Mean temperatures for RR Lyrae stars in 7 globular clusters (M3, M4, M5, M15, M107, ω Cen and NGC 5466) have been determined using the optical-infrared colour <V>-<K> as a temperature indicator. Where <K> has been relatively well determined, from means of 3 or more observations, the scatter in relationships such as Log P’ vs log (temperature) and log (temperature) vs (blue amplitude) is significantly reduced when IR-derived temperatures are used instead of those derived from (B-V). Within the observational errors, the gradient in the log P’ vs log (temperature) diagram is the same for each cluster. Temperatures derived from <V>-<K> should also be less sensitive to metallicity differences than their optically derived counterparts. The Sandage Period-Shift Effect has therefore been re-examined using 6 of the 7 clusters (NGC 5466 was excluded because of too few data). A strong correlation between period-shift and metallicity is found; a smaller shift (but in the same sense) is also found for the temperature – amplitude relationship.

UBVRI photometry and radial velocity measurements of EU Tau obtained during three weeks in November/December 1987 have been used to (1) confirm the period, P = 2.1025 days, and epoch established by Fernie (1987, P.A.S.P.,99, 1093) and earlier observers, and (2) investigate apparent phase and/or amplitude discrepancies in the light curve first reported by Gieren and Matthews (1987, A.J.,94, 431). Comparison of these new data with photometry collected by different observers covering over two decades suggests that the light curve of EU Tau undergoes short-lived (less than about 2 pulsation cycles) changes which appear to occur with a period of roughly 19 – 21 days. Various possible sources of modulation – including binarity, double-mode beating and nonradial pulsations – are considered and rejected as unsatisfactory.

During June 17-27, 1987 observers in Mexico, Chile, Italy, South Africa, Australia and the United States monitored the light output of the prototype variable star, δ Scuti. The goal of the network was to obtain this star’s frequency spectrum. During the campaign, over 17,000 observations were obtained in the Stromgren b and y filters. A preliminary analysis has been performed and the frequencies listed in Table 1 have been tentatively identified.

We have made precise radial velocity observations of α Boo (K1 III) which raise the possibility that giant stars redward of the Cepheid instability strip can pulsate. Our first measurements, made in early 1986 (Smith, McMillan and Merline 1987, Ap.J.,317, L79) revealed a systematic, apparently periodic, variation in radial velocity having a “sawtooth” waveform with a full amplitude of 160 m·sec−1 and period of 1.84 or 2.18 days. This discovery was qualitatively confirmed by Cochran (1987, B.A.A.S.,19, 1052). Additional data show that large night-to-night velocity variations of magnitude about 80 m·sec−1 persist over a 192 day baseline. A sharp velocity transition has been monitored on at least one night, which covers a range of 100 m·sec−1 with an acceleration of 19m·sec−1·hr−1. The positive slope during this transition argues for a 2.18 day rather than the 1.84 day period. Furthermore, the amplitude and shape of the variation may be changing, perhaps suggesting multiple periods. To test the likelihood that radial pulsation could be responsible for the observed variations, stellar pulsation models are being conducted with Dr. Arthur Cox at Los Alamos National Laboratory. Using a structure model based on our best estimates for the stellar parameters, we have found that the 2 day period would require the third or fourth radial overtone, which are driven (unstable) with relatively high growth rates.

We increase the metal abundance in the envelope of Double Mode Cepheids (DMCs) in order to remove the period ratio mass discrepancy. By increasing the value of Z, we are able to increase the opacity by a factor of 2 to 2.5. We find that for a star of 5 solar masses, with a temperature of 5957 K, and a luminosity of 1050 times that of the sun, an increase in the metal abundance starting at the Log (T) value of 4.79 and reaching a maximum Z value of 0.3 in the Log (T) range of 4.8 to 5.3 produces a period ratio P1/P0 = 0.7089, and a value of P0 = 3.001 days. The value of Z returns to 0.02 at Log (T) = 5.62. The enhanced zone (part of which convects) is located well below the H and He convective regions. The enhanced region is also stable against downward mixing since the radiative gradient is so sub-adiabatic.

The 2:1 resonance between the fundamental and the second overtone modes has received a great deal of attention in the context of Cepheids. It was clearly shown that it causes the Hertzsprung bump progression and brings about the very characteristic observed variation of the Fourier phases with period (Buchler & Goupil, 1984, Ap.J.,279, 394; Klapp, Goupil & Buchler, 1985, Ap.J.,296, 514; Buchler & Kovacs, 1986, Ap.J.,303, 749).

Classification of pulsating variables is based on a number of properties but the most important of them are the period and amplitude of the light variation. These quantities also play an important role in understanding the evolutionary status of the pulsating stars since the pulsation period is related to the mass and radius through the period--mean density relation while the amplitude of the light curve characterizes the efficiency of the mechanism responsible for the pulsational instability. In the present study we considered the period--frequency and light amplitude--frequency distributions for ten types of the most numerous pulsating variables from the fourth edition of the General Catalogue of Variable Stars. The following types of pulsating variables were considered: DSCT, DSCTC, DCEP, RR, RRAB, RRC, CWA, CWB, RV and M. Using these distributions we estimated the upper and lower limits of the period and light amplitude within which 95% and 99% of the pulsating stars of a given type are contained.

