In two-dimensional spectrographs, the optical distortions in the spatial and dispersion directions produce variations in the subpixel sampling of the background spectrum. Using knowledge of the camera distortions and the curvature of the spectral features, one can recover information regarding the background spectrum on wavelength scales much smaller than a pixel. As a result, one can propagate this better sampled background spectrum through inverses of the distortion and rectification transformations and accurately model the background spectrum in two-dimensional spectra for which the distortions have not been removed (i.e., the data have not been rebinned/rectified). The procedure, as outlined in this paper, is extremely insensitive to cosmic rays, hot pixels, etc. Because of this insensitivity to discrepant pixels, sky modeling and subtraction need not be performed as one of the later steps in a reduction pipeline. Sky subtraction can now be performed as one of the earliest tasks, perhaps just after dividing by a flat field. Because subtraction of the background can be performed without having to "clean" cosmic rays, such bad pixel values can be trivially identified after removal of the two-dimensional sky background.

Combining Hubble Space Telescope WFPC2 mosaics with extensive ground-based spectroscopy, we study the nature of E+A galaxies in three intermediate-redshift clusters (z = 0.33, 0.58, and 0.83). From a sample of similar to500 confirmed cluster members, we isolate 46 E+A candidates to determine the E+A fraction and study their physical properties. Spectral types are assigned using Balmer (Hdelta, Hgamma, Hbeta) and [O II] lambda3727 equivalent widths. For all members, we have galaxy colors, luminosities, Hubble types, and quantitative structural parameters. We also include measured internal velocity dispersions for 120 cluster members and estimate velocity dispersions for the rest of the cluster sample using the fundamental plane. We find that E+A galaxies comprise a nonnegligible component (similar to7%-13%) of the cluster population at these redshifts, and their diverse nature indicates a heterogeneous parent population. While cluster E+A's are predominantly disk-dominated systems, they span the range in Hubble type and bulge-to-total fraction to include even early-type members. Cluster E+A's also cover a wide range in luminosity [L-B similar to (0.2-2.5)L-B*], internal velocity dispersion (sigma similar to30-220 km s(-1)), and half-light radius [r(1/2) similar to (0.4-4.3)h(-1) kpc]. From their velocity dispersions and half-light radii, we infer that the descendants of E+A's in our highest redshift cluster are massive early-type galaxies. In contrast to the wide range of luminosity and internal velocity dispersion spanned by E+A's at higher redshift, only low-mass E+A's are found in nearby clusters, e. g., Coma. The observed decrease in the characteristic E+A mass is similar to the decrease in the luminosity of rapidly star-forming field galaxies since z similar to 1, i.e., galaxy "downsizing." In addition, we argue that our statistics imply that greater than or similar to30% of the E-S0 members have undergone an E+A phase; the true fraction could be 100% if the effects of E+A downsizing, an increasing E+A fraction with redshift, and the conversion of spiral galaxies into early type galaxies are also considered. Thus, the E+A phase may indeed be an important stage in the transformation of star-forming galaxies into early-type members.

We select E+A candidates from a spectroscopic data set of similar to800 field galaxies and measure the E+A fraction at 0.3 < z < 1 to be 2.7% +/- 1.1%, a value lower than that in galaxy clusters at comparable redshifts (11% +/- 3%). HST WFPC2 imaging for five of our six E+A's shows that they have a heterogeneous parent population: these E+A's span a range in half-light radius (0.8 h(-1) kpc < r(1/2) < 8 h(-1) kpc) and estimated internal velocity dispersion (50 km s(-1) less than or similar to sigma(est) less than or similar to 220 km s(-1)), and they include luminous systems (-21.6 less than or equal to M-Bz - 5 log h less than or equal to - 19.2). Despite their diversity in some aspects, the E+A's share several common characteristics that indicate that the E+A phase is an important link in the evolution of star-forming galaxies into passive systems: the E+A's are uniformly redder than the blue, star-forming galaxies that make up the majority of the field, they are more likely to be bulge-dominated than the average field galaxy, and they tend to be morphologically irregular. We find that E+A's make up similar to9% of the absorption-line systems in this redshift range and estimate that greater than or similar to25% of passive galaxies in the local field had an E+A phase at z less than or similar to 1.

