230 



KNOWLEDGE. 



[December 1, 1892. 



plate was exposed, but any motion of a bright mass of 

 matter in the tail in a direction away from the comet's 

 head during the exposure would tend to obliterate such 

 irregularities, and give a straighter appearance to the 

 streamers or jets. 



There can be no doubt about the existence of the 

 irregularities referred to. The southern (or right hand) 

 edge of the great streamer, photographed on the (ith of 

 April, is very distinctly notched or bent inwards at a 

 distance of about i° (that is, at a distance of about 3^^ 

 inches on the scale of our plate) fi-om the nucleus. A 

 little above the notch the narrow stream of luminous 

 matter which forms the southern edge of the great 

 streamer divides into two, or forks. This is clearly shown in 

 our plate at a distance of about 5 laches from the nucleus, 

 and it is still more clearly shown on the glass enlargement 

 from the original negative which Prof. Barnard has 

 kindly sent to me. There is also just traceable on the 

 plate, and distinctly visible on the glass enlargement, 

 a narrow stream on the inner or northern side of this 

 branching structure, and several faint structures which 

 appear to branch or fork somewhat after the manner of the 

 structure on the edge of the great streamer. 



There is also a distinct bend in the edge of the great 

 streamer of the 0th of April nearer the nucleus at the 

 jilace where it divides from the fainter group of rays 

 which form the southern half of the comet's tail. The 

 division between the great streamer which forms the 

 northern half of the tail and the group of rays which 

 forms the southern half is continued as a very narrow 

 black line towards the nucleus, and this line makes a very 

 obtuse angle with the rest of the southern edge of the 

 great streamer. On the northern side of this narrow dark 

 line is a bright forked structure, the branches of which trend 

 away from the nucleus, similarly to the branches of the 

 forked structure referred to above. 



The branching structure last described has, as shown on 

 Prof. Barnard's glass enlargement, a form which reminds one 

 of a solar prominence, and it gives a clue which may possibly 

 explain the irregularities in the edges of the streamers. If 

 during the rapid evolution of vapours in the neighbourhood 

 of the nucleus, gas is evolved in irregular or intermittent 

 quantities, and is projected outwai'ds, we should expect an 

 uprushing mass of gas in passing through a resisting 

 medium or atmosphere about the cometary nucleus to take 

 the tree-like prominence forms which uprushing masses of 

 gas take upon the sun, and such prominence forms, as they 

 are driven away from the sua and nucleus, would give rise 

 to irregularities in the edges of the streamers as well as to 

 the mottled appearance of the tail which we see in Prof. 

 Barnard's photographs. 



In a waterfall the foam on the surface tends to arrange 

 itself into parabola-like curves with the apex downwards, 

 because the stream is retarded at its edges, and a wave on 

 the surface or any other line across the stream moves more 

 rajjidly at the centre than at the edges of the stream, but 

 there are no such transverse markings in the mottling of 

 the tail of Swift's Comet. The mottlmg seems to be 

 disposed in irregular masses, with here and there an 

 appearance of branching in a direction away fi-om the 

 nucleus. Such irregularities are worthy of the closest 

 study, because deviations from general laws have always ' 

 formed stepping-stones to further discovery. 



Our first step is to make sure of facts. There can be 

 little doubt that there is a decided bend in the lower part 

 of the main stream photographed on the 7th of April, 

 about half-way between the nucleus and the curious knot or 

 branch in the main streamer, which Prof. Barnard describes 

 as a second comet ; possibly it may have Iieen a secondary 



or attendant comet, seen through the tail of the large one, 

 for such attendant comets have been observed before, though, 

 as far as I am aware, they have not appeared in the midst of 

 the tail of a large comet and, apparently, intimately asso- 

 ci.ited with one of its streamers. The fact which chiefly 

 weighs with me in concluding that the structure was pro- 

 bably a second comet is that the bright point which seems to 

 form the nucleus of the little comet is not stretched out into 

 a line by the motion of the camera in following the large 

 comet — as all the stars are. The nucleus of the little 

 comet was therefore a bright point which was moving with 

 the large comet. But the rest of the small cometary 

 structure has very much the appearance of being an irre- 

 gular branch from the main streamer of the large comet. 



It is much to be regretted that we have not other photo- 

 graphs taken an hour or two before and an hour or two 

 after Prof. Barnard's photograph on the 7th of April, and 

 it would also have greatly added to the interest of his work 

 if Prof. Barnard had taken photographs of the comet with 

 short exposures of a few minutes before and after the long 

 exposures. We should then, no doubt, have been able to 

 trace more structure in the head or nucleus of the comet, 

 which is now obliterated by the long exposure. But 

 Prof. Barnard did not expect to find the light of the 

 comet so actinic. Curious irregularities have been noticed 

 in the tails of other comets, notably in the tail of the great 

 comet of 1H82. There is a curious increase of brightness 

 and striation towards the end of its tail, which was drawn 

 by more than one observer and was photographed on 

 October 20th at the Cape Observatory. The original 

 negatives of this comet have been deposited by Dr. Gill in 

 the library of the Astronomical Society at Burlington 

 House, and are well worthy of examination. 



There is evidence which can hardly be doubted tending 

 to show that the tail of Donati's great comet, which 

 appeared in 1858, did not lie accurately in the plane of its 

 orbit. According to Prof. G. P. Bond, who collated the 

 drawings and observations of a great number of European 

 and American observers of this comet, and discussed them 

 in a very valuable monograph which was iJublished as 

 Vol. III. of the " Harvard Annals," the axis of the tail of 

 the comet was inclined at an angle of nearly -1° to the 

 plane of its orbit ; and during nearly the whole of its 

 apparition there was a contrast in the density of the two 

 branches of the tail, which remained unchanged when the 

 earth passed through the plane of the comet's orbit on 

 the 8th of September — a fact which points to the conclusion 

 that there was no revolution about an axis, of the nucleus, or 

 swarm of stones, from which the streams of matter forming 

 the tail issued. 



The curvature of the tail or tails (for Donati's comet 

 had three), as well as the gradual decrease of density of 

 the tail with increase of distance from the nucleus, points 

 to the conclusion that matter is continually streaming 

 away from the nucleus, and is driven away from the sun. 

 Matter driven backwards into space with very great 

 velocity, compared with the velocity of the nucleus in its 

 orbit, would evidently form a nearly straight tail in the 

 prolongation of the radius vector (or line joining the 

 nucleus with the sun), while matter driven baeliwards with 

 a velocity comparable with the velocity of the nucleus 

 would drop behind the radius vector, and form a tail 

 curving backwards in the plane of the orbit, the curvature 

 being more and more apparent the slower the velocity with 

 which the matter of the tail was driven away. Thus the 

 dift'erent curvatures of the tails of comets exhibiting more 

 than one tail is accounted for. 



Prof. Bredichin, whose theory has been much quoted, 

 is inclined to call in an unknown electrical repulsion, 



