EXPERIMENTS ON INCRUSTATIONS. 
311 
with the following results : When put into dilute nitric acid, a portion 
of them dissolved with great effervescence, leaving a residuum of fine 
sand of a white colour at the bottom of the vessel, covered by 
a layer of brown flocculent matter. The supernatant liquor was 
quite transparent. On treating this liquor with pure ammonia, no 
precipitation took place.* Treated with carbonate of ammonia, it 
threw down a white precipitate, which proved to be carbonate 
of lime. The layer resting on the sand was dissipated by heat, 
and burnt with the smell of vegetable smoke. The sand was com- 
posed of grains of quartz. These were the general results of the 
experiments, whether made with the outer coat or the interior 
part of the substance. When the outer part of the crust was used, 
the proportion of sand increased ; when the inner part was used, the 
dark-coloured flocculent layer increased. Subjected to the action of 
the blowpipe in their entire state, these substances, under a gentle 
heat, first blackened; at a higher heat threw off the smell of a ve- 
getable smoke, and became perfectly white. From these experi- 
ments it may, I apprehend, be deduced, that they consist of car- 
bonate of lime, quartz, and vegetable matter ; a conclusion that 
accords with the circumstances under which they are found, and 
their frequent arborescent character. The New Holland speci- 
men, subjected to a similar analysis, gave precisely similar results. 
As this specimen, for which I am indebted to Mr. Brown, who 
brought it from Bald Head, has a remarkable resemblance to coral 
both in form and closeness of texture, it may perhaps be considered 
a fair example of those substances considered coral by Vancouver, 
Flinders, and Peron. If this be admitted, it will follow that the 
reasoning is incorrect which is founded on their supposed submarine 
origin. 
* This statement requires to be somewhat modified ; for occasionally a little oxide of 
iron was thrown down, arising from its mixture with the sand forming the crust, and 
giving the supernatant liquor the colour of brandj\ 
