STUDIES ON WATER-DRINKING VI. THE ACTIVITY OF THE PANCREATIC FUNCTION UNDER THE INFLUENCE OF COPIOUS AND MODERATE WATER-DRINKING WITH MEALS P. B. H A W K, Ph.D. URBANA, ILL. REMOTE s t 6 / 2-. 3 H 3 1 a - Crp p/ 2~ (^Js^Qaas ~ * - STUDIES ON WATER-DRINKING VI. the activity of the pancreatic function under the influence OF COPIOUS AND MODERATE WATER-DRINKING WITH MEALS * * 1 P. B. HAWK, Ph.D. TJBBANA, ILL. Numerous tests are in vogue clinically for the determination of the functional activity of the pancreas. The best known of these are the glutoid test of Sahli , 2 the nuclei test of Schmidt , 3 and the Cammidge 4 reaction, the last mentioned test having been rather more freely criticized than either of the others. Comparatively recently, clinicians have come to a fuller realization of the aid which a knowledge of the pancreatic enzyme concentration of the feces may give them in their diagnosis of pancreatic insufficiency. The accurate determination of the enzyme content of the feces under normal or pathological conditions is still, how- ever, a line of medical research which has not been very generally investi- gated. Some of the investigators who have recently contributed to our knowledge in this direction either by suggestions as to method or by the collection of clinical data are Muller , 5 Wohlgemuth , 6 Wynhausen , 7 Boldy- reff , 8 Abderhalden and Schittenhelm , 9 Schlecht , 10 Goldschmidt , 11 Vol- hard , 12 Ury , 13 Frank and Schittenhelm . 14 One of the first enzymes to be *From the Laboratory of Physiological Chemistry of the University of Illinois. 1. For previous papers in this series of studies see, I. Hawk: Univ. Penn. Med. Bull., 1905, xviii, 7; II. Fowler and Hawk: Jour. Exper. Med., 1910, xii, 388; III. Rulon and Hawk: Jour. Am. Chem. Soc., 1910, xxxii, 1686; IV. Rulon and Hawk: The Archives Int. Med., 1911, vii, 536; V. Hattrem and Hawk: Ti-ie Archives Int. Med. 1911, vii, 610. 2. Sahli: Deutsch. Arch. f. klin. Med., 1898, p. 61. 3. Schmidt: Deutsch. med. Wchnschr., 1899, No. 45. 4. Cammidge: Proc. Roy. Soc., London (B), 1909, lxxxi, 372; also Proc. Roy. Soc. Med., iii, 164. 5. Muller: Arch. f. klin. Med., 1908. 6. Wohlgemuth: Berl. klin. Wchnschr., 1910, xlvii, 92. 7. Wynhausen: Berl. klin. Wchnschr., 1909, xlvi, 1406. 8. Boldyreff: Arch. f. d. ges. Physiol. (Pfliigers), 1905, cxxi, 13. 9. Abderhalden and Schittenhelm: Ztschr. f. physiol. Chem., 1909, lix, 230. 10. Schlecht: Miinchen. med. Wchnschr., 1908, lv, 725. 11. Goldschmidt: Deutsch. med. Wchnschr., 1909, No. 12. 12. Volhard: Miinchen. med. Wchnschr., 1908, liv, 403. 13. Ury: Biochem. Ztschr., 1909, xxiii, 153. 14. Frank and Schittenhelm: Ztschr. f. exper. Path. u. Therap., 1910, viii, 237. T3 36722 4 detected in the feces was pancreatic amylase. This enzyme, as early as 1875, was shown by Wegscheider 15 to be present in the feces of the nursling. This finding was later verified by von Jaksch, 16 Moro 17 and Allaria. 18 According to this last-mentioned investigator the manner of feeding the infant, i. e., naturally or artificially, had no influence on the amylase concentration of the feces. Strasburger 19 found this same condition to hold for the stools of adults. DESCRIPTION' OF EXPERIMENTS Three experiments were made, in each of which normal men (E and W) served as subjects, the same man serving as subject in Experiments 2 and 3. Each experiment was divided into three periods, a fore period during which the subject was brought into nitrogen equilibrium through the ingestion of a diet uniform as to quality and quantity from day to day, a water period during which the ration was supplemented by the drinking of additional water at meal-time, and finally an after period in which the fore period ration of low water content was again fed. The constituents of the diet apart from the water were graham crackers, butter, milk and -peanut butter. The diet of Subject W in Experiment 1 con- tained 85.8 gm. of protein per day whereas that of Subject E in Experi- ments 2 and 3 contained 92.3 gm. of protein per day. During the fore and after periods of each experiment 100 c.c. of water, in addition to the water content of the milk, was taken by each subject at each meal. In the water periods of Experiments 1 and 2, 500 c.c. of water above that previ- ously ingested were taken with each meal, whereas in Experiment 3 this volume was increased to 1% liters. Each individual stool was examined in the fresh condition. The feces of one period were separated from those of the next by means of charcoal, the material being ingested in gelatin capsules at the beginning of the first meal of each new period. The method used in the quantitative determination of the activity of the pancreatic function was that suggested by Wohlgemuth. 