1 Introduction

Late-type stars are known to sustain a dynamo which is powered by the combination of convective motions and rotation. The resulting magnetic field is thought to be responsible for various observational phenomena commonly referred to as "activity''. Stellar activity indicators are pervasive in all layers of the atmospheres of late-type stars. The best-studied magnetic field tracers include chromospheric H$\alpha$ and Ca II emission lines and coronal X-rays.

Strong and variable X-ray emission is observed from the early pre-main sequence (PMS) T Tauri stage to main sequence flare stars. Comparative studies of the emission levels of young stars at different ages may shed light on the origin and evolution of magnetic activity, which may be linked to angular momentum evolution.

T Tauri Stars (TTS) are divided into two subclasses according to the strength of their H$\alpha$ emission. Classical TTS (cTTS) are characterized by strong H$\alpha$ emission, while in weak-line TTS (wTTS) the equivalent width of H$\alpha$ is smaller than $10~{\rm\AA}$. In contrast to cTTS, wTTS do not show obvious signs of accretion and optically thick disks, and therefore are thought to represent a later evolutionary stage at which the circumstellar disk has become optically thin or dispersed. Nevertheless, some wTTS occupy the same region in the Hertzsprung-Russell diagram as the cTTS (see e.g. Walter et al. 1988; Alcalá et al. 1997), i.e. there seems to be no preferred disk lifetime between several 10⁵ and $10^7~{\rm yrs}$.

During the PMS contraction stars gain angular momentum and should spin up. However, cTTS show much slower rotation rates than wTTS (Bouvier et al. 1993) despite their high accretion rates. This can be explained by magnetic coupling between the star and the disk in cTTS which allows to regulate angular momentum transport without spinning up the star (Koenigl 1991; Edwards et al. 1993; Bouvier et al. 1997a).

The strength of activity is thought to decline with increasing stellar age. Skumanich et al. (1972) have proposed a $1/\sqrt{t}$-law for the decay of stellar activity in young stellar clusters based on CaII line observations. The power-law relation was confirmed by Feigelson & Kriss (1989) for a sample including PMS objects in Chamaeleon. However, the study of Walter et al. (1988) seemed to indicate that during the PMS phase the X-ray emission remains nearly constant, while for ages $\geq$ $10^8~{\rm yrs}$ it decays exponentially. Walter & Barry (1991) have argued that a power-law decay may be an artefact that occurs when the X-ray luminosity is used as activity indicator in a sample composed of stars with different stellar radii. Using the surface flux Walter & Barry (1991) have shown that the decay of various activity diagnostics probing the lower chromosphere, transition region, and corona can be described by an exponential.

An evolutionary decay of the X-ray emission is favored also by studies of the Einstein Observatory (EO) which have shown that the X-ray luminosity functions (XLF) of stellar clusters with different ages are displaced from each other (see e.g. Feigelson & Kriss 1989; Damiani et al. 1995). However, rather than being a pure age effect, the decreasing activity (i.e. the decay of the dynamo efficiency) might be explained by the slowing down of the rotation with increasing age on the main sequence (MS). A connection between the dynamo efficiency and the rotation rate is also supported by correlations between the X-ray activity and the rotational velocity (see Pallavicini et al. 1981; Bouvier 1990; Neuhäuser et al. 1995 = N95) found to hold for all kinds of active stars, featuring such different objects like dMe stars, RS CVn binaries, and TTS. For the fastest rotators among the Pleiads (Stauffer et al. 1994) and among the wTTS (Damiani & Micela 1995) the relation flattens out suggesting "saturation'' of the X-ray emission at very high rotation speeds. Stauffer et al. (1994) have proposed that the large spread observed in the X-ray luminosities of slower rotators are caused when more and more stars leave the saturation level thus increasing the dispersion. A later study by Micela et al. (1996) showed no significant correlation between $L_{\rm x}$ and ${v \sin{i}}$ for Pleiades stars, indicating that rotation can not be the only parameter governing the X-ray emission level of young stars. To date, no solution has been found to the question whether age or rotation determines the level of stellar activity.

If rotation does determine the activity, then cTTS should show lower X-ray luminosities than wTTS. XLF provide a statistical tool to compare the X-ray properties of different stellar samples. But so far, studies of the XLF of cTTS and wTTS have not led to a conclusive result concerning the evolution of X-ray activity during the PMS. N95 have shown that X-ray emission increases with age from cTTS to wTTS, but decreases after the wTTS phase. While N95 found that wTTS in Taurus-Auriga emit significantly more X-rays as compared to cTTS from the same region, Feigelson et al. (1993) found little evidence for systematic differences in the XLF of wTTS and cTTS in Chamaeleon, and argued that the differences in the X-ray luminosity between wTTS and cTTS in Taurus-Auriga might be attributed to as yet undicovered wTTS at the low-luminosity end of the distribution.

In this paper we examine the connection between X-ray activity, age, and rotation comparing different samples of young stars. The Taurus-Auriga-Perseus region is well suited for a study of the evolution of stellar activity because it hosts a large number of young stars at different ages: the molecular clouds of Taurus-Auriga are sites of ongoing star formation and have produced many TTS with ages between several 10⁵ to $10^7~{\rm yrs}$. Besides this star forming region, two young star clusters are located in the area, the Pleiades and the Hyades, at ages of $10^8~{\rm yrs}$ and $6 \times 10^8~{\rm yrs}$ respectively.

Our sample is larger and the sensitivity improved compared to all previous studies of this issue. The paper presented here is connected to an earlier analysis of archived ROSAT observations (Stelzer et al. 2000). Here we present a list of X-ray parameters for all known TTS in Taurus-Auriga, the Pleiades, and the Hyades as observed by ROSAT during pointed PSPC observations. Both detected and undetected sources are considered, i.e. for non-detections upper limits are given. In Sect. 2 we describe the data base and data reduction and give the results from source detection. In Sect. 3 the Hertzsprung-Russell diagrams (H-R diagrams) for cTTS and wTTS in Taurus-Auriga are presented. Our special interest (Sect. 4) is to compare the XLF of the different stellar groups with respect to the following issues: (i) Are the luminosity functions of cTTS and wTTS different, (ii) how does the X-ray luminosity evolve with stellar age, (iii) how does it depend on the spectral types of the stars and their binary character. Furthermore, we will explore the relation between X-ray emission and rotation rate for the largest sample studied so far (Sect. 5). In Sect. 6 we discuss and summarize our results.