11 
for surveying from 1920. (It must be mentioned here that the curves 
are extrapolated to some degree to make the principle ideas more clear.) 
The differences, which may be read from the two graphs, include 
errors from both the geodetic and the photogrammetric measurements. 
So they will give an information about the discrepancies to be expected, 
when geodetic measurements will be tied 1n to control-points, deter- 
mined photogrammetrically. 
In the graph, fig. 3, is also drawn a straight line, visualizing the 
accuracy-limit of the standard differences between distances measured 
in the field and those calculated from coordinates. It is now obvious, 
there is a definite distance for every negative scale of such a nature, that 
shorter distances are determined more precise by geodetic than by pho- 
togrammetric methods and longer distances on the contrary. That defi- 
nite distance is here called the critical distance. 1t is a function of the 
negative scale, of the special geodetic limit of accuracy, of the used pho- 
togrammetric instrument and method etc. In fig. 3. two such critical 
distances are drawn, C 7 for the negative scale of 1:7 000 and C 10 for 
the scale of 1: 10 000. 
To facilitate calculations concerning strip width, negative scale, fly- 
ing height, signalsize, distance between photo-control-points and bet- 
ween rows of them is constructed a nomogram (fig. 4.). 
Dy the aid of these three graphs it is possible to some extent to 
balance accuracy, negative scale and critical distance etc. in the most 
economical and least time-consuming way. As soon as the negative scale 
is determined, the obtainable accuracy also 1s determined, at given con- 
ditions concerning plotter and method etc., and also the critical distance 
for every limit of accuracy. 
However a special problem is illustrated in fig. 5. We presume first 
that the distance between the photogrammetrically determined points 
1 and 6 is the critical one. So the geodetic distance-angle-traverse can be 
tied in to these points within the geodetic and the photogrammetric 
limits of accuracy. Now we presume that the same distance 1—96 only is 
about one half of the critical one and that all the points 1—6 were 
determined photogrammetrically with an accuracy visualized by the 
large circles. We also presume that a geodetic distance-angle-traverse 
1—6 is measured with an accuracy visualized by the small circles. Now 
it ought to be possible under some conditions to adjust all the geodetic 
points, without altering any distance or angle, to all the photogrammet- 
ric points 1—6 and gain an accuracy within the geodetic and the photo- 
grammetric limits of accuracy, thus the same result as before. The 
accuracy must be a function of the geodetic and the photogrammetric 
accuracy, of the number of points and the distance (here 1—6). After 
the solving of this problem it must be much easier to plan the supple- 
mentary control with a foreseen accuracy and to combine the geodetic 
and the photogrammetric networks in the best way.