differences of the coordinates (the systematic contribution to the differences) 
then the root mean square values followed by maximum of the differences of the 
coordinates after eliminating the systematic part of the differences. Consider- 
ing the standard errors of unit weight for different adjustments in table 1 the 
differences in table 2 can be regarded as normal and they are so small that the 
results of the two computations of photographs at the same scale, A and B, can 
be considered equivalent. 
Standard error of unit weight of the adjustment 
The standard error of unit weight on which all investigations of accuracy are 
based, varies between 4.6 - 6.8 in microns and between 1.6 - 8.2 in centimeters. 
(Table 1). The figures vary a little for different computations in each series 
but these variations have practically no significance. One characteristic is 
that whereas the standard error of unit weight in microns usually decreases as 
the image scale becomes smaller, in this case the standard error increases and 
is at a minimum for the largest scale. 
Discrepancies between the given geodetic control points and photogrammetric 
  
field of points 
  
As was mentioned in the introduction the discrepancies between the control 
points, whose geodetic coordinates are known, and analytically determined pho- 
togrammetric coordinates can be computed after the adjustment from the coordi- 
nate residuals of image coordinates of each photograph. 
  
  
TABLE 3 
1:3500 i 1:6000 T 1:12 000 
A B I A B I A B 
Point | um um "Point | um um u Point um um 
z======d======d=====-h======$f====== ———— —_——_____-=--===—= 
Il Il 
EEESEESHSECCIS———EILLLLE === === a z===7z=z==5==25 
Mean |7.0 |6.9 1 e.2 (8.38 6.4 | 7.0 
A a ee nn a EE: se 
Mean T 1 
2.5 2.4 " 4.9 5.0 1 7.7 8.4 
  
  
  
  
  
  
  
  
  
e 9