On most basic seed saving matters, I would defer to Seeds of Diversity Canada’s excellent book, How to Save Your Own Seeds, which you can buy from them or from us. They give a good overview, simple but accurate, of how plants reproduce and what you need to do to save seed from many of the common plants you are likely to have in your garden. They also give recommendations on the isolation distances and population sizes you should have to maintain that balance between keeping a variety pure and uncrossed with others, and maintaining a healthy amount of variation between individuals within it to keep it vigorous and healthy. I will try to not repeat what they have already said.
However, that book is focused on vegetables, and as a result it does not give some of the staple crops as much attention as might be helpful for those new to them. So I will try to provide a little more detail on staples, and also describe some special tricks which are useful for seed savers.
One feature of many staple crops is that the food is the seed, and the seed is the food. This is important because it means that if you grow the food, you usually have already grown seed as well. For many people throughout history, this interchangeability between food and seed has been vital, not only for their survival, but also for their ability to acquire and adapt new foods into their cultures and cuisines. Countless people have just squirreled away some of the food they had and planted that to get their next crop. You can too.
However, if you want your crops to be as dependable, healthy, and productive as they can be, you will need a little more understanding of how they work, and a little more attentiveness when growing and processing your crop.
First, it is important to understand some of the mechanics of how plants reproduce. So here’s a little butterflies-and-bees lesson. I will simplify it to say that plant flowers have male and female parts; the male parts (called the anthers) produce pollen, which is carried to the female part of the plant (the stigma), and travels down to fertilize the ovary, forming one seed. Each seed in a fruit – whether a bean pod or a corn cob – needs its own grain of pollen. So far, this system is similar in all seed-bearing plants; but here begin a number of variations:
1) Some species of plant have perfect flowers; that is, each flower has both male and female parts. Others have imperfect flowers, where each flower is either male or female.
2) Most plants are monoecious; that is, one plant has both male and female parts (whether in the same flowers or separate); others are dioecious – the whole plant is either male or female.
3) Among monoecious plants, some are self-fertile; that is, they can pollinate themselves. Others are self-incompatible; they reject their own pollen, and will only produce viable seed if they receive pollen from another plant.
All these variations mean that different species of plant are pollinated in different ways. Some tend to self-pollinate; others tend to (or have to, if self-incompatible) cross-pollinate, usually either with the help of insects or the wind. Some self-pollinators have quite closed, less-ornate flowers, to discourage or at least not attract the pollinators they don’t need. Consequently, they need only small isolation distances to keep them pure. Wind or insect-pollinated plants, by contrast, need to be isolated by much larger distances, or you need to take special measures to keep them pure. They often also need crossing to remain vigorous over multiple generations, and so you need to have larger numbers of plants if you want to save seed.
So let’s take all that theory and start examining specific cases.
Legumes (Fabaceae family)
Legumes have perfect flowers. All of them are capable of self-pollination. Some have flowers which usually pollinate themselves before opening, and so only need a small isolation distance; common beans, peas, chickpeas, soy beans, and peanuts usually self-pollinate. Others have more open flowers, which let insects pollinate them, and need larger isolation distances; these include runner beans, favas, and cowpeas. Whether self- or insect-pollinated, dry legumes conveniently combine the seed and the food, so seed food harvesting is identical. Because of their tendency to self-pollinate, you do not need to save seed from many plants; 20 is the recommended number, but there are plenty of historical cases of people preserving a bean variety from a single plant.
Grains and Corn (Poaceae family)
Grains also conveniently combine food and seed crops. Barley, oats, and wheat are all self-pollinated and need only a couple of metres of isolation between varieties; rye and corn are wind-pollinated, and need a lot more space. There are few named rye varieties, so that is not much of a concern; but corn is a much bigger problem, particularly for those who want to grow and save their own corn seed in corn-growing areas, where a huge amount of corn pollen can be in the air at certain times.
Knowing a bit about corn’s reproduction is helpful to keeping it pure. Corn is monoecious with imperfect flowers; the male part of the plant is the tassel at the top, which has a number of branches containing tiny flowers which produce loads of yellow pollen, which is easily observed if you bump a corn plant when the pollen is ready to go. The female flower is the silk on the ear, with one silk connecting to each kernel to carry the pollen to it. The tassel will emerge a few days before the silks. A corn plant can pollinate itself, but because of this arrangement it tends to be pollinated by its neighbours. In fact, if you save seed from corn which has not crossed with lots of other plants each generation, you will notice it start to suffer from what is called “inbreeding depression,” a phenomenon which is not understood well but which is widely recognized; the plants will be smaller, yield less, and be less healthy. To prevent this, you should save seed from a minimum of 100-200 plants (the larger number is more widely recommended).
