1 Introduction

Observations of molecular clouds at millimetric wavelengths suggest that these structures are prone to a broad variety of kinematic disturbances. In particular, and at the largest scales, there appears to be frequent evidence for systematic gradients in velocity, extending (in certain cases) over tens of parsecs. Such gradients may arise through a broad variety of mechanisms, and it seems likely that dynamical shearing between clouds and the ambient medium (e.g. Arquilla & Goldsmith 1986; Goldsmith & Sernyak 1984; McCutcheon et al. 1986), chance superposition of kinematically differing cloud structures (Thronson et al. 1985; Wang et al. 1993; Ziurys et al. 1981), outflows from young stars (e.g. Tafalla et al. 1993), and the collective impact of winds emerging from HII regions, OB associations, supernovae and so forth (e.g. Gonzalez-Alfonso et al. 1995; Patel et al. 1993; Blitz 1993) may lead to similar variations.

In the large majority of cases, however, such systematic trends in LSR velocity have been attributed to cloud rotation.

The importance of rotation in the evolution of molecular clouds is far from adequately established. Whilst certain upper limit estimates for velocity gradient have been used to adduce a comparatively small contribution to cloud stability (Myers et al. 1991; Nozawa et al. 1991; Myers & Benson 1983; Loren et al. 1983; Dickman & Clemens 1983), other observations suggest that rotation may play a formative and important role in at least certain clouds, complexes (e.g. Field 1978; Blitz 1993; Arquilla & Goldsmith 1985, 1986), and disks (e.g. Jackson et al. 1988; Vogel et al. 1985).

The origins of such rotation remain uncertain, as is the importance of this contribution at various stages of cloud evolution. Thus, and most commonly, it is envisaged that there is a transfer of angular momentum from large to small scales as complexes contract and fragment, with radiation of Alfven waves and inter-clump interactions subsequently leading to various degrees of rotational braking. By contrast, Fleck & Clark (1981) have suggested that rotation in clouds may derive from the vorticity associated with IS turbulence.

These and other mechanisms imply differing distributions of angular momentum with cloud radii, and have implications for the comparative importance of angular momentum in maintaining cloud stability, and the orientations of angular momentum vectors with respect to the galactic plane.

Although there have been several previous attempts to assess the properties of cloud angular momentum from observations, systematic studies of this phenomenon have been restricted to comparatively few sources (e.g. Fleck & Clark 1981; Field 1978; Goldsmith & Arquilla 1985; Arquilla & Goldsmith 1986; Heyer 1988; Casali & Edgar 1987), and have come (in certain cases) to indecisive or contradictory conclusions.

In the following, we compile a data base of cloud rotational measures which is at least an order of magnitude greater than has been investigated heretofore. This is used to evaluate overall trends in angular momentum, the orientation of angular momentum vectors, the stability of molecular clouds, and the role of rotation in determining departures from cloud sphericity. We shall find that most current models for the evolution of rotating clouds are to varying degrees inadequate.