290 Jonnial of Comparative Neurology and Psychology. 



Labridae and Pleuroiiectidae. In general its development is found to 

 \a.xy pari passu with the development of the o})tic lobes and visual ap- 

 paratus, especially shown by its small size and simi)le structure in the 

 cave-inhabiting fishes. It is found, in a rudimentary state, in Amphi- 

 oxus, is present also in Cyclostomes, is more developed in ganoids, 

 but reaches its climax in teleosts. In the latter the large cells of its 

 homologues in other forms (Dachkern and nucleus magnocellularis) are 

 replaced by a greater number of smaller cells. 



The torus cells "are usually bii)olar, but may be unipolar or multi- 

 polar. In every case, however, three neurites ultimately arise from 

 the cell, either directly or indirectlf by the division of a chief ])ro- 

 cess." The dorsally directed neurites are non-meduUated and form 

 two tracts. One of these, the tractus toro-tectalis, breaks up in the 

 superficial fiber zone of the tectum, there coming in contact with optic 

 nerve terminals. The other tract, the tractus toro-cerebellaris, passes 

 laterad to the dorsal decussation and theuce ventrad parallel with the 

 posterior commissure. It is difficult to trace, but Sargent believes 

 it to be identical with Johnston's tractus toro-cerebellaris in Acipcuscr. 

 "The chief centrifugal neurite, or axone," of the torus cells is some- 

 what coarser. These neurites form several fasciculi which ultimately 

 "unite to form the fiber of Reissnek which runs i)OSterior into the 

 canalis centralis and thnnigh the posterior portion of its course gives 

 off branches which enter the ventral part of- the cord and jirobably run 

 to the musculature." 



These important relations of the processes of the torus cells are 

 only illustrated diagramatically by figures. Such schemata should al- 

 ways be supported by drawings of the elements in ijuestion as sho»yn 

 in the preparations that we may be sure how far and how precisely the 

 cell processes have been actually traced. It is often surprising upon 

 how few direct observations many an extensive neurological schema 

 rests. Such drawings are especially demanded where the relations are 

 so unusual. This defect in the present instance has been partially 

 supplied by the author's figures in previous articles and will undoubt- 

 edly be completely remedied in the more extended publication an- 

 nounced as in i)ress. 



The physiological significance, according to Sargen r, of the fiber 

 of Reissnek and of the toro-tectal tracts is that they constitute a short 

 circuit for (juick optic reflexes. Such a view rests in part ujion the 

 assumption that there are no other "one neurone" paths from tectum 

 to motor nuclei in the cord, which is probably not the case ; and that 

 the fiber of Reissnkk pa.sses out directly to the muscles. The latter 



