A good deal is known about the Sun from the measurement of its spectrum. The measurement of the ultraviolet spectrum has contributed to this knowledge and I shall try to summarize the results and problems of some of these investigations.

Stigmatic balloon spectra of the Sun in the vicinity of 2800 Å were obtained on September 22, 1968 and April 30, 1969. We compare the observed profiles of the H and K lines of ionized magnesium with the computed profiles of Athay and Skumanich and of Dumont. The discrepancy between observed and computed profiles of the Mgii lines is considerable. It is shown that this is a general fact for all chromospheric resonance lines of abundant elements. A brief review is given of different interpretations of this discrepancy.

As measured from space, the contribution of one moderate flare to the emission spectrum of the Sun in the far-ultraviolet (304 Å, Lyman continuum, etc.) is comparable with the emission of the whole undisturbed solar disc.

The distribution of the solar intensity in the Lyman α line has been measured close to the visible limb. It is compared to a computation including LTE departures (as evaluated by Y. Cuny). As far as the profile of the line and the intensity (integrated over the line) are concerned, the interspicular model of Coates is the only one which seems to agree with the observations.

A high-resolution solar spectrum in the range 2000–2200 Å was obtained in a recent flight of a sunpointing Skylark rocket. This was launched at 04.21 hr UT on April 22, 1969 from Woomera and reached an apogee of 178 km. An optical alignment system operating on the main vehicle pointing system gave a net stabilisation of ±3 arc sec in the position of the solar image relative to the spectrograph slit. The slit, of length 1.0 mm, was set in the north-east quadrant parallel to and 5 arc min from the north/south axis, its lower edge being 1 arc min from the equator. The roll control of ±2.5° was provided entirely by the standard Elliott Bros. type of vehicle stabilisation.

The resonance lines of Mgii occur at wavelengths (2802.7 Å, 2795.5 Å) just beyond the extinction limit of the Earth's atmosphere. At such wavelengths sophisticated optical techniques can now be employed and this fact, together with the high cosmic abundance of magnesium, makes these lines particularly important for study in UV Astronomy. In the case of the Sun, the lines consist of a broad absorption with a pronounced emission core.

Absorption at the Lyman-α transition from interstellar neutral hydrogen has been observed in the ultraviolet spectra of 18 nearby O and B stars. Radiation damping is the dominant cause of line broadening, which makes the derived line-of-sight column densities proportional to the square of the observed equivalent widths. An average hydrogen density on the order of 0.1 atom cm−3 has been found for most of the stars observed so far. This is in contrast to the findings from surveys of 21-cm radio emission, which suggest 0.7 atom cm−3 exists in the local region of the Galaxy. Several effects which might introduce uncertainties into the Lyman-α measurements are considered, but none seems to be able to produce enough error to explain the disagreement with the 21-cm data. The possibility that small-scale irregularities in the interstellar gas could give significantly lower values at Lyman-α is explored. However, a quantitative treatment of the factor of ten discrepancy in Orion indicates the only reasonable explanation requires the 21-cm flux to come primarily from small, dense, hot clouds which are well separated from each other. The existence of such clouds, however, poses serious theoretical difficulties.

The equivalent width of the blended line at Lyman α is given for 48 stars measured with the OAO-A2 scanning spectrometer. This provides an upper limit to the neutral hydrogen column density. In the Orion association these upper limits are significantly lower than the column densities determined from 21-cm emission line measurements. The determination of the Lyman α equivalent width for θ Ori by Carruthers is rediscussed and agreement between 21-cm absorption measures and Lyman α absorption is obtained for a spin temperature in the range of 40–70 K. It is suggested that the most likely explanation for the discrepancies found for the other Orion stars is that the 21-cm emission primarily occurs beyond the Belt stars.

The correlation between the OAO blended equivalent widths and color excess, 4430 Å absorption, and interstellar sodium absorption are examined. Excellent correlation between sodium and hydrogen column densities is found.

I would like to extrapolate beyond the present state of interstellar line spectroscopy toward the next generation of satellite-borne equipment, but with due recognition of the fact that more work still has to be done from the Earth's surface.

