The Retina and Optic Ganglia in Decapods, especially in Astacus. 31 
proximal ends of the groups of cone-cells, the presence of the single 
erect image just mentioned was easily verified, and, though its out- 
line could not be called sbarp, its general form could be easily 
recognised. It was, of course, only well seen in the central part of 
the retina; for, owing to the radial arrangement of the ommatidia, 
the sides of preparations made as I have described are always un- 
favorable for study. In this central region I never observed the small 
inverted Images claimed by Viallanes to be present at this level 
in each ommatidium. 
These observations are important in two respects; they show 
that, when the rhabdome is surrounded by pigment, its whole length 
can be penetrated only by very strong light, though its distal end 
can always be illuminated even by diffused daylight, and, secondly, 
they confirm the belief entertained by Müller, Grenacher, and 
Exner that the image in the compound eye is a single upright 
one for the whole retina, whose perceptive elements, the rhab- 
domes, receive each a single Impression. 
Granting this conclusion, it might fairly be asked, what, then, 
are the small inverted Images seen by so many investigators ? Tbe 
presence of these images under certain conditions is beyond question. 
When the corneal cuticula is carefully cleaned and properly mounted 
under the microscope, each facet produces a remarkably sharp, 
bright image of objects in front of it, even the comparatively flat 
faced facets of Astacus being no exception to this rule. These images, 
as Zenker surmised, can be made to occupy more proximal positions 
by filling the region in which they are formed with denser media, 
as the following observations on the facets of Carcinus maenas show. 
When the distal face of the cleansed corneal cuticula of this crusta- 
cean is immersed in water and the proximal concavity occupied with 
air, the image is seen at a distance of 9 (x behind the proximal face 
of the facet; with water in the proximal concavity it is formed at 
a distance of 17[i, and with glycerine, at 21p. I have made prepar- 
ations of the eyes of Musca and Carcinus as Gottsche and Vial- 
lanes did, and in both cases I saw the small inverted images described 
by these investigators. As is implied by Viallanes, these images 
are certainly produced by the corneal facets and displaced proximally 
by the denser transparent material of the retina, but that their for- 
mation is a normal process is by no means so certain. In both 
Musca and Carcinus^ the substance of the cones is so soft that when 
these bodies are freshly removed from the eye it flows like a thick 
