Corona

CORONA. During the period of totality of an eclipse of the sun there is seen surrounding it a halo, pearly-white in colour, of irregular shape, but roughly circular in form, about twice the size of the sun, with faint streamers extending several diameters from the limb of the sun. This is termed the Solar Corona. It was long uncertain whether the corona belonged to the sun or to the moon, or if it arose in the earth's atmosphere. It cannot originate in the atmosphere since it presents the same appearance when seen from widely separated positions on the earth's surface, and photographs show that the dark moon moves across the back ground formed by the corona, so that it cannot belong to the moon. It will be obvious that many of the phenomena that will now be described are inconsistent with any but a solar origin.

Phenomena.

The plate gives a good idea of typical coronae. There is an approximately circular, and not very large, bright inner corona around which are more or less extensive streamers proceeding in a roughly radial direction mainly from the neigh bourhood of the solar equator; and there are curved rays (the "polar brushes") proceeding from the north and south poles of the sun. The forms observed can be classified into two main types occurring alternately, at regular intervals, with intermediate forms between. At times of minimum sun-spot activity the polar brushes are short, and the equatorial streamers long; at times of maximum sun-spot activity the polar rays are longer, the equatorial streamers shorter and the corona is more circular in form. Near prominences the corona is brighter and shows an arched structure. It is uncertain how long the different forms endure, since observations are possible only during a total eclipse; thus there was a sun-spot maximum in Aug. 1917, but the corona of June 1918 was not a typical maximum type. In general after a maximum the streamers draw away from the poles and the longest rays are found in the sun-spot zones, making the corona rectangular in appearance. The extent of the corona varies greatly. Streamers have been followed out to the enormous dis tance of twelve solar diameters, i.e., over 1 o,000,000m. from the surface of the sun, but commonly three or four diameters is its extent.

Coronae near sun-spot maximum are found to differ more than those occurring near a sun-spot minimum. It is definitely estab lished that changes occur and that the broad features may be correlated with sun-spot activity. It remains at present undecided whether any changes occur during the short periods of time for which the corona can be observed at any one eclipse. Certainly no changes have been found during the time of totality at any one station, which is necessarily less than about seven and a half minutes. The interval of time between totality at different points of the earth's surface lying within the zone of totality may, how ever, be several hours, and it is of interest to compare observa tions of the corona made at widely separated stations. The evi dence is on the whole in favour of changes taking place. In 1905 plates taken in Egypt and in Spain with an interval of 7omin. showed no movement, but in 1918 stations 26min. apart showed changes in the coronal arches, indicating that they were receding from the sun with a velocity of about i6km. per sec., and Prof. Horn d'Arturo, comparing photographs of the corona of Jan. 14, 1926, taken in East Africa and in Sumatra, with an interval of I5omin., finds considerable changes and velocities of recession of the coronal material of the order of 'km. per sec. Of course, rapid motions along the lines of the streamers might exist and escape detection.

The Radiation from the Corona.

The greater part of the light from the corona shows a continuous spectrum; a number of bright emission lines are present, and also the Fraunhofer spectrum (the ordinary solar spectrum). The spectrum of the inner corona (from 8 to iomin. of arc from the limb of the sun, the diameter of which is about 3omin.) is mainly continuous and the bright lines are strongest here; farther out, in the middle and outer corona, the Fraunhofer lines are visible. It is important to know if, and how, the radiation from the corona is polarized. It is found that the bright-line spectrum is unpolarized, but the continuous radiation shows strong radial polarization. The polar ization rises rather rapidly to a maximum of about I 1% in the visible region of the spectrum, at 5 or 6min. from the limb, and thereafter decreases more slowly. Unfortunately little or nothing is known about the way in which the polarization varies over the spectrum. H. F. Newall found that there came from the veil of illuminated sky between the observer and the corona as much polarized light as from the corona itself at 3-4° from its centre.

