W. A. E. Ussher — Pleistocene Geology of Cornwall. 261 



(which come to within a few feet of the surface) and bottom (forty- 

 feet from the surface) of a clay-lined basin. This is a most excep- 

 tional phenomenon, and seems to show the great erosive power of 

 the stream tin floods rushing into and deepening a depression, very 

 much in the manner in which giants' kettles are produced by the 

 pestle-like friction of fragments swirled round hollows by subglacial 

 streams. A somewhat analogous phenomenon is mentioned by Dr. 

 Boase, which, although not relating to stream tin, I give here 

 (T. K. Gr. S. Corn. vol. iii. p. 131) : "A person surveying the Channel 

 took his station on Wolf Kock, where he observed a cavity resembling 

 a brewer's copper, and containing rubbish at the bottom ; it was 

 covered by the sea nine hours out of twelve." 



The occurrence of an oblique clay seam in the tin ground at 

 Tregilsoe (8 B), separating accumulations of slightly different 

 characters, suggests the existence of bedding, true or false. The 

 exceptional occurrence of clay shelf (4 G and perhaps 5 D) is worthy 

 of note. 



The changeable character of the deposits in stream tin sections 

 precludes the absolute correlation of individual beds. Inland streams 

 cannot be expected to furnish such sections as their estuaries, yet it 

 is scarcely safe to identify tin ground, when not overlain by sedi- 

 ments (as 9 G) ; when composed of fine material under a thin cover- 

 ing of sediment with no indication of a land surface (as in 4 G, H, 

 I, J, and 9B); or where it rests on outcropping tin veins (as 7 B), 

 with the stanniferous gravels of Par (3), Pentuan (5 A, B, G, D), 

 Carnon, etc. (6 A, B, G) ; whilst in some sections stanniferous deposits 

 occur at different horizons, as 4 E (probably 5 D), IB. 



To synchronize the forest remains in the various sections is unsafe, 

 because in many valleys deposition seems to have gone on con- 

 tinuously, or to have been interrupted by such very brief periods of 

 peat accumulation or undergrowth, that their relics became entirely 

 mixed up and incorporated with the succeeding deposits, as in 4 B, 

 G, D, E, and 5 F; also 4 J and 7 A. 



The deposition of stream tin gravels evidently extended over a 

 much longer period than is represented by the tin ground ; for the 

 very irregular wear of the sides and bottoms of their channels, and 

 the existence of false shelf (4 Z), E) here and there, and of masses 

 of the surrounding rock, the apparent debris of fallen cap or false 

 shelves (4 A, 5 A, 6 G), can only be accounted for by powerful 

 streams carrying their detritus to lower levels, and occupying the 

 energies of their upper and more torrential reaches in eroding their 

 banks and beds into such irregular shapes as the unequal durability 

 of the rocks permitted. 



In like manner, the duration of the forest growth is not to be 

 measured by the forest beds overlying stream tin in Marazion Marsh, 

 Pentuan (5 A, G), etc., which can only be regarded as synchronous 

 with a comparatively short part of the period ; whilst the recurrence 

 of peat beds with arboreal remains at different horizons in the stream 

 tin sections (4 A, B, 5 E, 7 B, 9 A, 10) shows that even after the 

 forests fringing the coasts were submerged and buried with the 



