14 GENEEAL EEMAEKS. 



one, and the corallites of M. hebes agree in form with those of M. millepora, M. prostrata, &c., 

 and not with M. cytherea and allied forms. The absence of M. corymbosa from Dana's list 

 will also be noted. Doubt has already been cast on Dana's identification of M. corymbosa, 

 Lamk. ; and the fact that he did not associate M. corymbosa, Dana, with M. cytherea in the 

 list quoted lends further support to the view that Dana's species is not identical with that of 

 Lamarck. 



Whether the radial polyps of M. aspera, which have equal tentacles according to Dana's 

 figure, differ structurally from those of M. prostrata is not at present known. The account 

 given by Fowler of the structure of M. aspera probably refers to the species M. pocillifera, 

 which is, however, closely allied, and is here regarded as a member of the same group. 

 A fragment of the material has been kindly placed at my service by Dr. Fowler, and I find 

 that it agrees closely with the ' Challenger ' specimens of M. pocillifera. The only other 

 species studied by Dr. Fowler does not belong to the same group as M. prostrata, so that we 

 have at present no means of comparison. It appears, however, probable that certain types of 

 arrangement of the septa are associated with morphological differences in the polyps which 

 produce them ; and a fuller knowledge of the structure of the soft parts will doubtless supply 

 valuable data for the purpose of classification. I may, however, point out that for this 

 purpose it is not sufficient to indicate general structure ; it will be necessary to give details 

 for the axial corallites, for the prominent radial corallites, and also for the immersed or other 

 forms which may difl^er from the usual type. 



Structure of the Polyps, and the Relation of Soft Tissues to the Skeleton. — Fowler * has 

 studied the anatomy of two ' Challenger ' species of Madrepora, in one of which a new and 

 interesting type of dimorphism occurs. The present section consists of a summary of his 

 results, together with a few supplementary observations of my own. 



The external body-wall consists of the usual three layers — ectoderm, mesogloea, and 

 entoderm, — and is supported on echinulations or rugae both on the radial corallite-walls and 

 in the inter-corallite regions, where the latter are of sufficient importance to possess inde- 

 pendent supports. Under the external body-wall and between the rugae {M. pocillifera) 

 a series of external longitudinal canals exist, which open into each other and also through the 

 corallum into a series of internal canals with radial and transverse connections; these, in 

 turn, communicate with the general coelentera of the polyps, and all communicate eventually 

 with the coelenteron of the axial polyp. These canals occupy the channels in the corallum 

 already described. The structure of the polyps is, in its general features, Actinian, but there 

 is a marked bilateral arrangement of parts. The mouth-disk, with its fringe of twelve 

 tentacles, is capable of retraction within the fossa of the eorallite. In vertical section of a 

 contracted polyp the soft tissues form a funnel-shaped structure similar to that seen, in 

 Alcyomum under similar conditions. The mouth is elongated in the sagittal axis. The 

 stomodaeum, which occupies the central portion of the coelenteron, is supported by twelve 

 mesenteries, which, however, differ in importance. The extremities of the stomodaeum are 



• " Anatomy of the Madreporaria.— Part II.," Quart. Journ. Microsc. Science, vol. xxvii. (1886) 

 pp. 1-16, pL i. 



