between H8 and CaII-K lines
As noticed in Paper I, we may remark in some cases a difference
between the radial velocity deduced from 
and that from CaII-K.
Our sample is now sufficiently large to see clearly from the data of
Table 3 (click here) that this is not a general property since nearly half of
the sample does not exhibit such a gradient, including 
Boo
itself. Moreover, no correlation has been found between the presence or
absence of this gradient and one of the criteria, or suspected criteria,
such as the IR excess or UV depression. However, as can be seen from
Table 1 (click here), nearly all the stars studied are spectroscopic binaries or
have a variable radial velocity. Thus we can assume that this difference is
due to the binarity of these stars.
between two periods of observations
A variation of between two or more periods of observations could
be an indication of binarity or of pulsation. There are too few
measurements for a clear conclusion to be reached from our
observations. We can only remark that of the six stars for which we
have data in the same spectral regions at two or three epochs only
one (HD 204965) exhibits a severe change of radial velocity in one
year and one (HD 38545) a moderate change, while the other four (HD
31295, HD 34787, HD 210418 and HD 220061) are seen to undergo no,
or only a small, change.
between the photospheric lines and the circumstellar lines
We can consider the difference of between the photospheric line
and the circumstellar line for stars for which a circumstellar
component has been detected on the CaII-K line. This difference may
be interpreted as being the radial component of the circumstellar gas
movement. This gradient is positive for the following stars, indicating
a fall of the gas on the star: HD 2904, HD 16811. For the following
stars the gradient is negative, indicating an expansion of the gas:
HD 5789, HD 34787 (in 1994 and 1995), HD 141851, HD 161868,
HD 217782. The gradient is slightly negative for HD 38545. Two stars
of the 1993 mission that also show an onfall, HD 16955 and HD
204965, can be added to the first of the above-mentioned group.
Bootis stars with a CS
component
Holweger & Rentzsch-Holm (1995) have shown that the
Bootis stars with detectable CS gas are to be found preferably among rapid
rotators. This is still true if we consider our results. With the
exception of HD 204965 and HD 221756, all the stars of Table 4 (click here)
possess 
values that are greater than 100 
.
However, on the basis of their photometric data, these two stars cannot be
considered to be Bootis stars. It may also be remarked that not
all Bootis stars with a value in excess of 100 exhibit a CS component.
Holweger and Rentzsch-Holm thus conclude that is the prime
factor responsible for the presence or absence of CS absorption in
CaII-K. They also make the assumption that Bootis stars are pre-
main-sequence objects that are still accreting material left over from
the protostellar phase. Our discussion in 5.3 shows that it is difficult to
confirm or rule out this assumption on the basis of the difference of
between the photospheric lines and the circumstellar lines.