I was having a conversation earlier about landrace cannabis and breeding, and one of the biggest challenges, to me, seems to be the stabilization of sex, taking a breeding line from wild dioecious to inbred/heirloom monoecious/sex stable. I’m having trouble visualizing the exact workflow to make that happen. Regular selection and culling seems like it would take an eternity.
That got me wondering, I’d really like to see the entire workflow, from start to finish, of someone taking a wild, dioecious landrace cannabis line and turning it into a ‘top shelf’ sex stable inbred line without the introduction of external genetics. There’s gotta be a long term grow journal of some one that’s tried that, a wild to ‘top shelf’ breeding project with no external genetics. Or maybe that’s not possible, and there just isn’t enough genetic diversity in a single landrace line like that.
If you were given 100 wild, landrace cannabis seeds that expressed male and female traits and many other undesirable traits (but also many desirable), what would your workflow be to turn it into the highest quality, indoor friendly cannabis you could, and how long do you think it would take?
It’s a ridiculous question, I know, but I am quite curious, because I think the answer may influence the manner in which I explore cannabis genetics.
Difference between stabilizing a strain and bringing a batch of seeds to market.
It could take 8 generations or more to stabilize a strain to grow true from seed. Many shortcuts are taken when making hybrids, and introducing “domestic” varieties. When a “breeder” finds two parents that make a good batch of plants that are relatively uniform (harder with mixing multiple hybrids) and has the traits they like, they can just keep making that batch with the same parents and sell the seeds as under a certain name. If someone takes that F1 batch and makes F2, the resulting batch of seeds can be all over the place with phenos, and then you’d select plants with the same traits that you’re looking for and keep breeding for only that trait/pheno.
That really makes sense. In that scenario, though, it’s all 1:1–is that how it’s done with cannabis, pairing ideal parents up with one another? I thought it was always supposed to be selective open pollination with guided culling over generations to get the population where you want, not the individual plants. Outside of cannabis, when I’ve done plant breeding, it’s always population based breeding, not 1:1 parental matching. This is very new to me.
Edit: Thinking about it more, I think that is what the cannabis community refers to when they say inbred line, population based breeding with a consistent end state.
I think that’s about shifting a population/field to your desires, like what they do with the hash fields. Still basic at a landrace, but cultivated. If your talking about a top shelf smoke that people will grow all over, that takes more individualized refining. Now all this is really beyond my experience, just based on my knowledge of selective breeding.
@Eudaemon - I think you may be confusing monocious with homo-zygotic.
I would underscore that homo-zygosity is on a gene by gene basis and breeding-true is only relevant for a stated set of genes. It is increasingly difficult to breed for multiple traits simultaneously. As you add desired traits the size of the necessary selection field increases exponentially. If traits t1 and t2 appear once each in four plants than finding a single desirable cultivar will require (statistically) 16 plants just to match those two traits…
Some breeders will work traits in parallel lines, crossing only after establishing that the individual desired characteristics are stable.
Most everyone chooses not to recreate the wheel and works lines that already largely match what they seek.
All cannabis is dioecious by default, meaning there are separate Male and Female Plants in the species. Not both organs on the same singular plant, that’s Monoecious, or what we lovingly call, hermies… Monoecious plant species have both male and female organs on the SAME plant and can self pollinate. Usually you don’t want that…
Landraces are not hermies. Sure there are some subspecies that will, if not pollinated by week 8 or so, start producing male flowers in order to save itself, but from what I understand, that has only been in select cultivars from select area’s around the planet. As a whole, cannabis doesn’t do that.
Now to actually get to the question lots of time and likely a mix of open and selective pollination. I’d select where you have to, in order to maintain the traits you want, and open pollinate a bunch of your selected cultivars as you can, to maintain a nice genetic spread while still working the traits you want/not totally messing up your selection process.
