342 
POPULAR SCIENCE REVIEW. 
cut away tlie apex, and attach to the cut surface a curved trans- 
parent rod, all the rays which previously came together at the 
apex of the cone would penetrate into the rod, and he conducted 
further on in this rod, just as light-rays can he carried on in 
the interior of curved glass rods. 0. Schmidt has himself 
made experiments of this kind. Some light will, of course, in 
general he lost in this process. This crystalline cone with the 
rod behaves now exactly like the curved crystalline cones of 
Phronima. Let us suppose that it conveys the light received 
from the direction a to the spot of the retina a, and then con- 
sider the neighbouring crystalline cone b. This may he so placed 
as regards its first portion and also its corneal facet, that, in 
accordance with the rules applying to the straight cones, it 
receives light from the direction b, which is close to the direc- 
tion a, and then it will convey this light further on, however 
circuitous its course may he ; and if its termination lies near a , 
it will deposit its light close to this upon the retinal spot b. In 
this way the different directions of space will be represented in 
the same order upon the retina. All that is necessary, there- 
fore, for the production of a ‘ mosaic picture 9 is that each 
crystalline cone shall serve principally light of a particular 
direction, and that their extremities shall he placed in an order 
corresponding to this direction. According to the figures given 
by 0. Schmidt these two conditions are perfectly fulfilled. 
Whether the light makes a circuitous route in the crystalline 
cone is a matter of subordinate consequence — nay, even the 
arrangement of their ends which are turned towards the light 
is quite indifferent, the direction of the axes of these ends is 
alone important. Could we seize an insect-eye with long fili- 
form crystalline cones by the retina, and disarrange the indi- 
vidual cones, each with its cornea, as one can disarrange the 
threads of a tassel, we should obtain a ‘ mosaic picture/ in spite 
of all curvatures and the displacement of the free ends, pro- 
vided always that the first part of the cone, with its corneal 
facet, had the same direction in space after the displacement as 
before. 
As regards the absorption of laterally falling rays in the 
surrounding pigment, the curvature of the crystalline cone may 
doubtless cause many rays not to he absorbed which would 
otherwise he absorbed, and vice versa ; the picture would thus 
lose somewhat in completeness, but why it should he destroyed 
is not easy to see. 
Last year there appeared a monograph upon the visual per- 
ceptions which are performed by the facetted eye, by JNbtthaft.* 
The author occupied himself with the question of the sharpness 
* Abhandl. der Senckerib. naturf. Gesellsch. Bd. xii. 
