so Prof. Bischo^ oji the Xatural History of 



phenomena must necessarily be closely connected with volca- 

 nic action, and cannot pass unnoticed. 



But these disengagements of carbonic acid gas could not 

 take place in the presence of atmospheric air in those vast sub- 

 terranean cavities without their mixing together. Yet, accord- 

 ing to Monticelli and Covelli,* the Mofettes of Vesuvitcs con- 

 tain but little atmospheric air, which seems not to intermix with 

 the carbonic acid gas until it reaches the surface. I have exa- 

 mined many such exhalations of carbonic acid gas, in the vi- 

 cinity of extinct volcanos (in the neighbourhood of the Laa^ 

 cher See and in the Erfel) as well as in places where there are 

 no immediate volcanic traces {Himdsruck, the eastern declivity 

 of the Teutoburger Wald), and, in general, have found a scarcely 

 measurable quantity of atmospheric air. According to Bous- 

 singault,-|- the elastic fluids, which are evolved from the vol- 

 canos at the equator in the New World, consist of a great 

 quantity of aqueous vapour, carbonic acid gas, sulphureted 

 hydrogen gas, and sometimes fumes of sulphur ; he considers 

 sulphurous acid gas and nitrogen, on the other hand, as acci- 

 dental. This philosopher! also found tlie same gases,, viz., caa*- 

 bonic acid and sulphureted hydrogen gas, in the springs which 

 rise in the vicinity of these volcanos. All this is by no means 

 favourable to the supposition of the existence of vast subter- 

 ranean cavities filled with air under the craters, and an equally 

 unfavourable circumstance is, that, according to Boussingault, 

 no nitrogen is evolved from the volcanos under the equator, 

 which must necessarily be the consequence of oxidation at the 

 expense of atmospheric air. 



Independently of all this, the metals of the earths have been 

 found by more recent experiments to be by no means so easy 

 of oxidation as Davy''s hypothesis assumed. Besides, this prone- 

 ness to oxidation must be supposed to be a property more es- 

 pecially belonging to the metals of silica and alumina, as these 

 earths together with oxide of iron, are the principal components 

 of volcanic products, — lavas, basalts, &c., generally amounting 

 to about O.S, whilst lime and alkalies, although never entirely 

 wanting, form but an inconsiderable proportion. But Berze- 



* Loco cit. p. 194. t Loco cit. v. lii. p. 5. X Ibid. p. 181. 



