74 



H. JACOBT, 



On account of the small number of images measured on the plate, it 

 did not seem desirable to attempt the determination of many y's. In equat- 



ions (22), I 



therefore omitted y 2 and y 



In 



words. I 



djusted the measures on the assumption that the path of the pole 



straight 



The following equations of the form (26) were therefore 



puted, omitting of course the terms in y 2 and 

 ficients of y l and y,', expressed in degrees, by 



and multiplying the coef 



27 

 .09 

 .09 

 .27 

 .45 



Equations of the Form (26). 



45 Y! -h.55 Y -1-1.00 y' — 0''59=0 



.08 

 .47 

 .47 

 .08 

 .55 



.83 



0.08=0 



.32 -+-0 . 1 8=0 



.32 n-0.32=0 



.83 -+-0.13=0 



1 .00 -hO . 05=0 



Res'l. 

 0.05 



.02 

 .01 

 .07 

 .05 

 .01 



A" ' 



.27 

 .09 

 .09 

 .27 



.45 



1 



1.00y -*-.55y'o 



*06=0 



.83 

 .32 

 .32 

 .83 

 .00 



.08 -*-0 . 54=0 

 .47 -+-0 . 03=0 



.47 

 .08 

 .55 



0.14=0 

 0.59=0 

 . 90=0 



Res'l. 



.08 



.10 

 .13 

 .16 

 .07 

 .08 



i 



The solution of these equations furnished the values : 





To 



ft 



0?06, 



T 



i 



0'.'44 



0.0101 ±0.0018, 



To 



n 



T 



i 



0:15 



0.0102 



O'.'O 6 , 



0.0018 



The probable error of 



— 



from the direct 



solution, and 



ti 



0T066 when computed, according to the directions on 



p. 59 



60. 



w 



The following values of d x and d y resulted from equations (24) 



d x 



It 



0'.29 



0?155, 



^y 



ft 



0:57 



ft 



0.155. 



It is interesting to note that these equations of the form (26) give the 



following solution, if we omit y, and y/. 



T 







0'.'7 5 , 



To 



Oil 5. 



These numbers, as they should be, are in exact accord with the results 

 obtained in the former solution on the supposition that the telescope did 

 not move. 



Another very interesting point to notice is the direction of 

 1 defined by the y's. This direction is within 4° of being exa 

 the plate. This may perhaps be explained as an effect of ten 



rtical 



gravity upon the telescope and its mounting. There does not appear to have 



been 



Having thus obtained values of djc and d$, we 



ploy equations 



(16) for the computation of 1 

 simply the means of C„ . and I 



$23.-MaT. ctp. 74. 



■pfi^A and v-*-pd <a. Z, 



and X, 



V 



are 



»* 



hich have been given in a former Table 



34 





