Royal Astronomical Society. 569 
of this problem was given by Gauss in his Supplementum Theorie 
Combinationis, &c. (Gottingen, 1828), and the method has been ap- 
plied by Bessel to the triangulation for the measure of the meridional 
degree in Prussia, and also to the computation of the extension of 
the French meridian through Spain, from Montjouy to Formentera. 
The triangulation which has been selected in the present case for 
an example of the method, includes ten stations (commencing with 
the base on Hounslow Heath), at which thirty-five independent 
angles were observed. For determining the corrections of those 
angles, nineteen equations of condition are furnished by the observa- 
tions, among which are instances of all the kinds which can occur 
in a trigonometrical survey. The final results differ extremely little 
from those given in the Survey, the greatest difference in the length 
of any side amounting only to about half a foot, and this in a 
distance of nearly eighteen miles. This close agreement must be 
attributed, however, to the smallness of the triangles, and the 
very great accuracy of the observations in this portion of the Ord- 
nance Survey. If the distances between the stations had been two 
or three times greater, the observations would probably have been 
less exact, and the differences between the results of the two me- 
thods of computation more considerable; but however this may be, 
it is only by following the method here explained that the whole of 
the precision which is attained by the observations is preserved in 
the results ; and for this reason the method should be adopted in all 
important surveys, particularly in those for determining the curva- 
ture of the earth. Besides giving a determinate result, and that re- 
sult the one which is most probably nearest the truth, it has the 
great advantage of superseding all arbitrary corrections, and ad- 
mitting only such as are rigorously deduced from the observations. 
The methods of deducing the most probable values of the angles 
from the observations, of assigning the weights, of forming the 
equations of condition, and all the steps of the process to the final 
determination of the corrections by which the equations are satisfied, 
are given at length. 
5. The President announced a communication that he had re- 
ceived from the Rey. Baden Powell, relative to an easy and convenient 
method of imitating the appearance of the corona, or glory, that sur- 
rounds the body of the moon, during the time of total darkness, in 
total eclipses of the sun; and also the appearance of the beads that 
occur not only in total eclipses, just prior to the time of total dark- 
ness, but likewise in annular solar eclipses. A sketch of the method 
was exhibited, which is merely this: a candle is placed in the focus 
of a lens fixed in a screen, with an aperture of about 3ths of an inch 
in diameter, on the opposite side of which screen is placed an opake 
circular disc, of equal (or even greater) diameter than the aperture, 
which may be placed at different distances, so as to produce an 
eclipse of any magnitude, as the spectator shifts his position. When 
it is central and total, there is a brilliant ring, or glory, even when 
it is so much nearer to the eye as to subtend a much greater angle 
than the aperture. Also, when there are any cusps, minute irre- 
Phil. Mag. 8. 3. No. 148. Suppl. Vol. 22. 2Q 
