O:^ SEISMOLOGICAL INVESTIGATION. 109' 



more likely to be still in progress than is one which can only be shown 

 to affect Palaeozoic or Secondary rocks. Not only have the older 

 movements in many cases ceased long since, and have given place to move- 

 ments in different directions ; but a fault which has long remained without 

 movement tends to become closed and re-cemented, so that there is a 

 considerable likelihood that any future movement may not follow exactly 

 the same line, even though the strain be in the same direction. 



3. The fault should crop out on ground fairly level, and in hard rocks, 

 otherwise the observations may be masked by the slight irregular ' creep '' 

 of the surface downhill, and no firm foundation for the apparatus be 

 obtained. 



4. It is desirable that the rocks on the two sides of the fault, though 

 geologically far apart, should be as like as possible in lithological charac- 

 ter, so that any surface movements due to change of temperature or 

 absorption of rain-water should affect the two sides alike. 



5. In order to avoid complications through slow solution of the rocks 

 by percolating rain, a fault bringing together insoluble silicious rocks 

 would be preferable to any other. 



6. As the records to be obtained may throw great light on movements 

 of the earth's crust, it is desirable that the fault selected for observation 

 should be one belonging to a set of disturbances of great magnitude, 

 having common characteristics, and affecting a considerable area. It is 

 therefore important that the district chosen should be one which has been 

 carefully studied geologically, and of which the structure is thoroughly 

 known. 



These various conditions, added to the consideration of convenience of 

 access of the locality, availability of a skilled observer, availability of the 

 land, and other minor points, made a series of requirements not easy to 

 satisfy, and I will now indicate in what respects the site finally selected 

 comes up to or falls short of the ideal set before us. 



Consideration No. 2 confines us at once to the only area in Britain in 

 which large earth-movements of Tertiary date can clearly be proved to 

 have taken place. This area may be taken to lie between the North 

 Downs and the English Channel, and to extend as far west as Weymouth 

 and Abbotsbury. But only the parts of it in which Tertiary rocks are 

 still preserved will do for our purpose ; the reason being that older move- 

 ments of the same general character affected the Jurassic and Lower 

 Cretaceous rocks. These intra-Cretaceous disturbances cannot always be 

 distinguished from the Tertiary movements, in the absence of the uncon- 

 formable Upper Cretaceous and Tertiary strata. Thus in the Wealden 

 area a good many faults are believed to affect the Lower Cretaceous rocks ; 

 but they are of no great magnitude, and it is impossible at present to 

 differentiate those of Tertiary date from the older series. 



We are thus confined, by a process of elimination, to the sharply folded 

 belt which occupies the southern part of the Hampshire Basin and 

 includes the northern half of the Isle of Wight. Even over this area it 

 would only be possible to use Mr. Horace Darwin's apparatus at certain 

 points ; for much of the country is sharply folded without faulting, and 

 any earth-movements now in progress could only be measured by careful 

 levelling and triangulation. Thus we are confined ultimately to a limited 

 highly disturbed and faulted belt, which extends east and west through 

 the centre of the Isle of Wight and reappears in Dorset between Studland 

 Bay and Abbotsbury. 



