MACDO'ALD ON HETEROCTATHUS AND SIPUXCULUS. 79 



of the columella. Tlie secondary septa are furnished with laminre of 

 tlie same description, which join those of the first set, at an acute 

 angle, without reaching the columella ; and the tertiary septa pass 

 into this point of union, having no supplementary laminae of their 

 own. The two sets of plates, just noticed, pi-esent a rounded shoulder 

 internally (more prominent in the primary ones) giving them the 

 character of lateral pali, or dismemberments of the septa. 



The columella is composed of a spongy tissue, with an oval and 

 slightly convex summit. 



All the plates of the disk are spongy, or minutely granular, on the 

 siu-face, but compact within. The body of the corallum is spongy at 

 the axis, in contiiuiity with the columella, more compact below and 

 around this, and again more porous towards the exterior, especially 

 above. 



The loeuli are circumscribed, but not crossed by synapticults or 

 interseptal dissepiments. They are just double the number of the 

 septa, lying one on either side of the latter, and are thus ai*ranged by 

 pairs in three distiuct circles ; the internal corresponding with the 

 primary, the middle with the secondary, and the external with the 

 tertiary rays. 



In the species taken at the Bellona Shoals the oral disk was dis- 

 torted, with a central constriction, as though a process of fission had 

 been going forward. In one specimen indeed the opposite margins 

 of the disk had actually coalesced. The primary septa were twelve in 

 number, and all the plates are so much compressed that the loeuli 

 are exceedingly narrow. Tlie external surface of the corallum is 

 beset with minute graniJations disposed in broken longitudinal liues 

 with porous channels between them ; on the other hand, iu the Pee- 

 geean species the disk is regular, with six primary rays and wider 

 loeuli, and the external surface of the corallum is coarsely gi'anulated, 

 without any very obvious linear disposition, as the first rudiments of 

 costce. 



In a recent visit to Moreton Bay we di'edged (ia a few fathoms 

 depth) two beautiful specimens of another species of this genus, 

 differing from the foregoiug iu having well marked longitudinal costce, 

 exactly forty-eight iu number, and corresponding, each for each, with 

 all the radiating sep)ta and lamincE, with which they are directly con- 

 tinuous at the margin of the disk. The principal lamince are falcate 

 towards the hollow of the cup and deeply notched, toothed and 

 echinate, as they pass into the spongy columella, whose actual limit 

 is thus rendered less defijiite than in the other species described. 



Of the soft parts of these polyps, I can say but little. They 

 appear to be very scanty, from the fact, that when the animals are 

 immediately taken from the water there is scarcely anythiug to be 

 seen but a brown, soft and tenacious matter, filling up the crevices of 

 the skeleton above described, and all the prominent points and ridges 

 become quite bare. Tlie whole surface of the corallum is covered 

 over with a thin ectodermic layer, which however is much worn at the 



