ON SEISMOLOGY AFTER THE WAK. 33 



lation. But among matters which appear to me the most pressing 

 at present are : — 



(1) The precise investigation of time curves and earth tides 



and their interpretation. 



(2) Investigations in manifestly seismic regions. 



The first is concerned with large-scale uniformities of the earth, 

 and the stations should be in non-seismic regions. The instruments 

 must be of the highest degree of precision as regards sensitiveness 

 and time accuracy, and it is vital that the vertical component, as 

 well as the horizontal components, of motion should be I'ecorded and 

 analysed. I do not think any great improvement in time curves will 

 be obtained without the use of the vertical component. A somewhat 

 analogous case occurs in terrestrial magnetism, where the vertical 

 component is the discriminating factor. 



Earth tides are important, among other reasons, because they 

 provide a check on the internal properties of the earth deduced from 

 the time curves. 



As regards the number of stations, the more the better, provided 

 they are really efficient. But much could be done by com- 

 paratively few stations. Is it too much to suggest that, as a start, 

 England and each of the Dominions — Australia, Canada, India, and 

 South Africa — should provide a fully-equipped first-order station? 

 These might be supplemented as occasion aiises in order to fill up 

 lacunae in the time cui'ves. 



For the second investigation more numerous stations are required 

 near the regions selected, but the instruments would not, in general, 

 require to be so sensitive, nor need three components be r-egistered 

 at all the stations. The object of such investigation is to some 

 extent local, and the co-operation of geologists would clearly be fruitful 

 in throwing light on the connection between volcanic action and earth- 

 quakes. One thinks of New Zealand as an obvious field for such work. 



Such a scheme requires close co-operation between parts (1) and (2) 

 for success, but I can ;imagine that the first-order station for (1) would 

 in its own region supervise (2) — e.g. Australia and New Zealand or 

 India and the East Indies. 



The first-order stations, limited in number, would in turn co-operate 

 with a central body, which I will assume is in England. 



This brings one to the question of apparatus and standardisation. 



Whatever instruments are used, standardisation is really essential. 

 It involves equipment for determining all the constants of the apparatus 

 and an experimental table by means of which a variety of artificial 

 motions can be given to the instrument and the results observed. 

 The necessity is almost self-evident, and its practical use was fully 

 demonstrated by Prince Galitzin. 



The question of instruments is somewhat more difficult, as we 

 may ex,pect these to be modified and improved in course of time. 



At present it is recognised that the most precise and most sensitive 

 installation for all three components is Galitzin 's system of aperiodic 

 pendulums, with galvanometric registration, and until something better 

 has been devised it would be a good thing if all the first-order stations 



