June 6, 1889I 



NA TURE 



137 



un. There are nebulae of whJch the apparent diameter is greater 

 ihan that of the sun. If we assume such a nebula, with the 

 -un's apparent diameter, 1924", it would be transformed into 

 a parabolic chain which would require 20,000 years to pass 

 perihelion, its transversal dimension still being such that the 

 earth could traverse it in one or two days at the most. In this way, 

 then, Schiaparelli shows not only that external swarms can be 

 attracted from external space info our system, but that when so 

 drawn out their constituent particles must fake the form which 

 we know such swarms as that of November to possess. 



More recently this subject has been treated by Prof. H, A. 

 wton, and some results at which he arrived have been thus 

 rated by Prof. A. S. Herschel : — ^ 



"The evidence so strongly and distinctly shown in favour of 

 I he theory of the original motion of mast, if not of all, of our 

 recorded comets in spaces far external to the solar nebula, rests 

 upon the assumption that the comet-yielding matter of the 

 primitive nebula, if it existed, was confined, like that which 

 tormed the planets, to the neighbourhood of the ecliptic plane. 

 Ibis ground for the conclusion may admit of an exception that a 

 -iniilar distribution of the inclinations of the orbits to that which 

 1 place's hypothesis requires, would have been produced were 

 IS matter otherwise spread uniformly on a very distant sphere, 

 instead of in the distant portions of a disk or annulus. But the 

 plane of the planetary motions in the solar system, and the 

 analogy which they present to spiral and disk-like nebul?e in the 

 heavens, scarcely allows us to as-ume with reasonable probability 

 such a different disposition of the matter of the outer part of the 

 nebula from what the courses of the planets show us must have 

 been its original mode of distribution and of gradual contrac- 

 tion near the centre ; and with no evidence before us of the past 

 or present existence of a distant spherical envelope of nebular 

 matter inclosing the solar system, we may certainly prefer to 

 accept, with Prof. Newton, the much simpler conclusion to 

 which he is finally conducted by his well-executed labours, that, 

 with the exception of a few, perhaps, of the zodiacal comets, 

 and comets of the shortest periods, all the comets which have 

 been recorded are originally denizens of the interstellar spaces, 

 pursuing unknown orbits like the stars, and separated at least 

 and dissevered in their primitive astronomical relations from any 

 connection with the nebular matter which, in the process of 

 concentration supposed by the nebular hypothesis, formed the 

 sun, the planets, and the asteroids." 



Most Comets have once existed as External Nebula. 



We now come to an important question : we have noted the 

 extreme probability that the comets which now form part of the 

 solar system did not always belong to it, but that they were 

 drawn into it by the sun's attractive energy in its course ,throui);h 

 regions of space which contained the meteorites of which they 

 are composed. 



Then, instead of considering the case of a cloud of meteorites 

 at a great distance from the sun, we have to consider one moving 

 in an orhit round it ; and we must attempt to inquire into the 

 conditions of that cloud, both before and after it began to fall 

 under the sun's attraction. Into what conditions must we in- 

 quire in order to compare comets with external swarms ? They 

 are mainly these — 



(l) It is agreed that a comet is a swarm of meteorites, each 

 meteorite being on an average far from its neighbours. This ' 

 follows from an inquiry into the masses of comets. These are \ 

 very small, for they have never been known to appreciably | 

 disturb any of the planets, or even the satellites, by their ! 

 gravitational attraction. I 



In 1776, Jupiter and his satellites were entangled in a comet, 

 yet the satellites pursued their courses as if the comet had no : 

 existence. The comet itself, however, was thrown entirely out 

 of its course by the gravitational influence of the enormous mass \ 

 of Jupiter, and its time of revolution changed from a long period ' 

 to a short one of twenty years or so. ' 



Biela's comet, first seen in 1826, appeared as a double comet i 

 m 1845. The extreme lightness of the two portions was shown 

 by the fact that their mutual attraction was imperceptible, and 

 each performed its revolution independently of the other. 



The mass of a comet probably never exceeds 1/5000 of that of 

 our globe. The meteorites composing them must therefore be 

 very far a])art, seeing that this small mass is distributed through 

 spaces millions of miles in extent. 



