378 DR. CHARLES DAVISON ON THE [Aug. 1902, 
to the Rossi-Forel scale, but have employed only one test for each 
degree.! At the head of the account of each earthquake, the 
position of the epicentre is given to the nearest tenth of a minute 
of latitude and longitude; and, in the account itself, I have stated 
the approximate position with reference to Dochgarroch, a place 
which lies close to the epicentres of nearly all the shocks. In the 
maps (Pls. XI & XII), the isoseismals are indicated by continuous 
lines when some confidence may be placed in their accuracy; and, 
in parts, by broken lines when their course is doubtful, owing to the 
scarcity or absence of observations. 
As the principal earthquake occurred shortly after midnight, many 
of the observers were asleep when it began. In most cases, how- 
aver, they were evidently awakened by the preliminary sound ; for, 
according to those who were awake, the beginning of the sound 
preceded that of the shock in 72 per cent. of the records, coincided 
with it in 21, and followed it in 8 per cent. According to these 
who were asleep, the corresponding figures were 72, 20, and 9. 
In other respects, also, the two classes of observers differ but 
slightly ; and the only point on which I have not availed myself of 
the evidence of those who were at first asleep, is the relative 
intensity of the two parts of the shock. | 
Tl. Favuts oF tHe Eptcentrap District. 
With a few exceptions, the earthquakes originated beneath the 
district lying between Inverness and the north-eastern end of Loch 
Ness. The chief structural feature of this district, is the great 
pboundary-fault which runs from Tarbat Ness along the eastern coast 
of Ross-shire, and follows the line of the Great Glen. The mean 
direction of this fault is about N. 35° E., and 8S. 35° W., and its 
hade is to the south-east. Its course within the epicentral district 
(for which I am indebted to Mr. J. Horne, F.R.S.) is shown in 
Pl. XIT and the text-figure on p. 392. 
Several of the after-shocks were observed only at Dalarossie and 
a few other places in the valley of the Findhorn ; and it is probable 
that they were due to movements along a fault in this district. 
The seismic evidence is insufficient to determine the position of 
this fault, except that it may run along the line of the valley. 
Mr. Horne informs me that the ground there has not yet been 
surveyed, but that he has proved the existence of faults along the 
Findhorn Valley near Drynachan Lodge, which lies about 11 miles. 
down the valley from Dalarossie. 
Ill. Fore-Suocks. 
The beginning of the series, which culminated in the earthquake 
of September 18th, seems to have occurred some time during the 
preceding summer’ months. No precise dates are available, but 
1 Phil, Mag. ser. 5, vol. 1 (1900) p. 51. 