6 This author considers that the extent to which pancreatic amylase appears in the feces may be taken as an index of the activity of the pancreatic function. He therefore suggests a procedure for the quantitative determination of the amylolytic power of sodium chlorid extracts of the fresh feces. His procedure has been fully discussed by us in another connection and a modification of the Wohlgemuth method proposed. 20 In the experiments 15. Wegscheider: Inaug. Diss., Strassburg, 1875. 16. Von Jaksch: Ztschr. f. physiol. Chem., 1888, xii, 116. 17. Moro: Jahrb. f. Kinderh., 1898, xlvii, 342. 18. Allaria: Progresso Med., 1905. 19. Strasburger: Arch. klin. Med., 1900, lxvii, 238 and 531. 20. Hawk: The Archives Int. Med.; to be publisned. 5 embraced in the present paper, however, the procedure used by ns was essentially that of Wohlgemuth except that the final amylolytic values of the stools were calculated in terms of dry matter instead of on the basis of the moist sediment secured by centrifugation as Wohlgemuth suggests. In order to demonstrate conclusively the origin of the fecal amylase Wohlgemuth 6 has made certain interesting tests. In these tests it was demonstrated that the ligation of the pancreatic ducts of dogs caused the feces as passed to be amylase-free or to contain at the most a minimal quantity of the enzyme. Wynhausen 21 has likewise found a much- decreased amylase output in cases with partial closure of the pancreatic duct. Table 1. — Fecal Amylase, Experiment 1, Subject W, Moderate Water-Drinking FORE PERIOD SIX DAYS Feces — Amylolytic Value — 1 °fo Starch Stool Fresh Dry Total Per Gram Dry No. (grams) (grams) (liters) Matter (c.c.) 1 62.8 16.1 11.4 709 2 98.8 26.4 10.8 409 3 109.8 30.2 18.1 600 4 185.8 44.8 26.3 587 5 41.8 13.1 6.7 513 6 104.9 25.5 20.9 819 7 24.0 7.9 3.3 416 Average per day. . . . . 104.6 27.3 16.3 597 WATER PERIOD TEN DAYS (1,500 c.c. HoO daily) 8 31.6 7.6 7.6 1,000 9 147.5 40.6 34.9 860 10 75.4 19.8 17.1 865 11 144.8 39.8 27.3 688 12 63.9 17.2 14.6 848 13 115.5 31.3 22.2 709 14 26.0 7.2 4.0 548 15 169.0 44.8 32.5 725 16 127.0 30.9 44.7 1,447 17 152.7 35.5 110.3 3,106 Average per dav. . . . . 105.3 27.5 31.5 1,145 AFTER PERIOD 18 60.0 14.8 14.6 983 19 119.6 28.0 29.4 1,052 20 51.1 14.1 20.8 1,477 21 142.7 35.8 41.1 1,147 22 81.4 22.8 15.0 657 23 53.4 16.4 19.1 1,162 Average per day. . . . . 101.6 26.3 28.0 1,065 DISCUSSION OF RESULTS 1. Moderate W at er -Drinking . — Data from the first experiment may be found in Table 1. The subject of this experiment was W and the purpose was the study of the influence of moderate water-drinking with meals on the activity of the pancreatic function. After a fore period of six days on a uniform diet, as already mentioned, the subject was found to be in nitrogen equilibrium as was shown by a nitrogen balance of 0.01 + gm. The average daily output of fresh feces was 104.6 gm., representing a dry matter content of 27.3 gm. per day during this period. 22 The total 21. Wynhausen: Berl. klin. Wchnschr., 1909, No. 30. 22. The data for fresh and dry feces were obtained by Dr. H. A. Mattill in connection with another research which has already been reported. See Mattill and Hawk: Proc. Soc. Biol. Chem., 1911. amylolytic value of the stools per day, expressed in liters of 1 per cent, starch solution which this feces would hydrolyze under the conditions of the Wohlgemuth method, was 16.3 liters. Placed on the basis of “1 gm. of dr}^ feces” we see by an examination of the data that the average daily amylolytic value for this period was represented by 597 c.c. of 1 per cent, starch solution. Under the influence of the 500 c.c. of water ingested at each meal for a period of ten days there was a marked increase in the amylolytic power of the feces. For example, the data from the first stool of this period indicate that the amylolytic value was 1,000 c.c. of 1 per cent, starch