570 William Patten 



tinuations of the crystalline cone cells consti tut e the «rhab- 

 doms«, in whose formation the retinulae take no part what- 

 ever. Ifthis Observation alone iscorrect, theentireresults 

 of Grenagher's work must fall to the ground. 



If the retinulae are not the essential elements, what then are? 

 Theyarethecrystallineconecell8,which,morphologically, 

 must he the essential elements, when compared with those 

 in the Mollusca and simple eyes of Arthropods. But hesides 

 the morphological necessity for considerine them as the 

 essential elements, we have, in addition, a physiological 

 one,sincewehaveprovedthatinthemthenervesterminate, 

 forming that system of cross fibrillae, so characteristic of 

 the Molluscan rods, called a retinidium. Thereis a third 

 reason for considering the crystalline cone cells as essen- 

 tial elements. for upon their retinidia there is a Constant 

 effort to throw the greatest amouut of light, often in the 

 form of a perfect image. The structure of the retinidium coincides 

 with that of the Mollusca, in that it consists of successive layers of cross 

 fibrillae, radiating from one or more axial fibres, and uniting with 

 nerves on the outer wall of the calyx. The fibrillae of the retinidium 

 may be supported , as in the rods of Molluscs, in a firm chitinous 

 secretion, the crystalline cone : or the secretiou may be fluid, or so in- 

 consistant as to be destroyed or dissolved by reagents, as in Mantis and 

 Musca. The facts we have enumerated furnish us with entirely new 

 data for comparisou, and without further consideration for the present 

 of Geenacher's views, we will see in what direction they lead us. 



It is, of course, no louger possible to compare the vitreous cells 

 of the simple eyes with the crystalline cone cells , or retinophorae , of 

 the compound eye. The undoubted homologue of the vitreous 

 cells, however, is to be found in the corneal hypodermis of 

 the compound eyes: the position and function of both layers 

 point to this conclusion. There is no course open except to 

 compare the two remaining layers with each other. Let us take , for 

 instance , one of the lateral eyes oi Se or pio, and it will be found, 

 according to Graber and Lankester, that the ommateum consists of 

 ommatidia each one composed of five centrai, colorless cells, or 

 retinophorae, hearing on their axial faces a cuticular secretion, or 

 rod. The retinophorae are surrounded by two or three circles of pig- 

 ment cells, or retinulae. Here then the colorless cells scerete the rods. 

 while the pigment cells play only secondary parts. It is exactly the 



