 samples from 15 pure bred Jersey cows, and 321 samples from 18 pure bred Guernsey cows. The results with reference to fat and protein (Nx6.38) have been averaged in groups covering a range in fat content of 0.5 per cent, with the results given in Table 28. The correlation between the fat and protein percentages in these series of analyses is not as high as that noted above. Overman has obtained correlation coefficients of 0.635 for the Guernsey-Holstein analyses, 0.727 for the Holstein analyses, 0.679 for the Guernsey analyses, 0.588 for the Ayrshire analyses, and 0.500 for the Jersey analyses. * Although samples of milk from 53 cows are represented in Haecker's analyses, it is significant that about 39 per cent of the samples (210 in number) were from 17 per cent (9 in number) of the cows, while 57 per cent (309 samples) were taken from 28 per cent (15 in number) of the cows. These relations were called to the attention of the author by Dr. O. R. Overman. ^ The data will be published in full, with a statistical analysis, as a bulletin from the Illinois Agricultural Experiment Station. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL OQ ^ ^ ^ S H. coeococoeO'^eort'Tji iqOoOTj;aJ»Ot>.rt; lo 00 o t^ '^ 05 (M (N 8 o to o o P3 Tti (M fC 05 -^ 00 ^_ CO CO ■^" r}< ^' to !>• O f Tti ■^ 05 Tf t>- CO ■— ' < « a J ^13- o o >« ■* 00 t^ >* 'S<. t>. p ^ CO to p CO CO CO rf Tt< Ti' Tji oO'^ioc^ooio-^oico ppiqQTf;pTt<_00iO COTfJ'^'tOiOiOOcDt^ ki'S-w Sc"So PO-^pT)<,— ITJ<0 <;'~o.(McococoTt< idtdcb •^ cD-^cocDoeo'O.-H o o >-g o <>! .-I rt; p 00 < t '^ CO CO CO CO CO i ^J C>^ 02 o o ^* ^^ ^^ ^^ t^ p iq p c^coco'<^'T}< t^ ci i^ ci t^ M t>: ti c'l <>> CO co' Tji Tt< id id vd CD r>^ t>^ 00 00 Sa" cDcOCDCDcDCDcDCDCDCOcicicO "C- CS» N CO CO ■^' TjJ id id CD CD t>^ t>^ 00 56 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL Concerning the homogeneity of these sets of data, it appears from the distributions given in Table 28 that no distinct discrepancies exist among the first four sets of data. The Jersey data, however, indicate a rather distinctly lower protein content for milk containing more than 3.50 per cent of fat. Since the greater discrepancies occur in those groups pos- sessing the smaller numbers of samples, it appears that no serious error would be committed by pooling all of the analyses together. This has been done in Table 29, and the average percentages in each array have been corrected to the nearest even 0.5 per cent of fat on the assumption of an equal ratio of protein to fat. For example, in the array of analyses from 3.26 to 3.75 per cent of fat inclusive, the actual averages of the 295 TABLE 29 A Comparison of Haecker's and Overman's Averages for the Protein Contents OF Milk of Different Fat Content and of Predicted Protein Percentages Derived from the Equation P = 2.0+0.4 F Per cent protein according to Haecker ()v erni;ni Per cent fat 2.50 Xo. of analyses Protein 2.55' No. of analyses 7 Protein 2.74 Mathematical prediction 3.00 3.00 47 2.68 116 3.06 3.20 3.50 55 2.81 295 3.37 3.40 4.00 57 3.08 443 3.58 3.60 4.50 116 3.27 486 3.81 3.80 5.00 103 3.45 318 4.00 4.00 5.50 89 3.65 191 455 4.20 6.00 39 3.82 89 4.45 4.40 6.50 24 4.02 38 4.46 4.60 7.00 13 422 7 4.84 4.80 1 According- to Haecker: " Tliere were so few .samples containing less than 2.5 per cent of fat, and so many more testing from 2.5 to 2.75, that no satisfactory average could be obtained for milk testing 2.5 per cent butter fat, so the averages were computed from the ratio of variation of milk testing from 3 to 3.5 per cent fat." analyses are 4.53 per cent of fat and 3.40 per cent of protein. The per cent protein corresponding to 4.50 of fat is computed from the proportion 4.50 _ _^ 4.53 ~ 3.40 ■ For comparison, Haecker's analyses ' are also given in Table 29, and in the last colmnn of the table will be found the results of predicting the pro- tein content of milk from the fat content by the equation p = 2.0-1- 0.4 f. 'It is a peculiar fact that, in the appendix of Haecker's bulletin containing presumably the original 543 analyses of milk, the average percentages of fat in each of the 9 classes are equal exactly to the mid-values of the class. If these were random selections of samples, such an ideal result would be statistically impossible with the small numbers of samples analyzed in each group. The obvious conclusion is that some selection of data • has occurred, or that the analyses have been modified in some unrevealed way, in order to secure these ideal averages. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 57 Overman's analyses for protein evidently are considerably higher than Haecker's for milk of all grades/ and for milk containing 3.50 per cent or more of fat, they are remarkably closely predictable by the simple linear equation above given. For lower grades of milk, the equatioii evidently over-estimates the protein content of milk from the fat content. A method of estimating the protein requirements for milk produc- tion.— In view of the systematic method by which Overman's samples were taken, assuring an equal representation of milk from all stages of lactation, and in view of their number and the number of cows and of breeds represented, it appears justifiable to use the Overman data as the basis of estimates of the protein requirements of cows for milk produc- tion. Since thus far protein requirements have been expressed in terms of nitrogen, and since in practice a pound of milk is a convenient unit, the prediction equation above given has been changed to N = 1.45 + 0.29f (38) in which N is the nitrogen in grams per pound of milk, and f is the per- centage of fat in the milk. This equation is applicable only to milks containing 3.50 or more per cent of fat. For milks of lower grade, the nitrogen content may be computed roughly from the Overman averages given ill Table 29. THE PROTEIN REQUIREMENT FOR MUSCULAR ACTIVITY The great variation in the muscular activity of farm animals, depend- ing mainly upon their temperament and upon the nature of their con- finement, offers a serious obstacle to the estimation of their require- ments for those nutrients necessarily consumed in muscular metabolism. It is a matter of considerable importance, therefore, to determine whether protein necessarily serves as a source of muscular work, either directly as the dietary amino acids coming to the muscles from the intestinal tract, or indirectly as the result of an increased catabolism of the muscle tissue itself. Does the muscle cell wear oait? — Since the days of Liebig, the relation of muscular contraction to protein metabolism has been shown to be of less and less importance. The classic experiments of Fick and Wis- licenus, which have been abundantly confirmed, showed that muscular work could be performed at the expense of carbohydrates and fats, and the trend of modern investigation into the chemical and calorimetric •Perkins (Ohio Mo. Bull., 1:304 (1916)) has also published a summary of a series of milk analyses (807), in which the protein to fat ratio is considerably higher than that of Haecker's analyses. 58 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL phases of muscular contraction (^^) is toward the view that carbohydrate is the immediate source of muscular energy. The entire series of events occurring during and subsequent to a muscular contraction can be described, apparently, without reference to protein, amino acids, or the nitrogenous extractives, with the exception of creatine, which seems not to be destroyed in the process (^^). Furthermore, the so-called fatigue products of muscles, in so far as available information indicates, are neither nitrogenous in character nor related to protein metabolism (^*). It is true that investigations are on record indicating changes in the con- tent of purines (^^) and of creatine (^^) in muscle as the result of activity, but the relation of these changes to the activity itself is obscure. The relation is probably indirect, and the function of the non-protein nitro- genous constituents of muscle relative to contraction is quite possibly regulatory only. Functioning in this way it is conceivable that they may not be consumable during muscular activity. To many investigators it has seemed almost axiomatic that muscle tissue must undergo disintegration as the result of contractile activity. And when experimental investigation has failed to indicate any such disintegration, or at least any considerable disintegration, it has been considered necessary to assume that most of the nitrogen thus degraded escapes excretion in the urine by some process of reutilization in the tis- sues (^''), a hypothesis apparently beyond the scope of experimental enquiry. It seems that this view of the necessary wastage of muscle tissue during activity has resulted from the analogy so often drawn between the mechan- ical motor and the animal motor. In the early history of physiology this analogy has served an admirable pur|)ose, and in pedagogy it is still extremely useful. But, like many other analogies, it is only partially true, and if pushed too far it will confuse rather than enlighten. It seems apparent that one phase of motor activity to which the analogy has an extremely doubtful application, if it applies at all, is the wearability of the motor. The mechanical motor undoubtedly wears out at a rate that bears a close relation to the amount or the intensity of the work per- formed. But this wearing out is due mainly to the friction of moving parts on bearings, that is, to a factor that has no l-noivn counterpart at least, in the animal body. Nevertheless, the opinion prevails in many quarters and is frequently expressed in print, that the animal motor also must wear out, and since the substance of the animal motor is largely protein, the conclusion has seemed reasonable, if not inevitable, that the protein catabolism must be increased as the result of muscular work. Eubner's term for the maintenance requirement of protein, the " wear- PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 59 and-tear quota," obviously is based upon this analogy of the mechanical motor and obviously implies that the catabolism of tissue protein will increase with increased motor activity. The term is an unfortunate one, particularly since it has attained a wide currency. Reviews of the literature — a critical comparison. — Since as early as 1855, the influence of work on protein metabolism has been a favorite subject of inquiry and a considerable number of experiments have been reported in the literature. However, the results secured have been variable and at the present writing opinion is divided as to whether work increases the catabolism of tissue proteins or is without appreciable effect. From time to time these experiments