A little bit of this is going to be technical, the smallest smidgen of genetics that I can get away with so that you can understand recessive deleterious traits; something important when saving numbers of plants and so you can understand also the nature of the traditional genetic variations in homozygous vs heterozygous.

Frank: Huh?

John: Gotta use your glossary if you're gonna be like that! $5 words that are accessible to anyone. Here it is: Selfers vs crossers. I'm trying to give a framework to distinguish differences between these two so that you understand why you can save relatively small populations in self-pollinated species and get away with it . (though I would advocate larger as better when possible) And why you can't get away with it in the long run in crossers, with minor exceptions. Once you get a grasp of this you become one of the initiated and watch from then on the kind of populations in crossers that you save.

Why is it important? It determines how you manage the variety and the population, how you both manage and select it, the numbers you use to manage that population, and how you treat it in general. The minimum number of plants you can grow and preserve genetic diversity, how much selection you apply. When you talk about selecting in a crosser, it is different from selection in a selfer. We have so many lousy ops in crossers because people have treated crossers like selfers.

F: Don't do that.

J: One of my missions in life is to train others how not to do that.

How do you derive new selections from a population? Again, selfers and crossers are different. I think the hardest thing of all in plant breeding is breeding a new op variety in cross-pollinated species. Almost as hard as creating a new one is maintaining a good one. It is hard work and most plant breeders worth their salt know that but very few plant breeders know how to maintain good ops any more. That's why there's some lousy crosser ops.

Population structures - thoughts on the differences between crossers and selfers.

Selfers' flowers remain closed, they are really evolved to tolerate inbreeding. It makes sense. If the flowers are closed and they're selfing all the time that is inbreeding. Inbreeding means the mating of individuals more closely related than mates chosen at random from a population. That means 'don't marry your cousin.' If you marry your cousin bad things might happen. Here's the difference between inbreeding and inbreeding depression: In 1909 E.M. East said 'inbreeding per se is not necessarily bad.' Inbreeding can lead to the revealing of deleterious traits and then become inbreeding depression. There are many famous examples of royal families that did inbreeding where deleterious things didn't necessarily happen. Certainly in animal husbandry people have mated sires to their mothers, etc. Inbreeding means the physical act of making closer than random outcrosses in any kind of cross situation and inbreeding depression means the deleterious manifestations thereof which may or may not occur. Selfers have evolved to tolerate inbreeding. That's why selfers are able to self successfully. Selfers can be selected and maintained narrowly with little or no resultant inbreeding depression. I think it is important to not maintain them too narrowly. It is good to have what are called multi-lines. But people get away with it. There are a lot of self-pollinated crops like lettuce, tomatoes, peas and beans where the plant breeding process selects down to one plant from which they make a whole new variety.

Crossers have floral mechanisms to promote outcrossing. The floral mechanisms have evolved to avoid inbreeding to avoid the deleterious traits that come up in selfers. Inbreeding depression is a real threat to fitness and varietal integrity. What is fitness? Fitness is the fitness to reproduce, which is what all is measured by in the biological world. Deleterious recessive genes can be revealed. Most deleterious genes are recessive. Why? If they were dominant they'd be ruled out immediately. So they hide in the shadows as recessives. You suffer loss of vigor, reduction in yield over-all, and very important, the first thing to go with inbreeding depression is loss of fecundity. There's a good $5 word! Fecundity is potential fertility.

Q: Why are deleterious traits more likely to be revealed in selfers?
A: Because selfers are always selfing, they get revealed constantly and kicked out by natural selection.

Most seed-propagated crops are diploid. Diploid means two sets of chromosomes. We eat a lot of polyploids-more than two sets of chromosomes, but almost all are asexually propagated. Potatoes have four sets, oats have eight sets, bananas are triploid. In plant breeding we work mostly with diploids. We can understand it with human examples because humans are diploid. Chromosomes are real things that are in your body. Each chromosome is very long with thousands of genes. We have 23 pairs of chromosomes, numbered 1-22 and the X and the Y chromosome. Except for the red blood cells, every cell in your body has a nucleus and has all of these chromosomes. When the chromosomes in humans are fully uncoiled and stretched out and coding for DNA that's being read by ribosomes, they are 6 feet long. Chromosomes coil and shorten to divide. They suddenly go from being 6 feet long to being fractions of microns long to make sex cells. Each divides and pairs in mieosis. Each of the 46 chromosomes is unique. You have 23 chromosomes from each of your Mom and Dad and the #6 chromosome from your Mom looks just like the #6 chromosome from your Dad. The true mystery of life is when the chromosomes go to make sex cells the two coil and find each other (#6 from your Mom and #6 from your Dad) even in this bowl of spaghetti with 46 and they line up in the embrace of life-all 23 pairs find each other. Sometimes its not exactly equal and that's when hereditary diseases like Trisomy 21 occur, which means three #21 chromosomes (that's Down's Syndrome). It's amazing how frequently they all line up perfectly and how infrequently they don;'t.

