APPENDIX. 



the dimensions of each, and the variation of the compass. 1 The remaining pages 

 should be ruled according to the form given below. The first column is for the 

 date, and the second for the time at which the earthquake was experienced ; and for 

 this it would be well to abolish A.M. and p.m., and consider the day as divided into 

 24 hours ; thus, 3-40 p.m. would be written 15 h. 40'. The third column is for 

 the number of cylinders overthrown ; the fourth for the direction in which they are 

 overthrown, and if, as is probable will be found to be the case when more than one or 

 two cylinders are overthrown, the direction is not the same in every case, the direction 

 in which each separate cylinder is overthrown should be noted ; in the fifth column 

 this direction should be entered, corrected for magnetic variation ; the sixth column 

 should contain the velocity of shock corresponding to the largest cylinder overthrown ; 

 and the last column is for any remarks as to the general nature of the earthquake, 

 or its effects, that may be noticed. These directions will, perhaps, be made clearer by 

 the examples given below, which, it is probably needless to remark, are purely 

 imaginary : — 



Record of Earthquakes during the month of January 188 at Cachar. 



Date. 



Hour. 



No. of cylin- 

 ders over- 

 thrown. 



Observed 

 direction of 

 overthrow. 



Corrected 



for mag. var. 



= 2° 30' E. 



Velocity of 

 hor. comp. 

 of shock. 



Remarks. 



January, 2nd 



3 h. 30' 



1 



N. 15° W. 



N". 12° 30' W. 



•6 





Ditto, 8th 



12 h. 27' 



3 



N. 5° E. 



N. 7° 30' E. 



1-5 





Ditto, 26th 



Oh. 40' 



10 



1 



2 



3,4,5 



6,7,8,9,10, 



N. 5° E. 



N. 4° E. 



N. 3 o, 30' E. 



N. 3° E. 



N. 7° 30' E. 

 N. 6° 30' E. 



N. 6° E. 

 N. 5° 30' E. 



5-0 





This instrument gives us the direction and velocity of the horizontal component 

 of the shock, the only other elements needful to form a complete seismometer are the 

 exact time and the velocity of motion of the vertical component ; of these, the former 

 necessitates an accurate time-keeper, while for the latter no simple instrument has 

 yet been proposed, but we would suggest that the following modification of one of 

 Mr. Mallet's self -registering seismometers will be found effective : Let a piece of wood, 

 say 6 inches square and 6 or 7 feet long, be taken ; then at each corner of one end a 



1 This may be obtained by so placing two plumb lines that a line drawn through both 

 of them may point to the pole star when the pointers of the Great Bear are either vertically 

 above or vertically below it, and then observing the angle between the direction given by 

 these two plumb fines and that given by the compass. 



( 96 ) 



