ox SEISMOLOaiCAL IXVESTItrATIOX. 



61 



direction, as inferred from experiments upon other similar columns, the 

 deflection per lib. pull will probably be one quarter of the above, or 

 0"-035. 



3. A pair of horizontal pendulums writing on smoked paper. The 

 booms of these pendulums, which are made of bicycle tubing, are each 

 1-030 ra. in length and carry at their outer ends 80-lb. weights. The 

 vertical support for these pendulums is a lamp-post bedded in concrete. 

 The vertical distance between the top of the ties which carry the weights 

 at the ends of the booms and the points where the inner ends of the 

 booms pivot against blocks attached to the lower part of the lamp-post is 

 7 feet IH inches. As at present adjusted a motion of the outer end of 

 the boom oriented north-south is by means of a light lever multiplied 

 six times, whilst the movement of an arm cranked to the east-west boom 

 is by a similar arrangement multiplied sixteen times. 



A plan of the arrangement is shown in fig. 1. The dimensions are in 

 millimetres. 



Fig. 1. — Measurements ia Millimetres. 





L evers 



--/■* 



oo - 



K J030--- 



N.S.Boom 



IS 



/6 



40- 



ftl 





From these dimensions it follows that if each boom were tilted equally 

 the diagram for th« north-south boom should be 3-1 times greater than 

 that given by the east-west boom. 



If the weights carried by the boom be regarded as centres of oscilla- 

 tion the multiplication of horizontal motion can be calculated from the 

 dimensions given in fiof. 1. 



This instrument takes the place of a pair of horizontal pendulums 

 which carried weights of 101b. This instrument, a spiral sprin" seismo- 

 graph, and a large balance arranged to record tilting, v.hich are referred 

 to in the Report for last year, are no longer in use. The oliaracter of 

 the records obtained from the last two of these instiumeiUs is referred 

 to on pp. 70 and 71. 



III. Tho Bc'cords of the Years 189!), 1900, and 1901. 



In the Report of the British Association for 1900, on p. 70, a map is 

 given showing the origins from which the earthquakes recorded in Britain 

 during the year 1S99 had radiated. These origins were determined by 

 methods explained on pp. 79 and 80 of that report. The accompanying 

 map, tig. 2 (Plate I.), gives a similar distribution — each carthtjuake being 



