DISCUSSION OF RESULTS. 



87 



and finally with the large heat production incidental to walking at the 

 rate of 144.1 meters per minute, the difference between the calculations 

 with the two base-lines becomes approximately only 1.5 per cent. This 

 small difference likewise holds in the running experiments. The values 

 computed from the Benedict and Cathcart base-line show the same 

 general characteristics, although the difference between the two results 

 is smaller than with the basal value of April 18, 1914. 



It thus appears that as the total amount of work increases it becomes 

 less important which of the three base-lines is used. For slow and 

 moderate speeds a possible difference of 3 to 6 per cent in the average 

 value of the heat per unit of work may be expected, while with the 

 highest speed this difference decreases to from 1 to 1.5 per cent. The 



TABLE 15.- 



-Comparison of heat-output during walking and running as referred to basal values 

 for the lying and standing positions. 



[Experiments with subject II without food.] 



table 14, p. 83, for explanation of basal values used. 



importance of the selection of a basal value is therefore greatest when 

 the amount of walking to be performed is smallest, i. e., when the sub- 

 ject is walking at moderate speed. This is somewhat unfortunate, for 

 a large proportion of walking for exercise, for the carrying of burdens, 

 and in marching is usually done at a moderate speed. Since, however, 

 the basal values obtained with the subject standing in the relaxed 

 position were daily and accurately determined, and since there is such 

 a high degree of constancy in the standing relaxed values, we consider 

 that they may be properly used as the basal values for these compu- 

 tations. The figures obtained per unit of work on this basis may there- 

 fore be considered to represent the increased metabolism necessary to 

 move 1 kilogram 1 meter in a horizontal direction. 



EXPERIMENTS WITH FOOD. 



A large number of experiments following the ingestion of food were 

 made with subject II. The results of all these experiments are pre- 

 sented in table 16, the data being grouped arbitrarily according to the 

 preceding meal and especially according to the character of the diet. 



