630 ME. C. DAYISON ON THE INVERNESS 



Taking into account the long- duration of the shock (at some 

 places as much as 30 seconds) and the elongated form of bot;Ji iso- 

 seismal lines, especially that of intensity V., it is probable that the' 

 seismic focus was of considerable length in a horizontal direction, 

 perhaps as much as four or five miles, if not more. Since the in- 

 tensity of the principal vibrations was greatest towards the begin- 

 ning in the neighbourhood of the south-west end of the focus, and 

 since a large number of the places where the sound preceded the 

 shock are grouped about the continuation of the fault-line towards 

 the north-east, we may infer that the amount of the fault-slip was 

 greatest near the south-west end of the focus, and died out more 

 gradually towards the north-east end. This is what we should 

 expect, the junction of the branch fault with the main fault being 

 to the south-west of the epicentrum. 



2. Subsequent Shocks. — Turning now to the later shocks, we have 

 the epicentra of three of the most important determined. Eeferring 

 to the map (p. 620), it will be seen that these three points (h, d, K) lie 

 on the south-east side of the main fault, but much nearer to it than to 

 the branch fault. The epicentra of the other accessory shocks can- 

 not be exactly ascertained, but it is obvious, from the positions of 

 the places at which they were felt, that, if connected with either, 

 they are more closely related to the main than to the branch 

 fault. 



In the first accessory shock (6, Kov. 15, 18*" 15"'), the vertical 

 component of the motion at Inverness was upward first and then 

 downward. This was also the case in the sixth shock ( /, iSTov. 19, 

 1'^ 40™) at Inverness, and, again, in the last shock of all (K, Dec. 14) 

 at Aldourie. Ealnafettack, Dores, and Inverness — all places to the 

 south-east of the line of the main fault. At Inverness, also, it was 

 noticed that the direction of motion of the last shock was more 

 nearly vertical than in any of the preceding shocks. 



In these three shocks at least, then, the fault-slip must have been 

 such that the rock-mass on the south-east side slipped downward 

 relatively to the rock-mass on the north-west side. Judging from 

 the positions of their epicentra, we infer that the depth of the 

 seismic focus of the first accessory shock (6) was greater than that 

 of the last shock (K), and the depth of the seismic focus of this, 

 again, greater than that of the third accessory shock (d). 



3. Conclusion. — The main fault and the branch fault include be- 

 tween them a great wedge of rock, capped for the most part by a 

 mass of Old Red Sandstone, and tapering ofi" towards the south- 

 west. Now, it is more likely that this wedge should subside as a 

 whole than that it should remain fixed while the great rock-masses 

 on either side of it are both elevated at practically the same time. 

 The positions of three known epicentra (h, d, and K), almost opposite 

 to that of the principal epicentrum (A), are also in favour of the 

 former view. I conclude, then, that the earthquakes were caused 

 by slight interrupted subsidences of this wedge of rock ; that the 

 first, and at the same time the greatest, slip took place along the 

 branch fault, though it may possibly have been preceded by a few 



