Cornell University Ubrary 
 RM 217.F5 1907 
 
 The effect of diet on endurance, based on 
 
 3 1924 003 520 701 
 
Printed from the transactions of the . 
 
 CONNECTICUT ACADEMY OF ARTS AND SCIENCES 
 
 Incorporated A. D. 1799 
 
 VOLUME XIII. PP. 1-46 
 
 MAY, 1907 
 
 Publications of Yale University 
 
 THE EFFECT OF DIET ON ENDURANCE 
 
 BASED ON AN^' EXPERIMENT, IN THOROUGH MASTICATION, 
 WITH NINE HEALTHY STUDENTS 
 
 AT Yale university, januAry to june,. i906 
 
 IRVING FISHER, Ph.D. 
 
 PsioFESsoR OF, Political EcoNQMY at Yale University 
 
 NEW HAVEN, CONNECTICUT 
 1907 
 
 THE TUTTLE, MOREHOUSE & TAYLOR PRESS 
 
Cornell University 
 Library 
 
 The original of tiiis book is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924003520701 
 
I. — The Effect of Diet on Endueance, Based on an Expbei- 
 
 MENT WITH NiffE HeALTHT STUDENTS AT YaLE UnIVEKSITT. 
 
 Januaey-Junb, 1906. 
 
 Introduction. 
 
 There appears to be very little literatare on the subject of endur- 
 ance. Since the epoch-making work of Mosso, much has been written 
 on fatigue, and many varieties of ergographs have been constructed 
 to record muscular fatigue ; but no systematic study of endurance 
 as such appears to have been made. Even the concept of endurance, 
 as related to strength and fatigue, has been lacking. No corre- 
 lations have been worked out between endurance and the factors 
 upon which it depends, except that it has been a matter of common 
 experience that endurance increases with exercise. In respect to 
 diet, opinions as to its relation to endurance, so far as the writer knows, 
 have rested on no better foundation than the personal impressions of 
 adherents of special dietary systems, such as those of Salisbury, 
 Dewey, Haig, Kellogg, and Fletcher. In Professor Chittenden's 
 painstaking study on "Physiological Economy in Nutrition" he has 
 shown that one result of a gradual and systematic reducti"on in proteid, 
 from the amount ordinarily consumed, has been an increase in strength, 
 but no data were obtained in respect to endurance. 
 
 The present experiment had a somewhat accidental origin. I was 
 engaged in collecting statistics of labor-power in relation to various 
 factors, among them especially diet. The data were collected because 
 of their economic bearings and without any intention at first of mak- 
 ing independent experiments. But some of my students, whom I had 
 engaged to make computations and diagrams, became interested in 
 the material with which they thus came in contact, and expressed a 
 strong desire to try dietetic experiments upon theihselv^s/ ^ Not 
 being a physiologist, I asked Professor Chittenden if he could, not 
 take charge of these experiments for them. It so'hapjsened that on 
 account of other similar work he was unable td do so, but suggested 
 that I should conduct them myself. I have done so with consider/- 
 able hesitation, not being equipped for physiologic studies. I have 
 therefore restricted my attention to the simpler practical aspects of 
 the problem, although some of the technical points have been inves- 
 
 Tkans. Conn. Acad.', Vol. XIII. 1 Mat, 1907. 
 
2 Fisher — The Effect of Diet on Endurance. 
 
 tigated through the very kind cooperation of able colleagues. My 
 thanks are especially due to Professor Chittenden and his co-workers, 
 Professor Mendel and Dr. Underbill, for the aid rendered by the 
 Sheffield Scientific School Laboratory in determining the nitrogen 
 excreted, and for much helpful advice and criticism. I wish also to 
 express my obligations to Dr. J. P. C. Foster for his services as 
 medical adviser to the students ; to Dr. W. G. Anderson, Director 
 of the Yale Gymnasium, and his corps of assistants, through whom 
 the endurance tests were conducted ; to Professor Rettger for fecal 
 tests ; and to the subjects of the experiment themselves, Messrs. 
 Bauer, Edwards, Lagerquist, Lawton, Mitke, Parmelee, Reeds, Taylor, 
 and Weyman, whose patient submission to the painful tests of endur- 
 ance was little short of heroic. 
 
 In January, 1906, the students above mentioned organized them- 
 selves into an eating club. The experiment began with an endurance 
 test on January 14, and consisted of two main parts, each of which 
 lasted about ten weeks. 
 
 The object of the first half of the experiment was to test the claims 
 which have been made by Mr. Horace Fletcher, as to the effects upon 
 endurance of thorough mastication combined with implicit obedience 
 to appetite. Our conclusion in brief is that Mr. Fletcher's claims, so 
 far as they relate to endurance, are justified. 
 
 Mr. Fletcher's method may be briefly ' expressed in two rules. 
 
 1. MdLStication. Thorough mastication of all food up to the 
 point of involuntary swallowing, with the attention directed, how- 
 ever, not on the mechanical act of chewing, but on the tasting and 
 enjoyment of the food ; liquid foods to be sipped and tasted, not 
 drunk down like water. There should be no artificial holding of food 
 in the mouth beyond the time of natural swallowing, even if, as is to 
 be expected at the start, that swallowing is premature. It is not 
 intended to " count the chews," or hold the food forcibly in the front 
 of the mouth, or allow the tongue muscles to become fatigued by any 
 unnatural effort or position, or in any other way to make eating a 
 bore. On the contrary, every such effort distracts one from 
 the natural enjoyment of food. Pawlow has shown that without 
 such attention and enjoyment of the taste of food, the secretion of 
 
 ' The reader who desires to pursue the subject is referred, as to mastication 
 and instinctive eating, to Higgins, Humaniculture, Stokes, N. Y. , 1906 ; as to 
 proteid, to Chittenden, Physiological Economy in Nutrition, Stokes, 1904 ; and 
 as to the general subject, to Horace Fletcher, The A. B.-Z. of our own Nutri- 
 tion, Stote^, iSds. 
 
Fisher — The Effect of Diet on Endurance. 3 
 
 gastric juice is lessened. The point of involuntary swallowing is 
 thus a variable point, gradually coming later and later as the practice 
 of thorough mastication proceeds, until the result is reached that the 
 food remains in the mouth without effort and becomes practically taste- 
 less. Thus the food, so to speak, swallows itself, and the person eats 
 without thought either of swallowing or of not swallowing it ; swal- 
 lowing is put into the same category of physiological functions as 
 breathing, which ordinarily is involuntary. 
 
 2. following instinct. Never to eat when not hungry, even if a 
 meal (or more than one, for that matter) is skipped. And when a 
 meal is taken, not to be guided by the quantity of food offered, or 
 by past habit, or by any theories as to the amount of food needed. 
 The natural taste or appetite is alone consulted, and the subject 
 selects, from the food available, only those kinds and amounts which 
 are actually craved by the appetite. After practice, the appetite 
 gradually becomes more definite and discriminating in its indica- 
 tions. 
 
 These two rules — thorough mastication and implicit obedience 
 to appetite — were alone employed dui-ing the ten weeks which con- 
 stituted the first half of the experiment. 
 
 Shortly after the beginning of the second half of the experiment, 
 there was an interim of six days at Easter recess, during which the 
 few men who remained in New Haven took advantage of the tempo- 
 rary absence of the cook to try the possibilities of living without one 
 entirely. During this brief period use was made not only of raw foods, 
 such as fruits, nuts and milk, but also of foods which could be pur- 
 chased" already cooked, such as the flaked breakfast foods. But all 
 the food was cold, and several of the men found it cheerless and 
 unsatisfactory. Judging from their feelings, they were losing in 
 weight and vigor. This part of the experiment was too brief, how- 
 ever, to justify any reliable conclusion as to the virtues of a raw, or 
 rather a cookless, diet. 
 
 The second half of the experiment lasted about nine weeks. The 
 same two rules which were employed during the first half were 
 continued during the second, but a third rule was added. This was 
 the use of suggestion, as follows : 
 
 3. When instinct is in doubt, use reason. — This rule consists of 
 acquiring and applying a little knowledge of foods and food elements. 
 For this purpose, in the present experiment two lists of food were 
 given. One was arranged in a tentative order of intrinsic merit, 
 beginning with fruits and ending with alcohol, and the other in 
 
4 Fisher — Tlie Eff'ect of Diet on Endurance. 
 
 the order of the proportion of proteid. The men were then asked, 
 when and only when the appetite was entirely willing, to choose the 
 better and purer foods and the low proteid foods in preference to 
 those high in proteid. In this way the men gradually shifted their 
 diet upward in the two lists, and thereby pursued a little faster the 
 same direction in which they had already been found to be uncon- 
 sciously moving under the influence of thorough mastication and 
 implicit obedience to appetite. 
 
 It would too greatly lengthen this report if any attempt were made 
 to repeat in detail all the specific advice given to the experimenters 
 under Rule 3. What has been said covers in a general way all the 
 points except the advice (subject always to the consent of appetite) 
 to eat light and quickly digested suppers in order to go to bed on an 
 empty stomach. 
 
 Careful record of the amounts of food eaten and the constituents 
 of proteids, fats and carbohydrates was kept for each man each day, 
 certain days being omitted if for any reason the record was incom- 
 plete, as when, for instance, the men were out of town or took their 
 meals away from the club.' To avoid weighing at the table, the 
 food was all weighed in the kitchen and served in "standard portions" 
 of 100 calories each, or simple fractions or multiples thereof, and the 
 men merely recorded the number of portions eaten. The proportions 
 of proteids, fats and carbohydrates were found by means of the 
 writer's "Mechanical Diet Indicator."" Atwater and Bryant's 
 tables were used as a basis for calculation. For the first few weeks 
 the figures were probably subject to some errors, and in all cases 
 more or less guessing had to be practiced with reference to the 
 amount of lean and fat of meats ; but the influence of any errors on 
 the results must necessarily be small, because meat supplied, at 
 the highest, only a small fraction of the total calories. It is believed 
 that the results are in general correct to two significant figures. 
 
 For the first two weeks of the first half of the experiment, the men 
 ate in their ordinary way. During the following eight weeks they 
 masticated more thoroughly and followed the leadings of taste more 
 carefully. Most persons, while nominally following taste, are largely 
 
 ^ The number of days each week on which the record of diet was kept was sel- 
 dom under six. 
 
 ^ For a description of this instrument, see the writer's ' ' A New Method of 
 Indicating Food Values,'' American Journal of Physiology, April, 1906. For a 
 description of its practical uses see "A Graphic Method in Practical Dietetics," 
 Jour, of the Amer. Med. Assoc. , Apr. 20, 1907. 
 
Fisher — The Effect of Diet on Jindurance. 5 
 
 controlled in their selection of foods by many other circumstances, — 
 as, conventionality, or the desire to eat what others eat and the unwil- 
 lingness to appear " different " ; politeness, the desire to please one's 
 host and hostess ifood notions, the opinion that certain foods and cer- 
 tain amounts of food are " wholesome " even if not palatable and that 
 certain foods should be avoided as injurious even if delicious to the 
 taste ; narrowness of choice, as at a boarding house table, which 
 often supplies what is not wanted and withholds what is ; and 
 habit, by which the particular kinds and amounts of foods which 
 have become customary through the previous, causes— conventionality, 
 politeness, food notions, and narrowness of choice — are repeated day 
 after day without thought. The subjects of the present study were 
 given a wide range of choice, the menu including fruits, nuts, cereals, 
 puddings and pastry, vegetables, milk, meats, etc. Meat if desired 
 was available three times a day. 
 
 The object of the experiment was to find what effects on diet and 
 endurance would follow from a strict obedience to the taste-instinct, 
 when this instinct was given a longer chance to act by prolonged 
 mastication and attentive tasting. Each man was therefore encour- 
 aged to choose his own food out of the menu for the day. Nothing 
 was set before him until it was ordered, and even after a food was 
 ordered it was not eaten if taste did not so dictate. The men were 
 specially warned, during the first half of the experiment, against any 
 conscious effort to decrease their food, proteid, or meat ; and while it 
 is possible that subconscious suggestion played a part, so far as could 
 be observed they were freer from its influence than any ordinary 
 experimenter who might take up the same experiment after reading 
 Mr. Fletcher's or Professor Chittenden's books. 
 
 That this conclusion as to the relative absence of subconscious sug- 
 gestion is correct was evidenced by the experiences both before and 
 after this part of the experiment. For a month prior to its actual 
 beginning (Jan. 14), the experiment had been fully decided upon, 
 and its plan and scope understood by the men. Had subconscious 
 suggestion played an important r61e, it would probably have shown 
 itself in a reduction of proteid during this month ; but determinations 
 of the grams of nitrogen daily excreted in the urine, taken at the 
 beginning and end of this month, indicated no substantial change, as 
 the following table shows. (M. does not appear in this table, owing 
 to the absence of any specimen for December.) 
 
6 Fisher— The Effect of Diet on Endurance. 
 
 TABLE I. 
 
 B E Lq Lw P E T W Average 
 
 Middle Dec. 11.3 11.1 13.8 12.3 11.3 13.9 14.3 15.9 13 
 
 Middle Jan. 10.4 13.7 14.3 14.3 11.1 14.8 13.3 15.4 13.1 
 
 On the other hand, during the second half of the experiment (Mar. 
 28-June 1), when the force of suggestion was consciously introduced, 
 the reduction of flesh and proteid went on rapidly, as is seen in 
 Table II. The facts, therefore, seem to show that the men followed 
 directions closely, avoiding largely the influence of subconscious sug- 
 gestion and following that of conscious suggestion in exact accord- 
 ance with the directions given them. 
 
 Changes in Diet. 
 
 During the first two weeks of the first period when no change of 
 habits was undertaken, the food showed little tendency to change in 
 amount or in kinds. On the other hand, for the remaining eight- 
 weeks, during which thorough mastication and instinctive eating 
 were practiced, there was a distinct though gradual tendency toward 
 reduction in the amount of food, in the quantity of proteid, in the 
 quantity of flesh foods, and in the quantity of liquids of all kinds — 
 water, tea, coffee, cocoa, and even soups. Exact figures were kept 
 for calories, proteid and flesh foods. These showed that the total 
 calories gradually fell about 10^, the proteids, 15^, and the flesh 
 foods, 40^, 
 
 In the second period, during which the force of suggestion to 
 reduce proteid and flesh foods was added, the same effects were 
 noted in a still greater degree. During this period the calories 
 dropped nearly 20^ the proteid over 25^, and the flesh foods about 
 70^. Comparing the diet at the close of the entire five months of 
 the experiment with the diet at its beginning, it was found that the 
 total calories had fallen about 25^, the proteid about 40^, and the 
 flesh foods over 80^, or to about one-sixth of their original amount. 
 A part of the reduction, at least of the calories, is probably due to 
 the change in season, as the experiment began in cold weather and 
 closed in hot weather. 
 
 These results are shown in the following table : 
 
Fisher — The Effect of Diet on Endurance. 
 
