Ionic abundances of N, O, S, and He are presented for the ES, E5, and NS condensations around Eta Carinae. The spectra$\sp1$ show strong (NII) $\lambda\lambda$6548,6583 while (OII) $\lambda$3727 is weak and (OIII) $\lambda$5007 is not clearly observed. Analysis is complicated by large velocity dispersions (250-750 km/sec) and spatial ionization variations. However, the (NII) and (SII) lines imply electron temperatures of about 10-20 $\times$ 10$\sp3$K and electron densities of 2-3 $\times$ 10$\sp3$cm$\sp{-3}$ (except NS). The source of ionization is not understood, but shock ionization may have a major role. Elemental abundances are not generally obtainable; however, an ionization curve similar to that of the S condensation would imply N(He)/N(H) = 0.14 $\pm$ 0.03 (for the condensations). The helium and nitrogen enrichments show that Eta Carinae is ejecting nuclear processed material. ftn$\sp1$Observations made by R. J. Dufour, visiting astronomer Cerro Tololo Interamerican Observatory, on 1984 March 15 and 16 using the 1.5m telescope with SIT vidicon.

Convective studies involving ion flow in the Venus dayside ionosphere are used to understand the nature of the observed O$\sb2\sp+$ distribution. Flows are modeled (not calculated) from previous data and theory which are incorporated to determine the ion distributions for O$\sp+$, CO$\sb2\sp+$, and O$\sb2\sp+$. Current understanding of ion chemistry and related parameters are used in solving the continuity equation in two dimensions to provide ion profiles at different solar zenith angles. To explain the O$\sb2\sp+$ distribution, downward flow with speeds of approximately 500 m/s are required.

Observations have been made of Comet Giacobini-Zinner over the wavelength interval 3000 A-5700 A and of Comet Halley over the wavelength interval 3000 A-10,000 A. A photoelectric spectrophotometer with a 40 A bandpass in the second order was used. Band fluxes from OH and CN have been obtained for Giacobini-Zinner. Band fluxes from OH, NH, C$\sb{2}$, and CN have been obtained for Halley. Swan band sequence flux ratios have been calculated for Halley and compared to resonance fluorescence theory and previous observations. The comparison suggests a reexamination of the molecular constants used in theoretical calculations and the validity of the statistical equilibrium assumption.

The results of an investigation of the optical spectra of thirty-four emission nebulae in thirteen irregular and late-type spiral galaxies are presented. The program galaxies, with the number of nebulae observed in each shown in parentheses, were: NGC55 (3), IC1613 (2), NGC1156 (1), NGC1569 (3), NGC2366 (2), IC2574 (2), NGC4214 (3), NGC4236 (1), NGC4449 (4), NGC4656 (3), GR8 (1), NGC6822 (8), and IC5152 (1). Spectrophotometric data were gathered during 1978 and 1979 at the Cerro Tololo Interamerican Observatory (CTIO) using the SIT vidicon on the 4.0-m and 1.5-m telescopes, and at the Kitt Peak National Observatory (KPNO) using the Intensified Image Dissector Scanner (IIDS) on the 2.1-m telescope. From measurement of diagnostic emission-line strengths, the relative abundances of He, N, O, Ne, and S (in some cases) were calculated with respect to H for each nebula. Electron densities were derived from the ratio of the {S II} line intensities, I(6716)/I(6730), while temperatures were obtained from the ratio of the {O III} line intensities, I(4363)/I(5007). The observations reported in this work confirm the results of earlier investigations involving much smaller samples of irregular and late-type spiral galaxies; the galaxies examined during this study were found to be consistently underabundant in the metals with respect to earlier-type galaxies. The average abundances (by number) for all of the galaxies studied are: log(He) = 10.88, log(N) = 6.72, log(O) = 8.23, log(Ne) = 7.64, and log(S) = 7.20 (log(H) = 12.00). Interpretation of the abundances in the context of the closed-box model for galactic evolution provided yields of 5.9 x 10('-3), 9.5 x 10('-4), 3.2 x 10('-4), and 2.1 x 10('-3) for He, O, Ne, and Z respectively. These yields are consistent with an initial mass function of the form (theta) (PROPORTIONAL) (1/M)('(alpha)) with (alpha) in the range 7/3 to 10/3 for stars more massive than 10 M(,(CIRCLE)). In addition, a value for the primary yield of N of 2.5 x 10('-5) was determined. Examination of the metallicity parameter, Z, versus the astration parameter, ln(1/(mu)), suggests that the gassiest galaxies have experienced the least astrophysical processing. This result, coupled with a correlation that was found for Z versus log(total galaxy mass), suggests that the average star formation rate decreases as galaxy mass decreases. Finally, the pregalactic He abundance, by mass, was determined (Y(,p) = .216) and interpreted in the context of models of the standard big-bang. From this value of Y(,p) the present average density of the Universe is found to be 1.06 x 10('-31) g/cm('3)--insufficient to close the Universe.