solution per gram of dry feces. This initial value of the period was far above that for any individual stool of the fore period and nearly double that of the daily average amylolytic value (597 c.c.) for the entire fore period. If the data from the analysis of the stools which were subse- quently dropped during this water period are examined it will be seen that the amylolytic values for these stools, with a single exception (No. 14), were above the average value for the fore period. It will furthermore be observed that the average amylolytic value for the water period, i. e., 1,145, was approximately twice as great as the average amylo- lytic value for the fore period. In other words the drinking of 1,500 c.c. of water per day, at meal time, through a period of ten days, was instru- mental in causing a virtual doubling of the anxiolytic activity of the fecal matter excreted during that period as compared with the fecal output of the fore period during which minimal amounts of water were daily ingested. The significance of this finding of an increased amylolytic power is further accentuated when the data from the after period are examined. Notwithstanding the fact that the water ingestion of this period was reduced to the low level of the fore period the stools as dropped were on a much higher amylolytic plane than were those of the fore period. This fact is clearly brought out when it is noted that the corresponding values for the fore and after periods were 597 and 1,065 respectively. We thus see that the stimulation of the factors which brought about the increased amylolytic power of the fecal output of the water period was not transi- tory in character. In other words the influence of the water was not limited to the time of its ingestion, but the factors leading to an increased fecal amylolytic value were so stimulated by the ingestion of the excess water as to continue their efficiency even after the water ration had been again reduced to that of the fore period. Whether this pronounced increase in amylolytic power observed to accompany and follow the inges- tion of additional volumes of water at meal time may be properly inter- preted as indicative of an increased activity of the pancreatic function is discussed in a later paragraph. 7 In the second experiment on the influence of moderate water drinking (Experiment 2) E served as subject. All conditions were the same as those in force in the similar experiment on W with the exception that the uniform diet was varied in a minor manner from that ingested by W. The data obtained from the fecal examinations of this experiment are tabulated in Table 2. In this instance approximate nitrogen equilibrium was secured in seven days, the nitrogen balance showing a plus value of 0.485 gm. per day. A ten-day water period then followed during which 500 c.c. of water were ingested with each meal in addition to the uniform water ingestion of, the fore period. The experiment ended with a four- day after period. Table 2. — Fecal Amylase, Experiment 2, Subject E, Moderate Water-Drinking FORE PERIOD SEVEN DAYS Stool Feces - Fresh Dry Amylc Total •lytic Value — 1 % Starch Per Gram Dry No. (grams) (grams) (liters) Matter (c.c.) 1 88.4 18.4 6.3 343 2 30.2 8.0 3.9 491 3 179.2 41.2 14.0 341 4 193.9 47.8 11.3 237 5 76.9 20.1 3.7 184 6 207.7 47.3 16.3 345 7 121.6 27.8 15.3 548 8 44.0 12.9 2.2 171 Average per day. . . . . 135.0 32.0 10.4 325 9 76.5 WATER PERIOD TEN DAYS (1,500 c.c. H 2 0 daily) 17.5 7.3 418 10 140.9 29.4 31.6 1,075 11 63.8 18.6 2.3 124 12 169.0 41.1 16.1 393 13 247.5 47.5 12.1 255 14 135.3 31.1 6.2 198 15 192.4 37.7 19.7 522 16 79.2 19.5 15.6 801 17 55.9 15.2 9.3 613 18 173.5 43.2 27.5 636 19 51.5 14.4 9.2 642 Average per day. . . .. 138.6 31.5 15.7 498 20 67.3 AFTER PERIOD FOUR 15.9 DAYS 14.0 878 21 117.8 30.7 19.9 649 22 147.7 34.3 21.8 637 23 145.9 38.7 25.8 667 24 31.2 9.8 9.4 961 Average per day. . . .. 