al 
  
With certain approximations we find for instance: 
  
S 
S. m pepe TE 
Qui (eris 
where (Qus; is the weight number of dx' and dy”, in the i:th circle. 
Table 4. Determination of corrections to the position of the principal point. 
  
Fagle III _ 
—— 
Camera LEE . R. C. 5 Aviotar 
  
| 
| | 
| 
| 
  
  
| 
| EET dx, dy', dx’, | dy', 
| Experiment mm mm mm | mm 
| 1 — 0.41 - 049 | + 0.01 —002 | 
| 2 | — 041 +047 + 0.01 004 | 
| 3 p e 040 |. s Eu F 0.48 ud + 0.08 |. +0® 1 
| Averages ........ J om Ol + 0.48 + 0.02 4 90f. — | 
| Stand. error. ...... | 9.5 8 25 8, 3.9s, 3.9 5 | 
assi gn ood. 270918 |: 0015 0.02 0.02 
In table 4 examples of the errors of the principal points in some 
cameras are demonstrated. The comparatively large values for the 
camera Eagle III are caused by the fact that a new glass plate had been 
inserted into the camera but without careful adjustment. In the camera 
RC5-Aviotar there is evidently no significant error in the position of 
the principal point. In the RMK P 10 the principal point could not be 
located with sufficient accuracy since one of the fiducial marks was 
not visible in the photographs. The corrections to the principal point 
in the Aviogon photographs were not determined because of the lacking 
flatness of the plates. 
In table 5 some results of the determinations of the standard 
errors of unit weight of the image coordinate measurements are de- 
monstrated. The standard errors were determined from the correspond- 
ing square sums [vv] in the usual way. 
Since the standard errors are determined mainly from combinations 
of five points only, the number of redundant observations is four. Con- 
sequently the standard error of the standard error can be expressed as 
or approximately 35 96 of the standard error itself. Since the measure- 
ments mainly were performed in the autograph A7 this instrument was 
25