436 DR. C. CHREE: ANALYSIS OF RESULTS FROM THE KEW MAGNETOORAPHS 
aiiioiiut of agreement between tlie individual monthly ranges observed i]i D, H and I 
and those calculated from the values of a and h in Table XL. The probable errors 
calculated from the differences between the eleven observed and calculated values for 
four mouths representing different seasons of the year Avere as folloAvs :— 
D. 
I. 
H. 
Greatest.... 
0-'9 
0-23 
3 -27 
Le.ast .... 
0-5 
0-13 
2-27 
Mean .... 
0-65 
0-18 
2-87 
Tliese pioljable eii’ors are l)y no means large, consideiing that the range for airc 
individual month deiAends on onh^ five davs’ results, and that the non-cvclic effects in 
X. aJ ^ 
T and H ai'e so consideirible. Still the results tend to confirm the conclusion to Avhich 
other considerations point, that the departures from the mean shoAAm by magnetic 
phenomena on quiet days cannot be solely determined by the simultaneously existing 
sun-spot frequency, unless Ave are prepared to hold that either Wolfee’s sun-spot data 
or KeAv magnetic data are habitually affected Ipy considerable errors. 
§ 76. There are a variety of Avays of explaining the phenomena, some of AA liich may 
be indicated briefly. Tlie Avhole solar SA^stem may be under the action of some 
external agency, Avhose efiect on the sun is made manifest by the occurrence of 
sun-spots. As the influence simu.ltaneously existent Avould naturally axua’ from part 
to part of tlie solar system, this hypothesis AA oidd not l)e inconsistent Avith the aucaa' 
that the influence on the sun itself is measured hy the sun-spot frequencA^ The 
influence might he of the nature of a radiatioJi making the Earth's atmosphere offer a 
lessened resistance to tlie electiic currents due to solar action to AA’hich A-arious 
phvsicists ascribe the diuriial magnetic inecpiality. 
Or the sun itself might be the sole agent, if Ave suppose that sun-spot frequencv is 
a qualitative )-ather than a quantitatiA’e measure of its activitA'. The sun-spot might 
stand to the really active cause someAifliat as tlie smoke tfoni a locomotiA-e to the heat 
of its furnace. When smoke issues Ave knoAv the fire is alight, but aa'c cannot deduce 
the actual heat. 
riiere is a third obA'ious alternatiA^e, auz. , tliat the sun-spot frequencA^ is a direct 
measure of the sun’s contemporaneous activity, hut that the efiect at the earth 
depends appreciably on Ai hat lias been happening at the sun for some time preA'iouslA'. 
I his IS AA'liat AA'e might expect to hapjien if the sun AA'ere the origin of a radiation 
AAdiich took a considerable tune to part Aiutli all its ionising poAA'er, or aaIucIi traA'elled 
so much sloAA’er than light that emanations lea\ung different parts of the suii 
simultaneously reached the eartli at apprecialily different times. If hoA\ever this 
third alternatiA'e A\ere true, Ave should e.xpect magnetic phenomena to show a general 
tendency to lag hehind sun-spot frequency. Mr. Ellis has made investigations on 
