Alternative Paths to Food Security - The Oxford Topic of Food, Politics, and Society - page 210

to the current crop improvement strategy based on genetic modification and applica-
tions of inorganic nutrients. In fact, the gains from these gene-centered and input strate-
gies have produced less yield improvement in recent years than are now being seen with
agroecological theory and practice.
Genetic Improvement
The gains in yield achievable through hybridization (heterosis), compared to those
attainable from inbred varieties, are theoretically 30 percent, and more practically about
15 percent (Yuan 1994; Zhong et al. 2005). Moreover, some part of the yield increases
reported for hybrid rice varieties should be attributed to the use of management prac-
tices that are part of SRI: transplanting single, young (13-day) seedlings, spaced 30x30 cm
in a square pattern, with reduced water application, and with increases in organic fer-
tilization; however, inorganic fertilizers and herbicides are still heavily relied upon
(personal communication from the originator of hybrid rice, Prof. Yuan Long-ping,
September 13, 2004).
On-farm comparison trials done in Bali, Indonesia, in 2006 found that SRI manage-
ment methods added 50 percent to the yields of Chinese hybrid varieties grown with
standard practices (13.3 tons per ha vs. 8.4 tons per ha; Sato and Uphoff 2007). That the
gains reported from hybridization are in part a consequence of changes in E , not just
in G , is clearer from a new world record for paddy yield set in Bihar State, India, in 2011
(Diwakar et al. 2012). One farmer using SRI methods got a paddy yield of 22.4 tons per
hectare, measured by state officials and accepted by the Indian Ministry of Agriculture;
four other farmers in the same village, also first-time SRI users, got yields or 19 or 20
tons per hectare. These “super-yields” were reached with modern hybrid varieties and
with integrated nutrient management, using a combination of organic and inorganic
fertilization). Relevant here is that these farmers, when using the same hybrid varieties
on their farms with conventional crop management methods, the same soil, and same
climate, had paddy yields of 7 tons per hectare, which is very good, but much less than
attained with SRI management (data from Dr. M.S. Diwakar, director of Directorate of
Rice Development, Govt. of India).
In the 1990s, scientists at the International Rice Research Institute (IRRI) reported
that they had developed a “new plant type” (NPT) that could give, on average, 25 per-
cent higher yields (Khush and Peng 1996). The NPT would have many fewer tillers
(stalks), but all would have large grain-bearing panicles. Some part of the claimed
increase should have been credited, however, to researchers' different management
practices, ones associated with SRI: 14-day-old seedlings planted singly in a square
pattern 25x25 cm apart. NPT plants were grown under standard flooded-field condi-
tions, however, with heavy applications of chemical fertilizer and herbicides. Despite
heavy investment in developing the NPT, this strategy for raising rice yields ostensibly
through plant-breeding improvements in G has remained little used, and is no longer
mentioned in IRRI's annual reports.1 The recent history of making improvements in rice
production by focusing on modifications of G with suboptimum E has thus not been
very successful.