MORPHOLOGY OF THE FOREBRAIN 455 



caudad than the ventral. Fig. 48 passes through the caudal 

 border of the interventricular foramen on the left side and im- 

 mediately behind it on the right. It passes through the anterior 

 commissure below and the dorsal sac rostral to the habenulae 

 above; cf. fig. 50, a sagittal section of an older embryo. From 

 the interventricular foramen, which is very wide immediately 

 rostral and dorsal to this level, two ependymal grooves extend 

 backward in the same relations as in urodcles, the sulcus dien- 

 cephalicus medius above and the sulcus diencephalicus ventral is 

 below. Between these is the pars ventralis thalami which is seen 

 to connect broadly with the lateral wall of the hemisphere. In 

 fact it is the precise equivalent of the eminentia thalami of am- 

 phibian larvae. It is completely separated from the dorso- 

 median part of the hemisphere by the membranous posterior 

 chorioidal fold and incompletely separated from the ventro- 

 median part by the sulcus ventralis. In the adult its rostral 

 end is carried forward, as in the frog, to form the nuclei of the hip- 

 pocampal commissure and tractus cortico-habenularis (see p. 

 461). 



Dorsal to the sulcus medius is the pars dorsalis thalami, and 

 above this the sulcus dorsalis and habenula. These relations come 

 out still more clearly in a section taken a little farther back (fig. 

 49). 



Fig. 50 is a sagittal section through the brain of an older embryo 

 of Chrysemys marginata 26.7 mm. long and fig. 51 a parasagittal 

 section from the same embryo illustrating the relations of the 

 diencephalic sulci. The pars dorsalis thalami is here, as in rep- 

 tiles and mammals generally, divided into a central major part 

 and a nucleus dorsalis containing more densely crowded cells. 

 Both embryological and phylogenetic development show that 

 these two parts have a common origin. 



The subdivisions of the diencephalon are shown with great 

 clearness in the models of the brains of gecko embryos figured by 

 Tandler and Cantor ('07, see especially figs. 10, 14, and 17). 

 The boundary between the pars ventralis thalami and the hypo- 

 thalamus in these embryos is less distinct than in the Amphibia 

 and the chelonian embryos here described. Tandler and Cantor's 



