PAL.-EONTOLOG V. 



present purpose, most appropriate to adopt his system as a basis. 

 It requires, however, important rectifications, which I will make as 

 I proceed with my descriptions. 



CLASSIFICATION OF POLYPES. 



1!V MILNE-EDWARDS. 



I. Corallia. 

 11. Hydroida : soft, not represented in fossil condition. 



The Corallia are divided into three orders: 



1. Zoantharia. 



2. Alcyonaria. 



3. Podactinai'ia. 



Of these, the first order includes the principal part of all the fossil 

 corals; the second is only represented by Graptolites, and the third 

 has no palaeozoic representative. 



The Zoantharia are divided into seven sub-orders : 



1. Malacodernia : soft, not fossil. 



2. Apora : recent coralline forms. 



Milne-Edwards placed the genus Palceocyclus with this sub-order, 

 but its afifinities are decidedly nearer to forms placed in another sub- 

 order, the Z. rugosa. 



3. Perforata. — Abundantly represented in the mesozoic and re- 

 cent coralline fauna, but not in the palaeozoic. The genus Protaraea, 

 placed here by Milne-Edwards, belongs to the next following order, 

 the tabttlata, and the genus Plcurodictiiini, likewise enumerated 

 among the Zoanth. perforata, is created by simply mistaking the 

 casts of a Michelinia for a particular type of organization, very ap- 

 propriately connected with the specific by-name of ^'' problcmatiaiviy 



4. Tabulata. 



5. Rugosa. 



These last two orders comprise nearly all palaeozoic corals, and 

 will form the special object of consideration in subsequent pages, 



6. Tubjilosa. — An order formed to include the genus Aulopora 

 and Pyrgia, both of which genera are in intimate relationship with 

 certain types placed under the Zoanth. tabulata, with which I am 

 going to describe them. 



