We have interpreted our observations as QTN in Bernstein mode, following the
works of [Meyer-Vernet, Hoang and Moncuquet, 1993] and [Moncuquet, Meyer-Vernet and Hoang, 1995]. Since, however, we are studying here a different region of the
Jovian magnetosphere, the consistency of this interpretation needs to be
further justified.
First, the assumption of plasma stability stems from an observational
fact: the ratio between the signal maxima and minima as a function of time is
roughly constant; the absolute minima take place always close to a gyroharmonic
at a level compatible with QTN, and the absolute maxima take place in most
cases at the expected Bernstein wave resonances. It is difficult to imagine
that an unstable distribution could yield the roughly constant amplitude
observed during about 5 hours. Otherwise stated, we have no evidence in our
data of distributions able to drive conspicuous instabilities, although Ulysses
was in the vicinity of the Jovian plasma sheet (at least until 13 
from
Jupiter and at 
magnetic latitude).
Second, the assumption that we are dealing with Bernstein modes
( 
) between the gyroharmonics not only explains the abrupt
drops observed in the radio spectra; it is also consistent with the observed
spin modulation of the peaks. Indeed the fact that 
has a maximum
when the antenna is perpendicular to 
(see Figures 2a and 2b) is just
what is
expected for Bernstein waves since the antenna is short for their wavelengths
( 
), so that has a maximum when the antenna
is parallel to 
[Meyer-Vernet, Hoang and Moncuquet, 1993].
This is opposite to the modulation expected for Langmuir waves ( 
) because again kL ;SPMlt; 3 (since 
where 
m is the Debye length), so that should be
maximum when the antenna is parallel to
[Meyer-Vernet, 1994].
This also explains why we do not ``see'' in our spectra a level increase due to
Langmuir waves at the plasma frequency. Such waves contribute to the minimum
thermal level, but they are damped, except at 
, whereas the Bernstein
waves are undamped between the gyroharmonics. At , owing to the above
mentioned property, Langmuir waves should be mainly detected when the antenna
is close to the direction of , which does not happen on Ulysses during
the studied period.