36 
tinguished by the number of internal longitudinal bars, their num- 
ber being 5—6 in M. tumulus , but 3—5 in M. amokurae. Further 
the shape of the intestinal loop and the length of siphons afford 
distinguishing characters. 
The present form does not fully agree with any other species 
hitherto described; accordingly there can be no doubt that it re¬ 
presents a new species. 
Halocynthia cavnleyensis n. sp. 
(Plate IV, figs. 3 & 4 ; textfigure 2). 
Localities: Carnley Harbour, Perseverance Harbour. Upon or 
between stones at low tide. 
Of this apparently new species a considerable number of spec- 
imens were collected. Some of them are solitary, others coalesced 
with their tests to such a degree that considerable force must be 
used to separate them from one another. The largest specimen 
measures ca. 5 cm, but many smaller sizes are represented. The 
shape is exceedingly varying, and the specimens have been attached 
to the stones in different ways. The colour is brown, but the un- 
even and wrinkled surface is more or less overgrown with Bryo¬ 
zoa, Hydroids etc. 
The body-apertures, which are situated upon very low 
siphons, are often very difficult to see from the outside. 
The test is very firm, moderately thick round the siphons, 
but coriaceous and strongly developed basally, in some specimens 
almost forming a peduncle, while in others it is moderately thick. 
On the inner side the test is dark coloured, a little opalescent. 
The bo dy-wall is muscular and without difficulty loosened from 
the test. The colour is reddish. On the preserved specimens no 
siphons are visible when the test is removed. In a specimen 
whose body was 3 cm long, test not included, the distance be¬ 
tween the body apertures was 1 cm. 
The ten ta cl es are present in a number of about 20—23. 
Their branches are rather short. The branches of 2nd order almost 
rudimentary. They are all of about the same size, and it was im- 
possible to State a regular alternation between smaller and larger 
ones. A number of Protozoa, attached to the tentacles, made them 
look more ramified than they really were (PI. IV, fig. 4). 
