94 REPORT—1905. 
four lunar hours ; and (3) the declination of the sun and the moon at the 
time of the shock. 
The results are tabulated according to the declination and time, the 
classification being such that each group of shocks represents a period 
during which the distribution of the tidal stresses was practically uniform. 
The grouping adopted has also the advantage that, by taking 11° as a 
limit, and dividing the year into three portions, according as the decli- 
nation of the sun is greater or less than that limit, we get three almost 
equal periods, corresponding closely to the four winter months, November, 
December, January, and February ; the four summer months, May, June, 
July, and August ; and the spring and autumn months, March, April, 
September, October respectively. 
The discussion of the record as a whole gave no indication of tidal 
periodicity, differing in this respect from that of the seismograph at 
Shillong, but this may well be due to the difference in the character of the 
record. At Shillong only true tectonic earthquakes of a degree of violence 
corresponding to at least III of the Rossi-Forel scale were registered ; at 
Tokyo, on the other hand, the greater part of the record is composed of 
shocks which could not be felt, which were so feeble that, even instru- 
mentally, they could only be detected, not measured, and which are very 
probably of quite a different nature to the tectonic earthquakes, in which 
alone the effect of tidal stresses is to be looked for. 
If only those shocks which had a measurable amplitude are con- 
sidered, the distribution is as represented in the second set of tables. 
The total number of shocks is too small for detailed discussion, but a 
comparison of the ratio of shocks occurring between the hours of 18 and 6 
on the one hand and between 6 and 18 on the other, when the declination 
is more than 11° north or south, and for the whole record, gives the 
following result :— 
Declination | Declination 
>11° N. | >11° g. All Shocks 
6h. to | 18h. to! 6h. to | 18h. to| Gh. to | 18h. to 
18h. | 6h. | 18h, | 6h. 18h. 6h. 
solar {No.of shocks ./ 81 lik G?* Paral St oegt 211 209 
; Ratio to mean .| 1-10 “90 ‘Oba 21505 1-01 1:99 
aes {No.of shocks . 67 | 64 | 65 | 86 | 210 210 
| | Ratiotomean .| 1-02 “98 “86 114 | 1:00 1:00 
From this it will be seen that in the case of both sun and moon the 
ratio of shocks occurring in the half-day containing the upper culmina- 
tion to those occurring in the half-day containing the lower culmination 
is above the average ratio when the declination is north, and below it 
when the declination is south. As the rate and range of variation of 
tidal stresses in Japan is greater in the half-day containing the upper 
culmination when the declination is northerly, while the reverse is the 
case when the declination is southerly, this result falls in with the 
supposition that tidal stresses have some effect, and tend to increase 
the frequency of earthquakes at the time when they have the greatest 
range and rate of variation. As regards the effect of the attraction of 
the sun, this principle may be more briefly expressed : that the ratio of 
day to night shocks is higher in summer than in winter, 
