Magnetic reconnection at current sheets or in current-bearing arches in the solar atmosphere is generally accepted as the mechanism responsible for the sudden energy release in solar flares. Attempts have so far been unsuccessful to isolate from the observations some unique preconditions which would be necessary and sufficient to ensure rapid conversion of energy by this process. Here we survey recent multi-wavelength observations which illustrate the variety of preflare activity. Multiple structures are now believed to participate in the energy release. Dynamic global coupling of the magnetic fields between a flaring site and the rest of an activity complex is seen from the data to be an important aspect of preflare activity.

We present spectral diagnostics for the fluxes of emission lines, in the spectral range 3600-4400 Å, during the cooling phase of stellar flares on dMe stars. Using these diagnostics, electron temperatures have been computed for flares on AD Leonis, Proxima Centauri and UV Ceti. This preliminary model assumes a single flare loop containing a homogeneous, stationary optically-thin flare plasma.

We report a simultaneous IUE, VLA and ground-based photometric observation of a flare on the dMe star, AD Leo, on 2nd February 1903. The optical flare was extremely impulsive, lasting in total only about 3 mins. A relatively longlived 6 cm flare was observed with the VLA which was initially 100% polarized. An IUE spectrum, taken ≈ 8 min after the onset of the optical U band flare, shows motu than a factor of 2 increase in the Ha II λ 1640Å emission line. The other mid-transition region lines such as C IV λλ154B/52Å show almost no response.

We review the results of simultaneous two-frequency imaging observations of solar microwave bursts with the Very Large Array. Simultaneous 2 and 6 cm observations have been made of bursts which are optically thin at both frequencies, or optically thick at the lower frequency. In the latter case the source structure may differ at the two frequencies, but the two sources usually seem to be related. However, this is not always true of simultaneous 6 and 20 cm observations. The results have implications for the analysis of non-imaging radio data of solar and stellar flares.

An overview is given of the observations of stellar radio ‘flares', defined as radio emission which is both variable in time and created by explosive releases of magnetic energy. The main sources of such flares are late-type Main-Sequence stars, classic close binaries, X-ray binaries, and pre-Main-Sequence stars.

We summarize the interpretations of these observations in terms of the various incoherent and coherent emission mechanisms. The possible importance of a coherent emission process in electrostatic double layers is pointed out.

We briefly indicate the diagnostic importance of radio emission for the flare process in classic and compact stars. In particular we discuss the possible production of radio flares from interactions between an accretion disk and the magnetic field of the central object.

While monitoring the flare star EV Lac with high time resolution using the Space Astrophysical Station ASTRON, a rather strong flare was recorded. During this event, flare emissions were detected in the C IV (λ1550 Å) UV line, in the narrow band continuum at λ2434 Å (28 Å bandwidth) and in the wide wavelength range from 1700 Å to 6500 Å, all emission enhancements taking place within 10 s. About 50 s after the flare start, a fast and very powerful burst of the C IV line took place.

This review of stellar magnetic field measurements is both a critique of recent spectral diagnostic techniques and a summary of important trends now appearing in the data. I will discuss both the Zeeman broadening techniques that have evolved from Robinson's original approach and techniques based on circular and linear polarization data. I conclude with an ambitious agenda for developing self-consistent models of the magnetic atmospheres of active stars.

Flare optical continuum with duration above one minute is emitted by a gas condensation with T ≈ 104 K, which may form in a gas-dynamic process. The light carve slope implies that this flare consists of several elementary events. The features of the initial C IV line burst during one elementary event are determined by numerical simulation. The comparison of theoretical intensity and duration of the C IV burst with the observations of the EV Lac flare on February 6, 1986 shows that the observed elementary event in the C IV line is consistent with the formation of a radiative shock wave with an explosive evaporation of the chromosphere. The possibility of the appearance of C IV doublet emission, accompanying the entire process, is also discussed.

Previous work relating flares to evolutionary changes of photospheric solar magnetic fields are reviewed and reinterpreted in the light of recent observations of cancelling magnetic fields. In line-of-sight magnetograms and H-alpha filtergrams from Big Bear Solar Observatory, we confirm the following 3 associations: (a) the occurrence of many flares in the vicinity of emerging magnetic flux regions (Rust, 1974), but only at locations where cancellation has been observed or inferred; (b) the occurrence of flares at sites where the magnetic flux is increasing on one side of a polarity inversion line and concurrently decreasing on the other (Martres et al, 1968; Ribes, 1969); and (c) the occurrence of flares at sites where cancellation is the only observed change in the magnetograms for at least several hours before a flare (Martin, Livi, and Wang, 1985). Because cancellation (or the localized decrease in the line-of-sight component of magnetic flux) is the only common factor in all of these circumstances, suggest that cancellation is the more general association that includes the other associations as special cases. We propose the hypothesis that cancellation is a necessary, evolutionary precondition for flares. We also confirm the observation of Martin, Livi, and Wang (1985) that the initial parts of flares occur in close proximity to cancellation sites but that during later phases, the flare emission can spread to other parts of the magnetic field that are weak, strong, or not cancelling.