have been reviewed, but it appears to be difficult to reconcile them with any definite conclusion. In 1909 they were reviewed by Tigerstedt(^^), who was not convinced that there is an increase of nitrogen metabolism resulting from muscular work. On the other hand, they were reviewed by Magnus-Lev}'(^^) at about the same time (1907) and the conclusion was reached that it is probable that such an effect occurs. In 1917 LuskC**) concluded from a study of the same literature not only that " muscular work does not increase protein metabolism," but also that " the character of the protein metabolism is unchanged by muscular activity." However, in 1925 Cath- cartC^) again reviewed the literature, which, in conjunction with recent experiments of his own, induced him to state that, in spite of this very definite conclusion of Lusk, "the accumulated evidence seems to me to point in no unmistakable fashion to the opposite conclusion, that muscle activity does increase, if often only in small degree, the metabolism of protein," although " there is no possible ground for the view that protein is the source of muscular energy." But the former statement may be more nearly true than Cathcart himself suspects. It would be far beyond the scope of this paper to inquire into the many reasons why critics weigh essentially the same evidence with such discordant results. It appears to be due in part to different criteria as to what constitutes competent experimental evidence. Thus, Atwater and Sherman ( ''2) insist that the diet during the working period should be increased in its energy content to allow for the increased energy require- ment; otherwise, an increase in the protein catabolism may mean simply that tissue protein is being destroyed as a source of muscular energy, rather than as the result of an inevitable wear on the muscular tissues. Cathcart, however, is not so discriminating, and does not disregard or even discount experiments in which no such assurance is given. Again, the results of some experiments are considered in- toto by some, and only in part by others. Shaffer's C^) well-known experiment offers a case in point. GO PROTEIN REQUIREMENTS OF CATTLE: MITCHELL Considering the average urinary excretions of rest and work periods, Shaffer and Lusk both conclude that work has had no effect on the nitrog- enous metabolism. Cathcart, however, notes a marked increase in uri- nary nitrogen on the fourth Avork day, and, without reference to the rest periods at all, implies that this is a result of the work performed. Cath- cart's recent experiments C*), although very carefully controlled and undivided in their obvious interpretation, seem no more convincing than the much earlier experiments of Wait^^) at the University of Ten- nessee, which yielded entirely negative results. These experiments of Wait are not considered in Cathcart's review. A definition of the problem and its bearing on experiniental metliods. — The maze of conflicting experimental results relative to the effect of work on protein metabolism suggests that a precise definition of the real prob- lem, followed by a consideration of the proper experimental conditions to impose, might be of value in deciding which of the recorded experiments are capable of the most exact interpretation. Since there is now no ques- tion of the availability of protein as a source of energy for muscular work, the real problem becomes that of determining whether or not there is an inevitable disintegration of muscle protein (or of other nitrogenous com- pounds) as a result of muscular activity. This definition of the problem imposes certain necessary conditions upon any experimental method of attack. In the first place, there should be an adequate intake of energy during the working periods, otherwise an increase in urinary nitrogen may mean a destruction of tissue protein merely to serve the unnecessary role of a source of energy. If no evidence is obtained of an increase in tissue destruction as the result of work, this in itself would seem to be satisfactory evidence that the energy intake was adequate. On the contrary, if the experimental results indicate an increased destruction of tissue, the investigator himself must assume the burden of the proof that the energy intake was adequate, and hence that the increased tissue catabolism was an inevitable consequence of muscular activity. Thus, the significance of negative results in the investigation of this problem would seem to be much more easily estab- lished than the significance of positive results. The only known method of investigating the extent of the destruction of tissue protein under any experimental conditions is to determine the nitrogen content of the urine on a nitrogen-free diet of adequate energy value, and even under these conditions the contribution of the non- protein nitrogenous constituents to the urinary nitrogen cannot be de- termined. It would seem to be impossible, therefore, to rule out the role of the latter constituents in the phenomena under investigation. How- PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 61 ever, it is equally clear that if protein is included in the experimental diet, at least in quantities above the absolute minimum required to replace endogenous losses of nitrogen, the significance of the urinary nitrogen is difficult if not impossible to establish definitely. Under sucli conditions the urine contains the products of an exogenous protein metab- olism, the intensity of which for a given protein intake will vary with the intensity of the prevailing anabolic processes and with the extent to Avhich the protein is utilized in anabolism. Hence, even though the pro- tein intake is kept constant in rest and work periods, the output of urinary nitrogen might be affected if muscular activity alters either the intensity of the anabolic processes or the extent to which the dietary pro- tein is utilized in anabolism. It is not inconceivable that an increase in muscular activity may do both, and hence a slight increase in the day's urinary nitrogen in work as compared with rest periods may bear no relation to the intensity of the minimum endogenous metabolism. On the other hand, if the energy intake is adequate, an increased catabolism of muscle due to increased activity may be entirely obscured if the protein intake exceeds the requirements, since dietary amino acids may be diverted from catabolic to anabolic processes with no effect on the output of urinary nitrogen. Returning now to past experiments on the effect of work on protein metabolism, as reviewed for example by Cathcart, the increases in the excretion of urinary nitrogen as a result of work, so frequently noted, are in most cases not accompanied by a demonstration that the intake of energy was adequate; in fact, in a considerable number of cases a suspi- cion that the contrary was true is almost inevitable. Furthermore, in the large majority of experiments, the protein intake was much above the minimum requirement, and in some cases was excessive. The precise interpretation of the positive results obtained with such diets seems im- possible. They may have resulted from a depression of the anabolic processes involving dietary protein, or from a depression of its biological utilization. Although it is tnie that work tends to favor muscular hyper- trophy, it seems more probable that this increase in anabolism is an after effect of work rather than a contemporary effect, since the immediate result of muscular contraction is undoubtedly catabolic. Even in the adult, some tissues are continually growing. If this growth is inhibited during muscular activity on a protein-containing dietary, the urinary nitrogen will be increased by a corresponding amount. The effect of mus- cular activity on the utilization of absorbed protein in metabolism is entirely unknown, but it is not inconceivable tliat a depression of utiliza- tion, ^nth a corresponding increase in urinary nitrogen (or sulfur) may 62 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL result. It might even be considered a matter for surprise if the great acceleration in the catabolic processes of the body during muscular work did not conscript some of the dietary amino acids that otherwise would be used for the growth of hair or other tissues, or for the replenishment of the digestive glands, or for the replacement of the nitrogenous losses resulting from the minimum endogenous catabolisms. Entirely aside from such admittedly hypothetical considerations as these, it would seem a futile undertaking to detect the effect of muscular work upon the rate of tissue catabolism that is measured normally, in the human subject, by the excretion of 1.5 to 3 grams of nitrogen in the urine daily, by imposing upon this an exogenous protein catabolism yielding 5 or 10 times as much urinary nitrogen. Such considerations as these lead inevitably to the conclusion that the characteristic effect of work on tissue catabolism should be investigated under conditions such that the exogenous catabolism of nitrogen is entirely eliminated or is reduced to an insignificant minimum. This can be accomplished by the feeding of a very low-nitrogen diet, or, if possible, a nitrogen-free diet. Furthermore, the experiment proper should be preceded by a feeding period of sufficient length to remove the " deposit protein " from the tissues, since this stored nitrogen will tend to vitiate the results of the experiment in the same way as would dietary nitrogen. On protein-containing but creatine-free dietaries, the excretion of creatinine nitrogen is generally considered to be proportional to the total endogenous nitrogen. Hence the effect of work upon the creatinine excre- tion is taken to indicate the effect upon muscle endogenous catabolism. Although the reasoning upon which this conclusion is based is not forti- fied throughout by adequate experimental data, it may be considered justifiable if not entirely convincing. The crucial experiments. — Of all the experiments that have been reviewed so thoroughly from time to time, there appear to be only two in which the diet was approximately nitrogen-free and at the same time was approximately adequate in energy value, and in which the period of nitrogen-free feeding was sufficiently extended so that the urinary nitro- gen had approximated the endogenous level. These are the experiments of ThomasC'^®) reported in 1910, and those of Kocher(2') reported in 1914. Thomas reduced his output of urinary nitrogen to approximately the endogenous level by subsisting on a diet of pure sugar. Following a fore- period of four days at this level, was a work period of three days during which work on an ergostat was performed amounting to 120,000 kilogram- meters. The experiment was