129 



from Rhodes, and was composed of the same material as the fragments found 

 by us. 



My own impression is that a Roman villa stood somewhere in the immediate 

 neighbourhood. The shape of the ground rather encourages this idea. Magna 

 Castra at Kenchester may be about a mile <>S as the crow flies ; and Bishopstone, 

 where, as you know, traces of such a villa were clearly found, somewhat less. 

 Credenhill Camp, and the site of a small Roman station, now occupied by Brinsop 

 Church and churchyard, are nearer still ; so that the locality was a favourite one. 

 A small fragment of Oolite, like the oolite of Bath, which we found in the well, 

 may or may not be suggestive of some better class of dwelling having been on the 

 spot, of which it formed a part. I should imagine that the well ^vas used by the 

 people who lived in such house as the place to shoot their rubbish in. The stones 

 nearer the bottom may have been taken from some ruder building pulled down ; 

 while the bones of animals and the broken crockery speak for themselves. 



So far, I am afraid that I have not said much to interest j'ou, and only the 

 reasons that I gave when I began have induced me to trouble you with so much 

 detail. But there are now two points which I should like anyone to inform me ou, 

 and which I suggest to you for your opinion. The first is : How did it happen 

 that the ground sank down in the regular way which I described to you? My own 

 theory is that the well rested on a rocky floor, to which we were unable to reach, 

 and that beneath that floor was a spring cavity, which had been always filled with 

 water ; but I think it possible that tiie excessive rainfall which we had in May, 

 1886, and which amounted at Brinsop to 7 '25 inches, may have produced some 

 change in the soil such as allowed this cavity to be drained. Then I suppose that 

 the flooring may have given way, and let the whole well sink bodily down. Now 

 you see that this theory demands this heavy draught on your credulity — that the 

 steening of the well, its contents of stone and earth, and the soil which backed it 

 outside the steening, must have all gone down uniformly together, like a cartridge 

 in a gun barrel. This is very hard to realise as being so. And what makes it 

 more strange, if it were so, is that, built as I told you the well was, with its rapidly 

 enlarging and then contracting space, the stones should not have been loosened by 

 the movement downwards, but should have allowed us to get down as far as we 

 did. Still, I cannot think of any other way in which the depression on the surface 

 of the field could possibly have been produced so regularly in its shape, and so 

 suddenly in the time of its being formed, as was the case. 



The other point which admits of some little interest relates to the recess 

 which I said we found in the side of the well at the depth of 30 feet. What in 

 the world was that for ? The well, from its narrow opening must have been hard 

 enough to construct as it was. What made its builders add this difficult piece 

 of building at such depth underground ? If the whole thing had been larger, and 

 not of such considerable depth, one might have fancied it designed for a store 

 place of some kind at a time when it was drier than now. Is it known that there 

 are instances of this peculiarity of construction elsewhere, such as might throw 

 light on this particular building, what it was built for, and who were tiie builders 

 of it ? I pause for a reply, aud thank you for your kind attention. 



