Coughie
Member
I've actually been sitting here, for the first time, after swearing I wouldn't...
I've been considering learning about and undertaking the idea of Selfing her..
Maybe some S1's would show me something more like what her mother was..
But I dont want to derail my 'project' as it is - I still want to make a heady-version of a Bubba-esque sort of line.. Something 'new', in combination anyway..
That's what the momma was, something-Bubba.. almost entirely indica
Two-toke Sleeper
I'm copying a respected approach to this Selfing idea, in hopes of stimulating some more thoughts and conversation;
With permission:
It's to intentionally bottleneck plants to reduce variation within each filial generation, so you have some aspect of control what you are reintroducing into the next generation. This breeds towards single phenotype stability.
****edit****
Each generation between backcrosses should be isolating something specific. Each time you self, you are removing gene traits and limiting variation, but also, "locking in" a single individual trait. The purpose of the backcrossing is to restore the variability of the original parent.
Example of something easy to grasp, using a single plant selection, to stabilize towards a single desired phenotype.
Say our goal, is to make a silly frosty OG. So we make -
F1 - OG x Deep Chunk
We want silly resin, but we want an OG type plant, so our F1 selection, is a single lanky, Piney OG pheno from 100 plants. Looks 80% OG with much more resin. Yields less. Poor calyx to leaf ratio - but the most frosty of the 100 plants that still grew like we expect an OG to grow.
We take this female to the same OG mom..
BC1 -OG x (OG x DC)
Now we seek an individual plant that again most fits our desired outcome. We sort 100 seeds - and find our OG has really gotten lost in the Bx1 but our resin is through the roof.
We select from the 100 plants, 5, that best fit our desired outcome - the most OG looking of the batch, with the most resin.
***This is where people mes up. They try and select plants that "have it all" when you should be selecting for individual traits.***
Because we are working with poly-hybrids and we want to maximize our path to stability, we want to take those 5 plants and self them, to see exactly how much diversity each of the 5 contain within themselves.
BC1 - A - 100 s1 seeds
BC1 - B - 100 s1 seeds
BC1 - C - 100 s1 seeds
BC1 - D - 100 s1 seeds
BC1 - E - 100 s1 seeds
Now, all you are looking for in these plants, is out of 100, what percentage have extreme resin output, even if this breeds you away from your OG type box.
Let's say, BC1 - D wins and had 40% plants that showed extreme resin output. (every other factor being irrelevant) You select the 5 best of that 40%, to make your S2 lines with.
BC1-D-A - 100 s2 seeds
BC1-D-B - 100 s2 seeds
BC1-D-C - 100 s2 seeds
BC1-D-D - 100 s2 seeds
BC1-D-E - 100 s2 seeds
BC1-D-A performs best with about 60% of the offspring showing extreme resin production. You select the 5 best from the 60% and self them to make your S3 lines to sort.
You do this until you hit a point of satisfaction in regards to stability for the single isolated trait.
So to follow the previous outline, assume this was achieved by S4. A plant in the S4 generation, when selfed, showed it's offspring (S5) to be 90% high resin content. Let's call this satisfactory.
***This S4 plant is your parent because it passes the desired trait forward. NOT some further random plant from the S5 line, unless of course, you want further stability, pick the best 5, sort 100 s6 each, find one to be 95%...and use the S5 as the parent instead.
The point being, don't use a plant in a backcross from an untested generation even if you believe it is a refined improvement, until you have actually tested it's selfed generation to see if it will actually perform as desired.***
BC2 - OG x S4
The reason you make this backcross is to reintroduce your P1 OG female to restore traits that were lost in the selfing process while isolating a single trait.
You sort 100 BC2 seeds and you KNOW any plant worth selecting should have the high resin content you desire, but now you want to select plants that also have high yield. So now, you select 5 plants that have the high yield, that also maintain high resin content...and you self those 5, again checking only to see which plant is going to pass BOTH traits on to the next filial generation with the greatest consistency....
Wash. Rinse. Repeat.
It is very time consuming, but if done right, you can theoretically build a very good, truly stable cannabis strain.
TomHill said once you start hitting F5 and F6 though, lots of latent gene traits begin to resurface, without cause or reason. I can not personally speak to this. I can however say - if you are trying to improve your personal stock via single plant selections, this method will work.
dank.Frank
I only have the single plant to work with, but wish to find and isolate phenotypes leaning more towards Coughie's mother than Coughie's personal expression...
I'm hoping there's enough of her mom in there, that I can find it.