The intensity and polarization of solar radio bursts at 9400, 3750, 2000, and 1000 Mc/s have been continuously observed since 1957 July. At 3750 Mc/s, the linearly polarized components were also measured during 4 months from 1957 October till 1958 January [1]. The analysis of observations to 1958 May reveals the following points:

1. Many bursts at these frequencies show a difference in the intensity of two circularly polarized components. Frequency distributions of maximum degree of polarization of bursts are shown in Fig. 1, where weak bursts are excluded to reduce errors. At 3750, 2000, and 1000 Mc/s, particularly at 3750 Mc/s, the bursts of small degree of polarization are predominant. Circularly polarized bursts are occasionally observed at 2000 and 1000 Mc/s, but at higher frequencies the degree of polarization is usually less than 50 per cent.

2. At 3750 Mc/s, no linearly polarized components could be detected in 117 bursts for a bandwidth of 10 Mc/s. The burst at this frequency is usually composed of a small degree of circularly polarized component and a randomly polarized one.

3. Sometimes the sense of polarization varies with frequency. An example is shown in Fig. 2. On this occasion, the degree of polarization is small at 3750 Mc/s and the sense of polarization reverses between 2000 and 3750 Mc/s. Roughly grouped in Table I are 129 bursts simultaneously observed at four frequencies. The reversal of sense was found in 64 per cent of such bursts. Sometimes the sense of polarization varies with time, as may be found in our monthly report. Some of such phenomena may be interpreted by the motion of the source from the preceding part of a bipolar spot group to the following one or vice versa [4].

4. In Fig. 3, the bursts associated with flares are classified in three types as regards the sense of polarization, and they are allocated to each of four quadrants on the solar disk according to the position of the flares. At higher frequencies, the sense of polarization seems to correlate with the position of the source as in the case of S component [2, 3]. A precise distribution map at 9400 Mc/s is shown in Fig. 4.

Bursts of solar radio emission which coincide in time with solar flares are occasionally observed over the whole frequency range up to 35,000 Mc/s. The purpose of this paper is to describe briefly some bursts observed at Pulkovo at wavelength λ = 3.2 cm.

Le rayonnement solaire sur ondes décimétriques et centimétriques est dû non seulement au rayonnement thermique de l'atmosphère solaire, mais aussi à un rayonnement dont la nature exacte est encore inconnue, et qui provient de régions émissives localisées sur le disque solaire. Le rayonnement global dans ce domaine de longueur d'onde augmente avec l'apparition de ces régions émissives sur le disque solaire et diminue avec leur disparition.

Le 20 Mars 1958, L. Peterson et J. R. Winckler [1] ont observé une brève augmentation du rayonnement cosmique d'une dizaine de secondes de durée; au même instant une éruption chromosphérique se développait sur le soleil et un sursaut radio assez exceptionnel était enregistré sur ondes centimétriques et décimétriques.

L'interféromètre a ses lobes principaux sensiblement parallèles au plan méridien [1]. Ils ont une largeur de 3′.8 et sont distants les uns des autres de 2 degrés. La détermination de l'heure et de la durée du passage des centres R dans les lobes permet de déterminer leur distance à un plan nord-sud passant par le centre du disque solaire et leur diamètre apparent dans l'hypothèse d'une symétrie circulaire et d'une luminance uniforme. Les différentes causes d'erreur introduisent une indétermination de l'ordre d'une minute d'arc au plus sur les positions des centres R.

The solar work at the radio observatory near Dwingeloo concentrates on the investigation of relatively small bursts, which, having the aspect of elementary phenomena, could give some insight into the way in which they are related to their immediate cause.

In 1957 high-speed records were obtained at a frequency of 200 Mc/s during solar radio noise storms at the Solar Observatory at Harestua. For the registrations a receiver of bandwidth 0.3 Mc/s and time constant 0.01 seconds was used in connection with a Brush recorder. Most records were secured at a paper speed of 25 mm/second.

It is known that the regions responsible for solar noise storms are associated with visible sunspot groups and that the diameter of such radio sources is not less than 7 minutes of arc at a wavelength of 3.7 m. Little, however, is known of the size and location of the individual short-lived bursts (duration ~1 second). Two interferometric systems have been set up at 3.7 m to measure these quantities, and some preliminary results have been obtained:

(a) The diameters of single bursts. To measure the diameters of individual bursts an interferometer of spacing 270λ was connected to three receivers. Receivers 2 and 3 gave records corresponding to total power interferometers in antiphase, while receiver 1 was arranged as a phase-switching interferometer in quadrature with the other two. From the relative sizes of an individual burst on the three records it is possible to determine the burst intensity, its angular size and position in one coordinate. Measurements carried out during a noise storm on 1958 March 21 gave an angular diameter of 6 ± 1 minutes of arc for about 25 individual bursts.

(b) The location of single bursts in two dimensions. In order to determine the instantaneous position of individual bursts, two phase-switching interferometers of spacing 135λ were aligned in E-W and N-S directions. A third aerial was used with a total power receiver to measure the burst intensity. Preliminary results indicate that the positions of the individual bursts during a noise storm show a scatter of less than 3 minutes of arc.

