RC: Re: Re: Amblin/Racking/what ever

[Deanna German <dgerman@earthlink.net>]

Truman Prevatt wrote:
>That's exactly the type of selection, i.e. among random horses be they
with some
>Arabian, that will wash out pretty quickly. There are about 100,000
different
>genes identified in the human gnome project. I would assume that the
horse
>carries about the same number. So given not breeding back pure Arabians
into
>morgans, then it would not take long before the Arabian influence would
be
>washed out.

Huh? The number of actual genes is irrelevant. The probable percentage
of genes from a parent IS relevant.

Assume that one parent is Arab (or unicorn, or whatever). Resulting
offspring is half-Arab. Half of its genetic material is more LIKELY to
be Arab than it is whatever the other half is. Simple logic and
statisitcs show that if you cross a half to a half, you'll probably end
up with a half (that is, assuming that each had half to start out with).
If you could cross a half to a half numerous times, you'd end up with a
bell curve as Heidi described.

Explain logically or mathmatically how this washing out thing happens.
Even if you are assuming a low percentage of Arab blood in each parent
(say a quarter and a quarter), the possible percentages of Arab blood
are simply adjusted downward and the bell curve becomes more compressed.

Individuals will have more or less Arab blood, but as a breeding group,
the Arab genes stay. Each succeeding generation becomes more homogenous
with respect to each other.



>I also remember that there (at least was) is not acceptance as to the
actual
>sire and dam of Justin Morgan's horse.  So if this is the case then
that even
>makes the whole situation fuzzier. Some of you Morgan fanciers out
there want to
>comment on that.

Now this IS relevant with regard to the assumptions upon which the math
and logic is based. If no Arab ancestry, no debate. A moot point.

Deanna