 TABLE II. 
 AVERAGE DIETETIC RECORDS OF ENTIRE CLUB 
 
 1st 
 Period' 
 
 2iid 
 Period 
 
 
 
 Average 
 
 Calories 
 
 Daily 
 
 
 
 no. of 
 
 of proteid 
 
 ' 'portions" 
 
 
 Average 
 
 "portions" 
 
 per lb. of 
 
 of flesh 
 
 Week 
 
 weight 
 
 daily 
 
 body wt. ' 
 
 foods 
 
 f Jan. 17-33 
 
 149.8 
 
 38.3 
 
 3.7 
 
 3.4 
 
 24-30 
 
 
 30.3 
 
 3.6 
 
 3.1 
 
 31-Feb. 6 
 
 
 37.8 
 
 3.5 
 
 1.6 
 
 Feb. 7-13 
 
 
 37.6 
 
 3.4 
 
 1.3 
 
 14-20 
 
 
 25.8 
 
 3.1 
 
 .9 
 
 21-27 
 
 
 36.4 
 
 3.3 
 
 1.1 
 
 28-Mar. 6 
 
 
 35.3 
 
 2.3 
 
 1.3 
 
 Mar. 7-13 
 
 
 34.6 
 
 3.1 
 
 1.3 
 
 14-20 
 
 
 35.9 
 
 2.2 
 
 1.1 
 
 31-27 
 
 148 
 
 26.7 
 
 3.3 
 
 1.4 
 
 38-Apr. 3 
 
 
 36.7 
 
 2.1 
 
 .8 
 
 Apr. 4-10 
 
 
 35.7 
 
 1.9 
 
 .5 
 
 11-17 i" 
 
 
 37.3 
 
 1.7 
 
 .4 
 
 30-26 3 
 
 
 36.1 
 
 1.7 
 
 .1 
 
 37-May 3 
 
 
 25.5 
 
 1.9 
 
 .5 
 
 May 4-10 
 
 
 35.7 
 
 1.9 
 
 .4 
 
 11-17 
 
 
 36.3 
 
 1.9 
 
 .4 
 
 18-24 
 
 
 34.9 
 
 1.7 
 
 A 
 
 35-31 
 
 
 32.3 
 
 1.4 
 
 A 
 
 June 
 
 144 
 
 
 
 
 Remembering that a "portion " is 100- calories, we see that, during 
 the first four weeks, the men consumed an average of from 2760 to 
 3030 calories per day, of which 120 to 240 were in the flesh foods, 
 such as meats, poultry, fish and shell-fish, and that 2.4 to 2.7 calories 
 of proteid were ingested for each pound of body- weight. Trans- 
 lating Professor Chittenden's figures for the physiological require- 
 ment of ingested proteid, we find it to be from 1.3 to 1.7 calories per 
 pound of body-weight. Thus the men were at this time consuming 
 nearly double the Chittenden allowance. During the last four weeks 
 of the experiment all these magnitudes were lower. The per capita 
 calories ranged from 2220 to 2620, of which only 40 were in flesh 
 foods, and the proteid had fallen to 1.4 to 1.9 calories per pound of 
 body-weight, which corresponds closely to the Chittenden standard. 
 
 Table II was constructed from the following three tables giving 
 separate data for the individual experimenters. 
 
 ' This column is calculated throughout on the basis of the body-weights on 
 Jan. 14. 
 
 2 Except E., M. and P. 
 
 3 Except E. The last two days of the Easter recess, Apr. 18, 19, are omitted 
 in tables II, III, IV, V. 
 
Fisher — The Effect of Diet on Endurance. 
 
 TABLE III. 
 
 DAILY QUANTITIES OF FOOD 
 
 (in "portions" of 100 calories each). 
 
 1st 
 Period " 
 
 3nd 
 Period 
 
 Week 
 
 B 
 
 E 
 
 Lq 
 
 Lw 
 
 M 
 
 P 
 
 R 
 
 T 
 
 W 
 
 Average 
 
 Jan. 
 
 19-23 
 
 25.4 
 
 36.1 
 
 23.4 
 
 32.6 
 
 33.9 
 
 37.0 
 
 33.0 
 
 36.1 
 
 29.4 
 
 28.3 
 
 
 34-30 
 
 27.9 
 
 39.1 
 
 34.3 
 
 32.6 
 
 26.8 
 
 36.3 
 
 34.5 
 
 37.3 
 
 33.9 
 
 30.3 
 
 
 81-Feb. 6 
 
 26.3 
 
 36.4 
 
 35.0 
 
 29.8 
 
 34.4 
 
 33.8 
 
 30.9 
 
 30.8 
 
 33.0 
 
 27.8 
 
 Feb 
 
 7-13 
 
 35.7 
 
 27.8 
 
 25.7 
 
 31.3 
 
 33.0 
 
 23.3 
 
 28.0 
 
 32.8 
 
 31.6 
 
 37.6 
 
 
 14-30 
 
 34.0 
 
 33.4 
 
 24.0 
 
 26.1 
 
 21.6 
 
 20.3 
 
 35.1 
 
 39.3 
 
 39.4 
 
 35.8 
 
 
 31-37 
 
 33.0 
 
 30.9 
 
 23.4 
 
 28.3 
 
 34.5 
 
 18.2 
 
 36.7 
 
 30.3 
 
 33.3 
 
 36.4 
 
 
 38-Mar. 6 
 
 33.3 
 
 25.7 
 
 23.6 
 
 27.4 
 
 34.1 
 
 18.0 
 
 25.4 
 
 29.1 
 
 30.8 
 
 25.3 
 
 Mar 
 
 7-13 
 
 23.3 
 
 33.0 
 
 24.5 
 
 27.4 
 
 26.4 
 
 19.3 
 
 23.5 
 
 33.7 
 
 80.1 
 
 34.6 
 
 
 14-30 
 
 21.9 
 
 31.6 
 
 24.5 
 
 39.1 
 
 30.2 
 
 20.8 
 
 35.0 
 
 30.8 
 
 28.9 
 
 35.9 
 
 L 
 
 31-37 
 
 21.1 
 
 34.3 
 
 25.2 
 
 81.6 
 
 35.9 
 
 21.2 
 
 36.5 
 
 34.4 
 
 30.4 
 
 36.7 
 
 r 
 
 38- Apr. 3 
 
 22.6 
 
 36.3 
 
 26.2 
 
 37.5 
 
 35.3 
 
 23.0 
 
 36.1 
 
 33.5 
 
 30.2 
 
 36.7 
 
 Apr 
 
 4-10 
 
 24.2 
 
 34.3 
 
 24.6 
 
 27.7 
 
 33.5 
 
 23.5 
 
 36.2 
 
 30.5 
 
 28.7 
 
 25.7 
 
 
 11-17 
 
 34.0 
 
 
 26.0 
 
 24.9 
 
 
 
 29.0 
 
 30.4 
 
 29.8 
 
 37.3' 
 
 
 30-26 
 
 26.5 
 
 
 24.1 
 
 34.5 
 
 23"6 
 
 33.3' 
 
 25.5 
 
 32.0 
 
 39.4 
 
 36.1'' 
 
 
 37-May 3 
 
 35.5 
 
 33.6 
 
 25.4 
 
 24.8 
 
 24.1 
 
 24.4 
 
 23.9 
 
 29.0 
 
 39.8 
 
 35.5 
 
 May 
 
 4-10 
 
 23.0 
 
 25.5 
 
 26.5 
 
 34.4 
 
 37.1 
 
 23.1 
 
 26.0 
 
 28.7 
 
 26.7 
 
 35.7 
 
 
 11-17 
 
 21.8 
 
 19.2 
 
 26.5 
 
 27.6 
 
 34.5 
 
 33.8 
 
 27.6 
 
 30.3 
 
 34.2 
 
 26.3 
 
 
 18-24 
 
 33.7 
 
 19.1 
 
 33.7 
 
 27.8 
 
 19.5 
 
 35.4 
 
 34.6 
 
 26.5 
 
 35.2 
 
 24.9 
 
 ^ 
 
 35-31 
 
 19.3 
 
 17.3 
 
 33.6 
 
 37.1 
 
 17.8 
 
 31.9 
 
 34.3 
 
 20.3 
 
 28.6 
 
 33,2 
 
 From this table we see that there were wide differences between 
 the men in regard to the change in the quantity of food. During 
 the first period the, men who reduced their calories conspicuously 
 were B., P. and R., the very men, as Table IX will show, who lost 
 weight during this period. 
 
 During the second period, reductions were noticeable in E., Lw., M. 
 and T. These, together with B. and P., were the men who lost 
 weight during the second period. We see here a distinct correlation 
 between quantity of food and loss of body-weight. 
 
 TABLE IV. 
 PSOTEID (in calories) PER LB. OP BODY-WEIGHT 
 
 (Body-weight as taken Jan. 14, 1906) 
 
 Week 
 
 B E Lq Lw M 
 
 _ Isfc , 
 Period ' 
 
 Jan, 
 
 Feb. 
 
 Mar. 
 
 17-23 
 34-30 
 31-Feb. 6 
 
 7-13 
 14-20 
 21-27 
 38-Mar. 6 
 
 7-13 
 14-20 
 21-27 
 
 3.6 
 2.5 
 2,2 
 2,1 
 2.0 
 1,8 
 1.8 
 1.9 
 1.9 
 1.6 
 
 2.9 
 2.8 
 3.1 
 2,9 
 3,1 
 3.2 
 3.0 
 2.5 
 2.8 
 2.6 
 
 1.8 
 2.2 
 2,4 
 2.2 
 1.8 
 1.9 
 1.8 
 2.0 
 1.9 
 3.0 
 
 3,8 
 3,8 
 3.9 
 3.6 
 2.0 
 3.3 
 3.3 
 2.3 
 2.5 
 2.7 
 
 2.6 
 2,7 
 2,5 
 3.0 
 1.9 
 3.0 
 3,4 
 2.5 
 2.7 
 2.4 
 
 2.8 
 2,5 
 3.0 
 3.1 
 1.7 
 1.5 
 1.5 
 1.7 
 1.7 
 1.7 
 
 2.5 
 2.5 
 2,3 
 3,0 
 1.8 
 1.9 
 1.8 
 1.7 
 1.9 
 1.9 
 
 3.8 
 8.0 
 
 2.7 
 2.9 
 2.7 
 3.8 
 3.6 
 2.1- 
 2.8 
 2.7 
 
 W Average 
 
 2.7 
 2.6 
 2.5 
 2.4 
 3.1 
 3.2 
 2.3 
 2.1 
 2.2 
 2.2 
 
 ^Not including E., M. and P. 
 
 E. 
 
Maker — The Effect of Diet on Endurance. 
 
 TABLE V^— Continued. 
 
 PROTEID (in calories) PEE LB. OP BODY-WEIGHT 
 
 (Body-weight as taken Jan. 14, 1906) 
 
 
 Week 
 
 B 
 
 E 
 
 Lq 
 
 Lw 
 
 M 
 
 P 
 
 R 
 
 T 
 
 W 
 
 Average 
 
 'Mar. 
 
 28-Apr. 3 
 4-l(f 
 
 1.8 
 
 2.6 
 
 3.0 
 
 3.3 
 
 3.3 
 
 1.9 
 
 1.8 
 
 3.7 
 
 3.3 
 
 3.1 
 
 Api\ 
 
 1.5 
 
 2.9 
 
 1.7 
 
 3.0 
 
 1.7 
 
 1.5 
 
 1.8 
 
 3.4 
 
 1.8 
 
 1.9 
 
 
 11-17 
 
 1.2 
 
 
 1.8 
 
 1.7 
 
 
 
 1.7 
 
 1:9 
 
 1.9 
 
 ,1.7 
 
 
 20-26 
 
 1.8 
 
 
 1.6 
 
 1.6 
 
 l"7 
 
 l"5 
 
 1.5 
 
 3.3 
 
 1.6 
 
 1.7 
 
 
 27-May3 
 
 1.7 
 
 3" 5 
 
 1.7 
 
 1.7 
 
 3.0 
 
 1.7 
 
 1.6 
 
 3.2 
 
 1.8 
 
 1.9 
 
 May 
 
 4-10 
 
 1.6 
 
 8.0 
 
 1.9 
 
 1.9 
 
 2.2 
 
 1.4 
 
 1.8 
 
 3.1 
 
 1.5 
 
 1.9 
 
 
 11-17 
 
 1.6 
 
 1.9 
 
 1.7 
 
 1.9 
 
 1.9 
 
 1.6 
 
 1.9 
 
 3.1 
 
 3.0 
 
 1.9 
 
 
 18-24 
 
 1.5 
 
 1.8 
 
 1.4 
 
 1.7 
 
 1.8 
 
 1.6 
 
 1.6 
 
 1.7 
 
 3.1 
 
 1.7 
 
 
 25-81 
 
 1.1 
 
 1.3 
 
 1.5 
 
 1.6 
 
 1.6 
 
 1.3 
 
 1.7 
 
 1.4 
 
 1.5 
 
 1.4 
 
 3nd 
 Period ' 
 
 We observe from Table IV that the men who reduced their proteid 
 the most during the first period were B., P., R., T. and W. Of these 
 the first three only lost weight appreciably, and this was partly ascrib- 
 able, as we have seen, to reduction in their calories. Careful 
 examination of the figures would seem to show, however, that there is 
 some correlation between reduction of proteid and loss of weight. 
 
 During the second period there was a decided reduction of proteid 
 in all cases except that of R., who had already brought his proteid 
 down considerably in the first period. E. reduced his proteid, but not 
 until the last three weeks, when he seemed to try to make up for lost 
 time. E., in fact, was the only man in the club, except possibly M. 
 who (through mere inadvertence) did not follow out the rules of the 
 experiment systematically. It need scarcely be said that this is not 
 stated as censure ; for the very fact of the moderation of E.'s and M.'s 
 mastication added to the value of the final comparisons. Even E.'s 
 sudden reduction in proteid at the end was not maintained two weeks 
 afterward, as was shown by the excretion of nitrogen in June, 
 given in Table VI. 
 
 It will be observed that the proteid at the end of the experiment 
 was reduced to a fairly uniform level for all the men. Moreover, the 
 proteid at the end corresponds closely with the results of Professor 
 Chittenden's experiments. This is especially significant in view of 
 the fact that this level was reached unconsciously— for only one of 
 the men, Lq., who used the mechanical diet indicator for the entire 
 club, knew regularly the exact character of each man's food propor- 
 tions — and without any food prescription as was employed in the 
 experiments of Professor Chittenden. This means that there is 'a 
 simple way of reducing proteid to the level of " physio logical econ- 
 
10 
 
 Fisher— Tlie Effect of Diet on Endurance. 
 
 omy," open to the ordinary man, without the necessity of special 
 knowledge of foods and without the necessity of weighing and measur- 
 ing food, either by the subject himself or by others. 
 
 Aside from the changes in proteid, the proportions of food elements 
 did not vary greatly, the percentages of fat and carbohydrate in the 
 total fuel value remaining very nearly constant. At the close of the 
 experiment it was found that for all of the men the proteid in propor- 
 tion to the total fuel value was very nearly 10^, having been reduced 
 from about 14^. This reduction in the percentage of proteid was 
 almost entirely offset by the increase in the percentage of fat, which 
 rose from about 30^ to about 33^ on the average. The percentage 
 of carbohydrate thus remained almost constant. Individual varia- 
 tions were much less than might have been expected. The proteid 
 at the close of the experiment among the different subjects deviated 
 very little from 10^ ; the proportion of fat varied from 28 to 36^) 
 and the carbohydrate from 51 to 62^. 
 
 The results of the experiment may throw some light on the 
 problem of the proper amount of food and food constituents for 
 healthy men eating in a natural manner. For the five men, Lq., Lw.i 
 M., R. and W., whose weights showed least tendency to fall and 
 whose average weight at the close of the experiment was 151.4, we 
 find the average total calories were 2620, of which 10.7$^ was proteid, 
 33^ fat, and 56.3^ carbohydrate. The number of calories agrees 
 closely with the estimates (for sedentary persons) of Atwater and 
 Benedict by means of the calorimeter. 
 
