26 



there was practical I3' an agreement between the quantities of protein 

 as estimated and as used. In the diet of uiedium cost the quantities 

 of fat were as close as could be expected, but the amount of carbo- 

 hydrates in the food used was appreciably smaller than in the estimate. 

 In the diet of low cost the quantities of both fat and carbohydrates in 

 the food used were noticeably smaller than the estimated amounts. 

 These differences are not surprising-, nor altogether unexpected, and 

 there are several reasons for them. For instance, it was impossil)le to 

 anticipate, in every case, the mental effect produced upon the family. 

 The appetites of some were undoubtedly affected ])y the knowledge 

 that their food was being studied and was of a delinite cost. Neither 

 could those making the study know, in detail, the personal tastes and 

 exactions of all the members of the family, and it appeared that they 

 differed widely in their acceptance of certain flavors and methods of 

 serving. It was also impossil)le to control the amoiuit of muscular 

 exercise taken, and, hence, this was irregular both in character and 

 amount. The possible effect of weather upon appetite was another 

 uncertain factor which nuist be recognized. These are some of the 

 elements which account for unexpected variations in quantities eaten. 

 The results of the investigation as a whole show plainly the impossi- 

 bility of providing a universally satisfactory bill of fare, as long as 

 the attitude of people toward their food is so largely a question of 

 personal whim. They do show, however, that it is possible to pro- 

 vide a diet of a definite cost which may vary within rather wide limits 

 and which shall at the same time supply the protein and energy called 

 for by the commonly accepted dietsiry standards. 



