FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. BY THOMAS B. OSBORNE and LAFAYETTE B. MENDEL, With the Co-operation of EDNA L. FERRY. (From the Laboratories of the Connecticut Agricultural Experiment Station and the Sheffield Laboratory of Physiological Chemistry of Yale University.) WASHINGTON, D. C. Published by the Carnegie Institution of Washington 1911 CARNEGIE INSTITUTION OF WASHINGTON Publication No. 156 3 i4-L 0 PRESS OF GIBSON BROS. WASHINGTON, D. C. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Although the proteins have long had attention centered upon them because of their commanding position in relation to nutrition, it is only in very recent years that the progress of chemical investi- gation, fostered by the introduction of new methods of research, has begun to make our conception of these fundamentally important substances more exact. The role of the nitrogenous substances in both plants and animals has gradually been brought to light by a series of effective researches; but even to-day it is still common, despite the newer knowledge in the field of biochemistry, to read of the part played by proteins, fats, and carbohydrates in nutrition, as if these groups of compounds were each chemically homogeneous and, within wide limits, physiologically interchangeable. Not only have these food-stuffs been discussed in the past without more than the scantiest consideration of possible specific differentiations within the individual groups, but in nearly every investigation on nutrition the nutrients have been employed in those complex and often little- understood mixtures which make up the common foods. That any such indefinite combination of known and unknown organic compounds is an unsatisfactory and unideal starting-point in any adequate study of the laws of nutrition was appreciated by Carl Voit, the foremost student of this subject of the past generation. In discussing the necessity for an accurate knowledge of the food intake in the study of metabolism he says: Zu dem Zwecke ware es unstreitig am besten, konnte man nur reine, chemische Verbindungen (die reinen NahrstofYe) z.B. reines Eiweiss, Fett, Zucker, Starkemehl, Aschebestandtheile, oder Gemische derselben geben. Da aber die Menschen und auch die Thiere nur selten solche geschmacklose Gemenge auf die Dauer aufzunehmen oder zu ertragen vermogen, so bleibt fiir die meisten Falle niehts anderes iibrig als schon durch die Natur zusam- mengesetzte Mischungen (die Nahrungsmittel) zu wahlen. Jedoeh ware es wohl moglieh und ganz verdienstvoll, die Grundversuche, nachdem vorher der Weg mit Hilfe der letzteren Mischungen gefunden worden ist, mit den reinen Stoffen zu wiederholen, obwohl sich dabei sicherlich im Wesentlichen keine anderen Resultate ergeben werden.* Note. — The expenses of this investigation were shared by the Connecticut Agri- cultural Experiment Station and the Carnegie Institution of Washington, D. C. In these experiments we have been further assisted by M. S. Fine and C. S. Leavenworth. * C. Voit: Hermann's Handbueh der Physiologie, 1881, vi, (1), p. 19. 2 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Judging on the basis of such information as was available at the time when the preceding was written, Voit further stated in regard to the proteins : Die verschiedene Modificationen des Eiweisses haben hochst wahr- seheinlieh sammtlich ganz die gleiehe Wirkung auf den Stoffumsatz, jedoch liegen hieriiber noeh keine genauen Untersuehungen vor.* Certain exceptions to the assumed physiological equivalence of the proteins of various origins have long been known, conspicu- ously in the case of gelatin. This nitrogenous compound, possessing many of the most characteristic chemical features of a typical pro- tein, was found to be inadequate as the sole source of nitrogen to the higher animals. Its lack of a constituent tyrosine complex and of sulphur was appreciated; and to this was often charged the in- sufficiency of its nutritive functions. It has remained for the newer researches on the proteins — those interesting studies of the past ten years which have unraveled so much of the structural complexity of the albuminous molecule — to raise new questions and place the problems of nutrition in a new light. It has been the structural dis- similarity rather than the likeness of the proteins which has aroused physiological comment. The rich fund which these investigations of the constituent amino-acid derivatives of the various proteins of both animal and plant origin have disclosed need not be presented here.f The more striking findings are rapidly becoming a matter of common knowledge in physiological circles. Thus it is appreciated that the zein of maize and the gliadin of wheat show distinctive features in respect to the yield of the different amino-acids obtain- able from them. Zein is distinctly deficient in comparison with most other proteins: it yields no tryptophane, glycocoll, or lysine. Glia- din is characteristically rich in the glutaminic acid complex. Other equally interesting illustrations might be cited. Are these different proteins, specific in respect to origin, biolog- ically differentiated and individually unique in chemical make-up, of equal value for the purposes of nutrition? If they are not, how are we to explain the uniformity, apparent or real, which is exhibited in the composition of the individuals of any species living under widely divergent dietetic conditions ? The flesh-eater and his herbiv- orous neighbor accumulate specific muscle-proteins, blood-proteins, and brain-proteins despite recognized differences in the nitrogenous intake. The problem here presented has been both simplified and com- plicated by the trend of recent studies in the chemistry of digestion. * C. Voit: Hermann's Handbuch der Physiologie, 1881, vi (i), p. 104. t Elaborate reviews of this subject will be found in the monographs by Plimmer, Sehryver, and T. B. Osborne, in the series of Monographs on Biochemistry, Longmans, Green & Co.; also Die Pflanzenproteine, Ergebnisse der Physiologie, 1910, x, pp. 47-215. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 3 So ' mg as it was still assumed that the proteins are only slightly modified within the alimentary tract, it was not easy to appreciate how these widely differing amino-acid complexes could be converted into a common protein or group of proteins. But with the introduc- tion of the evidence that proteins experience a profound cleavage prior to absorption — that the organism is equipped with an outfit of enzymes easily capable of effecting such intense hydrolyses, and that the proteins in all probability normally disappear from the alimentary tract as amino-acid fragments and relatively simple polypeptides— our conception of protein assimilation has changed notably. It is, perhaps, too early yet to determine to what extent the theories of protein metabolism and reconstruction, which Loewi and Abderhalden in particular have championed, accord with the facts of experience. Whether the organism synthesizes blood and tissue proteins from the amino-acid rests of digestion, and thus only; whether there are no rearrangements whereby one amino-acid may give rise to another; and whether under these circumstances the synthetic power of the organism is limited to a choice by the mini- mum of all essential protein building complexes — the so-called " Bau- steine" — can not yet be profitably debated in detail. The problem of protein synthesis is further complicated by a consideration of the activities of bacteria in relation to the food-prod- ucts which enter the alimentary tract. It is now well appreciated that microorganisms grow and die in large numbers throughout the lower parts of the digestive tube, so that it is not uncommon to find the faeces to be composed in a very considerable degree of the bodies of dead bacteria. It is not unlikely that the organisms which thus perish in the intestinal canal are subject to autolytic or other diges- tive degradation by which their protoplasmic constituents may become available as food sources to the organism of the host. Bear- ing in mind the synthetic possibilities inherent in plant cells, such as bacteria represent, it is by no means beyond the bounds of reason- able interpretation to assume that the amino-acids first formed by digestion of food proteins may experience a synthesis into new forms of protein complexes prior to a subsequent digestion and utilization. Viewed in this light, the immediate hydrolysis products of our food- stuffs may become available only after they have been in greater or less part reconstructed by these preeminently synthetic symbiotic bacteria into products of more uniform character, possibly widely different from the original intake. Nucleoprotein synthesis, for example, may thus become referable to bacterial intervention. No one can say at the present time to what extent, if at all, such synthetic possibilities enter into the problems here discussed. The subject has recently again been referred to by Luthje.* The *I„uthje: Ergebnisse der Physiologie, 1908, vn, p. 826. 4 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. important point for us is that we must deal with such possibilities ; we must plan and interpret our nutrition experiments to-day in the light of these newer ideas. If we accept the synthetic theory of metabolism, disregarding the cooperation of bacteria, it becomes possible to understand how the lack of some fundamental unit, like tyrosine, tryptophane, lysine, or cystine may lead to malnutrition or death through the inability of the organism to construct normal protein and normal protoplasm, because of a deficit of essential structural complexes. The errors of a one-sided diet present a new point of view. The im- portance of these theories is enormous, not alone for the nutrition of man but also for the welfare of our domestic animals. Rational and economical feeding is based upon a correct interpretation. Considerations such as the foregoing certainly justify an exten- sion of experimental work in nutrition along the lines suggested by the rapidly accumulating data on the structure of the individual pro- teins. The theories must be subjected to the rigorous test of experi- ment. The relative and absolute value of individual proteins must be determined by physiological trials as a preliminary to definite and more permanent, rational dietary programs. We have attempted the beginnings of such a study. At the outset we were confronted by the fact that there are on record few, if any, successful experiments in feeding isolated food- stuffs. Still fewer are those in which the protein compound was fed isolated, in a reasonable degree of purity. By "successful" we mean experiments in which the health and vigor of animals were main- tained under adequately controlled conditions for sufficient periods of time to permit of more than tentative conclusions. The total number of feeding trials in which the role of the protein food is in some way concerned is very large. We shall omit reference to all except those which bear specifically on the problems under consideration, viz., the significance of individual proteins in nutrition. It may not be amiss at the outset to point out the broader re- quirements which any adequate dietetic regime involves and upon which its nutritive success depends in good measure. In the first place the nutrients must be presented in a form that is digestible and thus available for physiological utilization. The physical texture as well as digestibility per se plays an important part in this respect. Again, the available parts of the diet must be adequate in amount to cover the calorific needs of the organism to which it is supplied, i. e., there must be sufficient metabolizable energy. One might be inclined to omit reference to such apparently obvious facts had they not been serious factors in previous experimental failures in feeding isolated food-substances. In recent years much emphasis has further been laid upon certain less evident considerations involving nutrition more FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 5 indirectly. Thus the question of the inorganic constituents of the diet is far from settled, and this aspect of dietetics can scarcely be said to rest upon any very rational basis. We realize the need of chlorine and phosphorus, of calcium and iron in certain vital functions; but the physiological requirement of other elements is none the less definite because it is not fully understood. On one point, namely, the power of the animal organism to employ the elements phospho- rus and iron in the form of either organic or inorganic compounds, we have assumed that the consensus of opinion and the preponderance of experimental evidence are in favor of either possibility. The debate on this topic need not be reviewed here. Closely related to these nutritive requirements is the subject of those food accessories which determine in large measure what is spoken of as the palatability of any ration. They act in manifold indirect ways to influence the digestion of the nutrients by their effect in promoting or retarding secretion into the alimentary tract ; they affect the appetite and the psychic element in digestion — all of which have received attention anew through Pawlow's efforts. After all one is as yet not justified in insisting that these incidental features of the dietary are absolutely indispensable. Quoting Tangl : Eine fordernde Wirkung der sog. Reiz — oder Wiirzstoffe — sagt Kell- ner — auf die Ausnutzung des Flitters ist bis jetzt bei keiner einzigen der auf diesen Punkt geriehteten Untersuchungen beobaehtet worden. Dasselbe gilt nicht nur fur die landwirtschaftliehen Nutztiere, sondern auch fiir den Menschen.* The undoubted biological importance of that rather vaguely defined and hetereogeneous group of compounds known as lipoids has raised the question as to whether they are at all necessary in any complete diet, or whether they can be synthesized by animals as they are by plants. The lipoids (phosphatides and cholesterols) are present as common cell derivatives in every familiar food and can only be excluded by special, laborious methods of extraction. It is not strange, therefore, that there is a paucity of evidence relating to their absolute significance as constituents of the food-intake. Recent experiments by Steppf on the indispensability of the ether-soluble constituents of the food for the life of mice are far from conclusive. Lastly, it seems worth while to point out that the nutritive conditions which pertain during the period of growth are in many respects — perhaps far more than is realized — different from those existing at a later period. During the active constructive phase new material must be supplied which differs both in quality and amount from the quota furnished as a maintenance ration. We realize well enough that the growing skeleton requires calcium; but to what *Tangl: Oppenheimer's Handbuch der Biochemie, 1910, in (2), p. 55. jStepp: Biochemische Zeitschrift, 1909, xxn, p. 452. 6 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. degree are the further nutritive demands modified by developmental changes? Are the same proteins which suffice the adult capable of supplying the essential building-stones to the young? Here again a wealth of new problems still confronts us. There are no recorded experiments on the larger animals, such as are commonly used in laboratory work, which furnish any con- clusive data on the broader questions involved in this research. The careful experiments of Rohmann and his pupils* on dogs fed with isolated proteins (edestin, vitellin, casein, myosin) extended over 50 days as a maximum period in one case. Such investigations, as well as those of Abderhalden and his collaborators,! have yielded much of interest in regard to the utilization of the food-stuffs inves- tigated and have given valuable hints for future work in nutrition. But the periods of observation have been far too brief to permit any permanent positive conclusions regarding the adequacy of the proteins fed. Minor deficiencies may fail to become conspicuous in even comparatively long periods in the case of animals whose size and span of life indicate a considerable store of reserve material. For various reasons the most successful investigations in this field have been conducted on much smaller animals, especially rats and mice. The preparation of suitable pure food supplies on a scale sufficient for long periods and in economically procurable amounts is thereby rendered possible. The necessary scientific measurements and analyses can be conducted on a scale impossible for larger ani- mals, and the problem of care and attention is equally simplified. Experiments can be duplicated without great effort and individual peculiarities eliminated by force of numbers. Furthermore, the various stages of growth and maturity are completed in the smaller animals within relatively short periods of time, so that the permanent effects of any dietary become apparent within a range of days or months that is not outside of ordinary experimental possibilities of observation. As illustrations of some of these features it may be recalled that the ultimate effects of complete inanition are apparent in four or five days in rats or mice ; in dogs the fatal outcome may be delayed for many weeks. The duration of life in the white rat is about three years; about 280 days suffice to complete the entire period of growth to maturity. To what extent, if any, the rigorous conditions, such as restraint in cages, etc., set by laboratory requirements may modify the normal physiological functions of the small animals is not apparent from such records as we have found. It should be noted that both rats *Marcuse: Archiv f. d. ges. Physiologie 1896, lxiv; 1897, lxvii, p. 373. Steinitz: Ibid, 1898, lxxit, p. 75. R. Deipziger: Inaugural Dissertation, Breslau, 1899. E. Ehrlich: Inaugural Dissertation, Breslau, 1900. tCf. papers by Abderhalden and collaborators in numerous recent volumes of the Zeitschrift f. physiol. Chemie. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 7 and mice are reported to have been kept in health on milk alone during periods of six months or more.* So far as we are aware no experiments in which "artificial" food mixtures were introduced have ever been carried on successfully for an equal duration. The most elaborate published nutrition experiments in respect to the variety of important details taken into consideration are those of Henriquesf on white rats. He determined the changes in weight and the food-intake, as well as the complete daily nitrogen balance, in a large number of animals fed on artificially mixed diets. The isolated proteins used were casein, zein, and gliadin, with additions of sugar, lard, cellulose, and inorganic salts. J Only with casein and gliadin were the nitrogen balances favorable for the few days during which each trial was continued. This was true despite the gradual loss of body-weight noted in each case. It serves to emphasize what is frequently overlooked, viz., that a favorable nitrogen balance over a short period of time is in no sense an adequate index to a satisfac- tory nutritive condition. We have found that the body- weight of the rat is a more reliable guide to the real nutritive equilibrium of the animals than is the nitrogen balance. jj Valuable as the experiments of Henriques and Hansen have been for the purposes of orientation in a new field of research, the failures can not be adduced as proof of the inadequacy of the pro- teins selected, nor can the apparently successful results be accepted as conclusive. For the latter point, the duration of the experiments is far too short, as will be seen from the work of subsequent experi- menters. The most serious criticism, perhaps, pertains to the food intake, which must have been within the lower limits of a maintenance ration for the animals which in many cases had not yet completed their growth. In respect to one conclusion the words of Henriques may be quoted : Wir linden also, dass es eine Wesensverschiedenheit zwischen der Bedeutung des Zeins und der des Gliadins fur den Stiekstoffumsatz im Korper gibt; das Zein vermag kein Stiekstoffgleiehgewicht herzustellen, was dagegen das Gliadin vermag wenn es in reiehlieher Menge gegeben wird. Der Grund dieser Verschiedenheit lasst sieh natiirlieh nicht mit *Falta and Noeggerath: Hofmeister's Beitrage zur chemischen Physiologie, 1905, vii, p. 320 (rats). Knapp: Zeitschrift fur experimentelle Pathologie, 1908, v, p. 150, says: "Bekanntlich konnen z.B. Mause mit Milch monatelang ernahrt werden und befinden sich vvohl dabei." Socin: Zeitschrift fur physiologische Chemie, 1891, xv, p. 93, fed mice 99 days on egg yolk alone. fHenriques and Hansen: Zeitschrift fur physiologische Chemie, 1904-5, xuii, p. 417; 1908, liv, p. 169; Henriques: Ibid., 1909. lx, p. 105. tit should be noted that these included "Knochenmehl" (p. 419-420), which is not entirely free from nitrogenous matter. Was bone-ash intended? ||It may be worth while to point out here that we have found very considerable losses of nitrogen, especially in hot weather, when urine is collected by the method adopted by Henriques and by us, which will be described later. Such losses, amounting to 10 per cent and over, would obviously lead to the assumption of a greater retention of nitrogen than actually occurred and a consequent incorrect nitrogen balance. 8 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Sicherheit angeben. Wir haben bereits angeflihrt, class es beiden genannten Proteinen an Lysin abgeht. Dieser Mangel scheint also nicht von ent- schiedener Bedeutung zu sein. Dagegen scheint der Umstand, dass das Zein kein Tryptophan enthalt, eine grosse Rolle zu spielen. Ob das Nicht- vorhandensein von Tryptophan, in Zein wirklich der Grund ist, weshalb bei Fiitterung dieses Proteins kein Stickstoffgleichgewicht eintritt, muss sich iibrigens durch eine Untersuchung entscheiden lassen, ob das Stick- stoffgleichgewicht sich herstellen lasst, wenn das Futter Zein -f- Tryptophan enthalt.* The preceding quotation has suggested that the deficiency in a protein dietary need not be one of quantity ; the structural character of the nitrogenous intake may determine its adequacy. Willcock and Hopkinst have approached the problem from this view-point. They fed mice on zein together with non-nitrogenous foods and compared their length of life with that of mice which received in addition some of the missing fragments of this "imperfect" protein, viz., tryptophane (and tyrosine for comparison). Zein was shown to be quite unable to take the place of a normal protein, like casein, in maintaining growth: The addition of the missing tryptophane group has no power to convert such loss (of weight) into equilibrium or gain — a fact possibly due to other deficiencies in the zein molecule, such as the absence of lysine. On the other hand, on the average, the loss of weight was slower with tryptophane than without it. But this result might well be expected, even if the tryptophane administered undergoes utilization without directly contributing to tissue formation or structural maintenance. If it serves as a basis for the elabo- ration of a substance absolutely necessary to life — something, for instance, of an importance equal to that of adrenaline — then, in starvation, or when it is absent from the diet, a supply is likely to be maintained from the tissue- proteids; the demand for it would become one of the factors determining tissue breakdown. In the case of young animals which directly benefit from the addition of a protein constituent, otherwise absent from their diet, to the extent of a well-nigh doubled life and marked improvement in gen- eral condition, but at the same time steadily lose, instead of gaining, weight, the utilization of the constituent would appear to be of some direct and specific nature, (p. 101.) The suggestion of a possibility of the direct formation of essen- tial hormones from amino- acid derivatives of proteins is timely. One can not draw any further conclusions regarding the value of the proteins (zein and casein) fed by Willcock and Hopkins, because in the absence of definite intake records, the question of a comparable and adequate supply of energy in the various cases remains undeter- mined. Special experiments showed that the prepared zein was "in no sense actively deleterious." The same uncertainty regarding the real participation of in- anition factors applies to the earlier widely quoted experiments of *Henriques: Zeitschrift fur physiologische Chemie, 1909, lx, p. 117. tWillcock and Hopkins: Journal of Physiology, 1906-7, xxxv, p. 88. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 9 Lunin* on mice. Some of the animals were fed on casein as the sole protein. They survived only a month or less on the "artificial" diet, the author attributing the decline to defects in the intake of inorganic salts and other accessory foods. Abderhalden and Ronaj fed mice with casein, sugar, and sodium carbonate. The animals lived only 8 to 24 days on this obviously inadequate diet. Falta and NoeggerathJ recorded the changes of weight in rats kept on dietaries containing the nitrogen in the form of isolated (commercial) proteins: ovalbumin, casein, serum albumin, serum globulin, fibrin, and haemoglobin. They fed no "roughage" in the form of cellulose. Neither these individual proteins nor mixtures of all were adequate to keep the animals alive longer than 94 days in the most favorable case. In most cases a gradual, steady decline was noted throughout the progress of the experiment. Death occurred when the animals had been reduced to two-thirds or three-fifths of their initial weight. The experiments clearly demonstrate the neces- sity of prolonged observation ; for rats were maintained on the casein food mixture during several weeks without loss of weight, just as in Henriques' briefer trials of three to four weeks. Only later did the untoward effects manifest themselves. On milk, milk powder, or lean horse-meat rats could be kept six months or more. The authors are properly guarded in their summary. They say : Ob in der Periode des Gewichtabfalles in unseren Versuchen die Tierc gentigend Nahrung aufgenommen haben, um ihr Kalorienbediirfnis zu decken, konnen wir nicht sicher angeben. Der Befund von Nahrungsresten im Digestionstractus der toten Tiere (Lunin) ist unseres Eraehtens hierfiir kein. zwingender Beweis. Hier miissten genaue Stoffweehselversuche mit Berticksiehtigung der Kraftbilanz einsetzen. Erst wenn der Einwand ungeniigender Nahrungsaufnahme oder ungeniigender Ausniitzung, fur welehe beiden Momente man vielleieht die Einformigkeit der Kost und den Mangel an Gewiirzen verantwortlieh maehen konnte, beseitigt ist waren andere Griinde zu erortern, wie z.B. der Mangel der notigen chemischen Bausteine oder ein abweiehendes ehemisches Gefiige der eingefiihrten NahrstofTe. (p. 322.) Utilizing the experience gained in the preceding research, Knapp|| succeeded no better in maintaining his animals. He estimates the calorific needs of a 200 to 250 gram rat at 50 to 60 calories. The curves of change of body-weight and caloric intake go more or less parallel in Knapp's experiments and he notes that the specific role of individual nutrients can not be suitably ascertained until animals can be induced in some way to eat the requisite amount of calories *Lunin: Zeitschrift fur physiologische Chemie, 1881, v, p. 31. fAbderhalden and Rona: Zeitschrift fur physiologische Chemie, 1904, xlii, p. 528. JFalta and Noeggerath: Hofmeister's Beitrage zur chemischen Physiologie,. 1905, vii, p. 314. ilKnapp: Zeitschrift fiir experimentelle Pathologie, 1908, v, p. 147. IO FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. in the form of the "artificial" mixtures. The danger of drawing erroneous conclusions is attested by the fact that he was unable to maintain his rats upon a variety of food articles : dog biscuit, graham bread, rice, etc., which one would assume to be adequate as mixed food. In explanation of his failures he says: Der Grund liegt hauptsaehlieh darin, dass die Thiere bei der reizlosen einformigen Kost den Appetit verlieren, im geringeren Grade wohl aueh darin, dass die Nahrung mit zunehmender Appetitlosigkeit im Darm auch weniger gut ausgeniitzt wird. (p. 168.) It might have been expected that the difficulties here encoun- tered could be obviated by the use of animals in which forced feeding could be satisfactorily instituted. C. Voit* kept a pigeon alive 124 days with peas by this method. Jacob! failed in repeated trials with pure food mixtures, and attributes his lack of success to the impos- sibility of adapting the physical texture of the introduced pure-food pellets to the requirements of the avian digestion apparatus, so that the pigeons died of inanition. He also attempted to advance beyond his predecessors in feeding "artificial" mixtures (with casein as the sole protein) to rats. He was scarcely more successful; the animals ate sparingly and he concludes that if it is possible to keep an animal alive 124 days, as he did, on a diet of the character noted, this must contain all the substances essential to life. The animals did not exhibit any gross pathological defects at autopsy, but all visible fat had disappeared. The next attempt, Jacob optimistically sug- gests, must be directed toward devising combinations which the ani- mals will eat: So gelingt es vielleicht doeh, eine Nahrung aus reinen Nahrstoffen ohne Genussmittel herzustellen, welche alle zur dauerndenErhaltungeinesTieres notigen Stoffe im richtigen Mengenverhaltnis enthalt. (p. 60.) McCollumJ fed both young and full-grown rats on complex artificial mixtures, in which edestin, zein, and sometimes casein, were the sole sources of nitrogen. They are the most successful experiments yet reported as regards maintenance of body-weight or growth on a restricted quality of protein intake. The chief difficulty encountered was that of anorexia, which the author attempted to overcome by frequent changes in the combinations of food-stuffs used and by addition of flavors. Some of the trials extended over more than 100 days without death; but the rats failed to maintain their weights, even with the most persistent coaxing of the appetite. Data regarding the food intake are wanting, so that the inanition factor (due to deficient calories) can not be excluded. *Voit: Zeitschrift fiir Biologie, 1866, II, p. 64. t Jacob: Zeitschrift fiir Biologie, 1906, xlviii, p. 19. +McCollum: American Journal of Physiology, 1909, xxv, p. 120. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. I I More interesting even are his studies on the growth of young rats. They made considerable gains in weight in experiments with the proteins mentioned — without casein in one series and with it in the other — and extending over from 56 to 127 days. But the curves of growth do not approach those obtained for normal diet by Donaldson, to which we shall have occasion to refer again. McCollum concludes that "the palatability of the ration is the most important factor in animal nutrition" and "the failure of previous efforts to maintain animals on a mixture of relatively pure proximate constituents of our food-stuffs was due to the lack of palatability of such mixtures." Finally reference must be made to the very successful attempts of Rohmann.* The details have not yet been published. Mice could be kept indefinitely on a mixture of casein, vitellin, egg albumin, and non-nitrogenous nutrients. They became mature and produced young which thrived. With only a single protein in the ration, a decline soon set in. If casein and egg albumin were used to replace each other the grown mice retained their weight, but the devel- opment of the younger ones was checked and they succumbed. Rohmann concluded from these facts that the different proteins possess a different significance for the nutrition and the development of young organisms. Although none of our predecessors has returned a decisive answer to the fundamental question whether any single protein or combination of proteins is incapable of supplying all the essential chemical complexes which the body is unable to furnish to itself by direct synthesis, the pursuit of a solution by no means appears futile. None of the difficulties — actual or assumed — which have arisen appears experimentally insurmountable at present. The digesti- bility and utilization of the artificial rations has never been demon- strated to be abnormal or even unfavorable. The texture and inclusion of "roughage," such as is ordinarily a part of every mixed dietary in the guise of cellulose, can be experimentally adjusted, if indeed it is of any serious moment in controlling the mechanical functions of the alimentary tract. Monotony of diet appears to have been overemphasized, if one may judge by the success with which milk or egg yolk have constituted the only food material for rats and mice. The failure common to all of the recorded experiments has been attributed to the difficulty of inducing animals to eat sufficient food. Strictly speaking, it has not been determined as yet whether the notable anorexia is the result of some unpalatable feature of the artificial food mixtures and thus the cause of the gradual inanition, or whether it is really a physio- logical sequence of an imperfect dietary. *R6hmann: Allgemeine Med. Central-Zeitung, 1903, No. 1; 1908, No. 9. Cf. Maly's Jahresbericht, 1903, xxxin, p. 823; 1908, xxxvm, p. 659, for the same abstracts. 12 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. The best hope of success — if such be possible — rests at present on the method of trial and error in which each variable is gradually eliminated by successive comparative experiments. This is the scheme which we have in large measure pursued. Our efforts have at first been directed toward devising a simple ration of isolated food components which should satisfy the numerous requirements set more adequately than any yet proposed. This established, substitutions could gradually be instituted in respect to the protein constituents. We learned before long that a diet which might be adequate for maintenance was by no means necessarily suited to the requirement of a growing animal. Hence our attention has become directed to some of the features of growth as well as those of the maintenance ration of the adult. METHODS EMPLOYED. The rats were kept in metabolism cages similar to that described by Henriques and Hansen.* A small door permits the introduction of food and water cups, such as are used in bird cages, through the side of the cage. Figure A shows the essential features. Instead of weighing the food (which was always fed in the form of a homogeneous paste) in the food cups, we devised the following very simple plan to avoid frequent weighings. The food is introduced into a glass cylinder about 25 cm. in length and 3 cm. in diameter. A rubber stopper in- serted into one end can be moved forward like a piston head and the food expelled from the other end of the cylinder into the food recep- tacle. The exit end of the cylinder is kept stopped when the food is not being expelled and the entire apparatus with its food content can be preserved in an ice-box for long periods without deterioration of the diet. The food eaten can thus be renewed at intervals and the quantity fed determined, when desired, by ascertaining the loss of weight of each food tube. Figure B illustrates our feeding-tube device. The urine and faeces were separately collected, the former in a receptacle containing boric acid and chloroform, and analyzed at intervals as indicated in the protocols. Control trials made by trickling known volumes of analyzed human urine over the cage bottom, and, after a suitable interval, washing with boric acid solu- tion, collecting the urine, etc. , just as in the rat experiments, indicated losses of 10 per cent or more. This must be borne in mind in consid- ering our results and presumably those of other investigators. We devoted great care to maintaining suitable environmental conditions (temperature, etc.), since the rats are sensitive to marked changes. With our diets they consumed large quantities of water. *Hcnriques and Hansen: Zeitschrift fur physiologische Chemie, 1904-5, xliii, p. 418. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 13 The success of the cage methods, as such, is shown by our ability to maintain animals in good health for very long periods in this way. Confronted at the outset with the necessity of ascertaining whether the conditions selected — the caging, laboratory environ- ment, consistency of the food and the mode of feeding, etc. — were B A. Sketch of cage used for feeding and collection of urine and feces. Upper figure shows outer view of food-and-water-receptacle. (Reduced to one-twelfth natural size.) B. Illustration of tube from which daily ration is discharged during each diet period. (Re- duced to one-fourth natural size.) endurable for the animals under any system of feeding, we under- took control trials with a mixed food in the form of dog biscuit and lard. This was prepared as follows : The dry dog biscuits were ground to a moderately fine powder in a mill and usually 70 parts by weight were mixed with 30 parts of melted lard . The mixture when cooled was reduced to a homogeneous paste by passing several times through a meat-chopping machine. In this way the paste was forced through small holes in numerous filaments, to which a rotary motion was imparted, insuring a very 14 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. complete and rapid mixing of the ingredients of the paste. This method of mixing was used for all of the foods described in this paper. A series of rats was fed with this food at the same time that other experimental diets were being investigated. In this way all the animals weie exposed to the same indeterminable variables of climate and environment which might perchance have exerted an unsuspected deleterious influence and which would exhibit them- selves in the control animals as well as those under special observa- 260 240 £20 too ISO ISO 140 / uo 100 90 so 40 CO / / 60 60 100 120 140 160 ISO 200 220 240 260 280 300 320 >40 360 330 400 Days Chart I. — Average Normal Rate of Growth of Male White Rat, according to Donaldson. tion. Like most of our trials during the first year of this work, they served chiefly for the purpose of orientation in respect to future procedure. It is scarcely necessary to record here the numerous individual experiments which resulted in a failure to maintain the rats in health and nutritive equilibrium. Failures, unless they are invariable in their occurrence, may well be due to accidents or inci- dents in no wise associated directly with the nutritive functions. Intercurrent parasitic diseases, incipient senility, hereditary defects, and other incidental features may be present or arise to interfere with the normal progress of an experiment. We have gradually learned to watch for such undesirable conditions and to exclude such animals as unsuitable for these studies, since proper allowance can FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 1 5 not be made for the perversions of function thereby introduced. For this reason we have been inclined to lay stress upon only those experiments which were either successful or which failed because of obvious causes. In selecting criteria of adequate growth the painstaking statis- tical studies of Donaldson* on the adjustment of size to body-weight and age in the white rat have been of great help. After birth the young white rat depends upon the mother for sustenance for about 20 days. The span of life is about three years. Sexual maturity is reached in about 60 days. The first year of rat life corresponds, according to Donaldson, to the first thirty years of human life; and the growth curve for this period has been published by him. Some of the details are reproduced in Chart I. The lack of appreciation of the salient features of these curves, representing graphically the gross normal increase in weight of white rats during the first third of their life, has led occasionally to con- clusions which appear to us as quite erroneous. If, for example, a rat weighing 250 grams maintains its body- weight for several weeks without marked variations one may properly conclude that a normal nutritive equilibrium exists in such an animal ; on the other hand a rat whose initial weight is 70 grams is in a period of most active growth. Normal nutrition for an animal in this phase calls for a measurable daily increment in weight and a gradual, yet detectable, increase in body-length. Within one month a 70-gram white rat ordinarily will double its weight when the diet is adequate. The illustrations cited suffice to indicate how different must be our cri- teria for the adolescent and the adult stages. *Donaldson: A comparison of the white rat with man in respect to the growth of the entire body. Boas Memorial Volume, 1906. i6 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. EXPERIMENTAL PART. CONTROL FEEDING. A somewhat detailed protocol of the metabolism experiments on two of the "control" rats fed on dog biscuit and lard, as already mentioned, will serve as a typical description of the conduct of all our feeding trials. Rat xn and rat xiii were caged separately on August 9, 1909. Fresh food-paste (see page 13) was introduced daily into the food dishes in excess of the amount eaten, wrhich was at first ascertained daily. The body-weights were at first deter- mined every other day, as was the nitrogen of the excreta (urine and faeces) . Subsequently it was found adequate to estimate the nitrogen balance in weekly periods. The data thus obtained are summarized Table I. — Summary of Data on "Control" Rat XII, fed on Dog Biscuit-Lard Diet for 147 Days. — Daily Averages. Intake. Nitrogen output. Date of experiment Body- weight. N-utilization. N-balance. Food. Nitrogen. Urine. Faeces. Total. 1909. gm. gm. gm. gm. gm. gm. p. Ct. gm. Aug. 8. 10. 143 . =; " J 150 0 9-5 0. 143 0 . 1 07 0 . 02 7 0.134 8l +0.009 12 ... 1 55 0 9.2 0.139 0. 103 0.029 0. 132 79 + 0.007 14 ... 157 7 8.7 0. 132 0.096 0.037 0. 133 72 — 0 . 00 1 16 .56 5 7.8 0. 124 0.098 0.030 0.128 76 — 0 . 004 18 ... 162 0 9 3 0. 149 0 . 1 06 0.012 0. 118 92 +0.031 20 ... .65 5 9.1 0. 144 0 . 1 06 0.031 0. 137 78 +0.007 22 ... 165 0 8.1 0.129 0.088 0.021 0. 109 84 + 0.020 24 ... 164 1 7.0 0.112 0 . 080 0 . 022 0. 102 80 +0.010 26 ... .65 5 6.6 0. 103 0.075 0.022 0.097 79 +0.006 28 ... 165 2 6.7 0. 105 0.073 0.021 0.094 80 +0.01 1 30 ... 162 6 5.8 0.091 0.069 0.027 0.096 70 — 0.005 Sept. 1 160 0 6.2 0.097 0.067 0.024 0.091 75 +0.006 3 ... 159 6 6.2 0.096 0 . 064 0.015 0.079 84 + 0.017 5 .... 159 7 7-3 0 . 1 1 3 0.073 0.025 0.098 78 +0.015 7 ... 161 4 8.1 0.133 0.076 0.038 0114 7' +0.019 9 ... .58 6 6.2 0. 102 0.079 0.027 0. 106 74 — 0 . 004 1 1 ... 160 7 7-5 0. 123 0.063 0.027 0.090 78 +0.033 13 ... 157 2 6.2 0. 101 0.063 0.024 0.087 /6 + 0.014 15 ... 157 9 6.1 0 . 099 0 . 06 1 0.023 0.084 77 +0.015 '7 ... 157 6 6.6 0. 107 0 . 066 0 . 029 0.095 73 +0.012 19 159 5 7-5 0. 123 0.073 0.028 0. 101 77 +0.022 26 •57 8 6.7 0. in 0.065 0.030 0.095 73 +0.016 Oct. 3 ... 154 8 5-5 0.091 0.080 0.022 0. 102 76 — O.OI I 1 1 ... .58 5 6.4 0.094 0 . 067 0.018 0.085 81 +0.009 '7 150 0 4-5 0.074 0.082 0.023 0. 105 69 — 0 . 03 1 24 . ... 147 5 5-i 0.084 0 . 067 0.015 0.082 82 +0.002 3' ... 142 2 4.8 0.080 0.052 0.020 0.072 75 +0.008 Nov. 7 ... 148 .0 5 4 0.090 0.055 0.016 0 071 82 +0.019 '4 .... 141 . 1 3-9 0.065 0.050 0.014 0.064 78 +O.OOI 21 . ... 142 •7 4.6 0 077 0.057 0.015 0.072 81 +0.005 28 • ■ • • '43 •7 4-7 0.079 0 . 049 0 . 020 0.069 75 +0.010 Dec. 5 .... 141 ■3 4 3 0.071 0.057 0.015 0.072 79 -0.001 12 . ... .47 •3 50 0.083 0.058 0.014 0.072 83 +O.OI I '9 .... 152 .2 5-4 0.090 0.058 0.015 0.073 83 +0.017 26 ... .50 4 5-1 0.084 0 . 063 0.016 0 079 81 +0.005 1910. Jan. 2 .... 151. 7 4.8 0.079 0.057 0.017 0.074 78 +0.005 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 17 in tables and also reproduced in graphic form. This experiment in common with many others was concluded after 153 days on Janu- ary 10, 1910, bya fire which destroyed all of our experimental animals. Tables I and II give the data for the "control" rats xn and xiii, which had been obtained up to January 3, 19 10, a period of 147 days, at the end of which time both rats showed a distinct gain in weight and a considerable gain of nitrogen : Table II. — Summary of Data on "Control" Rat XIII, fed on Dog Biscuit-Lard Diet for 147 Days. — Daily Averages. Intake. Nitrogen output. Date of experiment. Body- weight. N-utilization. N-balanee. Food. Nitrogen. Urine. gm. Faeces. Total. 1909. gm. gm. gm. gm. gm. p. Ct. gm. Aug. 8.. 255 . 5 10. . 266.O 12 4 O.187 0.145 0.039 0. 184 79 + 0.003 12 . . . 268.8 1 1 1 O. 168 0.197 0.041 0.238 76 — 0 . 070 14. . 275.7 12 7 0. 191 0. 126 0.038 0. 164 80 + O.027 16. . 280.O 12 1 0. 192 0.171 0.039 0.210 80 — O.Ol8 18.. • 287.5 13 4 0.214 0. 133 0.044 0.177 79 + O.O37 20. . 296. I 13 2 0.210 0.177 0.041 0.218 80 — 0 . 008 22. . 303 5 •3 6 0.216 0. 146 0.038 0. 184 82 + O.032 24. 307.O 13 8 0.219 0. 150 0.047 0.197 79 + 0.022 26. . ■ 3'7« >3 6 0.213 0.149 0.044 0. 193 79 + 0.020 28.. 325.O >3 8 0.216 0. 152 0.042 0.194 81 + 0.022 30. . 329.5 '3 7 0.214 0 . 1 44 0 040 0. 184 81 +0.030 Sept. 1 . . 333 ' 13 6 0.212 0. 141 0.046 0.187 78 + O.O25 3- 328.2 12 3 0. 192 0. 134 0.034 0.168 82 + O.O24 5- 327.0 13 3 0.207 0. 139 0.043 0. 182 79 + O.025 7- 327.2 '3 0 0.212 0.141 0.043 0. 184 80 + O.O28 9 • 3258 12 5 0.203 0. 146 0.040 0. 186 80 + O.OI7 11.. . 329.0 1 1 9 0. 194 0. 129 0.036 0. 165 81 + 0.02Q 13. 328.1 1 1 9 0 . 1 94 0 . 1 40 0.036 0. 176 Si + O.OI8 15.. 330.6 12 7 0.207 0. 138 0.038 0. 176 82 + O.O3I . '7 • 32Q.8 12 4 0.202 0. 140 0 038 0.178 81 + O.O24 19.. • 331-9 12 1 0. 197 0.147 0.030 0.177 85 + 0.020 26. . - 319-7 1 ! 2 0.187 0. 127 0.038 0. 165 80 + 0.022 Oct. 3.. . 322.8 12 5 0.207 0. 159 0.034 0.193 84 + O.OI4 11 . . • 334-7 13 9 0. 198 0.137 0.038 0.175 81 + 0.023 17.. ■ 335-8 12 2 0.200 0. 171 0.042 0.213 I9 — O.OI3 24. . • 334-5 IO 5 0. 175 0. 126 0 . 03 1 0.157 82 + O.OI8 31.. 330.5 IO 3 0. 171 0. 142 0.030 0.172 82 — O . OO I Nov. 7.. 332.0 IO 9 0.183 0. 131 0.033 0. 164 82 + O.OI9 14. . - 334-8 IO 8 0.181 0. 126 0.035 0. 161 81 + 0.020 21 . . . 328.8 9 7 0.162 0. 1 19 0.031 0. 150 81 + O.OI2 28.. 333-4 9 8 0. 163 0 . 1 1 0 0.028 0.138 83 + 0.02 5 Dec. 5. . 327 -3 8 4 0.138 0. 1 14 0.025 0.139 82 — 0 . OO I 12. . ■ 3238 9 3 0.154 0. 123 0.027 0. 150 82 + O.OO4 19.. 321.6 9 3 0. 153 0. 105 0.028 0.133 82 + 0.020 26.. 308.6 7 9 0 . 1 3 1 0 . 1 00 0.023 0. 123 82 + O.O08 1910. Jan. 2 . 307.5 8.3 0. 138 0. 1 15 0.022 0.137 S4 + 0.001 These and other analytical data have been introduced in this paper in part reproduced in graphic form. In all of the charts the abscissa units represent days, and the ordinate units food (broken line) or body- weight (solid line). The food-intake curve is plotted from the total amount eaten per week. The average daily nitrogen balance is indicated as above ( + ) or below ( — ) the heavy line (o). 180 160 14-0 60 50 40 30 ♦ 0-10 V Food eaten Nitrogen balance 0=tj- JWjfmfrn- E (0 i_ o -0-10 340 320 300 280 260 80 20 40 60 60 100 120 140 160 Days Chart II. — "Control" Rat xn fed on Dog Biscuit-Lard Diet for 147 days. 60 ♦ 0.05 • 0|=1 in E (0 t_ -0-10 A^ V Body weight 4 burn< id up s s * / / A \ Fo \ \ \ \ 3d eaten \ v.. \ \ \ Nitroge n balance 1 -"' Mx rr-mr J 1 ^ L 20 40 60 120 140 160 80 100 Days Chart III. — "Control" Rat xm fed on Dog Biseuit-Lard Diet for 147 days FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 19 DISCUSSION. It will be noted that Rats xn and xin, representing different ages, if we judge from the initial weights, were maintained in nutri- tive equilibrium without loss of body- weight during the period of 147 days — a not inconsiderable fraction of the life span of these animals. There is, however, a gradual decline in the amount of food which is eaten toward the end of the experiment, the quantity in the case of xii approaching limits which must have necessitated some demand upon the fat supply earlier accumulated. The utilization of the protein continued satisfactory, thus evincing unimpaired digestive powers It is far from likely that the ration used, with its large preponderance of energy in the form of fat, is an ideal one. The facts recorded, however, exclude the probability that monotony of diet is an insurmountable obstacle to nutritive success.* As to the possibility of prolonged feeding on a uniform unchanged diet, two illustrations are appended of experiments on rats 14 and 18 fed with a mixture of ground hempseed, starch, lard, and salts. These rats were first fed with a mixture of dog biscuit and lard for several weeks and then on the hempseed mixture. The composition of the food given during the different periods is shown in table III. Table III. — Composition of the Food in Percentages. Dog- biscuit. Lard. Nitrogen. Hemp- seed. Starch. Lard. Sodium chloride. Salt mix- ture I.f Nitrogen. Rat 14. Period i . . . . Period 2 . . . . Period 3 . . . . 58 A1 1.6 1.9 JO 30 46 46 50 42 42 38 ,0 IO 10 2 2 2 2.27 2.40 2.32 Period 4 Period 5 Rat 18. Period 1 . . . . Period 2. . . . Period 3. . . . i 58 42 70 30 ,6 1 .9 46 46 50 42 42 38 ,0 IO IO 2 2 2 2.27 2 39 2 45 Period 4. . . . Period 5 . . The figures for nitrogen are averages of the different batches of food which were made up from time to time. The hempseed meal was freed from the greater part of the hulls by sifting, but the different lots contained different proportions which escaped separation ; hence the actual nitrogen content of the individual batches of food varied somewhat. The figures given in the protocols and representing daily averages are based on the actual quantity of nitrogen fed, not on the averages given in table III. *Among the many often unapparent difficulties which beset such experiments, the frequent occurrence of intestinal parasites and the susceptibility of the animals o digestive disturbances are to be noted. fThe salt mixture 1, which contained organic and inorganic salts of the necessary bases and acids, is described on page 32. 20 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Rat 14 lived 200 days without marked change in weight (losing less than 10 per cent); rat 18 (still on the same diet after 322 days at the time of writing) weighs very nearly as much as at the begin- ning of the hempseed feeding. Table IV. -Summary of Data on Rat 14 fed on Hempseed-Starch-Lard Diet for 207 Days.— Daily Averages. Period I. Intake. Nitrogen output. Date of experimen Body- :. weight. N-utilization. N-balance. ! Food. Nitrogen. Urine. Faeces. Total. I9IO. gm. gm. ' gm. gm. gm. gm. p. Ct. gm. Jan. 24 ... 284.0 ... 274.2 3' 8.50. 134 0. 123 0.029 0. 1 52 78 — O.Ol8 Feb. 7 ... 268.3 8.2 0. 133 0. 133 0.023 0. 156 83 — O.023 Period 2. Feb. 14 262 . 5 8.7 0. 168 0.137 0.032 0. 169 8. — O . OO 1 21 ... 255.4 7-4 0. 143 0. 147 0.027 0.174 81 — O.O3I 28 ... 257.0 9 3 0. 178 0. 145 0.032 0.177 82 + O.OOI Mar. 7 ... 259.6 8-9 0. 168 0. 136 0.028 0. 164 83 + O.OO4 Period 3. - Mar. 14 208 . 0 .6 0.013 0.115 0.011 0.126 15 — O. I 13 21 212.2 54 0. 123 0. 131 0.016 0. 147 87 — O.024 Period 4. Mar. 28 ... 234.2 9 7 0.231 0 . 1 44 0 . 044 0 . 1 88 81 + O.O43 Apr. 4 252 .0 1 1 2 0.270 0. 157 0.056 0.213 79 + O.O57 18 ... 237.7 8 1 0 . 1 93 0 . 1 62 0.037 0. 199 81 — O.O06 May 2 ... 244 . 5 8 3 0 . 1 99 0 . 1 6 1 0.039 0.200 80 — O . OO I 16 260.3 9 4 0.288 0.168 0.056 0.224 75 +0.004 30 . . . 250.2 7 3 0.176 0.143 0.027 0.170 85 +0.006 June 6 ... 254.0 9 9 0234 0.158 0.046 0.204 80 +0.030 20 . . .! 242.0 8 3 0. 194 0. 150 0.044 0.194 77 0.000 Period 5. June 27 ... 251.8 8.5 0.205 0. 142 0.039 0.181 81 +0.024 July 4 248.0 8.1 0.200 0.155 0.026 0.181 87 i +0.019 1 1 ... 247.4 8.4 0.208 0. 156 0.043 0.199 79 +0.009 18 ... 251.5 8.7 0.210 0. 127 0.045 0. 172 79 +0.038 25 ... 253.0 8.9 0.213 0. 148 0.040 0.188 81 +0.025 Aug. 1 ... 261.8 10.5 0.252 0. 141 0.069 0.210 73 +0.042 8 ... 264.8 9.2 0.232 0. 103 0.018 0. 121 92 +0. 1 1 1 ■5 262 . 2 12.2 0.249 0.134 0.064 0. 198 74 +0.051 22 ... | 266 . 6 12.5 0.253 0. 108 0.076 0. 184 70 +0.069 29 ... 252.2 10.9 0.221 0. 105 0.069 0.174 69 +0.047 Sept. 5 ... 249.5 11. 8 0.262 0. 141 0.077 0.218 7' +0.044 12 ... 241.5 9-5 0.219 0.174 0.044 0.218 80 +O.OOI 19 ... 245.0 13-5 0.314 0.215 0. lOI 0.316 68 — 0 . 002 26. ... 237.8 10. 1 0.235 0.201 0.051 0.252 78 — 0.017 30. 1 80 . 2 Dead. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 21 280 260 24-0 220 200 80 60 40 0.10 0 E o 0.10 1 \2 3 Body weight 5 1/4- r>r\c\ Food eaten K • \ '* ao / \ Idead i 1 A \ /A # N i / V v/ \ r \ / V . i * i i / V 1 \ / -J N — "" ■** ^ 1— , i i i / 1 i 1 i 1 * Nitroger balance -+c ).I0J 20-r— -, 1 / i i i / w i r~r b= -LJ~ U_J 20 40 60 80 100 120 14-0 Days 160 180 200 220 24-0 Chart IV. — Rat 14 fed on Hempseed-Lard Diet for 207 days. Numbers on body-weight line indicate time at which each period began. Table V. — Summary of Data on Rat 18 fed on Hempseed-Starch-Lard Diet for 322 Days.- — Daily Averages. Period i. Date of experiment. Body- weight. Intake. Nitrogen output. N -utilization. N-balance. gm. Food. gm. Nitrogen. Urine. Faeces. Total. gm. I9IO. Jan. 24. . . . 31... Feb. 7... gm. 268.3 248.7 233-5 gm. gm. gm. p. ct. 6.2 4.9 O . 099 O . 1 2 ! O.O78 O.O95 0.024 0.015 0.145 0. 1 10 76 81 — O . 046 — O.032 Period 2. Feb. 14 21 ... . 28.... 237.0 248.0 251.8 8.5 9.2 8-9 O. 163 O. I42 O.I77 O. 145 O. 171 O. I I I 0.027 0.027 0.035 0. 169 0. 172 0. 146 83 85 80 — 0 . 006 +0.005 +0.025 22 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Table V. — Summary of Data on Rat 18 fed on Hempseed-Starch Lard Diet for 322 days. — Daily Averages— Continued. Period 3. Intake. Notiogen output. Date of experiment. Body- weight. N-utilization. N-balance. Food. Nitrogen. Urine. gm. Faeces. Total. 19 10. gm. gm. gm. gm. gm. p. Ct. gm. Mar. 7- • 236.O 3-2 0.072 O.O99 O.Ol8 0. 1 17 75 -O.045 14. 200.0 3-5 0.080 O. 127 O.O27 O.I54 66 -O.074 21 . . I94.7 6.2 0. 140 O. IO4 O.O34 O.138 76 + 0.002 28. . 182.O 3 5 0.079 O. I 17 O.O42 O.159 47 — 0 . 080 Period 4. Apr. 4- ■ 168.2 4.2 0.098 0. 131 O.OI8 O.I49 82 — O.O51 18.. . I59.O 5 -» 0. 129 0.135 0.028 O. 163 78 -O.034 May 2. . • >75-5 6 7 0. 161 0. 1 14 O.O26 O. 140 84 + 0.02I 16.. 216.0 1 1 8 0.287 0. 118 O.060 O.178 79 + 0. 109 30. . 230.0 7 7 0.187 0. 107 O.032 O.139 83 + O.O48 June 6. . 229.0 8 5 0.205 0. 102 O.O4I O.I43 80 +O.062 20. . 242.0 8 2 0.195 0. 109 O.O46 O.I55 76 + O.040 27.. 235.0 5 8 0. 135 0. 104 O.027 0 . 131 80 + 0.004 July 4 ■ 235.5 8 0 0.188 0. 127 O.036 O. 163 81 + O.025 11.. ■ 243.8 9 3 0.230 0.138 O.O4I O. 179 82 +O.05I Period 5. July 18... • 243.1 7-2 0.177 0.119 O.039 O.I58 78 + 0.019 25. .. • 242 . 5 7.6 0.183 0.139 O.O35 O.I74 81 + 0.009 Aug. 1 . . . 240 . 0 8.4 0.201 0. 142 O.O43 O.185 79 + O.Ol6 8.. . 255.0 9-3 0.223 0.144 O.044 O.188 80 + O.035 15... 261.7 12.7 0.264 0.138 O.086 O.224 67 +O.040 22. . . 257.3 11. 8 0.240 0 . 1 3 1 OO72 O.203 70 + O.037 29.. 261.2 '3 9 0.288 0. 117 O.O96 0.213 67 + O.075 Sept. 5. .. 256.0 10.5 0.240 0.137 O.086 O.223 64 + 0.017 12. . . 239.0 8.8 0.203 0. 189 O.O44 O.233 78 — 0.030 19... • 234.0 n. 8 0.275 0.193 O.085 O.278 69 — 0 . 003 26... 215.8 7-3 0. 171 0.191 O.O37 O.228 78 -0.057 Oct. 3 ■■• 220.5 1 1 .0 0.257 0.185 O.O45 O.23O 82 + O.027 10. . . • 231.5 11. 5 0.301 0.216 0.04I O.257 86 + O.O44 17... . 228.7 10.5 0.275 0.200 O.068 O.268 75 + 0.007 24... 225.0 8.7 0.227 0.187 O.060 O.247 74 — 0.020 31... 230.8 9.6 0.255 0. 171 O.O55 O.226 78 + O.029 Nov. 7... 230.0 10.2 0.281 0.185 O.O76 O.261 73 + 0.020 14.. 227.0 7.6 0.209 0.173 O.O48 0.22I 77 — O.OI2 21 . . . 218.7 8.3 0.229 0. 196 O.O42 O.238 82 — 0.009 28... 222.5 8.5 0.233 0. 164 O.O37 0.20I 84 + O.032 Dec. 5... • 233.2 10. 1 0.276 0. 196 O.048 0.244 83 + O.O32 12. . . 220.5 9.2 0.253 0.204 O.O58 0.262 77 — 0 . 009 19. •• • 233.3 10.6 0.292 0. 196 O.O58 O.254 80 + O.038 '9 Jan. 26... 1 m 9.0 0.246 0. 191 O.O55 O.246 78 O.OOO 2. . . 239.0 9 7 0.253 0.174 O.O58 O.232 77 + 0.02I 9... 232.7 6.6 0. 164 0.143 O.024 O. 167 85 — 0 . 003 16... . 239.8 7- 7 0.191 1 0. 148 O.O24 O. 172 87 + O.OI9 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 23 r Body weight 3 b J Z \ \ .A ^ A^ \ Food eaten \ \4 I * ^« /*, i 1 1 \ / \ / V 1 1 \ 1 / V \ / - V- ■^ t f 1 1 \ 1 l 1 l * A 1 / \/~ V \ V'' s 1 I 1/ V \ . — 1 \ V-' \; 1 l / Nitrogen balance | 1 1 I! n n 1 _l *— ' L 100 120 180 200 Days 220 240 260 £80 300 320 340 3fc0 Chart V. -Rat 18 fed on Hempseed-Lard Diet for 322 days. Numbers on body-weight line indicate time at which each period began. Other trials with the same diets frequently led to a decline in weight and a loss of body protein. In nearly all of these cases the insufficient food-intake was adequate to explain the incipient symp- toms of inanition. Our numerous attempts to vary the flavor of the food and thus increase its palatability have been without striking success. Under exactly similar conditions of diet and environment different rats may continue to exhibit markedly unlike appetite for the same food. It seems best in the present stage of our knowledge to exclude from the diet experiments all animals which exhibit what seems like a temperamental anorexia. Protocols from some of these experiments are recorded here for comparison. Rats xxviii and xxix were fed throughout the entire period on the dog biscuit-lard mixture with the results shown in tables VI and VII. 24 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Table VI. — Summary of Data on Rat XXVIII fed on Dog Biscuit-Lard Diet for 105 Days. — Daily Averages. Intake. Nitrogen output. Date of Body- weight. N-utilization. N-balanee. Food. Nitrogen. Urine. Faces. Total. 1909. gm. gm. gm. gm. gm. gm. p. Ct. gm. Sept. 19... 21 . . . 176 170 0 4 7-8 0. 126 O. 122 O.043 O.165 66 -O.039 23... 171 8 7-i 0. 1 14 O. I07 O.036 O.I43 68 — O.O29 25... 170 5 7-5 0. 125 O.O96 O.032 0. 128 74 — O . OO3 30... ■ 155 1 5.6 0.093 O. 106 0.020 0. 126 78 -O.O33 Oct. 3... • 152 8 5.8 0.096 O.O96 O.026 O. 122 73 — 0.026 10. . . ■ 152 5 5-9 0.095 O.082 O.OI9 0. IOI 80 — 0 . 006 17... '47 5 5.1 0.083 O.O75 O.OI8 O.O93 78 — 0.010 24... 148 5 6.9 0. 1 14 O.080 0.022 O. 102 81 +0.012 31... 141 3 4.8 0.080 O.067 O.OI2 O.O79 85 +0.001 Nov. 7... • '45 0 5-7 0.095 O.066 O.OI2 O.O78 87 +0.017 14... • >43 8 6.1 0. 106 O.O72 O.OlS O.O9O 83 +0.016 21 . . . • 143 4 5-4 0.099 O.O7O 0.0l6 O.086 84 +0.013 28... • '35 5 5.0 0.091 O.083 O.Ol6 O.O99 82 —0.008 Dec. 5 ... 134 5 5-3 0.097 O.080 0.017 O.O97 82 0.000 12. . . 131 5 5.8 0. 107 0.090 0.02I O. I 1 I 80 — 0 . 004 19... 126 6 5-2 0.096 O.O78 0.019 O.O97 80 — 0 . 00 1 26. . . 1 19 4 4.8 0.080 O.O75 O.Ol6 O.O9I 80 — 0.01 I 19 10. Jan. 2. . . 116. 9 5.6 0.093 O.087 O.023 O. I IO 75 -0017 Table VII. — Summary of Data on Rat XXIX fed on Dog Biscuit-Lard Diet for 103 Days. — Daily Averages. Intake. Nitrogen output. Date of experiment. Body- weight. N-utilization. N-balanee. Food. Nitrogen. Urine. Faeces. gm. Total. 1909. gm. gm. gm. gm. gm. p. Ct. gm. Sept. 19... 21 . . . ■ 1750 I 70 . 2 8.6 O.138 o..38 O.038 0. 176 72 — O.O38 23... 174.2 8.8 O. 141 O. 127 O.035 0. 162 75 — 0 . 02 I 25... 8.3 O.137 O. 103 O.O32 0.135 77 + 0.002 30... • 163.3 6.6 0. 107 O. 107 O.025 0. 132 77 — O.025 Oct. 3 . . . 164.4 8.2 0. 136 O. 106 O.028 0.134 79 + 0.002 10. . . 162.4 7-2 0. 1 19 O.094 O.028 0. 122 76 — 0 . 003 17... • 156.4 6.4 0. 105 O.083 O.026 0. 109 75 — 0 . 004 24.. 155.0 7-7 0. 128 O.079 O.026 0. 105 80 + OO23 3i • 151. 5 6.6 0. 1 10 O.O75 0.02I 0.096 81 + O.OI4 Nov. 7.. 150.6 6.9 0. 1 16 O.079 O.025 0. 104 78 + O.OI2 14 .. 151 .0 6.2 0. 1 1 1 O.O75 O.OI8 0.093 84 + O.OI8 21 . . 153-4 6.9 0. 125 O.077 O.023 0. 100 82 + O.O25 28.. 150.3 7.8 0. 142 O.076 O.023 0.099 84 + O.O43 Dec. 5 . . 146.8 6.4 0. 1 17 O.082 O.027 0. 109 77 + 0.008 12. . 142.2 6.0 0. 1 10 O.086 O.025 0. 1 1 1 77 — O.OI 1 19.. 124.7 32 0.058 O.069 O.OI4 0.083 76 -O.025 26.. 1 1 1 .9 3-7 0.062 O.078 0.015 0.093 76 — O.O3I 3i ■ 95 5 3-5 0.057 O.092 O.026 0.118 54 — 0 . 06 1 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 25 180 160 140 120 50 30 ♦ 0.05 - (0 E (0 L. o •0.10 Body weight \ Food eaten burne( J up \ \ \ L A ^^-^ • - L Nitrogen balance 1 1 I— J J- 4 > — 1 — 1 1 20 40 60 80 Days 100 120 Chart VI. — Rat xxvni fed on Dog Biscuit-Lard Diet for 105 days. Table VIII. — Composition of the Food in Percentages. Dog biscuit. Lard. Nitro- gen. Tru- milk.