Using wide-field Hubble Space Telescope WFPC2 imaging and extensive Keck LRIS spectroscopy, we present a detailed study of the galaxy populations in MS 2053-04, a massive, X-ray-luminous cluster at z = 0.5866 +/- 0.0011. Analysis of 149 confirmed cluster members shows that MS 2053 is composed of two structures that are gravitationally bound to each other; their respective velocity dispersions are 865 +/- 71 km s(-1) (113 members) and 282 +/- 51 km s(-1) (36 members). MS 2053's total dynamical mass is 1.2 x 10(15) M.. MS 2053 is a classic Butcher-Oemler cluster with a high fraction of blue members (24% +/- 5%) and an even higher fraction of star-forming members (44% +/- 7%), as determined from their [O II] lambda3727 emission. The number fraction of blue/star-forming galaxies is much higher in the infalling structure than in the main cluster. This result is the most direct evidence to date that the Butcher-Oemler effect is linked to galaxy infall. In terms of their colors, luminosities, estimated internal velocity dispersions, and [O II] lambda3727 equivalent widths, the infalling galaxies are indistinguishable from the field population. MS 2053's deficit of S0 galaxies combined with its overabundance of blue spirals implies that many of these late-type galaxies will evolve into S0 members. The properties of the blue cluster members in both the main cluster and infalling structure indicate that they will evolve into low-mass, L < L-* galaxies with extended star formation histories like that of low-mass S0 galaxies in Coma. Our observations show that most of MS 2053's blue cluster members, and ultimately most of its low-mass S0 galaxies, originate in the field. Finally, we measure the redshift of the giant arc in MS 2053 to be z = 3.1462; this object is one in only a small set of known strongly lensed galaxies at z > 3.

We present follow-up spectroscopy of the galaxy cluster MS 1054 03 (z = 0.83) confirming that at least six of the nine merging galaxy pairs identified by van Dokkum et al. in 1999 are indeed bound systems: they have projected separations of R-s < 10h(-1) kpc and relative line-of-sight velocities of Delta nu < 165 kms(-1). For the remaining three pairs, we were unable to obtain redshifts of both constituent galaxies. To identify a more objective sample of merging systems, we select bound red galaxy pairs (R-s <= 30 h(-1) kpc,Delta nu <= 300 km s(-1)) from our sample of 121 confirmed cluster members: galaxies in bound red pairs make up 15.7% +/- 3.6% of the cluster population. The color-magnitude diagram shows that the pair galaxies are as red as the E/S0 members and have a s homogeneous stellar population. The red pair galaxies span a large range in luminosity and internal velocity dispersion, to include some of the brightest, most massive members (L > L*,sigma(1D) > 200 km s(-1)); these bound 1D galaxy pairs must evolve into E/S0 members by z similar to 0.7. These results, combined with MS 1054' s high merger fraction and reservoir of likely future mergers, indicates that most, if not all, of its early-type members evolved from (passive) galaxy- galaxy mergers at. z <= 1.

We examine the distribution of stellar masses of galaxies in MS 1054.4 - 0321 and Cl 0152.7 - 1357, two X-ray-selected clusters of galaxies at z = 0.83. Our stellar mass estimates, from spectral energy distribution fitting, reproduce the dynamical masses as measured from velocity dispersions and half-light radii with a scatter of 0.2 dex in the mass for early-type galaxies. When we restrict our sample of members to high stellar masses, those over 10(11.1) M-circle dot (M* in the Schechter mass function for cluster galaxies), we find that the fraction of early-type galaxies is 79% +/- 6% at z = 0.83 and 87% +/- 6% at z = 0.023 for the Coma Cluster, consistent with no evolution. Previous work with luminosity-selected samples has found that the early-type fraction in rich clusters declines from similar or equal to 80% at z = 0 to similar or equal to 60% at z = 0.8. The observed evolution in the early-type fraction from luminosity-selected samples must predominantly occur among sub-M* galaxies. As M* for field and group galaxies, especially late types, is below M* for cluster galaxies, infall could explain most of the recent growth in the early-type fraction. Future surveys could determine the morphological distributions of lower mass systems, which would confirm or refute this explanation.

In this paper we analyze previously published spectra with high signal-to-noise ratios of E/S0s in the rich cluster CL 1358+62 at z = 0.33. Introducing techniques for fitting stellar population models to the data, we focus on the 19 Es and SOS with a homogeneous set of eight blue Lick indices. We explore the galaxy properties using six-parameter stellar population models from the literature and describe an approach for fitting the models differentially, such that the largest systematic errors are avoided. The results of the model fitting are accurate relative measures of the stellar population parameters. We find (1) no difference between the best-fit stellar population parameters of Es and S0s at fixed or; (2) the stars in Es and S0s are uniformly old, consistent with previously published results using the fundamental plane; (3) a correlation of [Z/H] with sigma, in a manner consistent with the observed B-V colors of the galaxies; (4) no strong correlation of [alpha/Fe] with sigma, discrepant with the correlation inferred from data on nearby galaxies at the < 3 sigma or level; and (5) a significant anticorrelation of [alpha/N] with sigma, which we interpret as a correlation of the abundance of secondary nitrogen with mean metallicity. While the differences between our conclusions and the current view of stellar populations may point to serious deficiencies, our deduced correlation of mean metallicity with velocity dispersion does reproduce the observed colors of the galaxies and the slope of the local Mg-sigma relation. More specifically, our data conclusively show that cluster S0s did not form their stars at significantly later epochs than cluster elliptical galaxies of the same mass, and the presence of secondary nitrogen indicates that both Es and S0s formed from self-enriching progenitors, presumably with extended star formation histories.

