560 THE SENSES AND SENSORY ORGANS. 



to i '55. In a similar manner I determined the refractive index 

 in the cornea of the Hornet, and found it to be 1*53 [219]. 



It is well known that a lens composed of layers increasing 

 in density towards the centre has a higher refractive index 



J 



than the highest refractive index of any of its layers, and 

 Exner accounts for the high refractive index of the entire 

 crystalline cone in Dytiscus on the same principle. He 

 believed it to be a cone consisting of a series of envelopes 

 increasing in refractive power towards its axis. There is little 

 doubt, I think, that this affords a true solution of the dis- 

 crepancies previously observed. 



Exner terms such a cylinder a refractive cylinder, and shows 

 that such cylinders act exactly as convex lenses do, and that 

 the focus of a refractive cylinder is a recurring function of ^ 

 where c is a constant, and / the length of the axis of the 

 cylinder within certain limits (see p. 572). Cylinders have 

 been made of glass annealed in a special manner which 

 exhibit all the properties of Exner's refractive cylinder. 



Exner believes that the subcorneal image lies in the middle 

 of such a cylinder, the crystalline cone, and that rays \vhich 

 were parallel before entering the refractive media of the insect's 

 eye leave the inner extremity of the crystalline cone as parallel 

 rays. If this were the case, it is difficult to understand the 

 optical purpose of the refractive media, as no other result would 

 be attained than a diminution of the amount of light falling 

 upon the retinal end organ, unless such parallel rays were sub- 

 sequently brought to a focal point, since the density of a pencil 

 of light can only be increased by the convergence of the rays of 

 which it is composed (see page 575). 



The Dioptric Theory, which I enunciated in 1884 [233], is 

 that the pencils leaving the crystalline cone are brought to a 

 second focus by the rhabdome, which acts as a lens ; these 

 pencils may be cither divergent, or parallel pencils as Exner 

 supposes, and in this way a magnified image of a part of the 

 subcorneal image is formed at the inner extremity of the 

 rlialulcinc, and as there is a double reversal, it is directed in 

 the same manner as the external object and is an erect image 