* Starch (arrow- root). Lard. Salt mix- ture I. Nitrogen. Rat 10. Period 1 Period 2 Period 3 Period 4 Period 5 Rat 1 1 . Period 1 Period 2 Period 3 Rat 12. Period 1 Period 2 Period 3 58.O 7O.O 58.O 70.O 58.O 70.0 42.O 30.O 42.O 30.0 42.O 30.O I.58 1 -93 ..58 '93 1.58 '•93 54-0*f 52.o*f 42.0*1 16.O 18.O 24.O 30.O 30.O 34 0 2.51 2.87 2.31 60.0 16.7 23-3 2-53 60.0 60.0 16.7 '5-7 23-3 23 3 I .O 253 2-47 Period 4 *"Trumilk" is a commercial milk powder. fThis was extracted in the laboratory once with 95 per cent alcohol, once with absolute alcohol, and four times with ether. 26 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 180 Days Chart VII. — Rat xxix fed on Dog Biscuit- Lard Diet for 103 days. Rats 10, 11, and 12 were first fed on the dog biscuit-lard mixture and later on one containing desiccated milk. The composition of their food is shown in table VIII. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 27 Table IX. — Summary of Data on Rat 10 fed on Dog Biscuit-Lard Diet for 98 Days and then on Milk Powder-Starch-Lard Diet for 84 Days. — Daily Averages. Period i. Intake. Nitrogen output. Date of Body- N-utilization. N-balance. experiment. weight. Food. Nitrogen. Urine. Faeces. Total. '9 10. gm. gm. gm. gm. gm. gm. p. Ct. gm. Jan. 24... 31... 237.6 216.3 4.8 0.075 O. I02 O.OI5 0. 1 17 80 —0.042 Feb. 7 ... 2IO.O 5-7 0.090 0. 107 O.OI4 O. 121 84 — 0.03 1 Period 2. Feb. 14... 208.O 7-9 0. 151 0. 127 O.O29 0. 156 8l — 0.005 21 . . . 21 I .3 8.1 0.155 0. 129 O.O27 O. 156 83 — 0.001 28... 212.7 6.9 0. 132 0. 105 0.022 O. 127 83 +0.005 Mar. 7... 208.7 6.0 0. 1 15 0. 107 O.Ol8 O.125 84 — 0.010 14... 200.0 5-1 0.096 0. 104 O.Ol6 O. I20 83 — 0.024 21 . . . 193-7 5.0 0.096 0. 104 O.Ol6 O. I20 83 — 0.024 Apr. 4... 186.5 5-7 0. 1 12 0.098 O.Ol8 0. 1 16 84 — 0 . 004 18... 177-5 5-6 0. 106 0. 103 O.OI4 0.117 87 — 0.011 May 2. . . 182.8 6-4 0. 120 0. 101 O.Ol8 0. 1 19 85 +0.001 Period 3. May 9... 189.7 5-8 0.145 0. 1 16 O.OI3 0. 129 91 +0.016 16... 196.5 6-3 0.159 0.084 O.023 0. 107 86 +0.052 23... "91-5 5-5 0.138 0.118 O.OI5 0.133 89 + 0.005 Period 4. May 30... 185.0 5' 0.134 0.134 O.OI4 0. 148 90 — 0.014 June 6... 175-5 5-9 0. 169 0. 169 O.OI9 0.188 89 — 0.019 13... 172. 1 6.4 0.183 0. 167 O.OI3 0. 180 93 +0.003 20. . . 176.2 4-6 0. 132 0. 146 O.OI 1 0.157 92 —0.025 Period 5. June 27... 165.5 4-5 0. 104 0. 1 13 O.OlS 0. 131 83 —0.027 July 4. .. 163.7 4 9 0. 1 14 O. I IQ O.OI4 0.133 88 —0.019 11.... 149.6 4-5 0. 103 0. 1 17 O.OI4 0. 131 86 —0.028 18.... 142.5 4 ' 0.095 0. 108 O.OI7 0. 125 82 — 0.030 25.... 134-3 4-4 0. 104 0. 131 0.022 0.153 79 — 0 . 049 Chloroformec 28 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Table X. — Summary of Data on Rat 11 fed on Dog Biscuit-Lard Diet for 84 Days and then on Milk Powder-Starch-Lard Diet for 94 Days. — Daily Averages. Period i. Intake. Date of experiment. Body- weight. Nitrogen output. N-utilization. N-balance. Food. Nitrogen. Urine. Fa?ces. Total. 1910. Jan. 24. 31 Feb. 7. gm. gm. 2243 213.7 6.4 208.8 7.3 gm. gm. o. 101 o. 1 16 gm. 0.018 gm. p. ct. 0.107 0.010 0.125 o. 1 16 0.020 o. 136 Period 2. 1910 Feb Mar. Apr. 14.. 196 21 . . 202 28.. . . 195 !■■ .. 198 14.. .. .87 21 . . 190 4-- .. .65 18.. 170 • / •5 .0 .0 •3 . 1 .0 .2 6.3 7-4 6.4 6.9 5 3 5-1 6.3 6.6 o. 120 o. 142 o. 123 0.131 o. 100 o. 100 o. 123 o. 129 o. 1 15 o. 126 O. 120 O. I 10 o. 102 0 102 0 102 0 1 1 1 0 020 0 135 0 018 0. 144 0 023 0. 143 0 021 0. 131 0 015 0. 117 0 018 0 120 0 019 0 121 0 020 0 .31 82 83 gm. — o . 024 —0.020 83 — 0.015 87 — 0 . 002 81 — 0.020 84 0.000 85 — 0.017 82 — 0 . 020 85 +0.002 84 — 0 . 002 Period 3. 1910. Apr. 25. . . . May 2 . . . . 9... 16 23.... 30.... June 6. . . . 13... 20. . . . 27.... July 4.... 11.... 18.... 21 ... Dead 188.0 194.6 '93 5 •94-3 200.0 192.2 .83.5 172.1 191.5 190.2 188.5 187.0 183.7 •54 5 7-i 7.0 7-3 6.8 6.9 5' 5-9 4i 7-3 5.8 6.1 5.8 6.2 0.179 o. 176 0.182 o. 170 174 129 15' 105 0.183 o. 146 0.153 0.145 0.155 0.097 0.010 0. 107 0. 120 0.023 0.143 0. 132 0.030 0. 162 0.137 0.021 0.158 0. 130 0.021 0. 151 0.118 0.013 0.131 i 0.147 0.021 0.168 0.159 0.019 0.178 0.148 0.020 0.168 0. 102 0.022 0. 124 0. 127 0.024 0. 151 0. 141 0.029 0. 170 0. 123 0.028 0.151 94 87 84 88 88 90 86 82 89 85 84 80 82 +0.072 +0.033 +0.020 +0.012 +0.023 — o . 002 —0.017 -0.073 +0.015 +0.022 +0.002 — o . 02 5 +0.004 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 29 Table XI. — Summary of Data on Rat 12 fed on Dog Biscuit-Lard Diet for 84 Days and then on Milk Powder-Starch-Lard Diet for 113 Days. — Daily Averages. Period i . Intake. Nitrogen output. Date of Bodv- N-utilization. N-balanee. experiment. weight. Food. Nitrogen. Urine. Faces. Total. 1910. gm. gm. gm. gm. gm. gm. p. Ct. gm. Jan. 24... 31... 231.7 212.0 5.7 O . 09 1 O . 096 0.021 O. I 17 77 — O.026 Feb. 7... 196.3 5.3 0.085 ° '°4 O.Ol6 O. 120 81 -O.035 Period 2. 1910. Feb. 14... 192.2 6.1 O.IlS 0. 1 16 O.OI9 O.135 84 — O.OI7 21.-... .! 187.6 6.6 O. 127 0.134 0.025 O.159 80 — O.032 28... . 182.5 6.1 O.II8 0. 124 O.OI9 O.I43 84 — 0 . 02 5 Mar. 7. . . .1 181.3 5-5 O. I04 0. 1 14 O.OI2 0. 126 88 — 0.022 14... 182.2 7-« O.I34 0. 1 17 0.020 0.137 85 — O.OO3 21 . . . . 184.6 7-3 O. 141 0. 108 O.OI5 0. 123 89 + O.Ol8 Apr. 4... . ' 1 76 . 2 5.8 O. 1 13 0. 109 0.02I 0. 130 81 — O.OI7 18... . 171.2 6.0 0.1I8 0. 109 O.OlS 0. 127 85 — O . OO9 Period 3- 19 10. Apr. 25. . . 188.7 6.3 O.I59 0.091 0.014 0. 105 Ql + O.O54 May 2 . . . 191 7 6 3 O.I58 0. 107 0.018 0. 125 89 + O.O33 9... 198 5 7 2 O. 180 0. 126 0.021 0.147 88 +O.O33 16... 196 7 7 1 O.I78 0. 126 0.028 0.154 84 + O.O24 23... 196 4 6 0 O. 151 0. 1 17 0.021 0.138 86 + O.OI3 30... 196 6 5 6 O. 142 0. 109 0.025 0.134 82 + 0.008 June 6. . . . 181 4 5 7 O. 146 0. 141 0.027 0.168 82 — 0.022 13... 176 3 5 3 O. 135 0. 132 0.014 0. 146 90 — O.OI I 20. . . '75 0 4 7 O.IlS 0. 131 0.013 0.144 89 — O.O26 27.. 173 3 4 5 O. I 12 0.099 O.OIO 0. 109 9> + O.OO3 July 4-- . .65 6 4 8 O. 122 0. 125 0.018 0.143 85 — O . 02 I n . . 162 5 5 0 O. 126 0.138 O.OI2 0. 150 90 — O.O24 18.. 155 0 4 6 O. I 15 0. 1 19 0.014 0.133 88 — O.Ol8 25.. 143 5 4 1 O. IO3 0. 163 0.009 0.172 9i — 0 . 069 Period 4- 1910. Aug. 1 . 137.0 4.6 O. 1 14 0. 104 0.017 O. 121 85 — O . OO7 8.. 115.5 3-7 O.O9I ; 0. 129 0.015 O.I44 84 -O.O53 9.. 110.5 Dead 220 200 180 160 50 30 ♦ 0.05 - E ID l_ J> -0.10 1 \2 Body weight 3y 4 5 Food eaten / / / \ N .- — ''' ^ ^^" ^ \ \ .' killed - Nitrogei i ba lane e J ' 1 1 i r — i J~~ 20 40 60 80 100 Days 120 140 160 180 Chart VIII. — Rat io fed on Dog Biscuit-Lard Diet for 98 days and then on Milk powder-Starch-Lard Diet for 84 days. Numbers on Body-weight line indicate time at which each period began. 220 200 180 160 50 40 30 '0.05 m E L. <^> -0.10 Chart IX. D 1 \2 Body w sight 3/ A • \ \ \ *-'' Food e aten \ \ f \ dead - V - w " \ \ / / - Nit ■ogen balance J 1 1 L-iJ 1 "" 20 40 60 80 120 140 160 130 100 Days — Rat 1 1 fed on Dog Biscuit-Lard Diet for 84 days and then on Milk powder-Starch-Lard iet for 94 days. Numbers on Body-weight line indicate time at which each period began. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 3 1 ♦ 0.05 Chart X. — Rat 12, fed on Dog Biscuit-Lard Diet for 84 days and then on Milk powder-Starch- Lard Diet for 113 days. Numbers on Body-weight line indicate time at which each period began. DISCUSSION. The rats xxviii and xxix showed a steady decline on the dog- biscuit-fat mixture (p. 13). In considering the quantity of food eaten it must be borne in mind that the nitrogen content was rather low (N= 1.65 per cent). Rats 10, 11, and 12 also showed, during the early period of the experiment, a steady decline on the same diet, although it contained more nitrogen . The temporary improvement shown after several weeks at the point marked 3 indicates the introduction of a change in diet. The improvement was, however, only temporary, as the charts in- dicate. The curves of body-weight in these animals recall those published by Falta and Noegerrath and correspond with the data of other investigators mentioned above. They indicate the type of experiment which is unsuccessful because of more or less obvious insufficiency in food- intake or stored supply. In the case of rats xxviii and 10, for example, this is pronounced and a steady and continuous decline is noted. The decline of rat xxix, at first gradual, became extremely marked with the striking decrease in the food- intake at the end of the experiment. The other illustrations (rats 11 and 12) show intermediate types. As a rule, older, full-grown animals exhibit slower decline than younger rats (of smaller weight) 32 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. because of the greater store of fat, etc. This helps to explain the long period of survival noted by various earlier investigators who have usually employed older (mature) animals for their feeding trials. THE CASEIN DIET. PRELIMINARY TRIALS. More experiments of this sort have been carried on with casein than with any other protein as the sole nitrogenous component. It can readily be isolated in a state of comparative purity; and inas- much as it contains phosphorus in an organic complex, known by long experience to be assimilable, one of the problematical features pertaining to most other proteins is eliminated by its use. Our efforts have been directed toward rinding a diet containing casein as the sole protein and which might meet the requirements of a long- continued experiment. In this we have been to a large degree suc- cessful in the case of mature rats. Our earlier trials were conducted with a variety of food-mixtures containing casein. It was soon apparent that the protein require- ment of the animals can be satisfied with comparative ease ; accord- ingly the ration was prepared with a nitrogen concentration of about 2.5 per cent. The necessity for the use of much fat to insure the requisite paste consistency (and thus avoid scattering of the food) has put distinct limitations on the range of choice. We have tried without success to avoid the use of so much fat. To indicate the variety and proportion of inorganic elements which we have attempted to introduce, some of the mixtures are given in table XII. Table XII. Salt mixture I (Rohmann). Salt mixture II (McCollum). Salt mixture III. grams. grams. Ash of milk 60.6 NaCl 33.4 Ca:i(P04)2 24.2 KC1 33.4 NaCl 12.1 Bone ash 25.1 grams. Caa(P04), 10.0 K2HP04 37.0 NaCl 20.0 Na citrate 15.0 Fe citrate 3.1 Na2C03 6.7 Mg citrate 8.0 Fe citrate 1.4 Ca lactate 8.0 100.0 Fe citrate 2.0 100.0 100.0 It is still debatable whether any "roughage," such as cellulose, is absolutely necessary. McCollum* fed egg-yolk alone to white rats for 18 weeks without unfavorable results. Nevertheless we have introduced an indigestible residue as conforming more nearly to the usual alimentary experience of the animals; and agar-agar was selected because it is more easily manipulated than cellulose and because experience with other animals has shown us how efficient it is for this purpose. f *McCollum: American Journal of Physiology, 1909, xxv, p. 127. fCf. Saiki: Journal of Biological Chemistry, 1906, II, p. 251. Swartz: Transactions of the Connecticut Academy of Arts and Sciences,i9ii,xvi, p. 247. Mendel and Swartz: American Journal of Medical Sciences, March, 19 10. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 33 In order to compare the suitability of casein diets containing the salt mixtures recorded above, a series of trials was begun on rats which had previously been under observation in cages so that any eccentricity of eating or of metabolism might be noted. During this preliminary period the diet consisted of the dog biscuit mixture already referred to in the "control" series. During the casein periods the following diets were used, which are indicated in the tables by the corresponding numbers. Table XIII. Pure casein Cane sugar Starch (arrow-root). Lard Agar Salt mixture I Salt mixture II Salt mixture III. . . . '3 4 20. 6 23 -7 35.0 5-2 18.0 15.0 29.5 30.0 5.0 10.0 4.0 46.0 30.8 4-4' 4.8 18.0 15.0 29.5 30.0 5.0 2-5 2. 1 25 Nitrogen. ioo.o 1.87 100.0 2. si 100.0 ..38 100.0 2.54 12.0 15.0 30.0 30.0 5.0 18.0 15.0 29.5 30.0 5.0 8.0 2-5 100.0 1 54 100.0 2.53 The records of the animals during the entire course of the experi- ment, until it was stopped by the loss of the laboratory by fire, are shown in Tables XIV-XXII and Charts XI-XIII. Table XIV. — Summary of Data on Rat XXX, fed on Pure Casein (as the only Protein) and Salt Mixture III for 42 Days. — Daily Averages. Period i. — Dog Biscuit-Lard Diet. Date of experiment. Intake. Body- weight. Food. Nitrogen. Nitrogen output. N-utilization. Urine. Faeces. Total. I909. Nov. I . 8. '5- 22. gm. 159.0 178. 8 191. 3 185.6 gm. gm. gm. gm. gm. p. ct. N-balanee. gm. 9.8 0.158 0.139 0.019 0.158 88 0.000 10.8 0.180 0.119 0.032 0.151 82 +0.029 9.3 0.154 0.1 14 0.028 0.142 82 +0.012 Period 2. — Casein DreT i. 1909. Nov. 29. Dec. 6. '3 167.0 5.1 0.096 0.091 0.023 0. 1 14 166.1 6.0 0. 1 12 0.091 0.020 0. 1 1 1 162.7 5-4 0. 101 0.091 0.015 0. 106 76 82 85 — 0.018 +0.001 — 0.005 Period 3. — Casein Diet 2. 1909. Dec. 20. 27- 1910. Jan. 3. 164.5 5-7 0.144 0. 123 0.015 0.138 170. 1 b.5 0. 164 0.139 0.014 0 153 174 3 7-1 0. 180 0. 149 0.017 0. 166 90 91 9' +0.006 +0.01 1 +0.014 34 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Table XV. — Summary of Data on Rat XXXI, fed on Pure Casein (as the only Protein) and Salt Mixture III for 42 Days. — Daily Averages. Period i. — Dog Biscuit-Lard Diet. Date of experiment. 1909. Nov. 1 . 8. 15- 22. 1909. Nov. 29. Dec. 6. '3- 1909. Dec. 20. 27- 1910. Jan. 3. Intake. Body- weight. Nitrogen output. N-utilization. N-balance. Food. Nitrogen. Urine. Faeces. Total. gm. gm. 181. 7 204.2 10.8 204 .5 8.2 197. 1 7.7 gm. gm. gm. gm. 0.173 0. 136 0.128 0. 140 0. 107 0. 1 13 0 . 02 5 0.01 1 0.024 0. 165 0.118 0.137 p. ct. Period 2. — Casein Diet i. 184.1 4-7 0.089 0.089 0.018 0. 107 186.1 6.9 0. 129 0.089 0.022 0. 1 1 1 1997 9.0 0.168 0.092 0.033 0. 125 Period 3. — Casein Diet 2. 208.5 202.7 7 9 9 4 0. 199 0.237 0. 1 1 1 0.023 0 . 1 60 0 . 026 0.134 0.186 198.5 6.5 0. 164 0. 146 0.030 0 . 1 76 86 92 81 80 83 80 82 89 82 gm. +0.008 +0.018 — o . 009 — 0.018 +0.018 +0.043 +0.065 +0.051 — 0.012 Table XVI. — Summary of Data on Rat XXXII, fed on Pure Casein (as the only Protein) and Salt Mixture III for 42 Days. — Daily Averages. Period i. — Dog Biscuit-Lard Diet. Date of experiment. 1909. Nov. I . 8. 15- 22. 1909. Nov. 29. Dec. 6. 13 1909. Dec. 20. 27- 1910. Jan. 3. Intake. Nitrogen output. Body- weight. Food. Nitrogen. Urine. gm. 180. gm. 5 N-utilization. N-balance. Faeces. : Total. gm. ?m. gm. 196.7 200 3 201 .4 9 3 7-4 9.2 o. 148 o. 120 0.145 0.099 o. 152 O. I I I 0.017 0.020 0.034 gm. p. ct. •m. o. 162 o . 1 1 9 0.145 Period 2. — Casein Diet i. 193 7 6 , 0. 1 13 0.095 0.018 0. 1 1 3 189.6 5-8 0. 109 0.089 0.020 0 . 1 09 191 .2 7 -i 0.133 0.093 0.024 0 . 1 1 7 Period 3. — Casein Diet 2. 193.8 199.3 6.1 7.6 0.153 0. 192 0. 1 13 0.154 204.7 7-8 0. 198 0. 160 0.015 o. 128 0.018 o . 1 72 0.020 o. 180 89 83 78 — 0.014 +O.OOI +0.007 84 82 82 0.000 0.000 +0.016 90 91 90 +0.025 +0.020 +0 018 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 35 Table XVII. — Summary of Data on Rat XXXIII, fed on Pure Casein (as the only Protein) and Salt Mixture I for 42 Days. — Daily Averages. Period i. — Dog Biscuit-Lard Diet. Date of experiment. 1909. Nov. 1 . 8. '5- 22. 1909. Dec. 13. . 20. . 27.. 1910. Jan. 3 . . Body- weight. Intake. Nitrogen output. Food. Nitrogen gm. I'rinc. Faeces. Total. gm. 206.0 220.2 226.3 220.2 gm. g)II. gm. gm. 8.4 8.6 8.2 O.I35 O.I37 O. 136 O.I38 O. I 16 O. 127 0.023 0.02 1 0.026 0. 161 0.137 0.153 N-utilization. p. Ct. 83 S5 206.3 7-5 0. 1 56 0.092 0.034 0. 126 219.5 8.8 0.223 0. 102 0.033 0.135 226.8 9 5 0.240 0. 165 0.040 0.205 233.0 9 4 0.235 0.168 0.01 5 J 0.183 78 85 83 94 N-balance. gm. — 0.026 0.000 — 0.017 +0.030 +0.088 +0.035 +0.052 Period 2. — Casein Diet 3. 1909. Nov. 29. . . . Dec. 6. . . . 205.5 199.2 5.6 6.9 0.077 0.095 0.086 0.071 0.023 0.042 0 . 1 09 0. 1 13 70 56 — 0.032 — 0.018 Period 3. — Casein Diet 4. Table XVIII. — Summary of Data on Rat XXXIV, fed on Pure Casein (as the only Protein) and Salt Mixture I for 42 Days. — Daily Averages. Period i. — Dog Biscuit-Lard Diet. Date of experiment. 1909. Nov. 8. 15... 22. . . Body- weight. Intake. Food. Nitrogen gm. gm. 157.0 170.4 169.3 gm. Nitrogen output. Urine. Faeces. Total. N-utilization. gm. gm. gm. p. ct. 9.3 0.157 8.2 0.137 o. 124 0.030 o. 154 0.092 0.030 0.122 Period 2. — Casein Diet 3. 81 78 N-balance. gm. + 0.003 — O.OI5 1909. Nov. 29. Dec. 6. 1909. Dec. 13. . . 20. . . 27... 1910. Jan. 3... 164.2 1575 5 9 5 9 0.081 0.081 0.067 0.064 0.017 0.024 0.084 0.088 Period 3. — Casein Diet 4. 172.5 8.0 0. 172 0.080 0.036 0. 1 16 181.8 7.8 0. 199 0. 108 0.039 0.147 198.5 9-4 0.236 0. 127 0.039 0. 166 206.2 9.0 0.226 0. 125 0.047 0. 172 79 70 79 80 83 — o . 003 — 0.007 +0.056 +0.052 +0.070 +0.054 36 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Table XIX. — Summary of Data on Rat XXXV, fed on Pure Casein (as the only Protein) and Salt Mixture I for 42 Days. — Daily Averages. Period i. — Dog Biscuit-Lard Diet. Intake. Date of experiment. Body- weight. Nitrogen output. 1909. Nov. 8. 15- 22. 1909. Nov. 29. Dec. 6. Food. gm. Nitrogen. Urine. gm. gm. gm. 1 16.5 122 .7 6. 5 o. 1 10 O. IOI 126.2 6.2 j 0.102 ! o. 074 Faces. Total. gm. gm. 0.020 0.017 o. 121 0.091 N-utilization. N-balance. p. Ct. 82 83 Period 2. — Casein Diet 3. 1 1 1 4 4 0 0 055 0 052 0 020 0 072 118 3 4 5 0 062 0 050 0 014 0 064 64 77 gm. — 001 1 +0.01 1 -0.017 - o . 002 1909. Dec. 13. 20. 27- 1910. Jan. 3 . Period 3. — Casein Diet 4. 127. 1 6.1 0. 126 0.060 0.028 0.088 142.9 8.8 0.225 0. 105 0.025 0. 130 1473 7-2 0.181 0. 1 16 0.029 0.145 152.5 6.7 0. 167 0. 108 0.026 0.134 78 89 84 84 +0.038 +0.095 +0.036 +0.033 Table XX. — Summary of Data on Rat XXXVII, fed on Pure Casein (as the only Protein) and Salt Mixture II for 42 Days. — Daily Averages. Period i. — Dog Biscuit-Lard Diet. Date of experiment. 1909. Nov. 8. '5- 22. Intake. Body- weight. FlM.d gm. gm. ■42.7 153.2 8.7 156.3 i 7-8 Nitrogen gm. o. 146 o. 128 Nitrogen output. Urine. Faeces. Total gm. gm. gm. N-utilization. N-balance. p. Ct. 0.1 15 0.023 O.091 O.026 O.138 o. 1 17 84 80 gm. +0.008 +0.01 1 Period 2. — Casein Diet 5. 1Q09. Nov. 29. Dec. 6. 136.7 136.0 4.6 5-7 0.070 0.088 0.072 0.071 0.024 0.034 0.096 o. 105 66 61 —0.026 — 0.017 Period 3. — Casein Diet 6. 1909. Dec. 13. 20. 27- 1910. Jan. 3. 135.5 5.6 0. 1 16 0.085 0.025 0. 1 10 1375 7-3 0. 184 0. 109 0.021 0. 130 140.8 6.2 0.159 0. 122 0.024 0. 146 143 3 5-9 0. 151 0. 121 0.021 0. 142 78 89 85 86 +0.006 +0.054 +0.013 +0.009 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 37 Table XXI. — Summary of Data on Rat XL, fed on Pure Casein (as the only Protein) and Salt Mixture II for 42 Days. — Daily Averages. Period i. — Dog Biscuit-Lard Diet. Date of experiment. Body- weight. I909. Nov. 10. ■5- 22. Intake. Food. Nitrogen gm. 150.O '35 4 147-3 gm. gm. 2-7 7-7 0.045 o. 130 Nitrogen output. Urine. gm. o. 106 O.O96 Faeces. gm. Total. gm. o . 007 0.113 o . 02 1 0.117 N-utilization. p. Ct. 84 84 N-balance. gm. -0.068 +0.013 1909. Nov. 29. Dec. 6. Period 2. — Casein Diet 5. 137 8 4 6 0 071 0 059 0 020 0 079 .* 5 6 3 0 100 0 068 0 033 0 101 82 67 —0.008 —O.OOI Period 3. — Casein Diet 6. 1909. Dec. 13. 20. 27- 1910. Jan. 3. 139 5 6.1 0. 129 0.088 0.027 0.115 130.0 4-3 0. 109 0.099 0.019 0.118 115. 6 3-4 0.087 0. 102 0.014 0. 1 16 1 10.0 4.0 0. 103 0. 105 0.017 0. 122 79 83 84 83 +0.014 — 0.009 —0.029 — 0.019 Table XXII. — Summary of Data on Rat XLII, fed on Pure Casein (as the only Protein) and Salt Mixture II for 42 Days. — Daily Averages. Period I. — Dog Biscuit-Lard Diet. 1909. Nov. 29. Dec. 6. Date of experiment. Body- weight. 1909. Nov. 10. . . . 15.... 22. . . . gm. I44.O 137-.8 I4I.4 Intake. Nitrogen output. Food. Nitrogen. Urine. Faeces. Total gm. gm. 4.9 6.3 l 0.082 I o . 1 06 gm. 0.090 0.069 N-utilization. N-balance. gm. gm. p. ct. 0.013 0.020 o. 103 0.089 84 81 Period 2. — Casein Diet 5. 130.2 130.6 3.8 5.6 0.059 0.086 0.062 0.072 0.019 0.031 0.081 0. 103 68 64 Period 2. — Casein Diet 6. gm. — o . 02 1 +0.017 -0.022 -0.017 1909. Dec. 13. 20. 27- 1910. Jan. 3. 126.4 4.8 0. 101 0.086 0.022 0. 108 128.4 5.0 0. 126 0. 105 0.021 0. 126 122.4 4-7 0. 120 0. 108 0.023 0. 131 1 16.2 4-4 0. 1 13 0. 109 0.025 0.134 78 83 81 78 —0.007 0.000 —0.01 1 — o . 02 1 38 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. £10 + 0.05 Chart XI. — Rats xxx, xxxi, and xxxn fed on pure Casein as the only protein and Salt mixture III for 42 days. Numbers on Body-weight line indicate the time at which each period began. ♦ 005 ■ Chart XII. — Rats xxxm, xxxiv, and xxxv fed on pure Casein as the only protein and Salt mixture I for 42 days. Numbers on Body-weight line indicate time at which each period began. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 39 ISO 130 ♦ 0.05 Chart XIII. — Rats xxxvir, xl. and xlii fed on pure Casein as the only protein and Salt Mixture n for 42 days. Numbers on Body-weight lines indicate time at which each period began. DISCUSSION. The essential differences in these series lasting 42 days involve the inorganic constituents of the diet. All of the animals selected, with two exceptions (xl and xlii)* gave a fairly satisfactory record during a three weeks preliminary trial, whereupon they were selected for the casein feeding. It will be seen that the most promising nutritive conditions were afforded by mixture 1, fed to rats xxxiii, xxxiv, and xxxv, in which the inorganic constituents were closely copied after Rohmann's successful ash mixture, with the addition of a little iron. During the first two weeks of casein feeding these three rats lost weight and nitrogen. This was caused by diarrhoea due to too great a proportion of inorganic salts in the food mixture. When this was reduced from 4.8 to 2.5 per cent a very rapid gain in weight and nitrogen at once took place. The superiority of the food mixture given to these three rats in contrast with the other two mixtures given to the other rats is plainly evident from comparison of these data. These experiments lasted 42 days and showed distinct gains in the nitrogen balance and also in body-weight during the period mentioned. f *These two animals were not pure white rats, but partly colored. They ate very poorly. We have gained the impression from observations on a large number of rats that these hybrid forms are not suited to our experimental needs and therefore have lately employed only the pure white races. fit may be noted that the apparently poorer utilization of nitrogen during some of the periods when little food was eaten is in part attributable to the fact that output of nitrogenous alimentary secretions continues despite the smaller intake of nitrogen. 40 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. THE CASEIN DIET — PROLONGED FEEDING TRIALS. The proportions of this food mixture (4, see page 33 ) thus tested were therefore employed in subsequent experiments as a typical "basal" ration. We record below the protocols of a few rats which for many weeks have been kept in good health on this dietary with pure casein as the sole protein. By an ill-advised plan these casein-fed rats were transferred dur- ing one month to cages containing sand, and the estimation of the nitrogenous excreta was omitted during this period. It was subse- quently discovered that the animals ate more or less sand during this time and we attribute the decline of weight and the subsequent death in some cases to a resulting damage of the alimentary organs, since sand was found in the faeces of some of the rats long afterwards and autopsy showed great congestion of the intestinal tract. Table XXIII. — Summary of Data on Rat 23, fed for 135 Days ox Pure Casein as the only Protein. — Diet 4, Page 33. — Daily Averages. Date of Body- weight. Intake. Nitrogen output. experiment Food. gm. Nitrogen. Urine. Fseces. Total. N-utilization. i\-oaia 1910. Feb. 7 gm. '93-7 209.3 222.0 220. 5 212.8 210.0 22 1 . 1 232.0 234 4 217.6 201 .8 211.3 219.0 225 .0 220.8 225.6 212.3 197.0 182.7 gm. gm. gm. gm. p. ct. gm. '4 21 28 Mar. 7 '4 28 Apr. 1 1 18 23 May 2 9 16 23 30 June 6 ■3 20 22 Dead 8.0 9-4 8.1 7.0 6.6 7-9 7-9 9.1 5-9 4-7 5-4 7-2 7-7 6.6 79 6.2 49 0.208 0.247 0.215 0.186 0.173 0.209 0.208 O. I 1 I O.IlS O. 108 O. 141 O.I45 O.I55 O.I47 O.026 O.041 O.039 O.03I O.024 O.029 O.030 0.137 0.159 0.147 0. 172 0. 169 0. 184 0.177 87 +0.07I 83 +O.088 82 +O.068 83 +O.OI4 86 +0.004 86 +0.025 86 .+0.031 0.185 0.197 0.168 0.201 0. 158 0. 126 O. I 12 O.I37 O.I3I O. I40 O.I55 O. I46 O.027 O.028 O.026 O.029 O.025 0.017 0.139 0. 165 0.157 0. 169 0. 180 0. 163 85 86 85 86 84 87 +0.046 +0.032 +0.01 1 +0.032 — 0.022 -0.037 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 41 Table XXIV. — Summary of Data on Rat 24, fed for 169 Days on Pure Casein AS THE ONLY PROTEIN. — DlET 4, PAGE 33- — DAILY AVERAGES. Intake. Nitrogen output. Date of Body- N-utilization. N-balanee experiment, weight. Food. Nitrogen. Urine. Faeces. Total. 1910. gm. gm. gm. gm. gm. gm. p. Ct. gm. Feb 7 "ni n >4 • 239.0 8.6 0.22() 0.145 O.023 0.168 90 + O.058 21 . . 244.0 IO. I O.266 0. 180 O.039 0.219 85 + O.047 28.. 239.0 9-4 O.248 0. 192 O.037 0.229 85 + 0.019 Mar. 7 . . 241 .0 8.2 O.2I7 0. 180 0.022 0.202 90 +O.OI5 14. . 242.0 8.8 O.234 0.181 O.032 0213 86 + 0.02I 28.. 259.6 9.6 O.254 0.195 0 . 03 I 0.226 88 + O.028 Apr. 11.. •■ 252.7 9.0 O.235 0. 176 O.032 0.208 86 + O.027 18 .. 235.8 218 8 6.5 4.6 4-5 5-8 7.0 25. . May 2 . . 9 • 16.. 205.2 214.8 216. 1 O.I78 0. 123 O.029 0. 152 84 + O.026 23. . . . 240.5 9-5 O.242 0. 150 0.030 0. 180 88 + O.062 30. 252.4 9.1 O.233 0. 163 0 . 02.3 0.186 90 + O.047 June 6. . •■ 257.1 8.8 0.226 0. 161 O.O24 0. 185 89 + O.04I 13. . .. 248.7 7-3 O 188 0. 163 0.02 1 0. 184 89 + 0.004 20. . 243.6 6.7 O. I72 0.155 O.OlS 0.173 90 — O . OO I 27. . 246.6 6.4 O. 164 0. 142 O.OI4 0. 1 56 91 + O.O08 July 4.. 243 . 5 7-3 O. 184 0. 152 O.OlS 0. 170 90 + 0.014 11.. 244.1 8.2 O.209 0. 166 0.031 0. 197 85 + O.OI2 18.. 230.0 8.6 0.220 0.133 O.049 0. 182 78 + O.038 25.. 1 69 . 1 1.6 O.O42 0. 103 O.O37 0 . 1 40 12 — 0 . 098 26. . 157.3 Chlorof'd... Table XXV. — Summary of Data on Rat 25, fed for 176 Days on Pure Casein as the only Protein. — Diet 4, Page 33- — Daily Averages. Intake. Nitrogen output. Date of experiment. Body- weight. N-utilization. N-balance. Food. Nitrogen. L:rine. Faces. Total. 1910 Feb: Mar. Apr. May June July Aug. Dead 0. gm. 7 ■ 177.0 14- 189.0 21 . 184.6 28. 199.5 7 ■ .. 198.3 U 199.2 28. 1 99 0 1 1 . 198.2 18. 200.6 25- .. 187.2 2. .. .67.3 9 ■ • 1 79 2 16. .. 181. 0 23- .. 186.0 30. 190.5 6. 196. 1 13 191 .0 20. 185.0 27- 185.0 4- 1 1 . 177.6 165.8 18. 153-5 25. 140.5 1 . 1 10. 0 2. 107.0 gm. gm. gm. gm. gm. p. ct. gm. 0.252 0.207 0.245 0.238 0.21 0 0.2I0 o. 196 o. 170 o. 164 o . 1 76 0.175 o. 181 o. 176 0.159 0.034 0.033 0.021 0.033 0.029 0.026 0.031 0.204 0.197 0.197 0.208 0.210 0.202 o. 190 87 84 9' 86 86 88 84 +0.048 +0.010 +0.048 +0.030 0.000 +0.008 +0 . 006 o. 161 o. 180 0.185 0.219 0.181 o. 160 0.157 o. 125 0.095 0.078 0.084 0.052 o. 125 o . 1 26 0.138 0.149 0.154 o. 146 o. 124 o. 106 0.098 0.076 0.090 0.086 0.020 0.029 0.024 0.022 0.024 0.023 0.023 0.025 0.015 0.014 0.017 0.018 0.145 o. 155 o. 162 0.171 o. 178 o. 169 o. 147 o. 131 o. 1 13 0.090 o. 107 o. 104 88 84 87 90 86 85 80 84 82 80 65 +0.016 +0.025 +0.023 +0.048 +0.003 — o . 009 +0.010 — o . 006 — 0.018 — 0.012 — 0.023 — 0.052 220 200 '+0.05 - 20 40 60 80 100 120 Days Chart XIV. — Rat 23 fed 135 days on pure Casein as the only protein, Diet 4, p. 33. 4-0 240 220 200 70 50 30 ♦0.05 -0-10 \ Bod 1 weight / -y Food f aten 160 V 1 1 1 — '16C 1 1 1 1 1 / ^ / > / / \ V * " — -, - / / \ \ \ s / / r 0- 1 kills d ^ ^ f * Nitroge r 1 b; lane e 1 1 1 , 1 1 1 T-Tl 1 1 20 40 60 80 100 120 140 160 180 Days Chart XV. — Rat 24 fed 169 days on pure Casein as the only protein, Diet 4, p. 33. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 43 190 170 150 •0.05 -0.10 20 100 Days Chart XVI. — Rat 25 fed 176 days on pure Casein as the only protein, Diet 4, p. 33. These experiments, with rats 23, 24 and 25, extending over 135-160 days without noteworthy alterations in the weight and with a very large gain of nitrogen, are, as far as we are aware, the most successful recorded attempts at artificial nutrition with a constant mixture of pure food- stuffs including a single protein. CASEIN AND VEGETABLE PROTEIN DIET. Without reporting here the numerous trials and failures to replace part or all of the casein by other proteins, a few data from our records may throw light upon the difficulties thus encountered. The proportion of the nutrients is the same in these experiments, one- third or more of the casein being replaced by the proteins indicated. A casual inspection of the succeeding pages shows that the failure to eat is frequently sufficient to account for the failure to maintain body-weight and tissue. The animals lost weight to the extent of their fat content and then speedily succumbed with indi- cations of inanition rather than any specific pathological metabolism. Those rats which ate less than 40 grams of the mixed food were unable to maintain their nutritive equilibrium. A further evidence that no permanent defect is induced by the character of the diet is found in the observation that a change to a mixed diet of seeds and vegetables often brought speedy realimentation and recovery. 44 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Despite numerous failures with other proteins than casein, sev- eral experiments have already been continued long enough to hold out promise of the possibility of ultimate success. In these the ani- mals were fed on a mixture in which the casein content was gradually decreased and then entirely replaced by vegetable protein.