Table of contents

In an Ω₀ = 1 universe, within the classical Eulerian theory of gravitational instability, the redshift evolution of a peculiar velocity field in a region with arbitrary initial density contrast is derived, for the first time, in real space and third-order perturbation theory. A vector proportional to the gravitational acceleration can also be expanded in terms of the redshift and the initial density contrast. The results are applied to isolated (spherically symmetric) superclusters and voids. Using reasonable models and the exact solution, we tested the accuracy of three (linear, second order, and third order) approaches. A numerical example showed that the relative error of the third-order solutions (average density contrast and peculiar velocity) is less than 5% when 0 < δ ≲ 1. In another example, a relative error was derived (at -1 < δ < 0) of less than 10% (average density contrast) to 2% (peculiar velocity). On the other hand, second-order environmental dynamical terms (supercluster-supercluster, supercluster-void, and void-void complexes) have been also obtained. In the complexes (which contain two large-scale structures with spherical symmetry at recombination), the global peculiar flow can be described as a natural (but not trivial) superposition of two effective peculiar flows. Given a member of a complex, its effective peculiar velocity field is the sum of a spherically symmetric radial field (which is equal to the peculiar velocity field obtained from an isolated evolution) and an environmental (due to the interaction with the companion) field. In general, the external tides can be comparable to the internal ones. The imprint of the environment fields in the mean radial effective peculiar flows is also studied.

We have laid down the formalism and techniques necessary for computing the multipole components in a spherical harmonic expansion for bursting sources given any specific power spectrum of density perturbations. Using this formalism we have explicitly computed and tabulated the expected first few multipole components for the γ-ray burst (GRB) distribution using a cold dark matter power spectrum. Unfortunately, our analysis leads us to expect that an anisotropy signal for this model of structure formation will be below the shot-noise level for the foreseeable future. Earlier studies by others had claimed that it may be possible within a decade or so to probe structure formation using the angular distribution of GRBs. We find that while it may be possible to probe the dipole due to our motion with respect to the cosmic rest frame, it does not seem feasible to probe the multipole components due to intrinsic fluctuations if the actual power spectrum is fixed to within an order of magnitude by the cold dark matter model of structure formation.

Detections of deuterium in high-redshift Lyman limit absorption systems along the line of sight to QSOs promise to reveal the primordial deuterium abundance. At present, the deuterium abundances (D/H) derived from the very few systems observed are significantly discordant. Assuming the validity of all the data, if this discordance does not reflect intrinsic primordial inhomogeneity, then it must arise from processes operating after the primordial nucleosynthesis epoch. We consider processes that might lead to significant deuterium production or destruction and yet allow the cloud to mimic a chemically unevolved system. These processes include, for example, anomalous/stochastic chemical evolution and D/⁴He photodestruction. In general, we find it unlikely that these processes could have significantly altered D/H in Lyman limit clouds. We argue that chemical evolution scenarios, unless very finely tuned, cannot account for significant local deuterium depletion since they tend to overproduce ¹²C, even when allowance is made for possible outflow. Similarly, D/⁴He photodestruction schemes engineered to locally produce or destroy deuterium founder on the necessity of requiring an improbably large γ-ray source density. Future observations of D/H in Lyman limit systems may provide important insight into the initial conditions for the primordial nucleosynthesis process, early chemical evolution, and the galaxy formation process.

We have developed a technique to systematically discover and study high-redshift supernovae that can be used to measure the cosmological parameters. We report here results based on the initial seven of more than 28 supernovae discovered to date in the high-redshift supernova search of the Supernova Cosmology Project. We find an observational dispersion in peak magnitudes of σ_MB=0.27; this dispersion narrows to σ_{MB, corr}=0.19 after "correcting" the magnitudes using the light-curve "width-luminosity" relation found for nearby (z ≤ 0.1) Type Ia supernovae from the Calán/Tololo survey (Hamuy et al.). Comparing light-curve width-corrected magnitudes as a function of redshift of our distant (z = 0.35-0.46) supernovae to those of nearby Type Ia supernovae yields a global measurement of the mass density, Ω_M=0.88^{+ 0.69}_−0.60 for a Λ = 0 cosmology. For a spatially flat universe (i.e., Ω_M + Ω_Λ = 1), we find Ω_M=0.94^{+ 0.34}_−0.28 or, equivalently, a measurement of the cosmological constant, Ω_Λ=0.06^{+ 0.28}_−0.34 ( < 0.51 at the 95% confidence level). For the more general Friedmann-Lemaître cosmologies with independent Ω_M and Ω_Λ, the results are presented as a confidence region on the Ω_M-Ω_Λ plane. This region does not correspond to a unique value of the deceleration parameter q₀. We present analyses and checks for statistical and systematic errors and also show that our results do not depend on the specifics of the width-luminosity correction. The results for Ω_Λ-versus-Ω_M are inconsistent with Λ-dominated, low-density, flat cosmologies that have been proposed to reconcile the ages of globular cluster stars with higher Hubble constant values.

The small scatter observed for the (U-V) colors of spheroidal galaxies in nearby clusters of galaxies provides a powerful constraint on the history of star formation in dense environments. However, with local data alone, it is not possible to separate models where galaxies assembled synchronously over redshifts 0 < z < 1 from ones where galaxies formed stochastically at much earlier times. Here we attempt to resolve this ambiguity via high-precision rest-frame UV-optical photometry of a large sample of morphologically selected spheroidal galaxies in three z ~ 0.54 clusters that have been observed with the HubbleSpaceTelescope (HST). We demonstrate the robustness of using the HST to conduct the morphological separation of spheroidal and disk galaxies at this redshift and use our new data to repeat the analysis conducted locally at a significant look-back time. We find a small scatter (<0.1 mag rms) for galaxies classed as E's and E/S0's, both internally within each of the three clusters and externally from cluster to cluster. We do not find any trend for the scatter to increase with decreasing luminosity down to L~L^*_V+3, other than can be accounted for by observational error. Neither is there evidence for a distinction between the scatter observed for galaxies classified as ellipticals and S0. Our result provides a new constraint on the star formation history of cluster spheroidals prior to z ≃ 0.5 confirming and considerably strengthening the earlier conclusions. Most of the star formation in the elliptical galaxies in dense clusters was completed before z ≃ 3 in conventional cosmologies. Although we cannot rule out the continued production of some ellipticals, our results do indicate an era of initial star formation consistent with the population of star-forming galaxies recently detected beyond z ≃ 3.

Two models of the gamma-ray burst population, one with a standard candle luminosity and one with a power-law luminosity distribution, are χ²-fitted to the union of two data sets: the differential number versus peak flux distribution of BATSE's long-duration bursts and the time dilation and energy shifting versus peak flux information of pulse duration time dilation factors, interpulse duration time dilation factors, and peak energy shifting factors. The differential peak flux distribution is corrected for threshold effects at low peak fluxes and at short burst durations, and the pulse duration time dilation factors are also corrected for energy stretching and similar effects. Within an Einstein-de Sitter cosmology, we place strong bounds on the evolution of the bursts, and these bounds are incompatible with a homogeneous population, assuming a power-law spectrum and no luminosity evolution. Additionally, under the implied conditions of moderate evolution, the 90% width of the observed luminosity distribution is shown to be ≲10², which is less constrained than others have demonstrated it to be assuming no evolution. Finally, redshift considerations indicate that if the redshifts of BATSE's faintest bursts are to be compatible with that which is currently known for galaxies, a standard candle luminosity is unacceptable, and, in the case of the power-law luminosity distribution, a mean luminosity ≲10⁵⁷ photons s^-1 is favored.

We examine the possibility that a population of relic massive black holes, perhaps constituting an important component of the dark matter, might be indirectly detected via their occasional very strong gravitational lensing of individual luminous stars in distant external galaxies. For plausible, and in some respects conservative, values of the relevant physical parameters, we show that such events might be detected either in wide area surveys reaching routine CCD magnitude limits (such as the Sloan Digital Sky Survey) or in small field, very deep images (such as the Hubble Deep Field). Thus, it would be a challenging but not impossible task to detect or place limits on a cosmic population of relic massive black holes.

