At Utrecht Observatory a 60-channel solar radiospectrograph has been in regular operation since the fall of 1968, in the bandwidth 160–320 MHz, with a time resolution of 0s.03 (DeGroot and Van Nieuwkoop, 1968). In addition to all generally known and classified emissions as noise storms, type IV bursts, type III bursts, type IV continua, a number of unclassified emissions have been observed. The spectrograph is particularly well suited to study short time scale fluctuations, even of small amplitude. One striking feature is an often recurring broad band (> 80 MHz) weak, quasiperiodic fluctuation superimposed on a type IV-like continuum (Figure 1). Since no interferometric measurements were available, it is not known whether it is a stationary or a moving type IV burst. However, due to the sometimes very long duration, we expect it to be a stationary type IV.

The longitudinal component of the magnetic field in quiescent prominences has been measured directly with magnetographs using the Zeeman effect on selected spectral lines (Rust, 1966; Ioshpa, 1968; Harvey, 1969). We know that as a general rule the magnetic field enters the, largely-vertical, sheet-like quiescent prominence on one side and exits on the other. The field traverses the prominence plasma with components both along and at right angles to the long axis of the prominence. It is the purpose of this paper to describe observations that may indicate the relative importance of the two components of the magnetic field, and to derive a distribution function for the magnetic field vectors.

Spectroheliograms of quiet regions obtained by the OSO-G satellite in chromospheric, transition zone, and coronal lines indicate that the chromospheric network as seen in Ca K can be resolved despite the low resolution (30 arc s) of OSO-G. The network pattern becomes more diffuse with increasing height, indicating that the magnetic field pattern spreads with height.

Coronal rotation is determined by means of the spectroscopic method. The mean value of the rotation rate vs. the heliographic latitude is found. At high latitudes the corona rotates much faster than the underlying photosphere. This fact confirms Waldmeier's and Billing's hypothesis that the high-latitude phenomena should depend on the magnetic fields of low latitudes and that the interchange of the matter between the active zone and polar coronal region along the field lines should take place.

In an attempt to summarize the present knowledge on the general magnetic field (gmf) of the Sun we pointed out the fine structure and the statistical nature of the gmf as one of its most important properties. The dipole-like behaviour of the mean polar field strengths is combined sometimes (since 1964) with a bias of the S-polarity flux for both poles. Highly uneven distribution of gmf with latitude and longitude, the disappearance of gmf at the South pole for months, and short period, almost synchronous at both poles, variations in the sign of gmf are pointed out. The fluctuations with time of the mean magnetic field of the Sun seen as a star (as well as mf at different latitudes) shows periodicity connected with the rotation of the Sun and very close agreement with the fluctuations of the interplanetary field (sector structure). The effect of faster rotation of N-polarities as compared with S-polarities as well as the bias of mean solar as well as interplanetary S-polarity fields are also pointed out. The possibility of short time-scale (hours) intrinsic changes in the local pattern of gmf is demonstrated.

When viewed separately in the two polarities, the large-scale pattern of magnetic fields on the Sun gives an appearance reminiscent of a very large cellular structure. These structures and their relation to ‘active longitudes’ are discussed.

The results of longitudinal magnetic field measurements in polar prominences (PP), obtained by Rust during 1964–1965 and by the author during 1966, 1969 and 1970, are considered. The magnetic field configuration in PP is similar to that of an arc. In opposite branches of such an arc the magnetic field has different polarities that do not coincide with the magnetic field polarity in current cycle BMR's. The magnetic field value increases as the angle between the line of sight and filament long axis grows. PP's reflect the behaviour of the border line between the magnetic fields of preceding and current cycles.

Results from observations of the polar magnetic fields made in 1968 with the Mount Wilson magnetograph are reported. The field was of south polarity near the heliographic north pole and of north polarity near the south pole. The inversion line of the field was at latitude +70° in the north hemisphere and −55° in the south, which coincides with the position of the polar prominence zones at that time. Observations made simultaneously at the Crimean Astrophysical Observatory are in good agreement with the Mt. Wilson data and confirm the latitude variation of the field.

A few hundred of Hale's plates from 1914, which he had used to determine the Sun's general magnetic field, have been remeasured. The plates were automatically scanned with a digitized microphotometer, and the treatment of the data was made with a computer. While the visual measurements by Van Maanen had given a strong general field similar to that of a dipole, the computer reductions did not reveal any significant field (stronger than 5 G) at any latitude.

Models relevant to numerical studies of the solar cycle are reviewed briefly with discussions of pertinent physical mechanisms. It is suggested that the observed surface activities are secondary in nature and an example of possible non-axisymmetric steady state solutions is given, together with the results of preliminary numerical computations.

A discussion is given of the dynamics of large-scale solar magnetic fields as a joint action of a transport by a regular velocity field, by a generalized diffusion, by coming to the surface and sinking and by sources of origin and destruction of magnetic fields. Also these processes are discussed using the observed picture of magnetic field dynamics.

The solar sector structure consists of a boundary in the north-south direction such that on one side of the boundary the large-scale weak photospheric magnetic field is predominantly directed out of the Sun, and on the other side of the boundary this field is directed into the Sun. The region westward of a solar sector boundary tends to be unusually quiet and the region eastward of a solar sector boundary tends to be unusually active. This tendency is discussed in terms of flares, coronal enhancements, plage structure and geomagnetic response.

Severny (1969) used a technique which enables the Sun to be observed like a magnetic star. Because of rotation, different sides of the Sun are visible at different times. A curve of variable magnetic field is obtained. During nearly three periods of rotation the Sun appears to be a quadrupole magnetic rotator. Generally, the integrated field of sunspots seems to be opposite to the field of the Sun as a whole. The curve seemed to remain similar throughout the year 1968.

The magnetic field is effectively frozen into the ionized gas in the Sun and it is therefore necessary first to describe the motion in the convective zone. Large scale motion in giant cells is strongly affected by Coriolis forces, giving a radial shear in the angular velocity, while the interaction of convection and rotation leads to the equatorial acceleration. Many hydromagnetic dynamo mechanisms have been proposed in the last few years. In particular, meridional fields can be generated from azimuthal fields owing to a preferred sense of helicity in the motion Such regeneration is included in Leighton's phenomenological model, which reproduces many features of the solar cycle. More detailed models will have to treat the concentration of magnetic flux into ropes by individual convection cells.

An outline of the mean-field magnetohydrodynamics suggested and developed by M. Steenbeck and the authors and its application to the dynamo theory of the solar cycle is presented. Four basic requirements are formulated which have to be satisfied by any dynamo model which claims to explain the solar cycle. The models investigated allow conclusions about the differential rotation. In this connection Leighton's work is criticized.

The expansion of the magnetic field in a series of force-free constituents for the internal field and that of ‘semi’ force-free constituents for the external field results in the skew symmetry of the coupling tensor of a magnetohydrodynamic dynamo. The periodic solution which exists in this case is deduced by the method of eigenvalues. Based on the numerical results a dynamo model is proposed to draw up the characteristic features of the magnetographic observations carried along the central meridian and at the polar areas.