tue of preserving a dry atmosphere. It was winter weather, when the stove would naturally be pretty constantly alight. How came the walls to be so damp? The answer seemed to be that the stove had not been constantly alight, but had been lighted only occasionally. This suggestion was borne out by a further examination of the rooms. In the kitchen there were practically no stores and hardly any arrangements even for simple bachelor cooking; the bedroom offered the same suggestion; the soap in the wash-stand was shriveled and cracked; there was no cast-off linen, and the shirts in the drawers, though clean, had the peculiar yellowish, faded appearance that linen acquires when long out of use. In short, the rooms had the appearance of not having been lived in at all, but only visited at intervals.
"Against this view, however, was the statement of the night porter that he had often seen a light in Jeffrey's sitting-room at one o'clock in the morning, with the apparent implication that it was then turned out. Now a light may be left in an empty room, but its extinction implies the presence of some person to extinguish it; unless some automatic device be adopted for putting it out at a given time. Such a device—the alarm movement of a clock, for instance, with a suitable attachment—is a simple enough matter, but my search of the rooms failed to discover anything of the kind. However, when looking over the drawers in the bedroom, I came upon a large box that had held a considerable quantity of hard
