Spectral data on early type galaxies is analyzed for chemical abundance with an emphasis on obtaining detailed abundances for the elements O and Si in addition to C, N, Na, Mg, Ca, Fe, and Ba. The abundance trends with velocity dispersion fit preconceptions based upon previous Mg conclusions, namely that larger galaxies have a higher alpha element to iron peak ratio indicative of a higher ratio of Type II to Type Ia supernova products. The heaviest alpha elements, Ca and Ti, do not participate in this trend, although this fact does not necessarily alter the basic picture given the uncertainties in nucleosynthetic yields. Elements that likely have significant contributions from intermediate-mass stars, namely C, N, and Ba, also gain ground relative to Fe in massive galaxies at a modest level, with the Ba conclusion uncertain from our data alone.

In a sample of elliptical galaxies that span a large range of mass, a previously unused Ca index, CaHK, shows that [Ca/Fe] and [Ca/Mg] systematically decrease with increasing elliptical galaxy mass. Metallicity mixtures, age effects, stellar chromospheric emission effects, and low-mass initial mass function (IMF) boost effects are ruled out as causes. A [Ca/Fe] range of less than 0.3 dex is sufficient to blanket all observations. Feature gradients within galaxies imply a global Ca deficit rather than a radius-dependent phenomenon. Some, but not all, Type II supernova nucleosynthetic yield calculations indicate a decreasing Ca/Fe yield ratio in more massive supernovae, lending possible support to the hypothesis that more massive elliptical galaxies have an IMF that favors more massive stars. No Type II supernova nucleosynthetic yield calculations show significant leverage in the Ca/Fe ratio as a function of progenitor metallicity. Therefore, it seems unlikely that the Ca behavior can be explained as a built-in metallicity effect, and this argues against explanations that vary only the Type II to Type Ia supernova enrichment ratio.

Using synthetic spectra to gauge the observational consequences of altering the abundance of individual elements, I determine the observability of new Lick IDS style indices designed to target individual elements. Then using these new indices and single stellar population models, I investigate a new method to determine Balmer series emission in a Sloan Digital Sky Surveys grand average of quiescent galaxies. I also investigate the effects of an old metal-poor stellar population on the near ultra violet spectrum through the use of these new indices and find that the presence of a small old metal-poor population accounts for discrepancies observed between index trends in the near UV and optical spectral regimes. Index trends for 74 indices and three data sets are presented and discussed. Finally, I determine the near nuclear line-strength gradients of 18 red sequence elliptical Virgo cluster galaxies for 74 indices.

We examine the spectrum in the vicinity of the NH3360 index of Davidge & Clark (1994), which was defined to measure the NH absorption around 3360 \AA and which shows almost no trend with velocity dispersion (Toloba et al. 2009), unlike other N- sensitive indices, which show a strong trend (Graves et al. 2007). Computing the effect of individual elements on the integrated spectrum with synthetic stellar population integrated spectra, we find that, while being well correlated with nitrogen abundance, NH3360 is almost equally well anti-correlated with Mg abundance. This prompts the definition of two new indices, Mg3334, which is mostly sensitive to magnesium, and NH3375, which is mostly sensitive to nitrogen. Rather surprisingly, we find that the new NH3375 index shows a trend versus optical absorption feature indices that is as shallow as the NH3360 index. We hypothesize that the lack of a strong index trend in these near-UV indices is due to the presence of an old metal-poor component of the galactic population. Comparison of observed index trends and those predicted by models shows that a modest fraction of an old, metal-poor stellar population could easily account for the observed flat trend in these near-UV indices, while still allowing substantial N abundance increase in the larger galaxies.

