THE MOON AND THE WEATHER, 



387 



amount of the Metonic cycle is not precisely nineteen years. But it is sub- 

 ject to a stronger objection founded on the principles which its supporters 

 themselves rely upon. The attraction of bodies in virtue of their gravitation, 

 increases in the same proportion as the square of the distance diminishes ; and 

 as we have already stated that the moon's distance from the earth is variable to 

 an extent not inconsiderable, it is evident, that her influence on the atmosphere 

 ought to be expected to depend much more on that variation of distance, than 

 on her relative position with respect to the sun. Now, although the cycle of 

 nineteen years corresponds with the changes of her relative position to the sun 

 as seen from the earth, yet it has no correspondence whatever with the varia- 

 tion of her distance ; and although, on each day of each succeeding period of 

 'nineteen years, she will have the same apparent position relatively to the sun, 

 she will not have the same distance from the earth, and, therefore, will not ex- 

 ert the same attraction on our atmosphere. Seeing, then, that the theory of the 

 moon's attraction does not lend its unqualified support to this assumed period 

 of nineteen years as a cycle of weather, let us see how far fact and ob- 

 servation countenance such a meteorological period. M. Arago (to whom 

 we are indebted for the most complete investigation of this question, and for 

 the collection of the labors of others upon it) has successfully shown that 

 observation affords no countenance or confirmation whatever to this hypothe- 

 sis. 



It has been said that the years 1701, 1720, 1739, and 1758, being cor- 

 responding years in successive intervals of nineteen years, show in the differ- 

 ent months the same characters of weather. Now to try this fact, it will be 

 necessary to adopt some distinct test of the characters of the seasons which has 

 nothing in it arbitrary, and about which two observers cannot differ. For this 

 purpose we shall take the highest and lowest temperature observed in each 

 ol the years, and the annual quantity of rain which fell in them respectively : 



Dates. 



1701 



1720. 



1739 , 



1758 , 



Temp. Max. 



...90-5 



...89-5 



...92-7 



...93-9 



Temp. Min. 



27-5... 



29-3... 



28-6... 



27-3 



Rain, inches. 



22-7 



18-3 



, 20-4 



Such is the kind of congruity on which the advocates for the Metonic cycle 

 rely. If any four years were taken indiscriminately at any given places, the 

 extremes of temperature and quantities of rain could scarcely be expected to 

 exhibit greater differences. M. Arago had extended the comparison to other 

 seasons separated by the same interval of nineteen years, or by multiples 

 of nineteen years. 



