106 PROF. P. MAETOT DUNCAK ON THE 



resting-place. The width apart of the calices has been remarked 

 upon, but it would be consonant with the truth to state that the 

 calices have been worn down, and that what are now seen are trans- 

 verse sections of corallites produced by wear and weather. If the 

 specimen is compared with the section of an aged form of ^. gihhosa, 

 the resemblance in the position of the lumen of the corallites becomes 

 evident. Some of the spaces marked by the septa are wide apart, 

 and others are closer, and there has been infiltration of calcareous 

 matter into the cavity of the corallite, which diminishes the calibre 

 of the lumen. 



In some places the lumen of a corallite may be separated from 

 that of another by a space equal to one quarter of its diameter, in 

 others by a space equal to a whole diameter. But no intermural 

 cellular coenenchyma can be traced, and it is evident that the whole 

 structure occupying the space is mural, that the thick walls are in 

 contact, and that they were inseparably united during life. The 

 circular appearance of the lumen of a corallite is deceptive, and it is 

 evident that the joined corallites were polygonal. Probably the 

 type was a slow -growing colony, and it is quite possible that a 

 number of individuals settled closely together, and that they united 

 by their walls ; for it is not explicable how gemmation from one 

 parent could have produced a wide and low form *. 



ASTEOCGEI^IA PLAl^A, UoHs, Op. cit. p. 19, pi. V. fig. 1. 



The type of the species is in the Museum of Practical Geology, 

 Jermyn Street. It is in sufl&ciently good condition to show the 

 polygonal shape of the corallites at the calicular surface, and to indi- 

 cate that the calices are rather deep and separated by a varying 

 width of more or less tumid, and invariably raised, united walls, over 

 which rather long costse pass. The breadth of the polygonal corallites 

 is less than in A. gihhosa, and the comparatively flat surface of 

 the corallum permits of great regularity of the lines of the calices. 

 In no instance are there circular corallites in the central and other 

 parts of the surface, where there is the usual crowding ; and usually 

 there is no great distance between the lumen of one corallite and that 

 of another, except where there is a space with three or four calices 

 in a group. The outlines of the cora^llites may be hexagonal or 

 nearly square ; and whilst three of them, with the walls, occupy a 

 length slightly over three millimetres, the intermediate walls may 

 be one third of the width of a calice, one half, or, in rare cases, two 

 thirds. In the majority of instances the wall is only one sixth of 

 the diameter of a corallite. 



The fossilization has filled up much of the corallites, and the colu- 

 mellas are therefore larger than normal. This may lead one astray 

 in dealing with the number of septa which normally unite with the 

 columella. The following are the numbers of septa in difi'erent 

 calices: — (1) 10 septa all uniting with an enlarged columella ; (2) 10 



* See Journ. Linn. Soc. Zool. vol. vii. p. 365, 1883, for some remarks on 

 the origin of similar colonies. 



