﻿GLABROUS MONTIPOR^. 39 



of a former growth. One section is hollow and lined with epitheca — a fact which links the 

 specimen closely with the foregoing. 



The calicles are rather more conspicuous (as pits of all sizes under 1 mm.) than in speci- 

 mens a to c. Probably on account of their branched form, the specimens are rather denser and 

 thus heavier than the more massive forms. 



fj. Macclesfield Bank, China Seas, wide lagoon, Coll. Bassett-Smith. 93. 9. 1. 75. 



sand. 



This is only a small fragment from quite a different locality, the method of growth of wliich 

 deserves separate description. In all essentials it agrees with the foregoing specimens, i.e. in 

 reticular character of the coenenchyma and size of calicles. A horizontal expanse, 2 mm. thick, 

 supported by an epitheca, rests across the tips of two pillars rising up perpendicularly 2'5 cm. 

 in height from the level surface of a previous growth of the same. The process is closely 

 similar to that exemplified by the specimen of M. oUusata (p. 33) from Fiji. The lower 

 growth is covered with white film, but the tips of the pillars have grown out into fresh growth 

 in several places. Indeed, it seems probable that the new stock is due to a table-like expan- 

 sion of the top of one or more of these pillars. 



21. Montipora divaricata. (PI. III. iig. 4; PI. XXXI. fig. 17.) 

 Mmtipora divaricata, Briiggemann, Phil. Trans., clxviii. (1879) p. 577. 



Description. — Corallum an irregular nodulated mass, built up of successive layers of coral, 

 each layer rising up at the free edges and on the summits of prominences into fresh tufts of 

 branching lobes. The ultimate branchlets are irregular in shape and distribution, and average 

 about 1 cm. in diameter and about the same in length. 



Calicles are conspicuous and numerous, slightly under 1 mm. in diameter and about the 

 same distance apart.* On the upper faces of the stock, the apertures are not very sharply 

 defined but are slightly sunk in the reticulum of the coenenchyma. On the lower faces, tliey 

 are sharply defined and open flush with the level interstices. Two cycles of short, stout sejita, 

 the secondaries being slightly shorter than the primaries. Fossa deep and open. 



Coenenchyma, a rather dense nodulated reticulum, open and slightly surging up in the 

 interstices on the upper surfaces of the stock, but more solid and smooth on the lower 

 surfaces. The axial streaming layer in sections of branches not sharply marked off from the 

 cortical layer. 



There is only one specimen of this coral, viz. Briiggemann's original type. In describing 

 it, it was said to be allied to M. rus and M. monasteriata, but to differ from these in " mode of 



* Briiggemann says the calicles are 2 mm. apart. It is true that round the bases of the tufts 

 and on the lower sides of slanting tufts, where the ccBnenchyma is becoming denser and conse- 

 quently narrowing the calicle apertures, the latter are further apart ; but on the upper faces of the 

 mass they are as often as not less than 1 diameter distant from one another.. 



