The Evolution of Optics. 



275 



near-sighted and possesses at best but imperfect vision. The 

 compound eye (Fig. 4) consists of a number of tubes some- 

 times as many as twenty- five thousand arranged side by side, 

 with such a disposition of the 

 lenses, pigment, etc., that 

 only those rays of light fall- 

 ing on the axis of the tube 

 may reach the retina. (This 

 is represented dlagrammati- 

 cally in Fig. 4 a.) There has 

 been much discussion as to 

 the method in which vision 

 is accomplished in these 



eyes, but it seems most probable that, as suggested by Miill- 

 er, the vision of the compound eye is a mosaic, each segment 

 of which corresponds to the field directly in line with the 

 axis of each tube. That each division of the compound eye 

 produces its own image seems highly improbable from the 

 fact that in many compound eyes the image would not fall 

 on the retina, and no arrangement for accommodation seems 

 to be present, while the nature of the crystalline cones and 

 the retina itself is such, in many cases, as to preclude the 

 possibility of a clear image. It seems probable that Miill- 

 er's theory is correct, that the compound eyes produce single 

 upright images, and that the ocelli, when present with the 

 compound eye, as is sometimes the case, are useful for vision 

 at very near range. 



A brief description of the eyes of spiders may not be un- 

 interesting in this connection, as a typical instance of the 

 manner in which the location of the eyes has been deter- 

 mined by the structure and habits of their possessors. " Spi- 

 ders which hide in tubes or lurk in obscure retreats, from 

 which they only emerge to seize a passing prey, have their 

 eyes aggregated in a close group in the middle of the fore- 



