226 A. E. Verrill — Corals of the Genus Acropora. 



branches showing almost exactly the structure of the type, though 

 other branches may be as usual. Therefore I conclude that the 

 difference is at most only a local variation. 



3f. laxa Lam., as described from the type by Brook, seems to 

 differ very little from the Singapore variety of this species, and it 

 may, perhaps, be identical. It has not been figured. 



Acropora austera (D.) Ver. See p. 313. 



Maclrepora mistera Dana, Zoopli., p. 478. Brook, p. 56. 



Plate XXXVI. Figure 10. Plate XXXVI B. Figure 1. 



Branches of Dana's type are in the Yale Museum. These show 

 that the axial corallites have thick reticulate walls and 12 wide, thin 

 septa, the 6 primaries nearl}^ meeting. No. 4190. 



The radial corallites also have 12 septa, the secondaries narrow. 



The walls of the larger corallites are very porous and costulate, 

 but the costulse are thin and spinulose, and more or less interrupted 

 by the lai'ger pores between them ; toward the margin, the walls are 

 often reticulate or fenestrate, owing to the large pores. But the 

 smaller radial corallites often have the walls strongly and roughly 

 echinulate, without costuloe. 



The surface of the coenenchyma is openly porous or reticulate, 

 and very scabrous, with coarse and irregular spinules. 



I have seen no specimens, except Dana's type. Brook, apparently, 

 had no specimens of it, unless he put them under some other species, 

 as is not improbable. He refers to it, doubtfully, some worn frag- 

 ments only. 



Acropora dissimilis Ver., sp. nov. 



Maclrepora echidncea Dana, Zooph., p. 458, pi. xxxi, fig. 9, pi. xxxv, fig. 3, 



1846, non Lam., Ehr., Stud. 

 Madrepora rosacea Studer (pars). Brook, op. cit., p. 84, (non M. rosea, as 



on pi. XV, nee rosacea Esper, p. 115). 



Plate XXXVI. Figure 9. Plate XXXVI A. Figure 9. 



Studer was undoubtedly correct in stating that the A. echidncea 

 (Lam., and of Ehr.) was a very distinct species from Dana's, but he 

 was wrong in uniting the latter with Esper's 31. rosacea, which has 

 very exsert radial corallites. 



Dana's species is closely allied to cerealis, appressa, and assimilis. 



It has unequal, tubular, loosely arranged, distal radial corallites ; 

 the longer ones but little appressed, and with oblique margins ; the 

 calicles open inward and upward. 