concluded by an after-period of three days PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 63 on the same diet. The daily excretions of urinary nitrogen were as fol- lows: fore-period, 3.05, 2.73, 3.32, and 2.85 gms. ; work-period, 2.47, 2.90, and 2.97 gms.; and after-period, 2.71, 2.22, and 2.31 gms. Judged by a comparison of the average excretions only the work had no effect on the ex- cretion of urinary nitrogen, the averages being, respectively, 2.96, 2.78, and 2.41 gms. If it is considered that in the latter period only was the endogenous level of tissue catabolism reached, Thomas concludes that the work might have effected a slight increase in muscle catabolism. However, if the endogenous catabolism were not attained until the last few days of the experiment, the increased excretion of nitrogen in the work period above this level would appear to be due more probably to the catabolism of " deposit protein " than to an effect of work. This experiment of Thomas, therefore, does not support the conclusion that muscular work increases the endogenous muscular catabolism. It is to be noted that definite proof of the adequacy of the diet in energy value during the work period, by a comparison of intake with outgo of energy, is not required when the urinary nitrogen shows no increase in this period. Kocher's experiment involved two subjects and a study of the nitrog- enous constituents in the urine. The diet used contained a minimal amount of nitrogen (1.01 gms. per day) derived entirely from cream and contained over 5000 cals. daily. In each case there was but one work day, during which the subject walked 60 kilometers (about 37.8 miles) in 10 hours. AVith both subjects the total urinary nitrogen and the creat- inine nitrogen increased on the day of work, and the increased excretion continued for the next day or two. The effect is much more noticeable in the first subject than in the second, a fact possibly correlated with the much lower creatinine coefficient of the first subject. Unfortunately there is not sufficient assurance that the calorie intake of the work days was adequate. On the basis of average values for the energy requirements at rest and during horizontal walking, Kocher estimates that the energy intake of Subject E. A. K. was about 4.0 per cent in excess of the requirements of the day of work ; with Subject J. G. F. the estimated excess was 4.6 per cent. However, the estimates of energy requirements might conceivably be in error by much more than this. Hence, the ex- periments do not demonstrate that an increase in endogenous catabolism is an inevitable consequence of muscular work ; in other words, that there is an inevitable wear on the muscle machine. In further experiments on the same subjects, using a high-protein high-fat diet, the urinary nitrogen showed no increase as a result of work when the calorie intake was high, but a distinct increase when the calorie intake was deliberately lowered so as to be obviously insufficient. 64 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL With the other subject, on a high-protein diet, even when the calorie intake was obviously insufficient, the urinary nitrogen did not increase as a result of work. Of more significance than the total urinary nitrogen on a high-protein diet, are the values for the creatinine excretion. These appeared to be unaffected by work, except for a slight increase on the day of work in one of the experiments in which the energy intake was delib- erately made insufficient. The experiments of Cathcart and Burnett ('*) also indicate a slight effect of work upon the creatinine excretion. The diets used in the differ- ent series of experiments varied in nitrogen content, but were all estimated to contain 2900 cals. Their specific dynamic effect probably approxi- mated 200 cals. In each series the diet was constant in the pre-work periods of 4 days, the work periods of 6 days, and the post-work periods of 4 days. The work was performed for one hour daily on a hand-lever ergometer and was equivalent to 25,000 kgm. meters (equivalent to 58.6 cals.). Depending upon whether the efficiency is taken at 20 or 25 per cent, the heat output due to this quantity of work may be taken as 295 or 236 cals. The subject weighed 79 kgms., and, with a surface area of 1.99 square meters, would have a basal requirement of 1920 cals. If the estimate of the energy intake is accurate, and the other activities of the working days did not exceed 500 cals., the energy intake may be con- sidered as adequate. During the twelve months of experimental feeding the subject gained 1.6 kgms., but in seven of the eleven working periods, a slight loss in w^eight occurred. The adequacy of the energy intake for the entire experiment cannot be doubted, but it is unfortunately not equally clear that the diet was adequate during the working days. Al- though it is difficult to interpret the figures for total urinary nitrogen and sulfur with reference to an effect of work, for reasons fully explained above, the slight increases in creatinine nitrogen during the work periods •were sufficiently consistent to indicate a direct or indirect effect of work upon the endogenous catabolism of muscle. The experiments of Kocher and of Cathcart and Burnett ' thus indicate an increased excretion of creatinine simultaneous with the performance of muscular work, but other investigators C'^^' '''') have obtained no such increases. If the creatinine excretion is followed at intervals as short as two hours, the output during a short period of work has been found by Shulz('8) to be invariably increased, often to a large extent; however, 'A more recent experiment by R.C. GaiTy (J. Physiol., 72:364 (1926-27)) possesses much the same significance as that of Cathcart and Burnett for work on an ergometer, but no clear eflfect of static effort on the excretion of total nitrogen or of creatinine was reported. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL C5 in some later period of the day, generally the following period, the creat- inine output is with equal regularity decreased, so that the total day's output of creatinine is not appreciably affected by the work performed. HartmannC'^^) found much the same fluctuations in the uric acid and phosphoric acid content of 2-hour urine samples, and suggests that such variations are not due to variations in the production of these compounds, but in the excretory activity of the kidneys as a result of muscular work/ That the excretory activity of the kidney is affected by muscular work has been clearly shown by Wilson and associates, by Dobreff, and by Asher and Weber (*®). Impressed by tlie scarcity of experimental investigations planned to determine directly the effect of work on tissue catabolism, and by the dis- cordance among the results obtained, Mitchell and Kruger(^^) have recently performed a number of experiments on rats in which the effect of work on the excretion of endogenous nitrogen in the urine was deter- mined. The results of 19 experiments clearly indicate that muscular effort, either static or motive, may be performed with no appreciable increase in the excretion of total endogenous nitrogen or of creatinine in the urine, and hence presumably with no appreciable increase in the catabolism of muscle tissue. This was true whether the diet was predominantly car- bohydrate in character, or whether it was predominantly fat and contained only traces of carbohydrate. These experiments on rats, as well as the experiment of Thomas on himself, point to the conclusion that an in- creased catabolism of muscle tissue is not an inevitable consequence of increased muscular activity.' It is equally clear, however, that if the diet does not provide sufficient non-nitrogenous nutrients to supply the work- ing muscle with energy, the muscle tissue itself may be sacrificed for this purpose. Whether an accelerated breakdown of muscle tissue results from any other contingency during contraction cannot definitely be decided from available information, although several possibilities suggest themselves. Intense muscular work, in which the rate of consumption of non-nitrog- enous nutrients by the muscles greatly exceeds the rate of replenish- ^ This explanation is supported by the experimental findings of Rakestraw (J. Biol. Chem., 47: 565 (1921)), as well as of Levine, Gordon and Derick (J. Amer. Med. Assoc, 82: 778 (1924)), who have noted consistent increases in the uric acid concentration of the blood during work, possibh^ due to diminished kidney function. 'In work experiments on dogs. Chambers and Milhorat (J. Biol. Chem., 77: 603 (1928)) have more recently shown that the urinary output of nitrogen in fasting was greatly increased by work, but that this increase could he entirely obviated by the administration of carbohydrates. 66 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL ment by the blood, may ultimately be shown to stimulate muscle break- down. Possibly, also, if the general body temperature, or the temperature of the working muscles, increases above a certain point, the endogenous catabolie processes, whatever they may be, are speeded up, with a result- ing increase in the output of endogenous metabolites in the urine. Ex- periments by Myers and Volovic and others (^^-^ g^^id by Linser and Schmid(*') show that increases in body temperature artificially induced in healthy subjects may increase the endogenous catabolism. The interesting results of Lee and Tashiro(^*) indicating an increased production of ammonia in excised muscles during contraction may have a bearing on the general problem of the effect of work on the endogenous catabolism. On the other hand, this production of ammonia may be due to the abnormal condition of the excised muscle, or to the lack of readily available sources of energy, or to anoxemia (*^). The protein requirement of ivorh may he considered practically non- existent.