438 Dr. Johnston- Lavis — Mechanism of Volcanic Action. 



liquid state, and probably solid, it is evident that those most volatile 

 would be the last to change their physical state. That some of the 

 more volatile ones were entangled or held in solution by the less 

 volatile is quite likely, but the amount must have been small. 



Another possible source of volatile matter in the deep-seated 

 igneous matter may well be due to a slow osmosis or diffusion 

 extending over vast periods of time and directed by the varying 

 affinities of one class of matter for the other. 



A quarter of a century ago, as a result of a careful and detailed 

 study of Yesuvius ^ and other volcanoes, I was able to show that 

 a volcano the more continuously active it was in the emission of 

 igneous material the more tranquil was the character of emission, and 

 that practically under such conditions lava was the only product. 

 I showed also that the longer were the intermissions in the extrusive 

 efforts of a volcano the more the ejecta tended to issue in a broken 

 up and fragmentary condition, from the larger and more violent 

 evolution of volatile or gaseous materials. "We thus had the whole 

 gamut of products — scoria, pumiceous scoria, scoriaceous pumice, 

 pumice, and pumice dust — bearing a distinct ratio to the time that 

 any volcano had been in a condition of ' repose '. 



Two explanations offered themselves to my mind for this state of 

 things. One was that the persistent evolution of volatile materials 

 primordially stored up in the original volcanic paste escaping 

 continually therefrom had collected in the volcanic chimney and 

 blew out the magma ; the other that the volatile materials were 

 acquired by the igneous magma where in contact with water-bearing 

 rocks in the upper strata of the earth's crust. 



Were the former the case one would expect that the first products 

 of an eruption should be less gas-filled or gas-bearing than the latter, 

 but this is not the case. My observations, which demonstrate the 

 fundamental facts upon which eruptive action of a volcano depends, 

 show unmistakably that the first materials yielded in a normal 

 eruption after a long period of 'repose' of a volcano are the richest in 

 volatile elements, and that as the eruption proceeds the amount of 

 gases in the issuing magma steadily diminishes, as shown by the 

 diminished vesicularity and increased crystalline individualization of 

 the essential ejecta.^ 



The class which interests us in the present question is the group of 

 the essential ejecta. I found that when one examines the stratified 

 deposits of the ejecta thrown out during an explosive eruption, that is, 

 an eruption of great violence taking place after a long period of repose 



1 "The Geology of Monte Sonoma and Vesuvius": Q.J.G.S., 1884, vol. xl, 

 pp. 35-119. 



2 In my paper " On the Fragmentary Ejecta of Volcanos" (Proc. Geol. Assoc, 

 vol. ix, pp. 421-32 and 3 figs.) I divided such ejecta into three classes. Essential 

 ejecta are those materials that issue in a fluid state, and consist either of the volatile 

 constituents or the magma in which these vrere contained, that produced the 

 particular emission in question. Accessory ejecta consist of the older volcanic 

 materials of the same vent torn away, expelled, and mixed with the essential ejecta 

 of an eruption. Accidetital ejecta consist of either volcanic materials from other 

 centres, or sedimentary or other rocks of the sub-volcanic platform, also torn out, 

 expelled, and mixed with the two before-mentioned ejecta. 



