20 
is no convincing trace of pinnules to be seen when the downward curved tentacles are split or 
when sections are made. On better preserved specimens, however, with the polyps protruding, 
there are distinct, though extraordinarily primitive pinnules, just like minute pointed lobes. It is 
difficult to count them, but there are about four or five longitudinal rows on each side, and 
at least a score in each row. 
The absence of pinnules is emphasised in Burchardt’s Clavularia amboinensis (Alcyonaceen 
von Amboina (II), Jenaische Denkschr. VIII, 1898, p. 657). He says that even with high 
magnification no trace of pinnules was to be seen. But Burchardt’s specimen was doubtless an 
Anthelia, and in any case it had nothing to do with ours. The stolons were reticulate, thin 
and flat; the polyps were limp, mostly half-retracted, and sometimes when fully retracted they 
were represented by small buds on the stolon. There were no spicules. Another form without 
pinnules, but almost certainly the same as Burchardt’s species, has been described by Bourne 
as Acrossota liposclera (Quart. Journ. Microscop. Sci. 1915, 1 pi.). 
As a polyp-portion with stomodaeum and tentacles is plainly drawn into a non-retractile 
stout calyx, our specimen should be referred to the genus Clavularia, not to Anthelia. Here 
we follow Kukenthal’s re-definition of the two genera, which seems to us to be thoroughly sound. 
Our first inclination was to include our specimen in Studer’s Clavularia rosea with 
which it agrees in size (9 — 10 mm.), colour, stiff calyx, longitudinal ridges, rough spindles, 
retractile tentacles and oral disc. But Studer described the tentacles of his specimen (from N.W. 
of Kerguelen) as ‘short’, while those of our specimens are notably long. Moreover, Studer 
makes no mention of the irregular spicules. As Studer’s description is very short, we do not feel 
confident in including his Cl. rosea with our specimens, but the resemblances are certainly striking. 
On the other hand our specimens agree entirely with P achy clavularia erecta Roule, and 
we propose to re-emphasize this genus and include our forms within Roule’s species. In so doing 
we would lay stress on the peculiarities of our specimens as distinguished from ordinary 
Clavularias. The notable resemblances between our specimens and Route’s are the following: 
(1) the thick composite nature of the basal membrane, (2) the large stiff polyps (calyces) 
longitudinally grooved, (3) the spindles with numerous compound warts; and (4) the rose- 
violet colour. 
Roule calls attention to the resemblance between his specimens and Studer’s C. rosea , 
but he does not mention the invaginated tentacles (nor any tentacles at all) nor the striking 
branched spicules. Through the courtesy of Professor Roule and of Professor Revilliod, Director 
of the Geneva Natural History Museum, we have compared our specimens with Roule’s, and 
they are certainly the same. 
The spicules — all purplish in colour — include : 
(a) straight and curved spindles with high, very compound warts, markedly in zones, and standing 
out very prominently on the margin, 0.6 X 0.07 mm. 
(b) very irregular, very warty forms, some bifid at one end, some triradiate, some with four 
branches and some more irregular still. Some of the warts are so compound that they 
might be called branched. The branches are from o. 1—0.3 mm - an< ^ up to 0.07 mm. in breadth. 
(c) small spindles with distant compound warts, 0.24 X 0.03 mm. 
Previously recorded from Amboina. 
