Excellent spectrograms in the region from 1780 to 1950 ängströms were obtained from the flight of a high resolution spectrograph in an Aerobee 150 rocket at White Sands Missile Range on June 14, 1971. The instrument’s dispersing system, consisting of a 79 line mm-1 echelle crossed with a 1200 line mm-1 grating, was incorporated in a Czerny-Turner arrangement using spherical mirrors of one meter focal length. Spectral resolution was limited to 13 mÅ by a slit width of 25 μ. The plate factor was 0.5 Å mm-1. Spectra were recorded on Kodak type 101-01 film.

The extreme ultraviolet (EUV) solar spectrum is considered in this review to cover a decade in wavelength from about 300 Å to about 3000 Å. The lower end is close to the practical limit of normal-incidence optics, and the upper end is the approximate limit of visibility from Earth’s surface. The solar plasma that gives rise to emission within this interval is very complex and covers a huge range of physical conditions. Temperatures range from about 4 × 103 K (the temperature minimum in the low chromosphere, observed in the 1600 Å continuum) to about 4 × 106 K. (corresponding to emission of the FeXVI doublet at 335 and 361 Å). The density of the emitting plasma ranges from 1015 cm-3 in the upper photosphere to 108 cm-3 in the quiet corona. Major types of energy transport and deposition within the plasma include not only radiation, but also acoustic waves, magnetohydrodynamic waves, thermal conduction, and convective flow. Magnetic energy often completely controls detailed structure and energy balances within the plasma. Inhomogeneities are not a small perturbation on the overall structure of the plasma, but rather may completely dominate that structure. Extreme departures from local thermodynamic equilibrium (LTE) are the rule rather than the exception. (It is interesting to contrast this complex situation with that seen in the next wavelength decade, from 3000 Å to 3 μ, which includes the visible spectrum. With the exception of a few strong chromospheric lines, this radiation emerges from a comparatively isothermal (5000 K<T<6000 K), horizontally homogeneous, atmosphere in hydrostatic and radiative equilibrium, in which LTE is the rule rather than the exception.)

The Harvard College Observatory UV spectroheliometer aboard the OSO-4 spacecraft has obtained the enhancement of active regions, i.e. the ratio of their intensity to that of the quiet Sun Ia and Iq, for several lines formed in the transition region between the solar chromosphere and the corona (The subscripts a an q indicate active and quiet regions, respectively.)

Observations of the solar spectrum have been made between 1200–2200 Å with high spectral resolution. The results were obtained with an all-reflecting echelle spectrograph carried by a stabilized Skylark rocket launched in April 1970. Measurements of the profiles of a number of emission lines due to SiII, CII, SiIII and CIV formed in the temperature range 104–105 K, indicate ion energies which are considerably in excess of the electron temperatures derived from the ionization balance. Since the ion/electron relaxation time is very short the observed ion energies cannot correspond to an ion temperature and hence a non-thermal mechanical energy component exists in the transition zone.

The problem of energy calibration of astrophysical apparatus is essential for every or almost every space experiment. The utilization of synchrotron radiation from an electronic accelerator should perhaps be taken as an ideal solution of this problem, if of course, such a possibility is available.

Special equipment for the extraction of synchrotron radiation has been devised at the circular electron accelerator with a maximum electron energy of 6 GeV, in the Physics Institute of Erevan (Gurzadyan and Ohanesyan, 1972). The equipment is designed primarily for the energy calibration of astronomical apparatus operating in a vacuum and hard ultraviolet and X rays. However, the equipment can also be applied to a wide range of experiments relating to the physics of solids, crystallography, physics of thin films, X rays, etc.

During the flight of the spaceship ‘Salyut’, manned by the spacemen G. T. Dobrovolsky, V. N. Volkov, and V. I. Patsaev, nine spectrograms of α Lyra and six spectrograms of β Cen were obtained with the help of the ‘Orion’ system, in the range of wavelengths 2000–3800 Å and with a spectral resolution of about 5 Å near 2600 Å.

The ‘Orion’ system is a telescope of the Mersenn system, with a large mirror of 280 mm diameter, and a slitless spectrograph of the Wadsworth system with a diffraction grating (Figure 1). The dispersion is 32 Å mm-1. The spectrograms were recorded on a perforated photographic film, 16 mm wide, covered with emulsion of the UFSH-4 type.

