Seed yield and quality of three foundation seed models under the formal seed system

Abstract

In Mali, smallholder farmers are generally not involved in foundation seed production which
is dominated by the conventional public sector based models. These dominant models have
so far failed to avail sufficient quantities of foundation seed especially for non-irrigated rice
systems, groundnut, and cowpea. During 2018 and 2019 cropping seasons, field trials were
conducted to test three models of foundation seed production, namely: Research Institutions
Model – RIM, Seed Companies Model – SCM, and Smallholder Farmers Model – SFM.
Single varieties of five crops vital to food security in Mali (rice, millet, sorghum, groundnut,
and cowpea) were used in a randomized complete block design with three replicates. The
objectives were to identify the best performing models in terms of seed quantity, quality and
model efficacy in realizing the yield potential of each crop variety. Significant differences
were detected between models and crop performances (P≤0.05). Owing to the trainings and
technical backstopping provided to smallholder farmers, the SFM realized the best performance
in terms of seed quantity and seed yield; followed by SCM, while the RIM realized the lowest
performance. No quality issue was reported for millet, sorghum, and cowpea even for the
seed produced by smallholder farmers. Among crops, millet realized the best performance for
seed quantity and seed yield and differed significantly from the two legume crops. None of
the three models realized the yield potential of the rice variety used in the trials. In addition,
there was problem with rice seed quality  all models due to variety contamination.