Search

Due to recent interest in molecular anions as possible interstellar species, we have carried out several laboratory studies of anion chemistry. The reactions of the series C(sub n)(sup -); and C(sub n)H(sup -) with H and H2 were studied to address the viability of such species in the diffuse interstellar medium and to address their ability to be carriers of the diffuse interstellar bands (DIBs). These same molecules were also reacted with N and O to show possible heteroatomic products. C(sub m)N(sup - was a particularly stable product from the reaction of C(sub n)(sup -) + N. C3N(sup -) was further reacted with H to study chemistry that could produce HC3N, a known interstellar species. The reactions were done in a flowing afterglow selected ion flow tube apparatus (FA-SIFT). The anions were generated in an electron impact or cold cathode discharge source and the anion of interest was then selected by a quadrupole mass filter. The selected ion was then reacted with the atomic or molecular species in the flow tube and products were detected by another quadrupole. While the C(sub n)(sup -) species do not appear to be viable DIB carriers, their possible presence could provide a mechanism for the formation of known heteroatomic neutral molecules detected in the interstellar medium (ISM).

Infrared spectral properties of silicate grains in interplanetary dust particles (IDPs) were compared with those of astronomical silicates. The approximately 10-micrometer silicon-oxygen stretch bands of IDPs containing enstatite (MgSiO3), forsterite (Mg2SiO4), and glass with embedded metal and sulfides (GEMS) exhibit fine structure and bandwidths similar to those of solar system comets and some pre-main sequence Herbig Ae/Be stars. Some GEMS exhibit a broad, featureless silicon-oxygen stretch band similar to those observed in interstellar molecular clouds and young stellar objects. These GEMS provide a spectral match to astronomical "amorphous" silicates, one of the fundamental building blocks from which the solar system is presumed to have formed.

The C10H8(+) cation and its dehydrogenated derivatives, C10H7(+) and C10H6(+), have been studied using a selected ion flow tube (SIFT). Reactions with molecules and atoms of interstellar interest show that C10H8(+) reacts with N md O to give neutral products HCN and CO, respectively. C10H6(+) and C10H6(+) are moderately reactive and reactions proceed through association with molecules. The implications of these results for the depletion of C10H(n)(+) in the interstellar medium are briefly discussed.

A profile fitting technique is employed to identify the velocities and Doppler b values for H I and H II clouds along the line of sight to Theta Car. Total abundances and depletions for 12 elements, plus column densities for the J = 0 to J = 5 rotational levels of H2 are obtained. Electron densities for both clouds are calculated from the ratios of the fine-structure levels of C II and N II, obtaining 0.08/cu cm and 1.2/cu cm. The fine-structure levels of C I, which led to 120/cu cm, are used to calculate the neutral hydrogen density for the H I region. D I is also present in the data from the Theta Car line of sight, yielding a D/H ratio of 5 x 10 exp -6. Elemental depletions are calculated for the H I region as well. Comparison of the results for Theta Car and those for Zeta Oph and Alpha Vir shows that the absolute depletions are different; however, the relative depletions are remarkably stable for different physical conditions.

From Copernicus observations of Sigma Sco, 57 individual lines of 11 elements plus the molecular species H2 and CO were identified. By using a profile-fitting technique, rather than curves of growth, it was possible to obtain column densities and Doppler b values for up to four separate components along this line of sight. Electron density in the major H I component was derived from the photoionization equilibrium of sulfur, obtaining, n(e) of about 0.3/cu cm. The neutral hydrogen density in the same component was also derived using fine-structure excitation of O I. An H II component is also present in which the electron density was n(e) about 20/cu cm. As a by-product of this analysis, previously undetermined oscillator strengths for two Mn II lines were obtained: for 1162.-017 A, f about 0.023 and for 1164.211 A, f about 0.0086.

The results of a UV survey of stellar winds and circumstellar shells in 40 B6-B9.5e stars covering luminosity classes V-III are presented. A graph is presented of the region from 1520-1560 A, which includes both the Si II UV multiplet 2 transitions and the C IV resonance transition for selected Be stars in the sample. The detection of shortward-shifted discrete component absorption features in nine of the program stars suggests that the material is produced in a stellar wind, which attains velocities of at least a few hundred km/s.

