It would probably take many thousands of plants to really develop, but I believe it is equally as possible to target traits backwards as forwards. Neither is perfect, but the former has the advantage of a reference point.
For me it started with observations rather than ideas.
The recurrence of selecting over generations and then suddenly there will be a plant that is identical or very near identical in everyway to a parent a decade ago or 5 gen back. As if everything that went into it recent crosses becomes recessive. Skipping multiple generations in the form of an outlier.
I’m just running with these observations and the back drop of experience I have while keeping open to new information. It is the framework of where I apply all of my time and resources because it is working so far and or adding knowledge.
So far the greatest success in this was crossing the 2 outliers in the 88 to make Alpine 1.0. the line as a whole was not good. Most of the plants I didn’t even try smoking. It was consistent with the Ammonia catpiss Nevil noted in inbred haze lines.
A few observations came out of that cross that do not align with common linear breeding perspectives. These perspectives are mostly from continually selecting the same traits of parents, not playing off the variable of the outliers which is central to the concept.
Both parents. Silk S and Buzzsaw were definitely outliers and had Nevil’s haze traits which were within the line, but not forward in this line or the other Nigerian Haze lines I ran aside from perhaps trichome production. If anything it was selected toward the Nigerian Silk side.
Also consider how long did it take for the original 5haze to become completely unrecognizably altered once it left Nevil’s keep. 1 year? A single cross, 2? Why wouldn’t the opposite be possible? Also consider the stage Nevil found his Male A and C. It was said that haze had already lost its spark. I believe there are cycles that are beyond the linear one generation at a time sequence.
The cross resulted in a very Nevil’s haze dominant line across many visual markers as well as profiles. The cross was completely different than the line it came from. The intersex traits were nearly wiped out. It effectively moved the whole prior population into recessive.
Vigor was increased through the cross. This was line work and generally we have the preconceptions that this would be neutral or a loss rather than an increase.
There are numerous other aspects and observations, but I don’t like wasting my time and outlier selection to revive ancestral traits is proving effective enough to give it my full confidence and years of practice.