Dreyer — A Surirf/ of the S^nral Nebula Jlessier 33. 9 



to the method employed, the errors of the lines at the ends of the 

 part examined are assumed = 0. 



Line 



r 





mm 



6 



0-000 





0-0000 



7 



- 0-004 





- 0-0008 





+ 0-006 





4- •0019 



9 



+ 0-027 





_L -00^4 



1 



+ 0-037 





-i- '0074 



1 \J\J t t: 



1 i 



+ u uio 







12 



+ 0-036 





+ -0072 



13 



4- 0-038 





4- -0076 



14 



+ 0-045 





+ -0090 



15 



+ 0-028 





+ -0056 



16 



+ 0-021 





+ -0042 



17 



+ 0-015 





+ -0030 



18 



0-000 





0-0000 



The errors of the intermediate lines marking the single milli- 

 meter-intervals (five revolutions of the scre^) have not yet been 

 definitively determined ; but I have compared all the intervals by 

 means of the screw, treating each 5 mm. space as a separate scale, 

 and in this vray I have satisfied myself that a curve, expressing 

 graphically the errors given above, will correspond sufficiently well to 

 the errors of the intermediate lines. 



As already mentioned, I have referred all my measures to a small 

 star (of about the 12th magnitude), situated about 3' south, following 

 the characteristic lozenge of four stars at the centre of the nebula, so 

 that the measures give rectangular coordinates with, this star as 

 origin. As I had only an approximate idea as to the right ascension 

 and declination of the point in the heavens to which the axis of the 

 telescope had been directed, and could only guess at the direction of 

 the parallel, it became necessary to have recourse to known stars 

 occurring on the plate, in order to determine the '-'constants of the 

 plate," and thereby deduce the " standard coordinates" of the 

 various objects measured, i.e. rectangular coordinates referred to the 

 projections, on a plane tangent to the sphere, of the hour circle and 

 the great circle at right angles to it, passing through an origin 

 supposed to represent the centre of the plate. In order to pass from 

 one system of coordinates {x, y), resulting from the measures, and 

 referred to an arbitrarily chosen origin and axes only oriented ap- 

 proximately, to another system (^, 77) in a tangent plane to the sphere, 



