536 THE AMERICAN NATURALIST [Vol. LV 



enter the second growth period (Fig. 7). It is difficult 

 to understand when Swingle says that E. Hertwig, 

 Kuschakewitsch and Witsohi ''concluded that all frog 

 tadpoles first develop as females." On the contrary I 

 have (1914) described the typical (or direct) develop- 

 ment of the testicle as it is to be observed in the alpine 

 Rana temporaria. Under optimal conditions of tempera- 

 ture in this variety of the grass frog already after the 

 twelfth day (larvsB 20 to 22 mm. total length) 50 per cent, 

 males are found; in such cultures transformation of sex 

 never occurs. 



In his material Swingle has not seen this typical de- 

 velopment of the testicle. The described first and second 

 year males are in reality hermaphrodites. His photo- 

 graph 33, plate 4 (1921) does not show a transverse 

 section through a male but through the female gonad of 

 a first year tadpole, characterized by the ovarial sac 

 (secondary genital cavity) and the peripheral germinal 

 epithelium. Photograph 34 likewise is not taken from 

 a male gonad but from a hermaphroditic one. It shows 

 the same stage of transformation of an ovary into a 

 testicle as Fig. 45 in our publication (1914) : representing 

 the gross structure of an ovary, but in the thickened wall 

 of the ovarial sac are several immigrated germ cells 

 (spermatogonia). Photograph 35 illustrates another 

 type of transformation, characterized by an excessive 

 proliferation of the sex cords, as is likewise described 

 in our publication (1914, Fig. 41). 



The cytological facts described in great detail by 

 Swingle will be discussed in another communication. 

 They do not alter our view of the significance of the 

 developmental changes. 



If there should still remain any doubt in regard 

 to the correctness of my interpretation the following ac- 

 count may help to dispel it. 