Nonlinear models of RR Lyrae variables were calculated in the region of the HR diagram containing double mode pulsation. In all models Teff = 7000 K, the luminosity was 60 times solar, X = 0.70, Y = 0.299, and P1/P0 = 0.746, characteristic of double mode RR Lyrae variables in M15 (Cox, Hodson, and Clancy, 1983). Models of 0.65 solar masses were calculated using standard opacities and models of 0.75 solar masses were calculated with opacities artificially enhanced by a factor of 1.2 between log(T) = 5.2 and log(T) = 5.9 (Andreasen and Petersen, 1988).

A derivation of Eötvös spheres and their pulsations is presented. These spheres incorporate a short range force in their gravitational potential. We used a Yukawa formulation to agree with other workers in this field, although a slightly different form would guarantee a gravitational force that is always attractive. Various relationships are obeyed by these obejcts, much like the polytropes to which they are similar. An attempt is made to test whether a short range force can resolve two outstanding problems in astrophysics. These are the Cepheid mass discrepancy and the frequencies of high-order solar p-modes. The disagreement in predicted and observed Cepheid masses cannot be explained by a repulsive force, but the solar p-modes have their asympotoic frequency decreased. Linear fits to sequences of models are presented for the radial eigenfrequencies and asymptotic frequencies for nonradial pulsations.

RR Lyraes that have nearby visual companions may be examined to decide if they do indeed belong to a real binary system. Identification of RR Lyraes in binary systems could lead to the determination of physical characteristics of the variable and to an understanding of the history and evolution of such systems in the Galaxy. Where possible, proper motion analysis, using astrograph plates from the Lick Proper Motion Program, for some visual binaries has been done and has enabled the rejection of some variables from consideration. The variables which have been shown not to be physically associated with their visual companions are YZ Cap, V445 Oph, BH Peg, and TX Vir. EZ Lyr, RW Ari, and V494 Sco still remain possible cases requiring relative radial velocities between the variable and companion. Limited plate material has meant that one must resort to radial velocity measurements for the RR Lyrae and close companion in order to determine the nature of the pair.

A new orbit has been derived for the binary Cepheid S Sagittae = HD 188727 = BD+16°4067, using both extant data and new radial velocities derived from radial velocity spectrometer observations at McDonald Observatory and the Dominion Astrophysical Observatory. We deconvoled both the pulsational and orbital components from the observed velocities in order to fit a truncated Fourier series to the pulsation velocities and to derive orbital elements respectively. We confirm the orbital elements given in Herbig and Moore (1952) and speculate that the secular trend evident in the orbital (O-C) diagram may be interpreted as evidence for a third component. (See Evans, Slovak and Welch, this volume, for IUE observations supporting this interpretation.)

BVR observations of the relatively metal-rich globular cluster NGC 6388 have been obtained with a CCD on the CTIO 0.9 m telescope. Eighteen possible short period variable stars have been discovered in or near the cluster. At least 10 of these are probable RR Lyrae members of NGC 6388. We confirm the finding of Hazen and Hesser that this cluster is one of the most metal-rich to contain a significant number of RR Lyraes. A program of CCD photometry of field and cluster variable stars has been initiated on the 0.6m telescope of the Michigan State University Observatory.

We have carried out high dispersion, long wavelength IUE observations and ground based photometry of the δ Scuti star, β Cas. Our ground based observations were used, together with the previous results of Antonello et al. (1986, I.B.V.S. 2958, 1986), to ensure that the IUE observations were correctly phased relative to the photometric variation. Fluxes for the emission core of the Mg II k line (2796 Å) were obtained from 23 ultraviolet spectra taken over several cycles in 1986 and 1987. The emission flux, if present, was measured with respect to the mean line profile of all of the spectra. Emission was present at phases between 0.4 and 0.5. This is in contrast to what has been observed for another δ Scuti star, ρ Pup, and for the classical Cepheids, where the emission appeared at maximum light.

Attention is called to the rather unusual distribution of the periods of the RR Lyrae variables in NGC 5897, a metal-poor halo globular cluster with a very low central concentration. Of the seven RR Lyrae stars known in the cluster, three have periods between 0.797 and 0.856 day and two have periods of 0.45 and 0.42 day. The other two have periods of 0.34 and 0.35 day with much lower amplitudes of variation. Due to the lack of crowding in this cluster photoelectric observations and Fourier decompositions of the resulting light curves should be possible for at least six of the RR Lyrae variables. In addition, the cluster appears to contain a non-variable horizontal branch star, SK 120, lying within the instability strip. As this is the only well documented case of such a star, photoelectric observations of this star would also be desirable.

The group of pulsating CP2-stars (also called “rapidly oscillating Ap stars” provides asteroseismology with oscillation spectra of high accuracy. The potential as a diagnostic tool for modelling stellar interiors is widely appreciated. The identification of pulsation modes is important for such an analysis. However, this is rarely possible in an unambiguous manner. To improve the situation and to make use of additional information, we observed HD 128898 simultaneously spectroscopically and photometrically at ESO. For each of our individual CAT-CES spectra (1 minute integration time) it was thus possible to determine the pulsation phase at mid-exposure. A total of 887 spectra (R = 50000) were binned according to their pulsation phase and coadded to improve significantly the signal to noise ratio.