ON THE EARTHQUAKE AND TOLCANIC PHENOMENA OF JAPAN. 113 



connection between earth shakes, tremors, and the issue of firedamp, and 

 refers to the work attempted at Takashima. 



Some years later the British Association appointed a committee to in- 

 vestigate their possible re]ationships, and I believe that seismographs and 

 delicate seismoscopes were set up in the Newcastle district. At the time I 

 pointed out that such instruments could only record earthquakes, and it 

 was unreasonable to suppose that there could be any connection between 

 earthquakes and the ordinary outpourings of firedamp. The instru- 

 ments required were tromometers, rather than seismometers and seismo- 

 scopes, and in this Report I have defined the nature of the phenomena 

 to be looked for, which are more like wave-pulsations than tremors. 

 The instrument by which these may be investigated is the light form of 

 conical pendulum already described. The records are not the same as 

 those obtained from the swinging of a pendulum tromometer. We know 

 that a curve showing the microseismic activity in Italy closely follows a 

 curve showing the colliery explosions between 1860 and 1881 which have 

 occurred in Germany. Unfortunately, we do not know anything about 

 microseismic activity in Germany. From tromometric observations 

 made at the ^Ecole de Douai, by M. Chesneau, during February and ]\Iarch 

 1886, and from observations on the escape of gas estimated by a Pieler 

 lamp at the Herin Mine (Compagnie d'Anzin), we learn that certain 

 relationships between these two phenomena were established (.see ' De 

 I'etude des mouvements de I'ecorce terrestre poursuivie particuliere- 

 ment au point de vue de leurs rapports avec les degagements de produita 

 gazeux,' par M. Raynal, Ministre des Travaux publics ; M. B. de 

 Chaucourtois, inspecteur general, assiste de MM. Ch. Lallemand et G. 

 Chesneau, ingenieurs au Corps des Mines, ' Annales des Mines,' Mars- 

 Avril, 1886). 



Earth pulsations, like escapes of firedamp, usually take place when 

 the barometer is low ; but a more general rule for the occurrence of the 

 former phenomena is, that they are observable whenever the district of 

 observation is crossed by a steep barometric gradient. It would be of 

 interest and of value to determine whether the escape of firedamp follows 

 a similar rule, and, for the welfare of our mines and miners, tromometric 

 observations oug-ht to be inaugurated in the neighbourhood of some of 

 the more fiery of our mines. The movements to be sought for are not 

 minute earthquakes, tremors, or vibrations, but pulsatory movements 

 which may be compared to the swell upon an ocean. 



On the Overturning and Fracturing of Columns. 



In the Report for 1891 a short account of the results of experiments 

 on the overturning and fracturing of columns was given. Partly in 

 consequence of the terrible disaster on October the 28th of that year, 

 when thousands of structures of all descriptions were overturned, 

 shattered, or utterly ruined, these experiments are being extended. 



With regard to overturning we can now state with considerable 

 accuracy the acceleration necessary to overturn a column-like structure 

 of given dimensions. One set of experiments, in which Mr. F. Omori 

 has continued to assist me, shows most clearly that a group of similar 

 columns, whether they are large or small, heavy or light, so long as they 

 have tlie same ratio of height to breadth, with horizontally applied motion 

 will fall simultaneously. Further than this, the acceleration which was 



1892. I 



