2000ApJ...544..176N

Using complete redshift catalogs, we have compiled a list of galaxy pairs based solely on a pair's projected separation, r_p, and velocity difference, ΔV. We have made high-velocity precision H I observations of each galaxy in the sample and have reported these in the literature. Due to the nature of the redshift catalogs, we are able to quantitatively evaluate the effects of isolation and number density of surrounding galaxies on each pair in the sample. For the close galaxy pairs (r_p<100 kpc), the degree of isolation (a measure of the number of near neighbors) has little effect on the median ΔV. This median is about 55 km.s^–1 for the 25 close pairs (if medium-density close pairs are omitted ΔV is even smaller, but the difference is not statistically significant). The effect of isolation is strong for the entire sample of galaxy pairs with separations as large as 1.0 Mpc. For these larger separation pairs, relaxation of strict isolation requirements introduces small groups into the sample, which dramatically increases the median ΔV. We find little evidence of an increase in the median ΔV with decreasing r_p, nor with increasing total luminosity. For our isolated pairs in low-density regions, the overall median ΔV is only 30 km.s^–1. For similar separations and isolation criteria, galaxy satellites with larger luminosity ratios (i.e., less dynamical friction) in higher density regions have ΔV approximately twice as large. We conjecture that our orbits are highly eccentric, so that the indirect effect of dynamical friction leads to predominantly small ΔV. However, the halos of our galaxies may also be of low density (although highly extended).