114 BEPORT— 1892. 



recorded as having caused them to overturn is practically identical with 

 that which may be calculated from their dimensions. The ratios of 

 breadth to height varied from 1 : 2^ up to 1 : 9, and in each group there 

 were at least six columns. These ratios are identical with the dimensional 

 ratios of gravestones and other bodies overturned by the last earthquake. 

 The actual sizes of the columns experimented on were not small ; for 

 example, one column, where the ratio of breadth to height is 1 : 2^, was 

 9^ in. square and 25| in. high. One result of this work is that we can 

 now go from place to place through the shaken district, and state with 

 considerable accuracy the accelerations that were experienced. As we 

 have one or two diagrams from the shaken area, we know, at least 

 approximately, the period of the motion, and we can therefore calculate 

 the amplitude and maximum velocity of motion which was experienced. 

 Such calculations are now being made by Mr. F. Omori. 



From the formula given in the last Report, which showed a relation- 

 ship between the dimensions, weight, and strength of a structure, and the 

 acceleration which would shatter the same, we are now able to calculate 

 accelerations from shattering phenomena. The greatest difficulty which 

 had to be overcome in making these investigations was the collecting 

 and subsequently determining the strength of portions of masonry 

 structures which had been ruined. Altogether, 26 cases of brickwork 

 have been obtained from the earthquake district, and the testing of their 

 contents has been completed. Reference is made to the results of the 

 calculations based on these experiments under the section relating to the 

 Great Earthquake. 



The only new development of the fracturing formula has been to 

 determine the form of a wall or pier which, rather than snapping at its 

 base — as was the case with many structures at the time of the last earth- 

 quake, and also with all the columns subjected to back and forth motion 

 on the earthquake track — would be as likely to break at any one 

 horizontal section as at any other. The effect of the last earthquake was 

 to cause brick piers of railway bridges to snap at their bases, and it 

 would therefore seem that the basal section required greater strength. 

 This might possibly be obtained by increasing the basal dimensions at 

 the expense of a portion of the brickwork in their upper parts, which, so 

 far as offering resistance to effects resulting from their own inertia, were 

 apparently more than sufficiently strong. As a result both of reasoning 

 and observation, it is evident that in earthquake countries, piers, walls, 

 chimneys, and the like should have vertical sections different from 

 those which they receive in ordinary practice. A full account of these 

 experiments will be published in a ' Journal of Seismology ' which is 

 shortly to be issued. 



The Gkeat Eakthquake op October 28, 1891. 



If we may judge from the contortions produced along lines of railway, 

 the Assuring of the ground, the destruction of hundreds of miles of huge 

 embankments which guard the plains from river-floods, the utter ruin 

 of structures of all descriptions, the sliding down of mountain sides and 

 the toppling over of their peaks, the compression of valleys, and other 

 bewildering phenomena, we may confidently say that last year, on the 

 morning of October 28, Central Japan received as terrible a shaking aa 

 has ever been recorded in the history of seismology. It is a subject 



