It's unfortunate that so many people feel the need to divide antennas
like so many other things into binary categories. When the radials are
horizontal, they radiate very little. As they're bent downward, they
radiate more and more. So the transition, if you must have one, occurs
wherever you want to consider the radiation to be "significant".

But even a ground plane with horizontal radials behaves like a dipole in
some respects, like current distribution. So if you make the radials
longer, it has an effect on the current distribution that's similar to
making the top longer, but in the opposite direction. And the impedance
change caused by lengthening the radials is similar to that of
lengthening the top. So in some ways it's still a dipole.

The point I was hinting at is that calling ground plane radials "ground"
doesn't impart magical properties to them. They are conductors with
current flowing on them, so they create fields like any other conductors
with currents. And the total current flowing in the radials is equal to
the current flowing in the top wire.

Roy Lewallen, W7EL

Rick wrote:
If the ground radials go from 45 degree sloping downward to straight
horizontal, at what point does it become a ground plane and not a dipole?

Roy Lewallen wrote:
The antenna you're describing is a dipole. Half of the dipole is the
upper vertical wire. The lower half is the radials. Changing the
length of one half has the same general effect as changing the other
half.

Roy Lewallen, W7EL

David Harmon wrote:
When making the standard VHF-UHF ground plane antenna as illustrated in
all the books, with four ground radials sloped down at a 45 degree
angle, are those radials a tuned length? How much difference would it
make if they were some random length considerably longer than specified?
(as, for example, using the antenna on a different band by merely
changing the center vertical element.)