ON SEISMOLOGICAL INVESTIGATION. Ill 



the ground is too steep and the rocks too soft to yield satisfactory fixed 

 points. Along the same line the junction of the Chalk and Eocene is 

 again slightly faulted near Lulworth ; but the fault is of small magnitude, 

 and the adjoining rocks are too much shattered for our purpose. The 

 Durdle fault runs parallel with and close to high cliffs, so that delicate 

 observations might be entirely masked by movements caused by the 

 gradual removal of large masses of rock by the sea on the south and the 

 consequent rise of the strata qn that side. At Bat's Head the Isle of 

 Purbeck Fault is finally lost beneath the sea, and the shattering of the 

 rocks is too great to allow of exact observations. This fault does not 

 reappear in the Weymouth area. 



There still remains one of the most important Tertiary disturbances 

 in the district, that known as the Ridgeway fault. This also is an over- 

 thrust fault cutting through a monocline, or through the north limb of a 

 sharp anticlinal fold. Its date is clearly later than the Bagshot period • 

 its magnitude is great, and if any of the Tertiary faults are still under- 

 going changes, this one is likely to partake in the movement. It brings 

 together rocks of very different ages and of varying character, so that the 

 choice of exact locality for the observations depended on the discovery of 

 a spot where the fault is a clean fracture, whei'e the rocks on each side 

 are hard and of fairly similar lithological character, and where the ground 

 is sufficiently level for the apparatus. Along a good deal of its course 

 there is much fault rock or broken ground, and in most parts the strata 

 on one or both sides are soft. These parts would not be convenient or 

 satisfactory for our purpose. For various reasons the choice narrowed 

 down to the neighbourhood of Poxwell, where Middle or Lower Chalk 

 abuts against Lower Purbeck ; or to the district between Upway and 

 Portisham, a distance of four miles, where Upper Chalk is faulted 

 against strata close to the base of the Lower Purbeck, or even against 

 Portland Beds. Of these localities Upway was chosen (fig. 2), for there 

 the deep lailway-cutting has laid open the structure of the disturbance 

 and within a reasonable distance, though not too near, was a piece of 

 fairly level ground, one end of which had been opened for chalk-pits and 

 the other for quarries in the Purbeck Beds. The railway-cuttintr itself 

 would not have been satisfactory, for in it a wide dyke of ' fault-rock ' 

 composed of Oxford Clay and Cornbrash, occurs, and south of the fault 

 there are soft rocks. Besides this, soft strata in a deep cuttinw will 

 almost certainly be subject to slow ' creep ' to such an extent as entirely 

 to mask any deeper-seated movement. 



The site finally selected proved by an unexpected series of coincidences 

 to be particularly convenient. It is broken ground, now only used for 

 rough pasture and not liable to be disturbed by the plough ; it beloni-s 

 to Gonville and Caius College, Cambridge, who have most kindly done all 

 in their power to help us in the experiment. Our thanks are not only 

 due to the College, but also to the tenant for his assistance in carrying 

 out the work. And last, but not least, it was conveniently accessible to 

 the member of the Committee who was prepared to undertake the 

 recording. 



While our excavations were being made I examined them, and noted 

 as exactly as possible the geological conditions in the immediate neigh- 

 bourhood, for the fault varies within very short distances, and has 

 changed completely in the two hundred yards between the railway 

 cutting and our selected site. In that short distance the dyke of Oxford 