The paper presents the results of an ultraviolet survey of stellar winds in 62 Be and 43 normal B stars covering spectral types B0.5-B5 and luminosity classes V-III. Excess wind absorption in Be stars over that detected in normal B stars is found in C IV and Si IV, confirming results of previous surveys. The stars showing excess wind absorption also show one or more shortward-shifted discrete absorption components, suggesting that much of the excess wind absorption is in the form of discrete absorption components. Evidence is presented that the high-velocity and highly ionized stellar wind observed in some Be stars is a function of latitude.

Since the launch of Copernicus in 1972, studies have been made of the depletion of gas-phase elements onto dust grains. A few stars have been studied in detail, resulting in a standard depletion pattern which has since been used for comparison. Recent developments, however, have suggested that this standard pattern may need to be re-examined. Some weak, semi-forbidden lines were detected recently which may be able to resolve some of the ambiguities. Studies of single elements have shown that depletion of carbon and oxgyen are much smaller than previously determined. The high resolution ultraviolet spectral scans of sigma Sco were originally made in 1973, but have only recently been analyzed. All these stars are bright and moderately reddened. All four stars will be analyzed in detail, but sigma Sco is the first one completed. The data has broad coverage of ions, making these stars excellent candidates for determination of accurate depletions. A profile-fitting analysis was used rather than curves-of-growth in order to determine separate abundances and depletions in components separated by several km/sec.

Using International Ultraviolet (IUS) archival low-dispersion spectra, ultraviolet spectral extinctions were derived for the circumstellar envelopes of two M supergiants: HD 60414 and HD 213310. The observed stellar systems belong to a class of widely-separated spectroscopic binaries that are called VV Cephei stars. The total extinction was calculated by dividing the reddened fluxes with unreddened comparison fluxes of similar stars (g B2.5 for HD 213310 and a normalized s+B3 for HD 60414) from the reference atlas. After substracting the interstellar extinctions, which were estimated from the E(B-V) reddening of nearby stars, the resultant circumstellar extinctions were normalized at about 3.5 inverse microns. Not only is the 2175 A extinction bump absent in the circumstellar extinctions, but the far-ultraviolet extinction rise is also absent. The rather flat, ultraviolet extinction curves were interpreted as signatures of a population of noncarbonaceous, oxygen-rich grains with diameters larger than the longest observed wavelength.

New observations of lines from neutral atoms in the lines of sight to three stars in the Rho Oph cloud are combined with literature data to look for evidence of density-enhanced regions with higher elemental depletions than those characterizing the cloud as a whole. The depletions inferred from neutral species, however, are comparable to those measured directly from observations of ions, suggesting that no strong density enhancements exist in the observed lines of sight. This conclusion is supported by electron densities derived from the observed ionization fractions, which also show no evidence for a strong variation of density with depth in the cloud. It appears that the overall density (and overall level of depletion) in the cloud is higher than in typical diffuse clouds, despite the apparent lack of strong condensations with enhanced relative depletions.

There are few aspects of interstellar grains that can be unambiguously defined. Very little can be said that is independent of models or presuppositions; hence issues are raised and questions categorized, rather than providing definitive answers. The questions are issues fall into three general areas; the general physical and chemical nature of the grains; the processes by which they are formed and destroyed; and future observational approaches.

A tentative synthesis of visual and IUE observations of 59 Cyg during 1978-1983 when the new Be phase began to develop is presented. The study reveals correlation between those spectral features originating in the cool H-alpha-emitting envelope and features originating in superionized regions exhibiting C IV and N V lines. The long-term variability patterns of the V/R ratio, the H-alpha emission, and the C IV absorption reveal striking similarities and a close correlation. The early development of the new Be phase is accompanied by remarkably large and striking C IV line-profile changes. It is also suggested that short-term V/R variations at H-alpha may be associated with short-term C IV changes.

Calcium depletions toward five moderately reddened (E/B - V/ greater than 0.35 and less than 0.45) stars and toward HD 147889, a heavily reddened (E/B - V/ = 1.08) member of the Rho Ophiuchi cloud complex is reported. A developmental Multi-Anode Microchannel Array Detector was used to collect the data; its performance is characterized. Comparing the results for calcium from one line of sight to another and to those for other species, it is found that calcium exhibits a depletion pattern remarkably similar to that of silicon. The depletion of iron, on the other hand, has smaller variations and therefore does not correlate as well. Also, the depletion results for manganese and sulfur, both normally undepleted in diffuse clouds, have been included for comparison.

