34 
metallic body in the neighbourhood of a magnet was made 
perfectly manifest. If then there is any magnetic relation 
whatever between the sun and comet, the latter will cer- 
tainly experience resistance. 
The question is thus resolved into one concerning the pro- 
bability that a comet would experience electric disturbance 
in approaching the sun. On this point we have the evidence 
now existing that there is a close magnetic relation between 
the sun and planets. If, as is generally believed, the sun- 
spot periods depend on the motions of the planets, a 
small fraction of the planetary energy must be expended. 
I find, indeed, that a very brief remark to this effect was 
given in the memoir of the original discoverers of the rela- 
tion, namely, Messrs. Warren de la Rue, Balfour Stewart, 
and B. Loewy. At p. 45 of their Researches on Solar 
Physics they add a small note to the following effect : “ It 
is, however, a possible enquiry whether these phenomena 
do not imply a certain loss of motion in the influencing 
planets.” As I conceive, no doubt can exist that periodic 
disturbances depending upon the motions of bodies must 
cause a certain dissipation of their energy, for if stationary 
the constant radiation of the sun could not produce any 
periodic changes, unless the sun were itself variable. Is 
there not then a reasonable probability that the light of 
the Aurora represents an almost infinitesimal fraction of 
the earth’s energy, and that in like manner the light of 
Encke’s comet represents a far larger fraction of its energy ? 
It is also worthy of notice that the tail of a comet is usually 
developed most largely at those parts of its orbit where the 
rate of approach or recess is most rapid, and where the 
electric disturbance would be correspondingly intense. 
I do not, of course, deny that the resisting medium may 
nevertheless exist, or may by other observations or experi- 
ments be made manifest. But I hold that so long as other 
physical causes can be pointed out which might produce 
