CORALS FROM MURRAY, COCOS-KEELING, AND FANNING ISLANDS. 97 



The general aspect of the corallum of this species resembles that of L. bottce, 

 except that the corallites are more crowded and not quite so prominent, but the 

 cahcular details are very different. The primary septa of the latter have not the 

 shoulder of the former, the septal edges are more clearly (but minutely) dentate, the 

 septal faces more roughly granulate, usually there are paliform teeth, and there are 

 a few papillae on the columella. L. immersa is distinctly smoother. The absence 

 of septal spines or conspicuous teeth near the columella and the absence or rarity 

 of columellar papillae strongly contrasts with the other species, except L. transversa, 

 but the presence of obscure dentations, especially on the secondary septa near the 

 columella, seems to indicate specific and not generic difference. 



The only discrepancies between the Cocos-Keeling specimen and the one which 

 Klunzinger described and figured seem to be that in the former a larger number of 

 septa (8 to 12) reach the columella, the septa are thick on the corallite walls, and 

 the walls are thick (not "dunn"). None of these characters is usually of specific 

 value. All the other characters stated by Klunzinger are present, but he unfor- 

 tunately did not describe the profile of the septal edges. His figure seems to show 

 a profile similar to the one exhibited in the Cocos-Keeling specimen, but I can not 

 be positive of this. 



Distribution. — Red Sea; Cocos-Keeling Islands. 



Genus ECHINOPORA Lamarck. 

 1816. Echinopora Lamarck, Hist. nat. Anim. sans. Vert., vol. 2, p. 252. 



Type species: Echinopora rosularia Lamarck, which according to Matthai is 

 a synonym of E. lamellosa (Esper). 



Echinopora lamellosa (Esper). 



Plate 32, figures 1, \a, specimen from Cocos-Keeling Islands; figures 2, la, Dana's type of E. undulata; figure 3, 



Dana's type of E. reflexa. 



1914. Echinopora lamellosa Matthai, Trans. Linn. Soc. London, 2d sen, Zool., vol. 17, p. 50, plate 8, fig. 6; 

 plate 14, figs. 2-6; plate 15, fig. 1; plate 16, fig. 6. 



The identification of the specimen described below is based on Matthai's 

 description and figures and a specimen identified by him and received by the 

 U. S. National Museum from Professor Stanley Gardiner. I have not checked the 

 synonym he presents. Echinopora horrida Dana is not a synonym of this species, 

 as suggested with a query by Matthai. 



In order to present the basis for referring Dana's E. undulata and E. reflexa to 

 the synonymy of lamellosa, the type of the former is illustrated by plate 32, figures 

 2, 2a; that of the latter by plate 32, figure 3. 



The following is a description of a specimen of Echinopora lamellosa from Cocos- 

 Keeling Islands: 



The species is represented by the distal portion of a folium, 55 by 70 mm. across; grow- 

 ing edges thin and translucent; inner broken edge 4 to 6 mm. thick. Living calices confined 

 to one surface, but some dead calices on the lower surface. 



Diameter of larger calices about 4 mm.; distance from one row to the next 4 to 6 mm.; 

 distance apart of calices in the same row ranges from adjoining up to 1.5 or 2 mm. Inter- 

 corallite areas longitudinally striate, with rough, stellately spinulose granulations along 

 the striae, about 10 stria to 5 mm. 



Septa in adult calice, 3 complete cycles, arranged as follows: the 6 primary septa are 

 somewhat thicker and average taller than the secondaries; they decrease somewhat in thick- 

 ness toward the columella, which they join. A prominent, thick, erect, spinulose granule 

 corresponds to the thecal end of each septum, or may rarely stand back a little outside 

 the thecal ring; within the calice is a thick, prominent, spinulose tooth, separated by a 

 notch from the mural granule; its inner edge falls perpendicularly, sometimes recurving to 



