Fruit taste vs offspring

With subjects like plant breeding, I inevitably get stuck on a picky detail which just doesn't make sense to me. Hopefully people who know more than me can reassure me...

As a mother, only 1/2 my genes went randomly into each kid. So, why do they say that you should save seeds from the fruit that tastes good, or has other characteristics you like, when only half those genes will be in the seed, and not necessarily into each seed. The spiral helix splits and 1/2 each goes into each seed, so how do you know which side of the helix has the trait you want and you can't tell which half is in which seed.... ???

So my brain goes to...   it seems as if you can't do this on a decent scale, it's quite possible that doing too much "choosing" might do as much harm as good?  I'm really hoping I'm either missing something, or just worrying unnecessarily, but I fear I'm feeling stuck on this point. (The vast differences between my two kids, isn't helping!)

ETA - maybe reading Joseph's book will help? Does it cover things like critical mass?

You're playing the odds, if it tastes good then it definitely has the gene/s for tasting good so it is possible it can pass on those gene/s if it tastes bad it may have the genes to taste good, but it may not.
So saving seed from the good tasting plant gives you a higher chance of getting the genes you want whereas saving seeds from the other plant has a lower chance of doing so.

What do you mean by a decent scale? it's easy enough to taste 100 squash plants and pick the ones you like, equally if you want earliness you can plant 1000 and save seed on a specific date, anything that isn't ready is automatically sorted out. Lets say you're trying to breed a red cabbage, anything that isn't red, or isn't cabbage shaped gets rouged out before it gets a chance to flower, the same with anything that can't cope with your growing conditions, you will be left with a load of plants that are that little bit closer to what you are looking for.

Also things like beans, peas, and tomatoes tend to self fertilize.  So if the tomato tastes good the seeds may very well have gotten the same genes or very similar genes from each "parent".

Do you remember learning about punnet squares in school? The simple version we usually see is kinda overly simplified, since it ignores traits that are determined by multiple genes, but it's still useful to estimate the odds of a trait showing up.

Suppose you want gene 'A' in your population, because fruits with the 'A' gene taste good. It's a dominant gene, so one copy from either parent is enough to give good flavor.

You find one tasty fruit in the whole population. You don't know this, but it inherited 'A' from one parent. The other parent tasted terrible.

		A	B
B		AB	BB
B		AB	BB

If you save seed from your tasty fruit, next season those seeds have a 50% chance of producing tasty fruit. If you only save seeds from tasty fruits, (and only allow pollination from tasty fruited fathers) the odds are better in the next generation.

		A	B
A		AA	AB
B		AB	BB

75% odds of tasty fruit. Keep culling the plants that have poor tasting fruit. Don't let 'em make seeds, don't let 'em contribute pollen. The occasional 'AA' (25%) will improve your odds even more.

		A	A
A		AA	AA
B		AB	AB

The goal is to reach the point of none of the parents being 'BB'. 100% tasty fruit! Since we can't actually see the genes, it's hard to know when we've made it. Sometimes some other trait is linked to the combination we want, so observation can really pay off. (Think wrinkled sweet corn seeds.)

Vigilantly only allowing tasty specimens to reproduce, guarantees that you'll get there. Possibly in three to four generations.

Awesome info. I think I felt the faintest wisp of a memory of these tables from biology class...

My suggestion, is to ignore everything you heard or read that makes it difficult to save seeds. Don't worry about genetics, population sizes,  isolation distances, or purity. Save seeds from what you love. Children tend to resemble their parents and grandparents. By saving seeds from great ancestors, we are selecting for great offspring.

If a population of plants are genetically diverse, and promiscuously pollinating, they are capable of becoming locally adapted. The gardener can select among them for the most pleasing textures and flavors. By saving seeds from what we love each year, the entire crop moves in that direction. If we can screen the crop before pollination occurs, then the selection is even quicker.

I don't find mutant monsters among my promiscuously pollinating crops. People and plants have lived in symbiotic relationships for thousands of years. The poisons and thorns of the wild species were left behind millennia ago.

T Melville wrote:Do you remember learning about punnet squares in school?

Lest I be misunderstood, these calculations aren't needed to breed desirable varieties or populations. Lots of people have done great plant breeding without them. They just serve to visualize likelihoods. A confidence builder that you can do it. But it's not math and charts that improve your crops. It's playing the genetic lottery enough times, observing what shows up, and making sure that you save more seeds from the best offspring & fewer or none from the worse results. Chance plus selection.