The Distribution of Intensity. The older measurements are quite inconsistent. More recent measurements agree more or less with the law found by Bergstrand in 1914. He found that the in tensities near the solar equator differ greatly from those near the poles at the same distance from the centre of the sun. The equa torial rays are three times as intense as the polar rays. He sug gested that we regard the corona as the superposition of an interior corona existing only in the equatorial zone, and a general outer corona. In both the intensity decreases as the inverse square of the distance from the edge of the sun, but the intensity of the equatorial corona is double that of the outer corona. However, on account of the complex detailed structure of the corona, any attempt to derive a law of intensity distribution of the kind just examined, which ignores variation round the sun and treats the corona as if it were a disc whose brightness varies only along a radius, is foredoomed to failure if regarded as anything but a rough approximation.

The Total Radiation from the Corona. In view of the great un certainty of the older measurements and the paucity of modern ones, caution must be used in giving any figures for the total radiation, but it seems fairly clear that the total energy radiated is from about a quarter to a half that received by us from the full moon, and that it is not constant but varies from one eclipse to another, the measures indicating, for example, that it was 40% greater in 1926 than in 1925.

The Continuous Spectrum of the Corona.

The eclipse of 1922 showed that the Fraunhofer spectrum of the corona does not come from terrestrial scattering of direct sunlight, since no trace of it was found beyond the limits of the coronal spectrum on the plates. It is found that the continuous spectrum and the Fraunhofer lines are strong, and the coronal green emission lines are weak, near a time of sun-spot minimum and this, on the whole, indicated that the coronal radiation is richer in red and infra-red rays than is direct sunlight, but the amount of the difference is small and uncertain. There seems to be no appreciable radiation from the corona in the infra-red region beyond about 55,000 A.U.

The Bright-line Spectrum of the Corona.

A number of bright lines are found in the spectrum of the corona, which are definitely not due to the higher regions of the chromosphere. None of them is very bright, and the older measures are very inac curate. The table gives a summary of the most accurate meas ures available of lines definitely believed to belong to the corona. None of them is identified with any known line in the spectrum of an element observed in the laboratory, and when they were first discovered the name "Coronium" was pro posed for the hypothetical element to which they were supposed to owe their origin. Their behaviour, however, seems hard to reconcile with the idea that they have a common origin. Thus the structure of the monochromatic images of the inner corona obtained by use of a prismatic camera indicates that the lines can be grouped in pairs, such as 3454 and 3643 and 3601 and 4086.

Of these lines 4231 and 4586 are near strong lines due to ionized iron, and 5536 is near a strong iron line and near a strong line due to ionized strontium. These lines are, however, possibly coronal in origin. Three lines in the far ultra-violet (3164, 317o and have been observed by Deslandres once only, and are probably of chromospheric origin due to TiII. A few other very doubtful lines are 3461, 4398, 4533 and 4567. See C. R. Davidson and F. J. M. Stratton, Mem. Roy. Ast. Soc., 64, 142 (1927).

The distribution of intensity of the lines across the corona groups these same pairs together, and also 3388 and 3987; the line appears to belong to the 3601 group and 6374 resembles and 3987 but not very closely.

A remarkable relation between the wave-lengths of certain coronal lines discovered in 1911 by J. W. Nicholson (resembling the relation between nebular lines also discovered by him), and accounted for in terms of a dynamical theory of a hypothetical atom, seems not to have stood the test of time. He found that most of the lines then classed as coronal lines could be fitted to a formula of the typeX where n is an integer. Increas ing accuracy of measurement has not improved the fit of this formula, moreover some of the lines used by him are of very doubtful coronal origin or even existence, and some have been identified as chromospheric lines due to ionized iron or titanium.

Rotation of the Corona.

It is generally assumed that the corona close to the sun rotates with it, but there is no definite evi dence of any motions in it that could be interpreted as rotational. We do not know independently the wave-lengths of any of the coronal lines, so we cannot say from the observed wave-lengths if there is any displacement that could be interpreted as due to motion toward or away from the observer in the line of sight. Attempts have been made to discover rotational effects by com paring the wave-lengths of a coronal line at two ends of a solar diameter. They have been unsuccessful on account of the great weakness of the lines at any appreciable distance from the sun's limb. It has been observed, though confirmation of this is desirable, that the Fraunhofer lines in the coronal spectrum are slightly displaced. This might indicate rotational movement but might, on the other hand, be due to the reflection of sunlight from particles moving radially from the surface of the sun, as sug gested by H. N. Russell.

