678 JUSTUS CARRIERE. 



are actually four ganglion layers successively in the ommatophor 

 and between each a crossing of fibres takes place. 



If one considers the ordering of the eye-units in Musca as 

 the one extreme of the form of the fan-shaped eye, the other is 

 found in the Cladocera, The eye approaches here to the 

 spherical form and exhibits its most developed form in Lep to- 

 ri ora hyalina, where the spherical eye is carried freely at the 

 front end of the body like a lantern on a thin stake. 



In a section which bisects the eye in the median plane (PI. 

 XLV^ fig. 2), there lie about twenty-four eye-units set radially; 

 the strongly pigmented retinulse lie quite closely to one another, 

 so that in the transverse section the thin rhabdoms appear like 

 clear points on a black ground. 



Tlie retinulae do not reach with their points quite to the centre 

 of the globe, but end so as to leave empty a little central space. 



The crystal cones are slender and considerably longer than 

 the retinulse are composed of five segments and do not lie very 

 closely together, and at the peripheral end show distinctly their 

 limits (a) whilst they fuse more and more intimately together 

 towards the cone-point {b, c, d). Their transverse section is 

 accordingly here rosette-shaped, there circular. 



At about the half height the cone shows a pyriform swelling 

 which stains intensely with colouring reagents. 



The crystal cone is surrounded throughout its length by a 

 number of cells, the " Umhiillungsschlauch " (investing tunic) 

 of Leydig; at the outermost end they form around the base of 

 the cone a sack-shaped envelope hanging together; in which 

 in one plane of section are visible four nuclei («), so that their 

 number may be estimated at five, and this part of the envelope 

 with the nuclei may be regarded as the remains of the crystal 

 cells (Semper^'s nucleus). 



The thinner part, of the cone is surrounded by long, narrow 

 cells, the nuclei of which are somewhat irregular. I am not 

 able to say if the nuclei which lie close round the cones before 

 the beginning of the pigmented retinula {d and h) belong 



