How many plants do you need to maintain a variety without inbreeding depression?

[aliceklar]

Greetings!

I'm an experienced gardener and seed saver, but relatively new to growing cannabis. I'm familiar with saving seed from chillies, tomatoes, beans and peas, and maintaining varieties that way, but they are all very happy to be inbred, and I am aware that cannabis is an outbreeder, and will suffer inbreeding depression if the genetic pool gets too small. The question is, how big a pool do you need to maintain a variety without getting loss of vigor or other undesirable mutations? I've some experience with saving seed from carrots, corn and squash, which are trickier, but I'm not an expert...

With corn - which is a strong outbreeder - you need to grow a minimum of about 100 plants per cycle, and save seed from the best 20. With carrots, to maintain a variety, its recommended to save seed from between 20 and 50 plants each time, and with squash its 5 - 10 plants (according to the good people at the Seed Savers Exchange). Now, I'm not sure how different cannabis is - for one, its male and female flowers are on different plants (more like squash), whereas carrots and corn have male and female flowers on the same plant (or male and female parts within the same flower). I'm guessing 5 mother plants would be a minimum? What about the fathers??

I'd really like to maintain favorite varieties myself, rather than buying seed each time - and I'm wondering how many males, and how many females, I need to use each generation to keep sufficient genetic diversity to avoid inbreeding depression. Very interested to hear of peoples actual experiences of this, and especially any tricks that can be used to maintain varieties when space is limited due to indoor growing in a small space.

Peace, AK x

 

['Boogieman']

Subbed

 

[Rodehazrd]

Great question I'm subbed

 

[aliceklar]

I also suppose that this is complicated by the fact that varieties are often preserved through cloning rather than sexual reproduction...

 

[unregistered190]

Interesting....now I am curious too

 

[tetragrammaton]

Great thread, I had no idea about this, so I look forward to someone coming forward with knowledge on this.

Hybrid vigor is great, but many "strains" (I put that in quotation marks because from what I understand, cannabis is the only plant in which we call different "varieties" strains, rather than cultivars), such as Skunk #1, Northern Lights #5, and countless other strains have been inbred and to this day still are great.

From what I understand, sometimes clones lose their vigor over years, but there just as many people who will tell you this is false, and just as many people who will tell you it's true. From what I understand, it really depends on the strain of the clone, and growing conditions, which causes it over time to become less potent, vigorous, hardy, etc... than initially.

Thanks for a great post, as I feel that a lot of cannabis botanists can learn a lot from other botanists, and vice versa.

Peace and Love,
Tetra

 

[Dropped Cat]

Now, I'm not sure how different cannabis is - for one, its male and female flowers are on different plants. I'm guessing 5 mother plants would be a minimum? What about the fathers??

I'm wondering how many males, and how many females, I need to use each generation to keep sufficient genetic diversity to avoid inbreeding depression. Very interested to hear of peoples actual experiences of this, and especially any tricks that can be used to maintain varieties when space is limited due to indoor growing in a small space.
Peace, AK x

Good post, and I'm certain as time goes by you'll find the program that works for your needs.

As for my contribution, I keep four mum females and one male,
I lost my second male and am looking for a replacement among my seed stock.

A micro set up, about two square feet flower and
a veg box half that size, with a shelf.

I, and the folks I share with, like a variety of effects
and with one narrow leaf type and three hybrid mums I can
satisfy the two main effects desired.

I have made bubble hashish, cobs and standard bud yields
from various combinations of the parent stock for many runs.

Bottom line, my nickle's worth of free advise would be
three proven females and two males to keep as P1 and do
F1's for seed lots.

Cuttings of the proven females and various crossings of the
P1 stock for variety.

Rinse, lather, repeat.

Check out my thread:

https://www.icmag.com/ic/showthread.php?t=294387

 

[@hempy]

Well depends on who you ask if you say a great mom n dad of a set stable or pure or inbreed lines works well odds are you will be attacked by people who are selfing a clone and claim that is great breeding yet will tell you that you need more than one female and male by them to breed with.

 

[aliceklar]

A micro set up, about two square feet flower and
a veg box half that size, with a shelf.

...

Bottom line, my nickle's worth of free advise would be
three proven females and two males to keep as P1 and do
F1's for seed lots.

Cuttings of the proven females and various crossings of the
P1 stock for variety.

Rinse, lather, repeat.

Check out my thread:

https://www.icmag.com/ic/showthread.php?t=294387

Thanks DC - that's good info. I've read your bagseed thread and am impressed with what you are achieving with such a tiny space. Flowering plants in such tiny pots too - nice work!

I hadnt thought about maintaining F1 lines - but keeping a selection the quality clone mums and dads does not seem like such a mission as they can be kept tiny and dont need much light (I've read the Oldtimer article about bonsai mums, etc), and genetic diversity can be maintained by selecting from the F2 seed. Its interesting - a different approach to what I'm familiar with (trying to maintain a "purebred" line of a vegetable). Lots to think about!

 

[q3corn]

Subbed. I'm curious as well.

