(244) 
2 minute intervals. The micrometer uses .1 second of arc units, so that readings 
to one second are very reliable. 
A graduated glass plate is located in the image plane of the camera. The 
maximum error of this graduation is 1.5 micron. 
With this instrument, the distortion could be measured with a mean error 
of one micron. 
Since this is a visual method, we can expect a departure from the results 
commonly obtained by photographic techniques. Experiments yet have to be 
conducted on this point. However, we may safely assume that the results will 
be identical with both methods, for the color correction of the Wild Aviogon 
is so good that a difference in the positions of the image, as judged by the 
human eye and as actually determined on the emulsion respectively, can be 
ruled out. Photogrammetric practice shows that this reasoning is warranted. 
The results of the distortion measurements are shown on Fig. 9a and b. 
The mean distortion for four lenses is indicated in Fig. 9a, i. e. for ob 
jectives Nos. A, B. C. D. The departures from mean are indicated on Fig. 9b 
for lenses “A”, “B”, “D”. (The difference for lens C cannot be shown on the 
figure on account of its extreme smallness.) From this, we clearly see that the 
distortion curve is very smooth, warranting the conclusion that the manufac 
turing processes used are very precise and consistent. 
The variation of the distortion according to various diaphragm settings 
(for lens “D”), are shown on Fig. 10. The greatest departure from the norm is 
under 1.5 micron. Knowing that the distortion of a lens does not vary with the 
opening is extremely useful to the practical photogrammetrist. 
Distribution of illumination. 
The distribution of the illumination in the image plane was determined 
indirectly by measuring the size of the entrance pupil for various angular fields 
a according to a well-known method. This gives what is known as the 
“geometric” illumination pattern, i. e. the pattern which is independent of 
absorption variations due to the angle of incidence of the light rays on the 
lens and of reflexion losses on the emulsion, itself also a function of the angle 
of incidence. Studies are under way in order to determine the effect of these 
last named elements and, thus, not only to get a clear picture of the 
“geometric” distribution of the illumination or brightness, but also of the 
absolute distribution. The results of the “geometric” measurements follow: 
According to the well-known technique, a thin metal sheet, with a very 
small hole, is mounted in the image plane (i. e. to the focal plane frame of the 
camera) and is illuminated from behind (Fig. 11). The cross-section Q' a of the 
bundle of parallel rays coming from this “object” is photographically deter 
mined (perpendicular to the optical axis) by placing photographic paper on 
the camera filter. For a distant object having a brightness B, illumination in 
the picture plane can be computed from Q' a by elementary methods, as 
follows: 
B • Q'« • cos 4 a 
•p = 
^ £2 
B • Qo 
E 0 =