The Retina and Optic Ganglia in Decapods, especially in Astacus. 35 
by which such superposition images may be produced. So far as 
my own Observation s are concerned, I know of notbing that is in 
disagreement witb Exner's belief tbat each dioptric organ acts as a 
j)lens-cylinderff, an optica! contrivance that this author (cf. Exnee, 
91, pag. 1) has ably described and that seems perfectly capable of 
producing the requisite images. 
A question more directly connected with the present studies 
concerns the character of the retinal images in Astacus: are they 
apposition or superposition images ? The fact that the pigment plays 
so important a part in determining the kind of image in a given 
eye, and, further, that in Astacus it presents two extremes in its 
arrangement, one for bright and one for dim light, requires that this 
question be asked for both conditions. Since the region of the retina 
between the cones and the rhabdomes is for the most part devoid 
of pigment, a superposition image might be expected; but, for a 
satisfactory answer, evidence of a more conclusive nature must be 
had. This I bave tried to obtain in the following manner. 
A frozen preparation of a fresh retina, such as I bave already 
described (cf. pag. 29), was made for the distai end of the rhabdome, 
placed in the usuai way under the microscope, and illuminated by 
rays from a lamp some two metres distant. It will be remembered 
that in a superposition image each rhabdome receives light from a 
number of cones, whereas in an apposition image the light entering 
any rhabdome comes from a single cone. With this difference in 
mind, I placed directly below the preparation, i. e., between it and 
the source of light, a sliding diaphragm composed of a glass piate 
carrying a piece of tinfoil in which there was an opening a little 
smaller in diameter than a facet. By moving this back and forth 
I could illuminate either one cone or the whole retina. In using 
the apparatus two precautions were to be observed; first, it was 
necessary to bring the diaphragm to rest with its opening opposite a 
facet and not between facets. The correct position could easily be 
judged from the rhabdomes, for it was possible to place the dia- 
phragm so that only one rhabdome showed light. The second pre- 
caution was to place in the eye piece of the microscope a diaphragm 
with a very small opening so that the light that passed through the 
sides of the retinal preparation when fully illuminated could not 
reach the experimenter's eye and thus interfere with accurate Obser- 
vation. Having set up the apparatus, I attempted to decide whether 
a given rhabdome, after being illuminated by light entering it 
3* 