 TABLE V. 
 
 QUANTITIES OF FLESH FOODS CONSUMED, (meat, fish, 
 shell-fisli, poultry) 
 
 (In "portions" of 100 calories, each). 
 
 1st 
 Period ' 
 
 Week 
 f Jan. 17-23 
 
 24-30 
 
 31-Feb. 6 
 Feb. 7-18 
 
 14-30 
 
 21-27 
 
 28-Mar. 6 
 Mar. 7-13 
 
 14-20 
 
 21-27 
 
 BELqLwM P E TW Average 
 
 2.7 
 
 2.3 
 
 1.8 
 
 1.3 
 
 .9 
 
 1,2 
 
 .8 
 
 .7 
 
 1.0 
 
 1.0 
 
 2.4 
 2.2 
 1.9 
 1.6 
 2.3 
 2.0 
 2.1 
 1.6 
 1.7 
 2.5 
 
 1.5 
 1.7 
 1.2 
 .9 
 .3 
 .2 
 .3 
 .1 
 
 2.6 
 2.0 
 2.1 
 1.7 
 1.1 
 1.6 
 1.5 
 2.4 
 
 .08 1.9 
 .0 2.5 
 
 2.4 
 
 1.8 
 
 1.6 
 
 .6 
 
 .7 
 
 .6 
 
 1.1 
 
 1.0 
 
 .9 
 
 1.7 
 
 2.6 
 2.2 
 1.3 
 
 1.2 
 
 1.1 
 
 1.2 
 .7 
 .7 
 
 2.3 
 2.1 
 2.1 
 1.5 
 .9 
 1.6 
 1.6 
 1.9 
 1.8 
 1.3 
 
 3.9 
 3.1 
 1.9 
 1.8 
 1.6 
 3.1 
 1.5 
 1.6 
 1.5 
 3.5 
 
 2.1 
 
 2.2 
 
 1.4 
 
 1.3 
 
 .5 
 
 .6 
 
 .8 
 
 .9 
 
 .7 
 
 .6 
 
 2.4 
 2.1 
 1.6 
 1.2 
 .9 
 1.1 
 1.2 
 1.3 
 1.1 
 1.4 
 
Fisher — IVie Effect of Diet on Endurance. 
 
 11 
 
 TABLE Y —Continued. 
 
 QUANTITIES OF FLESH FOODS CONSUMED (meat, flsli, 
 
 shell-fish, poultry) 
 
 (In "portions" of 100 calories each). 
 
 3nd 
 Period 
 
 Week 
 
 B E 
 
 Lq 
 
 Lw 
 
 M 
 
 P 
 
 E 
 
 T 
 
 "W 
 
 Aver 
 
 fMar. 38-Apr. 3 
 
 .9 1.7 
 
 .0 
 
 1.7 
 
 .6 
 
 .8 
 
 1.1 
 
 .8 
 
 .1 
 
 .8 
 
 Apr. 4-10 
 
 .4 1.1 
 
 .0 
 
 1.0 
 
 1.0 
 
 .1 
 
 .7 
 
 .8 
 
 .0 
 
 .5 
 
 11-17 
 
 .1 .. 
 
 .0 
 
 .9 
 
 
 
 .8 
 
 .7 
 
 .0 
 
 .4 
 
 30-36 
 
 .0 _. 
 
 .0 
 
 .1 
 
 "i 
 
 "6 
 
 .3 
 
 .4 
 
 .0 
 
 .1 
 
 37-May- 3 
 
 .16 1.3 
 
 .0 
 
 .6 
 
 .8 
 
 .3 
 
 .9 
 
 .5 
 
 .0 
 
 .5 
 
 May 4-10 
 
 .0 .9 
 
 .0 
 
 .8 
 
 .8 
 
 .0 
 
 .5 
 
 .6 
 
 .0 
 
 .4 
 
 11-17 
 
 .0 1.3 
 
 .0 
 
 .7 
 
 .8 
 
 .0 
 
 .4 
 
 .3 
 
 .0 
 
 .4 
 
 18-34 
 
 .0 1.3 
 
 .0 
 
 .5 
 
 .5 
 
 .0 
 
 .8 
 
 .3 
 
 .0 
 
 .4 
 
 35-31 
 
 .0 1.0 
 
 .0 
 
 1.0 
 
 .9 
 
 .0 
 
 .8 
 
 .4 
 
 .0 
 
 .4 
 
 Table V shows that during the first period all except E. and Lw. 
 reduced their consumption of flesh foods considerably. It is note- 
 worthy, as Tables XI-XIII will show, that these two were the men 
 whose improvements in endurance were probably among tlie least 
 during this period. 
 
 During the second period Lq.,'W., P. and B. virtually abandoned 
 flesh foods entirely, the " portions " consumed daily averaging nearer 
 zero than .1. These men improved greatly in endurance also. On 
 the other hand, E., Lw., M. and R. reduced their flesh foods the 
 least, and their ranking in respect to increased endurance was in 
 general relatively low. 
 
 Excretions, Body- Weight, Strength. 
 
 The following table of nitrogen excreted in the urine is interesting 
 in connection with the preceding table. It will be seen that the 
 reduction in nitrogen daily excreted corresponds in general to the 
 reduction in proteid consumed. 
 
 
 
 TABLE VI. 
 
 
 
 
 
 
 GRAMS OF NITROGEN F.XCRETED DAILY ^ 
 
 
 
 
 B 
 
 E Lq Lw 
 
 M 
 
 P 
 
 R 
 
 T 
 
 "W 
 
 Middle Jan. 
 First April 
 Middle June 
 
 10.4 
 6.6 
 6.3 
 
 13.7 14.3 14.3 
 14.7 9.3 11.1 
 13.1 8.4 ... 
 
 8.72 
 13" 7 
 
 11.1 
 6.3 
 6.1 
 
 14.8 
 11.6 
 
 13.3 
 
 13.4 
 
 8.8 
 
 15.4 
 9.0 
 9.4 
 
 N. in middle" 
 
 
 
 
 
 
 
 
 June per 
 kilog. of f 
 hody-weight- J 
 
 .093 
 
 .33 .13 ._. 
 
 .31 
 
 .09 
 
 ... 
 
 .13 
 
 .13 
 
 ' Each figure is obtained by averaging 3 or 3 consecutive days' specimens. 
 2 Jan. 33 and Feb. 10. 
 
12 Fisher — The Effect of Diet on Endurance. 
 
 This table shows that all the men excepting E. and M. greatly- 
 reduced their nitrogen excretion during the experiment, and that at 
 the close (with the two exceptions noted) the men were on about the 
 same nitrogen level as the subjects of Professor Chittenden's exper- 
 iment, namely, near one-tenth of a gram of nitrogen per kilogram of 
 body-weight. 
 
 Through the kindness of Professor Benedict of Wesleyan Univer- 
 sity, nitrogen analyses were made in December, 1906, six months 
 after the close of the experiment, to discover to "what extent the men 
 had adhered to their newly acquired diet after the eating club in 
 which it had been practiced was disbanded. The results were B. 
 11.0, Lq. 10.5, Lw. 7.9, M. 9.9, P. 6.8, R. 11.5, T. 11.9, W. 8.9. These 
 show that half of the men had reverted to some extent toward their 
 original diets. The men state that the reason for this reversion was 
 the difficulty in selecting food differing greatly in kind and amount 
 from that customarily served at their boarding houses. 
 
 The following table shows that the volume of urine daily excreted 
 was greatly reduced during the experiment : 
 
 TABLE VII. 
 VOLUME OP URINE DAILY EXCEETED (in cubic centimeters) 
 
 B ELqLwMPR TW 
 
 Middle Jan. 
 
 1425 
 
 1160 
 
 1130 
 
 1391 
 
 8171 
 
 706 
 
 1387 
 
 1793 
 
 1177 
 
 First April 
 
 630 
 
 985 
 
 900 
 
 1253 
 
 
 639 
 
 1035 
 
 980 
 
 797 
 
 Middle June 
 
 803 
 
 1130 
 
 823 
 
 
 
 785 
 
 480 
 
 .'-.. 
 
 696 
 
 970 
 
 From this table we see a striking reduction in the volume of urine 
 excreted, with the saine two notable exceptions, E. and M. These 
 two, who reduced their excretions least, were the men who were the 
 least assiduous in observing the rules of the experiment. 
 
 A careful examination of the feces was made by Professor L. F. 
 Rettger of the Sheffield Scientific School. A summary of his report 
 follows. In it was included a comparative statement for three sets 
 of specimens of two days each, taken in January, March and June, 
 referred to below as series I, II and III. These included data as to 
 the color, odor, quantity, consistency, approximate determination of 
 the number and predominant kinds of bacteria, putrefactive and 
 fermentative properties, and a true microscopic bacterial examination. 
 In brief, to quote from Dr. Rettger's report and letter : 
 
 ' Jan. 33 and Feb. 10. 
 
Fisher— The Effect of Diet on Endurance. 13 
 
 " The odor was very slight in almost every specimen in the last 
 series, a marked difference between these and series I and II, 
 particularly I. The average weight is less than in both series I and 
 
 II (av. wt. of series I = 13'7.3 grams; of II = 164.8 grams, and 
 
 III = 120.4 grams). 
 
 " The figures indicate considerable difference in the putrefactive and 
 fermentative properties of the three series, and the decrease is progres- 
 sive. In series I the amount of proteid dissolved was much larger 
 than in II and III. . . The specimens [of series III] were more 
 solid generally than in both previous series. I was unable to note 
 any appreciable difference in the microscopic appearance of the last 
 series as compared with the previous, except that in specimen B of 
 the last series a large number of moulds were present-. This has little 
 significance, however." 
 
 TABLE VIII. FECAL TESTS' 
 
 PUTBEFACTIVE DeuREE FERMENTATIVE PROPERTY 
 
 Mid. End Mid. Mid. End Mid. 
 
 Jan. March June Jan. March June 
 
 B 
 
 ! 
 
 « 
 
 H 
 
 + + 
 
 + 
 
 + 
 
 E 
 
 m 
 
 35^ 
 
 m 
 
 + + 
 
 + 
 
 + 
 
 M 
 
 30^ 
 
 10% 
 
 15^ 
 
 + 
 
 + 
 
 + 
 
 Lq 
 
 60^ 
 
 ? 
 
 50^ 
 
 + + 
 
 + 
 
 + 
 
 Lw 
 
 35% 
 
 15^ 
 
 t 
 
 + + 
 
 + 
 
 t 
 
 ;b 
 
 30^ 
 
 ? 
 
 ? 
 
 + 
 
 — 
 
 
 E 
 
 30^ 
 
 30^ 
 
 t 
 
 + + 
 
 + 
 
 T 
 
 T 
 
 30^ 
 
 . 1 
 
 5% 
 
 ? 
 
 + 
 
 + 
 
 w 
 
 30^ 
 
 \0% 
 
 5% 
 
 + + 
 
 + 
 
 + 
 
 We here observe that the degree of putrefaction in the last two 
 tests was usually considerably less than its magnitude in the first 
 test. The least change in the feces occurred in the cases of Lq., M. 
 and E., and the greatest changes in P., T. and W. Here again we 
 find some correspondence between the assiduity of the men and the 
 observed physiological changes ; for E. and M. were the least and P. 
 and W. the most careful among the experimenters. 
 
 A critic has raised the question whether the improvement in the 
 feces indicates lessened absorption of poisons, and whether, if the 
 feces were longer retained, the improvement in their character might 
 not be in consequence of the abstraction from them and absorption 
 
 ' In the table, " + "_ signifies the presence and " — " the absence of fermenta- 
 tive property ;" + -)-" represents a high degree of fermentative property ; " ?" 
 signifies that the putrefactive degree was doubtful, if not absent. 
 
 j; No specimen. 
 
14 Fisher — The Eff'ect of Diet on Endurance. 
 
 into the system of a larger amount of poisons. The length of time of 
 retention of the feces was not measured in any way. So far as can be 
 guessed from the impressions of the men, it was not lengthened, cer- 
 tainly not greatly, as in the case of Mr. Fletcher. ' As to the sig- 
 nificance of the improvement in feces, Dr. Rettger writes : 
 
 " The subject of intestinal putrefaction is one of which very little 
 is as yet known. A retention of feces may have the tendency of 
 lowering the amount of putrefactive products. This is due, I believe, 
 to two things : first, an absorption of such products as indol and 
 mercaptan ; and second, an unusual amount of antagonistic action 
 exerted on the evil-producing (putrefactive) bacteria by the ordinary 
 and presumably helpful bacteria. Recent work seems strongly to 
 emphasize the latter point. . . There is nothing to show that a 
 small degree of retention would make a very great difference. 
 
 " The absence of appreciable amounts of putrefactive bodies from 
 feces under the ordinary conditions of peristalsis does, beyond a 
 doubt, indicate a lessened production of the products (toxines) ; the 
 system must be the better off on account of this . . . the inter- 
 pretation of the facts must be dealt with rather cautiously." 
 
 The following table shows the body-weights of the men (after 
 deducting weight of clothing). 
 
 TABLE IX. 
 BODY-WEIGHTS IN POUNDS (withoul clotMng) 
 
 
 B 
 
 E 
 
 Lq 
 
 Lw 
 
 M 
 
 P 
 
 E 
 
 T 
 
 W 
 
 Average 
 
 Jan. 14 
 
 148 
 
 127 
 
 147 
 
 153 
 
 141 
 
 144 
 
 179 
 
 156 
 
 153 
 
 149.8 
 
 Mar. 38 
 
 144 
 
 138 
 
 147 
 
 154 
 
 143 
 
 136 
 
 176 
 
 155 
 
 151 
 
 148 
 
 June 16 
 
 138 
 
 133 
 
 146 
 
 149 
 
 138 
 
 181 
 
 175 
 
 148 
 
 149 
 
 144 
 
 We see that during the first period, the weights, except of P., 
 remained practically stationary, but that in the second period all of 
 the men lost somewhat in weight, though the loss was trifling in most 
 cases. The only substantial losses during the two periods combined 
 were : P. 13 lbs., B. 10 lbs., and T. 8 lbs. Of these it may be said 
 that B. was distinctly over his normal weight at the start. 
 
 The distinct correlation between the loss of weight and the reduction 
 in food, and to some extent in proteid, has already been noted. P.'s 
 loss is ascribable largely to overstudy. The general slight reduction 
 in weight of the entire club is probably explained in the same way, 
 for all the men, with possibly two exceptions, distinctly overstrained 
 
 ' See The A. B.-Z. of our own Nutrition. 
 
Fisher— The Effect of Diet on Endurance. 
 
 15 
 
 in their college work. Besides the influence of overwork, there was 
 also present the influence of the season,— at least if the common 
 impression is correct that persons usually lose weight with the 
 approach of warm weather. 
 
 Gymnasium tests were made at the beginning, middle and end of 
 the experiment. These tests were of two kinds,— tests of strength 
 and tests of endurance. The times of the tests were widely separated, 
 partly because those of endurance were too exhausting to be often 
 repeated, and partly because it was desired to avoid the influence of 
 "practice"; for not only does practice increase strength and en- 
 durance, but it also gives the users of the strength-registering 
 apparatus a facility or " knack " in manipulating it which produces 
 a false appearance of improvement. 
 
 The dates of the three tests were January 14, a week after the end 
 of the Christmas vacation ; March 28, just before the Easter recess ; 
 and June 16, ' just before the summer vacation. 
 
