74 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [JuilG 14, 



lamina had been moulded on the inferior, without being united to it ; 

 a pause having clearly intervened between the perfecting of one plate 

 and the secreting of the other ; and the adaptation in the irregulari- 

 ties, or in the projections and depressions, would account for a bi- 

 fold composition not being visible in thin slices. The characters of 

 the minor openings would lead likewise to the inference, that those 

 apertures were, for a time at least, not wholly closed ; but aiforded a 

 direct, often a large means of communicating between successive 

 layers. Among existing corals the Antliophyllum musicale of Ehren- 

 berg * possesses a cellular intermediate structure ; but the cells or 

 vesicles have no similar foramina hi their upper and lower laminae, nor 

 are they arranged in rows separable mechanically as in Choristopetalum 

 impar ; the whole composition being an irregular aggregate of vesi- 

 cular cavities, and produced, according to Ehrenberg, by appendages 

 of the mantle {op. cit. Gen. Char. p. 89). On the contrary, it is 

 believed, that the membrane, which secreted the outer surface of the 

 raised margin of the abdominal cavity, did not range far beyond that 

 structure, as it would be difficult to conceive, if an extension of the 

 mantle alone produced the cellular composition, why a relatively 

 large aperture should be kept open between a subjacent and superior 

 layer of cells in this case, and none should exist in Anthophyllum, 

 musicale, &c. It is conceived, therefore, that the cells contained 

 within themselves the secreting membranes by which they were 

 formed, nourished as well as supplied with calcareous matter from 

 the digestive organs, by means of the lateral pores in the abdominal 

 cavity and the sides of the cells likewise ; while at the fitting season 

 for developing another layer, the secreting membranes were extended 

 through the apertures in the pre-existing one, and when sufficiently 

 grown commenced forming the superstructure, a constant communi- 

 cation being kept up by aid of the necessarily corresponding openings. 

 Respecting the mode of developing additional cavities, very few 

 suggestions can be offered, and so far as was ascertained, they 

 were produced chiefly in the central area. In the middle of ho- 

 rizontal thin sections (fig. 10), small polygonal spaces were sur- 

 rounded Ijy others of full size ; but they were not numerous ; and in 

 a central boss of a fractured branch, similar minute cavities occurred : 

 vertical slices also, to the extent to which they could be trusted, 

 being necessarily more or less oblique, on account of divergence in 

 growth, sanctioned an interpolated origin. Not a trace, under any 

 circumstances, was noticed of a divisional process within a mature 

 cavity, resembling that which exists in the ChcEtetes^; of M. Fischer. 

 No satisfactory evidence was obtained of additional cavities in the 

 outer zone ; but this want of positive proof must not lead to the m- 

 ference, that they never exist in that portion of a branch. Among 

 existing corals, as De Blainville's SideroportB and Lamarck's Pocil- 

 loporcBy young visceral hollows are produced most abundantly in the 



* Beitrage, «&c. p. 89 ; Caryophyllia musicalis, Lamarck, 2nd Ed., t. ii. p. 350, 

 no. 6- Consult Esper's Pflauzeuthiere, Madrep. tab. 30. fig. 2-4. 



t Geology of Russia, by Sir K. I. Murchison, M. de Verneuil and Count Keyser- 

 ling, vol. i. Appendix A. p. 594. 



