4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 134 
2. PERIODS RELATED TO 273 MONTHS FOUND IN WEATHER 
In an important paper published in 1947,* I discovered, both in solar 
variation and in Washington temperature departure from normal, a 
period of 6.6456 days. This solar periodic variation recurred with 
perfect regularity in its phases from 1923 to 1944. The Washington 
periodic variation in temperature departures, though frequently out 
of phase, yet for any single month of the 12 months of the year in 
the entire interval from 1910 to 1945 averaged exactly the same length 
as the sun’s variation, namely 6.6456 days. I did not then understand 
why phase changes occurred from time to time in Washington tem- 
perature departure. It is now quite clear to me, as will appear below. 
Not until 1953 did I perceive that this period of 6.6456 days, so 
strongly evidenced, belongs to the family of submultiples of 22% 
years. For 22% xX 365.2564 days=8309.5831 days. Dividing by 
6.6456, we have 1250.38, or within %o percent of 1250. So it is prob- 
ably an exact submultiple of 273 months, to an accuracy far beyond 
the precision of the data. 
In the years 1952 to 1955 I published eight papers on the control of 
weather by the family of periods related to 273 months.® 
Before particularizing these weather investigations, I wish to 
emphasize that they stand entirely on meteorological records. Meteor- 
ologists are apt to say that the variations of the sun are too small to 
influence terrestrial weather appreciably. But solar-variation measures 
play no part in my studies just cited. The periodicities in weather 
which relate to 273 months are to be found in weather records them- 
selves. No further reference to solar measures is required. Periodic 
variations in precipitation are large. They range from 5 to 25 percent 
of normal for the individual periodicities. In temperature departures 
they range up to 5° F. 
These large periodic changes of weather related to 273 months lie 
buried in the published records and may be demonstrated from them. 
No reference to solar variation is required to find them. 
4 The sun’s regular variation and its large effect on terrestrial temperatures, 
Smithsonian Misc. Coll., vol. 107, No. 4, 1947. 
5 Periodicities in the solar-constant measures, Smithsonian Misc. Coll., vol. 117, 
No. 10, 1952; Important interferences with normals in weather records, associated 
with sunspot frequency, ibid., No. 11, 1952; Solar variation and precipitation at 
Peoria, Illinois, ibid., No. 16, 1952; Solar variation and precipitation at Albany, 
N. Y., ibid., vol. 121, No. 5, 1953; Long-range effects of the sun’s variation on 
the temperature of Washington, D. C., ibid., vol. 122, No. 1, 1953; Solar varia- 
tion, a leading weather element, ibid., No. 4, 1953; Sixty-year weather forecasts, 
ibid., vol. 128, No. 3, 1955; Periodic solar variation, ibid., No. 4, 1955. 
