Observational Constraints on the Origins of the Fundamental Plane

We review the concept of the Fundamental Plane (FP), and present new results on the near–infrared FP. We show that the K–band FP differs both from the optical form and the virial expectation under the assumption of constant (M/L) and homology. Systematic variations of only the stellar populations parameters (age, metallicity, IMF) cannot reproduce the slopes of the FP simultaneously at both the optical and near–infrared wavelengths. There appears to be an additional effect which could be due to a systematic departure of the velocity distributions of elliptical galaxies from a homologous family. In order to distinguish between the different possible stellar populations parameters, the FP and its projections can be observed at a range of redshifts. We describe several such evolutionary tests: in the intercept of the color–magnitude relation; in the intercept of the Kormendy surface brightness–radius correlation; and in the intercept of the FP itself. All three tests are fully consistent with the population of cluster elliptical galaxies having formed at high redshift; this contradicts large, systematic age variations as a significant contributor to the slope of the FP. 1. What is the Fundamental Plane? The Fundamental Plane (FP) is a bivariate family of correlations of the properties of elliptical galaxies (Djorgovski & Davis 1987; Dressler et al. 1987). The properties that are usually correlated for the FP are the half–light radius re, the mean surface brightness (SB) 〈μ〉e interior to that radius, and the central velocity dispersion σ0. Other parameters can be used in the FP, such as the substitution of a color or line index for σ0 (de Carvalho & Djorgovski 1989), or the substitution of luminosity for radius. Some monovariate correlations also represent important properties among ellipticals, such as the distance–independent Mg2–σ relation which may be a good age indicator (see Bender, this volume). The FP can be projected onto any two axes out of the three variables. Examples of these projections are the color–magnitude relation, the Kormendy radius–SB relation (Kormendy 1977), and the Faber & Jackson (1976) relation between luminosity and velocity dispersion. The Dn–σ relation is another example, as it was constructed as a nearly edge–on projection of the FP (Dressler et al. 1987); its residuals should correlate slightly with SB (Gunn 1988), as was observed by Lucey, Bower, & Ellis (1991a).