28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. IO7 



sions, I give these data in table 8. The figures are the amplitudes, in 

 degrees Fahrenheit, of the average temperature effects of the 6.6456- 

 day cycle, in the months specified in the first and second columns of 

 table 8. 



Table 8. — Amplitudes of temperature' effects of 6.6456-day cycle, contrasting 

 months, remote stations 



Year Couple Washington St. Louis Helena 



1924 Jan. 12.5 12.0 90 

 Feb. 6.0 6.5 5-0 



1925 Oct. 3-5 47 9-5 

 Nov. 8.7 12.3 9.0 



1926 Mar. 3-5 8.5 4-0 

 Apr. 14.7 9.0 2.0 



1927 Oct. 4.0 6.3 4-7 

 Nov. 16.0 13.3 10.3 



1928 Feb. ii.o 7-5 7-0 

 Mar. S.o 50 30 



1928 Mar. 50 5-0 3-0 

 Apr. II.O 12.5 5.0 



1929 Oct. 9-7 4-7 7-5 

 Nov. 8.3 9-5 9-3 



1930 Jan. 1 8.0 140 70 

 Feb. 6.7 10.7 5-3 



1931 Jan. 6.7 100 7-7_ 

 Feb. 3.0 6.5 90 



1936 Feb. 4-0 8-5 9-5 



Mar. 13.3 10.7 lo.o 



1939 Jan. 6.7 73 3-0 



Feb. 12.3 1 1.3 17-3 



1943 Feb. 6.3 lo.o 12.0 



Mar. 13.3 lo.o 6.3 



Total No. couples 12 



All agree in trend 7 



Washington-Helena agree 7 



Washington- St. Louis agree 10 



St. Louis-Helena agree 9 



20. A TRIAL FORECAST A MONTH IN ADVANCE OF WASHINGTON 

 TEMPERATURE DEPARTURES WITH VERIFICATION 



Although no means have been found as yet of predicting shiftings 

 of phase, and changes of amplitude, of the temperature effects attend- 

 ing the 6.6456-day period, the fact that a month or more frequently 

 elapses without much change in these respects, encouraged me to make 



I 



