XX. On the barometrical variation as affected by the moon's declination

Abstract T H E SE calculations were begun with the year 1815, and continued to 1832: the present paper continues them to 1841. The method used to bring out the averages may be thus described. Beginning as nearly as may be at the winter solstice, the days of a solar year are classed in what may be termed weeks of declination, mostly of seven days each, the middle day of each week corresponding,— 1, with the moon’s position coming north from the equator ; 2, with her position in full north declination ; 3, with her position returning to south over the equator; 4, with her full south declination.