1118 General Notes. [ November, 
the manner in which the primitive cumulus is formed at the 
germinal pole of blastodermic vesicle of Arthropoda indicates, it 
seems to me, taking into account the fact that the mesoblast is 
split off from the lower side of the neural plate, that the meso- 
blastic somites are here formed in essentially the same way as in 
the Chordata. The invagination or folding in of the germinal 
area, in insect embryos to form the amnion, at first posteriorly 
and at the sides, or according to the plan just the reverse of what 
holds in the formation of the amnion in the endocyemate types of 
Chordata, is to me conclusive proof that concrescence of the lips 
of the primitive elongated blastopore, or archistome, has taken 
place; for, in order to effect this sort of an invagination of the 
embryonic area the head end must for a time remain fixed, while 
the tail, continuing to grow in length, is thrust into the yolk, as 
in Calopteryx, carrying the amniotic limb of the blastoderm be- 
fore it. It also seems that paired cavities soon appear in the 
mesoblastic somites underlying and derived from the epiblast, as 
above described in arthropods. I therefore see no very essential dif- 
ference in the method of development in the two types. In boti 
it is obvious that a portion of the archenteric walls of the elon- 
gated archistome has given rise to the mesoblastic somites, by a 
process which differs in no respect from, but agreeing even in its 
abbreviation with that which takes place in Branchiostoma directly 
from the sides of the archenteron. 
We now come to the consideration of the most important part 
of the archistome-theory, namely, that portion of it which deals 
with the genesis of the limbs and their musculature. The readi- 
ness with which the view that the tentacles of an actinozoan 
ancestral form gave rise to the integument and musculature of 
the paired limbs of the Bilateralia is reconcilable with all the facts 
of embryology, is very remarkable. As is well known, the ten- 
tacle$ of Actiniz consist of an outer layer of epiblast into which 
a hypoblastic lining is thrust from the paired lateral gut-pouches. 
If the gut-pouches of the actinian were now shut off from the 
archenteron we would have mesoblastic somites developed and 
structures formed which are exactly recapitulated in the develop- 
ment of the Arthropoda. Thatis, the outer layer in the budding 
appendages of the embryos of the latter, which grow out from 
a segment, are constituted of the same two layers, the outer 
_of which gives rise to the hard, chitinous joints, and the inner to 
the muscles which move them. 
In the development of bilaterality through the actinozoa the 
_ circle of tentacles would be drawn out into an ellipse, or so as to 
enclose an oblong space surrounding the archistome. This 
would bring the primitive appendages, after a free existence had ` 
en assumed by the supposed ancestral actinozoan type, into 
t the position in which they grow out in arthropod em- 
around the archistome or furrow in the neural plate. The 
