W. Watson wrote:
There's a figure at the bottom of
http://www.tpub.com/neets/book10/42.htm that shows current
distribution on
an antenna. Is there some book or source that describes how figures B and C
and produced?

A lot of authors of antenna and electromagnetics texts avoid the issue.
There's a decent analysis in King, Mimno, and Wing, _Transmission Lines,
Antennas, and Waveguides_. The problem is much more difficult when the
antenna is longer than a half wavelength or when the wire diameter
becomes significant. In those cases, a much more complex calculation is
necessary, with only approximate results being available from closed
form solutions. Numerical solution of a triple integral equation is the
method usually used. That approach is necessary when there are nearby
current-carrying conductors, since that alters the current distribution
in ways which are usually too complex for other methods. A brief
discussion of several methods of approaching the problem can be found in
Sec. 14.11, "The Cylindrical Antenna Problem", in Jordan and Balmain,
_Electromagnetic Waves and Radiating Systems".

I'm guessing that one starts with an open ended transmission line, and
looks
at how the E and B fields are distributed along it. Then one peels the open
end back until the two lines are pointing away from one another. One then
ends up with E fields pretty much in a plane directed from one point on one
side of the wire to the similar point on the other wire. Similarly the B
fields encircle each wire. On one side they have a CC direction and the
other a CC direction (looking in the direction of the current).

Using a transmission line as an analogy for antenna operation works just
well enough to be dangerous. While it's a way to get an intuitive
understanding of antenna operation, taking it too far can lead to some
erroneous conclusions. One fundamental limitation is that classical
analysis of a transmission line depends on the assumption that no
radiation occurs, and that's certainly not the case for an antenna.

As a separate item, it would also seem that for a transmission line that is
open ended the current flowing in one wire induces a current to flow in the
opposite direction in the other wire. That is the current flows in the one
wire because it does so by induction rather than it being physically
part of
the other wire.

Yes, that's correct. If you assume that the entire field from one wire
couples to the other, which is strictly true for coax and approximately
true for twinlead, it follows from Ampere's law that the current on one
wire must be equal and opposite that on the other. This assumption of
field coupling isn't true for an antenna, although coupling does indeed
exist and it does have a profound effect on the currents.

Roy Lewallen, W7EL