Reg Edwards wrote:
Roy, surely you realise that all depends on soil conductivity and
permittivity which B, L & E forgot to determine before leaving the
site.

It just isn't sinking in, is it? It depends on the conductivity and
permittivity to a skin depth or more, which was impossible for them to
determine. A surface measurement wouldn't have provided the necessary
information. Measurement of ground wave attenuation to another location
would have included ground with a variety of possible characteristics
different from those in the immediate vicinity.

30 years previously, around 1905, Sommerfeld (and others) had produced
a significant report showing the importance of ground characteristics
on radiation and propagation at LF and below. Which B, L, & E ought
to have been aware of if they had known what they were about.

I'm sure they were. But the paper has nothing to do with propagation.
What makes you think it does?

They certainly did know about the effect of conductivity. In a
theoretical analysis at the beginning of their paper they calculated
expected radial ground currents for several different ground
conductivities, and explain how current is distributed in the ground
with conductivity being a factor. The radial ground current analysis was
later found to be in error(*), but it's still considerably closer than
the results I've seen from your analysis and program.

You should concentrate your thoughts on HF and above, not on LF and
VLF distractions. Different characteristics prevail at HF at which
frequencies amateurs are most concerned.

That's a bizarre admonition from someone constantly harping about how
many radials American AM broadcasters use. I am indeed most interested
in HF, where skin depth is on the order of 12 feet for average soil, and
measurement of surface conductivity and permittivity is pretty useless.
Incidentally, since you haven't read their paper, you probably don't
know that the BL & E measurements were done at 3 MHz, which is HF.

(*) To my knowledge, no one has developed a method of calculating radial
currents or ground system losses with even approximate accuracy other
than with numerical analysis such as used by NEC-4. Many years ago I
spent a couple of years of spare time in a technical library looking for
just such an analysis without success. Reg's method is delightfully
simple but gives results which are very wrong.

Roy Lewallen, W7EL