Form Regulation in Harenactis attenuata 371 
group bin No. V (Fig. 17, 6; Fig. 19). In the second series in which 
tentacles arise wholly on the oral side of the line of union, two 
groups with odd number of tentacles occur: these are group c 
in No. VII (Fig. 24) and group d in No. IX (Fig. 29). Evidently 
then, these cases are not common: they are probably the result of 
special conditions which determine the delay or inhibition of a 
tentacle, perhaps in some cases of more than one. In the case 
shown in Fig. 19, for example, the mesenteries on the two sides of 
the line of union do not correspond in position because of irregu- 
larities in closure, and the number of tentacles is determined by 
the number of intermesenterial spaces involved in the region of 
growth. Concerning conditions in the other two cases nothing 
definite can be said: probably in these also merely incidental fac- 
tors are concerned. 
The cases described were arranged so far as possible to show 
the gradation from groups of tentacles irregular in number and 
form (No. I, Fig. 8; No. II, Fig. 10) to the other extreme of sym- 
metrical discs (No. IX, Fig. 29). 
The question now arises as to whether these groups of tentacles 
are to be regarded as cases of heteromorphosis. ‘The cases in the 
first series (Nos. I to V) in which the tentacles arise in part from 
what was originally the aboral end of the body-wall are certainly 
heteromorphosis so far as these aboral tentacles are concerned. 
As regards the cases in the second series, where the tentacles 
are all formed on the oral end of the body-wall, polar heteromor- 
phosis in the usual sense certainly does not occur, since aboral 
structures do not arise from what was originally the oralend or 
vice versa. On the other hand, a given region of the oral end of 
the body-wall which originally represented only one side of the 
body produced structures characteristic of all sides. Properly 
speaking this is as truly heteromorphosis as are the polar reversals. 
In these cases the symmetry of parts of the region involved in the 
formation of the tentacle groups must have undergone reversal 
since it produces parts characteristic of the opposite side of the 
body, 1. e., a radial heteromorphosis has occurred in these cases. 
In other words in these rings changes of polarity and symmetry 
occur which lead to the formation, either on both sides or on the 