The International Ultraviolet Explorer (IUE) Satellite was used to collect interstellar extinction curves for 58 moderately reddened lines of sight. Spectral types of the background stars ranged from O7 to A0. All luminosity classes are represented, but early B main-sequence stars predominate. E(B-V) reddening values ranged from 0.14 to 1.13 mag, with an average of 0.63 mag. These extinction curves, less some 10 curves judged of lower quality, were co-added in order to reduce the random and fixed-pattern noise of the instrument. The resulting averaged curve was examined for evidence of small-scale structure, with negative results. The 2-sigma limits were found to be 0.09 mag for wavelengths between 1150 A and 2000 A, and 0.16 mag for wavelengths between 2000 A and 2300 A. Because of mismatch errors in weak stellar lines, it is unlikely that much improvement in these results is possible. The implications of these negative results are discussed with particular reference to the MgO theory of interstellar grains and diffuse bands.

An IUE satellite search is reported for chromospheric emission in the UV spectra of the brightest pre-main sequence stars the open cluster NGC 2264. Einstein X-ray observations to search for variability of the bright O7 V ((f)) star S Mon = W131, which may be the most massive member of the cluster, are also reported. Fifteen cluster members of spectral type B-K, ranging in age from a million to ten million yr, were observed in the ultraviolet. The only star exhibiting chromospheric emission is W79, a faint, heavily veiled T Tauri star that lies below the main sequence of NGC 2264. The observations are consistent with the generally accepted idea that chromospheric activity of late-type stars subsides with age.

The effect of a low-velocity shock on depletion in interstellar clouds is studied. High-resolution Copernicus observations of interstellar absorption lines toward four stars in the Rho Ophiuchi cloud complex were used to measure differential depletion in interstellar clouds separated by velocities of 10-15 km/s. Optical observations of CH and CH(+) were used as indicators of shock strength and direction of propagation. Observations indicate the presence of a shock with velocity about 10 km/s expanding away from the sun into the Rho Oph cloud. The existence of this shock is supported by the other observations quoted in the literature. Two distinct regions in the lines of sight have been found. A low-density, less depleted, predominantly atomic region is associated with preshock gas, while postshock gas accounts for a predominantly molecular, more highly depleted region. Apparently a weak shock, has the effect of enhancing grain formation or grain growth as a result of increased density in the postshock gas. It is possible for grain growth by accretion to occur in the shocked gas on time scales short compared with the cloud-crossing time of the shock.

Archival IUE data on the deeply embedded Rho Oph cloud star HD 147889 have been used to derive column densities for several species. The poor data quality and limited useful spectral coverage made unambiguous abundances very difficult to deduce, but firm limits could be set. Absolute depletions have been estimated, based on an assumed value of the total hydrogen column density, but they are uncertain due to uncertainty in N(H). The data appear to indicate that depletions are generally very large in this line of sight. There is also some indication that the relative depletions are enhanced in the densest portions of the cloud.

Data on neutral atomic species are analyzed for the purpose of inferring relative elemental abundances and depletions in diffuse cloud cores, where it is assumed that densities are enhanced in comparison with mean densities over integrated lines of sight. Column densities of neutral atoms are compared to yield relative column densities of singly ionized species, which are assumed dominant in cloud cores. This paper incorporates a survey of literature data on neutral atomic abundances with the result that no systematic enhancement in the depletions of calcium or iron in cloud cores is found, except for zeta Ophiuchi. This may imply that depletions are not influenced by density, but other data argue against this interpretation. It is concluded either that in general all elements are depleted together in dense regions so that their relative abundances remain constant, or that typical diffuse clouds do not have significant cores, but instead are reasonably homogeneous. The data show a probable correlation between cloud-core depletion and hydrogen-molecular fraction, supporting the assumption that overall depletions are a function of density.

The results of a monitoring program with IUE and Einstein are presented for three stars, delta-Ori A, l-Ori A, and 15 Mon. Line profile variability is observed in the UV profiles accessible to IUE and the relation between the variation in the different ions suggests that the ionization level is varying in the winds of these stars. This is consistent with Einstein observations of soft X-ray variability for two of the stars.