 Tests of strength, taken at the beginning, middle and end of the 
 experiment, show the following effects : 
 
 TABLE X. 
 STEENGTH TESTS (in lbs.) 
 
 
 Date 
 
 B 
 
 E 
 
 Lq Lw 
 
 M 
 
 E. Grip 
 
 ( Jan. 14 
 
 ■ Moh. 28 
 
 Jun. 16 
 
 100 
 93 
 91 
 
 135 
 130 
 130 
 
 95 98 
 110 100 
 110 93 
 
 145 
 113 
 105 
 
 L. Grip 
 
 { Jan. 14 
 \ Moh. 38 
 ( Jun. 16 
 
 94 
 
 83 
 75 
 
 115 
 115 
 115 
 
 83 112 
 
 110 100 
 
 97 105 
 
 107 
 105 
 105 
 
 Back Lift 
 
 ( Jan. 14 
 
 ■ Mch. 38 
 
 Jun. 16 
 
 375 
 260 
 380 
 
 400 
 440 
 300 
 
 280 340 
 375 380 
 390 250 
 
 375 
 275 
 330 
 
 Leg Lift 
 
 ( Jan. 14 
 \ Mch. 38 
 ( Jun. 16 
 
 530 
 515 
 400 
 
 600 
 600 
 545 
 
 370 400 
 455 450 
 440 445 
 
 460 
 410 
 400 
 
 Total 
 
 ( Jan. 14 
 \ Moh. 38 
 ( Jun. 16 
 
 989 1250 837 950 1087 
 950 1285 1050 1030 902 
 846 1090 937 892 940 
 
 P E T W Average 
 
 95 132 120 125 116 
 
 102 127 130 1)7 113 
 
 103 125 121 115 110 
 
 82 131 95 127 105 
 
 70 115 106 111 103 
 
 70 135 98 110 100 
 
 250 360 370 365 335 
 
 375 890 400 400 355 
 
 265 345 330 364 306 
 
 320 830 545 635 519 
 
 415 865 570 650 548 
 
 300 610 520 650 479 
 
 747 1443 1130 1253 1075 
 
 863 1497 1306 1378 1118 
 
 738 1205 1069 1289 995 
 
 In this table we see that during the first period there was a slight 
 increase in strength (from an average " total " strength of 1076 to 
 1118), and during the second period a slight fall to 995, which is 
 about 12^ from the mid -year's 1118, and about 8^ from the original 
 
 ^ But May 31 for E., Lw.,,R. and W., on account of eariier examinations than 
 the others, necessity to leave town, etc. 
 
16 Fitlier—The Effect of Diet on Endurance. 
 
 1076. Thus the strength of the men remained nearly stationary 
 throughout the experiment. The greatest losses were those of B., E. 
 and K., whose records fell respectively from 989 to 846, 1250 to 
 1090, and 1443 to 1205. 
 
 The loss of strength, like the loss of weight, seems most probably 
 explainable by the overstudy of the men. This cause was certainly act- 
 ively at work, and would apply in the case of all of the club with possi- 
 bly two exceptions. Overstudy applied conspicuously to B. and R., 
 both of whom not only overworked during the entire period of the 
 experiment, but had, just before coming to the last test, been through 
 the most exhausting and sleep-robbing week of all. There seems, 
 therefore, little reason to ascribe any part of the slight losses of 
 strength to the dietetic experiment itself. 
 
 This opinion is confirmed by two facts : One is that the man who 
 was least affected dietetically by the experiment was E., one of 
 the three largest losers of strength, while the men who were most 
 affected dietetically were P. and W., neither of whom lost strength 
 perceptibly, in spite of P.'s severe overwork and loss of weight. The 
 other fact is that in Professor Chittenden's experiment, which dietet- 
 ically was very similar, the subjects, who were soldiers and athletes 
 and not subject to pressure of work of any kind, showed large gains 
 in strength. From these two facts we may infer that, so far as the 
 diet is concerned, the effect would be to increase rather than to 
 decrease strength. 
 
 Changes in Physical Endurance. 
 
 It is fortunate that the strength of the men remained so nearly 
 stationary ; for it demonstrates the more clearly that the increase 
 in endurance which will be shown below was an increase in endur- 
 ance per se, and not in any degree due to an increase in strength. 
 Strength and endurance are entirely distinct and should be separately 
 measured. The strength of a muscle is measured by the utmost force 
 which it can exert once ; its endurance, by the number of times it 
 can repeat a given exertion well within its strength. 
 
 After much consideration and consultation it ■ was decided not to 
 place reliance on the ordinary ergographs as a means of measuring 
 endurance.' Instead, seven simple gymnastic tests of physical endur- 
 
 ''■ The reasons, in brief, were (1) because these ergographs are adapted to testing 
 only a few unimportant, and for the most part unused, muscles ; (3) because, in 
 operating these devices, the subjects do not simulate real work, since the mus- 
 
Fisher — The Efect of Diet on Endurance. \1 
 
 ance were employed, and one of mental endurance. The seven phys- 
 ical tests were : 
 
 (1) Rising on the toes as many times as possible. 
 
 (2) Deep knee-bending, or squatting as far as possible and rising to 
 
 the standing posture, repeating as often as possible. 
 
 (3) While lying on the back, raising the legs from the floor to a 
 
 vertical position and lowering them again, repeating to the 
 point of physical exhaustion. 
 
 (4) Raising a 5-lb. dumb-bell (with the triceps) in each hand from 
 
 the shoulder up to the highest point above the head, repeating 
 to the point of physical exhaustion. 
 
 (5) Holding the arms from the sides horizontally for as long a time 
 
 as possible. 
 
 {6) Raising a dumb-bell (with the biceps) in one hand from a position 
 in which the arm hangs down, up to the shoulder and lower- 
 ing it again, repeating the motion to the point of physical 
 exhaustion. This test was taken with four successive dumb- 
 bells of decreasing weight, viz., 50, 25, 10 and 5 lbs. respectively. 
 
 C?) Running on the gymnasium track at a speed to suit the subject, 
 to as great a distance as possible. 
 
 The mental test consisted of adding specified columns of figures 
 as rapidly as possible, the object being to find out whether the 
 rapidity of performing such work tended to improve during the 
 experiment. 
 
 From the wisdom born of experience it may be stated that the 
 physical tests were too numerous and too severe. But after they 
 
 eles are placed in an awkward and unnatural position in wMoh " no purchase " 
 is felt ; (3) because experience has shown that subjects waste their effort by 
 expending it not only while raising but while lowering the weight, and that this 
 waste during the period of relaxation varies greatly with different subjects ; (4) 
 because a fixed weight is used instead of a weight proportionate to the different 
 strengths of the various subjects. One might as well attempt to test the walk- 
 ing powers of a woman weighing 100 lbs., as compared with those of a man weigh- 
 ing 300 lbs., by compelling the woman to carry a 100-lb. weight so that she 
 might walk with the same weight as the man. Some of these objections have 
 been met in special instruments, such as that of Prof. W. S. Hall of Northwest- 
 ern University. 
 
 After the experiment was half over, and too late to make use of it, the writer 
 devised an ergograph which, it is believed, meets all of the above objefetions. 
 He was led to do so by the fact that the tests employed were so frightfully 
 exhausting to the men. A description of the new ergograph will be published 
 later. It is to be employed in further tests. 
 
 Tbans. Conn. Acad., Vol. Xin. , 3 Mat, 1907. 
 
18 Fisher — The Effect of Diet on Endurance. 
 
 were once adopted in January, it was necessary in subsequent tests 
 to adhere to them, so far at least as always to begin with the same 
 test and follow the same sequence of tests as far as the series was 
 repeated. It is clear that one's ability to succeed in an individual 
 test would depend greatly on what and how many tests had immedi- 
 ately preceded ; consequently the only modifications in the January 
 tests which could legitimately be adopted in IMarch consisted in omit- 
 ting all tests after the first two or three. These first two or three, 
 being taken under the same conditions as before, reflected correctly 
 any change in endurance so far as those particular tests were con- 
 cerned. 
 
 At the final series of tests in June, no omissions from the January 
 program were made ; to save time, however, the last two parts of 
 test 6, together with test 7 (which came at the end for all the men),^ 
 were repeated only up to the point at which they had been carried in 
 January, although the men were able in June to carry them much 
 further, and in many cases did so of their own accord. One man, for 
 instance (W.), who in the run in. January was glad to stop at 10 
 laps, went on in June to 34, running at the same speed until near 
 the end ; and this was done after having more than doubled his 
 former records in almost all of the other tests. The unlooked-for 
 increase in endurance made the June tests much more time-consuming 
 than the tests in January and March. Had the men in June taken 
 test 7, and the two last parts of test 6 up to the same fatigue limit 
 as in January, some of them would have had to remain in the gym- 
 nasium (supperless) until bed time. One of the men, who in January 
 in the last two parts of test 6 raised the 10-lb. dumb-bells 318 times 
 and the 5-lb. dumb-bells 1,863 times, without doubt could have raised 
 them in June double and probably treble these numbers, but to have 
 done so would have consumed of itself an hour and a half of extra 
 time. 
 
 In view, therefore, of the only partial repetition of test 7 and the 
 last two parts of test 6, these records are omitted from Table XI. 
 The first part of test 6 (lifting the 50-lb. dumb-bell) is also omitted, 
 being given separately below. 
 
 The following table (XI) shows the results of the three sets of tests 
 in January, March and June. ' This table will repay careful study. 
 
 ^ The order in which the tests weref taken was not the same for all of the nine 
 men, owing to the lack of a suflScient number of gymnasium assistants in taking 
 the tests. Bnt care was taken that each man should himself preserve the same 
 order in all three series of tests. Thus, for the March series he took the first two 
 
Fisher — The Effect of Diet on Endurance. 19 
 
 From it we see that witli one exoeptiou (E.) all of the men had im- 
 proved in the March and June tests as compared with the January 
 tests, and the eight men who did improve showed improvement in 
 every test, except Lq., Lw. and T., who showed slight falling off in 
 individual cases. 
 
 As inspection will show, some of the increases are remarkable. 
 The recorded increases in the 60 odd cases were, with a few excep- 
 tions noted below, all true increases and not due to increased effoi't 
 to break a previous record. In anticipation of such possible effect 
 of ambition, the men were urged in the January tests to the utmost 
 limit they could or would stand. The original intention had been to 
 work each muscle tested until it was physically unable to repeat the 
 motion, but this was not usually found practicable, except in tests 3, 4 
 and 6, and in some cases 2. In the other tests the will gave out before 
 the muscles. The March and June tests were so managed that when 
 a man had surpassed his January record he was not allowed to pro- 
 ceed beyond the degree of fatigue which he had reached in the first 
 test. This was usually not a difficult matter, as the fatigue in 
 January had been excessive and the men had no desire to suffer 
 again the painful after-effects. Hence, with the exceptions to be 
 noted, the March and June records not only . exceeded those of Jan- 
 uary, but were accomplished with much less fatigue. The actual 
 improvement was therefore greater than the recorded improvement. 
 
 or three tests wMoh. he had taken in January. This explains why, in the March 
 series, the tests as shown in the tables are not the same for all the men. The 
 order of the January and June tests for the diffei-ont men is given below. The 
 tests which were taken in March are in italics. 
 
 B 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 E 
 
 1 
 
 3 
 
 3 
 
 4 
 
 5 
 
 Lq 
 
 1 
 
 2 
 
 Ji- 
 
 3 
 
 5 
 
 Lw 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 M 
 
 1 
 
 5 
 
 2 
 
 3 
 
 4 
 
 P 
 
 3 
 
 1 
 
 5 
 
 3 
 
 4 
 
 E 
 
 1 
 
 3 
 
 3 
 
 4 
 
 5 
 
 T 
 
 1 
 
 3 
 
 3 
 
 4 
 
 5 
 
 W 
 
 3 
 
 5 
 
 1 
 
 3 
 
 4 
 
20 
 
 Fisher — The Effect of Diet on Endurance. 
 
 (1) 
 
 Rising 
 on 
 
 Toes 
 
 TABLE XI. 
 
 TESTS OF PHYSICAL ENDUEANCE. 
 
 B E Lq Lw M P 
 1482 
 
 Deep 
 
 (8) -j Knee 
 Bending 
 
 (3) 
 
 Leg 
 Raising 
 
 5-lb. 
 
 (4) -j Dumb-bell 
 ( (triceps) 
 
 f Holding 
 
 (5) -j Arms 
 
 [ Horizontal 
 
 ( 25-lb. 
 
 (6) ] Dumb-bell 
 ( (biceps) 
 
 Jan. 
 Mar. 
 June 
 
 Jan. 
 Mar. 
 June 
 
 Jan. • 
 
 Mar. 
 
 June 
 
 Jan. 
 Mar. 
 June 
 
 300 
 400'' 
 500 = 
 
 83 
 
 300" 
 
 35 « 
 
 '33« 
 
 75 s 
 
 137" 
 
 E Lq 
 
 1007 333 ' 
 1365' 3630 ■i 
 1061' 3000° 
 
 143 
 
 "si' 
 
 53 « 
 
 '38 « 
 138" 
 
 '59 » 
 
 70 
 191' 
 303 !! 
 
 93 
 
 '30" 
 
 78 « 
 
 106 « 
 
 80" 
 
 Lw 
 69' 
 65' 
 
 80 > 
 
 48 
 
 47' 
 
 58' 
 
 22 « 
 33" 
 35" 
 
 38" 
 "51 « 
 
 M 
 
 137 
 400' 
 
 E 
 
 703 
 831 5 
 
 1500* 1800' 1363' 
 133 308 374 
 
 T "W 
 
 900 1263 
 
 1500* .... 
 
 1800= 3350'' 
 
 404 
 
 155' 
 30 » 
 
 "31 « 
 51 « 
 
 "75" 
 
 330' 
 
 27' 
 34" 
 37" 
 
 44" 
 "56" 
 
 453' 
 50" 
 
 103" 
 100" 
 
 104" 
 
 139 
 
 350* 
 33" 
 
 508* 
 
 30" 
 40" 
 53" 
 
 MS MS MSMS MS MS MS 
 
 Jan. 5- 1-33 4- 7 3-37 3-30 5-39 2- 5 
 
 Mar 5-49 ._. 
 
 June 9-36" 3-56' 3-50' 3- 0' 6- 5 10-1' 3-16' 
 
 19" 
 
 83" 185" 
 
 ioi" 501" 
 
 MS MS 
 
 3-33 11- 
 .-.. 15-35 
 3-34' 33-45' 
 
 Jan. 50" 18" 
 June 105" 10" 
 
 16" 6" 
 36" 33" 
 
 30" 11" 
 30" 39" 
 
 10" 
 
 373 
 
 35" 
 758 
 
 54" 
 108 3 
 
 Criticism of Records of Physical Endurance. 
 
 That the fatigue after the March and June tests was in general 
 touch less than after the January test was made evident by three 
 substantial proofs. The first was the feelings of the men themselves 
 as recorded in the foot-notes to Table XI. After the March and J une 
 tests, every man of the eight who showed improvement felt " not 
 tired," or " less tired than in January test," which is the same as 
 saying "not exhausted"; or else he had gone "to limit" as in Jan- 
 uary, which means that the muscle itself refused to continue work. 
 The last was usually true of the "leg-raising," "raising 5-lb. dumb- 
 bell (triceps)" and "raising 25-lb. dumb-bell (biceps)". The only tests 
 in which there was the possibility of being mistaken as to the degree 
 of fatigue were the "rising on toes" and "holding arms horizontal." 
 In the former fatigue comes so slowly, and in the latter the pain is 
 so intense that they prove to be tests of will power or " grit " quite as 
 
 ^ Cramped. 
 
