Notes on Terry's observations
Some signal generators are altered by the circuit under test especially at
resonance as the current drawn increases sharply. Assuiming you were using
a good low impedance instrument as you indicated the readings are valid.
A 1/2 wave resonators peaks I (current) at resonance while a 1/4 waves
resonator peaks voltage at resonance. The secondary coil acts like a
transmission line if the upper terminal is small to medium size.
If you connect a sensitive current meter in series with your signal
generator as you take the measurements you will see the 1,3,5 multiples
peaking while the voltage peaks from your cell phone antenna are peaking at
1,2,4,6. This is normal and to be expected.
The toroid reacts with the secondary distributed capacitance which is
comprised of the many small series capacitances between turns. The
harmonics are supported by the turn to turn capacitances, ie, distributed
capacitances. This effect is what produces all the multiple harmonics you
are observing. If you connect a super huge top load capacitance you will
see only one fundamental with any multiple resonances produced attenuated
down by 10-20 dB, and, in effect, do not become part of the active
secondary circuit due to their attenuation. This is why TC tuners should
always use a very large topload, ie, to prevent all the undesireable
multiple harmonics that cause strange effects which some experimenters call
"racing sparks" along the secondary coil. Use a huge topload and "racing
sparks" will disappear altogether assuming the primary is in proper tune.
Racing sparks are caused by too small topload and multiple series
resonances with the distributed capacities along the secondary inductor
reacting with the topload capacitance.
Hope this info helps you out with your experiments.