Eyes of Mollusca and Arthropods. 701 



is evident that Carrière either failed to recognize this similarity , or 

 regarded it as of uo consequenee , for lie called the eolorless cells of 

 the Coelenterate letiiia, sense cells, and similar elements of the retina 

 of Molluscs and Worms, secretiug cells ! but , so far as I could find, he 

 gave no reason for reg-arding them as morphologically different ele- 

 ments. In such simple eyes as tliose oi Nereis cultrifera, the optic 

 vesicle has been described as being filled with a vitreous body, but we 

 may be sure that a part of it is , in reality, composed of a layer of 

 rods, just as in the Molluscan eye. Indeed, in Alciope, the rods are 

 very highly developed, and do not differ essentially from those of 

 Pecteti^ or of the Cephalopods. 



The same modification of the ommatidia has taken place in Alcio- 

 pidaeasin the Cephalopods. Theretinulae havebeenreducedto anarrow 

 collar of rodless pigment cells which surround the neck of each retiuo- 

 phora. The double rod of the latter is very highly developed and con- 

 tains an axial nerve fibre, which canbe traced iuto the body of theretino- 

 phora. These facts render it almost certain that the retinophorae are 

 double, as in Mollusca, and therefore we should expect to find two 

 nuclei. Graber Claims to have found three nuclei in the »Retinal- 

 schläuchen«, as he calls the retinophorae , but it must be understood 

 that I do not lay any stress on this fact . for there appears to be no 

 doubt that the extra nuclei he saw did not belong to the retinophorae, 

 but to the surrounding pigment cells. The second nuclei of the retino- 

 phorae, that I have shown to be present in Area, Pecten, and Haliotis^ 

 are bodies very difficult to observe, even in the most carefully pre- 

 pared sections. The cells must be isolated by maceration, in order to 

 demonstrate, with certaiuty, the presence of the second nucleus. 



The open ends of the rodsinthe Alciopidaeare very interestiug, for 

 they offer a further confirmation of the supposition that they are double; 

 they may be compared with the rods found in Pecten, where we have 

 evidence of a former opeuing , now completely closed , in the fact that 

 the central nerve issues from the end of the rod to form the »loops of 

 the axial nerve«. It is very probable that the axial nerve of Alciopidae, 

 and also that of other highly developed forms, extends through the 

 opening of the rods to form »axial loops« similar to those in Pecten. 



The slightly modified ommatidia, found even in the highly develop- 

 ed eyes of some Wonns, render it very probable that the isolated omma- 

 tidia are present in the scattered pigment areas so often found there. 

 This is still more probable when we consider the frequency with which 

 eye-like pigment spots occur. I have examined superficially, without 



