My contention was that there was no vertical component in the dance
language. I never said bees don't forage in trees! Obviously, bees
find flowers in trees and they do this in part by their sense of
smell. As I said before, if they fly above the canopy, they would have
no trouble locating flower patches both visually and by the odor.
Also, it is known that if the nectar  plants are especially abundant
and odoriferous, the information in the dances is generally ignored.
(see reference at end of email)

Many years ago I went hiking on Mt. Shasta using only a map. This was
not a topographical map, in that it did not indicate elevations, but
only features such as streams and a few peaks and ridges. We had
picked out a stream to camp by and proceeded to attempt to hike there.
When we got to the actual location, we saw the stream from atop a
cliff, in the bottom of a thousand foot chasm. We ended up camping on
a barren mountainside with only melted snow for water, unable to reach
the stream below.

Without the third dimension, the map failed at that point. A flat map
is useful for many or most situations but it would fail to show
whether a given address, for example, was on the ground or 50 stories
up in an apartment building. So in addition to the street address, one
would need the floor number (vertical component) to find the
apartment.

If one can visualize what a bee sees in traveling out, one can see
that given distance and direction, it could find most types of forage.
Just getting nearby would be adequate, if the floral source was
fragrant and/or abundant. The feeding source on the tower was neither.
The bees flew near the ground and could not find the feeder at the top
of the tower. Probably if the feeder was at the bottom of the forest
floor where the bees had to fly over the canopy, they would have the
same problem.

The point I was trying to make was that in certain situations, a two
dimensional representation will fail. It failed for the bees at the
tower and it failed for me at the precipice. James Nieh's work is very
interesting but wasn't about Apis mellifera at all, so I don't see how
it is relevant, except to show how the height information *could* be
encoded into a bee signal. So far as is known, it isn't present in the
Apis mellifera dance. From his web site:

>  Melipona panamica foragers can communicate the three-dimensional location of food sources. To achieve this, foragers use a combination of mechanisms. Results from a series of removal experiments (segregating all feeder-experienced foragers from potential recruits as they left the nest) suggest that direction is communicated outside the nest whereas height and distance are communicated inside the nest.

>  A recruiting forager produces a series of pulsed sounds when she unloads her food to other bees and when she begins to make clockwise and counterclockwise dance movements (Nieh 1998B). During the food-unloading phase, she produces longer sound pulses for a food source on the canopy floor than for one 40 m up in the canopy. During the dance phase, sound pulse duration is positively correlated with increasing distance of the food source from the nest. Thus M. panamica foragers appear to use sounds to communicate food height and distance.


* * *

Honeybee colonies achieve fitness through dancing
Gavin Sherman & P. Kirk Visscher

During summer and autumn, there was no significant
difference in the food collected, as measured by mass changes, when
colonies did and did not have directional information from the
dance.

However, in conditions like those in winter in this study, communication
of food source location does increase a colony's food intake.

The evolution of recruitment communication in social insects is
presumably steered by whether, for particular colony sizes and
habitats, a recruitment mechanism increases food collection more
than it decreases it.


-- 
Peter L. Borst
Danby, NY  USA
42.35, -76.50

******************************************************
* Full guidelines for BEE-L posting are at:          *
* http://www.honeybeeworld.com/bee-l/guidelines.htm  *
******************************************************