So what happens when you self-pollinate?

First of all, homozygosity increases. That's how a geneticist sees it. The more a species is inbred (i.e., the more self-pollinated), the more genes at which the pairs of alleles are the same. Therefore, the more homozygous the overall population of genes is, and the higher the level of homozygosity. You can turn the concept over, too, and state it this way: HETEROZYGOSITY DECREASES Self-pollination or inbreeding leads to an increase in homozygosity, and a decrease in heterozygosity. It's actually two ways of looking at the same thing. What happens when you increase homozygosity? UNIFORMITY INCREASES Characters become fixed (uniformity increases), because variation decreases (and variation in a population results in part from the sorting out of heterozygosity).

Look at it this way. Remember Mendel's F2? The F2 had the highest amount of variability, because all of the F1 heterozygosity got transferred from the F1 sperm and egg cells into the F2 plants. If every F1 plant has the highest possible amount of heterozygosity, when those plants make pollen and eggs, every pollen cell will be as different as possible from every other pollen cell, and likewise, each egg cell. When the pollen and egg are combined in every possible random combination, the F2 results, and shows the results of those random combinations. If you took individual plants out of the F2 (remember, every plant in an F2 population is different) and selfed them, what would you get? Each F2 plant would produce a group of seeds which we would call an F3 family. In an F3, there is more homozygosity than in an F2, and therefore, more uniformity. The characters that were evident in the individual F2 plant become fixed (or more uniform) in the F3 family. If you continue this logical progression, in each selfed generation (F4, F5, F6, etc.), the characters become more and more uniform within each family, and more and more different between families.


Let's review the concept of inbreeding. Characters become fixed during inbreeding. What is happening at a genetic level? Homozygosity increases. What does this mean? More and more gene pairs become homozygous. If a character is caused by a recessive allele, that character will be exposed during inbreeding. If this is a desirable character, inbreeding is therefore a positive event. But recessive characters can be negative events, too, and in outbreeding species, the recessive alleles are frequently negative, even lethal. What happens during inbreeding is that the recessive lethal combinations occur, and the plant dies, sometimes as early as the embryo within the seed.


Inbred lines do eventually stabilize. Why? Because the lethal recessive combinations die off, leaving only those homozygotes which are strong enough to survive. Inbreds of typically OB species tend to be very weak, and would probably never survive in Nature.


The negative effects of inbreeding are referred to as inbreeding depression. The classic example is in maize. For the first 8 to 10 generation of inbreeding, the height of maize inbreds decreases with each cycle. Then, the height levels off, and the inbreds (although not as vigorous as the original plant) stabilize. In OB species, inbreds tend to be weak, are often poor pollen producers, and are frequently poor seed parents. In IB species, inbreds are generally healthy, vigorous plants.


This phenomenon reverses when inbreds are crossed. The F1 between two highly inbred lines of an OB species tends to be extremely vigorous, in addition to being extremely uniform. Again, using maize as an example, there is a 2X or more increase in yield when the F1 is compared to its inbred parents. The ears of corn are larger, heavier, bear more seeds, and the plants produce more leaves, more forage, and heavier stalks. This increase in overall plant size, yield, and health is known as hybrid vigor. In contrast, there is little hybrid vigor observed in F1s between inbreds of an IB species. However, even in this case, the F1 is very very uniform, because every individual in the F1 has the same genetic composition (i.e. the same genotype). In hybrid maize breeding, the lack of vigor in the inbred parents can be a serious problem. The marketable product is hybrid seed. If the inbred seed parent is weak, it produces little seed, which makes the hybrid seed very very expensive. The solution is to make the seed which goes to market a three-way or a four-way cross. In a three-way cross, an F1 hybrid is used as a seed parent, and an inbred is used as a pollen parent. The breeding company gets a strong, vigorous seed parent, and the seed sold to the farmer still shows a lot of hybrid vigor. In a four-way hybrid, F1s are used as both seed and pollen parents, because inbreeding depression has a negative effect on pollen production, too. With both seed and pollen parents being vigorous, maximum seed yields occur. In general, however, true F1s will outperform both three-way and four-way crosses in a crop like maize.


In an OB species, the preferred state is heterozygosity. Why? It may give the plant or the species a greater adaptability to environmental change. In an OB species, the preference for heterozygosity is extremely strong. There are recorded examples where a species would not tolerate more than one or two generations of inbreeding, without a tremendous loss of vigor. If you are trying to save seed by inbreeding, you need to be aware of this phenomenon.


