Again, Dr. Dunn can show case after case where this lactic acid feeding has not been unavailing. These cases have done better than any set of cases that we have previously treated.

Dr. Caillé says only a few persons know how to use percentage feeding; he is mistaken. There are many men who know how to use it and are doing so. No one should attempt to throw cold water upon these investigations. I do not ignore in any way the great possibilities of human milk. We are beginning to understand more and more why it is better than any modified milk, when it is good, but we cannot get it over and over again, and we should endeavor, when we cannot get good human milk, to get the best modification we can.

DR. JACOBI.-I do not understand the difference that has been claimed between chemistry and biology. I do not see that there is any boundary line between the two. If there were we should not be in a position to speak of and study biological chemistry. We should not have the very efficient and learned journal coming regularly on the subject of biological chemistry. I do not recognize that boundary line, and therefore I think Dr. Chapin is not justified from that point of view in bringing up that issue. I believe that Dr. Rotch and Dr. Chapin work on the same basis. It is an increase in our knowledge of organic chemistry that we do not make this distinction any more. Biology cannot exist without chemistry; and biochemistry will become the biology of the future more and more.

DR. CHAPIN.-I did not intend to put biology against chemistry, but to protest against putting a chemical value on foods to the exclusion of their biologic value.

DR. JACOBI.-We should not forget that we cannot all work in the same line. There is not a man in the profession in America that has done more original work on just these points, the study of the chemistry and biology of milk, than our friend Dr. Rotch. There are two classes of men among us: those who do original work and those who have more critical minds. We have here a few who do original work; the majority of us, myself included, have only the critical mind. Let us be satisfied that both of these classes, and all of us here who belong to one or the other, have the same purpose in view.

THE HARD CURDS OF INFANT STOOLS; THEIR ORIGIN, NATURE AND TRANSFORMATION.

BY THOMAS S. SOUTHWORTH, M.D.,

AND

OSCAR M. SCHLOSS, M.D.,

New York.

The discovery that the firm rounded curds, occurring not infrequently in the stools of infants fed upon cow's milk, are composed in part of fatty acids and soaps has led to considerable discussion and seems to warrant a further inquiry into the origin of such bodies and their subsequent transformation during their passage through the digestive tract. The claim has frequently been made of late that these masses are simply aggregations of soaps and fatty acids containing no proteid and therefore by implication are not entitled to be called curds. Although doubtless founded upon some individual tests, this view of the hard curd seems to be largely based upon the categorical statement of Czerny and Keller* in discussing "casein stools" that "these flakes and masses which have been taken for casein in the feces are not casein, but fatty soaps."

It is not an uncommon experience to find in the vomited matter both of sick adults and of infants with disturbed digestions, when upon a milk diet, large firm rounded and somewhat elongated masses of rubbery consistency which are assumed to be the result of the coagulation of cow's milk under abnormal stomachic conditions. It is at least inconceivable that these vomited masses have been formed anywhere else than in the stomach, and they correspond, moreover, very closely in appearance to the curds which may be formed from cow's milk by the addition of rennetpepsin and acid, heating the mixture to body temperature and, when the whole has coagulated, pressing the coagulum repeatedly together so as to thoroughly expel the whey. Such masses formed artificially no one certainly hesitates to call a curd, for when thus formed experimentally outside the body by the process above outlined the product differs in no essential particular from the early stages of the curd of ordinary commercial cheddar cheese. The conditions which are necessary for the formation of a cheesy milk

Des Kindes Ernährung, Erhährungsstörungen und Ernährungstherapie, s. 17; Sechste abteilung, zweite hälfte, 1906.

curd, namely, rennet-pepsin, acid, warmth and pressure, may also be present in an infant's stomach, especially during those types of disturbed digestion which increase the amount of abnormal acids. The only real difference then between the factors in the commercial and physiological production of curds lies in the fact that while lactic acid is chiefly active in the commercial process, hydrochloric, butyric, acetic and other acids may also play a part in the stomach.

The curds formed from milk, although often spoken of as acidparacasein curds, or loosely as "casein curds," because of the major part necessarily played in their formation by the coagulation and curding of casein, are not, however, composed exclusively of paracasein. In the formation of a cheese curd the fat of the milk is included in the meshes of the paracasein coagulum, and only a small part of this fat (less than 0.50 per cent.) escapes in the whey.

