Carol Deppe's book on breeding vegetables has several chapters on seed saving. Partly, the number of plants to save seed from will vary quite a bit depending on how seriously you take it. A "professional germplasm curator" would apparently want to save from a minimum of 100 inbreeding plants or 200 outbreeding plants, or even more if the variety is very heterozygous. A serious amatuer might want at least 20 and 40.

Of course, some people just save from one plant, generation after generation, which might work well for the casual home gardener. One problem is that an accidental cross could wipe out the whole line.

You say, "I understand that seeds of the choicest plants should be selected for saving. For example, if large size of fruit is desirable, seeds of the largest fruit should be saved." I've read the caution somewhere that if there is an accidental cross, then the contaminated plants could exhibit hybrid vigor, so selecting the most vigorous plants with the biggest fruit could have the effect of causing you to discard the pure old variety.

You continue, "In doing so is it possible to deteriorate another attribute of the strain, say flavor, if seeds are saved only from the one or two plants having the largest fruit?" Some characteristics are linked (such as when their genes occur very near each other on the chromosome) so selecting for one could inadvertently cause the selection for (or against) the other. On average, however, that is why you save from many different plants, so you don't lose anything by accident.

There's a lot more to say on this topic.

Well, Jim, as MisterCross notes, the answer could have a lot to do with how serious you are.

The answer is also controversial. Various experts have different responses to how many plants should contribute to the seed stock, with 10-15 about average for most vegetable plants.

In theory, in-breeding self-pollinators (i.e., tomatoes)only need one fruit. But here, again, as Jimster points out, accidental crosses or mutations can have effects if you do that. Personally, I always recommend that seed be saved from at least five such plants.

For the just-folks gardener, some insights can be gained from the SSE Guidelines. Which are:

"Seeds: All seed requests should be filled with at least 25 seeds. Requests for corns, which require larger plantings in order to maintain sufficient genetic diversity within a populations, should be sent a minimum of 200 seeds. Some outcrossing species---Brassicas (cabbage family), seed-propagated Alliums (onion family) and Umbelliferae (carrot, celery, dill, parsnip)--require a population of 30 or more plants to maintain viability. Requests for these species should be filled with at least 60 seeds to allow for low germination and losses during planting."

So, we're looking at from 12-100 plants _minimum_ depending on genus.

How does the average gardener accomplish this? One way is to play the recombination game. Get together with five or six other seed savers who are willing to pay attention to purity issues. Divide the seed among them. Let's say there are six of you. That means each only has to grow five cabbages, for instance.

When seed is produced, you combine all of it from all six growers. Then redivide it. This assures that each member of the group now has seed that, in total, contains the entire genetic diversity of the population.

Many thanks, MisterCross and GardenLad. Those are two very thoughtful and well written replies. I received more information than expected. I will read them over several times to try to understand the implications for my gardening.

Looks like it's time to get a book or two. As MisterCross says, "There's a lot more to say on this topic".

Jim

Since my original posting I have gotten Carol Deppe's book and read it cover to cover. Wow! There is a lot to know about seed saving and preservation of varieties. Fortunately, her book, although challenging at times, is well enough written so that there is no ambiguity to the information. If you stick with it, everything can be understood. Good stuff. Everything I wanted to know.

She refers a number of times to Susan Ashworth's book, so I eventually will get that one too, since it undoubtedly has a different emphasis.

Jim

IM glad you got through that book! Im a 8th year college student but more humanities than sciences... phew. I couldnt do it. Im sure its interesting if you can understand it!

some one recomended to me with bean seeds to buy the smallest pkg of the same kind and mix it in to my own home saved seed every couple of years. It may help to buy them from a different sourse.

Garnetmoth,

Carol Deppe's book is quite challenging in parts, namely the parts on genetics, which are quite complex. Much of the book is easier going, however, and contains lots of illustrative stories about amateur vegetable breeding projects. She also suggests ways to simplify projects which might otherwise exceed the amateur's resources. Nevertheless, I am finding I need to reread the book because there is too much information to be retained in one or two readings.

Jim

Jimster: Im glad im not completely inept ;-)
Im engaged to a biologist, and he just wants tomatoes. and more tomatoes. and cherry tomatoes. so.... im up for trying to keep varieties pure and true, and saving sports if i ever come across them. Ill keep breeding to those with more room! or when I get tired of regular gardening ;-)

I keep some mixed beans because that's the way I recieved them, and it makes it easy to keep a lot of genes without having to keep them all separated and pure. I have never had much in the way of diverse tomato strains, mostly the tomatoes seem to be line bred. When I do get variation, it's usually a cross from another variety and just messes up the characteristics I want the tomato for. (I would like to do some crosses on purpose, to get the good flavor of Green Grape into my most dis res tomato! Green Grape gets the worst Late Blight of anything I've got)
However, as I understand it, anything in the Brassica family needs a lot of individuals to keep the seed from "running out". This implies some diversity in any strain. I keep Long Standing kale, which was dropped by SESE about 15 years ago. Three of us here keep it, and now and then I get some seeds from one of them, in fact one woman and I are planning to swap some seeds of it. I try to make sure I have at least 25, more like 50 plants to make seed. This is no real hardship, as I have a big garden. Not only does Long Standing provide winter greens, but we eat the flowering tops in salads and cooked like broccoli. The plant is edible for about 8 months of the year! What I have seen about other kales is that they bloom later, after the summer heat gets going, and the flowering stems tend to rot rather than set seed (and they don't amke good eating that way either). I see the same effect in mustard too, and would like to find a var that is resistant to it, as mustard wont' cross with my kale. I do have to watch out if I have any collards or cabbage left in the garden in spring. I don't grow broccoli, because by the time it would come in, we've eaten so many kale tops we don't want anything like broccoli. Brussels sprouts and kohlrabi would be the other possible crosses. Oh, and something to keep in mind with your turnips is that they will cross to chinese cabbage. Yes, Bok Choi and Napa cabbages are just glorified turnip greens. I like to let them go to seed, as the crosses to 7 tops turnips make extremely nice greens that are MUCH more resistant to bugs, disease, and cold than the pure choys and napa.