Study of Madreporarian Types of Corals. 225 



with the Oculinidte and Pocilloporidse into a very natural 

 affinity with the Turbinolidre. The coenenchyme of these 

 colonies is treated by the author as an elaboration of a primi- 

 tive eostra-calycinal deposit around individual polyps. Bourne, 

 in one of his papers, suggests the possibility of coenenchyme 

 being epithecate in certain of these types. 



In order to elucidate the " costate " portions of Astrasid and 

 Fungid colonies the author demonstrates the exact homology 

 of skeletal parts in the calyces of ancient Cyathophyllids and 

 of recent Astrseids and Fungids. The pseudotheca which 

 appeared in Acervularia and other Cyathophyllid types 

 marked out an inner from an outer area of the calyx and 

 septa. The exact counterpart of this is found in the type- 

 genus of the Astr£eida3, Hdiastrcea. In it, however, only the 

 inner part is called calyx, while the outer area is spoken of as 

 a " costate " — ewtra-calicinal — area. It is on this outer area 

 that the so-called " Randplatte " (which the author translates 

 as " edge-zone ") is supported in the living polyp, and the 

 author takes it that the typical edge-zone has mesenteries and 

 mesenterial loculi simply because it was originally an inherent 

 part of the polyp. It is clear that such costate parts in 

 Astrseid colonies have an entirely different evolutionary 

 history from the coenenchyme in the Pocilloporidaj &c, where 

 no edge-zone surrounds the polyp. The author traces back 

 this difference in recent colonial types to a difference already 

 well-marked in (Silurian Madreporaria — viz. the difference 

 between the calyces of a typical Cyathophyllid and a typical 

 Zaphrentid respectively. In the former a broad calycinal 

 outer zone with dissepimental base surrounds an inner tabulate 

 area ; in the latter there is no such outer zone or the very 

 slightest indication of it. 



The author found that the families of Edwards and Haime's 

 classification must undergo considerable changes ; she limits 

 herself here to one or two of the most important changes sug- 

 gested by her on the basis of microscopic septal structure and 

 generally of the morphology of the skeleton. The family of 

 AstreeidaB, E. & H., hitherto included two main subfamilies, 

 the Astrseinse and Eusmilinse. The former is made by the 

 author the sole representative of Astraidse, while the latter is 

 entirely broken up. The genera Trochosmilia, Placosmilia y 

 and their allies are referred to the family of Turbinolidse ; the 

 genera Rhipidogyra, Pectinia, Dendrogyra, Euphyllia, and a 

 large number of fossil genera are placed in a new family, 

 Amphiastrseidse, erected by the author. The Mesozoic repre- 

 sentatives of this new family are proved to be direct colony- 

 building descendants ot Palaeozoic Zaphrentids, while the 