To try to reduce the risk of our corn being pollinated by others’ corn, we follow two strategies. Some of the corn varieties we grow are early-season; this means that they finish pollinating before our neighbours’ corn starts, effectively keeping them pure. This is risky, since the dates when corns flower vary depending on conditions; as our corn tassels, I keep nervously watching our neighbours’ corn fields and hoping they stay behind us; so far, they have. Full-season corns pollinate at the same time, so there is a high risk of contamination. Our current strategy is trying corn varieties with colours which will show crosses. White is the easiest: commercial corn here is yellow, and crosses of yellow corn pollen onto white ears will consistently show up as yellow kernels, which we eat and remove from the gene pool. Blue corn is a bit of a riskier proposition for this strategy: my experience so far is that I can effectively keep it pure by removing all yellow or greenish kernels from my seed supply, but I cannot guarantee this. However, I consider it worth the risk because I love the blue corn we grow and don’t want to abandon it just because it’s blue. As far as I know, red, yellow, orange, brown, black, and purple corns will all hide yellow crosses, and so we only grow them as early corns.
It is also possible to bag corn tassels and silks and hand-pollinate them. It requires a bit of specialized equipment and patience, but isn’t that difficult. Seed Savers Exchange offers a handout explaining how to hand-pollinate if you want to.
Squash (Cucurbitaceae family)
Squash have imperfect flowers and are normally insect-pollinated; however, they are self-compatible, and so don’t need the large numbers of plants which outcrossers need to avoid inbreeding depression. Six to 20 plants should be fine to maintain your seed. Thanks to the insects, you have to isolate different squash varieties of the same species from each other by quite a distance. However, you can grow one squash variety of each species beside each other, and they won’t cross. So pay attention to species when choosing squash.
If, like me, you just can’t limit yourself to only one variety per species, it is also quite possible to hand-pollinate them. Many sources have directions on how to do this; we have an article giving step-by-step directions for the method we use.
Tricks
General Advice to Decrease Cross-pollination
If you follow the recommended isolation distances you shouldn’t have trouble, though it is important to remember that they are always educated guesses, chosen to give a high probability that you won’t have crosses. There is always the possibility of a bumblebee going out for a jaunt on a breezy day, landing on a supposedly self-pollinating pea flower, chewing through the side just for fun, wallowing in pollen, and then going berserk and deciding to see how far it can fly before sundown, only to land on a pea flower two kilometres downwind where out of sheer spite it chews into that flower and crosses it. There is no knowing the lengths to which a malicious insect may go; but generally, the isolation distances work if followed.
…Oh yes, and pollinators are our friends, by the way.
To decrease the risks of insect-pollinated crops not being accidentally crossed, some recommend planting another high-nectar crop in between two varieties of the same species. This is particularly advantageous if bees are major pollinators. The idea here is that the bees aren’t flying randomly, and don’t just collect pollen by accident. They usually work over a patch fairly systematically (after all, flying takes a lot of energy, so they don’t want to waste it), and when they are full, they go back their hive to unload, both nectar and pollen. If you have something in full flower between your seed crops, the bees should fill up and go wipe their feet before landing on your second variety. I like buckwheat for this purpose; it is easy to grow, flowers early, and then keeps flowering for weeks. However, if there is a beekeeper in your area it would be thoughtful to ask permission before planting a patch of buckwheat. Buckwheat honey has a distinctive flavour, which some love and some detest. You can also just plant patches of flowers in your gardens (though of course that can rapidly add more complexity to your seed saving).
It is also worth looking outside your garden when considering seed saving. The more foods for pollinators there are between your garden and the next place they might find the same species, the lower the risk of crossing. If there is nothing but a parking lot between your garden and your neighbour’s garden, you will need a lot more space than if there are trees and bushes.
“Cheating” on Population Size
Some crops need large population sizes, and also take a lot of space per plant. What if you don’t have enough space to grow as many as are needed this year? The solution is in seeds’ botanical fine print: you have to grow that many plants, but you don’t have to grow them all in one year. Many seeds remain viable for multiple years, which allows you to grow only part of the required population each year and then mix the seeds from multiple years to achieve your required population size, as depicted in the table below for corn:
| Year | # of plants | Seed Lot Planted | Seed Lot Harvested |
| 1 | 100 | A | B |
| 2 | 100 | A | C |
| 3 | 100 | (B+C) | D |
| 4 | 100 | (B+C) | E |
| 5 | 100 | (D+E) | F… |
In this example, we are growing half the required population each year; for some crops, you can probably stretch it to three years. You will want to test the germination of you seeds to help you decide whether you can go another year or not.
