In what I read and believe, tomatoes are MOSTLY self-pollinating before the flower fully opens, with a few varieties being more promiscuous.
Also, if you have lots and lots of very aggressive, very hungry bees who will force their way into flowers even before they fully open, you can get more cross-pollination than you would otherwise.
It seems to very from less than 1/2% cross-pollination to a few %, or in some rare cases maybe up to 5%.
I think the confusion comes from differing goals that are seldom stated clearly.
1.
Seed preservationists like gene banks and conservators demand CLEAN lines like museum curators. 0.1% cross-pollination would defeat their goals, so
they do whatever it takes to get pure, pure, puuuuure seeds. Even so, they have to "rogue out" each crop and only collect seed form the MOST standard individuals. And they probably have to deal with genetic drift anyway, as the tomatoes self-select to fit better into local conditions of climate, soil pests, diseases, and curator preferences.
2.
Commercial seed producers have to meet federal standards, and would get bad reputations if they sold stuff that was noticeably contaminated. So they plant big blocks and select from the center, and HOPEFULLY rogue out the oddballs. I wonder how much better they do than the "1/2% to a few %" cross-pollination you usually get with zero effort?
3.
People who save seeds at home are very much at the mercy of genetic drift, unless they save seeds from dozens of plants and do a lot of work to save only from "typical" individuals.
Probably most seed savers would be
very hard-pressed to even SEE ANY cross-pollination unless it got up above 10% (very rare) and grew more than 10 plants of the same variety. Even then, they would have to notice the oddball, even if one seed they planted WAS cross-pollinated.
If they had 2% cross-contamination, they would still have to grow out 50 plants to have an even chance of seeing even ONE cross. (I THINK. Statistics are not Zathras' strong point.)
If the cross-pollination is less than you could detect in several years of growing a few plants per year from one batch of saved seeds: WAS there any cross-pollination? I think that's "Schrodinger's cross-pollination". It might BE there, but it would take me many generations of plants to even detect it.
Many technically-inclined home seed savers "just want to do it right" and they bag or isolate or bring cuttings indoors. So they can get less than 0.1% cross-pollination by exerting themselves. They may be planning to preserve their favorite varieties cleanly through an apocalypse, or despite Monsanto buying even more legislators.
However, to even SEE the difference in one year of growing, compared to just growing side-by-side without bags, I think they would have to plant out between 30 and 200 plants each way (from bagged seed, or from side-by-side-seed).
That's just to have an even chance of seeing ONE cross-pollinated plant from side-by-side-seed, and hopefully zero in the bagged seed.
Would that even have a 67% confidence level that the difference was real?
My thought is that when I save tomato seeds, I'll use a LITTLE isolation, like 5-20 yards, then simply avoid growing out 30-300 plants of one variety and micro-scoping them to look for crosses and rouges.
But first and foremost: if I noticed even 1-2 definitely cross-pollinated plants over a few years of growing, I would put that variety back onto my list to
buy some more clean seeds of this variety from a vendor who does the hard work and already has to prove compliance with standards".
I don't mind supporting ethical seed vendors. In fact I will feel a little bad about multiplying Frank Morton's neat new lettuce varieties and giving them away. But by then, they won;t be "new"!
If they ALL started going out of business, and hobbyists were the ONLY ones preserving heirloom varieties "clean", I might start doing the bags and THEN telling trade partners I jumped through hoops "to keep it clean".
The easiest way to mark your seeds "not totally isolated" is to just say "
ex 'Oranje van Goeijenbier'" which means:
"these seeds came FROM a seed parent that was 'Oranje van Goeijenbier', but I make NO claim about who the father was".
("Luuuuke! I aaam your FATHER!")
But I concede: if my goal was to multiply my seeds
every year for 20 years, and wind up with a strain at least as genetically uniform as what I started with, and no discernible genetic drift, I would fool around with bags and numbers of parents and painfully precise selection procedures and something to protect against local-climate-drift.
But if I just want to have more seeds for the next few years without buying again, or sharing with other hobbyists, I wouldn't bag tomatoes or lettuce. But then i would not sell them commercially or donate them to high-tech seed bank.
Whether or not I would give them to a local "seed lending library" depends on the library's policy about isolation.
Before I read
@Joseph's articles about modern mini-landraces, and then looked into sources of genetic drift, I had more interest in "pure varieties", and still have some twitches of interest. But if it's a full-time job to prevent ALL genetic drift, maybe i would RATHER that my snow peas and snap peas are better adapted to my climate, and slightly different from an out-of-the-box generic OP commercial variety that works "pretty well for any climate".
This may be the first time on any gardening site that I've ever advocated the LESS gadget-y, LESS tinkering approach.