384 
ASTRONOMY: C. G. ABBOT 
tion of long period, having a range of from 3% to 5% during the period 
of the sunspot cycle. Second, a short interval irregular variation run- 
ning its course in intervals of a few days, weeks or months, and having a 
range often as great as 3%, sometimes as great as 7% and even in excep- 
tional cases of 10%. 
In the years 1911 and 1912, the reality of this supposed short irregular 
variabihty of the sun was tested by observing simultaneously at Mount 
Wilson and at Bassour, in Algeria, a station about 50 miles south of 
Algiers, situated at an elevation of about 3600 feet. The results of the 
two stations on the whole supported one another and indicated that the 
variations were due to causes outside the earth's atmosphere. 
In the year 1918, the Smithsonian Institution sent an expedition, at 
the cost of the Hodgkins Fund, to Calama, Chile, to obtain during a term 
of years, daily measurements as far as possible of the intensity of the 
solar radiation. The station is situated about 150 miles northeast of 
Antofagasta at an elevation of 7500 feet, in one of the most cloudless 
regions of the world, where rainfall is almost unknown. The expedition 
is in charge of Mr. A. F. Moore, Director, assisted by Mr. L. H. Abbot. 
The observers began actual observing on July 27, and in the six months 
next following had observed on about 70% of the days. 
During a part of this interval measurements were made on Mount 
Wilson by my colleague, Mr. Aldrich, and the computations have now 
reached such a stage of advancement that we are beginning to be able to 
tell that the two stations mutually support one another as to the short 
irregular periodicity of the sun. 
In order to make a proper comparison it is evidently not necessary that 
the days observed should be consecutive but only that satisfactory obser- 
vations should be obtained on each day of the comparison at both sta- 
tions. In order to exhibit the dependence in the clearest possible man- 
ner, we may indicate Calama values by abscissae or horizontal distances, 
and Mount Wilson values by ordinates or vertical distances. If the 
measurements were without error and there was no variation of the sun, 
evidently all results should fall upon the same point, whether observed 
at one station or the other. But if there was a true variation of the sun, 
the point would be stretched out into a Kne at 45 degrees to the axes. As 
all measurements are subject to error it is not to be hoped that the sev- 
eral points corresponding to the several days of observation|will all be 
found strictly upon such a line, but they ought to arrange them- 
selves about it in such a way that the line will give the best mean 
representation. 
