520 Reviews — The Eruption of Sakura-jima. 



and probably more, of these vents have not altered their positions 

 to any appreciable extent since that date. He compares with this 

 the constant position of the lava lake, Halemaumau, in the crater of 

 Kilauea, and suggests that there is some evidence for a similar 

 constancy at Vesuvius and Etna, though here the information is 

 much less definite. 



From these facts the author draws the following conclusions : 

 (1) The vents are the openings of conduits which have pierced 

 through the solid rock for the greater part of their courses. (2) The 

 lengths of these conduits must be considerable, for, if this were not 

 so, we should expect vents to break through the thin crust at various 

 places at different times. (3) The persistence and relatively small 

 size of the vents, which must be wider than the conduit, indicate 

 that there has been very little corrosion of the vent by the uprising 

 magma, which is presumably kept fluid at a lower temperature than 

 the melting-point of the surrounding walls, owing to its dissolved 

 gases. (4) The complete absence of synchronism in the activity of 

 the different vents shows that they communicate individually with 

 the magma reservoir. (5) Judging by the size of the vents the 

 conduits cannot be more than " a few tens of metres " wide. 



These conclusions, together with the fact that the vents occur at 

 the top of a scarp more than 2,000 feet high (probably caused by the 

 subsidence of the north-west portion of the original crater), lead the 

 author to adopt Daly's "gas-fluxing" hypothesis to explain their 

 formation. This hypothesis is briefly described as follows : Gas 

 bubbles rise to the top of the magma in the reservoir and accumulate 

 in irregularities (" cupolas ") in the roof. These gases are supposed 

 to be at a higher temperature than the magma owing to their exothemic 

 interactions and so " blowpipe " their way up through the overlying 

 rock. The action of these gases will only extend over a small area, 

 so that many narrow independent pipes will be formed. This 

 hypothesis is especially applicable to the formation of the vents on 

 the Stromboli crater terrace, if this has been formed by the sinking 

 of a large mass to the north-west, since the magma reservoir may 

 be assumed to have maintained its original level under the remaining 

 centre part of the original mountain, which would then be the most 

 favourable place for the " gas-fluxing " action to take place. 



The paper is illustrated by some very interesting photographs, 

 sketches, and reproductions of illustrations from the old books dealing 

 with the volcano. 



W. H. W. 



IV. — The Eruption of Sakura-jima on January 12, 1914. By 

 Charles Davison, Sc.D., E.Gt.S. Science Progress, No. 45, 

 July, 1917, pp. 97-110. 



IN Science Progress for July, 1917, Dr. Davison gives a very 

 interesting summary of Professor Omori's investigations 1 into 

 the eruption of Sakura-jima which took place in January and 



1 "The Sakura-jima Eruptions and Earthquakes": Bulletin of the 

 Imperial Earthquake Investigation Committee (Tokyo), vol. viii, pp. 1-34 

 (1914), 36-179, 181-321 (1916). 