We examine the morphology-density relation for complete mass-limited samples of galaxies from five X-ray luminous clusters from z = 0.023 to z = 0.83. We find no change in the morphology-density relation for our complete sample of 366 galaxies with masses > 10(10.8) M-circle dot (roughly 0.4 M-star at z = 0). In contrast, we find a change in the early-type fraction with redshift that is in good agreement with previous work in a luminosity-selected sample (0.4L(star)) of 478 galaxies from the same parent data. Our results indicate that the galaxies that dominate the morphology-density relation, and its evolution, have masses < 10(10.8) M-circle dot.

Using a magnitude-limited, spectroscopic survey of the X-ray-luminous galaxy cluster MS 1054 - 03, we isolate 153 cluster galaxies and measure MS 1054's redshift and velocity dispersion to be z = 0: 8307 +/- 0: 0004 and sigma(z) = 1156 +/- 82 km s(-1). The absorption-line, poststarburst ('' E+ A ''), and emission-line galaxies, respectively, make up 63% +/- 7%, 15% +/- 4%, and 23% +/- 4% of the cluster population. With photometry from HST ACS, we find that the absorption-line members define an exceptionally tight red sequence over a span of similar to 3.5 mag in i(775): their intrinsic scatter in (V-606 - i(775)) color is only 0: 048 +/- 0: 008, corresponding to a (U - B)(z) scatter of 0.041. Their color scatter is comparable to that of the ellipticals (sigma(Vi) = 0: 055 +/- 0: 008), but measurably smaller than that of the combined E+ S0 sample (sigma(Vi) = 0: 072 +/- 0: 010). The color scatter of MS 1054's absorption-line population is approximately twice that of the ellipticals in Coma; this difference is consistent with passive evolution where most of the absorption-line members (> 75%) formed by z similar to 2, and all of them by z similar to 1: 2. For red members, we find a trend (> 95% confidence) of weakening H delta absorption with redder colors that we conclude is due to age: in MS 1054, the color scatter on the red sequence is driven by differences in mean stellar age of up to similar to 1.5 Gyr. We also generate composite spectra and estimate that the average S0 in MS 1054 is similar to 0.5-1 Gyr younger than the average elliptical; this difference in mean stellar age is mainly due to a number of S0's that are blue (18%) and/or are poststarburst systems (21%).

We examined the morphology-density relations for galaxy samples selected by luminosity and by mass in each of five massive X-ray clusters from z = 0.023 to 0.83 for 674 spectroscopically confirmed members. Rest-frame optical colors and visual morphologies were obtained primarily from Hubble Space Telescope images. The visual morphologies ensure consistency with the extensive published results on galaxy evolution in dense environments. Morphologydensity relations (MDRs) are derived in each cluster from a complete, luminosity-selected sample of 452 galaxies with a magnitude limit M-V < M-V(star) + 1. The change in the early-type fraction with redshift matches previous work for massive clusters of galaxies. We performed a similar analysis, deriving MDRs for complete, mass-selected samples of 441 galaxies with a mass limit of 10(10.6) M-circle dot. Our mass limit includes faint objects, the equivalent of similar or equal to 1 mag below L-star for the red cluster galaxies, and encompasses similar or equal to 70% of the stellar mass in cluster galaxies. The MDRs in the massselected sample at densities of Sigma > 50 galaxies Mpc(-2) are similar to those in the luminosity-selected sample but show larger early-type fractions, with a weak indication of a shallower slope. However, the trend with redshift in the fraction of elliptical and S0 galaxies with masses > 10(10.6) M-circle dot differs significantly between the mass-and luminosityselected samples. The clear trend seen in the early-type fraction from z = 0 to similar or equal to 0.8 is not found in mass-selected samples. The early-type galaxy fraction changes much less and is consistent with being constant at 92% +/- 4% at Sigma > 500 galaxies Mpc(-2) and 83% +/- 3% at 50 galaxies Mpc(-2) < Sigma < 500 galaxies Mpc(-2). Given the mass limit in our sample, this suggests that galaxies of mass lower than > 10(10.6) M-circle dot play a significant role in the evolution of the early-type fraction in luminosity-selected samples; i. e., they are larger contributors to the luminosity-selected samples at higher redshifts than at low redshifts.