14(6 REPORT— 1897. 



3. The times at which various maxima of motion are attained, and the 

 tilting they represent expressed in seconds of arc. 



4. The total duration of the disturbance. 



5. A tracing of the photogram. 



IV. Observations at Carisbrooke Castle and Shide. 

 By John Milne, F.R.S., F.G.S. 



In the report for last year it was stated that at about the end of June, 

 through the kindness of Mr. A. Harbottle Estcourt, Deputy-Governor 

 of the Isle of Wight, I had been enabled to establish a second horizontal 

 pendulum at Carisbrooke Castle, and a description of this installation, 

 together with that at Shide, was given in some detail. The object of the 

 second installation was to see how far the records of two similar instru- 

 ments at some distance apart coincided in character. The Shide records, 

 as already reported upon, consist of movements due to earthquakes which 

 have originated at some distance — disjdacements, which show that the boom 

 of the instrument has suddenly been caused to swing or change its zero 

 points ; tremors, which are irregular swingings of the boom extending over 

 many hours or several days ; pulsations, which are regular back and forth 

 movements of a pendulum, which movements have periods of two or three 

 minutes ; diicrnal waves and seasonal waiiderhigs. 



In the following report these movements will be discussed in the order 



in which they are here mentioned, the Carisbrooke records being taken 



first. 



The Carisbrooke Records. 



The Carisbrooke records were obtained between June 16 and August 31, 

 1896. Because the journey to Carisbrooke and back entailed a walk of 

 four miles, it was only visited once every twenty-four hours. For this 

 reason, together with the fact that the clockwork arrangement often 

 failed to drive the photographic paper — an imperfection which has since 

 been remedied — there were very many interruptions in the continuity of 

 the records. Notwithstanding this, a sufficient number were obtained to 

 compare with corresponding records at Shide, and to indicate the character 

 of Carisbrooke as an observing station. 



The earthquakes recorded were as follows : — 



July 5. — Four exceedingly small, elastic switchings of the boom, the first 

 at 3 hrs. 6 mins. 47 sees., and the last at 3 hrs. 44 mins. 7 sees. 



July 21. — At 7 hrs. 3 mins. 53 sees, there was a small elastic disturbance 

 with 5 maxima. 



August 30. — A very heavy disturbance (see fig. 2), corresponding in 

 time, points of maxima, and other detail with the Shide record, No. 36. 

 This earthquake had its origin in Japan. 



The first two records, which have amplitudes of "5 to 1 mm., do not 

 •correspond with records at Shide, whilst there are similar nSnute dis- 

 turbances recorded at Shide which are not visible on the Carisbrooke 

 photograms. The conclusion, for the present, at least, is that these small 

 tremors, which suggest an elastic switching of the end of the boom, are 

 very often of local origin, whilst earthquake movements of a pronounced 

 character are recorded in a similar manner at both stations. The reason 

 that no record was obtained at Carisbrooke on August 26 (No. 35 in the 

 Shide list) was because on that day the recording apparatus was not in 

 operation. The days of such interruptions are indicated on the general 

 list of disturbances, pp. 147, 148. 



i 



