Hybridizing Plants: The Creative Adventure of Designing Your Own Flower Varieties

I like the idea of creating my own plants with the colors and patterns I personally get to choose to my own liking. While doing some of this myself over the years and observing a lot of other people doing the same, you start to notice various limitations, flaws in your understanding of how to get where you want to go with a flower "design," doing something unique, and a litany of unexpected quirks with plant breeding. You might think it's as simple as taking pollen off one daylily and dabbing it onto the stigma of another one, and the result is that all the good characteristics you see in both plants will appear in a new and improved plant from the seeds you harvest.

I'm sure that has happened before, but the process can be as elaborate and involved as you could ever want, or not want, it to be. Here are some points I wish I had known when I first started hybridizing flowers—things you either find out on your own through trial and error or another hybridizer tells you:

• 90–100% of your resulting crosses may not be of any use for various reasons. It's not unusual to hear a flower company talking about planting hundreds of thousands of seeds—or more—just to find two new varieties for a future year. What this can mean is that if a new tulip variety is discovered in a field and maintains stability for a couple of years to be propagated into more plants, it might take 20 years or more for that variety to arrive on the market. It not only has to prove itself worthy of keeping around in the first place, but it also has to be easy, hardy, and special enough to stand out. Then it has to go from one plant to thousands of plants in order for you to be able to buy one at your local big-box store. Many of these survivors are moved through various stages over several years of testing, as the best in the group move to the next test stage.

If I make a cross at home of, say, an African violet that looks interesting, it will take maybe 4–5 months for the seed pod to develop. Then, if I immediately plant those seeds, it might be another 6 months before any of them bloom. After that, I would want to keep growing those plants and then eliminate many of them for a list of reasons ranging from "looks just like one everyone has," to "the blooms flop over," to "it doesn't propagate easily," to "it died a couple of months later." You're basically looking at a year just to see what you might be dealing with on this kind of cross. Then, when you realize how bad most of them are, the ones you have left (if there are any) need to go another year or two to confirm the blooms don't change into something different. It's amazing how a promising flower can look totally different a year later.

There are quirks all over the place that are universal, and some are just specific to that one flower. If you want to raise daylilies, you have to make sure you're crossing a diploid with a diploid and a tetraploid with a tetraploid (the number of chromosomes in a flower). If you cross a diploid with a tetraploid, it's likely to be an infertile cross, and you won't get seeds.

Many flowers are missing the reproductive parts. You might wonder why you are unable to get any seeds from a cross—it might be because some flowers produce pollen, some can only receive pollen, some are fertile both ways, and some have stigmas and pistils that are trying to become petals, so they won't accept pollen. What often ends up happening is that some of your plants can only be used for pollen, and it might turn out that you have one good pollen-producing flower variety. This might be the flower you use to pollinate all of your flowers because it not only makes a lot of pollen but also has a lot of good plant characteristics. As a result, many of your flowers end up looking alike because your champion pollen maker has a specific set of traits.

Many double flowers with all those extra petals can have serious problems getting pollinated because either a bee—or you—has to get through all those petals, hopefully locate a stigma, and pollinate it with pollen from another flower. Self-pollinating can work, but it can also fail. You won't know until someone either tells you or you figure it out a year or two after first trying it.

If you have a really cool pattern or color on your flower, often the rest of the plant is not ideal. You're lucky if you can get a fresh color, a great stem, hardiness, and a productive new plant. Colors on flowers don't operate the same way they do when you mix watercolors. The dominant color erases the recessive color most of the time but can mix with that color. You could also get a color that was being used by a hybridizer 50 years ago to create this particular flower. So, you must focus on only crossing similar colors if your desired color is recessive or making second or third crosses to bring that color out. This process can take years.

Some colors are not even possible, such as a blue dahlia. A yellow African violet is actually a flower that is white but absorbs yellow pollen into the petals, which makes it appear yellow. Most people would just think it was a yellow African violet. There are quirks, and there are tricks.

When I look at most daylilies, they seem to have a ghost of the original yellow and orange daylilies in them. Even though I have a "black" daylily, it might actually be a very dark red with some yellow showing through. But to the casual observer, it looks pretty black. Many "black" flowers are actually just very dark red or dark purple flowers.

You need balance. If there are many flowers—a flush of flowers—and you try to breed for a variation that has everything the same but with bigger flowers, you'll have problems. You will then discover that the plant only puts so much energy into flower production. You might get bigger flowers, but you may end up with 7 larger flowers instead of 12 smaller ones. The flowers might also flop over because the stems aren't sturdy enough. Then you think, "Well, next year I can work on crossing this with another flower that has sturdy stems to support those blooms." Notice how this process can easily span many years.

Accidents happen. You'd be surprised at how many of your favorite plant varieties were just random discoveries by someone. A single sunflower appears with double petals for no reason. A man saves it, plants the seeds, culls the single plants, and starts a line of double sunflowers. Then, a few years later, they are everywhere. Or a single ivy leaf appears with a different pattern, gets propagated and patented, and becomes the latest houseplant craze. It's fun to go to www.Garden.org and look up some of your favorite flowers to see where they came from. Often, the two plants used for the cross look nothing like the resulting flower—or they look almost identical to one of the parent flowers.

You can often get the best results when crossing two plants by figuring out the goal you want to accomplish and using a selection of plants that fit that description to start with. If you want a large cactus dahlia, your best choice is to use existing cactus dahlias but pick the biggest ones. Then you're much more likely to get a range of cactus dahlias that might include a few that are larger than the parents.

If it's feasible, hand pollination is the way to go. But if you're smart about it, you can let the bees do all the work for you. If you're trying to grow a new red rose, you'd probably want a couple of good red roses in your hybridizing patch, letting the bees bounce around between them, then planting the seeds to see what you get. If you're dealing with a lot of double roses, that might be easier than removing all the petals and trying to pollinate them yourself. But here's another secret: many hybridizers cross semi-double flowers with each other, knowing a certain percentage will end up fully double. Double flowers can often be very difficult to work with and even infertile. So it can actually be easier to work around this with semi-double flowers.

Often, I'll see a hybridizer using the pollen from a certain favored semi-double flower, then removing petals from a double to pollinate it. Often, you will get no seeds, but sometimes you'll be lucky and get... three seeds. One might germinate. You can see how things easily get complicated.

There's nothing wrong with pollinating simpler plants. A beginner might have fun hybridizing daylilies because they tend to be easy to pollinate. You don't need magnifying lenses and tweezers to perform the cross. I randomly took pollen from a pink double daylily and dabbed it on the stigma of a "black" single daylily earlier this year. Later, I noticed a seed pod full of about 20 seeds. I don't know what I'll get, but it could be fun to see. However, I likely won't know what they look like for another two years if I plant them now.

The best time to start a cross is six months ago. But if you didn't make a cross last July, the next best time is now. And a year or two will eventually arrive, so you might as well have something in progress. The thing about hybridizing is that once you get going, you notice the slow progress less because you have several things happening. If you're regularly seeing new things blooming, you won't care that the seeds you just planted won't bloom for a couple of months (at least).

That's why I love houseplants. You don't have to bring everything to a complete stop every autumn, wait six months, then start again. Things can be ongoing.

© Copyright Terry Aley

The Aley Acres seed shop on Etsy.

Dahlias, Notes from a Gardener book on Amazon.

Floral Art and Landscape Painting Etsy shop.

TikTok, some gardening posts.