The Eetina and Optic Ganglia in Decapods, especially in Astacus. 59 
than the ventral part. It is probable indeed that the dorsal part 
represents what may be called the embryonic optic nerve, i. e., that 
part of the nerve formed during embryonic development, a time when 
individuai fìbres are less likely to gain mesodermic sheaths than 
later. This Interpretation, though somewhat gratuitous, does not 
clash with any facts known to me about the optic nerve, and 
offers an explanation for several peculiarities otherwise diffìcult to 
nnderstand. 
In this connection it is of considerable importance to observe 
the relation» between the various parts of the retina and of the optic 
nerve as established through their nervous connections. These 
connections have already been described (cf. Fig. 59), and it need 
only be mentioned bere that the fìbres which leave the part of the 
retina ontogenetically oldest (green) pass into the oldest part of the 
optie nerve, and, conversely, fìbres from the ne west part of the retina 
are connected with fìbres in the newest part of the nerve; in other 
words, the retina and the optic nerve have increased hand in band. 
It is also of interest to observe that that part of the optic tract which 
is ontogenetically the oldest presents a very different condition from 
the rest. As can be seen in the diagrammatic fìgure 59, an impulse 
starting from the newer parts of the retina (red, blue, or black) 
would necessarily pass over fìve neurons in reaching the brain, whereas 
one Coming from the older part (green) would necessarily pass over 
only three neurons. In Branchipus an impulse may reach the brain 
from any part of the retina by passing over three neurons. This 
primitive condition is reproduced in Astacus only in the onto- 
genetically oldest part of the optic tracts, a position, however, in 
which, if a primitive condition occurred at ali, one would expect to 
fìnd it. 
The essential similarity in the method of growth followed by 
the retina and the optic nerve suggests at once the idea that the 
intervening ganglia probably likewise follow the same lines of 
growth. In fact by tracing the course of the newest fìbres it is 
comparatively easy to state which is the growing end of most of the 
ganglia. Thus the fìrst ganglion (cf. Fig. 59) grows like the retina 
anteriorly, the second posteriorly, and the third again anteriorly. 
The direction of growth in the fourth ganglion is less easily defìned, 
and the most that I can state is that the growth must be in a ven- 
tral direction, at least where the optic nerve emerges. 
In Branchipus^ as might be expected, the growth of the optic 
