100 C. M. JACKSON AND C. A. STEWART 



The rats were well cared for throughout the experiments, 

 being kept during underfeeding in a warm room on account 

 of unusual susceptibihty to cold at this time. They were 

 weighed daily at the beginning of the underfeeding and the 

 refeeding periods, the time of weighing being later extended to 

 intervals of several days. The males and females of the same 

 litter were usually not separated, and the number of litters 

 born to each female is recorded in table 1. While no care was 

 taken to keep males with the females constantly in all cases 

 (and hence the number of litters is not the maximum possible), 

 it may be stated that the test rats had at least as good an oppor- 

 tunity for breeding as did the controls. 



At the end of the experiments, the rats either died or were 

 killed, as indicated in table 1. In forty-five cases (28 test rats 

 and 17 controls) autopsies were performed, and the weights 

 of the various parts, systems and organs recorded. The tech- 

 nique used was the same as described in the earlier papers by 

 Jackson and Stewart. In this series, Stewart was responsible 

 for the care of the animals and the performance of the autopsies. 



The data for the organ weights (also body length and tail 

 length) are given in table 2 under four groups: those refed after 

 underfeeding from, 1) birth to 3 weeks; 2) birth to 10 weeks; 

 3) 3 weeks to 20 weeks of age, and, 4) 3 weeks to nearly a year 

 (average 342 days) of age. The controls in each group are not 

 all from the same litters as the test rats, some being added 

 (selected from the other groups) to make the average body weight 

 of the controls about equal to that of the corresponding test 

 rats in each group. In the fourth group, however, controls 

 were not available so low in body weight as the test rats, and a 

 different mode of comparison was used. For the body parts 

 and systems of this group data from Jackson and Lowrey ('12) 

 were used for the normal in rats of corresponding body weight. 

 For the individual organs, the control rats (four males, eight 

 females) from the corresponding litters were used, although 

 they were much heavier than the test rats. Following a method 

 suggested by Donaldson ('15), both controls and test rats were 

 compared with the Wistar norm for rats of the same body weight 