The purpose of this is to understand why selfers don't get inbreeding depression and why crossers do and why you treat crossers the way you do. Genes come in pairs in a diploid. People say 'oh I've got the gene for blue eyes. Well if you have the gene for blue eyes then you really have one for blue eyes from your Mom and one for blues eyes from your Dad. How many people here have blue eyes but one of your parents has brown eyes? What gene for eye color did that parent who has brown eyes have? What do you know about that parent, guaranteed, because you have blue eyes? That they had a big B for brown and they had a little b for blue and nobody knew about it until they procreated and you came out a blue-eyed child. That is a form of genetic testing. You are a test victim and we found out that they are heterozygous, big B for brown, little b for blue, and it was right on one spot on the proper chromosome. It is called the gene for eye color but there are two different alleles, there's the blue-eyed allele and the brown allele. Now this is a very important point for a lot of things: Does that mean that there are only two alleles for eye color? No! How many people know about sweet corn and the gene that confers the sugariness in sweet corn? There's the traditional (you can read a nice description in the Johnny's catalog) su, there's the se and there's the sh2. Those are the three that we most commonly talk about. Those are all alleles for sweetness at one gene. Some crops have sixty different alleles for self-incompatibility within a good op population. Fruit color in tomatoes: you can have orange tomatoes that are beta-carotene, orange tomatoes that are delta-carotene, there's a tangerine gene, those are all different versions of a gene, but they are the same genes.

When you get two copies and they're the same, if you're blue-eyed, blue-eyed is the most recessive point, and so you have two copies of blue, you are then homozygous for that gene for that one trait. We talk about that plant is homozygous for that trait. That person is homozygous for blue eye color or that person is homozygous for eye color. Now if you get a copy of brown from one parent and a copy of blue from the other parent, then you are heterozygous, and most importantly, big B brown is dominant to little b blue and this was the big thing that Gregor Mendel found out.

I'm going to talk about biotypes. Most people when they think about self-pollinated species think about what they are getting from a commercial seed source. When you get a new lettuce variety from Royal Sluis, they have selected one plant, one plant only. They say 'this is it', and as they say in the seed business, 'we'll blow it up from that one plant' That thing has already been self-pollinated for several generations and it is homozygous for almost all of the genes-not completely. It goes to homozygosity for a selfer but it takes at least seven generations to get all the traits the same and there are still some traits that you're not selecting at all that are still heterozygotes and that is very important to remember. But for all of the important traits that you can see, the phenotype of the plant (the visual appearance of the plant: what you see is what you get) are homozygous on any one chromosome. Often because there are more than two copies (alleles) of any one gene, they often give you A1, A2, A3. In self-incompatibility they'll give you 30 versions, they are all just variations on that one gene. There are other ways that genes can be manifest post-DNA. A gene sequence makes a protein which confers a trait or part of a trait. There are two copies of everything and more than two varieties of any one copy of that gene which makes life interesting, that's what makes variation and sometimes the recessive one can hide.

When you first cross two different plants if they're diverse enough they're fully heterozygous. Then you take a plant and you try to self it, every generation that you self a plant, it becomes 50% more homozygous. Then you go to 75%, 87.5%, 93.75% etc. More uniform. By the 7th generation you are over 99% homozygous.

