May, 1929] FOODS IN A COLLEGE COMMUNITY 23 



rolls alone and the salad alone being estimations. In every instance 

 in Table 6 the weight of the rolls or bread served with the salad was 

 determined separately from the weight of the salad. The column for 

 calories for ten cents is the only one in which the food value of 

 butter (See page 38) is taken into consideration, and here the results 

 are recorded to the nearest 5 calories, because it is believed that the 

 estimations for the butter do not justify reporting the results more 

 closely. 



The energy value of the rolls served with the salads purchased 

 at Restaurant B is found from the data for fresh weight, total calories 

 and total protein per portion in Table 1 (page 14) to be 3.1 calories 

 per gram of fresh weight and the protein content 8.5 per cent (aver- 

 age values for Samples 29 and 75 in Table 1). The rolls served with 

 the salads purchased at Restaurants C and D were considered to 

 have an energy value of 2.9 calories per gram of fresh weight and 

 a protein content of 8.8 per cent (average values for all the white 

 rolls listed in Table 1). The bread (Sample 84) served with Salad 

 No. 322 contained 2.9 calories per gram of fresh weight and 10 per 

 cent protein. It is believed that these values may be used in general 

 in estimating the calories from the fresh weight of rolls and bread. 



Comparison of the fresh and the air-dry weights of the salads 

 alone shows that there is a tremendous loss in drying, in some in- 

 stances nearly 82 per cent. The total energy content and the total 

 protein content of the salads (without rolls) vary widely, as would 

 be expected from the variations in sizes of serving and the great 

 difference in composition. In five out of ten instances the rolls (not 

 including the butter) furnished as much or somewhat more energy 

 (See Table 6) than the salad itself. The data for Salad 322, served 

 with sliced bread rather than rolls, suggest that one can obtain nearly 

 150 calories more for the same price when rolls are included with the 

 salad than when slices of bread are included. 



Samples 380A and 38pB (Table 6) give a good illustration of the 

 variations which may be expected in the servings of the same salad. 

 These two salads wei'e identical in composition, and an order of rolls 

 was served with each. The fresh weights of the rolls were only 

 slightly different and the fresh weights of the salads were practically 

 the same; yet Sample 380B contains 12 per cent more energy than 

 Sample 380A on the basis of salad plus rolls and 15 per cent more 

 energy in the salad alone. The same comparison holds true with 

 regard to the protein content of the two samples. 



The energy per gram of air-dry matter in the salads alone 

 (Table 5) varies from 3.7 to 4.7 calories in six instances and from 

 6.0 to 7.5 in five other instances. High values were expected in 



