THE EYE AND OPTIC TRACT OF INSECTS. 
241 
We cannot, however, trace the same layers of nerve-cells, 
ganglion-cells, &c., layer for layer, in Arthropoda, that we 
find in the retina of Yertebrata, any more than we can trace 
them in Cephalopods, in Alciope, or in Pecten. All that we 
can say is that there is present in all animals with highly- 
organised eyes certain complicated nervous structures situated 
between the nerve-end cells and the brain which are probably 
for the purpose of elaborating and combining the sensations 
received by the nerve-end cells. 
How much, then, of the optic tract of insects should be 
considered to belong to the retina ? It seems to me that to 
be consistent we should consider all the nerve structure lying 
between the crystalline cone-layer and the true optic nerve to 
be analogous with the retina of other animals. 
According to my view, in fact, the retina of insects consists 
of the retinulse, peri-opticon, epi-opticon, opticon, and all the 
intermediate nerve-tracts. 
Both Ciaccio (4) and Berger (2) considered the retina proper 
to end at the region of the decussating fibres ( N.f .). If this is 
the case the decussating nerve-fibrils should represent the optic 
nerve, which they do not, as Berger himself has shown. 
The true optic nerve lies between the opticon and the 
cerebral ganglion, and it is this which is so enormously 
elongated in the podophthalmatous Arthropoda. 
It may seem strange at first to have to consider those 
structures in the eyes of Arthropods, which have hitherto 
been somewhat loosely described as “ optic ganglia,” to be 
really part of the gigantic retina ; but if the nervous layers of 
the retina are really for the purpose of combining and elabo- 
rating the impressions received by the nerve-end cells, as I 
suppose they must be, it is only natural to find them larger 
and more complicated in the Arthropod eye than in the 
Vertebrate eye. 
In the Vertebrate eye the nerve-end cells are, comparatively 
speaking, extremely small and very closely packed, and con- 
sequently the nervous effort of translating the sensations of 
the nerve-end cells into a picture is very much less than it is in 
