256 COMPOUND NATURE OF GRAVEL AT OXFORD AND LONDON. 
Similar varieties of gravel, the one angular, and the other completely 
rolled (the latter being derived from the adjacent pebble beds of 
the plastic clay formation), occur in the valley of the Thames near 
London. These rounded pebbles, like those from Warwickshire, had 
apparently received their attrition from the long continued action of 
violently agitated waters, during more early revolutions that have af¬ 
fected our planet; whilst the imperfectly rolled fragments are referable 
to the diluvian waters, which drifted them only from the neighbour¬ 
ing hills to their present place ; and from the angular state of this and 
similar beds of diluvial gravel, we may infer that the inundation which 
produced them was of short duration. 
On the south-west side of the Evenlode, the valleys that intersect 
the Cotswold Hills in Gloucestershire are the effect of deep denuda¬ 
tions produced on the oolite limestone, by a volume of waters rushing 
downwards over strata composed of uniform and moderately yielding 
materials. Any irregular projections that might have existed on the 
original surface would cause these waters to descend with accelerated 
velocity over the intermediate depressions, and to excavate that series 
of sweeping combs and valleys that wind with the regular flexures of a 
meandrous river, and present masses of land alternately advancing 
and retiring with all the uniformity of the salient and re-entering 
angles that mark the course of running water. 
Striking examples of such valleys extending upwards far above 
the highest springs that take their rise in them, and forming vast 
diluvian furrows along the back of the inclined planes of the great 
oolite formation, may be seen in passing along the line of the Roman 
Fossway, from Bath to Stov r in the Wold: this line, being parallel to 
