Luse 
and 
lary 
ial 
- 1909 - 
an indication. But creeper forest is also found in hilly terrain and along 
streamlets and is easily mistaken for secendary forest - 
Shrub forest on rocky soil can be separated from secondary forest if rock 
outcrops are detectable and if the lecation is aleng ridges. The presence 
of landslides can also be an indication that the low forest type observed 
on the aerial photograph is à natural one and not due to human influence. 
Summarized, one can state that for the differentiation between secondary 
forest after shifting cultivation and low forest types, physiographic 
features and location are an important key. In an area without roads the 
separation of low forest types from secondary forest becomes doubtful, 
especially if no recent shifting cultivation occurs near the streamlets 
and rivers. 
A normal procedure of photo-interpetatiom for forest type mapping is to 
classify scrub and scrubforest on dryland together as one class disregarding 
the possibility that these types of vegetation are either due to poor soil 
conditions or are a degenerated forest type after prolonged shifting 
cultivation. 
Secondary forest may be subdivided into young and old. Because of the 
almost one hundred procent crown cover, the subdivision is not made with 
the help of height measurement (parallax method) on aerial photographs but 
is based on a subjective classification relating the height and crown 
diameters of the secondary forest to the natural forest. The scale of the 
aerial photographs can influence the interpreter in classifying young and 
old secondary forest; there may be overestimates of young secondary forest 
on large scale and old secondary forest on small scale photography. 
This rough differentiation into young and old secondary forest is not 
detailed enough for a calculation of the rotation period. Reynders (1962) 
in West Irian calculated a 25 years rotatien period of shifting cultivation 
using aerial photographs combined with fielddata. For the different stages 
of secondary forest age classes of 1 to 4 year were used. In forest type 
mapping, fielddata on the age of secondary forest of former fields, accurate 
enough to be delineated on the aerial photograph, are normally not available. 
Because in the tropical rain forest no amnual rings are formed by trees, 
there is no accurate determination of the afe of secondary forest. With 
experience the age can be assessed for the first 4 to 6 years but an 
estimation of the age of older secondary forest is pure guesswork; 
information on this data from the settler can be, for several reasons, 
unreliable. A subdivision of secondary forest according to age classes and 
its delineated on the aerial photographs has to be considered as a system 
with a limited accuracy in the tropical rain forest regions.