ORGANIZATION AND GROWTH 209 
described this experiment in the first volume of my 
Physiological Morphology. During my last stay in Naples 
I repeated it, and found that in one case the polyps at a 
were formed about twenty-four hours earlier than at d. 
But this difference is not great enough to explain the differ- 
ence in time in all cases between the formation of polyps at 
the oral and at the aboral end, which not infrequently 
amounts to one or more weeks. 
4. Another possibility is conceivable. The substances 
necessary for the formation of polyps are formed in the 
stems of the Tubularia, and move in both directions, but 
the movement occurs first in the direction from root to oral 
pole (upward), and more easily than in the reverse direction. 
If this assumption were correct, then when we divide the 
stem ab as above, the polyp at the oral cut end d of the 
aboral piece bd ought to be formed distinctly earlier than at 
the aboral end c of the oral piece ac, because the formative 
substance of the polyp has to move in the first case in the 
direction from root to polyp (upward); in the second, in the 
reverse direction. 
I picked out eight healthy specimens of Tubularia from 
acolony, amputated root and polyp, and divided the remain- 
ing pieces ab by a transverse incision between cand d. The 
pieces ac were fixed in an inverted position, the pieces bd in 
an upright position in narrow glass tubes standing vertically, 
so that only the ends c and d were surrounded by water and 
could form polyps. After three days two polyps were formed 
on the d ends; upon the following day all the d ends carried 
polyps. On this day the formation of polyps only just 
began on the c ends in two of the specimens; not until four 
days later was regeneration complete in all the specimens. 
In a second experiment I used nine Tubularian stems. 
After three days all the d ends had polyps, while regenera- 
ation had only begun in two cends; only three days later 
Digitized by Microsoft® 
