BACKGROUND

The first concepts of star formation via gravitational collapse were established by Pierre-Simon Laplace, whose Exposition du Système du Monde (1796) gives a vivid scenario of the formation of the Sun and the solar system from a rotating ``nebula''. In the first half of this century, Edward Barnard and others speculated that dark clouds were the sites of stellar birth (Trimble 1996). Albert Joy (1945) reported a class of unusual emission-line and variable stars near dark clouds representing newly formed stars, now called classical T Tauri (CTT) stars, which were found to be frequently grouped in ``T associations'' (Ambartsumian et al 1982).

During the 1980s and 1990s, a revolution took place in infrared and millimeter astronomy allowing direct observational search for the earlier infall and protostellar phases embedded deep in the clouds. Although the simple Laplacian picture lays at the foundation of contemporary models, the detailed astrophysics of YSOs proved to be much more complex than a simple self-gravitating nebula, and its understanding has not proceeded in a simple fashion. Much of the thinking can be viewed as the interweaving of five themes.

1.

Pre-main sequence stellar evolution The basic picture established by early analytical treatments of YSO stellar interiors (Hayashi 1966) is still in force. T Tauri stars have fully convective stellar interiors with monotonically decreasing luminosities but nearly constant surface temperatures, powered principally by gravitational contraction rather than nuclear reactions. Models basedf on different assumptions regarding atomic opacities and convective structure give somewhat different isochrones on the Hertzsprung-Russell diagram. The rotational evolution of stars along the Hayashi tracks may be complex, and the resulting magnetic field generation is not well-established. A YSO with a differentially rotating interior could generate a core field as high as 4#4 G in 10³ yr, but turbulence without differential rotation would dissipate the field (Levy et al 1991).

2.

Outflows The broad emission lines of CTT stars often exhibit P Cygni-type profiles and were originally interpreted as dense hot winds (Herbig 1962, Kuhi 1964). Collimated outflow bowshocks seen as Herbig-Haro objects, small-scale optical jets, and molecular bipolar outflows were later found to be common among the younger YSOs (reviews by Bachiller 1996, Reipurth & Bertout 1997). It was readily perceived that the YSO outflows are not accelerated by radiation or coronal gas pressure, but required the intervention of magnetic fields and the circumstellar disk (e.g. DeCampli 1981, Pudritz & Norman 1983, Uchida & Shibata 1984). Although there is a consensus that magnetic fields confine and accelerate outflows, the detailed acceleration mechanism close to the star is still the subject of lively discussion (Pudritz et al 1999).

3.

Disks While the optical obscuration of most YSOs was originally attributed to a spherical cocoon of unaccreted dusty material, the discovery of intense emission from the micron to the millimeter bands from CTT stars over the past 15 years requires that large amounts of dust must be present in a flattened disk. These protoplanetary disks can now be directly imaged in emission with millimeter interferometers (e.g. Dutrey et al 1994) and the Hubble Space Telescope in visible light silhouette (McCaughrean & O'Dell 1996) or in near-infrared emission (Stapelfeldt et al 1998)

4.

Accretion While star formation theory predicts stellar growth by accretion from a large circumstellar envelope (Shu et al 1987), direct evidence for infall proved elusive for many years. Doppler signatures of gas infall are now seen in the earliest protostellar phases (e.g. Zhou et al 1993, Mardones et al 1997). Ballistic infall models including rotation indicate that the envelope feeds a central accretion disk several hundred AU in size (Terebey et al 1984), consistent with recent disk imaging. Some molecular line evidence of disk accretion has also been found (Ohashi et al 1996). After the envelope is cleared by outflows, this disk remains and the material accretes onto the central star on timescales of 10⁵-10⁶ yrs. This process can proceed in both a relatively steady fashion, seen as CTT stars, or in short-lived episodes of high accretion with associated ejection of material, seen in YY Ori stars (Bertout et al 1996) and at very high levels as FU Orionis stars (Hartmann et al 1993). Detailed models of T Tauri permitted line profiles and continuum veiling - historically attributed to outflows or star-disk boundary layers - are now explained as emission in magnetically-confined accretion columns (Hartmann et al 1994, Hartmann 1998, Calvet et al 1998). Accretion also affects pre-main sequence evolutionary tracks (Siess et al 1997).

5.

