REPORT OF COMMISSION VII 
GVII-19 
a genuinely permanent basis, and their streets were generally not surfaced. 
In spite of such hindrances, the study here summarized would have been 
much more difficult and costly without the aerial photographs. They served 
reliably as primary sources of information, and if not available and used, some 
phases of the work would probably have remained undone. Air photos will, in 
the future, enjoy much wider general employment among students of cultural 
landscapes. 
Mr. Jean Gandillot, in charge of a course in aerial photography as applied to 
geology, is studying the utilization of emulsions and colored filters for making 
a better contrast with gray tints and, by studies about assemblies and stereo 
grams, he has discovered some geological facts which had not been seen by 
geologists examining the earth. 
These studies extend to lithology and tectonics, hydrography, volcanism, 
glaciology and littoral oceanography. 
In the field of archeological research, it has permitted the discovery of the 
Roman centuriations of Tunisia, where several very significant systems totalling 
almost 30,000 centuries have been identified, and in France, the centuriation 
of the region of Narbonne. 
RESEARCH LEADS TO NEW IMPROVED DEVELOPMENTS 
In New York State, investigations and studies have indicated the best scales» 
film, season, and kinds of air photos to use for forest survey purposes. Methods 
and techniques in timber cruising have been developed to fit local situations. 
One method is used for large areas of even aged softwood forests, another ex 
plains a plot scheme for hardwood areas, and still a third tree method is recom 
mended for large softwoods scattered in hardwood types. This study used 
crown width and tree heights measured on aerial photographs. Tables showing 
accuracy are given. Comparison of theoretical volume are made with volumes 
obtained from field samples. 
In Canada a research project was initiated in 1952 to test the value of large 
scale photography for detailed forest surveys. Two experimental areas, one of 
40 square miles and one of 10 square miles, were photographed with snow on the 
ground at 1'. 8,000 scale and with strips at 1:1,600 taken on the same flight lines. 
A method of accurately determining stand density from the photographs was 
developed, and suitable techniques of measuring average stand height and ar 
riving at a volume estimate were worked out. According to tests of the methods 
against ground surveys the desired accuracy could be obtained at costs less 
than those for equivalent ground surveys. A final test against the amount of 
wood actually cut from a 3 square mile area is under way at the present time. 
Other developments in Canada include the successful use of large-scale sam 
pling photographs to facilitate the interpretation of the general coverage photo 
graphs; experiments in the use of moving film magazines to minimize the effect 
of the forward motion of the aircraft; and trials of the application of special 
camouflage film to the determination of amounts of insect-killed timber. 
Studies of the systematic errors in measuring tree heights indicate that inter 
preters have different bias for tree sizes, species, photo scales, and seasons. In 
Canada crown diameters could not be satisfactorily measured at 1:7,200 scale, 
but results were quite satisfactory at 1:12,000. New methods are being devel 
oped for measuring crown density and the results are very good. Preliminary 
reports on volume estimating at large scales appear satisfactory. Similar studies 
in Pennsylvania show that crown diameters can be measured with a 6 to 8 foot 
error, crown cover within 10 to 20%, and tree heights range from 14 to 20 foot 
errors at .95 probabilities using 1:12,000 photos.