Eyes of Mollusca and Arthropods. 067 



traced back to an invaginate eye in whieh the retinulae also played 

 primary roles: it is therefore easy to determine what conditions are 

 essential. and what are the results of a simple moditieation in form; 

 secondly, by a comparison of the Molluscan and Arthropod compound 

 eye, betweeu whieh there is not the slightest genetie relationship, 

 we may possibly arrive at some conclusiou as to what elements of 

 the Arthropod eye are essential , and what secondary to vision ; or in 

 other words . what characters may be regarded as morphological , and 

 what analogous. 



A primitive ommatidium may be regarded as a eolorless (single? 

 or Compound) cell, provided with a terminal rod and an underlying 

 reflective part, and surrounded by a circle of pigment cells. When 

 the eolorless cell, or retinophora, of an evaginate eye assumes 

 a more important part, the pigment cells become modified into 

 circles, one surrounding the rod, and another the reflective part; e. g. 

 Area. An analogous condition prevails in Arthropods, where the 

 retinulae are modified into two principal circles; an outer one of 

 comparatively light-colored cells, surrounding the calyx , and au inner 

 row surrounding the reflective part, or pedicel; other pigment cells 

 may be present, but they show a less Constant and uniform arrangement. 



A layer of irregulär cells is quite widely present in the compound 

 Arthropod eye, around the bases of the ommatidia, just above the basal 

 membrane. They contain little or no pigment, but are often filled with 

 highly refractive crystals, milk- white in reflected, yellowish or eolorless 

 in transmitted, light ; they are often so refractive that when seen in the 

 latter manner they may appear nearly black. The outer row of retin- 

 ulae may be di vided into secondary circles , of whieh the very outer 

 one often contains crystals similar to those over the basal membrane 

 (Penaeus). The middle row of retinulae, the retinulae of Grenachek, 

 are most Constant in number and shape, seven being usually present; 

 four, and less often six, cells surround the calyx. 



The style of the retinophorae consists of a structureless tube whieh 

 towards its inner end often breaks up into numerous, striated, refractive 

 plates, the Striae running at right angles with each other in alternating 

 plates of different refrangibility. The faces of the plates are never pa- 

 rallel with each other. The frequent absence of the pedicels in 

 those Arthropods that can See well, together with the absence 

 of nerve fibrillae, indicates that they are not essential struc- 

 tures. It appears that the pedicel is more frequent in nocturnal, than 

 in diurnal, Insects. I must confess that I am very loath to make any 



