ON SEISMOLOGICAL INVESTIGATION. 81 



IV. On the Comparison of Records from three Milne Horizontal 

 Pendulums at Shide. 



At Shide three Milne horizontal pendulums are installed on two 

 similar brick piers, 2 ft. 6 in. distant from each other. Each pier is 

 1 ft. 6 in. square, and rises 4 feet above its footings, which rest on 

 concrete. One pier was built in May 1897, and the other in November 

 1902. The instruments are described in the ' British Association Report' 

 for 1902, p. 60. The older of the tAvo piers carries the type instrument, 

 which has a period of 16 seconds and records east and west movements. 

 This is referred to as pendulum A. Pendulum B has the same period, 

 and is oriented in the same direction as pendulum A. Pendulum C, 

 which with B forms the Yarrow instrument, has a period of 20 seconds 

 and records north-south motion. 



The following results refer to seismograms obtained between November 

 21, 1902, and March 24, 1903, or in the Shide register Nos. 659 to 693. 



Times of C om,mencement. — Out of twenty-six cases the times of com- 

 mencement of A and B have in eleven instances never differed more 

 than one minute. When this limit has been exceeded the movements to 

 be measured have usually been slight thickenings or blurs. Comparing 

 C with A or B, out of nineteen cases there are ten instances falling within 

 the one-minute limit. 



Times of Maxima. — The times at which maxima have occurred as 

 recorded by A and B have not differed more than two minutes in ten 

 instances out of fourteen records. When this limit has been exceeded 

 the records usually refer to slight thickenings in traces in which one out 

 of several points might be selected as a maximum. 



The maxima for C agree within the two-minute limit with those of 

 A and B eight times. 



AmjMtudes. — The amplitudes recorded by A and B have in twenty- 

 five cases only once differed 1 mm. from each other. The records 

 obtained for C have not differed greatly ('5 to 1 "5 mm.) from those shown by 

 A and B. Out of twenty instances the C records were eleven times larger, 

 three times smaller, and nine times equal to those shown by A and B. 



Durations. — Out of twenty-one instances the records of C were three 

 times greater, six times smaller, and twelve times practically equal to 

 those obtained from A and B. 



These comparisons are similar to comparisons of records from two 

 similar seismographs made by Dr. Charles Chree, F.R.S., at Kew.' 



V. On the Coinparison of Earthquake Registers from, Shide, Kew, 

 Bidston, and Edinburgh. 



In the 'British Association Reports,' 1901, pp. 44-50, and 1902, 

 pp. 73, 74, references are made to series of earthquake records obtained 

 I at Kew, Shide, Bidston, and Edinburgh, stations which are respectively 

 I situated on alluvium, chalk, sandstone, and volcanic rock. 



The following notes chiefly refer to observations made between July 1 

 and December 31, 1902, dui'ing which period the instruments at the 

 different stations have been so adjusted that 1 mm. deflection of the 

 [outer end of the boom corresponded to a tilt of the bed plate of 0"'5. 



' See B.A. Eejjort, 1901, p. 51. 



1903. a 



