CORALS FROM MURRAY, COCOS-KEELING, AND FANNING ISLANDS. 167 



ently only 6 septa, the directive much broader than the others. In the immersed corallites 

 the septa are more nearly equal. Corallum moderately porous; surface reticulate and 

 echinulate; wall striate and fragile. (The specimen appears worn.) 



"Another specimen in the Paris Museum, from the same locality, is laheWed M . pustulosa, 

 and if this should prove to be the type of M. -Edwards's species it is certainly not distinct 

 from the above. M. -Edwards, however, gives Seychelles as the habitat of bis M. pustulosa; 

 but so far as I can ascertain there is no specimen of the species from that locality in the 

 collection, and the description given agrees fairly well with this specimen. 1 he specimen 

 here referred to is a fine well-preserved form with the following characters: Main branches 

 3 cm. thick, much divided, with numerous spreading and tapering branchlets. Axial 

 corallites 2.5 to 3 mm. diameter, 3 mm. exsert. Radial corallites simple, tubular with 

 oblique aperture, of variable length up to 4 mm. and 1.5 mm. diameter, at an angle varying 

 from 60° to 80°, with short, nariform, labellate and mimersed ones between. This is the 

 arrangement on the younger branches; in other parts the tubular corallites become elongate 

 and proliferous, up to 2.5 cm. in length, 5 mm. diameter at the base, with tubo-labellate 

 bud-corallites at an angle of about 60°; the majority of the proliferations are, however, only 

 from S to 7 mm. long. This specimen appears to me to give the real characters of the 

 species better than the type. 



"A third specimen of enormous size, over i m. diameter, shows the proliferations still 

 more elongate, forming branchlets averaging 15 mm. in length. All three specimens are 

 from the Red Sea and were collected by Botta in 1837. 



"The moderate variation in size of the a.xial corallites and the variation in length of 

 the proliferous corallites or branchlets, according to age, have probably led to this species 

 being described under several names. It is significant that all the types of all the forms 

 {1M. pustulosa Edw. and H.) come from the Red Sea. 



"Indian Ocean; Red Sea; Keeling Island." 



Dr. Wood Jones has brought from Cocos-Keeling fragments of an interesting 

 series of specimens, which will be described serially: 



Specimen No. i {plate 6g, fig. 2): 



A branch, 137 mm. long; diameter of lower end, 12 mm.; diameter of axial corallite, 

 3.75 mm.; 3 branchlets given oft" about 36 mm. below tip. Axial corallite with thick costu- 

 late walls and 2 well-developed cycles of septa. Proliferous corallites perpendicular to 

 surface of branch confined to one side; from 3 mm. up to 8.5 mm. tall; diameter of axial 

 corallite of the proliferations from 2 to 3.25 mm.; lateral corallites labellate or tubo-labellate; 

 2 cycles of septa with prominent directives. The other protuberant radial corallites nearly 

 perpendicular to the surface of the branch, tubular, with the inner wall shorter than the outer, 

 or nariform; height, 1.5 to 3 mm.; diameter about 2 mm.; distance apart, 2 to 2.5 mm.; wall 

 thick, porous, apertures elliptical, but not greatly compressed; 2 cycles of septa, directives 

 prominent, secondaries small or rudimentary; walls costulate, perforate. Immersed coral- 

 lites usually show 2 cycles of septa. Coenenchyma costulate, granulate, perforate, flaky. 



Regarding this specimen Dr. Wood Jones says: "Fragment of a colony growing in a 

 site in which much sediment is carried in the water." 



Speci?nens No. 2 (plate 6g, figs, j, ja).- 



Two specimens experimented with by Dr. Wood Jones in studying the repair of injury 

 to corals. The larger specimen is So mm. long; diameter of lower end, 25 by 28 mm.; of 

 upper end, about 21 mm. These differ from No. i chiefly by having the corallites so crowded 

 that over large areas the walls of adjacent corallites touch. There are many labellate coral- 

 lites between the protuberant ones. The texture of the walls is more delicately reticulate 

 than in No. i. 



Specimen No. j: 



This is another of Dr. Wood Jones repair specimens. It is similar to No. 2, except 

 that there are no protuberant corallites; all are crowded and tubo-labellate or immersed. 



Specimen No. 4: 



This specimen is one obtained by Dr. Wood Jones in the still water of the lagoon and 

 twice figured by him.^ It is a branch about 81 mm. long; diameter of lower end, 13 by 



'Proc. Zool. Soc. London for 1907, plate 27, figure 2b (left-hand figure). Coral and Atolls, p. 84, figure 24^1 

 (left-hand figure), igio. The right-hand one of the two figures represented by "h" is another species, Acro- 

 pora pulchra (Brook). 



