| Abstract | We analyze pairs of galaxies selected from the CfA magnitude limited and SSRS diameter-limited redshift catalogs. The pairs are chosen only from regions of low galaxy density and have projected separations of up to 1.0 Mpc. This is almost an order of magnitude larger than the typical projected separations of samples of binary galaxies used in earlier studies. Accurate H I velocities for these pairs were obtained at both Arecibo and Parkes Observatories, and we tabulate these new velocities along with total H I flux and other quantities of interest. The velocity measurements are very homogeneous, and we estimate the typical internal error in the measured systemic velocity to be under 5 km s-1. The velocity difference distribution f(∆V) for our sample of isolated wide pairs of galaxies decreases monotonically with increasing velocity difference A V. In particular, we do not find a peak near 70 km s-1 as seen in earlier, more compact samples. This tends to confirm the suggestion of Schneider & Salpeter (1992) that the nonmonotonicity of earlier samples of binary galaxies is a consequence of a selection bias. The distribution of ∆V for galaxy pairs in low-density regions is found to have two components, a peak around zero and a tail extending to larger velocity differences. The tail of the ∆V distribution has a strong dependence on the degree of isolation. As the stringency of the isolation criterion is increased, pairs are preferentially lost from the tail of the distribution. The width of the peak around zero, on the other hand, has little dependence on the isolation criterion. For our sample of isolated pairs of galaxies, the median of the velocity differences is ̃30 km s-1. The velocity difference of the galaxy pairs decreases with increasing projected separation. The characteristic 30 km s 1 velocity difference for our isolated sample is less than half the typical 70 km s-1 velocity difference found in previous studies using relatively compact pairs. We thus confirm the suggestion of Charlton & Salpeter (1991) that there exist physically associated galaxy pairs even at separations as large as 1.0 Mpc. Unlike earlier samples of binary galaxies, the crossing time rp/∆V for our pairs is comparable to the Hubble time, so unique mass derivations are not possible. We nonetheless compute the dynamical mass of the galaxies as given by a variety of estimators, to get values of up to a few times 1012 Msun. This is consistent with both dynamical mass estimates for the Local Group, and also the recent mass estimates of spiral galaxies from the dynamics of faint satellites by Zaritsky et al. (1993). |
