GEOLOGY western half of it that allowed water to enter from without, appears to have existed between Kilsby and Crick, and through this gap a portion of the glacier water poured after reaching a certain level, and in it left a deposit of sand continuous with the Hillmorton beds. We thus have a simple explanation of the disastrous quicksand encountered near the south end of Kilsby tunnel when the latter was made, of the enormous volume of water pumped from it, and the remarkable way in which the trial holes failed to detect it. Southward of the Kilsby tunnel gap the water appears to have been disposed of partly to the westward, around the southern end of the Marl- stone outlier on which Welton stands, towards Braunston, and so into the Learn valley ; partly along the valley through which the L. & N. W. Railway runs from Watford to Weedon (see map), and thence along the channel previously referred to as extending from Heyford to Courteenhall, and on into the low-lying land constituting the valley of the Ouse. This old channel was, there is much reason for believing, a branch of the Ouse, and not the Nene as the nearest stream to it now is. The sand beds which we rely upon for identifying the course of flow are, between Wilton and Daventry, about 40 feet thick under 10 feet of gravel, and nearly the same at other places south of the Nene (see p. 23). Lower Glacial Deposits The sand beds (not including the overlying gravel) although formed by water action antecedently to the period of complete glaciation of the county, no doubt ultimately passed upwards into an earth and boulder- laden ice, and by this ice, valleys at a higher level, and otherwise less accessible to glacier flood water were gradually choked, and the general surface of the ground covered. The infilling Boulder Clays, or dirty gravels of certain pre-glacial valleys may be regarded as an imperfectly washed residue of this first ice sheet, and are therefore Lower Glacial deposits, but on the whole the previous presence of an extensive ice sheet can only be inferred from the modified Drift to be considered next. Mid-glacial Gravels. Inter-glacial Period The first glaciation was followed by a comparatively long interval of time during which a mild climate prevailed. The retreat of the ice, like its advance, was accompanied by great floods, implying rather rapid changes of climate. It would appear that subaerial melting produced superglacial floods sufficient in intensity to carry away all the finer argillaceous matter previously included in the ice, and ultimately left a well-washed ground moraine of coarser material spread over much of the county, but especially in the larger valleys, where the ice had been thickest and the consequent floods greatest and most prolonged. These Drift Gravels or Mid-glacial Gravels rest either upon the denuded surface of one of the Mesozoic rocks, or unconformably upon the earlier sand or gravel beds (as may be well seen at Hillmorton) ; they are usually well 25