Detecting a weak linearly polarized component in an isolated solar burst is very difficult. It is a unique event and it is not possible to make a series of observations to reduce the error. Interference and reflections confuse the result and it is extremely difficult to determine the equipment parameters precisely, especially those pertaining to the antenna.

1. Solar radio emission is usually characterized only by the intensity, i.e. by a single value at a given frequency.

In polarization measurements the radio emission is characterized by four values (for example, by the unpolarized-component intensity and three components of the polarization ellipse).

Here we see that observations of solar radio emission with a polarimeter give us considerably more extensive information than the usual observations give. It is a very important fact that the polarization of the radio emission is determined by the strength and direction of the magnetic field of emitting regions. Thus, we connect the polarized radio emission with a magnetic field that is noticeably higher than the field in the region which we can study by optical methods.

It should also be noticed that one of the most interesting unsolved problems is establishing the nature of the disturbed solar radio emission. It is clear that new information on the polarization of radio emission will contribute to the solution of this problem; but in spite of its importance and urgency, investigations to solve it are being carried on only in a small number of observatories. The observations in Australia and the work in Japan may be mentioned.

On 1957 November 4, an unprecedented solar radio event was observed at meter wavelengths. Observations at 169 Mc/s were obtained by the Nançay Observatory (great interferometer and simple radio telescope) and by the Humain Observatory. At 200 Mc/s the event was observed by the Nera Observatory.

The angular diameter of the Crab nebula is considerably increased when it is viewed through the solar corona. This effect is due to the scattering of the waves by coronal irregularities and its study gives some indication of the nature of the outermost corona. This paper is a summary of observations carried out each year at Cambridge for the period 1952 to 1958 inclusive.

On peut se demander si la grande extension de la couronne solaire demandée par l'hypothèse de Chapman [1] pourrait se manifester par les phénomènes de réfraction lors d'une occultation, par exemple celle de la source du Taureau.

1. This paper reports on the results of the investigation of the outermost regions of the solar corona (which for brevity I shall call the supercorona) that have been obtained during recent years.

Seven years ago, in 1951, a new method of observation was published in Doklady of the Academy of Sciences of the U.S.S.R. The idea of the method, which at the present time is well known, is that the radio emission of the Crab nebula is received at the time when the source is covered by the solar corona. This original eclipse takes place yearly in the middle of June.

By receiving the radio waves that pass through the solar corona regions, we can estimate the effects of attenuation, scattering and refraction and reach some conclusions about the structure of the supercorona.

En 1957 et 1958 des observations de la nébuleuse du Crabe lors de son passage derrière la couronne solaire ont été effectuées sur 169 Mc/s à l'aide de l'interféromètre méridien à 32 antennes de Nançay [1, 2]. Cet appareil ayant un pouvoir séparateur de 3′.8 dans la direction est-ouest et un lobe principal tous les deux degrés, nous permet de distinguer le rayonnement de la radiosource même quand elle est très proche du soleil. Le flux maximum reçu lors du passage de la source non élargie atteignant 10 pour cent du flux donné par le soleil calme, la source sera très facile à distinguer sur les enregistrements.

During this part of the Symposium, so carefully prepared by Messrs. Denisse and Pawsey, a great amount of new data has been presented, almost all obtained since the last international meeting on these subjects only one year ago. It is clear that this is a part of astronomy that develops impetuously. Hypothetical theories have been definitely confirmed by direct observation; on the other hand, new facts have been disclosed, increasing the complication of the phenomena; but we have reason to hope that we are in a transitory stage and that the complications will gradually be explained on clear and simple principles. It is of course impossible to mention in this summary the contents of all communications; only the more essential results will be reviewed.

The investigations into cosmic radio noise that I made with J. W. Phillips and S. J. Parsons arose from operational studies of the performance of radar equipment at about 5-m wavelength in 1944–45. To obtain greater detection ranges, preamplifier stages of low-noise factor were being tested, but they did not result in the expected improvement in radar performance. J. M. C. Scott, a theoretical physicist now at Cambridge, suggested to me that the limitation in effective noise factor might be attributable to cosmic noise, about which something was already known from the work of Jansky and Reber.

I have already reported on the spectra of 31 radio sources [1], but since then, new observations have become available, and, in addition, I have corrected my own 38-Mc/s measurements for a nonlinearity in the response of my receiver. The qualitative picture remains the same, but from the more recent information I have found values for the spectral indices of 85 sources, and have revised some of those given previously. From the larger sample of sources now available, it is possible to suggest some relationships of sufficient interest to warrant further investigation.

In October and November 1957 at the Physical Institute of the Academy of Sciences of the U.S.S.R. a search was conducted to detect the plane polarization of the radiation from the Crab nebula at 9.6-cm wavelength [1].

The brightness distribution across the Cassiopeia A source in position angle 90 degrees consists of a primary region of emission 4.1 minutes of arc in width, with a much fainter extension offset from the main region of emission and having a brightness of only 10 per cent of that of the main component. Measurements were made up to the third maximum of the transform (2160 λ), and the position of the first zero was determined at 840 λ.