A HISTORY OF DEVONSHIRE 



boulders of flint breccia in the pits on the hills near Sidmouth, may be 

 easily accounted for by gradual subsidence over the pipes which have 

 been dissolved out of the. subjacent Chalk or Greensand. In a quarry 

 on Salcombe Hill one of these pipes was seen penetrating Greensand 

 with beds of chert to a depth of nearly 20 feet ; the narrow termination 

 of the pipe for 5I feet was filled with sand, above this to the surface 

 with clay with nodular chalk flints. In the same section are pipes of 

 sand which do not extend more than 3 or 4 feet in the Selbornian. 



The accumulation is confined to the plateaux ; its thickness varies 

 from a few to over 30 feet. Where the plateau has been broken by 

 fault, as in the valley on the west side of Stockland Hill, the stony clay 

 has been displaced with it.' The upheaval and denudation of the 

 Cretaceous sea bed, the deposition of argillaceous and loamy Tertiary 

 materials, whether Woolwich and Reading beds or London Clay, and 

 the subsequent solvent action of percolating water corroding the Chalk 

 remnants or calcareous Selbornian beds, and substituting clay or loam 

 derived by filtration from the overlying materials, explain the local 

 variations of the accumulation. 



There are some old marl pits on Brown Down (between Yarcombe 

 and Otterford) which seem to have been opened in dark drab-brown 

 clay without stones. 



Between Otterford and Church Staunton small quartz pebbles occur 

 in red clay, which was seen resting on reddish loamy sand with occa- 

 sional pebbles and fragments of chert. 



South of Stockland Hill, toward Dalwood Down, 5 to 6 feet of 

 coarse brown and grey quartzose sand with whitish spots and a few 

 chert stones occurs on the summit. 



A patch of gravel rests on the Selbornian north of Burnworthy 

 farm (near Staple Hill) ; its southern limits are within the Devon boun- 

 dary. A pit gave the following downward section : — 



Red-brown loam with irregular whitish streaks, with occasional quartz pebbles 

 throughout, 3 to 4 feet. 



Pale buff and red-brown sand with whitish mottling and an impersistent seam of 

 fine quartz gravel, 2 to 3 feet. 



Red-brown gravel, mostly of small irregular quartz pebbles, with angular frag- 

 ments of siliceous rock and chert, close together in a clayey matrix ; seams of 

 finer gravel and coarse sand are present, 5 feet exposed. 



Large pebbles of quartz, flint, and dark grey slaty rock occur in the 

 lower part of the deposit. 



This gravel was doubtfully ascribed to the Tertiary period in 1878." 

 The Haldon Gravels were pronounced to be Bagshot by Reid in 1898.' 

 He describes the plateau gravel on Great Haldon as ' mainly composed 

 of large boulders of " annealed " or toughened chalk-flint weathered to 

 a considerable depth. . . . Mixed with these is much Greensand chert 



* De la Beche's Report, p. 311 — faults near Wambrook. 

 ' Ussher, ^art. Joum. Geol. Soc. 1878, p. 452. 

 ' Reid, ibid. 1898, pp. 234-6. 



36 



