17 
less distorted; the edge is more raised above the surface of the coral than in S. ¢Zéatus. The 
diameter of the cyclosystem is 1 by 0,8 mm. The septa are very thick, the space between them 
narrow. The ampullae appear first on the posterior surface of the flabellum, one below each 
cyclosystem; later they are formed on the anterior surface or sides of the branches, but always 
in connexion with a cyclosystem. The surface of the ampullae is very irregular. 
This species may be defined: 
Hydrophytum flabellate; surface covered with large blunt spines; cyclosystems directed 
towards one surface; gastropore deep and curved down the stem. 
At first sight S. ¢z@eatus and S. umbonatus, owing to the peculiar position of their cyclo- 
systems and the shape of the gastropore, appear to belong to a different genus to the other 
species of Stylaster. But in S. multiplex and S. minimus we have intermediate forms connecting 
these with the ordinary types of Stydaster. In the former S. ma/tiplex, the cyclosystems are 
partially turned towards one surface of the flabellum and in the latter, S. mzx¢mus, although the 
cyclosystems face towards one surface only, the gastropores are not very deep and curved, as 
in S. ¢zdzatus. These two characteristics cannot then be depended upon for generic distinctions. 
The genus Stexohelia was defined by MoseEtey in the following terms: “Corallum delicate, 
branching, flabelliform; pores in regular cyclosystems only. Cyclosystems all turned towards one 
surface of the flabellum. Dactylopores without a style or with a very rudimentary one. Gastro- 
pore very deep and curved, so as to tubulate in all but the older branches, the entire lengths 
of the axes of the branches, with small styles seated at the bottoms of those tubes and directed 
parallel to the axes of the branches at right angles to those of the mouths of the cyclosystems”’. 
If we are right in concluding that the shape and position of the cyclosystems as they are here 
described (which description also applies exactly to S. tdzatus and S. umbonatus) are not good 
characters for generic distinction, then the only character that divides the genus Stexohelia from 
Stylaster is that the dactylopore in the former is either without a style or with a very rudimentary 
one. The small size of Stenohelia profunda must have made it difficult to ascertain this point 
with certainty. S. wmdéonatus is a larger and S. t2/zatus a very much larger species and these 
both show a distinct style in the dactylopore, which can in neither case be called rudimentary. 
It seems therefore that there is not sufficient reason for keeping separate a genus Stexohedia, 
and so the new species S. édzatus and S. wmbonatus have been included in the large and variable 
genus Sty/aster and according to us the only two species of Stenohelia, S. madierensis Kent 
and S. profunda Moseley, must also be added to it. Jonnsron, who first described the species 
S. madierensis placed it in the genus Sty/asfer: it was afterwards redescribed by Savitte Kent. 
Errina Gray. 
The genus Zvrzma may be defined as follows: 
Coenosteum arborescent, irregularly flabelliform; gastropores and dactylopores separate, 
not in cyclosystems, regularly or irregularly arranged; style in gastropore, no style in dactylopore; 
dactylopore on a nariform projection; gastropore with or without a scale (the raised margin of the 
pore or fused dactylopore projections). Gastrozooids with four or five short tentacles: dactylozooids 
SIBOGA-EXPEDITIE VIII. 3 