To understand the subsequent changes in these curds if formed in the stomach of the infant, it is therefore important to keep in mind that they are milk curds, including both fat and paracasein. If curding has taken place in small flakes these can be more or less thoroughly penetrated and affected by the proteolytic and fat-splitting ferments present in the stomach. But if large and compact these masses frequently cannot be disintegrated in the stomach, although it is possible that both fat and proteid may undergo some changes more especially upon the exposed surface. When not disintegrated in the stomach these masses, unless vomited, must in time be passed on into the intestines, where they are subjected to the further influences of an alkaline medium, the fat-splitting and saponifying agencies of the hepatic, pancreatic and intestinal secretions, and the destructive action of the varied bacterial flora there present. In the intestine, fats are split by ferments into glycerin and fatty acids, and the latter, in an alkaline medium, may be more or less transformed into soluble or insoluble soaps according to the nature of the bases with which they combine. When the proteid network is accessible it is subjected to digestive action, and bacteria which flourish upon proteid material may penetrate to some extent into its structure. Small soft flakes, if intestinal conditions are normal, may thus be readily disintegrated, while the larger masses may not be disintegrated, and may eventually appear in the stools.

Howland has recently advanced the view that the occurrence of such masses in the stools is an evidence of excessive fat in the infant's food. The formation of such curds in the stomach is not limited to cases receiving a high percentage of fat in their food, since observation directed to this point will show that they occur even with very considerable dilutions of plain milk, and the explanation of their occurrence in Howland's type of cases need not be a difficult one. Since high fat percentages frequently cause stomachic disturbances, accompanied by the production of excessive quantities of abnormal acids, the most favorable condition is present for the formation of large curds, namely, excessive acidity. Furthermore, the larger the amount of fat in the milk food, the larger the fat inclusion in the proteid coagulum, the larger the mass formed, and the less the probability that it can be disintegrated during its passage through the digestive tract, since the greater preponderance of fat will tend to weld it the more firmly together.

If further argument is needed to prove that the formation of these masses depends primarily upon the coagulation and curding of the paracasein, it may be found in the effect of efficient peptonization. Peptonization of milk which transforms such portions of the casein as its action reaches into soluble and non-coagulable forms of proteid is one of the best recognized of available methods for causing the disappearance of such masses from the stools. It should be remembered, however, that this result is not always attained because the partial peptonization so often employed may leave considerable casein unaltered and still capable of entering into the formation of large curds. Citrate of soda, lauded for this purpose by Wright, and certain alkalies may also be effective, the former by preventing the formation of paracasein and the latter by delaying curding and by neutralizing excessive and abnormal acidity. However, all these must be used in sufficient quantity.

Although the foregoing theoretical and clinical grounds for considering the masses found in the stools to be identical in origin with those found in vomited matter seemed reasonably conclusive, a series of observations and tests were instituted at the Nursery and Child's Hospital to definitely settle certain disputed points. Foremost among these was the claim that the masses in the stools contain no proteid, but only fatty acids and soaps.

Typical firm curds occurring in seventy-five stools passed by thirty-eight infants were subjected to qualitative chemical tests to determine the presence of proteids, soaps, fatty acids and neutral fat in these masses. In the tabulation of the results the tests upon the stools of older children who were receiving whole milk together with other diet consisting of eggs, cereals, etc., have, with the exception of two representative instances (numbers 25 and 26), been excluded in the belief that the findings among younger infants fed exclusively upon milk formulas would prove of greater interest. In the table as appended appear the tests upon masses found in fifty-one stools of twenty-six infants. The youngest were two months old. Ten were from two to six months old. Eleven from seven to thirteen months old. One fifteen months old and two three years old. The fat in the formulas varied from 2 per cent. to 4 per cent. In only five cases did it exceed 3 per cent. The proteid percentages ranged from 0.90 per cent. to 4 per cent. In but eight of these, however, did the proteid exceed 2.50 per cent., and the majority of these were older infants fed upon plain milk diluted with barley gruel. Few, if any, of the formulas seem open to criticism as excessive if fed to normal infants at the ages given.

The technique employed was as follows:

(1) The masses to be tested were ground in a mortar and thoroughly treated with successive portions of ether and boiling alcohol to dissolve out the neutral fat and free fatty acids. These solvents were then evaporated and their residue examined microscopically for fatty acid crystals (often after recrystallizing from boiling alcohol) and also tested with osmic acid and Sudan III. for fat.

(2) The residue remaining after ether and alcohol extraction was then treated with 5 per cent. hydrochloric acid which would split the insoluble soaps into fatty acids and form chlorides of the bases, as well as remove any inorganic salts. The residue was then again treated thoroughly with boiling alcohol to remove any fatty acid thus formed and the remainder washed with ether, thoroughly dried and set aside for the proteid tests. The washings (alcoholic) were examined for fatty acids, the finding of which at this stage would indicate the presence of insoluble soaps. in the original masses.

(3) The washed and dried residue was then in each instance