APR. 4, 1923 CLARK: ORIGIN OF VERTEBRATES 131 
there is no budding at all, there can be no doubt that fundamentally 
the coelenterates are phanerogam-like colonial animals. 
There is nothing that can be assumed to connect the sponges with 
any other animal type except, perhaps, with certain of the Protozoa. 
It is evident that the gastrula stage in the development of the bilateral 
animals cannot represent any sponge-like progenitor. It is possible, 
however, to interpret the bilateral animals in terms of the colonial 
coelenterate. Indeed, it is not possible to interpret them in any other 
way, for any other explanation of their origin would assume the 
presence of a fundamental bilateral tendency, an unknown and un- 
determinable variable not common to all animal types. 
If a colonial coelenterate with radially symmetrical polyps should 
develop a persistent defect in the ontogeny whereby the units became 
bilaterally symmetrical, bilateral animals of four main types would 
at once appear: 
1. Bilateral animals in the form of a linear more or less unified 
colony. 
2. Bilateral animals in which the colony formation was inverted, 
the budding of the new elements taking place within the original unit. 
3. Bilateral solitary animals each representing a dissociated coelen- 
terate unit; and 
4. Bilateral animals with the colonial habit, though independent 
of each other. 
These four main types, between which there would be numerous 
intergrades, all represent definite types occurring among the coelen- 
terates themselves, and therefore none of them can be said actually 
to represent anything new in animal structure other than the novelty 
consequent on the developmental defect which resulted in the loss of 
the radial and the assumption of the bilateral body form. 
Among the animals of today all four of these main types are 
represented: 
1. The tape-worms or segmented cestodes form a linear colony of 
continuous growth so like a partially unified strobila as to leave little 
doubt of the fundamental similarity of type. The scolex of the ces- 
todes is radially symmetrical, but the proglottides are strongly flat- 
tened and bilaterally symmetrical, though the difference between the 
dorsal and the ventral surface is but little marked. 
2. The flukes have a peculiar development which is essentially 
similar to strobilization, except that the buds are formed within the 
original unit instead of in a linear series. 
