Years ago I participated in what turned out to be an epic tomato breeding project. It started with an enthusiast who found a dwarf tomato variety in a USDA collection (iirc). He grew it out and sure enough, it was a stocky indeterminant plant that needed no staking and didn't take over the yard.
It was crossed with an heirloom varietty with superior flavor and the project was born. It takes about 9 generations for tomatoes to stabalize after cross pollination. So to speed up the process, seeds were shared in the northern and southern hemispheres, getting at least two growing seasons in per year. Second generation seeds segregate for parental traits and the selection begins.
During the selection process, we planted out received seeds, say 1-2 dozen and it quickly becamse apparent which picked up the dwarf trait. They were half the size of the other seedlings. These were kept and grown out searching for something worthy, the other's culled.
Fast forward and there are about 100 dwarf varieties that were created - big, flavorful heirloom quality in a manageable grow-in-a-pot tomato plant. These were shared with small scale seed companies and have become widely available now.
So the selection process left me wondering about mangoes. I know the genes are different (heterozygous vs. homozygous) and but paradoxically, selection might be faster with grafting onto already mature trees. I have an ultra low branching Julie seedling in my yard and I'm curious, what % of Julie seedlings carry the dwarf 9or slow growing) trait.
And what % of those would provide an interesting improvement on Julie?
Cuttings from the dwarf reverse grafted onto an already mature tree would tell within a few seasons...
https://www.dwarftomatoproject.net/about/Footnote I contributed to Dwarf Kangaroo Paw