134 DESCRIPTION OF CORALS. 
Lamarck as a variety of Oc. virginea, but Lochner compares the specimen 
represented with the Mediterranean coral or Dend. ramea). In no case how- 
ever do stellular cavities appear to have been developed in the animal sub- 
stance which permanently invested the coral, or in irregular extensions of the 
mantle*. 
In well-known Oculine, on the contrary, no persistent central lamelliferous 
cavities occur at any period of growth, normal or otherwise ; and Ehrenberg 
states, that the essential mode of developing young tubes is by a swelling in 
the parent-tube. In Oc. pallens, a separation from Oc. hirtella made by that 
authority, the additional polype-receptacles are, however, produced on the side 
of those previously existing, and are usually so crowded together at the ex- 
tremity of the branches as to occupy nearly the whole intervening spaces : 
they originate also clearly in the mantle, or external animal substance. In the 
same species, extensions of this outer covering take place where the original 
polypes have been destroyed, and young tubes are freely developed in the 
expanded portions. In these respects there are considerable agreements with 
the Bracklesham coral under immediate examination ; but it must be stated 
that Ehrenberg is of opinion, that the slender variety of Oc. pallens, believed to 
be the one just noticed, ‘‘ draws near” to his genus Stephanocora, established on 
a coral found in the Red Sea, and possibly not yet figured (op. cit. pp. 76, 78). 
In mode of growth that polypidom agrees moreover with Oculina virginea, as 
defined by Ehrenberg, Oc. pallens and the Bracklesham coral, forming both 
branches and expanded or overlying gemmiferous layers ; and still farther, a 
* Dendrophyllia cornigera, M. de Blainville’s third recent species, agrees with Dend. ramea in 
being permeated by central lamelliferous tubes ; but it differs apparently in the whole of the exterior 
not being constantly covered, while living, by animal matter. On this account, it is presumed, the 
coral should be removed to Ehrenberg’s genus Cladocora (Beitrage, p. 85). A comparison of the 
structures of Dend. ramea with those of a polyparium believed to be Dend. ( Clad.?) cornigera led to 
the detection of subordinate distinctions, valuable nevertheless in the study of imperfect fragments of 
analogous fossils. In Clad.? cornigera the centre of the stem is not rapidly consolidated, or but par- 
tially ; the open stellated structure is traversed by numerous well-developed diaphragms ; and the ex- 
terior is not thickened by continuously applied secretions, except at the base, or near the union of a 
branch with a principal stem. ‘The additional polype-cavities are produced by germs (Ehrenb. loc.cit.) 
on the side of the parent branch, but a certain limited connexion apparently exists for a time between 
the interior of each. The fossil figured by M. Michelin under the same name (Dend. cornigera, Icon. 
pl. 10. fig.9) agrees perfectly in generic characters with Clad.? or Dend. cornigera. It is also be- 
lieved that the extinct coral, Dend. irregularis of M. de Blainville, is generically if not specifically 
identical with M. Michelin’s fossil. 