127.5 32.3 22.7 703 During the preliminary interval of low water ingestion covered by the fore period the average daily total amylolytic value of the feces was equivalent to 10.4 liters of 1 per cent, starch solution. Placed on a dry- matter basis we find that the amylolytic value of the feces of this period may be represented by 325 c.c. of 1 per cent, starch solution per gram of dry matter. If we now examine the data for the period during which the extra volume of water was ingested we find that the conditions are similar to those already discussed in connection with the experiment in which W served as subject. In other words, water caused an increase in the amylolytic value of the feces. The average total daily amylolytic % 8 value for the water period was 15.7 as against 10.4 for the fore period whereas the value on the dry matter basis was 498 as compared with a value of 325 for the preliminary interval. The influence of the moderate water-drinking in causing an increase in the amylolytic power of the feces was much less pronounced with E than with W. This statement is borne out by the observation that there was an increase of nearly 100 per cent, in the case of W as against an increase of little more than 50 per cent, in the case of E. However, when we examine the data for the after periods of the two experiments in question, we note that the amylolytic value in E ? s experiment was above that of the water period, whereas in W’s experiment this value was slightly lower than that of the water period. By calculation we see that the amylolytic value for the after period of the test on W was increased only 80 per cent, above that of the fore period whereas the increase in the test on E was about 120 per cent. It would seem, therefore, that the water caused a more pronounced immediate stimulation in the case of W than in that of E but that the influence of the stimulation was more persistent in the case of E than in that of W. Table 3.- -Fecal Amylase, Experiment 3, Subject E, Copious Water-Drinking FORE PERIOD — -SIX DAYS — Feces - Amylolytic Value — 1% Starch Stool Fresh Dry Total Per Gram Dry No. (grams) (grams) (liters) Matter (c.c.) 1 35.2 10.2 3.8 375 66.0 18.8 8.6 458 3 202.2 49.8 23.6 473 4 3 29.2 32.2 14.9 464 5 161.3 34.6 8.8 256 6 171.8 38.8 15.8 406 7 34.6 9 4 4.2 448 Average per day. . . . . 133.4 32.3 13.3 412 WATER PERIOD— -FIVE DAYS (4,000 c.c. H 2 0 daily) 8 90.3 18.2 10.5 578 9 37.2 10.2 8.3 816 10 249.4 58.8 26.5 451 11 74.7 3 9.9 6.1 306 12 258.0 38.2 342.3 8,961 13 52.6 15.9 7.8 491 Average per day. . . . . . 152.5 32.2 80.3 2,494 AFTER PERIOD THREE DAYS 14 128.3 29.6 12.3 414 15 86.4 21.0 15.6 743 16 206.5 44.5 25.0 561 17 50.6 5.6 7.0 1,257 Average per day . 157.3 33.6 20.0 595 2. Copious Water-Drinking . — The amylolytic power of the feces was studied in but one experiment in which copious water drinking was prac- ticed at meal time. In this experiment (3) E served as subject. All the experimental conditions were similar to those in force with this same subject in the second experiment on moderate water drinking. In a fore period covering an interval of six days a very satisfactory balance for income and outgo of nitrogen was obtained, the data indicating a plus balance of. only 0.03 gm. The data from this experiment are given in Table 3. An examination of that table will show, in the first place, that 9 the average total daily amylolytic value for the fore period was 13.3, whereas the amylolytic value on a dry matter basis was 412. These values simply mean, as heretofore explained, that the average output of feces per day during the six-day fore period possessed the power to hydrolyze 13.3 liters of 1 per cent, starch solution, whereas 1 gm. of dry feces possessed the power of transforming 412 c.c. of such a starch solution. The water period of this experiment was five days in duration as against the ten-day periods utilized in the experiments already discussed. However, the daily ingestion of water was much higher in this instance than in either of the afore mentioned studies. In the moderate water- drinking tests we caused the subjects to ingest 500 c.c. additional at each meal or a total daily ingestion of 1,500 c.c. in excess of that customarily taken. In certain other water-drinking studies made by us (Fowler and Hawk 1 ; Mattill and Hawk 22 ; also Wills and Hawk, unpublished), in which the influence of copious amounts of water was under investigation, the total volume of water added daily to the normal ration had been 3,000 c.c. However, in the present investigation we were dealing with a subject, E, who was accustomed to drinking rather larger volumes of water than individuals ordinarily ingest and for this reason his water ration