Shcherbakov (1979) first reported that the S star HR 1105 shows a strong and variable He I X10830 line. We confirm this report and present spectra of HR 1105 around 10830 A taken over approximately 0.3 orbit, from a program to obtain full orbital coverage for this 600-day system. Using Griffin's (1984) orbital solution, we have phased in the data presented by Shcherbakov with our own and found that for two coincidences of phase, the 10830 profiles at different epochs but same orbital phases are identical within the limitations of the data. The earlier observations were taken eight orbits (13 years) before ours.

A number of extensive chemical equilibrium calculations have recently been performed for temperatures below 2000 K, with mixtures containing over 200 gaseous species and the allowance for the formation of over 60 condensates. These calculations were based on the minimization of the Gibb's free energy.

During our 1612 MHz survey of IRAS sources in Parkes, Australia, we found two objects with very large expansion velocities : IRAS15405-4945 and IRAS16342-3814 (P. te Lintel Hekkert et al. 1988, Astron. Astrophys. (Letters), in press.). These two sources and the well known source, OH231.8 + 4.2 , were discovered to have almost identical IRAS colours. The spectra of the three sources could be fitted with a 100 (±15) K black body curve. We than selected, using the IRAS colour-colour plot, a number of bright sources ( F (60 μm) > 25. Jy ) with similar colours to OH231.8 + 4.2 and IRAS16342-3814 . A small survey of about 150 sources was started in the last week of June and will be finished at the end of July 1988. Using the radio telescope in Nançay , we observed all four OH maser transitions, with the hunderd meter radio telescope at Effelsberg only 1612 MHz was observed, the sources found at Nançay were confirmed in Bonn and vice versa. From the preliminary analyses of the first 40 observations six sources were found of which three were not known before.

In the outer atmospheres of cool giants and supergiants there is a competition between heating by shock waves which develop from noise lower down in the convection zone and radiative cooling. The post-shock cooling is most effective in keeping temperatures low and in damping shocks if molecules (especially CO) form. CO is important both because of its high dissociation energy (11 eV) and its large IR opacity.

A formula for stellar mass-loss rate has been derived using conservation equations of mass and momentum for coronal and continuous radiation driven wind. The derived mass-loss rate formula has been found to be consistent with the observed mass-loss rates for stars from 0 to M spectral type. Two constant parameters appearing in the mass-loss rate formula have been found to have values for special groups of stars like Be-stars and Wolf-Rayet stars different from each other and from the majority of stars. The effect of mass-loss according to the formula for the majority of stars on stellar evolution has been examined.

Long-period variable stars (LPV's) are considered. Pulsations of the surface of such a star result in formation in the stellar atmosphere of a shock wave in each cycle of the star's variability, ionizing the circumstellar gas which, recombining, gives rise to emissions in optical lines hydrogen and metals. I show the recombining layer behind the shock front to be optically thick in the free-free continuum at the radio wavelengths as short as 1 cm. At the gas temperatures behind the front T = 20000 K, the radio flux density at 1 cm from the front surface at a distance of a few hundred parsees will be several or several tens of mjy. So far, the only positive result of searches for LPVS’ radio continuum is the detection of radio flux from R Aql, obtained at different epochs by several authors. In October 1978 R Aql showed radio emission of 5.3 mjy at 14.9 GHz (Bowers and Kundu, 1979), and in August 1985 - 0.54 mjy at the same frequency (Drake et al.. 1987); these figures are consistent with our model. Besides that, R Aql experienced stronger radio flares at longer wavelengths, up to a few hundred mjy (Woodsworth and Hughes, 1973; Estalella et al.. 1983); these cannot be explained by thermal radio emission and require a nonthermal mechanism (synchrotron or cyclotron maser).

IRAS data for high-galactic-latitude F supergiants, a few luminous F-G stars, and a few peculiar (forbidden) emission-line stars reveal they have dust shells with characteristics similar to those observed in planetary nebulae. The objects described here appear to have experienced severe mass loss in the recent past on the AGB. These stars are most likely post-AGB stars or evolving from the tip of AGB towards the left in the H-R diagram and may be described as possible proto-planetary nebulae. From the IRAS data we find ten additional possible proto-planetary nebulae. From the IRAS data of HD 56126, -53° 5072, HD 168625 and -59° 6723 the luminosities, temperatures and masses of the dust shells are derived. HD 56126 (F5I), HD 168625 (B8Iae), and -59° 6723 (B8Iae) appear to be in an evolutionary stage (post-AGB) similar to that of HD 161796 and HR 4049.

