As a result of research carried out with rocket-borne grating spectrographs, the nature of the extreme ultraviolet spectrum of the Sun is now known to a short wavelength limit of 33.7 Å, the Lyman-alpha line of C VI. Most of the emission lines of wavelengths greater than 400 Å have been identified, as have those from 80 Å to 33.7 Å. Between 149 Å and 400 Å, however there are many intense emission lines whose identity has not as yet been established. Twenty or more have been proved to be from iron, since they appear in spectra obtained from high temperature plasmas into which iron has been introduced, but the stages of ionization have not yet been established. Lines from the elements most abundant in the Sun, H, He, O, N, O, Ne, Mg, Al, Si, S and Fe, in most of the stages of ionization requiring 500 eV or less for production have been found. The outstanding exceptions are the lines in the fluorine and neon sequences.

Spectroheliograms, photographed with normal incidence spectrographs, show that the emission lines Fe XV 284 Å, Fe XVI 335, 361 Å, originate principally from active regions, in contrast to He II 304 Å, which is emitted with great intensity from the disc also. Continuum emission, in the wavelength range 170–300 Å, has been recorded from intense centers of activity.

Expected relative prominence of elements and ionization stages is estimated from existing recordings of the solar XUV spectrum. The following 3 groups of elements, apart from H and He, need be considered : (C, N, O), (Ne, Mg, Si, Fe), (Na, Al, S, Ar, Ca, Ni), the relative abundance descending by approximately an order of magnitude from one group to the next, and all other elements being at least another order of magnitude less abundant. Lines of low and medium stages of ionization are expected only from most abundant elements. The observed presence in the solar spectrum of lines from the neon-like Fe XVII suggests that very high stages of ionization must not be a priori excluded.

These considerations indicate that a discussion of laboratory data may be confined to atomic systems containing from 1 to 18 electrons. Observed data on these 18 isoelectronic sequences, and the possibilities of their extrapolation, are reviewed. It is concluded that required data for the first 10 sequences are well in hand, but considerable laboratory work remains to be done on the third-period sequences.

A comprehensive paper is being prepared for later publication.

The Scylla plasma which is contained in a ceramic discharge tube has been purposely contaminated with known impurities. The spectrum of the discharge with Fe (CO)5 added shows most of the solar lines in the 150–220 Å region. Some of these strong lines may be produced by iron in intermediate stages of ionization (Fe X to Fe XIV).

Experimental problems of absolute intensity measurements of solar electromagnetic radiation in the extreme ultra-violet (EUV) and soft X-ray region of the spectrum (designated cumulatively as “XUV” for brevity) are reviewed. From an evaluation of existing rocket and satellite data together with existing projects and goals of future observations from space vehicles, it is suggested that the numerous associated practical problems be devided into two major areas, namely (a) the general experimental art of heterochromatic absolute XUV spectrophotometry in the laboratory and (b) specific problems characteristic of the objectives of solar physics as well as existing restrictions of space technology. Within the first area (a) emphasis is placed on recent progress toward justified reliance on ionization detectors without necessary connection to source standards. For the second area (b), emphasis is placed on the immediate need to have existing exploratory observations followed by a new phase of more systematic experiments of increased accuracy.

This paper reviews solar extreme ultraviolet observations made in the spectral region λ170 — λ400 by the first Orbiting Solar Observatory during March, April, and May, 1962. Application of the data to the identification of emission lines in this spectral region is discussed. The observations are compared with concurrent 2800 MHz radio emission and other measurements of solar activity.

X-ray bursts are defined as electromagnetic radiation originating from electronic transitions involving the lowest electron shells; gamma rays are of nuclear origin. Solar gamma rays have not yet been discovered.

According to the origin we have :

1. 1. Quasi thermal X-rays, emitted by (a) the quiet corona, (b) the activity centers without flares, and (c) the X-ray flares.

2. 2. Non-thermal X-ray bursts; these are always associated with flares.

The following subdivision is suggested for flare-associated bursts :

In July 1963, the U. S. Naval Research Laboratory flew an Aerobee rocket equipped with a Bragg Crystal Spectrometer in an attempt to record an x-ray spectrum of solar emission. A pointing control was used to keep the device accurately pointed at the Sun. A spectrum of about 12 lines between 12 and 25 Å was recorded, made up mainly of lines from O VII, O VIII, N VII and Fe X VII. The O VII and N VII lines appeared to come from the entire solar disk. The Fe XVII lines originated almost entirely in a single plage region.

Observations from the first Orbiting Solar Observatory have set an upper limit of 3.40 ± 0.95 photons/cm2.s for the 20–100 keV X-ray flux from the “quiet” Sun. Eight impulsive and short-lived 20–100 keV X-ray bursts were observed which were associated with optical flares and cm radio bursts. The 2–8 Å X-ray flux from the “quiet” Sun was observed to be associated with plage groups on the Sun. The intensity for this 2–8 Å X-radiation was found to be quite variable, changes of 5% being observed almost hourly.

Solar X-ray data obtained over the past year by scientists in the United Kingdom are reviewed. These include several new X-ray events reduced from the Ariel satellite results, soft X-ray flux measurements from NRL satellite 1963–21 C and notes on recent rocket studies in this field. A brief discussion is added on the nature of solar flare X-rays and their measurement.

This paper contains the following subjects; a) a description of the telemetry set operating in Arcetri (Florence, Italy) in order to record the measurements of solar X and UV radiation, via satellite:

b) the values of solar flux at 44–60 Å during the period March-April 1964, as deduced by telemetry of NRL 1964–01 D satellite:

c) a typical example of atmospheric absorption at optical and X-ray wavelengths during the sunrise on the satellite:

d) a particular feature of recording when the satellite passes through the sub-auroral belt.

