y, descried the coalescence of the
cusps, and founded on the observation a valuable discussion of such
effects.[865] Taking account of certain features in the case left
unnoticed by Neison[866] and Proctor,[867] he inferred from them the
presence of a Cytherean atmosphere considerably less refractive than our
own, although possibly, in its lower strata, encumbered with dust or
haze.

Similar appearances are conspicuous during transits. But while the
Mercurian halo is characteristically seen on the sun, the "silver
thread" round the limb of Venus commonly shows on the part _off_ the
sun. There are, however, instances of each description in both cases.
Mr. Grant, in collecting the records of physical phenomena accompanying
the transits of 1761 and 1769, remarks that no one person saw both kinds
of annulus, and argues a dissimilarity in their respective modes of
production.[868] Such a dissimilarity probably exists, in the sense that
the inner section of the ring is illusory, the outer, a genuine result
of the bending of light in a gaseous envelope; but the distinction of
separate visibility has not been borne out by recent experience. Several
of the Australian observers during the transit of 1874 witnessed the
complete phenomenon. Mr. J. Macdonnell, at Eden, saw a "shadowy nebulous
ring" surround the whole disc when ingress was two-thirds accomplished;
Mr. Tornaghi, at Goulburn, perceived a halo, entire and unmistakable, at
half egress.[869] Similar observations were made at Sydney,[870] and
were renewed in 1882 by Lescarbault at Orgeres, by Metzger in Java, and
by Barnard at Vanderbilt University.[871]

Spectroscopic indications of aqueous vapour as present in the atmosphere
of Venus, were obtained in 1874 and 1882, by Tacchini and Ricco in
Italy, and by Young in New Jersey.[872] Janssen, however, who made a
special study of the point subsequently to the transit of 1882, found
them much less certain than he had anticipated;[873] and Vogel, by
repeated examinations, 1871-73, could detect only the very slightest
variations from the pattern of the solar spectrum. Some additions there
indeed seem to be in the thickening of a few water and oxygen-lines; but
so nearly evanescent as to induce the persuasion that most of the light
we receive from Venus has traversed only the tenuous upper portion of
its atmosphere.[874] It is reflected, at any rate, with comparatively
slight diminution. On the 26th and 27th of September, 1878,