S46 Mr. Nixon on the Measuremeni of the 



Method of Observation. — The telescope being in a line with 

 and of the height of the cylinder, with the proper object-glass 

 of the latter at a short distance from that of the telescope, the 

 division of the pearl slip was got in a line with the horizontal 

 wire of the cylinder, before, and (as an additional precaution) 

 after the cylinder had been inverted; the reading of Fortin's 

 level following each observation. The cylinder being reversed, 

 the telescope was moved southwards until its object-glass stood 

 at the proper distance from the additional object-glass; when 

 the process of making the division of the slip level with the 

 wire previous and subsequent to inverting the cylinder, to- 

 gether with the simultaneous noting of Fortin's level, was re- 

 peated. (It is almost superfluous to state that the latter was 

 never read off until the bubble of the reference-level stood be- 

 tween its marks.) 



There were now sufficient data to determine the difference 

 of inclination of the axis of the cylinder resulting from the 

 latter being reversed within its Ys ; which difference it has 

 been demonstrated is equal to double the instrumental error. 



To prove that the great excentricity of the additional object- 

 glass did not vitiate the results, the two last measui'ements 

 (marked in the subjoined list with an asterisk) were made with 

 the line of collimation of the additional object-glass rendered 

 parallel to the axis of the cylinder, on reversing it within its 

 Ys, by altering the cross wires with the utmost care. 



List oftlie Hesults of the Measurements. 

 April 8th, 1831. Temp. 57°. Error = 13"-2 



57 13 -0 



- — 58 13 -7 



59 11 -6 



April 9th, 1831. 5Q 13 "2 



11th. 62 12-4* 



63 13 -3* 



The mean of the whole is 12"-8, but if we reject the fourth 

 measurement it will become 13"'l. 



By the Tenth Method. 



Theory. — Having demonstrated that the axis of the faulty 

 cylinder deviates, when reversed, from the opposite of its 

 previous direction by double the instrumental error, it is evi- 

 dent that the line of collimation of a telescope made parallel to 

 the axis of the cylinder before, and that of another telescope 

 made parallel to it after the reversing took place, would be 

 inclined to each other by twice the error. 



Apparatus. — In addition to the apparatus first employed, 



two 



