The Waggle Dance of Bees - A Symptom, Not a Signal?

We have had a lively exchange this past couple of months on BEE-L about how
recruited bees find a new source of food or how swarms manage to move to a
new location.  The swarm relocation problem allows us to assess the
relative credibility of the odor-search and bee language hypotheses.

As we all know, when a swarm issues from a hive it most often clusters on a
nearby object (e.g., tree, bush).  Experienced foragers inspect potential
nest cavities, return to the swarm cluster, and execute waggle dances.
Eventually, the swarm takes flight and travels to only one of the
prospective sites.

Consider first an odor-search model.  Scout bee activity actually occurs
not only on the swarm cluster but also is quite frenzied at each potential
nest cavity.  Those of us fortunate to have watched that behavior at a
potential destination know that scouts repeatedly expose their Nasanov
glands, both inside and outside those cavities.  Some of them also execute
waggle dances on flat surfaces near cavity entrances.

How does the swarm eventually manage to move to only one of those cavities
instead of splitting, with some portion going to each of the sites?  An
explanation that relies solely on odor is very simple.  With experienced
bees flying back and forth between cluster and likely sites, one of those
sites becomes easier to locate for other recruited bees from the swarm
cluster.  They, then, also expose their Nasanov glands at such a site.
That odor, drifting downwind, provides a "beacon" for newly recruited bees.
Searching bees can execute a zigzag flight upwind once they get within the
downwind odor plume.

In time, the easiest of the sites to find has many more bees exposing their
glands at that site than at the other sites, etc. ("positive feedback").
Eventually, the less favored sites get no new recruits at all and are
abandoned.

When many bees frequent a single site, and weather permits, all scout bees
from that site return to the cluster, the swarm cluster disbands, and the
swarm is led through the air by the few hundred bees who know the way.  The
experienced travellers do so by opening their Nasanov glands as they
repeatedly fly out ahead of the swirling bees in flight.

One can find a partial description of this process in the following
article:  Wenner, A.M.  1992.  "Swarm movement: A mystery explained."
AMERICAN BEE JOURNAL.  132 (1):27-31.

James Cowan, a perceptive beekeeper in Aberdeen, Washington, later provided
additional information in his letter to me of 13 January 1992, as well as
in his letter on page 819 in the last November issue of ABJ and in another
letter to me this month.

   How much did research and writing of that article cost beekeepers?  Nothing.
*********
Consider now the bee language explanation for the same event.  In his
posting of January 15th, Peter Borst emphasized "decision making" by
dancing bees on the swarm cluster and included the following comment from
one of the publications (*3) that he referenced:
*********
"The main focus of the article is on how a decision is made as to which
site a swarm will go to among the several choices located by the scouts.
The way in which the bees communicate information about the site is as
follows:"  'Scout bees fly throughout the surrounding countryside,
searching for new nest-site cavities. When a scout returns after inspecting
a high-quality cavity, she performs waggle dances which encode the distance
and direction to the site. Most bees that danced for nest sites also
followed the dances of other scouts.'  The method of communicating
information is similar to that used by foraging bees to recruit more
foragers to a productive site."
*********
On February 1st, he expanded on the theme of "animal consciousness," a
belief system embraced by Donald Griffin (ANIMAL THINKING) and also by
Richard Dawkins in his popular works.

Can scout bees "build consensus" on the surface of the swarm cluster?  Some
folks think so, as in a recent publications by a group at Cornell: "Group
decision making in swarms of honey bees"  (1999. BEHAVIORAL ECOLOGY AND
SOCIOBIOLOGY. 45:19-31).  In the authors' words, it was a "...study that
renews the analysis of honey bee swarms as decision-making units."

Their interpretation of what happens among scout bees on the swarm cluster
(in part):

   "...in consensus building ... bees that dance initially for a non-chosen
site [tend] to cease their dancing altogether, not to switch their dancing
to the chosen site..... even though a swarm is composed of tiny-brained
bees it is able to use the additive weighted strategy of decision making
because it distributes among many bees the task of evaluating numerous
potential sites and the task of selecting one particular site for its new
home."
*********
Nowhere in that publication could I find any mention of the very visible
Nasanov gland exposure at the prospective cavities, even though they wrote:
"Once the scouts have completed their deliberations, they stimulate the
other members of the swarm to launch into flight and then steer them to the
chosen site (...Michener 1974; Seeley 1982; Winston 1987)"  Nor could I
find any mention of the wind direction that prevailed during their
experiments.

Why did they not mention my 1992 article that appeared in the AMERICAN BEE
JOURNAL (an article that includes much of the information at the beginning
of this posting)?  I strongly suspect it was because they do not want to
draw attention to any element that does not fit with their favored
hypothesis, the notion of bee language.
*********
My overall interpretation of all of the above?  Swarm movement depends on
Nasanov odor produced by scout bees.  That odor drifts downwind from
suitable cavities and also enables scout bees to lead swarms through the
air to the new site.

The dance maneuvers on the surface of the swarm clusters and at the new
site are thus merely a SYMPTOM (not a cause) of what is happening during
swarm relocation.  Julian O'Dea hinted as much when he termed the dance
"idiothetic behaviour" in this extended exchange --- as Rosin did recently
in the AMERICAN BEE JOURNAL (p. 98 in the February issue).

Bee language proponents thus concentrate too much on a study of the symptom
(i.e., waggle dance), rather than on the total event of swarm movement.
They ask us to believe that a group of scout bees can form a consensus on
the surface of the swarm cluster, when a simple Nasanov (and odor-search)
explanation for swarm movement is sufficient.

Having made the comparison, we can ask:  Which is the more credible
hypothesis?

And how much did the complicated research on "decision making" cost the
public?  We cannot know for sure, but that Cornell research was supported
by grants from both the National Science Foundation and the U.S. Department
of Agriculture.  Did taxpayers get their money's worth?  That is for all of
us to decide.

Swarming season is upon us.  Hopefully, some of you may find time to repeat
James Cowans' careful observations (as covered in his November letter and
in his letters to me directly).

                                                                Adrian

Adrian M. Wenner                    (805) 963-8508 (home phone)
967 Garcia Road                     (805) 893-8062  (UCSB FAX)
Santa Barbara, CA  93106

********************************************************************
*
*  "When we meet a fact which contradicts a prevailing theory,
*     we must accept that fact and abandon the theory, even when
*     the theory is supported by great names and generally
*     accepted."
*
*                                       Claude Bernard --- 1865
********************************************************************