96 B. V. Marsh on the Distinguishing Features of Comets. 
that a small nucleus may readily escape detection. : 
- Doubtless comets derive a portion of their brilliancy from theif 
reflection of the sun’s light, but there are strange discrepancies 
between the observed and calculated brightness which can only 
be explained by admitting that they are principally self-lumr 
nous. If we consider their light electric, we shall cease to be 
surprised at sudden and irregular variations. As regards the 
tail, we know that the total amount of matter contained in it 
must be extremely small—whence, as it occupies an immens 
space, it must be inconceivably rare—so rare that it can scarcely 
be supposed capable of reflecting enough of the sun’s light t0 
render itself visible even under the most favorable circumstances 
—so that its being self-luminous seems to me scarcely to admit 
of doubt. 
Recurring to the analogy between the comet’s tail and the 
auroral streamer we find the two identical in appearance, 
tion and in proportions, and not very unlike in form—and both 
are traversed by waves of light which in each case have thet 
origin in the base and travel outwards with great rapidity to the 
extremity.* The auroral streamer has been proved to be a 
_.* The following Seoeces from Commr. Gilliss’s article in the last number of 
pr ‘ : ; : 
this Journal, not only furnish an instance of this pulsation, but contain 
proof of pid movements and changes in i rts of the lumi 
ous sectors, streams composing the tail, déc., as seem ily to involve ele 
trical velocities, and afford sisikiaee aedicaien bo. movements of streams of ele 
“The whole appearance of this sector, at fos brillant and well defined, wndr 
of the light near was interrupted at. intervals by 
a a seed 
_. ™" 
sata :ciliiiietcciniadetsiciteabinain 
