SummaryVelocities of the sources of type II bursts are derived from rates of frequency drift using standard density models, both statistically for 21 bursts, and individually for 5 bursts extending over wide frequency ranges. The derived velocities exceed the speed of sound in the magnetic-field.free corona: on the average the velocity decreases with increasing height to a minimum of ,....., 750 kmjs at a little below I Ro' and j,hereafter slowly increases with height. The nature of the type II source is discussed in relation to these velocities, and also in relation to detailed measurements of harmonic ratios and band splitting for the five individual bursts. It is suggested that the type II source is either a strong parallel shock (direction of propagation of sho~k parallel to magnetic field) or a perpendicular shock. Magnetic field strengths of 2-20 G at 0·5 R 0 above the photosphere, decreasing to 1-10 G at 2 Ro, are derived. Finally, it is shown that theories by which fundamental emission arises in front of the shock, whilst harmonic emission originates in the interior of the shock, are untenable.
I. INTRODU.OTIONAlthough the solar radio burst of spectral type II has been recognized as a distinct type for nearly 15 years, the precise nature of the exciting disturbance is still unknown. Considerable attention has been given in recent years to the hypothesis that the disturbance is a shock front generated by a flare explosion and propagating along a coronal streamer, and since the corona is pervaded by magnetic fields, such a shock may well be magnetohydrodynamic (Westfold 1957;Uchida 1960;Tidman 1962). Also a magnetic origin for one of the characteristic structural features of type II bursts, namely band splitting, has been explored (Roberts 1959; Sturrock 1961). The question of the validity of these magnetic theories for the type II source is important since, if they prove to be well founded, observations of type II bursts will enable us to estimate magnetic field strengths high in the corona, in regions hitherto inaccessible to optical observation.In the present paper we attempt to infer the nature of the type II disturbance and to estimate coronal magnetic field strengths. For this purpose we shall discuss the velocities of the sources (deduced from frequency drift rates using standard coronal density models) and also harmonic ratios and band splitting. Although much of the discussion is statistical, the conclusions are checked by appeal to several outstanding individual bursts covering wide frequency ranges. The bursts used were observed between 1952 and 1963 with the Dapto dynamic spectrograph. Because of the very low frequency limit of the Dapto spectrograph (now 5 Mc/s) we have been able to extend the observational domain to considerable heights in the corona, into