A friend of mine who is a corn breeder at one of the universities told me: (university breeders in the classical sense have always bred things to be able to distribute freely to people who can put them to good use. He was one of those ilk) He would breed sweet corn populations and distribute them to sweet corn breeders. When he got good populations that he liked enough-he's just doing the bare-bones beginning breeding getting something kind of good that's got a bunch of traits in it­­he would start to make inbred lines out of that and when he found the ones that he really liked he would get requests from different people around the country and he told me that 'if it was somebody that I really liked and they were really good and cooperative with me and they helped me out in my program, I would give them F4s and F5s. Somebody for whom I didn't care that much I would give F7s. The moral to that story is that the F7 is more homozygous, more locked in, more set up, there's not as much variation still coming out, while an F4 or an F5 that's really looking good, has some good stuff there that you can take several different ways, you still have enough heterozygotes to give you some breathing room. The more generations you self it the more it gets locked in.

Q: Alan Kapular used to offer several things in the F3 and F4 generations. He was fairly cavalier about them, describing them as 'fairly stable.' Had he any justification for calling something F3 or F4 'fairly stable?'
A: Not really, There's no magic except purely the numbers for statistical basis. He was probably saying he'd done some selection for this melon, they were probably all oblong. I'm giving you something now you can play with, select from that.' I don't know if I'd call it fairly stable at that point.
F: He means that he's thrown out the worst of it. The diversity remaining in there is going in the right direction, it tastes pretty good, its about the right shape. Definitions of stable vary.
George Moriarty (Cornell U.): It depends on your starting point. If you're crossing a bush dark zucchini and you have another bush dark zucchini over here that has CMV resistance, they're both dark zucchinis, you cross them, select for the virus resistance, at the F4, they're fairly uniform.
Will: What they're coming from wasn't greatly variable to start with.

Beans are a self-pollinated species but there's some variation here, C1, C2. That particular gene C is heterozygous but most of them are homozygous and what happens in beans was shown in a very elegant experiment 100 years ago by scientist Wilhelm Johannsen in Sweden went to the market, bought a bag of black beans, and separated them into different size categories by weight. He started to grow out those as different groups, and he found that there would be distinctly different size beans in each group. After a few generations he had the biggies, not so bigs and smalls, he had 7-9 categories of size and they pretty much always fell into their proper category of size. The size they eventually became didn't always correspond to the size they were in the beginning. He bought at the market what is called a multi-line, with a lot of genetic diversity for seed size, shape, weight, and by selfing them he got them into distinct biotypes. When you buy a bag of garden bean seed from a seed company today of a new variety that Asgrow has just released, if you were to run that experiment you wouldn't get that result because that's a pure line that all came from one plant, you're getting that one biotype. When you get an older type you're getting a multi-line. If you grow the same multi-line enough years even in a selfer if you don't exert much pressure it will segregate out into a multi-line, there's genetic diversity that will come out, there's an occasional cross that'll make things interesting and even things that were pure lines 30 or 40 or 50 years ago will eventually become multi-lines again. Because selfers usually only self, they carry down through the generations like that in that bag in that mix. You pick any one bean out of there, you're going to get that biotype. Have you ever seen Yellow Eyes that are a lot more blotchy, one than another? If you started to select for levels of blotchiness in a good diverse Yellow Eye population, you could make five new Yellow Eye varieties out of that because those are different biotypes to some degree within that population. Through the generations selfers stay the same. The evolutionary mechanism is conserving: when they latch on to genetic traits that are good and have good fitness in the environment, the selfing is a self-preservation of that combination of traits, and any deleterious trait that arises, whether a mutation or from an outcross or wherever it came from, will be revealed and kicked out.

There's a little inherent variation because we have all the different levels of blotchiness. In a selfer that's preserved because that's an evolutionary mechanism, but there are occasional matings, and under stressful conditions they intermate more to come up with new combinations. This is what I call very conservative: We don't have any matings unless its justified by class and privilege. Occasionally these new combinations set up something new and evolutionary. I like to call it an evolutionary trial. These new combinations may be able to withstand stress better. NOVA had a show years ago-where they showed a big bowl of beans with all sorts of color variations and they said, 'these are the kind of beans that they grow in this certain area of the Sudan.' When there's a drought 40% of these beans have enough drought tolerance to make it to the next generation and when it's a really wet year, monsoon, another 50% make it through and the drought-resistant ones don't do as well. This is one of the classic genetic variation stories. What I thought at the time, when you get a real drought year with only 40% living through and the other 60% dying from extreme drought, what did you lose in that 60% that was really valuable in a monsoon year? How the heck does that make sense? In that basket of beans that they are showing you, 40% are the drought tolerant, the homozygous biotype that's drought tolerant, but there's a percentage in that basket of heterozygotes where there were outcrossers between a monsoon-tolerant and a drought-tolerant biotype in that same basket. Every generation there's a certain number of outcrossers and this is nature's way of sneaking through some drought tolerance genes in a year when there's a monsoon and that's how you preserve that genetic diversity.

