238 
POPULAR SCIENCE REVIEW. 
some uncertainty about the disposition of the palaeozoic rocks 
under them. Admitting, however, the basin to be complete 
and isolated, that is no proof that the older Palaeozoic rocks 
prevail exclusively to the east; for the coal measures of the 
Somerset basin maintain their full development to the edge of 
the basin, and are there cut off by denudation, and not brought 
to an end by thinning out. They form part of a more extended 
mass, of which we have there one fragment, while on the west 
another portion exists in the Welsh basin, and another in the 
newly-discovered small basin of the Severn valley ; and there is 
no reason why on the east the same disposition should not pre- 
vail. 
The Severn Valley basin is entirely covered by the New Red 
Sandstone ; and as the Welsh basin is bounded on the east and 
the Bristol basin on the west by Mountain Limestone, the same 
objective argument might have been used in either case to show 
the impossibility of Coal Measures occurring in this intermediate 
area, or of their extending beyond the boundaries of either 
great basin. 
But the fact is, it is the very nature of the great line of dis- 
turbance to have minor rolls and flexures of the strata at, or 
nearly at, right angles to it, and so causing breaks in the coal- 
trough, which would otherwise flank it without interruption ; 
thus the Aix-la-Chapelle coal-field is separated by older rocks 
from that of Liege, which is again separated by a ridge of 
Mountain Limestone from that of Hainaut. So, in the case of 
south-western England, we have the separate basins of South 
Wales, Severn Valley, and Bristol — the extremes of the inter- 
vening belts of older rocks being two miles at Nameche and 
eighteen miles in Wales. These barriers are clearly only local ; 
and the division of the coal measures into separate basins ap- 
pears to be their ordinary condition along this great line of dis- 
turbance. The length of the two known portions of the axis 
included between Pembrokeshire and Frome, and between Calais 
and Westphalia, is 472 miles ; and in this distance we find eight 
separate and distinct coal-fields. The combined length of these 
eight coal-fields is about 350 miles, leaving about 122 miles 
occupied by intervening tracts of older rocks ; so that nearly 
three quarters of the whole length is occupied by coal strata. 
I consider that a structure which is constant above ground, so 
far as the axis of disturbance can be traced, is, in all probability, 
continued underground in connection with the range of the 
same line of disturbance ; and I see no reason why the coal- 
strata should not occupy as great a proportional length and 
breadth in the underground and unknown as in the above- 
ground and explored area. It is certain the basin-shaped form 
of the Somerset coal field is no reason why other coal basins, 
