420 Modern Agricultural Reforms
and available to deal with any new increases in pest populations. Keeping sufficient organic mat-
ter in the soil to support populations of neutrals is becoming part of an IPM strategy (personal
communication, Peter Kenmore, during Bellagio conference).
15 See Fernandes et al (1997). On infertile acid soils, farmers often need to use certain chemical
nutrients such as phosphorus and calcium to prime biological processes such as nutrient recycling
and nitrogen fixation. Research in Costa Rica found that when cultivating beans, mulches of
organic matter prevent phosphorous fertilizer from becoming bound to aluminium and other
ions in the acid soil, making it more available for plant nutrition. Phosphorus applied in conjunc-
tion with organic material produced as good or better yields as when three times as much phos-
phorus was applied directly to the soil (Schlather, 1998).
16 There is research indicating that the application of inorganic nitrogen fertilizer suppresses poten-
tials for biological nitrogen fixation by reducing micro-organisms’ production of the enzyme
nitrogenase which enables soil microbes to transform nitrogen from the atmosphere into forms
usable by the roots (Van Berkum and Sloger, 1983). This suggests that naturally occurring nitro-
gen can be made unavailable by the application of nitrogen fertilizers, but it does not negate the
point that organic and inorganic sources of nutrients are best managed in a complementary man-
ner. It is worth contemplating the fact that since 1950, applications of nitrogen fertilizer have
increased about 20-fold (Smil, 2000, p109), while crop yields have gone up at most three-fold.
While nitrogen is often a limiting factor for plant growth, if it were of overwhelming importance
for plant production, we should see more proportional increases in yield, rather than such sharply
diminishing returns.
17 ‘The importance of water-control techniques in contrast with irrigation is consistently underesti-
mated in the literature. There is a wide range of these techniques, including those that just hold
water in the sandier soils [by increasing soil organic matter] as well as a series of measures to
reduce runoff where crusting is the problem. These are not just indigenous techniques. The most
important ones in the next decade have large potential yield effects (when combined with inor-
ganic fertilizers) and need to be undertaken during the crop season, generally with animal trac-
tion, and not just as emergency measures on the most degraded or most easily degraded regions
(hillsides)’ (Sanders, 1997, p19). On this point generally, see FAO (1994).
18 In the rice–wheat rotation systems widely used in the Indo-Gangetic Plains of South Asia, certain
kinds of ploughing techniques, adjusted by depth and timing, can retain enough water from the
rice season for the following wheat season, so that the amount of water needed for the latter crop
is reduced (personal communication, Craig Meisner, CIMMYT/CIIFAD). Seeding wheat in the
standing rice crop towards the end of its growing season enables the wheat crop to benefit from
residual soil moisture, reducing the need for irrigation (personal communication, Peter Hobbs,
CIMMYT). These low-till methods are being promoted by CIMMYT and IRRI because they can
save water, raise yields, lower production costs, reduce weeds and herbicide use, plus reduce green-
house gas emissions (‘New Movement Among Farmers to Give Up the Plow Takes Root’, press
release from Future Harvest, The Hague, 2 October 2001, http://futureharvest.org/new/lowtill.
shtml).
19 Most of the ideas in this chapter have been prompted from the co-authors’ interactions with col-
leagues at Cornell University and in developing countries where CIIFAD has been engaged in
collaborative, interdisciplinary programmes since 1990 to further the prospects for sustainable
agricultural and rural development (Uphoff, 1996a). It is hard to know where ideas come from,
and to give full or proportional credit where it is due. We take responsibility for presenting these
ideas for critical consideration by researchers and practitioners, not claiming personal credit for all
of them, and acknowledging our indebtedness to colleagues at Cornell and elsewhere for the
stimulation and challenge they have contributed to this thinking. Critical review by Rainer Assé
and Christopher Barrett of the whole manuscript was particularly helpful.