22 THE ECHINODERMS OF TORRES STRAIT. 
There remain 18 names which designate recognizable forms of Ophiocoma, but how 
many of these are really valid species has yet to be determined, and in some cases at least 
can be determined only on the reefs, where the individual diversity in a definite area under 
identical conditions can be carefully studied. As an illustration of the usefulness of such 
work, I beg to call attention to the observations made at Mer on erinaceus and scolopen- 
drina. Such study is particularly needed on the highly variable brevipes, of which at least 
10 forms have been given names. When at Mer, I was ignorant of the great diversity 
shown by this species, and I collected two forms under the supposition that they were 
quite distinct, and later, at Hilo, Hawaii, I collected a third form, not even suspecting it 
was brevipes. Only careful observation on the reef can determine the true relation of these 
forms to each other, but I am using varietal names to designate them in the present report. 
In attempting to draw specific lines in Ophiocoma, one soon discovers that the shape 
of the oral shields, upper and under arm-plates, and the arm-spines is almost worthless, 
for the forms assumed by these various parts is subject to extraordinary individual diver- 
sity, associated with a certain amount of growth-change. The number of tentacle-scales 
furnishes a character of great value, though in certain cases one is sorely puzzled to know 
how much weight should be attached to it. Yet I have rarely seen an Ophiocoma of which 
the true number of tentacle-scales could not be positively asserted. Some individuals of 
species having only one tentacle-scale show two scales on some or even on many basal arm- 
joints, but that one is the true number does not admit of question; the presence of two 
seems to be clearly a case of reversion. Yet it does not follow that having a single tentacle- 
scale is an important or even a valid specific character (see p. 128, under schenleinii). The 
number and arrangement of the dental papille, the character of the granulation of the 
disk, and its extent orally, seem to be characters of considerable constancy and importance, 
while the number of arm-spines, the proportion of disk to arms, and the coloration are 
much less reliable. 
Koehler (1907) has drawn attention to the unsatisfactory character of the line sup- 
posed to separate Ophiocoma from Ophiomastiz. The character of the disk covering and 
the form of the upper arm-spines furnish the distinctions supposed to make one separable 
from the other, but a careful comparison of all the species of the two genera shows that 
the line has to be a very arbitrary one. Since the character of the disk covering seems to 
me a more fundamental feature than the form of the arm-spines, I am retaining in Ophio- 
coma the species wendtii, and am also including Ophiomastix pusilla Brock, in spite of 
their claviform arm-spines. Such spines indeed occur in some individuals of scolopendrina 
and other Ophiocomas. 
All the various features being duly taken into account, the following key will make it 
possible to distinguish the 19 forms, which are here recognized by name. 
Key to the Species and Varieties of Ophiocoma. 
A. Disk granules nearly or quite spherical and of more or less uniform size. 
B. Claviform spines when present, uppermost of series. 
C. Disk granulation rather coarse, 9 to 64 granules per square millimeter, covering only part, 
and often a very small part, of oral interbrachial areas. 
D. Tentacle-scales 2 (distally there is usually only 1, and individual pores proximally may 
have but 1). 
E. Granulation coarse, 9 to 36 per square millimeter; adoral plates small at sides of 
oral shields; arm-spines 4 or 5; coloration black or blackish, or some combina- 
tion of blackish and whitish. 
F. Oral shields square or oblong with rounded corners as wide proximally as distally; 
under arm-plates not noticeably wider than long, often longer than wide, espe- 
cially basally; granulation moderately coarse 25 to 36 per square millimeter; 
uppermost arm-spines often short and swollen; West Indian................ echinata 
FF. Oral shields wider distally than proximally, often markedly so; under arm-plates 
usually much wider than long; granulation coarser, often only 9 to 16 granules per 
square millimeter; uppermost arm-spines rarely short and swollen; Indo-Pacific. 
