THE THEORY OF ARTHROPOD VISION. 555 



garded as the physiological end of the nerve.' In this form 

 the original theory of Miiller has been very generally accepted. 



I have already discussed the morphological and develop- 

 mental grounds which lead me to reject this view, and it 

 remains for me to discuss the optical principles on which, I 

 hold, it must be laid aside. 



Gottsche [201] first attacked Miiller's hypothesis on the 

 ground that a real inverted image is formed behind each 

 corneal lens. The existence of such an image is easily demon- 

 strated and was even known to Leeuwenhock. But various 

 views have been held in relation to its significance. If we 

 admit that the position of this image does not correspond with 

 any receptive or retinal surface, which has been abundantly 

 demonstrated, it can only be regarded as a source of radiant 

 light. It has been argued by some writers that the image is 

 adventitious, that every oil drop produces a similar image ; but 

 how this argument can avail those who believe in its non- 

 existence, or assist in getting rid of a collection of real focal 

 points from which light radiates in diverging pencils, is beyond 

 my comprehension. The sub-corneal image cannot be thus 

 lightly disregarded, as every convex facet must produce pencils 

 of converging rays, just as every spheroid of oil has a focal 

 image of objects behind it. Nor could such a focal plane be 

 neglected if an oil drop were interposed in any optical 

 apparatus. The presence of a sub-corneal image is in itself 

 absolutely destructive of Miiller's hypothesis, since all the light 

 proceeding from it must consist of widely divergent pencils. So 

 that the result of the presence of a lens in front of each of 

 Miiller's radial tubes would be that less light would penetrate 

 it than if no such lens existed. 



Grenadier's Hypothesis. Grenacher, who is at present gene- 

 rally admitted to be the authority on the compound eye, 

 adopts Miiller's view, with the modification already quoted 

 from Huxley, that the great rods are light-transmuting organs. 

 It is difficult to trace the origin of this modification: I think 

 Grenacher first enunciated it in so many words, but there is 

 evidence that it was in the minds of many writers before 1877, 