 ' Not as tired as in January test. 
 
 8 Not to limit. 
 
 * Not nearly as tired as in January test. 
 
 ' Not tired. 
 
 " To limit of muscle's Ga,paoity. 
 
 ' About same fatigue as in January, 
 
Fisher — The Effect of Diet on Endurance. 21 
 
 much as of muscle power. In these cases the men hacl some difficulty 
 in remembering the original degree of fatigue. But the increases 
 were so great and the men were so positive as to their feelings that 
 there remains little room to doubt the substantial correctness of the 
 results. In a few other individual cases, as of Lw. and Lq., whose 
 records in test 1 were sometimes stopped by cramps, there is some 
 room for doubt as to the correctness of the recorded improvement. 
 
 The second proof that the fatigue of the men in the June tests was 
 less than that in the January tests was found in the fact that the 
 stiffness and soreness which followed in June were markedly less 
 than in January and of niuch shorter duration. This was true of all 
 the eight men who showed improvement, except R. 
 
 The third proof of less fatigue in June than in January for the 
 eight men is that in June the men finished the ordeal of the endurance 
 tests with more strength left than in January, although, as we saw 
 from Table X, they began the two testp with slightly less strength. 
 The fact that they had more strength left after the June test is made 
 evident by the first part of test 6, given below, which in each case 
 came after the endurance tests were nearly or quite finished. This 
 consisted in lifting a 50-lb. dumb-bell. The weight being so great, 
 this was practically a test of strength rather than of endurance. 
 Now all of the eight men who showed improvement in the endurance 
 tests of Table XI, showed improvement in this strength test also, as 
 the following table shows : 
 
 TABLE XII. 
 LIFTING (by biceps) 50-lb. DUMB-BELL. ^ 
 
 
 B 
 
 E 
 
 Lq 
 
 Lw 
 
 M 
 
 P 
 
 E 
 
 T 
 
 W 
 
 Jan. 
 
 0' 
 
 13 
 
 13 
 
 03 
 
 13 
 
 03 
 
 43 
 
 23 
 
 13 ^ 
 
 June 
 
 13 
 
 03 
 
 53 
 
 83 
 
 133 
 
 13 
 
 10 3 
 
 132 
 
 26 = 
 
 But, as we have seen in Table X, the strength tests taken before 
 the endurance tests showed a slight falling off in June as compared 
 with January for all but one (Lq.) of these eight men. In other words, 
 in June the men began their endurance tests weaker than in Jan- 
 uary, but finished them stronger. The larger residuum of strength 
 
 ^ This part of test 6, being one of strength rather than of endurance, was not 
 included in the endurance Table XI. Had it been included it would have 
 increased even more the percentage of improvement shown, for it shows an 
 average increase from 2.4 to 8.4, or 250^. 
 
 2 Not to limit. 
 
 3 To limit of muscle's capacity. 
 
22 Fisher — llie Effect of Diet on Endurance. 
 
 left after the June tests as compared with the January tests indicates 
 that the June tests, in spite of being far more severe, fatigued the 
 men less. 
 
 The 50-lb. dumb-bell test resolved the last doubts in my own mind 
 whether, for some of the men, the recorded results might not exag- 
 gerate the true improvement. The two men of whose records I 
 should have felt a little doubt were B. and R. Both of them came 
 to the June test after prolonged mental exertion, and their exhaustion 
 at the end was far more evident than that of any of the others. That it 
 was great is clear from their own statements given below, though 
 only R. reported himself as having been about as stiff and sore after 
 the June as after the January tests. But both B. and R., whereas 
 they had less strength (Table X) before the June endurance tests 
 than before the January tests, had more strength left (Table XII) 
 after the June tests than after the Januai-y tests. At the close of 
 the January tests they were so exhausted that B. could not raise 
 the 50-lb. dumb-bell at all and R. could raise it onlj- 4 times. Had 
 their exhaustion after the June tests been as great, it seems certain 
 that B. would still have been unable to raise it, and R. would have 
 been unable to raise it more than 4 times ; but as it was, B. raised 
 it once and R. 10 times. 
 
 The value of such a positive proof that the June tests were more 
 easily endured' than those of January was not perceived until the 
 figures were analyzed. Had it occurred to me in time, all the strength 
 tests taken before the endurance tests would have been repeated after 
 them. It is true that the strength tests at the beginning were not 
 of the same muscles as those (the biceps) used in the strength test by 
 dumb-bells at the end, but, as Table X shows, the strengths of different 
 muscles for the most part vary in unison with each other.' 
 
 It is significant that the only man whose strength, as shown by the 
 above table, was less at the close of the June experiment than at the 
 close of the January experiment was E., who was also the only man 
 whose endurance showed any reduction. The facts, therefore, in his 
 case are not discordant with those already stated ; for, as has been 
 stated, E.was the least assiduous in following the experiment. This was 
 
 'Out of the 108 comparisons of strength (i. e., comparisons for each of nine 
 men in each of four tests for January vs. March, March vs. June, and January 
 vs. June), only 20 are discordant with the general trend as shown by the totals. 
 Thus, for B. the general trend between March and June as shown by the total 
 was downward, and this downward trend is found in all but one of his four tests 
 the discordant case being the "back lift.'' 
 
Fisher — The Effect of Diet on Endurance. 23 
 
 often remavked, both to him and to me, by the other members of the 
 club, and it was suggested more than once that I should " nudge " 
 him. But,. as I had been desiring a "control," or a subject in which 
 all the conditions except mastication were the same as for the other 
 men, I decided to say nothing. The result was instructive, for E.'s 
 case stood out as exceptional in almost all respects. His reduction 
 in quantity of food (Table III), except for a spurt at the end, was 
 less than of most of the men ; his reduction in proteid (Table IV), 
 with the same exception, was the least of all ; his reduction in 
 quantity of flesh foods (Table V) was the least of all ; his nitrogen in 
 June (Table VI) was one of the highest ; his reduction in volume of 
 urine (Table VII) was one of the lowest two ; his improvement 
 in the fecal tests (Table VIII) was third to lowest ; his loss of 
 strength (Table X) was second greatest ; and as to endurance, he 
 was the only one who failed to show improvement. 
 
 There was only one other man, M., who was thought, though in a 
 smaller degree, to masticate less carefully than the experiment called 
 for ;^ and for him we find corresponding peculiarities, though in a 
 smaller degree. Thus, his reduction in total daily food (Table III) 
 was less than of most of the men ; his reduction in proteid (Table 
 IV) was less than the average reduction ; his reduction in flesh 
 foods (Table V) was the third smallest ; his June nitrogen was the 
 highest (Table VI) ; his reduction in quantity of urine (Table VIl) 
 was one of the lowest two ; his improvement in fecal test (Table VIII) 
 was second to lowest ; his loss of strength (Table X) was the third 
 greatest (or fourth, if measured in percentage) ; and his increase in 
 endurance, though great, was (except in test 1, which is subject to 
 some doubt) less than the average. 
 
 The shortcomings of these two men, E. and M., as to mastication 
 were not intentional, but due to carelessness and force of habit, as 
 well as, in the case of M., to the fact that he waited on table and felt 
 naturally more pressed for time. Their experience is valuable in 
 showing that, in a general way, the changes in diet and endurance 
 were proportionate to the thoroughness of mastication and the follow- 
 ing of natural appetite. 
 
 The men kept diaries in which are recorded their sufferings after 
 the various tests. These show a decided lessening in stiffness and sore- 
 ness in the later tests, though in the June tests the men had generally 
 done double the amount of work that they had done in January. It 
 would have been a physical impossibility to do as much in January 
 as was easily accomplished in June in tests 3, 4 and 6 ; and granted 
 
24 Fisher — The Effect of Diet on Endurance. 
 
 that it had been even possible in January to goad the men to do as 
 much in tests 1, 2 and 5 as they did without urging in June, they 
 must certainly have been ill. 
 
 The following are statements from the men themselves : 
 
 Personal Impressions at end of Experiment. 
 
 (B) I was very sore [after the June test, Saturday, June 16, 1906] ' 
 However, I think the soreness was not so severe or lasting as it was 
 after the January test. The muscles of my right arm were swollen 
 considerably and I was unable to straighten it for two or three 
 days. But the swelling was not so severe as it was after the 
 earlier test and the arm was much more usable. The muscles of the 
 thighs were the sorest ; they were nearly worn out. Dui-ing Sunday, 
 Monday and Tuesday after the test I had difficulty in walking ; and 
 going down stairs was quite a difficult and severe undertaking. 
 Wednesday morning the soreness had not left, though it had de- 
 creased considerably. I took a considerable tramp that day, and by 
 night I could scarcely feel the soreness at all. By Thursday I had 
 practically regained my normal endurance ; walked six or seven 
 miles that day. The calf-muscles too were quite sore, but much less 
 so than after the January test. There was another particular dis- 
 tinction. After the earlier test the calf-muscles were hard and 
 knotted for several days ; but this time, while they were sore, they 
 were almost normally soft. Saturday evening when I went to bed 
 they were quite hard, but Sunday morning they were normal and 
 practically remained so. . . . 
 
 I have no doubt that in my case there was great increase in 
 endurance, though I think that I lost in amount of energy that I 
 could exert at any given moment. This loss is due perhaps to two 
 things ; (1) I took, on the whole, less exercise than during the time 
 preceding the January test ; (2) I had been working quite hard for 
 three months steadily, while the January test followed a three weeks' 
 vacation during which I did little or nothing. As to increase of 
 endurance there can be no doubt. For example, in the deep knee- 
 bending, I began to get tired at '50 and had no idea of going above 
 lOO! When I reached this I set my goal at 125, then 150, 160, and 
 was able to reach 200 before I was exhausted. In January, after I 
 was tired I was not able to go on very long before I became com- 
 pletely exhausted. This shows increased endurance. I had the 
 same experience in the other hard physical tests. In case of the run, 
 I was sure I could not go more than three laps after my first lap ; 
 
Fisher — The Effect of Diet on Endurance. 25 
 
 but I made 11 or 1 2 and could have gone several more. Considering 
 everything, I have no doujjt that I was able to hang on much longer, 
 after I began to get tired, than in January. 
 
 I am at a loss to ascribe the increased endurance to anything else 
 than to the diet. My way of living otherwise continued about the 
 same after the January test as it was before. . . . Personally I 
 am convinced that the increased endurance must be due to diet and 
 manner of eating ; all other factors that I can think of are unfavor- 
 able rather than favorable to more endurance. I am convinced to 
 the extent that I shall certainly continue " Fletcherizing " and using 
 a low-proteid diet. 
 
 (E) All effects of [June] test disappeared entirely within four 
 days. [Effects of January test lasted six days.] 
 
 (Lq) The stiffness and soreness had entirely disappeared in four 
 days. It was not nearly so severe as the test in January. After I 
 was through in January I could hardly go down the stairs of the 
 Gymnasium, and three days after the test going up and down stairs 
 was accompanied with a great deal of pain. ... I was stupid 
 mentally for a whole week the first time, but in the last test I passed 
 that stage in a couple of days. . . . Had it not been for the late 
 hours and long stretches of work, I should have been able to make a 
 better comparison with conditions in January, though as it was 
 results show improvement. . . . 
 
 I cannot say as to the help mentally I have derived, for I have, 
 always gone to my limit and I would be unwilling to make any posi- 
 tive statement. As for the physical, I know there is an improve- 
 ment there, for my stomach, which was never so very strong, has 
 been greatly helped. 
 
 (Lw) There was no stiffness or soreness felt in the triceps or the 
 stomach muscles as the result of the last tests. The thigh muscles 
 were a little stiff on the second day only — about such stiffness as one 
 might expect from a long walk. The calves of my legs began to 
 stiffen on Friday [June 15, 1906, the day after the test] and continued 
 to do so on Saturday, after which the stiffening began to lessen, and was 
 scarcely felt on Monday. The biceps of my right arm gave me the 
 most trouble. These were sore on Friday A. M. and continued to 
 increase in soreness till Sunday evening, feeling worst, however, 
 Sunday A. M. When I arose Monday A. M. all the soreness and 
 stiffness had disappeared. A peculiarity about the latter which im- 
 pressed me was the fact that although my arm was very sore it did 
 not seem to be very stiff. After the tests in January I could not 
 
26 Fisher — The Effect of Diet on Endurance. 
 
 straighten my arm, but I could after the last tests in spite of the 
 extreme soreness. I had entirely recovered by Monday from the 
 tests. At no time after the tests did I feel any pain in proceeding 
 up and down stairs, and if I remember rightly I couldn't say the same 
 in January; neither did I feel particularly uncomfortable at any time. 
 After the half-mile run and the lifting tests which I took later, I felt 
 no soreness or stiffness afterward. . . . 
 
 I think the credit must be given to the diet experiment. I have 
 worked harder from January to June than ever before, and have taken 
 less exercise. As my mental work was so different from that previous, 
 I cannot form an estimate of any increase or decrease in efficiency, 
 but as I have said before, I always rested up more quickly. During 
 the spring I have not felt that " all gone feeling " which usually has 
 appeared in the past. The diet which we have had has relieved me 
 of the sour stomach after meals, and I have felt better and worked 
 harder on less exercise than ever before. . . . After a moderate 
 amount of exercise, I have felt no such stiffness as used to come. 
 
 (M) The stiffness and soreness were entirely worn off in two 
 days. I did not feel it nearly as much as I did last January. In fact, 
 I did not exert myself to the utmost this last time because I had sev- 
 eral examinations to take a day or two later. 
 
 My general impression is that the experiment was an all around 
 benefit to me. I fully believe that during the tests, they reflected 
 the true state of the case in showing my efficiency in June compared 
 with that in January. I believe that there was a decided improve- 
 ment in efficiency and could ascribe it partly to my exercise and the 
 other part to the new manner of eating. I believe, however, that my 
 exercise played a very small part because I think what I gained in 
 exercise I lost in sickness [mumps]. 
 
 My exercise this year was practically the same as years preceding. 
 After April 1st I had very little exercise, on account of the mumps. 
 This left me in a weak condition over a month. I had lots of work 
 to make up and studied harder from April to June than any other 
 period of my course. My exercise was neglected these three months 
 and I studied almost constantly eveiy day and until 12 at night. 
 
 My experience has shown me that I was at my best in mind and 
 body when I ate meat four times a week. I have tried both more and 
 less and found the above to be the medium. I also found that I could 
 do more when I had the largest meal at noon. The greatest benefit 
 of the experiment to me personally is that last year I broke down .in 
 the spring term and this spring I kept up my work and health in a 
 much better condition. 
 
Fisher — The Effect of Diet on Endurance. 2V 
 
 (P) I went into the second test with some trepidation, kno^ying 
 that I had lost considerable weight the preceding ten weeks. . . . 
 Physicallj'-, I cannot say that I felt stronger before the second test 
 than before the first ; nor did I feel weaker. As the test developed, 
 however, I soon saw that my endurance, both mental and physical, 
 had increased. 
 
 Generally speaking, the soreness was less extensive, less trying or 
 acute, and (I tl\ink) shorter-lived- than in January. ... It seems 
 to me, as I finished the test much fresher' than in January, a clear 
 gain in efficiency is proven. The test seemed certainly to make a 
 true report. . . 
 
