HETEROMORPHOSIS 155 
Five minutes later the foot reached the bottom of the test- 
tube (Fig. 34, c). The bending spread gradually to elements 
lying nearer the bead; as the foot could no longer advance 
vertically, it was pushed horizontally over the bottom of the 
test-tube; and at the same time the head, which until now 
had played no active part, was slightly raised (2:35 P. M., 
Fig. 34, d). The bending then passed from one element of 
the body to another, until the head was brought into an 
erect position (Fig. 34, e). Finally the entire animal righted 
itself so that at 1 o’clock it had the position shown in Fig. 
34, f). The whole righting process had therefore occupied 
an hour. The animal remained in this position for two days, 
when it crawled out of the test-tube. I have repeated the 
experiment many times, but always with the same result. 
5. If a Cerianthus is divided transversely in the middle, 
and both pieces are laid upon the wire screen, they work 
their way (often immediately after the division) through the 
screen with their aboral ends directed downward. If the 
head and foot of an animal are amputated, the middle piece 
may still show this reaction. When this occurs, the aboral 
end always bends downward, and works its way through 
the wire screen. Never have I seen the reverse occur— 
that such an animal assumes a position in which the oral 
pole is directed downward and the aboral pole upward. I 
wished to determine whether light or gravity had any effect 
upon the position of the new organs formed in these headless 
and footless animals when fixed vertically in sand, with 
their aboral ends directed upward; in no case, no matter how 
often I fixed the animals in an inverted position in the sand, 
did I succeed in retaining them in this position longer than 
two days. Nor did they remain with their aboral cut ends 
directed downwerd in a narrow test-tube the long axis of 
which stood vertically. In all cases they turned their oral 
poles upward. 
Digitized by Microsoft® 
