The most difficult thing I encountered was their method of telling the time. During the day it was reckoned rationally enough by the passage of the Sun, which was never obscured by clouds and could always be seen. Every house had a small hole in the roof, at a fixed distance from the floor, and the daily track and varying shape of the spot of sunshine thus admitted gave names to the periods of the day. There seemed to be a settled superstition that no house was fortunate unless this spot of sunshine entered by the door in the morning. For this reason the principal door in nearly every house was built in the west, so that the rising Sun would cast its spot first on the porch outside and then gradually creep in through the door, across the floor, and up the opposite wall late in the afternoon. Of course there were daylight periods in the early morning and late afternoon when the Sun was too low to cast a spot, and these were known by terms which are best translated "before the clock" and "after the clock." No one dared to make a social call while the Sun was still outside the door, but friends were best welcome when the Sun was just entering it. Moreover, whoever slept until the Sun had entered the door was looked upon as an irredeemable sluggard. The track of the spot from the door-sill to the wall opposite was measured by linear distance from the centre or noon-position of the spot. As in different houses the apertures through which the clock-light was admitted were always the same distance from the floor, such expressions as "two feet before noon," or "a foot and a quarter after noon" (which I translate from the Kemish) always had a definite and exact meaning. The nearer the spot drew to noon the more exactly circular it became and the more slowly it moved. Therefore, very fine measurements were needed in the middle of the day, and an inch near noon represented nearly as much time as a foot in the morning or evening. But the daylight methods were simplicity itself compared with the night methods, which were calculated on an entirely different system, based on the combined movements of the two moons, neither of which agreed or coincided with the movement of the Sun in any close degree. I urged upon the doctor, as one of his earliest duties, the necessity of reforming their calendar and establishing a uniform method of denoting the time, to extend throughout the day and night. But on this point he failed to agree with me.