Revisiting source versus sink limitations of wheat yield during grain filling

To further increase wheat (Triticum spp.) yield, we need to understand whether it is source or sink limited. Earlier papers suggested that wheat yield is source limited in modern cultivars, including during the grain-filling stage. Many recent papers support this interpretation, showing that yield is strongly related to environmental conditions that affect source capacity. In contrast to this, other authors working on source–sink manipulations have concluded that wheat yield is only or predominantly sink limited during grain filling. The objective of this forum paper was to examine this contrasting literature and revisit the assumptions and the interpretation of the results. We found that the arguments for a major sink limitation to wheat yield during grain filling arose from a common approach to quantitatively assess the degree of source–sink limitations, based on relativizing the change in grain weight (in response to source–sink manipulations) to the estimated change in source availability per grain. We show that the estimated changes in source availability with source manipulations are often overestimated in the literature, thus underestimating source limitations. Most importantly, we discuss why relativizing the change in grain weight to the estimated change in source availability biases the interpretation of source versus sink limitations. We conclude that wheat yield is mostly source limited during grain filling, and thus strongly dependent on environmental (including agronomic) conditions. A new model to interpret wheat yield limitations is proposed, describing yield as source limited during the whole crop cycle, including during grain filling.