 I can ascribe gain in endurance to nothing but the diet and thor- 
 ough mastication. Every other factor in the situation was against 
 this gain — exercise, of which I took certainly no more than usual 
 and in the latter weeks much less ; work, of which I had had a long, 
 bard pull as against the three weeks' rest preceding the January 
 test ; sleep, much decreased for most of May and June. You stated 
 last December that you wished every factor to be in favor of the 
 first test and against the second. This condition has been true in 
 high degree for my case. . . Whatever the efiicacy of the two 
 tests in proving the superiority of low proteid and thorough masti- 
 cation for the other members of the club, I feel convinced that they 
 prove that superiority with considerable force in my own case. 
 
 I have tried meat and chicken a number of times in the last two 
 weeks, partly from curiosity and partly from necessity. But in every 
 case anticipation has been pleasanter than realization, and my low- 
 proteid tendencies bid fair to remain for some time to come. I may 
 say that I had no opinion on the diet question when the experiment 
 started, but am now a hearty low-proteid exponent. . . 
 
 I went into the test with considerable foreboding as to my endur- 
 ance showing ; for I have worked now without a break for twenty- 
 two weeks at hard mental labor, the last two weeks being especially 
 confining and involving large losses of sleep and exercise. I may 
 say that I have been unusually well for .six or eight weeks, and 
 bowels have been running with greater ease and constancy than for 
 several years. . . The endurance-tests, showing a good increase 
 in every test, consequently came as a complete surprise ; and my 
 self-confidence, largely absent at the start, returned in increasing 
 measure as the test went on. . . 
 
 Thursday, June 21 [5 days after test]. Played golf this morning 
 and afternoon (9 holes each time) with perfect ease, no difficulty 
 
28 Fisher — The JE^ect of Diet on Endurance. 
 
 with walking or driving the ball. Soreness wholly gone at present 
 writing, no touch of it noticeable anywhere. 
 
 (R) The outcome of the last endurance test was about the same 
 as the one held January 14, 1906. I was fearfully sore for about one 
 week, reachmg the climax at the middle of the week. 
 
 (T) Throughout the test I passed from one event to another 
 with much shorter periods between than I did in January. With the 
 exception of lying on my back and raising my feet, I at no time ap- 
 proached as near exhaustion as I did in January. In January, in I'ising 
 on my toes and in the deep knee-bending, I continued till I fell to 
 the floor. I was not exhausted at the close [of the June test], but 
 marked papers for 2^ hours before going to bed. Sunday I scarcely 
 felt any the worse, though my muscles felt a little queer when I poked 
 my finger into them. Monday my leg muscles were a little stiff 
 after a period of rest, but not painful in the least. My right shoulder 
 was a trifle lame, due wholly I think to hitting it once in a while in 
 the last test with the 10 and 5-lb. weights. My right arm at the 
 elbow was decidedly lame and would not admit of being completely 
 straightened, though it was undoubtedly better than in January. 
 By Tuesday all the other stiffness had practically left me except the 
 right elbow, which was, however, better. By Thursday I was unable 
 to detect any soreness whatever in any part. 
 
 (W) May 31. After the test I felt fairly tired and ready to quit 
 — however, not nearly so exhausted as before in January. Could 
 walk down stairs Avith more confidence and could raise my supper to 
 my mouth much more easily than after the first test. . . . The 
 results certainly far surpassed any expectation I had, especially as 
 in the morning I did not feel quite as sprj^ and active as usual, due 
 to a little unusual over-exertion the previous day. 
 
 June 1. Sore in thighs and biceps, also felt my abdominal muscles. 
 
 June 2. Expected to be much worse on this the second day, as in 
 January, but not so. About same as yesterday. Later in the day 
 could run up-stairs two steps at a time as I could yesterday— a thing 
 undreamed of in January for over a week after the test. 
 
 June 3. Felt pretty well today, much improved over yesterday, 
 still felt my thighs in walking down hill or down stairs, but not 
 nearly as bad as yesterday. 
 
 June 4. Feel my thighs only very little, other muscles not felt at 
 all. Rode a bicycle 5^ miles ; did not feel it. 
 
 June 5. Seem to be all )vell, haven't noticed a soreness all day. 
 
Fisher — The Effect of Diet on Endurance. 29 
 
 The following table expresses the percentage of improvement in 
 the records of Table XI. 
 
 TABLE XIII. 
 IMPEOVEMENT IIST PHYSICAL ENDURANCE IN PEECENTAGES. 
 
 BELqLwMP ETW 
 
 Jan.-Mar. 33+ 36± 686- -5± 315± ... 18+ 66+ ... 
 
 Jan. -June 66+ 5± 800- 33 ± 1081 ± 31+ 79+ 100+ 165 + 
 
 .„x (Jan.-Mar. ... ... 173± -2± .._ 
 
 ^'^' ■) Jan. -June 144+ -43± 188+ 31± 17+ 10+ 31± 94+ 36 + 
 
 a, I 
 
 Jan. -Mar ... 50 ... 36 ... ... 33 
 
 Jan. -June 33_ -37 123- 59 3 37 106 -17 77 
 
 Jan.-Mar ._ 36 
 
 Jan. -June 69 -57 3 34 47 37 4 33 170 
 
 (3) j 
 
 (4) j 
 
 ,-, (Jan.-Mar. ... 66± ... ... ... 43± 
 
 ^^' iJan.-June 93- 89 ± -7± -17± 74 ± 77+ 56 ± 1± 115 ± 
 
 (6) \ Jan. -June 110 -44 62 450 50 163 170 300 100 + 
 
 . (Jan.-Mar. 33+ 26 ± 398 ± 14 ± 140 ± 26 18+ 66+ 37 ± 
 
 ^'^- iJan.-Juue 85 ± -13 ± 194 ± 95 ± 312 ± 56+ 73 ± 66 ± 109 ± 
 
 In the preceding table most of the figures are succeeded by a " + ", 
 which signifies that the true improvement was greater than the 
 figures indicate. Thus, the first entry in Table XIII, " 33 + ", means 
 that B.'s improvement between January and March in test (1) 
 (rising on toes) was more than 33^. Similarly, "686-" for Lq. in 
 same test means that improvement was less than 686^. Again, 
 "215± " for M.'s same test signifies that his improvement in this 
 test may have been greater or less than 215^. Finally, when any 
 figure is not followed by a sign, as for instance, B.'s (3) (leg raising), 
 the meaning is that the figure given is, humanly speaking, correct. 
 This accuracy applies only to those tests in which the muscles were 
 worked till they were physically Unable to repeat the movement. 
 The reasons for the various suffixes may be found by studying the 
 foot-notes of Table XI.' 
 
 1 For instance the " + " after 33 for B.'s (1) is ex,plained by the fact (as indi- 
 cated in the foot-note to Table XI) that after his March test he was not as fatigued 
 as after his January test, although he had improved upon his January record 
 by 33^. The only cases in which the explanation of the suflixes will not be 
 found from the foot-notes to Table XI are the following; E.'s (1), 26 ±, in 
 which case the " — "is inserted owing to the fact that E. had come to the March 
 test after the refreshment of a nap ; and M.'s (1),- 1081 ± , in which case the " — " 
 is inserted owing to the fact that this high figure is inconsistent with the other 
 results of the test, it being thought that M. may have been mistaken in his 
 
30 Fisher — The Effect of Diet on Endurance. 
 
 The table shows enormous differences in the figures even of the 
 same m;ui for tlie same period. Thus, the June improvement of 
 W. reads 165+, 26 + > 77, 170, 115±, 100 + . Such wide differences: 
 between the improvements in different tests seem puzzling at first,, 
 but they ai-e explained, partly if not wholly, by two reasons. The 
 first is the obvious one that many of the figures are not exact records, 
 but understatements, and naturally their margin within the truth 
 will vary widely. Thus, the records for deep knee-bending (2) for 
 W. show merely that the improvement is over 26^ ; the true figure 
 may well be lOOj^, which would be more consistent with the other 
 figures. But the deep knee-bending test had been found in January 
 very painful and inconvenient in its after-effects, and there was there- 
 fore less inclination in the June tests to approach closely to the limit 
 in this particular test. 
 
 The other reason is that in some tests a larger fraction of the total 
 strength of the muscle tested was called into play than in others. 
 Thus, "leg raising" requires a very large fraction of the strength of 
 the abdominal muscles, while " rising on toes " requires only a small 
 fraction of the strength of the calf muscles. This may explain why,, 
 in general, the improvement in the test of the calf muscles seemed 
 so much greater than in that of the abdominal muscles. This expla- 
 nation is, however, purely hypothetical. It would be interesting to 
 find out experimentally how much an improvement in the endurance 
 of a muscle shows itself when it is exerted in different degrees, say to 
 1f>ic, 50^ and 25^ of its strength-capacity." 
 
 Bearing in mind these two possible reasons for the variations in the 
 figures, and also the fact that there must have been more or less actual 
 differences in the improvement of different muscles, we need not be 
 surprised at the disparities which the table shows. 
 
 If we omit the cases in which the records are at all doubtful (with 
 suffix ±) or exaggerated (suffix — ), we have left the following table 
 for the eight men who showed improvement : 
 
 remembrance of his January test. The " — " has been inserted whenever there 
 was the slightest ground of any kind for thinking the figures might be overstate- 
 ments. With these figures weeded out, the remaining ones cei-tainly understate 
 the actual improvement. 
 
 'The original object of using the graded dumb-bells, 50-lb., 25-lb., 10-lb.,. 
 and 5-lb. , for testing the biceps, was to throw light on this problem ; but for 
 reasons previously stated, these tests were not fully carried out. 
 
Msher — The Effect of Diet on Enchirance. 31 
 
 TABLE XIV. 
 
 PERCENTAGE OP IMPROVEMENT (exact or understated) 
 OF EIGHT MEN. 
 
 B Lq Lw M P R T "W 
 
 «x 5 Jan.^Mar. 
 ^ ' \ Jan.-June 
 
 33 + 
 66 + 
 
 --- 
 
 --- 
 
 ... 
 
 '31 + 
 
 18 + 
 79 + 
 
 66 + 
 100 + 
 
 165 + 
 
 ,g^ f Jan.-Mar. 
 ^ ' 1 Jan. -June 
 
 144 + 
 
 188 + 
 
 ... 
 
 "17 + 
 
 15+ 
 
 ... 
 
 "94 + 
 
 '36 + 
 
 ,0-, ( Jan.-Mar. 
 ^ ■' i Jan.-June 
 
 "33 
 
 --- 
 
 50 
 59 
 
 "3 
 
 36 
 
 37 
 
 106 
 
 -17 
 
 83 
 
 '77 
 
 , J, j Jan.-Mar. 
 ^ ' { Jan.-June 
 
 '69 
 
 36 
 
 3 
 
 "34 
 
 "47 
 
 '37 
 
 "I 
 
 "33 
 
 170 
 
 ,,, j Jan.-Mar. 
 *' ' \ Jan.-June 
 
 --- 
 
 --- 
 
 ... 
 
 ... 
 
 "77 + 
 
 ... 
 
 --- 
 
 --- 
 
 (6) ■ -{ Jan.-June 
 
 110 
 
 63 
 
 450 
 
 50 
 
 163 
 
 170* 
 
 300 
 
 100 + 
 
 Av i Jan.-Mar. 
 ■ ( Jan.-June 
 
 33 + 
 
 84 + 
 
 36 
 
 84 + 
 
 50 
 181 
 
 "39 + 
 
 36 
 
 56 + 
 
 18 + 
 89 + 
 
 66 + 
 80 + 
 
 33 
 
 107 + 
 
 The figures of Table XIV show an undoubted increase in endur- 
 ance, both for the first half and more especially for the whole 
 period of the experiment. 
 
 But, for an accurate presentation, we may carry our criticism one 
 stage further. The figures given hitherto represent a conglomerate 
 sort of endurance, made up of endurance of different muscles subject 
 to different degrees of strain. As pointed out before, the calf mus- 
 cles were called upon for only a small fraction of their strength- 
 capacity, whereas the abdominal muscles were called upon for a very 
 large fraction. Moreover, the fraction must have varied somewhat 
 in different tests, according to the variation in strength and weight. 
 An ideal test would be one in which the same fraction of strength 
 was used. ' 
 
 Fortunately, such an exact test is afforded by the 25-lb. dumb- 
 bell. It followed immediately after the 5'0-lb. dumb-bell had been 
 raised until the biceps was unable to repeat the motion. At the 
 moment the 5 0-lb. test ended, the -SS-lb. test began. At this moment 
 the strength of the biceps was just at or barely below the fifty lbs. 
 required to raise the heavier dumb-bell. In other words, in raising the 
 25-lb. dumb-bell the muscle needed just fifty per cent, of its strength 
 at the time the test began. The uise of the 25-lb. dumb-bell grad- 
 ually reduced this strength from 50 to 25 lbs! The test was there- 
 
 1 It is on this principle tHat the new ergograph, before referred to, is con- 
 structed. 
 
35! Fisher — The Effect of Diet on Endurance. 
 
 fore perfectly uniform for all the men ; it showed how many con- 
 tractions were necessary in each case to bring down the strength of 
 the biceps from 50- to 25 ; it showed how much the muscle could 
 endure before being robbed, by fatigue, of half its strength. Thus at 
 the beginning the strength is 50 lbs.; after the first contraction it is, 
 say, 49 ; after the second, 48, etc. But the contractions continue until 
 the strength sinks below 25 lbs. The loss of strength maybe said to 
 measure fatigue. The slowness of this loss may be said to measure 
 endurance and is well indicated by the number of contractions 
 necessary to tire a muscle from a strength of 50 lbs. to a strength of 
 25 lbs. 
 
 Four exceptions, however, need to be noted. Three men, B., Lw. 
 and P., were unable in January to raise the 50-lb. dumb-bell at all 
 (see Table XII). Consequently their January test with the 25-lb. 
 dumb-bell did not begin at 50^ of the strength, but af a higher frac- 
 tion. This explains their high apparent improvement. Thus, Lw. is 
 credited with an improvement of 450^, because in January he could 
 raise the 25-lb. dumb-bell only 6 times, and in June, 33 times. 
 But the 33 contractions in June began at just 50^ of the strength of 
 the muscle, owing to its previous exhaustion to the 50-lb. level by 
 the 50-lb. dumb-bell, whereas the six contractions in January began 
 at a higher level; for at that time the biceps could not raise the 
 oO-lb. dumb-bell at all. Its strength was at that time less than 50 
 lbs., say 40 lbs., in which case the lifting of the 25-lb. dumb-bell 
 required not 50^ but &2^fo of its strength. To compare a 50^ test of 
 June with a 62J^ test in January gives a record of improvement 
 which is not one of pure endurance, but which includes the element 
 of increased strength. This is "endurance" in the crude sense in 
 which we may say a man has more endurance for carrying trunks 
 than a boy ; but for a comparison of pure endurance, the boy should 
 be given smaller trunks to handle than the man. 
 
 The fourth case is E., to whom the reverse reasoning applies. In 
 June when he reached test 6, he was unable to raise the 50-lb. dumb- 
 bell at all, though in January he had raised it once. Hence, while 
 the 25-lb. dumb-bell was a 50^ test in January, it was a more 
 severe one in June, and the -44^ which records his falling off does 
 not represent a pure loss in endurance, but partly also a loss of 
 strength. To reckon pwre endurance we need to bring -44 up toward 
 zero. 
 
 Making the four omissions just mentioned, we may use the remain- 
 ing records from the last line of Table XIII, as a barometer of pure 
 endurance. 
 
