339 



in the lava when it was raised and poured out by an iron ladle. A 

 portion was thrown into a glass bottle, which was then closed with a 

 ground stopper ; and on examining the air in the bottle some time 

 afterwards, it was found not to have lost any of its oxygen. Nitre 

 thrown upon the surface of the lava did not produce such an increase 

 of ignition as would have attended the presence of combustible mat- 

 ter. The gas disengaged from the lava, proved on examination to 

 be common air. When the white vapours were condensed on a cold 

 tin plate, the deposit was found to consist of very pure common salt ; 

 and the vapours themselves contained 9 per cent, of oxygen, the rest 

 being azote, without any notable proportion of carbonic acid or sul- 

 phurous acid gases ; although the fumes of this latter gas were ex- 

 ceedingly pungent in the smoke from the crater of the volcano. On 

 another occasion the author examined the saline incrustations on the 

 rocks near the ancient bocca of Vesuvius, and found them to consist 

 principally of common salt, with some chloride of iron, a little sul- 

 phate of soda, a still smaller quantity of sulphate or muriate of potassa, 

 and a minute portion of oxide of copper. In one instance in which 

 the crystals had a purplish tint, a trace of muriate of cobalt was de- 

 tected. From the observations made by the author at different pe- 

 riods, he concludes that the dense white smoke which rose in immense 

 columns from the stream of lava, and which reflected the morning and 

 evening light of the purest tints of red and orange, was produced by 

 the salts which were sublimed with the steam. It presented a stri- 

 king contrast to the black smoke, arising from the crater, which was 

 loaded with earthy particles, and which in the night were highly lu- 

 minous at the moment of the explosion. The phenomena observed 

 by the author afford a sufficient refutation of all the ancient hypo- 

 theses, in which volcanic fires were ascribed to such chemical causes 

 as the combustion of mineral coal or the action of sulphur upon iron, 

 and are perfectly consistent with the supposition of their depending 

 upon the oxidation of the metals of the earths upon an extensive 

 scale in immense subterranean cavities, to which water, or atmosphe- 

 ric air, may occasionally have access. The subterranean thunder 

 heard at great distances under Vesuvius, prior to an eruption, indi- 

 cates the vast extent of these cavities ; and the existence of a subter- 

 ranean communication between the Solfaterra and Vesuvius is esta- 

 blished by the fact, that whenever the latter is in an active state, the 

 former is comparatively tranquil. In confirmation of these views, the 

 author remarks that almost all the volcanoes of considerable magni- 

 tude in the old world are in the vicinity of the sea : and in those 

 where the sea is more distant, as in the volcanoes of South America, 

 the water may be supplied from great subterranean lakes ; for Hum- 

 boldt states that some of these throw up quantities of fish. The au- 

 thor acknowledges, however, that the hypothesis of the nucleus of the 

 globe being composed of matter liquefied by heat, offers a still more 

 simple solution of the phenomena of volcanic fires. 



