We have been considering an extensive country more or less covered with gravel; such is England south of Yorkshire; both upon the east and west sides of the island. This country having no high mountainous part in the middle, so as to give it a considerable declivity towards the shores and rivers, the gravel has remained in many places, and in some parts of a considerable thickness. But in other parts of the island, where the declivity of the surface favours the transportation of gravel by the currents of water, there is less of the gravel to be found in the soil, and more of the fragments of stone not formed into gravel. Still, however, the same rule holds with regard to tracing the gravel from its source, and finding particular substances among the gravel of every region, in proportion to the quantity of country yielding that substance, and the vicinity to the place from whence it came. Here are principles established, for the judging of a country, in some respects, from a specimen of its gravel or travelled stones. In this manner, I think, I can undertake to tell from whence had come a specimen of gravel taken up any where, at least upon the east side of this island. Nor will this appear any way difficult, when it is considered, that, from Portland to Caithness or the Orkneys, there are at least ten different productions of hard stone in the solid land which are placed at proper distances, are perfectly distinguishable in the gravel which is formed of them, and with all of which I am well acquainted. Let us suppose the distance to be 600 miles, and this to be divided equally into 10 different regions of 60 miles each, it must be evident that we could not only tell the region, which is knowing within 60 miles of the place, but we could also tell the intermediate space, by seeing an equal mixture of the gravel of two contiguous regions; and this is knowing within 30 miles of the place. If this be allowed, it will not seem difficult to estimate an intermediate distance from the different proportions of the mixed gravel. This is supposing the different regions to be in all respects equal, which is far from being in reality the case; nevertheless, a person well acquainted with the different extent and various natures of those regions, may make allowances for the different known circumstances that must have influenced in those operations, although it is most probable there will be others which must be unknown, and for which he can make no allowance. The author of the Tableaux de la Suisse has entered very much into this view of things; he has given us some valuable observations in relation to this subject, which I would here beg leave to transcribe[7]. [Footnote 7: Discours sur l'Histoire Naturelle de la Suisse, p. 27.]