4.M! F. A. Per ret — Lava Eruption of Stromboli. 



of vapor and returns in a wave, thus instituting a recurrent 

 action which continues for a certain time. 



The sudden contact of water and incandescent material, over 

 a considerable area, results in a violent evolution of steam with 

 a rushing, blowing roar, and the projection outward of tine 

 scoria? and dust which gives a dark color to the lower portion 

 of the cloud of steam. The vapor is often evolved in the 

 form of high pressure spiracles proceeding from the interstices 

 of the scoria*, of which the temperature is such that the vapor 

 is invisible for some centimeters above the source, thus render- 

 ing difficult the precise photographing of the jets. 



But all this commotion — spectacular and impressive though 

 it may be — is confined to the water surface at the point of 

 entrance of the lava, and there remains the outstanding, all- 

 important fact that the stream of compact liquid lava continues 

 on its course beneath the waters of the sea. There is no sur- 

 face indication of its existence as it disappears into the depths 

 with a smooth, unagitated surface, and we are confronted by 

 the fact of a mass of highly heated liquid moving through water 

 without conflict — a sub-aqueous lava comporting itself as 

 quietly as when sub-aerial. 



That this is' possible is due to the formation — by rapid cool- 

 ing through contact with a cold, conducting liquid of great 

 heat capacity — of a porous lava sheath of low heat conductivity 

 which intervenes to hinder the free passage of heat into the 

 water. This may be assumed to be quasi-flexible throughout 

 most of its thickness (that portion which includes the temperature 

 gradient between 500° and 900°), and is probably very tough. 

 The " icicles " which fringe the entrance to the spatter-grottoes 

 at the Kilauea lava lake are often several meters in length, but 

 so tough and flexible as to remain suspended though buffeted 

 in and out by belches of gas from the caves. 



A sheath of water vapor (spheroidal state) will also suggest 

 itself as a heat insulator interposed between the hot lava and 

 the water, but this well-known phenomenon, under the condi- 

 tions here prevailing, is of such short duration that its influence 

 to protect the flowing lava stream must be accounted secondary. 



The protective lava sheath will have a total temperature 

 gradient from the temperature of the outside water (say 30°) to a 

 thousand or more within. Rapid cooling is essential to the forma- 

 tion of such a skin, for only in this way is a surface of such 

 tough viscous quality and little crystallization obtainable in 

 most lavas. Through the interposition of such a non-conduct- 

 ing sheath an incandescent flowing lava will protect itself from 

 reaction with water, while a hot crystalline rock has no protec- 

 tion beyond the momentary check afforded by the spheroidal 

 state of the water. At Etna the writer has seen a lava which 



