1994A&A...284..369E

In the present paper we study the appearance of velocity-distance diagrams, when the distances are calculated by either the direct or the inverse Tully-Fisher (TF) relation and when the velocity field is generated by a spherically symmetric mass concentration, and try to understand some contradicting reports on the kinematics close to the Great Attractor (GA). Since the application of Malmquist corrections for the distance moduli is not a straightforward task in the presence of a general velocity field, we also searched for methods for obtaining a minimally biased subsample from the data with direct TF-distance and tested the influence of the chosen inverse relation. We first conducted a series of random number simulations by generating direct and inverse TF-distances and line-of-sight velocities for a synthetic data set. We conclude that instead of the V vs. R_TFdir and V vs. R_TFinv diagrams (both being difficult to correct for the general Malmquist bias, the R vs. V diagrams are the more basic ones when used together in such a way that in the R_TFdir vs. V diagram one excludes galaxies having logV_M values below some beforehand assigned limit (maximum bias allowed) and in the R_TFinv vs. V diagram one calculates the <R_TFinv>_V values being careful with the galaxies in the central cone. These methods are - however - sensitive to the progressive incompleteness of the sample and to the uncertainty in the inverse TF-relation, respectively. We compared our model results with existing real data (Mathewson et al. 1992b). The logV_M-class analysis reveals the frontside infall with the expected angular dependence. However, though there are galaxies that would seem to follow the "S-curve" in the central region, it seems to be impossible to extend the method beyond the putative center, because of the incompleteness of the data. On the other hand, by using a realistic slope for the inverse relation, also the backside infall was revealed up to distances far larger than in any previous sample studied. This is though only a tentative result, the uncertainty being traced back to our knowledge of the inverse TF-relation.