* Rat i 6. — This rat, after a preliminary feeding on dog biscuit- lard diet, was fed through Period 2 with a mixture containing casein 12, excelsin 6, sugar 15, starch 29.5, lard 30, agar 5 and salt mixture I 2.5 per cent. This contained 2.74 per cent of nitrogen. During Period 3 the proportions of excelsin and casein were 9 per cent, the rest of the mixture remaining the same and containing 2.75 per cent of nitrogen. During Period 4 the proportion of excelsin was made 12 and casein 6 per cent, and the diet (otherwise as before) contained 2.8 per cent of nitrogen. Throughout Period 5 the casein was wholly replaced by excelsin, but otherwise the mixture was that above given. The nitrogen content of this diet was 2.94 per cent. Table XXVI. — Summary of Data on Rat i6, fed for 210 Days on Mixtures Containing Casein and Excelsin as the Only Proteins. — Daily Averages. Period i, 28 Days. — Dog Biscuit-Lard Diet. Intake. Date of experiment. 1910. Jan. 24. 3i Feb. 7. '4 21 . Body- weight. gm. 238.8 235.2 225.O 226.8 219.5 Nitrogen output. Food. Nitrogen. Urine. Faeces. gm. gm. gm. gm. Total N-utilization. N-balance. gm. p. Ct. gm. 8.4 6.5 8.5 5-7 o. 132 o. 106 o. 163 O. I 10 o. 103 0.084 o. 102 0.088 0.025 0.128 0.024 o. 108 o . 029 o . 1 3 1 0.018 0.106 81 77 82 84 +0.004 — o . 002 +0.032 +0.004 Period 2, 42 Days. — Casein 12 per cent, Excelsin 6 per cent. 1910. Feb. 28. Mar. 7. '4 21 . 28. 4 Apr. 227.8 6.1 0. 164 0. 141 0.017 0.158 230.0 7.8 0.211 0. 150 0.033 0.183 225 .6 7.0 0.189 0.154 0.029 0.18? 22< 6.6 0.180 0. 128 0.018 0. 146 234.0 7 4 0.205 0. 1 56 0.024 0 . 1 80 227.8 6.9 0. 191 0. 148 0.026 0.174 90 84 85 90 88 86 +0.006 +0.028 +0.006 +0.034 +0.025 +0.017 Period 3, 35 Days. — Casein 9 per cent, Excelsin 9 per cent. 1910. Apr. 1 1 18. 25- May 2 . 9- 225.5 224.0 229.5 2342 240.0 7-5 7.0 7.6 8.5 8.7 0.207 0.193 0.21 1 0.234 0.238 o. 160 o. 180 o. 172 o. 169 0.175 0.031 0.021 0.023 0.027 0.025 o. 191 0.201 0.195 o. 196 0.200 85 89 89 88 89 +0.016 — o . 008 +0.016 +0.038 +0.038 *For a description of the vegetable proteins used see Osborne; Die Pflanzenpro- teine, Ergebnisse der Physiologie, 1910, x, p. 47. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 45 Table XXVI. — Summary of Data on Rat 16, fed for 210 Days on Mixtures Con- taining Casein and Excelsin as the Only Proteins. — Daily Averages. — Cont'd. Period 4, 14 Days. — Casein 6 per cent, Excelsin 12 per cent. Intake. Date of exp<_riment. 1 9 1 O. May 16. 23- 1910. May 30 June July Aug. Sept. Dead Body- weight. Nitrogen output. N-utiJization. Food. Nitro-ien. Urine. Faeces. Total. gm. 232. 1 233-5 gm. 7-5 7-2 gm. 0.210 0.201 gm. o. 167 0.177 gm. 0.028 0.020 gm. 0.195 0.197 p. a. 87 90 Period 5, 119 Days. — Excelsin 18 per cent. 235.6 235.0 231.8 228 218 216 216 220 207 214 220 212.7 219.4 216.3 209.5 197.0 160 .3 7-2 0.215 8.1 0.238 7-2 0.211 7 3 0.215 4 7 0.139 7-4 0.217 6.3 0.186 7 6 0.225 6.6 0.193 7-3 0.214 7-9 0.23 1 8.2 0.242 8-3 0.244 8.6 0.252 8.9 0.262 7-9 0.234 t>. 1 0. 180 165 192 192 '73 141 150 155 158 181 '59 141 196 170 '59 212 229 223 0.024 0.020 0.022 0.026 0.01 5 0.025 0.020 0.025 0.027 0.017 0.016 0.025 0.020 0.025 0.037 0.022 0.039 o. 189 0.212 0.214 o. 199 o. 156 0.175 0.175 0.183 0.208 o. 176 0.157 0.221 o. 190 o. 184 o . 249 0.251 0.262 89 92 90 88 89 88 89 89 86 92 93 90 92 90 86 9' 78 N-balance. gm. +0.015 +0.004 +0.024 +0.026 —0.003 +0.016 — 0.017 +0.042 +0.01 1 +0.042 — 0.015 +0.038 +0.074 +0.021 +0.054 +0.068 +0.013 — 0.017 — 0.082 dead 100 120 wo Days 180 200 220 240 Chart XVII. — Rat 16 fed 210 days on Casein and Excelsin as the only proteins. See p. 44. Numbers on Body-weight line indicate time at which each period began. 46 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Rat 70. — This rat was fed during Period 1 with a mixture con- taining casein 9, pea legumin 9, sugar 15, starch 29.5, lard 30, agar 5, and salt mixture No. 1, containing 2.5 per cent. This contained 2.75 per cent of nitrogen. During Period 2, the diet contained 18 per cent of legumin as the only protein, the proportion of the other con- stituents being unchanged. The nitrogen content was 2.97 per cent. 160 140 120 60 40 20 ♦ 0.05 - CO E U o -0.10 1 2 Body weight 1 dead / / / / 1 1 \ \ \ \ \ \ / / / -I \ \ \ / / / / > Food eaten \ I 1 1 1 t 1 - L= , 1 t 1 1 ! Nitrogen balance 20 4-0 60 80 100 Days 120 14-0 160 Chart XVIII. — Rat 70, fed 160 days on Casein and Pea-Legutnin as the only proteins. See p. 47. Numbers on Body-weight line indicate time at which each period began. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 47 Table XXVII. — Summary of Data on Rat 70, fed for 160 Days on Mixtures Con- taining Casein and Pea Legumin as the Only Proteins.— Daily Averages. Period i, 69 Days. — Casein 9 per cent, Pea Legumin 9 per cent. Date of experiment. I9IO. Apr. 5 1 1 18 25 May 2 9 16 23 30 6 '3 June 1910. June 20 July Aug. Sept. Dead 27- 4 1 1 18. 25- I 8. 15 22 . 29. 5 12 Intake. Nitrogen output. Body- weight. gm. I7I-7 160.O 157-8 162.5 151 .0 141 .2 128.6 144-5 152.8 157-8 160.6 N-utilization. Food. Nitrogen. Urine. Faeces. Total. gm. gm. gm. gm. gm. p. ct. 9.6 5-9 8.9 5-9 4-4 4.2 5-3 6.8 6.9 5-3 0.267 o. 164 0.249 164 122 1 1 5 145 0.186 0.187 0.143 159 181 179 153 118 0.075 o. 130 o. 131 0.139 o. 124 0.026 0.015 0.016 0.016 O.OI I 0.017 0.009 0.022 0.021 0.015 .85 196 195 169 129 0.092 0.139 0.153 o. 160 0.139 90 91 94 90 9' 85 94 88 89 90 Period 2, 91 Days. — Pea Legumin 18 per cent. 161 3 5-2 O.I55 0. 141 0.013 0.154 160 0 5-3 0.158 0. 128 0.020 0. 148 152 3 5-3 0. 156 0.134 0.019 0.153 '53 8 5-5 0. 163 0. 125 0.017 0. 142 154 0 4-7 0. 140 0.096 0.014 0. 1 10 151 5 5-< 0. 151 0. 107 0.019 0. 126 149 0 5.0 0. 1 50 0 . iii 0.015 0. 126 148 7 5-5 0. 163 0. 103 0. 1 16 0. 1 19 145 3 6.1 0. 180 0. 140 0.018 0.158 145 5 6.0 0.179 0.133 0.017 0. 1 50 142 6 5-7 0.168 0. 1 16 0.017 0. 133 .38 5 6.0 0. 176 0. 135 0.025 0. 160 102 0 1 .2 0.034 0.082 0.013 0.095 92 87 88 90 90 87 90 90 90 91 90 86 62 N-balance. gm. -f O.082 — O.032 + O.O54 — 0.005 — o . 007 + O.023 + O.O06 + O.033 + O.027 + O.OO4 +O.OOI + O.OIO + O.OO3 + 0.02I +O.030 + O.025 + O.024 + O.O44 + 0.022 + O.029 + O.035 +O.OI6 — o . 06 1 Rat 7 1 . — This rat was fed during Period 1 with a mixture con- taining casein 12, glutenin 6, sugar 15, starch 29.5, lard 30, agar 5, salt mixture I 2.5 per cent. This contained 2.69 per cent nitrogen. During Period 2 the diet contained casein 12, glutenin 6, sugar 15, starch 24.5, lard 35, agar 5, salt mixture I 2.5 per cent, with 2.68 per cent of nitrogen. During Period 3 the diet consisted of glutenin 16.4, sugar 13.6, starch 22.3, lard 40.9, agar 4.5, salt mixture I 2.3 per cent, and contained 2.60 per cent of nitrogen. Throughout Period 4, the diet contained glutenin 18, sugar 15, starch 14.5, lard 45, agar 5 and salt mixture I 2.5 per cent. This diet contained 2.83 per cent of nitrogen which belonged wholly to glutenin. Rat 71 is still alive at the present writing after 217 days of exclusive diet containing glutenin as its only protein, and 286 days including the casein and glutenin period. 4§ FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Table XXVIII. — Summary of Data on Rat 71, fed for 244 Days on Mixtures Containing Casein and Glutenin as the only Proteins. — Daily Averages. Period i, 13 E AYS.— Casein 12 PER cent, Glutenin 6 per cent. Dat e of Body- ment. weight. Intake. Nitrogen output. N-utilization. N-balance.* experi Food. gm. Nitrogen. Urine. Faeces. Total. '9 0. gm. gm. gm. gm. gm. p. Cl. gm. Apr. 5 . . . 2<;7. <; 18. . Per 210.5 6.8 0. 184 O. 183 0.030 0.213 84 — O.O29 iod 2, 56 Days. — Casein 12 PER . cent, Glutenin 6 per cent. 19 10. Apr. 25. . 238.7 1 1 .0 O.297 O.249 0 . 03 1 0.280 90 + O.OI7 May 2 . . 225.5 4.6 0 . 124 O.187 0.013 0.200 90 — 0 . 076 9 • 222 .2 9.6 O.259 O. 163 0.017 0. 180 93 + O.O79 16. . 227. 1 6.1 O. 166 O. 130 0.014 0.144 92 +0.022 23.. 234.5 7-3 O. 196 O. 139 0.020 0.159 90 + O.O37 30.. 242.3 8.2 O.218 O. 1 52 0.025 0 . 1 77 89 + 0.04I June 6.. 256. 1 10.9 O.290 O. 196 0.040 0.236 86 + O.O54 13. 255.0 6.5 O.I73 O. 169 0.033 0.202 81 — O.O29 Peri od 3, 35 Days . — Glutenin i 5.4 PER CENT. 19 10. June 20. . . . 248.5 6.0 O.I55 0.134 0.026 0. 160 83 — O.OO5 27 .. 252.6 7 4 O.I9O 0.133 0.035 0.168 82 + 0.022 July 4 ■ 238.1 4-5 0.II5 0. 105 0.016 O . I 2 I 86 — O . O06 11.. 240.5 5-9 O.I53 0.091 0.029 O. 120 81 + O.O33 18.. 236.5 5.0 O. 132 0.087 0.017 O. I04 87 + 0.028 Peri od 4, 182 Dai 's. — Gli JTENIN l8 PER CENT. 1 9 10. July 25.. 241 .0 6.6 O.I74 0. 129 0.022 O. I 51 87 + O.023 Aug. 1 . . 261.8 8.5 O.24O 0. 140 0.02I O. l6l 9' + O.O79 8. . 259.2 8.0 O.226 0.138 O.O3O O.168 87 + O.058 15.. •■ 255.1 6.8 O.I94 0.135 O.OI7 O. 152 9' + O.O42 22. . 247.5 7.0 O.I95 0. 107 0.020 O. 127 90 + O.068 29.. 2450 6.7 0. 190 0.075 O.OIO 0.091 92 + O.099 Sept. 5-- 241.5 7-5 0.212 0. 132 O.O3O O. 162 86 + O.050 12. . 235.2 6.3 0.179 0. 171 O.OI4 O.185 92 — 0 . 006 19.. 244.8 9 3 0.263 0.209 0.040 O.249 85 +O.OI4 26. . 240.0 8-3 0.237 0. 191 OOI9 0.2I0 92 + O.027 Oct. 3 ■ .. 238.5 7-7 0.218 0.186 O.OI9 O.205 9' + O.0I3 10. . 239.2 6.7 0. 191 0.151 O.028 0 . 1 79 85 + O.OI2 17.. 228.0 6.5 0.186 0. 163 O.O23 O.I86 88 O.OOO 24 . 223.5 6.9 0. 198 0. 162 O.O37 0. 199 81 — O . OO I 3i ■ .. 238.7 9-4 0.269 0.202 O.037 O.239 86 + 0.030 Nov. 7- ■ 253.4 9.8 0.280 0.188 O.O38 0.226 86 + O.O54 '4 248.0 7.6 0.217 0 . 1 76 O.O33 O.209 85 + 0.008 21 . . 250.8 7.8 0.221 0.158 O.O45 O.203 80 + O.OI8 28. . 254.0 7-4 0.2 1 1 0. 144 O.O37 O.lSl 82 + 0.030 Dec. 5- 253.6 7 9 0.226 0.174 O.O39 O.213 83 + O.OI3 12 . . 254.0 9.0 0.252 0.208 O.O4O O.248 84 + 0 004 19. 262 . 5 1 1 .0 0. 301 0.214 O.064 O.278 79 +0.023 26.. 10.8 0.296 0.220 O.O52 O.272 82 +0.024 191 1. Jan. 2 .. 279.8 11. 7 0. 32 I 0.222 O.O49 O.27I 85 +0.050 9 . ■ 273.5 8.7 0.237 0.208 0 . 03 5 0.243 85 — 0 . 006 16.. .. 275.5 11.5 0.317 0.270 O.O38 O.308 88 +0.009 "In respect to the relatively high nitrogen balance compare last foot-note on page 7. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 49 ♦ 0.10 Chart XIX. — Rat 71 fed 286 days on Casein and Glutenin as the only proteins. See p. 48. Numbers on Body-weight line indicate time at which each period began. 220 200 180 60 4-0 20 + 0.05 - 10 E (D o -0.10 /~~^^i--^\ \ Body weight 3^--' \2/ < \ \ \ \ Food eaten 160 . \ v dead \ \ ^ -"\ \ X — - "»— *■ X \ V^ ,-— -—. "" 1 V \ \ 1 X ■J V \ \ Nitrogen balance \ 1 __ c , 20 40 60 80 Days 100 120 14-0 Chart XX. — Rat 72 fed 141 days on Casein and Glutenin as the only proteins. Numbers on Body-weight line indicate time at which each period began. 5° FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. Rat 72. — This rat was fed with the same mixtures as those fed to rat 71 during the corresponding periods. Table XXIX. — Summary of Data on Rat 72, fed for 141 Days on Mixtures Containing Casein and Glutenin as the only Proteins. — Daily Averages. Period i , 13 Days. Casein [2, Glutenin 6 PER CENT. Intake. Nitrogen output. Date of experiment. Body- N-utilization. N-balance. weight. Food. Nitrogen. Urine. Faeces. Total. 1910. gm. gm. gm. gm. gm. gm. p. Ct. gm. Apr. 5... 18.... 209. 5 195-7 8.7 0.233 ° '45 °-°45 0. 190 8l +O.O43 Perio d 2, 56 Days.— -Casein 12 per cent, Glutenin 6 per cent. 1910. Apr. 25. . . . 205.0 10.4 0.281 O.I86 0.053 0.239 8l + O.O42 May 2 . . . . 193 3 8 7 0.235 O.I47 0.049 0. 196 79 +O.O39 9.... 186.4 5 5 0. 150 0. 126 0.012 0.138 92 + O.OI2 16.... 181. 0 4 0 0. 123 O. I 14 0.009 0. 123 93 O.OOO 23.... 187.2 6 4 0. 170 0. 1 2b 0.024 0. 150 86 + 0.020 30... 196.4 7 2 0. 192 0. 120 0.038 0.158 80 + O.O34 June 6. . . . 198.0 6 9 0.214 O. 123 0.025 0. 148 88 + O.066 „.... 202.8 7 4 0. 196 O. 131 0.019 0. 1 50 90 + O.O46 Period 3, 4 2 Days. — Glutenin 16.4 ter CENT. 1910. June 20. . . . 205 .0 6.9 0. 170 O. I IO 0.028 0.138 84 +O.O38 27.... 213.0 7-3 0.187 O. I 12 0.045 0.157 76 + 0.030 Ju.y 4... 216. 1 6.3 0. 162 O.O97 0.032 0. 129 80 + O.O33 11.... 213.0 5.8 0. 152 O.O92 0.029 0. 121 8l + O.03I 18.... 217.7 6.2 0. 164 O.085 0.035 0. 120 79 + O.O44 25.... 214.9 6.3 0. 166 O.O72 0.032 0. 104 81 + O.062 Period 4, 30 Days. — Glutenin 18 per < ;ent. 1910. Aug. 1 217.3 6. 1 0. 170 O.O9O 0.025 0 . 1 1 5 85 + O.O55 8. ... 207.7 5.8 0. 165 O.O78 0.025 0. 103 85 + O.062 15.... 202.5 5-9 0. 166 O.O99 0.016 0. 1 1 5 90 +0.05I 22. . . . 165.0 2.9 O . 08 ! O.O79 0.037 0. 116 54 -O.O35 24.... 1575 Dead Rat 73. — This rat was fed during Period 1 on a mixture con- taining casein 12, zein 6, sugar 15, starch 29.5, lard 30, agar 5, salt mixture I 2.5 per cent, and nitrogen 2.38 per cent. In order to secure good utilization of the zein it was necessary to hydrate it by incor- porating in the food 10 cc. of water per 100 grams of the mixture. The nitrogen content of the different batches therefore varied from 2.30 to 2.48 per cent. The actual nitrogen content of each batch fed was used in calculating the nitrogen balance which is given in table XXX. During Period 2, the diet contained pea legumin 18, FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 51 sugar 15, starch 29.5, lard 30, agar 5, salt mixture I 2.5, and nitrogen 2.97 per cent. Table XXX. — Summary of Data on Rat 73, fed for 181 Days on a Mixture Containing Casein and Zein as the Only Proteins and for 53 Days on one containing Pea Legumin as the sole Protein. —Daily Averages. Period i, 181 Days. Casein 12 percent, Zein 6 per cent. Intake. Nitrogen output. Date of Body- N-utilization. N-balanee. experiment. weight. Food. Nitrogen. Urine. Faeces. Total. 1910. gm. gm. gm. gm. gm. gm. p. Ct. gm. Anr ? 179.0 174.0 1 1 . 3 9 0.092 0. 177 0.024 0.201 74 — 0 . 1 09 18 165 .0 7 5 0. 177 0. 171 0.018 0. 189 90 — O.OI2 25- 167.0 8 0 0. 189 0. 163 0.013 0. 176 93 + O.OI3 May 2 . 156.0 5 9 0.143 0.147 0.013 0. 160 9" — O.OI7 9 140.0 4 0 0.098 0. 123 0 . 008 0. 131 92 -O.033 16. 127.8 3 2 0 . 080 0 . 1 07 0.009 0. 1 16 89 — O.036 23 ■37 4 5 2 0. 123 0. 1 1 1 0.006 0.117 95 + O.O06 30. 126. 1 3 1 0 . 072 0 . 084 0.007 0.091 9° — 0.019 June 6 126. 5 4 4 0 . 1 02 0 . 1 08 0.005 0.113 95 — O.OI 1 13 129.5 4 8 0.112 0 . 09 1 0.008 0.099 93 + O.OI3 20. 128.2 4 0 0.094 0.084 0.006 0.090 94 + 0.004 27. 128.7 3 7 0 . 086 0 . 075 0.004 0.079 95 +0.007 July 4 123.3 3 4 0.080 O.081 0.006 0.087 92 — 0.007 I ! . 123.3 3 9 0.093 O.080 0.007 0.087 92 + 0.006 18. 125.0 3 7 0.088 O.064 0.007 0.071 92 + 0.017 25- 118.3 3 6 0.084 O.076 0.005 0.081 94 + 0.003 Aug. 1 . 119. 8 4 1 0.094 0.063 0.006 0.069 94 + O.025 8. 124.9 4 9 0.1 12 0.075 0.005 0.080 96 + O.032 "5- 127.5 5 0 0.133 0.092 0.009 0. 101 93 + O.032 22. 132.3 2 8 0.070 0. 105 0.012 0. 1 17 83 -O.O47 29. 135.2 8 1 0 . 200 0 . 082 0.01 1 0.093 94 + 0. 107 Sept. 5. 131.5 5 4 0. 134 0.096 0.011 0. 107 92 + O.027 12. 1325 5 1 0. 126 0. 1 10 0.007 0. 1 17 94 + 0.009 19. 138.2 6 1 0.151 0. 128 0.014 0. 142 91 + O.OO9 26. 142.8 7 1 0. 170 0. 139 0.017 0. 156 90 + 0.014 Oct. 3. 143.6 7 1 0. 167 0. 144 0.01 5 0.159 9' + O.O08 Perio d 2, 53 Days. — Pea I vEGUMIN 18 PEF . CENT. 1 9 10. Oct. 10. 136.3 5-2 0.153 O.I68 0.036 0.204 76 — O.05I >7- 127 5 5 1 0. 150 O. 165 0.020 0.185 87 -O.035 24. 121 1 4 6 0. 136 O. 142 0.018 0. 160 87 — O.024 3' 1 10 / 4 3 0. 129 0. 133 0.023 0. 156 82 — O.027 Nov. 7. 107 2 4 6 0.135 0. 132 0.019 0. 151 86 — O.Ol6 14 99 0 3 9 0. 1 1 5 O. 121 0.012 0.133 90 — O.Ol8 21 . 97 5 4 1 0. 12 1 O.IlS 0.01 1 0. 129 9' — 0 . 008 29. 70 0 These observations have thus shown the sufficiency of the arti- ficial dietaries to maintain full-grown small animals for long periods of time (from 50 to 286 days) in nutritive equilibrium. In many experiments, such as Nos. 11, 14, 16, and 72, here reported, the animals died rather suddenly, without any previous period of notable 52 FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. -oio =J dead 240 Chart XXI. — Rat 73 fed 181 days on Casein and Zein, and for 53 days on Pea-Legumin as the only proteins. See p. 50. Numbers on Body-weight line indicate time at which each period began. decline sufficient to explain their ultimate death. This fact strongly suggests that death was not necessarily attributable to any primary nutritive defect, but rather to incidental causes. Occasional fatali- ties will not surprise those who have experienced the difficulty of protecting a large number of rats, under the conditions noted, from the appearance of infectious or parasitic maladies which may become fatal. In the same surroundings sudden death has also come to not a few of our animals on ordinary mixed diets consisting of seeds and vegetables. In some cases obvious causes were revealed at autopsy ; but systematic post-mortem examinations have not been attempted. With young rats fed similarly we have succeeded in maintaining weight, although with little if any growth. In our preliminary studies of growing animals the food intake was not determined with appropriate care to correlate our findings with the altered curves of growth; hence the insufficient diet rather than any chemical deficiency may have been a possible cause of arrested development. Rats of 30 grams initial weight have been kept by us for many days without gaining weight when fed with a mixture containing a single protein ; with desiccated milk in the food they subsequently attained a perfectly normal growth. Despite the obstacles encountered we are inspired to the belief that with modifications in the feeding suggested by our first year's experiments still further progress can be made. Meanwhile further conclusions respecting the inadequacy of the individual proteins for nutritive functions are not justified. FEEDING EXPERIMENTS WITH ISOLATED FOOD SUBSTANCES. 53 SUMMARY The problems of nutrition have been reviewed in this paper in the light of the newer knowledge of the chemical structure of the proteins. The possibilities of protein synthesis in animals and the conditions which this postulates ; the significance of the availibility, palatability, and physical texture of the food-intake; the suggested role of various accessories — inorganic salts, lipoids, etc. ; the distinc- tion between the nutritive demands during the period of growth and those of later adult life, are brought within the range of discussion. The literature on experiments in which isolated food-substances have been fed to animals is discussed in some detail, with a critical con- sideration of some of the essential conditions of investigation which are demanded in successful research in this direction. The methods of metabolism study with white rats used in this research are described and illustrated. Control feeding trials showed that the animals can be maintained in nutritive equilibrium and health for periods of many months under the conditions of experi- ment adopted. The failure to eat sufficient food is indicated as a cause for the unsuccessful termination of numerous experiments. The facts presented exclude the probability that monotony of diet is an insurmountable obstacle to nutritive success. Numerous experiments are reported in which casein formed the sole nitrogenous constituent of the dietary. In this connection it is shown that the make-up of the inorganic constituents of the diet ex- ercises an influential effect on the nutritive efficiency of the dietary. From the experience thus gained a "basal" ration was constructed on which rats were kept many months in good health. Some of these experiments in which the animals exhibited no noteworthy altera- tions in weight and showed a good gain in nitrogen are, as far as the authors are aware, the most successful recorded attempts at artificial nutrition with a constant mixture of pure food-stuffs, containing only a single protein. Satisfactory experience also followed the gradual complete substitution of the casein by other proteins, one animal continuing more than 217 days on a diet in which the sole protein was glutenin. With young rats it has been possible to maintain weight with dietaries like those just mentioned, although with little if any growth. The limitations of the method are discussed and plans for continued investigation indicated. January 191 i. FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. BY THOMAS B. OSBORNE and LAFAYETTE B. MENDEL, With the Co-operation of EDNA L. FERRY. (From the Laboratories of the Connecticut Agricultural Experiment Station and the Sheffield Laboratory of Physiological Chemistry of Yale University.) WASHINGTON, D. C. Published by the Carnegie Institution of Washington 191 1 CARNEGIE INSTITUTION OF WASHINGTON Publication No. 156, Part II. PRESS OP GIBSON BROS. WASHINGTON, D. C. TABLE OF CONTENTS. PAGE. Introduction 55 Influence of various conditions on nutrition of white rats 55 Effect of long caging 55 Monotony of diet 56 Palatability of focd 56 Physical texture of the food 56 Digestibility of the food 56 Need of " roughage " 57 Inorganic constituents of the food 57 Effect of extraneous and accidental factors 58 Earlier experiences of the authors 59 Prolonged maintenance on isolated food substances 59 Relation of nitrogen balance to weight of rat 60 Changes in the method of caging and feeding 60 Alimentary bacteria and nutrition 61 Addition of faeces to the diet 61 Nutrition and growth 63 Normal rate of growth of male and female white rat 63 Normal growth as influenced by nutrition 64 Relation of energy supply to growth 65 Maintenance versus growth 65 Relation of protein to growth 66 Growth with insufficient food supply 66 Experiments of Waters on cattle 67 Experiments of Aron on dogs 68 Measurements of poorly nourished children 70 Suspension of growth on a maintenance diet 71 Detailed measurement of stunted rats 72 Underfeeding contrasted with a maintenance diet 74 Effect of stunting on the growth impulse 75 Realimentation of stunted rats 77 Disproportionate growth 77 Effect of partial starvation on body-weight 77 Effect of partial starvation on nervous system 78 Comparison of milk and mixed diet 79 Preparation of "protein-free" milk 79 Experiments with isolated proteins and "protein-free milk" 82 Critique of the non-protein factors in the diet 82 Adequate and inadequate proteins 83 Discussion of the results and their bearings 84 The charts and their explanations 86 Index of charts with reference to food-mixtures and proteins fed 86 in FEEDING EXPERIMENTS WITH ISOLATED FOOD- SUBSTANCES. PART II. INTRODUCTION. In Publication 156 of the Carnegie Institution of Washington* we have discussed some of the problems of nutrition which have been raised by the newer investigations in the field of protein chem- istry. The literature bearing on the feeding of isolated proteins was there reviewed in some detail, together with critical considerations of previously available experimental data. We described a plan for the study of metabolism and illustrated a method of investigation in which white rats were the experimental animals. For the details involved, our earlier paper must be consulted. A few protocols were there presented to show that the outlined mode of investigation offered a promising means for attacking certain questions in the field of nutrition. INFLUENCE OF VARIOUS CONDITIONS ON NUTRITION OF WHITE RATS. Numerous contingencies may arise to modify or vitiate the re- sults of experiments in which animals are kept in cages and fed upon artificially prepared mixtures of isolated food-stuffs, quite independ- ent of the factors inherent in the food-stuffs themselves or the com- binations in which they are exhibited. Among these possibilities, the caging itself, involving continued restraint and limited opportunity for exercise, suggests an unfavorable environment. This factor can at length be disposed of. Donaldson has concluded, from the best data obtainable, that "the three-year-old white rat is very old, and is justly comparable to a man of 90 years. "f Rats have been kept in our cages in appar- ent good health and without difficulty during periods of more than 14 months — a very considerable part of the span of life in these animals (cf. Charts XXIII, XXIX, XXX). *Feeding experiments with isolated food-substances, by Thomas B. Osborne and Lafayette B. Mendel, with the eo-operation of Edna L. Ferry. 191 1. Pp. 53. fH. H. Donaldson: A comparison of the white rat with man in respect to the growth of the entire body. Boas Memorial Volume, New York, 1906, p. 6. 55 56 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Monotony of diet has been urged as an obstacle to success where the same food mixtures are daily furnished^ without change over long periods of time. Very closely associated with this is the question of the palatability of the diet. The two factors need, however, to be distinguished. The palatability of the diet has, perhaps, been over- emphasized in recent years in its bearing on the real nutritive value of foods. It applies primarily to the individual with highly organized nervous system and psychical functions. The quality found in foods which are unpalatable because they disgust or nauseate is something positive; the negative property of lack of palatability, i. c, absence of stimulating taste, etc., is not necessarily a serious obstacle. In any event the palatability of the diet is difficult to determine or regulate and in attempting to control it experimentally in animals physi- ologists have been guided very largely by anthropomorphistic con- siderations. We have now gathered observations which lead us to dismiss the idea that monotony per se leads to anorexia or other forms of nutri- tive failure in our animals, despite the comment which this feature has received from other investigators. There is no convincing reason why a continued unvaried diet should necessarily be unphysiological ; one need only recall the fact that the diet of all sucklings is the same from day to day, and that many of the domestic animals are satis- factorily maintained on rations which are scarcely altered in quali- tative make-up except at long intervals. We have observed rats in the same cage for considerably more than a year, during which the daily diet was invariably furnished in the form of our food-pastes. In some of these the composition of the paste was practically the same during these very long periods (cf. Charts XXVII, XXVIII, XXIX, XXX). It is true that we could point to many failures to maintain rats on an unchanged diet continued over much shorter periods. One must not, however, here confuse monotony with the real cause of decline. In these latter cases some deficiency or defect in the monot- onous feeding sooner or later brings on a physiological state where anorexia occurs ; and the advantage which a change ofdiet initiates ought primarily to be ascribed to the alteration in the food ingredi- ents rather than the relief from the sameness of the intake. Among factors referring more directly to the nature of the food itself, the physical texture and digestibility of the nutrients must be taken into consideration. The structure of the food materials may, under ordinary conditions of diet, influence its utilization in no small degree; and the low "coefficients of digestibility" shown by many foods of plant origin testify to this fact. In our experiments the products fed were isolated and reduced to a state of very fine com- minution. At most, therefore, some inherent indigestibility of the individual foodstuffs employed might be concerned. Experiments INTRODUCTION. 57 by M. S. Fine,* while they do not completely do away with this possibility, make it more evident than before that incomplete diges- tion is, in the case of plant products, for the most part associated with the peculiar vegetable tissues therein contained, rather than a specific resistance of the isolated nutrients. The need of "roughage" to facilitate the normal evacuation of the gut has also been debated. We have, as a general procedure, added the indigestible polysaccharide carbohydrate agar-agar to food- pastes in order to approximate more nearly the conditions which prevail where cellulose enters into the mixed dietary. It can not be maintained, however, that this is necessary for satisfactory nutrition ; for we have maintained animals over a year on foods (cf. Chart XXIX) devoid of indigestible principles, if perhaps an exception be made of some of the inorganic ingredients. It is well known that inorganic salts, notably bone ash, may exert the same influence as cellulose in giving bulk to the faeces ; and they are often so employed in the technique of metabolism experiments at the present time.f Aside from the proteins, in which our experimental interest has been primarily centered, our attention has been drawn more and more to those components of the diet which are not sources of energy, yet fundamentally indispensable — namely, the inorganic compounds. It is possible that further investigation will compel the inclusion of some of the more vaguely defined and unknown members of the groups spoken of as extractives, lipoids, etc., in this category. Every attempt made by us to approach the solution of the problem of inorganic salts in the dietary has brought fresh surprises. When Forsterj fed dogs and pigeons on salt-free foods he made the interesting observation that the animals speedily died — more rapidly even than when all food was withheld. He concluded: Der im Uebrigen in Stickstoffgleichgewicht sich befindende thierische Organismus bedarf zu seiner Erhaltung der Zufuhr gewissen Salze; sinkt die Zufuhr unter einer gewisse Grenze oder wird sie ganzlich aufgehoben, so gibt der Korper Salze ab und geht daran zu Grunde. The classic experiments of Lunin§ on mice led to a somewhat different interpretation of the need of salts. He showed that the animals survived longer on a diet containing an addition of sodium carbonate to the ash-free food than when sodium chloride was added. In the latter case the duration of life corresponded approximately with that observed on a salt-free dietary. From these facts it was argued that the foremost value of the sodium lies in its capacity to neutralize the acids (sulphuric, phosphoric) formed in the metabolism *M. S. Fine: Dissertation, Yale University, 191 1 (unpublished). Cf. Mendel and Fine: Journal of Biological Chemistry, 191 1, vols. X and XI. fCf. Lothrop: American Journal of Physiology, 1909, xxiv, p. 297. JForster: Zeitschrift fur Biologic, 1873, ix, pp. 297-380. §Lunin: Zeitschrift fiir physiologische Chemie, 1881, v, p. 31. 