638 H. GRAHAM CANNON 
cenital rudiment in the two forms, and from its relation to 
the mesendoderm and ultimate fate, it would seem that the 
differences are of small significance and that the two structures 
described as genital rudiment are really homologous. 
A comparison of the mode of development of 5. vetulus 
with that of Polyphemus as described by Kihn reveals 
some very close analogies. ‘Text-fig. 1 (a) shows a diagram 
of the ventral mass of 5. vetulus before the formation 
of the mesendoderm. In the posterior region are the large 
primordial germ-cells bordered laterally and posteriorly by 
ectomesodermal cells. In front is the group of yolky cells 
which are mesendodermal. The mner layers of this latter 
cell-mass spread out over the anterior part of the blastoderm 
as mesodermal cells, and from the outer layer is developed the 
very definite mesendoderm. While this is growing backwards 
mesoderm cells are still being proliferated inwards at the anterior 
and lateral edges of this group and possibly at the posterior 
edge. The fact that these latter cells originate by proliferation 
of cells at the edge of this mesendodermal group, together 
with the fact that they form mesoderm distinct from the 
mesoderm included in the backwardly growing mesendoderm, 
suggests that possibly they are a separate source of mesoderm, 
that they are ectomesodermal cells—a continuation forwards 
of the ectomesodermal cells which are formed at the periphery 
of the genital rudiment. If this were so, an analogy might 
be drawn with the development of Cyclops as described by 
Urbanowicz (18), where he states that larval mesenchyme 
arises from cells surrounding the primitive endoderm cell 
while the secondary mesoderm arises from the gut. The more 
recent work of Fuchs (4) on Cyclops has, however, failed 
to confirm the findings of Urbanowicz, and has, on the contrary, 
demonstrated an extraordinary resemblance between the 
development of Cyclops onthe one hand and Polyphemus 
and Moina on the other, in neither of which is there any 
larval mesenchyme as distinct from secondary mesoderm. 
But in S$. vetulus when the mesendoderm is growing 
backwards, although its hinder end is very sharply separated 
