Yeah, my parents aren't very tall. My mom is 5'6'' and my dad is 5'10''. My grandparents are about the same height. My great grandfather was a tall ass bastard, over six foot.
Both my brother and I, and one of my cousins, are tall. I'm the tallest, but my brother is 6 foot and my cousin, who is a she, is 5'8'' if not a little taller.
For breeding, isn't that a gamble though? Relying on the great grand children to reproduce the trait? I'd figure you'd just pull the original good offspring, instead of keeping the shitty runt plants.
Unfortunately it's not this simple. In biology classes across America people have been led to believe that it's as simple as XY and xy, so that possible combinations are XX XY xx xy Xx Xy xY.
In reality it gets much more convoluted than this--and there are other things to consider such as expression.
Sometimes mutant traits are desirable when incorporated into healthy lines--other times effects can be detrimental.
A great example of this in humans is the sickle-cell trait. If a person has only one of the two alleles for the sickle-cell mutation--they have a protective effect against malaria and similar parasites.
Malaria is thought to have been the selective pressure which caused this mutation to form and proliferate (much in the same way we've created MRSA--malaria has created the sickle-cell trait, you can think of malaria as an "antibiotic" against humans).
So having one copy of the trait is very good and offers disease resistance--while having two, and thus sickle-cell disease--actually makes you very susceptible to malaria, even moreso than a normal individual (and of course it poses additional risks associated with sickle-cell disease).
So if we think of our plants as the people in this situation, we might notice one that is particularly sick (the one with sickle cell disease)--and another is doing just fine (the normal plant), however if we were to breed the two--we might find that the hybrid is twice as vigorous as the normal plant under a given set of conditions (i.e. the plant has only one of the sickle-cell alleles and is adapted to its environment). It's really up in the air.
Most of this would be moot, but for the fact that we're able to easily clone plants. Because we can do this, it allows us to probe further into what MJ genetics have to offer. If we couldn't do this, the act of breeding would be directionless.
Another way of saying things is that there's nothing to suggest that breeding some mexican brick with a bomb as strain won't produce something completely jaw dropping--it's just going to depend on the genetic profiles and how they mix.
The example you gave of short parents with tall offspring describes this perfectly. However--as with humans, many times it WILL be the case that the offspring resemble the parents. Sometimes the genetic benefit won't show itself till several generations out--and it's for this reason that successful breeders grow large numbers of plants, and generally carry them several generations out--many times randomly.
We can think of tall has being a desired trait (resin production, cola size, etc). We can't always judge the traits which will arise based on the traits of the parents--especially not without a genetic sequence and 7 million percent more scientific data on cannabis.
In essence, a blue cat and a red cat don't necessarily yield a purple cat. You have a few possibilities here:
1. Red
2. Blue
3. Purple.
and the best one:
4. Who the fuck knows.
Probably the best thing about genetics, and evolution, is that number 4 always holds true. We don't fucking know--and we're probably never going to. In terms of genetics, the answers are changing faster than we can ask them.
Life is essentially the most masterful password cracker of all time--it just keeps trying genetic sequences at random until it figures out something that WORKS in its environment, and then that thing proliferates it's nuts off.
Probably the coolest thing is how uniquely adaptive life is. For most of the species on earth--if you pick it up and move it 100miles, it will die--fast. The only exceptions really are the heartier mammals--and situations where you move a predated animal to a place where it has no natural predators and an abundance of food. If you rather introduce an animal into a similar food chain as where it came from--so long as it was not at the top of its previous food chain, it's going to have a rough time adapting.
These changes, when done naturally, occur over a great deal of time. They follow climate, chemical, and geological patterns. A cool example is the recent interbreeding we're starting to see between polar bears and grizzlies--as well as some species of whale we'd previously never seen mixing.
This is because their environments are changing such that they are overlapping now. This is where humans came from in the grand scheme of things, and MJ as well.
Something that is ALWAYS ALWAYS left out of genetic considerations for cannabis is environment--and it is the single most important thing with regard to genetic expression and genetic drift.
Think of it this way. Let's say you're trying to harvest some pollen. The TINIEST change in temperature--bumping the plant this way or that--can influence how it recombinates its dna for incorporation into pollen. This means that your offspring is not only determined by the pollen--but that also your pollen is determined by the environment.