49 



genus, which may also be an exception. The difference in the genus Stcphanophyllia is only 

 a slight one; the axes of symmetry are wanting here, except the first one, i. e. the septum 

 of the fourth order. But in Leplopsammia new septa are formed, or can be formed, on both 

 sides of an older septum and this is a well-marked difference from all the other genera of the 

 family. Lacaze-Duthiers has pointed out that difference very clearly in hïs hgures 3 and 12 

 on Plate XI (Lac.-Duth. 1897). 



For systematical purposes I will make a difference between the terms order and cycle, 

 the former indicating the genetic sequence of 

 the septa as I have pointed out previously. The 

 septa of different orders are never complete 



xioaa 1 6 



\ 



\ 



ijl 8 J t> J 38)03 



tit'i f 3 5 t 5 1 S <■ 5 i S U5fcl 



Fig. 1. ortlers. 



Fig. 2. cycles. 



in the Eupsaiiimidae whenever it is possible 



that there are more septa of the same order. 



The term cycle — in systematical descriptions 



of more importance — I will use in the sense 



employed by most authors. The cycles may 



be complete or incomplete. The difference between the orders and cycles and their respective 



sequence are illustrated in the figures 1 and 2. 



The definition of the genus Thecopsammia is according to Pourtalès (187 i) "corallum 

 simple, attached, without costae, covered with a complete epitheca". In the genus Balanophyllia 

 there are species without any tracé of an epitheca, other species with a thin epitheca, leaving 

 the ribs still visible, and others again (e. g. Bal. gigas) with a dense epitheca covering about 

 three quarters of the corallum, through which no ribs are visible. So I am wholly in accordance 

 with Duncan (1873) who says: The epitheca, which is strongly developed in some species of 

 the genus (Baiauopl/ villa), affords but secondary characters for differentiation ; consequent! v 

 the genus must include Thecopsammia of Pourtalès. 



Dendrophyllia de Blainville. 



This genus is nearly allied to the genus Balanophyllia ; it is practically impossible to 

 distinguish young Dendrophv Ilias with only one calicle from true Balanophvllias. In the older 

 Dendrophyllia the corallum is alvvays compouncl ; the young corallites, formed by budcling, 

 remain connected with the parental stem. In Balanophyllia budding can take place too, but 

 the newly formed corallites tend to separate from the parental corallite and often do so, though 

 they can remain connected for a long time, may be during their whole life. 



The identification of the incrusting species is very difhcult. Many species are described 

 but not always satisfactorily. I have before me five different incrusting species - - at least I 

 think they are different. But I could identify only three as formerly described species and even 

 that with some hesitation. 



1. Dendrophyllia micranthus (Ehrb.). 



Oculina micranthus Ehrenberg 1834. 

 Dendrophyllia nigrescens Dana 1846. 



