The discovery of extragalactic jets must be considered one of the triumphs of astrophysics, as it involves one of the relatively few actual predictions in this fundamentally observation driven science. Early radio telescopes revealed the existence of extragalactic double radio sources such as Cygnus A (Jennison & Das Gupta 1953), and led to the discovery of quasars in the 1960's. The combination of non-thermal, essentially power-law, radio spectra, and often substantial polarization, very quickly led to the emission mechanism being identified as synchrotron radiation from relativistic charged particles in helical orbits around magnetic field lines. In the early 1970's a few theoretical papers (Rees 1971, Longair et al. 1973, Scheuer 1974, Blandford & Rees 1974) proposed that these extremely large (typical scales of 100 kpc) and powerful (typical LR ~ 1044 erg s-1) sources were fed by jets emerging from the centers of elliptical galaxies. Only with the dynamic range provided by the Very Large Array and other radio telescopes coming on line later in the 1970's were many of these jets actually detected. Competing models, involving blobs (or ``plasmoids'' or ``plasmons'') of radio emitting plasma (e.g., Pacholczyk 1977) or gravitational-slingshots that eject independent engines and sources of plasma (e.g., Saslaw, Valtonen & Aarseth 1974) are now rarely considered. This is because of the convincing evidence of nearly co-linear emission extending from less than a pc to hundreds of kpc in many sources. While we shall not discuss these alternative scenarios any further, it is still worthwhile to note that many of the observations of radio galaxies can be accounted for by the slingshot model (Valtonen & Heinämäki 1999).
The vast majority of these extragalactic radio sources can be easily classified into one of two categories, as first noted by Fanaroff & Riley (1974), who carefully examined the first set of synthesis maps. The weaker ones, or FR I's, with radio fluxes P178MHz < 2 × 1025 h50-2 W Hz-1 sr-1 (where h50 is the Hubble constant in units of 50 km s-1 Mpc-1), tend to have their brightest regions within the inner half of the total extent of the radio source; i.e., much of the flux emerges from the jets themselves, and the fluxes emanating from the two jets are almost always comparable. The FR I sources can be further sub-classified into fairly symmetrical twin-jets, Wide Angle Tail, and Narrow Angle Tail (or head-tail) sources (e.g., Bridle & Perley 1984). Many of the details of these diverse structures can be understood in terms of interactions of the jets with the motions of, or irregularities in, the density and magnetic fields of the external media (interstellar, intracluster and intergalactic) through which they propagate. Other FR I morphologies, particularly S-type symmetries, could be explained by the precession of the jets axes or ejection of the jets from one component of a binary AGN.
More powerful sources, or FR II's, emit the bulk of their radio photons from the outer portions of their structures. Almost all of these are the classical double sources, with most emission coming from lobes which are usually well separated and extend outside the stellar extent of their host galaxy. These lobes contain hot spots near their outer edges. The hot spots are identified with the locations where the bulk velocities of the jets undergo rapid deceleration in shocks (Mach disks); the magnetic fields are compressed in these shocks and the individual electrons (and perhaps positrons and/or protons) are accelerated to very relativistic speeds at those locations, thereby explaining the extraordinary emissivities (e.g., Blandford & Rees 1974). Only very careful observations allow the detection of a jet in FR II sources since they are often thousands of times weaker than the lobes; very seldom is a second (counter-)jet found. These FR II jets are usually very well collimated in comparison with FR I jets (e.g., Jeyakumar & Saikia 2000).
Readers interested in more thorough discussions of most of these topics should first consult the older, but still very useful, reviews of Bridle & Perley (1984) (primarily for the observations), and Begelman, Blandford & Rees (1984) (primarily for the theory). A superb summation of the entire subject can be found in the volume edited by Hughes (1991) and a substantial recent review is by Ferrari (1998). Many excellent conference proceedings could be consulted, including those edited by Hardee et al. (1996), Biretta & Leahy (2000) and Laing & Blundell (2001). In light of the availability of these summaries, only a limited number of classical contributions will be cited in this review, which emphasizes some new work selected from a vast literature.
Radio emitting jets have also been identified emanating from a few binary systems within our Galaxy; these have been dubbed ``microquasars''. In addition, jets have been found emerging from many protostars. This brief review cannot discuss these local manifestations of jets; the reader interested in relativistic jets within our galaxy should consult an excellent recent review by Mirabel & Rodríguez (1999).