... (1996)^3.1

Stein (1996) only identifies the galaxies by his own sequential ID numbers and the position $(\alpha,\delta)_{1950.0}$. Cross-identification with the Richter (1989) `R' ID numbers for the 70 galaxies selected as described above was made by forming all possible (Stein,Richter) galaxy pairs. For each pair the coordinate distance between the two galaxies in the pair was computed, and the pairs were then sorted after coordinate distance. 37 pairs had a distance of less than 4''. The following pairs had distances of 5.3'', 6.0'', 13.3'', and 24.5''. The first 3 were clear matches, since the heliocentric velocity from Stein differed by at most 2% from that of Richter. The fourth was a clear mismatch, since that Stein ID had already been matched to a Richter ID further up the list. Thus 40 galaxies from the Stein catalog were also in the above subset of the Richter catalog. The established cross-identification for these 40 galaxies can be seen in Table 

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... stars^4.1

Poulain & Nieto (1994) write: ``Foreground stars, when visible through the eye-piece, were measured through the smallest aperture available (10.9 arcsec [in diameter /BMJ]) and their contribution subtracted at a later stage.'' In their Table 1a, they note whether the given galaxy had one or more stars subtracted. The two galaxies in common with our study had no stars subtracted.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... stars^4.2

Burstein et al. (1987) write: ``Separate observations of contaminating stars are noted [in their Table 2 /BMJ] by asterisks in the columns headed by ``Run'' and ``N'' (notes); star observations always refer to the immediately preceding galaxy observation.'' One of the four galaxies in common with our study, R256, had one star subtracted.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... stars^4.3

Sandage (1975) writes: ``Contaminating stars were were present in some of the apertures as notes in Table 1. Measurements of the stars alone, using smaller apertures, are listed where appropriate. Contamination-free colors and magnitudes were calculated for the final reduction.'' One of the two galaxies in common with our study, R256, had one star subtracted.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... ftp^5.1

ftp://samuri.la.asu.edu/pub/redcats

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... spectroscopy^6.0

More precisely: what was used was the section [282:682,1:1024] of the photometry bias and dark images.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... spectroscopy^6.1

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... slit)^6.2

Note, that shifting the slit profile image is not strictly correct, since it contains the y-average of the pixel-to-pixel variations intrinsic to the CCD. However, the error is negligible.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... km/s^6.3

Throughout this section, if nothing else is stated, the conversion from Å to km/s is done at 5177 Å, the (rest frame) wavelength of the magnesium triplet.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... interval^6.4

To get a rest frame wavelength interval that is common to the galaxy and the star, their relative radial velocity needs to be known. The program uses the input guess on the radial velocity to do this. We computed an accurate such input guess by cross-correlating the two spectra using the task stsdas.contrib.xcor.

A common galaxy/star rest frame wavelength interval of the same length was used for all the galaxies. It was defined by always skipping the 130 most blue columns in the galaxy spectrum, and then skipping the number of columns in the template star spectrum required to shift it onto the galaxy spectrum rest frame wise. This latter number depends on the redshift of the galaxy. 830 columns were used for the fitting, skipping 64 columns in the red end of the spectrum to stay clear of the position of the very strong 5577 Å sky line. The resulting wavelength interval was of length 534 Å.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... J97^6.5

The JFK95b and J97 data for the galaxies R338/RMH48 and R343/RMH50 were switched, cf. the caption to Table , p. .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...$y \approx -0.56 x + 4.13$^7.1

This corresponds to the line $\kappa_1 + \kappa_2 = 8.0$, which is what is shown in Fig. 2b in Bender et al. (1992). These authors define a new 3-space $(\kappa_1,\kappa_2,\kappa_3)$ in which the $(\kappa_1,\kappa_2)$ projection shows the FP nearly face-on. Note, that in the text of Bender et al., the line is given as $\kappa_1 + \kappa_2 = 7.8$, a line that fall just inside the the distribution of galaxies, not at the edge. The relation $\kappa_1 + \kappa_2 = 8.0$ is also what is used recently by Burstein et al. (1997).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... 24^10.1

No overlap photometry on stars using phot was done involving field 24, but for the galaxy R243 located in the overlap between field 24 night 4 and field 23 night 3, an offset $\Delta m_{\rm circ}$ is available. At the 10.08'' aperture, it is $-0.007 \pm 0.006$ in GR and $-0.007 \pm 0.003$ in JB.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... JB^10.2

To these magnitude differences at a finite aperture r, $\Delta m(r)$, should be added the difference in aperture correction induced by the difference in seeing, $\Delta {\rm apcor}(r)$, to get the intrinsic magnitude difference, $\Delta {m_{\rm T}}$; i.e. $\Delta m_{\rm T} = \Delta m(r) + \Delta {\rm apcor}(r)$. $\Delta {\rm apcor}(r)$ for those particular images and the used 9 pixel aperture is -0.002 for GR and -0.003 for JB, and thus negligible. For reference, the seeing values are listed in Table  (p. ), the aperture correction vs. seeing is plotted in Fig.  (p. ) and (p. ), and the pixel scale is 0.5073 ''/pixel.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... file^11.1

It should be noted, that the output from the mknobsfile in digiphotx.photcalx and from the mknobsfile in digiphot.photcal is not in the same format: the x version has eight columns with OTIME as the third, the non-x version has only seven columns, and does not have OTIME. Here, the x version was used.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.