$94 Scientific Intelligence. 
matical theory of the luminous or coloured circles which form 
around the Sun or Moon, and such a one as shall equally agree 
with the results of observations, and with the known proper- 
ties of light in the atmosphere.” The possible influence of the 
inflexion and the polarisation of light is to be considered. The 
memoirs to be sent in before the end of March 1824. 
6. Method of forming Three Haloes artificially round the Sun , 
or any luminous object — The following experiment, which illus- 
trates in a pleasing manner the actual formation of haloes, has 
been given by Dr Brewster : — Take a saturated solution of 
alum, and having spread a few drops of it over a plate of 
glass, it will rapidly crystallise in small flat octohedrons, scarcely 
visible to the eye. When this plate is held between the ob- 
server and the sun, or a candle, with the eye very close to the 
smooth side of the glass-plate, there will be seen three beautiful 
haloes of light, at different distances from the luminous body. 
The innermost halo , which is the whitest, is formed by the 
images refracted by a pair of faces of the octohedral crystals, 
not much inclined to each other. The second halo , which is 
more coloured, with the blue rays outwards, is formed by a pair of 
faces more inclined ; and the third halo , which is very large and 
highly coloured, is formed by a still more inclined pair of faces. 
Each separate crystal forms three images of the luminous 
body, placed at points 120° distant from each other, in all the three 
haloes ; and as the numerous small crystals have their refract- 
ing faces turned in every possible direction, the whole circum- 
ference of the haloes will be completely filled up. 
The same effects may be obtained with other crystals, and 
when they have the property of double refraction, each halo 
will be either doubled, when the double refraction is consi- 
derable, or rendered broader, and otherwise modified in point 
of .colour, when the double refraction is small. The effects 
may be curiously varied, by crystallising, upon the same plate 
of glass, crystals of a decided colour, by which means we 
should have white and coloured haloes succeeding each other. 
7. Theory of Haloes. — The first rational explanation of haloes 
was given by Mariotte, and was improved and extended by Dr 
Young. 1 st 9 Mariotte and Dr Young state it as a hypothesis, 