Fisher — The Effect of Diet on Endurance. 33 
 
 We therefore have three methods of estimating the increase of 
 endurance between January and June. These may be put together 
 in the following table : 
 
 TABLE XV. 
 PERCENTAGE OF INCEEASE OF ENDURANCE, JANUARY TO JUNE. 
 BY THREE METHODS. 
 B ELqLwMPRT W 
 
 "Vtesfs^ \ ^^± -1^± ^^4± 95 ± 313 ± 56+ 73± 66 ± 109 ± 
 
 Omitting; \ 
 
 doubtful \ 84+ ... 84+ 181 39+ 56+ 89+ 80+ 107 + 
 
 cases " + " ) 
 
 "Pure" ) 
 endurance [■ ... ... 63 ... 50 ... 170 800 100 + 
 
 of biceps ) 
 
 The first line of this table tells us the average of the recorded im- 
 provement in endurance shown for each man. But as each such aver- 
 age is made up from the figures of Table XIII, some of which, as 
 indicated in that table, are possibly too high, some doubt necessarily 
 attaches to it, though practically the only real cases of doubt are 
 Lq. and M. The average of these averages is 101^ for the entire 
 club, and is probably within the truth ; for most of the individual 
 figures which go to make up this result are understatements, not 
 overstatements. 
 
 The second line shows the average improvement in tests in which 
 there is no daubt that the figure is at least not too high, though it 
 may be too low. The average of these is 89^, and is therefore cer- 
 tainly too low an estimate of the average improvement for the eight 
 men who improved at all. 
 
 The third line shows the increase of pure endurance (that is, en- 
 durance considered apart from strength) for the five men for whom 
 the figures were available. The average of these is 116^. 
 
 We are quite safe in saying therefore that the average improvement 
 of the eight men who improved was 90^. As to the degree of retro- 
 gression of E., it is difficult to say, though it is believed that the fig- 
 ures exaggerate it. This is certainly true of the 25-lb. dumb-bell test, 
 for reasons given. My o wndmpression, and E.'s also, is that. he actually 
 gained in endurance from the dietetic experiment, but that, his gain 
 was not enough to offset the loss occasioned by (1) the hard term's 
 work, which, as in the case of the other mien, was a decided handicap, 
 and (2) the omission of hie customary .exercises,' which must have 
 
 Trans. Conn. Acad., Vol. XIII. 3 Mat, 1907. 
 
34 Fisher— The Effect of Diet on Endurance. 
 
 been a greater handicap in his case than in any other of the men ; 
 for he had been accustomed for six years to heavy gymnasium train- 
 ing, but during the year of the experiment this training was given 
 up, largely because of the difficulty in finding time for it. If this 
 interpretation is correct, we may liken the experiment to nine men try- 
 ing to swim against a current. The eight who exerted themselves 
 the most succeeded in forging ahead ; the one who tried the least 
 drifted backward, though the effect of the swimming (dieting) was to 
 propel him forward. Whether or not E. was actually propelled for- 
 ward by diet must remain a matter of conjecture or inference ; but 
 that the other eight men gained is an established fact. 
 
 Changes in Mental Endurance. 
 
 The mental test consisted in adding a specified number of figures. 
 The following tables show the time during which the addition was 
 performed and the number of errors committed : 
 
 TABLE XTI. 
 
 TIME OF PERFORMING A UNIFORM AMOUNT OF ADDITION. 
 
 BELqLwM P ETW Average 
 
 MSMS MSHSMS MS MSMSMSUS 
 
 Time ( Jan. 14 5 40 4 49 6 15. 4 54 6 46 3 17 6 6 41 4 6 5 29 
 
 of \ Mar. 28 5 16 4 37 4 35 4 15 5 47 3 43 6 32 7 18 4 34 5 3 
 
 adding ( ' June 16 4 50 5 9 4 40 4 23 5 50 2 58 7 3 6 5 4 7 5 
 
 This shows that during the first period seven had improved and 
 two had fallen off, and on an average there had been a decrease from 
 5m. 29s. to 5m. 3s., an average improvement of 26s. W. showed an 
 increase in time of adding, although he would naturally have been 
 expected to improve on account of having taken up clerical work 
 involving adding. 
 
 During the second period there was an average improvement of 
 only 3s. ; three retrogressed ISs. to 42s., three retrogressed 3s. to 8s., 
 and three improved 26s. to 73s. The fact that the men held their 
 own in the June adding test is probably indicative of actual improve- 
 ment, for they were fatigued mentally by examinations, etc., on the 
 day when they entered the June test. During the entire experiment 
 there was an average improvement of 29s. ; seven had improved and 
 two had retrogressed (Is. and 20s.) 
 
 'E., Lw., R. and W. taken on May 31. 
 
Fisher — The Effect of Diet on Endurance. 35 
 
 Tlie following table shows that the number of errors committe.d 
 was remarkably constant for most of the men and for the average : 
 
 TABLE XVII. 
 NUMBER OF ERRORS OF ADDITION. 
 
 
 
 B 
 
 E 
 
 Lq 
 
 Lw 
 
 M 
 
 P 
 
 R 
 
 T 
 
 W Average 
 
 Errors 
 
 Jan. 14 
 
 ■ Mar. 38 
 
 June 16 
 
 10 
 16 
 
 8 
 
 8 
 
 5 
 
 13 
 
 12 
 8 
 5 
 
 4 
 4 
 4 
 
 1 
 3 
 
 3 
 
 1 
 
 1 
 
 3 
 3 
 6 
 
 1 
 3 
 
 
 1 4.4 
 I 4.6 
 
 3 4.5 
 
 The adding test was too short to be of great value. In future 
 tests a larger number of figures will be employed, and a diilerent 
 method. After the specified amount of adding has been done, it 
 will be at once repeated on another equivalent set of examples. The 
 excess of time required for the second set over that required for the 
 first may be called the " fatigue time," and this fatigue time, taken 
 as a percentage of the total time of adding, may be used as a criterion 
 of endurance — the less it is, the greater the endurance, and vice 
 versa. This plan was developed too late to be put into operation at 
 the beginning of the experiment. It was, however, employed in the 
 March and June tests, and confirmed the conclusion reached above, 
 that there was little diffei-ence between the mental endurance in 
 March and June. Five of the men showed a less "fatigue time" in 
 June than in March, and four a greater. 
 
 The following statements of the men themselves will show that 
 their feelings as to working power were in harmony with the conclu- 
 sion that it had improved : 
 
 Subjective Impressions as to Mental Working Power. 
 
 B. (March) " Not decreased at any rate, seems to have increased." 
 (June) I did more work during the latter part of year than I 
 ever did before in an equal period of time. But, I had the 
 work to do and compelled myself to do- it. However, I was 
 mentally tired at the close of the year, particularly so at the 
 time of the test, for it came after the siege of exams for which 
 I did my own work besides a couple of days of hard tutoring. 
 This much, at any rate, is positive : There was no. decrease 
 of mental power resulting from the experiment. I was no 
 more tired at the close of last year than I was at the close of 
 the year before. After a week's rest I felt quite normal and 
 then did considerable mental work all summer. 
 
36 Fisher— The Effect of Diet on Endurance. 
 
 E. (March) " Working power improved. Can concentrate atten- 
 tion for a longer time." 
 (June) " I accomplished a greater amount of mental work than 
 in previous years during the corresponding period of the col- 
 lege year. , I do not think that my feeling of fitness for it 
 was any greater, however, a.nd I cannot say that my experi- 
 ence of fatigue after the work was any less. I learned to eat 
 slower than had been my custom during previous years. 
 Though not subject to indigestion, I experienced less stomach 
 disorders during the period of the experiment." 
 
 Lq. (March) " I have put in more long hours during this term 
 than any previous term, consequently have had a good deal 
 less sleep. I do not know that I can work any better, except 
 that I can work a longer period at one time without feeling so 
 tired from it." 
 (June) " Of course a great deal of the extra work was outside 
 work which was an extra tax. I, however, did a great deal 
 more work on papers that I prepared than I ever had before. 
 . . . Although I spent longer hours than before I did not 
 feel the effect of the work so much as before." 
 
 Lw. (March) " Have been working harder during the past four 
 months and have taken less exercise than at any other equal 
 period during past 2^ years. The character of the work has 
 been so different that I am unable to say whether there is any 
 increase in working power, but I find that I rest up very 
 quickly after becoming tired (mentally). 
 
 " When March tests were taken I did not feel as ' fit ' for test 
 on that particular day as at time of January tests. Had been 
 working hard and had been under nervous strain, which un- 
 doubtedly affected the tests." 
 (June) " I have worked harder from January to June than ever 
 before, and have taken less exercise. As my mental work was 
 so different from that previous, I cannot form an estimate of 
 any increase or decrease in efficiency, but as I have said 
 before, I always rested up more quickly." 
 
 M. (March) " I think on the whole a slight improvement." 
 (.June) "Do not notice any change." 
 
 P. (March) " I have never worked so steadily, or with so little 
 necessity to exercise the will to work, as in the central six 
 weeks of the test. The work I was- doing was chiefly research 
 in the Library, poring for three or four hours at a time over 
 old records — not labor of the most interesting kind." 
 
Msher — The Effect of Diet on Endurance. 31 
 
 (June) My work from March to June was of a more confining 
 sort than ever before in the spring. From March to the Easter 
 recess I was occupied with library research ; from Easter on I 
 was engaged on an essay and the marking of some 260 Social 
 Conditions theses. On these last I spent considerably more 
 time and effort than in the preceding year. Thus my work 
 was harder and more time-filling than usual. I had much less 
 out-door exercise than in previous springs, and missed that 
 relaxation from effort which all prior springtimes have bred. 
 Yet there waa no feeling of overwork, or even of work as a 
 burden, till the first of June. From then on I did feel tired, 
 and examination time found me with a mind very diflBoult to 
 keep in harness. Undoubtedly I had over-pushed myself, but 
 did not realize it till June. . . 
 
 R. (March) Felt an increase in efficiency. 
 
 (June) My power for mental work was "greater between the 
 March and June tests than between the January and March 
 tests and the latter was greater than before the experiment 
 began in January. I can state without hesitation that my 
 mental working power increased in consequence of " Fletcher- 
 izing." 
 
 T. (March) Felt that he had at least held his own, but " surprised 
 to find that every one of my tests (physical) had improved." 
 For the first test came after the rest and recreation of the 
 winter holidays when he " was in splendid condition. Since 
 then I have had to work extremely hard with little regular 
 exercise and rarely in bed before about midnight." Surprised 
 also that the mental test showed no improvement, probably 
 because " the confusion around me was considerably greater 
 than in the first test." Can do his ordinary mental work 
 faster than before, though not sure that he can work longer. 
 (June) " I consider I did more work last year daring the period 
 of the experiment than any other year. During the whole 
 nine months of the college year I was praptically working up 
 to my limit of endurance. I did not grow sleepy as early 
 evenings as previous years and my attention was not as easily 
 distracted from my work as previous years. 
 
 " The lack of improvement in the second mental test may 
 have been due largely to the fact that I was mentally fagged 
 out after the examinations and was feeling the need of my 
 holidays." 
 
38 Fisher— The Effect of Diet- on Eiidur\m^^>^ 
 
 W. (March) No definite impressiou either of gath or loss. 
 
 (June) " On the whole I felt quite as workish as ever I did in 
 
 the spring months and did not feel the hot days as much of 
 
 a' drag as usual." 
 As to illness, in the course of the experiment there were the usual 
 winter colds, though apparently these were less common than before. 
 One man had grip for a few days, another the mumps, and several h;id 
 constipation. In general, the men expressed themselves as better 
 than usual and in some cases they were very enthusiastic. None of 
 the ailments suffered by the men were ascribable to the test itself, 
 unless it be a case of what appeared to be slight rheumatic sensations 
 of T., who had always been a heavy meat-eater, and who during the 
 experiment introduced at first much acid fruit. That the acid in 
 conjunction with the high proteid might occasion such symptoms is 
 at least consistent with some of the numerous theories of rheuma- 
 tism. On avoiding very acid fruits he soon lost all these symptoms. 
 The following extract from the diary of B. is, I think, typical of 
 the facts in this respect to general health : " Have now, March 2-3, 
 slight sore throat. In regard to colds, I have been troubled less 
 this year than at any time for years ; but this fact may be due to 
 great change in climate, Nebraska to Connecticut. Have usually had 
 colds more or less all winter ; therefore my freedom has been indeed 
 remarkable. From September last till the beginning of the experi- 
 ment I experienced frequent attacks of indigestion, 'heart-burn.' 
 Have been almost free from that, though two or three times I had the 
 same experience after eating bananas." 
 
 Summary. 
 
 The phenomena observed during the experiment may be summar- 
 ized as a slight reduction of total food consumed, a large i-eduction 
 of the proteid element, especially for flesh foods, a lessened excretion 
 of nitrogen, a reduction in the odor, putrefaction, fermentation 
 and quantity of the feces, a slight loss of weight, a slight loss of 
 strength, an enormous increase of physical endurance, a slight 
 increase in mental quickness. These phenomena varied somewhat 
 with different individuals, the variations corresponding in general to 
 the varying degree in which the men adhered to the rules of the 
 experiment. 
 
 That we are correct in ascribing the results, especially in endur- 
 ance, to dietetic causes alone, cannot reasonably be doubted when it is 
 considered that no other factors of known significance were allowed 
 
^'^^^si^r—'TAe ijffect of Diet on Endurance. 
 
 39 
 
 to aid in this result. On the contrary, so far as the operation of 
 other factors was concerned, these must have worked against rather 
 than for the results achieved. Exercise was in no case indulged in 
 to a greater extent than had previously been the custom, and in most 
 cases it was less. The men were warned not to take up exercise, 
 except so far as they had been accustomed to before the experiment 
 began, and if they varied their exercise at all, to lessen rather than 
 increase it. They were very conscientious on this point, as on others, 
 — so much so that some of them at first gave up exercising until they 
 began to feel "logy." This over-zeal was corrected ; but in no case 
 have I reason to think that the exercise taken was more, or more sys- 
 tematic, than previously. M. was probably the most systematic in 
 taking exercise. His statement on this point, as previously given, 
 the reader may care to review. 
 
 The men did not practice on the endurance tests between times. 
 This was expressly forbidden, and the men were too trustworthy to 
 admit of a doubt on this point. The tests themselves, needless to 
 say, were too far apart to have given any chance for repetition to 
 give " knack," and were too severe to count as beneficial exercise. 
 
 Nor were the men more regular in their hours of retiring or other 
 habits. On the contrary, they were rather more reckless in burning 
 the midnight oil. It developed that, with thefr increased freedom 
 from fatigue, they indulged more freely than ever their propensity 
 to work in the lines of their respective ambitions. At first they felt 
 justified in doing this, as it accorded with their instructions not to 
 remove any handicaps to their chance of improving their endurance, 
 but to increase rather than decrease such handicaps. But this liberty 
 became license, and I was forced to remonstrate with the men for 
 their late hours and overstudy, which tended to rob them of their 
 surplus endurance almost as fast as it accrued. Long before the 
 experiment was finished the men had given every appearance of 
 improved working power, but I was not at all sure that they would 
 have any of it left to show in the final test, because of their tendency 
 to use it- up in work-. Had the extent of their working proclivities 
 been realized in advance, it is doubtful if the experiment would have 
 been undertaken at all. It should be stated that all except M. were 
 graduate students, and almost all of them, in addition to their uni- 
 versity work, were earning their own way. 
 
