SARGENT: THE OPTIC REFLEX APPARATUS OF VERTEBRATES. 143 



masses {pp. cit., Fig. 9), or sometimes aggregated about Reissner's fibre, 

 whose course it may obscure, and whose diameter it may appear to 

 increase. Sometimes, especially in the ventricles, it becomes so coagu- 

 lated about a blood corpuscle or some disentegrating epithelial cell as to 

 produce artifacts strikingly like cells in appearance. In fact, it would 

 be possible to obtain photographs of such cell-like artifacts showing an 

 apparent nucleus, nucleolus, and chromatin, as well as long streaming 

 or branching dendritic processes. "When such an artifact occurs about 

 Reissner's fibre, it is easy to misinterpret it as a cell connected with the 

 fibre. If the animal had been dead several hours before the cord and 

 brain were placed in fixing fluid, the canal was found filled with coagu- 

 lated granular matter, and the fibre was so far disintegrated as to be 

 indistinguishable. 



(2) Detaclied epithelial cells. (See Sargent, :00, Fig. 4.) These cells, 

 I believe, are thrown out from the walls into the canal normally. I find 

 them in all stages of being pushed out from between other cells, and 

 in the canal they may be found in all stages of disintegration.^ In the 

 ventriculus terminalis some of these cells, thrown off at an early stage, 

 persist and develop into the posterior canal cells (Plate 4, Fig. 27). 



(3) Blood corpuscles. These are of course frequently numerous w'here 

 the brain has been removed from the animal, due to artificial lesions of 

 the blood-vessels, produced in taking out the brain ; but they have also 

 been observed in lesser number in small animals sectioned wdiole. 

 Their presence may be due to minute hemorrhages produced by the 

 fixing fluids. 



(4) Artifacts due to the. fixing fluids. These seldom occur, except 

 in the use of corrosive sublimate, and then only when the after-treat- 

 ment with iodine is not thorough. Such artifacts may have their shape 

 more or less determined by the cavities in which they are formed, but 

 then they are not easily misinterpreted. 



1 Peter (:01) has described a similar process in the canal and ventricles of 

 larval Rana. He finds that the cells of the inner pigment layer, which arises from 

 the ' Deckschicht,' degenerate and wander into the canal or ventricle. He fre- 

 quently found the broad end of the pear-shaped nucleus projecting into the 

 ventricle, pressed out between the cells of the nerve tube, while the pointed end 

 lay between the deeper cells. He has observed each stage of this ' Durchwande- 

 rung,' and believes this process is a dehiscence of the cells of the ' Deckschicht/ 

 so that " das Centralnervensystem einzig und allein auf Kosten der Sinnesschicht 

 entsteht." He thinks tliere is probably some relation between this process and the 

 formation of the ' Teloderm ' from the ' Leiterepithel ' described by Melmert (:96) 

 in Emys. 



VOL. XLv. — so. 3 2 



7 



