In the evolution of wheat, there are two instances of chromosomal doubling, when Emmer wheat Triticum turgidum was formed from Einkorn wheat, and when Triticum aestivum was formed from Emmer wheat.
When allopolyploids are artificially made, chromosome doubling is done by chemicals. Here, it occurred by natural endoreplication. But for the entire plant to turn out with allotetraploidy and allohexaploidy respectively, wouldn't it be necessary for endoreplication to occur just after zygote formed--quickly, so all, or most, cells of the plant would have this genotype?
Isn't the probability of that happening twice in the process of evolution, both times in those incidences when the wheat crossed with the goatgrass species, very low? Is it just coincidence that it happened this way, at the right time, to allow for the allohexaploid species to develop naturally?