422 MR. H. M. BERNARD ON THE [Aug. 1 894, 



show might be due to the presence or absence (presumably through 

 post-mortem destruction) of a thiu membranous cornea, -which he 

 assumes covered the eyes of the trilobites, similar to that which 

 covers the eye in Branchipus, I have always considered this mem- 

 branous cornea of Branchipus as indicatory of the former presence 

 of a water-sac, which secondarily disappeared as the eye became 

 stalked ; otherwise it seemed difficult to explain why the eye itself 

 did not belong to the external cuticle represented by the cornea. 

 In the same way, among some of the later trilobites, the water-sac 

 probably degenerated secondarily, leaving the eye in contact with, 

 but not strictly belonging to, the outer cuticle, which may have 

 covered the eye like a thin membrane. This whole subject is, 

 however, beset with great difficulties, so that it is impossible as yet 

 to come to any definite conclusion : for while, on the one hand, the 

 so-called faceted eyes of trilobites, showing round projecting single 

 eyes arranged at some distance from one another, remind one 

 strongly of the tips of crystalline cones, 1 such as occur in the eyes of 

 Apus (see fig. 8, p. 421), on the other it is clear from Clarke's 2 

 researches that these were certainly in some cases true corneal 

 lenses, apparently belonging to the outer cuticle, and, indeed, 

 somewhat elaborate structures. 



Further, the eye of Limulus is a great difficulty ; here we have 

 no trace of a water-sac, nor of corneal lenses, while the bodies 

 which appear analogous to the crystalline cones are simply inward 

 projections of the outer cuticle. In discussing the eye of Apus, I 

 was led to the conclusion that the eye of Limulus was the more 

 primitive, a conclusion also arrived at by Watase. 3 If this is so, 

 then these eyes certainly belong to the external cuticle primarily, 

 and not secondarily by the degeneration of a water-sac. The only 

 way out of the difficulty, it seems to me, is to assume that while, in 

 some cases, the eyes, in travelling backwards, passed beneath a fold 

 of the cuticle into deep pockets, as above described, in others the 

 folds themselves degenerated secondarily, leaving the eyes once 

 more on the free exterior surface of the head. 



V. Behind the eyes of Apus there occurs, in all species of the 

 Apodidse that I have examined, the well-known ' dorsal organ,' 

 which in Apus appears to be an excretory organ. 4 It is often raised 

 on a slight plateau above the surrounding cuticle ; the cuticle of the 

 plateau itself is extremely thin, and likely to collapse easily during 

 the early stages of fossilization. Did such an organ occur on the 

 dorsal surface of the head of the trilobites ? Fig. 9 (p. 423) shows 

 us that, in the Cambrian trilobite Olenellus asaphoides, there was such 

 an organ, of essentially the same shape as that in Apus, but apparently 

 shifted farther back than in Apus, that is, on to the fifth segment. In 



1 See Packard, ' The Structure of the Eye of the Trilobites,' in the ' American 

 Naturalist' for 1880. 



2 ' Structure and Development of the Visual Area in the Trilobite Pkacops 

 liana. Green,' Journ. Morph. vol. ii. (1889). 



3 • Morphology of the Compound Eyes of Arthropods,' Journ. Roy. Micr. 

 Sue. 1890, p. 318. 



4 ' The ApodidaV p. 304. 



