Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 134 
I then tabulated all the data. The mean march of them is shown in 
curve A, figure 5. A period of 45%4+3, equaling 15% years, seemed 
indicated, as shown in curve B. Subtracting its smoother ordinates 
(42+smoothA ), the final column of the mean table, A, results, and 
is plotted in curve C. Though rough, a period of 45% years, with an 
amplitude of 8 days, is found. The smoothed curve B for 15% years 
has the amplitude of 7 days. From curve C, one would expect a snow 
coverage at Tokyo averaging a week earlier for the years 1955-1970 
than the average which prevailed from 1930-1945. 
Casting the eye along any of the 45 lines of the table (fig. 5), one 
sees no well-marked tendency for a change in amplitude at any phase 
of the 45'4-year period from the seventeenth to the twentieth century. 
GLACIAL ADVANCES RELATED TO 273 MONTHS 
I recently received from the author, Herbert Griinhagen of Stadt- 
oldendorf, Germany, a paper entitled “Die Klimawellen der Fiszeit.” * 
He refers to a beautifully printed small book by W. Soergel, professor 
of geology and paleontology at the University of Freiburg, entitled 
“Die Vereisungskurve.” ® In this paper Soergel gives a curve to 
represent the fluctuation of latitudes of ice penetration in central 
Europe as glaciation advanced and receded from the direction of 
Sweden. 
Griinhagen smooths Soergel’s curve by using mean values for each 
successive 65,000 years. The curve thus treated he plots in 102 points, 
covering the interval from minus 565,000 to minus 30 years earlier 
than the year 1800 of our era. 
Griinhagen’s prior researches had disclosed to him periodic varia- 
tions related to various phenomena. He had noted that when such a 
period was found, the double of it was also apt to be a recognizable 
period in the phenomena. 
In these circumstances, a copy of my paper “Sixty-year Weather 
Forecasts” came to him, and some of the solar periods I used agreed 
closely with some periods he had found. He calls attention to families 
among my solar periods, in one of which the periods go in the order 
1, %, %, %; and another in which the order 1, %, %, 442 is found. 
It occurred to Griinhagen to see if longer periods than mine, in- 
creasing by powers of 2, might perhaps be found in the smoothed 
Soergel curve. He puts forward six such periods as follows: 15 years 
multiplying by 2", 2'*, 2%*, yielding 61,440, 122,880, and 245,760 
7 Separate from Niedersachsen, a periodical for home and culture, Heft 1, 1956. 
8 Published by Borntraeger Brothers, Berlin, 1937. 
