THE PROPERTIES OF POOR GROUPS OF GALAXIES - I - SPECTROSCOPIC SURVEY AND RESULTS

We use multifiber spectroscopy of 12 poor groups of galaxies to addres s (1) whether the groups are bound systems or chance projections of ga laxies along the line of sight; (2) why the members of each group have not already merged to form a single galaxy, despite the groups' high galaxy densities, short crossing times, and likely environments for ga laxy-galaxy mergers; and (3) how galaxies might evolve in these groups , where the collisional effects of the intragroup gas and the tidal in fluences of the global potential are weaker than in rich clusters. Eac h of the 12 groups has fewer than about five cataloged members in the literature. Our sample consists of 1002 galaxy velocities, 280 of whic h are group members. The groups have mean recessional velocities betwe en 1600 and 7600 km s(-1). Nine groups, including three Hickson compac t groups, have the extended X-ray emission characteristic of an intrag roup medium (see Paper II). We conclude the following: 1. The nine poo r groups with diffuse X-ray emission are bound systems with at least s imilar to 20-50 group members with absolute magnitudes as faint as M-B similar to -14 + 5 log(10) h to -16 + 5 log(10) h. The large number o f group members, the significant early-type population (up to similar to 55% of the membership) and its concentration in the group center, a nd the correspondence of the central, giant elliptical with the optica l and X-ray group centroids argue that the X-ray groups are not radial superpositions of unbound galaxies. The velocity dispersions of the X -ray groups range from 190 to 460 km s(-1). We are unable to determine if the three non-X-ray groups, which have lower velocity dispersions (<130 km s(-1)) and early-type fractions (= 0%), are also bound. 2. Ga laxies in each X-ray-detected group have not all merged together becau se a significant fraction of the group mass lies outside of the galaxi es and in a common halo. The velocity dispersion of the combined group sample is constant as a function of radius out to the virial radius o f the system (typically similar to 0.5 h(-1) Mpc). The virial mass of each group (similar to 0.5-1 x 10(14) h(-1) M.) is large compared with the mass in the X-ray gas and in the galaxies (e.g., similar to 1 x 1 0(12) h(-5/2) M. and similar to 1 x 10(13) h(-1) M., respectively, in NGC 533). These results imply that most of the group mass is in a comm on, extended halo. The small fraction (similar to 10%-20%) of group ma ss associated with individual galaxies suggests that the rate of galax y-galaxy interactions is lower than for a galaxy-dominated system, all owing these groups to virialize before all of their galaxies merge and to survive for more than a few crossing times. 3. The position of the giant, brightest elliptical in each X-ray group is indistinguishable from the center of the group potential, as defined by the mean velocit y and the projected spatial centroid of the group galaxies. This resul t suggests that dominant cluster ellipticals, such as cD galaxies, may form via the merging of galaxies in the centers of poor group-like en vironments. Groups with a central, dominant elliptical may then fall i nto richer clusters. This scenario explains why cD galaxies do not alw ays lie in the spatial and kinematic center of rich clusters but inste ad occupy the centers of subclusters in non-virialized clusters. 4. Th e fraction of early-type galaxies in the poor groups varies significan tly, ranging from that characteristic of the field (less than or simil ar to 25%) to that of rich clusters (similar to 55%). The high early-t ype fractions are particularly surprising because all of the groups in this sample have substantially lower velocity dispersions (by a facto r of similar to 2-5) and galaxy number densities (by a factor of simil ar to 5-20) than are typical of rich clusters. Hence, the effects of d isruptive mechanisms like galaxy harassment on the morphology of poor group galaxies are weaker than in cluster environments. In contrast, t he kinematics of poor groups make them preferred sites for galaxy-gala xy mergers, which may alter the morphologies and star formation histor ies of some group members. If galaxy-galaxy interactions are not respo nsible for the high early-type fractions, it is possible that the effe cts of environment are relatively unimportant at the current epoch and that the similarity of the galaxy populations of rich clusters and so me poor groups reflects conditions at the time of galaxy formation. 5. The fraction of early-type group members that have experienced star f ormation within the last similar to 2 h(-1) Gyr is consistent with tha t in rich clusters with significant substructure (similar to 15%). If some of the subclusters in these rich, complex clusters are groups tha t have recently fallen into the cluster environment, the similarity be tween the star formation histories of the early types in the subcluste rs and of those in our sample of field groups indicates that the clust er environment and associated mechanisms like ram pressure stripping a re not required to enhance and/or quench star formation in these parti cular galaxies. If the recent star formation is tied to the external e nvironment of the galaxies and not to internal instabilities, it is mo re likely that galaxy-galaxy encounters have altered the star formatio n histories of some early-type galaxies in groups and in subclusters.