' Monthly Notices, vol. xxxix. p. 279. 



(2) We next assume that a comet's luminosity is to a large 

 extent produced by collisions of meteorites. 



It is certain that one of the principal causes of the increase of 

 temperature of a comet during its approach to perihelion is the 

 increased number of collisions due to the greater tidal action 

 which takes place. Hence the larger the swarm, the greater the 

 difference between the attractions of the sun upon opposite sides 

 of it, and therefore the greater the disturbance set up. Also, 

 the shorter the perihelion distance, the greater fraction of it is- 

 the diameter of the swarm, and the greater therefore the 

 differential attraction. 



The initial movements of the individual members of the 

 swarm, and these superadded by tidal action, may be defined as 

 producing internal work. 



If ull the heat of a comet is produced by such internal work^ 

 it is clear that the temperature of the comet will depend (i)upoii 

 the velocity of orbital motion of the particles, (2) upon the size 

 of the swarm of which it is composed, and (3) upon its perihelion 

 distance. It will practically be independent of the velocity of 

 the comet in its orbit round the sun. 



If the luminosity be due entirely to internal collisions brought 

 about by the increase of solar action, then large comets, 

 or those best visible, should begin to be brilliant long before 

 smaller or more distant ones. But this does not seem to- 

 be so. Mr. Hind has pointed out that proximity to -the 

 earth is not so important a condition for visibility of a comet 

 in the daytime as close approach to the sun (Nature, 

 vol. X. p. 286); and M. Faye is the authority for the statement 

 that no comet has been seen beyond the orbit of Jupiter {ibid. 

 p. 228). "It is assuredly not on account of their smallness 

 that they thus escape our notice in regions where the most dis- 

 tant planets, Saturn, Uranus, and Neptune, shine so clearly with 

 the light which they borrow from the sun ; this is because the 

 rare and nebulous matter of comets reflects much less light than 

 the solid and compact surfaces of the planets of which we speak, 

 much less even than the smallest cloud of our atmosphere." 



On the latter part of this quotation it may be remarked that it 

 is not necessary to assume that comets at a great distance from 

 the sun, any more than nebul?e, are visible by means of reflected 

 light. 



Another possible cause is that of collisions with bodies external 

 to the comet. If external work is done on a comet by meteorites 

 in space — that is to say, if there are collisions with external bodies- 

 — the velocity of the comet must be considered in the first place, 

 and the equal or unequal distribution of the masses which it 

 encounters can be tested by the phenomena observed. 



The discussion of the recorded observations shows, indeed, 

 that in addition to the constantly increasing action which takes 

 place in a comet during its approach to perihelion passage, there 

 are at times temporary increases in temperature. 



We know that meteorites are scattered through space, and 

 here and there are gathered into swarms. It is only to be ex- 

 pected, therefore, that at times a comet will meet with such 

 swarms just as our own planet does, and in that case its tem- 

 perature would be increased by the collisions which would occur. 

 The increase of temperature would depend upon (i) the dimen- 

 sions and density of the swarm ; and (2) upon its velocity. The 

 larger and denser the swarm the more collisions would be likely 

 to occur, and the greater the velocity of the comet, the greater 

 the amount of kinetic energy available for transformation into 

 heat energy. 



If we assume that the increased brightness of comets as the 

 sun is approached depends to any extent on collisions with 

 meteorites external to the swarm, we must conclude that such 

 meteorites exist nearer together nearer the sun. This we should 

 expect. A test of this view would be great and irregular varia- 

 tions of intensity, as we know that the meteorites which the 

 comet is liable to meet are not equally distributed. Such a 

 variation was noticed in Sawerthal's comet in 1888, amounting 

 to three magnitudes (Nati;re, vol. xxxviii. p. 258) in two days. 

 Such variations, however, would be more likely to be observed 

 in the tails in consequence of the enormous dimensions of some of 

 them ; and indeed they have been observed from the time of 

 Kepler. 



The fact that these variations so str ^ngly resemble at times 

 auroral displays is an additional argument in favour of the 

 meteoric origin of the latter. 



Another result of a different order produced by a comet moving 

 through a meteoric plenum would be the gradual shortening of a 

 comet's periodic time, and this shortening should not be abso- 



