Most of the photographs which are used for forest inventory have been taken 
for other purposes (geology and topographic mapping). They therefore have too 
small a scale which is not well suitable for forestry interpretation. Another great 
disadvantage in most tropical regions is the very heterogeneous forest composi 
tion and the absence of sufficient passable roads to study the particular com 
position of certain areas. 
Nevertheless we are able to state that the use of aerial photographs for tropi 
cal forest survey has made some progress during the last four years as this 
material has proved to be an excellent help to obtain more general information 
of the forest composition in tropical regions. To prove this statement some 
examples of application of aerial photographs for forestry purposes are given 
from results obtained in Surinam, Uganda and Dutch New Guinea. Though 
other information in this particular field is rather scarce, these examples are not 
meant to indicate that in other regions nothing has been done. However, to 
improve this special technique, better cooperation between the different sur 
veyors working in remote parts of the world, is absolutely necessary in order to 
find a way to improve the results in the future. 
The Forestry Division of the International Training Centre for Aerial Survey 
at Delft is willing to help to stimulate a further development of this very valu 
able technique. 
The paper by Mr. J. Veenenboos, of Netherlands, was also read by title only, 
Analysis and Interpretation of Aerial Photographs for Soil Survey and 
Land Classification Purposes: 
The article deals with a method of using aerial photographs to speed up the 
work of soil- and land-classification surveys. This technique has been accepted 
by soil scientists as a justified means for reliable semi-detailed and small-scale 
mapping. It is based on photo-analvsis in combination with field work, re 
stricting photo-interpretation to certain and well defined conditions. 
The analysis is performed with the help of many different elements which are 
possible indicators of differences and similarities in soil conditions and land 
classes. By this method maps of higher quality can be made at shorter notice 
with less cost. 
J. Veenenboos: Analysis 
and Interpretation of 
Aerial Photographs for 
Soil Survey and Land 
Classification Purposes. 
(Publ. XV 2) 
In the discussion on the papers above in session d Discussion 
Dr. P. Tham of Sweden, made the following comment. 
A question of great importance in forest photogrammetry is of course the con 
tact scale of the photographs. Using wide angle cameras, scales of 1:15 000 to 
1:30 000 have been recommended, and 1:30 000 should be the limit to-day for 
forest surveys in Sweden. 
The Swedish Committe of forest photogrammetry (Namnden for skoglig foto- 
grammetri) reported in 1951 on the study of scales from 1:8 000 to 1:15 000. 
An English specialist who visited Sweden recently stated that scales smaller 
than 1:20 000 should not be used. I believe that it could be of great interest to 
discuss this problem now, and learn what experience other countries have had 
in this respect. 
Mr. S. T. B. Losee of Canada: Scale in certainly an important question. In 
Canada we began with scales similar to those which I understand are being 
used in Sweden to-day, that is, scales in the neighbourhood of 1:15 000, and as 
our work became more intensive we turned to large scales. At the present time 
my company is doing considerable work at scales of 1:8 000 with partial 
coverage at scales of 1:15 000. 
345