> After all, there's no guarantee that "hybrid vigor" actually improves the metrics humans care about (such as yield), rather than causing some other traits (s

> After all, there's no guarantee that "hybrid vigor" actually improves the metrics humans care about (such as yield), rather than causing some other traits (such as leaf size) to be expressed stronger. I can assure you in strongest terms that, yes, hybrid vigor is unique in this regard. 'Solving' hybrid vigor in a way that you can capture the benefits in a stable inbred is a multi-billion dollar innovation. The most innovation in breeding is happening in maize. It's responsible for the most high-throughput sequencing, the most automated greenhouse metrology, the most funding in general (esp. in China). About 10-15 years ago any plant biologist was basically guaranteed funding if they put the words 'epigenetics' and 'heterosis' in their grants. Everyone figured that the recent understanding in DICER, the plant RNA Pols, DNA methylation and histone modifications, etc. would lead to a solid working model of heterosis. It didn't happen. Every conceivable growth and yield trait in maize has been analyzed. The qTLS have been found, the regulatory elements mapped, the chromatin environment understood, etc. It's very difficult to explain how well maize has been studied; in many ways the field is vastly ahead of any other model despite the absurd difficulty of doing genomics in such a repetitive and large genome. The notion that leaf width/area hasn't been analyzed to death with regard to yield is laughable. Sidenote: the plant breeders have been wiping the floor with animal breeders because they naturally perform group selection in crop rows. Things like broad leaves are bad bad bad because they cost a lot and just end up shaded by their neighbors and higher leaves on the same plant. Animal breeders fail here and have been selecting for hyper-aggressive animals that fight and stress everything to all hell, much to their detriment. The fact is that heterosis matters, and any high-performance inbred will be beat by even a moderate-performing hybrid by any reasonable metric.