can improve the relevance of their research by drawing upon farmers' own informal methods of
experimenting with unfamiliar cultivars and practices. Farm innovators over thousands of years
have enabled the human population to double ten times since agriculture began, including eight
doublings before industrialisation and the use of fossil fuels:

 The human population expanded as traditional agricultural societies learned to domesticate
 animals, select crop varieties, manage weeds and insects, and enhance nutrient recycling. Both
 ecosystems and social systems were modified to sustain improved agricultural technologies.
 The transformations occurred through experimentation, fortuitous mistakes, and natural selec-
 tion (Norgaard, 1985).

African farmers are less likely than scientific breeders to seek a single best cultivar for any given
crop. Instead, an accepted new cultivar usually joins other valued genotypes of the same crop in
a farmer's fields. Mixed stands (of cultivars as well as species) are conventional. Plant breeders
can ease their own task by combining groups of relatively compatible traits into different cultivars
in the knowledge that farmers will readily manage more than one.

Yield stability in Africa, unlike that in industrial economies, depends on a patchwork of many
different varieties planted on the same farm, rather than on a continuous supply of new cultivars
(Plucknett and Smith, 1986). In the West, rapid evolution of new races of pathogens prompts a
frequent turnover of cultivars of such crops as wheat, for which the average lifespan of a new
variety in northwestern US is only five years (Plucknett and Smith, 1986). Wheat mixtures have
recently been rediscovered as a means of managing pathogens. The biological hazards of genetic
homogeneity in the US are demonstrated by the speed with which Florida's citrus crop succumbed
to citrus canker bacterial infection in the mid-1980s, and by the devastating southern corn leaf
blight in 1970. In 1983, for example, 86% of Florida's commercial orange harvest consisted of
just three varieties, while two-thirds of its grapefruit crop was made up of a single strain
(MacFadyen, 1985).

In developing countries, cultivar specialisation may increase short-term profits for a few large
farmers, but threaten the long-term environmental and economic sustainability of production.
The IARCs can help national programmes to reduce the likelihood of epidemics caused by
breakdown of monogenic resistance in popular cultivars.

In addition to epidemiological reasons for monitoring cultivar specialisation in Africa, the local
relevance of breeding agendas depends upon understanding farmers' everyday strategies of
cultivar diversification. Some maize and potato examples illustrate this point.


How Rwandan Farmers Use Potato Cultivar Diversity

Farmers in Rwanda recognize several dozen different potato varieties, which they distinguish
according to plant and tuber traits, as well as agronomic and culinary characteristics. Most grow
three to eight different cultivars at once. They mix cultivars within fields, and use variability in
traits such as the length of the growth cycle, dormancy (time elapsed between physiological
maturity and sprouting), disease resistance (particularly late blight), tolerance of rainfall excesses


GATEKEEPER SERIES NO. SA30