My Dear Sir, —This evening, at half-past seven o’clock, I received notice from one of my servants of a luminous appearance in the sky, visible towards the S. W., which I immediately ran out to observe, and which, as it differed in some remarkable particulars from any phenomenon of the kind I have ever before observed or seen described, I think it not unlikely to prove interesting to the Royal Society. The evening was one of uncommon serenity and beauty: the moon, only thirty-eight hours after the full, having considerable south declination, was not yet risen. In consequence, the sun being already far enough below the horizon to leave only a faint glow of twilight in the west, the stars shone with unsubdued brilliancy, no cloud being visible in any quarter. Orion in particular was seen in all its splendour; and commencing below that constellation, and stretching obliquely westward and downwards, nearly, but not quite to the horizon, was seen the luminous appearance in question. Its general aspect was that of a perfectly straight, narrow band of considerably bright white cloud, thirty degrees in length, and about a degree and a quarter, or a degree and a half in breadth in the middle of its length; its brightness nearly uniform, except towards the ends, where it faded gradually, so that to define its exact termination at either end was difficult. However, by the best judgement I could form, it might be considered as terminating, to the eastward or following side, at, or a very little beyond, the stars
t, k
, λ Leporis, which stars (being of the fifth, or at most 5.4 magnitude) were pretty conspicuously visible; from which circumstance the degree of brightness of the ground of the sky in that region may be well estimated. Between these stars and
μ
Leporis, the luminous band then commenced, involving neither of them, but more nearly contiguous to
k
and
λ
than to
μ
. From thence its course was towards π Eridani, which star must have been covered by it, and was not seen; this judgement of its direction having been formed by noticing that it passed clearly above
γ
Eridani, and as clearly below and parallel to the direction of
δ, ε
Eridani, which two stars being dimmed by the vapours of tie horizon and the twilight, were so little conspicuous as perfectly to account for π not having been noticed. At the point of its passage between
γ
and
δ
it was still considerably bright, and as it terminated with somewhat more abruptness at a point beyond
ε
(then about 12° high) than at its upper extremity, I am rather disposed to consider this end as somewhat curtailed by the vapours Making no allowance, however, for this, and estimating its visible termination at a point on a celestial globe nearly opposite ζ Eridani (which star however was not noticed at the time), the length above assigned to the luminous band (30°) has been concluded by measurement on the globe.
XIII. The Bakerian Lecture.—Researches in spectrum-analysis in connexion with the spectrum of the sun.—No. III