We have yet to understand the magnetic activity cycles of cool close binary systems of sunlike stars. Mutual tidal interactions, as well as magnetic ones, may result from a regime of dynamo models not yet tested, because these have been developed for single stars. To arrive at the basic physics, though, requires that we first examine the phenomenology of magnetic activity for binary systems. In particular, we would like to discover if such activity has a clearly-defined cycle, such as the sun exibits.

Among the proxy indicators of magnetic activity are the Ca II H and K lines. Strassmeier et al. (1988) used the strength of these lines as the primary criterion for the inclusion of systems in The Catalog of Chromospherically Binary Stars. Of the RS CVn stars in the catalog, 12 have orbital periods of one day or shorter; 9 are eclipsing systems. As part of a decade-long program, we have focussed our observations and models on eight of the short-period group (Hall, 1976): XY UMa, UV Psc, SV Cam, RT And, CG Cyg, ER Vul, BH Vir, and WY Cnc. These close systems are tidally-locked in synchronous rotation and tidally-distorted into Roche lobe configurations.

Quantitative analysis of the variations of EV Lac as being caused by spottedness leads to the conclusion that the total areas of starspots vary strongly – up to two or threefold – from one season to another but the starspot temperatures remain constant.

New photometric data and two older data sets for AC Boo can be represented with a common set of Roche model parameters and two starspot pairs which drift only in longitude and are located on the cooler component.

New starspot photometry of BY Dra and EV Lac, extending over a decade is analyzed in combination with data from the literature to study the long-term behaviour of spottedness on these stars. Both stars show brightness suppression as well as brightness modulation so spots exist both near the poles and at lower latitudes. Polar spots have been continuously present on BY Dra for the last 14 years. (O–C) diagrams for lightcurve minima reveal systematic, sinusoidal variations in BY Dra with a period of 13–14 years. In EV Lac the spots have been located on the same hemisphere for the last ten years.

The active regions of FK Comae Berenices show a flip-flop behaviour, i.e. the concentrated part of spot-activity shifts exactly to the other side of stellar surface, and then remains on the same longitude for a time interval from a few years to a decade. The activity shows excellent phase coherence with respect to these two active longitudes separated 180 degrees from each other. FK Comae may provide a physical example of a non-linear dynamo, which shows surprisingly simple observational changes in the pattern of the magnetically induced spot configurations.

Cyclic variations of the parameter O-C (the difference between observed and calculated times of the light minimum in eclipse) have been detected for Nova 1934 DQ Her in the U BV RI-bands during 1982-89 with a recurrence time of about 5-6 years and amplitudes of about 2-4 min. No O-C oscillations for the nova-like AC Cnc have been discovered with the resolution of 45 sec in 1985-89. Both systems consist of a white dwarf and a late-type dwarf component. Some possibilities to explain their O-C-behaviour are discussed.

At McDonald Observatory we have been monitoring the relative radial velocities of a sample of K giants. The technique employed uses the telluric O2 lines near 6300 Å as a reference for measuring the stellar line shifts. We demonstrate that precisions of 10 m s−1 are possible with this technique. We present radial velocity data covering a 2 year time span for α Boo, α Tau, and β Gem. All of these stars show both long term variations (~ several hundred days) with a peak-to-peak amplitude of about 400 m s−1 and short term variations (~ few days) with a peak-to-peak amplitude of about 100 m s−1. The long term variations may be due to the rotational modulation of surface active regions whereas the short term variations may be indicative of pulsations.

There has been considerable progress recently in the study of magnetic fields on late-type stars. Advances in the theory include investigation of systematic effects, new and improved methods of analysis for unpolarized and circularly polarized lines, new models of stellar dynamos and of broadband linear polarization, and the first studies of the thermodynamic nature of stellar magnetic regions and their vertical structure. Observationally, there have been new measurements, particularly of young and active stars (including the first detection of a field on a pre-main sequence object), a new monitoring campaign, the first indications of the relative temperatures in stellar plages, and the first measurements of fields in a single stellar active region. I discuss the results in the framework of stellar activity and surface structure.

We attempt to detect a magnetic field on the weak T Tauri star, TAP35, via the enhanced equivalent widths of Zeeman-broadened absorption lines. We synthesize 25 Fe I lines, having a range of Zeeman sensitivities, using an LTE Stokes line-transfer calculation. The oscillator strengths of all lines are empirically determined a priori using the same line-transfer code applied to the spectrum of the magnetically quiet star, τ Ceti. The Fe abundance of TAP35 was established by synthesizing lines that are insensitive to Zeeman splitting. We find that the equivalent widths, Weq, of Zeeman-sensitive lines in TAP35 are systematically enhanced relative to the Zeeman-insensitive lines, consistent with the presence of widespread, kilogauss fields. The excess Weq can be explained by a product of field strength and surface filling factor (Bf) of 1 kiloGauss. A strong upper limit can be placed on the product of those two quantities, Bf < 2 kG. This measurement bears on the physics of T Tauri coronae, chromospheres, dynamos, and accretion-disk boundary layers.

