A survey for infall motions toward starless cores.: III.: CS (3-2) and DCO+(2-1) observations

We present CS ( 3 - 2) and DCO+ (2 - 1) observations of 94 starless cores and compare the results with previous CS ( 2 - 1) and N2H+ ( 1 - 0) observations to study inward motions in starless cores. Eighty-four cores were detected in both CS and DCO+ lines. A significant number of CS ( 3 - 2) profiles and a small number of DCO+ ( 2 - 1) lines show the classical infall asymmetry'' similar to that seen in CS ( 2 - 1) observations. The DCO+ ( 2 - 1) lines, however, usually show a single Gaussian peak. The integrated intensity of N2H+ correlates well with that of DCO+ ( 2 - 1) but poorly with that of CS ( 2 - 1) and CS ( 3 - 2), suggesting that CS suffers significantly more depletion onto grains than do either DCO+ or N2H+. Despite these depletion effects, there is evidently enough optical depth for the CS ( 3 - 2) and CS ( 2 - 1) spectral lines to exhibit infall asymmetries. The velocity shifts of the CS ( 3 - 2) and ( 2 - 1) lines with respect to N2H+ correlate well with each other and have similar distributions. This implies that, in many cores, systematic inward motions of gaseous material may occur over a range of density of at least a factor similar to4. We identify 18 infall candidates based on observations of CS ( 3 - 2), CS ( 2 - 1), DCO+ ( 2 - 1) and N2H+ ( 1 - 0). The eight best candidates, L1355, L1498, L1521F, L1544, L158, L492, L694-2, and L1155C-1, each show at least four indications of infall asymmetry and no counterindications. Fits of the spectra to a two-layer radiative transfer model in ten infall candidates suggest that the median effective line of sight speed of the inward-moving gas is similar to0.07 km s(-1) for CS ( 3 - 2) and similar to0.04 km s(-1) for CS ( 2 - 1). Considering that the optical depth obtained from the fits is usually smaller in CS ( 3 - 2) than in ( 2 - 1) line, this may imply that CS ( 3 - 2) usually traces inner denser gas in higher inward motions than CS ( 2 - 1). However, it is also possible that this conclusion is not representative of all starless core infall candidates, because of the statistically small number analyzed here. Further line observations will be useful to test this conclusion.