Form Regulation in Harenactis 85 
may call qualitative, certain quantitative features exist. Usually, 
for example, the rapidity of development or the size of oral or 
anterior structures decreases as the distance between the level 
from which they arise and the originaloralor anterior end increases, 
and the same is true, mutatis mutandis, for aboral or posterior 
structures. As was shown in the first paper of this series, the 
rapidity of tentacle-formation in Harenactis decreases with increas- 
ing distance of the level of restitution from the original oral end: 
in Tubularia (Child ’07c, ’07e) and in Planaria maculata and P. 
dorotocephala (Child ’06) similar differences exist, though in 
these planarians a complicating factor exists in the presence at the 
posterior end of a region physiologically specified as a new zooid. 
And finally, in cases of primary heteromorphosis quantitative 
differences between the two ends of the axis often exist, as in 
Harenactis, where the aboral tentacles always appear later than 
oral tentacles in an cesophageal piece, and not only that, but later 
than oral tentacles at the same level. Similar differences often 
appear in Tubularia and Planaria. 
Summing up what has been said, we find that the restitutional 
reactions show four distinct characteristics with respect to the 
principal axis: first, different reactions may occur at the two ends 
of the ‘axis of a piece (qualitative polar differences or heteropolar 
phenomena); second, in cases of primary heteromorphosis the 
heteromorphic reaction very commonly occurs more slowly than 
the “normal” one (quantitative polar differences); third, the 
character of the reactions in primary heteromorphoses may differ 
completely according as the pieces are taken from one or the 
other end of the axis (qualitative regional differences); and fourth, 
in most cases the rapidity or energy of the reaction characteristic 
of either pole decreases as the distance between the level from which 
it occurs in a given case and the pole where it originally occurred 
increases (quantitative regional differences). In considering the 
problem of what we call polarity we cannot neglect any of these 
phenomena, for all of them have to do with the axis. If we con- 
sider polarity with Driesch as a directive organization we cannot 
account for primary heteromorphoses, in which the two ends of 
the axis are alike except perhaps quantitatively. If, on the other 