We simulate the radiative transfer of Zeeman-broadened lines in an active star containing both quiet and flux-tube regions. A traditional one-atmosphere Zeeman analysis of these synthetic lines yields magnetic field measurements systematically inaccurate by up to 40%. However, two atmospheric components alone cannot reproduce the observed line strengths, luminosities or photometric constancy of active stars.

I present preliminary results of a multi-line Zeeman analysis of the active star ξ Boo A and the inactive star τ Cet. High S/N echelle spectra are compared with LTE model line profiles for 25 Fe I lines to deduce surface magnetic fields.

We explore the use of the flux difference between the lines Fe I 5250 Å and Fe I 5247 Å as a technique to measure photospheric magnetic fields in late-type stars. The technique developed takes into account all the LTE physics of the problem, assuming a radial and homogeneous magnetic field distribution over the stellar surface. Some test calculations, in order to prove the feasibility of the method, are shown.

We present a very preliminary and simplified analysis designed to measure photospheric magnetic fields in late-type stars, using the FeI 5247.06 Å and 5250.22 Å lines. We show how the use of the equivalent widths of the lines and differences in their depth can give a rapid estimation of the magnetic flux.

Observations and theoretical models of broadband linear polarization (BLP) in cool stars are discussed. Two effects, 1) differential saturation in magnetically sensitive absorption lines and, 2) Rayleigh and Thomson scattering are considered as the possible causes of the BLP. Results of theoretical models, and comparisons with observations are summarized.

The intrinsic polarisation parameters P, θ, Q and U of the star μ Cep, in the V spectral region during the period 1979-1989, are presented. The data were obtained from the Belgrade polarimetric observational program. The polarization percentage of the star exhibits change is of about 2.5% during this interval and the position angles change by about 90 degrees. These very large long-term changes of polarization parameters are discussed briefly.

For a sample of RS CVn binaries Busso and collaborators (Busso et al., 1987, 1988; Busso and Scaltriti, 1990) compared the observed energy distribution from the UV to the near-IR (and to IRAS bandpasses for some objects) with those calculated from the known radii and spectral types of the stellar components. About 25% of the sample gave evidence of an IR excess up to 0.7 mag, generally starting from the I-band. The excess is significantly larger than observational error. These same authors ascribe the excess to a thin dust shell surrounding the binary. Using the “dirty silicates” model by Jones and Merrill (1976) they find for the shell a temperature range of 1500 K to 2000 K, an optical depth of about 0.01 at 10 micrometers in the shell, and a distance of the shell from the stars to be typically about 100 times the radius of the larger star.

In this paper we examine the link between observed magnetic fields and filling factors in main-sequence stars of types G and K, and the results obtained for these parameters from a simple dynamo model. We explore how the predicted magnetic fluxes, fB, for a selected sample of stars, vary with rotation, adopting two theoretical approaches to estimate the filling factor, and considering different expressions for the variation of the stellar angular velocity, ω, with depth.

Whereas the rotation-activity relationship for single cool stars can be described in terms of fundamental stellar properties, the activity in relatively close binaries is enhanced with respect to that of single stars by an amount which depends strongly on the properties of the companion. We consider mechanisms which could cause an excess in activity of binaries, and conclude that it is likely that the presence of a companion affects the interior structure of the stars in such a way that either the efficiency of the dynamo or of the atmospheric heating is enhanced.

Several years ago we proposed a method for the analysis of X-ray observations of late-type stars. It allowed the determination in a uniform manner of coronal base electron densities for more than 40 late-type stars, in terms of a one-temperature consideration of homogeneous spherically symmetric coronae (Katsova et al., 1987). Fig. 1 shows the results as a function of spectral type. Comparison of our results with values for different kinds of solar regions shows that physical characteristics of F and G star coronae correspond to densities less than those in active regions on the Sun. Values for the active K-M0 stars are comparable with those of dense steady condensations found directly above large sunspots.

On this basis, activity can be explained as an increase in that part of the stellar surface that is occupied by strong local magnetic fields. This is illustrated in the table where we compare magnetic field measurements by Saar and Linsky (1988) with our calculations.

We have observed seven main sequence stars in the Pleiades, with B – V between 0.98 and 1.41 (5100–3900 K), and with a wide range of rotational velocities, in Ca II H and K, Hα and have derived Li abundances. Our results, combined with literature data, indicate that the most chromospherically active stars are rapid rotators, and that at a given effective temperature the Li-rich stars show the highest chromospheric activity. A different surface coverage of active regions could influence the dichotomy observed in the Li abundance distribution.