Theories of the Corona.

The presence of the Fraunhofer lines in the spectrum of the corona suggests strongly that there is an atmosphere of dust particles or a gaseous envelope scattering the solar radiation, as does also the polarization of the coronal light. The density of the coronal material must be very low since there is no appreciable refraction produced by it, nor do comets, which sometimes almost graze the solar surface, suffer any de tectable perturbations in passing through the corona. One of the oldest theories of the corona is that it consists of a swarm of meteoric particles revolving round the sun. This theory has the disadvantage of explaining none of the observations, and was probably only put forward because this was the only known means, at the time, of maintaining any kind of atmosphere against the force of gravitation. Various so-called mechanical theories have been put forward from time to time, based on the original one of Schaeberle. According to these the corona is due to light emitted and reflected from streams of matter, which are ejected from the sun by forces acting along lines roughly normal to the solar sur face and localized chiefly in sun-spot zones (to account for the correlation with sun-spot activity). The main difficulty in these theories has been that of providing a suitable expelling agency; moreover the explanation of the continuous spectrum is not thereby completed, and the problem of the bright line spectrum is left quite unsolved.

Probably the view to take at the present time is something like this. We have a satisfactory theory of the chromosphere, whereby it turns out to be a gaseous atmosphere consisting largely of elec trons and ionized atoms supported by radiation pressure (the competent force lacking in older theories). At the top of the chromosphere the density is very low, pressure is of the order of 10-12 atmospheres, and since we can scarcely suppose the density to increase outwards from the sun, there is no difficulty in fitting on to it a tenuous coronal envelope. The mechanism of the radia tion pressure supporting the atoms is that of line absorption, which is very different from that of pressure on small particles. In par ticular it is selective in its action. Only those atoms are capable of existing in such an atmosphere that are ionized to the appro priate degree under the existing conditions of temperature and pressure, so that they can absorb the requisite amount of radia tion from the supply provided, which last is determined by the surface temperature of the sun. This is quite welcome, however, for since we have to account for the coronal lines we must sup pose them due to some atoms, and since there are not many of them only a few atoms should be concerned in their production. The coronal region must consist of gases in a rather higher degree of ionization than in the chromosphere, since the pressure is so much lower, and though there may not be possible a static equi librium for such an atmosphere, yet, owing to recombination of ions (allowing suitably powerful absorption to occur from time to time) there may be a dynamical equilibrium. We do not re quire to suppose the corona quiescent. If we regard coronal ma terial as shot up from below, owing to excessive radiation pres sure, we have a good reason for the observed connection be tween coronal arches and streamers and prominences, since, near prominences and sun-spots, locally superheated regions of the solar surface are found to occur ; moreover the connection with sun-spot activity is no longer mysterious. There will be present a considerable proportion of free electrons ; these can scatter light and, it turns out, would do so in the right kind of way, so far as observation is possible ; and the scattering by atoms present is not inconsistent with observation.

Finally with regard to the origin of the coronal lines, it seems most likely that they are to be explained in the same way as the nebular lines, i.e., as due to known atoms in circumstances which cannot so far be imitated in the laboratory ; so that transitions can occur sufficiently often under the conditions of low density (i.e., long undisturbed life of the individual atoms) and appro priate stimulation, so as to produce "forbidden" lines of appre ciable intensity and to suppress the ordinary lines. The atoms involved would presumably be those capable of existing (possibly at a lower stage of ionization) in the high chromosphere, and in deed doubly ionized calcium has been suggested as the source of these lines, with some plausibility, but so far it is not more than a suggestion; ionized titanium seems also very likely to be involved.

BIBLIOGRAPHY.-S. A. Mitchell,

Eclipses of the Sun (New York, Bibliography.-S. A. Mitchell, Eclipses of the Sun (New York, 1923) ; F. J. M. Stratton, Astronomical Physics (1924) ; Eberhard, Kohlschiitter and Ludendorf, Handbuch der Astrophysik, vol. iv.; Memoirs and Monthly Notices of the Royal Astronomical Society, London; The Astrophysical Journal (Chicago University Press) ; Lick Observatory Bulletin (1918) . (J. A. CA.)