58 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. of proteins. Sodium chloride obviously has no potential neutraliz- ing power. If the usefulness of the salts were associated solely with their specific character as salts, the salts of sodium ought to be some- what comparably efficient. The function of the inorganic salts is by no means exhausted, how- ever, by the simple action of chemical equilibrium. It would lead us too far afield in this place to discuss the problem in detail. Charts XI, XII, and XIII, Part I, pp. 38-39) showing the marked differences induced by alterations in the inorganic salts of the diet, the other food components remaining unchanged, are highly suggestive. We have since then made numerous attempts to improve upon the salt mixture empirically selected and prepared somewhat in imitation of the ash of milk. Rats were kept alive (while they steadily declined) 84 days on a food mixture which analysis showed to contain only minimal, inevitable traces of ash (0.16 per cent, a considerable part of which was phosphoric acid derived from the casein) . Chlorides were entirely lacking, distilled water being furnished for drinking. In view of this it is necessary to proceed with extreme caution in draw- ing conclusions from observations extending over brief periods. We shall refer to the subject again, it being sufficient here to emphasize the subtle and specific value of the salts. The lack of knowledge in this field has furnished an obstacle which we have only lately suc- ceeded in overcoming in part. Even when all these varied conditions are taken into account, there still remain, as we have pointed out before, extraneous inci- dents and accidental factors apart from nutrition itself, which may complicate or vitiate experiments like those projected. Disease, old age, injury, may be mentioned in illustration. Failures to maintain nutrition successfully under such extreme conditions do not neces- sarily imply a deficiency or inadequacy of the dietary. Accordingly, successful experiments must be given far greater weight than failures, where so many possibilities of detrimental influences, aside from the diet itself, are liable to arise over prolonged periods of observation. Some of the uncertainties have been eliminated by the experience previously gained. For example, the intercurrent diseases of our animals have been almost entirely excluded by the use of rats raised in the laboratory for this research. By the prompt elimination of diseased animals, by scrupulous attention to the conditions of the cages and feeding arrangements — in other words, by painstaking attention to hygienic factors — we have succeeded in maintaining a large number of animals in exceptionally good health, so that they have become the more suitable to permit of accurate conclusions regarding the effects of the diets studied. Furthermore, the age and hereditaryfactors in our animals are nowknown to us, so that another source of uncertainty has disappeared. KARUKK EXPERIENCES OF THK AUTHORS. 59 EARUER EXPERIENCES OP THE AUTHORS. As the result of the first year's experiments, it was found possible to maintain rats in health and apparent nutritive equilibrium over considerable periods of time on a mixture of isolated food-substances containing isolated proteins as the source of nitrogenous intake. For example, one protocol (Chart XXX) shows that a full-grown rat* was maintained satisfactorily in this way for more than 2 1 7 days on glutenin, the animal continuing on this regime at the time when the earlier report was prepared for publication. Rats were likewise main- tained on diets in which other proteins, notably casein alone or in combination with isolated vegetable proteins, formed the sole nitrog- enous food component, over periods of time exceeding any previously reported, at least under conditions in which the "purity" of the dietary substances was carefully maintained unchanged over equally long periods of time. By maintenance we do not merely mean that the animals remain alive. No feeding experiment is to be regarded as successful in fulfilling the nutritive requirements unless the animals approximately maintain their weight and health (or make normal growth if they are at a stage where this is still to be expected). Although these apparently successful experiments indicated that the combinations of isolated food-stuffs employed satisfied the nutri- tive requirements of the rats and consequently constituted a com- plete food for the maintenance of mature animals, a prolongation of the observations has led to a less favorable outcome. A continuation of the experiments over longer periods has shown that in every case, sooner or later, the animal declined ; and unless a change in the diet was now instituted within a comparatively short time the animals died. The Charts XIV, XV, XVI in our earlier paper illustrate this very well. The rats 23, 24, 25 were maintained without noteworthy alterations in weight over 130 to 160 days on a constant mixture including a single protein. The animals ate well, as the food records show, until the final period of decline. These records can be duplicated, especially in respect to the de- cline, by many others, as for example Charts XLI, XLII, LXXVIII, LXXIX, LXXX, CII, CXV, CXVI appended to this report. The history of rat 71 is particularly instructive on this point. f This animal (see Chart XXX), weighing 257 grams on April 5, 19 10, was put upon a diet containing casein (12 per cent) and glutenin (6 per cent) as the only proteins. Subsequently glutenin alone (16.4 per cent after 69 days and 18 per cent after 104 days) formed the protein of the diet. The rat continued in excellent nutritive *The earlier data regarding this animal, rat 71, are given in Publication No. 156, Carnegie Institution of Washington, p. 47 ff. jThe earlier data will be found in Publication No. 156, Carnegie Institution of Wash- ington, pp. 47-48. 60 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. condition, eating well and exhibiting favorable nitrogen balances, until the end of q§ months, when a gradual decline was observed. When the animal, at the end of a total feeding period of 335 days (42 days after the onset of the decline) was reduced to 162.5 grams in weight and near death, an attempt was made to see whether the decline was due solely to improper food or to the onset of old age or disease. With mixed food realimentation took place at once and the rat regained its weight in a week. A resumption of the former glutenin food during 35 days gradually led to a second decline, which was promptly checked by a change in the diet involving only the non-protein components of the food mixture. Here, then, is a record of the feeding of a full-grown rat, with the exception of 7 days, during a period of 454 days on a diet of isolated food-stuffs and on a diet containing a single protein, glutenin, for 371 days. This obser- vation is remarkable because of the exceptional duration of the experiment. It is apparent, therefore, that as a maintenance diet our food lacked something other than protein and energy. It remains to be shown precisely what the lacking component of our earlier diets is, whether some organic constituent or a peculiar proportion of inorganic ingredients. In any event it is evident that our original artificial food mixtures are incapable of supporting life indefinitely. Aside from this, however, records like that of rat 71 living on glutenin as the sole source of protein (see Chart XXX) , or rat 133 (Chart LXX) on edestin, in contrast with rats xi, xiv, 146. and 157 (Charts CXXVI, CXXVII, CXXVIII, and CXXIX) on zein indicate the possibility of nutritive inequalities among the proteins themselves. Marked deficiencies tend to manifest them- selves in comparatively short periods of time. In all of these cases the food actually consumed supplied sufficient energy for the imme- diate needs of the rats under investigation. In the continuation of our experiments we have tried to profit by the first year's experiences. The methods have not been materially altered, except that the determination of the nitrogen balance has been omitted for the present. We learned from very numerous trials that it runs parallel with gain or loss of weight, and that the food intake varies closely with the weight of the animal, thereby making a record of the nitrogen unnecessary for judging the nutritive status of the rats employed. The same cages as heretofore have continued to prove very satisfactory. Instead of being rested on glass funnels for the collection of urine, they are now placed over a frequently changed sheet of absorbent paper (paper napkin) upon an enameled tray or pan. The fluid excreta thus promptly absorbed are frequently removed. It has already been pointed out that the food mixtures, prepared in paste form to prevent scattering by the animals and make it possible to obtain accurate records of the quantities eaten, ALIMENTARY BACTERIA AND NUTRITION. 6 1 are not ideal in composition. The inclusion of 20 to 45 per cent of fat in the diet — a condition necessitated by the requirements of the experiments as outlined — seems like an excessive amount; neverthe- less the utilization appears to be satisfactory and attempts to devise less objectionable modes of feeding have been unsuccessful in our hands. ALIMENTARY BACTERIA AND NUTRITION. In the course of our later studies we have been forced to take cognizance of the possible role of the bacterial flora of the alimentary tract in relation to appropriate nutrition. The water-free, fat-rich food characteristic of our experimental dietaries is not, broadly speaking, a particularly favorable medium for the development of cer- tain groups of bacteria. The food of our animals therefore probably introduces into the digestive tube of the experimental animals bac- terial invaders somewhat different from those which normally inhabit the alimentary tract of rats living on a free mixed diet. It is quite conceivable, therefore, that the bacterial conditions may be altered markedly as a result of the restriction in the growth of certain groups or the facilitation of the development of still others in the alimentary tract under these changed and sustained conditions of altered diet.* It is well known, for example, that in higher animals the preponder- ance of acid-producing organisms — to use a single illustration — may lead to an inhibition of the growth of the putrefactive group. Guided by such considerations and the observation that those rats that have been maintained for long periods on diets with isolated food-stuffs become koprophagists, we have initiated the plan of feed- ing small quantities of the faeces of rats living on ordinary mixed food to some of our experimental animals, particularly in cases where symptoms of nutritive decline had become manifest. In nearly every instance the occasional addition of a small amount of the faeces from a normally fed rat at once stopped the decline in weight of the experi- mental animals to which a single protein was being fed. The results in almost all of these cases have been sufficiently striking to warrant a further pursuit of this topic. In our experiments there appears to be an unmistakable favorable influence induced by the occasional addition to the dietary of normal faeces with their high bacterial content. It must not be overlooked that other components besides bacteria, notably inorganic salts and unknown compounds, are also furnished by this means; but the quantities involved have always been very small. Further investigation will be necessary and is already projected. The procedure in the case of these faeces-feeding trials consisted in introducing small amounts (about 0.5 gm.) of air-dry excrement *Cf. Herter and Kendall: Journal of Biological Chemistry, 1910, vn, p. 203; Kendall: Journal of the American Medical Association, April 15, 191 1. 62 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. of rats on mixed food into the cages twice a week. It is an interesting observation that when the rats kept on a mixture of isolated food- substances were offered a choice between their own faeces and those of rats on mixed diets, they invariably chose the faeces of the latter. In many cases we have noticed a marked improvement in the nutri- tive conditions of animals maintained on a single-protein dietary when other rats were introduced into their cages for breeding pur- poses. In view of the favorable influence exerted by feeding the faeces of rats living on mixed food, it is quite likely that the presence of the strangers in the cages furnished a suitable opportunity to obtain "normal" faeces. This may explain the favorable results noted, in contrast with the negative effects seen where several rats living on the same single-protein diet have been maintained in the same cage. The extent of the influence exerted by what we have, in the absence of a better explanation, assumed to be bacterial influences, is illustrated in some of the appended charts, the periods at which the faeces feeding was begun being indicated. The favorable effects have not been confined to experiments wTith one protein, but are mani- fested with casein (see Charts XXXIX, XL, XLL and XLII), with edestin (see Charts LXVI, LXVII, LXVIII. and LXIX), and with gliadin (see Charts CI, CII, and CIII). Two failures may likewise be recorded, viz, an ultimate one with casein (Chart XLI) and a com- plete one with edestin (Chart LXXVII) as the protein component. These were not due to incapacity of the animals to grow, since fur- ther alteration of diet brought marked improvement. The influence of faeces feeding is especially striking in the case of the gliadin tests, since without the addition of the faeces it has been almost impossible to attain satisfactory nutritive condition with this protein plus the special non-protein components of the food here employed. It is instructive therefore to compare such failures (cf. Period 2, Charts CXV and CXVI) with Charts CI and CIII, in which faeces feeding was resorted to. In four of the experiments with edestin-food alluded to and re- corded on Charts LXVI, LXVII, LXVIII, and LXIX, fresh faeces were not actually introduced into the cages; but the improvement, and even growth, in these young rats is coincident with the oppor- tunity afforded to obtain "normal " faeces when other rats were daily introduced into the cages for a few hours. In Chart CII is seen the result of an attempt to determine whether the favorable influence of the faeces is actually of bacterial nature. Faeces were fed as in the comparable gliadin experiments (Charts CI and CIII) ; but they were previously sterilized by thrice repeated heating in an atmosphere of steam. The decline of the animal was not prevented to the same extent with sterilized as with normal faeces. Further trials are necessary in this direction; and our NUTRITION AND GROWTH. 6$ experience, though limited, invites attention anew to the possible nutritive functions of bacteria in the alimentary tract. Some of the aspects of this problem are referred to in our earlier paper.* NUTRITION AND GROWTH. The criteria of adequate nutrition are quite different in the case of growing animals from those applying to adults of the same species. During the period of adolescence it is not sufficient to maintain a condition of nutritive equilibrium and constancy of form or body- weight. In this stage of an animal's existence there should be evi- dences of development, and growth should manifest itself in a change of size. The curve of growth, expressed in changes of body-weight, is remarkably constant and characteristic for each species under the ordinary conditions of nutrition and environment. The individual values may at times fluctuate about a mean ; but in the majority of cases the excursions from the average are not extensive. In Chart XXII are reproduced curves illustrating the average normal rate of growth of the white rat, both male and female. The statistics for two of the curves are taken from Donaldson,! whose observations we have repeatedly verified in their general features. A third curve on the same chart represents the results of our own observations on the growth of the female white rat, regarding which data are less abundant. It will be noted that the curves of growth for the two sexes do not completely coincide in type. After an age of 70 days, represented by a body- weight of about ioo grams, the rate of growth is somewhat slower in the female than in the male. In- deed, the females rarely attain the large weight and size exhibited by the normal adult males of the same age, even in the case of animals from the same litter. We gain the impression that our "breed" of rats may in general be somewhat smaller than those measured by Donaldson and his collaborators. At any rate, the data available for statistical purposes are not very extensive and the curves here presented must have only a provisional value until more numerous measurements are made. In connection with certain of our experi- ments it may be stated that "the effect of mating on the growth- curve for the males can probably be neglected. "J In the case of females, the effect of the bearing of young is, according to Watson, § "to render the mated rats slightly heavier than the unmated — some of the excessive weight being due to the larger amount of fat present in the mated animals." Two charts (XXIV, XXV) are appended *Carnegie Institution of Washington, Publication No. 156, p. 3. fDonaldson: A comparison of the white rat with man in respect to the growth of the entire body. Boas Memorial Volume, New York, 1906. JCf. Donaldson: ibid, p. 8. §Watson: Journal of Comparative Neurology, 1905, xv, p. 523. 64 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. to illustrate the influence of the course of pregnancy on the growth- curve of female rats of different sizes. Making allowance for these minor divergencies, the striking uniformity in the progress of development in an animal nevertheless is a specific racial characteristic, and gives to the curve of growth a unique value as an index of the conditions which determine it. Growth is affected by two factors : nutrition, and what Rubner has termed " Wachstumstrieb " or growth-impulse. The latter factor is inherent in the animal. The limits are determined by heredity and can not be altered materially by the most abundant diet. " Eine noch so reichliehe Ernahrung vermag die in derRasse und derenVererbung gelegenen Grossen- und Massenbegrenzungen nicht zu mehren."* We are not prepared, at this time, to discuss the nature of the hereditary factor or impelling "force " in growth. f Arou writes : Die Natur des Waelistumstriebes ist dunkel. Sie ist eine Funktion der Zellen, im besonderen der jugendliehen Zellen. Welehe Faktoren diesen Zelltrieb regulieren, wissen wir nicht, vor allem nicht, warum er allmahlich aufhort. Ob hier die Zeitdauer seiner Wirksamkeit, ob die erreiehte Grosse des Individuums den Ausschlas: fur das Abklins^en des Waelistumstriebes gibt, ist bis jetzt nicht entschieden.J Rubner has attempted to formulate its character: Die eine grosse Unbekannte auf dem Gebiete der Wachstumsphysiologie ist der Wachstumstrieb, der in gesetzmassiger Weise den Gang der Entwiek- lung, Massenzunahme, durch die Regelung der Ernahrung leitet. Den Urgrund hat dieser Wachstumstrieb in der Geschwindigkeit derKernteilung; wie wir noch sehen werden leitet sich hieraus der ganze Prozess des Stoff- umsatzes ab. Die Kernteilungsgeschwindigkeit ist offenbar etwas der Spezies Eigentiimliehes, somit sind wir nicht in der Lage, vorlaufig tiefer in dieses Problem vorzudringen.§ The second factor in growth, namely, n utrition , can be approached more easily by the experimental method. It is along this line that we have hoped, therefore, to be able to attack some of the problems of the relative value of the individual foodstuffs. It is well known that growth can be retarded by means involving the nutrition of the individual. Waters has well summarized the situation in these words : The upper limit of the size of an animal is determined by heredity. The stature to which an animal may actually attain, within this definitely fixed limit, is directly related to the way in which it is nourished during its grow- ing period. vSome of our approved theories have been so extreme as to hold, in effect, that the animal must grow at its maximum rate practically every *Rubner: Archiv fiir Hygiene, 1908, lxvi, p. 82. fCertain aspects are considered in C. S. Minot: The problem of age, growth, and death. New York, 1908. JAron: Biochemische Zeitschrift, 1910, xxxp. 207. §Rubncr: Archiv fiir Hygiene, 1908, lxvi, p., 86. NUTRITION AND GROWTH. 65 day from birth to complete maturity in order to reach its normal size, or the full stature fixed by heredity. In other words, it is assumed that the animal has but one way of reaching its full stature and full development, viz., by developing to its upper limit through its entire growth period. This assumes that the organism is utterly incapable of compensating for any retarded development at any time in its growth period, either by a subse- quently increased rate of growth, or by extending, even in the slightest degree, the growth cycle, much less by growing for a time at least when so sparsely fed that no gain in weight occurs.* Rubner has expressed the role of nutrition in growth as follows : Kanndie Ernahrung auch keinen Wachstumstrieb sehaffen, so kann sie, wenn ungiinstig und unzweckmassig, doch zu einem Hemtnnis des natur- lichen Wachstums werden. Wachstumsbehinderung ist innerhalb gewisser Grenzen noch keine Ursache einer Existenzgefahrdung, ein Kind, dem die Nahrung normales Wachstum hindert, stirbt deswegen durchaus nicht, es holt spater leicht wieder ein, was es versaumt hat . . . Nur das steht sicher, dass die Behinderung des Wachstumstriebes, wie dies wirklich vorkommt, nicht wahrend der ganzen Wachstumsperiode andauern darf, da sonst allerdings die Grosse des Individuums dauernd Schaden leidet. Verlorene Korpergrosse in der Jugendzeit kann nach Vollendung der Wachstumsperiode nimmermehr abgeglichen werden . . . Eine optimah Ernahrung, wie die W achstumserriahrung sein muss, stellt an die richtige Auswahl der Stoffe ganz andere Anforderungen als eine einfache Erhaltungsdiat.f Obviously the energy problem plays an important part in the nutrition of growing animals. For the present we are primarily concerned with the qualitative aspects of the diet rather than the quantitative features of the food-intake. These two factors may at times, stand in intimate relation to each other; improperly consti- tuted food may, for example, modify the amount eaten and therefore the energy available for growth. As was intimated in our first report we have been able to arrest development in rats by feeding mixtures containing a single protein; but inasmuch as the food intake was not measured at that time, it was impossible to say whether the chemical character of the diet or a quantitatively inade- quate food consumption was responsible for the dwarfing. The fact brought out was that in these young animals there could be a main- tenance of weight, precisely as in older rats. Waters has appropriately emphasized the necessity of a more exact definition of what is meant by maintenance, in contrast with growth. He writes: 0 It has long been assumed that the body of an animal, when supplied with only sufficient nutriment to maintain its weight, remains constant in composition and that no growth or production or change of any sort occurs. *H. J. Waters: The capacity of animals to grow under adverse conditions. Proceed- ings Society for the Promotion of Agricultural Science, 1908, xxix, p. 3. fRubner: Archiv fiir Hygiene, 1908, Lxvr, pp. 82-83. 66 FEEDING EXPERIMENTS WITH ISOLATED EOOD-SUBSTANCES. It is true that the term maintenance has been used somewhat loosely, but in general we have been in the habit of regarding the animal in maintenance when its live weight was constant. A more correct definition of the term would perhaps be to say that the animal was in maintenance when its body was in energy balance, but the live weight has been the conventional measure of our maintenance values.* It is generally admitted that the proteins satisfy several functions in a growing organism as well as in the adult. The first is that of maintenance, corresponding with what has been termed the " Abnut- zungsquote, " or wear-and-tear, by Rubner. This makes good the inevitable losses occasioned by the processes of metabolism, cellular and secretory processes, etc. It is a small yet ever present need for protein (as well as energy), representing in a general way the minimal protein need of the stationary organism. Any excess of protein beyond this maintenance requirement may, in the adult, experience temporary storage (" Ansatz") or be devoted to dynamo- genie purposes ; but in the organism capable of development it con- tributes a share toward growth. It should be emphasized that the rate of growth is not by any means proportional to the excess of protein available. It is surprising, indeed, how small a content of protein in the dietary suffices to make growth possible. Rubner and Heubnerf found, for example, that in suckling infants a protein intake equivalent to 5 per cent of the total calories satisfies the protein needs of maintenance, while 7 per cent permits of growth. Rubner writes: Das Wachstum ist erne Funktion der Zelle, es kann durch unzureich- ender Eiweisszufuhr latent werden, aber Eiweiss vermag nicht die Wach- stumsschnelligkeit iiber die von der Natur gestreckten Grenzen zu heben, daher wird mit steigender Eiweissmenge in der Kost prozentisch weniger verwertet und das tiberflussig zugefiihrte Eiweiss wird einfach als Brennstoff verbraucht der isodyname Mengen N-freier Stoffe einspart. Diese starke Anziehung von Eiweiss zum Wachstum nimmt im Laufe der Entwickhmg ab und ist am grossten in der ersten Zeit des Lebens.^ Waters has found in his extensive studies on cattle that growth, in the sense of changes of size and form, may occur even under adverse nutritive conditions. Fundamentally such investigations touch upon the much controverted question as to the relative impor- tance of breeding and feeding in determining the shape and activities of mature animals. It is well known that by limiting the food supply of an ungrown individual, its development may be retarded. If the underfeeding is prolonged through the cycle of growth, the full stature limited by heredity may not be reached. *H. J. Waters: The capacity of animals to grow under adverse condition. Proceed- ings Society for the Promotion of Agricultural Science, 1908, xxix, p. 3. fRubner and Heubner: Zeitschrift fiir experimentelle Pathologie, 1905, I, p. 1. JRubner: Archiv fiir Hygiene, 1908, i,xvr, p. no. NUTRITION AND GROWTH. 67 Waters asked the question: Will this animal of smaller stature be in the same proportion with re- spect to all the organs and the different parts of its body as though it had been nourished to its full capacity and had attained its normal size and max- imum development? Or will in this period of sparse nourishment a more complete development occur in certain parts of the body than in other parts? In short, when there is not sufficient food supplied to the growing animal to develop all of the organs and all parts of the body to their full limit and extent, will the rate of development of certain of these organs or parts diminish earlier than others and will the development of certain parts cease altogether before the development of other parts is diminished in rate and is it possible that some parts may cease their development before that of other parts?* In actual experiments at the Missouri Agricultural Experiment Station, Waters found that ungrown cattle may remain at a constant body-weight for a long period of time, and yet increase in height and apparently decrease their store of fat. In other words, the skeleton has grown, or at least the bones have lengthened. Two interesting illustrative protoeolsf are reprinted here, one, Table XXXI, in which a stationary body-weight was maintained, the other, Table XXXII, in which there was actual decline on a starvation ration. Table XXXI (from Waters, Table II). — Showing Increase in Height at Withers, Length of Head, Depth of Chest, Width of Chest, and Loss of Fat in a Yearling Steer when Kept at a Stationary Body- Weight. No. 595. Grade Hereford. Born May 15, 1907. Nine and a half months old when experiment began. Full fed four months previous to beginning of trial. Condition when put on maintenance, medium. Weight at beginning of trial, 609.2 lbs. Weight at close of trial, 595.6 lbs. Average of ten daily weights. Date. Height at withers. iqo8. cm. Feb. 8 ioq Mar. 13 1 12. 5 Apr. 11 1 15 . 5 June 2 . , 116 July 1 117. 5 Aug. 1 1 17. 5 Sept. 2 1 1 7 . 5 Sept. 2Q 1 ICJ Oct. 30 1 IQ.25 Nov. 30 : 1 iq. 5 IC)OQ. Jan 1 1 iQ.75 Jan. 30 > 119.75 Totalheightin 12 months. 10.75 Per cent gain ; q . 86 Length of head. Depth of chest. cm. cm. 38 40 56 58 4' 57-5 42 44 59 58.5 44 59 44 59-5 45-50 59-5 45-75 59-5 45-75 59-5 46.50 45.50 60. 5 60.75 7.50 ■9-73 4-75 8.48 Width of chest. cm. 35 36.5 Condition. 35- 33- 34 33 33 33- 31 3' 30.75 30.75 25 1 Medium. Medium. Medium to thin. Common. Common. Common. Common to fair. Fair. Fair. Fair to thin. Thin. Thin. - 4 — 12 Notk. — When slaughtered, carcass wasclassed as poor canner. All visible stibdermal and intramuscular fat had disappeared. * — Denotes a loss. *H. J. Waters: The influence of nutrition upon the animal form, for the Promotion of Agricultural Science, 1909, xxx, p. 71. Proceedings Society jFrom H. J. Waters: The capacity of animals to grow under adverse condition, ceedings Society for the Promotion of Agricultural Science, 1908, xxix. Pro- 68 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Table XXXII (from Waters, Table VI). — Sub-Maintenance. Steer No. 591. Grade Hereford. Born May 15, 1907. Experiment began Feb. 26, 1908. Age of animal at beginning of experiment, nine and a half months. Full fed four months before trial began and was in good condition. Weight at beginning of trial, 572.7 lbs. Weight at close of trial, 490.4 lbs. Total loss in weight, 82.3 lbs. Average daily loss 0.43 lb. — Denotes loss. Date. I C)08. Feb. 8 Mar. n Mar. 28 Apr. May- 1 1 2 June 1 June July Aug. 20 3' 3' Gain Height at Length of withers. head. cm. cm. 110.5 30 113 41-5 "5 42 1.4.5 41 116 42 118. 