Although once it was thought that main-sequence stars are remarkably homogeneous with respect to their chemical composition, the upper main-sequence stars (30000 > Te > 7000) show a variaety of chemically peculiar stars besides the so-called normal stars. Those include the Am, Ap, λ Bootis, He-deficient, and He-rich stars. This review summarizes the current data, which are necessary to construct and test the theoretical models of these stars. In the second half of the review we concentrate on Li. In the lower main-sequecnce stars abundances of Li have been determined in hundreds of stars. Some of the remarkable results are: (1) A uniform upper abundance value irrespective of stellar effective temperature, (2) abundance gap in the F stars of the Hyades, and (3) increasing depletion with smaller stellar mass for the Hyades.

Constraints that abundance anomalies observed on main sequence stars put on turbulence, meridional circulation and mass loss are reviewed. The emphasis is on recent observations of Li abundances.

Upper limits to turbulence are obtained from the Be abundance in the Sun and from underabundances of Ca and Sc in FmAm stars. The Li abundance in G type stars suggests the presence of turbulence below convection zones.

The abundance anomalies, both over and underabundances, observed in FmAm and λ Booti stars can be explained by diffusion in the presence of mass loss. A mass loss rate of 10−15 Mo yr−1 is required to explain the FmAm stars while a mass loss rate of 10−13 Mo yr−1 is required by the λ Booti stars.

The position and width of the Li abundance gap observed in Hyades and other open clusters is explained by diffusion. A detailed reproduction of the Li(Teff) curve seems to require a mass loss rate of slightly more than 10−15 Mo yr−1, of the same order as the mass loss rate required by the FmAm stars. In the presence of such a mass loss only small overabundances of heavy elements are expected. The observed variations in the Li abundance as a function of the age of clusters suggests that the Li abundance observed in old halo stars does not represent the cosmological abundance.

Detailed two dimensional calculations of diffusion in presence of meridional circulation for HgMn and FmAm stars lead to a cut-off of about 100 km s−1 for the maximum equatorial rotational velocity at which abundance anomalies are expected in these objects. This agrees with observations. A similar calculation for the F stars of the Hyades where Li underabundances are observed leads to a contradiction, unless meridional circulation patterns are modified by the presence of convection zones once they become as large as in late F stars. There remains a possibility that meridional circulation would be responsible for some of the reduction of the Li abundance as observed in the Hyades and UMa. Further observations are suggested to distinguish the effects of settling and nuclear destruction.

The “lithium gap” observed in the Hyades and other galactic clusters by Ann Boesgaard and her collaborators (Boesgaard and Tripicco 1986, Boesgaard 1987, Boesgaard, Budge and Burck 1987) gives a challenge to theoreticians. Indeed a good fit between the theoretical results and the observations will give a clue for our understanding of the stellar internal structure and evolution.

A theoretical explanation of the “lithium gap” by gravitational and radiative diffusion has been proposed by Michaud 1986. In G type stars, the convection zone is too deep for gravitational settling to take place: the density at the bottom of the convection zone is so large that the diffusion time scale exceeds the age of the star. Increasing the effective temperature leads to a decrease of the convection zone, and consequently to a decrease of the diffusion time scale. In F stars it becomes smaller than the stellar age, leading qualitatively to a lithium abundance decrease as observed. When the convection zone is shallow enough, the radiative acceleration on lithium becomes important as lithium is in the hydrogenic form of li III (while it is a bare nucleus, li IV, deeper in the star). This radiative acceleration may prevent lithium settling for hotter F stars. This is a very attractive explanation, which leads to a minimum of the lithium abundance nearly at the place where it is observed in effective temperature. However it suffers from some difficulties: the theory predicts an increase of the lithium abundance larger than normal in the hottest F stars, which is not observed, and the predicted minimum lithium abundance is one or two orders of magnitude higher than the minimum observed in the Hyades. The former may be overcome if mass loss occurs in these stars (Michaud 86). Let us focus on the latter.

