SUPPLEMENTARY REPORT ON METEOROLOGY. 131 
(1). Three halos round the sun, having Radii of 223°, 46°, 
and 90° nearly. The two smaller circles are generally coloured, 
the red being innermost. The circle of 90° is a rare appear- 
ance *, and is colourless. 
(2). A horizontal circle passing through the sun in which the 
parhelia or mock suns occur, usually at (or rather a little be- 
yond) the points where the halos intersect the horizontal (or 
parhelic) circle, and sometimes also in the point exactly opposite 
to the sun (anthelion). 
(3). Arcs of circles with reversed curvatures, touching the ha- 
los of 223° and 46° at their highest and lowest points. Mock 
suns sometimes appear also in the halos vertically above the 
true sun. 
278. Even more complex phznomena are occasionally re- 
corded, as that observed at Petersburg, 29th of June, 1790, by 
Lowitzt. For a history of such appearances we must refer to 
Brandes’s article in Gehler’s Dictionary, Kamtz’s Meteorology, 
the article Meteorology in the Encyclopedia Metropolitana (with 
an excellent plate), Fraunhofer’s paper on the subject {, and to 
Dr. Young’s invaluable catalogue of references§. In the mean- 
time we proceed to the theory of the fundamental appearances. 
279. In the 17th century Mariotte referred the halos of 224° 
to refraction through triangular prisms of ice. Minute icy 
spiculz being conceived to float through the air, or rather to de- 
scend slowly through it in all possible directions, a diffused re- 
fracted light must be seen, as is actually the case when the air 
is in this state; but those prisms which chance to be in a po- 
sition such as to refract the sun’s image to the eye in the posi- 
tion of minimum deviation, will (as in the rainbow, where the de- 
viation is a maximum) affect the eye more intensely on account 
of their parallelism and accumulation. The least deviated rays 
(the red) will fall upon the eye at the smallest angle with the 
sun, and consequently will form the smaller ring of the halo|. 
280. Huyghens proposed a different theory 4], which for a time 
superseded that of Mariotte. He attributed the phenomena 
to refraction through spherical and cylindrical particles of hail 
having opake nuclei of determinate magnitudes. But the ar- 
bitrary nature of his hypothesis is contradicted by the constancy 
* In fact, the observation of Hevelius was unique in this respect until the 
circle of 90° was recently witnessed by Erman (see Poggendorff’s Annalen, 
1840, xxxix. 255. note.). 
+ Figured by Young (Lectures, i., plate xxix. fig. 483) and by Kamtz. 
${ Schumacher’s Astron. Abhandlungen, iii. § Lectures, vol. ii. 
|| Mariotte, Zuvres, 1686, quoted by Young. 
§| Huyghens, Phil. Trans., 1670, and Opera feliqua, ii. (Young). 
K 2 
