The lakes in which the fish lived and died date back to late Triassic time. Much younger were the lakes that followed the continental ice sheets, and in many valley localities these younger water bodies have registered their brief span of geologic life. For they, too, were settling basins for clays, which are characterized by annual depositional bands like the growth rings in a tree. These clays may be examined best in the clay pits at any of the brickyards, particularly at South Hadley Falls, or beside the high river banks rising above the Connecticut flood plain just south of Hadley. The clays consist of alternating thin, dark, fine bands and thick, light, coarser ones. The coarser bands are sandy, and some of them have ripple-marks. The total number of pairs of beds is the number of years that glacial Lakes Springfield and Hadley inundated the valley, but it is not a simple matter to count them. Actually the lake bottom deposits are but a small fraction of the total volume of material brought to the lake. Lake shore deposits and deltas grew outward and buried the bottom deposits after a few hundred years had passed. Thus in the pits at South Hadley Falls, the clays rest upon glacial gravels, and a scant hundred layers intervene between them and the sands above. Shore encroachment is not encountered at Hadley, but the shallow depth of the present water table hinders deep exploration, and the Fort River has removed many of the upper beds. Long winters result in thick winter deposits, and heavy spring floods cause thick sand layers. If the winter of any year is long at Northampton, it is usually long everywhere in New England; and if the Connecticut has floods, most neighboring drainage systems have them, too. In this way, similar layers, or similar successions of layers, are formed at different places at the same time; and the lake deposits at White River Junction, Deerfield, Hadley, South Hadley Falls, Chicopee and Springfield may be matched and dated with respect to each other. The complete record in the valley shows that, in the vicinity of the Holyoke Range, the lake came into existence about 18,000 years ago. In each of the clay pits every set of lines exposed on the working faces represents a year, and the deposit as a whole is a calendar—in fact, it is also a thermograph for part of the region’s post-glacial history. Some bands at South Hadley Falls and along the Hadley river bank are highly distorted, and the distorted layers are planed off smooth. Spring sand covers the distorted beds. The disturbance can be attributed to ice which froze to the lake bottom and dragged the clay layers as it expanded and contracted with changes in temperature.