222 
BIOLOGY: A. H. CLARK 
Proc. N. a. S. 
radial symmetry first to a symmetry in which the body is the same on either 
side of two planes at right angles to each other (as in the polyp giving rise 
to the strobila), then to a similar but much more pronounced symmetry 
of the same kind with a faintly indicated dorsal and ventral surface (as in 
the cestodes), later to a bilateral symmetry with well differentiated dorsal 
and ventral surfaces, the former usually somewhat more extensive than 
the latter (as in the annelids), and finally to a very pronounced bilateral 
symmetry with the dorsal surface always greatly in excess of the ventral 
(as in the crustaceans and insects). The two tentacles of the polyp from 
which the strobila develops persisted as the lateral extensions of the pro- 
glottides in the tape- worms (which sometimes reach a considerable length), 
and these reappeared as the parapodia of the annelids, and in their most 
perfected form as the legs and other segmental appendages of the arthro- 
pods. 
Strobilization comparable to the type characteristic of the coelenterates 
also occurs in another form. The units, instead of budding off in a linear 
series, are formed on the walls of the interior of a closed polyp which finally 
disintegrates, liberating all the enclosed units as free living individuals. 
These have a symmetry which is apparently bilateral, but always shows 
marked radial features. Such a development is found in the liver fluke 
and its allies, though today in no non-parasitic forms. 
This method of growth by which young are actually seen to be budded 
off from the interior of a closed sack developed from the ciliated larva, 
combined with the more or less extensive infolding within the gastrula 
mouth of part of the outer body wall as seen in the more specialized coelen- 
terates, furnishes a clue to the origin of the unsegmented animals. 
In those colonial coelenterates with division of labor the polyps are 
divided primarity into nutritive, reproductive, and "defensive," but the 
last are characterized by an abundance of cells containing an excretion and 
a coiled tubule, and some or other of them are constantly discharging the 
included liquid. They are probably fundamentally excretory and only 
secondarily adapted for defensive purposes. 
It is easy to imagine that in the early stage of some coelenterate type 
the infolding of the outer cell wall at the gastrula mouth was sufficient 
to bring within the inner cavity the region from which the new polyps 
are budded off, and to assume that there then were given off internally 
three buds, one of the nutritive (sack-like), one of the reproductive, and 
one of the excretory type, which developed into the perivisceral, the gon- 
adial, and the nephridial coelome. 
The so-called coelomate animals with unsegmented bodies, none of which, 
in sharp contrast to the segmented animals, have any true appendages, 
thus represent a colonial coelenterate in which the colonial development 
has been transferred within the original unit; and the development of 
