The consequence is that for a small-scale map with a contour line interval of 25 m, 
we must know the lateral tilt of each photograph with a mean square error of abt. 5 
centigrades. This precision can easily be obtained by means of a solar periscope or a 
horizon camera. Particularly the horizon camera, which can take photographs in all four 
directions like the one Col. Löfström of Finland ordered at and obtained from the Wild 
factory is very promising. Even if the horizon itself would be invisible and only clouds 
or a mountain horizon could be used, the system works very well. The mean Square error 
in the difference of flying height between successive exposures which can be obtained 
with a modern statoscope is abt. 1.5 m. This corresponds to an inclination of the base line 
of 1 centigrade and to the introduction of bz of 0.05 mm in a plotting machine. Since the 
A.P.R. will provide us with sufficient ground data in the flight line, it will be possible, 
when using these data, to impose directly the angular elements of absolute orientation on 
each correctly working plotting machine without the necessity of any aerial triangulation. 
The only conditions are the execution of absolute orientation of a few pairs in the area 
of each daily mission, to check the constant and systematic errors in the isobaric surface 
and to determine the scale. 
To determine the relative planimetric position and scale of each photograph there 
are two solutions with different precision. Knowing the position of isocentres and nadir 
points, it is possible to carry out a precise slotted templet lay-out which provides us with 
the planimetric position and scale of all pairs. It is possible however, to go one step 
further and to plot directly with the use of the elements of the exterior orientation on the 
scale on which we want to make a stereotemplet only those points we need for a templet 
lay-out. This requires only a few minutes for the relative orientation with x and by. We 
read also the heights of those points. Then the points derived from the stereotemplet are 
plotted on the map sheets 1 : 50,000. The heights are adjusted, which can be done either 
by computation or graphically. Now the plotting can start. 
This whole procedure can still be improved by the application of the I.T.C.-Jerie 
blockadjustment. We then read machine coordinates of four corner tiepoints in each 
section, transform them all to one system to which the leastsquare blockadjustment is 
applied. This will reduce the necessary number of groundcontrolpoints for planimetry as 
compared with the use of stereotemplet. 
When considering such a procedure, it is obvious that the question of a determination 
of planimetry by means of Hiran ete. cannot arise because these methods are in general 
far too expensive. Competition with these methods however, can only be succesful if we 
really introduce the camera orientation instruments which enable us to determine the 
elements of absolute orientation of each pair with a precision, satisfactory for this map 
scale. 
It may be mentioned that a higher precision than corresponding to a mean square 
value of the error in tip or tilt of 5 centigrades has no sense as long as we need contour 
line intervals of 25 m. 
The great advantages of a full use of this camera orientation equipment can be fully 
understood if we try to apply the method proposed above on for instance the “Photo- 
grammetric Mapping of the Brock Range" as described by Paul Blake in Photogram- 
metric Engineering 1959, p. 679-685. In particular about his entire ground survey for 
height control would have been unnecessary! 
d. The use of the small scale photographs. 
The problem remains in how far a scale of photography between 1 : 70,000 and 
1:100,000 will be sufficient from the point of view of identification of those details which 
must be represented on the map. It is obvious that there are certain terrains in which 
this will not be possible. I am convinced, however, that in the near future photogram-