30 A. C. D. CROMMELIN, ESQ., D.SC., F.R.A.S., ON 
diameter of the particle is of the same order as the wave length 
of light it pees important, being equal to gravity when the 
Pas IS sptoy Inch and twenty times gravity for a diameter 
of =sgaa Inch, after which it appears to diminish again. There is 
a difticulty about this theory for explaining all the facts in that 
it would involve an almost constant acceleration of the tail 
matter through the whole length of the tail. (ii) That it is 
electrical repulsion from the sun acting on the charged particles 
emitted by the head. Mr. Eddington has pointed out that we 
might on this assumption explain the cessation of repulsion at 
a certain distance by gradual neutralisation of the charge. (a1) 
The third explanation, which has been put forward in a number 
of slightly varying forms in recent years, suggests that the sun 
and not the comet’s head is the originator of the greater part of 
the tail matter; the function of the head being the repulsion of 
this matter to form the envelopes and also rendering it Juminous ; 
the theory involves the large assumption that the discharge of 
these ions or electrons is unceasingly going on in all directions 
round the sun, for comets emit tails whatever their direction 
from the sun may be; in this they differ from the streams of 
matter forming the corona or producing magnetic storms and 
aurore on the earth; for these, as Mr. Maunder has shown, are 
ejected along definite stream-lines, and could only produce some 
momentary excitement in a comet’s tail, not the long-enduring 
phenomena with which we are familiar. 
To my mind both the telescopic and photographic, and I 
may add the spectrographic, study of comets seems to show 
that the head, coma, and tail form a single entity, and that the 
tail belongs to the head and is emitted by it, not by the sun. 
Halley’s comet itself may add to our knowledge of cometary 
physics, for at the return of 1835 it was the scene of very 
active changes; on October 10th Smyth noticed a curious 
brush of light issuing from the nucleus, resembling the 
luminous sector drawn by Hevelius in 1682. The next day it 
had developed into a lucid sector, with two rays spreading on 
either side of the nucleus across the direction of the tail. On 
October 12th Struve saw it attended by two delicately-shaped 
appendages of light, one preceding, the other following the 
nucleus. At other times, he says, it was surrounded by a 
semicircular veil, which extended back in a double train of 
light to a vast distance. Bessel tried to explain some of the 
changes of shape by supposing a rotation of the comet in five 
days, and Professor W. H. Pickering adopted the same explana- 
tion for the changes of shape of some recent comets. There 
