14 E'Loomis — Results derived from an examination of the 



appreciable. At the time of the observations shown in the 

 table on page 11, the average height of the barometer was less 

 than thirty inches, while the mean height throughout the year 

 at the same stations was nearly 30-10 inches. That a cycloidd 

 movement of the air should produce so little depression of the 

 barometer is due to three circumstances ; first, the small velocity 

 of the wind (averaging less than nine miles per hour) : second, 

 the small area of the cycloidal movement (generally about 500 

 miles in diameter), and third, the proximity of the stations to 

 the equator. According to Ferrel's formula any great barome- 

 tric depression requires a strong wind ; the cycloidal movement 

 must extend over a large area ; and the depression of the 

 barometer is proportioned to the sine of the latitude of the 

 place. In about half of the preceding cases there was an area 

 of low barometer prevailing in the northern pprtion of the 

 United States, and about as frequently, there was an area of 

 high barometer on the northern side of the station. These 

 different conditions appear to be about equally favorable to 

 heavy rain-falls. 



The distribution of these fifty-two cases by seasons was as 

 follows : 



Spring, 8 ; 



Summer. 14 : 



Autumn, 14 and 12 ; Winter, 4. 



We thus see 

 heating effect 



The diatribe 



i that great n 

 of the sun's 



ition of these 

 lows : 



dn-falls are m 

 rays is greatc 



it of vapor. 

 cases accord ii 



ost frequent when th 

 «t and the atmospher 



]g to the hours of th 



At the 7.35 a. 



m. obs. 19 cas 



tea. 



4.35 P. M. 25 " 



1 1.00 P. m. 8 " 



The intervals between the observations, which for convenience 

 I have called eight hours, are in fact unequal. Making, how- 

 ever, due allowance for inequality of intervals, it seems clear 

 that in the United States, south of latitiuh .i*> . In i\ \ rain fails 

 are less frequent from 4.35 P. M. to 11 p. m., than during the 

 remainder of the day. If we knew the precise time of beginning 

 and end of each rain-fall we might determine the precise hour 

 of maximum and of minimum frequency of heavy rain-falls. 



On Plate III are shown the curves of equal rain-fall for 

 June 10, 1873, being No. 29 of the table on page 11. The 

 outer curve shows the extreme limits of the rain-area ; the next 

 curve shows the limits of a rain-fall of one inch ; and the inner 

 curve shows the limits of a rain-fall of two inches. The arrows 

 show the direction of the wind as reported at 4.35 P. M., which 

 is presumed to have been about the time of the commencement 

 of the rain-fall. There was a center of low pressure (29-71) 



