232 Dr. Tempest Anderson [Jan. 23, 



tliese and other indications, it is most probable that radial cracks 

 entered deeply tbrougb tlie substance of the mountain, and penetrated 

 even the submarine portion of its cone. 



The local distribution of erupted material in Martinique is ac- 

 counted for by the great fissure at the top of the valley of the Kiviere 

 Blanche, which communicated with the main pipe of the volcano, 

 and out of which the eruptions took place. This fissure, which was 

 mentioned as existing in the eruption of 1851, pointed almost directly 

 towards St. Pierre, and as the erupted material flowed out almost like 

 a fluid, it was directed straight down on the doomed city. The lowest 

 portion of the lip of the crater of the Soufriere was much broader and 

 more even, so the incandescent avalanche which descended from it was 

 spread much more widely. 



The latest accounts from Professor Lacroix indicate that the recent 

 small eruption of Mont PeUe has filled up the highest parts of the 

 fissure and formed a cone, the foot of which covers up the former 

 crater ring. In any further eruption, therefore, the avalanche of in- 

 candescent sand will not be confined to the district of the Riviere 

 Blanche, but may descend on any side of the mountain. 



The accompanying photograph of Mont Pelee in eruption was 

 obtained from a ten-ton sloop in a sea way, and is therefore not quite 

 sharp. Attention was directed to the eruption by a peculiar black 

 cloud which appeared over the volcano and then rolled down the side 

 of the mountain to the sea. The cloud was formed of surging, roll- 

 ing, expanding masses, in shape much like those of the previous 

 cauliflowers, but quite black, and full of lightning-flashes and scintil- 

 lations, while small flashes constantly struck from its lower surface 

 on to the sea. The upper slopes of the mountain cleared somewhat, 

 and some big red-hot stones were thrown out ; then the triangular 

 crack became red, and out of it poured a surging mass of incandes- 

 cent material, reminding us of nothing so much as a big snow- 

 avalanche in the Alps, but at a vastly different temperature. It was 

 perfectly well defined, did not at all tend to rise like the previous 

 cauliflowers, but flowed rapidly down the valley in the side of the 

 mountain which had clearly been the track of previous eruptions, 

 until in certainly less than two minutes it reached the sea, and was 

 there lost to view behind the remains of the first black cloud, with 

 which it appeared to coalesce. There and on the slopes of the moun- 

 tain were doubtless deposited the greater part of the incandescent 

 ash, while the steam and gases, with a certain portion of still entan- 

 gled stones and ash, came forward in our direction as a black cloud, 

 but with much greater rapidity than before. The cloud got nearer 

 and nearer ; it was well defined, black and opaque, formed of surging 

 masses of the cauliflower type, each lobe rolling forward, but not all 

 with one uniform rotation ; bright scintillations appeared, some in 

 the cloud itself and some like little flashes of light vertically between 

 the cloud and the sea on which it rested. These were clearly the 

 phenomena described by the survivors of the St. Vincent eruption as 



