82 W. J. UAKIX. 



of the distal nei've could be found. The isolated cells 

 obtained by macerating the retina in chloral hydrate solution 

 are illustrated in fig. 15 (PI. 7), but these were only obtained 

 on a few occasions and after a long search, for it is most 

 difficult to separate them from the distal cells. 



The cell-body is very small and there is but little cyto- 

 plasm left surrounding the nucleus, but from this extend 

 long branched processes. The nucleus retains the blue stain 

 after Mallory when it has been taken from the nuclei of the 

 distal sense-cells, and generally it may be said that the inter- 

 stitial cell-nuclei stain darker and are more homogeneous, it 

 being more difficult to resolve the granules. They are further- 

 more flattened and are only about half the size of the sense- 

 cell nuclei. The processes lie in close contact with the distal 

 sense-cells, there being often two clasping them and extending 

 between them towards the septum (PI. 7, fig. 15, a.). 



From the proximal end of the cell may arise one or more 

 irregular processes which branch and penetrate some distance 

 between the rod-cells. It is quite easy to understand how 

 these long processes, which in my opinion tie and support 

 the sense-cells, have been for a long time considered as 

 nerve-endings, either of nerve-cells or of the fibres from the 

 distal nerve. In many ways the interstitial cells resemble 

 in shape and staining the neuroglia cells found clasping the 

 nerve-cells in the various ganglia of Pecten and other lamelli- 

 branchs. The outer ganglionic layer of Patten is composed, 

 therefore, of two types of cells — sensory cells forming an 

 outer layer and connected with the distal nerve, and support- 

 ing non-sensory cells interpolated between them. Miss Hyde 

 did not recognise the latter at all. I have had no success 

 with methylene-blue methods, but I do not think they would 

 be of much advantage unless fixation was very good (a thing 

 not by any means easy to attain with many special methods), 

 ■for it would be almost impossible to check the results and to 

 determine whether, in the confusing mass of fibres, nervous 

 or both these and non-nervous processes had taken the stain. 



We have now to consider the second sensory part of this 



