284 
THE PHYSICAL GEOGRAPHY OF THE PAST. 
again beyond those of the Millstone Grit. This is the 
phenomenon of overlap, which has been so largely made use 
of in determining the original boundaries, or in other words 
the coast-line of any particular member of the Series. 
Up to the present time we have been dealing with 
sedimentation which took place in salt or brackish water; 
but, after the deposition of the masses of sandstones and 
shales of the Millstone Grit, the great Inland Seas, now so 
shallowed by coarse sandy shoals and mudbanks, became 
wholly or partially cut off from the ocean, and the water 
threw down finer sand and clay. We now begin to find traces 
of old land surfaces, which ever become more and more 
frequent. Subsidence was still going on, but slowly and 
intermittently, and the fine clay deposits of the alluvial flats 
were often for long periods together so near the surface of the 
water as to support a thick mass of vegetation, the remains of 
which we now have in our Coal Seams.* 
In order to find anything at all approaching the morasses 
which covered a great part of the surface of the British Isles, 
and of Northern Europe, in the Coal Period, we must look to 
the gloomy cypress swamps of the Mississippi. In the Great 
Dismal Swamp accumulate immense thicknesses of vegetable 
matter, the product of generation after generation of growing 
trees and semi-aquatic plants. These masses of peaty matter 
owe their wonderful freedom from any admixture of sand or 
silt, to the filtering agency of the marginal belt of reeds and 
brushwood, which effectually prevents any sediment from 
mixing with the vegetable mass. Doubtless some such cause 
as this produced the extraordinary purity of some of our coal 
seams. 
In picturing to ourselves the appearance of those huge, 
swampy flats of the Coal Period, which covered hundreds of 
thousands of square miles, we must not imagine a forest 
growth like that of the present day. The predominating 
forms, and those whose remains had most to do with coal¬ 
forming were Cryptogams, and consisted of trees related to our 
Lycopods or Club Mosses, and to our Equiseta or Horse 
Tails. These acquired proportions which were truly gigantic 
as compared with their lowly and degenerated modern repre- 
* These land deposits seem to have taken place in the deltas of 
large rivers, even at a very early period in Carboniferous times, for 
thin beds of coal are found in Northumberland in sandy and detrital 
deposits, which are actually contemporaneous with the Mountain 
Limestone of the Midlands; it is manifest, however, that these could 
have had but a local extension, and that the conditions favourable for 
the growth of vegetation over extensive areas were long subsequent. 
