Form Regulation in Harenactis attenuata 385 
certain that the ability of the oral end to form tentacles at all is 
much less in the rings than in other pieces; the time between 
the operation and the first appearance of tentacles is usually more 
than twice as long in the rings as in other pieces. 
But, on the other hand, in many of the tentacle groups the ten- 
tacles all appear at the same time and are of equal length whether 
the groups are situated on the line of union (Figs. 15 and 16, 18 and 
20) or wholly on the oral side of it (Figs. 28, 29 and 30). In 
general such groups are the most perfectly radially symmetrical, 
while those in which there is the greatest difference between ten- 
tacles on the oral and aboral sides are usually irregular (Figs. 8 
and 10). In the radially symmetrical groups with equal tentacles 
the old polarity has simply been more completely obliterated than 
in the others before the formation of the tentacles, so that no appre- 
ciable traces of it remain in the region of union after the establish- 
ment of the new polarity. 
These tentacle groups, then, show us various stages in the pro- 
cess of the obliteration of the old and the establishment of the new 
polarity. In the early stages of some rings, e. g., Figs. 6 and 11, 
the original polarity is still present in something like its original 
condition, for tentacles appear only on the oral side of the cut sur- 
face: later in these same pieces (Figs. 8 and 12) aboral tentacles 
develop. In many other cases the difference in size and time of 
appearance between the different tentacles is not as great, and they 
often become almost or quite equal in size in the later stages. 
And finally we come to such cases as No. V (Figs. 17 and 20) in 
which scarcely a trace of the original polarity is apparent in the 
tentacle groups, and No. [X (Fig. 29), where the original polarity 
and its effects are not at all visible in the groups: the only feature 
in such cases which can possibly be referred to the original polar- 
ity is the localization of the tentacle groups on the oral side of the 
line of union, although, as was shown above, this localization 
probably occurs only under certain experimental conditions. 
I believe that the union of the two ends with each other and the 
physiological effect of this union upon each part are the chief 
factors in obliterating or decreasing the original polarity in this 
region and so producing conditions favorable for the establish- 
