BONNE Y : ON THE ORIGIN OF THE BRITISH TRIAS. 
7 
the majority form casts in a pale greyish or reddish sandstone, 
usually so hard as to be almost a quartzite. The pebbles are 
generally, but not always, about as well rolled as the others, 
and range with them in size. After what has been written, 
I need only say that at least the majority belong to the Upper 
Llandovery. A few other Silurian fossils have been found in 
more argillaceous rocks, and two or three bits of silicified wood, 
perhaps Permian. 
Passing to the crystalline rocks, we find that granites, 
gneisses, and mica-schists occur but rarely and are usually much 
decomposed. Felstones, however, to use a convenient field term, 
are rather common. After the descriptions which I have given 
of these, it will be enough to repeat that they have evidently 
travelled far, have very little resemblance to the so-called 
Uriconian felstones of the Midlands, or to any known to me 
in the Lake District, and none to those in North Wales. Many 
of them, however, correspond closely with the felstones so 
abundant in the Highlands and other parts of Scotland, though 
they often contain tourmaline, which is a rare mineral in that 
country. 
I pass on to the contents of the pebble bed in the southern 
Bunter. I described those at Budleigh Salterton in 1895, and 
may refer for particulars to the paper published in the Geological 
Magazine* 
Most of the rocks mentioned above also occur there, but not 
in the same proportion. The fossihferous hard grits or quartzites 
which, though more numerous, are not abundant, attracted 
the attention of palaeontologists many years ago. They were 
found to represent Lower Devonian, Bala, and Arenig grits 
(the last occurring very sparingly), and often have a closer 
lithological resemblance to corresponding deposits in Normandy 
than to those in Britain. The non-fossiliferous grits or quartzites 
are perhaps rather more frequently speckled with minute grains 
of felspar than in the Midlands, but the most marked difference 
is one of form. " Pebbles which are nearly prolate spheroids 
occur in both deposits, but that is the common shape in Stafford- 
shire, while those in Devonshire are usually oblate." Corre- 
*Volame for 1895, p. 75. 
