362 MR. S. PACE ON THE 
generally arise at the pointstof contact; and the specimen may 
thus acquire an appearance rather suggesting that of the Banyan 
tree with its numerous false stems. This type of growth has 
deen described as ZT. radicalis by Bernard*. : 
Not infrequently a cup becoming accidentally broken from its 
stalk continues to live in this detached condition ; and, if it has 
at the same time been inverted, subsidiary cups will often be 
formed upon its upturned lower surface. 
In addition to these modifications, due mainly to position, 
there are others which direct observation on the reef has enabled 
me to trace to another cause: namely, the danger of becoming 
silted up or clogged with the fine mud which is always a pro- 
minent feature on a coral-reef. This is apparently the greatest 
evil which a coral has to dread, and the structure of the corallum 
is frequently much modified in such a manner as to adapt it to 
life on muddy ground. Under such conditions the cup of a 
Turbinaria is often flattened out, the “ peltate ” type of Bernard, 
and the colony may even assume a convex form ; or else the cup 
may be cleft on one side, or perforated at its base, so as to 
render it impossible for any silt to lodge within it. 
Where the side of the original cup becomes cleft, one of the 
lobes thereby formed may extend round the outside of the cup, 
and, the growth of the other lobe being arrested, the corallum 
may take on a roughly spiral form, and silting will be obviated 
by the presence at the base of the corallum of a con- 
tinuous gutter by which any foreign matter will be carried off. 
Those forms which Bernard has termed the ‘“ Turbinarie 
JSrondentes” belong to this type of growtht. A very common 
method by which a Turbinarian defends itself against silt is that 
in which at an early stage the margin of the cup, or rather disc, 
becomes bent down at regular intervals or frilled{, so that with 
further growth, the details of which are susceptible of various 
modifications, a very perfect gutter system results. In what 
* Brit. Mus. Catalogue, pl. ix. 
| The specimens of 7. awrtcularis and T. calicularis figured by Mr. Bernard 
(Brit. Mus. Catal., pls. x. & xi.) are poor examples of this type. It is shown in 
its most perfect form by a specimen of Montipora (in which genus this modifi- 
Cation is much more common than in Turbinaria) figured by Mr. Saville Kent 
in his ‘ Naturalist in Australia,’ pl. xxiv. p. 146. 
{ The specimen of 7. peltata figured by Mr. Bernard (Brit. Mus. Catal., pl. vi.) 
shows this very well. 
