Some general effects of size, shape, and material on the scattering properties of small particles are considered with a view to determining their relative importance to interstellar dust interpreatations.

It is shown that mixtures of graphite core-ice mantle grains with large and small mean size and graphite grains provide good fits to the observed interstellar extinction curve.

OAO-2 near UV spectral scans of 75 early-type stars have been examined for evidence of the UV extinction bump. The bump has been detected in 40 of these stars. The position of the maximum of the bump is within ± 40 Å of 2175 Å in all cases. A color excess between the wavelengths 2175 Å and 3500 Å was determined for all 40 stars. This color excess, which is a measure of the bump strength, correlates very well with E(B – V) except for stars associated with nebulosity.

The ratio of total to selective extinction has been determined for the lines-of-sight to six planetary nebulae.

One hundred and fifty six open clusters with modern photometry, and apparent diameter measures by Trumpler, have been used to measure the mean ratio of total to selective extinction, R̄ It is found that cluster diameter is related to both concentration class and richness class. Diameter also seems to depend on the background stellar density and to decrease with reddening. With such effects measured and removed R̄ = 3.15 ± 0.20. This result is not significantly dependent on the cluster's position in space.

From star counts the extinctions in five spectral bands were determined for four areas of the Southern Coalsack. The wavelength dependence of the extinction values follows the Whitford curve. Minimum masses for six globules are given.

Recent observations of interstellar extinction and diffuse galactic light in the ultraviolet are interpreted in terms of a bi-modal size distribution of interstellar grains.

Integrals over the observed spectrum of interstellar extinction to 9 μm–1 are related to known laboratory properties of specific compounds. Using Lα hydrogen densities and cosmic abundance ratios, many suggested grain materials can be eliminated.

We have searched for diffuse features at λ 4430, 5780 and 6613 in the spectra of 17 stars which show infra-red excesses or other grounds for believing that they have dusty envelopes. In no case could we identify diffuse bands that may be ascribed to the circumstellar envelope. We conclude that the grains in circumstellar envelopes are not identical with the grains in the general interstellar medium.

Eleven diffuse interstellar features have been studied in the spectra of dust-embedded stars in Cyg OB2 (VI Cyg) (including a Wolf-Rayet star) and field stars in the same direction. The equivalent widths of the lines are all strongly correlated with each other and with EB – V, but there is virtually no correlation with D line strength for these stars. The wavelength positions and half-widths of the diffuse lines appear to be the same for all stars studied.

It has been reported by several authors that the extinction profile near the wavelengths of unidentified diffuse features shows fine structure (Brück and Nandy, 1970; Walker, 1971; York, 1971; Bromage, 1972); the question arises whether these features also exhibit fine structure in wavelength dependence of polarisation. It was first suggested by Greenberg et al. (1970, 1971) that interstellar bands produced by grain impurities may be polarized differently from the neighbouring continuum. Similar conclusions were also derived by Wickramasinghe and Nandy (1971) and Kelly (1971). Our first attempt to derive a profile of polarization against wavelength across the 4430 Å band was reported earlier (Nandy and Seddon, 1970). In this method we have used a Wollaston prism at the Cassegrain focus of the 36″ Telescope of the Royal Observatory, Edinburgh to produce orthogonally polarised spectra on the photographic plate. The results for the star 55 Cyg (Figure 1) indicated a definite profile. It was felt than an extension of the programme to acquire further profiles and to establish instrumental polarisation to a greater accuracy was necessary.

We have analysed observations of the interstellar extinction in the range 3400 Å–l 1000 Å. The observations have high photometric accuracy and wavelength resolution, and allow a detailed examination of broadband structure as well as the general shape of the wavelength dependence of the extinction curve. The broadband structure has a characteristic size of several hundred ångströms, and may be as important as the diffuse bands in indicating the physical nature of the grains.

We describe our preliminary theoretical interpretation of this broadband structure, and the new observation which it predicts.

At a resolution of 0.17 Å, the profile of λλ 5780, 5797 were found to be asymmetric and steeper to the blue. The FWHM of λ 5780 and λ 5797 are respectively 2.7 Å and 1.1 Å. There is also a broad, shallow feature shortward of λ 5780. Equivalent widths of λλ 5780, 5797 were measured at a resolution of 0.26 Å, and they present a very tight correlation. Therefore, it is highly likely that these two diffuse lines have a common origin. The correlation between the equivalent width of λ 5780, EW(λ 5780), and E(B – V) is good for normal field stars. But there is a systematic weakening of λ 5780 in the spectra of Be stars, and the reverse is true for stars in the Scorpius-Ophiucus region. Excluding Be stars, a least square fit gives EW(λ 5780) = 0.75 E(B – V) Å mag.–1. From the profiles and structures of the diffuse features, and the correlations with various parameters of the interstellar medium, it is suggested that: (1) the diffuse band λ 4430 may be produced by the preionization of H– and, (2) λλ 5780, 5797 may be due to the pure electronic transitions of the impurity centers in solid grains.

Two methods of determining the state of polarization of the integrated light from the Milky Way are discussed. In the symmetry method it is assumed that the polarized foreground zodiacal light possesses north/south symmetry relative to the ecliptic: an average polarized intensity of at least 10 S10(R) is derived for the Milky Way at l ≃ 90° and 270°, which is consistent with the result of Sparrow and Ney (1972) who obtained 18 S10 (R) at l = 35° and 210°. In the method of two epochs the above assumption is not needed but instead time variations of the zodiacal light are assumed to be insignificant: the results obtained by this method (in S10 (V) units), although compatible with the values found by the symmetry method, do however show that time variations may have influenced this result. Although the evidence for a definite polarization of the Milky Way is persuasive it cannot yet be considered conclusive.

Observations and Mie theory calculations of colors and polarization in several reflection nebulae are compared.

Satellite observations of the Merope and Barnard's Loop nebulae in the 1050 to 4250 Å spectral region show that the increase in the albedo of interstellar grains below 2000 Å occurs in individual reflection nebulae as well as in the general interstellar medium. Evidence for differences in the composition and size distribution of grains due to variations in the local radiation field is presented.

The Curtis Schmidt telescope at the Cerro Tololo Observatory has been used for a two-color survey of reflection nebulae in the Southern Milky Way. Major concentrations of reflection nebulae are found in Carina and in the zone 250° < lII < 270°. A conspicuous gap in the distribution of reflection nebulae occurs at lII ≃ 275°. A detailed photometric and spectroscopic investigation of the stars imbedded in southern reflection nebulae is now in progress.

The depolarization effects in reflection nebulae are caused by the size distribution, multiple scattering and irregularity of the grains. In the following abbreviated version of a study of the depolarization effect it is shown that the multiple scattering in reflection nebulae can lead to considerable decrease of the linear polarization. A sample of results obtained by the Monte Carlo method is tabulated.