We numerically investigate the dynamical evolution of a rapidly collapsing and self-gravitating fluid composed of collisionless stars and collisional gas with a particular emphasis on the roles of star formation in the formation of a stellar bar. We adopt idealized and generalized initial conditions of the collapsing two-component fluid and rather simplified models for star formation and investigate the dynamical roles of gas consumption by star formation in stellar bar formation. We found that the rapidity of gas consumption by star formation predominantly determines whether or not a stellar bar forms after the collapsing two-component fluid reaches virial equilibrium: A stellar bar is more likely to form in a two-component fluid with rapid star formation, whereas inhomogeneous structures (e.g., a few clumps) are more likely to form in a fluid with gradual star formation. This is principally because the growth of gaseous clumps, which is a key determinant for suppressing the growth of global bar instability, is inhibited more significantly by gas consumption in the two-component fluid with more rapid star formation. Furthermore, we found that this particularly important role of gas consumption by star formation does not depend strongly on the initial virial ratio and initial fraction of gas mass in the two-component fluid. These results demonstrate that the rapidity of gas consumption by star formation during collapse is one of crucial parameters for the formation of a stellar bar in collapsing and self-gravitating two-component fluid. Although our numerical models are rather idealized and less realistic, these results provide a clue to the understanding of stellar bar formation in forming disk galaxies.

We present a schematic model for the formation of baryonic galactic halos and hot gas in the Local Group and the intergalactic medium. We follow the dynamics, chemical evolution, heat flow, and gas flows of a hierarchy of scales, including protogalactic clouds, galactic halos, and the Local Group itself. Within this hierarchy the Galaxy is built up via mergers of protogalactic fragments. Hot and cold gas components are distinguished, with star formation occurring in cold molecular cloud cores, while stellar winds, supernovae, and mergers convert cold gas into a hot intercloud medium. We find that early bursts of star formation lead to a large population of remnants (mostly white dwarfs) that would presently reside in the halo and contribute to the dark component observed in the microlensing experiments. The starbursts and mergers heat the gas and lead to powerful evaporation-driven winds. This outflow is crucial, as it drives gas out of the clouds and eventually into the intergalactic medium. The model thus suggests that most microlensing objects could be white dwarfs (m ~ 0.5 M_☉), which comprise a significant fraction of the halo mass. Furthermore, the Local Group could have a component of metal-rich hot gas similar to, although less than, that observed in larger clusters. We discuss the known constraints on such a scenario and show that all local observations can be satisfied with present data in this model. The most stringent constraint comes from the metallicity distribution in the halo. The best-fit model has a halo that is 40% baryonic, with an upper limit of 77%. Our model predicts that the hot intragroup gas has a total luminosity 1.5 × 10⁴⁰ ergs^-1 and a temperature of 0.26 keV, just at the margin of detectability. Improved X-ray data could provide a key constraint on any remnant component in the halo.

We investigate the hypothesis that Lyα absorption lines arise in two populations of halos—minihalos of small circular velocity (V_c ≲ 55 km s^-1), in which star formation and metal production are inhibited by photoionization of the UV background radiation, and large galactic halos (55 ≲ V_c ≲ 250 km s^-1) that contain stars and metals. Based on the model of Lyα-absorbing gas confined in both populations of halos, we attempt to explain the recent observations of (1) associations of visible galaxies with Lyα lines at low redshifts z ≲ 1 and (2) metal lines associated with a nonnegligible fraction of low H I column density Lyα lines at z ~ 3. For galactic halos, we find that photoionized gas clouds confined in the pressure of ambient hot gas can produce Lyα absorption for H I column densities as low as 10¹⁴ cm^-2 and that the impact parameter of a sight line for such absorptions matches well with the observed radius of the gaseous envelope of a typical luminous galaxy. Using the Press-Schechter prescription for the mass function of halos, we also show that the fraction of Lyα lines with associated metal lines can be understood in terms of the fraction of Lyα absorbers associated with galactic halos. In particular, the reported fraction of ~0.5-0.75 at z ~ 3 is reproduced when the boundary value of V_c to separate minihalos or galactic halos is 40-60 km s^-1, which is consistent with the theoretical prediction of galaxy formation under photoionization. The average metallicity of both Lyα forest and damped Lyα systems at z ~ 3 is explained in terms of the model of halo-formation history combined with the age-metallicity relationship of Galactic halo stars. Possible methods to test this hypothesis and alternative scenarios are also discussed.

We consider the hypothesis that galactic magnetic fields are primordial. We also discuss the various objections to this hypothesis. To test this hypothesis properly we assume that there was a magnetic field present in the galactic plasma before the galaxy formed and explore how such a field would evolve assuming a specific model for the interstellar medium in the galactic disk. After the galactic disk formed, the lines of force thread through it and remain connected to the external cosmic medium. They enter through one side of the disk, proceed horizontally a distance l in the disk, and then leave through the other side. We find that the lines of force are stretched by the differential rotation of the galactic disk, amplifying the toroidal component of the field and increasing l. When the magnetic field is strong enough, it produces ambipolar velocities that try to lift the line out of the galactic disk but in opposite directions on different parts of the line. The result is that, instead of the line being expelled from the disk, its horizontal length l is shortened, both in the radial and in the toroidal direction. This leads to a reduction of the rate of horizontal stretching and finally a reduction in the magnetic field strength. After a sufficient time, the magnetic field at all points goes through this reduction and the field strength approaches a universal function of time. This function is slowly decreasing and only depends on the ambipolar properties of the interstellar medium. At any given time the magnetic field is toroidal and has the same strength everywhere. On the other hand, it turns out that its direction varies rapidly with radius, changing sign every 100 parsecs or so. However, if the initial cosmic magnetic field is not uniform, the areas of one sign of the toroidal field dominate over the other. The resulting field has a net Faraday rotation. If such a field were observed with low resolution in an external galaxy, then the field would appear toroidal in between the spiral arms. The spiral density wave would turn it so that the lines appear to trace out the spiral arm, although the apparent lines really are the sum of pieces of magnetic lines as they cross the disk. They do not necessarily extend very far along the arms. We contend that this model of the magnetic field, which arises naturally from a primordial origin, can fit the observations as well as other models for the magnetic field, such as those arising from the mean field dynamo theory. Finally, because the field lines are topologically threaded through the disk, they cannot be expelled from the disk. This counters the objection against the primordial origin, namely that such a field could not survive very long in the galaxy.

We have obtained sensitive long-slit spectra of diffuse ionized gas (DIG) in the Andromeda galaxy, M31, covering the wavelength range of 3550-6850 Å. By co-adding extracted DIG spectra, we reached a 1 σ uncertainty of 9.3 × 10^-19 ergs s^-1 cm^-2 arcsec^-2 corresponding to 0.46 pc cm^-6 in emission measure. We present average spectra of DIG at four brightness levels with emission measures ranging from 9 to 59 pc cm^-6. We present the first measurements of [O II] λ3727 and [O III] λ5007 of the truly diffuse ionized medium in the disk of an external spiral galaxy. We find that I_{[O II]}/I_Hα = 0.9-1.4. The [O III] line is weak (I_{[O III]}/I_Hβ = 0.5), but it is stronger than found for the Galactic DIG. Measurements of [N II] λ6583 and [S II] (λ6717 + λ6731) are also presented. The [S II] lines are clearly stronger than typical H II regions (I_{[S II]}/I_Hα = 0.5 compared to 0.2), confirming various imaging studies of spiral galaxies. Overall, the line ratios are in agreement with predictions of photoionization models for diffuse gas exposed to a dilute stellar radiation field, but the line ratios of the DIG in M31 are somewhat different than observed for Galactic DIG. The differences indicate a less diluted radiation field in the DIG of M31's spiral arms compared to DIG in the Solar Neighborhood of the Milky Way. Turbulent mixing layers can contribute at most 20% of the ionization budget of the DIG, with lower percentages producing better fits to the observed line ratios. We have also detected He I λ5876 emission from the brightest DIG in M31. The He I line appears to be stronger than in the Galactic DIG, possibly indicating that most of the Helium in the bright DIG in M31 is fully ionized. However, this result is somewhat tentative since bright night sky lines hamper an accurate measurement of the He I line strength.

