POPULAR SCIENCE 109 



movement is invariably one of depression by amounts ranging 

 from about 1 ft. to 5$ ft. At three places on higher ground, 

 the movement, however, is in the opposite direction. At 

 Kamano (147 ft. in height) the measured rise is 5^ in. ; at 

 the top of Hikinohira (2,178 ft.), it amounts to 30 ft. ; and 

 at the top of Gongen-yama (1,339 ft.) near Nabe-yama, to 

 41 ft. Even making every allowance for the depth of ashes 

 at these places, it is evident that the rise near the centre of the 

 island must have been considerable. 



The horizontal movements, if somewhat less in amount, are 

 still of great importance. For instance, near the south-west 

 coast, the displacements of three points amount to 6 ft. 8 in., 

 10 ft. 1 in., and 11 ft. 11 in., all to the south or a few degrees 

 west of south. Two points close to the north-west coast were 

 shifted 3 ft. 6 in. and 14 ft. 10 in. (the latter being the greatest 

 horizontal displacement observed) in a nearly north direction, 

 while a point in the extreme north of the island was moved 

 8 ft. 7 in. towards the north-east. The north and south portions 

 of Sakura-jima thus appear to have been displaced outwards in 

 opposite directions. Again, along the west coast of Kagoshima 

 Bay, in the immediate neighbourhood of Sakura-jima, displace- 

 ments of 1 ft. 10 in., 2 ft. 2 in., 3 ft. 1 in., and 2 ft. 8 in. were 

 observed. These vary slightly in direction, but the important 

 point about them is that all four directions, together with those 

 in the north of Sakura-jima, converge towards an elliptical area, 

 the position of which agrees roughly with the area of greatest 

 depression. 



Sound-Phenomena. — The sounds which accompanied the 

 eruption may be divided into three groups : (i) the early sounds 

 heard during the morning and afternoon of January 12 ; (ii) 

 the strong detonative sounds from 6.30 p.m. on the 12th to 

 6 a.m. on the 13th ; and (hi) the weaker sounds of the after- 

 explosions which occurred during the ten days following the 

 great eruption. 



Observers of the gun-firing in France and Flanders have 

 shown on numerous occasions how strangely the sound-waves 

 are propagated by the atmosphere. The sounds of very 

 distant guns may be heard with clearness, while those of equally 

 large guns close at hand are quite inaudible. When the places 

 of observation are mapped they are found to lie within two 

 detached areas, one surrounding the source of sound, the other 



