1 Introduction

One of the most important clues concerning the early evolution of dynamically hot galaxies (DHGs: ellipticals, dwarf spheroidals, and bulges of spirals) in the fundamental plane of galaxies is the existence of a well-defined relationship between metallicity and mass (e.g., Bender et al. 1993). The fundamental lesson taught by this relation is that star formation in DHGs stopped because of gas loss, with less massive systems losing greater fractions of their gas. Outflow probably begins when supernovae have raised the internal energy of the gas enough to allow it to escape the potential well (e.g., Brocato et al. 1990).

Most commonly, the metallicity in DHGs is measured via the Mg₂ index. While the Mg₂ index is an excellent means of ranking galaxy metallicities, it does not yield an abundance directly, i.e., the number density of a particular element relative to hydrogen, and calibrations of the Mg₂ index (model-dependent) are usually in terms of the iron abundance, an element whose production is notoriously difficult to model. Though this may be best for some purposes, e.g., studies of stellar populations, it is not sufficient for all purposes. To study the chemical evolution of DHGs requires the abundance of an element whose production is well understood. Were such abundances available, there would be some hope of quantifying the gas fraction at which DHGs of different masses begin to lose mass. Knowledge of the abundances would admit studying the yield of heavy elements, and hence the slope of the stellar initial mass function during the star formation epoch. Given the known photometric and dynamical properties of DHGs today, abundances would also allow us to study the global energetics involved during their star formation phase.

This paper is one of a sequence investigating the oxygen abundances of DHGs. Here, we present oxygen abundances for samples of planetary nebulae in M 32 and in the bulge of M 31. These two nearby systems are good representatives of typical DHGs. Though M 32's light profile may be truncated compared to isolated ellipticals, its structural, dynamical, and spectral properties are perfectly typical for an elliptical of its luminosity (Kormendy 1985; Bender et al. 1993). Similarly, recent work on the DHG fundamental plane has shown that the photometric, dynamical, and stellar population properties of bulges follow those of pure ellipticals (Bender et al. 1992, 1993).

Oxygen is an excellent element with which to study the evolution of galaxies. Oxygen is a primary element whose sole significant production site is type II supernovae (Wheeler et al. 1989), so its abundance is tied directly to the history of massive star formation, and the enrichment time scale is short compared to the gas consumption time scale. Oxygen abundances are also easily observable in planetary nebulae. Planetary nebulae have high electron temperatures, so the temperature-sensitive [O III]$\lambda$4363 line is observable, making it possible to determine accurate electron temperatures in high metallicity environments. Further, the dominant ionization stages of oxygen, O⁺ and O⁺⁺, have observable lines, while other ionization stages are easily accounted for using ratios of readily detectable helium lines (e.g., Kingsburgh & Barlow 1994).

Planetary nebulae are good sites in which to probe the oxygen abundance, and they are the only sites that are directly accessible in DHGs. Since planetary nebulae are bright in strong emission lines (e.g., [O III]$\lambda$5007), they are easily located within their parent galaxies using emission-line and continuum-band imaging (e.g., Ciardullo et al. 1989). Observational and theoretical evidence indicates that the stellar precursors of most planetary nebulae do not modify their initial oxygen abundance (Iben & Renzini 1983; Henry 1989; Perinotto 1991; Forestini & Charbonnel 1997). Hence, a planetary nebula's oxygen abundance reflects that in the interstellar medium at the time of its precursor's formation. Finally, most of the stellar populations in DHGs are old, so they will produce planetary nebulae at comparable rates per unit mass. As a result, planetary nebulae sample the oxygen abundances in DHGs according to the mass in each stellar population. The resulting mean oxygen abundance for the planetary nebula population in a DHG should then be a mass-weighted mean of the oxygen abundances in its stellar populations.

Apart from their utility for studying the chemical evolution of M 31 and M 32, the spectroscopic data for the planetary nebulae we present are interesting for what they reveal about the evolution of the planetary nebulae themselves. Though there may exist a good qualitative understanding of planetary nebula evolution, it is unclear how well it stands up to quantitative scrutiny. This situation arises primarily because the distances to planetary nebulae are difficult to establish within the Milky Way. Traditionally, this constraint has made it difficult to study such absolute properties as the luminosity and size of planetary nebulae, as well as the temporal evolution of these quantities. Extragalactic planetary nebulae are especially valuable in this regard because their distances are known. The addition of the data sets for M 32 and the bulge of M 31 is particularly helpful since these planetary nebulae arise from old stellar populations. They will thus provide an intriguing contrast with the planetary nebula populations in the Magellanic Clouds, which are the product of recent star formation (Richer 1993). Whether the evolution of planetary nebulae depends upon the progenitor mass or metallicity are among the questions that we may hope to answer through a comparison of the properties of planetary nebulae in M 31 and M 32 with those elsewhere. A better quantitative understanding of planetary nebula evolution would be a great help in understanding and using the planetary nebula luminosity function as a distance indicator.

In this paper, we present our spectroscopic data for our samples of planetary nebulae in M 32 and in the bulge of M 31. The observations and their reductions are described in Sect. 2. The line intensities and reddenings we deduce are presented in Sect. 3. The reddening-corrected line intensities are then used to calculate electron temperatures and oxygen abundances in Sect. 4. Summary comments are given in Sect. 5.

In companion papers, we will use the data we present below to study the chemical evolution of DHGs and the evolution of planetary nebulae in different environments.