192 



OBSERVATIONS ON NEBULA 



h. 1991, in my catalogue, corresponds to A. S. C. 2063, and is in the same field 

 witli the nebula. The places of the stars of h. 1991 may therefore be regarded 

 as tolerably accurate. The intervals between the other two nebulae, and the 

 respective stars to which they were referred, were much larger, the one occu- 

 pying l^ the other 12"" of sidereal time; and, of course, as great exactness was 

 not attainable by the means employed. The necessity of selecting a calm 

 evening, and of maintaining perfect stillness in the observatory during these 

 long intervals, on account of its height and the unsteadiness of the telescope, 

 yet, at the same time, of correcting the uncompensated clock which was em- 

 ployed, by coincident observations with the transit instrument, rendered these 

 comparisons less frequent than they otherwise would have been. The small 

 uncertainty which this circumstance introduces into the absolute right ascen- 

 sions of the stars is a constant error, and is of slight importance, since the 

 places of the standard stars, as assumed above, would alone give rise to uncer- 

 tainty of the same nature. 



The reduced differences in Article 32 were made such as they would have 

 been at the epoch 1830.0. The measured stars, previously arranged in the 

 form of a catalogue, by means of the three already mentioned, were projected 

 on paper, and compared with the original charts described in Article 7, by 

 pricking the stars through from one upon the other. 



34. To form some idea of the amount of necessary error involved in the pro- 

 cess of estimation described in Article 7, the differences of R. A. and Decl. be- 

 tween the measured and estimated places in h. 1991 were carefully measured 

 by a scale and vernier. They are as follows : 



star. 



Error in 



K.A. 



Decl. 



1 



2 

 8 

 9 

 18 

 21 

 22 

 24 

 25 

 28 



P.7 



0.4 



0.10 



0.00 



0.26 



0.17 



0.68 



0.42 



0.23 



0.03 



22".7 



14 .5 



0.9 



.0 



.0 



1 .4 

 3.8 



19 .5 

 .0 

 0.9 



It is to be noticed, that star (24) was excessively difficult of measure- 

 ment, not being seen on or very near the wire. The large errors apparently 



