ABBOT: SOLAR RADIATION MEASUREMENTS 229 



observation itself, but the five days next following. He found 

 that the largest temperature effects occurred generally from the 

 third to the fifth day after the solar event. 



Correlation coefficients numerically as large as 0.50 are prac- 

 tically always certain evidence of strong dependence between 

 the variables. Clayton, therefore, very justly concluded that for 

 some regions, at least, the observed variations of the sun so 

 greatly influence temperatures that the effects are not masked 

 by terrestrial influences. Furthermore the delay of several days 

 between the solar cause and the terrestrial effect gave promise 

 for useful forecasting. 



A remarkable result is the opposing signs of correlation. An 

 increase of solar radiation is attended at some stations by posi- 

 tive and at others by negative temperatiure departures. Clay- 

 ton marked his stations on a world map and seemed to find that 

 while in the tropics and polar zones positive correlations pre- 

 vail, negative ones are found generally in both north and south 

 temperate zones. 



These early results of Clayton's seemed so interesting and 

 promising that the Smithsonian Institution appeared justified 

 in estabhshing a new solar radiation observing station in the 

 most cloudless available region in the world, in order to furnish 

 solar values regularly through the entire year. After a disap- 

 pointing expedition to North Carolina, a region chosen only be- 

 cause of war conditions, the station was located in July, 19 18, at 

 Calama, Chile. Two observers, Mr. A. F. Moore, Director, and 

 Mr. L. H. Abbot, Assistant, have occupied this desert station 

 continuously up to the present time. Their zeal and success 

 have been remarkable. The station has not quite satisfied our 

 hopes for cloudlessness but determinations of considerable weight 

 have been made on about 75 per cent of all days since July 27, 

 1918. 



In the meantime, Mr. Clayton and his colleagues in Argentina 

 have diligently continued their computations of the terrestrial 

 effects produced by solar variations. The results they reached, 

 up to June, 19 1 9, have just been published by the Smithsonian 

 Institution.^ They are indeed remarkable, though confined al- 



3 Misc. Coll. Vol. 71, No. 3. 