Of the 129 symbiotic stars in Allen's (1984) catalogue, 42 were found to be IRAS sources. Of these 42 IRAS sources, 22 are D-type (symbiotic Miras), 5 are D'-type (yellow symbiotics) and 15 are S-type. The separation of S, D and D’ types into three distinct groups is clearer in the log[fλ(25μm)/fλ(12μm)] versus (H-K) diagram. The IRAS fluxes of S-type symbiotics are consistent with that observed from normal M giants. This result suggests that mass-loss rate from most of the S-type symbiotics is similar to that from normal M giants. The IRAS data of D-type symbiotics shows evidence for the presence of dust envelopes. The masses of the dust envelopes (10-6 to 10-7 Mo) around Miras in D-type symbiotics are similar to that observed in field Mira variables. These results suggest that mass-loss rates in symbiotic Miras are similar to those from field Mira variables and also that the mass loss from symbiotic Miras is pulsationally driven similar to that found in field Mira variables by Whitelock, Pottasch and Feast (1987). Analysis of IRAS data of yellow symbiotics Ml-2, AS201, Cnl-1, Wray 157. and HD149427 suggests that they are young planetary nebulae containing a binary nucleus. Ml-2, AS201 and Cnl-1 show evidence for the presence of evolved hot companions. The evolutionary stage of the late type (F-G) companions is not clear.

Due to the atomic structure and properties of neutral helium, its spectrum can be an excellent diagnostic of the hot components of the atmosphere of a late-type giant star. Having this in mind, we have studied the formation of the helium lines and continua in the K giant star β Cet. For that purpose different NLTE atmospheric models were constructed. We have also solved the NLTE problem for the formation of lines and continua for a model helium atom consisting of twelve bound levels and three ionized stages.

We have measured the excess IR emission from carbon mira and SR variable stars from IRAS LRS spectra. The 8-22μm excess is defined as the ratio of flux above a 2500K energy distribution fit to the LRS spectrum at about 8μm. The carbon star LRS spectra show both emission and absorption features, which are incorporated into our 8-22pm excess. The most prominent feature in carbon stars is the 11.2μm SiC dust emission feature extending from 10μm to 13.8 μm. We observe another emission feature of unknown origin which peaks between 8.4-8.7μm. The SiC emission feature is occasionally blended on the red side by an absorption feature (attributed to gaseous HCN + C2H2) which extends from about 12-16μm. Many of the spectra appear to turn down at the 8μm end due (?) to an HCN + C2H2 absorption feature located at 7.1μm. Carbon stars do not generally show as large an excess as the M mira variables do. The figure below shows our measured excesses for both carbon miras and carbon semi-regular variables. There appears to be little correlation of excess with period, however the mira variables show about twice the range of variation of excess that the semi-regular variables do. We find little correlation between our measured 8-22μm excess and the excesses of Jura (Ap. J., 303, 327, 1986) based on the ratio of 12μm flux to 2μm flux. Our data do support hi s conclusion that longer period variable stars show larger average excesses, but this is only true for mira variables in our analysis.

Observations of RCrB obtained during the 1986 and early part of 1963 light minima show the following changes relative to the spectrum obtained at maximum light.

The molecular band of C2 (Swan System), CN bands at λ4215 and low excitation metal lines are enhanced in absorption.

Since March 1988 we have monitored the 22.2 GHz H20 maser line of 12 long period variable stars (LPV) with the 37 m telescope at the Haystack Observatory. We include the two Carbon stars V778 Cyg and EU And. The maser flux from V778 Cyg has varied by at least a factor of 5 from its detection level of 1.9 Jy on 1987 March with a possible period of a year. Figure 1 plots the 22GHz flux as a function of Julian date. We do not yet know whether the period will repeat. V778 Cyg also has shown variations in the intensity (by a factor of - 2) on a time scale as short as 15 hours. The water maser flux from EU And has decreased from its detection value of 8 Jy in 1986 November staying relatively weak at the 2-3 Jy level during the last year. We interpret both V778 Cyg and EU And as binaries, each with an M star component with a thick circumstellar shell and a C star component. The C star is brighter in the visual region where the system is classified whereas the M star is brighter at wavelengths > 5 μm. The circumstellar shell is the source of the strong silicate emission seen in the IRAS LRS spectra and of the water maser emission.