We present CCD photometry and low-resolution spectra of the spectroscopic Type Ic supernova SN 1983V in NGC 1365. Photometry in the B band spans nearly 250 days after maximum light, while the spectra cover the photospheric epoch, from 9 days before up to 38 days after B maximum. We also compile and discuss infrared photometry that has been published elsewhere. The photometric evolution of SN 1983V is analyzed in comparison with other Type Ic supernovae and with SN 1983N and SN 1993J. The spectroscopic evolution is described, line identifications are proposed, and evidence for H, He, and Si is discussed. The photometric evolution of SN 1983V suggests that it is a close relative of SN 1983N and SN 1993J, events for which it is supposed that the external layers comprising most of the He/H and H shells, respectively, were lost before the explosion. The mass-to-energy ratios of SN 1983V and SN 1993J are substantially different if estimated from the expansion velocities. However, the similarity of the B light curves suggests that if the masses and mass-to-energy ratios are not similar, then the mass of SN 1983V has to be large enough to compensate for the higher mass-to-energy ratio. The latter possibility contradicts the standard scenarios that describe these events and implies a precise fine-tuning of the masses and mass-to-energy ratios of the two supernovae. This contradiction, however, could be resolved if a significant fraction of the inner ejecta containing ⁵⁶Ni does not follow the decrease in density expected for a spherically symmetric, homologous expansion.

We present long-slit optical spectroscopy of three high surface brightness Magellanic irregular galaxies. This paper draws attention to our detection of He I λ5876 line emission from the ionized gas outside the H II regions or the warm ionized phase of the interstellar medium. We measure a mean reddening-corrected intensity ratio of He I λ5876/Hα ≈ 0.041 independent of spatial location. This ratio is much higher than that measured in the diffuse, warm ionized interstellar medium of the Milky Way (Reynolds & Tufte).

The high value of He I λ5876/Hα implies the helium ionization fraction is approximately equal to the hydrogen ionization fraction in the diffuse ionized gas (DIG). If the DIG is powered by young stars, then stars hotter than 40,000 K must contribute to the Lyman-continuum radiation reaching the DIG. Since optical and ultraviolet spectra confirm the presence of such massive stars in these galaxies, stellar photoionization remains the most likely power source. The contrast with the low helium ionization in the Galactic DIG, however, is intriguing and provides strong evidence that the physical state of the DIG, not just its presence, varies among galaxies.

We present spectroscopic observations in the near-ultraviolet, optical, and near-infrared of the super- star clusters (hereafter SSC A and SSC B) in NGC 1569. Previous studies have suggested that they are in a poststarburst phase. However, our data suggest a younger mean age and a significant age spread. The spectrum of SSC A shows the Wolf-Rayet (W-R) feature at 4686 Å, with a luminosity equivalent to 20-40 WNL stars. This qualifies NGC 1569 as a W-R galaxy. The small Balmer jump detected in both clusters also suggests strongly the presence of very massive (and therefore very young) stars. Evolutionary synthesis models are used to constrain the star formation scenario by explaining the spectral energy distribution, the Balmer jump, the equivalent width (EW) of the W-R bump, and the near-infrared Ca II triplet in absorption. No single-age stellar population fits all these constraints. We propose a two-burst model, with the younger burst having an age of 3 Myr (2 Myr for SSC B) and the older one having an age of 9 Myr (8 Myr for SSC B), to explain the simultaneous presence of hot massive stars and red supergiants. We speculate that sequential star formation is taking place in the SSCs of NGC 1569, with the younger burst probably located in the surroundings of and initiated as a consequence of the energetic stellar activity of the older central compact cluster. A deficit of ionized gas was found around the SSCs, which we ascribe to the strong stellar winds and supernova explosions of the older burst removing the gas from the vicinity of the clusters.

We present the results of a program to monitor the absorption lines in the spectrum of the Seyfert 1 galaxy NGC 4151 caused by outflowing gas from the nucleus. Spectral observations were taken in the UV with the GHRS and FOS of HST. GHRS G160M and G270M exposures were taken of the wavelength regions centered on C IV and Mg II, respectively, extending over a nearly 4 yr period. Fits to the observations at all epochs but one require at least eight distinct components spanning a range of outflow velocities from 0 to ~1600 km s^-1 with respect to the nucleus. (NGC 4151 has a redshift of ~1000 km s^-1).

Although we see subtle changes in GHRS spectra of the broader of the absorption features, the wavelength constancy of all the features is remarkable. The limits on the secular acceleration suggest that either (1) The absorbing clouds are well beyond the broad emission line region; (2) The clouds are experiencing significant drag from an intercloud medium; or (3) We are observing an extremely complex flow pattern rather than the motions of individual clouds. The exception to this constancy occurred during one of the epochs of our monitoring when a broad shallow C IV trough appeared at an outflow velocity of 3750 km s^-1 and then subsequently disappeared.

An archival G160M exposure taken by Ulrich et al. was studied in order to measure the population of fine structure levels of Si II via the lines at 1526 and 1533 Å. The ratio of the upper to lower state appears to be less than the ratio of statistical weights, though additional observations are needed to secure this. This ratio is used to set upper limits on the electron density and lower limits on the distance of one of the absorbing clouds.

Simultaneous FOS observations, covering a wider range of lines, but at a lower resolution, show significant changes in the Si IV absorption profile.

We discuss the NGC 4151 absorption phenomenon in light of other observations of "intrinsic" absorption systems in both Seyfert nuclei and quasars.

We find the main families of simple periodic orbits in and around the bar of NGC 4314 and examine their stability. In many ways, our results agree with those found for model barred galaxies, yet our realistic potential allows us to go further in a comparison with the galaxy morphology. In particular, we underline the importance of the families of periodic orbits that are asymmetric with respect to the bar minor axis.

The x₁ family provides the building blocks for the bar. In the inner parts we find orbits that are roughly perpendicular to the bar, although their shape and orientation vary along the corresponding families. As in previous studies, we find a symmetric unstable 3:1 family, but we also find an asymmetric and stable 3:1 family. We also find asymmetric diamond-like orbits near corotation. We pay special attention to the orbital behavior at the ultraharmonic resonance region, and we investigate all possibilities offered by our study in explaining the boxy structure at the end of the bar.

We calculate the distance to NGC 4478, a neighbor of the giant Virgo elliptical M87, using the surface brightness fluctuations method on parallel data from the HST WFPC2 camera. The exposures considered were taken through the F814W filter, which approximates the Johnson I, and through the F606W filter, which is effectively a broad V filter. We describe the use of Fourier methods for separation of the fluctuations due to statistical variations in the projected number of stars per pixel (used to determine the distance) from fluctuations from other sources, which include photon noise, foreground stars, cosmic rays, and the variations due to the overall structure of the galaxy. The mean apparent magnitude of stars responsible for the measured fluctuations is _I=29.78 ± 0.03, with an additional systematic uncertainty of 0.05 mag. Using the latest calibration of Tonry (1997), we obtain a distance of 15.6 ± 1.0 Mpc, which is consistent with the distance determined through the use of this method in the infrared K band by Pahre & Mould (1994). We discuss the globular cluster (GC) luminosity function and color distribution, based on the ~130 candidates identified in the field. We find a roughly Gaussian luminosity function centered at m_V = 23.82 ± 0.38 with σ = 1.16 ± 0.21. The GC color distribution is also roughly Gaussian, centered at V-I = 0.91 ± 0.03, with σ = 0.20 ± 0.03, and it is coincident to within the error bars with the blue peak of the bimodal color distribution found in M87. Comparing our data with recent evaluations of the GC color distribution at two different locations in M87 we find a trend of the red peak disappearing with radial distance. We speculate that this could be understood if the blue portion of the M87 GC population were acquired from its smaller companions. The specific frequency of globular clusters is among the lowest for elliptical galaxies, comparable only to the value found in M32. In both cases this may be a result of tidal truncation by a giant neighbor. This further indicates that clusters have been stripped from NGC 4478 by M87 and likely now are part of its globular cluster population.

