44 Prof. Bischof o?i the Natural Historij of 



fissures in the rocks,* by which the waters are admitted in 

 other places. But in doing so, it may frequently happen that 

 these fissures do jiot communicate with the channels by which 

 the water is admitted, and that the volcanic action is conse- 

 quently for a time suspended, but that on its revival the 

 slightest shock is sufficient to break through the walls, and 

 thus to reopen the communication.! This may even be caused 

 by the expansion of the cooled walls of the focus by the heat 

 communicated to them from all sides ; in the same manner as a 

 small crack in a crucible increases when exposed to a red heat. 

 The more the temperature of the lava is reduced by the 

 water and the generation of the steam, the longer will be the time 

 recpiired for the refusion of the solidified lava. In this manner 

 a long period may elapse, as the lava is so very bad a conduc- 

 tor of heat.:]: The repose and activity of a volcano are, therefore, 

 the alternate solidification and liquefaction of the lava, and the 

 interruption and renewal of the supply of water to the volcanic 

 focus. II If the store of lava in the volcanic focus should at 

 first become exhausted by repeated discharges, the volcano is 

 entirely reduced to a state of rest, or at least until it receives a 



portion as the elasticity of the vapours diminish, the substances will be 

 tlirown to a less distance, so that the black rapilli, which are the first eject- 

 ed after the lava has ceased to flow, will be cast farther than the white 

 ones. Von Humboldt's Keise, t. i. p. 245. 



* It is well known tliat considerable fissures are formed in lava during 

 its cooling, especially when it is on the surface of the earth. The streams 

 of lava in the coimtry surrounding the Laachcr-Sec, offer many instances of 

 this kind. Hamilton also mentions great fissures in the lava-streams of 

 VesiiTius. Gilbert's Annal. t. vi. p. 23. See also Necker, loco cit. 



t "We may here notice the well-known phenomenon, that among the 

 ejected masses from a volcano, pieces of rock occur, which neither belong to 

 the substances composing the edge of the volcanic cone, nor to those found in 

 the vicinity, and therefore must be derived from masses concealed very deep 

 under the volcano. Vesuvius particularly, furnishes remarkable instances 

 of this kind. Such ejected masses, however, are now found much more 

 rarely than formerly on this or other volcanos, from which it seems to 

 follow, that the channels of the ejections have been by degrees widened. 



Ij: Monticelli and Covelli, loco cit. p. 15 and 39. 



11 Experiments hitherto made shew, that long spaces of time are requi- 

 site to produce the strongest effects, viz. the elevation of lava to the great- 

 est height. Von Humboldt (Reise, &c. t. i. p. 261.) calls our attention to the 

 circumstance, that long intervals of quiescence seem to characterize the 



