Chase] 532 



9. The solar-hourly^ ordinary and heavy rains have also an 

 inclination to maxima and minima, which appear to be mainly 

 dependent upon the thermal currents. 



10. The principal anomalies in the lunar-daily rainfall at 

 Philadelphia occur near the quadratures, and are attributable to 

 occasional heavy storms. For instance, the abnormal flexures 

 in the curve for 1825-44 (fig. 2), were produced hy ten storms, 

 of more than 2.5 inches each, on the 9th-14th and lTth-20th 

 days of the lunar month. It is well to notice that these flexures 

 are added to the curve, without obliterating the traces of baro- 

 metric nnediation. 



11. The middle day of the most rainy week in each of the 

 ten years' groupings at Philadelphia occurred after full moon, 

 or during that portion of the lunar month when the moon's 

 action is intensified by a falling barometer and increasing con 

 densation of atmospheric vapor. Schiaparelli likewise found a 

 rainy maximum in Northern Italy, about the time of the last 

 lunar quarter, and Mr. Dines observed that the rain which fell 

 on the 22nd day of the moon's age, in Surrey, " is in all cases 

 above the average, when a period of five years is taken," but I 

 am not aware that ixuy one has hitherto attempted to show the 

 dependence of such a maximum upon any obvious law. 



12. A careful examination of the local aerial currents resulting 

 from the moon's barometric and tidal actions, may perhaps help 



TABLE XL— 



