124 THE ECHINODERMS OF TORRES STRAIT. 
well, some even extending to Western Africa. In each of the three geographical areas 
there seems to be going on a similar specific evolution, affording an interesting illustration 
of parallel differentiation. In the West Indian region this differentiation has reached its 
most definite results; here we have 3 well-marked species, one (pumila) in the pumila 
group and two (echinata and riiset) in the scolopendrina group. There is no present ten- 
dency towards any intergradation of these species; any West Indian Ophiocoma can be 
referred at once and without difficulty to its proper species. In other words, specific evolu- 
tion in Ophiocoma in the West Indies has attained definite and clear-cut results. The 
same is true in the Panamic region, where, however, only two species are involved—alex- 
andri of the pumila group and ethiops of the scolopendrina group; the latter is the more 
interesting, for, while it possesses the general appearance and coloration of the West Indian 
echinata, it has nearly lost one of the two tentacle-scales, as the West Indian riisez has 
done, and meanwhile it has become much the largest species of the genus. 
When we turn to the Indo-Pacific region we find that specifie differentiation is much 
less perfect and there are numerous forms-whose phylogenetic significance is as yet very 
obscure. This is least true in the pumila group, of which but two! forms are involved— 
one of these is valenci@ of the East African region, much like alexandri of the Panamic 
region, but with short arms, and the other is parva of the southern tropical Pacific islands, 
which has retained the 6 arms of young pumila, the species to which it is most closely 
allied. In the brevipes group, differentiation has gone on very markedly in color, size, 
and proportions, and to some degree in arm-spines and mouth-parts, but nothing fixed 
seems to have been reached. Hence we have a perplexing series of forms which look very 
unlike each other, but which are bound together by numerous intergradations. It is quite 
possible that each of the four forms herein recognized by name is an incipient species, but 
the evidence is very dubious at present. In the scolopendrina group there is clearly a dif- 
ferentiation similar to that which has taken place in the West Indies, by which we should 
have a more or less variegated species with 2 tentacle-scales, and a uniformly black species 
with a single tentacle-scale. But as yet there are many connecting links left, and hence 
we find variegated individuals with 2 tentacle-scales, which we call scolopendrina, varie- 
gated or dark individuals with 1 tentacle-scale (and claviform upper arm-spines), which 
are called wendtii, black forms with 2 tentacle-scales, known as erinaceus, and other black 
specimens with a single tentacle-scale which have been named schenleinii. An isolated 
and well-marked species (canaliculata) has become differentiated on the southern coast of 
Australia, but nowhere in the Indo-Pacific region is there a well-defined indisputable species 
of the scolopendrina group. 
As Matsumoto (1917) has pointed out, brevipes is apparently the most primitive 
member of Ophiocoma. From it, the scolopendrina group has been differentiated by the 
tendency to loss of disk granulation, associated with increase of size of granules and an 
increasing number and more definite arrangement of the dental papille. Further specializa- 
tion leads to such species as rivset and schenleinii, in which a tentacle-scale has been lost 
and the color has become uniformly black. The evolution of the pumila group seems to 
have been along a somewhat different line. Here the color has tended to become lighter, 
and green is an evident feature of young individuals and of recently regenerated arms. 
Now the color green is known in the genus otherwise only in brevipes, a fact which suggests 
that the pumila group has originated directly from brevipes and not, as one might infer 
from the long arms and certain other features, from scolopendrina. If this is the case, dif- 
ferentiation has gone as far as in rise? in fixing a single tentacle-scale and a distinct color- 
pattern, while in the character of the disk-granulation it has gone much further. As 
1T am leaving bollonsi out of account, as I have never seen an example of the species, and while it has granules 
which “become spiniform,”’ it has 2 tentacle-scales and a uniformly dark coloration which precludes the idea of its 
having any close genetic connection with the pumila group. 
