of the trajectory. The precision 3-axis mount allows for 
rotations in azimuth, elevation, and roll. Interchangeable 
lenses of 5, 7, and 10 inch, nominal focal lengths are fitted to 
the cameras. Film measurements are made to 3 fiducial crosses 
which are internally projected onto the film, and which define 
the film coordinate system. Timing data are indicated along one 
edge of the film by a 1000 cycle per second flashing bulb. Full 
second time is indicated in binary code. 
In normal operational use, orientation of the CZR-1 camera 
may be performed by using target boards, dials, or a combination 
of the two methods. Triangulation is performed using 2 through 
} camera combinations with the best geometry obtainable under 
ordinary field conditions. Systematic errors for this system 
are normally considered to be 20 to 40 seconds of arc using 
. target board orientations, and 180 to 240 seconds of are using 
dial orientations. In actual operation the errors of the sys- 
tem are practically doubled. On the other hand, the system has 
an ultimate accuracy capability of 10 seconds of arc or better. 
This value was established in an earlier test using star orien- 
tations and reconstruction of the directions of rays to the 
stars. 
Why this tremendous discrepancy factor of 50 between the 
theoretical capability, and the actual operation of the instru- 
ment? This becomes the story of the systematic errors in the 
system, and as stated earlier, the major source of systematic