Every dock excavation, however, does not necessarily cut through the submerged forests, for channels in an estuary are constantly shifting, and many of our docks happen to be so placed as to coincide with comparatively modern silted-up channels. Thus at King’s Lynn they hit on an old and forgotten channel of the Ouse, and the bottom of the dock showed a layer of ancient shoes, mediaeval pottery, and such-like—interesting to the archaeologist, but not what we are now in search of. At Devonport also the recent dock extension coincided with a modern silted-up channel. In various other cases, however, the excavations have cut through a most curious alternation of deposits, though the details vary from place to place. The diagram (fig. 1) shows roughly what is found. We will suppose that the docks are placed, as is usually the case, on the salt marshes, but with their landward edge reaching the more solid rising ground, on which the warehouses, etc., are to be built. Beginning at or just above the level of ordinary high-water of spring tides, the first deposit to be cut through is commonly a bed (_A_) of estuarine silt or warp with remains of cockles, _Scrobicularia_, and salt-marsh vegetation. Mingled with these we find drifted wreckage, sunk boats, and miscellaneous rubbish, all belonging to the historic period. The deposits suggest no change of sea-level, and are merely the accumulated mud which has gradually blocked and silted up great part of our estuaries and harbours during the last 3500 years. [Illustration: Fig. 1.] This estuarine silt may continue downward to a level below mean tide, or perhaps even to low-water level; but if the sequence is complete we notice below it a sudden change to a black peaty soil (_B_), full of vegetable matter, showing sallows, alder, and hazel rooted in their position of growth. In this soil we may also find seams of shell-marl, or chara-marl, such as would form in shallow pools or channels in a freshwater marsh. This black peaty soil is obviously the same “submerged forest” that we have already examined on the foreshore at the mouth of the estuary; the only difference being that in the more exposed situation the waves of the sea have cleared away the overlying silt, thus laying bare the land surface beneath. In the dock excavations, therefore, the submerged forest can be seen in section and examined at leisure. The next deposit (_C_), lying beneath the submerged forest, is commonly another bed of estuarine silt, extending to a depth of several feet and carrying our observations well below the level of low-water. Then comes a second land-surface (_D_), perhaps with trees differing from those of the one above; or it may be a thick layer of marsh peat. More silt (_E_) follows; another submerged forest (_F_); then more estuarine deposits (_G_); and finally at the base of the channel, fully 50 feet below the level of high-water, we may find stools of oak (_H_) still rooted in the undisturbed rock below.