Repeated imaging observations have been made of NGC 3627 with the Hubble Sp
ace Telescope in 1997-1998 over an interval of 58 days. Images were obtaine
d on 12 epochs in the F555W band and on five epochs in the Fs14W band. The
galaxy hosted the prototypical, "Branch normal," type Ia supernova SN 1989B
. A total of 83 variables have been found, of which 68 are definite Cepheid
variables with periods ranging from 75 to 3.85 days. The dereddened distan
ce modulus is determined to be (m - M)(0) = 30.22 +/- 0.12 (internal uncert
ainty) using a subset of the Cepheid data whose reddening and error paramet
ers are secure.
The photometric data of Wells and coworkers combined with the Cepheid data
for NGC 3627 give M-B(max)= - 19.36 +/- 0.18 and M-V(max)= -19.34 +/- 0.16
for SN 1989B. The previous six calibrations in this program combined with t
wo additional calibrations determined by others gives the mean absolute mag
nitudes at maximum of (M-B) = - 19.48 +/- 0.07 and (M-V) = - 19.48, 0.07 fo
r "Branch normal" SNe Ia at this interim stage in the calibration program.
Using the argument by Wells et al, that SN 1989B here is virtually identica
l in decay rate and colors at maximum with SN 1980N in NGC 1316 in the Forn
ax cluster, and that such identity means nearly identical absolute magnitud
e, it follows that the difference in the distance modulus of NGC 3627 and N
GC 1316 is 1.62 +/- 0.03 mag. Thus the NGC 3627 modulus implies that On - M
)(0) = 31.84 for NGC 1316.
The second-parameter correlations of M(max) of blue SNe Ia with decay rate,
color at maximum, and Hubble type are reinvestigated. The dependence of (M
(max)) on decay rate is nonlinear, showing a minimum for decay rates betwee
n 1.0 <Delta m(15) <1.6. Magnitudes corrected for decay rate show no depend
ence on Hubble type, but a dependence on color remains. Correcting both the
fiducial sample of 34 SNe Ia with decay-rate data and the current eight ca
librating SNe Ia for the correlation with decay rate as well as color gives
H-0 = 60 +/- 2(internal) km s(-1) Mpc(-1)
in both B and V. The same value to within 4% is obtained if only the SNe Ia
in spirals (without second-parameter corrections) are considered.
The correlation of SNe Ia color at maximum with M(max) cannot be due to int
ernal absorption, because the slope coefficients in B, V, and I with the ch
ange in magnitude are far from or even opposite to the canonical reddening
values. The color effect must be intrinsic to the supernova physics. "Absor
ption" corrections of distant blue SNe Ia will lead to incorrect values of
H-0.
The Cepheid distances used in this series are insensitive to metallicity di
fferences. The zero point of the P-L relation is based on an assumed LMC mo
dulus of (m - M), = 18.50. Because this may have to be increased by 0.06-0.
08 mag, all distances in this paper will follow, and H-0 will decrease by 3
%-4%.