— Until evidence to the contrary is forthcoming, it appears justifiable to assume, however, that ordinary work carried out under favorable conditions does not accelerate the breakdown of muscle sub- stance, and that, even with working animals, the slight increase that might occur would be more than covered by the increased protein intake necessarily accompanying the increased intake of total food. Hence, the balancing of the rations of farm animals with respect to protein need not involve any consideration of the amount of muscular activity that the animal will undertake voluntarily, or that would be imposed upon it under the usual conditions. ESTIMATING THE PROTEIN REQUIREMENTS OF CATTLE UNDER ANY SET OF CONDITIONS The problem factored. — The problem of the nutrient requirements of animals is solved satisfactorily only when it is factored into its ultimate and independent terms. The amount of protein required to nourish a pregnant heifer in milk cannot be applied directly to another heifer dif- fering either in size, stage of gestation, or rate of milk production, nor can a satisfactory correction be made for such differences unless the requirement for each independent function is known and some informa- tion is at hand relative to the manner of its change with size, time, or intensity of functioning. It has been the purpose of this paper to attempt such factoring of the protein requirements of cattle in so far as available experimental data will permit. The use of the information thus obtained may be readily illustrated and an approximate method of putting the results in a form for practical PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 67 application to feeding problems may be devised, but the tentative nature of the data must be constantly kept in mind and the existence of certain interrelations between animal functions occurring simultaneously must be pointed out, even though nothing quantitative can be said about them at present. The need for further information on many points throughout this scheme of measuring the protein requirements of cattle suggests con- tinually a number of profitable lines of reasearch. Using the factors in any particular case. — An 800 pound Holstein heifer requires for maintenance an amount of nitrogen equivalent to 0.030 gm. per kilogram of weight per day (see page 11), or 1.36 gms. per 100 pounds, or a total of 10.88 gms. For growth it would require daily 4.11 gms. of nitrogen (Table 23), making a total requirement of 14.99 gms. of nitrogen, or 0.206 pound of protein (N x 6.25). If the heifer is in calf and is at the end of her sixth month of gestation, her daily nitrogen requirement should be increased by 4.0 gms. daily (Table 27), and if she is in addition producing 15 pounds of 4 per cent milk," a further consider- able quota of 39.1 gms. of nitrogen must be allowed daily (Equation 28), thus raising her total nitrogen requirement to 58.1 gms. of nitrogen, or 0.80 pound of protein (l^xQ.25). The composite nitrogen (or protein) requirements obtained in this way for any combination of conditions represent the amounts of nitrogen that are expended by the animal and used by the animal in the elaboration of new tissue or of milk. The amounts of digestible dietary nitrogen needed to cover these requirements must allow in addition for the wastage of dietary nitrogen in metabolism. The biological value of a given source of nitrogen measures the mini- mum wastage of absorbed nitrogen in tissue syntheses. With information concerning the biological values of the digestible nitrogen of different feeds and rations, it would be possible to convert the nitrogen require- ments of the animal, compounded in the fashion just illustrated, into re- quirements for digestible nitrogen or protein. Thus, a requirement of 0.8 pound of protein, would, for a source of nitrogen possessing a biologi- cal value of 50 at approximately the level of protein feeding appropriate for the specified conditions, call for 1.6 pounds of digestible protein. It should of course be realized that the biological value of a given source of nitrogen may vary, depending upon the purpose for which it is to be used in the animal body. The biological value of feed protein for ruminants. — In a tentative way, can it be safely assumed that the biological value of ordinary sources of nitrogen in cattle rations is 50 or more ? If so, the values Cor the nitrogen ^ Admittedly an extreme, though not an impossible, illustration. 68 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL requirements of cattle derived above can be roughly used in estimating the reqiiirements in terms of digestible protein. Laboratory animals have given biological values for feed proteins of 60 or more, provided the protein is not fed at a level higher than 10 per cent (^^' ^*). Will rumi- nants give values of approximately the same order of magnitude? Hart, Humphrey and Morrison (^®) have reported nitrogen metabolism studies on four growing calves consuming rations containing approxi- mately 10 per cent of protein, either entirely from the corn plant or largely from the alfalfa plant. The data obtained appear to be suitable for estimating roughly the biological values of the dietary nitrogen upon the basis of certain assumptions concerning the magnitude of the endog- TABLE 30 EsTiMATKD Biological Values of THE Nitrogen of Alfalfa and Corn Rations FROM Data Obtained on Growing Calves BY Hart, Humphrey AND Morrison Animal No. Hody \vl. Ration kgnis. Oailv N intake gms. Daily fecal N g-ms. Daily urine N gms. Daily iiietaliolio' \ trills. Divily cii (logons \ = Kins. Biolcsical value 2 pet. 1 Alfalfa 258 65.8 29.3 22.8 19.0 7.7 73 Corn 235 59.9 27.3 24.5 16.0 7.1 &4 2 Alfalfa 202 58.0 26.0 25.4 16.7 6.1 60 Com 230 66.4 33.2 26.5 18.0 6.9 62 3 Alfalfa 159 49.0 19.9 15.0 14.3 4.8 76 Corn 179 60.8 24.5 21.6 18.0 5.4 70 4 Alfalfa 197 65.9 25.0 21.4 19.3 5.9 74 Corn 179 59.0 21.3 16.3 17.5 5.4 80 * .\ssunied to equal 0.5 gni. per 100 gnis. dry matter consumed. All rations were assumed to contain 90 pet. dry matter. - Assumed to equal 0.030 gm. per kgm. body weight. , „. , . , , N intake — (Fecal N — met. N) — (Urine N — end. \) * Biological value = lOOx „ . ^ , t^^ r^r; , .„ N intake — (Fecal N — met. N) enous nitrogen in the urine and the metabolic nitrogen in the feces. The former is taken to be 0,030 gm. of nitrogen per kilogram of body weight, and the latter 0.5 gm. of nitrogen per 100 gms. of dry matter consumed (^' p- ^^). The calculations will be found in Table 30. The biological values average 71 for the alfalfa rations and 69 for the corn rations. These values are in good agreement with what would be expected from laboratory animals, and may be taken to justify the tenta- tive use of the value 50 in converting protein requirements in terms of animal expenditures and storages to protein requirements in terms of digestible protein. Probably this value includes a considerable margin of safety, particularly with reference to the use of dietary ])rotein in milk production. Future experimental work, it is hoped, will serve to differ- entiate the different farm feeds and rations with respect to tlie biological values of their protein constituents. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 69 Tentative estimates of the total requirements of calves for digestible protein of definite biological value, compared with Armsby's stand- ards.— It will be of interest to compare estimates of the digestible protein requirements of calves of different weights and breeds, as thus obtained from the data developed in this report, with the digestible protein re- quirements of the Armsby feeding standard, which in turn were taken from Kellner's standard. Such a comparison is made in Table 31. It should be emphasized that the estimates in Table 31 based upon the information developed in this report (see Table 23, and page 11) are TABLE 31 Thk Requirements of Digestible Protein by Growing Calves According to the Data Developed in this Report, as Comp,\red with the Armsby Standards Hereford-Shorthorn calves: Dairy cal ves: Daily requirement of digestible protein Daily requirement of d igestible protein Armsby's standard * lbs. Tlie data of this report 2 Armsby's standard* lbs. The data of this report^ lbs. Body weight lbs. Holstein Ib^. Ayrshire lbs. Jersey lbs. 200 0.90 0.67 0.60 0.51 0.66 0.63 300 1.07 0.65 0.72 0.48 0.51 0.49 400 1.20 0.62 0.80 0.46 0.44 0.43 500 1.32 0.60 0.83 0.44 0.40 0.39 600 1.40 0.59 0.85 0.43 0.37 0.36 700 1.40 0.57 0.85 0.42 0.36 0.35 800 1.37 0.56 0.85 0.41 0.35 0.34 900 1.30 0.55 0.41 0.36 0.34 1000 1.30 0.54 0.42 0.38 1100 0.54 0.43 1200 0.54 0.45 * In terms of true protein. ^ On the assumption of a biological value of .50 for the dietary protein. Since the biological value is determined from the utilization of the total nitrogen of the ration, the estimated requirements obtained by the use of biological values refer to crude protein. estimates of protein requirements for rates of growth equal to those ex- hibited by the Missouri Hereford-Shorthorn calves of Group II, and by Eckles' various groups of dairy heifer calves. If these calves did not grow at normal rates, then the estimates of protein requirements are corres- pondingly too low. It will be noted that the estimates given are much lower than the Amisby standards, generally 50 per cent or more lower.' Can such low estimates of the protein requirements of calves be reconciled with the general trend of modern investigations, or must they be considered the * The difference is really greater than appears from the figures themselves, since the Kellner- Armsby standards refer to digestible true protein, but the estimates from this report refer to digestible crude protein (NX6.25). 5 70 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL result of inaccuracies in the information upon which they are based, or of an incomplete summation of the independent factors involved? In his discussion of the protein requirements of farm animals, Armsby himself (^' p- 412-414^ j^^^g referred to a number of investigations, including some of his own, that appear to show that animals can support what seems a normal rate of growth upon a supply of protein little greater than the maintenance requirement plus the amount actually stored. In fact, in embodying the Kellner protein standards in his own feeding standards for animals, Armsby says : On the whole, one can hardly fail of the impression that the requirements for protein as such in growth have been over-estimated and that the organism may- utilize its protein supply more economically than the current feeding standards would indicate; in other words, that the actual protein supply may be made considerably smaller than has been supposed before it becomes a limiting factor in growth. Until this impression is confirmed by more extensive investigation, however, it appears the safer course to adhere provisionally to the accepted standards A comparison of the tentative estimates in Table 31 with the results of the Armsby cooperative experiments. — On the initiative of Armsby, a series of cooperative investigations on the protein requirements for the growth of cattle was undertaken by eight agricultural experiment sta- tions in this country under the auspices of the National Eesearch Coun- cil (^''^). The plan of these investigations called for the testing of two rations, identical in respect to the source of protein, but differing in the content of this nutrient by reason of a variable inclusion of starch. The low-protein ration was to contain 20 to 35 per cent more digestible true protein than an estimate