mentions the passage of the star to the centre of the nebulosity, but not its passage behind the luminous nucleus.

From the foregoing facts we are forced to conclude that the matter of cometary tails and nebulosities, if gaseous, is of extreme tenuity; but it is perhaps so discrete-i.e. the particles are so far apart as not to occasion any perceptible occultation of a light seen through them. This was the opinion of M. Babinet, who, from the calculations above quoted respecting this extreme tenuity of cometary matter, has come to the conclusion that 'the substance of comets, therefore, is a kind of very divided matter, consisting of isolated particles, without mutual elastic reaction.' An observation made by Bessel helps to confirm this view, as Humboldt, when recording it, justly remarks. On September 29, 1835, Bessel saw, about 8" from the centre of the head of Halley's comet, a star of the tenth magnitude. At this moment its light was traversing a considerable portion of the nebulosity; but the luminous ray was not deflected out of its rectilinear course, as the illustrious astronomer satisfied himself, by measuring the distance between the occulted star and a star visible on the verge of, but outside, the nebulosity. So complete an absence of refracting power,' says Humboldt, 'scarcely allows us to suppose that the matter of comets is a gaseous fluid. Must we have recourse to the hypothesis of a gas infinitely rarefied, or are we to believe that comets consist, of independent molecules, the union of which constitutes cosmical clouds, devoid of power to act upon the luminous rays that pass through them, just as the clouds of our own atmosphere do not alter the zenith distances of the stars?'

It, therefore, still remains an open question whether the cometary nucleus, the luminous and brilliant central portion of a comet-that part of it, in short, which gives to the comet the appearance of a star-is opaque or transparent. In any

case, let us repeat, it is clear that we must refrain from generalising, for it would be absurd to identify, from this point of view, the faint nuclei, hardly visible in the telescope, of many of the smaller comets with those of comets which have shone like stars of the first magnitude, and have been luminous enough to appear in broad daylight and shine in the most brilliant regions of the heavens in the vicinity of the sun.

In support of the opacity of cometary nuclei various anciently recorded facts have been adduced; but these facts are either apocryphal or at least very doubtful. Thus the eclipse of the year B.C. 480, mentioned by Herodotus, and the eclipse mentioned by Dion Cassius, which took place in the year in which Augustus died, not admitting of explanation from the movement and interposition of the moon, were supposed to be due to the intervention of comets, a supposition altogether without foundation and very improbable. In the Cométographie of Pingré, under the date of 1184, we find the following: 'On the 1st of May, about the sixth hour of the day, a sign was visible in the sun: its lower portion was totally obscured. In the centre it was traversed by what appeared to be a beam! The rest of its disc was so pale that it impressed the same pallor upon the faces of those who looked upon it. Was this phenomenon the effect of a comet situated between the sun and ourselves? I do not know, but I consider it possible.' The total obscuration of the lower part of the sun would be, on this hypothesis, the partial eclipse produced by the opaque nucleus, and the beam traversing the disc the densest portion of the tail. Lastly, the pallor of the sun could be explained by the interposition of the vapours composing the nebulosity. But this is mere supposition.

Not an eclipse of the sun but an eclipse of the moon would appear to have been caused by the comet of 1454, according to Phranza, the protovestiare, or master of the wardrobe, of the

Turkish emperors. But it has been proved that the Latin version of the text is corrupt, and that Phranza has simply chronicled the fact of the simultaneous presence in the heavens of the comet and the full moon at the time when the latter was eclipsed.

SECTION III.

COLOUR OF COMETARY LIGHT.

Different colours of the heads and tails of comets-Examples of colour taken from the observations of the ancients: red, blood-red, and yellow comets-Difference of colour between the nucleus and the nebulosity-Blue comets-The diversity of colour exhibited by comets is doubtless connected with cometary physics, and with the temperature and chemical nature of cometary matter.

THE light of many comets has been sensibly coloured. The comet of B.C. 146 exhibited a reddish tinge, according to Seneca: A comet as large as the sun appeared. Its disc was at first red and like fire.'...

A little further on Seneca again observes: 'Comets are in great number, and of more than one kind; their dimensions are unequal, their colours are different; some are red, without lustre; others are white and shine with a pure liquid light. ...Some are blood-red, sinister presage of that which will soon be shed.' The ancients had, therefore, observed the difference of colour in the light of comets. And we shall mention a number of similar examples taken from the chronicles of the Middle Ages and from modern observers.

The comets of 662 and 1526 are cited by Arago as having been 'of a beautiful red;' and we have seen that Pliny in his classification speaks of comets whose mane is the colour of blood.' Such was the comet which appeared in November 1457; according to an ancient chronicle 'its coma or tail

resembled the colour of flame.' The horrible comet which, according to Comiers, appeared in 1508 was very red, representing human heads, dissevered limbs, instruments of war, &c.' The first comet of 1471 'was very large and of a reddish colour; it rose before the dawn.' In 1545 'a comet whose coma was the colour of blood burned for several days; it then became pale and soon disappeared.' Gemma, when speaking of the comet of 1556, thus expresses himself: Although Paul Fabricius has stated that the comet appeared small to him, I can affirm that, from the commencement of its apparition, I found it not less than Jupiter in size; the colour of the comet resembled that of Mars; its ruddy colour, however, degenerated to paleness.' This remark refers more especially to the nucleus, for, according to another eyewitness, 'the colour of the tail towards its extremity continued pale, livid, and similar to that of lead.' The opposite was the case with the comets of 1577 and 1618. Tycho relates of the first that its head was round, brilliant, and remarkable for a certain leaden whiteness, whilst the tail, turned towards the east, darted in a direction opposite the sun rays of a more ruddy colour. As regards the second, its tail appeared, says Arago, of a very bright colour.

The comet of 1769 had a slightly reddish nucleus, as also had that of 1811, observed by Sir William Herschel; but the nebulosity of the latter was of a bluish green, which caused Arago to conjecture that this appearance of colour might be due to the simple effect of contrast. It is clear, however, that the colour of the nucleus and that of the nebulosity were very sensibly different. A brilliant zone, narrow and semicircular, surrounding the head of the comet of 1811 on the side nearest to the sun, was of a decided yellow colour.

Amongst observations of earlier date we find mention made of comets which have shone with a golden yellow light.