Correlations between UV extinction parameters and diffuse band strengths are south, using a data base of 50 stars with recently measured extinction curves. A novel aspect of this study is that the basic dependence of most interstellar parameters on E(B-V) (or, nearly equivalently, on the total hydrogen column density) has been largely canceled out. After this cancellation, a weak correlation is found between the 4430 A diffuse band and the 2175 A extinction feature. No other correlations are found. These results support a molecular origin for the 5780 A and 6284 A bands. It is speculated that the correlation pattern found for the 4430 A band can be explained if this band originates in reactions on disordered grain surfaces.

Results are reported from the combined CfA Stellar Survey of selected bright A stars and an Einstein Guest Observer program for Ap and Am stars. In an initial report of results from the CfA Stellar Surveys by Vaiana et al. (1981) it was noted that the spread in observed X-ray luminosities among the few A stars observed was quite large. The reasons for this large spread was studied by Pallavicini et al. (1981). It was found that the X-ray emission from normal stars is related very strongly to bolometric luminosity for early-type stars and to rotation rate for late-type stars. However, an exception to this rule has been the apparently anomalous behavior of A star X-ray emission, for which the large spread in luminosity showed no apparent correlation with either bolometric luminosity or stellar rotation rate. In the present study, it is shown that the level of emission from normal A stars agrees with the correlation observed for O and B stars.

Nightly variations in He II 4686 A have been observed in four Of stars, HD 152408, HD 151804, 9 Sge, and Lambda Cep. Three of the stars were not previously known to show such variations. In one case, Lambda Cep, simultaneous observations with IUE showed variations in ultraviolet (UV) resonance-line profiles during the same time period as those in He II 4686 A. This implies that the perturbations causing the He II 4686 A variability affect the entire wind acceleration zone.

It has been found for the Rho Ophiuchi cloud, and suggested for diffuse cloud cores in general, that H2 has a lower formation rate than normally expected in diffuse clouds. It is proposed that this may be due to a reduction in grain surface area per unit volume, if the grains in these regions are enlarged due to coagulation. For the Rho Oph cloud, there is independent evidence that grain coagulation has occurred, so this hypothesis regarding H2 formation has a more substantial basis than for other clouds, where the evidence for suppressed H2 formation is less certain, and for which there is little information on grain sizes.

Observational technology is now reaching the point where optical absorption-line studies of dark interstellar clouds are becoming feasible. Such studies are of great interest. In particular, it will be important to discover the degree of depletion which has occurred in dark clouds. In order to assess depletions on the basis of observations of a limited range of ionization stages, it is necessary to analyze the ionization balance. The present investigation is concerned with such an analysis. The two key parameters necessary for deducing the ionization balance are the electron density and the radiation field intensity. A summary is provided regarding the existing data on the cloud density using data from the literature, and the radiation field intensity is derived from the observed H2 rotational excitation. The implications of both the density and radiation field data for the ionization balance of several species are discussed. It is found that for all species treated the first ion is dominant in the line of sight toward HD 147889.

A telescope-spectrometer on the Copernicus spacecraft made possible the measurement of many ultraviolet absorption lines produced by the interstellar gas. The present survey provides data on ultraviolet absorption lines in the spectra of 88 early-type stars. The stars observed are divided into four classes, including reddened stars, unreddened bright stars, moderately reddened bright stars, and unreddened and moderately reddened faint stars. Data are presented for equivalent width, W, radial velocity V, and rms line width, D, taking into account some 10 to 20 lines of N I, O I, Si II, P II, S II, Cl I, Cl II, Mn II, Fe II, Ni II, Cu II, and H2. The data are based on multiple scans for each line. Attention is given to details of observations, the data reduction procedure, and the computation of equivalent width, mean velocity, and velocity dispersion.

Neutral nitrogen and oxygen column densities or limits are derived for 53 path lengths through the diffuse interstellar medium and compared with column densities of neutral hydrogen. For neither N nor O is a systematic increase of depletion found as reddening increases. The value found for forbidden O/N is 8, and the abundances of N and O are both between 40% and 70% of the solar values. Implications of these results for models of interstellar grains are briefly discussed.

