Ian White wrote:
"The real technical question is how many, and how long, will be "just
enough" for "here"?"

Reminds me of the trailer house designer pulling out reaces until the
whole projkect collapses, then rebuilding with only the last brace
removed reinstalled.

You don`t need any radials with a horizontal dipole. Broadcasters are
launching ground waves to reach a local audience. Amateurs may want ro
reach DX with sky waves.
For radials under an earth mounted vertical, there`s no magic number or
length. You can add to the count until resistance elimination fades. You
can lengthen them until the current at their tips is almost zero. Or,
you can just decide how much you will spend on wire. I read that
quantity of radials is better than longer radials.

If you want some idea of how conductive soil is, I read that a Variac,
ground rods , ammeter and voltmeter is the way to go. Adjust current
between the rods to one ampere with the Variac. Resistance between the
rods is then the volts between them.

Terman has already supplied Table 22-1 "TYPICAL GROUND CONSTANTS" on
page 308 of his 1955 edition for estimating dielectric constant and
ground conductivity.

At high frequencies, poor ground conductivity takes eome energy from the
reflected wave and the combined direct and reflected wave can`t be a
complete cancellation. The direct wave alone is stronger than its
combination with an out of phase wave. Ground wave propagetion fades in
a short distance at high freqiuencies. So what use does the DXer have
for soil conductivity? Isn`t something like Terman`s Table good enough,
unless he is considering laying his antenna on or near the soil?

Best regards, Richard Harrison, KB5WZI