INTRODUCTION 
The forest areas of the humid tropics are diminishing because of human 
influences such as illegal exploitation and very destructive shifting 
cultivation. By repeated shifting cultivation the forest may degenerate 
into savannalike shrub forest or grassland. 
In many countries with tropical rain forest there are only obsolete or 
unreliable data available about the extent and the rate of increase of 
shifting cultivation. 
Interpretation of aerial photographs can be most important for quickly 
obtaining data concerning the extent of shifting cultivation in forest 
areas. The rate of increase of shifting cultivation may be calculated if 
sequentional photography is available. Mistakes in area calculation may 
occur when the results of photo-interpretation of up to date areal 
photographs are compared with data obtained from obsolete forest type 
maps and different photo-interpretation systems are used for the map 
compilation. Also caution has to be taken when comparing SLAR and LANDSAT 
MSS image interpretation results of shifting cultivation with the data 
obtained from aerial photographs without ground control. 
For forest surveys in the tropics panchromatic aerial photographs at scales 
ranging from 1:20,000 to 1:60,000 are normally used. Large scale photography, 
true colour, infrared black and white and false colour films are used mainly 
for research purposes. New remote sensing techniques such as side looking 
airborne radar (SLAR) and LANDSAT MSS images are not included in this paper. 
On these small scale images large clearcut areas can be observed clearly, 
but small plots and the differention between secondary forest and natural 
low forest types cannot usually be detected at all. 
In normal photo-interpretation work the emphasis is on delineation of 
forest types of potential economical value and delineation of shifting 
cultivation is mostly generalized. To delineate shifting cultivation, 
the interpreter must be trained and groundtruth data must be available to 
obtain a reliable differentiation between secondary forest after shifting 
cultivation and the forest types with a low tree height growing on poor 
ecological sites. 
These low forest types may results either from topographical or from 
soil influences. Also a distinction has to be made between secondary 
forest caused by natural calamities and by human influences.