6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 'J'] 



stations between 40° and 50° X. in figure 2, and for stations between 

 30° and 40° N. in figure 3. The continuous curve in the upper part of 

 the diagram shows the mean solar radiation, and those below it show 

 the mean atmospheric pressure preceding and following high values 

 of solar radiation. The broken curve in the upper part of the diagram 

 shows the mean solar radiation, and those below it show the mean pres- 

 sure preceding and following low values of solar radiation. 



The first thing to be noted in these curves is that the pressure fol- 

 lowing high values of solar radiation oscillates in opposition to the 

 pressure following low values of solar radiation. The high points 

 in the continuous curves correspond in general to low points in the 

 broken curves. This is particularly true of the northern stations 

 shown in figure 2, and extends to 12 days following the observations. 

 This fact clearly indicates a relation between solar radiation and 

 pressure in the United States and Canada. The only marked excep- 

 tion is in the case of the mean pressure at Salt Lake City, following 

 low values of solar radiation. 



The next thing to be noted is that the maxima and minima of 

 pressure appear first at stations in the central United States and 

 Rocky Alountain region, and occur later at stations in the eastern 

 United States, showing a progressive movement from west to east. 



The third thing of importance is that the first maximum of- pressure 

 occurs at western stations of the United States and Canada on the 

 same day as the observed high values of solar radiation. This is 

 true for Salt Lake City, Portland, Ore., and Winnipeg, Canada, 

 showing that there is a center of action in the general region between 

 Winnipeg and Portland, and the primary result follows the change 

 of solar radiation with surprising rapidity. This maximum is indi- 

 cated by the letter "A" in figures 2 and 3. The secondary maxima, 

 " B " and " C," probably follow the normal sequence of phenomena 

 on the sun after the occurrence of high solar values, but this cannot 

 be determined with certainty owing to the broken series of solar 

 observations. These observations are so interrupted as to make it 

 very difficult to obtain accurate means for succeeding days. Except 

 at the far western stations the relation of the minima of pressure 

 to low values of solar radiation was in general the same as that of 

 the maxima of pressure to high values of solar radiation. This evi- 

 dence is more striking, however, for Buenos Aires. Argentina, as 

 shown by figure 4. The figure gives a comparison of the mean solar 

 radiation for successive days following high observed values and 

 mean temperatures at Buenos Aires for the same days. The point 



