506 REPORT OF OFFICE OF EXPERIMENT STATIONS. 



colloctod were sent to Middlctown to be culculiitcd, t{il)iilated, and 

 ciassilied there where there are facilities for such work and a force 

 skilli'd in the performance of it. By this cooperation the results 

 accomplished at the hospital during- the j)ast year are large and furnish 

 a number of interesting- and suggestive facts. At the same time the 

 carrying on of the studies has not been in any way a })urden or a drjig 

 upon the institution; it did not interfere in the least with the regular 

 routine of the work, and the expense to the hospital was compara- 

 tively small. 



The more important part of the investigations during the past j'^ear 

 consisted of dietarv studies in diti'erent wards and dining rooms. As 

 a part of each study determinations were made of the amoimts of dif- 

 ferent food Ujaterials brought into the kitchen and used in preparing 

 the different meals, the amounts of the cooked foods sent to the dining 

 room, the amounts served at the ta})le, and those left uneaten, both 

 on the plates and at the serving tables. Record was also kept of the 

 number of persons at each meal during each study. This furnished 

 data for the calculation of the quantities of nutrients and energy per 

 person and per man in the food served, eaten, and rejected. Obser- 

 vations were also made of the methods of pre[)aring, cooking, and 

 serving the food, and of utilizing that which had been sent into the 

 dinino' room but had not been served. 



FOOD CONSUMPTION. 



Twenty-seven such studies, each one week in duration, were com- 

 pleted; 23 with patients and 4 with employees. The total number of 

 persons included approximateh^ 1,000 male patients and 125 employees, 

 botii male and female. The data obtained were submitted to the Office 

 of Expei-iment Stations for classification, tabulation, and calculation of 

 results. In public institutions, as elsewhere, men eat more food and 

 need more than women. The computations can be made " per man," 

 " per woman," or "per person." All the studies of the patients at 

 St. Elizabeths were made with men and the results are expressed 

 "per man" per da3\ The studies with employees were with both 

 men and women, l)ut are computed on the " per man " basis, it being 

 assumed that on the average a woman eats eight-tenths as much as a 

 man. On this assumption the ratios per man, per person, per woman 

 would be 10 : 9 : 8. The results of the experiments showed that on an 

 average the food actually eaten b}^ the patients furnished 93 grams of 

 protein, 103 grams of fat, 301 grams of carl)ohydrates, and 2,705 calo- 

 ries of availa))le energ}^ per man per day. In a few of the wards the 

 food consumption was somewhat lower than this, and in some it was 

 higher; but in the majority of cases the variation of the individual 

 studies from the average was not unusually wide. The results of the 

 four studies with employees showed that they ate on an average 125 



