ßgg William Patten 



in those nocturnal, or deep sea fishes, which are so universally supplied 

 Avith reflecting- surfaces. 



I must anticipate the theoretical remarks to be given in the fifth 

 chapter, in order to say that I regard these bodies as organa adapted 

 for the absorption of light energy , rather than for the perception of 

 objects. These accessory organs may with propriety be called eyes 

 since , originally , all eyes had such a function , and may still have it 

 even in the more highly perfected forms. 



The careful and valuable studies of Grobben (55) upon the 

 eyes of Phyllopod Crustacea , have shown that in this group the 

 lateral eyes are overgrown by a double, membranous fold, in some 

 cases enclosing the eye in a complete sac, which in other instances 

 remains open. He regards these folds as protective, and considers 

 that they are not developed in BrancJiipus] but the fact, as I have 

 shown, that the latter g-enus is supplied with a corneal hypodermis, 

 would indicate that a similar fold was present in the embryos. If 

 this is so , we must suppose that the hypodermic folds over the Clado- 

 ceran eye represent the corneal hypodermis of the higher Crustacea, 

 and that the folds have a deeper phylogenetic signification than 

 has been supposed. The resemblance betweeu this foldiug of the 

 hypodermis, and the invagination to form what I have considered 

 as the primitive Arthropod eye, is so striking that oue cannot fall to 

 notice it. But on the other band the resemblance may be a super- 

 ficial one, and we'should not be warranted in regarding it as haviug 

 a phylogenetic meaning, without a more accurate knowledge of the 

 eye in this group of Crustacea. 



Development. 



I can see no reason for supposing that the developmeut of the 

 visual Organs of Arthropods should be any different, in principio, from 

 that ofMolhiscs. Carrière states, on the authority of Kennel, that 

 the eyes of Peripatus are formed by the modification of invaginated, 

 and in the early stages, opeu-mouthed, optic cups. The strongest ob- 

 jection that can be urged against the supposition that all Arthropod 

 eyes are modifications of primitive ojitic vesicles is that embryology 

 ^ves no evidence of such an invagination. Although the embryological 

 evidence is very scanty and entircly confined to the compound eyes, it 

 is pretty certain that in most, if not all, compound eyes, there is no trace 

 of an invagination; but this ought not to be an insurmountable objec- 