A brief report is given on the launching of the first Sun pointing stabilized Skylark rocket from Woomera, Australia on August 11th 1964.

The conditions of production of the solar X rays are briefly discussed. The comparison of the results of recent calculations and experimental measurements supports the assumption of a thermal origin of the X ray radiation. In the “tail” of the solar spectrum below 15 Å, the main contribution comes from the free-bound radiation of electrons and ions of carbon, nitrogen, oxygen and other elements.

The solar X ray radiation consist of a “quasi-stable” component, produced in undisturbed coronal regions and a “slowly varying” component generated in the active regions of the corona.

This paper reviews our understanding of (a) the earth's ionosphere, and (b) the solar atmosphere, in relation to the recent observations of solar XUV.

The expected ionospheric characteristics are derived as directly as possible from the XUV observations and then compared with the well known E and F layer formation. Agreement does not come very naturally. There are indications of a high ionospheric recombination coefficient decreasing rather rapidly with height. It is found that radiations from the high ions have greater solar cycle variations than radiations from the low ions. This leads to some difficulty in explaining the constancy of the E to F critical frequency ratios. Both UV and X-rays appear to contribute to the E layer.

There are about 250 identified XUV solar emission lines from about 50 ions. The flux measurements for each line gives information on the number of corresponding ions in the solar atmosphere. Each radiation comes from a region with a well specified temperature selected from the whole available range 20000° to 2000000°K. This makes it possible to derive, (a) the amount of solar material in each temperature range, (b) the chemical abundances, and (c) the physical differences between normal solar atmosphere, centres of activity, and flares. The analysis is held up at the present moment until the problem of the dielectronic recombination coefficient is cleared. In due course a good distribution of matter with temperature should be available but the fitting of such results to chromospheric and coronal models provides many problems.

The full importance of dielectronic recombination in plasmas has only recently been realized. The process is discussed in detail for the case of low density plasmas, and formulae for the rate of the process are derived using Correspondence Principle and Detailed Balance arguments. Recent work in the derivation of a simple Gener ral Formula for estimation of dielectronic recombination coefficients and on the ionization balance for Iron and Calcium ions in the corona is reviewed. The new ionisation balance curves are in much better agreement with observations on the Doppler widths of coronal lines.

An important problem in solar physics is the identification of the intense system of spectral lines lying between 170 Å and 220 Å. These lines have also been observed in the high temperature plasma, Zeta, and also in a number of other sources. These sources are reviewed and the experimental evidence relating to the problem of identification is discussed.

On the assumption that the ultraviolet resonance lines are formed by collisional excitation followed by direct radiative transition of a quantum which eventually escapes from the atmosphere, it is shown how the chemical composition may be found. These assumptions are discussed, as is the accuracy of the determinations. The effects of non-radial inhomogeneities are also discussed. Finally the question of the determination of the hydrogen abundance is raised, for which purpose the solar radio observations are used.

The results of this analysis are compared with other solar abundance determinations.

This is a revised version of the work reported to the COSPAR Symposium of 1962. In the first place, the basic components of the radio intensities ranging from 1000 MHz to 9400 MHz derived by H. Tanaka were used to establish the relation between the electron temperature and integrated number of electrons for 1960 and 1961 flights. The result justifies our former assumption that the radio intensities with which the EUV line intensities from the transition layer are to be compared should, as a first approximation, be very close to those at sunspot minimum.

In the second place, the integrated numbers of electrons from the EUV line intensities were derived based on the same data as were used by Pottasch. This is to eliminate any arbitrariness in adopting parameters which are still uncertain.

The result is that we still have a large amount of disagreement between the two sets of intensities in the sense that EUV line intensities are at least ten times stronger as might be inferred from the radio intensities. If one assumes an inhomogeneous model in which the solar disk is bright only in patches covering about 0.15 of its total area, the discrepancy would be eliminated.

Resonance lines of coronal ions of silicon are prominent in the spectral ranges 40–62 Å and 254–356 Å.

An unexpected feature of the soft X-ray spectrum is the weakness or absence of the resonance lines of iron in ionization stages XI through XV.

A second feature is the prominence of lines of the type (3d → 2p) relative to the resonance transitions (3p → 2s) in Li-like and Beryllium-like spectra. It is suggested that the upper levels (3d) are excited by quadrupole collisions from the ground 2s or 2s2 levels.

The intensity of the soft X-ray lines relative to the resonance lines in the 300 Å region seems to be more consistent with temperatures well above one million degrees than with temperatures as low as 700000°K, but the data are not adequate for a precise comparison. The relative intensity of the line emission from the various stages of silicon ionization may be interpreted as indicating that the ionization of silicon peaks in stages IX and X.

The abundances of C, Mg, S, and Al relative to silicon do not seem to be greatly different from the chromospheric abundances reported by Pottasch or with the photospheric abundances.

Collisional excitation of autoionizing levels is found to contribute significantly to the ionization rate for certain ions at temperatures characteristic of the solar corona. Preliminary calculations have been carried out for O IV, O V, O VI, Fe XV and Fe XVI. Autoionization may be of importance for O IV and increases by approximately a factor of two the ionization rate of Fe XV and XVI.

The ultra-violet flux radiated by hot stars is found to be smaller than the value deduced from theory. The possible origins of this discrepancy are discussed : interstellar absorption, deviation from local thermodynamic equilibrium, molecular absorption, blanketting effect. Observations of planetary nebulae provide data about UV radiation from very hot stars.