Dick, you bring up excellent questions on bee breeding.

In the first place, there is little reason to select for queen traits (so
long as she lays a good pattern)--what you select for is colony traits.
 Half of those traits come from the 15-40 drones that the queen mother
mated with.

When you graft from a queen, you only get half the genes of the queen, plus
the genes of only one of those 15-40 drones.  Any single daughter queen
only carries a fraction of the genes of the successful parent colony (on
average about 33%).  And then if you mate that virgin out to a different
population of drones than her mother mated with, you  further reduce the
proportion of alleles compared to the parent colony.

A mated daughter of your best colony can theoretically share only 12.5% of
the alleles of the parent colony.  This is an inherent mathematical problem
involved in trying to breed a small population of bees.  Far better, IMHO,
to forget individual queens, but rather to work with the entire bee
population in the mating range of your breeding program.

> How many daughters would you need to study in a control group to find a
> valid superiority?
>

Typically, you'd need at least 12 in each group to detect a substantial
difference.  The more the better!

>I suspect that nature made drones haploid as a way of controlling for
> some sort of wild variability.


Honey bees have the highest genetic recombination rate of any known
organism.  Forcing drones to chase down virgin queens, in competition with
each other, selects for advantageous recombinations, and eliminates
unfavorable combinations.  This strategy speeds up the process of natural
selection over the typical diploid/diploid breeding of vertebrates.

In addition, the forcing of drones to exhibit resistance to disease from
parasites (such as nosema) rapidly selects for overall resistance to
parasites in the bee population.

Randy Oliver
Grass Valley, CA
www.ScientificBeekeeping.com

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html