The optical emission lines of the Cygnus Loop, IC 443, Vela X, Puppis A, LMC N49, and LMC N63A were observed at very high spatial and spectral resolution with 4-m echelle spectrographs. Results indicate that the velocity dispersions of knots in the filaments of the nearby galactic supernova remnants (SNRs) result primarily from turbulence within the emitting regions. The line-of-sight velocity dispersions (half-widths at half maximum) within the observed knots are on the order of 10-30 km s('-1), including associated thermal velocity dispersions of roughly 15 km s('-1) for hydrogen, 6 km s('-1) for oxygen, and 4 km s('-1) for nitrogen. In general, the knots move randomly relative to each other with speeds of 10-30 km s('-1), although relative speeds as large as 80 km s('-1) are observed in Puppis A. Occasionally, diffuse components of the filaments are observed. These diffuse components have velocity dispersions between 55 and 85 km s('-1), but the relative contributions of turbulent and thermal motion to the dispersions are unknown. The galactic observations are interpreted in terms of the "microscopic" structure of SNR filaments and knots. Velocity dispersions between 5 and 140 km s('-1) are observed in the distant SNRs of the Large Magellanic Cloud (LMC). Velocity dispersions of 5-20 km s('-1) are associated with spatially extended areas that possess the same radial velocities as the LMC. The associated thermal velocity dispersions for these areas are approximately 13 km s('-1) for hydrogen, 3.5 km s('-1) for nitrogen, 3.3 km s('-1) oxygen, and 2.1 km s('-1) for sulfur. Larger velocity dispersions of 30-140 km s('-1) are associated with smaller areas that are moving relative to the spatially extended areas. These relative velocities may be as large as 250 km s('-1). The contributions of thermal and bulk motion to the velocity dispersions of the smaller areas cannot be determined with kinematic methods. The LMC observations are interpreted in the context of SNR evolution in cloudy interstellar media.

