Early farmers picked up some fruit of wild plants to home, eat them, and spited out the bigger seeds or defecated the tiny size seeds to human latrines. And then the seeds germinated in the early lab. But the key points were how the early farmers decided to pick up which fruit? It’s not hard to make some hypothesis:
Fruit size:
Strawberry, early farmers went into thorny thicket amid the mosquitoes on a hot, humid day to pick up strawberry; the only criterion is the size. The early farmer would choose only bigger size ones to compensate the while to get sunburned and mosquito bitten.
Bitterness:
Many wild seeds evolved to be bitter, bad-tasting or even poisonous to keep human or animals away from eating them. Natural selection functioned as getting the fruit tasteful dispersed by animals and continue the process of reproducing, meanwhile if the seed remained tasteful, the animal will chew them up as well.
Occasional individual almond trees have a mutation in a single gene that prevents them from synthesizing the bitter-tasting amygdalin, which breaks down to yield poison cyanide. Such trees die out in the wild, while the trees lived near humans survived by being picked up by curious children or early farmers who noticed the non-bitter fruit.
Fleshiness, Oiliness and fiber length are all similar in the selection process.
Initially the wild fruit had less fruit around their seeds, but early farmers chose the fruit mutated to be with more Fleshiness, less seeds, such as squashes, pumpkins, bananas oranges, grapes, watermelons.
Olives were cultivated earliest around 4000 B.C. Ancient farmers also selected sesame, mustard, poppies and flax for oily seeds. Modern plant scientists have done the same for sunflower, safflower, and cotton.
Before recent development of cotton for oil, it was for its fibers, used to weave textiles. Flax, hemp.

Seed dispersal