2 (> 
rorULAR SCIENCE REVIEW. 
really reached the posterior surface of the crystal-like body; 
uor has any anatomist since his time doubted the nerve charac- 
ter of the fibre behind it. The various form of this fibre (as 
before described) was not, however, known or clearly demon- 
strated until Gottsche first, and after him Leydig, described the 
peculiar swellings, ridges, and knot-like expansions of the nerve 
so unlike anything hitherto observed in nerve structures. Nor 
was it until after long-continued and laborious investigations 
that Leydig finally expressed his opinion that the whole ap- 
paratus Avas nothing more than a peculiar modification of 
terminal nerve fibre, and, in fact, the homologue of the rod or 
cone in which the fibres of the vertebrate retina end. These 
cones or rods of the vertebrate retina are exceedingly minute 
in the mammalian eye, but in the lower vertebrates (e.g., frog, 
fish, &c.) are much larger and coarser. In the insect eye they 
are not only of a greater size and much more easy to prepare 
and examine, but they are also enclosed in separate sheaths, 
and have additions (such as muscle fibrils and tracheal tubes) 
which render them at first sight very unlike the rods and cones 
of the vertebrate ^retina. Nevertheless they agree in being 
terminal extremities of the nerve fibre, as also in the peculiar 
transformation of nerve substance into a highly refractive and 
transparent matter. They are, therefore, as much “ percipient 
elements ” as are the rods and cones of a vertebrate retina ; and 
they further agree in this, that they are placed just where the 
images formed by the corneal facets fall. In the first part of 
this article, we alluded to the fact that true optical images were 
formed by the corneal lenses, and that these images must ne- 
cessarily be produced in the plane immediately behind the 
cornea wliere the crystal-like expansions of the terminal nerves 
are situated. There is, therefore, no optical necessity for a 
second formation of images, but it is none the less certain that 
in this crystalline nerve end a series of refractions must occur. 
In the vertebrate eye, images formed by the lens are trans- 
ferred to tlie back of the globe, where the rod or cone ends of 
the retinal filn-es are disposed so as to meet the surface on 
whicli tlie image is formed. This disposition is similar in 
]>oth kinds of eye ; the ends of the fibres are opposed to the 
]>ictnre, and the direction of the bacilli or rods is radial to the 
centre of the eye. Hut, in the one case, tlie images and the 
“ j»ereif)ient bacilli ” are in the front of the eye — that is, vision is 
directly forwards — whilst in the other case the images and 
j>ercij)ient bacilli are at the back of the eye, and each nerve 
fibre entering the eyeball from behind turns back upon the 
inner surfat-e and looks towards the bottom of the eye; that is, 
visifuj is dirifctcd backwards. This inversion of sight is a 
special characteristic of the vertebrate eye. 
