Whether the H II regions of dwarf emission-line galaxies are self-polluted by the nucleosynthetic products of massive stars remains an open question, despite recent claims to the contrary. We show that the finding by Kunth and collaborators that the neutral interstellar medium of I Zw 18 has a metallicity one order of magnitude lower than that of the ionized gas is highly uncertain. The range of possible values of (O/H) admitted by the saturated O I absorption line used in their analysis is very large, spanning a factor of the order of 1000. We suggest that future observations of the intrinsically weaker S II 1256 multiplet offer the best prospects for measuring the true degree of metal enrichment of this dwarf galaxy.

The damped Lyman alpha systems seen in the spectra of high redshift QSOs offer the means to determine element abundances in galaxies observed while still at an early stage of evolution. Such measurements, which have only recently come within reach, complement and extend the data provided by studies of different stellar populations in our Galaxy and of extragalactic H~II regions which have up to now formed the basis of galactic chemical evolution models. In this paper we demonstrate the potential of this new approach with high-resolution echelle observations of several elements in the z(abs) = 2.27936 absorption system in the bright z(em) = 2.940 QSO 2348-147. The absorbing galaxy appears to be chemically unevolved, with heavy element abundances only 1/100 of solar; if it is the progenitor of a spiral galaxy like our own, it is unlikely to have collapsed to form a thin disk by z = 2.3 (corresponding to a look-back time of approximately 13 Gyr for H_0 = 50 km/s/Mpc and q_0 = 0.01). Our data allow us to measure the nitrogen-to-oxygen ratio at a metallicity lower than those of the most metal-poor dwarf galaxies known. We find that, relative to the solar scale, N is more underabundant than O by at least a factor of 15. This result is broadly in line with current ideas on the relative importance of primary and secondary production of N; future measurements in several damped Lyman alpha systems will permit more stringent tests of models for the evolution of the N/O ratio with time. Oxygen and other alpha-elements are overabundant relative to Fe by no more than the factor of about 3 typical of metal-poor stars in the disk and halo of the Milky Way.

Our knowledge of galactic chemical evolution is currently limited to observations of Milky Way stars and H II regions of nearby galaxies. Damped Lyman~$\alpha$ systems offer a new approach for tracking the evolution of normal galaxies from early epochs to the present day. Here we report the first measurements of nitrogen abundances in galaxies with less than 1/100 of solar metallicity, a range unexplored by previous observations.

We present observations of Ti~II~$\lambda 3384$ absorption towards 15 distant stars in the Galactic halo and the Magellanic Clouds. These new data extend existing surveys of the distribution of Ti$^+$ to larger distances from the plane of the Galaxy than sampled previously, allowing the scale height of titanium to be determined for the first time. We find $h_{Ti^+} = 1.5 \pm 0.2$~kpc, a value which although greater than those of other tracers of neutral gas, is not as large as had been suspected. We interpret the extended distribution of Ti$^+$ as an indication that its severe depletion in interstellar clouds in the disk is reduced at the lower densities prevailing in the halo. The data are consistent with a simple power-law dependence of the Ti abundance on the ambient density, with exponent $k \simeq -1$. If the model is correct, it implies that refractory elements like Ti are fully returned to the gas phase at distances beyond $\sim 1$ kpc from the plane of the Galaxy.

We present spectroscopic observations of supernova 1994D in NGC 4526, an S0$_3$ galaxy in the Virgo cluster 15 Mpc distant. The datasets consist of the interstellar Ca\II\ and Na\I\ lines towards the supernova at high spectral resolution (FWHM 6 \kms), \halpha\ and [N\,II] observations at lower resolution (FWHM 33 \kms) of the nucleus of NGC 4526 and the supernova, obtained with the WHT at La Palma, and 21 cm spectra obtained with the 100 m Effelsberg Radiotelescope in the field of NGC 4526. The velocity of the gas in NGC 4526 determined from our \halpha\ spectra is +625 \kms\ at the centre and +880 \kms\ at the supernova position. Our value of the systemic velocity is higher than the previous value of +450 \kms. In our high resolution spectra we detect Ca\II\ and Na\I\ absorption at +714 \kms\ which is produced in interstellar gas in NGC 4526. To our knowledge this is the first detection of interstellar absorption originating in a galaxy of early morphological type. We detect multi-component Ca\II\ and Na\I\ absorption lines in the range from +204 to +254 \kms\ which originate in a complex of High Velocity Clouds (HVCs) located at a distance $\ll$ 1 Mpc, in the surroundings of the Milky Way. The gas appears to have near solar abundances, low dust content, and a diluted UV radiation field. At $-$29 \kms, we find weak Ca\II\ absorption and weak H\I\ emission. This component may originate in gas infalling onto the Galactic disk. Finally, close to rest velocity, we find both warm and cold gas located beyond 65 pc, probably associated with high latitude gas at the border of Loop I. The total reddening of the supernova, estimated using the standard Milky Way gas-to-dust ratio, is $E$(B$-$V) $\simeq$ 0.05.