FROM THE ZONE OF AMMONITES ANGULATUS. 25 



The reproduction by gemmation cannot occm' from the walls of the corallites, except 

 at the edge of the corallum. The close contact of the walls, and the existence of dense 

 coenenchyma, prevent any budding from the wall ; but where the outside corallites are 

 partly free, there gemmation may occur outside and below the calicular margin. 



Fissiparity does not occur, but the young buds arise either from the top of the calicular 

 edge or margin, or just within the calice. When there is some distance between the 

 calices on account of the thickened walls or coenenchyma, buds may arise on the coenen- 

 chymal or inter-calicular surface. 



Many of the species have an epitheca, some are pedunculated, and others are 

 massive, encrusting, or dendroid. 



The septa vary greatly in their numbers and cyclical arrangement, and very often 

 they have a large paliform tooth close to the columella. There are no pali. 



A styloid columella projecting more or less, is an essential generic requisite. The 

 endotheca is scanty, but it always exists. 



The calices are small, and vary in depth ; but, as a rule, they are arranged with great 

 symmetry, and are polygonal in outline. Transverse sections show the complete consoli- 

 dation of the walls, and the space between the costal ends, in these sections, is often 

 marked with granules. 



The species without any coenenchyma, and whose walls are thin, are distant in their 

 alliance to Stylocosnia, and had they no columella, they would be considered to belong 

 to the genus Isastraa. The genus Cyathoccenia (Duncan) comprehends Astroccenice 

 without columellas. 



The fossil condition of the specimens must be considered during the specific deter- 

 mination of AstrocosnicB. Usually, the columella is represented by a flat, central, and 

 more or less circular mass with the ends of thick septa adherent to it. In these in- 

 stances a calcareous deposition has occurred around the columella and between the 

 septal ends, the columella having been broken off. It happens, however, that the 

 columella may be broken off without the deposition having taken place, and either the 

 structure retains its normal size at the point of fracture, or is absent altogether. 



On examining doubtful specimens which have lost their columellse, much attention 

 should be paid to longitudinal sections produced by weathering, fracture, or by artificial 

 means. A small projection at the base of the calice is more readily determined to exist 

 in longitudinal views than in those which simply show the open calice. 



There are eleven species of the genus Astroccenia special to the Welsh Lias, and one 

 species found with these has been described byD'Orbigny &?, Stephanocoenia Sinemuriensis. 

 M. D'Orbigny obtained his specimens from the Lower Liassic deposits of France. M. de 

 Fromentel and MM. Terquem and Piette have found the species in several localities, 

 and the first-named palaeontologist has determined it to belong to the genus 

 Astroccenia. 



The Liassic AstrocoenicB occur as large and massive, small and dendroid, or as irregu- 



