On the Seventh Section o/°Bessel's A&tronomicalObservations. 1 73 



With regard to this table, I observe that the greater differ- 

 ences which sometimes occur, as between the periods 1 1 and 

 13, 14 and 18, 43 and 45, 44 and 46, can hardly be ascrib- 

 ed to accidental imperfections of the observations, for they 

 commonly agree very well, as may be seen in the journals ; 

 these differences appear to arise from small partly temporary, 

 partly permanent changes of the line of collimation. Where 

 two periods are separated by a line, something in the instru- 

 ment has been altered either by accident or by design. From 

 these data the height of the pole is found without further cor- 

 rection for flexure, error of division and refraction, on the sup- 

 position of an equal value of each period. 



1 to 3 ... 



7—8 ... 



9—13 ... 



14 — 22 ... 



23 _ 24 ... 



25 — 26 ... 



28 — 31 ... 



32 — 33 ... 



34 — 38 ... 



39 — 41 ... 



42 — 46 ... 



Mean. 



54° 42' 50"-80 

 51 -81 

 51 -92 

 51 -32 

 51 -70 

 50-92 

 51 -29 

 52-18 

 51 -59 

 51 -55 

 51 -14 



weight = 2-0 periods 



2-0 



4-8 



8-9 



2-0 



2-0 



4-0 



2-0 



4-8 



2-7 



4-8 



54 42 51 "456 probable error = + 0"*0365. 



The same observations have given the correction of the 

 declinations of my tables of the two pole stars, of which I am 

 obliged here to give that of a. JJrsce Majoris, which I shall im- 

 mediately have occasion to use, viz. 



663 observations on the east side ... — 0"-452 



782 west side ... — 0"*219 



from which the declination for 1820 follows: 

 = 88° 20' 54"-157 and 54"-390. 



The reflected image of the same star has been observed in 

 64 passages, in each from 4 to 6 times, both above and below 

 the pole; 16 times in the eastern, and as often in the western 

 position of the instrument. It would have been impossible to 

 make so many observations of this kind in about 18 months, 

 were it not possible to see the star likewise in the day-time by 

 reflection from water. In the following table of these obser- 

 vations the first column contains the results of the circle cor- 

 rected by the refraction of my table, and the above determined 

 thermometrical correction; the second, the reduction to 1820; 

 the third, the place of the pole; the fourth, the distance of the 



reflected 