of the minimum requirements obtained by add- ing a maintenance requirement of 0.5 pound daily per 1000 pounds live weight and an estimate of the daily protein retention per 1000 pounds from Armsby's equation given above (p. 41, Equation 21). The high- protein ration was to conform to Haecker's standards, and contained from 50 to almost 100 per cent more protein than the low-protein ration. The calves were to be fed in pairs, selected for similarity of age, sex, weight, and breeding, both to receive the same intake of net energ}', one from the low-protein ration and one from the high-protein ration. The growth of the calves was to be followed by nitrogen balance studies and by deter- minations of the increase in body v/eight and body measurements. For one reason or another, it was found impossible to conform to the plan as outlined, mainly because the digestible nutrient and net energy contents of the rations fed were found to be considerably lower than were expected from average analyses and digestibilities, and because the high-protein rations were generally more acceptable to the calves and were I' ROT BIN REQUIREMENTS OF CATTLE: MITCHELL 71 more readily consumed than the low-protein rations. The results of the experiments were not considered to be sufficiently significant and con- cordant to warrant positive statements concerning either the adequacy of the low-protein ration for a normal growth rate, or the superiority of the high-protein over the low-protein ration in growth-promoting value. In the metabolism experiments that were undertaken upon a minority of the calves, the high-protein calves in general seemed to be storing nitrogen at a considerably greater rate than the low-protein calves, though the significance of the nitrogen balances obtained is seriously questioned by the investigators themselves, either because the collection periods or the periods of preliminary feeding were too short, or because the bal- ances were quite inconsistent with the live weight increases. On the basis of gains in live weight, differences between high-protein and low-protein calves were not consistent, and in a large proportion of comparisons the low-protein calf gained faster than its mate on the high- protein ration. However, significant comparisons cannot readily be made between high-protein and low-protein calves, since, either intentionally or in- advertently, the net energy intakes of the paired calves were not equal- ized, the result being that differences in growth cannot be interpreted with reference to differences in protein intake only. It was very clearly shown in these investigations that the planning of experimental rations and the control of the consumption of food by experimental animals can- not safely be based upon the use of average analyses of feeds and average digestion coefficients. The performance of the low-protein calves is of particular interest to this discussion, since their intake of crude protein was of the same order of magnitude in most cases as the estimates of this report summarized in Table 31. The data on these calves pertinent to the question at issue are given in Table 32.^ The computed normal growth of the calves was ob- tained from the growth equations given above on the basis of the initial weight of the calf, rather than the initial age. Values for true protein have been converted into values for crude protein on the basis of the find- ings in the digestion trials relative to the ratio between digestible true protein and digestible crude protein. The markedly subnormal growth of the first four Massachusetts calves, receiving slightly less than the estimated requirements of protein is not surprising in view of the large underestimation of the energy intake. ^ Two of the low-protein calves in the North Dakota experiments are not included in this table, since their intake of digestible crude protein was considerably greater than the requirements as estimated in this report. 72 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL The high-protein calves in this experiment also gained at a subnormal rate. The subnormal growth of the two Virginia steer calves, of two of the Pennsylvania calves, and of the one Maryland calf is readily explain- able on the basis of the low intake of protein, markedly lower than the estimated requirements. The following seven calves made approximately normal or supernormal gains in weight on amounts of protein but little above the estimated requirements, in some cases in spite of an energy intake much less than the intended: Massachusetts Nos. 18, 20, and 23,^ Virginia ISTos. 3 and 4, Pennsylvania No. 974, and North Dakota No. 3A. It is true that Massachusetts calf No. 24 and Pennsylvania calf No. 1033 made subnormal gains on amounts of protein almost exactly equal to the estimated requirements, but on the other hand, the Ohio calf grew at a rate much faster than normal on an intake of digestible crude pro- tein that must have been considerably less than the estimated require- ments. No calf in the entire series of studies receiving more protein than the estimated requirements grew at a subnormal rate. On an average of 18 digestion trials, the actual intake of digestible crude protein by the low protein calves was found to be only 68 per cent of that expected from average analyses and average coefficients of digesti- bility, a fact readily explainable by the inclusion of large amounts of starch in their rations.^ The starch thus added would increase the excre- tion of metabolic nitrogen in the feces without increasing the nitrogen intake. Hence, the estimated intakes of digestible crude protein in the Ohio, North Dakota, South Dakota, Maryland, and Nebraska experiments are probably much greater than the actual. From this study it seems fair to conclude that calves can grow at a rate as great as that expected from available data on normal growth with intakes of digestible crude protein but little if any greater than the estimates developed in this report. In fact, with more liberal energy in- takes, it may be possible that such growth could be attained upon even smaller intakes of protein than these. A comparison of the tentative estimates of protein requirements in Table 31 with the Missouri data on dairy calves. — Another extensive investigation of the protein requirements of growing calves was conducted at the Missouri Agricultural Experiment Station from 1913 to 1931 by Swett, Eckles, and Ragsdale(^8). A total of 34 Holstein and Jersey * In the second Massachusetts experiment no consistent difference was noted between the gains of the high-protein calves and those of the low-protein calves. ^ The low-protein ration in the North Dakota experiments was not composited according to the plan of these studies and contained no starch. Hence, the estimated crude protein intake is not subject to this considerable error. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL O C 3 t. o!T3 c a o) B o ** «; aj o. ^ -2 CO Pi H i:'2 x: S hf c ■a b. c; °'i ^ OJ cr o X « Q COCDt^t-^CO?DOO!>J 00 CO o o o o o o Min. r quirein accord toth repo lbs io d d d d d d d d d d lO d d d d d ■3 S |ci:5x CO CO CO o T»< lO 03 05 lO lO CO 00 tK 5 Cl- o o o o o o o o • O o o o o o r- t- CO M CO o »o CD c^ tq o -"^ Ti< CD CO CO CO ■^_ « p g . — < (M 05 O ■-i:n^ ddod •<* lO Tf OOOOOOO"— IrH oooooo^o-^ooo ?, u S § a< X (N CO W H g 1-1 H < S H H o o o c g^o i^-^ppo>oqooo(^^lOO«pQO ^^^,_i^,_;,-<^,- (-5oooooOor-'OiooooococacocooTt<'^oocooocDcDt^C3 t3-c:;-oooooooooooooooot^cDcDi^t^coco-^-<*. CO .2-&rf (NCDTt. ITS O CD OOCOO>OOCOOCDOl(M Ttoioco(Meo CDt^CO'^'— lOiO'^iC p^ o p:J O t, t, O {^ t^ OJ PQPhCBC«K02C»Ph cu a •a ■t5 Vi ^ -S CO IB CQ > o <5 K w ^H ^M j-i _^ ,^^ o .tl .t:3 .tj *S .S^ -i^ -i^ rt^ ^\ n\ A\ f\\ ni •"! ni fll rt^ fli fl_) flJ dl .s .s .s '53 'S 'S "o "o o o o o fl (3 fl fl fi C fl O O fl^ O O Oj o .S c o .s .s . >. >. o o o o o o o WMWWWW^WWfflWWWM^W^^WWO^^^^W X -• O T}< CD 00 00 CO Tti CO -^ "^ IM 2 ^ g^ o o ,i^ ^ c« 03 O Q -C ^ -*— » -^ o o C^ rt< CO CO o o ^i£^J£i£J£i£^i£"g?=;--oS^^-2 =-5 n a 3 3 c-o- ft; O) ^^ %% o o a o. 74 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL heifers was under experimental feeding for periods ranging from 90 to 840 days. Different levels of protein feeding were tested on these ani- mals in periods ranging in length from 30 to 750 days. The level of protein intake was varied by varying the proportions of timothy and alfalfa hay, and of corn, oats, bran, cottonseed meal, dried skim milk, starch and sugar in the rations. The lower protein rations thus contained more timothy hay, corn, oats, starch, and sugar than did the higher pro- tein rations. The number of different protein levels tested on each animal as well as the order in which they were tested was irregular for most of the animals. The feeds were analyzed during the course of the experi- ments, but average digestion coefficients were used in computing the intakes of digestible crude protein. The energy intakes were computed by the use of Armsby's tables. The measurements of growth included body weights and body dimensions. The results are summarized according to a number of different plans, and upon these summaries the final conclusions are based. It was found that dairy heifers could make normal growth, in accordance with Eckles' standards, upon amounts of protein far smaller (56 to 77 per cent for Holstein and Jerseys, respectively) than those called for by the Armsby standards; in fact, with an increase in energy intake, normal growth was attained on still smaller intakes of protein. However, an increase in protein intake above that requisite for normal growth increased the rate above the expected. It is also concluded that the Holstein heifers were able to make dis- tinctly better growth with reference to their normal, on lower levels of protein, than were the Jersey heifers. However, this unexplainable dis- tinction between the breeds is not clearly shown in all of the summary tables (Nos. 42, 43 and 44). The lowest estimated levels of protein intake in these Missouri ex- periments were equal to or lower than the intended low protein intakes in the Armsby cooperative experiments, and on the average were about 40 per cent higher than the estimated requirements as given in Table 31 of this report. However, the actual intake of digestible crude protein in the low-protein periods of the Missouri studies probably approximated closely in the average to the estimated requirements in Table 31 in the same manner and for the same reasons as obtained in the Armsby co- operative experiments. The use of average digestion coefficients for pro- tein with rations containing considerable amounts of starch and sugar will always overestimate the content of digestible crude protein, and for large amounts of these carbohydrate materials, such as must have been used in the low-protein Missouri rations, the overestimation might well amount to 40 per cent or more. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 75 Since normal growth was never obtained on these low intakes of protein, the Missouri data afford no support for the estimates of the protein re- quirements of growing calves developed in this paper. It may be per- missible, however, to go behind the summary tables and to examine the original data and the observations upon individual calves so lavishly recorded. A great number of disturbing factors immediately become evi- dent. Besides the probable progressively increasing overestimation of the digestible protein intakes from the higher to the lower protein levels, and the changes in the source of protein, quite possibly further handicapping the low-protein rations, there are indications, mentioned by the authors themselves, that the low-protein rations were inadequate in vitamins or minerals. These indications are based not only upon a study of the rations themselves, but upon the pathological symptoms (stiffness, irreg- ular appetite, ill-defined sickness, bhndness) not infrequently observed in the calves while subsisting upon them. These periods of sickness and refusal of feed apparently have not been eliminated from the experimental data relating to the low-protein periods. The change from one ration to another was made abruptly and the data obtained during the periods of adjustment to the new ration are included in the summaries. The change from a prolonged regime on an inadequate diet that has induced a stunted and unthrifty condition, to "an adequate ration, will tend to occasion a rapid resumption of growth that, for a time at least, will exaggerate the nutritive value of the second ration. It would appear that such transitional periods should not be con- sidered in the comparative evaluation of experimental rations. Most of the calves in the Missouri experiments were bred during the period of experimental observation and calved shortly after the termina- tion of the experiment. The disturbances in body weight thus produced must have been considerable and their effects in all cases wall complicate the interpretation of the weight data secured for the last few periods. These irregularities and disturbances in the Missouri experiments may be fairly considered as detracting from the significance of the summarized data upon which the conclusions are based. A number of them will oper- ate definitely in detracting from the apparent value of the low-protein rations, and all of them will interfere with effective comparisons between rations. The conclusions apparently based so securely upon the various summaries of data, must be discounted accordingly, so that the disagree- ment between these summaries and the estimates of protein requirements given in Table 31 may be more apparent than real. Uncertainties in the use of factored requirements. — In conclusion, it should be pointed out that there are uncertainties in the method of 76 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL compositing protein requirements from the factors studied separately in this report. For example, the occurrence of pregnancy in a growing heifer may retard growth somewhat, while lactation is known to do so. Hence, the protein requirement of a pregnant lactating heifer would presumably be somewhat less than that indicated by the sum of the separate requirements. On the other hand, the protein requirements for pregnancy considered above have not included any estimates for the hypertrophy of the uterus, no basis for such estimates being available. This error, therefore, would tend to compensate for any depression of growth caused by pregnancy. The occasional occurrence of twin births in cattle (99) is another special problem not touched upon by this report. The effect of pregnancy upon the nitrogen metabolism of the mother has been the subject of a number of researches, and the results obtained have lent support to two theories. One theory pictures pregnancy as a distinct sacrifice of the mother to the perpetuation of the species. This theory is based upon a number of studies on dogs and rabbits, which have been ably reviewed by Murlin(^9^^ whose own data upon dogs are taken to support the theory. According to this theory, the total nitrogen stored by the maternal organism during pregnancy is not sufficient to form the fetus and its membranes; thus, the maternal tissues themselves are sac- rificed. As a result, during the first half of the period of pregnancy a negative nitrogen balance persists, especially during the third and fourth week in the dog ; not until the last half of pregnancy is a positive balance attained. It may be said, however, that not all of the observations upon dogs agree in indicating a negative balance as a characteristic of the first half of gestation, and that it is difficult to believe that the mere transference of protein from the maternal to the embryonic organism by whatever means before the complete establishment of the placenta should occasion such marked and persistent losses of nitrogen in the urine, particularly in view of the infinitesimal size of the embryo at this time, and of the absence of other signs of a marked disturbance of the maternal metabolism, particularly the basal metabolism. In direct opposition to the view that pregnancy is destructive of the maternal organism, is the view developed from a number of somewhat fragmentary observations on the nitrogen metaboHsm of human preg- nancy, according to which the nitrogen stored in pregnancy is in great excess of the needs of the fetus for its own growth and for the growth of its protective and nourishing membranes. A nitrogen reserve is thus built up to carry the mother over the puerperium and the initial period of lactation, removing the necessity for a sacrifice of the maternal tissues. This view has been discussed and subscribed to by Wilson (9°) and has received support from his experimental observations. PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 77 In attempting to reconcile these two views, Harding in his recent article on metabolism in pregnancy (^^) believes that the former applies to the lower animals and the latter only to the human organisms. How- ever, it appears that much more investigational work should be done be- fore this compromise position can be considered with complacence. Apparently the only evidence of this nature bearing on bovine preg- nancy has been contributed by Crowther and Woodman (^2), who deter- mined the nitrogen balance of a pregnant cow (D) and of a dry cow not in calf (C) over long periods of time. The data bearing on the question at issue may be briefly referred to. Cow D was bred on September 6, after which the collection of excreta was unfortunately suspended until November 8. On resumption of the metabolism experiment, the cow was found to be in slight negative nitrogen balance which persisted for three weeks, at which time a positive balance was established and main- tained until parturition. Before being bred, Cow D had shown through- out a uniformly higher positive nitrogen balance than Cow C, but when the experiment was thus resumed, the reverse relation was observed. The authors, therefore, conclude that the cow, like the dog and the rabbit, suffers a disturbance in its nitrogen metabolism in the first stage of pregnancy, and this conclusion is accepted by Harding in the review of the subject previously cited. However, another conclusion from this extensive study is that " even after nitrogen equilibrium is established and a relatively constant nitro- gen consumption is maintained, there may arise from time to time con- siderable deviations from equilibrium either in the positive or negative direction. It would appear therefore that for reliable work of this char- acter long experimental periods are essential." Such an apparently fortuitous disturbance in the nitrogen equilibrium of Cow C occurred about December 1 and a considerable negative nitrogen balance persisted for five weeks on an intake previously adequate for nitrogen storage. The point may therefore be raised that the similar disturbance of Cow D, to which such great import is attached, may be of this fortuitous nature and thus may bear no relation to conception or pregnancy. The investigations thus briefly reviewed evidently have an important bearing upon the problem of the protein requirement for pregnancy in cattle, since if the condition of pregnancy disturbs the nitrogen economy of the maternal organism, either by inducing excessive tissue catabolisra or a storage of nitrogen greatly in excess of the needs of the fetus, the rate of nitrogen retention by the fetus and its membranes would not represent the total nitrogen requirements of pregnancy, but would be something less than these. The subject is of great physiological interest as well as of practical importance. 78 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL Another opportunity for future research relates to the nitrogen metabo- lism of the cow after parturition throughout her puerperium. Crowther and Woodman have shown that their one subject was in a condition of negative nitrogen balance for several weeks after parturition. Might not the extent of this excess nitrogen catabolism measure the involution of the uterus (^^) and hence the nitrogen needed for the hypertrophy of this organ during pregnancy? Obviously the initiation of lactation during this period will complicate the study of this important question. Is the optimum requirement of protein different from the minimum? — The significance of " deposit protein " in animal nutrition also has a bear- ing on the method of computing protein requirements that has been advocated and exemplified in this report. When the nitrogen intake of an animal is suddenly raised, a temporary increase in the nitrogen balance may occur extending over one, two or three days; if it be suddenly de- creased, the reverse temporary disturbance in nitrogen balance occurs. The phenomenon is explained by assuming that there is a variable and labile form of protein in the body, different from that of the organized tissues, and probably retained in the cellular fluids. This circulating reservoir of protein rises and falls with the level of protein feeding. How- ever, this explanation is not based upon the results of direct experimental inquiry. It is interesting to speculate whether the amount of deposit protein in the tissue bears any relation to the well-being of the animal or to its physiological efficiency. If it does, then an animal should thrive better when nourished on amounts of protein greater than its requirements, as