For purposes of stellar population analysis, emission corrections for Balmer series indices on the Lick index system in Sloan Digital Sky Survey (SDSS) stacked quiescent galaxy spectra are derived, along with corrections for continuum shape and gross stellar content, as a function of the Mg $b$ Lick index strength. These corrections are obtained by comparing the observed Lick index measurements of the SDSS with new observed measurements of 13 Virgo Cluster galaxies, and checked with model grids. From the H$\alpha$ Mg $b$ diagram a linear correction for the observed measurement is constructed using best fit trend lines. Corrections for H$\beta$, H$\gamma$ and H$\delta$ are constructed using stellar population models to predict continuum shape changes as a function of Mg $b$ and Balmer series emission intensities typical of H{\sc II} regions. The corrections themselves are fairly secure, but the interpretation for H$\delta$ and H$\gamma$ indices is complicated by the fact that the H$\delta$ and H$\gamma$ indices are sensitive to elemental abundances other than hydrogen.

Spectral feature index diagrams with integrated globular clusters and simple stellar population models often show that some clusters have weak H beta, so weak that even the oldest models cannot match the observed feature depths. In this work, we rule out the possibility that abundance mixture effects are responsible for the weak indices unless such changes operate to cool the entire isochrone. We discuss this result in the context of other explanations, including horizontal branch morphology, blue straggler populations, and nebular or stellar emission fill-in, finding a preference for flaring in M giants as an explanation for the H beta anomaly.

We report on the discovery of strongly variable FeII and MgII absorption lines seen at z=1.48 in the spectra of the z=4.05 GRB 060206 obtained between 4.13 to 7.63 hours (observer frame) after the burst. In particular, the FeII line equivalent width (EW) decayed rapidly from 1.72+-0.25 AA to 0.28+-0.21 AA, only to increase to 0.96+-0.21 AA in a later date spectrum. The MgII doublet shows even more complicated evolution: the weaker line of the doublet drops from 2.05+-0.25 AA to 0.92+-0.32 AA, but then more than doubles to 2.47+-0.41 AA in later data. The ratio of the EWs for the MgII doublet is also variable, being closer to 1:1 (saturated regime) when the lines are stronger and becoming closer to 2:1 (unsaturated regime) when the lines are weaker, consistent with expectations based on atomic physics. We have investigated and rejected the possibility of any instrumental or atmospheric effects causing the observed strong variations. Our discovery of clearly variable intervening FeII and MgII lines lends very strong support to their scenario, in which the characteristic size of intervening patches of MgII ``clouds'' is comparable to the GRB beam size, i.e, about 10^16 cm. We discuss various implications of this discovery, including the nature of the MgII absorbers, the physics of GRBs, and measurements of chemical abundances from GRB and quasar absorption lines.

We report on two recent z~4 gamma-ray bursts (GRBs), GRB 060206 and GRB 060210, for which we have obtained well-sampled optical light curves. Our data, combined with early optical data reported in the literature, shows unusual behavior for both afterglows. In R-band GRB 060206 (z=4.045) experienced a slow early decay, followed by a rapid increase in brightness by factor ~2.5 about 1 hour after the burst. Its afterglow then faded in a broken power-law fashion, with a smooth break at t_b=0.6 days, but with additional, less dramatic (~10%) ``bumps and wiggles'', well detected in the densely sampled light curve. The R-band afterglow of GRB 060210 (z=3.91) is also unusual: the light curves was more or less flat between 60 and 300 sec after the burst, followed by ~70% increase at ~600 sec after the burst, after which the light curve declined as a t^{-1.3} power-law. Despite earlier reports to the contrary, we find that for GRB 060206 X-rays also more or less follow the optical decay, but with significant variations on short timescales. However, the X-ray afterglow is contaminated by a nearby, variable source, which especially at late times obscures the behavior of the X-ray afterglow. We argue that ``anomalous'' optical afterglows are likely to be the norm, and that the rapid variations often seen in Swift-XRT data would also be seen in the optical light curves, given good enough sampling. As a result, some of the often employed procedures, such as deriving the jet opening angle using a smooth broken power-law fit to the optical light curves, in many cases might have a poor statistical significance. We argue that the early increase in brighness for both bursts might be due to the turn-on of the external shock. Existence of such features could provide valuable additional information about the burst. (Abridged)