L1544: A Starless Dense Core with Extended Inward Motions

We present a multiline study of the dense core L1544 in the Taurus molecular complex. Although L1544 does not harbor an embedded star, it presents several characteristics of cores that have already undergone star formation, suggesting that it may be rather advanced in its evolution toward becoming a star-forming core. The spectral lines from L1544 present an interesting dichotomy, with the thick dense gas tracers suffering very strong self absorption while CO and its isotopes are not being absorbed at all. The presence of the self absorptions allows us to study both the density structure and kinematics of the gas in detail. A simple analysis shows that the core is almost isothermal and that the self absorptions are due to very subthermal excitation of the dense gas tracers in the outer layers. The density has to decrease outward rapidly, and a detailed radiative transfer calculation that simultaneously fits three isotopes of CO and two of CS shows that the density approximately follows a r^-1.5 power law. The self absorptions, in addition, allow us to measure the relative velocity between the inner and outer layers of the core, and we find that there is a global pattern of inward motions (background and foreground approaching each other). The relative speed between the foreground and background changes with position, and we use a simple two-layer model to deduce that while the foreground gas has a constant velocity, the background material presents systematic velocity changes that we interpret as arising from two velocity components. We explore the origin of the inward motions by comparing our observations with models of gravitational collapse. A model in which the infall starts at the center and propagates outward (as in the inside-out collapse of Shu) is inconsistent with the large extension of the absorption (that suggests an advanced age) and the lack of a star at the core center (that suggests extreme youth). Ambipolar diffusion seems also ruled out because of the large amount of the inward speed (up to 0.1 km s^-1) and the fact that ionized species move with speeds similar to those of the neutrals. Other infall models seem also to have problems fitting the data, so if L1544 is infalling, it seems to be doing so in a manner not contemplated by the standard theories of star formation. Our study of L1544 illustrates how little is still known about the physical conditions that precede star formation and how detailed studies of starless cores are urgently needed.