330 CHARACTERS AND DIVISIONS OF ALCYONARIA. 



occasionally present ; and no traces of proper " tabulae " (in the strict 

 sense) have been noticed to occur. There are also no structures 

 which admit of being regarded as septa. 



As regards the microscopic structure of the corallum of Tubipora, 

 the entire calcareous skeleton is permeated throughout by a system 

 of minute parallel tubules, which sometimes branch, and which 

 open on the surface by well-defined rounded apertures (fig. 2 1 3, a 

 and b). These canaliculi run at right angles to the walls of the 

 corallites and also to the connecting-floors, and they render the 

 skeleton of Tubipora completely and minutely porous. Moreover, 

 thin sections show that the skeleton is made up of a network of 

 irregular fusiform spicules which are firmly united with one another 

 (fig. 213, c). 



The genus Tubipora is not known to occur in a fossil condition, 

 but the structure of its skeleton is a matter of considerable palaeon- 

 tological importance, as high authorities regard the genus as the 

 closest living ally of the Palaeozoic genus Syringopora. Reasons 

 have been previously given (p. 320) for considering that Syringopora 

 and its allies are closely related to the Favositida, and that they 

 truly form a group of the Madreporaria Perforata. It is sufficient, 

 therefore, to point out here that the skeleton of Tubipora differs 

 fundamentally from that of Syringopora in the following points : — 



a. The entire skeleton is traversed by minute canaliculi, no 

 traces of which are found in the compact corallum of Syringopora. 



b. The skeleton is composed of distinct spicules. 



c. The axial tube, when present, is not formed of invaginated 

 " tabulae," as in Syringopora, and there is no evidence of the exist- 

 ence of proper " tabulae " at all. 



d. No proper septa exist, whereas well-developed spiniform septa 

 are present in Syringopora. 



Upon the whole, therefore, it would appear that there is no suffi- 

 cient evidence at present available which would support the reference 

 of Syringopora and its allies to the Tubiporidce. 



Gorgonid^e. 



The family of the Gorgonidce, as defined by von Koch, includes 

 those Alcyonarians in which the colony is fixed, and is sup- 

 ported by an axial sclerobase, which is secreted by a continuous 

 layer of epithelium. The sclerobase may be purely horny (as in 

 Gorgonia itself), or it may be composed of alternate horny and 

 calcareous joints (as in Isis). The polypes are all similar to one 

 another, and polymorphism has not been observed. This family is 

 of little palaeontological importance ; but the genus Isis has been 

 recognised in deposits as ancient as the Cretaceous. Remains re- 