The solar atmosphericmodel calculations incorporating absorptionby hydrogen and metals and Rayleigh scattering by hydrogen give larger specific intensity in the solar ultraviolet spectrum than that observed. This discrepancy has been interpreted as due to some unknown source(s) of opacity in this spectral region. We have, therefore, compared the absorption coefficient due to bands of nine diatomic molecules with that of metals and hydrogen in the ultraviolet spectral region at temperatures and pressures appropriate to the solar photosphere. The molecular band absorptions due to A-X bands of CO, SiO and OH and to B-X band of CH have been found to be larger than or equal to the total absorption by metals and hydrogen below 4000 Å and may be able to account for the missing opacity except possibly between 2500 Å and 3000 Å. Whether absorption due to some other molecular band or bands is responsible for the opacity in this wavelength interval is not certain.

Satellite lines, situated on the long wavelength side of the helium-like ion resonance line, can be observed in highly-ionized ions both in laboratory sources and from the Sun. Although seen for more than 30 years, these lines have only recently been classified in detail as inner-shell transitions in lithium-like ions. Laboratory experiments have shown that under steady-state conditions these satellites are produced by dielectronic recombination, although in transient ionizing plasmas direct inner-shell excitation can be important. Detailed calculations have been carried out for high Z ions up to copper, and the results can be compared with solar flare spectra in iron. Such comparisons allow both the electron temperature and the transient state of the plasma to be determined. Laboratory spectra from such high-Z ions are different in appearance, and may be dominated by processes resulting from the transient ionizing state of the plasma.

A number of previously unclassified multiplets of FeXIV, XIII, XII, and XI produced by transitions of the type 3s23pn–3s3pn+1 are identified in the XUV spectrum of the Sun. The iron lines account for most of the previously unidentified strong lines between 330 and 370 Å. Solar observations of especial value for the investigation of the 300–400 Å region were the slitless spectroheliograms of September 22, 1968 (Purcell and Tousey, 1969) and November 4, 1969 (Tousey, 1971) – on which the image of a flare was recorded.

Other solar identifications in the same spectral region include the resonance lines of NiXVII and NiXVIII, and one 3p–3d multiplet of FeXIII. The solar blend at 417 Å involving the FeXV inter-combination line and SXIV is resolved.

Over two hundred spectral line intensities between 1 and 60 Å from ions in the solar corona have been computed for electron temperatures ranging between 105 and 109 K (Mewe, 1972a). Also the continuum (ff+fb) radiation intensity has been computed. The calculations were set up to interprete soft X-ray measurements aboard the ESRO 2B satellite (IRIS): Wavelength bands 44–55 Å (experiment of Space Research Laboratory, Utrecht) and 1–3, 3–9, 6–18 Å(expt. of University College, London) (Brinkman and Shaw, 1972).

An experiment is in course of preparation at the Astrophysics Research Unit at Culham for flight on a Sun-pointing rocket. It is designed to determine the ionization temperature and electron density as a function of height in the temperature range of about 8 × 104 K to 3 × 106 K by measuring limb to disk intensity ratios of extreme ultraviolet emission lines in the 170 to 850 Å region. The work is an extension of current experiments in which normal-incidence spectrographs are used to determine the structure lower in the chromosphere-corona transition region.

Starting from the formula for line intensities E for a spherically symmetric atmosphere, considering the pressure parameter P0 = NeTe to be constant the inverse temperature gradient can be determined by taking it out from the integral for E. In this way Dupree and Goldberg (1967) found that the temperature gradient varies considerably whereas the conductive flux factor F=Te5/2(dTe/dh) is nearly constant for Te above 105 K. The best fit was obtained by using photospheric abundances. As the inverse temperature gradient varies considerably in the line forming region, it is, however, not satisfactory to take it out from the integral.

The coronal spectrum provides a unique opportunity to probe atomic processes in a high temperature plasma. The corona is optically thin, at low density, and the coronal ions are highly stripped. These characteristics result in an emission line spectrum with a rich multiplet structure, with forbidden lines prominent in the spectrum, and deexcitation of metastable states by two photon decay making significant contributions to the continuum. The multiplet structure is further complicated by the presence of satellite lines from autoionizing states formed during recombination. With few exceptions, the true coronal lines are below 300 Å, and the general necessity for and complexity of grazing incidence optics has made it difficult to obtain high resolution spectroheliograms.

On October 3, 1970 the astrophysical rocket observatory K-2 was launched from the Soviet Union, designed to investigate the Sun in the region of short- and X-ray waves Gurzadyan et al. (1971). During the launching a series of X-ray pictures of the Sun was received, in the period 04 h 22 min to 04 h 30 min universal time, through an aluminum foil 5000 Å thick, covering a polistriol film 1000 Å thick. The pictures were taken by means of a pinhole camera under the following conditions: the focal length was 150 mm, the angular resolution – from 8’ to 0.’92, which corresponds to the dimensions of the input diaphragm from 0.28 mm to 0.04 mm, effective exposure – 520s, the pointing accuracy by a stabilised platform carrying the instruments – 0.′7-0.′8. The picture was taken by a film of the UF-R type (Kalinkina et al, 1965).

