Copyright © 1981 by Annual Reviews. All rights reserved
|
| Annu. Rev. Astron. Astrophys. 1981. 19:
77-113 Copyright © 1981 by Annual Reviews. All rights reserved |
Direct spectroscopic analyses of individual stars are available in a few cases only which are confined - to our knowledge - to high-luminosity stars in the Magellanic Clouds and have given rather discrepant results in the past (Przybylski 1968, 1972, Wares, Ross & Aller 1968, Wolf 1972, 1973, Osmer 1973, Fry & Aller 1975, Foy 1977). Low-resolution spectra of giants in the Draco and Ursa Minor dwarf satellite systems resemble those of the most metal-weak galactic clusters (Cowley, Hartwick & Sargent 1978), and scanner measurements by Zinn (1978) of 23 giants in Draco give <[Fe/H]> = -1.9 with a range over a factor of 3. Stetson (1980) argued that the presumed range is due to inclusion of stars on the asymptotic branch, but a still bigger range has been found from spectral scans of the Ca+ K line (Kinman, Kraft & Suntzeff 1981). Other techniques applicable to individual stars in satellite systems of our Galaxy are DDO photometry (e.g. Gascoigne et al. 1976; but cf. Gustafsson, Bell & Hejlesen 1977) and the broader-band (and hence more rapid) Washington photometry (Canterna & Schommer 1978). Of these methods, the scanner technique (Searle & Zinn 1978) seems to be the most precise and reliable, at least for old populations resembling those of the galactic globular clusters, but all methods are in need of recalibration at the high-metallicity end (cf. Zinn 1980b).
Many of the statistical characteristics of stellar populations are
abundance-related and can be used (with due precautions) as abundance
indicators within (and occasionally beyond) the Local Group. These
have been reviewed by
van den Bergh (1975)
and will not be discussed
in detail here, but they include the systematics of color-magnitude
diagrams, the period and color distributions of cepheids (cf.
Iben & Tuggle 1975)
and other variables, and the color distribution of supergiants (cf.
Humphries & Davidson
1979),
with the numerical ratio
B/R increasing with metallicity although in the metal-weak
SMC no supergiants are detected with spectral type
later than M1
(Humphries 1979).
Most of these population characteristics are affected (in
principle at least) by age and initial mass function as well as
composition (the latter in itself possibly involving more than one
parameter), but in some cases there is hope that these can eventually
be sorted out (cf.
Kraft 1979
on the "second parameter" in globular
clusters). Thus, the "thick" giant branches in the Sculptor, Fornax,
and Ursa Minor dwarf systems have been taken as an
indication of a metallicity range similar to that in
Cen
(Norris & Bessell
1978,
Demers, Kunkel & Hardy
1979),
while the colors of SMC cepheids at a
given period indicate a uniform metal deficiency by a factor of about 4
(Gascoigne 1969,
Bell & Parsons 1972,
1974).
The H-R diagrams of clusters in the Magellanic Clouds (cf.
van den Bergh 1975)
have been
subjected to many conflicting interpretations which we do not have
space to review here. The brightest and reddest stars of intermediate
age in the Clouds are carbon stars
(Feast & Lloyd Evans
1973,
Blanco, Blanco &
McCarthy 1978,
Mould & Aaronson 1979,
1980)
with Mbol
- 5, as
are those in Fornax
(Demers & Kunkel 1979,
Aaronson & Mould 1980)
and Sculptor
(Westerlund 1979).
Infrared photometry of these very red
stars provides a powerful age indicator based on the systematics of
the AGB which implies that these stars are more massive and hence
younger than stars of similar metallicity in Galactic globular
clusters and that the metallicity spread in Fornax and the Clouds is
associated with a significant age spread (cf.
Norris & Zinn 1975);
this is in agreement with the classification of spectral scans of the
integrated light of Magellanic Cloud clusters by
Searle, Wilkinson &
Bagnuolo (1980)
which implies a quite smooth one-to-one correspondence
between these two parameters.