For what it's worth ....
The comments below regarding liklihood of hybrids / integrades with a
more migratory sp being more likely to occur out of range remind me
of the situation in AZ (where I lived 'til recently) regarding R-B
Sapsucker. The majority of apparent R-B Sapsuckers turned out to have
various signs of hybridization / integradation with R-N or Y-B, even
though pure R-B breed much closer to AZ than do R-B integrades.
Presumably, R-B with some R-N or Y-B genes were much more prone to
long distance migration and vagrancy, R-N and Y-B being much more
long distance migrants than R-B. The parallel with G-W Gull x W Gull
is prob a good one, since both the sapsuckers and these particular
gulls do have substantial integrade / hybrid populations where the
different "species" come into contact. As others have pointed out,
GBBG hybrids are much rarer, but presumably the same principle could
apply; i.e. that a GBBG with some (for example) Herring genes might
be more prone to long distance migration and, hence, vagrancy.

>The simple odds is not the argument here. I would think that the argument
>is that another species (e.g., Herring Gull) would introduce genes with
>different migration length, strategy, timing, etc., that might make a hybrid
>MORE likely than a typical GBBG.

...

>Another example: Though we have not ever recorded a gull in Colorado that
>has Western Gull genes (either pure or otherwise), which do you think more
>likely: a pure Western Gull with those fairly sedentary genes or a Western x
>Glaucous-winged hybrid that has those long-distance migrant Glaucous-wing
>genes? I know which I'd pick and, being on the Records Committee, I'd want
>to be absolutely positive that the first CO record of Western Gull was a
>pure one.

Cheers
Steve

--
Dr Stephen Barlow,
Senior Research Scientist
School of Chemistry and Biochemistry,
Georgia Institute of Technology,
Atlanta, GA 30332-0400, USA
Phone: 404-385-6053
Fax: 404-385-6057
Email: stephen.barlow at chemistry.gatech.edu