was fixed at 4,000 c.c. per day during the water period in an attempt to place the daily volume at such a figure as should be copious for his organism. He felt no personal discomfort at any time during the period in which this excessive amount of water was being daily introduced into his system. Under the influence of copious water-drinking through a five-day period, the amylolytic activity of E ? s feces was increased in an extremely emphatic manner. Particularly was this true of Stool 12, passed on the fourth day of the period. Expressing the amylolytic values in the same manner as heretofore followed we find that the average amylolytic power of the feces was equivalent to a force sufficient to hydrolyze 80.3 liters of 1 per cent, starch solution. Furthermore, if we transfer the comparison to the dry-matter basis we learn that 1 gram of dry matter represents sufficient amylolytic activity to hydrolyze 2,494 c.c. of 1 per cent, starch solution. When we compare these volumes with those obtained during the fore period we are astounded at the very material increase in the amylolytic activity which has taken place under the influence of the copious water-drinking. For example, if we consider the dry-matter basis we see that the amylolytic value has increased from 412 to 2,494, a most surprising increase of more than 600 per cent, in the amylolytic activity. The question as to the proper interpretation of this finding, in so far as it relates to the functional activity of the pancreas, is discussed later on. In this experiment for the first time, we observed an amylolytic value for the after period which was lower than that of the water period. Evi- 10 dently the stimulation of the copious water-drinking was so pronounced in character in this particular instance as to rather militate against any further added stimulation after the organism was released from the imme- diate influence of the water ingestion. That the influence of the water extended somewhat beyond the time interval during which it was actually being ingested is gathered from the recorded average amylolytic values for the after period. These values are nearly 45 per cent, above those of the fore period thus indicating an after effect which persisted for a time, at least, after the period of high water intake had closed. INTERPRETATION OF FINDINGS The foregoing discussion of the various experiments embraced in this study has indicated clearly that when either small or large volumes of water above those customarily ingested were taken at meal time that the resultant fecal matter possessed greater amylolytic activity than did the feces passed during periods in which these extra volumes of water were not taken. How is this finding to be interpreted? It has been estab- lished that an increase in the activity of the pancreatic function will be indicated by the appearance of an added excretion of pancreatic amylase in the feces. This fact would of itself cause the feces in question to possess an increased power to hydrolyse starch solution. On such a basis Wohlgemuth has suggested a method for the quantitative determination of the amylolytic power of the feces. This power, according to Wohlge- muth, is due entirely to the ability of dilute solutions of the fecal amylase to transform starch into substances which no longer give a blue color with iodin. From this point of view, therefore, we may interpret our results as indicating that the pancreatic function has been greatly stimu- lated under the influence of the water-drinking at meal-time and that consequently larger quantities of pancreatic amylase are present in the feces passed during the water period, thus giving to the fecal matter a higher amylolytic power than that possessed by stools dropped during previous periods in which the water ingestion was minimal. We began our investigation under the impression that Wohlgemuth’s method would furnish us with accurate data, and nothing occurred in the course of the experiments to cast any doubt on the validity of Wohlge- muth’s claim until an examination was made of the final stool (17) in the water period of Experiment 1. Here for the first time in the course of our studies the entire series of seven tubes of 1 per cent, starch solution showed complete digestion. As soon as this fact was determined another