(410) 
made and it is only necessary to measure short distances to the nearest reseau 
intersection. 
The computation necessary for this technique is quite considerable when 
undertaken with manually operated computing machines. This work has now 
been much simplified. The most significant development however, is in the use 
of electronic computing machinery to undertake these computations. The com- 
putations have been found particularly suitable for electronic computing which 
can undertake the stereogram solution in a matter of three minutes, and may 
also be able to undertake the adjustment of the stereograms in large blocks as 
separate operations. This is a very recent development which promises great 
things. It is obvious from fundamental considerations that the accuracy achieved 
by this technique must exceed that which could be obtained by plotting machin- 
ery since there are no sources of error from complicated mechanisms, nor ap- 
proximations involved in the matching of optical systems such as are necessary 
in the plotting machine technique. Further, the use of the electronic computor 
now makes the whole process extremely rapid. 
Consideration is being given to the possibility of equipping the observing 
instrument with an automatic booking device which could actually record the 
readings and perhaps punch the cards for feeding into the computing machine, 
and if further experiments develop as smoothly as it is anticipated, the whole 
process will become not only the most accurate photogrammetric process extant 
but also one of the speediest and it may well become economically possible to 
use it not only for large scale work but also for much smaller scales where it 
would recommend itself on account of its rapidity. Much of this development is 
due to the work of Mr. J. G. F. Lawrence at the Ordnance Survey. 
I would hasten to point out that the development of this technique could 
not in any way spell doom to plotting machines which still remain necessary for 
the delineation of the detail and the contours, where this is undertaken by 
photogrammetric means. 
A new plotting machine is being developed in this country by E. H. Thomp- 
son, now Professor of Photogrammetry at University College London. This 
instrument is conceived to be in every way as accurate as any existing machine 
but simpler in construction, less bulky, easier to erect and adjust, and may well 
be less expensive. 
6. The major requirements for research. 
On the continent of Europe a great deal of effort is being given to consid- 
cration of the adjustment of aerial triangulations observed in plotting machines, 
and it has been advanced that this is the major problem confronting photo- 
grammetrists. However, as Van der Weele has pointed out, one cannot design 
an adequate adjustment technique without first understanding all the sources 
of error. Many of these arise in the plotting instruments themselves so that it is 
likely that there would be a different adjustment technique for every different 
type of observing instrument used. 
I am strongly of the opinion that the most pressing need for research is not 
to try to adapt our photogrammetry to any existing plotting machines, but to 
concentrate on understanding and reducing the many different sources of error 
in photogrammetry which limit obtainable accuracy. The adjustment technique 
must fit the system one chooses, so that there could never, in my opinion, be