SARGENT: THE OPTIC EEFLEX APPARATUS OF VEIiTEBRATKS. 153 



terior commissure shows the horns of the recessus as small circular 

 orifices with thick walls composed of this radiating epeudyma (Plate 1, 

 Fig. 6, rec. /'.). 



The ependyma of these grooves is conspicuously distinguishable from 

 the ependyma lining the other portions of the ventricles, and the de- 

 marcation between the two is always clear and sharp. It is thicker, of 

 very compact structure, and composed of cylindrical or columnar cells 

 with prominent nuclei (Plate 1, Fig. 6). The pei-ipheral ends of the 

 cells are continued peripherad in Jong ependymal fibres, which, collected 

 into fasciculi, radiate from the groove. This ependymal thickening 

 doubtless serves as a supporting structure for the axons of the toctal 

 reflex cells which pass between its columnar cells before emerging into 

 the ventricle. In the 2G-day larvae studied no evidence of the grooves 

 or ependyma was found. Only in the more advanced individuals of the 

 30-day stage could the beginnings of these structures be distinguished. 

 It is remarkable that such definite and conspicuous structures have 

 escaped the attention of the numerous investigators who have studied 

 the brain of Petromyzon.^ 



Ahlborn has vaguely shown in P. planeri, in his Figure 26, a structure 

 under the posterior commissure that may be taken to represent this 



1 Since the above description was written (September, 1901), tliere has ap- 

 peared a paper by A. Dendy, February, 1902, " On a Pair of Ciliated Grooves in 

 tlie Brain of tlie Animoccete apparently serving to promote the Circulation of the 

 Fluid in the Brain-cavity." Dendy first found these lateral longitudinal grooves in 

 tlie Ammocoetes stage of the New Zealand lamprey (Geotria australis), and later 

 in the Ammocoetes stage "in one of the European species" (Petromyzon ? sp. ?). 

 In neither of the species examined by him are the grooves so deep as I have found 

 tlieni in adult P. marinus. In Geotria " the inner margins of the two grooves 

 touch one another in the middle line beneath the posterior commissure. Anteri- 

 orly the two grooves diverge from one another on the roof of the recessus sub- 

 I)inealis, and disappear in the deep crevices between the ganglia habenulae and the 

 side walls of the brain. Posteriorly they terminate at the hinder margin of the 

 posterior commissure." In neither of the species examined did he find the epithe- 

 lium prolonged around and above the commissure so as to form the lining of the 

 liorns of the mesocoelic recessus, as I have described it. 



Dendy has described the grooves as ciliated ; he distinguished, however, between 

 the cilia of the ventricular walls and those of the grooves, the latter being longer. 

 Though the ventricular walls are abundantly ciliated, I believe that there are no 

 cilia in the grooves. He has perhaps mistaken the axons entering the grooves for 

 cilia. He found these grooves only in the Amnioca'tes stage and doubted their 

 existence in the adult. Dendy has chosen to see some connection between these 

 grooves and a septum formed by a fold of the choroid plexus, and believes that 

 these perform some function in maintaining the circulation of the cerebro-spinal 

 flui<l, though he fails to make it clear in what way they could do this. 



