50 MADKEPORARIA. 



reticulum, thus disguising [he radial syinuuitry. The pali are sometimes regular, but mostly 

 very irregular as small knobs ; here and there they are replaced by a stout, prominent, skeletal 

 ring, which, owing to the fact that it may surround a deep, open fossa, is very conspicuous even 

 to the naked eye. These columellar rings seem to be most frequent where the axial reticulum, 

 which comes to the surface at growing tips, is passing into adult calicle formation. Small 

 central tubercles appear where the columellar tangle has no such open fossa. 



The sections of the basal stump and of the branches show the comparatively thin axial 

 reticulum surrounded by dense coral perforated by but a few canals. The skeleton is coloured 

 a faint blue-grey. 



The blui.sh colour of this coral is interesting. As is well known it is characteristic of 

 the alcyonarian Heliopora, but according to my experience it is rare in the Stony Corals. I have 

 found one other Poritcs, seo P. West Indus x.'24,\). 101, from this region, and I recall one from 

 the Indo-Pacific area, P. Great Barrier Reef 37, Vol. V. p. 140, which, curiously enough, has a 

 growtli-form very commonly seen in Heliopora. Besides these, a few Madrepores are known 

 showing various shades of blue, and there is the well-known figure of Turhinaria dnerasccns of 

 Ellis and Solander. I am not aware that any investigations have been made as to the cause 

 of this variation in the colour of the Stony Corals which on being bleached are typically sn(jw 

 white. 



The locality of the other specimen referred to (see Porites West Indies x. !34) is not 

 recorded. This is most unfortunate, for although its calicles and its skeletal texture are 

 entirely different, yet it has one startling peculiarity in common, namely the conspicuous 

 columellar or palic rings surrounding deep fossa?. In both corals these minute rings are visible 

 to the naked eye, but in the form above described they occur only in clusters, especially near 

 the growing tips ; in the other form they are almost universal (see further. Table III. E, a, 

 and E, b. 



a. Zool. Dept. 99. 6. 26. 8. 



32. Porites Barbuda 2- {f- Barhida. snunda.) (PI. II. fig. 9 ; PI. X. fig. 6.) 

 [Barbuda, coll. Gregory ; British Museum.] 



Description. — The corallum is small and the forking is so close and irregular that the 

 cluster appears to have been low and contoited. In the single specimen the main stem, 9 mm. 

 in diameter, swells suddenly into a knob 1'5 mm. in diameter. From this swelling, in 

 addition to two blunt projections (? aborted attempts to fork), two constricted stems carry up 

 the coral. These commence at once to swell again and to fork, and so on. The terminals are 

 flattened, from 5 to 7 mm. thick and 12 mm. broad, and point in all directions. The extent of 

 the living layer is unknown : at least 3 cm. 



The calicles are small, 1 mm. in diameter, shallow and for tiie most part as irregular 

 breaks in a ragged smface of flaky reticulum ; here and there, e.g. on swelling sui-faces, they are 

 deeper and more regular. The walls seem everywhere composed of horizontal, flaky reticulum, 

 the meshes showing a tendency to be neat, circular- pores ; they vary in thickness, here and 



