1898,] OOBALS FROM THE SOUTH PACIFIC. 535 



elongated vertically and close set, which rises almost to the surface 

 of the corallum as in P. haimiana. There is also a row on either 

 side, distinct near the surface, but much thickened below, forming 

 with the central row a broad wall. Additional rows, still deeper, 

 have likewise thickened, so that the corallum appears in section 

 to be extremely dense. 



I have compared P. savigniensis (PI. XLV. fig. 4) with j specimen 

 of Ooscinarcea uioniU, Forsk. (syn. 0. mceandrina Ehr.) in the 

 British Museum. The superficial resemblances between the two 

 species are very great, P. savigniensis differing mainly in the size 

 of its calices. The basal wall differs, however, from that of the 

 specimen described by Duncan ^ in being without costje, and is, I 

 consider, an epitheca. Prom the surface no central row of 

 synapticula can be distinguished between the fossse in either 

 P. savigniemis or Coseinarcm monile, but fractured and ground 

 surfaces of the former show that there is a distinct central row 

 with a well-marked row^ on either side. The synapticula are not 

 so thick or so close set as in P. haimiana, and, further, scarcely 

 thicken at all deep down in the corallum, so that the spaces between 

 the rows are little obliterated, in this resembHng P. ohtusangula. 

 The septa, too, in P. savigniensis are very distinctly ridged, and 

 much perforated in the valleys, as in Mcmndroseris hotta;, while 

 the axial fossse are closed below in precisely the same way as in 

 P. haimiana. 



The genus Plesioseris was separated by Duncan " from Mcmn- 

 droseris on the ground that its calices have a distinct wall and very 

 slightly trabeculate and imperforate septa. The examination of 

 the gemmation in P. profundacella and P. superficialis has shown 

 me that budding usually takes place by the body- wall of the polyp, 

 somewhere over the septa, forming a new calicular centre. The 

 septa, then, between the nQW and the old fossse are built up and 

 joined at the same time by synapticula, so that finally a distinct 

 though slightly perforated wall is formed ; ridges result by the 

 budding again and again of the mother-polyp before the w^all has 

 had time to grow up. In P. superjicialis are a number of such 

 ridges, but the calicular centres are not arranged in any determinate 

 order in respect to them, while in P. jirofundacella the calices are 

 single or in short series, completly separated by such ridges. 



From a consideration, then, of the resemblances of the above 

 species I propose to absorb the genera Mcmndroseris, Plesioseris, 

 and Ooscinarcea into the geuus Pscnmnocora, Dana. 



The diagnosis of Psammocora in accordance with the hard 

 parts would be as follows : — Colonial Fungida primarily incrusting, 

 but later massive or foliaceous. An epitheca is present as an 

 imperforate basal wall, but there is no true theca. Calicular fossa) 

 distinct, closed in below by trabeculse arising from the septal 

 edges. Synapticula numerous and stout, in vertical series, often 

 appearing to form a wall between the calicular fossse. Septa 



' Loc. cit. p. 314. '^ Loc. cit. p. 3U9. 



