New radial velocities and uvby photometry for Polaris are presented. T
hese reveal that the pulsational amplitude of Polaris had declined to
0.59 +/- 0.19 km s-1 and 0.010 +/- 0.002 mag in V by 1992. Comparison
with earlier determinations reveals an exponential change with time ra
ther than any sudden onset of the decline, and suggests that Polaris h
ad an amplitude near 0.12 mag (6.0 km s-1) prior to 1900. There is no
sign of the amplitude changing in a manner consistent with a beat phen
omenon; instead, the star seems destined to stop pulsating in 1994. Ou
r data add new points to the O-C diagram of Polaris and show a continu
ation of the period increase known previously. We point out that depar
tures of this diagram from a true parabola do not necessarily imply th
at factors other than stellar evolution are at work. Finally, we lay t
o rest the myth that the amplitude of Polaris has declined because it
is evolving across the red edge of the Cepheid instability strip. We s
how Polaris to be well within the strip and surrounded there by Cephei
ds of substantial amplitude. In particular, Polaris is virtually ident
ical to RT Aur in M(V) and color, yet RT Aur has an amplitude of 0.8 m
ag. We conclude that there is as yet no adequate understanding of what
sets the amplitude of a Cepheid.