8th INTERNATIONAL WHEAT CONFERENCE

SPIke gRoWTh ReguLATINg RATe of deATh
ANd SuRVIVAL of fLoReT PRImoRdIA Seem key
PRoCeSSeS deTeRmININg gRAIN NumBeR IN WheAT
Fernanda G. González1 , Gustavo A. Slafer2 ,
1, 3
Daniel J. Miralles
1 Consejo Nacional de investigaciones científicas y técnicas (CONICET), EEA Pergamino,
INTA. Ruta 32 km 4.5, B2700WAA Pergamino, Buenos Aires, Argentina.
2 ICREA, Department of Crop and Forest Sciences, University of Lleida, Centre UdL-IRTA.
Av. Rovira Roure 191, 25198 Lleida, Spain. 3 Departamento de Producción Vegetal, Facultad
de Agronomía, Universidad de Buenos Aires, Av. San Martín 4453, C1417DSE Buenos
Aires, Argentina.
fgonzalez@pergamino.inta.gov.ar
The most widely accepted model to understand wheat yield physiology identifies a critical
period from the emergence of the penultimate leaf to anthesis. During that period
the model considers that the growth of the spike is limited by the assimilate supply due
to the competence made by the stem, which is growing at its maximum rate at the same
time, and interplant competition is high (and therefore individual resources limited) in
well managed crops. This competence would result in the death of most of the florets that
had been previously differentiated. The model has been widely validated (with experiments
modifying radiation, photoperiod, soil N, genotypes) by the positive relationship
observed between the number of reproductive organs (either fertile florets or grains) and
the spike weight at anthesis. However, the importance of the number of fertile florets
or grains for wheat yield determination as well as the assimilate limitation focused on
this critical period have been questioned recently. The main argument is that a good
relationship between two variables does not necessary indicates causal relationships,
unless detailed studies showing association between the assimilate availability and the
development and survivalof floret primordia were available. As it is important to settle
whether the bases of the model are mechanistic and not just casual, we aimed to validate
it from the analyses of the detailed processes of floret development and dynamics of spike
growth that led to the strong relationship frequently observed at anthesis. We analyzed a
set of experiments including different cultivars, both spring and winter, semidwarf and
conventional-height, photoperiod-sensitive and -insensitive types, form different breeding
programs (Australia, Argentina and UK). Field and growth chamber experiments
were included, covering six different growing seasons (1997 for the growth cabinet, and
2000, 2001, 2002, 2005, 2006 for field experiments), and different treatments were applied
during the spike growth period to alter its growth. Our analysis at three different spikelet
positions along the spike (apical, central and basal positions) showed that there was no
relationship between the number of living florets at anthesis and the maximum number
of differentiated florets, for a range in floret survival (ratio between fertile to total number
of differentiated florets) from 0 to 0.64. In contrast, a unique (for the three spikelet positions)
and strongpositive relationship was observed between the number of fertile florets
342