THE STRUCTURE OF FREE-MARTIN GONADS 109 



cerned, have already been considered in the first part of the 

 general discussion as due to an early anastomosis of the blood- 

 vessels between the different-sexed twins, and therefore an 

 early introduction of the male sex hormones or the sex hormones 

 may have been unusually potent or the duration of their action 

 may have been prolonged. These factors are responsible at least 

 for the early changes in the transformation of the gonad. How 

 much of the subsequent events is due to the initial stimulus of 

 the sex hormones and how much is due to other factors is diffi- 

 cult to determine. However, a few suggestions as to the factors 

 responsible for the later changes may be considered. In the 

 first place, the sex hormones may continue to act until birth, 

 providing the vascular connections persist (the foetal membranes 

 are still united at birth, according to Tandler and Keller, '11). 

 This necessarily involves the further assumptions that the inter- 

 stitial cells of the testis of the male twin are active throughout 

 foetal life or at least periodically active, and that these cells 

 produce sex hormones. Secondly, owing to the absence of the 

 cortex, it is difficult to state how much of the male development 

 may be the result of the absence of normal ovarian secretions. 

 It is known that in the absence of such secretions male char- 

 acters tend to develop. Pearl and Surface ('15) described a 

 cow with cystic degeneration of the ovaries which led to the 

 development of male secondary sexual characters. Spayed 

 females in birds take on male secondary sexual characters, 

 (Goodale, '16), and the castration of the hen-feathered Sebright 

 male results in the development of the plumage characteristic 

 of the cock-bird (Boring and Morgan, '18, Morgan, '19). In 

 the third place, the interstitial cells of the free-martin gonad 

 may furnish sex hormones which aid in its own differentiation. 

 Certainly, some relation exists between the high degree of male- 

 ness and the hypertrophy of the interstitial cells, as in all of 

 the most highly transformed gonads they are abundant. If 

 these cells are to play any part in differentiation of a gonad, 

 they would be expected to occur at least in some of the foetal 

 stages. This is not the case, at least in the early foetal stages, 

 as no traces have been found. 



