We report radial velocities for 99 galaxies with projected positions within 30 arcmin of the center of the cluster A3733 obtained with the MEFOS multifiber spectrograph at the 3.6-m ESO telescope. These measurements are combined with 39 redshifts previously published by Stein (1996) to built a collection of 112 galaxy redshifts in the field of A3733, which is used to examine the kinematics and structure of this cluster. We assign cluster membership to 74 galaxies with heliocentric velocities in the interval 10500-13000 km/s. From this sample of cluster members, we infer a heliocentric systemic velocity for A3733 of 11653{+74}{-76} km/s, which implies a mean cosmological redshift of 0.0380, and a velocity dispersion of 614{+42}{-30} km/s. The application of statistical substructure tests to a magnitude-limited subset of the latter sample reveals evidence of non-Gaussianity in the distribution of ordered velocities in the form of lighter tails and possible multimodality. Spatial substructure tests do not find, however, any significant clumpiness in the plane of the sky, although the existence of subclustering along the line-of-sight cannot be excluded.

In two previous papers a semi-analytical model was presented for the hierarchical clustering of halos via gravitational instability from peaks in a random Gaussian field of density fluctuations. This model is better founded than the extended Press-Schechter model, which is known to agree with numerical simulations and to make similar predictions. The specific merger rate, however, shows a significant departure at intermediate captured masses. The origin of this was suspected as being the rather crude approximation used for the density of nested peaks. Here, we seek to verify this suspicion by implementing a more accurate expression for the latter quantity which accounts for the correlation among peaks. We confirm that the inclusion of the peak-peak correlation improves the specific merger rate, while the good behavior of the remaining quantities is preserved.

Based on the photometric survey of the inner region of the Centaurus cluster (Jerjen & Dressler 1997a) we measured redshifts for a deep, surface brightness limited sample of galaxies using the MEFOS multifibre spectrograph at the ESO 3.6m telescope. With the new data set radial velocities for 120 centrally located cluster members become available which is equivalent to 78% of all known cluster galaxies in the region brighter than B_T=18.5. The relevant aspect of this investigation is that new redshifts for 32 dwarf galaxies have been measured, rising the total number to 48. We investigate the prominent bimodal velocity distribution of Centaurus in more detail, discussing the very different characteristics of the velocity distributions for the main Hubble types E&S0, spirals, Im&BCD, and dE&dS0. The nucleated, bright dwarf ellipticals are the only galaxies with a Gaussian-like distribution centred at 3148+/-98 km/s. The remarkable coincidence of this velocity with the mean velocity of Cen30 and the redshift of NGC 4696 in particular strongly suggests a connection of the dE&dS0s to the gravitational centre of the Centaurus cluster and/or to the cluster dominant E galaxy. The application of statistical tests reveals the existence of a population dwarf galaxies bound to NGC 4696. The dynamical parameters for the two velocity components suggest that Cen30 is the real Centaurus cluster whereas Cen45 can only be a loosely bound group of galaxies. This conclusion is followed up with a type-mixture analysis. All results are fully consistent with the cluster-group scenario. Whether Cen45 is merging with the cluster or is located in the close background remains unclear. We show that the poorness of Cen45 represents an intrinsic problem which makes it difficult to approach this question.

A homogeneous sample of galaxy redshifts in the core regions (R < 0.5 h$^{-1}$ Mpc) of 12 clusters is used to measure the frequency of substructure with different tests. In 50 % of the cases substructure is detected, a frequency which agrees well with previous studies of cluster cores. Magnitude information and rough morphological classes are also available for 80 % of the sample, allowing us to confirm that bright galaxies (M < -22 mag) have a significantly lower velocity dispersion than the rest. Elliptical galaxies are the main responsibles for this luminosity segregation in velocity space, whereas no segregation can be seen for spiral galaxies. Given the coincidence of substructure and luminosity segregation, a cluster model with an old population of ellipticals which are under the effect of dynamical friction in each subcluster is thus favoured by these observations. Spiral galaxies seem to be late arrivals, or are passing in front or behind the core of the cluster.

An extensive redshift survey has been conducted on a sample of 15 nearby (0.01 < z < 0.05) clusters of galaxies. A total number of 860 redshifts were determined by fitting of emission--lines and/or cross-correlation techniques. Of this sample, 735 galaxies are within 0.2--0.8 Mpc ($H_0$ = 50 km/s/Mpc) of the center of clusters. Approximate morphological types are available for most of the galaxies. A comparison of the present redshifts with published data allows an extensive error analysis. The agreement is excellent with the most modern data, showing a zero point error of 5 km/s and an overall consistency of the measurements and their uncertainties. We estimate our redshifts to have mean random errors around 30 km/s. A population analysis of the clusters will be given in a forthcoming paper.