Observations are reported of the Ca II resonance lines and II α in dK and dM stars, made with high S/N ratio and high spectral resolution. Ca II emission is found in all stars observed, and those having weak Ca II exhibit marked Hα absorption. It is found that the strengths of the two kinds of chromospheric, lines are not tightly correlated, an effect which can be shown to be independent of the effective temperature of the stars. The result implies that a one-parameter description (e.g. heating rate) of the chromospheres is not viable. While lateral inhomogeneities are likely to be an important second parameter, we also suggest that the Hα line may be formed in a region considerable higher that in which the Ca II lines are formed.

Mass motions are a principal means by which components of solar flares can be distinguished. Typical patterns of mass motions in Ha are described for chromospheric flare ribbons, remote chromo-spheric flare patches, flare loops, flaring arches, surges, erupting filaments and some expanding coronal features. Interrelationships between these phenomena are discussed and illustrations of each are presented.

It has been shown for solar flares (Donati-Falchi et al. 1985) that the continuum emission at the Balmer discontinuity (the blue “pseudo-continuum”) is a very sensitive tool to determine the electron density. In order to use the same interpretative scheme for stellar flares (Falchi et al. 1988), spectroscopic observations of various flare stars have been performed in June 1987 at ESO observatory (Chile). In this paper we report the analysis of spectra of the star n. 729 (V 1216 Sgr) of the Gliese catalog.

High time resolution, moderate spectral resolution spectra were taken during the decay phase of a ΔU > 3 magnitude flare on the dM5.5e star Wolf 424. This flare shows intense, broad Balmer line emission with extended wings, narrow Call lines and numerous weak emission lines from neutral and singly ionized metals. The time history shows substantial, short-lived enhancements in the line emission. These variations are not always seen in association with continuum enhancements.

Although progress has been made in understanding certain aspects of the physics of solar and stellar flares, there are a number of topics which, in the author's opinion, still pose a problem. We summarize these topics here.

The spectra of a flare on YZ CMi, obtained with a temporal resolution of 60 seconds on March 4 1985 and over the range 3600-4400 A, are analysed using a gas-dynamic model. In this model, the optical radiation in the U-band, outside of flare maximum, is produced by a condensation formed during the gas-dynamic process. With the optical continuum described by a Planck function for a temperature of T ≈ 104K, the emitting source area S ≥ 5 10l7cm2. The hydrogen plasma kinetics of an “8 levels plus continuum” model atom are calculated, and it is shown that the low slope of the Balmer decrement, just after the flare maximum, is connected with a large population in the second level of the hydrogen atom.

The basic results of white-light flare (WLF) photometric and spectrographic observations are reviewed. WLFs represent the most extreme density conditions in solar optical flares and are similar to stellar flares in many respects. It is shown that WLFs originate in the low chromosphere and upper photosphere, and that their huge radiative losses remain difficult to explain within the context of known mechanisms of energy transport.

Relative energies are given for the U, B, V, R and I bands for a-3.8 magnitude U-band flare observed on the dwarf dMe star Gl 234 AB on 28 Feb 1985. This flare had a 45 second rise time and 20 minute decay time. The total flare energy from all five bands during the flare was 7 1031 erg, 34% of this total was from the U-band and 20% from the two near infrared R and I bands. The energy density (per frequency interval) implied a rising continuum towards the red, however this only lasted for approximately 20-40 seconds, i.e. during the impulsive phase, after-which the excess flare emission could not be detected in tlie near infrared bands. Of the various models fitted to the flare data (i.e. optical synchrotron, bound-free emission and free-free emission), bound-free emission seems the most promising.

This presentation reviews selected ideas on the origin of the magnetic field of the Sun, the dynamical behavior of the azimuthal field in the convective zone, the fibril state of the field at the photosphere, the formation of sunspots, prominences, the spontaneous formation of current sheets in the bipolar field above the surface of the Sun, coronal heating, and flares.

This paper discusses the hydrodynamic modeling of flaring plasma confined in magnetic loops and its objectives within the broader scope of flare physics. In particular, the Palermo-Harvard model is discussed along with its applications to the detailed fitting of X-ray light curves of solar flares and to the simulation of high-resolution Caxix spectra in the impulsive phase. These two approaches provide complementary constraints on the relevant features of solar flares. The extension to the stellar case, with the fitting of the light curve of an X-ray flare which occurred on Proxima Centauri, demonstrates the feasibility of using this kind of model for stars too. Although the stellar observations do not provide the wealth of details available for the Sun, and, therefore, constrain the model more loosely, there are strong motivations to pursue this line of research: the wider range of physical parameters in stellar flares and the possibility of studying further the solar-stellar connection.

A detailed analysis of 73 flares on UV Cet type stars observed at the 6 m telescope of the Special Astrophysical Observatory with 3x10–7 s time resolution shows no fine structures on flare light curves with time scales from 10–6 to 10–1 s.