The relative intensities of the 2s–2p and 2l–3l′ (l=0, l′= 1; l= 1, l′ = 0, 2) solar XUV lines of ions in the lithium isoelectronic sequence are sensitive to electron temperature because of the large energy separation of the 2p and 3/’ levels. Recent observations of these lines for three members of the sequence, O+5, Ne+7 and Mg+9 (Heroux and Cohen, 1971), consequently provide a potentially powerful means of studying the temperature structure of the upper chromosphere-corona transition region. These observations have been examined in the light of recent calculations of the electron excitation cross-sections for the relevant transitions (Flower, 1971). It is found that the observed intensity ratios are systematically greater than values calculated assuming that all the lines of a given ion are produced at essentially the same temperature, namely, the temperature for which the ionization equilibrium calculations of Jordan (1969) predict that emission in the lines is most highly favoured. Part of the discrepancy is removed if this assumption is dropped and the emission in each line is calculated separately before taking the ratio. These two calculations do not yield the same result because there is a high temperature tail in the ionization curve of the lithium-like ions which results in the transitions of higher excitation energy, the 2l–3l′ transitions, being enhanced relative to the 2s–2p transition. The unsatisfactory agreement between theory and observation could have important implications for current theories of the ionization equilibrium, but, before definite conclusions can be reached, further observations of the same type need to be made and remaining uncertainties in the cross-sections to be resolved.

New results are presented from high resolution Bragg crystal spectrometers flown in late 1970 on two Skylark rockets. The first instrument, launched on 24 November 1970 at 22 13 UT from Woomera, South Australia, contained two crystal spectrometers, each with an effective area of 50 cm2 and field collimation to 3’ FWHM. This instrument obtained the X-ray spectrum of the quiet corona in the wavelength range 5–14 Å. The second instrument was launched on 6 December 1970 at 11 13 UT from Sardinia, Italy, and contained four crystals of 6 cm2, each collimated to 4’ FWHM. This instrument was pointed at a non-flaring active region near N20 W40(McMath region 11060), and obtained an X-ray spectrum between 5 and 23 Å. This first use of a collimator to limit the field of view has considerably increased the spectral clarity compared with earlier observations by excluding the contributions of other active regions.

Some preliminary results of a rocket experiment, which utilized collimated Bragg Crystal spectrometers to map solar X-ray line emissions from He-like ions of oxygen and neon are presented. The two spectrometers used KAP crystals with a common field of view 1.7’ FWHM in the narrow dimension. The spectrometers were scanned in wavelength once each second to observe the resonance, intersystem and forbidden lines of OVII and NeIX.

A review is given of continuing efforts to generate laboratory spectra at photon energies above 1 keV from highly ionized atoms. Emphasis is placed on the most easily-interpreted highest ion stages such as the hydrogenic, helium-like and lithium-like species. The close similarities to solar flare observations of certain signatures, such as the line and continuum X-ray spectra and microwave emission as well as their sequence of occurrence, are pointed out and scaling factors are given.

The polarization and angular distribution of solar hard X radiation above 10 keV was calculated under the assumption that the X rays originate as bremsstrahlung from energetic electrons moving in a preferred direction. The source electrons are supposed to have a power-law spectrum. These conditions are to be expected in the impulsive phase of an X-ray burst. The spiral orbits of the electrons around the magnetic field lines are taken into account.

This paper reports recent progress in the classification of the spectra of iron XVIII to XXIV. Emission lines due to these stages of ionisation were reported in the spectra of solar flares by Neupert et al. (1967) who pointed out that the 2s2 2pn–2s2 2pn-1 3d configurations were mainly responsible. The present work verifies this and in addition gives individual line classifications of some of the strongest lines. Earlier studies of these transitions were reported by Fawcett (1965), Fawcett et al. (1967), and Feldman and Cohen (1968, 1970). Spectra of Fe XVIII to XXIV dueto 2s2 2pn–2s 2pn+1 transitions lie near 90 Å. Satisfactory solar and laboratory observations of these spectra are not yet available. The new classifications in these isoelectronic sequences are extended almost as far as iron and enable accurate extrapolations to iron. Isoelectronic spectra of both the forementioned transitions have been studied and the wavelengths tabulated (Fawcett, 1970, 1971a, b; Fawcett et al., 1971) for all the elements in the period between silicon and iron. The wavelength data enables the calculation of the 2s2 2pn energy levels and hence provides confirmation of the identity of solar forbidden lines in silicon, sulphur, argon and calcium. These energy levels are in close agreement with the theoretical calculations of Edlén reported at this Conference.