UNKNOWN ATLANTIC OR WEST INDIAN PORITES. 101 



smooth, except wliere short, stout lobes arise, wliich takes place where the encrusting base 

 reaches the edge of its sujiporting substratum.* These lobes are 2 cm. thick where they rise 

 from the base, but they may begin at once to flatten, and when about 2 cm. high, to fork. 



The calicles, slightly indented and subcircular, are about 1 mm. in diameter, absent on the 

 rounded tops of the lobes which are undifferentiated, flaky-filamentous reticulum. The walls 

 everywhere consist of loose, open uprisings of this reticulum, whicli, when forming a wall, 

 is mostly angularly filamentous, showing thin trabeculse, and thin, smooth, horizontal threads, 

 here and there running together into small flakes. The wall-tliread running in this reticulum 

 is often incomplete, frequently has a pronounced zigzag. Its trabecular elements rise all over 

 tlie surface as tliin smootli rods, swelling into finely echinulate granules. The septa have tliis 

 same character as they project from the reticular walls. The pali are everywliere well 

 developed, as is also the central tubercle when present. This latter is, however, frequently 

 absent, and instead, a small, deep, open fossa is conspicuous to the naked eye. 



In sections the trabeculte are seen to be flaky beneath the surface, and running 

 continuously as twisted interrupted lamelljE. Tlie colour of the stock is blue-grey, with the 

 finely echinulate tips of trabeculie, septa and pali as whitisli specks. 



Tliis coral has other interests l^esides its colour, on which see remarks on P. Barbuda 1, 

 p. 50, and note on p. 143. 



One specially interesting problem relates to its method of growth. The larger part of the 

 specimen is encrusting, but it sends up short processes which begin to fork. The question 

 arises as to whether this encrusting portion is only a base of what would ultimately have 

 been a branching Pontes. It certainly looks like it, not only in the fact of the thick lobes 

 beginning to fork, but also in the facts (1) that tlie tips are of undifferentiated streaming 

 reticulum, obviously in rapid growth, and such as is seen at the tips of most branching forms ; 

 and (2) That tlie character of the skeletons of the calicles, a loose, ojjen, foaming or tossing 

 reticulum, almost Ijaffling one's powers of description, is quite typical of branching Poriics, l)ut 

 is not at all typical of the encrusting and massive forms — cp. the astrsoid group. 



If tliis reasoning is correct, it brings the specimen still nearer to the l^lue branching form 

 P. Barbuda 1, with which it has one other character in common, namely the deep circular 

 fossaj. On the other hand, the brancihes are very different in shape and thickness, and the 

 characters of the calicles differ in their skeletal elements. 



Although there is no record of the locality of this coral, its characters proclaim it as 

 belonging unmistakably to the West Indies. See further Table III., Ea and E6. 



a. Zool. Dept. 1906. 1. 1. 10. 



' In this case, a chip of a brick, and most of the processes seem to arise where the corallum is 

 flowing over one of the edges of the brick. 



