NO. I PERIODS RELATED TO 273 MONTHS—ABBOT 3 
A single supposed periodic appearance from mean monthly data, 
with an amplitude four times as great, or %3 percent, would have some 
claim to veridity. Using monthly means from 1924 to 1952, 348 in 
number, all periods shorter than 45 months would have 8 or more 
repetitions. The criterion for probable veridity would be a percentage 
range exceeding 43+ WV8=%s percent of the solar constant for a 
45-month period. The requirement decreases as the square root of 
the number of repetitions increases. Thus for a period of 4% months, 
my shortest used in syntheses, it would be %2 percent. The %3 percent 
of the solar constant is 0.0008 calories. 
RANGES OF PERIODS IN SOLAR-CONSTANT MEASURES, WHICH ARE 
ALIQUOT PARTS OF 273 MONTHS 
In my paper “Periodic Solar Variation,’ * I list in table 2 64 
periods. They range in amplitude as follows: 
TABLE 2.—Ranges of periods discovered, in percentage of the solar constant 
ietiods ATITEIONEMS s xs) o:sfae ies 136.5-45 30-25 23-15 14-10 10-6 5-4.5 
BUEN Sele facicte,cuscé c #5, 0-c0.0 « 5 5 5 5 5 5 
Wmplitudes, % ............ 0.09-0.18 0.05-0.2I 0.03-0.08 0.02-0.I1I 0.06-0.13 0.02-0.08 
Periods in months.......... 4.4-3.7 3.5-2.9 2.7-2.4 2.1-1.7 1.7-1.3 <it.3 
yas 5 aise oc sav o's 5 5 5 5 5 9 
Amplitudes, omen eee: 0.03-0.06 0.03-0.04 0.02-0.08 0.02-0.04 0.02-0.04 >0.02 
All these 64 periods, so far as ranges of amplitude indicate, fall 
within the criterion for veridity based on accidental error of observa- 
tion. 
OTHER EVIDENCES OF VERIDITY 
Referring again to the last-cited paper, figures 1 and 2 therein show 
how the curves of observed periods which are aliquot parts of 273 
months stand out more and more plainly, as superriding periods whose 
lengths are also aliquot parts of 273 months are removed. Figure 3 
of that paper shows how strongly the smoothed curves resemble sine 
: curves when cleared of such superriders. 
Still more convincing indications of the veridity of a large number 
of the periods found in solar-constant measures will appear in follow- 
ing sections, as we note how periods in quite different phenomena are 
identical with them. 
8 Smithsonian Misc. Coll., vol. 128, No. 4, 1955. 