series of ten tubes was prepared and again the entire series was com- pletely digested. Hot caring to rely on data obtained from further extracts because of the possibility of changes having taken place in the feces, we examined no further series. A similar observation was made in 11 connection with Stool 12 in Experiment 3. In onr search as to the cause of this surprisingly pronounced increase in the amylolytic activity of the stools mentioned, we observed that they each possessed a very pronounced acid reaction. We were familiar with the findings of Chittenden and Griswold, 23 which were later confirmed by Vernon 24 to the effect that amylase (salivary) is inactive in the presence of 0.009 per cent, hydro- chloric acid. At the same time we were familiar with the further fact that it requires an acid concentration much above 0.009 per cent, hydro- chloric acid to hydrolyze 1 per cent, starch solutions under the conditions of our experiment. Furthermore, in the experiments of Vernon to which reference has already been made, as well as in others by Schierbeck, 23 it has been demonstrated that certain proper acid concentrations below 0.009 per cent, hydrochloric acid will facilitate the activity of amylase. This power was shown to be possessed by organic acids as well as by inorganic acids. In the case of hydrochloric acid a concentration of 0.004 per cent, was found to cause an increase of 400 per cent, in the amylolytic power of the enzyme, whereas 0.0083 per cent, lactic acid increased the amylo- lysis nearly 500 per cent. In light of the above findings, we might interpret the fivefold increase in the amylolytic power of acid Stool 17 of Experiment 1 and the sixfold increase in the amylolytic power of acid Stool 12 of Experiment 3 as due to the stimulating influence of the acid reaction of the sodium chi or id extracts of the feces on the contained amylase. It obviously needs no argument to support the claim that the acidity of these extracts could not of itself have caused the hydrolysis of the 1 per cent, starch solutions. Even had the feces possessed a surprisingly high acid concentration, this concentration must of necessity have been so much lowered by the time the series of dilutions of the salt solution extracts had been made as to render the acidities, of the latter part of the series, at least, of no influence in so far as their power to hydrolyze starch is concerned. It is entirely possible that the amylase was passed into the intestine in increased amount on the days in question under the influence of the high water intake, that it there fulfilled its function of bringing about an augmented anrylolysis and that subsequently the reaction of the surrounding media was so altered by certain factors as to render impossible any further activity of the enzyme. Therefore the reaction in which the enzyme was finally excreted from the organism by way of the feces was not, of neces- sity, the reaction which the intestinal substrate of the pancreatic amylase possessed. Be that as it may, however, it is of course impossible to come to any definite conclusion as to the extent of the stimulation of the activity of the pancreatic function on the basis of data collected from acid 23. Chittenden and Griswold: Am. Chem. Jour., 1881, iii, 305. 24. Vernon: Jour. Physiol., 1902, xxvii, 174. 25. Schierbeck: Skand. Arch. f. Physiol., 1892, iii, 344. 12 stools by means of the Wohlgemuth method. This being so, we have suggested a modification of that method which should obviate the trouble- some features associated with the examination of stools which possess a pronounced acid or alkaline reaction. This method appears elsewhere. 20 If we examine the data in Tables 1, 2 and 3 it will be seen that the complete elimination of all consideration of the acid stools in question which gave evidence of the surprisingly pronounced amylolytic power, will not prevent us from drawing the conclusion that the activity of the pan- creatic function was stimulated under the influence of water-drinking at meal time. Certain other experiments may be cited as tending to sub- stantiate the theory of an increased activity of the pancreatic function under the influence of water-drinking with meals. For example, Pawlow 26 has shown that if 150 c.c. of water be introduced into the stomach of a dog