112 PRESENTATION OF THE GENERAL REPORT OF SUB-COMM Iv-l 
tion of the research work; I think they are of 
importance. We have the impression from our 
work that up to now the importance of the 
procedure in the autographs has been over- 
estimated. It seems to be of far greater impor- 
tance for us to concentrate upon earlier stages 
of the photogrammetric process, for example, 
the changes of dimension of the photographic 
material, the signalisation and we mean the size, 
form and contrast to the surrounding ground, 
the identification of the points in the model and 
several other questions, if you want to extend 
the accuracy of the measurements. 
Further, we have had the most excellent 
co-operation with Mr Härry and his colleagues 
in Switzerland, as a matter of fact, without his 
valuable support it would not have been possible 
to make any research at all. Nevertheless, we 
recommend that for the future the test field 
should be localized to the same country that is 
responsible for the error investigation and the 
statistical treatment of the observations. 
For the continuing researches within Com- 
mission IV-1, it is of course necessary to dispose 
a test field with a very accurate geodetic net- 
Work. A test field is a very expensive establish- 
ment and can serve many purposes in combina- 
tion. It is also an advantage, in our opinion, to 
give it a permanent character, a place where 
you can repeat tests continuously and under 
identical conditions. For the special purposes of 
Commission IV-1 it ought to combine the 
character of Oberriet and Monti di Revdira, 
thus including both flat and hilly terrain, and it 
should be as varied and diversified as possible. 
No doubt it would be valuable if a test could 
include larger material of measured points than 
Monti di Revoira. A statistical treatment of this 
material would probably give more valuable 
results. 
Let me recommend once more that we 
recommend every test in the future should be 
combined with a special investigation about the 
cause of gross errors and a study of the possi- 
bilities of eliminating them.