ON EARTH TREMORS. 145 



Earth Tremors. — Report of the Committee, consistinrj of Mr. G. J. 

 Symons, Mr. C. Davison (Secretary), Sir F. J. Bramwell, Pro- 

 fessor G. H. Darwin, Professor J. A. Ewing, Dr. Isaac Roberts, 

 Mr. Thomas Gray, Sir John Evans, Professors J. Prestwich, 

 E. Hull, G. A. Lebour, R. Meldola, ami J. W. Judd, Mr. M. 

 Walton Brown, Mr. J. Glaisher, Professor C. G. Knott, Pro- 

 fessor J. H. PoYNTiNG, Mr. Horace Darwin, and Dr. R. Copeland 

 (drawn up by the Secretary), appointed for the Investigation of Earth 

 Tremors in this Country. 



APPENDIX PAGK 



I. Account of Observaiwns made with the Horizontal Pendulum at Nicolaieiv. 



By Professor S. Kortazzi 155 



II. The Blfilar Pendulum at the Royal Observatory, Edinburgh. By Professor 



R. Copeland 158 



2Ir. H. Darwin^ Bifilar Pendulum. 



The preliminary trial of the bifilar pendulum last year led to the dis- 

 covery of one or two possible sources of error, chiefly resulting from altera- 

 tions in the distribution of temperature near the instrument. In order to 

 eliminate these as far as possible, Mr. Darwin has made several changes in 

 the latest form of the pendulum.' 



When the gas-jet was kept burning for some time, it was found 

 that the expansion of the tube produced an apparent tilting to the east, 

 i.e., away from the source of heat. As soon as the flow of heat through 

 the instrument became nearly steady, a far more considerable movement 

 of the mirror in the opposite direction became evident, which was perhaps 

 due to the action of convection currents in the surrounding oil. 



The expansion of the tube is greatest on the side towai-ds the gas-jet. 

 Its disturbing effect is therefore a maximum when the gas-jet is in a plane 

 at right angles to that in which the silver wire lies. In the new instrument 

 the mirror is held in a frame so that the plane of the mirror is perpendicu- 

 lar to that of the silver wire, and the principal effect of the expansion is 

 merely an inappreciable change in the sensitiveness of the pendulum. At 

 the same time we should expect that this method of mounting the mirror 

 would diminish the disturbing action of convection currents, as the surface 

 exposed to them lies chiefly in a plane at right angles to that in which the 

 movements of the ground are being measured. 



In order to avoid any straining of the tube the lever used in deter- 

 mining the angular value of the scale divisions is prolonged above the 

 tilting- screw. To this upper portion is attached a movable weight, which 

 can be adjusted so that the centre of gravity of the lever coincides with 

 the axis of the tilting-screw. The lever is moved by a rocking-arm worked 

 from a distance by a pair of pneumatic bellows. 



The instrument rests on three foot-screws, two of which are in a line 

 parallel to the plane of the silver wire. A tangent-screw is connected 

 with these two, so that one can be raised and the other depressed by an 

 equal amount, and so enable the sensitiveness to be varied. A second 

 tangent-screw is attached to the third foot-screw, or ' back-leg,' by means 



' For the account of these improvements I am indebted to notes supplied by Mr. 

 Darwin. See also a paper, « Bifilar Pendulum for Measuring Earth-tilts ,' Nature 

 vol. 1. 1894, pp. 246-249. 



1894. r 



