In this paper, I review most of the current knowledge about magnetic field topologies and surface differential rotation of cool active stars like the Sun. I concentrate in particular on the most recent observationalresults, derived from studies on dynamo field generation in stellar convectiveenvelopes from magnetic imaging of very active rapidly rotating low mass stars, and present the elements of evidence that dynamo processes in these active stars are distributed throughout the whole convective zone, rather than being confined in a thin interface layer as in the Sun.I finally discuss the new discovery that dynamo magnetic fields generate feedback effects on the global dynamics of stellar convective envelopes.

Many stars exhibit strong magnetic fields, some of which are thought to be of primordial origin and others a sign of magnetic dynamo processes. We briefly review the results of observational studies of solar-type stars seeking to evaluate the linkage between rotation rate and possible magnetic cycles of activity. Clearly turbulent convection and rotation within spherical shell geometries provide ingredients essential for dynamo action. However, intensive efforts over several decades in solar research have demonstrated that it is no easy matter to achieve cyclic magnetic activity that is in accord with observations. Helioseismology has revealed that an essential element for the global solar dynamo is the presence of a tachocline of shear at the base of the solar convection zone, leading to the likely operation of an interface dynamo. We review the crucial elements for achieving a cyclic magnetic activity. We then discuss some of our current 3-D MHD simulations of solar turbulent convection in spherical shells that yield differential rotation profiles which make good contact with some of the helioseismic findings. We show that such turbulent motions can amplify and sustain magnetic fieldin the bulk of the convective zone whose strength are sufficient to feedback both upon the convection and its global circulations.

During the last decade, it has been suggested that the structure of the solar photosphere, in the quiet sun, varies over the 11-year activity cycle but this variability of various quiet sun parameters (intensity, granule sizes, overshoot, etc.) has not been fully demonstrated yet. The recent advance in the knowledge of the solar granule properties (flow evolutions, size distribution, spatial organization, interaction with magnetic field, line asymmetry, penetration in stable layers, morphological properties, etc.) let us optimistic to try to understand and to measure the cyclic variations of the photosphere.

This paper is devoted to review the development and the results of the program "solar prominences" that has been aimed to observe the Hanle effect at the Pic-du-Midi during the ascending phase of Cycle XXI (1974-1982). This aim had been defined and the observations have been performed by Jean-Louis Leroy. The Hanle effect is the effect of a weak magnetic field on the scattering linear polarization: its main features are, for some field orientations, a depolarization and eventually a rotation of the polarization direction. The magnetic field diagnostic from polarization measurements requires a modelling of the polarized line formation, that has been achieved in Meudon in the well-adapted formalism of the atomic density matrix. It is shown how the program has been developed to determine the 3 components of the field vector and the electron density, by setting multi-line polarimetric observations. Particular attention has been devoted on the solution of the 180 degrees ambiguity, which has been solved by 3 independent methods. By using this solution, one unique average magnetic field vector has been determined in each of 296 quiescent prominences, leading to results on the field strength, direction, vertical gradient, cyclic variations. The future perspective opened by the low scattered light level of THEMIS and other spectropolarimeters is to increase the spatial resolution of the measurements.

Measurements of magnetic fields in the solar corona are needed toimprove our understanding of structures, dynamics and heating of the corona and of the acceleration of the fast and slow modes of the solar wind expending into the heliosphere.However, those measurementsare very difficult andstill very scarce.I revue the main methods proposed to access the direction and the strengthof the coronal magnetic field and discuss what seem the main avenues toimprove its determination in the future.

Rapidly rotating solar type stars display varying amounts ofchromospheric emission in their Balmer lines. The profiles of theselines often display rapid variability. Since the mid-1980s,high-cadence echelle spectroscopy of numerous such stars has revealedthat some of the variability is caused by concentrations of neutralhydrogen, trapped in the stellar coronal magnetic field at distancesup to a few stellar radii from the rotation axis. In this paper wereview the observations and the tomographic techniques that can beapplied to mapping these "slingshot prominence" systems. We discussthe possible mechanisms for formation and mechanical support of thesecondensations, in the light of recent efforts to map 3-dimensionalstellar coronal magnetic fields and X-ray emission measuredistributions, with the help of Zeeman-Doppler imaging.