Data on the interstellar Ca II H and K lines toward HD 147889, a star deeply embedded in the rho Ophiuchi dark cloud, were obtained at the Mauna Kea Observatory of the University of Hawaii, using a multi-anode microchannel array (MAMA) detector. An upper limit on the Ca II column density of 1.7 x 10 to the 11th/sq cm was derived, which, together with ionization considerations and an estimated lower limit on the total hydrogen column density, implies that the calcium-to-hydrogen ratio in this line of sight is 0.00000795 or less, relative to the solar value. The logarithmic depletion is up to -5.10, the largest yet measured, and implies that gas-grain interaction in this region has proceeded to a much greater degree than is typical of diffuse clouds.

An intensive search for OH and H2O in the directions of sigma Sco, alpha Cam, and micron Per was undertaken with the Copernicus satellite. Multiple scans were carried out over the wavelength region for the expected absorption features due to the OH D-X and H2O C-X transitions. The feature due to OH was detected marginally towards sigma Sco, and only an upper limit can be given towards alpha Cam. H2O was not detected in any of the stars at the signal level accumulated. The OH abundance towards sigma Sco and the respective lower limits for the OH/H2O ratios are discussed with regard to the extant models for the steady state abundances of OH and H2O, and shown not to be inconsistent with ion-molecule schemes.

Twenty stars classified as either Ap or Am have been imaged by the Imaging Proportional Counter on the Einstein Observatory in a search for evidence of coronal emission. It is found that an Am star will not be a source (at the 10 to the 28th ergs/sec level) unless it is in a spectroscopic binary with a period below about 10 days. The X-rays in the close binaries can be explained by the RS CVn effect in the late-type secondary, and the evidence is against strong emission from Am stars. Among the nine Ap stars observed, four were detected. While it is possible that the X-rays are being generated by undetected late-type dwarf companions, the large fraction of detections makes the Ap stars themselves the likely source of emission.

(Previously announced in STAR as N82-30203)

The presence of narrow, shifted absorption features superposed on the wide P Cygni profiles of the UV resonance lines in the spectra of O and B stars is studied and possible explanations of the narrow components are discussed. Spectra from 26 stars of spectral types O4f to B1 Ib were examined for the presence of narrow components in the absorption part of the profiles, and positive evidence was detected in 17. The central velocities of the features are found to be similar for different ions of the same star, and typically equal to 0.75 times the terminal velocity of the stellar wind. The width of the narrow absorptions is about 18% of the terminal stellar wind velocity, with a column density generally not correlated with the effective temperature or mass loss rate. Possible explanations for the narrow components include peaks in the degree of ionization, a stationary stellar shell, a plateau in the stellar wind velocity law, a decelerating stellar wind, a variable mass loss rate or a two component stellar wind with low-velocity, low-density material found at a distance greater than 2 stellar radii, which explanation appears most likely.

Mass loss rates for a number of early-to-mid-B main-sequence stars are compared with extrapolations of recently derived empirical correlations between M and stellar parameters for luminous OB stars. The order of magnitude of the rates for the B stars is consistent with a simple dependence on luminosity over the range from Of to mid-B, and is not consistent with previously derived correlations involving a dependence on mass or radius, which implies that luminosity is the dominant factor in driving the mass loss throughout the range of spectral types from mid-B main-sequence to Of. The winds in Be stars may represent a straightforward extension of the OB wind phenomenon to lower luminosities, and there appears to be no significant difference between B and Be stars in terms of their wind characteristics.

The International ultraviolet Explorer high dispersion spectra of 59 Cyg obtained as part of the long term monitoring program have shown that noticeable variation can occur in C 5 and N 5 on timescales 3 hours t24 to 28 hours. In order to begin to resolve whether these changes occur continuously or sporadically, 48 hours were devoted to monitoring this star in January 1982. The January spectra show no short term variation, which may be consistent with sporadic rather than continuous variation.

High signal-to-noise profiles of the broad diffuse interstellar band at 4430 A were obtained on the 2.2-m telescope at the Mauna Kea Observatory, using the newly-developed pulse-counting multi-anode microchannel array detector system in an effort to determine whether the band profile varies with mean grain size as expected if the band is produced by absorbers embedded in grain lattices. The lack of profile variability over several lines of sight where independent evidence indicates that the mean grain size varies shows that lambda 4430 is probably not formed by the same grains that are responsible for interstellar extinction at visible wavelengths. The possibility that this band is created by a population of very small ( approximately 100 A) grains is still viable, as is the hypothesis that it has a molecular origin.

