AND THE WEST INDIAN EUUI'TIONS OF 1902. 



225 



lake water, that will then cause the greater explosive effects of 

 eruptions, and produce the vast volumes of steam that ascend 

 above the eruptive craters. 



Should the lava not find a conduit extending to the surface, 

 Plutonic dykes may be formed far below; and should lava 

 reacli the surface without meeting with water, a purely 

 emissive eruption will be the result. 



The great paroxysmal explosive eruptions, such as the 

 recent West Indian outbursts, may, therefore, be regarded as 

 due to the formation and rise of an unusually large body of 

 lava, togetlier with the supply to the volcanic conduit of 

 sufficiently large bodies of water, to transform the whole into 

 solid fragmentary ejectamenta. As the fusion temperature of 

 rocks is above the critical point of water, some of the water 

 may be decomposed by the disassociation of its elements, and 

 free hydrogen being thus evolved some of the effects stated in 

 the reports of the recent eruptions might be produced by the 

 inflammability of that gas. 



The time of an eruption may, I consider, be determined by 

 one or more of several factors, that will be sufficient to give 

 the requisite relief of vertical pressure. Amongst these factors 

 will be lateral pressure, secular elevation, planetary or lunar 

 attraction, and hygrometric atmospheric conditions. In illus- 

 tration of the relief of pressure consequent upon the last-named 

 factor, it may be mentioned that a fall of the barometer of two 

 inches will remove pressure from the area of the base of Mount 

 Etna alone, to the extent of two thousand millions (2,000,000,000) 

 of tons. The islands of Martinique and St. Vincent, with the 

 other islands of the Lesser Antilles, are on the crest of a long 

 ridge that has been elevated in comparatively recent geological 

 times, and the elevatory movement has apparently not yet 

 altogether ceased. It is this elevation with its consequent 

 relief of pressure that has most probably been the cause of 

 the renewed volcanic activity in the Windward Islands of the 

 West Indies. 



The explanation of volcanic action wliich I have here ventured 

 to give is in accordance with an hypothesis I brought before the 

 British Association so long ago as the I]ath Meeting of 1888. 

 As I have not since become acquainted with anything to shake 

 that hypothesis, perhaps I may be pardoned for having some 

 confidence in its soundness, which appears to me to be sup- 

 ported and illustrated by the West Indian eruptions of 1902. 