We present high spatial resolution 21 cm H I maps of the interacting galaxy pair NGC 7714/7715. We detect a massive (2 × 10⁹M_☉) H I bridge connecting the galaxies that is parallel to but offset from the stellar bridge. A chain of H II regions traces the gaseous bridge, with Hα peaks near but not on the H I maxima. An H I tidal tail is also detected to the east of the smaller galaxy NGC 7715, similarly offset from a stellar tail. The strong partial stellar ring on the east side of NGC 7714 has no H I counterpart, but on the opposite side of NGC 7714 there is a 10⁹M_☉ H I loop ~11 kpc in radius. Within the NGC 7714 disk, clumpy H I gas is observed associated with star formation regions. Redshifted H I absorption is detected toward the strong starburst nucleus. We compare the observed morphology and gas kinematics with gas dynamical models in which a low-mass companion has an off-center prograde collision with the outer disk of a larger galaxy. These simulations suggest that the bridge in NGC 7714/7715 is a hybrid between the tidal bridges seen in systems like M51 and the purely gaseous "splash" bridges found in ring galaxies like the Cartwheel. The offset between the stars and gas in the bridge may be due to dissipative cloud-cloud collisions occurring during the impact of the two gaseous disks.

We present the data and preliminary analysis for a series of 90 ROSAT HRI and two ASCA observations of the broad-line radio galaxy 3C 390.3. These data were obtained during the period 1995 January 2 to October 6 as part of an intensive multiwavelength monitoring campaign. The soft X-ray flux in the ROSAT band varied by nearly a factor of 4 during the campaign, and the well-resolved light curve shows several distinct features. Several large-amplitude flares were observed, including one in which the flux increased by a factor of about 3 in 12 days. Periods of reduced variability were also seen, including one nearly 30 days long. While the HRI hardness ratio decreased significantly, it is apparently consistent with that expected from the detector during the monitoring period.

The two ASCA observations were made on 1995 January 15 and May 5. The 0.5-10.0 keV spectra can be described adequately by an absorbed power law. There is no evidence for a soft excess in the ASCA spectra, indicating that the ROSAT HRI is sampling variability of the X-ray power law. A broad iron line was observed in a longer 1993 ASCA observation, and while there is statistical evidence that the line is present in the 1995 spectra, it could not be resolved clearly. There is evidence, significant at greater than 90% confidence, that the photon index changed from 1.7 to 1.82 while the flux increased by 63%. The spectral change can be detected in the spectra below 5 keV, indicating that the origin cannot be a change in the ratio of reflected to power-law flux. A compilation of results from ASCA and Ginga observations show that on long timescales, the intrinsic photon index is correlated with the flux.

We report the presence of an absorption feature in the X-ray spectrum of the BL Lacertae object H1426+428 (z = 0.129). H1426+428 was observed at three separate epochs with BBXRT, ROSAT, and ASCA. In the BBXRT spectrum an absorption feature is clearly present at ~0.66 keV, consistent with absorption by highly ionized oxygen. The phenomenological model that best describes the absorption feature is an inverted Gaussian with width FWHM ≲ 43,000 km s^-1. This feature is not present in the ROSAT and ASCA data, both taken when the source flux was fainter by a factor ≈ 2. Instead, these data show some evidence for absorption edge(s) at lower energies. These results are consistent with partially ionized, low-density gas surrounding the central engine of H1426+428.

As a first approximation the X-ray continuum can be described with a power law with photon index Γ ≈ 2. The BBXRT and ASCA data are better fitted with a convex (downward-curved) broken power law. The exact shape of the continuum depends on the assumed profile for the absorption feature. We show that, using a specific warm absorber to fit the data, there is no need for a continuum model more complex than a single power law.

We used the Berkeley-Illinois-Maryland Association array to map Cygnus A at 87 GHz with 04 angular resolution. The hot spots at the working surfaces in the radio lobes show spectral index gradients from -0.6 to -1.1 between 87 and 5 GHz, which suggests significant synchrotron aging in the hot spots. The data do not show a marked spectral break between 5 and 87 GHz, however, which suggests a continuous injection of high-energy relativistic particles into both compact and extended hot spots on timescales less than 3 × 10⁴ yr.

We include new data in an updated analysis of helium in low-metallicity extragalactic H II regions with the goal of deriving the primordial abundance of ⁴He (Y_P). We show that the new observations of Izotov et al. are consistent with previous data. However, they should not be taken in isolation to determine Y_P, owing to the lack of sufficiently low-metallicity points. We use the extant data in a semi-empirical approach to bounding the size of possible systematic uncertainties in the determination of Y_P. Our best estimate for the primordial abundance of ⁴He assuming a linear relation between ⁴He and O/H is Y_P = 0.230 ± 0.003(stat) based on the subset of H II regions with the lowest metallicity; for our full data set we find Y_P = 0.234 ± 0.002(stat). Both values are entirely consistent with our previous results. We discuss the implications of these values for standard big bang nucleosynthesis (SBBN), particularly in the context of recent measurements of deuterium in high-redshift, low-metallicity QSO absorption-line systems.

Spectra at 16-45 μm of several regions within the central 80'' of the Galaxy have been obtained at 20'' resolution using the Goddard Cryogenic Grating Spectrometer No. 2 on the Kuiper Airborne Observatory. A broad band of excess emission extending from 24 to 45 μm is present in the spectra at positions covering the "tongue" and the inner edge of the circumnuclear disk. A similar dust emission feature has been observed in some carbon-rich evolved stars and in a nitrogen-rich evolved massive star. The observations reported here are the first detection of this dust emission feature in the interstellar medium.

We find that the dust properties in the "tongue" and the inner edge of the circumnuclear disk are different from the dust in the cavity, which shows no evidence for the ~30 μm feature in its spectrum. The relation between the tongue feature and the inner disk has been a subject of considerable debate, with suggestions that the tongue material is either falling in or has been expelled from the Galactic center. Our observations suggest that if there is a physical connection between the two regions, then the tongue material is more likely falling in toward the Galactic center.

Two kinds of dust components and corresponding heating sources are argued to be responsible for the dust emission in the Galactic center: (1) the hot dust component (140-300 K) heated by IRS sources that are distributed throughout the cavity; and (2) the warm dust component (60-90 K) heated by a He I emission-line star cluster that give rise to a local interstellar radiation field that heats the dust in the central several parsecs. We find a lower limit of ~1 × 10⁶L_☉ for the He I emission-line star cluster.

Finally, we find that the line-of-sight extinction across the cavity is not uniform. It is diminished and roughly constant in the northeastern side of the cavity but increases to the southwestern side. Based on this result, together with the 1995 study by Zylka et al. who found that there is a cold dust component along the line of sight to the central 30'' of the Galaxy, we propose that there is a cool dust envelope with nonuniform distribution covering the cavity of the Galactic center.

We present a theoretical investigation on the pulsational behavior of metal-rich RR Lyrae variables over the range of evolutionary parameters suitable for stars with metallicities Z = 0.006, 0.01, and 0.02. With the addition of similar results for metal-poor pulsators, we discuss the theoretical pulsational scenario covering the metallicity range Z = 0.0001-0.02.

By connecting pulsational constraints to evolutionary prescriptions for He-burning stars, we discuss the observed behavior of the RR Lyrae population in the Galactic field. We find that the distribution of field ab-type RR Lyrae stars in the period-metallicity plane can be easily understood within the framework of the present theoretical scenario, suggesting that the Oosterhoff dichotomy also affects field variables.

Theoretical predictions concerning the amplitude-period diagram are discussed and compared with observational data. We find a general agreement for metal-poor ([Fe/H] < -1.4) RR Lyrae stars, whereas more metal-rich variables show amplitudes smaller than those predicted for pulsators originated from old, low-mass evolving stars. Alternatively, the agreement between theory and observations would require that a substantial fraction of metal-rich RR Lyrae variables in the Galactic field were younger than ≈ 2 Gyr.

The comparison between the pulsational behavior of RR Lyrae either in the Galactic field or in the Galactic bulge discloses the evidence that, at least as far as RR Lyrae variables are concerned, the metal-rich components of the bulge and of the field population appear quite similar. We finally suggest that the peak at log P ≃ -0.55 in the period frequency distribution of first-overtone RR Lyrae stars in the Large Magellanic Cloud, considered as possible evidence for second-overtone pulsators, could be more simply taken as evidence of a metal-rich stellar population.

