190 REPORT—1896. 
minutes followed by a sudden return to the normal position, Minute 
spiders have sometimes found their way inside a case, but it is very doubt- 
ful that they should be able to cause the sudden disturbances shown and 
finally leave the boom in its normal position and free to swing. With 
records from nineteen installations in Japan I never remember observing 
movements of this character. Whatever may be the cause of these dis- 
lic. 2.—Displacements on September 10. 
Sept 8 
1895 
placements it is probably very local in its operation, and therefore they 
cannot be regarded as earthquakes. 
The duration of a displacement is evidently the length of time it takes 
a pendulum which has been slightly deflected to come to rest. Witha 
light boom this is about 15 minute but with a heavy boom it may be 5 
minutes. A group of disturbances may extend over 20 or 30 minutes. 
One group of 40 occupies 3 or 4 hours. 
An earthquake originating at a distance has the appearance of fig. 3, 
which is probably the Shide record of the commencement of shocks which 
shock Cyprus on June 29, 1896. 
Between August 19, 1895, and March 27, 1896, or during 202 working 
days, 485 sudden displacements and earthquakes were recorded. 
In the following list the records referring to sudden displacements are 
those which succeed each.other at short intervals, and are marked 
‘sudden’ or ‘strong.’ Those which are followed by the remark ‘slight’ 
or ‘moderate’ may be due to actual earthquakes, the origins of which in 
some instances have been at great distances. 
Records (August to November) marked A approximately correspond 
in time to disturbances noted by Professor Agamennone in the ‘ Bulletin 
Météorologique et Seismique de lObservateur Impérial de Constantinople.’ 
T refers to records published by Professor Pietro Tacchini (for September 
and November) in the ‘ Bollettino della Societa Sismologica Italiana.’ G 
refers to records received from Professor Gerland at Strassburg, and K to 
those from Professor Kortazzi at Nicolaiew. 
These references, it will be observed, are very incomplete, and are 
only made up to the end of March 1896. Ina subsequent report it is 
hoped that these will be completed, whilst the list itself will be extended 
up to date, and include the observations made at Carisbrooke. 
The corrections are given in minutes and seconds, and are to be added 
or subtracted as indicated. From August 19 to October 27 the times 
