n 
Independent models are widely used, for data acquisition by 
analogue instruments and as computational units for strip- and 
blockadjustment. Also the fully analytical methods for aerial 
triangulation and comparators are increasingly applied. We wit- 
ness a second phase of development of analytical photogramme- 
try. 
Surprisingly enough, some essential practical questions have 
still not been thoroughly investigated, in spite of increased 
application of aerial triangulation. For instance, comparison 
of mono- and stereocomparators is left pending, although there 
is a tendency towards monocomparators, evidently for elemen- 
tary financial reasons. There are few studies only on the de- 
termination and the accuracy of perspective centres of analogue 
instruments. Some investigations treat the problem of point 
transfer. Amongst them interesting suggestions have been sub- 
mitted for omitting artificial point marking altogether. Such 
practical problems remain recommended for further research, 
including the possibilities offered by analytical plotters. 
There are also remarkably few new developments concerning 
computer programs for blockadjustment. New programs have been 
presented which essentially follow the principles and strate- 
gies well known from the previous period. Apart from the spe- 
cially important program development for selfcalibrating block- 
triangulation with additional parameters (see section 2.3) 
some modifications of established methods can be mentioned, 
Such as the joint 7 parameter solution for independent models, 
the combined adjustment of plate coordinates and of model coor- 
dinates in one block, or the "bundle-model" method by which 
bundles are treated by spatial similarity transformation. As 
to the joint adjustment of photogrammetric and terrestrical 
data improved algorithms have been developed for efficiently 
structuring normal equations. Applications have remained rather 
limited, however. 
There are many references about tendencies to treat blockad- 
justments by mini-computers. A number of such programs are in 
use, implying however considerable limitations. A general 
breakthrough of mini-computers in aerial triangulation seems 
not to have been achieved up to now. Rather, regular practical 
blockadjustments with large or medium capacity computers have 
been highly successful in many organisations. Nevertheless, 
implementation of blockprograms on small size computers remains 
an essential task. Success can be expected since the technolo- 
gical development of computers has removed previous limitations. 
The polynomial method of strip- and blockadjustment has been 
widely used during the past decade, notwithstanding some theo- 
retical objections. It continues to be applied because of the 
computer programs in operation, although it is gradually re- 
placed by more rigorous. and generally applicable methods of 
blocktriangulation. Thorough theoretical investigations into 
the accuracy of polynomial methods of block-adjustment are 
still lacking. However, empirical tests have clarified their 
accuracy performance in many respects and compared with other 
methods. Especially the high accuracy of polynomial strip- 
adjustment has been confirmed for the standard arrangement of 
control.