 

[El Timbo]

Bear in mind that your starting stock - if you buy seeds from a breeder or seed maker - will most probably all be seeds from the same 1 female x 1 male. So you are already off to a bad start if avoiding inbreeding depression is your aim.

 

[zif]

Vigor and inbreeding depression are slippery concepts. We do know when viability falls off - Cannabis tends to start making sterile pollen and few/bum seeds when it’s carrying too many deleterious recessive alleles.

Let’s imagine you can’t or don’t want to use clones. With seed alone:

If you’re making 1:1 M:F crosses, you should easily be able to maintain viability for at least a good half dozen generations. (Even selfed lines can be pushed to S3-5 before petering out.)

Using something like ten plants per generation will extend that dramatically.

Or make several sibling lines, recombining them when only they’re flagging, and you can probably maintain viability from 1:1 crosses indefinitely.

What you probably won’t maintain is ‘the variety’, even assuming you were fortunate enough to start with something close to one. You can chase a few traits, but as Allard put it, heavy inbreeding is like swimming in a rushing river. You’re going somewhere fast, but it’s not all up to you. The random background fixation of alleles will have a much bigger effect than your selection.

On the other hand, it will also rapidly make your line unique.

We know you need a couple of thousand plants per reproduction if you want to preserve (close to) the full set of genes in a given variety.

Selecting for an F1 is a very wise approach to striking the vigor:numbers balance. Reciprocal recurrent selection is a very neat way of doing so.

If you do take clones, anytime you like the outcome of a pairing, keep the parents. They will produce the same outcome essentially forever.

 

[Im'One]

I grew out some island sweet skunk and found a pheno that lets me sleep all night with no pain. It doesn't relieve my other symptoms however. I didn't clone but seeded it with a Taskenti indica plant and kept the seeds.
If I grow those out and find a good male then find another iss female with similar traits should I breed those two for a backcross?

 

[Dropped Cat]

Anytime you like the outcome of a pairing, keep the parents. They will produce the same outcome essentially forever.

Exactly.

Many commercial breeders strive to have keeper P1 stock
as the F1 seeds from the proven parent stock are key to
maintaining those desired lines.

Many sad stories of good breeders loosing a prized P1.

I intentionally culled the P1 mum from a selection of seeds
from a stash found from 1979.

I regret that decision from time to time.

 

[aliceklar]

Reciprocal recurrent selection / selecting for an F1

Reciprocal recurrent selection / selecting for an F1

Vigor and inbreeding depression are slippery concepts. We do know when viability falls off - Cannabis tends to start making sterile pollen and few/bum seeds when it’s carrying too many deleterious recessive alleles.

Let’s imagine you can’t or don’t want to use clones. With seed alone:

If you’re making 1:1 M:F crosses, you should easily be able to maintain viability for at least a good half dozen generations. (Even selfed lines can be pushed to S3-5 before petering out.)

Using something like ten plants per generation will extend that dramatically.

Or make several sibling lines, recombining them when only they’re flagging, and you can probably maintain viability from 1:1 crosses indefinitely.

What you probably won’t maintain is ‘the variety’, even assuming you were fortunate enough to start with something close to one. You can chase a few traits, but as Allard put it, heavy inbreeding is like swimming in a rushing river. You’re going somewhere fast, but it’s not all up to you. The random background fixation of alleles will have a much bigger effect than your selection.

On the other hand, it will also rapidly make your line unique.

We know you need a couple of thousand plants per reproduction if you want to preserve (close to) the full set of genes in a given variety.

Selecting for an F1 is a very wise approach to striking the vigor:numbers balance. Reciprocal recurrent selection is a very neat way of doing so.

If you do take clones, anytime you like the outcome of a pairing, keep the parents. They will produce the same outcome essentially forever.

Thanks zif - that's awesome - a lot to take in there! I'm very happy working with clones by the way, and I'd certainly settle for maintaining a healthy population in which desirable characteristics improve over time. Am googling reciprocal recurrent selection now... ... (found some info in botanylibrary.com)



OMG... I hope its OK if I come back and ask some more questions about exactly how that might work in practice?? Jeez that looks like so much fun. Tho I may just have to give up my day job and spend the rest of my life growing weed now

I'm currently growing out quite a range of different varieties as an initial step, to see what grows well for me / what I like, so I have the raw materials for a diverse base population.

So, if I understand correctly, I'd start by generating "Two heterozygous populations" (A and B), which I could do by:

• taking my favourite two female plants out of the selection I'm currently growing
• pollinate each with something different,
• grow out the two sets of F1 seeds (not sure how much variability I'd see in this group?)
• cross the best female in each F1 population with the most interesting male in the same population. If the F1s are fairly uniform (which is what I'd expect from my experience with peas and tomatoes... although I know that cannabis is an outbreeder, so may be very different!) I should be able to save seed from just one or two of the F1s.

This should provide two sets of F2 seed that would be wildly heterozygous (assuming that the ancestry of the original 4 parents was sufficiently diverse).

