28 PROF. P. M. DUNCAN ON A NEW GENUS OF MADEEPORAEIA. 



be counted, it will be found that in some with 55 calices 35 have 

 the long lamina and 10 have an indistinct septal arrangement, while 

 the rest have no long structure crossing the calicular fossa. In a 

 patch with 75 calices of many shapes and sizes there are 55 with 

 the elongate structure. The orientation of the lamina differs even 

 in neighbouring calices. 



The septa are bilamellar, and the evidence of a very irregular and 

 narrow interlamellar space is apparent, sometimes superficially, and 

 invariably in microscopic sections of corallites near the calices. 



A perfect, fall-grown calice has the two opposite primary septa 

 united at the axis by means of a narrow discontinuous columella, 

 which is ornamented in the same manner as the septa, but which, in 

 some instances, has a raised edge (figs. 3, 6, 9, and 10), or they 

 may unite by their inner ends and close partly the axial space with or 

 without the assistance of thickening or of any extra growth (figs. 12, 

 13, and 14). In buds the columella may be distinct, and there are 

 the usual six primaries (fig. 2, d). In large immature calices the 

 six primary septa may converge and unite with a small columella or, 

 what appears to be the more common case, they do not unite. The 

 ends elongate or twist (fig. 16) and, with the dissepimental tissue, 

 close the axial space. It is this closure, either by a columella 

 which sections prove to be occasionally discontinuous and always 

 non-essential in its method of growth, or by united septal ends, 

 or by twisting and elongation of the septal ends, assisted by the 

 tabula-like upper dissepimental structures, that forms the main 

 distinction between the new genus and Septastrcea. 



The endotheca is variable in amount, and whilst it is close and 

 thin in some corallites, or parts of corallites, it is distant and stout 

 in others. In some cases the horizontal stout platforms are fairly 

 regular ; but, as a rule, although the dissepiments completely close 

 the interseptal loculi and act as tabulae, they lack the regularity of 

 those endothecal structures. Thin dissepiments often close the 

 whole of the interseptal loculi at indefinitely placed intervals in the 

 height of the corallites, and as completely as any tabulae (fig. 6). The 

 uppermost dissepiments are those which close in the calicular fossa 

 and reach up close to the free edge of the septa. There are no deep 

 interseptal spaces in the form, and the dissepiments moreover come 

 close to the top of the axis, and, by joining with the septal ends and 

 the intermediate structure (when it exists), help to close the axial 

 space and to form a columella. The upper surface of the dissepi- 

 ments at the bottom of the calices is ornamented with snarsely 

 placed granules, and this very exceptional ornamentation is not seen 

 on the lower dissepiments, having been absorbed during growth. 

 The dissepiments are often very close near to the calices, and the 

 visible one at the bottom of the fossa is close above several others, 

 which seem to have followed the same lines of curvature. It is this 

 festooning of the dissepiments which, when seen in a vertical section 

 of the colony, adds to the old-fashioned appearance of the more 

 or less tumid mass at the base of the calicular fossa (figs. 7 and 8). 

 Probably it was the very tabulate appearance of the endotheca of 