“Cheating” on Isolation Distance
Sometimes you won’t be able to isolate your plants far enough from others to keep them pure by distance alone. If that is a problem, you have several options.
The simplest is to use the pattern described above, except to grow one variety each year:
| Year | Variety Grown |
| 1 | Arikara Yellow bean |
| 2 | Jacob’s Cattle bean |
| 3 | Orca bean |
| 4 | Arikara Yellow bean |
| 5 | Jacob’s Cattle bean |
| 6 | Orca bean |
Bagging and Caging
If the multi-year schedule described above doesn’t work, you can take more extreme measures. With plants which have relatively few large flowers or fruits, you can bag the flowers and hand-pollinate them, as mentioned for corn and squash.
For plants which have many small flowers, this is impractical; instead, some people “cage” the whole plants, growing them inside a screened enclosure (I use a mosquito net, like you might use for camping, for some legumes). If they are capable of self-pollinating but are sometimes interfered with by insects, you can simply keep them shut in their cage the whole time they’re flowering. But if they require insects for pollination, you will have to be a little cleverer. I have heard of two techniques (though I have not tried them personally).
If you need to keep your seed plants caged the whole time they’re flowering, because of neighbours’ plantings or wild relatives of your crop, you will have to introduce pollinators to your cage. These pollinators must be “clean,” and uncontaminated with foreign pollen. So some people grow their own houseflies, which are an easily-raised insect capable of pollinating many species of plant; all you need to do is put a scrap of meat in an open container and wait a few days. When you see maggots, cover the container with a screen. When the flies emerge, release them in your cage and wish them happy pollinating… the lengths we go to to save seed, eh?
If you are only trying to keep your seed plants from being contaminated by another group of your own seed plants, you can do alternate- day caging. Cage both plantings. Then open and close the cages on alternate days, so that wild pollinators have access to only one variety per day, as in the schedule below.
| Day | Cage A | Cage B |
| Monday | Open | Closed |
| Tuesday | Closed | Open |
| Wednesday | Open | Closed |
| Thursday | Closed | Open |
| Friday | Open | Closed |
This saves you the bother of raising pollinators, but it means you need an extra cage, and your seed yield will be lower per plant because of all the flowers the pollinators couldn’t access. If that seems worth it to avoid dealing with maggots, go ahead…
Always Watch Your Seed Crops
Assuming that you follow the theory, most of your seed should be pure and true-to-type. However, you also will want to do some observation in the garden each year when working with your own seed. Watch for plants which are obviously not like they should be. If they are diseased, remove them. If they appear to be a cross or a mutation, however, take a closer look and think a little. Every cross or mutation presents the seed-saving gardener with a choice: if you like the variety you already have, you should remove it so that it doesn’t accidentally get mixed in with the normal plants you want to save seeds from. But an unusual plant is also an opportunity; many heirloom varieties are the results of gardeners who noticed just such plants, found that they were better in some way than the original variety they were growing, saved their seeds, and stabilized them over several years to make new varieties.
Germination Testing
If you save seed, you may want to test its germination rate to ascertain its quality. This is probably unnecessary if you harvest good-quality, healthy seed one year and plan to plant it the next spring; but if it doesn’t look so healthy, or is several years old, a germ test can be helpful.
There are various techniques used for germination testing, depending on the crop and how professional one wants to be about it. For most of the staple crops, I would suggest the following test:
Take two sheets of paper towel. Write the variety name on them with a pencil. Count out a sample of seeds randomly selected from your seed supply (25-200, depending on their size, how many seeds you have, and how patient you are). Line them up on one towel, then fold the other over them and roll it up.
Soak the seeds in room-temperature water, then drain until water stops running out of the towel and starts to drip. Put them in a plastic bag or plastic container.
Find a place which has a similar temperature to what you would expect when they are germinating in your garden, and set them there. After about a week, unroll the towel and count how many seeds have sprouted. If there are seeds which still look healthy, but have not sprouted, leave them for a few more days and check again. Calculate your germination rate:
sprouted seeds/ total seeds x 100 = % germination
The small grains need prechilling to germinate properly. Set up the test, and then put the seeds in a place which between 5-10º C for 3-5 days before moving them into a warmer place to simulate early spring.
If you are interested in more of the details of germination testing, you can study the Canadian Food Inspection Agency’s Canadian Methods and Procedures for Testing Seed.