Several types of ultraviolet observations of Be stars are briefly summarized, with a view towards identifying the characteristics, if any, that distinguish Be stars from other early type objects. The present weight of the evidence is that, from the point of view of ultraviolet data, Be stars are nearly indistinguishable from normal B stars. Both groups have similar characteristics, with differences only in degree. Further comparative studies of Be and nonBe stars have the potential to reveal the causes of the Be phenomenon, and possibly also the origins of winds in early type stars. Several areas for future ultraviolet research are suggested.

An analysis for the effects of stellar winds concerning Copernicus scans of Si III and Si IV resonance lines in the spectra of 22 early B stars, of which 19 have Be characteristics, yields line asymmetries in all but one of the stars to which theoretical profiles were fitted. The resulting wind parameters were used to calculate mass-loss rates. Most of the silicon in the winds of the cooler stars is in the form of Si III and Si IV, leading to a mass loss rate value ranging from 10 to the -11th to 3 x 10 to the -9th solar masses/year. Equatorial gravity darkening is seen in the relative velocity widths of ultraviolet and visible photospheric lines. The measured photospheric Si III and Si IV resonance-line equivalent widths do not agree with those predicted by current stellar atmosphere calculations. Be star mass loss rates are probably insufficient to affect either their own evolution or that of the galaxy, since their winds do not significantly contribute to the enrichment of the interstellar medium.

Surveys of abundances for species well suited to IUE's wavelength coverage and spectral resolution are undertaken to acquire interstellar line data for a large number of stars and column densities greater than those of previous instruments. A complex set of interactive computer programs has been developed for the efficient analysis of the IUE data, which is described along with results on Si II column densities toward two moderately reddened stars. Silicon depletions are inferred in the cases of the two stars, HD 42088 and HD 151804, and their H I column density determinations are used to estimate the total hydrogen column density.

Intensive Copernicus scans of the 1(11)-0(00) line of the F-X band of H2O were carried out, resulting in a 2 sigma upper limit of 1.04 mA for this line at 1114.225 A. The recently published f-value implies a column density limit of 5.3 x 10 to the 12th/sq cm, marginally above the level expected from recent chemical model calculations. A 1.8 sigma feature appears at the expected position of the H2O line, implying a column density for the ground rotational state of 2.8 x 10 to the 12th/sq cm.

The 6196.0 A diffuse feature is observed in various directions towards moderately and slightly reddened stars at high spectral resolution and signal to noise ratio. The feature is found to be symmetric with no fine structure with an equivalent-width upper unit paralleling that of H2. The feature is as narrow as 0.4 A, although much broader in other stars due to cloud blending in the line of sight, and is absent in all observed stars with E(B-V) less than 0.1, and upper limits more than an order of magnitude below values extrapolated from more reddened stars. Upper limits are determined by multiplying the square root of the number of channels in which the line may lie, and another narrow feature, although highly asymmetric, is observed at 6194.6 A. It is concluded that the results increase the possibility of at least one of the diffuse band carriers being a molecule.

Ultraviolet line profiles are fitted with theoretical line profiles in the cases of 25 stars covering a spectral type range from O4 to B1, including all luminosity classes. Ion column densities are compared for the determination of wind ionization, and it is found that the O VI/N V ratio is dependent on the mean density of the wind and not on effective temperature value, while the Si IV/N V ratio is temperature-dependent. The column densities are used to derive a mass-loss rate parameter that is empirically correlated against the mass-loss rate by means of standard stars with well-determined rates from IR or radio data. The empirical mass-loss rates obtained are compared with those derived by others and found to vary by as much as a factor of 10, which is shown to be due to uncertainties or errors in the ionization fractions of models used for wind ionization balance prediction.

The resonance line of F I at 954.825 A is detected in the interstellar spectra of Delta Sco, and possibly Delta Ori, and upper limits are obtained toward five other stars. It is noted that the equivalent width for Delta Sco of 10.9(+4.5, -3.1) in milliangstroms corresponds to a column density of 1.52(+0.60, -0.43) x 10 to the 13th/sq cm and a depletion of -0.53(+0.15, -0.14) dex. For the other lines of sight, depletions of at least 0.1-0.5 dex are inferred. The results are discussed in the context of current theories of the depletion process.