possessing a pancreatic fistula, that the flow of juice begins after an interval of two or three minutes and in case there is already present a distinct flow of juice that the output is distinctly increased. That this flow of juice was not brought about through the stimulation of acid chyme entering the duodenum was shown from the fact that, if the stomach of the animal be emptied, the contents of the organ are found to be neutral or faintly alkaline in reaction. This experiment by Pawlow indicates clearly, then, that water introduced into the stomach causes a direct stimulation of the nervous mechanism of the pancreas which is followed by an outpouring of pancreatic juice. As further basis for our belief that water ingestion stimulates the pancreas, we would cite certain other experiments of Pawlow 27 and the earlier work of Heidenhain 28 and Ssa- nozki. 28 The more recent experiments of Foster and Lambert 29 may also be adduced as furthering this claim. These various investigators just enumerated have shown that an increased flow of gastric juice follows the entrance of water into the stomach. The studies of Foster and Lambert go even farther than this and show most clearly that there is not only an increased flow of juice but that this increased volume possesses a higher acid concentration than that possessed by juice secreted under other conditions. We have obtained results in this laboratory 30 which go to verify the findings of Foster and Lambert. On the basis of the well- established theory as to the mechanism of pancreatic secretion as evolved by Bayliss and Starling 31 we may logically expect a pronounced increase in the flow of pancreatic juice to follow the entrance of the strongly acid chyme into the duodenum. That the pancreas is more active under a 26. Pawlow: The Work of the Digestive Glands, Second Edition, 1910, p. 144. 27. Pawlow: The Work of the Digestive Glands, p. 112. 28. Quoted by Pawlow, p. 112. 29. Foster and Lambert: Jour. Exper. Med., 1908, x, 820. 30. Wills and Hawk: Proc. Soc. of Biol. Chem., 1911. 31. Bayliss and Starling: Jour. Physiol., 1902, 28. 13 high water ingestion is also evidenced from other experiments made in this laboratory 22 which have demonstrated a more complete digestion and absorption of ingested fats and carbohydrates under the influence of water-drinking at meal time. It has further been found by us 32 that the carbohydrate content of the actual samples of feces listed in Tables 1, 2 and 3 of this article was lower during the water periods than during the period of minimal water ingestion. This fact demonstrated definitely then that the intestinal content of these subjects was of a higher amylase concentration during the period of increased water intake. Other things being equal we would expect a fair degree of uniformity between the availability of ingested carbohydrate and the content of fecal amylase. The observation of Brad- ley 33 to the effect that pancreatic lipase possesses greater fat-splitting power when its reaction mixtures are diluted with several volumes of water also goes far toward a substantiation of the theory of a pancreas activated through water ingestion. We have much to learn as to the alterations in the reaction of the intestine under different conditions. In the case in point we have an excessive quantity of hydrochloric acid passed into the intestine, under the influence of the high water intake, a quantity which probably, at times, is far in excess of that which may be neutralized by the pancreatic juice and bile. The greater portion of this excess acid we may suppose to be neutralized by the ammonia which is produced in the deamidation of protein material. This acid would therefore appear in the urine as ammonium chlorid. However, if there should occur a very copious out- pouring of hydrochloric acid at a time when there was but little protein material available for deamidation and consequently a minimal amount of ammonia present for the neutralization of such excess acid, it is entirely possible that we might have an acid reaction due to hydrochloric acid extending the entire length of the large intestine. In that event the stools would probably possess a relatively high acid concentration. If the food residues remained in the intestine for an interval shorter than the normal because of increased peristalsis, diarrhea, etc., acid stools might also result. The normal reaction of the intestinal contents is acid down to the ileocecal valve under normal conditions. Furthermore, carbohydrates which escape absorption may undergo acid fermentation through the action of certain acid-forming bacteria such, for example, as Bacillus coli and Bacterium welchii. A number of organic acids are formed which include acetic acid, lactic acid, butyric acid and succinic acid. The extent to which this acid reaction may be carried is indicated by Macfadyen, 32. Mattill and Hawk: Unpublished results. 