The dominant photodissociation process for H2 under interstellar conditions is absorption in the lines of the ultraviolet Lyman bands at 912 Å < λ < 1108 Å and re-emission into the continuum at λ ~ 1600 Å. It is shown that 40% of all upward transitions in this band lead to emission into the continuum of the ground state and therefore dissociation of the molecule. The photodissociation rate is calculated and the theory is developed for the abundance of interstellar H2. It is shown that interstellar space will be divided into atomic and molecular regions in a manner analogous to the division between Hii and Hi regions. The molecular regions will exist only where the rate of formation of H2 is sufficiently great to build up a self-shielding layer. This is expected in cold clouds with number densities greater than 100 atoms cm−3. Normal regions of interstellar space will have an extremely small abundance of molecular hydrogen.

1. (1) The minimum measured brightness of the night sky (in the visual region) is about 100 stars of the 10th magnitude per square degree near the galactic poles.

2. (2) The ratio of fluxes in the ultraviolet (2300–3000 Å) and visual regions is approximately in agreement with expected theoretical data based on the models of stellar atmospheres for spectral classes B0 to G5 and on the distribution of stars of the different spectral classes over the sky.

3. (3) The nature of possible deviations (theory minus observation) is discussed.

The Mariner 5 ultraviolet measurements obtained while the spacecraft was in interplanetary flight are interpreted as Lyman-α radiation. This radiation may arise from (1) the scattering of solar Lyman-α radiation by interplanetary atomic hydrogen, (2) the scattering of solar Lyman-α by atomic hydrogen that is present at the edge of the solar system, or (3) diffusely scattered Lyman-α radiation from the Galaxy. The Mariner 5 measurements show a symmetry with respect to the galactic equator which suggests that the major source of the observed emission is the diffuse galactic radiation. In the Münch model for the diffuse Lyman-α radiation from an HI region, the sources of the Lyman-α photons are stellar chromospheres, the photons are scattered a large number of times by interstellar atomic hydrogen, and are absorbed by interstellar dust. Some of the Mariner 5 measurements may be attributed to Lyman-α emission from an HII region since the field of observation scanned the edge of the Gum nebula.

1. (1) Observations. – A survey is made of observations of the background radiation at UV wavelengths from above the atmosphere. Sources of the background radiation and ways of determining the extragalactic component of the background are discussed. Future observations are also discussed.

2. (2) Cosmology. – Limits to the properties of the intergalactic gas follow from observations of the UV background. The problem of detecting galaxies at early stages in their evolution is considered.

3. (3) The Galaxy. – Observations and theoretical estimates are given for the integrated brightness of the Galaxy at UV wavelengths beyond the Lyman-α line. Also discussed are the nature of the Lyman-α emission from the Milky Way and the principal ways of constructing the luminosity function of stars in the Galaxy from observations of the integrated UV spectrum.

4. (4) Subcosmic Rays in the Interstellar Medium. – Fast neutral excited hydrogen atoms can be formed from charge-exchange interactions between subcosmic-ray protons and neutral interstellar gas. Upper limits are given to the energy density of subcosmic rays having E ~ 100 keV.

5. (5) Limits on the Background Radiation in the range <912 Å. – The distribution of neutral hydrogen in the peripheries of galaxies allows limits to be obtained for the flux of metagalactic ionising radiation. The heating and ionisation of the interstellar medium by X-rays is considered.

For a simplified model of a gaseous nebula, that is spherical, isothermal, homogeneous and composed of pure hydrogen, the amount of Lyman-α radiation emitted from the surface of the nebula is evaluated for various physical conditions. The problem of observability of the Lyman-α emission line is examined, taking into account absorption by dust and by interstellar neutral hydrogen.

One of the most important results of the present Symposium is the obvious necessity of close cooperation between ground-based and UV stellar spectral observers. Either of the two observational techniques yields only part of the total information about the stars and the interstellar medium; together they enable the scientist to construct the imposing image of modern stellar spectroscopy that we have seen growing during this Symposium.