Faint magnitude BVI CCD photometry of the central region of the old open cluster Berkeley 17 (Be 17) has been undertaken to investigate the claim that it is the oldest open cluster yet discovered (Phelps, Janes, & Montgomery). This study reveals Be 17 to have a metallicity -0.30 ≤ [Fe/H] ≤ 0.00; a reddening 0.52 ≤ E(B-V) ≤ 0.68 or 0.61 ≤ E(V-I) ≤ 0.71; a distance modulus, (m - M)₀ = 12.15 ± 0.10, corresponding to a distance of 2.7 ± 0.1 kpc; a diameter of ~10 pc; a minimum mass of 400 M_☉; and an age of 10-13 Gyr. With an adopted age of 12^{+ 1}₋₂ Gyr, these results suggest that Be 17 is indeed the oldest open cluster yet discovered. The inferred old age for Be 17 indicates that the cluster is an important one for studies of the differences between open and globular clusters, the transition from the halo/thick disk to the thin disk of the Galaxy, and even for establishing the minimum age of the universe.

We examine two scenarios for formation of the planetary nebula K648: a prompt scenario in which the planetary nebula is ejected and formed immediately after a helium shell flash and a delayed scenario in which a third dredge-up occurs and the envelope is ejected during the following interpulse phase. We present models of both scenarios and find that each can produce K648-like systems. We suggest that the prompt scenario is more favorable but cannot rule out the delayed scenario.

We report the detection using Advanced Satellite for Cosmology and Astrophysics (ASCA) data of a hard X-ray diffuse emission around the gamma-ray pulsar PSR 1055-52. The pulsar is a middle-aged radio pulsar (P = 0.197 s, τ=P/ ≈ 5.3 × 10⁵ yr) with a pulse profile similar to that of the Crab pulsar and is one of the EGRET gamma-ray pulsars. In the soft X-ray band, ROSAT detected the pulsar as a point source indicating radiation from the neutron star. The ASCA hard band image (2-8 keV), however, showed that the diffuse emission around the pulsar dominates the pulsar itself. Morphology suggests a ring around the pulsar with a cavity in between, although the source is clumpy. The size of the suggested ring is 20' (5.8 D_kpc pc, where D_kpc denotes the distance in kiloparsecs) in diameter. The luminosity of the diffuse source is ≈ 2.0 × 10³²D ergs s^-1, which corresponds to 6.8 × 10^-3 of the rotation power of the pulsar. Spectral analysis has been performed for the two brightest clumps. The spectrum can be fitted by a Raymond-Smith thermal plasma model with temperatures of 9 keV and 3 keV, respectively, for the two clumps and by a power-law model with photon indices of 1.7 and 2.2. Since the thermal model gives a very high pressure, we prefer the power model. We suggest that the discovered X-ray emission, together with morphology is the synchrotron nebula excited by the pulsar wind of PSR 1055-52.

This is the first of two papers devoted to physical interpretation of Rybicki's quadratic and bilinear relations. The fundamental equations obtained on the basis of Ambartsumian's invariance principle and regarded as its extension to all depths in the atmosphere, imply that the Q and R relations have with more general structures than those known up to the present. These equations allow a simple probabilistic interpretation. Some bilinear relations are derived to connect the transfer problems of different sorts. For the sources distributed in the semi-infinite atmosphere by exponential law, the separate Q and R relations are obtained. It is shown that the two-point quadratic Q relation may be derived directly on the base of physical considerations used in the invariance technique.

An earlier examination by B. Miller and B. J. Eastlund (1996) of coherent radio emission from particles confined to closed and filled magnetospheric shells that approach the light cylinder of a pulsar is extended to include incoherent emission from energetic charged particles, similarly confined. The emitting particles are now also postulated to be electrostatically mirroring in the pulsar's magnetic field. This model generates time-averaged pulse shapes that resemble closely those from the Crab pulsar, over 14 orders of magnitude of frequency, and matches the observed power spectrum over the same frequency range.

We report the observations of the X-ray lobes of the large Galactic source W50 associated with the two-sided jets source SS 433. The observations were made with the Position Sensitive Proportional Counter on board ROSAT. The eastern lobe has a knotty structure, with X-ray emission starting at ~15' from SS 433 and peaking at ~35'. Both thermal bremsstrahlung spectra with high temperatures (≥5 keV), as well as power-law spectra with flat photon index (Γ ~ 1-2), gave adequate fits. With ROSAT, we also detected an additional soft X-ray emission from the eastern lobe, at ~1° east of SS 433 and coincident with the radio "ear." This region is associated with the terminal shock of the SS 433 jets. The western lobe has a similar morphology; however, the spectrum is softer, and no emission was found coincident with the western radio "ear."

We also report the observations of W50 with the AdvancedSatelliteforCosmologyandAstrophysics (ASCA), and combine the ROSAT and the ASCA data for a wide coverage of the spectra in the 0.1-10 keV energy range. We find that the spectrum from the brightest region (at ~35' from SS 433) requires an additional component to account for an excess of photons in the 1-3 keV band. The total X-ray luminosity from both lobes is ~10³⁵-10³⁶D²_5.5 ergs s^-1.

We discuss the origin of the X-rays in the model of emission from the supersonic jets of SS 433 interacting with the ambient medium, accelerating particles to very high energies, and producing a cocoon of "hot" shocked material. We present our interpretation in light of the correlation of X-ray emission with the map of W50 at other wavelengths.

Recent numerical simulations of accretion disk boundary layers have shown qualitatively the importance of advected energy in the inner region of the disk. In this short paper we present quantitative results of advective boundary layers in the optically thick regime. Numerical results are obtained for various systems, by means of a one-dimensional time-dependent numerical code. At high accretion mass rates, ≈ 10⁻⁴M_☉ yr^-1, or low values of the viscosity parameter, α ≈ 0.001-0.01 (characteristic of FU Orionis systems and some symbiotic stars), the optical thickness in the inner part of the disk becomes very large (τ ≫ 1). The disk, unable to cool efficiently, becomes geometrically thick (H/r ≈ 0.5). The energy dissipated in the dynamical boundary layer is radiated outward to larger radii and advected into the star. The boundary-layer luminosity is only a fraction of its expected value; the rest of the energy is advected into the star. The fraction of the advected energy is ζ = L_adv/L_acc ≈ 0.1 in symbiotic stars (accretion onto a low-mass main-sequence star) and ζ ≈ 0.2 in FU Ori systems (accretion onto a pre-main sequence star).

A model of AG Peg is presented, focusing on the physical conditions in the emission nebulae. The model accounts in a consistent way for photoionization by the star and ionization by shocks. The SUMA code is used in the calculations of the spectra. We show that two regions contribute to the UV and optical line spectra. The broad lines are emitted from photoionized gas surrounding the hot star, while narrow lines are emitted by shocked gas—which is also reached by the ionizing flux from the hot star—near the red giant.

At an early phase (ϕ = 2.33) the emitting region is between the stars but already very close to the red giant. The nebula surrounding the hot star is not exactly symmetric. At a later phase (ϕ = 7.05) the shock created by "head-on-back" collision of the winds propagates outward and slightly accelerates throughout the giant atmosphere characterized by a decreasing slope of the density. The strong shock caused by head-on collision of the winds from the two stars cannot be recognized in the UV-optical line spectrum.

The spectral energy distribution of the continuum is well fitted by blackbody emissions from the stars. The fit of the optical-UV observed continuum at the early phase presents some problems connected with the reddening correction of the data. On the other hand, the fit of the data at a later phase is consistent with the parameters deduced by the modeling of the system. The far-IR data are well fitted by reradiation by dust, which could not be recognized without modeling, and indicate a dust-to-gas ratio similar to the Galactic one.

Photometric and spectroscopic observations of the classical nova OY Ara were carried out using the Cerro Tololo Inter-American Observatory telescopes. We found that OY Ara is an eclipsing system with an orbital period of 0.155466 ± 0.000003 days. The radial velocity semiamplitude of the Hα emission line is 118 ± 9 km s^-1. Assuming the secondary is near the main sequence, the dynamical analysis shows that the primary and the secondary have masses of M₁ ≈ 0.82 M_☉ and M₂ ≈ 0.34 M_☉, respectively. We also find that the accretion disk of OY Ara is 2.5-6 mag more luminous than that of the X-ray novae, presumably due to the much higher rate of mass transfer in classical novae.