If you grow out 100 plants and only 10 have any traits you like, I wouldn’t open pollinate then, I’d do a select pollination between those 10 or even more selectively if I have to. Then grow out some of the F2’s and see what you get, you might only get 5 you like from those, yet another kind of a selective breeding. Then at F3’s you might get 50+ you like, can open pollinate all those then since everything is showing up nicely and you’ll keep some variation in the line, then go for F4’s. Or you could do selective at F3 to further reinforce the traits and open pollinate the F4’s. Rinse and repeat ad nauseum until you either get it how you like or give up on it. It’s really something you just have to do and then make an assessment on it when you get there. But like everyone else has said, it’s really just a matter of time and lot of crossing and selecting.
That really helps clear that up, the monoecious/dioecious clarification. I do find it interesting that it is dioecious by default yet there pockets where monoecious seems to be the default (landrace thai). I had honestly just assumed cannabis was a monoecious plant that we somehow bred to be dioecious, but I guess that wouldn’t be entirely possible because most monoecious plants can’t go dioecious by breeding? That is a little bit outside the scope of my knowledge and experience.
That workflow you posted is exactly one of the types of things I am looking for. What sort of results to expect at each generational leg of the breeding journey. It sounds like the key to speeding things up is to open up the pollination numbers as the traits get more stable, leading to an inbred line. I was sort of approaching it backwards, thinking I had to maintain as much of the F1 population as possible, even if they have undesirable traits, to try and breed those out as slowly as possible, and then whittling down the breeding population each generation until I had a small population of plants with desirable traits instead of a large population of plants with desirable traits. The way I was envisioning it, though, would have taken a centuries practically.
How much vigor do you add and consistency do you lose by backcrossing one of those F4s or F5s with an F1 or F2?
This is all still pretty new to me. Wild how much there is to it.
Do you think that’s part of why the modern cannabis gene pool is so shallow, or does that have more to do with a lack of access to novel landraces and heirlooms that make up new breeding stock? Re-inventing the wheel sucks, for sure, but it seems like it would have made for a more interesting modern cannabis scene.
Ah gotcha. I mean honestly it just depends what your goals are. If you’re wanting to kind of create your own thing, I would say that in my opinion, especially if starting from a landrace or similar, the F1’s will already be so varied that there wouldn’t be any benefit in doing a open pollination. I would be super selective at the start and open it up later as the trait(s) you want stabilize in the line. The only reason I’d do open first or anywhere in the beginning really, is if I’m hunting for a specific combination of traits or plant to use. Just have to play it by ear but yes generally you’d select first, and open later once things are mostly how you want. Otherwise you’re just spinning the wheels and not really making headway.
For vigor there isn’t really a set thing anyone can say. Some don’t lose any as you inbreed up to a point, some lose it immediately at the f2 cross, some gain it back immensely from a backcross or outcross. it just totally depends on the plant and/or line itself.
This is so much fun and so exciting! It just really sucks I have to wait so, so many months to see the results. I am not a patient person, so I am learning just as much about that as I am cannabis. I need results now, not in a year :-/
You’re exactly right about spinning my wheels. I kept thinking about light selection over generations of open pollinations, and I kept thinking it would take probably 50 years to gently drift a population even the smallest of directions! It’s the definition of wheelspin.
I think cannabis evolves with the culture. I do not experience the gene pool as shallow. You can do a lot with a couple strains. In the '70s when I first smoked weed there were three kinds. Pot that got me high, pot that didn’t (oregano? Ditchweed? Maybe post extraction) and hash which I believed was “just the tips of the leaves”.
Now we have access to weed from around the globe. Lack of diversity is a symptom of constant hybridization.
Want diversity? You won’t get that from a solo breeding project. Best to cross two very different inbred lines and grow the vigorous F1s for smoke only.
We don’t seem to talk about phenotype manipulation much but size, color, taste, node length, high* and a host of other attributes usually thought of as genetic can be adjusted with environmental changes.
*time to harvest, method of consumption, nutrition, quantity etc.
That’s a whole career/lifetime & a book or two.
lots of time and likely a mix of open and selective pollination. I’d select where you have to, in order to maintain the traits you want, and open pollinate a bunch of your selected cultivars as you can, to maintain a nice genetic spread while still working the traits you want/not totally messing up your selection process.