Developing methodologies to understand farmer-managed maize folk varieties and farmer seed selection in the Central Valleys of Oaxaca, Mexico

摘要

Collaborative plant breeding (CPB) is an approach to crop improvement incorporating close attention to local biophysical and sociocultural environments and interaction between farmers and plant breeders. CPB may have particular potential for improvement in highly stress-prone environments and for low-resource, traditionally based agricultural communities, situations where more conventional approaches have not been effective. However, CPB will require methodological adjustments or innovations relevant to the smaller scale of its target area and its participatory approach. This study investigated methodologies useful to CPB, working with maize farmers from two communities in the Central Valleys of Oaxaca, Mexico. A method for rapid estimation of broad sense heritability (H) was applied in farmers' maize fields. H estimates for morphophenological traits were compared with narrow sense heritability (h2) from half sib family analysis of five of the same populations and with published estimates. Absolute values of H were larger than h2 from this study and the literature, however trait ranking was the same as in the literature, but differed from h2 rankings from this study. With an understanding of their limitations, these rapid, economical estimations provide useful information for CPB work on-farm, where empirical information is frequently lacking. Collegial interaction based on the knowledge and skills of farmers and breeders will depend upon understanding those in terms relevant to each group. Methods from social and biological sciences were integrated to understand selection and its consequences from farmers' perspectives but based on concepts used by plant breeders. Information was elicited regarding farmers' perceptions of their maize populations, growing environments and expectations for response to selection. Farmers' decisions about varietal repertoires imply assessments of local genetic and environmental variation. Traits of high and low heritability are distinguished, as reflected in expected selection response. Farmers' selection practices were not always effective yet they understood the reasons for this and had no expectations for selection response in some traits given the methods available. Farmers' statements, practices and perceptions regarding selection and the genetic response of their maize populations to their selection indicate selection objectives different than may be typically assumed, suggesting a role for breeder and farmer collaboration.