FKOM THE CRATER OF VULCANO. 3S3 
water did not develope the octohedral form, but evolved 
air-bells. Nitro-muriate of platinum did not alFect a satu- 
rated solution of this alum. Its composition, therefore, is 
an ammoniaco-sulphate of alumina. 
The Baryta was not at any rate terra ponderosa^ for it 
was exceedingly lig-ht and tender ; the minute folife which 
composed the mass were of a pearly semblance, and felt 
somewhat unctuous to touch. The scales had every appear- 
ance of boracic acid; dissolved in distilled water, it j-ed- 
dened litmus paper ; and, with alcohol, exhibited, when in- 
flamed, the green colour which characterizes boracic acid. 
That the waters of the ocean have some subterranean 
communication with the source of volcanic fires, has ever 
appeared to me a conclusion perfectly warrantable; but 
the spring whence the muriate of ammonia has flowed, is 
a problem of more difficult complexion., I greatly deceive 
myself, however, if an ammoniacal combination does not 
obtain in marine salt. If sea-salt be finely powdered and 
triturated with a solution of caustic potassa, or even with dry 
quicklime, muriatic acid will announce the escape of am- 
monia. 
This circumstance, combined with the researches of 
KouELLE and Proust, respecting the existence of a mer- 
curial salt in the oceanic waters, would lead us to conclude, 
when conjoined with the discovery of muriate of potassa 
in them by Dr Wollaston, that their chemical constitu- 
ents are more complex than had been hitherto supposed. 
TOL IV. 
z 
