A STUDY IN POLARITY 169 
normal for the regeneration from such oblique surfaces, for in- 
stance, approximately mid-way between the potential head and 
tail. But, since the new material along the posterior cut fused 
with that along the anterior cut to form a head, the pharynx 
was forced to take a direction practically perpendicular to the 
original long axis of the worms, instead of the usual angle of 
about 60° to the old median axis—the position which obtains 
during the early stages of regeneration of triangular pieces which 
retain their original polarity. 
Two other anomalous specimens must be mentioned. Two 
pieces cut from the pharyngeal region of different worms failed 
to produce heads, though they lived some 20 days and each | 
regenerated a pharynx. Morgan (’98) found that in an exces- 
sively acute triangular piece, whose acute angle was at the most 
anterior part of the piece, the contraction of the tissue after cut- 
ting caused this sharp point to bend over and fuse with the side, 
thus preventing regeneration at the most anterior portion of the 
piece. Similar headless forms were obtained from cross-pieces 
in which a notch was made on the anterior surface, the fusing 
together of the edges of this notch preventing regeneration of a 
head. Child (15) produced forms of varying degrees of head- 
lessness by allowing pieces to regenerate in various solutions 
which retarded metabolism. The two headless specimens in 
my experiments did not arise in any of these ways, but both 
developed from triangular pieces of the same size and appearance 
as those which regenerated heads, and were treated in exactly 
the same manner as the others. The history of one of these 
pieces is shown in figure 12. <A similar headless specimen re- 
generated in one instance from a piece cut from a planarian as 
in figure 13. These three headless worms were very sluggish, 
and although each had a pharynx, attempts to feed them were 
unsuccessful. 
Morgan (98) and Child (15) have both commented on possible 
causes for the peculiar regeneration of the triangular pieces 
which develop heads at right angles to the long axis of the worm 
from which they are taken. Morgan (’98, p. 374) says, “The 
