CORALS FROM MURRAY, COCOS-KEELING, AND FANNING ISLANDS. 133 



von Marenzeller, is the same as P. cactus (Forsk.). (I doubt the correctness of 

 Milne Edwards and Haime's statement that P. knorri has no columella.) As it is 

 important that this synonymy should be straightened out, I am publishing (plate 

 56, figs. I, I a) figures of the terminal part of a frond of P. formosa Dana, on natural 

 scale and twice natural size, so that it may be compared with plate 42, figure 77, 

 of von Marenzeller's Pola corals, and with Klunzinger's plate 9, figure 2. Pavona 

 danai M. Edw. and H., P. complanata Verrill, P. angularis Klz., and P. la.xa Klz., 

 will be discussed under P. danai, of which they are synonyms. 



The collection of specimens of Pavona in the U. S. National Museum is very 

 large, and probably contains all the recognized living species. Some years ago I 

 made a catalog of the species apparently referable to the genus, listed the types in 

 the U. S. National Museum, and prepared a synoptic table for the determination 

 of the species. As some of these notes may be of service to others, they are here 

 published. 



The following list shows the species assigned to Pavonia by Dana, the genera 

 to which at present referred, and the types or original specimens of those in the 

 U. S. National Museum, the latter being preceded by an asterisk: 



Verrill described Pavona foliosa and P. complanata, the types of which are in 

 the U. S. National Museum. 



An examination of a large series of Pavona at once shows that the species are 

 divisible into two large groups, viz: (i) frondose or branching forms; (2) forms 

 primarily incrusting, but which later may be massive. Group i is subdivisible 

 into three subgroups: (la) fronds with wide, flattish or gently curved sides, mar- 

 gins not greatly dissected; (i^) fronds crispate, summits divided into relatively 

 narrow, more or less curled lobes; (ic) branches narrow, angular, keeled, and 

 coalescent. These subgroups mtergrade and must not be considered as fi.xed. The 

 young ot a species may belong to subgroup lb and the fully developed colony to 

 subgroup la; for instance, P. danai is based on a young colony, while P. complanata 

 is the fully developed corallum. Without an adequate suite of specimens for com- 

 parison Verrill could not have known that the two belong together. Subgroup ic is 

 typified by P. divaricata Lam. 



Special bibliographic references need be given only for the species described 

 by Verrill, Brueggemann, and Ridley: 



Pavonia varians Verrill, see p. 141 of this paper. 



Pavonia foliosa Verrill, see Proc. Essex Inst., vol. 5, p. 44, Jan. 1867. This is a precise 



synonym of P. frondifera Lam., according to the specimens labeled by Dana. 



Distribution: Singapore; enormously abundant in the Philippines; Loo Choo 



Islands; Caroline Islands; Fiji Islands. 



