ANNOTATED LIST. 111 
range, since it is reported from the east coast of South Africa, northward to the Persian Gulf 
and eastward to the Society Islands. Oddly enough, it doesnot occur at the Hawaiian Islands, 
nor do I know of any specimens from Samoa. The northernmost locality is Kominato, 
Japan, and the southernmost records are from Natal, and Port Curtis, Queensland. 
Ophiothrix martensi australis subsp. nov. 
(Plate 15, Figure 4.) 
The specimens of O. martensi which I found at Mer and at Thursday Island show a 
very distinctive type of coloration by which they are readily distinguished from specimens 
from the Philippine Islands (the type locality) and East Indies. With 6 Australian speci- 
mens at hand and 15 from the Philippines, the differences have been carefully weighed, 
and it seems desirable that the Australian form should, for the present at least, bear a 
subspecific name. I have accordingly selected one indicative of its geographical relation- 
ship. Excellent photographs of australis are given by Déderlein (1896, Jena. Denkschr., 
8, pl. xvi, figs. 13, 13a), based on a specimen from Thursday Island. If these photographs 
and the colored figure given herewith (pl. 15, fig. 4) are compared with Lyman’s original 
figure (1874, Bull. M. C. Z., 3, pl. iv, fig. 10), the character by which the two forms are 
distinguishable will be seen at a glance. In typical martensi there is a deep blue-purple 
longitudinal median line on the upper side of the arm, bordered on each side by a broader 
white or light-colored band or area. Judging from living specimens of australis, this lighter 
color is in life orange or orange-yellow, but all trace of yellow is lost either in alcohol or 
on drying. On the lower surface of the arm in typical martensi is a median longitudinal 
light stripe, presumably yellow or orange in life, of greater or less width, often more or less 
interrupted and broken. Now, in the southern form, the light (yellow or orange) stripes 
both above and below are so broken and irregular as to give a very different appearance. 
In typical australis (pl. 15, fig. 4) the two halves of each upper arm-plate are of contrasting 
colors; if the right half of one plate is purple or deep indigo, the left half of the same plate 
and the right half of each adjoining plate is orange, and vice versa. Hence we have 
striking alternation of purple and orange (more or less bleached in preserved material) 
for a greater or less distance on the arm. Similarly on the ventral surface, instead of a 
median line or band of yellow, we find the two halves of each plate of alternating and 
sharply contrasting colors. In no specimen of australis that I have seen is the alternation 
of colored plate-halves perfect; another form of alternation also occurs in which alternate, 
whole plates contrast, and all our specimens show more or less of an irregular combination 
of these two forms. The specimen from which Mr. Grosse’s figure (pl. 15, fig. 4) was drawn 
is one of the most typical, but many plates are unicolor, though purple and light almost 
always alternate. The difference, then, between martensi and its sub-species australis 
may be expressed thus: in martensi the colors tend to a definite longitudinal arrangement, 
while in australis they tend to a transverse arrangement in which the two contrast- 
ing colors are apt to alternate on the two sides of the arm. All of the East Indian and 
Philippine specimens are distinctly typical martensi, while our 6 Australian specimens are 
clearly australis. 
The Alert took martensi, according to Bell, at Port Curtis, Queensland, and Port 
Darwin, Northern Territory, as well as at Thursday Island. He refers to the diversity of 
color shown, commenting on a dark and a light form. These dark and light forms occur 
in both typical martensi and in australis, and are probably due simply to the extent to 
which the purple (or dark-colored) pigment is developed. They may perhaps be associated 
with age or with food. 
Mr. Lyman records martensi as taken by the Challenger, ‘‘ August 7, 1874,” and hence 
somewhere among the Fiji Islands. It would be interesting to know whether these speci- 
mens are typical or whether they belong to the subspecies. To judge from Bell’s comments 
