Functional f.ictfjis in the inorpholot^y of Ihe foreljrain of lishes 197 



tionships prepared the way for the difterentiation within this área of true 

 hippocampal cortex in higher forms. 



The essential functional factor in this process of cortical differentiation 

 is the progressive elimination of direct fibrous connections with lower 

 and simpler reflex centers and the elaboration of associational connections 

 of the primordium hippocampi with a variety of other complex correla- 

 tion centers of the hemispheres, none of which is concerned exclusively 

 with any single sensory system. A special medial efferent tract from 

 the primordium hippocampi, the columna fornicis, was developed early 

 in this history. This tract also appears in selachians, in which the topo- 

 graphic relations of the primordium hippocampi resemble more closely 

 the amphibian than in any other fishes except the Dipnoi. 



In the lateral wall of the hemisphere the development of true cerebral 

 cortex within the lobus pyriformis occurs in reptiles under conditions 

 somewhat similar to those in the medial wall, but this cortex remains of 

 lower type for it is in ali cases under more direct olfactory influence. 

 The general cortex and neopallium emerged in the dorso-lateral wall 

 under the influence of the non-olfactory components of the lateral fore- 

 brain bundle. 



Studies now in process upon the early development ot the forebrain 

 of Amblystoma in connection with the growth of the fiber tracts and 

 the development of reflex patterns indícate that the same functional 

 factors \\hich are revealed by morphological study of various lower brains 

 are operative in the ontogeny of the amphibian brain and play a major 

 rale in its morphogenesis. 



The physiological factors which are considered in this paper are evi- 

 dently adaptations to diverse modes of Ufe which have been developed 

 by natural selection or otherwise relatively late in phylogenetic history, 

 that is, they are cenogenetic rather than palingenetic characters. Primi- 

 tive metamerism, the longitudinal zones of His, and similar very ancient 

 characters were probably defined much earlier in the evolutionary history, 

 and these adaptive characters were then superposed upon the more 

 ancient patterns. This analysis, therefore, sheds no üght upon the 

 problems of primitive metamerism, etc., and it is not inconsistent with 

 any hypothesis whatsoever relating to such questions. 



The remarks made in the preceding paragraph apply also the con- 

 clusions to which I was led in an examination of the morphology of 



