574 THE SENSES AND SENSORY ORGANS. 
the cone as parallel pencils p’ p’ and m’ m’. If, however, the 
lenticular surface of the cone is considered, these pencils will 
unite in focal points at p’ m’, Fig. 2, D, and the only effect 
produced by a second refracting surface 7 7, bounding a denser 
medium beyond it, will be to increase the convergence and 
shorten the focus, i.c., reduce the magnifying power of the 
system. 
Exner gives an elegant proof of the second inversion of the 
image by the cone [252], to which I must refer the reader who 
is curious in such matters. 
The Evolution of the Compound Eye.—For the above reasons 
it is clear that the lenticular function of the rhabdome is not 
necessary except when the crystalline cone is absent, or its 
focal length is greater than the distance of the retina behind 
it. In Lampyris the rhabdomes are not apparently lenticular ; 
in the noctuid Moths there is apparently a large lenticular 
rhabdome, and in all insects in which the cone is absent the 
lenticular rhabdome serves as a second refractive medium. 
We thus see the stages by which a euconic eye, the more 
primitive form of the compound eye, may have gradually 
become an aconic eye; the Moths form an intermediate link, in 
which both forms of refractive media are present. 
It is possible that, in the most primitive form of compound 
eye, that of the Phronimide, total reflection is largely con- 
cerned in the visual function; from the thread-like refractive 
structures of these animals it is not difficult to conceive the 
origin of refractive cones, such as those of the King-crabs and 
Crustacea generally. In some insects this kind of eye is but 
little altered ; in the Moths two lenses, a conical lens and a 
rhabdial lens, appear, with a segregate retina, whilst in the 
Diptera and Neuroptera the cone has disappeared, the retina 
has become continuous, and the only refractive media are lenti- 
cular; and this may be regarded as the most perfect form of 
compound eye. 
Such an origin of the various forms of compound eye from 
a simple type other than that which persists in the ocelli 
appears probable. The aggregation of simple eyes to form a 
