410 E. O. HOVEY STRIATIONS AND U-SHAPED VALLEYS 



liquid rock. As the lava reached the orifice of the conduit, pressure was 

 relieved suddenly, and the steam expanded and continued to expand with 

 explosive violence, breaking up the lava to all degrees of grain, from 

 great blocks 10 meters or more across down to impalpable dust. The 

 fragments when formed were angular, but attrition in the eruption cloud 

 and against the surface of the ground rounded the contours of the grains, 

 particularly the larger ones and the great masses. Under the microscope 

 the fine dust is seen to be highly and sharply angular. In the field the 

 ejected blocks usually show bruised and rounded angles and corners. 

 The first eruption clouds, furthermore, reached the surface in the bottom 

 of a vast pit-like crater about 1 kilometer across. Their force of explo- 

 sive expansion was confined on three sides by practically vertical walls 

 from 300 to 650 meters high, but the southwest side of the rim was 

 breached to the bottom of the crater by a great Y-shaped cleft. These 

 two factors 2 prevented the free expansion of the exploding cloud and 

 gave direction to it, confining it to a narrow sector of a circle, bounded 

 on the northwest by the high bluffs along the right bank of the Riviere 

 Claire, but without any sharp surface delimitation on the other side. 

 This was the "zone of annihilation" often referred to in my own and 

 other early reports upon the eruptions. The content of dust in the cloud 

 was such as to make part of it heavier than the atmosphere, so that it 

 clung to the ground, rolling and sliding over its surface. 



Striations 

 sand-blast action on mount pel£ 



Within the sector thus described all the bluffs facing the crater were 

 smoothed, scored, and grooved as if by the action of a stupendous sand- 

 blast or the passage of a well sanded glacier. This sand-blast action at 

 Pele has been noted by Professor Lacroix, 3 me, 4 and Doctor Anderson, 5 

 but only in a preliminary or very cursory manner, whereas the occurrence 

 is worthy of more extended notice. 



2 Some authors have argued for the existence of an inclined opening to the volcano's 

 conduit to explain the marked orientation of the eruption cloud. The assumption of 

 such specific inclination of the vent seems unnecessary, since the two factors mentioned 

 would be sufficient to produce the effect, hut a discussion of the cause of the confinement 

 of the destructive blast to a narrow zone is beside the purpose of the present article. 



3 A. Lacroix : Rollet de l'lsle and Giraud. Comptes Rendus de l'Acad., cxxxv, p. 387. 

 Meeting of September 1, 1902. 



A. Lacroix : La Montagne Pelee et ses Eruptions, 1904, p. 217. 



* Bull. American Museum of Natural History, vol. xvi, October, 1902, p. 363, pi. xlix, 

 fig. 2. American Journal of Science, IV, vol. xiv, November, 1902, p. 345, fig. 15. 

 Comptes Rendus, ix, Congr. Geol. Internat, 1904, p. 716. 



6 Philosophical Transactions of the Royal Society of London, series A, vol. 208, 1908, 

 p. 300, pi. 25, figs. 1, 2. 



