>Those colonies with the heaviest mite loads died first, and they failed to
move upwards as they consumed honey.  That didn't happen in the controls
and colonies with the lowest mite loads.

I re-read Heran (1952) who conducted temperature preference studies at my
alma mater in Graz, a stone-throw from where I am sitting right now. He
mentioned as a side note, that hunger increased the temperature preference
by 2.47°C. So maybe Varroa is messing up the nutritional status perception
of individual bees by feeding on them or diminishing their body reserves?

For certain varroa is affecting the thermal fluctuations in the cluster. (
https://doi.org/10.1603/AN10142) Maybe by offsetting the target value for a
subpopulation of bees?

60 years later at the same institute Heran's work with temperature
gradients and honey bee aggregation was continued, (
http://zool33.uni-graz.at/artlife/bee https://youtu.be/q3AXjJGwlao ) where
they found that clustering behaviour could be simulated by the following
algorithm:
1. move randomly preferably towards T-gradient.
2. if you meet a nest mate stop
3. Compare ambient ot preferred temperature and stay in place longer the
smaller the difference.

My working theory at a possible synthesis of those findings is that
"hunger"/decreased nutritional status is normally a mechanism/trigger to
start clustering in fall (gravitate towards center or deter straying from
group) that is balanced against what Heran did find lowers the T preference
of individuals: CO2 exposure and previous exposure to low temperatures.
CO2(active heating with flight muscles or prolonged stay in cluster
center), previous low T exposure (having been on outside of cluster
previously).
Or simply put, the lower the perceived gradient within the cluster, the
more mobile the cluster as a whole is. With brood present or too many bees
having an artificially raised t-optimum the t gradient and therefore
cluster mobility drops so far that contact to honey reserves can be lost.

Similar to the setup here https://youtu.be/lTUypl1Ub9U where the "social
gradient" ankers the "cluster" in place in suboptimal conditions.

Of course there is also the possibility I am trying to see
connections/patterns where there are none :-)

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