THE MOON AND THE WEATHER. 



3S 5 



to perigee it rained 1,169 times; and during the seven days nearest apogee it 

 rained 1,096 times. Thus, cateris paribus, the nearer is the moon to the earth 

 the greater would be the chances for rain. 



From observations of Pilgrim at Vienna (which, however, are much less to 

 be depended on), it appears that the proportion of the prevalence of rain be- 

 tween perigee and apogee is that of nine to five an improbable result. 



From all that has been stated, it can scarcely be denied that there exists 

 some permanent and regular correspondence between the prevalence of rain 

 and the phases of the moon. What that exact correspondence is, remains for 

 more extended and accurate observations to inform us ; meanwhile, that rain 

 falls more frequently about four days before full moon, and less frequently about 

 four or five days before new moon than at other parts of the month, seems to 

 be a conclusion attended, to say the least with some degree of probability. 



Schubler also examined the question of a correspondence between the di- 

 rection of the wind and the lunar phases, and found that winds from the south 

 and southwest, became more and more frequent at those periods of the month 

 at which rain was also observed to increase, and that such winds were more 

 and more rare, while winds in the contrary direction occurred oftener toward 

 those epochs of the month when least rain was observed to prevail. These 

 results, it will be seen, are quite in accordance, and the question respecting 

 the mode of action by which the periods of rain are produced, would be re- < 

 duced to the question tf the physical action by which the moon affects the 

 currents of the atmosphere. 



The connexion of barometric indications with atmospheric phenomena is so 

 obvious, that the inquiry as to a correspondence between the lunar phases and 

 the variations of the barometer, could scarcely escape the attention of meteo- 

 rologists. M. Flaugergues accordingly made a series of observations at Viviers 

 (in the department of Ardeche), in France, which were continued from 1808 

 to 1828, a period of twenty years, on the heights of the barometer in relation 

 to the lunar phases : that the influence of the sun might be always the same, 

 the observations were made at noon, and the heights of the barometer were 

 reduced to what they would be at the temperature of melting ice. The fol- 

 lowing are the mean heights of the barometer, deduced from these observa- 

 tions : 



New moon 29*743 



First octant 29-761 



First quarter 29-740 



Second octant 29-716 



Full moon 29-736 



Third octant 29-751 



Last quarter 29-772 



Fourth octant 29-744 



Hence it appears that the height of the barometer is least about four days 

 before full moon, and greatest six or seven before new moon. Now these are 

 about the times at which the investigations of Schubler give the greatest and 

 least quantity of rain : and, since the fall of the barometer generally indicates 

 i a tendency to rain, these results are in accordance. Although it must be ad- 

 ' milted that the variation of the barometer is in this case so minute, that a sen- 

 sible effect could hardly be expected from it, still, though minute, it is quite 

 listinct and decided. 



M. Flaugergues also observed the mean height of the barometer when the 

 t noon was at her greatest and least distance from the earth, and found that at 

 perigee it was 29-713, and at apogee 29-753. 



So, far, therefore, as this small difference can be supposed to indicate any- 



35 



