60 
positions and introduce in the instrument the average values of the orientation elements 
thus obtained. 
The residual parallaxes are, as for Comm. B, in general lower than 0.02 mm and 
only in a few cases do they reach 0.03 mm. 
Yet, occasionally some large local parallaxes have been found, of which so far the 
source could not be traced; they are, for instance, more numerous in strip 2A (plates) 
than in strip 1A’ (film) according to the information received from the Centre in Milan. 
This subject should be thoroughly studied, because the troublesome influence of this fact 
on the precision of the aerial triangulation is evident. Above all, we believe it necessary 
to repeat the triangulation by using the original negatives. 
The azimuth of the first model was determined for all Centres by the Bureau Exé- 
cutif; this made it possible to execute the triangulations practically without changing 
the b.. 
For the scale transfer the Centres had to use the pricked nadir pionts; the pricked 
holes have a diameter of 0.1 mm, i.e. larger than that of the floating mark. This has 
caused difficulties which, in general, could be overcome by using next to the pricked 
points, other adjacent points. After each connection — according to the instructions — 
the coordinates of the pricked lateral points and of the control points have been read; 
altogether an average of 16 points for each couple in strip 1A’. 
The time needed for each connection has been as follows: 
IGM, Brussels 3h25m 
Centre in Milan 2h 
EIRA 2h35 m 
ITC, Delft 1h46 m (solar triangulation). 
The differences in time are mainly due to the method of orientation and to the in- 
sufficient training of the operator who, in certain cases, did not have enough experience 
in the use of the numerical method. It must also be borne in mind that a large part of 
this time, i.e. abt. 1/,-1 hour, has been used for measuring the coordinates of the control 
points; difficulties have also been encountered when identifying certain control points, 
which may be explained by the density of the points which have been determined in the 
two polygons. 
LARGE-SCALE TRIANGULATIONS (commission B) 
In the following table the distribution of the executed strips has been indicated. 
Their total number is 19. Of the 10 tests 7 have been executed at least twice; test IV, 4 
times. Only two strips have been carried out both ways. 
The number of pairs for each strip varies between 7 and 36; however, of the strip 
of test IV it has only been possible to use part 18-41 i.e. 23 pairs, because two intermedi- 
ate couples did not have a sufficient overlap. The length of the tested strips varies between 
5,5 and 16 km. 
The Centre in Vienna could utilize the original negatives, whereas the other Centres 
had diapositives at their disposal. 
The seale of the model varies according to the tests and the Centres; generally one 
has tried to have the largest possible scale allowed by the focal distance of the camera 
and the instrument concerned. 
For the focal distance of 10 em the proportion between the scale of the models and of 
the negatives varies between 4 and 5; for the focal distance of 17 cm between 1,7 and 2,5 
according to the restitution instrument. 
The scale of the model, which depends on the enlargement in the machine and the 
scale of the photographs, is between 1 : 2.000 and 1 : 5.000. 
For the relative orientation three Centres have adopted the numerical method on six 
points (classical location) after an approximated optical-mechanical orientation; the 
fourth Centre (EIRA) has, however, chosen the optical mechanical method (symmetrical 
method).