No. 420.] REGENERATION IN THE EGG. 971 
several other instances of.the same phenomenon in hydroids, 
and has made the additional and important discovery that in 
these forms the external conditions may determine the kind 
of regeneration that takes place. However, even in these 
forms differences are present in the two ends- of a piece, for, 
even in the most favorable forms it is found that, as a rule, a 
new head or hydranth forms sooner on the anterior end of the 
piece than on the posterior, and from the posterior end of the 
piece a posterior end is more easily developed than from 
the anterior end. In other and higher forms the factors that 
determine the kind of regeneration seem to come from within 
the piece itself. In some of these forms we also find occasion- 
- ally that the new part may be different from the part removed 
(heteromorphosis), and in some cases, as in Bonnet’s fresh- 
water worm, a posterior end develops on an anterior surface. 
Other cases of the same sort are also known; for instance, in 
the earthworm, Allolobophora, a short posterior piece (less 
than half the length of the worm) generally produces at its 
anterior end another tail; a very short piece of the head-end 
may produce a new head on its posterior surface, and the same 
holds for short pieces of the head of a planarian.! 
Sachs has vigorously combated this idea of polarity, that he 
refers to as a metaphysical notion of the morphologists, and 
offers as a substitute his hypotheses of formative stuffs. 
have attempted to show that the idea of formative stuffs is 
insufficient to explain the phenomena of regeneration, and at 
present I see no escape from the idea of a polarity of some sort 
existing in the organism. On the other hand, we cannot be too 
careful in clearly defining what we intend to include in this 
idea of polarity, for I think we can see even at present that 
we must form a conception of the polarity different from 
that in which the term is used by the physicist. 
The term “polarity” when used in connection with phenomena 
connected with the magnet means that there is a difference 
in the two ends of the magnet. The idea also involves the 
further one that any piece of the magnet also shows this same 
1 We may call that kind of heteromorphosis in which the new part is more or 
less a mirror figure of the old part “ amphimorphosis.” 
