In many parts of Africa, farmers periodically fallow their land, which is a
llowing land to lie idle for one or more seasons primarily to restore its f
ertility. This paper assesses the feasibility, profitability, and acceptabi
lity of improved tree fallows, which are the deliberate planting of trees o
r shrubs in rotation with crops to improve soil fertility. Improved tree fa
llows are assessed at different stages of intensification, drawing on farme
rs' experiences in three different settings. In extensive systems where lan
d is plentiful and existing fallows with natural regeneration of vegetation
restore soil fertility (southern Cameroon), farmers have little incentive
to invest labor in establishing improved fallows. Where population density
is higher and fallow periods are decreasing and farmers perceive a decline
in soil fertility (eastern Zambia), improved fallows have great potential.
In intensive systems where land is unavailable and cropping is often contin
uous (western Kenya), many farmers find it difficult to fallow land. Even h
ere, there is scope for introducing improved fallows, especially among farm
ers who have off-farm income. Labor constraints and institutional support w
ere found to greatly influence the feasibility of improved fallows. In inte
nsive systems, low returns to cropping, low base yields, and a high opportu
nity cost of labor increase the returns to improved fallows. Principal fact
ors associated with acceptability include past perception of soil fertility
problems, past use of measures for improving soil fertility, current fallo
wing, economic importance of annual cropping, and wealth level. Adoption po
tential may be increased by reducing fallow periods, intercropping trees an
d crops during the first season, reducing establishment costs, producing hi
gher value by-products, and by encouraging farmers to test improved fallows
on high-value crops.