234 Prof. Dr. Fr. Ycjdovslcy on 



connection between the intlividual cones ; they are isolated and 

 mostly situated at the periphery of the pigment-cells. The 

 total number of the cones in the specimen represented amounts 

 to only fourteen, and the variable form and size of the com- 

 ponents can be well seen in the drawing. 



The number, form, and size of the crystalline cones vary, 

 however, not only in different individuals, but also in the 

 right and left eyes of one and the same animal. The eye 

 just described belongs to the left side of the head (fig. 17). 

 In fig. 18 is shown the eye from the right side of the same 

 animal. In the centre of the pigment-cells there is only one 

 crystalline cone ; the others are situated at the periphery of 

 the pigment-cells, irregularly scattered, and two even lying 

 not far from the frontal margin of the head. In this case 

 there are only eleven crystalline cones of the most diverse 

 size and form. 



The structure of the crystalline cones here reproduced is 

 certainly very peculiar and difiicult to correlate with the 

 details of an ordinary compound eye of Gammarus. In each 

 cone there is a dark, finely granular, sharply defined matrix, 

 at whose margin, when in a favourable position, a little body 

 can be seen which I am inclined to consider as a nucleus. 

 In this matrix an irregular number of smaller rounded bodies 

 are present. In the simplest case there is a large central 

 sphere and with it one or more little spheres. There can, 

 however, be such an increase of both kinds that the matrix 

 appears to be filled with vacuoles. These inner vacuolar 

 spherical bodies are filled with a homogeneous slightly re- 

 fractive substance which is only stained by carmine in a very 

 feeble and diffuse way. 



In the above-mentioned serial sections into which one 

 specimen was cut, I have unfortunately found little explaining 

 the connection of the appearance in side view of the crystal- 

 line cones, as described, with the nervous organization. In 

 fig. 19 one such section is reproduced. Nothing is here to 

 be seen of the nerves. The cuticle of the head (c) stands 

 off a long way from the eye. The hypodermis {hp) forms a 

 thin layer wliich contains only the regularly placed nuclei. 

 Underneath there are three groups of large pigment-masses 

 without nuclei, between wliicli occur rounded vacuolar bodies. 

 These are the rudiments of the crystalline cones, and they 

 are not nearly so refractive in these sections of 5 /a in thick- 

 ness as in the preparations showing the animals from the 

 side. A direct connection between the outer larger pigment- 

 masses and the deeper layer of smaller pigment-cells {rt) 

 does not exist in the serial sections. 



