1168 
MR. C. DAVISOX OX THE ANNUAL 
specific gravity of mercury, li the maximum range of the barometer, and 2rrh the 
wave-length of tiie barometric undulation. To obtain an estimate of the maximum 
displacement ad tlie surface, namely, gwhhj^v, Professor Darwin takes iv = 13‘6, 
h = 5 centims., = 5000 miles = 4'8 X 10® centhns., nearly, and vjg = 3 X 10®, 
which is equivalent to assuming that the superficial layers of the earth are more 
rigid than the most rigid glass. With these data he finds the maximum displace¬ 
ment of the surface to be 4'5 centims., so that the ground is 9 centims. higher under 
the barometric depression than under the elevation. 
The ratio of the vertical displacement at depth x to that at the surface is 
(1 -f xjh) e “ , 
or 
1 — .t^/ 2/P fi- x^/Sh^. 
Thus, the displacement at first diminishes very slowly as the depth increases, and is 
independent of the modulus of rigidity and the barometric range; being, with the 
above value of h, only one-fifth per cent, less at a depth of fifty miles than it is 
at the surface. 
Now, since most or all earthquakes probably originate at a depth much less than 
this, and since the fault-slip which produces even a moderately strong shock must be 
very small, for even in the most violent earthquakes it is rarely perceptible at the 
surface; also, since the work to be done is, not the compression of solid rock, but the 
slight depression of a fractured mass of rock whose support is very nearly, but not 
quite, withdrawn ; on these accounts it seems possible that the annual variation in 
barometric pressure may be comjDetent to produce the annual variation in seismic 
frequency. 
It is not to be expected that the amplitude of the annual seismic period in different 
districts should be proportional to that of the annual barometric period; but if there 
is a connection between the two periods the epoch of the seismic maximum should 
coincide with, or follow closely after, that of the barometric maximum. 
88. Over the greater part of the northern hemisphere, the annual barometric 
maximum occurs as a rule either in November or December, except at loft}^ stations 
or in high latitudes. In Europe the maximum occurs nearly everywhere in November^ 
except along the western border in latitudes north of about 45°, over which the 
Atlantic conditions prevail for some distance inland. In Asia the maximum is nearly 
always in December. In North America it occurs in November and December, and 
along part of the western coast in January. 
In the seismic districts of the Southern Hemisphere, the maximutn occurs as a rule 
in the corresponding months :—in New Zealand in April, in the south-east of 
Australia in May or June, in South America in June or July, with some few 
exceptions in each case, depending either on the altitude or neighbourhood to the 
coast. 
