Search

MANY attempts have been made recently to match the wavelength dependence of the observed interstellar extinction curves with calculations for theoretical models. These attempts (reviewed in refs. 1 and 2) have met with varying degrees of partial success. It is now generally recognized that a pure dielectric particle will not be able to explain the entire wavelength range of the extinction observations. In particular the very strictly linear portion (0.8µ−1< λ−1< 2.3µ−1) of the mean extinction curves of Nandy3,4and the continued increase in extinction towards the far ultraviolet5would seem to rule out homogeneous dielectric particles as a significant constituent of interstellar material. Calculations by Stecher and Donn6and by ourselves7have indicated that a size-distribution of pure graphite particles could produce general agreement with observations in the visible and ultraviolet spectral regions up to ∼ 2200 Å, but no agreement is possible further in the ultraviolet. This agreement includes the reproduction of an apparent hump in the observed extinction at ∼ 2400 Å—which Stecher and Donn have referred to as the “signature of graphite”. Of the grain models so far proposed the graphite core–dice mantle grain seems most favourable in being able to reproduce the detailed extinction curves in the visible spectral region as well as the overall features of the extinction up to ∼ 1500 Å (ref. 8). For such particles, however, the hump in the extinction at ∼ 2400 Å characteristic of graphite is washed out because of the effect of the outer ice mantle. If the structure in the extinction curve at the same wavelength is confirmed by future observations, the graphite core–ice mantle grain cannot be regarded as an entirely satisfactory model. A graphite component of the grains is, however, strongly indicated both by this ultraviolet feature as well as by the kink at λ−1= 2.3µ−1of the visible extinction observations, the latter being probably associated with a change in the refractive index of graphite. It is therefore interesting to explore the possibility that a graphite core with a dielectric mantle of refractive index different from that of ice may produce better agreement with the observations.

REDDISH1has suggested that there is a major discontinuity in the stellar and interstellar features of the galaxy at lII= 140°. The evidence comes from large changes in the space densities of Wolf–Rayet stars, dust-embedded OB stars and globules, in the rotation with wavelength of the position angles of interstellar polarization of starlight, and in the distribution of the position angles of the polarization of the radio emission. These interstellar features appear to rule out the apparent termination of the Perseus Spiral Arm as the cause of the discontinuity2, a conclusion supported by an absence of a discontinuity in the surface density of neutral hydrogen at this longitude3.

ACCURATE measurements of the wavelength dependence of interstellar obscuration in the spectral range from 10,000 Å to 3400 Å (refs. 1–4) show a distinct change in the slope of the extinction curve at 4300 Å. Extinction observations have been made for a number of galactic regions, and it is found that a change in slope always occurs at 4300 Å but that the magnitude of the change differs significantly from one region to another. At the same time, an unidentified interstellar band of half-width approximately 20 Å appears at 4430 Å, but the discontinuity in the slope of the extinction curve does not affect the profile of this band5.