24 BERTRAM G. SMITH. 



Considering the low taxonomic position of Cryptobranchus 

 amongst the Urodeles, the occurrence of external fertilization in 

 this form indicates that this is the primitive method for the group ; 

 internal fertilization by means of spermatophores, characteristic of 

 so many Urodeles, is a later development. 



Nothing conclusive is known concerning the method of fertili- 

 zation in the Crossopterygian Ganoids, the ancestral stock from 

 which the amphibia are believed to be descended. In the higher 

 Teleostomi, with a few notable exceptions, fertilization is external. 

 In the Dipneusti, which, like the Amphibia, are supposed to be 

 derived from a primitive Crossopterygian form, nothing definite is 

 known concerning the manner of fertilization. In view of the 

 diversity of habits within such groups as the Teleostomi and the 

 Urodela, and the wide gaps between the Urodeles and the nearest 

 living representatives of ancestral groups, it is probable that even 

 were our knowledge of the habits of existing animals complete, a 

 comparison between the Urodeles, the Crossopterygii, and the 

 Dipneusti would have little phylogenetic significance. On the 

 other hand, the occurrence of a simple method of external fertili- 

 zation as the primitive condition of the Urodeles suggests the pos- 

 sibility of finding within the group stages in the development of 

 the complicated breeding habits, involving internal fertilization by 

 means of spermatophores, characteristic of most of its repre- 

 sentatives. 



Brooding Habits. — It has been noted previously that the male 

 usually remains with the eggs after fertilizing them. That the 

 male may render efficient protection to the newly-laid eggs is 

 shown by the following incident of September I, which I quote 

 almost verbatim from my notes. 



A large " red " hellbender (afterwards identified as a female) 

 crawled across the stream toward a flat rock submerged in about 

 IO inches of water. This rock was about 3 feet by 2^ feet, 

 slightly tilted so that the down stream side was a little raised 

 from the bottom. As the female approached this rock, the large 

 fiat head of another Cryptobranclms (afterwards found to be a 

 male) of unusual size was thrust out from under the down-stream 

 edge of the rock and the newcomer was seized at the side of the 

 head by the jaws of the male. She drew back and was released. 