The interaction of Birkeland (magnetic-field-aligned) current carriers and the Earth's magnetic field results in electrostatic potential drops along magnetic field lines. The statistical distributions of the field-aligned potential difference (phi)(,(PARLL)) were determined from the energy spectra of electron inverted "V" events observed at ionospheric altitude for different conditions of geomagnetic activity as indicated by the AE index. Data of 1270 electron inverted "V"'s were obtained from Low-Energy Electron measurements of the Atmosphere Explorer-C and -D Satellite (despun mode) in the interval January 1974-April 1976. In general, (phi)(,(PARLL)) is largest in the dusk to pre-midnight sector, smaller in the post-midnight to dawn sector, and smallest in the near noon sector during quiet and disturbed geomagnetic conditions; there is a steady dusk-dawn-noon asymmetry of the global (phi)(,(PARLL)) distribution. As the geomagnetic activity level increases, the (phi)(,(PARLL)) pattern expands to lower invariant latitudes, and the magnitude of (phi)(,(PARLL)) in the 13-24 magnetic local time sector increases significantly. The spatial structure and intensity variation of the global (phi)(,(PARLL)) distribution are statistically more variable, and the magnitudes of (phi)(,(PARLL)) have smaller correlation with the AE-index, in the post-midnight to dawn sector. A strong correlation is found to exist between upward Birkeland current systems and global parallel potential drops, and between auroral electron precipitation patterns and parallel potential drops, regarding their mophology, their intensity and their dependence of geomagnetic activity. An analysis of the fine-scale simultaneous current-voltage relationship for upward Birkeland currents in Region 1 shows that typical field-aligned potential drops are consistent with model predictions based on linear acceleration of the charge carriers through an electrostatic potential drop along convergent magnetic field lines to maintain current continuity. In a steady state, this model of simple electrostatic acceleration without anomalous resistivity also predicts observable relations between global parallel currents and parallel potential drops and between global energy deposition and parallel potential drops. The temperature, density, and species of the unaccelerated charge carriers are the relevant parameters of the model. The dusk-dawn-noon asymmetry of the global (phi)(,(PARLL)) distribution can be explained by the above steady-state (phi)(,(PARLL)) process if we associate the source regions of upward Birkeland current carriers in Region 1, Region 2, and the cusp region with the plasma sheet boundary layer, the near-Earth plasma sheet, and the magnetosheath, respectively. The results of this study provide observational information on the global distribution of parallel potential drops and the prevailing process of generating and maintaining potential gradients (parallel electric fields) along auroral magnetic field lines.

Since the Goldreich and Julian model for a pulsar magnetosphere was introduced attempts to form a self-consistent theory based on their model have met with failure. This thesis contends that the Goldreich and Julian magnetosphere is not unique, that more plausible magnetospheres exist and that they are characterized by vacuum gaps separating the charge-separated plasmas in the pulsar magnetosphere. This work models a number of such magnetospheres using discrete charges that are generated from the pulsar surface charge. The models are physically differentiated by their net charge, with one model having the net charge of the Goldreich and Julian model and the others less than this in regularly spaced intervals. It is demonstrated that static solutions exist which possess corotational regions above the polar regions and both a corotational and superrotational region about the equator. The result is in accord with Pilipp's theorem concerning the impossibility of having only a corotational magnetosphere with vacuum gaps. The finite size of the space-charge configuration places the magnetosphere well within the light cylinder of all but one of the known radio pulsars (assuming that a pulsar has the standard radius). The confinement of the space-charge within the light cylinder raises questions about the present assumptions made for pulsars. The configurations are totally static as the surface has been made emission-free in the aligned case, and the results suggest that one must at least look at the non-aligned case to explain the pulsar emission process. Possibly some of the assumptions about radio pulsars and their environment must be revised.

New observational data on the Cygnus Loop supernova remnant (SNR) include: (1) A detailed high resolution comparison of X-ray and optical emission for a field in the SE; (2) A map of the O III electron temperature for the field previously studied by Hester, Parker, and Dufour (1983); and (3) CCD imagery of the NE limb in the light of four emission lines. A wide range of new and existing observations of the Loop are for the first time interpreted within the context of a single physical description. The Cygnus Loop is not an evaporative SNR evolving into the McKee and Ostriker (1977) ISM, nor are tiny cloudlets necessary to explain its morphology. The data show the Cygnus Loop to be evolving into a medium consisting primarily of an intercloud phase with n(,0) (TURNEQ) .1 cm('-3) containing clouds with parsec dimensions and n(,0) (LESSTHEQ) 10 cm('-3). The optical emission arises from extensive sheet-like radiative shock fronts driven into the clouds. These fronts locally form the outer boundary of the remnant. The appearance of X-ray emission outside the optical emission on the limbs is due solely to projection effects. The distorted and bumpy shock front is shown to give rise in projection to the filamentary morphology of the remnant. Variations in spectral characteristics of the optical emission are due to a general increase in completeness of the recombination region and decrease in v(,s) with increasing distance away from the edge of a cloud along the face of the front. Models show that the sheet geometries needed to explain the shapes and extents of observed filaments lead to surface brightness and kinematic properties of the emission which are in very good agreement with observation. These geometries require density inhomogeneities within clouds of only (TURN)10% on scales of (TURN)10('17) cm. A thick ((TURN)2") recombination region, resulting from field limited compression, is resolved for the first time. The expected relationship between nonsteady flow and high O III electron temperature is confirmed. Evaporation of clouds cannot account for the X-ray data. The brightest X-ray emission in the Loop comes just interior to young optical emission, and the gas has a pressure of at least six times the average for the Loop. Regions around clouds which should have undergone significant evaporation are not bright in X-rays. The X-ray data can be explained within the context of the large cloud picture by the combined effects of (1) a denser than average intercloud medium near clouds; (2) inertial pressurization of the X-ray emitting shell behind rapidly decelerating portions of the blast wave; and (3) strong compression of the hot gas by reflected and bow shocks around clouds.

