145 No. 3 opening started as an adit, partly on the seam and partly in the glacial till, but with the increasing dip of the coal the adit developed into a slope, as shown in Figure 20. Coal Seam. The Wilson seam varies from 4 feet to 18 feet in thickness where measured, but it is not certain that the full thickness is exposed where the smaller measurement was obtained. It is certain, however, that the seam is of varying thickness, and that the maximum figure given above does not represent the true average. The floor is a green, fine to medium-grained sandstone, con- taining scattered, well rounded pebbles up to an inch in diameter. Under the microscope the sand grains are seen to consist of rounded volcanic fragments, with quartz and decomposed felds- par grains, in a calcareous cement. The seam itself, of which a diagrammatic section is given in Figure 21, is best exposed in No. 1 opening, and was there sectioned and sampled. It is divided into two benches by a 5- inch band of soft, grey sandstone, occurring about 5 feet above the floor. At the base of the seam are usually found lenticular bands of granular, yellowish-grey, soft clay shale; but in some places the coal rests directly on the sandstone, or is separated from it by a layer of bone or by a bedded vein of calcite. In other places, bands of the same soft, plastic, clay shale occur in the lower part of the seam. The lower bench is dirty and of inferior quality, as shown by the analyses of the coal. The upper bench, about 12 feet thick, is composed almost wholly of coal in several layers, not all of the same quality. The upper part of the upper bench is crushed, dirty, and poor looking, and the analysis of this portion runs 37 per cent ash. The roof of the seam is a greenish, medium-grained sand- stone, more pebbly than the floor, and in places almost a con- glomerate. In this opening (No. 1) the seam shows no evidence of local thickening due to folding, the laminations being continuous and even; but a slight amount of thickening due to faulting occurs. The type of faulting is illustrated in Figure 22. The seam shows strong, continuous slickensides, parallel to the bedding, which are probably due to slight differential movements caused by the