We obtained Hubble Space Telescope Goddard High-Resolution Spectrograph medium-resolution (G160M grating) phase-resolved spectroscopic observations of the prototype dwarf nova U Geminorum during different stages of two different outbursts. The spectral wavelength ranges were centered on three different line regions: N V (1238 Å, 1242 Å), Si III (1300 Å), and He II (1640 Å). The spectrum corresponding to the early decline phase of outburst 1 is essentially featureless except for weak N V absorption and narrow interstellar lines, while the spectrum at the peak of outburst 2 reveals broad emission peaks separated by narrow central absorption. The double-peaked emission-line profile structure with low-velocity central absorption seen in the second outburst suggests a disk origin, but the emission velocity widths appear narrower than the widths of the optical disk emission features. We interpret the high-excitation emission lines, with central absorption below the continuum, to be due to photoionized material (coronal?) above the disk plane with the thickened outer disk absorbing the boundary layer or inner disk radiation. The possibility of a wind origin for the profiles is also discussed, as well as the possibility of an ejected optically thin shell. The N V absorption velocity versus orbital phase traces the motion of the white dwarf, but the He II absorption velocity appears to deviate from the white dwarf motion. We present the results of synthetic accretion disk spectral fitting to the data of both outbursts and derive accretion rates for the two outbursts of 6 × 10^-10M_☉ yr^-1 and 2 × 10^-9M_☉ yr^-1. Implications are discussed.

We have studied the temperature distribution and other physical properties of the circumstellar envelope of the prototypical high mass-loss carbon star IRC +10216 by calculating the thermal balance and the radiative transfer in the envelope self-consistently. Cooling is dominated by CO line emission and adiabatic expansion, and heating by dust-gas collisions throughout most of the envelope. Heating by the grain photoelectric effect is important in the outer part of the envelope. The radiative transfer is calculated by using a Monte Carlo method. The mass-loss rate, the CO abundance, the dust-gas momentum transfer efficiency, and the distance to the source are free parameters in our model. These physical parameters are constrained by the comparison of our model results with the observations of various ¹²CO and ¹³CO lines. In particular, recent submillimeter-wavelength observations of moderately high excitation transitions, such as the J = 6 → 5 line, put very important constraints on the temperature distribution in the inner part of the envelope, and they do not support the presence of very high temperatures (T ~ 500-1000 K) in the inner part of the envelope (at about 5 × 10¹⁵ cm from the central star) suggested by a previous study. We also find that a mass-loss rate of 3.25 × 10^-5M_☉ yr^-1 and a distance of 150 pc provide the best agreement between our model results and observations.

The HubbleSpaceTelescope 2000-8000 Å spectrum of SN 1987A observed on 1995 January 7 (7.87 yr after the explosion) is dominated by Hα and UV lines, including Mg II-Mg I λ2825 (equal to Hα in luminosity), Fe II UV 2 (two-thirds the intensity of Hα), Fe II UV 3 (one-half the intensity of Hα), and a 3730 Å emission feature identified with a blend of [O II] λ3727 and Fe I emission lines. [O I] λ6300 and lines of [Ca II] and Na I, as well as some Fe II optical forbidden and permitted lines are present at visual wavelengths. Also present are a number of weak emission features, which are presumably metal lines produced by photon degradation as a result of reprocessing of UV radiation into metal lines. Modeling the Mg II-Mg I lines provides the velocity of the outer visible radius of the envelope, 9000 ± 500 km s^-1 in the Mg II λ2800 line, which is consistent with the earlier direct HST imaging at near-UV wavelengths. The UV/optical emission lines originate from the radioactive luminescence of the cool gas (T ≈ 130-160 K). The metal lines reflect the instantaneous reprocessing of the energy deposited from ⁴⁴Ti radioactive decays through collisions with fast electrons, while the Hα emission primarily comes from the recombination of previously ionized hydrogen. The overall luminosity of the Fe II emission lines, ~10³⁵ ergs s^-1, can be explained if the bulk of the positrons from a mass (1-2) × 10^-4M_☉ of ⁴⁴Ti release their energy in the iron-rich material, which suggests the presence of a magnetic field B > 5 × 10^-13 G prohibiting the escape of positrons into oxygen and hydrogen gas. The ionized fraction in the iron-rich material is small (0.2-0.3), and the total UV/optical emission from Fe I should be comparable to that from Fe II. Most of the 10³⁶ ergs s^-1 deposited by the ⁴⁴Ti positrons should be emitted in the Fe II 26 μm line. The observed Hα luminosity decrease, by 5 orders of magnitude between the ages of 1 to 8 yr, is reproduced in a time-dependent model of ionization and cooling with the "standard" amount of radioactive nuclides. However, an additional source of energy at the present epoch with a deposition rate 30 ergs s^-1 g^-1 (≈ 10³⁶ ergs s^-1 in the whole envelope) is not ruled out. The present average temperature in the hydrogen envelope predicted by the time-dependent model is 130 K, which is lower than the value T ≈ 350 K obtained from the observed Balmer continuum shape. However, the shape is affected by a possible contribution of metal lines to the Balmer continuum. The luminosity of the [O I] λ6300 doublet is consistent with that expected for the deposited energy of γ-rays from (1-2) × 10^-4M_☉ of ⁴⁴Ti for an assumed 1.5-2 M_☉ of oxygen. If the oxygen mass does not exceed 2 M_☉, 1 × 10^-4M_☉ is a lower limit for the amount of ⁴⁴Ti in SN 1987A. The maximum fraction of the ⁴⁴Ti positron energy deposited into oxygen-rich material does not exceed 5%, which is consistent with positron trapping in Fe-rich material. The [O I] λ6300 line intensity rules out the presence of a central source of γ-radiation (hν > 100 keV) with a luminosity L_γ > 4 × 10³⁶ ergs s^-1.

Line asymmetries and shifts are powerful tools for studying velocity fields in the stellar photospheres. Other effects, however, could also generate asymmetries, blurring the information of the velocity patterns. We have studied the shifts and asymmetries induced in the profiles of spectral lines by pressure effects. The best theoretical and experimental data on line broadening and shifts caused by collisions with atomic hydrogen were used to analyze the Na I D and three Ca I lines. Line bisectors of synthetic spectra computed with accurate data for the Na I and Ca I lines are compared with very high resolution, high signal-to-noise ratio solar spectra and indicate that pressure broadening reproduces the wings of the observed lines, but pressure shifts introduce neither asymmetries nor shifts comparable to the observed ones.

We have used the ASCA and ROSAT X-ray satellites to probe the coronae of a sample of nine solar-like G stars. These stars are all ostensibly single with ages ranging from 70 Myr to 9 Gyr and have X-ray luminosities ranging from 1 to 500 times that of the quiet Sun. Specifically, we investigate the dependence of the coronal temperature and emission measure structure of these stars on age and rotation period.

In the younger stars, a considerable portion of the volume emission measure resides at very high temperatures, reaching up to ~20-30 MK in EK Dra. Such temperatures are comparable to temperatures that are achieved on the Sun during short flaring episodes. In two-temperature fits to ROSAT data, the higher temperature decays rapidly within the first few 100 Myr; the decay may be described by an inverse power law, T_hot ∝ age^-0.3. We also find a power-law dependence between the total X-ray luminosity and the higher temperature L_X ∝ T. We interpret this as evidence of a decrease in the efficiency of high-temperature coronal heating as a solar-like star ages and its rotation slows down. A reconstruction of the coronal differential emission measure (DEM) distribution in three of the stars using ASCA data indicates a bimodal distribution in temperature, with the hotter plasma at 12-30 MK and the cooler plasma below 10 MK. We infer, for the first time, a consistent evolution of the DEM structure in a solar-type star. The emission measure of the hotter component rapidly decreases with age and becomes unimportant at ages beyond ~500 Myr. The emitted X-ray emission of the young Sun thus rapidly softened, which had important implications for the young planetary atmospheres. We suggest that the high-temperature component is the result of superimposed but temporally unresolved flaring events and support this picture by reconstructing the time-integrated (average) emission measure distribution of a typical solar X-ray flare. Radio observations of active stars fit well into this picture and suggest that the presence of nonthermal electrons in coronae is linked to the presence of hot ( > 10 MK) plasma, very much the same situation as in solar flares. We find, however, that radio emission saturates, if at all, at smaller rotation periods than does X-ray emission.

