90 PAPERS FROM THE DEPARTMENT OF MARINE BIOLOGY. 



usually shows a papillate columella, without such a partition, while a few calices 

 appear to show one. A lamellate columella is rather constantly present in Klun- 

 zinger's figure of L. transversa, but he explicitly states: "Die Columella besteht 

 nur aus wenigen Papillen, welche zumeist verwachsen sind und eine quer durch 

 den Kelchgrund ziehende, zwei der Septa verbindende Lamelle bilden * * *" 

 Milne Edwards and Haime indicate a papillate columella for L. roissyana. Matthai 

 says of L. roissyana "septa swollen in the theca," of L. ehrenbergiana "septa not 

 swollen in the theca." His figure (plate 17, fig. 6,) of the latter shows septa thick- 

 ened in the theca. 



There are in the U. S. National Museum 58 specimens of Leptastrea, which 

 I am classifying in the following species: Leptastrea purpurea (Dana), L. trans- 

 versa Klz., L. bottce ( M. Edw. and H.), L. immersa Klz. 



I believe Matthai is right in referring Cyphastrea hotta and Baryastrea solida 

 to the same species, but the name must be Leptastrea hotter, notwithstanding that 

 "the second larger example referred by Milne Edwards and Haime to the same 

 species is a true Cyphastrea (most probably C. serailia)," for their figured specimen 

 must be taken as the type, and Klunzinger in 1879 eliminated the Cyphastrea from 

 the species, should there have been confusion of forms. The name bottce antedates 

 solida by a month, as is shown in the synonymy on page 94. I doubt Milne Edwards 

 and Haime having used the specimen of Cyphastrea as a cotype, because they 

 query the reference of bottce to Cyphastrea, and in their description state: "Cloisons 

 un peu debordantes, inegales, peu serrees, comme tronquees en haut."^ They espe- 

 cially noted the truncate upper margins of the septa, one of the striking characters 

 of the particular variant of the species represented by typical L. bottce, and one 

 which in my experience does not occur in Cyphastrea serailia. I agree with Matthai 

 that Leptastrea incequalis Klz. is a variant of L. bottce, as the suite of specimens in 

 the U. S. National Museum shows that they intergrade. 



SYNOPSIS OF CHARACTERS OF SPECIES OF LEPTASTREA. 



Before considering in detail the four species here recognized, their more salient 

 differential characters will be indicated. 



Two of the species, L. purpurea and L. transversa, have conspicuously polygonal 

 corallites and calices (which are often deformed) and no free corallite limbs. In the 

 former the grouping of the higher cycles of septa around the lower is usually obvious. 

 The septal edges are conspicuously dentate, the septal faces show distinct granula- 

 tions, and the columella is papillate. In the latter, septal grouping is rare, the septal 

 edges are entire or only with microscopically fine dentations, the septal faces are 

 nearly smooth, and the columella is either without distinct papillae or they are only 

 slightly developed. Usually there is a lamella extending lengthwise of the calice 

 between the directive septa, while the principal septa outside the directive plane 

 fuse to its sides. 



The other two species have subcircular corallites and except in typical L. 

 bottce they have free coraUite Umbs. The septa of L. bottce are dentate, although 

 the dentations may be very fine; the septal faces are beset with distinct granula- 

 tions; and there is a papillate columella. L. immersa has septa with subentire 

 margins, almost smooth faces; the inner edge of the principal septa fuse loosely 

 in the bottom of the calice; and there are few or no papillae on the upper surface 

 of the columella. In its smooth septal surfaces L. immersa bears to L. bottce a rela- 

 tion similar to that which L. transversa bears to L. purpurea. 



'Not italicized in the original. 



