6 SHALER ON THE CLASSIFICATION OF LAVAS 



movement, else the eruptions of one volcano could not cause the others to become 

 dormant. It does not seem to me necessary to assume that this horizontal movement 

 of the gases takes place through rocks in a state of complete fusion. It is a well 

 recognized fact that certain substances, iron for instance, are more permeable to gases 

 when hi ghly heated than when cold, nor need we assume that the movement of the 

 gases is very rapid ; it may be, and probably is, rather slow, otherwise the repetition of 

 periods of repose and explosion could not well be explained. 



3. The lavas of ordinary volcanos do not represent the melting of the whole 

 section, where they originate, but only that part of the section which it is pos- 

 sible to melt at low temperatures. 



This proposition is deduced from the fact that all our lavas are characterized by the 

 large proportion of silex which they contain ; none of them save the basalts fall below fifty 

 per cent, of silex, and most of them exceed this amount. Now at the low temperature at 

 which all of our lavas were formed, it is pretty certain that the limestones and clay slates, 

 or, indeed any other elements of our ordinary geological sections, except the sandstones 

 and other very siliceous rocks, would not become melted, though they might slowly part 

 with their vaporizable substances. The result would be that the production of lavas would 

 depend in a large degree upon the existence, in the section through which the gases made 

 their way to the vent, of siliceous deposits, which could become fused at temperatures 

 as low as probably prevailed in the region whence they came. This will enable us to 

 account for the invariably siliceous character of our lavas. At the same time, it suggests 

 that in part, at least, the peculiarities of character of volcanos may be explained by the 

 irregular distribution of fusible deposits in the earth's crust. It has long been observed 

 that the quantity of lavas voided by volcanos varied greatly ; those in the Eifel district 

 being singularly wanting in lavas, while those in other regions, as for instance, in the 

 Sandwich Islands, throw out great quantities of lava. The ashes, which are in good part, 

 at least, the product of the friction of fragments of melted rock upon each other, the 

 rapilli, volcanic bombs, etc., which vary in quantity with the amount of lava, Avill 

 not be thrown out in large quantities when the molten rock does not abound. In this 

 way we see that the height of a volcano or the mass of the cone is in no way a 

 fair measure of the energy of the escape of gases ; the Eifel volcanos, such as Laacher 

 See, with their low walls of debris reminding us in their form of the lunar volcanos, 

 may have discharged gases in exceeding abundance, yet have produced little lava, 

 possibly on account of the absence of siliceous materials in the section Avhence they are 

 derived. 



It is easily seen that this view aids us in understanding the absence of subsidence in the 

 foundations of the great volcanos, notwithstanding the vast ejection of fluid and solid mat- 

 ter from their cones. Taking only the very siliceous deposits which do not usually form a 

 large part of the section, the lava-making forces would not be likely to remove a consid- 

 erable vertical thickness from the region just beneath the cone. 



4. The conversion of solid rocks into lavas is in part at least due to the energy of the 

 movement of the imprisoned gases. 



This seems to me to be fairly well shown by the phenomena of volcanic 

 eruptions. Whenever a new volcano breaks out, or whenever a cone long in 



