Julian O'Dea said:

> The problem with the new Esch et al. theory is that if it
> is true that optic flow is the basis for distance estimation,
> this demands great accuracy as regards direction.

I'd like an explanation of how an estimate of a distance
makes "demands" upon an estimate of direction.

Aren't the two completely isolated vectors, presented
by the dancer in two unique movements that can be
measured independently?

Regardless, why not explain the details of the mechanics
of the "locality odor" theory you support, rather than talking
exclusively about the "dance" theory you do not support?

You could start by answering some of the specific
questions I posed in a prior message in this thread,
addressing each point in turn.  I'd sure like to know,
because the conditions I cited are common daily events.

As for dance, any error in distance or direction, unless
massive, would be acceptable due to the simple fact that
flora tend to grow in groups.  Although an error in direction
would get "worse" with longer distances, a 5 mile flight
with a one-degree error would put the bee "off target" by
a mere 460 feet.

Below is a table of distances, degrees of error in
direction, and maximum number of feet "off target"
as a result:

  Distance   1 Deg   2 Deg    3 Deg error
  1 Mile        92 ft    184 ft    276 ft
  2 Miles     184 ft    368 ft    553 ft
  3 Miles     276 ft    553 ft    830 ft
  4 Miles     368 ft    737 ft   1106 ft

While it is obvious that larger direction errors over longer
distances result in larger drifting of "target points", note
that the more reasonable foraging distances where bees
are easy to find most any day (less than 3 miles) allow for
quite a bit of error in direction without enough "target drift"
to matter if one is speaking of stands of trees, fields of
clover, and other flora.

Errors in distance also have minimal impact due to the
fact that a bee dance represents distance by a "length"
of a short "run" of a bee.  Therefore, the error can only be
expressed as a percentage, and when the distances are
small, so are the errors.

 Another table, this time distance percent error, and
feet "off target":

    Distance     1%       2%      3%       4%      5% error
    500 feet      5 ft     10 ft     15 ft       20 ft        25 ft
  1000 feet     10 ft     20 ft     30 ft      40 ft        50 ft
      1 Mile      52 ft   105 ft   158 ft    211 ft       264 ft
    5 Miles     264 ft   528 ft   792 ft  1056 ft     1320 ft

It follows that far-off sources will be rarely exploited as well
as nearby sources, simply due to "target drift" from errors
in measuring or communicating distances.

It also follows that most bee dances will be "about" distances
that are within a few miles of the hive, since more foragers
will find the advertised source, return to the hive impressed
with the forage, and dance themselves, recruiting more foragers.

The above may explain why so many experts share the opinion
that most foraging takes place within a few miles of the hive.
Longer distances, combined with imprecise measurement and/or
communication, result in more foragers getting "lost", and returning
without gathering anything.

It also should be noted that feeding stations are the worst-case
"small target" for bees.  They are unnatural and unrealistic, in that
there is no such thing as a single plant at a single point that can
be a worthwhile source of forage for a colony.

So, I'd guess we should scrap the feeding stations, and start tagging
the bees with small RF-reflective targets to correlate to maps of what
grows where.  Jerry Bromenshenk may have lots of this sort of data
already, as I recall he was tracking bees with radar.

In short, yes, bees CAN loose their way, but only when longer
distances are involved, and target areas are small.  The evidence
of common experience seems to support a conclusion that there
is error in both measurement and communication.

Live with it.  The bees seem to.

        jim