Thanks for the interest in this issue.

Steve Olney wrote:

We tend to
concentrate on radiation resistance alone, when it is the ratio of
radiation resistance to losses which determines how much goes into ether
and how much warms up the local environment.

In an article in the Swedish magazine QTC mr Dagaas, chief engineer at SAQ, 
briefly explains the working principle of the MTA:
A single vertical with top loading and earth system would have a feed impedans 
of approx. 2,5 ohm.
If six such antennas are parallelled, the total loss resistance becomes 1/6 
while the radiation resistance remains unchanged. The total impedance is now 
0,45 ohm due to reduced losses.
The antenna input power is fed by 1/6 of the total current (feed impedance is 
raised by a factor 36).
By varying the current distribution between the verticals (by changing the 
tuning coils) a suitable feed impedance can be achieved. 20 ohm in the SAQ case.

It seems like the ratio of  Rrad/Rloss is increased. Corresponding thougths can 
be found in the ARRL Antenna book edition 17 p4-21.

In parallell with this discussion on the LF reflector, the subject has been 
treated in the news group rec.amateur.radio.antenna.

/Christer