Eruptive solar events like flares and coronal mass ejections are thought to involve the release of energy stored in magnetic fields. In a mass ejection, the coronal magnetic field opens to interplanetary space and some 10¹⁶ g of material are propelled outward along the open field lines at a speed of typically 350 km s^-1. A considerable body of research shows that the energy in force-free magnetic fields, in which electric currents flow parallel to the field, may be sufficient to open the coronal magnetic field but not additionally to accelerate the ejected material nor to lift it against solar gravity. Thus a purely magnetic explanation for coronal mass ejections must involve cross-field currents—that is, currents with components perpendicular to the magnetic field direction.

This paper explores the energetics of a family of simple solutions for an axisymmetric corona in magnetostatic equilibrium including cross-field currents. We extend previous treatments to show that it is possible to build up magnetic energy in excess of that in the presumed fully open state that follows a mass ejection, even if the latter still contains horizontal pressure gradients. We show further that energies in excess of the energy of a fully open state without horizontal pressure gradients are attainable over a wider range of parameters than previously reported. However, our results are tempered by our demonstration that excess mass associated with pressure gradients that balance the magnetic forces of cross-field currents always results in negative total energy. This is because the excess mass is great enough that its negative gravitational energy exceeds in magnitude the sum of the magnetic and thermal energies that also arise from the presence of excess mass. We suggest that nonlinear solutions, involving more realistic density distributions that model the helmet structure and cavity of coronal streamers, may overcome this problem of negative excess energy.

Time-dependent dynamical calculations of the radiating gas in solar coronal flux tubes are used to identify features of UV spectral-line profiles that can reveal the direction in which wave energy flows through the solar transition region. The profile features survive spatial and temporal averaging through nonlinear dependencies of line emission coefficients on thermal properties of the plasma that are correlated with fluid velocities. This approach can be applied to stellar and other unresolved sources, as well as the solar atmosphere. It can be regarded as a new angle of attack on the long-standing problem of determining coronal heating mechanisms. The approach requires low noise data of high spectral resolution. Therefore, it can take advantage of some unique properties of the SUMER instrument on SOHO. We make specific predictions for SUMER data that, in principle, can test whether energy propagates upward or downward in coronal flux tubes, thus allowing one to discriminate between competing theories of coronal heating. We are acquiring SUMER data in an attempt to do this.

Erratum to ApJ, 464, 765 (1996)

We show that evolution of the luminosity density of galaxies in the universe provides a powerful test for the geometry of the universe. Using reasonable galaxy evolution models of population synthesis that reproduce the colors of local galaxies of various morphological types, we have calculated the luminosity density of galaxies as a function of redshift z. Comparison of the result with recent measurements by the Canada-France Redshift Survey in three wave bands of 2800 Å, 4400 Å, and 1 μm at z < 1 indicates that the Λ-dominated flat universe with λ₀ ~ 0.8 is favored, and the lower limit on λ₀ yields 0.37 (99% C.L.) or 0.53 (95% C.L.) if Ω₀ + λ₀ = 1. The Einstein-de Sitter universe with (Ω₀, λ₀) = (1, 0) and the low-density open universe with (0.2, 0) are, however, ruled out with 99.86% C.L. and 98.6% C.L., respectively. The confidence levels quoted apply unless the standard assumptions on galaxy evolution are drastically violated.

We have also calculated a global star formation rate in the universe to be compared with the observed rate beyond z ~ 2. We find from this comparison that spiral galaxies are formed from material accreted over an extended period of a few Gyr, while elliptical galaxies are formed from an initial starburst at z ≳ 5 that supplies enough metals and ionizing photons in the intergalactic medium.

New Hubble Space Telescope WFPC2 observations of the lensed double QSO 0957+561 will allow stronger constraints on the lens mass distribution and hence will refine the derived value of H₀. We first present improved optical positions and photometry for the known components of this lens. The optical separation between the A and B quasar images agrees with VLBI data at the 10 mas level, and the optical center of the primary lensing galaxy G1 coincides with the VLBI source G' to within 10 mas. The best previous model for this lens (see the recent work of Grogin and Narayan) is excluded by these data and must be reevaluated.

Several new resolved features are found within 10'' of G1, including an apparent fold arc with two bright knots. Several other small galaxies are detected, including two that may be multiple images of each other. We present positions and crude photometry of these objects.

We study the possible correlation of a selected sample of 87 gamma-ray bursts that can be localized to ≤6 deg² with 3616 Abell clusters of galaxies (with |b| > 30°) for which we have determined angular sizes. The intention is to confirm whether the bursts are related to Abell clusters of galaxies, as has been claimed by researchers who found a 95% confidence level correlation between Abell clusters and a subsample of 136 events from the BATSE 3B catalog.

We find no correlation between the positions of gamma-ray bursts and those of Abell clusters. In fact, the number of observed events toward Abell clusters is even smaller than that expected to occur at random.

We consider possible interpretations of the recently detected X-ray afterglow from the gamma-ray burst source GRB 970228. Cosmological and Galactic models of gamma-ray bursts predict different flux and spectral evolution of X-ray afterglows. We show that models based on adiabatic expansion of relativistic forward shocks require very efficient particle energization or postburst reacceleration during the expansion. Cooling neutron star models predict a very distinctive spectral and flux evolution that can be tested in current X-ray data.

We report on a comprehensive radio monitoring program of the bright gamma-ray burster GRB 970111. These VLA observations were made at a frequency of 1.4 GHz (λ = 20 cm) and span a range of postburst timescales between 28 hr and 1 month. Despite extensive sampling at submillijansky sensitivities, no radio source was detected above 0.5 mJy in the current best error box (~14 arcmin²) for GRB 970111. A highly unusual radio source, VLA J1528.7+1945, was seen to drop in flux density by a factor of 2 in our monitoring period, but it lies outside the error box, and thus it is unlikely to be related to GRB 970111. Cosmological fireball models of gamma-ray bursts make predictions of late-time emission occurring at longer wavelengths. The absence of a flaring or fading radio counterpart to GRB 970111 provides strong constraints on these models.

We present the results of OSSE observations of the soft gamma-ray continuum emission from the Galactic plane at longitude 95°. Emission is detected between 50 and 600 keV where the spectrum is fitted well by a power law with photon index -2.6 ± 0.3 and flux (4.0 ± 0.5) × 10^-2 photons s^-1 cm^-2 rad^-1 MeV^-1 at 100 keV. This spectral shape in this range is similar to that found for the continuum emission from the inner Galaxy, but the amplitude is lower by a factor of 4. This emission is due to either unresolved and previously unknown point sources, or diffuse electron bremsstrahlung, or a combination of the two. Simultaneous observations with OSSE and smaller field-of-view instruments operating in the soft gamma-ray energy band, such as X-ray Timing Explorer or Beppo-SAX, would help resolve this issue. If it is primarily diffuse emission due to nonthermal electron bremsstrahlung, as is the >1 MeV Galactic ridge continuum, then the power in low-energy cosmic-ray electrons exceeds that of the nuclear component of the cosmic rays by an order of magnitude. This would have profound implications for the origin of cosmic rays and the energetics of the interstellar medium. Alternatively, if the emission is diffuse and thermal, then there must be a component of the interstellar medium at temperatures ~10⁹ K.

We have obtained simultaneous two-dimensional spectroscopy of the Ca II triplet lines in the central 24'' × 20'' of NGC 1068, using an optical fiber system. The stellar velocity field derived from these data shows a kinematic off-centering between the inner (r < 3'') and outer (r > 5'') regions, indicative of a nonsymmetric gravitational potential. Both regions exhibit kinematic minor axes aligned with PA ~ 0° and could have similar systemic velocities. However, their kinematic centers are displaced by ~25. We propose that two kinematically different stellar systems rotating around parallel but shifted axes are present in NGC 1068. The origin of this peculiarity is discussed in relation to the gravitational potential of the galaxy and the hypothesis of a minor merger event.

