NO. 3 SOLAR RADIATION AND THE WEATHER CLAYTON 



45 



The results of the computations are given in table XVI for 

 periods running from about 3 to 53 days. For the periods longer 

 than 26 days the means cannot be considered as exact, owing to the 

 small number of periods embraced in the time interval co veiled by 

 the observations. The mean results are shown in the last columns 

 (i) and (2) of table XVI. Comparing these with the periods indi- 



Table XVI. — Solar Periods in Days Calculated From '' Phasograms" of Trial Periods 



cated by the plot in the lower part of figure 16, it is seen that the 

 periods are almost identical. Comparing with the periods shown in 

 the bottom part of table XII it might be that the periods 18. i, 23.7, 

 26.9, and 33.6 of table XVI were the same periods as the 18.5, 25.3, 

 26.8, and 32.4 in table XII, differing somewhat on account of diffi- 

 culties of determining the lengths or to the fact of actual changes 

 in the length of the periods. 



SUMMARY 



1. I think we may reasonably conclude from this research that 

 there is an intimate relation between the abnormal changes in tem- 

 perature in Buenos Aires and changes in solar radiation. 



2. Periodic or semi-periodic changes in solar radiation are reflected 

 in similar periodic changes in the temperature of Buenos Aires. 



3. The means of the solar radiation following the largest observed 

 values and of the coincident temperatures at Buenos Aires show the 

 chief maxima at intervals of about 10.5 to 11 days. If this period be 

 considered as one-third of a solar synodic rotation, it would place the 

 cause of the increased radiation in the solar latitude of the promi- 

 nences where the solar rotation period is about 31 to 35 days. If the 

 period be one-fourth of a solar rotation, it would place the outbreaks 

 in latitudes near the poles. 



