11 



. After each meal the "left overs'" of any sort were weighed and the 

 weights recorded. The quantities were usually too small to appear 

 again upon the school family table. When this was the case the food 

 was later used by the emplo\"ees. This made for the experimenters 

 a complication peculiar to establishments where two distinct tables 

 are supplied. It was necessary to calculate from the recipes of all 

 " made dishes ■" the exact amount of each ingredient in the material 

 not consumed, and deduct it from the quantities originalh' set aside 

 for the dietary. 



Wa,'^te. — During some of the experiments, the amount of waste and 

 refuse was weighed and recorded. In these cases the amounts of 

 kitchen waste, consisting of such articles as potato parings, coffee 

 grounds and the like, were recorded after each meal. At the same 

 time account was kept of the kinds and amounts of uneaten fragments 

 left upon the plates at table. 



When the study w^as over an inventory was taken of raw materials 

 which had not been used. The amounts on hand, like the ingredients 

 of the "left overs," were deducted from the weights recorded at the 

 beginning or purchased during the study. From this result the 

 amount of waste might be subtracted, in which case the figures 

 obtained would represent food actually eaten. 



Coinputdtlon of rtmdts. — On the basis of the real, as distinguished 

 from the " estimated ^ weights, new tables were made out, showing the 

 cost and nutritive value of what had been consumed. These tables 

 constitute what we called the "actual" or "final" dietary, and in a 

 condensed form thej^ are given beyond. 



None of the food materials from these studies was analyzed. The 

 composition of all materials used was assumed to be the same as that 

 of the average values for similar materials recorded in a previous 

 publication of the Office of Experiment Stations," except in a few 

 cases in which analyses were taken from an unpublished compilation. 

 The values used in computing the results of the studies are given in 

 Table 21) of the Appendix. The reference numbers in the first column 

 of this table correspond with those in parentheses following the weight 

 and cost of the food materials in the table for each study, and thus 

 indicate the composition assumed for each material. The fuel value of 

 the materials was computed by use of the folio wmg factors : For pro- 

 tein and carbohydrates 4 calories per gram, and for fat 8.9 calories. 



DIETARY UNDER USUAL, CONDITIONS (No. la). 



The study began November 7, 1901, and lasted fourteen days. The 

 members of the family wbo were eating the regular food, and could 

 therefore be considered in all experiments, were three teachers and 



«U. 8. Dept. Agr., Office of Experiment Stations Bui. 28, revised. 



