You CAN get the important (subjective) traits stabilized into the offspring of a given line as true breeding traits.
In this case I strongly doubt it.
Again, we're simply not starting with an inbred line here, but a hybrid, and its a question of heterosis.
I can't say specifically which traits (which is why I said "probably not") but with a unique polyhybrid like ECSD its a fairly safe bet that at least some of the interesting traits of the original (particularly those dealing with scent/flavor and growth) REQUIRE heterozygosity to be expressed.
No amount of crossing can "stabilize" an "Aa" heterozygote into a homozygous/true-breeding form. That's the problem. If you make the locus in question homozygous (true breeding) you lose the trait. And if you take ALL the heterozygous traits of the original and make them homozygous, you're probably going to be leaving behind at least some of what makes the original "clone only" strain interesting.
Put more simply, one of the interesting traits of ECSD is probably "hybrid vigor", which simply cannot be stabilized.
This is done various ways (not going back there, is irrelevant) but through dominant to recessive pairing. Again, it is through selection and experimentation but the correct pairing absolutely sets these paired alleles. Successive generations (again means are subjective), given specific selection criterion increase the statistical observation of the given traits selected for.
I think you're still missing my point. You're quoting from the textbook talking about finding genes and creating heterozygotes.
Yes, of course you can isolate any given trait with enough selection. The point is that by their very nature some traits are not "true breeding".
Sure, if you look hard enough you can find an AA plant, and find an aa plant, and then cross them to consistently yield "Aa"s.
But once you have that Aa, more inbreeding won't generate only "Aa"s. Try it, and you'll always throw off 25% each AA and aa.
More specifically, if AA is red and aa is blue and Aa is purple, then you can NEVER generate a true-breeding purple line, because purple simply doesn't breed true.
Cross any two purple plants and you'll always get 1/4 each red and blue ones. Cross a purple and a red and you'll always get half-red plants half-purple ones. Cross a purple and a blue, and you'll always get half-blue and half-purple.
With careful parent selection you should be able to put out a pack of ceeds that grow out all purple plants, they just won't be "true breeding" in the commonly accepted definition of that term (ie many or most of the offspring will not be similar to the parents). They'll be hybrids.
For clarity, I'm using a deliberately simplified example of codominance with one allele. Some of the genetic factors that control traits of interest are probably more complicated, making selection harder and expression in offspring rarer.
For example, now say to make a purple plant you need not just Aa, but AaBbCcDd. With that requirement, if you crossed two purple plants, only 1 in 16 offspring would be expected to be purple.
To this "purple" trait, now lets add a "lemon" trait that also requires a specific genotype with four heterozygous allelles (you could call it EeFfGgHh). Cross two AaBbCcDdEeFfGgHh "purple lemons" and you'd expect only 1 in 256 of their offspring to be a "purple lemon". The other 255 could range from "highly similar" to "highly different" depending on what phenotypes were associated with the various allelic combinations.
In this example the "purple lemon" pheno is a rare hybrid. Not only is "purple lemon" not true breeding, there is really NOTHING you can do to create a true breeding purple lemon line. All you can hope to do is to identify or create two parents that will always throw off purple lemons.
Actually doing it might be easy or hard, depending on your ability to do selections for A, B, C, and D. Some of those particular genotypes may not have selectable associated phenos, and then you're stuck with your punnet squares, math, and "dumb luck".
Get it?
That being said, even "stabilized", "true breeding" inbred lines still exhibit gene pool influences and are only one off pairing away from showing the entire gene pool.
Well, in a case like ECSD you're literally talking about a genetic individual. The "gene pool" in question is at MOST two different copies of each allelle, which although not "zero" genetic diversity, isn't very much.
Purity is as you say "subjective" since all cannabis in it's root state would be hemp, all forms of "drug cannabis" are the result of selective pressure going back thousands of years as your aware.
To be clear, hemp itself has also been selectively bred by man for fiber. As far as I know, nobody really knows what the "original" cannabis strain looked like or even if it still exists. But yes, in the case of a specific individual polyhybrid plant like ECSD, which is in effect just a special "mutt", talking about genetic "purity" is a bit bizarre. The whole "problem" here is that ECSD is NOT "pure", which is why recreating it in ceed form isn't a trivial exercise.
**Identical is not a worthy goal of target breeding, you may select for some traits but identical is a goal that nature doesn't allow. thank goodness ..... master race?!?!
Depends what you're trying to do, I think. For example, if you were breeding a line intended for lab research, extreme homogeneity could be useful.
I'd say that in THIS context, the goal is to create a pack of ceeds where some or even better all of the plants have flavor, scent, high, and growth characteristics approximating those of "the real sour diesel". Again, accepting certain limitations here, I definitely think it can be done.