SURVEY OF THE ARACHNIDAN EYES. 71 



fusion has gone still further in the lateral eyes of the Scorpiones, 

 and has finally led to the formation of the common lens now found. 

 In this view each lateral eye of the Scorpion is regarded as the sum 

 of a number of single eyes. The rhabdoms contained between the 

 retinal cells correspond to those in the eye of Limulus. The latter 

 is somewhat highly developed, its character as a compound eye 

 being recognisable from the formation of rhabdoms in the retinulae. 

 The lateral eye of the Scorpion thus also appears as a degenerate 

 facet-eye. The depression which ontogenetically represented it must 

 therefore not be regarded as the primary optic pit. 



In Scorpio, as is Avell known, there are several lateral eyes on each 

 side. Each of these is to be considered as a complex of single eyes, 

 and all taken together as representing a lateral eye of Limulus. They 

 have, perhaps, been derived from one large facet-eye by the separation 

 of individual complexes of single eyes. An altogether analogous 

 process seems to occur in the Trilobites. 



While the Trilobites usually have a facet-eye on each side, the genus Harpes 

 has, in place of the facet-eyes, two or three prominences {H. vittatus 2, H. 

 ungula 3, Bakeaxde) with perfectly smooth surfaces, which are simple eyes, 

 strongly resembling in outward appearance the lateral eyes of Scorpio. Palaeon- 

 tologists have appropriately described them as ocelli, although, from a zoological 

 point of view, they do not deserve this name, having most probably arisen in a 

 way similar to that conjectured in connection with the lateral eyes of the 

 Scorpiones. 



Careful examination of the surface of Trilobite eyes perhaps allows of our 

 forming further conclusions as to the origin of these facet-eyes, which coincide 

 with the view just put forth. The whole surface of the facet-eye usually differs 

 in its structure from the rest of the body covering. In some genera, however 

 (Phacops, Dalmania, Zittel), the covering of the visual surface is identical with 

 the rest of the shell, while the single lenses lie in rounded or polygonal de- 

 pressions. This looks as if, in the last case, the single eyes were not close to one 

 another, while in the first case they were crowded together. As our knowledge 

 of the structure of the Trilobite eye is so small, this can merely be advanced as 

 a conjecture. 



The median eyes of Scorpio, by the distinct formation of retinulae 

 and rhabdoms, are more evidently proved to be compound eyes. 

 They are multilaminar (Fig. 10 C, p. 14). The same is the case 

 with the median eyes of Limulus, the elements of which are also 

 comparable with those of the eye of Scorpio. This multilaminar 

 character raises the eyes under consideration to a higher ontogenetic 

 level, further evidence of this being afforded by their complicated 

 method of development. Both these characters make it impossible 

 for us to trace back the median eyes of Scorpio and of Limulus to 

 lateral eyes, although the two kinds of eyes otherwise show great 



