Proefschrift--Utrecht. / "Literature": p.[77]-78.

Thesis (Ph. D.)--University of Michigan, 1938. / Caption title. Thesis note on p. 1.

Thesis (Ph. D.)--University of Michigan, 1943. / "Preprinted from the Astrophysical journal, 98 (November, 1943)."

Tesis doctoral--Universidad de Madrid. / At head of title: Consejo Superior de Investigaciones Cientificas. Instituto Nacional de Geofisica. no. 6. Spanish, French, English and German summaries. Bibliografía: p. xi-xiii.

Thesis (Ph. D.)--Ohio State University, 2005. / Title from first page of PDF file. Document formatted into pages; contains xv, 286 p.; also includes graphics (some col.). Includes bibliographical references (p. 275-286). Available online via OhioLINK's ETD Center

Thesis (Ph. D.)--University of Florida, 1981. / Description based on print version record. Typescript. Vita. Includes bibliographical references (leaves 73-75).

Thesis (Ph. D.)--University of Chicago, 1913. / "Reprinted from the Astrophysical journal, vol. XXXVII, no. 1." Includes bibliographical references. Also available on the Internet.

Observations of four massive x-ray binary stars are presented, based on data accumulated between February 1989 and August 1991. Using modern techniques of spectroscopic data analysis, velocity curves are derived for three of these systems. Two of these curves (SMC X-1, QV Nor) yield very precise mass estimates for the component stars, while the third (Cen X-3) offers a constraint on the possible masses. The fourth system (X Per) is not shown to exhibit periodic variations, despite an extensive study conducted over more than two years. For the two systems that yielded precise masses, the component neutron stars are shown to lie within the theoretical mass range based on theories of their formation via the supernova explosion of a helium star in a close binary system. This is a marked improvement on previous studies where both stars had estimated masses which lay well outside of the expected range. The derivation of these masses incorporates the use of non-Keplerian velocity corrections, arising from the non-spherical, asymmetrically illuminated primary stars. A study of the line profiles showed that the temperatures around both primary stars were consistent with the parameters in these calculations. For the third system, the inaccuracy of the published ephemeris resulted in a lack of observations at the times of maximum and minimum velocity. The semi-amplitude is thus not well constrained, but it is shown that the observations are consistent with the assumption of a normal mass neutron star secondary. The system is shown to have undergone a gradual decrease in its orbital period which follows a parabolic trend, suggesting substantial mass-transfer. For the fourth system, a periodicity analysis of 130 spectroscopic velocity measurements of a Be star, via Fouriergram and string-length techniques, failed to highlight any strong periodicity. The scatter in the data appears larger than would be expected for a non-variable B star. The absence of periodic velocity variations at the expected period is discussed in terms of the binarity (or otherwise) of the Be star. A transition from Be to shell-star or ordinary B star phase occurred during the study, which is not evident from the spectral variations observed in the blue.

Spectroscopic and photometric observations of eight contact/near-contact binaries are presented and analysed. Spectroscopic observations were obtained at 4200 Å (radial velocity spectra) and 6563 Å (hydrogen-alpha line profiles). New photometric observations were obtained at visual and infrared wavelengths, and other previously published light curves are also re-analysed. Absolute dimensions have been obtained for five systems; TY Boo, VW Boo, BX And, SS Ari and AG Vir, and their evolutionary positions discussed. Four of the systems are found to be in marginal but poor thermal contact, exhibiting regions of apparent "excess luminosity" in their light curves. A qualitative analysis of these "hot spot" regions has been attempted for the first time using spot models now incorporated into a light curve synthesis programme. Substantial time for this project was awarded on telescopes funded by the United Kingdom Science and Engineering Research Council (SERC), comprising 14 nights at the Issac Newton Telescope (INT) on La Palma, and 4 nights at the United Kingdom Infrared Telescope (UKIRT) on Mauna Kea. Additional observations were made during an 8 night commissioning run on the Jacobus Kapteyn Telescope (JKT) on La Palma, and extensive observations were made with the Twin Photometric Telescope (TPT) at St Andrews University Observatory between 1985 and 1989. These resulted in over 100 spectra at 4200 Å and over 50 spectra at 6563 Å (INT and JKT observations), over 300 infrared photometric observations (UKIRT), and over 3500 visual photometric observations (TPT). Of the five systems analysed in detail in this work, TY Boo appears to be a normal shallow-contact W-type system. Both VW Boo and BX And exhibit regions of "excess luminosity" around the ingress and egress of secondary minimum which are well modelled by a warm spot on the cooler component sitting symmetrically around the neck joining the pair. Such a phenomenon may be expected to arise naturally in systems which have come into contact but are not yet/currently in thermal contact, exhibiting a temperature difference between the components. BXAnd like other B-type systems seems to be reaching this contact state for the first time, but the position of VW Boo is uncertain, and whilst evidence that it could be in the "broken contact" state predicted by the TRO Theory is far from conclusive, its lower orbital angular momentum clearly marks the system as worthy of further study. SS Ari and AG Vir exhibit light curves with unequal quadrature heights. Attempts to treat the higher quadrature as a region of "excess luminosity" due to an energy transfer "warm spot" does not however provide a good model of this phenomenon. Since invoking a dark starspot model also does not provide a good explanation for such systems, it may be that this form of light curve distortion is due to an entirely different form of distorting surface phenomenon. Like BX And, AG Vir appears to be just reaching contact for the first time, but like VW Boo, the slightly lower angular momentum of SS Ari warrants further study.

Spectra were obtained from several sources for a fine abundance analysis of the extreme helium-rich star HD 168476. The atmospheric parameters are found using theoretical models and abundances determined for the ions identified in the spectra. The star is found to have an effective temperature of 14000°K, log(surface gravity) of 1.5 and a microturbulent velocity of 10 km/s. Over 1400 lines are identified on the spectra ranging in wavelength from 3100A to 4925A and 5490A to 6585A, and about 530 lines are suitable for the abundance analysis. It is confirmed that helium, carbon, nitrogen and neon are overabundant, with hydrogen and oxygen underabundant. An ultraviolet spectrum of the star is also obtained and lines identified. In addition to the ions found in the visible region of the spectrum, neutral ions are also present, indicating the possibility of a cool outer shell to the star. Photometric and spectroscopic observations are made to study the light and radial velocity of the star, to determine if it is variable. Statistical tests showed that the star is variable in its V magnitude, and most probably in its colours, as well as being variable in its radial velocity. No period is found and it is suspected that the variability may be complex. Several theories for the causes of the variability are discussed and a possible origin for the extreme helium-rich nature of the star mentioned.