442 C. JUDSON HERRICK 



species illustrate arrests at different stages in this developmental 

 process. Even though the very generalized forms like Necturus 

 give evidence of retrogression, yet the definitive histological 

 pattern of the adult brain is strikingly similar to that of larvae 

 of other forms and of adult cyclostomes. The histological 

 structure of larval urodeles and of adult Necturus has been 

 described in considerable detail (Herrick, '14, '14 a, '17) and that 

 of the tadpoles of the frog is probably essentially similar. 



From these considerations it is probable that the line of descent 

 through extinct ganoidean types to the earliest Amphibia of the 

 Devonian was generalized not only as regards the external form 

 of the brain, but also in its histological structure. The preser- 

 vation of this plastic, undifferentiated, or 'young' type of tissue 

 was doubtless an essential factor in making possible progressive 

 evolution in the new amphibian direction when the time was 

 ripe for this evolutionary movement. 



In the meantime most of the other groups of fishes referred to 

 in this sketch had diverged from the generalized ancestral form 

 of vertebrate brain, not only in the form relation of the forebrain 

 already described, but also in the direction of much more highly 

 specialized histological structure. Their tissues were matured 

 or 'senile' in the sense defined by Child ('15). The estabhsh- 

 ment of the sharply circumscribed and rigid reflex patterns 

 characteristic of the various species of higher fishes stabilized 

 their organization at the expense of plasticity and thus deter- 

 mined the direction of future evolution of each phylum within 

 certain wide hmits. Dedifferentiation permitting a new start 

 in a different direction was a biological impossibility. The 

 probable mechanism of this orthogenetic trend I have elsewhere 

 discussed ('20). 



The amphibian hemisphere is evaginated on a quite different 

 plan from that of the selachians already described. In the sharks 

 (and more conspicuously still in Chimaera) the most highly devel- 

 oped secondary olfactory centers are found as close as possible 

 to the bases of the fully evaginated olfactory bulbs. The 

 terminal plate and adjacent regions are greatly thickened (figs. 

 8, 9, 10), and the hollow cerebral hemispheres described in this 



