582 Royal Astronomical Society. 



means of hanging levels. The meridian mark on the northern wall 

 of a. house, directly to the south of the transit instrument, was 

 never observed to deviate in position ; though the pillar, carrying 

 the northern mark, was liable to deviations when the sun was power- 

 ful or the frost severe. 



On the Advantages to be attained by a Revision and Rearrange- 

 ment of the Constellations, with especial reference to those of the 

 Southern Hemisphere, and on the Principles upon which such Re- 

 arrangement ought to be constructed. By Sir J. F. W. Herschel, 

 Bart., K.H., V.P.R.A.S., &c. &c. 



The stars must have been divided into groups, which, as also 

 single stars, would have received names long before they began to 

 be considered with reference to the seasons, or the sun's motion in 

 the ecliptic. The best-defined groups would be the first named, but 

 in the stars of the zodiac this principle would be afterwards modified, 

 for the convenience of dividing the circle into equal parts. In the 

 extra-zodiacal regions no principle seems to have prevailed, either 

 of subdivision or nomenclature ; and many of the figures of the con- 

 stellations scarcely resemble the forms intended, with the excep- 

 tion of a few, as Scorpio for example. Names, however, would be 

 imposed from other associations than mere form. The author pro- 

 poses to consider how far the present system of constellations is 

 adapted to the purposes of astronomy, and to examine by what mo- 

 difications it may be made more serviceable ; and also to inquire into 

 the circumstances which render a systematic revision of them de- 

 sirable, or even necessary, at the present advanced epoch. 



The use of constellations to the astronomer is to enable him to 

 refer to a particular star. For this purpose, a distribution on any 

 principle would serve ; yet even on this point the present system 

 often leads to confusion : for, 1st, the similarity of the names of se- 

 veral constellations, and the bestowing on a new constellation the 

 name of an old one, with an adjunct for the sake of distinction, 

 render it necessary in many cases, in order to avoid confusion, to 

 write the names at full length : thus, for example, we have (with 

 several others) Ursa major and minor, three Triangles, Pisces and 

 Piscis, Apis and Apus, Telescopium repeated three times, Quadrant, 

 Sextant, and Octant, &c. &c. 2ndly, Some constellations are so ex- 

 tensive that they exhaust three or more alphabets, and therefore it 

 is sometimes necessary, beside the letter of the star, to specify also 

 its right ascension and declination : for example, in Argo three stars 

 are marked a, and seven A. Again, such constellations extend over 

 so many hours of right ascension, that the name of the constellation 

 is of no use in finding, in the catalogues, one of the stars composing 

 it : thus, Argo occupies eight hours of right ascension. 3dly, The 

 imperfect and uncertain boundaries of the present constellations lead 

 to confounding the Greek letters of one constellation with those of 

 another contiguous to it. Moreover, the boundaries are not always 

 the same in different maps — a circumstance which alone is decisive 

 of the necessity of some system which might be more favourable to 

 a general understanding. 



