Hapvpon—The Actiniaria of Torres Straits. 477 
and separate from the marginal tentacles by a naked space, are numerous tuber- 
culiform or lobed tentacles, arranged more or less radially. Sphincter muscle, 
endodermal, circumscribed, diffuse and very weak, or absent. The musculature 
throughout is very weak. 
I have combined the definition of this family by Andres and M*Muwrrich with 
the addition of calling the marginal tentacles ‘‘ minute,” and of recording the 
presence of a sphincter muscle in some genera. M*Murrich points out that the very 
characteristic of the family, as a whole, is the excessive proportion of glandular 
cells in the endoderm, and the very slight development of nematocysts. As 
in the Discosomidx, some of the forms are stated to be without a sphincter, 
others have a feeble, while others again have a well-developed, endodermal 
sphincter. 
The following genera may be relegated to this family :— 
Rhodactis. Ricordea. 
Actinotryx. Heteranthus. 
RHODACTIS, M. Edw. et Haime.* 
Rhodactidze, in which the edge of the oral disk is produced into subtentacul- 
form crenulations; the disk is covered with short tentacles, of which the outer 
rows and a row round the mouth are simple and conical, while those of the central 
zone haye a very short, thick stem, surmounted by several short finger-like pro- 
cesses ; each of these latter, and apparently some of the large simple ones, are set 
in depression of the disk, surrounded by a ring-like thickening of the disk. Cir- 
cumscribed endodermal sphincter muscle ; two gonidial grooves. 
I have recast the definition of this genus, laying emphasis on the collared pits 
in the oral disk from which the larger complex tentacles arise and on the circum- 
scribed muscle. For the knowledge of the latter fact I have to thank Dr. Carlgren, 
who is investigating &. rhodostoma, and who will, I believe, also describe a new 
species of this genus. Duchassaing and Michelotti describe two species—R. Dane 
and R. musciformis (Mem. R. Acad., Turin, xxili., 1866, pp. 131, 132); but so far 
as I can understand their account, neither species belongs to this genus. 
R. rhodostoma (Ehr.), 1834, for bibliography cf. Andres, 1884, p. 283. 
? R .Howesti, S.-Kent, 1893. 
*1851; M. Edw. 1857, ‘‘ Hist. Nat. Coral.” i. p. 293. 