The abundance of lithium in stellar atmospheres presents an important observational constraint to the hydrodynamical models of the outer layers of stars. It can be considered as a cumulative measure of the extent of matter exchange between surface and deeper layers during the stellar evolution.

From the observed large scatter of lithium abundances in evolved stars it follows that the efficiency of mixing has been highly variable from one object to another. At present, it seems to be difficult to find any satisfactory explanation to the lithium abundances of individual stars. We suppose that at this stage the statistics of lithium abundances in different types of stars can give some insight into the character of mixing processes operating in stars. In this report some observational results about the distribution of the lithium abundances in normal late-type giants are presented.

Nature has filled the upper right quadrant of the Hertzsprung-Russell diagram with more varieties of peculiar stars and odd chemical compositions than even our most speculative observers and theorists could dream up. To bring some structure to this vast subject I will categorize the phenomena we observe according to our model of stellar evolution, dividing the stars among the first ascent of the giant branch and the core-helium burning phase, the asymptotic giant branch (double shell-burning) phase, and the post-AGB and pre-planetary nebula stars. The types of stars found in these three groups are summarized below.

The Magellanic Clouds are sufficiently close that evolved stars which exhibit chemical peculiarities and the effects of mass loss can be readily observed. Such objects include carbon stars, S stars, long-period variables, OH/IR stars and planetary nebulae. Because of the relatively well-known distances of the Magellanic Clouds, the intrinsic luminosities of these objects can be accurately determined, in contrast to the situation in the Galaxy where the great majority of asymptotic giant branch (AGB) stars occur in the field population. In this review, observations of AGB stars in the Magellanic Clouds will be discussed with particular reference to those features which can shed light on mass loss and chemical peculiarities resulting from stellar evolution.

The existence of carbon stars brighter than Mbol=-4 can be understood in terms of dredge up in thermally pulsing asymptotic giant branch (AGB) stars. As a low- or intermediate-mass star evolves on the AGB, the large fluxes engendered in a helium shell flash cause the base of the convective envelope to extend into the radiative, carbon-rich region, and transport nucleosynthesis products to the stellar surface. Numerical models indicate that AGB stars with sufficiently massive stellar envelopes can become carbon stars via this standard dredge-up mechanism. AGB stars with less massive stellar envelopes can become carbon stars when carbon recombines in the cool, carbon-rich region below the convective envelope.

Neutron capture occurs on iron-seed nuclei during a shell flash, and the products of this nucleosynthesis are also carried to the stellar surface. The conversion of 22Ne into 25Mg can initiate neutron capture nucleosynthesis in largecore mass AGB stars, but only if these stars can survive their large mass loss rates. The current estimates of nuclear reaction rates do not allow for appreciable neutron capture nucleosynthesis via the 22Ne source in lower mass AGB stars. The carbon recombination that induces dredge up in AGB stars of small envelope mass, however, also induces mixing of 1H and 12C in such a way that ultimately a 13C neutron source is activated in these stars. The 13C source can provide an abundant supply of neutrons for the nucleosynthesis of both light and heavy elements. While the existence of neutron-nucleosynthesis products in AGB stellar atmospheres can be understood qualitatively in terms of an active neutron source, the combination of nuclear reaction theory and evolutionary models has yet to provide quantitative agreement with stellar observations.

It has been found by Utsumi(1985a,b) that in J-type carbon stars of C4-5 and WZ Cas(C9,2J Li), abundances of s-process elements with respect to Fe are nearly normal, while in normal carbon stars of C5-8, heavy metals are overabundant by factors of 10-100, and rare-earth elements are overabundant by a factor of about 10.

In the MK system, most J-type stars are classified as C4-5,4-5 stars which show very strong C2 and CN bands. Yamashita(l972,1975) classified many C7-9 stars most of which are CS or SC stars. His classification of C7-9J stars is mainly based on Cl2C13(0,1)band at 6168 A, C13N(4,0)band at 6260 A, and LiI 6708 A line. In most of C7-9 stars, lines of s-prooess elements are greatly enhanced. It is a question if in all J-type stars abundances of s-process elements are nearly normal or not.

