THE ERUPTIOX OF \-E5rVTUS IX .VPBIL. 1906. 233 



material.s ejected. The material once started advances with accel- 

 erated velocity. 



The slopes have been deeply furrowed by them, and in places the 

 furrows or " barrancos " possess a great regularity of form, bemg 

 equally spaced and separated by sharp-edged talus ridges. They 

 serve to explain similar forms of other volcanoes. 



The avalanches have built up breccias with a most chaotic struc- 

 ture, identical not only with those of the Antilles, but with trachitic 

 and andesitic breccias of the central massif of France. The fur- 

 rows have outlined the drainage, and the mud flows have accented 

 and modified their structure. The dry avalanches were not confined 

 to this eruption, but have been noted by observers at many previous 

 eruptions. 



Partial destruction of Otta'jano and San Giuseppe. — On the night 

 of A]jril 7. at about 12.30 a. m.. a shower of lapilli, accompanied by 

 intense electrical phenomena, commenced in the area northeast of 

 the mountain. The fall increased and lasted until 4 a. m. Windows 

 were Vjroken. and the lapilli accumulated on the roofs until these 

 were crushed, killing many who had been unable to flee or those who. 

 as at San Giuseppe, had taken refage in a church. All the victims. 

 about 200, lost their lives in this way. The lapilli at the time of 

 falling were cold. The quantity of material at Ottajano averaged 

 7 meters deep: in places it was thicker. The thickness diminished 

 toward the edges of the area and increased toward the mountain. 

 The fine dust which followed the great paroxysm covered the lapilli 

 with a uniform Ijed several centimeters thick. 



The average dimensions of the projectiles ranged from those of a 

 hazelnut to those of a walnut, and some pieces were 15 to 20 centi- 

 meters in diameter. They consisted of black or reddish scoria, rather 

 light, but accompanied by angular fragments of denser rocks of 

 various nature, including ancient lavas and metamorphic rocks. The 

 mineralogical and chemical study of this material shows that, apart 

 from a small amount of lapilli resting on the surface and due to the 

 Strombolian explosions, the greatest part of the material has a differ- 

 ent composition from that of the recent lava and is the product of 

 the Yulcanian explosions, which have caught up the ancient debris of 

 the volcano. This conclusion conforms with the facts that they 

 arrived cold and that the ejections changed their nature at midnight 

 Vjy becoming completely dark. 



The fall at Ottajano, 5 kilometers from the mountain, was greater 

 than at the observatory, only about half that distance. The wind 

 would be unable to transport coarse material to such a distance, and 

 the direction of the ejection instead of being vertical must have been 

 more or less horizontal. The history of Vesu^'ius shows that erup- 

 tions have taken place in greatly inclined or horizontal directions, 



