630 MR. H. H. THOMAS ON THE MINERALOGICAL [Nov. 1902, 
An examination of the Pliocene sands from St. Erth and St. Agnes, 
in Cornwall, which contain local minerals’ such as andalusite, 
cyanite, cassiterite, and brown and blue tourmaline, shows that 
staurolite is extremely rare or absent; and in view of the 
essentially local origin of these sands, it should be present if derived 
from a local source. It seems, therefore, that we must look to 
some other area to supply us with staurolite. Evidence already 
given points to a southerly source, and the question arises what 
land-area existed at the time capable of supplying it. 
In Triassic times, the Armorican massif is generally believed 
to have been high land. That it formerly extended much farther 
north-westward is also probable. Indeed, most geologists hold the 
view that the present Armorican peninsula was connected with 
the South-west of England.* In the schists and gneisses of 
Britanny ° and of the North-west of France generally, rock-masses 
occur in which staurolite is fairly abundant. This area therefore, 
or its northerly extension in Triassic times, seems to be the probable 
source of the staurolite, and many of the other heavy minerals 
associated with it. 
It seems likely that the brown tourmaline is partly of local 
origin, and partly derived from a more distant area. This would 
account for the different degree of rounding noted in grains of 
identical size. The idea of a distant origin is possibly strengthened 
by the fact that Mr. E. B. Wethered* seems to have obtained 
detrital tourmaline from rocks of Middle Devonian age, which were 
deposited prior to the intrusion of the tourmaline-bearing granites 
of the West of England. 
The blue tourmaline was probably derived from the metamor- 
phosed sedimentary rocks in which it isso common. It is abundant 
in the Permian Breccias of the centre of Devon. Much of the 
rutile has probably had the same source. 
To summarize, therefore, I put forward the following points :— 
(i) That the distribution of material, and individual mineral 
species, indicates a main southerly sediment-bearing 
current, joined by a westerly minor current, which makes 
itself apparent about 20 miles north of the coast. 
(ii) That a mass of material was furnished by a highly meta- 
morphosed area, and that this differed widely from any 
now exposed in the South-west of England. 
(iii) That the probable source of much of the material was the 
Armorican massif of Triassic times, and the high land of 
Devon and Cornwall to the west, feeding the southerly and 
westerly currents respectively. 
In conclusion I wish to express my thanks to Mr. Ralph Morgan, 
of Exeter, for his kindness in supplying me with many samples of 
New Red Sands from Devon; to Prof. Sollas, for the kindly interest 
1 J. J. H. Teall, ‘Brit. Petrogr.’ 1888, pp. 386-87. 
° A. de Lapparent, ‘ Traité de Géologie’ 4th ed. (1900) p. 1004 (may). 
3 A. Lacroix, ‘ Minéralogie de la France’ vol. i (1893-95) p. 6. 
4 Quart. Journ. Geol. Soc. vol. xlviii (1892) p. 383. 
