114 PERCY SLADEN TRUST EXPEDITION 



1899. Astrma rotumana, Gardiner, Proc. Zool. Soc. London, p. 750, pi. 47, fig. 3. 

 1899. Astrma affinis, Gardiner, Proc. Zool. Soc. London, p. 750. 



The two large specimens (the larger rDeasuring 19x15x9 cm.) from Funafuti, which 

 Gardiner referred to Astrma denticulata (Ell. and Sol.), the two small convex ones (the 

 larger measuring 9x7x6 cm.) from Rotuma, which he assigned to Astrcea pallida (Dana), 

 and the fragment from Wakaya named Astrcea affinis (Ed. and H.) resemble one another 

 so closely that there is hardly any doubt they belong to the same species. They do not 

 agree with any of the type specimens I have examined. It is, however, likely that this 

 species may have been previously recorded by Dana, but no satisfactory determination 

 is possible from his descriptions or figures. 



In all these specimens the corallum is massive and convex, the corallites are polygonal, 

 often irregularly compressed, close together but usually separated by narrow grooves. 

 Sometimes the walls are fused. The average diameters of the calices are 8 — 11 mm. and 

 6 — 8 mm. and the depth 5 or 6 mm. The septa are vertical, with rough sides and either 

 toothed or entire edges, their total number varying from 25 — 35 ; of these 9 — 15 (average 

 11 — 13) meet the columella, their upper two-thirds being narrower than their lower one- 

 third. At the calicular margins the principal septa are much broader (about 2 mm.) and 

 thicker ("5 mm.) than the subsidiary septa and are more highly exsert (l "5 — 2 mm.). (This 

 feature, as has been pointed out, is characteristic of many of Gardiner's Pacific specimens.) 

 The exsert ends of the septa of neighbouring corallites either meet in notches or alternate. 

 The columella is spongy, being formed of thin septal trabeculse and is from ^ to ;|^ the 

 width of the calices, sometimes quite rudimentary. New corallites are formed by equal 

 or sub-equal fission, which is usually effected by two of the thickened principal septa 

 meeting across the calices. 



The small specimen (10 x 6 x 4'5 cm.) from Rotuma, which Gardiner has made the 

 type of a new .species, Astrcea rotumana, is closely related to the above-mentioned 

 specimens but its corallum is lighter, the corallites somewhat further apart and the septa 

 fewer. These differences are perhaps to be accounted for by the fact that the specimen 

 was an edge-piece. It has also a general resemblance to my Salomon examples of Favia 

 clouei. 



24. Favia sp. ? 

 1899. Prionastrcea tenella, Gardiner, Proc. Zool. Soc. London, p. 761. 



A small convex specimen from Rotuma with an incrusting corallum 4 '8 cm. long and 

 4 '2 cm. broad, which Gardiner referred to Prionastrcea tenella, Dana, appears to be a good 

 species of Favia, being quite different from any other specimen I have examined. Its 

 corallites are polygonal, the diameters of the calices measuring on an average 5 mm. and 

 4 mm. and depth 3 mm. ; the inter-corallite walls are ridged, being aboiit 1 mm. thick. 

 The septa are sloping with uneven edges and rough sides, their total number varying from 

 20 — 26 ; alternating with these is a cycle of very narrow septa. 12 — 15 septa meet the 

 columella, of which 6 — 8 are broader than the others, these giving the corallites a star- 

 shaped appearance. All the septa are veiy slightly exsert, those of adjacent corallites 

 meeting over the inter-corallite walls. The columella is thin and spongy. Fission is 



