XU THE VOYAGE OF H.M.S. CHALLENGER. 



In another direction, Cornularia seems to be a starting-point for Clavularia 

 and Sarcodictyon. In Clavularia the polyps spring from stolon-like processes, 

 or from a basal membrane. In this latter case the bases of the polyps are not 

 surrounded by the membrane, but ai-e seated upon its surface and continued into its 

 substance by stolon-like endodermic tubes. Furthermore, in this genus the oral region 

 is retractile within a rigid calyx portion. Similar conditions are, according to Herdman,^ 

 to be observed in Sarcodictyon, where the polyps, provided with retractile mouths, are 

 seated upon a basal expansion. The latter is, however, narrow and stolon-like, and 

 extends from each polyp mainly in two directions, so that the individuals of the colony 

 are united in rows, and only at rare intervals are aggregated on a common basal surface. 

 In Cornularia the polyps bear no spicules and the stolons have a horny ectodermic 

 layei'. In Gymnosarca, which is perhaps closely related to Cornularia through Cornu- 

 lariella, the creeping stolons, which are numerous and thick-walled, anastomose and give 

 oflF free cylindrical stolons on which polyps are seated. Finally, in Telesto, upright 

 elongated polyps rise from the stolons, and exhibit thick, horny walls with spicules, from 

 the canal-system of which there arise cylindrical individuals with shorter cavities. This 

 genus also includes Clavularia prolifera, v. Koch, perhaps identical with Gorgonia 

 trichostemma, Dana, and also Clavularia rupicola, described by F. Miiller^ under the 

 generic title Carijoa. Telesto exhibits marked affinities with many Gorgonacea, 

 towards which Ccelogorgia forms a transition link. 



The genus Cyathopodium diverges in another direction. This is Verrill's generic 

 title for Aulopora tenuis, Dan.,^ in which the walls of the stolons become calcified, 

 and thus suggest tlie condition met with in Tuhipora. With Clavularia-\\\iQ forms 

 having flat basal expansions, certain genera are connected in which the tendency to 

 upright branching appears. This is the case in the genus Scleranthelia, and in Antho- 

 podium, which latter, according to Verrill, is related on the one hand to Telesto, on 

 the other to Callipodium. ■ 



A grouji of forms, more or less related to Sarcodictyon, is formed by the genera 

 Sympodium, Erythropodium, and Callipodium. Of these the latter, according to 

 Verrill's description and figures, very closely resemble Sarcodictyon. On the other 

 hand, the whole group is related to the family Briareidas, which has perhaps been 

 derived from similar forms. 



1. Cornularia, Lamarck. 



2. Rhizoxenia, Ehrenberg. 



3. Clavidaria, Quoy and Gaimard. 



4. Sarcodictyon, Forbes. 



5. Anthelia, Savigny. 



6. Gymnosarca, Saville Kent. 



7. Cornulariella, Verrill. 



8. Telesto, Lamouroux. 



' Proc. Roy. Phys. Soc. Edin., vol. viii. p. 31. 



^ Archivf. Naturgesch., Jahrg. xxxiii. p. 33(1, Anm. fig. 56, 1867. 



3 Zoophytes, p. 630, pi. lix. fig. 5. 