Over a range of spatial scales, the onset regions of the fast solar wind are influenced or controlled by the structure of the local magnetic fields. This is demonstrated for the large scale structures that determine coronal holes, as well as for the supergranulation network. It is likely that reconnexion effects in small-scale emerging fields play a role in the initial accelerating mechanism. The role of magnetic fields in outflowing solar plumes is at present less clear.

This review presents a general overview of the magnetic fields foundin main sequence A and B stars, and of some of the phenomenaassociated with this magnetism. An attempt is made to place theobserved magnetic fields into an evolutionary context.

Observations of magnetic Ap stars taken at the 2m Telescope BernardLyot (TBL) of Pic-du-Midi Observatory are of the greatest importancein the study of the stellar magnetism. With the help of the polarimeterSTERENN, in the '90s, Leroy and collaborators performed a vigorouscampaign of observations of broadband linear polarization of Ap stars,virtually disclosing a new window on the transverse components of thestellar magnetic fields. Less than 10 years later, Donati, Landstreet,Wade and other collaborators are routinely using MuSiCoS in polarimetricmode to observe Stokes profiles of Ap and other kinds of magnetic stars.I will review the impact of TBL observations on the study of magneticfields of Ap stars, and the progresses in the modelling that have beenaccomplished in the last ten years.

Ultra-high precision measurements of Stokes V Zeeman signatureshave been obtained using the MuSiCoS spectropolarimeter (Donati et al. 1999) at the Pic du Midi obseravtory, aimed at studying the distribution of weak andtangled magnetic fields in F, A, and late-B type stars.We will discuss recent results obtained from these data, including thecomplete absence of magnetic fields (with an upper limit of around 25 G)in the non-Ap/Bp sample, and the relatively easy detection of magnetic fieldsin nearly all of the Ap/Bp stars.In addition polarimetric data in all four Stokes parameters (I, V, Q, and U) has beenobtained for a sample of roAp stars, a pulsating subgroup within the magnetic Cp stars, with the aim to study vertical (stratification) and horizontal (via Zeeman Doppler Imaging) abundancecharacteristics and their relation to the magnetic field geometry. We will presentresults of our data analysis, using the Least Squares Deconvolution (LSD)technique (Donati et al. 1997) and with the goal of determining the surface abundance distribution of rare earthelements and the magnetic field geometry of these stars.

We have searched for magnetic fields in several early B-type stars:βCephei, Slowly Pulsating B (SPB) and in particular Be stars. We recallthe properties of these stars and highlight the problem of the Be phenomenon. Wediscuss two clues towards understanding this phenomenon in terms of pulsations andmagnetic fields. We summarize results from spectropolarimetric observationsobtained at the Pic du Midi (France) and highlight the results for the Be starω Ori and the β Cep star V2052 Oph.

We report on our discovery of a magnetic field in the B2IV star ζ Cas. We determine the rotation period from the UV wind behaviour. We summarize the results from spectropolarimetric observations of ζ Cas obtained in 2001and 2002 with the MuSiCoS échelle spectropolarimeter at the 2 m Télescope Bernard Lyot (TBL). We also find a non-radial pulsation mode and confirm that the star is N-enriched. This is the third detection of a magnetic field in an early B-type variable star and the first one in a SPB star.

We report on spectropolarimetric observations of 7 A-type supergiantstaken with the MuSiCoS echelle spectropolarimeter mounted at the 2 m Bernard Lyot telescope at the Pic du Midi, during June-July 1999 and June-July 2002. We present the values of the derived longitudinal component of the magnetic field. The significance of the results isdiscussed.

We discuss the interplay between radiatively driven winds andmagneticfields in hot stars.In late main sequence B stars, radiatively driven winds are likelytooccur but with a mass loss ratestrongly affected by the magnetic field structure at the stellarsurface.Away from the stellar surface, the interaction of a radiativelydrivenwind with a dipolar magnetic field leadto the formation of a large magnetosphere consisting of anextended hotpostshock region as well as a dense coolingdisk forming near the magnetic equator (Magnetically Confined WindShockModel, MCWS).In the context of this model, we try to interpret an extensive setof observations (X-ray, UV and optical data) available for the O7VstarΘ1 Ori C as wellas the pulsating B star β Cep. We finallyshow how observations of X-ray lines with the XMM or Chandraspacecraftscould help us to obtain a clearer pictureof the interaction between magnetic fields and the winds of hotstars.

