386 DR. CHARLES DAVISON ON THE [Aug. 1902, 
(September 30th). The records are comparatively numerous, but 
many amount to little more than a statement that the shock was 
felt; and it is impossible to draw the isoseismal lines with any 
-approach to accuracy. On the map (Pl. XII) the isoseismal 5 alone 
is shown, and even the course of this must be regarded as doubtful. 
Its length, as drawn, is 38 miles, width 25 miles, and the contained 
area about 750 square miles. ‘The centre is situated 2 miles east- 
‘south-east of Dochgarroch, and 1-7 miles on the south-east. side of 
the fault, and the longer axis is roughly parallel to the fault-line. 
Outside this isoseismal, the shock was felt at Little Scatwell, 
Relugas, Dunphail, and Grantown, which are respectively 18, 24, 
25, and 26 miles from the centre of the isoseismal 5. There are 
also records of its occurrence at Deskford, Ordiquhill, and Banff, 
places near to one another, and respectively 57, 59, and 67 miles 
from the centre ; and at Comrie and Crieff, distant 73 and 76 miles. 
The absence of observations at intermediate localities makes these 
records somewhat doubtful, but there is nothing impossible in an 
after-shock of this intensity being felt so far from the epicentre. 
The shock consisted of two or three distinct oscillations, the 
average of seven estimates of its duration being 21 seconds. At 
Dochgarroch and Inverness, and in the immediate neighbourhood, 
‘these vibrations seemed to be nearly vertical. 
The sound was certainly heard by 62 per cent. of the observers. 
It was compared to passing waggons, etc. by 14 per cent. of those 
who described it, to thunder by 43 per cent., wind by 14, the fall 
of a heavy body by 7, explosions by 14, and to miscellaneous sounds 
by 7 per cent. The beginning of the sound preceded that of the 
shock in 62 per cent. of the records, coincided with it in 25, and 
followed it in 12 per cent. The end of the sound coincided with 
that of the shock in 17 per cent., and followed it in 83 per cent., of 
the records. The duration of the sound was generally greater, and 
never less, than that of the shock. 
The direction of the longer axis of the curve and the position of 
the centre favour the connection of this earthquake with the 
boundary-fault. Moreover, the sound was heard at two places— 
Garve and Little Scatwell—on the north-west side of the iso- 
seismal 5, and respectively distant 7 and 6 miles from it. As the 
sound-vibrations from the upper margin of the focus would be more 
readily heard than those from the lower margin, this fact confirms 
the inference that the hade of the earthquake-fault must be to the 
south-east. 
September 18th, about 6.25 a.m.: Inverness.—A noise was heard, 
but much slighter than that which accompanied the shock at 
3.56 A.M. 
h. September 18th, 9 a.m. 
Intensity, 5; epicentre, lat. 57° 27-0' N., long. 4° 15:1’ W. Number of 
records 26, from 18 places (Pl. XII). 
The disturbed area of this shock is elliptical in form, and is 
