172 



ANNUAL REPORT SMITHSONIAN INSTITUTION, 1933 



ture, but the phases and the magnitudes of the periodicities change 

 according as sun spots are few or many. The same is true at other 

 places. I shall not detain you with many instances of the periodici- 

 ties in solar variation reflected in weather but I will ask you to note 



FiGORB 15. — Phases of weather periodicities changed by sun-spot prevalence. 



that the 11-month period in Bismarck temperatures holds practically 

 with unchanged phase from 1875 to 1925, provided we limit ourselves 

 in comparisons to intervals of equal sun-spot numbers. On the other 

 hand the phase of this weather wave reverses if we compare intervals 



