Plant breeding for organic and sustainable, low-input agriculture

Evolution of diversity during 20 years of mass selection on ‘Pigarro’, a Portuguese
improved maize population with fasciation.
Pedro Mendes Moreira 1 , Carlota Vaz Patto 2 Silas E. Pêgo 3 , Arnel R. Hallauer 4
1 Escola Superior Agrária de Coimbra, Departamento de Fitotecnia, Coimbra, Portugal;
2 Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa, Portugal;
3 Instituto Nacional de Recursos Biológicos/EAN, Oeiras, Portugal;
4 Iowa State University, USA.
Maize was introduced in Portugal, after America’s discovery by Columbus. Maize adaptation to
a diversity of landscape, climate conditions and people’s needs, took place during five centuries.
These adaptations led to flint-type open pollinated landraces with technological ability for
production of the traditional maize bread called “broa”. “Broa” production still has an important
economic and social role in Central and Northern Portuguese rural communities and this is
probably why traditional maize landraces have not been yet totally replaced by hybrids.
In 1984, Pêgo started, with the CIMMYT support, an on-farm participatory maize breeding
project at the Portuguese Sousa Valley region (VASO). This project was intended to answer the
problems of small farmers; i.e., increasing yield without loosing the parameters defined by
farmers for bread making quality, potential for polycropping systems, and use in sustainable
agriculture. During 20 years of participatory maize breeding (PMB) at Sousa Valley, mass
selection and recurrent selection were applied on ‘Pigarro’, a maize landrace from VASO.
Concern has been expressed that genetic diversity might be reduced by natural and artificial
(human) selection. Our present objective is to compare mass selection data of Pigarro’s diversity
and erosion over an interval of 20 years, using: morphological data and SSR molecular markers.
Morphological data evaluation (e.g. yield gain, ear length, fasciation level) was conducted in
Portugal (3 locations in 2 years) and in the USA (4 locations in one year) using seven different
mass selection cycles. ANOVA comparisons and regression analyses on the rate of direct
response to selection was done.
Samples from three different mass selection cycles were fingerprinted with SSR molecular
markers. Thirty randomly selected individuals per cycle were fingerprinted with a set of 10 SSR
markers uniformly distributed across the maize genome.
The results from morphological data revealed that: ear length significantly decreased and
simultaneously, ear diameter, kernel row number and fasciation significantly increased. This
selection also led to significant increase of days to silk and anthesis.
The results from SSR molecular markers revealed that no effective loss of genetic diversity has
occurred during the selective adaptation to the farmer’s needs and the regional growing
conditions. 96.26% of variation was attributable to within-selection cycles diversity indicating
that a great proportion of the genetic diversity is maintained in each selection cycle.
Besides morphological and SSR molecular markers comparison; an insight of PMB on VASO
project (e.g. location, germplasm, farmer) will be discussed.
68