92 REPORTS ON THE STATE OF SCIENCE.—1912. 
to in the above groups I find eight instances in which the direction of 
the N.-S. motion has exceeded the E.-W. motion by intervals of from 
10 to 69 minutes. The azimuths of the origins of these earthquakes 
were 41, 36, 45, 8, 20, 18, 15, and 5 degrees east of north, which is 
the direction of Japan or the Central Pacific. In six instances where 
the K.-W. motion exceeded the N.-S. motion by intervals of from 10 to 
46 minutes the azimuths of the origins were 70 E., 79 W., 90 W., 
90 W., 69 E., and 12 W., which with the exception of the last suggest 
origins in Central Asia or Central America. It would therefore appear 
that marked differences in the duration of two rectangular components 
of motion are possibly associated with the azimuth of its origin. 
VI. Megaseismic Activity and Periods of Quiescence. 
In the British Association Report, 1910, p. 54, I gave a nole on 
megaseismic activity and rest. The resuli showed that a large group 
of megaseisms was followed by a long period of quiescence, while small 
groups were followed by comparatively short periods of quiescence. 
This result was based on the examination of twenty-eight groups of 
large earthquakes. The present discussion is based upon eighty groups, 
found in the Registers for the years 1899-1908 inclusive. The number 
of earthquakes in these groups varies from two or three to fifteen. In 
{wo cases, however, the number of earthquakes is forty-six and fifty- 
one. If an earthquake has been recorded over the whole world I have 
considered its intensity double that of a disturbance which is only 
recorded over a hemisphere. The intensity of a group is assumed to be 
the sum of the intensities of each earthquake it contains. Groups 
usually extend over from one to+three days, and it is seldom they extend 
over more than six days. The intensity per day is the intensity of a 
group divided by the number of days over which it extended. This quan- 
tity does not appear to show any relationship to the number of days of 
rest which preceded or followed the group which it represents. 
The number of days which have elapsed between the centre of one 
group and the centre of the group which follows has usually been from 
fifteen to fifty days. In the accompanying figure the number of earth- 
quakes in different groups are plotted in relation to these intervals. 
At first sight it would appear that these two quantities had a rough 
relationship, but it must be remembered that the intervals between 
centres of groups have frequently been increased by the duration of the 
groups. 
VII. Megaseismic Frequency in Different Seasons. 
In the Report for 1906 for the seven years 1899-1905 I compared 
the frequency of large earthquakes in the following three districts :— 
1. Districts A, B, C refer to the East Pacific coast north of the 
Iiquator, including the Antillean fold. 
2. Districts E, F refer to the West and South-west portions of 
the North Pacific. 
3. District 1X, or the various folds extending from the Balkans to 
the Himalayas. 
The ratios of the numbers of disturbances which were noted in 
