58 
G. H. Parker 
curvature contained 28 ommatidia, the anteroposterior 39, or 12 more 
than in the corresponding dimension of the previous stage. In a 
still larger and presumably older crayfish measuring 3 cm. in length, 
the number of ommatidia in the dorsoventral rows was 29, while in 
the anteroposterior ones there had been an increase to 43. In both 
these animals the growing edge of the retina was the anterior one, 
and it therefore seems safe to conclude that after the crayfish has 
been hatched the retina continues to grow only on its anterior 
margin. This method of growth probably also characterised the 
development of the retina previous to hatching; at least, such is the 
case in the closely allied genus Homarus^ and I believe we would 
not be far wroug in assuming that the anterior edge has always 
been the growing edge and that the posterior one represents approxi- 
mately the region from which the retina has grown ; in other words, 
the posterior edge is ontogenetically the oldest and the anterior the 
newest part of the retina. 
In dealing with the growth of the optic nerve, it is necessary 
to keep clearly in mind the two parts of which it is composed: the 
smaller dorsal portion (PI. 2 Fig. 38 n.d) and the larger ventral one. 
In a young crayfish whose length was 3 cm. the area of the ventral 
part of the nerve as seen in transverse section was about =2.3 times 
that of the dorsal portion (Fig. 39). In the adult these proportions 
are considerably changed (Fig. 40) ; for, while the area of the dorsal 
part has increased by about one third (from 1 in the young to 1.34 
in the adult), the area of the ventral part has nearly tripled (from 
2.3 in the young to 6 in the adult). This general increase has pro- 
bably been produced by two causes: first, the individuai fibres, like 
the retinal element, have simply each increased in calibre slightly, 
thus enlarging the whole nerve; and, secondly, the addition of new 
nerve fibres has certainly increased parts of the nerve. There is no 
evidence that new fibres are added to the dorsal part of the nerve, 
and I believe that the slight increase in that region is due entirely 
to the enlargement of individuai fibres; in the ventral part, however, 
the fibres have multiplied nearly three times, from 1291 in the young 
to 3430 in the adult; i. e., the increase in the number of fibres has 
been about in the same proportions as the increase in the area of 
transverse section. These observations lead to the conclusions that 
the optic nerve grows by the addition of new fibres to its ventral 
part, a unilateral method essentially like that in the retina, and that 
the dorsal part of the nerve may be regarded as ontogenetically older 
