1 Section for Organismal Biology, Department of Plant and Environmental Sciences, Faculty of Science, Københavns Universitet2 Lanzhou University3 Section for Organismal Biology, Department of Plant and Environmental Sciences, Faculty of Science, Københavns Universitet
Plants produce biomass and then allocate some of this biomass to reproduction. The pattern of reproductive allocation is an important aspect of a plant’s reproductive strategy in nature and is closely linked to yield and Harvest Index in cereal crops. Recent research has concluded that reproductive allocation should be analyzed and interpreted allometrically because ratios or fractions such as Reproductive Effort or Harvest Index are size dependent. We investigated reproductive allocation of individuals in 6 varieties of Triticum (wheat) grown at a wide range of densities. We harvested leaves, stems, spikes and grains of individual plants and analyzed the relationship between grain mass and vegetative mass allometrically. The large variation in density created large variation in plant mass and reproductive output. Most of the variation in individual yield (grain mass) was due to variation in plant size. There were significant differences among the varieties in the allometric exponent (slope of log–log relationship) of grain versus vegetative mass, such that some varieties produced higher yield (and therefore had a higher Harvest Index) than others when plants were small, while others had higher yield at larger sizes. Thus, the Harvest Index and its rank among varieties changed with plant size, which puts into question the practice of selecting for Harvest Index when crop performance varies greatly among individuals, years or environments. Selection for a high Harvest Index when individuals are large may mean unintentional selection for a lower Harvest Index when individuals are smaller. We conclude that cereal breeders should focus on reproductive allometry when interpreting Harvest Index, and select for allometric patterns that are most advantageous in a given agronomic context, especially when there is large variation in productivity among individuals, locations or years.