judged by a thorough study of all possible ways in which protein is utilized. This conception implies that there is an optimum intake of protein greater than the minimum, which in turn implies that the mere presence of an excess of protein (or amino acids) in the body fluids exerts a favorable physiological effect. The apparent existence of an optimum requirement cannot be postulated until it has been shown that all of the requirements for protein have been included in the minimum requirement with which the assumed optimum has been compared. The theory of optimum nutrition has been applied particularly to the dairy cow. Although early investigations in this country and in Europe may be cited in support of the view that an excess of protein above the requirements of maintenance and the production of milk stimulates milk secretion, even up to very high levels of protein intake, and although current practice in feeding cows on test is based upon the correctness of this assumption, more recent investigations have not afforded any con- siderable support for it. Besides establishing the fact that the protein PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 79 requirements for milk production are much lower tlian the current feed- ing standards would indicate, the experiments of Haecker(^^), Hills and associates (^*), Ellett, Holdaway, and Harris (^^), Fries, Braman, and Kriss(®^), Perkins(*), and particularly Buschmann(^) indicate that little if any increase in milk production may result from increasing the protein intake above the minimum requirements. Seven's (^') self- feeding experiments on dairy cows also offer no support — in fact quite the contrary — for the belief that nitrogenous concentrates are stimu- lants to the mammary glands/ It is probably still premature to deduce any final conclusions on this point, since there is great need for a careful definition of " minimum " and " optimum " and for crucial, carefully planned and carefully controlled investigations concerned specifically with this question. It is equally true, however, that the theory of opti- mum protein nutrition, exceeding the minimum, is difficult to explain on physiological grounds, and has suffered rather than benefited by the later experimental work in milk production. BIBLIOGRAPHY 1 Mitchell, H. H. The determination of the protein requirements of animals and of the protein values of farm feeds and rations. Bull. Nat. Res. Counc, No. 55, 44 p. (1926). 2 Armsby, H. P. The nutrition of farm animals. New York. P. 326-330 (1917). 3 Sherman, H. C. Protein requirement of maintenance in man and the nutritive efficiency of bread protein. J. Biol. Chem., 41: 97-109 (1920). 4 Perkins, A. E. Protein requirement of dairy cows. Ohio Agr. Exp. Sta. Bull., 389, 35 p. 5 Buschmann, A. Untersuchungen liber den Eiweissbedarf der Milchkuh und den Einfluss eiweissreicher und Eiweissarmer Fiitterung auf die Menge und Zusammensetzung der Milch. Landw. Vers.-Sta., 101: 1-216 (1923). ^Hindhede, M. Adequate protein minimum in dietaries. Ugeskrift for Laeger, 79: nos. 13, 14 and 15. Abstracted in Chem. Abst., 11: 2344 (1917). ''Forbes, E. B., Fries, J. A., and Kriss, M. The maintenance requirement of cattle for protein as indicated by the fasting katabolism of drj' cows. J. Dairy Sci., 9: 15-27 (1926). * Such negative evidence as that cited may, in a certain sense, be of more significance than the positive evidence obtained in earlier investigations not specifically referred to. When positive indications of a stimulating effect of excess protein are obtained in practical feeding tests, it is, in fact, difficult if not impossible to prove from the data obtained that the increase in milk production was catised by the increase in protein consumed, since other variable factors have operated. On the other hand, when large increases in protein intake above the apparent requirements fail to cause an increase in milk production, such evidence is directly contradictory to the theory of protein stimulation, except on the assumption that there are counteracting factors at work. This complicating assumption is obviously not the most rational nor the simplest interpretation of the data. 80 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 8 Carpenter, T. M. Prolonged fasting as affecting the composition of steer's urine. Proc. Nat. Acad. Sci., 11: 155-160 (1925). 9 Bull, S., and Grindley, H. S. Nitrogen metabolism of two-year-old steers. J. Agr. Res'., 18:241-255 (1919). 10 Titus, H. W. The mutual influence of the proportion of the several nutrients in feeds on their digestibihty. New Mexico Agr. Exp. Sta. Bull., 153: 52 p. (1926). "Copenhagen Laboratory. 63 de Beretning fra den Kgl. Veterinar og Land- bohojskoles Laboratorium f. landokonomiske Forsog 1907. Cited by Busch- mann. A., Landw. Vers.-Sta., 101: 16 (1923). See ref. (6). l2Honcamp, F., Koudela, St., and Miiller, E. Harnstoff als Eiweissersatz beim milchgebenden Wiederkiiuer. Biochem. Z., 143: 111-155 (1923). 13 Hart, E. B., Humphrey, G. C, and Morrison, F. B. Comparative efficiency for growth of the total nitrogen from alfalfa hay and com grain. J. Biol. Chem., 13: 133-153 (1912). l4Steenbock, H., Nelson, V. E., and Hart, E. B. Acidosis in omnivora and herbivora and its relation to protein storage. Wis. Agr. Expt. Sta. Res. Bull., 36: 19 p. (1915). 15 McColIum, E. v., and Hoagland, D. R. Studies of the endogenous metabolism of the pig as modified by various factors. I. The effects of acid and basic salts, and of free mineral acids on the endogenous nitrogen metabolism. J. Biol. Chem., 16: 299-315 (1913). 16 , and Steenbock, H. Studies of the nutrition of the pig. Wis. Agr. Expt. Sta. Res. Bull., 21 : 53-86 (1912). l^Pfeiffer, Th. Beitrage zur Frage iiber die Bestimmung der Stoffwechselprodukte im thierischen Koth. J. Landw., 33: 149-192 (1885). 18 Mitchell, H. H., and Kick, C. H. The supplementary relation between the proteins of corn and of tankage determined by metabolism experiments on swine. J. Agr. Res., 35: 857-864 (1927). 19 Morgen, A., Beger, C, and Westhauser, F. Die stickstoffhaltigen Stoffwechsel- produkte und ihre Bedeutung fiir die Bestimmung der Verdaulichkeit des . Proteins in den Futtermitteln. Landw. Vers.-Sta., 85: 1-104 (1914). 20 , , and . Weitere Untersuchungen iiber die Verwertung des Ammonacetats und des Asparagins fiir die Lebenserhaltung und Milch- bildung. Landw. Vers.-Sta., 75:265-320 (1911). 21 Voltz, W. Der Ersatz des Niihrungseiweisses durch Harnstoff beim wachsenden Wiederkiiuer. Der Futterwert des nach dem Beckmannschen Verfahren aufgeschlossenen Strohs und der Spreu. Biochem. Z., 102: 151-227 (1920). 22Scheunert, A., Klein, W., and Steuber, M. t)ber die Verwertbarkeit des Ham- stoffs als Eiweis-squelle fiir Wiederkauer, zugleich ein Beitrag zur Frage der exkretorischen Funktionen der Haut. Biochem. Z., 133: 137-191 (1922). 23Folin, O. Laws governing the chemical composition of urine. Amer. J. Physiol., 13:66-115 (1905). 24 Landergren, E. Untersuchungen iiber die Eiweissumsetzung des Menschen. Skand. Arch. Physiol., 14: 112-175 (1903). 25 Klemperer, G. Untersuchungen iiber Stoffwechsel und Emahrung in Krank- heiten. Z. klin. Med., 16: 550-605 (1889). 26 Roehl, W. Uber die Ausnutzung stickstoffhaltiger Nahnmgsmittel bei Storungen der Verdauung. Deut. Arch. klin. Med., 83: 523-557 (1905). 36 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 81 27Kocher, R. A. Uber die Grosse des Eiweisszerfalls bci Fieber und bei Arbeits- leistung. Untersuchungen mittels des Stickstoffminimums. Deut. Arch. klin. Med., 115:82-123 (1914). 28 Thomas, K. t)ber die biologische Wertigkeit der Stickstoffsubstanzen in ver- schiedenen Nahnmgsmitteln. Beitrage zur Frage nach dem phy.siologischen Stickstoff mini mum. Arch. Anat. u. Physiol., Physiol. Abt. (1909) : 219-302. 29 Graham, G., and Poulton, E. P. The influence of high temperature on protein metabolism with reference to fever. Q. J. Med., 6: 82-124 (1913). 30 Robison, R. Distribution of nitrogenous constituents of the urine on low nitrogen diets. Biochem. J., 16: 131-133 (1922). 31 Klercker, K. O. af. Beitrag zur Kenntnis des Kreatins und Kreatinins im Stoff- wechsel des Menschen. Biochem. Z., 3: 45-87 (1907). 82Siven, V. O. Ueber das Stickstoffgleichgewicht beim erwachsenen Menschen. Skand. Arch. Physiol., 10: 91-148 (1900). 33 Smith, M. The minimum endogenous nitrogen metabolism. J. Biol. Chem., 68: 15-31 (1926). 34Hindhede, M. The biological value of bread-protein. Biochem. J., 20:330-334 (1926). 35 Moulton, C. R., Trowbridge, P. F., and Haigh, L. D. Studies in animal nutri- tion. I. Changes in form and weight on different planes of nutrition. Missouri Agr. Exp. Sta. Res. Bull., 43: 111 p. (1921). — , , and . Studies in animal nutrition. II. Changes in propor- tions of carcass and offal on different planes of nutrition. Missouri Agr. Exp. Sta. Res. Bull., 54: 76 p. (1922). '"^ , , and . Studies in animal nutrition. III. Changes in chemical composition on different planes of nutrition. Missouri Agr. Exp. Sta. Res. Bull., 55:88 p. (1922). 88 Ritchie, W. S., Moulton, C. R., Trowbridge, P. F., and Haigh, L. D. Studies in animal nutrition. IV. The nitrogen, ash, and phosphorus distribution in beef flesh as affected by age and condition. Missouri Agr. Exp. Sta. Res. Bull., 59:78 p. (1923). 39 Moulton, C. R., Kempster, H. L., Hogan, A. G., and Brody, S. Normal growth of domestic animals. Missouri Agr. Exp. Sta. Res. Bull., 62: 58 p. (1923). *0 Robertson, T. B. The chemical basis of growth and senescence, Monographs of Experimental Biology. 389 p. Philadelphia and London (1923). *l Brody, S. Time relations of growth. I. Genetic growth constants of animals. J. Gen. Physiol., 8: 233-251 (1926-7). 42 Murray, J. A. Meat production. J. Agr. Sci. [England], 9: 174-181 (1919). *3 Moulton, C. R. Age and chemical development in mammals. J. Biol. Chem., 57:79-97 (1923). 4* Brody, S., and Ragsdale, A. C. A weight-height-age cune as a measure of the state of nutrition and of growth of the dairy cow. J. Dairy Sci., 5: 479-484 (1922). *5 Bell, E. T. I. On the occurrence of fat in the epithelium, cartilage, and muscle fibers of the ox. II. On the histogenesis of the adipose tissue of the ox. Amer. J. Anat., 9: 401-438 (1909). . The interstitial granules of striated muscle and their relation to nutri- tion. Internat. Monatschr. Anat. u. Physiol., 28:297-347 (1911). — . The interstitial granules (liposomes) in fatty metamorphosis of striated muscle. J. Pathol, and Bact., 17: 147-159 (1912-3). 46 82 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 47Armsby, H. P. Feeding for meat production. U. S. Dept. Agr., Bur. Anim. Ind. Bull., 108, 89 p. See also Armsby, H. P., Nutrition of farm animals, New York (1917), p. 375-378. 