NO. 12 MOUNT ST. KATHERIXE STATION ABBOT 5 



Except for the large positive values found sporadically in January 

 1935, and April and May 1934, there seems to be no evidence of ap- 

 preciable yearly periodicity. For the differences are no larger than 

 would be expected as the result of unbalanced experimental errors. 

 During the months just excepted somewhat unsatisfactory conditions 

 prevailed at one or both stations. This general conclusion is highly 

 satisfactory. It means that on good days the differences of exposure 

 of instruments caused by unequal altitudes of the sun and the differ- 

 ences of procedure and magnitudes in allowing for atmospheric losses 

 at two independent stations in opposite hemispheres, separated by 

 nearly a third the circumference of the earth, produce no differential 

 periodicity in excess of two or three tenths of i percent of the solar 

 constant of radiation. Accordingly the yearly range of systematic 

 error for one station, being half as great as the combined ranges of 

 two stations, is surely negligible. Summer with its increased heat, 

 haziness, and humidity, opposed by winter with its greater cold, clear- 

 ness, and dryness, and besides these the observation of the sun at 

 different angles above the horizon at the constrasted stations, have 

 altogether failed to produce differences in the results which indicate 

 that systematic errors are certainly appreciable. 



Finally, the individual daily values at the two stations during the 

 five best months, June to October, 1934, have been tabulated in 

 table 3, and plotted in figure i. In computing the mean values and 

 deviations in the table, I have included a few fairly good values 

 marked " unsatisfactory " which were excluded in tables i and 2. 



A heavy line in the figure gives the best value of the march of the 

 solar constant of radiation.^ The independent results of the two sta- 

 tions are indicated by a lighter line for Montezuma and a dotted line 

 for St. Katherine. The close accord shown by these two remote and 

 contrasting stations cannot but encourage the belief that the observa- 

 tions of the variability of the sun hitherto reported from Montezuma 

 are very close to the truth. 



Every day but six within this interval of 152 days from June to 

 October 1934 is covered by good observations at one station or both. 

 The two stations obviously support each other in displaying in com- 

 mon many variations of the intensity of solar radiation. The most 

 conspicuous variation of long period shown has a periodicity of a 

 little more than 40 days, perhaps even 45 days, and has an amplitude 

 of about I of I percent. This effect is doubtless to be associated with 

 that solar periodicity of about 45 days to which I once drew attention 



^ The table and diagram were prepared independently and may differ slightly. 



