recent experience in this field is reported. Stereocomparator observations on 
prints made from aerial films photographed at 15 000 feet reduced the standard 
error of height determination to about 1/10 000 of the flying altitude. Experi 
mental work with the parallax bar on photographs taken at 30 000 feet with 
sparse ground control gave H/6 000. And there is still room for improvement 
in both techniques. Furthermore, the internal consistency and homogeneity 
promise good results in extensive aerial triangulation. 
It is however important in the present stage of development to realize that ten 
years ago the analytical method, like the plotting machine method, showed 
abrupt changes in the discrepancy curves (cassures). This observation led the 
writer into a series of fundamental researches which disclosed the intrinsic 
shortcoming of the analytical method which was then in current practice, namely 
that it relied on observations made on the least number of symmetrically- 
distributed minor control points. The use of groups of minor control points was 
therefore suggested. It has since become the practice to use double minor control. 
The results have improved considerably, and the lateral discrepancy curves are 
now very much smoother. The statistical reasoning underlying these views is 
presented in Part I of this communication which includes the results of a recent 
investigation into the outstanding issue of the influence of the presence of 
correlation on inference from the residuals. 
Part II of the paper is devoted to a description of the methods used in Egypt, 
including the analytical evaluation of profiles from stereocomparator observa 
tions. Sample calculations are given; and the results are analysed with respect to 
both the internal consistency and absolute error. 
Part III is a continuation of the author’s communication to the last Congress 
entitled »Some recent progress in the design of machinery for analytical photo- 
grammetry». A full description is given in the present paper of (a) the author’s 
simple design of an attachment to electric desk calculating machines which would 
endow them with a storage capacity and means of automatic exchange of 
numbers between the machine and the store, as well as automatic or semi 
automatic performance of the sequencies of arithmetic operations involved in 
the various stages of analytical work; and (b) his design of a self-recording stereo 
comparator which would automatically identify and record the readings of the 
four coordinates to the nearest one thousandth of the pitch of the micrometer 
screws. 
The concluding Part IV is a critical analysis of current literature and trends. 
It is observed that recent publications on the use of electronic computing engines 
quote percentage improvements of efficiency without giving any details about 
the way in which data had been handled prior to the introduction of these 
engines. In other words, the standard of comparison is missing. Furthermore, no 
serious attempt is made to examine the possibilities of the less spectacular 
approaches which would be more in line with the requirements of the smaller 
establishments. In general, much of what has been published in this field lacks 
those characteristics of scientific research which the author submitted to the last 
Congress in his paper on the »Design, analysis and presentation of photogram- 
metric experiments». In the writer’s opinion, it is high time to re-examine the 
whole problem of analytical photogrammetry if we wish to make this potentially- 
powerful method acceptable and accessible to everyone, not a speciality of a few 
organizations. This will better serve the advancement of photogrammetry. 
Photogrammétrie analytique en pratique 
Il est évident que la photogrammétrie est en train de prendre une tournure, car, il ne sera pas 
longtemps avant que la méthode analytique devient le moyen normal d’établir le contrôle pour 
l’établissement des cartes par les méthodes de l’aérophotogrammétrie. L’application de la méthode 
analytique aux photographies prises aux hautes altitudes dans le but de contrôle a des possibilités 
immenses. Une expérience nouvelle à cet effet est donnée par la présente communication. Les 
241 
le