48 Brody, S., Sparrow, C. D., and Kibber, H. H. Time relations of growth. II. The equivalence of age in mammals estimated on the basis of their growth constants. J. Gen. Physiol., 9: 285-308 (1926). 49 Haecker, T. L. Investigations in beef production. Minn. Agr. Exp. Sta. Bull., 193: 110 p. (1920). SOHaigh, L. D., Moulton, C. R., and Trowbridge, P. F. Composition of the bovine at birth. Missouri Agr. Exp. Sta. Res. Bull., 38: 47 p. (1920). 51 Eckles, C. H. The nutrients required to develop the bovine fetus. Missouri Agr. Exp. Sta. Res. Bull., 26: 36 p. (1916). 62 Murray, H. A., Jr. Physiological ontogeny. A. Chicken embryos. III. Weight and growth rate as functions of age. J. Gen. Physiol., 9: 39-48 (1925). 53 MacDowell, E. C, and Allen, E. Weight of mouse embryos 10-18 days after conception, a logarithmic function of embryo age. Soc. Exptl. Biol, and Med. Proc, 24: 672-74 (1927). 54 Eckles, C. H. Dairy cattle and milk production. 591 p.. New York, p. 237 (1923). 55 Popov, I. S. The food requirement of pregnant animals [title translated from the Russian]. Nauch. Agron. Zhur., 1:205-210. Abstracted in Exp. Sta. Rec. (1926), 54:562 (1924). 56 Haecker, T. L. Investigations in milk production. Minn. Agr. Exp. Sta. Bull., 140:7-79 (1914). 57 Gaines, W. L. Relative rates of secretion of various milk constituents. J. Dairy Sci., 8:486-496 (1925). 58 Overman, O. R., and Sanmann, F. P. The energy value of milk as related to composition. Formulas for the computation of the energy. 111. Agr. Exp. Sta. Bull., 282: 207-218 (1926). 59 Timpe, H. Gesetzmassigkeiten in der Zusammensetzung der Kuhmilch und der Nachweis von Verfalschungen derselben. Chem. Zeitg., 23: 1040-1043 (1899). 60 Van Slyke, L. L. Conditions affecting the proportions of fat and proteins in cow's milk. J. Amer. Chem. Soc, 30: 1166-1186 (1908). 61 Andersen, A. C, and Langmack, P. V. F. P. Unders0geleser over den danske Komaelks gennemswitlige Sammensaetning — 113 Beret. Forsogs, Koben- havn. (1923). Cited by Gaines(57). 68 Hill, A. V. Muscular activity. 115 p., Baltimore (1926). 63Tiegs, O. W. The function of creatine in muscular contraction. Australian J. Expt. Biol, and Med. Sci., 2: 1-19 (1925). Fiske, C. H., and Subbarow, Y. The nature of the " inorganic phosphate " in voluntary muscle. Science, n. s., 65: 401-403 (1927). 64 Scott, E. L. The present status of our knowledge of fatigue products. Pub. Health Repts. [U. S.], 33: 605-611 (1918). 65 Macleod, J. J. R. Zur Kenntniss des Phosphors im Muskel. Z. physiol. Chem., 28:535-558 (1899). Burian, R. Die Herkunft der endogenen Hampurine bei Mensch und Sliugetier. Z. physiol. Chem., 43: 532-546 (1905). PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 83 66 Brown, T. G., and Cathcart, E. P. The effect of work on the creatine content of muscle. Biochcm. J., 4: 420-426 (1909). Pokelharing, C. A., and van Hoogcnhuyze, C. J. C. Die Bildung de.s Kreatins im Muskel beim Tonu.s und bei der Starre. Z. physiol. Chem., 64:262-293 (1910). Uyeno, K., and Mitsuda, T. Creatine formation during tonic muscle contrac- tion. J.Physiol. [England], 57:313-317 (1923). Schlossmann, H. Uber das Verhalten des Kreatingehaltes des Froschmuskels bei der Arbeit. I. Z. physiol. Chem., 139: 87-94 (1924). . iJber den Kreatingehalt des Froschmuskels bei Reizung. Arch, exptl. Path. Pharmakol., (14): 105 (1925). 67 Cathcart, E. P. The physiology of protein metabolism. London, p. 134 (1921). 68 Tigerstedt, R. Handbuch der Physiologie des Menschen, by W. Nagel. Bruns- wick, 1:441-451 (1909). 69 Magnus-Levy, A. Metabolism and practical medicine, vol. I, by C. von Noorden. Amer. translation, Chicago, p. 352-355 (1907). 70Lusk, G. The science of nutrition, 3d. ed. Philadelphia, p. 316 (1917). 71 Cathcart, E. P. The influence of muscle work on protein metabolism. Physiol. Rev., 5: 225-243 (1925). 72 Atwater, A. O., and Sherman, H. C. The effect of severe and prolonged muscular work on food consumption, digestion, and metabolism. U. S. Dept. Agr., Office Exp. Sta. Bull., 98: 67 p. (1901). 73 Shaffer, P. A. Diminished muscular activity and protein metabolism. Amer. J. Physiol., 22: 445-455 (1908). 74 Cathcart, E. P., and Burnett, W. A. The influence of muscle work on metabolism in varying conditions of diet. Proc. Roy. Soc. [London], Series B, 99: 405- 426 (1926). 75 Wait, C. E. Experiments on the effect of muscular work upon the digestibility of food and the metabolism of nitrogen. U. S. Dept. Agr., Office Exp. Sta. Bull., 117:43 p. (1902). 76 Thomas, K. Physiological nitrogen minimum [trans.]. Arch. Anat. u. Physiol., Physiol. Abt., 1910 Suppl. Bd.: 249-285 (1910). 77 Van Hoogenhuyze, C. J. C, and Verploegh, H. Beobachtungen iiber die Kre- atininausscheidung beim Menschen. Z. physiol. Chem., 46:415-471 (1906). 78 Schuiz, W. Die Verlauf der Kreatininausscheidung im Ham des Menschen mit besonderer Beriicksichtigung des Einfiusses der Muskelarbeit. Pfiiiger's Arch. Physiol., 186: 126-172 (1921). 79Hartmann, C. tJber den Einfluss der Muskelarbeit auf die Harnstiure und Phosphorsiiureausscheidung. Pfliiger's Arch. Physiol., 204: 613-628 (1924). 80 Wilson, D. W., Long, W. L., Thompson, H. C, and Thurlow, S. Changes in the composition of the urine after muscular exercise. Soc. Exp. Biol, and Med., Proc, 21:425, 426 (1924). Dobreff, M. Experimenteller Bcitrag iiber den Einfluss von Affckten und Muskelarbeit auf die Urinausscheidung. Pfliiger's Arch. Physiol., 213:511- 534 (1926). Asher, L., and Weber, A. Beitriige zur Physiologic der Driisen. Nr. 95. Uber die Einwirkung kurzcr, intensiver Muskelarbeit auf die Harnabsonderung. Biochem. Z., 173: 69-94 (1926). 81 Mitchell, H. H., and Kruger, J. H. The effect of muscular work upon the endogenous catabolism of the tissues. J. Biol. Chem., 76: 55-74 (1928). 84 PROTEIN REQUIREMENTS OF CATTLE: MITCHELL 82 Myers, V. C, and Volovic, G. O. The influence of fever on the elimination of creatinine. J. Biol. Chem., 14: 489-508 (1913). Shaffer, P. A. The excretion of kreatinin and kreatin in health and disease. Amer. J. Physiol., 23: 1-22 (1908). McClure, C. W. The excretion of creatine, creatinine and uric acid in acute febrile conditions. Arch. Intern. Med., 22: 719-749 (1918). 83 Linser, P., and Schmid, J. tjber den Stoffwechsel bei Hyperthermie. Deut. Arch. klin. Med., 79: 514-538 (1904). 8* Lee, O. P., and Tashiro, S. Alkaligenesis in tissues. II. Ammonia production in muscle during contraction. Amer. J. Physiol., 61: 244-253 (1922). 85 Brunquist, E. H., Schneller, E. J., and Loevenhart, A. S. The effects of anoxemia on nitrogen metabolism. J. Biol. Chem., 62:93-115 (1924). 86 Mitchell, H. H. The nutritive value of the proteins. Physiol. Rev., 4: 424-478 (1924). 87 Cooperative experiments upon the protein requirements for the growth of calves. Bull. Nat. Res. Counc, no. 12:219-288 (1921). ' Ibid., no. 42: 1-44 (1924). 88Swett, W. W., Eckles, C. H., and Ragsdale, A. C. The minimum protein re- quirement for growing dairj'^ heifers. Missouri Agr. Exp. Sta. Res. BuU., 66: 155 p. (1924). 89 Murlin, J. R. Metabolism of development. II. Nitrogen balance during preg- nancy and menstruation of the dog. Amer. J. Physiol., 27: 177-205 (1910). 90 Wilson, K. M. Nitrogen metabolism during pregnancy. Bull. Johns Hopkins Hospital, 27: 121-129 (1916). 91 Harding, V. J. Metabolism in pregnancy. Physiol. Rev., 5: 279-302 (1925). 92 Crowther, C, and Woodman, H. E. A study of nitrogen metabolism in the dairy cow. J. Agr. Sci. [England], 12: 40-56 (1922). 93Slemons, J. M. The involution of the uterus and its effect upon the nitrogen output of the urine. Bull. Johns Hopkins Hospital, 25: 195-200 (1914). 94 Hills, J. L., Beach, C. L., Borland, A. A., Washburn, R. M., Story, G. F. E., and Jones, C. H. The protein requirements of dairy cows. Vermont Agr. Exp. Sta. Bull., 225: 119 p. (1922). 95 Ellett, W. B., Holdaway, C. W., and Harris, W. G. Feeding standards for milk production. Virginia Agr. Exp. Sta. Techn. Bull., 23: 52 p. (1921). 96 Fries, J. A., Braman, W. W., and Kriss, M. On the protein requirement of milk production. J. Dairy Sci., 7: 11-23 (1924). 97 Nevens, W. B. Experiments in the self-feeding of dairy cows. 111. Agr. Exp. Sta. Bull., 289: 26 p. (1927). 98 Deuel, H. J., Jr., Sandiford, I., Sandiford, K., and Boothby, W. M. A study of the nitrogen minimum. The effect of sixty-three days of a protein-free diet on the nitrogen partition products in the urine and on the heat production. J. Biol. Chem., 76: 391-406 (1928). 99 Cole, L. J., and Rodolfo, A. Seasonal distribution of twin births in cattle. Proc. Amer. Soc. Animal Prod., (1924): p. 116-118. The National Research Council Membership and Organization. — The National Research Council is a cooperative organization of scientific men of America, including also a representation of men of aflFairs interested in engineering and indus- try and in the " pure " science upon which the applied science used in these activities depends. Its membership is largely composed of accredited representatives of about seventy-five national scientific and technical societies. Its essential purpose is the promotion of scientific research and of the application and dissemination of scientific knowledge for the benefit of the national strength and well-being. The Council was established at the request of the President of the United States, by the National Academy of Sciences, under its Congres- sional charter to coordinate the research facilities of the country for work on war problems involving scientific knowledge. In 191 8, by Executive Order, it was reorganized as a permanent body. Research Fellowships The Council maintains, with the financial assistance of the Rockefeller Foundation and General Education Board, three series of post-doctorate research fellowships. These fellowships are primarily intended to aid young investigators. Fellowships in Physics, Chemistry and Mathematics. — Candidates must already have the doctor's degree or equivalent quahfications and must have demonstrated a high order of ability in research. Address applica- tions to Secretary, Research Fellowship Board in Physics, Chemistry and Mathematics, National Research Council, Washington, D. C. Fellowships in Medical Sciences. — Both graduates in medicine and doctors of philosophy in one of the sciences of medicine, or in physics, chemistry, or biology, are eHgible for these fellowships. Address applica- tions to Secretary, Medical Fellowship Board, National Research Council. Washington, D. C. Fellowships in the Biological Sciences (including Botany, Zoology, Anthropology, Psychology, Agriculture, and Forestry). — Candidates must already have the doctor's degree or equivalent qualifications and must have demonstrated a high order of ability in research. Address applica- tions to Secretary, Board of Fellowships in the Biological Sciences. Na- tional Research Council, Washington, D. C. )