thing, it would indicate a prevalence to rain at perigee and at apogee, which is 

 in accordance with the observations of Schubler. 



" In spite, therefore," says M. Arago, " of the distance which separates Stutgard 

 from Viviers, and ia spite of the different methods pursued, and the difference 

 of instruments used, MM. Flaugergues and Schubler have arrived at analogous 

 results." It seems very difficult, therefore, at present, not to admit that the 

 moon exercises upon our atmosphere an action very small, it is true, but which 

 is nevertheless appreciable even with the instruments which meteorologists 

 commonly use. 



We have shown that the theory of the moon's attraction, applied to explain 

 atmospheric tides similar to those of the ocean, would lead to the conclusion 

 that the height of the barometer observed at noon, when the moon is in her 

 quarters, would be less than its height at noon at new and full moon. Obser- 

 vation, however, shows the very reverse as a matter of fact. The observation 

 of M. Flaugergues gives the mean height at the barometer quadratures 29-756, 

 and at new and full moon 29- 739 ; the height quadratures being in excess to 

 the amount of 0'017. This result has been further confirmed by the more recent 

 observations of M. Bouvard, at the Paris observatory : he has found the mean 

 height of the barometer at the quarters 29-786, and at new and full moon 

 29-759 ; the excess at the quarters being 0*027. 



Although, therefore, it cannot be denied that there exists a relation between 

 the barometric column and the lunar phases, yet it is not the relation which 

 the theory of atmospheric tides would indicate ; and by whatever physical in- 

 fluence the effect may be produced, it is certainly not the gravitation of the 

 moon affecting our atmosphere in a manner analogous to that by which she af- 

 fects the waters of the ocean. Any physical effects which depend on the rel- 

 ative positions of the sun and moon, as seen from the earth, would necessarily 

 occur in the same order throughout the year, when these two luminaries them- 

 selves have corresponding positions in the heavens on the same days of the 

 year. At a very early period in the history of astronomical discovery, it was 

 known that, after the lapse of nineteen years, the sun and moon assume on suc- 

 cessive days of the year relative positions. 



Thus, for example, if the moon were 90 behind the sun on a certain 

 day of a certain month in the year 1800, it would be 90 behind the sun on 

 the same day of the same month in the year 1819, and again in the year 1838, 

 and so on ; but on the same day of the same month in any intermediate year 

 it would have a different relative position with respect to the sun. This cycle 

 of nineteen years was known to the Greeks, and was called the Mctonic cycle, 

 from Melon, its reputed discoverer ; and it has always been used as a conve- 

 nient method of calculating eclipses and other phenomena depending on the 

 relative positions of the sun and moon. In a solar eclipse, the sun and moon 

 must occupy nearly the same position in the heavens ; and in a lunar eclipse, 

 nearly opposite positions : it is evident, therefore, that if an eclipse occur on 

 any day in any given year, an eclipse of the same kind must occur on the cor- 

 responding day in every nineteenth succeeding year. The tides, depending as 

 they do on the relative positions of the sun and moon, would be calculated 

 with facility by means of the same cycle ; and meteorologists who hold the 

 doctrine that atmospheric vicissitudes depend solely or chiefly upon the rela- 

 tive aspects of the sun and moon, have favored the doctrines, that there is a 

 general cycle of weather, the period of which corresponds with that which we 

 have noticed. Thus they hold, that the general changes of weather succeed 

 each other in the same, or almost the same order, throughout every successive 

 period of nineteen years. 



We shall not here object, on theoretical grounds, to the doctrine that the true 