 The advance of warm weather must have tended, had not their 
 diet counteracted it, to tire the men, if, at least, we may trust com- 
 mon impressions as to " spring lassitude." 
 
40 Fisher — The Effect of Diet on Endurance. 
 
 Again, the conditions immediately preceding the March and June 
 tests, as compared with those preceding the January test, were such 
 as to give the advantage to the January test. The latter came soon 
 after the Christmas holidays, when the men, as they themselves 
 stated, felt refreshed and at their best, whereas the March test came 
 just before the Easter recess, after a hard term's work, and the June 
 tests came after a like period of hard work, — in some cases, as of B. 
 and R., immediately after exhausting examinations. 
 
 Finally, the tests themselves were serious drains on vitality. Each 
 required a period of from several days to two weeks for recuperation, 
 and each robbed the men temporarily of several pounds of weight. 
 The cookless diet experiment for six days also cost something to 
 those who took part in it. 
 
 In addition to the tests mentioned in this report was one on Jan- 
 uary 23 of leg-raising, deep knee-bending and arm-stretching, taken 
 after a night from which two hours of sleep were purposely cut off. 
 After consideration, it was decided not to repeat this test as being 
 too fatiguing. It therefore has been omitted from this report ; but 
 it added one more burden for the men. 
 
 When, therefore, we observe the known handicaps, — the over-study, 
 the strain of the. tests, the advance of warm weather, the fact that 
 the first test came after rest and the other tests after woi'k, and 
 when we are unable to find any other cause than diet— such as exer- 
 cise, regularity of bed-time or other habits — we are forced to 
 conclude that the only causes which produced the endurance \vere 
 dietetic. 
 
 Possibly some persons may be disposed to find a convenient escape 
 from this conclusion by ascribing the improvement to suggestion. 
 Under this theory, the men improved because they expected to. It 
 is quite true that there may be more force in autosuggestion than 
 most of us realize. But, fortunately, for the present case we scarcely 
 need to argue the point ; for as a matter of fact it was not true that 
 all of the men expected to improve. This was certainly not true 
 before the March test. In fact, the men were about equally divided 
 in their predictions as to the outcome, and used to have animated 
 discussions. Yet, both the confident and the skeptic faction im- 
 proved in endurance in the March test ; and so far as I am acquainted 
 with their prognostications and have noted- their improvement, there 
 was little if any correlation between those prognostications and their 
 improvement. 
 
 It is of course still possible that some unobserved element has crept 
 
Msher—The Effect of Diet on Endurance. 41 
 
 into the case, to which, and not to the diet, the improvement in 
 endurance was due; but in view of all the facts recited, this is 
 extremely improbable. What slight doubt remains should be resolved 
 by further studies. I earnestly hope that other and more careful 
 studies may be made by more competent investigators than I. 
 
 We conclude that the improvement in endurance was exclusively 
 due to dietetic causes. The only dietetic causes at work were (1) 
 thorough mastication, (2) implicit obedience to appetite, (3) (during 
 the second half of the experiment) when appetite did not clearly 
 determine the choice, the voluntary selection of the non-flesh and 
 low-proteid foods, and (4) an ample variety of good foods, well 
 cooked. 
 
 So far as cooking is concerned, this cause, as has been said, entered 
 unintentionally. But there is no evidence that it was a prime factor 
 in the experiment, while there is some evidence to the contrary. 
 Thus, E., who especially remarked the culinary virtues of the cook 
 and who missed her services more than any one else during the brief 
 ,13eriod of her absence, was the one member of the club who failed to 
 improve in endurance. 
 
 If we allow ourselves to speculate as to the changes in the charac- 
 ter of diet which were produced by thorough mastication, we may 
 draw an inference from the fact that the carnivorous animals are fast- 
 eaters, whereas the grain-eating animals are slow-eaters. It would 
 seem, therefore, when man changes his habits from fast eating to 
 slow eating he naturally changes his food from the food of a fast- 
 eating to that of a slow-eating animal. The question, therefore, 
 which is the natural food for man, may possibly be associated with 
 the question, which of the two methods of eating is natural to man. 
 Was the slow eating of the nine men an artificial and unnatural prac- 
 tice, as would be indicated from the fact that the majority of men 
 eat far faster ? Or, are the ordinary habits of man in respect to the 
 manner of fast eating themselves unnatural ? I have not attempted 
 to gather the facts necessary to solve this problem, but it certainly 
 constitutes an interesting one for the physiologist and anthropologist. 
 The few facts upon which I have chanced to fall would seem to 
 indicate that man is naturally a slow eater, and that the hurry-habit 
 to which most of us are prone is a consequence of the artificial high- 
 pressure to which modern civilization has subjected us. Certain it is 
 that the conditions which give rise to quick-lunch counters and to the 
 short stops of trains for refreshments, were produced, not in order to 
 meet any natural propensity to eat fast, but on the contrary, in the 
 
42 Fisher — The Effect of Diet on Endurance. 
 
 interest of the more rapid transaction of business, with which meal- 
 times are regai'ded as an interference. 
 
 We may therefore at least conclude that whatever the speed of 
 eating which is natural to the human animal, his actual speed under 
 civilized conditions is greater than natural. It is noteworthy also 
 that children are very deliberate in eating their cookies. It is only 
 after they are reproved for keeping their elders waiting that they 
 begin to imitate the latter and bolt their food. Dr. Higgins * and 
 Dr. Hasse " have pointed out also some physiological considerations, 
 based on the anatomy of the human throat compared with the 
 throats of the carnivores and of " poltophagic " animals, which would 
 indicate that man, to a large extent at least, is naturally a slow-eating 
 animal. Dr. Henry Campbell '' has also given some evidence, based 
 on a study of the primitive tribes, to show that chewing is more 
 thorough among uncivilized races, and that the hurry habit to which 
 we are accustomed is largely promoted by the use of prepared and 
 ' mushy " foods, — which, in fact, appear to have been devised ex- 
 pressly for the purpose of being quickly swallowed. 
 
 The evidence, however, on the natural food-habits of man is as yet 
 very meager, and it is only provisionally that we may consider the 
 thorough mastication advocated by Mr. Fletcher as " natural." With 
 this reservation we may say that the experiment here described may 
 be called an experiment in natural eating, or an effort to restore a 
 blunted or lost food-instinct, so that it may serve as a safe guide to 
 the proper quantities and kinds of foods. If it be asked in what 
 way this natural eating tended to improve endurance, whether it was 
 because of the finer sub-division of food through mastication ; the 
 increased " insalivation " ; the increased flow of " appetite juice " ; 
 the better adaptation of foods to the particular needs of the individual 
 and the moment ; the lessened quantity of food ; the lessened proteid ; 
 or the lessened amount of flesh foods, no satisfactory answer can be 
 given, although, as the previous discussion shows, there is more or 
 less evidence on some of these points. There are certainly some very 
 fascinating problems for the physiologist to solve in regard to fatigue 
 as related to diet. Are the " fatigue poisons " due, for instance, 
 chiefly to the combustion of proteid in excess of the physiological 
 
 ' See Humardeulture, N. T. Stokes, 1904. 
 
 ' See AroMv fur Anatomie (Waldeyer's) 1905, p. 321. 
 
 2 " Observations on Mastication," London Lancet, July 11, 18, 25 and Aug. 8, 
 1903. Reprinted in Horace Fletcher's The, A. B.-Z. of Our Own Nutrition, 
 Stokes, 1903. See pp. 126-135. 
 
Fisher — The Effect of Diet on Endurance. 43 
 
 needs, as the theory of Chittenden would explain them ? Or, are they 
 largely due to the ingestion of these poisons with ilesh foods, as the 
 vegetarians and Dr. Haig have maintained ? Or, do both explana- 
 tions have a share ? 
 
 The results of the experiment demonstrated so great an increase 
 of endurance as to seem at first incredible. It certainly was a sur- 
 prise, both to the men and to me. But statistics which I have been 
 collecting during the last two years have prepared me to find great 
 differences and changes in endurance. The special result of the 
 present experiment is to show that diet is an important factor in 
 producing such alterations. The fact that endurance, even among 
 persons free from disease, is one of the most variable of human fac- 
 ulties — far more variable than strength, for instance — is evident to 
 any one who has made even a superficial examination. Some persons 
 are tired by climbing a flight of stairs, whereas the Swiss guides, 
 throughout tlie summer season, day after day spend the entire time 
 in climbing the Matterhorn and other peaks ; some persons are 
 " winded " by running a block for a street car, whereas a Chinese 
 coolie will run for hours on end ; in mental work, some persons are 
 unable to apply themselves more than an hour at a time, whereas 
 others, like Humboldt, can work almost continuously through eight- 
 een hours of the day. Among statistics gathered independently of 
 the present experiment, I have found measurable differences between 
 persons far greater than the change of endurance of the eight students 
 which we have seen.^ Among some 50 tests of different persons 
 holding their arms horizontally, many were found whose arms actually 
 dropped against their will inside of ten minutes, whereas several were 
 able to hold them up over an hour, and one man held them 3 hours 
 and 20 minutes, or a round 200 minutes, and then dropped them 
 voluntarily. Similarly with deep knee-bending, some persons were 
 found physically unable to rise again from the stooping posture after 
 accomplishing less than 600 bondings, whereas several succeeded in 
 stooping 1,000 times, and in one case, 2,400. Again, in leg-raising, 
 the legs positively refused to rise to the vertical in some cases before 
 40 times were reached, whereas in two cases this motion was per- 
 formed 1,000 times or over. On the new ergograph previously 
 referred to, among the 16 preliminary tests there was a range in 
 endurance between different persons from 18 to 145 and in the same 
 person at different times from 29 to 110. 
 
 ' For an account of some of these statistics see " The influence of flesh-eating on 
 endurance." Yale MediealJourndl, March, 1907. 
 
ii Fisher — The Effect of Diet on Endurance. 
 
 It is, to say the least, remarkable that hitherto so little eflfort has 
 been directed toward discovering the factors which explain such 
 differences in endurance. That exercise is one of the most and per- 
 haps the most important factor has alone been recognized. A 
 correspondent assures me that by means of moderate re(7Mtor exercise 
 he succeeded in increasing his endurance between 100 and 200^ in 
 three weeks as measured by leg-raising and "dipping." The 
 influence of diet has always been regarded as small or negligible, and 
 the opinion has been almost universal, until recently, that a diet rich 
 ^n proteid promotes endurance. Even among those whose researches 
 have led them to the opposite conclusion, there is very little concep- 
 tion of the extent to which diet is correlated with endurance. Such 
 a person, a medical friend of the writer, stated, when the present 
 experiment was planned, that he did not think the dietetic factor strong 
 enough compared with others to produce any marked effect. We 
 have all heard, of course, of the enthusiastic reports of vegetarians 
 as to their increased endurance, but these we have discounted as 
 exaggerations. The result of the present experiment, however, 
 would seem to indicate that one's improvement in endurance is usually 
 not less, but greater, than he himself is aware of. I'robably it is 
 also true that we may lose a large fraction of' our working power 
 before we are distinctly conscious of the fact. 
 
 While the results of the present experiment lean toward " vegeta- 
 rianism," they are only incidentally related to that propaganda. 
 Meat was by no means excluded ; on the contrary, the subjects were 
 urged to eat it if their appetite distinctly preferred it to other foods. 
 
 The sudden and complete exclusion of meat is not always desir- 
 able, unless more skill and knowledge in food matters are employed 
 than most persons possess. On the contrary, disaster has repeatedly, 
 overtaken many who have made this attempt. Pawlow has shown 
 that meat is one of the most and perhaps the most "peptogenic" 
 of foods. Whether the stimulus it gives to the stomach is natural, or 
 in the nature of an improper goad or whip, certain it is that stomachs 
 which are accustomed to this daily whip have failed, for a time at 
 least, to act when it was withdrawn. 
 
 Nor is it necessary that meat should be permanently abjured, even 
 when it ceases to become a daily necessity. The safer course, at least) 
 is to indulge the craving whenever one is " meat hungry," even if, as 
 in many casc'^, this be not oftener than once in several months. The 
 rule of selection employed in the experiment was merely to give the 
 benefit of the doubt to the non-flesh food ; but even a slight preference 
 for flesh foods was to be followed. 
 
Fisher — The Effect of Diet on Endurance. 45 
 
 Under flesh foods are included all meat- and " stock-" soups. It 
 has been shown that although these extracts of meat contain a large 
 amount of nitrogen, it is not in the form of proteid which can be 
 utilized, but only of waste nitrogen which must be excreted. Ap- 
 parently the sole virtue of such soups is that they supply the " pepto- 
 genic " stimulus above referred to. 
 
 The experiment will be seen to harmonize with and supplement the 
 experiment of Professor Chittenden, on which it was founded ; but 
 the objects of the two experiments were quite different. Professor 
 Chittenden's was aimed to ascertain the physiological requirements 
 as to proteid, and did not touch upon the question of endurance. 
 Moreover, Professor Chittenden, in order the better to measure the 
 proteid and nitrogen, artificially reduced the quantities ingested, 
 whereas in the present experiment, test was made of Mr. Fletcher's 
 claim, that thorough mastication leads naturally to the adoption of 
 the physiological amount of proteid. This we found to be true, espe- 
 cially after the introduction, at the middle of the test, of the sugges- 
 tion that when appetite was in doubt, the lower proteid foods should 
 be selected. But the tendency was quite marked during the first 
 period also, and might have been expected to lead to the same results 
 without the introduction of even the suggestion of voluntary choice, 
 had the experiment been long enough. This was the experience of 
 others, notably Mr. Fletcher himself, whose case, in fact, first called 
 Professor Chittenden's attention to the possible virtues of low 
 proteid. , 
 
 The practical value of the experiment consists in the fact that any 
 layman can apply it, with or without a knowledge of food values, 
 though with more advantage if he possesses than if he lacks such 
 knowledge. 
 
 If the dietetic rules of the present experiment are followed, no self- 
 denial as to foods is required. It is, however, absolutely necessary 
 that there should be self-control enough to break up the habit of 
 hurried eating to which modem civilization has brought us, habitu- 
 ating us, as it does, to eat against time. 
 
 Experience indicates that appetite does not lead to a diet fixed in 
 amount or constituents, but moves in undulating waves or cycles. 
 The men who took part in the experiment were encouraged, after 
 any of the symptoms which seemed to be associated with high pro- 
 teid (such as heaviness, sleepiness, stiffness or soreness after exer- 
 cise, or catching cold), to cut down on their proteid and substitute 
 fat to restrain the gastric juice. This advice was intended to make 
 
46 Fi&her — The Effect of Diet on Endurance. 
 
 application of the theories of Folin' that we usually carry a reservoir 
 of proteid, enough to supply our needs for body-building for a fort- 
 night. If this reservoir is exhausted, proteid starvation occurs and 
 the body feeds on itself ; if it is filled too far it overflows and causes 
 the evils of excessive proteid. If this theory is correct, the art of 
 eating may consist largely in maintaining a golden mean such that 
 the proteid reservoir is neither empty nor overflowing, or at any rate, 
 not overflowing much. Many persons fear to reduce their proteid to 
 the Chittenden minimum for fear of proteid starvation ; but the 
 experience of those who have tried it would seem to show that this 
 fear is groundless, provided no violence is done to natural appetite. 
 This may be trusted, so it would appear, to raise a warning in the 
 form of " nitrogen hunger " before the danger point is reached. 
 
 '"A Theory of Protein Metabolism." American Journal of Physiology, March, 
 1905.