GROWTH OF AMPHIBIA AFTER THYROIDECTOMY 27 



dicated in the drawings, and this adds to the difficulty of esti- 

 mating the volume of the gland. During metamorphosis the 

 actual size of the hypophysis remains practically unchanged, 

 but there may be a sHght decrease. Any difference in size to 

 be noted is less than the normal variability and less than the 

 experimental error in any attempt made to determine the size. 

 Figure 40 shows a representative specimen of average size of an 

 hypophysis after metamorphosis. Some of the glands of these 

 young frogs were smaller than those of the control larvae, but 

 more of them were larger. Rogers ('18) states that the hy- 

 pophysis shrinks during metamorphosis, but his data (p. 594) do 

 not support his conclusion. He apparently examined only three 

 specimens of the young frog's hypophysis, of which one (no. 13) 

 is larger than that of three out of six control tadpoles, another 

 is larger than one of his controls, and the difference between 

 the average of his experimental and control specimens is much 

 less than the variability shown within either group, and is much 

 less than the experimental error of his method of comparison. 

 Rogers' hypophysis no. 14 is only one-third as large as his no. 13. 

 His specimens are so few in number that the average size is 

 determined by the accidental selection of material. Selecting 

 at random he might have examined three specimens of the size 

 of no. 13 and drawn the conclusion that there is no shrinkage at 

 all during metamorphosis. In the large number of specimens 

 which we examined there was the same sort of variability shown. 

 However, a slight growth or a slight shrinkage of the hypophysis 

 during metamorphosis would be of no importance, because, as 

 shown in table 2, there is a shrinkage of one-half to two-thirds 

 in the size of the entire animal during this process, so that the 

 hypophysis increases considerably in relative size at this time. 

 Hence it is obvious that a young frog has a much larger hypophy- 

 sis in proportion to the size of the entire body than does the 

 larva from which it is transformed. 



It so happened that in the first larvae we autopsied the hy- 

 pophysis of the female in every case was larger than that of the 

 male, and we so reported the matter (Hoskins and Hoskins, '18). 

 In a much larger series of autopsies we have since observed that 



