damage than H-210 hybrid, when intercropped with beans.
Clearly, much further work is needed before appropriate crop
mixtures and row spacings are to be achieved.
 The manipulation of weed abundance and composition in in-
tercrops can also have major implications on insect dynamics
(Altieri, 1983). When weed and crop species grow together, as it
is commonly observed in traditional cropping systems, each plant
species hosts an assemblage of herbivores and their natural
enemies, thus trophic interactions become very complex. Many
weeds offer important requisites for natural enemies such as alter-
nate prey/hosts, pollen or nectar as well as microhabitats which
are not available in weed-free monocultures. Relevant weeds that
support rich natural enemy faunas include the perennial stinging
nettle (Urtica dioica), Mexican tea (Chenopodium ambrosioides),
camphorweed (Heterotheca subaxillaris) and goldenrod (Solidago
altissima) (Altieri and Whitcomb, 1979). In the last 20 years,
research has shown that outbreaks of certain types of crop pest are
more likely to occur in weed-free fields than in weed-diversified
crop systems. Crop fields with a dense weed cover and high diver-
sity usually have more predaceous arthropods than do weed-free
fields. Ground beetles, syrphids, lady beetles (Coccinellidae) and
other predaceous insects are especially abundant in weed-diversi-
fied systems. Relevant examples of cropping systems in which the
presence of specific weeds have enhanced the biological control of
particular pests can be found in Altieri and Letourneau (1982).
These observations suggest that selective weed control may
change the mortality of insect pests caused by natural enemies.
The ecological basis for obtaining crop-weed mixtures which
enhance insect biological suppression awaits further development.
 In traditional agroecosystems, the dispersion of crop plants in
species rich or genetically diverse mixtures restricts the spread of
pathogens. Such diversity gives a measure of stability in that the
failure of some species or genotypes due to diseases may be com-
pensated by the improved performance of others (Thresh, 1982).
Proper inclusion of immune or resistant crop plants in the mix-
tures can impede pathogen spread and increase the separation
between susceptible plants. Growing of tall plants together with
shorter crops can significantly decrease the spread of diseases by
either acting as a barrier to the free spread of propagules or by in-
terfering with the movement of insect vectors. For example, in
Japan, radish mosaic decreased when radishes were sown between
rows of rice, and pigeon peas in Haiti were protected from virus
diseases when grown between rows of tall sorghum (Palti, 1981).
 There is also evidence that some plant mixtures adversely affect
nematode populations. Marigolds (Tagetes spp.) offer great
potential for nematode reduction through toxic action in inter-
crops (Visser and Vythilingam, 1959). Intercropped Crotalaria,
itself susceptible to nematode attack, diverts nematodes
(Radopholus similis) from other crops and then interferes with
the nematode life cycle within its roots (Palti, 1981).
 Although traditional intercropping appears to offer con-
siderable potential as a means of increasing crop dominance over
weeds, there is much room to improve the effectiveness of weed
control in intercrops by manipulating crop diversity, spatial ar-
rangement, soil fertility, relative proportions of component
crops, and use of competitive cultivars. In shifting cultivation
systems, the natural regeneration of forest vegetation can be
replaced by a legume cover crop such as Psophocarpuspalustaris,
Centrosema pubescens and others which provide excellent
vegetation cover to the exclusion of weeds (Akobundu, 1980).
Similarly, weeds can be effectively suppressed in intercropping
systems by the use of low-growing crops (i.e., melon and sweet
potato) which quickly shade the soil surface thus minimizing
weed growth. Legumes intersown between maize rows offer the
opportunity for improving soil fertility, crop yield and weed con-
trol on otherwise impoverished soils of the humid tropics
(Akobundu, 1980). Crop density can be easily manipulated to


promote crop dominance over weeds in intercrops. Highest Com-
bined crop yields and the greatest degree of weed suppression
were obtained from a sorghum/pigeon pea mixture with a nor-
mal density of pigeon pea sown with a twice normal population
of sorghum (Shetty and Rao, 1981).

Extension of Appropriate Pest Management Practices
to Small Farmers
 Generally in developing countries, the extension approach
consists of researchers developing recommendations, preferably
with continuous suggestions and critical input from farmers via
the extension service. Thus, extension agents are the principal
link between researchers and farmers. They reach farmers
through demonstrations, training courses, follow-up visits, often
with local pilot farmers as examples and demonstrators for others.
Seeds, pesticides, equipment, subsidies, credit, etc. are part of
the "package deals" and must be available at the proper time and
at an accessible place (Matteson et al., 1984). So far, the few
technological breakthroughs made in peasant farming, have in-
evitably been accessible to those peasants of recognized ability
and to those most favored in terms of control of resources and ac-
cess to markets, roads and credit (de Janvry, 1981). Moreover,
due to the heterogeneity of peasant farmers, global recommenda-
tions have proven to be seriously unfit for the majority of small
farmers who are usually confined to marginal areas. Thus, the
recommendations have only been, confined to accommodated
peasants that enjoy better soils, natural resources, and institu-
tional support.
 If rural development is indeed successful among small farmers,
technical and organizational strategies must emphasize:
 1. improvement of use-efficiency of local resources;
 2. minimization of dependency on purchased inputs and in-
 dustrial technology;
 3. satisfaction of self-sufficient production; and
 4. organization of peasants to enhance their cooperation for
 economic and social survival.
There are several non-profit groups in the developing world em-
phasizing the "bottom up" or "grassroots" approach to rural
development. These groups, meagerly funded and isolated from
the mainstream agricultural colleges and ministries, have an
ecological vent relying on resource conserving technologies that
promote nutrient cycling, natural pest control and soil conserva-
tion (Altieri, 1983). The establishment of modular systems in
Tabasco, Mexico (Gliessman et al., 1981) and of improved high
land cropping systems in Bolivia (Augstburger, 1983) are suc-
cessful examples. The establishment of a self-sufficient ex-
perimental small farm (0.5 ha.) in Chile (CET, 1983), where
most of the food requirements of a family of scarce capital and
land can be met, has had great impact. Groups of peasants com-
ing from local and distant areas live in CET's farm for variable
periods of time, learning through direct participation in the far-
ming operations, all the organic production practices (i.e., com-
posting, raised beds, intercropping, etc.), farm designs, planting
dates, proper varieties, etc. After their training farmers are given
a basic package of seeds and then return to their communities to
teach their neighbors the new methods, and thus apply the
model in their own lands.
 In the Peruvian Andes, "Grupo Talpuy" has been rescuing and
recording the knowledge of local farmers about farming practices
(i.e., mixed cropping, use of potato varieties, crop rotations, fer-
tilization, etc.), traditional crops utilized, use of non-crop plants,
etc., which is then synthesized and later disseminated in written
form (a low cost magazine called Minka) throughout the rural
areas (Minka, 1981). Each issue treats a different subject (i.e.,
mixed cropping, Andean crops, local herbal medicine, soil con-
servation, agricultural tools, low-cost house construction, etc.) in
a very simple manner, illustrated with a number of drawings and


VOL. XX-PROCEEDINGS of the CARIBBEAN FOOD CROPS SOCIETY


49