166 



MOSAIC VISION. 



the difficulty, because, if any definite picture is to be 

 formed, the seositive rods, cones, or other structures 

 must lie in the plane of the image, 

 and this is not, in fact, the case. 



Dor suggested that the crystalline 

 cones are nervous structures, and cor- 

 respond to the rods of the vertebrate 

 eye (Fig. 79). He admits, however, 

 that, as a matter of fact, the image is 

 not formed at the anterior surface of 

 the crystalline cones.* 



And yet in his final summary, 

 having shown that the image is formed, 

 not at the anterior surface, but deep 

 down in the crystalline cones, he 

 expresses quite a different view, 

 compares the crystalline cone to 

 the vitreous body, and considers that 

 the true retina is to be found in an 

 envelope which surrounds the cone. 



Plateau j regards the mosaic theory 

 of Miiller as definitively abandoned, 

 but rather seems to have had in his 

 mind that of Gottsche. At least, he states that, accord- 

 ing to Miiller, the mosaic is formed by a number of 

 partial images, each occupying the base of one of the 

 elements composing the compound eye. This, how- 

 ever, is not Miiller's theory. 



* " La cornee avec sa convexite posterieure correspond a la cornee 

 et au cristallin des vertebres, le corps cristallin (avcc le soi-disant 

 corps vitre) et la fibre nerveuse qui s'y attache a la couche des 

 batonnets, enfin le ganglion optique a celles des couches de la re'tine, qui 

 sent compose'es des granulations, des cellules, et des fibres nerveuses." 



t " Eech. Exp. sur la Vision des Arthropodes." Bruxelles : 1887. 



Fig. 113.— One of the 

 elements of the eye 

 of a fly (afier 

 Gottsche). A;fc, Crys- 

 talline cone ; x, posi- 

 tion of the image; 

 s, rod ; &c, sheath ; 

 sew, outer sheath; 

 r, retina ; y, seat of 

 vision. 



