THE BEE. 35 



eyes of all iasects, and our inference in this respect 

 is drawn — 



1. From the exceptional character of hexagonal or 

 any other than circular lenses in the eyes of all ani- 

 mals, and from the fact of the simple eyes of insects 

 themselves being circular (PI. IV, fig. 4). 



3. From the circumstance that, in the insect races, 

 the conical lenses of the ocelli (PI. IV. fig. 1, b) (to be 

 described presently), which do not impinge one upon 

 another, are not hexagonal, but round. 



3. Because, in the posterior angle of the compound 

 eye of the worker-bee, we often find some of the 

 corneal or external lenses of a smaller size, and not 

 adherent (PI. IV. fig. 5), but having a little inter- 

 mediate space surrounding each, and these facets are 

 invariably round. 



4. From the fact that in one insect at least, the Sheep 

 tick, Mehphagus ovinus, which ranks very low in the 

 scale of development, we find all the external facets 

 of the compound eyes non-adherent and circular*. 



So much, then, for the corneal lens of the oceUus 

 of the Bee, a compound hexahedral prism with double 

 convex surfaces. Following the course of a ray of light 

 after it has passed through this lens, we find that it 

 traverses a vacant space (PL IV. fig. 1, c) before enter- 

 ing the conical lens (6), this space being surrounded 

 by the dark pigment already referred to {d), and con- 

 stricted or narrowed midway into the form of a round 



* A careful examination of the eye in the pupa, whilst in 

 process of development, confirms the opinion here expressed, 



c 



