1 Introduction

The origin, evolution and even physical reality of Hickson Compact Groups (Hickson [1982]) are still a matter of debate. Their morphology suggests that many individual HCG galaxies present signatures of interaction (see, e.g., Mendes de Oliveira & Hickson [1994]: hereafter MdOH). But it has been argued that a high fraction of HCGs are rather the result of chance alignments within loose groups (Mamon [1986], [1987], [1992], [1994]) or within filaments (Hernquist et al. [1995]; Ostriker et al. [1995]). If studies on the evolution of HCGs have produced apparently conflicting conclusions (White et al. [1990]; Hickson et al. [1992]; Zepf [1993]; Pildis [1995]; Sulentic & Rabaça [1994]; Athanassoula et al. [1997]), this probably reflects the wide diversity in the nature and evolution of these groups.

Studying early-type galaxies in such environment will be helpful to constrain the actual status of their host group, and also to characterize the reactions of early-type galaxies to gravitational perturbations. In this paper we focus on the kinematical aspects.

The presence of diffuse X-ray emission from the hot intra-group gas provides a strong evidence for the boundness of a system (Ponman et al. [1996], and references therein), but it is still difficult to ascertain the presence of such a diffuse background (dos Santos & Mamon [1999]). H  I deficiency in many HCG spirals (Williams & Rood [1987]; Huchtmeier [1997]), and concentration within their central regions (Menon [1995a],b) are other signatures of the existence of the intra-group medium. In addition, a group can be interacting but yet be unbound (and free of intra-group medium) if we observe it at an early stage of its evolution. In this case signatures of interaction can be searched for in an excess of star formation, from broad-band colors, CO and FIR properties (Sulentic & de Mello Rabaça [1993]; Moles et al. [1994]; Venugopal [1995]; Verdes-Montenegro et al. [1998]; Leon et al. [1998]); an alternative is to study the distribution of AGNs (de Carvalho & Coziol [1999], and references therein). There is also the kinematics of spiral galaxies (Rubin et al. [1991]: hereafter RFH; Mendes de Oliveira et al. [1997]). In only a minority of cases it is possible to obtain clear evidence of interactions. For example, kinematical evidence have been found in HCG 16 (Mendes de Oliveira et al. [1998]), HCG 67 (Bettoni & Buson [1999]: hereafter BB99), HCG 90 (Plana et al. [1998]), and HCG 96 (Verdes-Montenegro et al. [1997]).

Because of the above ambiguities, we selected early-type galaxies in HCGs regardless of their X-ray status or previously reported evidence for interaction.

The effects of interactions on early-type galaxies are perhaps more subtle than on spirals, and stellar kinematics have been less systematically studied. In addition to morphological signatures such as shells or ripples, distorted isophotes (Davoust & Prugniel [1988]), these galaxies are expected to show transient features in their kinematic profiles, as in pairs of galaxies (Bonfanti et al. [1995]; Combes et al. [1995]). But the difficulty of connecting particular kinematic characteristics to interactions has been emphasized by Rampazzo et al. ([1998]). There are, however, some exceptions: in HCG 90, for example (Longo et al. [1994]), strong perturbations in the velocity patterns could be unequivocally linked to direct interactions within the group.

With the aim of increasing the small amount of available data in stellar kinematics, we have looked for kinematic signatures in the early-type galaxies of three Hickson groups; these are HCG 67 (a strong source of diffuse X-rays), HCG 74 (whose members are embedded in a common envelope), and HCG 79 (with a high density of galaxies). In Sect. 2, imaging and spectroscopic material is presented. Our results are detailed in Sect. 3, and summarized in Sect. 4.