X-ray spectra of a significant fraction of cooling flow (CF) clusters of galaxies indicate the presence of large columns of "cold" absorbing material. The physical nature of the absorbing medium remains a mystery. Searches for H I in this medium using the 21 cm hyperfine structure line yielded null results in most cases. The Lyα absorption cross section is ≥10⁷ times larger than that for the 21 cm line, and it can therefore be used as a much more sensitive probe of H I in clusters. This method is applied to the Perseus CF cluster, where a medium resolution (~250 km s^-1) UV spectrum is available. The upper limit on the H I column obtained using Lyα is at least ~2000 smaller than implied by X-ray spectra, indicating that the X-ray absorber is exceedingly devoid of H I. Higher resolution UV spectra with the Hubble Space Telescope may improve the H I column limits by an additional factor of up to ~4000 and thus put severe constraints on the nature of the X-ray-absorbing medium. Most CF clusters have emission-line nebulae at their center, and this method can therefore be applied in a significant number of CF clusters.

Nebular spectra of Type Ia supernovae are modeled by including important atomic processes and reliable atomic data for a representative set of Type Ia explosion models. The calculated spectra are compared with the observed optical spectra of spectroscopically normal Type Ia supernovae, leading to strong constraints on the explosion models. It is concluded that sub-Chandrasekhar-mass models are favored, but Chandrasekhar-mass models might not be entirely ruled out.

We report the discovery of near-sinusoidal radial velocity variations of the G0V star ρ CrB, with period 39.6 days and amplitude 67 m s^-1. These variations are consistent with the existence of an orbital companion in a circular orbit. Adopting a mass of 1.0 M_☉ for the primary, the companion has minimum mass about 1.1 Jupiter masses and orbital radius about 0.23 AU. Such an orbital radius is too large for tidal circularization of an initially eccentric orbit during the lifetime of the star, and hence we suggest that the low eccentricity is primordial, as would be expected for a planet formed in a dissipative circumstellar disk.

Using the Rossi X-Ray Timing Explorer, we have measured ~27 μs time delays in 830 Hz quasi-periodic oscillations (QPOs) between 4-6 and 11-17 keV in 4U 1608-52, with high-energy photons lagging low-energy photons, and found upper limits to the time delays of 45 μs between 2-6.5 and 6.5-67 keV in ~730 Hz QPOs in 4U 0614+091 and 30 μs between 8.7-12.4 and 12.4-67 keV in ~870 Hz QPOs in 4U 1636-53. We also find that the cross-coherence function between QPOs at different energies is greater than 0.85 with 95% confidence in 4U 1608-52 and 4U 1636-53. If Compton upscattering of low-energy X-rays in a region with an optical depth of a few is responsible for the delays, then the Compton upscattering region is between a few kilometers and a few tens of kilometers in size.

We have detected high-frequency (HF) quasi-periodic oscillations (QPOs) from the low-mass X-ray binary 4U 1820-303 during observations performed in 1996 October using the Rossi X-ray Timing Explorer. The QPOs are visible when the source occupies the low-state luminosity range L_X = 2.4-3.1 × 10³⁷ ergs s^-1 (2-20 keV, at 6.4 kpc); the centroid frequency of the main QPO peak varies between 546 ± 2 Hz and 796 ± 6 Hz and is tightly correlated with the source count rate. The measured QPO widths are typically ~20 Hz, with mean rms amplitude 4.1% ± 0.3%. At the upper end of this luminosity range a second significant QPO peak appears with frequency 1065 ± 7 Hz, width 40 ± 20 Hz, and rms amplitude 3.2% ± 0.8%. When both QPOs are visible simultaneously, the difference between their frequencies is 275 ± 8 Hz. When the source brightens beyond L_X = 3.1 × 10³⁷ ergs s^-1 (~10% of the Eddington limit for a helium-rich envelope), neither QPO is detected. Neither the magnetospheric beat frequency model nor the sonic point model of HF QPOs provides a perfect explanation of the phenomenology we observe.

These results represent the first detection of kilohertz QPO activity in a globular cluster X-ray binary, and provide a new method of directly comparing the properties of cluster and noncluster neutron star binaries. If the highest QPO frequency we observe is identified with the marginally stable orbit in the accretion disk, the neutron star mass may be ~2 M_☉, 35%-50% more massive than usually assumed. This may have consequences for the current evolutionary scenarios for this source and also for the debate about the evolution of millisecond pulsars in globular clusters.

We present new calculations that strongly reinforce the idea—originally proposed by Charpinet et al.—that pulsation modes are driven through an opacity bump due to a local enhancement of the iron abundance in the envelopes of sdB stars. Our improved models incorporate nonuniform iron abundance distributions obtained through the condition of diffusive equilibrium between gravitational settling and radiative levitation. They also include special Rosseland opacity tables that take into account the large variations of the iron abundance about the cosmic value that are predicted by equilibrium radiative levitation theory. For representative models with M = 0.48 M_☉ and log g = 5.8, we find strong instabilities for low-order radial and nonradial (p and f) pulsation modes in the range 36,500 K ≳ T_eff ≳ 29,000 K. The four pulsating sdB stars currently known all have effective temperatures in that range. In addition, one of our models with T_eff = 34,000 K has a band of unstable modes with periods in the range 116-195 s, in excellent agreement with those of the known pulsators. We therefore claim that our proposed iron bump mechanism provides a natural explanation for the instabilities found in the newly discovered class of pulsating sdB stars.

We have made high-resolution (4'') observations of the 2.6 mm continuum and CO (1-0) line emission associated with the young stellar object DG Tau B. The aperture-synthesis CO map clearly shows that redshifted emission is symmetrically distributed about the DG Tau B optical jet and extends at least 6000 AU to the northwest of the unresolved continuum source. By contrast, blueshifted CO emission is confined to a compact region, which is less than 500 AU in radius. Within 1000 AU of the star, the redshifted CO emission is elongated, brightest along its central axis, and unresolved in the transverse direction. Beyond 1000 AU, the flow broadens and begins to bifurcate. The morphology of the CO emission provides compelling evidence that the extended molecular outflow is driven by the highly collimated atomic jet. The spatial correspondence between knots in the optical jet and successive broadenings of the outflow supports the hypothesis that the molecular outflow is produced by the action of multiple working surfaces in a time-varying jet.

We report on laboratory measurements of molecular hydrogen formation and recombination on an olivine slab as a function of surface temperature under conditions relevant to those encountered in the interstellar medium. On the basis of our experimental evidence, we recognize that there are two main regimes of H coverage that are of astrophysical importance; for each of them we provide an expression giving the production rate of molecular hydrogen in interstellar clouds.

The highly polar ring-chain C₅H₂, a carbene with a singlet electronic ground state, has been detected with a Fourier transform molecular beam microwave spectrometer in the same diacetylene-neon discharge in which a carbene with the same elemental formula but a linear backbone was recently found. Thirteen a- and b-type rotational transitions between 6 and 27 GHz were measured to 1-2 kHz, and precise values of the three rotational constants and two centrifugal distortion constants were determined. With these, the entire microwave and millimeter-wave spectrum of C₅H₂ can be calculated to an accuracy of 0.2 km s^-1 or better. The identity of the new molecule as C₅H₂, which is fairly certain because of the close agreement between the measured rotational constants and those calculated ab initio, is conclusively confirmed by detection of the totally deuterated and two partially deuterated isotopic species at exactly the expected isotope shifts.

Following the recent laboratory detection of the ring-chain carbene C₅H₂, we have now detected C₇H₂, the next longer member of the sequence formed by substituting a carbon chain for one of the hydrogen atoms of cyclic C₃H₂. The strongest lines, again obtained with a diacetylene-neon discharge, were roughly 15 times weaker than those of C₅H₂. Enough rotational lines were detected in both the normal and doubly deuterated isotopic species to be certain of the identification and to obtain the rotational and leading centrifugal distortion constants accurately enough for calculation of the entire rotational spectrum below 300 GHz to a fraction of 1 km s^-1 in equivalent radial velocity. The ring chain here is calculated to be the most stable isomer of C₇H₂ and extremely polar, so it is of considerable astronomical interest. Several other isomers of C₇H₂ may be detectable in the laboratory and in space with present techniques.