Happon—The Actiniaria of Torres Straits. 453 
PHELLIA, Gosse. 
With the characters of the sub-family. 
The following members of the genus Phellia have been investigated anatomi- 
eally. P. limicola, Andr. (Le Attinie, 1884, pp. 73, 74); P. Ternatana, Kwietn. 
Zool. Anz., 1896, No. 512, and Abhandl. Senckenbergischen naturf. Gesellsch., 
xxill., 1897, p. 328; he also alludes, p. 327, to P. Ambonensis, and P. decora (?), 
Klunz., from Ambon and the Red Sea, respectively. P. Sollasi, Hadd., from 
Funafuti, Ellice group, has been described anatomically by Maguire (Proc. R. 
Dubl. Soc. viu., Part vi., 1898, p. 717), and I have also studied P. vermiformis, 
n. sp., and a species of Phellia from Gare Loch, Scotland, which is probably 
P. gausapata, Gosse. 
Andres gives a long list of species which have been described as belonging to 
this genus; to these must be added eight new species named by Danielssen 
(Actiniidze, Norske Nordhays-Exped., 1876-78 (1890) ). 
As the genus Phellia was instituted by Gosse in 1858 (Ann. Mag. Nat. Hist. (3) 
I1., p. 192) for two new species of sea-anemones, P. murocincta, and P. gausapata, 
it is important to determine the anatomical characters of these species. Dr. John 
Murray entrusted to me a form that he obtained from low water near the head of 
Gare Loch on the Clyde. The length of the specimen when contracted and 
preserved in alcohol was 12 mm. long, and 7 mm. in diameter. The body was 
transversely wrinkled, and the upper third was thickly covered with cuticular 
scabs, these were thinly scattered elsewhere. It is always a difficult matter to 
determine the species of an Actiniarian from a single preserved specimen, but I 
have very little doubt that this is P. gausapata, Gosse (cf. Hist. of Brit. Sea 
Anemones, 1860, p. 140). 
So far as the general anatomy is concerned this species agrees with the above- 
mentioned forms, and therefore we may safely regard them as belonging to the 
genus Phellia. The main point of difference is that, in P. gausapata there are in 
my sections only one pair of imperfect mesenteries in each of the exoccels, that is, 
there are only 12 pairs of mesenteries altogether. The form and pattern of the 
transverse sections of the retractor muscle also differs in details from that of 
the other species. 
The only point calling for special notice is the number and disposition of the 
imperfect mesenteries. In P. Ternatana, and P. vermiformis, the arrangement in 
each sextant is as follows:—Py ,,||,, 1) P, that is, next to each pair of perfect 
mesenteries is a pair of small imperfect ones; then follow a minute pair (PI. xxvut., 
fig. 11), and in the centre of the exoccel is a pair of moderately large imperfect 
mesenteries. ‘The same arrangement appears to be characteristic of the small 
