where n nl are the angles made by the ray before and after refraction 
at the equivalent prism. They are related by Snells Law 
N is the refractive index of the prism. 
There are important consequences of these two equations: 
(a) The centre of symmetry varies in position with the angle 6 
of the incident rays, i.e. no single centre can be found for 
which distortions at all angles are symmetrical 
(b) The separation of the centre of symmetry and principal point 
is directly related to the residual tangential distortions. 
Table 2 gives some results for a camera which possess asymmetry. 
It can be seen that adjustment of the principal point of auto- 
collimation to the fiducial centre will lead to serious residual radial 
distortion, if other processes such as corrector plates are aligned on 
the centre. Neither adjustment to the principal point nor to the centre 
of symmetry affects the tangential distortion which is large. Unless 
measured experimentally, this may be quite unsuspected and may lead to 
considerable errors in mapping. The location of the principal point and 
centre of symmetry is therefore a vital stage in the calibration process 
as a control on tangential distortion, unless the latter js to be measured 
directly. 
In the air, negatives taken with an asymmetrical camera adjusted 
with centre of symmetry on the centre cross will be indistinguishable, 
except for tangential distortion, from those taken with a symmetrical 
camera, if the nadir point is in both cases recorded on the centre cross. 
In the former case the register glass of the camera will not be exactly 
horizontal, departing by perhaps 1-2 minutes of arc. This is so small 
that in the general case, it is swamped by the unknown tilt of the camera, 
although it may be of importance if accurate gyro-stabi lised mountings 
become available. 
5, DRAFT INTERNATIONAL STANDARD ON CALIBRATION 
The draft standard! was prepared by Commission I of the International 
Society of Photogrammetry and was adopted for trial by a Congress held in 
September 1952. Further consideration of the standard has suggested that 
it contains weaknesses, and that it is not sufficiently detailed or speci- 
fic in many cases to be of practical value to new comers to calibration or 
to surveyors confronted with a calibration certificate. 
The present author believes that the standard would be improved by a 
change of layout, to divide it into three main sections: Definitions, 
Methods of Test, Presentation of Results. 
In the first section, the following should be defined 
I. The fiducial centre. 
The principal point of auto-collimation. 
Distortion - radial and tangential. 
D Ceo n 
Standard distortion of lens type. 
  
rise 
  
  
   
  
  
  
  
  
  
  
  
  
MIT Tr tt NOPHOTO Rt 
TUM HEU Ute 
i Ó—— 
do 
Mme 
  
  
vu 
m 
— ——— 
  
  
s 
SE 
  
  
X 
ctp 
  
  
  
   
      
  
  
  
i 
1 
» 
$ 
t 
1 
  
  
  
  
  
  
  
  
   
  
  
  
Fargo