340 PROCEEDINGS OF THE AMERICAN ACADEMY 



was carried upward, falling into the sea, — nearer or farther off 

 in proportion to the size of its fragments. The finer parts rose 

 above the current of the trade winds, and, taking the upper 

 and opposite flow, spread over the sea and the island of Bar- 

 badoes, obscuring the light of a tropical sun, and causing the 

 greatest consternation on land and sea. This ash I have ex- 

 amined from several parts of its course, and it differed in no 

 respect from the fine parts of the trachyte, undergoing decom- 

 position by atmospheric agents on the spot. There followed 

 after this explosion no flow of lava, but a shower of rude 

 balls of half-fused, tumefied trachyte, succeeded by fragments 

 of rocks, earths, and finally mud and water. The final action 

 took place obviously within the crater, formed more than 

 eight hundred feet below the surface of the top of the 

 mountain, and resulted in the production of a regular cone 

 of sand and gravel, which remained. Twenty years after 

 (1812) a similar explosion took place, and the point of great- 

 est interest is, that a new centre of action appeared. A 

 smaller crater was formed, so near the older one that the rim 

 of the later one breaks its continuity. The action which fol- 

 lowed the dispersion of the disintegrated covering in this 

 case was of a kind among the most remarkable on record. 

 A large part of the force was expended in discharging from 

 the crater rocks broken into fragments, from the size of a 

 cubic inch to that of grains of sand ; nearly every fragment 

 and grain being bounded by straight lines, square or rectangu- 

 lar, with sharp angles and edges. As the volcanic vents of 

 the West Indies, and indeed whole islands, have been elevated 

 from below a deep ocean which surrounds them, they offer 

 the best examples of that secondary effect, resulting from 

 chemical action taking place within the aggregates formed, 

 which I could adduce. 



Professor Horsford suggested that, as the volcanic ashes are 

 silicates of alumina, it might be possible for the mixed chlo- 

 rides of aluminium and silicium to be shot as a bolt from 

 a crater, and at a distance from that point find moisture and 



