492 ERNEST W. h. HOLT, 



cephalon, which is seen (t. th.) as a single layer of columnar cells 

 passing forwards from the front wall of the pineal stalk (infra-pineal 

 process). Where the roof of the thalamencephalon passes into that of 

 the cerebrum, the cells rapidly diminish in size, the latter structure 

 {pa.) consisting of a single layer of very small roundish cells. 



Below the broadest part of the pineal body a slight downward 

 fold (/".) is noticetl in the thalamencephalic roof. 



Transverse sections show a very narrow commissure joining the 

 white matter of the tips of the optic thalarai at the origin of the roof from 

 the pineal stalk. The same commissure (fig. lis. c.) is also revealed 

 by horizontal sections in the early post-larval stage, and a few fibres 

 are, seen to pass up from it into the pineal body (fig. 11 pn). This 

 is evidently the homologue of the commissure recognised by Balfour 

 and Parker in Lepidosteus (3, p. 379) and by Balfour (2, pp. 355 

 — 356) in Scyllium as representing the grey or middle commissure of 

 higher vertebrates. 



It has also been recognised by Mc Intosh and Prince in Anar- 

 rhichas (6, p. 909). In the young Ammocoete Shipley (8, p. 357) 

 noticed what is evidently the same commissure, and adopted for it 

 the name of superior commissure, given by Osborn to a similar 

 structure described by him in some Amphibians. The former observer 

 suggests that it is similar in topographical relationships to the com- 

 missure of the pineal stalk in the mammalian brain. However this 

 may be, the term is convenient. 



The fold of the thalamencephalic roof previously noticed appears 

 to be the first trace of the shutting off of the Recessus infrapinealis 

 of Hoffmann (5, p. 102), who has given a very clear account of its 

 development in the Salmonidae. In the herring it would appear to 

 be much later in appearing. It is evidently homologous, as suggested 

 by Mc Intosh and Prince in Anarrhichas, with the great vesicle in 

 the roof of the thalamencephalon in Lepidosteus (cf. Balfour and 

 Parker, 3, p. 376), the development of which appears to commence 

 in a similar manner (cf. 3, p. 379). 



In view of Rabl-Rückhard's (7, p. 127) researches the fold 

 must also be regarded as the homologue of that, which in Elasmo- 

 branchs acquires a vascular (choroidal) plexus, and becomes the tela 

 choroidea of the thalamencephalic roof. For convenience of definition 

 I would propose to term it, in Teleosteans at least, the Velum infra- 

 pineale. 



However, while Balfour (1, p. 399) shows that in Scyllium the 



