Steve wrote in message ...
A couple of points I thought of last night but it was near midnight and
I had written more than enough:

1. Ted, I read your post quickly. I don't have time right now to go back
through it carefully but I think there is one point you didn't take into
account. In nature, all your numbers may be valid. In the world of
orchid hybridizing, each plant doesn't have its chance to stay in the
gene pool. In fact, going from, say, the third to fourth generation we
may select a single parent to go on to the next level. The cross MAY
never be made again with a different parent.
By the third generation every plant has major chunks of any one ancestor
missing. Your numbers, the odds, may still be valid except that we are
selecting for certain traits, not the average plant. The plant selected
for further breeding is probably the one with the less likely
combination of genetics.

Hi Steve,

Actually, my numbers apply only to the probable genetic relationship
between a given individual and its grandparent. To get numbers
applicable in either the wild or in a captive population, one would
have to take into account a wide variety of other issues, including
but no limited to selection pressure, genetic effects on survival and
reproduction as influenced by a host of environmental parameters. I
had said nothing about survival, or indeed anything about gene
frequencies as affected by selection (natural or artificial). We must
remember that the probabilities that can be used to assess likely
genetic relationships between individuals of a family are only a
subset of those required to model gene frequencies in a population. I
looked only at probabilities associated with independant assortment.
This tells us precisely nothing about how gene frequencies will change
through time.

Cheers,

Ted