33. Bradley: Journal Biol. Chem., 1910, viii, 251. 14 Nencki and Sieber’s 34 finding of an acidity at the ileocecal valve equiva- lent to 0.1 per cent, acetic acid. This observation was made on a patient with a fistula at the end of the small intestine. SUMMARY The problem studied was the activity of the pancreatic function under the influence of water-drinking at meal time. Normal men were used as subjects and were required to ingest a diet uniform in all respects from day to day. Three experiments were made, each experiment being divided into three periods; a fore period in which the subject was brought into nitrogen equilibrium through the ingestion of the uniform diet sup- plemented by a minimal water ingestion; a water period during which the water ingestion at meal time was increased from 1,500 to 4,000 c.c. per day ; and an after period in which the dietary conditions of the fore period prevailed. Two of the experiments were on the influence of moderate water-drinking and in these the volume of the extra water ingested during the water period was 1,500 c.c. per day or 500 c.c. per meal. In the third experiment the influence of copious water-drinking was studied and in this instance the subject was required to ingest on each day of the water period 4,000 c.c. above that ingested in the fore and after periods. The amylolytic activity of the feces, denoting, according to Wohl- gemuth, the content of pancreatic amylase present in the feces, was taken as the index of the activity of the pancreatic function. The amylolytic values for the stools dropped during the periods of moderate and copious water-drinking at meal-time were much higher than the similar values as determined for the stools dropped during the periods of minimal water ingestion. This finding may be interpreted as indicating that the drink- ing of water with meals had stimulated the pancreas. We make this interpretation with certain reservations regarding conditions not con- trolled by the Wohlgemuth method. Two stools were encountered in the course of the investigation which exhibited a pronounced acid reaction. Inasmuch as the basic principle of Wohlgemuth’s method is the hydrolysis of starch solutions through the medium of unneutralized fecal extracts, the method does not give dependable data as to the amylase content when such stools are under examination. This is true especially in view of the fact that it has been shown that 0.004 per cent, hydrochloric acid will increase the activity of amylase 400 per cent., whereas 0.009 per cent, hydrochloric acid will cause absolute inhibition. The power of feces extracts, therefore, to hydrolyze starch cannot be taken as a measure of the amylase present unless precautions are taken to neutralize the fecal extracts and then make the conditions uniform for the action of the enzyme if present. 34. Macfadyen, Nencki and Sieber: Arch. f. exper. Patliol. u. Pharmacol., 1891. xxviii, 311. 15 On the basis of the data gathered in this and in associated investiga- tions made in our laboratory and elsewhere, we are prepared to draw the general conclusion that the ingestion of quantities of water at meal-time ranging in volume from y 2 to 1% liter stimulate the pancreatic function in two ways : first, a direct stimulation of the nervous mechanism of the pancreas brought about while the water is still in the stomach and, second, an indirect stimulation brought about on the entrance of the increased volume of acid chyme into the duodenum. The drinking of water with meals ought therefore to bring about a more rapid and complete digestion and absorption of the fat and carbohydrate constituents of the diet, two observations verified by experimentation in our laboratory.