516 T. A. STEPHENSON 
be much less safe than in the case of Gonactinidae or even 
Ptychodactidae. 
The Actiniaria as freed from extraneous skeletonless corals 
show general tendencies towards more complex individuality 
rather than towards colonial development, towards a special 
development of musculature in some way or another, towards 
different histology and on the whole more activity. They go 
in for expression of permutations and combinations of various 
characters, leading to great diversity—this diversity affecting 
differences among polyps, whereas it is perhaps more connected 
with variation of skeleton and colony-form, among corals, 
which may to some extent be compared with the Aleyonaria, 
although of course the latter much surpass them both in 
uniformity of the individual and diversity of the colony. 
§ Bt. 
The discussions so far have dealt with curious forms which, 
whatever their fate, are special cases, coming outside the 
main mass of anemones. ‘Those that follow are concerned 
with forms the general position of which is fairly clear, i. e. they 
all come under the main tribe (Nynantheae in the sense taken 
on p. 540) of the sub-order Actiniaria, excluding Edwardsians, 
Zoanthids, Gonactiniids, Ptychodactids and corals whether 
hard or soft—or to put it another way, they are presumably 
the descendants of a muscular Halcampa-like stage 
(cf. Text-fig. 8) with ciliated tracts on its filaments. Among 
these forms there seem to be four main sets which can be 
followed, and in the following sections the exceptional sets 
will be considered before the majority-forms. 
§F. The Ilyanthidae. 
There has been a family Ilyanthidae in use for a long time 
(Actinies pivotantes’), for the more or less vermiform 
creatures with no adherent base. It has been subdivided 
somewhat arbitrarily—that it needs subdivision is not in 
question, but how to doit. Although, however, we are obliged 
