ARTIFICIAL PAKTHEXOGFXESIS IN ECHINUS ESCULENTUS. 545 
of the few larvaj we separated in this manner to passthrough 
metamorphosis, although they remained alive for several 
mouths. 
This method would seem to offer considerable advantages 
over the other methods we have tried, if the difficulty of 
washing the eggs thoroughly could be properly overcome. 
It can hardly be said to have been given a fair trial by us 
this season, on account of so much of our material this year 
being unfavourable. 
VI. Conclusion and Summary. 
Contrary to the experiences of Delage (5), we have always 
found that the parthenogenetic larvae are readily distinguish- 
able from the normal ones. This is most marked in the length 
of the arms and other minor featui*es. Comparison of fig. 1, 
which represents a normal larva of E. esculentus, with 
that of fig. 8 brings out this point clearly. These two larva^ 
are of the same age approximate!}’ as regards development. 
That repi’esented in fig. 1 is somewhat more advanced than 
that of fig. 8. The arms of the larvae shown in fig. 8 are a 
third longer than those of fig. 1. The age of the larva of 
fig. 1 is twelve days, while that of fig. 8 is eight days. This 
illustrates the second point of difference between the two 
kinds of larvie — their rate of growth. 
In the later eight-armed condition, fig. 4, which represents 
a parthenogenetic larva twenty-five days old, this slender 
long condition of the arms is not so marked as in the earlier 
stage shown in fig. 8. At this stage the growth of the par- 
thenogenetic larvae is much slower than that of the normal 
ones. This is again brought out by a comparison of figs. 4 
and 10. Fig. 4 represents a parthenogenetic larva twenty- 
five days old, while that of fig. 10 represents a normal larva 
twenty -two days old. It will be seen that the lai-va of fig. 10 
is far in advance of that of fig. 4. The Echinus-rudiment in 
that of fig. 10 is well formed, while that of fig. 4, if present, 
is hardly distinguishable. 