5 44 120 44 HO 44-5 nO-5 44-5 0 5-5 8. ,4 14.10 Depth of Width of chest. chest. cm. 57 57-5 5S 57 57-5 58 50-5 58 n cm. 38.5 34-5 35 33 33 33 31-5 20.5 20 - 9-5 — 24.6 The following is from Waters, in regard to a series of compara- ble cattle maintained by him on different nutritive planes, desig- nated as sub-maintenance, maintenance, and super-maintenance: It is to be observed that there is no appreciable difference in the rate of growth in height of these three animals on widely different nutritive planes, from the beginning of the experiment (February) to the end of June. At this time the curve of the sub-maintenance animal flattens perceptibly. A month later, the maintenance animal is apparently responding to the in- fluence of the low nutritive plane. As would be expected, in the case of the super-maintenance animal, the rate of growth remains unchanged. It may be surprising to many [Waters writes elsewhere] that an animal on maintenance, much less on sub-maintenance, should show any increase whatever in the width of hip or length of leg . . . Apparently the animal organism is capable of drawing upon its reserve for the purposes of sustain- ing the growth process for a considerable time and to a considerable extent. Our experiments indicate that after the reserve is drawn upon to a con- siderable extent to support growth the process ceases, and there is no further increase in height or in length of bone. From this point on the animal's chief business is to be to sustain life. This law applies to animals on a stationary live weight as well as those being fed so that the live weight is steadily declining, and indeed to those whose ration, while above main- tenance and causing a gain in live weight, is less than the normal growth rate of the individual. Such an animal will, while gaining in weight, be- come thinner, because it is drawing upon its reserve to supplement the ration in its effort to grow at a normal rate.* More recently Aronf has made comparable studies on growing dogs. He formulated his problem in the following words: "Was wird geschehen, wenn furkurzereoderlangereZeitinderNahrung nur so viel Energie usw. zugefiihrt wird, wie erforderlich ist, umden Erhalt- *H. J. Waters: How an animal grows. Kansas State Board of Agriculture, Seven- te?nth Biennial Report, 1909-1910, 1, p. 208. jAron: Biochemisehe Zeitsehrift, 1910, xxx, p. 207. NUTRITION AND GROWTH. 69 ungsbedarf des wachsenden Organismus zu befriedigen, aber kein Ueber- schuss, der als Wachstumsenergie dienen konnte? Die naehstliegende Annahme ist, dass dann kein Waehstum stattfmdet, dass der Waehstums- prozess stillsteht. Konnen wir nun wirklich den Waehstumstrieb durch Nahrungsbesehrankung unterdriieken? Wie lange? Und was gesehieht spater mit einem wachsenden Organismus, dessen Waehstum eine Zeitlang hintan gehalten worden ist? (p. 208.) Aron succeeded by restricted feeding in attaining constancy of body-weight in practically all of his dogs, in some cases during a period of nearly a year. The daily gains or losses fluctuated within a few grams. The description of the animals during the experiments is of interest to us: Bei alien Hun den konnte man deutlieh beobaehten, wie die Tiere trotz des Gewiehtsstillstands wuchsen, d.h. an Hohe und Lange zunahmen. Dabei wurden die Tiere zusehends magerer, Fett und Muskeln sehienen an Masse abzunehmen, die runden Formen sehwanden, die Knoehen traten eekig unter der Haut hervor, und schliesslieh sehienen die Tiere nur noch aus Haut und Knoehen zu bestehen. Trotzdem waren die Hunde nieht etwa sehwaeh. Im Gegenteil, sie waren lebhaft, liefen und sprangen umher, oft mehr als ihre normalen Brudertiere, die ein zwei- oder dreimal zo grosses Korpergewieht zu bewaltigen batten. Dieser Zustand zunehmender Abmagerung unter standiger Grossen-, d.h. Langen- und Hohenzunahme beiKonstantbleibendesGewiehtes dauerte je nach dem Grade derNahrungs- entziehung (ungefahr 3 bis 5 Monate an. Wurde jetzt, wenn das Tier vollig abgemagert war, . . . , die Nahrungsmenge writer so gering belassen wie vorher, so ging das Tier unter geringem Gewiehtsverlust in volliger Inanition zugrunde. Wurde aber jetzt die Nahrungsmenge etwas erhoht, wie bei Hund A, so hielt sieh das Tier zwar vollkommen abgemagert, aber auf konstantem Gewieht. Und jetzt erweist sich dieser Gewichtsstillstand als identisch mit Wachstumsslillstand! Der Hund A ist noeh weitere 5 Monate auf dem gleichen Gewieht gehalten worden, ohne dass sich nun in seinem Aussehen nennenswerte Aenderungen konstatieren liessen. Dureh geeignete Nahrungsbesehrankung gelingt es also, waehsende Hunde beliebig lange auf konstantem Gewieht zu halten. Naturlich darf man nieht allzu junge Tiere nehmen. Wahrend dieses Gewiehtsstillstandes gehen aber gewaltige Umwandlungen im Tierkorper vor, die sich ausserlieh in dem fortschreitenden Langen- und Hohenwaehstum und der Abmagerung dokumentieren. OfTenbar ist trotz des Gewiehtsstillstandes das Skelett weiter gewaehsen und hat nieht nur an Grosse, sondern aueh an Masse zugenommen. Folg- lich miissen andere Korpergebilde (wie Haut, Fleiseh, Organe usw.) an Gewieht verloren haben; denn sonst konnte ja das Gewieht des Tieres nieht das gleiche geblieben sein. Ebenso wie die Massenverhaltnisse der einzelnen Korpergebilde haben sich nun hoehstwahrseheinlich aueh die Mengenverhaltnisse der einzelnen Korperbestandteile, wie Fett, Eiweiss usw., betrachtlieh versehoben. (p. 212.) Aron's analyses of the underfed dogs showing stationary weight in comparison with well-fed control animals indicate that in addi- tion to the bones, the brain also was protected from loss of weight, while the adipose and muscular tissue suffered notable losses. Most striking is the degree to which water has replaced the tissue substance 7o FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. utilized to compensate for the lack in the food, the blood especially becoming distinctly "watery," as the selected protocol shows:* Table XXXIII.— Content of Dry Matter in Various Tissues. Control dog. Underfed dog. Blood Brain. . . . *i8.8 24.6 57-2 2Q. I *5 . I IQ-3 40.0 15.2 Bones Muscle *Protem = NX6.';. It is apparent here, as in Waters's experiments, that the energy deficit has been furnished by the body. "Sind alle verfiigbaren Reservestoffe aufgebraucht, dann gewinnt der Erhaltungstrieb die Oberhand iiber den Wachstumstrieb, und das 'Wachstum' stockt." (Aron, p. 222.) In relation to our own later observations it is desirable to quote Aron's view regarding the impulse to growth. He concludes: . . . dass die innere treibende Kraft zum Waehsen iiberhaupt in dem Kerngeriist des Korpers, dem Skelett, ruht. Die Muskulatur verfiigt anseheinend iiber gar keinen riehtigen Wachstumstrieb. Sie folgt dem wachsenden Skelett nur dann, wenn die Ernahrungsverhaltnis.se es erlauben, vielleicht auf Grund rein mechanischer Krafte (Zug). Recht interessant scheint zum Schluss noeh die Frage, wie sich bei den durch lange fortgesetzte Unterernahrung im Wachstum zuriickgehaltenen Tieren die Entwicklung und die Entwieklungsfahigkeit verhalt. Mein Tiermaterial war nicht ausreichend, um ein Studium der Gesehlechtsorgane der zwar im Alter der Geschlechtsreife stehenden, aber im Wachstum weit zuriickgebliebenen Tiere zu gestatten. Dagegen scheint mir die Beobaeh- tung der Stimme auf ein wirkliches Zuriickbleiben der Entwicklung auf dem infantilen Stadium zu deuten. Die Unterschiede zwischen den Bruder- tieren der ersten, zweiten und vierten Versuchsreihe waren auffallig. Die im Gewicht zuriickgebliebenen Tiere schrien kreischend wie junge Hunde, wahrend ihre normalen Brudertiere mit tiefem Tonfall bellten. In ganz dem gleichen Sinne spricht die von Waters festgestellte Tatsache, dass seine in Gewicht und Wachstum zuriickgebliebenen Tiere ein Fleisch, das fiir 'Kalbfleisch' charakteristisch war, aufwiesen, wahrend sie dem Alter nach schon "Rindfleisch" besitzen sollten. (pp. 222-223.) Studies of the relation of weight to the measurements of children during the first yearf have also given evidence of "disproportionate " growth in the case of poorly nourished infants. Whereas there is, in the normal infant, a fairly constant relationship between body- weight and height, circumference of head, chest, etc., this is not true where proper increase of body- weight is retarded by poor nutrition. For example, in children whose weight at the end of the third month *Arcn: Biochemische Zeitschrift, 19 10, xxx, p. 220. fE. C. Fleischner: Archives of Pediatrics, October 1906. SUSPENSION OF GROWTH ON A MAINTENANCE DIET. 7 1 is only equal to that of a normal child at birth, the height has been found above that of the latter, illustrating, as Fleischner remarks, "that age plays some part in the growth of the infant, independent of the weight." This corresponds with the cases of the animals already cited. Fleischner concludes from his measurements of 500 children of whom 25 per cent were well nourished, 35 per cent fairly well nourished, and 40 per cent poorly nourished : It is in the poorly nourished children that age plays its most important part ... In the poorly nourished children, most of whom are probably somewhat premature, when the weight is below normal, all the measure- ments are correspondingly below normal. The height and circumference of the head reach the normal birth measurements a little ahead of the weight, while the chest and abdomen are two months later in reaching the measure- ments of a normal child at birth. When the weight is stationary the in- crease in the measurements is very small, depending upon the slight in- fluence which age has upon the growth of the infant notwithstanding the weight. The measurements of infants of the same weight, notwithstanding the age, are very similar, the small difference depending, as when the weight of a child is stationary, upon the very slight influence of age upon growth. The final conclusion can be drawn that during the first year of life the primary factor in the increase of the measurements of the body is steady, consistent increase in the weight, the influence of age being secondary and much less important.* SUSPENSION OE GROWTH ON A MAINTENANCE DIET. Early in the course of our investigation we noted that young rats could remain in apparent good health while living on some of the mixtures of isolated food-stuffs, without giving any evidence of growth. In some instances the animals ultimately declined and died where the diet was not changed ; but in numerous cases body- weight, which we used as our guide, remained practically unchanged or showed a minimal slow increase (cf. Charts XXXVII, LXIII, and LXIV). The experiment showing the greatest growth under these dietary conditions is recorded in Chart XXXVIII. Other investigators have met with this stationary condition and accepted it as evidence of satisfactory nutritive equilibrium. We soon became convinced, however, that a diet which will not induce real growth at the proper age is unquestionably defective from the standpoint of perfect nutrition. Furthermore, inasmuch as the ungrown rat has a far smaller reserve of available energy and manifests the utilization of a suitable diet both speedily and conspicuously by its measurable changes in size, the animal becomes an exceptionally appropriate subject at this early stage for the study of the nutritive requirement. The most precise evidence which we can present at this time of the stationary condition of the animals which we have stunted by *E- C. Fleischner: Archives of Pediatrics, October 1906. J2 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. the particular dietaries adopted is derived from measurements on three young rats of the same litter maintained for 124 days without noteworthy growth, on a diet of Per cent. Glutenin 18.0 Starch 14.5 to 34.5 Sugar 15.0 to 20.0 Agar 50 Salt mixture 1 2.5 Lard 20 . o to 45 . o The curves of growth of these animals as well as three others from the same brood fed on mixed food or the milk-food mixture (and showing a normal growth) are reproduced in Charts LXXXI, LXXXII, LXXXIII, LXXXIV, LXXXV, and LXXXVI. The animals were killed at the age of 178 days and measurements were made by Dr. S. Hatai, of the Wistar Institute. The tabulated data are given on the following page, together with a report from Dr. Hatai, to whom, as well as to Dr. Donaldson, we are greatly indebted for helpful cooperation. The statistics of body-length, weight of brain, spinal cord, etc., of the stunted animals at an age of 178 days are comparable with those characteristic for normally growing rats of the same body- weight, which is attained at an age of approximately 54 to 63 days. Here, then, are illustrations of maintenance without growth. Dr. Hatai further reports as follows: vSince it seems to be'the least variable character, I have selected the body- length as the basis for computation. When the other characters which we can measure are calculated from the formulas based on body-length, it is seen that the observed weight of the brain and of the spinal cord agrees closely with the calculated in both the control and the stunted rats. Thus both series have a growth of the nervous system normal to their body- length. In the control series, the percentage of water observed in both the brain and the spinal cord agrees with that calculated according to the body- length. In general then the control rats agree with the general population in these characters. Since the stunted rats have an abnormally small body- length for their age, they can not be treated by the formula for determining the percentage of water from body-length. When, however, we take the estimated percentage of water for 178 days (see Donaldson*) we find that this value agrees with that observed in the stunted series. It may be further noted that the ratio between body-length and tail-length is the same in both series. We therefore conclude that in both scries the body-weight is normal to the body-length; the brain and spinal cord weight normal to the body- length; and the percentage of water normal for age. Concerning other organs we have no data, but we may infer from the foregoing that they also have weights normal to the body-length. You will see from the above that the stunted rats though small have the general relative development of the Controls and that in the only case where it is possible to follow the maturing process, that is in the percentage of water in the nervous system, they have matured in accordance with their age (see Donaldson*). *Donaldson: Journal of Comparative Neurology, April 191 1. OSBORNE AND MENDEL PLATE 1 B A. Rat 238, female. Age 1 40 days, weight 144 grams, which is normal for a rat of same age as 240. B. Rat 240, female. Age 140 days, weight 55 grams. Same brood as Rat 238. C. Rat 305. Age 36 days, weight 55 grams. Showing the appearance of a normal rat of same size as 240. A and B show the contrast between two rats of the same age, one of which (Rat 240) has been stunted. The lower two pictures afford a comparison between two rats of the same weight, but widely differing in age. The older, stunted rat, B, has not lost the characteristic proportions of the younger animal, C. OSBORNE AND MENDEL PLATE 2 D. Rat 168, male. Weight 235 grams, which is normal for a rat of the age of 220 shown below. E. Rat 220, male. Age 148 days, weight 58 grams. F. Rat 305. Age 36 days, weight 55 grams. Showing appearance of a normal rat of same weight as 220. D and E show the contrast between two rats of the same age, one of which (Rat 220) has been stunted. The stunted rat is not essentially altered in its bodily proportions from those of a much younger rat of the same weight. SUSPENSION OF GROWTH ON A MAINTENANCE DIET. 73 Table XXXIV. — Hatai's Measurements of Stunted Rats from Experiments of Osborne and Mendel, 1910-191 i. Control Rats. Diet. Rat 96... Milk Rat 97. . . Milk Rat 99. . . Mixed Sex. Age in days. Weight in grains of — ■ Body- Brain. I Cord. Hypo- physis. Percentage of water. Brain. Cord. Body Length in mm. of — Tail. Fern. I 178 1 54.9 1 .73460.49340.007378.30671 . 139 176 Fem. 178 164.51.69740.50070.009378.47371.2201 183 Male j 178 175.0,1 .85150.48100.005278.62371 .809 181 Average 164.8 1 .76120.49370.0071 78.46771.389 1 No Calculated from body-length Ii .76450.5004 78.374 Estimated percentage of waterfrom age1 ! i 78.4 71 . 192 71 .2 180 146 164 144 151 Body-length to tail-length 1 : 0.83 Stunted Rats. Rat 100. . Rat 101 . . Rat 102. . Glutenin Fem. | 178 Glutenin Male j 178 Glutenin Male I 178 85 .0 1 .63230. 40890. 003578. 141 70.775 148' 129 71 .8'i . 50220. 37810. 002278. 27271 .701 139; 108 85. 7! 1 .62800.39770.003378. 13371 . 134 148 125 Average..1 80.8 Calculated from body-length Estimatedpercentageof waterfrom age 1 .58750.39290.003078. 18271 .203 145 121 1 .58960.3639 78.4 71.2 145 Body-length to tail-length 1 : 0.83 Brain weight Formulas. , Body-length+134 \ , :o. 569 log (10- 23.7J+0.554 •43 o • 1 1 • u4 01 1 Body-length+134 Spinal cord weight = 0.585 log (10— — + 6) — 0.795 '43 Percentage of water (brain) =82.62 — 2 log (Body-weight— 10). Percentage of water (spinal cord) =85.20—6.5 log (Body-weight). Photographs of other rats which have been dwarfed in like ways give evidence of the similarity of the stunts in general appearance with normal animals of the same weight at a much earlier age. Thus, in Plate 1, rat 305,0, weighing 5 5 grams at the age of 36 days, compares favorably with rat 240, B, dwarfed on a gliadin food mixture, at the age of 140 days, when it weighed 55 grams (cf. Chart CXIII). It is interesting to contrast B with the uppermost photograph A of rat 238, likewise 140 days old and from the same brood but weighing 146 grams, the normal weight for this age. Each was raised under 74 FEEDING EXPERIMENTS WITH ISOLATED EOOD-SUBSTANCES. identical conditions from the age of 38 days, except that rat 238 (see Chart LVI) was fed with a paste containing casein and pro- tein-free milk, while in the food of 240 (see Chart CXIII) the casein was replaced by gliadin. Plate 2 shows rat 220, K, fed on gliadin and protein-free milk but weighing only 58 grams, although 148 days old, and, for con- trast, rat 168, D, of approximately the size normal for the age of rat 220, is also shown. Figure F shows a normally nourished rat of the same weight as rat 220. This picture is introduced to show that rat 220 has the appearance of a normal rat of corresponding size and weight. All these pictures were taken on exactly the same scale and afford a ready comparison of the relative sizes of the animals. The interesting photographs of underfed cattle published by Waters, on the contrary, make the change of form in his under- nourished animals of stationary weight quite apparent. We are, however, not prepared to assert that careful measurements of our stunted rats will not disclose some trace of similar changes in skeletal form. They must be slight at most; for we have often compared animals long maintained at small stature with properly grown animals which have just reached the same weight, without detecting any devi- ation from the youthful form in so far as one could judge by mere visual inspection. The photographs speak in the same sense. The point on which we lay great stress in the foregoing experi- ments is the fact that the stunting is not attributable primarily to under-feeding. Our dwarfed rats have as a rule eaten as adequately as normally nourished animals of the same size. The energy factor, as such, thus drops out of the problem. In this respect the experiments are not comparable with those of Waters and of Aron, both of whom accomplished their results by underfeeding with adequate food mate- rials. In our experiments the ' ' energy requirement for maintenance' ' and the "energy requirement for growth," which together are essen- tial to the developing organism, were both supplied. The rats did not grow primarily at the expense of stored tissue materials : they failed to grow in any sense. We are obviously dealing with some other feature than insufficient energy supply. The numerous illustrative experi- ments which will be cited later are accordingly to be interpreted as instances of maintenance without growth. If it is true that growth can only continue when the energy intake exceeds the mere main- tenance requirement, it is equally true that an excess of calories does not per sc insure growth in a suitable animal. Here then is the opportunity to ascertain and differentiate some of the essential qual- itative factors: protein, inorganic salts, etc. — their minimum and optimum values. EFFECT OF STUNTING ON THE GROWTH IMPULSE. 75 EFFECT OF STUNTING ON THE GROWTH IMPULSE. Before proceeding to study the influence of dietary variations on (a) maintenance and (b) growth, respectively, it became necessary to learn whether a more or less temporary inhibition of growth checks or in any degree modifies the capacity to grow (Wachstumstrieb) . The literature on this subject by no means reveals a unanimity of opinion, although familiar experience will bring to mind many illus- trations of compensated retardation of growth in children.* A few typical experiments may be cited. Rat 36 (male) kept stunted 49 days on a diet of gliadin foodf (37 days) followed by casein food mix- turef (12 days), showed complete recovery of growth on a mixed diet (see Chart XCVI). The "mixed diet of our animals consists of dog biscuit, sunflower seed, and fresh carrots (with occasional changes and addition of lumps of rock salt) . Rat 65 (female) stunted, during 33 dayson adiet of casein-zein food,! likewise resumed anormalrateof growth as soon as the mixed diet was instituted (see Chart XXXVII) . Special interest is attached to experiments in which after a pre- liminary stunting period the resumption of growth was accomplished on a diet containing milk as the effective component. Two protocols of the diet during the stunting period are reproduced in Table XXXV, with reference likewise to Charts XXVIII and XXIX. Table XXXV. Duration of stunting Rat 64 (female), 33 days. Rat 51 (male), 46 days. Stunting diet. Casein *Zein Starch Sugar Agar fSalt mixture I Lard per cent 12.0 6.0 29.5 15.0 5.0 2.5 30.0 Casein Starch Sugar Agar fSalt mixture I Lard per cent 18.0 29.5 15.0 5.0 2.5 30.0 *The zein was hydrated by the addition of a little water. |Cf., p. 86. The curves in these cases are seen to be quite comparable with those of the normally growing rats. Bearing inmind that the animals here studied were continually kept in small cages under actual experi- mental conditions, the "normal" character of the growth curves makes it evident that the environment is no wise detrimental. *Cf. Condereau: Recherches chimiques et physiologiques sur I'alimentation des enfants, Paris, 1869; Pagliani: Giornale della reale societa italiana d'igiene, Milano, 1879, 1. (Quoted by Hatai: American Journal of Physiology, 1907, xvm, p. 320.) fSee p. 122. I See p. 98. Water was added to this mixture until the zein was well hydrated. 76 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Normal growth, as judged by curve of increase in body-weight, was resumed on a diet consisting of per cent. "Trumilk" 60.0 Starch 16.7 Lard 23.3 Similar experiences are shown after feeding gliadin (Charts XCIX, C) or edestin (Chart LXV). In the case of rat 37 (Chart XCVII), a stunting period of 49 days on a diet of gliadin food for 37 days, followed by casein food mix- ture for 12 days, was followed by normal resumption of growth under a dietary regime in which a period of feeding on the above milk-food was alternated with mixed food. Judging by the typical character of the curve of growth in this animal the two types of resuscitation diet, though radically different in origin, are equally efficacious in promoting growth. The growth curve shows little deviation from its usual course incidental to the changes in the dietary. It may be remarked that the early stunting does not neces- sarily impair the capacity to breed at a later period when growth is again established. Furthermore, we have found that the milk-fat- starch mixture continued from early life in no wise impairs the potency of rats as breeders. Its nutritive efficiency will be referred to again. Experiments such as those recorded above give unmistakable evidence of the fact that a considerable period of stunting by no means impairs the " Wachstumstrieb " of these animals. As soon as an appropriate diet is instituted growth begins anew and proceeds with practically the same speed as under normal conditions. By this we mean that a definite increment of gain from some fixed weight requires approximately the same period for its accomplishment as in the case of uninterrupted growth. A rat which will ordinarily grow from 60 grams to 180 grams in body- weight in 60 days will make the same gain even when its growth has been inhibited days or even weeks and its size and form retained at a maintenance level. This will be apparent by comparing, for example, the normal growth curve for both male and female rats with that of the realimented rats, during the same period of time, in Charts CXXII and CXXIII. It should be emphasized that the situation is here quite differ- ent from that developed by Waters and Aron in the experiments on cattle and dogs. With their conditions of underfeeding the animals increase in size (height, etc.) while starving; and during the earlier period of such trials a poorly fed animal may actually gain in height as rapidly as a highly nourished one, fed to the limit of its appetite.* *Cf. Waters: The capacity of animals to grow under adverse conditions. Proceedings Society for the Promotion of Agricultural Science, 1908, xxix, p. 15. EFFECT OF PARTIAL STARVATION ON BODY-WEIGHT. 77 The duration of the period of growth of the undernourished animal depends upon the constitutional vigor of the individual and the store of fat which it has accumulated. Quoting Aron : "Dem Einschmelz- ungsprozess fallt neben dem Fettgewebe in erster Lime die Musku- latur zum Opfer, wahrend die Organe ihm widerstehen, wohl weil sie lebenswichtiger sind." The results of realimentation in animals which show this "dis- proportionate" growth, i. c, growth of one part at the expense of another, are not yet satisfactorily ascertained. Waters believes that physiological compensation may result " by an increase in the rate of growth in a period of liberal feeding following a period of low nourish- ment and low gain. In other words, an animal that is below the normal in size at a given age, through poor nourishment, apparently has the capacity, when liberally fed, to compensate for this loss, in a measure at least, by an increased rate of gain." He also suggests the possibility that growth may be accomplished on a more economical basis — a view which we are not yet ready to accept. EFFECT OF PARTIAL STARVATION ON BODY-WEIGHT. Hatai* has studied the effect of partial starvation followed by normal diet on the growth of white rats. The "partial starvation" consisted in feeding a diet that is practically devoid of protein, viz, starch and water, during 21 days to animals about 40 days old. The realimentation was continued to the age of maturity, at the end of 200 days. The statistics thus obtained and reproduced in Table XXXVI are presented graphically in Chart XXVI. Table XXXVI. — Hatai's Measurements of Underfed and Reaumented Rats. Body-weight. Ratio T •,. , After ImtlaL 21 days. Final. gain. initial and final. Male, controls Male, experimented Female, controls Female, experimented. . . . gm. gm. 35.2 63.1 37.6 28.4 36.3 67.8 34.3 27.0 gm. 224.4 242 .0 1 1 72 - 6 fi67.8 gm. 1 89 . 