The Gaia DR2 reference frame should be without relative rotation to the
quasars (QSOs) and consistent with the International Celestial Reference
System (ICRS). For the faint part of DR2 (stars with Gaia magnitude G ? 16)
that task was done via Gaia's observations of QSOs (G ? 17 mag), but the
bright DR2 (G ? 13 mag) is difficult to validate and it rotates relative to
the faint DR2 at rate of the order of 0.1 mas/yr. Very bright DR2 stars (G ?
6 mag) mostly have inferior astrometry. Here, the aim is to determine two
spin components (?X and ?Y) of the bright DR2 using International Latitude
Service (ILS, for 387 stars) and independent latitude stations (INDLS, for
682 stars) catalogs of proper motion in declination ??; both are referred to
the Hipparcos reference frame and their stars are mostly from 4 to 8 mag in
the V-band (critical part of DR2). Also, using the new Hipparcos (NHIP)
values ?? for ILS and INDLS stars, we can see that the merit of the ILS and
INDLS is the long time baseline (?t ? 90 years) important for ?? because the
standard deviation of ?? is opposite to ?t. Applying the least squares
method (LSM) to the differences of ?? between two catalogs (ILS-DR2,
INDLS-DR2, etc.), our results support the mentioned spin. The 3? criterion
and Tukey's fences method were used to reject some stars, the Abbe criterion
to explain the variability in ILS-DR2 and other ?? differences, and the
Shapiro-Wilk test to check the standard distribution of differences. The
obtained ?Y is significant at the 2 ? level, and the ILS and INDLS catalogs
could be useful for validation of the bright reference frame of Gaia DR2.