302 PERCY SLADEN TRUST EXPEDITION. 
The difficulty of deciding these questions is made greater by the small number of 
specimens usually obtained: in only one case in the present collection was a large 
amount of material (Antipathes ceylonensis) brought up. As a rule only a single 
specimen occurs, often to be made the type of a new species, which, if a sufficient 
number of specimens were available for comparison, might easily prove to be one of a 
eradational chain of forms. 
The action of external influences on another branch of the Ceelenterates, the corals, in 
producing various types of growth from a single species, has been shown recently by 
Dr. F. W. Jones, whose evidence serves to support the view that similar causes are pro- 
ducing similar results in the group under consideration. It is unfortunately impossible 
to test the problem in these forms in the same way, ?. e. by watching their growth, so 
that we cannot gauge the extent to which these external agents work. Nor, from 
paucity of material, can we say whether the differences are of the nature of mutations, 
in which case we could regard them as species, or whether they are continuous, 7. e. with 
connecting links binding the forms together as variations of one species. Caution must 
therefore be exercised in using as an index of specific rank slight differences in the shape 
of colonies which otherwise do not greatly differ from one another. 
(2) SPINEs. 
In the past species have been determined frequently by means of the spines only, 
especially when dried specimens alone were available for study. In some cases the size, 
shape and arrangement, as well as the presence or absence of secondary spines (in the 
Indivise) may be of sufficient value to separate forms from one another. At the same 
time it must not be forgotten that spines are manufactured by the living tissues for their 
support, and that any cause tending to mould the tissues one way or another will through 
them react on the shape of their supporting elements, the spines. Until, therefore, we 
know more about the life-history we must be as cautious in using spines as specific 
characters as in the case of method of growth. 
It has often been pointed out by authors that the spines can vary greatly within the 
limits of a single colony, so much so that if fragments of the colony were to be collected 
and examined separately, using the spines alone as a guide, they would be classed as 
separate species. Again, it often happens that species which undoubtedly differ in other 
characters, even to the extent of belonging to different orders, have spines indistinguish- 
able from one another both in size and arrangement. 
Nevertheless there are cases where the spines are distinguishing features for want of 
any better guide, and must be used in describing a species. 
(3) Potyes. 
Setting aside the broad division of the Antipatharia into three groups according to the 
number of the mesenteries, i. e. Leiopathes with 12, Cladopathes with 6, and the rest 
with 10, there are in the latter three main types of polyp. These are: the round or oval, 
as in Antipathes, Cirripathes, and Stichopathes; the divided, as in Schizopathes and 
Bathypathes; and the elongate type, halfway between the two, such as is found in 