2 1 204.4 136.3 1335 6.37 6.43 4-75 4.89 Hatai concluded that, as far as body-weight is concerned, "the experimented rats have completely recovered from the effect of 21 days of partial starvation . . . The recovery in the weight is most astonishing, especially during the first 3 or4 days, within which time the starved rats regain the weight lost during the 21 days of starvation. Later the increase in weight is very steady, though not as rapid as during the first few days, until the rat has reached the age *Hatai: American Journal of Physiology, 1907, xvin, p. 310. |The body-weight in both control and experimented is small for the age. 78 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. of 150 days, and after this age increase in weight is relatively slow. What will happen to such rats during the later portions of the span of life has yet to be determined in order to answer the question whether this partial starvation in early life has any influence either on longevity or the onset of old age." (p. 314-315.) EFFECT OF PARTIAL STARVATION ON NERVOUS SYSTEM. Though the period of retarded growth was eventually completely compensated in Hatai's animals, in so far as the weight of the body and central nervous system are concerned, the chemical composition of the brain and spinal cord was not entirely free from the effect. As the result of an extended investigation of the effects of underfeeding on the nervous system, Donaldson* has arrived at the conclusion that one of the characteristics of growth, the change in the water content of the brain, has not been arrested like the increase of the animal in size and body-weight, but apparently accelerated. He states: The underfed group are in this character similar to somewhat older ani- mals. Evidence further points to the continued formation of the medullary sheaths with advancing age even in rats which are underfed, i. e., underfeed- ing does not arrest medullation. Underfeeding which stops growth of the body and retards that of the nervous system does not modify the percentage of water in the spinal cord, while it does reduce it in the brain — the amount of this reduction being less in the cases where the underfeeding is less severe . f With respect to the possible psychological effects of such under- feeding and return to normal diet Donaldson says : So far as our tests show, such an experience does not modify the rat's ability to learn, for, by a series of experiments, it has been possible to deter- mine that such a rat can learn to get its food under complicated conditions just as well and as rapidly as a normal animal (Hayes). t The preceding facts as to resuscitated rats are recorded here — despite the fact that this temporary stunting was produced by under- feeding (rather than unsuitable feeding as in our experiments) — because they suggest that the real story of the condition of the animals may perhaps not be revealed by the external evidences of growth. It is not at all impossible that the rats which we have dwarfed for months may have experienced some continued subtle changes in the make-up of the nervous system despite the appear- ance of unchanged youth which they manifest. Measurements of size and weight alone may not suffice to disclose the real physio- logical status of the animal, especially in respect to the development of the nervous functions and structures, which are singularly pro- *Donaldson: Journal of Comparative Neurology, 191 1, xxi, p. 139. fDonaldson: ibid., p. 169. J Donaldson: Journal of Nervous and Mental Disease, 191 1, xxxvm, p. 262. COMPARISON OP MILK AND MIXED DIET. 79 tected even during starvation. This is seen to be true in the series of stunted animals fed on the glutenin mixture in our experiments (p. 72). There is a large field of investigation still open here with important bearings on the problems of retarded growth in man. According to Donaldson* "the progressive diminution of the per- centage of water in the central nervous system with advancing age is to be regarded as an index of fundamental chemical processes, which take place in the more stable constituents of the nerve cells. These processes are but little modified by changes in the environment and taken all together constitute a series of reactions which express not only the intensity of the growth process in the nervous system, but also the span of life characteristic for any given species." Pos- sibly, then, the further study of the nervous system in connection with our experiments may throw light on the phenomena of malnu- trition (which our stunting experiments primarily represent) as well as those of undernutrition or starvation. It may be well here to note that the experience of Donaldson f indicates the main features of human growth to be well represented in the albino rat. So good is the essential correspondence that there is every reason to continue the work on this form. The striking difference is that the rat grows some thirty times as rapidly as man. COMPARISON OF MILK AND MIXED DIET. The failure either to induce substantial growth in young rats or to satisfy completely the maintenance requirement of older animals during very long protracted periods on the mixtures of isolated food- stuffs thus far reported raises the question as to what constitutes an ideal nutriment for a rat. The suitability of mixed diet is beyond question. The favorable experiences with dried milk powder (some of which have been recorded on pages 75 and 76) early directed our attention to this product. Rats were not only resuscitated after nutritive decline and suitably maintained, but also grown from early age on pastes in which the milk powder (with lard and starch) con- stituted the mixture. The commercial brand " Trumilk"! employed by us has been analyzed at the Connecticut Agricultural Experiment Station withthe following results : Per cent. Water 3-8 Total solids 96 . 2 Protein (NX6.38) 25 . 6 Fat 27.4 Lactose 37-2 Ash 6.0 *Donaldson: Journal of Comparative Neurology, 1910, xx, p. 143. fCf. Donaldson: Journal of Nervous and Mental Disease, 191 1, xxxvn, p. 258. |This product was kindly furnished to us in powder form by the Merrell-Soule Co., Syracuse, N. Y. 80 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. The preparation apparently contains a small excess of iron over that found in cow's milk — probably as a contamination from the desiccating process used. It is obtainable in easily manipulated form and with the addition of a small amount of nitrogen-free lard and starch forms a food paste readily consumed by rats. These pastes have been used, either with or without our earlier standard salt mixture (I),* as follows: "Trumilk" Starch Lard Salt mixture I Nitrogen content. Per cent. Per cent. 60.0 60.0 16. 7 i5-7 23-3 23-3 0.0 1 .0 1 00.0 1 00.0 2.5 2.5 We have carried rats through the period of growth as well as pregnancy on this diet alone, from the time that they were removed from the mother (cf. Charts XXXI, XXXII, and XXXIII). As a further illustration of the excellent nutritive properties and physiologically appropriate ' ' combination ' ' of food ingredients in the milk food-mixture, illustrative charts are appended to show the re- covery of rats moribund after prolonged periods of malnutrition, with lack of inorganic salts in the dietary (Charts XXXIV and XXXV). Many similar illustrations might be reproduced, giving evidence of the perfect realimentation of rats by the use of the milk food (cf. Charts XXVIII, LXV, XCIX, and C). Remembering that our earlier trials with casein, the chief protein ingredient of the milk powder, and with combinations of casein and other proteins were at best successful only in maintaining nutritive equilibrium — and that not indefinitely — and were never adequate for the manifestation of real growth, we directed our attention to the non-protein constituents of milk. After numerous failures to modify the inorganic and non-protein ingredients of our dietaries by altering the relation of proportions of the various ions as well as the character of the carbohydrates and fats, it occurred to us that the protein-free portion of the milk might give the clue to the successful feeding of pro- teins which did not appear to be the inefficient factors in our cases of malnutrition. Accordingly a product was prepared as follows: Perfectly fresh centrifugated milk, nearly free from fat, was pre- cipitated in lots of about 36 liters by diluting with 7 liters of distilled *This mixture, prepared in imitation of Rohmann's successful product and empirically found by use to be the most satisfactory of the different combinations tried, has the following composition: Grams. Grams. Ca3(P04)2 10.0 Mg citrate 8.0 K2HPO4 37.0 Ca lactate 8.0 NaCl 20.0 Fe citrate 2.0 Na citrate 150 100. o (Cf. our previous report, Feeding experiments with isolated food-substances, Publi- cation No. 156, Carnegie Institution of Washington, p. 1,2.) COMPARISON OF1 MILK AND MIXED DIET. 8 1 water which contained 1.64 c.e. of concentrated hydrochloric acid. The flocculent precipitate of casein was strained out on cheesecloth and the very nearly clear solution was filtered through a pulp filter. The filtrate, which at the most was very slightly turbid from sus- pended fat, was tested carefully by the alternate addition of dilute alkali and acid to determine whether any more casein could be sepa- rated from it. The addition of alkali caused a slight precipitate which did not increase on adding more alkali or dissolve on the addi- tion of even relatively large amounts of alkali. This was presumably chiefly calcium phosphate. The addition of acid in no case caused any further precipitation. The filtered milk serum was then heated to boiling for a few minutes and filtered. The filtrate, which was in all cases water clear, was then neutralized to litmus with a dilute solution of sodium hydroxide and evaporated to dryness on a steam bath at a temperature of about 700. The product thus obtained formed a friable, pale yellow mass which was easily reduced to a fine powder by grinding in a mill. Several grams of this powder were tested for protein by dissolving in about 30 c.c. of water containing a little hydrochloric acid and warming gently. The solution was then saturated with ammonium sulphate. The precipitate, which appeared to consist chiefly of calcium sulphate, was separated by centrifugation, dissolved in a little water, and potassium hydrate solution and copper sulphate added. The solution showed no evidence of the biuret reaction until it was saturated with potassium hydroxide and shaken with alcohol. It then separated into two layers, the upper alcoholic layer showing a slight but positive biuret reaction. Millon's reaction tried on portions of 2 or 3 grams of the substance did not give a posi- tive reaction. Nitrogen determinations in several lots of the protein- free milk powder thus made showed them to contain 0.66, 0.59, 0.60, 0.72, 0.71, 0.67, 0.75 per cent of nitrogen. Munk* states that if the proteins of milk are precipitated by alcohol, or separated according to Hoppe-Seyler, from one-thirtieth to one-fifteenth of the protein remains dissolved. All the proteins can be precipitated only by tannin in the cold or by copper hydroxide on heating. He further states that cow's milk contains about one-sixteenth of its nitrogen in non-protein form. Since our protein-free milk powder was equal to 50 per cent of the total solids of the milk, it should, if Munk's state- ments are correct, contain 0.48 per cent of non-protein nitrogen, thus leaving at the most only 0.28 per cent of protein nitrogen, equal to 1.69 per cent of protein. Since 100 grams of the food mixture employed in our experiments contained 28.2 grams of protein-free milk powder, we can assume that at the most the food pastes thus made contained only 0.48 per cent of milk protein. The protein-free *Munk: Virchow's Archiv fur pathologische Anatomic, 1S93, 134, p. 501. 82 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. milk powder thus produced as above described left about 14.5 per cent of inorganic matter on ignition. This includes not only the inorganic constituents of the milk, although by no means in the com- bination in which they occur in the mammary secretion, but also the inorganic salts which were formed by the addition of the hydrochloric acid used to precipitate the casein and also the sodium salts which resulted from neutralizing the milk serum with sodium hydroxide solution. EXPERIMENTS WITH ISOLATED PROTEINS AND " PROTEIN-FREE " MILK. The use of this product (which we shall designate as protein-free milk) as an adjuvant to isolated proteins to furnish the inorganic elements of the diet has succeeded beyond our expectation. When employed, for example, in combination with various proteins, in the proportion in which its ingredients occur in the complete milk food already used (see page 76), it induces normal growth. Added during the periods of nutritive decline to food mixtures which no longer suffice to maintain rats, recovery has manifested itself in practically every case. Where, as in the case of zein, gliadin, or hordein feeding, no advantage has been obtained by the use of the protein-free milk, it has become obvious that the protein per se is the defective food constituent. Thus at length we have found a method of controlling or furnishing some of the most essential non-protein factors in the diet, so that the value of the individual proteins can be investigated under much more favorable conditions than formerly. Numerous charts (see p. 103 fig ) present the graphic records of feeding experiments with casein, edestin,* glutenin,* glycinin,* gliadin,* hordein,* ovalbumin,! and lactalbumin,| showing appropri- ate growth, or maintenance, according to the age at which the animals were started on the use of the protein-free milk as the non-protein component in place of the earlier inorganic salt mixture. It might be objected, after superficial consideration of these re- sults, that the favorable outcome (especially for growth) is due to milk protein contaminating the ' ' protein-free milk ' ' component of the diet. Aside from the fact that the amount of possible contamination is at most small, evidence of the untenability of such a theory is available from several sources. In the first place, growth has not followed the use of all proteins when the protein-free milk was added to them. *For the preparation of these vegetable proteins see T. B. Osborne: Darstellung der Proteineder Pflanzenwelt, Abderhalden's Handbuehderbioehemischen Arbeitsmethoden, 1909, ii, p. 270. fThis was prepared by Hopkins's method and was free from conalbumin. Cf . Osborne, Jones, and Leavenworth: American Journal of Physiology, 1909, xxiv, p. 252. JThe preparation of this is described on p. 81. ISOLATED PROTEINS AND " PROTEIN-FREE " MILK. 83 The results can be grouped in two series, viz : Diet = Isolated protein, protein-free milk, starch, agar, fat. Group I. — Young rats. Group II. — Young rats. Active growth with — Little or no growth with — Casein (Charts xlvi, xlvii, lii, liii, Gliadin (Charts cvnr, cix, ex, cxi, liv, LV, lvi, LVTi, LVin, ux, and LX. cxir, cxin and cxiv). Ovalbumin (Charts xc and xci). Hordein (Charts cxxiv and exxv). Lactalbumin (Charts xcn and xciii). Edestin (Charts lxxi, lxxii, lxxiii, lxxiv, lxxv, and lxxvi). Glutenin (Charts lxxxvii, lxxxviii, and lxxxix). Glycinin (Charts xciv and xcv). The failures in group n lead to the conclusion that the proteins, gliadin and hordein, are inadequate for the functions of growth. We are presumably dealing with a chemical inadequacy rather than any toxicity and consequent lack of growth, judging by the fact that the gliadin and hordein rats are maintained in good, nutritive condition even in the absence of growth . Their body-weight is scarcely changed at all. Without the use of the protein-free milk or faeces-feeding gliadin rats have usually declined (Charts XCVIII, XCIX, and C). A second reason why the success of these trials is not due to the presence of possible minute contaminations with milk protein is discoverable in Charts XLJII, XLJV, XTV, XLVIII, XLIX, L,, LI, CVIII, CIX, CX, and CXI. Here the addition of not inconsiderable portions (5 to 30 per cent) of the actual milk food to the earlier inefficient protein mixtures is incapable of bringing about growth in any degree equal to that at once initiated when the protein-free milk is added in relative abundance. Further evidence that a trace of milk proteins is not responsible for the growth of the rats fed with mixtures containing our protein- free milk powder is furnished by experiments in which successively larger quantities of the milk food are added to the gliadin food. Here we see that growth gradually increases with the larger additions of the milk food, although with even as much as 30 per cent in the food the rate of growth is much below normal. With additions of 5 or even 20 per cent of the milk food, the rate of growth is very slow, as shown by Charts CIV, CV, CVI, and CVII. That this result is to be attributed to the proteins introduced in the milk food and not to a combination of a small quantity of milk proteins together with a sufficient quantity of the inorganic or other constituents of the milk is shown by experiments now in progress in which the addition of the milk food to the gliadin and protein-free milk food is producing normal growth. In this mixture we have all of the constituents of 84 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. the protein-free milk present in the same proportions as in our milk food, but less than one-third of the protein constituents of the milk. It is therefore evident that only a small proportion of the protein constituents of the milk are required to produce normal growth, and it may be assumed that the presence of a small quantity of milk proteins in our protein-free milk powder would manifest itself by at least some slight growth. DISCUSSION OF THE RESULTS AND THEIR BEARINGS. We have stated that by our plan a biological comparison of dif- ferent proteins in respect to their role in growth can at length be made. Our work in this direction must be regarded as barely begun. Nevertheless it is of interest to speculate as to the indications already gained and the outlook for future work. A comparison of the two groups of proteins — those adequate and those inadequate for growth purposes — at once reveals the fact that the latter category comprises proteins (gliadin, hordein, zein) commonly spoken of as chemically "incomplete." They lack one or more of the amino-acid complexes which are obtainable from the so-called "complete " proteins. None of them furnish glycocoll or lysine, and zein in addition is devoid of tryptophane. By feeding relatively small quantities of proteins like casein with gliadin growth begins at once. Here we can determine the minimum of suitable protein to satisfy this growth requirement — a study already begun (cf. Charts CXX, CXXI, CXXII, and CXXIII). The addition of amino-acids to "complete," as it were, the inadequate proteins can now be studied amid controllable factors ; the biological role of hydrolyzed proteins and the significance of complete hydrolysis or digestion in nutrition can be examined anew. The experiences which have demonstrated the striking differ- ences in value of the individual proteins and the small proportion of casein which suffices to induce growth instead of standstill (cf . Charts CXX, CXI, CXXII, and CXXIII, for example) emphasize the impor- tance of the purity of the protein fed. We have devoted much labor and incurred a very considerable expense to obtain proteins in a form as uncontaminated as present methods will permit. The products used were as pure as one would expect them to be if employed for purposes of refined protein analysis. Had less perfect products been employed it is quite conceivable and indeed likely that the admix- tures would have sufficed to alter completely the outcome of many experiments. For example, gliadin is prepared free from glutenin only be very careful purification methods ; and although the nutritive properties of these two companion proteins are extremely unlike, as clearly indicated by our trials, a failure to effect a complete separa- tion of a little glutenin from gliadin would have been sufficient to prevent the deficiencies of the latter from exhibiting themselves. Or DISCUSSION OF THE RESULTS AND THEIR BEARINGS. 85 again, failure to purify carefully a protein like casein will vitiate the study of a problem like the synthesis of amino-acids. Pure casein is glycocoll-f ree ; and the continued feeding of such a product as the sole protein of the dietary enables one to make deductions respecting the synthesis of glycocoll. The use of crude commercial protein preparations can never satisfy the requirements of refined study in this domain, where small effects continued over long periods are of great importance. We believe, therefore, that such considerations justify the energy and expense which have been put into the work. In relation to the much-discussed problem of the relative value of organic vs. inorganic phosphorus in nutrition, our data after feed- ing phosphorus-free edestin to growing rats (cf . Charts LXXV and LXXVI) show a success quite as great as that with phosphorus-con- taining casein (cf. Charts LVI, LVII, LVIII, UX, and LX). The animals must here have synthesized their phosphorus compounds from inorganic phosphorus. Whether milk production and other functions calling for such synthetic reactions will continue adequately is open to investigation. It is also noteworthy that all of our animals grow on a dietary that is purine-free, or essentially so. Here the ques- tion of purine synthesis suggests itself. It is apparent, e. g., in the case of gliadin, that the grown as well as ungrown rats may be main- tained for long periods on single proteins. With such an ideal non-protein dietary component at hand amino-acid substitutions can be attempted in the adult as well as in the growing animal. The protein minimum (or minima) is also open to accurate investigation. With a method of feeding devised which will permit a differentiation between growth and maintenance, which furnishes an energy-yielding protein-free component that is appro- priate, and leaves the protein as the sole variable in the dietary, we believe that further contributions can be made to the problems of nutrition. In the preparation of the large quantities of carefully purified proteins required for these experiments, we have been assisted by Mr. Charles S. Leavenworth, Mr. Owen Nolan, Mr. Leigh I. Hol- dredge, and Mr. Lawrence Nolan, whose valuable cooperation we are glad to acknowledge here. 86 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. THE CHARTS AND THEIR EXPLANATIONS. In the following charts, to which reference is made in various places in the text, the abscissae of the curves represent days and the ordinates actual body- weight (solid line) or food-intake (dotted line) in grams. In some of the charts the average (normal) curve of growth, plotted from body- weight data available for normally grow- ing animals of the same sex, is represented by a broken line for com- parison. The food-intake curve is plotted from the quantities of food eaten per week. The numbers on the body-weight curves indicate the time at which changes in the character of the feeding were insti- tuted. All curves in this paper are plotted on the same scale, so that they are directly comparable. Salt mixture I, to which reference is frequently made, was composed of — Grams. Ca3 (P04)2 io.o K2HPO4 37 -o NaCl 20 . o Na citrate 15.0 Mg citrate 8.0 Ca lactate 8.0 Fe citrate 2.0 100. o INDEX OF CHARTS WITH REFERENCE TO FOOD-MIXTURES AND PROTEINS FED. [Numbers refer to pages in the text.] Casein, 93, 96, 97, 99, 100, 101, 103, 104, 105, 106, 107, 108, 122, 123, 133, 134. Casein +glutenin, 94. Casein +legumin, 97. Casein+milk, 102, 104. Casein -f-zein, 92, 98. Edestin, 109, no, in, 112, 113, 114, 115, 116, 117. Edestin+milk, 112, 113. Feces, 99, 100, 101, no, 111,115,125,126. Gliadin, 122, 123, 124, 125, 126, 128, 129, 130, 131. 132, 133, 134- Gliadin -|-milk, 127, 128. Glutenin, 94, 119, 120. Glutenin-f-edestiu, 94. Glycinin, 121. Hempseed, 91. Hordein, 135. Lactalbumin, 121. Milk, 92, 93, 95, 96, 97, 109, 118, 123, 124. Mixed food, 87, 88, 89, 90, 94, 96, 97, 98, 99, 100, 101, 106, 108, 112, 115, 116, 117, 118, 122, 123, 125, 126, 130, 131, 132, 136, 137, 138. Ovalbumin, 120. Protein-free milk, 94, 101, 103, 104, 105, 106, 107, 108, 112, 113, 114, 115, 116, 117, 120, 121, 128, 129, 130, 131, 132. 133, 134. 135, 137, 138. Zein, 136, 137, 138. CHARTS AND THEIR EXPLANATIONS. Chart XXII. \ \ \ •~\ c\ o\ lD\ "ci ^2\ ail S\ o\ col o\ CI o Dl 0)| e\ o\ si \ \ -i •* -& . \ % 1 • 0 20 40 60 80 100 120 140 160 180 Days Charts XXIV (rat 166, female) and XXV (rat 156, female) show the typical growth of female rats, including pregnancy, under cage conditions. The animals were fed on mixed food. Chart XXV. 90 FEEDING EXPERIMENTS WITH ISOLATED EOODSUBSTANCES. Chart XXVI 260 240 220 200 180 160 140 120 100 80 60 40 20 Male starved /i Male control Female control ^ * -" it starved s ' , - ■* /•'* // / ■■ .■' / / / , ' / / ' / / / * / , / / / > / / / * / / / / / ' / / // / /. f f '/ / // J / / / / ,i - / / ,f t/ y // */ 'I /> t // t/ f ^T>v 1 c ... ,_ j -■ ■ Period o f* starvation 20 40 60 80 100 Days 120 140 160 200 Chart XXVI (curves, from Hatai, American Journal of Physiology, 1907, xviii, p. 311) shows the body-weights of albino rats at different ages. C, conception, and O, the date of birth 2 1 days after conception. An illustra- tion is given of the influence on growth of a period of 2 1 days of starvation during early life. CHARTS AND THEIR EXPLANATIONS. 91 Chart XXVII. ,0 ■v^ O O O O O O to . O O O 00 O N ^ h 10 ro ■* -^OOOOOO to . O 0 0 0) 0 N ■^ m ^1- ^f ro ~ 0 0 0 0 0 0 . O OO N W O £h ^ IH ^f ^f ei ~ 0 0 0 0 0 0 to .000000 ^,t-ro M -: -*no 0000 . CO M O O O O S PM «,«f HH W V c 1) 3 3 a 3 O 3 4J X Cfl 00 "2 E bOv 0 g u » sr: 0 0 *d o o re In >-. O Gj d ° TO o co <+H ■ »-( 4J bJO to • *-• iu d TO ^ to O >■, 4-> IS . T-l co CO O a cu 43 +-> CO o ,d CO s 00 S CU ■(-> TO ft 42 * re TO +j O 43 H> n aj xn . 43 £_ aH 3 ft d > X £.2 t3 43 a) CO CO ft re ffl •d d d TO Td Ui a> ,o ^^ d c re a d d d w d 92 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart XXVIII. O SLUEJO ro ■** o r* o ro u. . O u~) m r^ Pu, ■^•O m M 013 ^ONOfO "S M . O O © CO j^,o m tN PU « t/i HH ■m 9 M 3 4-1 a o O Trumilk Starch . Salt mix Lard... H ■^OOi/iOOino . oi so ainv;N o ^ hi (Mm CO trt >— i V q QJ 3 3 *J ■4-» G 0 U .5 & 4, tarch ugar. gar. . alt m O N t/3 t/} < 73 hJ cu *d bJO o ^ ,_ d d be cu cu -M .3 +J • d 03 a +J d S|3 03 H to o '3 * d s 2o »*-! -TH CU ai d . d 5ft w d cu >, £ K c? . >vd 5"*° o ^ &*■ CU 03 O a3 > XI X cu cu n 03 <" a) U -a d 'S <" d 03 im M ^ « 9 ^ 03 rt ^ a,.- dji^5 •sd o •d o t-H 0 E^ *a o o^X '^ ^ _- +-1 <» W T3 <4_ a w .2 o Ui o3 >-i Xfl cu CU -i_> +-» s were longp stituen 03 '•B -d cu rj cu O XI o c; V ^ o +j CU be o d a d -a • *H ■r" d u< eral dui orga "3 be 03 > ^ c cu £f-d u O w d m 03 e CU X -g CU be ■£ ° be£ ing ntial ifyin be d c u dur esse mod o cu t/) cj o •^,a c d 1U jj -rt en a3 T3 -m t3 o3 cu cy j3 +-> ,^d ^.S5 > c CU u -4-1 C o d »\ v u CU PL, N M tn V a S3 3 M ti '■*-> en a 0 3 a H b O ^(Bffiht H -:ouioo«o .00 0\ v> lO CN O Ph ^ m CO 03 L- o H be •d cu 05 o 2 +^ CO >, oj cu • t-H •d fl O o a o o _*d CU XI s-s o XI CO "cu 3 (N CO O in QJ Oh CU X u ° be >-> +-> X t-H X XI +-> L. o3 X a CO >■> s o3 "-1 -d a o o -rx — ^ CO O d cu a oj o3 ^ (U > o u OJ d o *d d o o I) u u CO X 03 CU L. -4H d CU cu oj ? X CO &A d o X d o d o »j c d c a d CU Ih 03 Oh a d a3 o 03 94 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart XXX. CHARTS AND THEIR EXPLANATIONS. 95 Chart XXX (rat 71, male) shows long-continued feeding of isolated foodstuffs and also long-continued maintenance on glutenin from wheat as the onlyprotein. The history of the animal is on p. 59. The diets were as follows: Constituents. Glutenin Casein Starch Sugar Agar Salt mixture I Lard Per. 1. Per. 2. p.ct. p.ct. 6.0 6.0 12.0 12.0 29. S 24-5 150 15.0 SO SO 2.5 2.5 30.0 35-0 Pei . 3. p. ct. 16.36 0.0 22.27 13- 63 4-54 2.27 40.9-3 Per. 4. p. ct. 18.0 0.0 14-5 ISO SO 2.5 450 Periods 5 and 8. p. ct. 18.0 0.0 34-5 20.0 5.0 2.S 20.0 Constituents. Glutenin Edestin Starch Sugar Agar Salt mixture I. Lard Per. 6. Per. 7. p. ct. 9.0 Mixed 90 food. 335 18.5 5.0 2.5 23. S Constituents. Glutenin Protein-free milk . Starch Agar Lard Per P. P. :t. 18 0 28 2 23 8 5 0 25 0 Chart XXX further shows the possibility of maintaining an animal satisfactorily under our cage conditions for 458 days. Attention is particu- larly directed to period 9, during which the only change in the diet consisted in substituting protein-free milk for some of the non-protein components of the dietary. The lowest line represents the nitrogen balance of the rat. Chart XXXI. Chart XXXII. terminated 220 200 Chart XXXI (rat 222, male) shows early growth curve of male on milk diet, having the following composition: Trumilk, 60.0 p. ct.; starch, 15.7 p. ct. ; salt mixture I, 1.0 p. ct. ; lard, 23.3 p. ct. Chart XXXII (rat 195, male) shows normal growth curve of male on milk diet, having the following composition : Trumilk, 60 p. ct. ; starch, 16.7 p. ct.; lard, 23.3 p. ct. 96 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart XXXIII. 170 150 130 110 90 70 E bO n if- W" >/ B°dV « ■ / / / / ■ s V^ / // // \/ // // ._ ^ 1i!k foot 20 40 60 80 100 Days 120 140 I60 180 Chart XXXIII (rat 181, female) shows growth and normal pregnancy of female on milk food, consisting of Trumilk, 6o p. ct.; starch, 15.7 p. ct.; salt mixture I, 1.0 p. ct.; lard, 23.3 p. ct. Chart XXXIV. terminated 240 220 200 200 Chart XXXIV (rat 106, male) shows malnutrition induced by lack of inorganic salts in the dietary and subsequent perfect recovery on milk-paste. The diet was mixed food for period 1 ; for the remaining periods as follows : Constituents. Per. 2. Per. 3. Constituents. Per. 4. P.ct. 18 .... 25 to 32 . 5 17 29.5 0 5.0 p. ct. 18.0 32. 5 21.9 0.0 2.6 25.0 p. ct. 60.0 157 1 .0 23-3 Starch Salt mixture I. . . . Salt mixture I Lard 0 0.0 20 35-0 CHARTS AND THKIR EXPLANATIONS. Chart XXXV. 97 200 180 160 140 120 60 40 ) \ 2 r^ Hermin. 'l \S V f— Mix<"d food -x ;in food - >ut salts ._^_. with sails / ilk food -> with 3 \ 4 ^ _ od eate n \ V .''"" - \ - 20 40 60 80 100 120 Days 140 160 180 200 Chart XXXV (rat no, female) shows malnutrition induced by lack of inorganic salts in the dietary and subsequent perfect recovery on milk-paste. The diet consisted of mixed food for period i , and as follows for the remain- ing periods : Constituents. Per. 2. Per. 3. Constituents. Per. 4. p. ct. 18 .... 25 to 32. 5 17 29.5 0 50 0 20 350 p. ct. 18.0 32.5 21.9 0.0 2.6 25.0 Trumilk p. ct. 60.0 15-7 1.0 23 3 Starch Salt mixture I . . . . Lard Agar Salt mixture I Lard Chart XXXVI. 30 B )dy weig ht Z A "\ 20 fc-- -Ca: ein food 0 0 E QJ -J in £ + a) O dead; cause unknown 20 40 Days 60 Chart XXXVI (rat 54, male) shows the maintenance for 46 days of a very small rat, without growth, on a diet in which casein formed the sole protein. The composition of the food was as shown herewith: .Constituents. Per. 1. Per. 2. Casein p. ct. 18.0 0.0 29-5 150 5-0 2.5 30.0 p. ct. 9.0 9.0 29-5 15.0 5.0 2.5 30.0 Pea legumin Starch Sugar Agar Salt mixture I . . . Lard 98 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart XXXVII. Chart XXXVII (rat 65, female) shows stunting for 33 days during early life, followed by normal growth and preg- nancy on mixed food. In addition to the typical growth during 317 days, the curve emphasizes the unaltered "ca- pacity to grow " after stunting by improper diet. The diet during period 1 was as shown herewith. p. a. Casein 12.0 Zein 6.0 Starch 2Q.5 Sugar 15.0 Agar 50 Salt mixture I. 2.5 Lard 300 CHARTS AND THEIR EXPLANATIONS . 99 Chart XXXVIII. Oead.causei unknown Chart XXXVIII (rat 50) shows main- tenance for 158 days on a diet in which casein formed the sole protein. The com- position of the food was as shown here- with: p. ct. Casein 18.0 Starch 29.5 Sugar 15.0 Agar 5.0 Salt mixture I. 2.5 Lard 30.0 20 60 80 100 120 Days "' 140 160 Chart XXXIX. 140 ■N-— -2 Body v\ eight 120 /W 1 3 100 A - Casein r , ~ma! fae 80 60 1 70 50 30 / / <*c ^ / IUU / / / 0s- tf- OO TS O . 