There's more than monsoon tolerance and drought tolerance. Some have stronger stems in windstorms, some are shorter. There are vining lines in the bowl, and especially if you get into landrace varieties of these self-pollinated species or what I call farmer varieties, this is rampant. They always talk about all the different forms in the native rices of Southeast Asia. What happens because of this? You don't ever get 100% yield but you get a buffering between all the biotypes for the environment. This is the pioneer species to the nth degree where you have all of this buffering against different environmental stresses. If we want to have buffering in our self-pollinated species to go down through the generations, we do not want pure lines. I am now convinced of it. You're going to have to baby pure lines and the year that is bad for that pure line you're going to lose the whole shooting match whereas if you had a multi-line, you'd get 40% yield. When you get a good year and everything is right you might not get as much yield in an agronomic crop, you're not going to get the exact prettiness in every head of lettuce, nevertheless that's where I'd put my eggs when growing under any kind of environmental stress. The different biotypes buffer the environment.

Crossers are all about population buffering. This is where the population buffers itself. Most individuals in a cross-pollinated species don't have the same #6 chromosome. Mom and Dad didn't have the same #6 chromosome because they were outcrossed every generation so they always have tons of heterozygosity in a natural cross-pollinating situation. They're always picking up a gene that's a little different from my Mom and a gene that's a little different from my Dad. In a good op squash population, some plants in the population, if it is diverse enough, have more drought tolerance than others, some plants are flowering a week to ten days earlier than others, some will do better against damping off in the soil, some will be more heat-resistant. Regions share a higher percentage of genes, its more like a cousin-cousin marriage instead of being pure outcrossers, but every generation that you put them out there they make crosses. One with good eating quality may cross way over here with something that's genetically unlike because a bee just carried the pollen with a late flowering plant or a heat-resistant one. There's nothing pure line down through time about this. There's nothing stable about it: it's bumping and shaking and moving down through the generations. The selfer situation is much more stable where it has the ability to keep good drought tolerance. Here with the crossers its much more elusive. There are selfs in this population, if its squash they'll self naturally in the field, there are cousins and there are wide crosses between the different varieties, so crossing predominates and that's what I call free love, it really is the sex orgy of the natural world, good amalgamations are fleeting.

This was a good amalgamation of genes for drought tolerance in this bean and because of the ability to self that amalgamation was held to go down through the generations. Crossers don't have that if you keep them diverse. That's where being able to select a crosser so that it is uniform for fruit type and flowering and drought resistance and still have enough genetic diversity within is hard. Fortunately, in some crossers like squash you can select a little more narrowly than you can in beets or carrots. If you try to select beets and carrots narrowly you get in trouble really fast.

Frank: What would be the consequences if I just went in there and selected for late flowering?

John: If you just selected for late floweringwhy in the hell would you ever want late flowering?
Frank: Okay, if you just selected for eating quality and that's all
John: If you just selected for eating quality and that's all, you'd lose everything. You'd make this grow into a whole new ameoba, it's like cutting off the head. You'd get good eating quality squash for a while-I've seen people do this in carrots-but you'd get less and less seed yield, that's the first thing to go, the biggest problem in inbreeding depression is fecundity. The first thing to go is the reproductive quality. Pretty soon you're not producing as much pollen, you're not producing as much seed, the seed is late, it might start flowering early, you might get damping off, you'll get stuff that's not as vigorous coming out of the ground, it's not going to grow as big a plant. A lot of the really good tasting squash, the old 'Buttercup,' the standard by which we compare everything, has got inbreeding depression. Its had it ever since Dr. Yeager pulled it out of his plot in North Dakota in 1925. That was a narrowed-down one-plant selection. He got away with it but that doesn't make very many leaves, not what I want for squash. It doesn't make that many fruit per plant. We've held on to it. As my old boss Larry Satterly used to say, 'Sometimes you get lucky.' Yeager got very lucky with that. You shouldn't discount those. You shouldn't say 'I can never do this because John said don't do it. Try it, maybe you'll get lucky.'