ULTIMATE EFFECTS OF INANITION ON SIZE 101 



(rather than body length), and the percentage difference estab- 

 lished for each organ. The algebraic difference between these 

 percentage results for the control and the test rat in each case 

 is taken as indicating the percentage of change in the corre- 

 sponding organ. Therefore, the average percentage changes 

 for the organs in the fourth column in table 1 are calculated 

 by a method different from that in the first three columns of 

 the table, where the absolute data for test rats and controls 

 are directly compared. In making the comparisons of organs, 

 the very slight differences in body w^eight (not exceeding 2 per 

 cent) between test rats and controls in each group were ignored. 



For economy of space, only the average data are given in 

 table 2. The sexes were at first grouped separately, but were 

 later combined in all cases excepting those with marked sexual 

 difference in weight (hypophysis, suprarenals, and gonads). 

 The individual data will be filed later in The Wistar Institute 

 of Anatomy, Philadelphia, where they may be obtained if desired. 



On account of the relatively small number of observations, 

 it is evident that the results for the various organs may be 

 influenced by individual variation and experimental error, hence 

 they are riot to be considered final. Where the changes are 

 marked and constant, however, they indicate the result with a 

 considerable degree of probability; and vice versa, when the 

 tests and controls differ but slightly, we may assume it as prob- 

 able that the earlier inanition has had no marked effect upon 

 the relative or proportional size of the organ at the end of the 

 refeeding period. 



EFFECTS ON BODY WEIGHT 



The effects of the underfeeding and subsequent refeeding 

 upon the body weight in the various groups are shown by the 

 curves in charts 1 to 5. Chart 1 shows the composite curves 

 for the eight litters underfed from birth to three weeks of age, 

 including 11 test males, 12 test females, 6 control males and 4 

 control females. The normal difference in weight according 

 to sex is evident. The sudden rise in the curves for the con- 

 trols at about 300 days of age is due to the death of individuals 