Si II oscillator strengths for eight resonance lines from the ground state have been determined using a curve-of-growth analysis of interstellar data from the Copernicus and International Ultraviolet Explorer satellites. The oscillator strengths for 1526 A and 1808 A increase by factors of 3.0 and 1.5, respectively, from previously tabulated values; those for 989 A and 1020 A decrease by factors of 0.55 and 0.60, respectively. These new values result in better silicon abundances for interstellar clouds and resolve a long-standing puzzle in the interpretation of quasar absorption line data.

Data from the International Ultraviolet Explorer (IUE) have been used to derive extinction curves for several stars chosen because of unusual diffuse band strength or location within dense clouds. Both the strength and position of the 2200 A extinction maximum were assessed, as well as the level of extinction in the far-ultraviolet. A variety of combinations of these features were found, ranging from near or complete absence of a 2200 A bump (HD 29647), to low far-ultraviolet extinction (Rho Oph), to unusually steep, far-ultraviolet extinction (HD 169454). Anomalous extinction apparently shows up frequently in stars associated with nebulosity or dense clouds. The apparent independence of 2200 A and far-ultraviolet extinction supports earlier suggestions that these two features are created by distinct populations of grains.

New ultraviolet colors are presented for 110 hot stars. These data are combined with infrared colors and diffuse-band measurements to study the relationship of diffuse interstellar bands (4430, 5780, 6284 A) to the overall extinction curve. Equivalent widths of 5780 A and 6284 A are not well correlated with infrared, visible, or ultraviolet extinction measurements for stars in this sample. The central depth of 4430 A is well correlated with visible and infrared extinction, but less well correlated with UV extinction at 1800 A. The wavelength 4430 A is strongly correlated with the strength of the 2200-A bump. The data suggest that if small grains account for the general rise in UV extinction, the diffuse bands are not formed in these grains. The wavelength 4430 A may well arise in large grains and/or in the material responsible for the 2200-A bump. Correlations with UV extinctions derived by other authors are discussed in detail. It is suggested that definitions of extinction parameters and band shapes, as well as selection effects in small samples of stars, may still compromise conclusions based on correlation studies such as are being attempted.

Seven O stars known to have strong, and sometimes variable, stellar winds have been observed repeatedly with the Imaging Proportional Counter on the Einstein Observatory, in a program designed to determine whether the X-ray fluxes from these stars are variable. In three cases, definite changes were seen, either on a time scale of a year (Iota Ori and Delta Ori) or five days (15 Mon). In two of these cases, the X-ray spectrum was harder when the overall flux was higher, indicating that some of the fluctuations may take place in a hot (approximately 10 to the 7th K) emitting region at the bottom of the winds.

Analysis of the silicon interstellar abundances was made for more heavily reddened lines of sight than were accessible to the Copernicus satellite. Silicon rarely had accurate column densities determined from Copernicus data because the available lines all lie on the flat portion of the curve of growth. With International Ultraviolet Explorer (IUE) it is possible to reach color excesses of E(B-V) approximately 0.5-0.7, and in addition obtain data on the weak SiII line at 1808 A, so that a wide range of oscillator strengths is available. The lower resolving power of the IUE causes difficulties in that several of the SiII lines are blended with strong lines of other species. Data on the lines of sight analyzed suggested that some of the absorption lines fall on the damped portion of the curve of growth, implying that silicon may not be as highly depleted as expected.

Observations of variability in stellar winds or envelopes provide an important probe of their dynamics. For this purpose a number of O, B, Be, and Wolf-Rayet stars were repeatedly observed with the IUE satellite in high resolution mode. In the course of analysis, instrumental and data handling effects were found to introduce spurious variability in many of the spectra. software was developed to partially compensate for these effects, but limitations remain on the type of variability that can be identified from IUE spectra. With these contraints, preliminary results of multiple observations of two OB stars, one Wolf-Rayet star, and a Be star are discussed.

Data from IUE have been used to derive an ultraviolet extinction curve for HD 29647, a B star embedded in the Taurus dark cloud complex. The curve appears normal, except that the 2200 A extinction bump is absent. Arguments are developed to show that this is most likely due to a modification of the grain optical properties through the accretion of mantles in this dense cloud. If this conclusion is correct, it will have some impact on models for the formation of the 2200 A bump.