ANTHOZOA. ZOANTHARIA 87 



agree in that respect with living Madreporaria, so that 

 it is probable that both have descended from the same 

 ancestors; the difference in the mode of development of 

 the later septa, however, seems to indicate that the two 

 groups soon diverged. If this view is correct then the 

 Rugosa must be regarded as a natural division. 



Other writers, however, maintain that the Rugose corals 

 do not form a natural group, since the bilateral symmetry, 

 the fossulse, etc., which were regarded as characteristics, are 

 not in all cases found in adult specimens ; and further, 

 some of the Mesozoic genera of Aporose corals are stated 

 to possess fossulae and a bilateral symmetry, and to show 

 in the young stages a pinnate arrangement of septa like 

 that found in Rugose corals. Also some of the families of 

 Aporose corals, particularly those found in the Trias and 

 Jurassic, are said to possess features similar to those of 

 certain Rugose families of the Palaeozoic period, from 

 which it is inferred that they are the descendants of the 

 latter. Thus, for example, the Astraeidae (Aporosa) are 

 believed by some authors to be closely related to the Cyatho- 

 phyllidae (Rugosa), and the Turbinolidae (Aporosa) to the 

 Zaphrentidae (Rugosa). If this view of the affinities of 

 Aporose and Rugose families is correct, it is obvious that 

 the Rugosa cannot be regarded as a natural group which 

 became extinct or nearly extinct at the close of the 

 Palaeozoic period, but that different Rugose families are 

 the ancestors from which a number of Aporose families 

 have sprung independently. 



Until more is known of the affinities of the Rugose 

 corals it will be convenient to divide the genera described 

 in the following pages into Rugose, Aporose, and Perforate 

 groups. 