Calibrated digital surface photometry in the standard UBV pass- bands has been used to prepare two dimensional maps to examine the structure and stellar populations in NGC 5128 (Cen A). The main body of the galaxy exhibits colors and stellar populations slightly bluer than those of an old elliptical galaxy out to r (TURN) 176". Beyond this radius, excess emission in the inner halo contributes significantly to the U passband only, probably due to line emission from material swept into the halo from the dust lane. A gradient in B-V of (TURN)0('m)01/ 1.5 (D/3) kpc is evident along the major axis in both hemispheres. Four large diffuse arcs have been detected on the B-V raster. The outermost two are in the northern hemisphere and have similar B-V colors of (TURN)-0.11. The innermost pair are on opposite sides of the nucleus at r (TURN) 107" with B-V = 0.08, suggesting that a ring or sphere may exist at this radius, centered on the nucleus. The dust lane is found to be a hollow cylinder of inner and outer radii 2.2 (D/3) kpc and 2.6 (D/3) kpc, respectively, tilted at 80(DEGREES) (+OR-)2 to the plane of the sky. The most intense star formation regions lie on the northern rim on the near side of the lane and the southern rim on the far side. The circular shape of the lane constrains the main body of the galaxy to being prolate or only very mildly elliptical. There is evidence that the lane is being fed from the intergalactic medium along the SE edge at a rate of (TURN)1 M(,o)/year. A minimum mass for the lane is calculated to be 10('8) M(,o) based on the observed extinction in the lane to the SE of the nucleus. The presence of the inner optical jet has been confirmed and shown to coincide with features in the x-ray jet. A parallel dust ridge is present on either side of the jet, extending at least the full length of the jet. No counterjet has been detected in the southern hemi- sphere. Excess optical emission from the edges of the radio bridge connecting the nucleus and the lobes has probably been detected.

Magnetometer and plasma science data from the Voyager 1 and Voyager 2 encounters of Jupiter are combined with earlier theoretical analysis by T. W. Hill and V. M. Vasyliunas to determine the total mass outflow rate and radial bulk velocity of the magnetospheric plasma in the Jovian plasma sheet, as well as the height integrated Pedersen conductivity of the Jovian ionosphere. The analysis requires accurate determination of the orientation of the plasma sheet using magnetometer data surrounding the plasma sheet crossing points. This analysis has been applied to fifteen crossings of the Jovian plasma sheet on both the day side and the night side. Results obtained from this analysis are believed to be accurate within a factor of two, and confirm previous order-of-magnitude estimates. Evidence of an enhanced outflow of plasma in the "active hemisphere" of System III longitude is observed on the dayside. On the night side, evidence of plasma inflow and superrotation is seen in the "inactive hemisphere." These two observations lend support to the corotating-convection model of plasma transport in the Jovian magnetosphere.