Eyes of Molluscs and Arthropods. 649 



as a spongy-like core, in which the longitudinal fibres with tlieir 

 cross branches may be seen with perfect ease. It is not within my 

 power to give. by means of drawings, an adequate idea of the marvel- 

 lous complexity, yet at the same time wonderful clearness and simplic- 

 ity, of this structure, which at once furnishes such a complete proof 

 that the seat of vision is located at that place , as it has seldom been 

 the lot of a Student to experience, in an equally difficult subject. Sec- 

 tions of another eye , prepared at the same time in this manner, show 

 essentially the same structure , except that many of the finest fibrillae 

 have become so strongly varicose as to disguise their real nature ; under 

 unfavorable optical conditions , a mass of fine granules are seen , with 

 comparatively few, distinctly visible, cross fibres. But even then a care- 

 ful examination would not fail to disclose the same general structure 

 that is so plainly visible in the dissected eyes. However indistinct the 

 cross fibrillae, at times, may be, the four central bundles always remain 

 visible. In cross sections of the hollow style, may be seen, with a 

 distinctness that leaves no doubt in the mind, very fine fibres radiating 

 from the axial bündle towards the periphery (figs. 128, 124 and 125). 

 The effect is similar to that seen in cross sections of the retinidia of 

 Pecten. 



If a cross section of the retinidium oi Mantis be compared with one 

 oiPenaeus, Area, or Galathea, a slight difPerence will be observed: In 

 the last three, the axial nerve bündle remains exactly in the central 

 axis, between the adjacent walls of the four cells. In Mantis, there 

 are four bundles , one inside of each cell ; this can only be explained 

 by the fact that , at the inner end of the calyx and indeed through- 

 out the w^hole length of the style , the axial walls of the four cells 

 have disappeared, and thus the way was left open for the fibres 

 to pass i n 1 the centre of each cell ; this is evidently a clear gain in 

 the distribution of the horizontal fibrillae, and may possibly account 

 for the absence of a crystalline cone. The wall of the style is surround- 

 ed by six nerve fibres, which appear in cross sections as so many 

 small dots (figs. 125 and 124). They may be followed as far as the 

 neck of the calyx, where they disappear, breaking up into numerous 

 smaller branches , continuous with those inside the calyx by means of 

 minute cross fibrillae. Towards the outer end of the ommatidium . the 

 retinulae are easily seen to consist of seven distinct cells, closely sur- 

 rounding the style; towards the inner ends, the space enclosed by the 

 pigmented axial faces of the retinulae, remains about the same, while 

 the enclosed style with the six nerve fibres becomes smaller and thus 



