I'm not trying to be a dick here, but I am pretty sure that none of Subcool's strains are actually F1 hybrids.
I'd certainly hope not, and just to be clear, when I gave my description earlier of "pollen chucker", I didn't mean to imply that all Subcool does is take two lines cross them, and sell the seeds.
I'm pretty sure he is still doing some selection, even if its clear he's not doing it over many generations to stabilize the offspring.
An F1 hybrid is very strictly the combination of two true breeding lines, which results in a plant with all dominant genetics, and the justly revered 'hybrid vigor'.
Well, again, just to be clear here, in classical genetics nomenclature the term F1 is just a way of describing the offspring of any two parents (hybrid or inbred). It refers to any member of the first offspring generation of any cross.
The F1 is typically a hybrid (if the parents are different) but it might not be (if the parents are similar inbred lines).
You're right that typically the term "F1 hybrid" is used specifically to refer to the offspring of two true breeding lines, but I don't think it absolutely has to. It depends on the context.
Anyway, this minor semantic quibble aside, I agree entirely. The "best" F1s come from crossing two inbred lines, and optimally not only will the offspring all look alike or nearly so, but they'll hopefully express the best (or at least some of the best) traits from each of the two parents.
When you cross non-stabilized lines, which is generally any hybrid less than F7, the result is actually an F2 hybrid, even if it is first generation in this specific circumstance, and this generation (and all those after) will not have hybrid vigor.
Again, in classical genetics nomenclature, F2 refers to any second generation cross (regardless of the genetic lineage of the parents). Its just a frame of reference from your original crosses.
What degree of hybrid vigor (or "heterosis", to use the fancy genetics term) the direct offspring of hybrid plants will have will be dependent on the exact genetics involved.
For example, if each of the original parents are themselves perfect F1 hybrids each formed by crossing two unique inbred lines, and you were to cross these two parents, each of their F1 offspring could look totally different. . .yet each would still be expected to have hybrid vigor.
In other words if you were to cross (AxB) x (CxD), where A-D were each unique true-breeding strains, every relevant gene locus in the offspring would be heterozygous and you'd expect hybrid vigor in these plants.
Forgetting for the moment what filial generation (F1, F2, etc) you "should" call these mutts, this is where the phenomenon of polyhybrids comes into play. Generate enough random variation with enough genetic potential, and sure. .. some of these plants "should" be absolutely outstanding.
Real seed breeders, if they want to breed from hybrid stock, stabilize the hybrid first to arrive at a true-breeding parent line, and it is from the cross of these stabilized parent lines that will result in an actual F1 hybrid, which generally show relatively little variation (usually 2-4 phenotypes) and contain no paired recessive genes. F2 generations will show the most radically different phenotypes, as they will have the most genetic variation, and will be the first generation to contain paired recessive genes.
Agreeing with this, the problem here is that some of the more interesting phenotypes REQUIRE heterozygosity (kind of like green eyes), and therefore CANNOT be stabilized into true breeding lines. This is part of the reason elite "clone only" strains exist. . .you simply can't backcross them to stabilize them.
Yes, in theory it should be possible to create unique true-breeding parents that when crossed will reproduce ANY specific genotype (and therefore phenotype) that you like, but in practice, actually doing this can range from difficult to practically impossible.
Since some of the genetic factors that create interesting plants may not be externally visible, it simply may not be possible to select for them using conventional breeding techniques. Also, not every phenotype is controlled by simple Mendellian genetics where one gene locus controls one phenotype. Some traits may be controlled by multiple genes that can interact in non-obvious ways, again confounding efforts to do normal selection.
With enough genetic knowledge, and enough selections, and enough generations, these sorts of problems can be overcome to some extent, but again, this sort of thing is what separate the the "real" breeders from the wanna-bes.