Three reasons. First, because of hybrid vigor. Realize that this will occur mostly in OB species. Two: For uniformity. Every plant in an F1 is identical (for practical purposes: there can be some genetic variation in the inbreds that gets multiplied when the inbreds are crossed). This uniformity can be advantageous when you are trying to harvest a field at one time, by a machine. But it may not be an advantage to a home gardener. After all, how many bushels of tomatoes can you use all at one time? Three: Because of the availability of certain gene combinations that are only present in a commercial F1. Because of the price of F1 seed, breeding companies develop F1s whenever they can. This means that many of the most desirable combinations (VFFNTT-resistant tomatoes, for example) are only available as F1 hybrids. Yes, you could theoretically self down that F1 and recover an open-pollinated line that was also VFFNTT-resistant. But some of those combinations are likely to be heterozyous, which means that (due to independent assortment), not all of the selfed seed would be VFFNTT! How could you sort them out? A home gardener could not, and even a professional breeder would have a difficult time.


The key to the producing F1 hybrids is controlled pollination. To make an F1 hybrid, you need to develop at least two inbred lines, one for the seed parent, one for the pollen parent. In practice, you make many more than two, make all possible combinations, and choose the best performing hybrid combinations. Inbreds are made through self-pollination, which can happen naturally (most garden legumes, lettuce, tomatoes, impatiens); by hand (cucurbits, pansies, petunias); or not at all (most Brassicas). The actual cross that produces the seed from which you grow an F1 plant is often done by hand (many vegetables and flowers); by wind (maize --- take a trip through the Midwest when they are detasseling --- the process is pretty amazing [no pun intended]); or by insects (carrots, onions, some Brassicas). Using insects requires caged or screenhouses, and raising pollinators in isolation, but it's do-able. I watched hybrid ageratum production once: the workers used modified Dustbusters to collect the pollen from greenhouse grown pollen parents, and just dusted the vacuum-collected pollen onto the seed parent flowers with a brush. (I once tried the same technique with astilbe, and got very poor pollen collection). That's the F1. Seed produced by an intentional cross on an inbred seed parent will produce F1 plants.


Next, you need to make sure that there is no contaminating pollen around. Plants can be pretty promiscuous: wind blows pollen for miles, and insects love to interfere with your work as a pollinator. Maybe they are jealous. Maybe they simply see you as a big competitor. Maybe they are just hungry. If you are working with a perfect (and self-compatible) flowering species, the protection process starts with emasculation. You must remove the male portion of the flower (the anthers) before they shed pollen. Typically, this occurs before the flower opens, and this means emasculation must occur when the flower is an unopened bud. How early? It depends on the species, because the important event is pollen shedding, or anthesis. Typically, if anthesis occurs with flower opening (and this is common), emasculation occurs the day before the flower opens. How do you know when the flower will open? By observing over a period of days. Typically, the flower opens the day following the appearance of petal color in the flower bud. To emasculate, you carefully remove the petals of the unopened bud, then gently pull off the anthers (tweezers are good). You then cover the emasculated bud with a paper envelope or cloth bag to keep out insects and to protect from stray windborne pollen. Protandry is the most typical condition, and pollen is shed before the stigma is receptive. If this is the case, and the pollen is known not to germinate until the stigma is receptive, contaminating pollen can be washed away in some species. Lettuce is a good example. Emasculation is tedious and difficult. Therefore, a lettuce breeder waits until the flower opens, washes away the self-pollen with a squeeze bottle or squirt gun), allows the flower to dry. The next day (and for the next few days), cross-pollen is carefully applied to the now receptive stigma, and the cross is made.


This assumes that there is always fresh pollen available when the stigma becomes receptive. Such convenience may not exist. Consider a cross between an early variety, and a late variety. The early-bloomer may not be in flower when the late one begins. If you cannot control the timing of the flowering, you may need to learn to store pollen for an extended period of time. The rule of thumb is: binucleate pollen can be stored cold and dry (in a refrigerator, within a dessicator, over activated silica gel) for a long time, while trinucleate pollen cannot be stored for much more than a day. In reality, it is all trial and error. Tomato pollen can be stored for years; maize pollen can be stored for a day.


There is usually a structural change at the point when a stigma becomes receptive. If the stigma is split into two parts (forked like a snake's tongue), the stigma at anthesis may be tightly together, but open and spread when receptive. In some species, the stigma is "up" when non-receptive, and "down" when receptive. "Up" would usually mean "out of the pollinator path", whereas "down" would indicate "in the path of the pollinator". Think about female maize flowers (the silks). Receptive silks are green and hanging down in the breeze. Early silks are not extended; post-pollination silks are brown and dry. Another clue about a flower is that the swelling of the ovary in a fruit occurs quite rapidly after pollination in many species. If the ovary is swollen in relationship to the style and stigma, pollination may have already occurred!

Return to:  Genetics for Seed Savers Index

Go to:  Chapter 5:  Genetics for Seed Savers FAQ