Studies of the carbon and nitrogen abundances in metal poor giants have generally supported the models of Sweigart and Mengel (1979) which indicate the action of meridional circulation in CN-processing metal poor red giant envelopes. Carbon et al. (1982) find the trends in the carbon and nitrogen abundances in M 92 stars to be at least qualitatively consistent with CN-cycle processing of the envelope, but stars at all phases of giant branch evolution have unusual, and unexplained, carbon and nitrogen abundances. The oxygen abundance is the missing piece of information which may explain what’s going on. Sweigart and Mengel’s models suggest that ON-processing might also occur in the most metal poor giants. The sample of giants in M 92 for which Carbon et al. have provided carbon and nitrogen abundances is ideal for the determination of oxygen abundances to look for the effects of ON-processing. Once the oxygen abundances are known, the sum C+N+0 can be computed to see if it is constant along the giant branch or varies from star to star.

The carbon star is one of the best probes for the galactic study;

1. (1) it is intrinsically bright (Mbol = − 2 to − 6) especially in the red and infrared wavelength regions,

2. (2) it has spectral features readily detectable on objective prism plates due to their strong carbon molecular bands,

3. (3) it is an evolved star distributed abundantly (∼1 star per square degree) along the galactic plane.

We can detect it in the Galaxy up to several kpc from the sun on objective prism plates of the Schmidt telescope.

We have been making survey observations of faint cool carbon stars using the Kiso 105-cm Schmidt telescope. Kodak IN and 103aF plates are respectively taken behind the 4-degree objective prism (700 Åmm−1 at Hα) for the detection and for the spectral classification. V-band plates are utilized to obtain the position and the brightness of the stars detected.

The survey areas are distributed along the northern galactic plane. Seven fields in the Cassiopeia region (l = 115° to 133° and eight fields in the Taurus-Auriga-Gemini region (i = 170° to 188°) have been observed and processed up to now (Maehara and Soyano 1987a,b).

The surface distribution of M stars is studied by differentiating them according to whether they show a circumstellar dust shell (CS) or not. Analysis shows that galactic latitudinal and longitudinal distributions are not determined by spectral subclasses alone. The study also indicates that the M type stars with CS have higher intrinsic luminosities in the K band than those without CS. The M stars used in the study are obtained from the Two Micron Sky Survey catalogue (IRC) which is an unbiased sample with respect to the interstellar extinction. The CS feature is identified by the ratio of flux densities at 12 and 25 μm in the IRAS point source catalog.

The J=2-1, v=0 emissions of 28SiO, 29SiO, and 30SiO from three late-type stars were simultaneously observed with the Nobeyama 45-m telescope in Jaunuary 1987. The relative intensities of [29SiO] / [30SiO] were measured to be 2.4 for χ Cyg, 1.5 for NML Tau, and 2.9 for V1111 0ph. These values are lower limits for the relative isotope abundance of [29Si] / [30Si], and are larger than the terrestrial value of 1.51. Recext theoretical studies suggest that hydrostatic nucleosysnthesis and supernova explosions result in smaller values of [29Si / [30 Si] than solar. These models cannot explain the observed excess of 29Si.

Infrared spectra of evolved stars are generally dominated by the radiation from their circumstellar shells. M stars are characterized by the 10 μm emission feature from silicate dust grains, while C stars by the 11 μm SiC band. However, some C stars have been found to show the 10 μm feature indicating the oxygen-rich property of their circumstellar dust (Willems and de Jong 1986, Little-Marenin 1986).

In order to investigate the gas phase chemistry of the circumstellar envelopes around these peculiar objects, we have observed radio molecular lines of H2O, SiO, HCN, and CO towards three of them BM Gem (C5, 4J), V778 Cyg (C4, 5J), and EU And (C4, 4).

