I've not read Dr. Smart's impressive dissertation cover to cover, Charles, but I took a look at the section on land use. You are right that Dr. Smart's partitioning of the agricultural landscape into different land use patterns vis-a-vis honey bee health is exactly the kind of study I had in mind with my previous comment. His findings make perfect sense, and they are consistent with what I've observed anecdotally here in Ohio. 

It's worth pointing out an interesting paradox, I think. The types of land cover that seem to most favor honey bee success--uncultivated clover, sweetclover, goldenrod, dandelions, sunflowers, etc.--are simultaneously dependent on agricultural land use and threatened by it. These flowering "weeds" are early successional species--and in several cases "invasive" species--that thrive on the frequent and large-scale disturbance caused agricultural land use. Let an ecosystem proceed to mature forest and/or grassland, and the vast majority of floral resources would disappear.* On the other hand, if we continue intensifying our cultivation to the point of eliminating the crop margins, roadsides, and old fields that support all the weeds that bees depend on, we will obviously cause severe nutritional deficits in bees and other wildlife. 

This is what I mean by saying that blanket statements about agriculture being "good" or "bad" for honey bee health are unhelpful. Instead, let's support an approach to agriculture that values pollinator forage and wildlife habitat alongside agronomic production, maximizing the kind of pollinator forage and wildlife habitat that agricultural land use naturally provides when not taken to extremes of intensity.  

Now, let's pose the same question to the matter of urban land use. There's a dissertation topic for somebody (maybe me). 

Doug 

* I should clarify that I value mature forest and grassland very much and think they should be preserved and restored wherever possible; but agricultural is a good thing, too, and it's not going away, so let's make the most of it.
             ***********************************************
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