264f Mr. Lyell on the Tubular Cavities^lled 'with Gravel 



of springs charged with carbonate of lime, such as are common- 

 ly seen to issue from chalk. This hypothesis of the adequacy 

 of pluvial waters was first pointed out to me by Mr. Blackad- 

 der of Glammiss, and Mr. De la Beche separately expressed 

 to me the same opinion. But it struck me as an objection to 

 this view, that rain-water would in that case be now in the act 

 of shaping out cylindrical hollows everywhere, both where chalk 

 comes to the surface and where it is overspread by gravel. 



But Mr. Strickland, in reply to this objection, has com- 

 municated to me in a letter, dated August 31, 1839, the fol- 

 lowing very interesting remarks. " During a I'esidence of 

 about seven years in the neighbourhood of Henley-on-Thames, 

 I frequently observed subsidences to take place in the gravel 

 above the chalk. They occurred on the top of a chalk hill 

 between 200 and 300 feet above the Thames, and far removed 

 from the action of any running water which might be sup- 

 posed to have undermined the gravel. The latter formed a 

 stratum from 10 to 20 feet thick above the chalk, and the 

 subsidences appeared to take place quite suddenly, leaving a 

 nearly circular cavity with upright sides from 3 to 6 feet 

 wide, and from 2 to 4 feet deep. As no mechanical action of 

 running water could possibly operate in these situations, it 

 appears to me that the true explanation of the phaenomenon 

 must be the corrosive action of acidulated water acting on 

 the surface of the chalk at the particular points to which it 

 may percolate through the incumbent mass of clay and gravel. 

 "We have a further evidence of this in the fact, that these sub- 

 sidences never occur, as far as I am aware, in those places 

 where the chalk is exposed to the day, the rain being there 

 absorbed equally over the whole surface, instead of being con- 

 ducted to particular points, as is the case where the clay and 

 gravel covers the chalk. If this view be correct, we may infer 

 that many, if not all, of those gravel-filled cavities so common 

 in the chalk, may have resulted from atmospheric agency." 



According then to the theory above-stated, we may attri- 

 bute the larger size of the upper extremity of each sand-pipe 

 to the longer time during which the rain-water has there 

 acted ; as the corroding operation proceeded from above 

 downwards, and the percolation for ages of acidulous waters 

 will account for the absence of shells, except as casts, in the 

 contents of the tubes. 



In those cases where the tube penetrates the overlying sand 

 and gravel for a certain distance, and then ends abruptly, and 

 is capped by perfectly undisturbed strata, which occasionally 

 consist of chalk rubble, we have only to suppose that the 

 upper portion of the deposit traversed by the tube has been 