Magnetic activity Several independent lines of evidence indicated that YSOs exhibit unusually high levels of magnetic activity. The original perception by Joy (1945) that YSO spectra share characteristics with the solar chromosphere led to the development of T Tauri stellar atmospheres with large plage regions and deep chromospheres (Herbig & Soderblom 1980; Calvet et al 1984; Finkenzeller et al 1987). Hundreds of flash variable stars in star-forming regions were found, suggesting analogy with dMe flare stars (Haro & Chavira 1966, Ambartsumian & Mirzoyan 1982). Powerful X-ray flares from T Tauri stars were found with the first satellite-borne imaging X-ray telescope, which was generally interpreted as enhanced solar-type emission (Feigelson & DeCampli 1981a; Walter & Kuhi 1981; Montmerle et al 1983; Feigelson et al 1991). Gyrosynchrotron radio continuum flares, similar in character but again orders of magnitude stronger than solar levels, were found in some YSOs (Feigelson & Montmerle 1985; White et al. 1992b).

Although this review concentrates on the theme of magnetic activity, we trace recent efforts to synthesize all previous approaches into a coherent understanding of YSO astrophysics. For example, we discuss the possible origin of flares in magnetic field lines connecting the star to the disk, and possible roles of energetic radiation for ionizing the disk and promoting accretion and outflow. We summarize here some central concepts that underlie our discussion of magnetic and high energy processes, recognizing that our treatment does not adequately present the rich phenomenology and astrophysics of YSOs (cf. Levy & Lunine 1993, Mannings et al 1999).

Figure 1 illustrates the principal phases of YSO evolution: protostars, classical T Tauri (CTT) stars and weak-lined T Tauri (WTT) stars. To reduce the effects of foreground extinction (TTS are often optically visible whereas protostars are deeply embedded in their parent cloud), the evolutionary phase of YSOs is generally classified by their infrared-millimeter spectral energy distributions (Lada 1987, André & Montmerle 1994).

Class 0

infrared-millimeter sources are young protostars with massive, cold (5#5 K) envelopes which collapse towards the central regions. A collimated outflow and a disk rapidly form within the envelope which is 10³-10⁴ AU in size (Figure 2 left). The age of Class 0 sources is estimated to be 3#3 yrs.

Class I

sources have ages around 6#6 yr. Most of the material in the envelope has accreted onto the disk or star, and the disk is a few hundred AU in extent (Figure 2 middle). Outflow activity is still present, but with a larger opening angle and lower mass-loss rate that at the Class 0 stage (Bontemps et al 1996).

Class II

is the infrared designation of CTT stars. Most of their complex phenomenology can be modeled as a star interacting with a circumstellar accretion disk (Figure 2 right). The youngest members of the class drive outflows, and all drive strong winds with 7#7 yr^-1 and 8#8 km s^-1. Contemporary models are based on magnetically confined accretion from a magnetosphere extending out to the corotation radius (Figure 3). When Class II sources are unobscured, they can be placed on the Hertzsprung-Russell diagram and compared to theoretical evolutionary tracks. The derived ages are mostly between 0.5 and 3 Myr, though some stars retain CTT characteristics as long at 9#9 Myr.

Class III

infrared sources, or WTT stars, are simple blackbody spectral energy distributions, implying little or no accretion disk (Wolk & Walter 1996). Many occupy the same region on the Hertzsprung-Russell diagram as CTT stars, though some are approaching the zero-age main sequence (ZAMS). It is tantalizing to surmise that the loss of disks from the Class II to III phases is accompanied by planet formation, as roughly 1/3 of CTT stars have disks with masses larger than the primitive solar nebula (e.g. Beckwith et al 1990, André & Montmerle 1994). For stellar ages > 20-30 Myr, all indications of circumstellar disks disappear and we enter the regime of 'post-T Tauri' stars. These stars, long missing from YSO samples (Herbig 1978), are now emerging from wide-field X-ray surveys (§6). They are distinguished by their location above the ZAMS (though absence of accurate distances frequently impedes accurate placement on the Hertzsprung-Russell diagram) and by photospheric lithium abundances above those seen in ZAMS stars, because the initial lithium is easily destroyed on the way to the main sequence destroyed as a result of convective mixing (Martín 1997).

The interested reader can consult a number of related reviews. Broad treatments of YSOs can be found in Annual Review articles by Shu et al (1987) and Bertout (1989), a monograph by Hartmann (1998), and in conference volumes edited by Lada & Kylafis (1991), Levy & Lunine (1993) and Mannings et al (1999). Various aspects of magnetic activity and flaring in YSOs have been reviewed by Feigelson et al (1991), Montmerle (1991), Montmerle et al (1993), and Glassgold et al (1999). Radio emission is reviewed by André (1996), and recent X-ray results are summarized by Neuhäuser (1997c). Some meteoritic implications are discussed by Woolum & Hohenberg (1993).