No. 560] INFLUENCE OF FASTING ON GROWTH 



483 



cell-body. We shall attempt now to demonstrate this 

 point further by comparing the results obtained for long 

 periods with differently nourished animals. The data 

 which are given in the subjoined tables are so arranged 

 that the number of times when either the starved or the 

 periodically fasting salamanders received food is just 

 one half of the number of feedings of the control speci- 

 mens for some definite length of time. We take the 

 number of feedings which could be obtained from the in- 

 dividual records as indicating the approximate amount 

 of consumed food, since the actual quantities — except in 

 a few instances — have not been measured directly. It is 

 clear, of course, that animals fed ad libitum do not always 

 take the same quantity of food, nor is it likely that dif- 

 ferent animals consume each similar amounts, but in the 

 run of weeks it may be expected that the positive and 

 negative variations will compensate for each other. We 

 may, therefore, accept the number of times at which the 

 animals received food as a measure for the total quantity 

 of food consumed during a certain period. Furthermore, 

 to make the weights of the different animals comparable 

 with one another they have been computed on the assump- 

 tion that the initial weight of all animals was one gram. 



In Table C we have the data of four groups of starved 

 salamanders and of their corresponding controls. The 

 first two groups, each comprising four individuals, are 

 compared at the end of 112 days, during which period the 

 control specimens were fed 96 times, while the others 

 (after 7-8 weeks of complete inanition) were fed 48 

 times. The ratio between the number of feedings being 

 l-yO.5, it follows that by the end of 112 days the control 

 animals have probably consumed twice as much food as 

 the starved animals. The final body weight at the close 

 of this period was 3.823 g. and 4.265 g. (1st group), and 

 3.092 g. and 3.694 g. (2d group), respectively. Taking the 

 mean of these two groups, we get 1 -f- 1.165 as the ratio 

 between the final weights, and 1-4-2.45 as the ratio be- 

 tween the daily increase in the control and starved 



