238 
VOLCANIC LIGHT. 
Oan these flames he attributed to the decomposition of 
water, entering into contact with the pyrites dispersed 
through the schistose marl ? or is it inflamed hydrogen that 
issues from the cavern of Cuchivano ? The marls, as the 
smell indicates, are pyritous and bituminous at the same 
time ; and the petroleum spi'ings at the Bueu Pastor, and in 
the island of Trinidad, proceed probably from these same 
beds of alpine limestone. It would be easy to suppose some 
connexion between the waters filtering through this calca- 
reous stone, and decomposed by pyrites and the earthquakes 
of Cumana, the springs of sulphuretted hydrogen in New 
Barcelona, the beds of native sulphur at Carupano, and the 
emanations of sulphurous acid which are perceived at times 
in the savannahs. It cannot be doubted also, that the 
decomposition of water by the pyrites at an elevated tempe- 
rature, favoured by the affinity 'of oxidated iron for earthy 
substances, may have caused that disengagement of hydrogen 
gas, to the action of which several modern geologists have 
attributed so much importance. But in general, sulphurous 
acid is perceived more commonly than hydrogen in the 
eruption of volcanoes, and the odour of that acid principally 
prevails while the earth is agitated by violent shocks. When 
we take a general view of the phenomena of volcanoes and 
earthquakes, when we recollect the enormous distance at 
which the commotion is propagated below the basin of 
the sea, we readily discard explanations founded on small 
strata of pyrites and bituminous marls. I am of opinion 
that the shocks so frequently felt in the province of Cu- 
mana are as little to he attributed to the rocks above the 
surface of the earth, as those which agitate the Apennines 
are assignable to asphaltic veins or springs of burning petro- 
leum. The whole of these phenomena depend on more 
general, I would almost say on deeper, causes ; and it is not 
in the secondary strata which form the exterior crust of our 
globe, but in the primitive rocks, at an enormous distance 
from the soil, that we should seek the focus of volcanic 
action. The greater progress we make in geology, the more 
we feel the insufficiency of theories founded on observations 
merely local. 
On the 12th of September we continued our journey to 
the convent of Caripe, the principal settlement of 'the 
