9 
Assuming a block of 100 X 100 km?, this would consist of 8 strıps of abt. 16 pairs 
each. Applying the block adjustment according to the method of least squares, using the 
1.T.C.-Jerie analogue computer, we need for planimetry no more than 8 ground control 
points of which 1 on each of the 4 corners and 1 in the middle of each side. With normal 
classical aerial triangulation we can expect a mean square error in the absolute position 
of the control points of abt. 40 microns in the plane of the negative, which is 2.8 m in the 
terrain in a scale of photography of 1 : 70,000. In à map 1 : 50,000 this is no more than 
0.06 mm, an error which can be entirely neglected, even in a map 1:25,000. How is 
now the situation with the photographs 1:20,000? Here 2.8 m is 0.14 mm in the plane 
of the negative. Even in case we assume the mean square error in the transfer of points 
to be 0.04 mm in the negative of the 1 : 20,000 photographs, the total mean square error 
in the position of points in these negatives is not much more than 0.14 mm. The con- 
sequence of this is that even for a map of 1:25,000 the mean square error of the control 
points, determined in this way will be smaller than that of the draftsman. 
This fact makes the aerial triangulation of photographs 1 : 70,000 entirely acceptable 
for the determination of control points for a map 1 : 25,000. In case, however, we wish to 
use the photographs 1 : 20,000 for the production of maps of a larger scale, as mentioned 
before we have to determine a new system of control points in these photographs, very 
likely by means of aerial triangulation of the photographs 1 : 20,000 themselves. 
In case of the production of forest maps in mountainous terrain, as is a usual job 
in some countries, the value of the triangulation as described above will be, that we 
provide the photographs with an abundance of control points. In this way we can transfer 
the forest locations from the photographs 1 :20,000 to a base map without too much 
difficulties caused by relief displacements ete. Since also the heights are known, this can 
be done in an almost correct way by using such instruments as the Stereoflex, Stereotop 
and Santoni Stereomicrometer. 
This raises the question which precision for height can be obtained from small-scale 
photography. The experience gained so far with the I.T.C.-Jerie blockadjustment of wide- 
angle photographs without any use of statoscope or other camera orientation instruments 
is, that a mean square error in height of all control points can be obtained, which lies 
between 0.25 and 0.30% of the flying height using ground control points at both ends 
and in the middle of each strip (block of 8 strips of 20 models each). This means that 
with the assumed flying height of 6,200 m a mean square error in height between 1.5 and 
2.0 m can be obtained. It is obvious that in this case even contour lines with an interval 
of 10 m can be produced from the small-scale photographs in precise plotting machines. 
Stereoscopic transfer of control points from the scale 1 : 20,000 to 1 : 70,000 will have 
less influence on the precision of height than on that of planimetry. 
This precision of the height control points in the photographs 1 : 20,000 will make 
the transfer of forest details to a forest map even on a larger scale than 1 : 20,000 by 
means of a Stereotop, an easy job. 
In this way the combination of these two types of photography with an aerial 
triangulation of the small-scale photography can solve all mapping problems even of a 
forest service, which in general meets such more difficult mapping problems than an 
agricultural service. 
g. The use of the photographs 1 : 70,000 for a map 1 : 25,000. 
We arrive, however, at another very important conclusion from these considerations 
in countries, where a regular topographic map in a scale 1 : 25,000 is produced: the 
control points for the photographs 1 : 20,000 (needed of such areas in which not enough 
details can be distinguished in the small-scale photographs) can be obtained with suf- 
ficient precision from the aerial triangulation of photographs 1 : 70,000 on condition that 
the contour line interval in the map 1 : 25,000 is not smaller than 10 m.