Eyes of Molluscs and Arthropods. 017 



needles. aud examined with a high magnifying- power. Although I haveuot 

 always beeu equally successful, some most brilliant i)reparatious were 

 ohtained, in whieh eouldbe seen half a dozeu or more nerve fibres exteud- 

 ing aloug the wall of the cell, over the outer euds of which they projeeted 

 in long festoons, whose leugth often exceeded that of the cells them- 

 selves. The prineipal fibres were quite large and perfectly distinct, 

 with very few side brauches; biit towards the extremit3^ they broke up 

 into many smaller l)rauches, which eontinued to siibdivide more and 

 more rapidly. imtil they at last ended in myriads of the finest fibrils, 

 not terminating freely , biit unitiug with each other to form a perfect 

 network of coutinuous fibrillae. In some cells, all the fibres, with their 

 mass of terminal fibrillae , remained so united as to retain the shape of 

 the rod before the cuticular substance was dissolved; in other cases. 

 the large branehes had become separated from each other , and were 

 turned and twisted in all directions ; fig. 63 is only one out of hun- 

 dreds of similar cells, each provided with the long festoons of nerve 

 fibres, the ultimate ramifications of which, with all their confusing com- 

 plexity, could be seen infinitely more clearly in the original prepara- 

 tions, than I have been able to indicate in the figure , where only one 

 of the fibres is entirely drawn. By selecting a small and well isolated 

 portion of the retinidium, it will be seen to consist of innumerable, 

 equally large branehes, which become continuous with each other in all 

 directions to form an inextricable network of fibrillae (fig. 61). Although 

 nearl}' perfect , isolated retinidia are often obtained, they are usually 

 united into larger or smaller groups in such a manner that there can be 

 no doubt that, at the extremity at least, the individuai retinidia are 

 connected with each other by fibrils, in the same manner as the various 

 nerve branehes of the same cell. The bases of the rods are reduced 

 to slender stalks, separated from each other by clear Spaces through 

 which the nerve fibres of the colorless cells pass outwards to form a si- 

 milar network of fibrillae which unite with those described above. In 

 preparations of isolated cells, from whose rods the cuticular substance 

 has not been dissolved, I have never been able to find retinophorae with 

 rods at their outer end; the latter was always eontinued outwards as one, 

 or several fibres, that soon divided up into smaller branehes, and termi- 

 uated in the same mauner as those of the surrounding pigment cells. It 

 is probable that the rods of the colorless cells are so inconsistant as to 

 lose their shape when the surrounding cells are removed, or that they 

 are dissolved in the processes of maceration. The retinidial layer of the 

 cuticula does not entirely fili the cavity of the eye ; it passes quite grad- 



