-8- 
be absolutely unacceptable. Should a modern cartographic camera and 
stable reproduction materials (Aluminium laminated paper and 9star 
or crohar bases) be available, this hypothesis would have been tried 
out. 
Therefore, rectification and mounting had to be made 
at l/l0,000, which also implied that the necessary triangulation 
control had to be obtained on the same scale. 
Aerial triangulation would be, certainly, the best 
way and, on account of the length of the strips, preferably by ae- 
ro-nivellement. 
Here a graphical or numerical approach could be chosen, 
4.1 - Graphical 
With a model scale of 1/10,000, the first stereo-pair 
is oriented and the carry over and control points are projected on 
the plotting table. For this a band of transparent and stable ma 
terial, 20 inches wide, must be used. The following models are then 
successively orientated, the scale being transferred on the common 
carry over points. Thus, all the triangulation and control points of 
each strip will be obtained graphically at 1/10,000, approximately. 
^he error propagation in the process, would, however, require a fas 
tidious graphical adjustment. Besides, strip sections would very of 
ten have to be bridged in order to determine the scale of the first 
model. It is thought that the expected results would not justify the 
use of a first class stereoplotter. 
4.2 - Numerical - ^he instrument coordinates obtained^ in the aerial 
triangulation of each strip could be quickly adjusted by means of a 
straight forward Helmert transformation, applied in sections of up 
to ten models. 
If electronic computing facilities were available a 
faster and better procedure would indeed be through analytical plani- 
metric block adjustment, 
4.3 - Radial Triangulation - Since the Autograph Wild A7? equipped 
with an EK-2 Coordinate Printer, was fully engaged in bridging, mainly 
for the l/50,000 map series, and the use of a plotting machine - A8 
or Kelsh - would take us back to the graphical method, implying the 
further disadvantage of working with independent models, radial 
triangulation by means of slotted templet had to be used. 
^he following are a few considerations on this step of 
the work.