> the only work I have seen where they have addressed the 
> question of honey bee preference for nesting sites all agrees 
> on this point: they have no preference re: tree type.

I'm sorry, but to be crystal clear **NONE** of the works cited in this thread made ANY conclusions about "tree preference".
To state that these works support the claim that "bees have no tree preference" is flatly incorrect.

a)  The Ithaca paper had far too small a dataset (39 nests) to yield any insight on the issue of tree species, and overtly DID NOT offer any such conclusions .

b)  The CT paper stated that of 108 trees with bees, 36 were maple, and 27 were oak, and also DID NOT draw any conclusion about tree species "preference".

c)  The Polish paper cited in the article posted found that the only significant factors were "trunk size" or perhaps "size" with "health".  There's very clear data, that when subjected to AIC ("Akaike Information Criterion"), show that the data collected is INSUFFICIENT to say anything about tree species.  See table 4.  The "wi" (weight) column means what it says.

The Polish article that referenced paper (c) stated 23.6% hollow trees among 15,000 trees, which is a unusually large percentage.  With such a wealth of cavities, one would have a wealth of choices, and the bees would likely be better served by the more basic consideration of "cavity size", per the Seeley criteria, as cavity size was found to matters more than even entrance size/direction, as I remember.  


To summarize, men offering NO OPINION on the issue of "tree preference" does not allow one to conclude that bees have no "tree preference". 

> Further, my motto is absence of proof is never proof of absence. 

But here there is a total absence of evidence, not evidence of any absence.

> If anyone has anything to the contrary, they should post it.

No, the burden of proof is to show us something to support the divergent viewpoint proposed, not show us things that overtly DECLINE to opine one way or the other, each for very valid statistical reasons.

             ***********************************************
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