80 60 0~3 £- / - / ,™5 / X A, O J 0. / / 40 0 20 40 60 80 100 120 140 160 180 Days Chart'XLVII. Charts XLVI (rat 177, female) and XLVI I (rat 191, male) show mainte- nance on a diet in which casein formed the sole protein during 83 days followed by growth when protein-free milk was substituted for a part of the non-protein constituents of the diet. The diet was as shown herewith. Constituents. Casein Protein-free milk . Starch Sugar Agar Salt mixture I . . . . Lard Per. 1. p. ct. 18.0 0.0 325 170 tO 20.0 50 2.5 22.0 25.0 Per. 2. p. ct. 18.0 28.2 23.8 0.0 SO 0.0 25.0 104 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart XLVIII. Chart XLIX. Charts XLVIII (rat 210, female), XLIX (rat 209, male), L (rat 215, male), and LI (rat 216, male) show inadequate growth, during period 1, on the casein food with a small admixture of milk, fol- lowed by resumption of growth on a diet containing casein and protein-free milk in a quantity equivalent to that of our milk- paste diet which has proved sufficient to promote normal growth. The compo- sition of the food was as shown in table. Chart L. 140 120 100 60 60 40 zo / / / / t 0 f/C J T7^ .Casein f Milk 3od90% a 10% "if 1 7 <-- Ca ein + Pro tein-frec milk-—* k / "ood eat en — t \ \_ ' / ,00 120 Constituents. Per. 1. Constituents. Per. 2. Casein food (casein, 18.0; starch, 32.5; sugar, 17.0; agar 5.0; salt mixture I, 2.5; p. a. 00 10 Protein-free : Starch Agar Lard p. ct. 18.0 28.2 238 5-0 25.0 Milk food, (Trumilk, 60.0; starch, IS-7; salt mixture I, 1.0; lard, 23.3) Chart LI. 20 40 60 80 Days 100 120 CHARTS AND THEIR EXPLANATIONS. I05 Chart LII. 160 140 120 100 BO 60 *,' c • << Body *e'ghl O T3 r- 1> ___' // ]/ n, Edestir alternat // - - /■/- - Cc // / / se n + Pr Dtein -free milk - - ... > 4— /? ■ ---Case n i-Prote n-free m ilk I0U 80 60 tn E 0 1/ 0 20 40 60 80 100 120 140 m Days i Period 2. p. ct. Casein 18.0 Protein-free milk. 28.2 Starch 23 . 8 Agar SO Lard 25.0 Chart LIV (rat 204, female) shows uninterrupted growth when a diet of isolated food-stuffs containing casein as its sole protein was substituted for mixed food. The requisite inorganic salts were furnished in the added protein-free milk. The experiment is of exceptional inter- est inasmuch as the animal successfully passed through two periods of pregnancy on a purine-free food containing a single protein. This obviously affords a method of studying various synthetic processes in the animal body. The diet during period 1 consisted of mixed food. During period 2 as shown herewith. Chart LV (rat 203, male) shows uninterrupted growth when a diet of isolated foodstuffs containing casein as its sole protein was substituted for mixed food. The requisite inorganic salts were furnished in the added protein-free milk. The diet during period 1 consisted of mixed food; during period 2, as shown herewith. Period 2. p. ct. Casein 18.0 Protein-free milk . 28 . 2 Starch 23.8 Agar 5-0 Lard 25.0 CHARTS AND THEIR EXPLANATIONS. 107 Chart LVI. Chart LVII. 140 120 100 80 60 40 i c--'- Casein + Protein-free milk > 20 40 60' 80 100 Days Chart LVIII. 20 40 60 Days Chart LIX. Chart LX. Charts LVI (rat 238, female), LVII (rat 269, female), LVIII (rat 247, male), LIX (rat 252, male), and LX (rat 268, male) show a normal growth on a diet containing a single protein, ca- Casein is.o sein. The requisite inorganic salts were furnished in the SSS^?.^". H'.l added protein-free milk. This experiment illustrates arti- Ajj* 2S-» ficial nutrition with isolated food-substances from a very- early period of life. The diet was as shown herewith. i Io8 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. 200 180 160 140 120 100 BO 60 40 20 Chart LXI. Body w eight r^ 1 ' z/*/ I *£ ^ > / 1 / r / 'V V / ^ J 4 fl ^ / / s l \ Food eaten ' \ \ \ i i I I v^ -'"' I v / \ 1 I 20 40 60 80 100 120 140 160 180 200 220 240 260 Davs Chart LXII. CHARTS AND THEIR EXPLANATIONS. 109 Charts IyXI (rat 141, female) and LXII (rat 139, male) show recovery of animals maintained on a diet containing casein as the sole protein. The preliminary nutritive condition of the rats is shown to be satisfactory in period 1 on a mixed diet. The ultimate decline on the casein diet during period 2 could not be checked by increasing the content of casein during period 3. This shows that the nutritive failure of the animals was not attributable to the protein per se. Speedy recuperation and maintenance attended the substitution of protein- free milk for the inorganic salt mix- ture contained in food previously used. Note the influence of this dietary change on the appetite of the animals. In period 1 mixed food was used. The composition of food, during the other periods was as shown in table. Chart LXIII. Constituents. Per. 2. Per. 3. Per. 4. Casein p. ct. 18.0 0.0 32. S 21.9 to 26. 9 0.0 S.o 2.5 2.6 20 . 0 25 . 0 p.ct. 36.0 0.0 22.5 13-9 0.0 2.6 25.0 p. ct. 18.0 28.2 23.8 0.0 5.0 0.0 25.0 Protein-free milk Starch Sugar Agar Salt mixture I. . . Lard terminated Charts LXIII (rat 60, male) and LXIV (rat 58, female) show maintenance and slight growth of a rat on a diet in which edestin constituted the sole protein for 67 days. The experiment was ter- minated because of the death of another animal, which was found partly eaten, in the same cage. The diet was as shown herewith. p. ct. Edestin 18.0 Starch 29.5 Sugar 15.0 Agar 5.0 Salt mixture I. 2.5 Lard 30 . 0 Chart LXV. terminated Chart LXV (rat 189, female) shows failure of rat to grow or be maintained on a diet containing edestin as the sole pro- tein during 72 days (period 1). There is no loss of capacity to grow, as will be seen by the curve of growth on the milk diet in period 2,32 days. The diet consisted of — Period I. p. ct. Edestin 18.0 Starch 29 . 5 Sugar 15.0 Agar 5.0 Salt mixture 1 2.5 Lard 30.0 Period 2. Trumilk 60.0 Starch 15 . 7 Salt mixture 1 1.0 Lard 23. 3 no FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart LXVI. 140 120 100 60 60 E Edestin food- ■-k - died; cause Ldestin food + normal faeces- 4 unknown 20 40 60 80 Days 100 120 140 160 Charts I/XVI (rat 169, male) and I^XVII (rat a p. a. 190, male) show maintenance on a diet in which starch";.'.'.'.'.'.'.'.'. 2^5 to 32^ edestin formed the sole protein. The influence of v^u^r xs-o 170 faeces of normally fed animals in preventing de- salt mixture, 1..'. 2.5 ...'. cline in body- weight for some time is shown during Lar ' ' period 2. The faeces were obtained from rats temporarily introduced into the cage each day. The diet is given above. Chart LXVII. 60 80 Day 5 CHARTS AND THEIR EXPLANATIONS. Ill Chart LXVIII. 80 100 Days Charts LXVIII (rat 196, female) and LXIX p ct (rat 193, female) show maintenance on a diet in Edestin 18.0 '.... which edestin formed the sole protein. The influ- sugar.'.'.'.!!"! is!ot0i7!o ence of faeces of normally fed animals in preventing sa^inixturei'" is "" decline in body-weight is shown during period 2. Lard '.'.'.' 25^0 30.0 The giving of faeces was discontinued during period 3. The faeces were obtained from normally fed rats temporarily intro- duced into the cage each day. The diet is given above. Chart LXIX. 112 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart LXX. Period 2. p. ct. Edestin 18.0 Starch 29 . 5 to 32 . S Sugar 150 17.0 Agar 50 Salt mixture I 2.5 Lard 25.0 30.0 120 140 Days Chart LXX (rat 133, female) shows mainte- nance on a diet in which edestin was the sole protein during 161 days. Period 1 on a mixed diet shows normal growth. Period 2 is of interest because the food was also purine-free and devoid of organically combined phosphorus. All growth ceased during the edestin feeding (period 2), in contrast with other experiences where protein- free milk was present in the dietary. Chart LXXI (rat 218, female) shows inadequate growth on a diet of edestin + milk-paste (period 1) followed by growth during period 2, chart lxxi. in which the food contained pro- tein-free milk and edestin as its sole protein. In growing to several times its original weight the animal must have synthesized its purine- and phosphorus-containing com- plexes from purine-free food. The influence of size on food require- ment is shown by the food-intake curve. The diet consisted of — Period I. Edestin food (edestin, 18 0; starch. 32.5: sugar, 17.0; agar, 5.0; salt p. ct. 90 Edest'.n food ?.Q%^_ . . toest,n + Protein-free milk- Milk u 1 0 % mixture I, 2.5; lard, 25. 0) Milk food (Trumilk, 60.0; starch J5-T, salt mixture I. 1.0; lard, 23.3.) 10 Period 2. Edestin 18.0 Protein-free milk 28 . 2 Starch 23. 8 Agar 5 0 Lard 25.0 20 40 60 80 Days 100 120 CHARTS AND THEUR EXPLANATIONS. Chart LXXII. 160 140 120 100 80 60 40 ii3 Chart LXXIV. / 40/ ^ r~ o J;/ $6 J ^1? 7 (l II i J 1 II 1 1 / / 2 Food eaten r-^ 1 / Y Edestin r Milk rood 90% /"-v/ ic 10 V ' / '' \ \ r.~~ «r -Edestin + Protein-free milk---* i ■ Chart LXXIII. 190 170 150 130 110 90 / / / / / / r-1 fa J^ A A// y / y 70 50 30 E & o J 2 "ood eate n /-«. N / ^Edestin f MJk ood90% u 10% H- / Ldestin + f 'rotein-fr semilk-* 0 20 40 60 80 100 120 Days Charts LXXII (rat 217, male), LXXIII (rat 211, male), and LXXIV (rat 212, male) show inadequate growth on a diet of edestin + milk-paste (period 1) followed by growth during period 2, in which the food contained protein- free milk and edestin as its sole pro- tein. It should be noted that the ani- mals in growing to several times their original weight must have synthesized their purine- and phosphorus-containing complexes from purine-free food. The influence of size on the food require- ment is shown by the food intake curve. The diet consisted of — ■ Period I. Edestin food (edestin, 18.0; starch, 32.5; sugar, 17.0; agar, S.o; salt mixture I, 2.5; lard, 25.0) Milk food (Trumilk, 60.0; starch; 15.7, salt mixture I, 1.0; lard,, 23.3) Period 2. Edestin Protein-free milk. Starch Agar Lard p. ct. 90.0 10. 0 18.0 28.2 238 S.o 250 20 40 60 Days 80 100 120 ii4 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart LXXV. Chart LXXVI. p. ct. Edestin 18.0 Protein-free milk . 28 . 2 Starch 23.8 Agar 50 Lard 25 . 0 Charts LXXV (rat 248, female) and LXXVI (rat 253, female) show growth from an early age on a diet containing protein-free milk in which edestin formed the sole protein. It should be noted that the animals in growing to several times their original weight must have synthesized their purine- and phosphorus-containing complexes from purine- free food. The influence of size on the food requirement is shown by the food-intake curve. The diet was as shown herewith. Chart LXXVII (rat 114, male) shows the failure of edestin (period 2) to maintain previous satisfactory nutritive condition of the animal during period 1, on mixed food, even after adding faeces to the diet (period 3). Immediate improvement and satisfactory nutritive condition followed addi- tion of protein-free milk to edestin food (period 4) . The diet consisted of mixed food for period 1, and for periods 2,3, and 4 was as shown in table. Constituents. Periods 2 and 3. Constituents. Per. 4. p.ct. 18.0 29.5to32.S ISO 17.0 s° 2.5 25.0 30.0 Edestin p. ct. 18.0 28.2 238 5-0 25.0 Starch Protein-free milk .... Starch Agar Agar Salt mixture I Lard Constituents. Per. 2. Per. 3. Chart LXXVIII (rat 140, female) shows the failure of maintenance on a diet in which edestin formed the sole protein (period 2), until protein-free milk was added to the diet (period 3). Period i, on mixed food, is introduced to show the previous satisfactory nutritive condition of the animal. The diet consisted of mixed food for period 1, and for periods 2 and 3 it was as shown in table. p. ct. Edestin 18.0 Protein-free milk ... . 0.0 Starch 29 . 5 to 32 . 5 Sugar 15.0 17.0 Agar . Salt mixture I . Lard 5.0 2.5 25. 0 30.0 p. ct. 18.0 28.2 23.8 0.0 5-0 0.0 25.0 CHARTS AND THEIR EXPLANATIONS. 115 Chart LXXVII. 280 260 240 220 200 180 160 / ' /v s/V-^ J 2 A ' ■^\ E ody wei ght / ^ O 0) E 4 1/ / c ■D O O <4- / « Mixs - Edesti i food- u ■0 LkJ rotein-free milk- --> 140 70 50 in E TO t_ O V F ood eat en V 1 i / ^ ,' \ / t V-~* / ~\ / / / 0 20 40 60 80 100 120 140 160 180 200 220 240 260 Days Chart I,XXVIII. 20 140 Days 220 240 260 Il6 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart LXXIX. 180 / ~~A Body weight f V k 160 A.^ v/ / / 3 140 120 Ay r r\J KT 100 ) 80 1 60 60 \ s. Food eaten - \ / / / / / -• \ / 40 N / ^— - E ra id 20 40 60 80 100 120 140 Days 160 180 200 220 240 260 Chart LXXIX (rat 152, female) shows the failure of maintenance on a diet in which edestin formed the sole protein (period 2), until protein- free milk was added to the diet (period 3). Period 1, on mixed food, is introduced to show the previous satisfactory nutritive condition of the animal. The diet consisted of mixed food for period 1 and for periods 2 and 3 was as follows : Constituents. Edestin Protein-free milk. . . Starch Sugar Agar Salt mixture I Lard Per. 2. Per. 3 p. cl. 18.0 0.0 29St0 32.5 ISO 17.0 S.o 2.5 25.0 30.0 p. a. 18.0 28.2 23.8 0.0 5.0 0.0 2S.0 CHARTS AND THEIR EXPLANATIONS. 117 Chart LXXX. 120 300 280 260 240 220 200 180 160 140 120 100 80 60 E a u O r v\ A. / /z \ /— — — — — _ — - ** ^"" / ^T / / / 4 V Boc y weight J 4 ^ / h P 3 / / ■'Mixed f Eld estin foe d -- - " Edestin + Protein -free mil * > ( 1 *. 1 1 1 1 / / A 1 \ 1 v __/ \ / I 1 1 \ \ V \ f 1 \ / / / / / / \ \ Fc od eater 1 1 1 1 1 1 / / // \ 1 1 < 1 \ / V 20 40 60 80 100 120 140 Days 160 180 200 220 240 26C Chart LXXX (rat 148, male) shows the failure of maintenance on a diet in which edestin formed the sole protein (period 2), until protein-free milk was added to the diet (period 3). Period 1, on mixed food, is intro- duced to show the previous satisfactory nutritive condition of the animal. Note the influence of changes in diet on the food consumption. The diet consisted of mixed food for period 1 , and for periods 2 and 3 was as follows : Constituents. Per. 2. p. ct. Edestin 18.0 Protein-free milk .... 0.0 Starch ' 20 . 5 to 32 . 5 Sugar I 15.0 17.0 Agar J 5-0 Salt mixture 1 2.5 Lard 25.0 30.0 Per. 3. p. ct. 18.0 28.2 23.8 0.0 5-0 0.0 25.0 Il8 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart LXXXI. 20 140 Chart LXXXII. Killed for measurement 60 80 Days Charts LXXXI (rat 96, female) and LXXXII (rat 97, female). Control animals for the glutenin dwarfs, Charts LXXXI V-LXXXVI. For other data see page 73. The diet consisted of Trumilk, 60 p. ct.; starch, 15.7 p. ct. ; salt mixture I, 1 p. ct., lard, 23.3 p. ct. Chart LXXXIII. 170 150 130 no 90 70 in •£ id i_ o / j / killed ^ meas urement V Qj I ^ ^c si 1 / / / / / ~» tj 0 20 40 60 8 J2ays 0 100 120 140 Chart LXXXIII (rat 99, male). Control animal for the glutenin dwarfs, Charts LXXXI V-LXXXVI. For other data see page 73. The diet consisted of mixed food. CHARTS AND TH^IR EXPLANATIONS. 140 Chart LXXXVI. Killed. for measurement Charts LXXXIV (rat ioo, female), LXXXV p. a. (rat i o i, male), and LXXXVI (rat 1 02, male). These starch*.'.''.'.'.'.'.'.'. i4.sto34 animals, from the same family as the control rats, |us*r js ° 2° Charts LXXXI-IyXXXIII, were maintained on a Salt mixture i.'.'.'. 2.5 diet of glutenin from wheat 124 days, when they were " 20 ° 45 killed for measurement. The chart illustrates maintenance without appre- ciable growth. For other data see page 73 . The diet was as shown herewith. 120 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart LXXXVII. Chart LXXXVIII. Chart LXXXIX. 120 100 150 130 20 40 Days Charts LXXXVII (rat 293, female), LXXXVIII p ct (rat 284, male), and LXXXIX (rat 279, male) show Giutenin 18.0 growth from an early age on a diet containing protein-free starch" ."*. m' . . '. 23.I milk, in which giutenin from wheat formed the sole protein. £^ 25 ° The animals in growing to several times their original weight must have synthesized their purine- and phosphorus-containing complexes from purine-free food. The influence of size on the food require- ment is shown by the food-intake curves. The diet was as shown in table. Chart XC Chart XCI. 140 120 100 80 60 40 'Ovalbum n+Proteii -free mill ■ ■a / / Si* 1 / ? Food =aten / / / / s \ / 180 0 2.0 .40 60 80 DAys Charts XC (rat 258, female) and XCI (rat 250, male) show growth from an early age on a diet contain- ing protein-free milk, in which ovalbumin formed the sole protein. The animals in growing to several times their original weight must have synthesized their purine-con- taining complexes from purine-free food. The influence of size on food re quirementis shown by the food-intake curves. The diet was as shown above p. ct. Ovalbumin 18.0 Protein-free milk . 28.2 Starch 23.8 Agar 5.0 Lard 25 . 0 200 180 160 140 CHARTS AND THEIR EXPLANATIONS. Chart XCII. 140 121 Chart XCIII. 120 120 100 Charts XCII (rat 251, male) and XCIII (rat 259, female) show growth from an early age on a Lactaibumin... . is'o **'.... diet containing protein-free milk, in which lactal- starch'™......', il.ltoik'.k bumin formed the sole protein. The animals in ^|^ growing to several times their original weight must have synthesized their purine- and phosphorus-containing complexes from purine-free food. The influence of size on the food requirement is shown by the food-intake curves. The diet was as shown herewith. Chart XCIV. Chart XCV. p. ct. 5.0 .... 30.0 32.0 140 120 100 160 140 120 100 80 60 40 to E ro t_ O 1 / •Glycinin Protein- free mil / <> / / <■ 1 SI — -\ / t / 20 40 Days 60 80 Charts XCIV (rat 257, female) and XCV (rat 241, p ct male) show growth from an early age on a diet contain- Giycinin 18.0 ' , . P -11 • 1 • 1 1 • • £ j it 1 Protein-free milk. 28.2 ing protein-tree milk, in which glycimn formed the sole starch 23.8 protein. The animals in growing to several times their ^ar ^50 original weight must have synthesized their purine-con- taining complexes from purine-free food. The influence of size on food require- ment is shown by the food-intake curves. The diet was as shown herewith. 122 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart XCVI. Terminated 280 Chart XCVI (rat 36, male) shows the failure of inhibition of growth to cheek the "capacity to grow." The rat was stunted on gliadin food for 37 days (period 1) and on casein food for 12 days (period 2) and completely recovered growth on mixed diet during 217 days (period 3). The diet for periods 1 and 2 was as follows: Constituents. Per. 1. p. a. Gliadin (from wheat) ... 18.0 Starch 29.5 Sugar 15.0 Agar 5.0 Salt mixture 1 2.5 Lard 30.0 Constituents. Per. 2. p. cl. Casein 18.0 Starch 29 . 5 Sugar 15.0 Agar 5.0 Salt mixture I . . . 2.5 Lard 30.0 Chart XCVI I (rat 37, male) shows unimpaired capacity for growth on mixed diet and milk diet after an earlier period of stunted growth on gliadin diet for 37 days (period 1) and casein diet for 12 days (period 2). Part of the period of growth was accomplished on milk food, part on mixed food, the change being made at 3 to mixed food, at 4 to milk food, and at 5 to CHARTS AND THEIR EXPLANATIONS. 123 mixed food again. Note that this has not affected the typical character of the curve of growth. The diet was as follows: Constituents. Per. [. p. ct. Gliadin (from wheat). . . 18.0 Starch ' 29.5 Sugar 15.0 Agar 5.0 2.5 Constituents. Casein. Per. 2. Salt mixture I. Lard 30.0 p. ct. 18.0 Starch j 29.5 ISO SO 2.5 30.0 Sugar . Agar Salt mixture I. Lard Periods 3 and 5. Constituents. Mixed Trumilk food. ! Starch I.ard Per. 4. p. ct. 60.0 16.7 233 Chart XCVII. 290 Chart XCVIII. 110 90 70 50 \<5> \ Or \4 V *dead ■- Gliadi i food ■ -> Chart XCVIII (rat 185, male) shows the failure of a rat to be maintained on a diet composed as shown herewith. p. ct. Gliadin (from wheat) ... 18.0 Starch 29. 5 Sugar 15.0 Agar 50 Salt mixture 1 2.5 Lard 30. 0 20 40 60 Days 124 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart XCIX. Chart C. 180 160 140 120 100 80 60 tSi E tu L " / / / / / ? J < 1 c «3 »7 */ / «/ L_tV_ r~ 1 mi // V-- v 1 / 1 1 1 / V ( — Gliadi n food- 3 -- Milk 150 30 90 70 •0 o .A 09 j 1 i I CD / i i //' ■■ h- Gliadir food - - 3 -Milk f o 3d 20 *0 60 Days 80 100 120 20 40 60 Days 80 100 120 Charts XCIX (rat 186, male) and C (rat 188, female) show the failure of the rat to be maintained during periods i and 2 on diets mentioned below. The perfect resumption of growth when the diet consisted of milk-paste (period 3) illustrates that the "capacity to grow" normally is not visibly impaired by previous large loss of body-weight. The food consisted of — Constituents. Gliadin (from wheat) Edestin Starch Sugar Agar. Salt mixture I Lard Per. 1. Per. 2. p. ct. p. ct. 18.0 0.0 0.0 18.0 29. 5 32. S 15.0 17.0 So 5-0 25 2.5 30.0 25.0 Constituents. Per. 3- Trumilk Starch Salt mixture I Lard p. ct. 60.0 IS- 7 10 23-3 Chart CI (rat 147, female). The animal, well r... .. ,, . .. ,Qp-cL v . ^'' . '. . . ., Ghadin (from wheat) . 18.0 nourished on a mixed diet during period i, tailed starch 29.5^34.5 to maintain its body-weight on a diet in which Agar* '.'.'.'.'.'.'.'.'. '.'.'.'.'.'. ^.'o I7.° gliadin was the sole protein (period 2), until faeces L^tdmixture l 22- 0 300 were added in period 3. The diet consisted of mixed food during period 1 ; for periods 2 and 3 it was as shown herewith. Chart CII (rat 142, female). The animal, well Gliadin(fromwheat). lSp0ct;... nourished on a mixed diet during period 1, failed starch 20.5 to 34-5 to maintain its body-weight on a diet in which Agar!'.'. '....'.'. '.'.'.'.'.'.'. ^'.o gliadin was the sole protein (period 2) . The addi- £^tdmixture J 2j'o 30 tion of fasces to the diet in periods 3 and 4 checked the decline. During period 3 the faeces added were thoroughly sterilized and seemed to be less efficient than the unsterilized faeces in period 4, or in other similar experiments. The diet consisted of mixed food during period 1 ; for periods 2, 3, and 4 it was as shown herewith. CHARTS AND THEIR EXPLANATIONS. 125 Chart CI. 170 -SZ 150 Boc y weight /\ A A f --'V^■', 130 ^ \ 3 v H _/ \y 110 90 70 50 70 < /---- IV ixed feu )d — X — •-Gliadi n food-- .-*--* •- Gliadi n food + Normal f aeces — ~* 1 Food eaten 50 A ,- -\ r / I 1 \ / \ V / * - 30 E 0 20 40 60 80 100 120 140 160 180 200 220 240 260 Oays Chart CII. Cihadin f jt Oiiadm food ■* 160 140 120 100 80 60 40 norn lal faeces 1 Body v veight 0 n 0 **" s S 'NA 4 1/ K ( . \ ■s ;*- ' Food 1 laten \ 1 1 \ • -** 20 in E (Q O S 20 40 60. 80 100 120 140 Oays. 160 180 200 220 240 - 260 126 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart CHI. p. ct. Chart CIII (rat 130, female). The animal, Gliadin(fromwheat). l8.0 well nourished on a mixed diet during period 1, starch 29s to 34. s failed to maintain its body-weight on a diet in Agar/. '.'.'.'.'.'.'.'.'.'.'.'.'.'. l$'.o l.7.'.° which gliadin was the sole protein (period 2), until J*ardmixture * 250 300 faeces were added in period 3. The diet consisted of mixed food during period 1 ; for periods 2 and 3 it was as shown. 130 110 90 70 $0 s ' / 0^' 3 • ;/ ^3 1/1 £ «3 2 20 40 60 Days CHARTS AND TH£lR EXPLANATIONS. Chart CIV. 140 120 100 eo 60 40 in E ra i_ o 127 Chart CV. 80 100 120 / / / / / / / / / J? 5? min / dvov COf\l i> ^ \ 0 - 1 0 - / .E / h 0 J 0 - ^ jy ' 52/ 3 — Gliadin f Milk / / ■«- u 30°/c -> 2 3 . 20 40 60 Days 80 100 120 Chart CVI. Chart CVII. Days Charts CIV (rat 234, female), CV (rat 228, female), CVI (rat 235, male), and CVII (rat 227, male) show the effect of successively larger addi- tions of milk-paste to gliadin food mixture which has been shown in other experiments to be inadequate for the purposes of growth. Note the more rapid growth as the content of milk is increased. The diet consisted of — Constituents. Per. 1. Per. 2. Per. 3. 'Gliadin food tMilkfood p. ct. 95 S p. ct. 80 20 p. ct. 70 30 *G!iadin food: gliadin (from wheat) 18.0; starch, 29.5 to 32. .">; sugar. 17.0; agar. 5.0: salt mixture I, 2.5; lard, 25 to 28. fMilkfood: Trumilk. 60.O; starch, 15.7; salt mixture I, 1.0; lard, 23.3. 128 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart CVIII. Chart CIX. Chart CX. Chart CXI. Charts CVIII (rat 214, female), CIX (rat 219, male), CX (rat 220, male), and CXI (rat 213, male) show the failure to induce more than slight growth when gliadin forms the sole protein of the dietary, even under conditions in which most other proteins have been found effective. That the failure to grow is not due to insufficient food intake is evident. The character of the diets is given in the table below. Period I: *■ cL Gliadin food: gliadin (from wheat), 18.0; starch, 32. Si sugar, 17.0; agar, 5.0; salt mixture I, 2.5; lard. 25. 0 90 Milk food: Trumilk, 60.0; starch. 15-7; salt mixture I, 1.0; lard, 23-3 I0 Period 2: p. cl. Gliadin (from wheat). 18.0 Protein-free milk 28 . 2 Starch 20.8 Agar 5-0 Lard 28.0 CHARTS AND THEIR EXPLANATIONS. Chart CXII. 140 120 100 80 60 40 in E ° A / -HH-.G iadm+ P rotein-f ree milk » Fo< yd eater 1/ \ r— fin r — / / / / ^.-' ' -»v / \ \ \ \ V CHARTS AND THEIR EXPLANATIONS. *33 Chart CXX. Chart CXXI. Day-s Charts CXX (rat 208, female) and CXXI (rat 206, female) show, in period 1, failure to grow on the diet indicated below; and, in period 2, nearly normal growth on a diet containing protein-free milk in which one-quarter of the gliadin, previously found inadequate to induce growth, was replaced by casein. Note the small quantity of casein which suffices to promote growth instead of standstill. This emphasizes the different nutritive value of casein and gliadin. The diets consisted of — Constituents. Per. I. Constituents. Per. 2. Casein or 1 Edestin or | . . . Gliadin J Starch Sugar Agar Salt mixture I. . . p. ct. 18.0 325 17.0 jS.o I2.5 25.0 Gliadin food (gliadin (from wheat), 18.0; protein-free milk, 28.2; starch, 20.8; Casein food (casein, 18 0; protein-free milk, 28.2; starch, 23.8; agar, 5.0; lard, 25 0).. p. ct. 75 25 Chart CXXII. 80 100 Days Chart CXXII (rat 179, female). Period 1 shows maintenance without growth on a diet containing salt mixture I (no protein-free milk) and casein as the sole protein. This should be contrasted with numerous similar experi- ments in which the inorganic constituents of the diet were present in the 134 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. form of protein-free milk. Period 2 shows the influence of the substitution by casein of one-fourth of the gliadin in a dietary repeatedly shown to suffice for maintenance but not for growth. This emphasizes the different nutri- tive value of casein and gliadin. The composition of the diets was as shown below. Constituents. Per. 1. Constituents. Per. 2. Casein p. ct. 18.0 32.5 17.0 to 20.0 S° | 2.5 22.0 to 25.0 Gliadinfood (gliadin (from wheat), 18.0; protein-free milk. 28.2; starch. 20.8; agar, 5.0; lard, 28.0). Casein food (casein, 18.0; protein- free milk, 28.2; starch, 23.8; agar, 5.0; lard, 25.0) p. ct. 75 25 Starch Sugar Agar Salt mixture I. . . Lard Chart CXXIII. 180 160 140 120 100 80 m E / Casein'food- xGliadin + Protein-free milk 75%^ ^Xaseih+ « ,-u 41 I 25% 20 40 60 80 100 120 Days 140 160 180 Chart CXXIII (rat 173, male). Period 1 shows imperfect maintenance without growth on a diet containing salt mixture I (no protein-free milk) and casein as the sole protein. This should be contrasted with numerous similar experiments in which the inorganic constituents of the diet were present in the form of protein-free milk. Period 2 shows the influence of the substitution by casein of one-fourth of the gliadin in a dietary repeatedly shown to suffice for maintenance but not for growth. This emphasizes the different nutritive value of casein and gliadin. The composition of the diets was — Constituents. Per. 1. Constituents. Per. 2. p. ct. 18.0 32. 5 I7.0tO20.0 5.0 2.5 »22.0 25.0 Gliadinfood (gliadin (from wheat), p. ct. Starch 18.0; protein-free milk, 28.2; starch, 20.8; agar, 5.0; lard, 28.0.). 75 Casein food (casein, 18.0; protein- Salt mixture I . . . . Lard , free milk, 28.2; starch, 23.8; agar, 25 CHARTS AND THUIR EXPLANATIONS. Chart CXXIV. Chart CXXV. 135 Charts CXXIV (rat 256, female) and CXXV (rat 255, female) show maintenance without growth of medium-sized rats on a diet of protein-free milk and hordein, from barley, as the sole protein. Note the undiminished appetite during course of experiment. Precisely similar mixtures contain- ing other single proteins have sufficed to induce growth. This experiment demonstrates the different nutritive value of hordein and most other pro- teins and its resemblance in this respect to the chemically similar protein gliadin. This is a marked instance of the relation of the chemical consti- tution of the protein to nutrition. The composition of the food was as shown herewith. p. ct. Hordein 18.0 Protein-free milk ... . 28.2 Starch 16.8 to 18 Agar 5.0 Lard 300 32 136 FEEDING EXPERIMENTS WITH ISOLATED FOOD -SUB STANCES. Chart CXXVI. Chart CXXVII. terminated 160 140 120 100 60 40 20 2 \J -.Body weight 3>\ 3 • \a food ■0 T> T3 0 "4- O (*- c Q> N O \s (J te •minated 0 Q \ \F \ ■00 ood eaten * \ i v \ 1 20 40 60 80 Days Charts CXXVI (rat xi) and CXXVII (rat xiv) show the failure of well-nourished animals (see period 1) to be maintained on a diet in which zein formed the sole protein. The diet consisted of — Constituents. t a'nd , Constituents. Per. 2. Per. 4. p. ct. Dog biscuit ^8 17 Zein p. ct. 16.89 10.14 8.78 338 10.14 SO. 67 p. ct. 10.77 23-70 21. 54 2. IS 5.17 36.63 Lard 41 .66 Sugar Salt mixtuie. . Agar Lard CHARTS AND THEIR EXPLANATIONS. 137 Chart CXXVIII. 260 240 220 200 180 '160 HO 120 100 80 60 40 . 2 t / \ N Zein food • [1 F ood eat en ,--\ / • 00 -^ / 1 0 20 40 60 80 100 120 140 IbO 180 200 220 240 260 Days Chart CXXVIII (rat 146, male) shows the failure Period 2. of a well-nourished rat (see period 1) to be maintained zein 1I.0 on a diet containing protein-free milk and zein as the sole f™^""1"6 milk' 238 protein. It should be noted that precisely similar mix- Agar tures in which zein was replaced by any of the other pro- m teins studied, sufficed either to induce growth or at least to maintain body- weight for an equally long period. Attention is directed to the continued fall in weight despite the large food intake. The composition of the food was mixed for period 1 ; for period 2 it was as shown herewith. 138 FEEDING EXPERIMENTS WITH ISOLATED FOOD-SUBSTANCES. Chart CXXIX. ■) 270 250 "S* 1 ~?7 V\ 230 210 V J -V-J 190 A i 170 1 " 150 130 110 90 Food eaten 70 E TO 1_ 13 II — -^ * 20 40 60 80 100 120 140 Days J 60 180 200 220 240 260 Chart CXXIX (rat 157, male) shows the failure of a well nourished rat (see period 1), to be maintained on a diet containing protein-free milk and zein as the sole protein. It should be noted that precisely similar mix- tures in which zein was replaced by any of the other proteins studied, sufficed either to induce growth or at least to maintain body-weight for an equally long period. Attention is directed to the con- tinued fall in weight despite the large food intake. The composition of the food was mixed for period 1 ; for period 2 it was as shown herewith. Period 2. p. cl. Zein 18.0 Protein-free milk. 28.2 Starch 23 . 8 Agar 5.0 Lard 25 . o New Haven, Connecticut, U. S. A., July i, 191 i. D y Lo MBL/WHOI LIBRARY H IflJS J nm MXUS H ■ r - -