Lo 
NS. 
00 
ion 
a, 
roads. 
ngs 
ng 
= 1907 = 
forest. If permanent agriculture or husbandery is not possible, the 
situation can become even worse, shortening the fallow time of the soil 
under secondary forest in order to obtain the same yield of agricultural 
products as from larger fields. This may result, in à rather short time, 
in an abandoned area with degenerated secondary forest, shrub forest or 
even a savanna type grassland. 
A system commonly used in Latin America is to use secondary forest as an 
interface between natural virgin forest and grassland for animal husbandery. 
Because of the short dry season and the high percentage of hardwoods the 
first burning of the fallen trees and the standing stumps is only partial. 
Agriculture by haphazard planting between the criss-cross layer of timber 
is possible, but this prevents good animal husbandery. The natural 
regeneration consists mainly of Cecropia; & quick growing light tolerant 
species. After 6 to 8 years the Cecropia is cut, mainly to get enough 
fuelwood for burning the decaying hardwood for à second time, Then follows 
grass sowing sometimes mixed with a second crop. Whether such a system can 
still be regarded as shifting cultivation is a matter of opinion. The 
result is an open grassland with a pseudo-gallery forest along the 
streamlets. 
Detection of vegetation and forest types occurring with shifting 
cultivation on aerial photographs 
  
As shifting cultivation is a dynamic process the individual agriculture 
plots are not delineated individually and secondary forest is not 
classified into age classes. When the aerial photographs are not up to 
date and recent forest encroachment is not yet registered, it may be 
advisable to include a zone around the existing shifting cultivation 
indicating a possible extension of the human influence. In forest 
inventory shifting cultivation will be normally excluded for volume 
assessment. 
Secondary forest on aerial photographs can give the same tonal and crown 
structure impression as natural low forest types. A differentiation between 
these two categories is therefore not always made. 
Summarizing, one can state that a forest type map, made for a timber volume 
inventory, will have less information about shifting cultivation than can 
be deducted from the aerial photographs. Using such a forest type map for 
problems related to shifting cultivation can result in an incorrect 
determination of the current or future area under shifting cultivation.