The rotation of sunspot penumbrae has been investigated on the longitudinal magnetic and velocity fields, observed in the photospheric line Fe I λ5253Å of five lone sunspots. We reconstructed the entire vectors of both fields from their line-of-sight components. All three components of both vectors revealed that the rotation of the sunspots was, in fact, a torsional oscillation. All components of each sunspot had the same rotational period. The penumbrae oscillation periods were distributed in the range from 3.4 days to 7.7 days. The phase of the velocity azimuthal component oscillation was ahead of the phases of all other components of both vectors. If the penumbra plasma density had been equal to the photospheric plasma density (10 −7 g cm −3 ) then the oscillation magnetic energy of the components exceeded their kinetic energy approximately by a factor of 10-200. The obtained results led to the conclusion that these oscillations were constrained.

The discrepancies in the values of longitudinal magnetic field obtained from magnetographic records in different spectral lines are considered. On the basis of extensive data including 60 pairs of magnetographic maps for 11 spectral lines, obtained simultaneously for one of these lines and 26103 with the aid of the Crimean double channel magnetograph, the following conclusions have been reached. The relative field strength 6(20 = Hil(2i)/Hil (6103) depends partly on the distance from the center of the disk ( Figure 4) and mainly on the magnetic sensitivity of the line g22 (Figure 3), pointing to the primary role of saturation effect. The possible influence of line asymmetry on these discrepancies is also suggested.The difficulties arising for the quantitative comparison of magnetographic response with the strength of a magnetic field were pointed out by Severny (1967). A new reason for a critical analysis of the magnetographic measurements of magnetic fields comes out of , who revealed (outside sunspots) that the strengths of longitudinal field Hal in the line Fe/I 5250 A are systematically twice as low as in the line Fe/I 5233/~. In accordance with the results of Sheeley (1967), this striking discrepancy was ascribed to the higher temperature sensitivity of the line 5250/~, and thus in 'magnetic knots', which are supposed to be hotter than the surroundings, there should be a local weakening of the spectral line with a small excitation potential of the lower term (0.12 eV). This statement was supported by Frazier (1970), Wiehr (1970), Howard and Stenflo (1972) all expressing some doubt in the confidence of magnetographic measurements in the line 5250 A and in other photospheric lines with small excitation potential because the use of such lines is thought to lead to an underestimate of the true field.In fact, several authors, e.g. Tsap (1969) andFrazier (1970), have established a definite correlation between the field strengths and the central residual intensities. Therefore with accurate measurements of magnetic field, the variations of the line profile in magnetic regions should be explained. The best example in this respect is presented by measures in the line K z where one can get not only a false value of the field strength but even a false polarity (Severny, 1966).Nevertheless the tentative explanation suggested by Harvey and Livingston (1969) does not seem to be credible from several points of view. First, the profile of 5250/~ line, like the whole effect, does not show the dependence on the value of Hll. It is hard to reconcile this fact with the statement about the temperature as a decisive factor, as if that were the case we should expect different values of temperature to correspond to different degrees of line weakening, i.e. to various values of the ratio 5 = Hll (5250)/Hrl (5233), whereas the observations indicate the value of 6 independent of the Solar Physics 31 (1973) 307-316. All Rights Reserved

No abstract

We investigated the structure of magnetic field and vertical electric currents in two active regions through a comparison of the observed transverse field with the potential field, which was computed according to Neumann boundary-value problem for the Laplace equation using the observed H:-value. Electric currents were calculated from the observations of the transverse magnetic field.There exist two systems of vertical electric currents in active regions: a system of local currents and a global one. The global current is about 2 x I0 ~2 A. In the leading part of the active regions it is directed upward, and in the tail downward.Flare activity is closely connected with the value and direction of both local and global currents: the flares tend to apear in places with upward currents. The luminosity of H~. flocculi is also connected with vertical electric currents; the brighter the flocculi, the more frequently they appear in places of upward electric currents.The sensitivity of H7 emission to the sign of the current suggests that charged particles accelerated in the upper parts of magnetic loops may be responsible for these formations. Joule heating might be important for flocculi, if plasma conductivity is about 5 x 108 c.g.s.e.A model of a flare is suggested based on current redistribution in a system of emerging loops due to changes of loop inductance.

No abstract