Ever since Curtis (1918) photographed the nucleus of M87, (see Biretta, these proceedings, for a spectacular sequel) astronomers have known that the universe contains elongated features which they have commonly called jets. As Pringle (these proceedings) has warned us, this handle has been used casually for a quite disparate set of objects and it would be inexcusably naive to expect that only one physical process is at work in all these cases. In this talk, I shall restrict my attention to the well-collimated and bipolar outflows that we observe associated with many active galactic nuclei (AGN), and young stellar objects (YSO), two or three X-ray binaries (XRB) and may find in cataclysmic variables (CV). I will argue that magnetic acceleration and collimation are jointly responsible for most of what we observe. I will start by summarizing, cursorily, the phenomenology of jets in these different environments.
1.1. Active Galactic Nuclei
Double radio sources are now known to be fueled by the associated active galactic nuclei via narrow jets. (Recent reviews include Belvedere 1991, Hughes 1991, cf. also Laing, these proceedings.) A fluid model is generally adopted and radio emission (and in some instances also optical and X-ray emission) is produced by synchrotron radiation from relativistic electrons accelerated in situ. Within the extended radio source, the lower power, "Type 1" jets are typically two-sided and thought to be sub- or trans-sonic; the higher power, "Type 2", one-sided jets are believed to be supersonic and, at least, mildly relativistic. VLBI observations of the compact radio sources often reveal miniature jets within which features appear to be moving superluminally with speeds vob ~ 5-10ch-1, where h is the Hubble constant in units of 100 km s-1 Mpc-1, (Zensus & Pearson 1991, Marscher, these proceedings). These are conventionally interpreted as relativistic jets directed towards us with Lorentz factors 10. Even larger Lorentz factors may be present if "intraday" variability is caused by hyper-relativistic expansion (Quirrenbach et al. 1989). Jets are also observed at optical and X-ray energies. Recent Compton Gamma Ray Observatory detections of "blazars" also implicate relativistic jets.
1.2. Young Stellar Objects
Of more recent vintage are the bipolar outflows now commonly associated with YSO (e.g., Königl & Ruden 1992, Reipurth, Ray, Stahler, these proceedings). These comprise poorly collimated molecular outflows, higher velocity neutral outflows and well-collimated, ionized jets with speeds in the range 200 - 700 km s-1. The optical jets often exhibit "Herbig-Haro" objects, which are rapidly moving knots that are associated with traveling shock waves. The relative thrusts and discharges in these different components are difficult to estimate.
1.3. X-ray Binaries
The most famous Galactic jet is, undoubtedly, SS433 (Vermeulen, these proceedings). The measured jet speed has the enigmatic value of 0.26 c and processes on a cone with vertex angle 20°. Despite our detailed understanding of the kinematics, the dynamics remains a mystery. Cyg X-3 is also a powerful radio source that exhibits bipolar outflow with a similar speed (Strom et al. 1989). As reported here by Mirabel (these proceedings), the source 1E1740.7-2942, in the Galactic Center, produces a pair of jets. (It is, though, possibly, a chance superposition of an extragalactic source.)
1.4. Cataclysmic Variables
Although they contain the best studied accretion disks, CV (low mass, white dwarf binaries), have so far produced the weakest evidence for bipolar outflow. They seem to lose appreciable amounts of mass through high speed (~ 5000 km s-1) winds accelerated well away from the white dwarf (e.g., Drew 1987, Mauche & Raymond 1987). The UV line profiles are consistent with bipolar outflow but are also attributable to bipolar illumination of a spherical outflow.
The question I have been asked to address is how can these very different types of source produce such morphologically similar outflows?