00 MADREPORARIA. 



specimens, it is highly probable that the specimen would turn out to be only an older and 

 coarser example of the small specimen described above as A. horizontalis. It is also to be 

 regretted that the habitat of this coral is unrecorded. This is another reason why it is 

 advisable to keep the specimens apart as distinct species. Both sides of the coral are shown 

 on PI. XXVII. 



a. Locality not recorded. [Register No. 93. 7. 1. 18.] (Tjrpe.) 



Species 5. Astrseopora arenaria. (PI. XXIX. ; PI. XXXIII. fig. 11.) 



Description. — CoraUum encrusting, forming with the substratum a solid irregular mound. 

 The pendent edges either hanging right down, or turned outwards by a well-developed 

 epitheca. 



Calicles very irregular in shape, 2 mm. and under in diameter (new minute calicles 

 budding all over the surface among the older), either immersed or u-regularly protuberant, 

 either bluntly conical, or globose, pointing in all directions ; protuberant margin, irregularly 

 granular and jagged. Two cycles of septa very irregularly developed as slight ridges, some- 

 times apparent only near the margin. Here and there, however, the primaries are well 

 ileveloped. 



Coenenchyma solid, rough and granular. The two typical elements, costal and synap- 

 ticular, greatly disguised by their thick nodulated appearance, although, here and there, the 

 typical echinulations and floors can be well made out. General appearance as if built up of 

 sand-grains cemented together with varying compactness. 



The single specimen of this new type, which is named after the character of its 

 ccenenchyma, is a high mound, nearly regularly rounded at the top, while the pendent edges 

 show a tendency to turn outwards. These creeping edges show clearly that the species 

 belongs to the e.xplanate type of growth, but at the same time the specimen shows an affinity 

 with the globular members of the genus. There is, unfortunately, no section exposed, but, 

 .judging from the great relative height of the mound, and also from the fact that young calicles 

 appear in all the interstices between the larger calicles, we may safely conclude that the 

 corallum is increasing in thickness at the centre with great rapidity, the existing calicles 

 lengthening and separating as they lengthen, making room for new ones to develop. This 

 allies the form with the globular members of the genus, while the turned-out edges ally it 

 with the explauate forms. In the true globular forms, the epitheca seems no longer to have 

 the power of turning the advancing edges of the corallum outwards ; the coraUum thickens not 

 only at the centre, but also at the edges, rolling over and depressing the epitheca. 



The specimen, coming from the Red Sea, was labelled Astrwopora inyriopMhalma, 

 Lamarck, but it is very distinct from the specimens above identified as Lamarck's myri- 

 Oiihthalma. 



a. Red Sea. (From the Paris Museum, as A. myriophthalma, Lamarck.) (Type.) 



