Contrary to previous spacecraft observations
which were restricted to near the ecliptic plane
where different types of flow interact, Ulysses remained for a long time in
high-speed steady flow. In such a flow, we accurately found a
radial profile of the electron density in 
, in the steady state fast
solar wind coming from the south polar coronal hole.
We find there a radial variation of the core temperature
in 
(between 1.52 and 2.31 AU from the Sun), which
is in the range of previous near ecliptic studies in the high-speed wind, near
solar activity minimum. This suggests that
the electron density and core temperature scaled to 1 AU
are independent from high southern latitudes. In contrast, they exhibit
a temporal and/or
latitudinal variation in the northern hemisphere, as noted by
[Phillips et al., 1995a]; note that these variations in the northern
hemisphere which are absent in the southern one might possibly
explain the north/south
asymmetry on the
proton temperature radial profile obtained by [Goldstein et al., 1996].
Histograms
of both parameters poleward of 
show that (1)
the scaled electron density and core temperature are roughly normal
although not exactly so,
with the four corresponding cumulants given in Tables 2 and 3, (2)
in the southern hemisphere, the histograms show a mean density
of 
and a mean core temperature of 
K whereas in the northern one
the scaled electron density and core temperature histograms are
centered respectively at 
and 
K. This indicates
an asymmetry slightly smaller than 10%, with a southern
hemisphere slightly
denser and warmer than the northern one. Likewise, various authors have
reported north-south asymmetries on different plasma parameters
[Marsden et al., 1996].
In particular, [Goldstein et al., 1996]
find
that at latitudes greater than , the average solar wind speed is 
larger in the north than in the south; this is qualitatively
consistent
with our result, given the
well-known anticorrelation of density and
speed in the solar wind.
This may be explained by a
genuine asymmetry between the north and the south polar coronal holes;
for example,
polarimetric coronal hole observations made in white light during the 1994
eclipse suggest such an asymmetry [Koutchmy & Bocchialini, 1997];
in addition, the expansion factors of
coronal field lines close to the Sun were different in the south and in the
north [Goldstein et al., 1996].
This may, however, also be due to a
temporal variation in the recorded data
since they have been obtained over a 10
months period during which the solar activity slowly decreased
to minimum.
acknowledgments The Ulysses URAP investigation is a collaboration of NASA/GSFC, Observatoire de Paris-Meudon, University of Minnesota, and CETP, Velizy, France. The French contribution is supported by CNES and CNRS. We thank P. Lantos, S. Koutchmy, and J.C Vial for helpful discussions regarding the solar corona, and also J. Lemaire and J.-L. Steinberg for helpful comments on this paper.