The original title of this contribution was "Magnetic Fields in theVicinity of Young Stars". One of the goals was to show the impact ofthe large scale magnetic field on the gravitational collapse leadingto star formation, and also to discuss some of Jean-Louis Leroy'scontributions to the measurement of the polarisation, hence thedirection of the magnetic in the interstellar medium made withSTERENN, the polarimeter he and Jean-François Leborgne built for thePic-du-Midi observatory. Another goal of the talk was to survey thestate of our knowledge of the magnetic field in the immediatecircumstellar environment of young stars. In doing that we discoveredan interesting correlation. Therefore, this paper presents only abrief overview of the original talk presented at theconference. Instead, it emphasizes largely on what we consider to bethe most interesting aspect of what was presented, namely, thealignement of YSO, and in particular T Tauri stars, with the localmagnetic field in Taurus.

The magnetic field of T Tauri stars is believed to be strong enough to disrupt the inner region of their accretion disk. At the inner disk edge, matter flows along the magnetic field lines at free fall velocity and accretes at the magnetic poles onto the star yielding hot accretion shocks on the stellar surface. I review the observational evidence supporting this magnetospheric accretion process in T Tauri stars and highlight recent results which point to a highly dynamic and time variable interaction between the inner disk and the stellar magnetosphere on timescales ranging from hours to months.

I discuss in this contribution constraints on the origin of mass-lossfrom young stars brought by recent observations at high angularresolution (0.1'' = 14 AU) of the inner regions ofwinds from T Tauri stars. Jet widths and collimation scales, the largeextent of the velocity profile as well as the detection of rotationsignatures agree with predictions from magneto-centrifugal disk windejection models. However dynamically cold disk wind solutions predicttoo large terminal velocities and too low jet densities and ionisationfractions, suggesting that thermal gradients (originating in anaccretion heated disk corona for example) may play an importantrole in accelerating the flow.

The first evidences for magnetic activity in low-mass protostars and brown dwarfs were found in the mid-1990's by the old generation of X-ray satellites, namely ROSAT and ASCA. I will show that these first X-ray detections have opened a promising observational window to study the activity of protostars, deeply embedded in their molecular cloud and optically invisible. However these early observations gave only access to the tip of the iceberg, detecting protostars mainly during phases of high activity. I will review recent results obtained with the new generation of X-ray satellites, namely Chandra and XMM-Newton, which now give access to the X-ray spectral study of protostars also during their quiescent phase, and which are also enough sensitive to detect young brown dwarfs in star-forming regions. I will also review evidences of activity from radio observations, and for brown dwarfs, evidences of activity from optical observations.

A recent observation of the Orion Nebula Cluster withthe Chandra X-ray Observatory detected 1075 sources, providing auniquely large and well-defined sample to study the dependence ofmagnetic activity on bulk properties for pre-main sequence (PMS) starsdescending the Hayashi tracks. The principal result is that X-rayemission does not show the activity-rotation relation which is thedominant correlate of magnetic activity indicators in main sequencestars. Instead, other effects are seen in PMS stars: X-rayluminosities are correlated with stellar bolometric luminosity, massand size for M <2-3 M⊙, but the emission drops precipitouslyin some or all PMS stars with M >2-3 M⊙. The absence of an X-ray/rotation correlation in PMS stars, andparticularly the high X-ray values seen in some very slowly rotatingstars, is a clear indication that the mechanisms of magnetic fieldgeneration differ from those operating in main sequence stars. Themost promising possibility is a turbulent dynamo distributed throughoutthe convective zone, but other models such as a "supersaturated"α - Ω dynamo or relic core fields are not immediatelyexcluded. The evidence does not support X-ray production inlarge-scale star-disk magnetic fields. This work is described indetail by Feigelson et al. (2003).

As potential progenitors of the Ap stars, the pre-main sequence intermediate mass Herbig Ae/Be stars present us with the opportunity to investigate theearly history of magnetic fields at these masses. Although we have nodirect evidence for the presence of magnetic fields in thesestars (except in one case), many clues for such fields have been identified in the past years. They include active phenomena which areoften attributed to surface magnetic fields. Rotational modulation ofspectral lines formed at the photosphere and in the wind of these starsconstitutes yet another indication for the presence of such fields. Future prospects for the detection and mapping of hypotheticalmagnetic fields in Herbig Ae/Be stars are very encouraging, andin particular spectropolarimetric observations of these stars withinstruments like Espadons on CFHT will allow us to test thehypothesis that magnetic fields similar to those found on Ap starsare indeed present in these pre-main sequence objects.