e = ss = 
  
Williamson Eagle III (Ross Xpress c — 136 mm) pom oM and 
13 X 13 cm 
Zeiss RMK P 10 (Topogon c = 100 mm) 18 X 18 cm | Rt An pde 
cal use. 
Wild RC5 (Aviotar c — 210 mm) 18 X 18 em 
Wild RC5 (two Aviogon c — 150 mm) 23 X 23 em 
In some cameras different filters were used. In other cases film and 
plates were used. 
At least three photographs were taken in each series with each ca- 
mera. The camera Eagle III was preliminarily tested in different tem- 
peratures, The camera was cooled in a freezing box and was transport- 
  
ed up in the tower as quickly as possible. The lowest temperature was . 
about —18? C. Photographs were taken in this temperature and then 
during the warming of the camera up to the normal temperature which 
was about + 8° C. The other cameras were tested only in this tempera- 
ture. The cameras and the lenses were regarded as too valuable for 
such experiments. The lens (Ross Xpress) of the Eagle III is of such 
a simple type, however, that the temperature differences may have no 
effect upon the geometrical quality. 
3.1 Determination of the radial distortion 
The radial distortion was determined in accordance with the grid 
method ref. [3] after measurements of image coordinates in the auto- 
graph A7 no. 310. The influence of the earth's curvature and the refrac- 
tion can be neglected since they together amount to the magnitude one 
micron, only. 
The results of the determinations of the radial distortion are de- 
monstrated in Diagrams 1—8. Each curve is an average of three or 
four independent determinations, The individual determinations are 
also demonstrated when the diviations are sufficient. In Diagrams 
1 and 2 the distortion curves of the Eagle III camera are demonstrated 
for normal and low temperature respectively. Evidently there is no 
significant difference between the curves. The deviations between the 
individual determinations are somewhat larger for the camera in low 
temperature, however. For the Aviotar and the Aviogon lenses the ra- 
dial distortion curves according to the manufacturer are also available 
and are demonstrated in Diagrams 4 and 5. Evidently the differences 
between the corresponding curves are comparatively small and are hard- 
ly significant with respect to the standard errors of the basic measure- 
ments. 
The standard error of the radial distortion amounts according to the 
grid method is about the half of the standard error of unit weight of 
the basic image coordinate measurements. This latter factor will be de- 
monstrated below, Table 5. Evidently the standard error of the radial 
distortion curve is of the magnitude 2—3 microns. 
19 
  
CEE EE 
  
  
ee 
E