104 SCIENCE PROGRESS 



top of the plateau, large mandarin-orange trees were uprooted 

 and carried westwards up a slope. The blasts were, however, 

 directed principally against the north-east corner of the plateau 

 and the neighbouring village of Koiki. The destruction here 

 was general, the tree-trunks being overthrown or broken in 

 directions varying between lines which, when produced back- 

 wards, pass through the highest and lowest craterlets of the 

 western series. 



On the eastern side, no distinct trace of the blast could be 

 detected. Here, however, the craterlets are formed on ridges. 

 On the opposite side, they lie for the most part along the trough 

 of a V-shaped radial valley, which probably facilitated the 

 partial westward concentration of the explosive effects. 



Precipitation of Ashes. — Over the Japanese empire, westerly 

 winds usually prevail in the upper region of the atmosphere. 

 They are no doubt accountable for the abnormal precipitation 

 of ashes in an easterly direction from the Japanese volcanoes. 

 Thus, in the recent eruption, ashes were carried to a distance 

 of 765 miles towards the north-east and only nine miles in the 

 opposite direction. Similarly, the layer of ashes reached a 

 depth of 3 ft. to a distance of fourteen miles on the east 

 side of the volcano and to somewhat less than two miles on 

 the west. At Kagoshima, the depth was less than one-eighth 

 of an inch. On the low ground on the eastern side of Sakura- 

 jima, ashes and pumice were piled to a depth of 7 ft., and in 

 one valley to a depth of more than 13 ft. The total volume 

 of ashes is estimated by Prof. Omori to amount to "15 of a 

 cubic mile. Adding this to the amount of lava ejected, namely 

 •37 of a cubic mile, the total volume of material ejected is thus 

 slightly more than half a cubic mile, or about one-twelfth of the 

 volume of the whole mountain. 



Earthquakes of January 12, etc. — At 6.30 p.m. on January 12, 

 an earthquake occurred differing widely from the local shocks 

 which so frequently precede and accompany volanic eruptions. 

 It coincided, as regards time, with a marked increase in the 

 eruptive activity of Sakura-jima, and its epicentre, though not 

 exactly known, cannot have been far distant from the island. 

 On these accounts, it may be regarded as a volcanic earthquake. 

 But, while the typical volcanic earthquake is of brief duration 

 and is strong with a limited area, and rarely recorded by 

 distant seismographs, the Kagoshima earthquake was of 



