4826.] Astronomical Society, - 233 



this diameter Xy and the latitude on the planet, /, we have 

 X = — -'— 7> and the numerical result is x = 42'^'38, 



differing only 0''*04 from the measurement. Most probably it is 

 the slanting position of the axes of the eUipse, with regard to the 

 vertical circle, which causes this illusion. 



Lastly, there was terminated on the same evening, an 

 " Account of some Observations made with a Twenty-feet 

 Reflecting Telescope, by J. F. W. Herschel, Esq. Sec. RS. and 

 For. Sec. of this Society." This valuable communication is 

 divided into four sections. The first contains descriptions and 

 approximate places of 300 new double and triple stars. The 

 telescope with which the observations were made, is one of the 

 "front view" construction; aperture 18 inches, focal length 20 

 feet. It was constructed in the year 1820, under the joint 

 superintendence of Mr. Herschel and his venerable father. Its 

 light, with its full aperture, enables it to reach the faintest 

 nebulae of the third class, while, with an aperture of 10 or 12 

 inches, it serves to define double stars of the first class of an 

 average degree of closeness. Mr. Herschel briefly describes the 

 method of differences employed in sweeps of the heavens, the 

 modifications introduced into the process on account of Mr. 

 HerscheFs being deprived of the valuable assistance of his aunt. 

 Miss Caroline Herschel, his classification and characteristics of 

 the magnitudes of the stars from the 7th to the 20th inclusive, of 

 which none of the last three can be seen with the least illumina- 

 tion, but comprehend the stars seen or suspected in resolvable 

 nebulae. Mr. H. then presents an example of the method, in 

 which the business of " a sweep" is conducted, and of the 

 method of obtaining from it the approximate right ascensions 

 and polar distances of the objects which it comprises ; accompa- 

 nied by several instructive remarks. The table exhibits, in 

 eight columns, the approximate places of 321 new double and 

 triple stars, for Jan. 1, 1825, with their estimated angles of posi- 

 tion, distances, magnitudes, and other particulars. A great 

 many of the double stars tabulated in this paper, exhibit the 

 highly interesting and curious phsenomenon of contrasted 

 colours ; in combinations of white and blue or purple, yellow, 

 orange, or red, large stars, with blue or purple small ones: red 

 and white combinations also sometimes occur, but with less 

 frequency. In all these cases, the excess of rays belonging to 

 the less refrangible end of the spectrum falls to the share of the 

 large star, and those of the more refrangible portion to the small. 

 Another fact not less remarkable, and rendering highly probable 

 some other relation than that of mere juxta-position, is, that 

 though red single stars are common enough, no example of an 

 insulated blue, green, or purple one has yet been produced. 



The three remaining sections of this paper comprise observa- 