 

[aliceklar]

Reciprocal recurrent selection #2

Reciprocal recurrent selection #2

For info, here's the description I found of reciprocal recurrent selection: (from botanylibrary.com)



"Reciprocal Recurrent Selection: A form of recurrent selection that is used to improve both gca and sca of a population for a character using two heterozygous testers is known as reciprocal recurrent selection (RRS). It is also termed as recurrent reciprocal half sib selection. This scheme was proposed by Comstock, et al. in 1949.



Main features of this method are given below:
i. This scheme is also used for the improvement of polygenic characters.
ii. Selection is made on the basis of test cross performance.
iii. Two heterozygous populations each of which is the tester for other are used in this method. The two populations may be designated as A and B.
iv. This method is used for improving a population both for gca and sca for specific character.
v. This method is equally effective with incomplete, complete and over dominance.
vi. This method is used for the improvement of those characters which are governed by both additive and non-additive gene action.
vii.This method requires three seasons or years for completion of each cycle of selection as given below:

First Year:
Several phenotypically superior plants are selected from population A and B. The pollen of some selected plants of A population is used to cross large number of randomly selected plants of population B. Similarly, pollen of some selected plants of B population is used to cross large number of plants of population A. All the plants of population A and B used as pollen parents in the crosses are selfed.



Second Year:
The progeny of test crosses made with pollen parents of A and B populations are evaluated in separate replicated trials. The superior progeny are identified.



Third Year:
The selfed seeds of those A and B plants whose progeny were found superior in replicated trials are grown in separate block. All possible crosses are made among the progeny of A plants and also among the progeny of B plants. The crossed seeds of A block are composited in equal quantity to raise A generation. Similarly, crossed seeds of B block are bulked to raise B1 generation. This completes original cycle of selection.


Fourth Year:
The A1 and B1 populations are grown from the composite crossed seeds of respective population obtained in third year. Then operations of first year are repeated.



Fifth Year:
The operations of second year are repeated.



Sixth Year:
The operations of third year are repeated.



The last three years constitute first cycle of reciprocal recurrent selection. Such selection cycles may be continued till the desired improvement is achieved."





I dont think this is quite right for working with cannabis, as it is talking about "selfing" pollen parents - which only makes sense for monoecious plants (that have both male and female flowers on the same plant). For cannabis this would mean turning male plants hermi and pollenating themselves?? That is beyond what I'm able to do!



What I'm wondering now is how this might work in practice for a dioecious plant like cannabis (and how could it be modified to be workable with small populations, grown indoors)

 

[aliceklar]

Selecting for an F1 is a very wise approach to striking the vigor:numbers balance. Reciprocal recurrent selection is a very neat way of doing so.

Zif, could you expand on what you mean here, and how it might work in practice? This stuff is fascinating, but I'm out of my depth, especially with the reciprocal recurrent selection stuff...

 

[zif]

Zif, could you expand on what you mean here, and how it might work in practice? This stuff is fascinating, but I'm out of my depth, especially with the reciprocal recurrent selection stuff...

I’d recommend avoiding getting bogged down in the specific details of the plan, at least to begin with. (Corn breeders came up with this stuff - they could self any plant. But couldn’t easily take cuttings.)

The general idea is to improve the “combining ability” of two parent lines. In other words, to make more desirable F1s between them, while inbreeding those lines.

One (of very many) ways to do this in cannabis:

1. select a decent number of superior (F)emale plants in A & B. Save F selections as clones. Collect and bulk pollen from all but cull (M)ales in A & B. Store bulked pollen. Seed A selections with B pollen, and B with A.
2. test progeny. The best results identify a F in each of A & B. If everything is great/all progeny are what you want - you’re done with rrs!
3. pollinate best A F with A M bulk, B w/B. These seeds are your new parental lines.
4. repeat.

The main point is that we’re selecting in A and B mainly based on the test crosses. Our selections to inbreed the lines are chosen because they make better F1s. We also intentionally slow the inbreeding by using all of the males - we don’t want to let accidentally fixed traits to swamp our intended changes. (Note you can also use a number of new males in 3 - we just want the ‘best’ F to set seed with many sib Ms.)

In corn, lines made this way often become very good parents against any other inbred line. This is the g in gca - that is, general combining ability. The two lines by design work very well together. This is sca - that is, specific combining ability.

It’s just a start, but hopefully that helps a bit.

 

[aliceklar]

That helps a lot! Thank you Zif!

So, when you say:

2. test progeny. The best results identify a F in each of A & B.

Do you mean that I would be growing out the seeds from the selected superior "A" females, that had been pollinated by half of the mixed "B" pollen (and vice versa), and I would then be choosing the single female that made the best offspring when fertilised by the opposite line?

And then I'd complete the cycle by going back to the clones I'd kept of the winning female in each line (A and B), and pollinate each clone this time with the remaining half of its "own" pollen (A with A, B with B) to generate a new test population of A's and B's again. etc..


I'm definitely going to give this a try!

 

[Bud Jones]

Subbed and thx for posting . I have those 80’s Super Skunk beans I just got . Only a ten pack and there are several different schools of thoughts on the best tactics to preserve them . One being open pollination..