A spectroscopic diagnosis on M1-5, M1-9, K3-66, and K3-67 is presented. Our sample were chosen from the catalogue of radial velocities (Schneider et al 1983) by the reason of not only their kinetic peculiarity, but also apparently compact images. Main purpose is to analyze the chemical properties which should give us an information about galactic chemical abundance distribution in the direction of galactic anti-center region based upon kinetic peculiarity. Another one is to study on an expansion characteristics by which we can recognize intrinsically compact planetary nebulae in young phase.

Spectroscopic observations were made at the Okayama Astrophysical Observatory (Shibata and Tamura 1985), the Steward Observatory, and Lick Observatory (Tamura and Shaw 1987).

It is generally accepted that the helium flash occurs when the 3α reaction commences in the degenerate helium core of low mass stars. In this core, original CNO isotopes have been converted into 14N and the electron Fermi energy becomes large enough to approach the threshold energy for e-capture on 14N. Hence Kaminisi et al. (1975) have pointed out that in these circumstances the 14N(e−, v)14C(α,γ)18O (NCO) reaction may play an important role for igniting the helium flash.

We, therefore, examine the effects of the NCO reaction on the evolution of low mass stars. A key ingredient of the NCO reaction is that the density reaches the threshold for e-capture (ρth≃ 106 g cm−3). Evolutionary sequences are presented for the cases of accreting helium white dwarfs (Hashimoto et al. 1986) and a 0.7 M⊙, Population II star ascending the giant branch.

In the upper left corner of the HR-diagram various stars have been found that are clearly not members of the hydrogen main sequence. The majority of them lie to the left and below the main sequence, indicating that they are highly evolved stars close to the extinction of their thermonuclear power source and hence to rapid cooling towards the white dwarf domain. The implication is that a “short” time ago they were red giants and as such experienced phenomena like mixing, dredge-up and heavy mass loss. Therefore, one expects the hot evolved stars to display the consequences of the afore mentioned processes that are up to now not satisfactorily understood.

The ubiquitous evidence for processed material in the atmospheres, winds, and circumstellar ejecta of massive stars will be reviewed. A broad array of normal and peculiar evolutionary stages is considered, up to and including Type II supernova progenitors. The quantitative analysis of these spectra is difficult, and until recently for the most part only qualitative or approximate results have been available. However, several important current programs promise reliable abundance determinations, which will enable detailed comparisons with recent evolutionary calculations. A significant emerging result is that the morphologically normal majority of both hot and cool supergiants may already display an admixture of CNO-cycle products in their atmospheres. It may become possible in this way to identify blue supergiants returning from the red supergiant region, as appears to have been the case for the SN 1987A progenitor.

For main sequence stars, the central nuclear processing generally has no effect on surface abundances. Later in the evolution, the newly synthetized elements may be revealed at the stellar surface by processes such as mass loss, convective dredge-up, overshooting, diffusion, rotational and tidal mixing, etc. The changes of CNO abundances are the most conspicuous and the easiest to observe spectroscopically; some abundance ratios like C/N, O/N may undergo changes by more than 102. On the whole, surface chemistry is a most powerful diagnostics of stellar evolution, model assumptions and nuclear cross sections.

Since light variability in white dwarfs was first discovered twenty years ago, eighteen DA white dwarfs, several pulsating DB white dwarfs, and hotter pre-white dwarfs have so far been found to be pulsating variables. The most conspicuous characteristics of pulsations in these stars are that they seem to consist of multiple g-modes of nonradial oscillations. Attention should be paid to multiplicity of modes. Stimulated by the success of helioseimology, a research field called ‘asteroseismology’, in which we may probe the internal structure of stars by means of observations of their oscillations, is going to develop. How well such a seismological approach succeeds is dependent on how many modes are observed in each of stars. Since the number of modes of an individual pulsating white dwarf is larger than those of other types of pulsating stars but for the Sun, the seismological study may be the most promising as to the white dwarfs. In fact, by applying the asymptotic relations among eigenfrequencies of high order g-modes with low degree, the degree l, and the radial order n, Kawaler(1987a,b,c) succeeded to get some constraints on the physical quantities of some of pulsating white dwarfs.