I here present my search for benchmark systems with main sequence stars widely separated from ultracool dwarfs or white dwarfs. Different near-infrared and optical surveys have been used to create samples of L dwarf candidates from colour cuts and by exploring the faintest limits of available surveys. I started by using the Two Micron All Sky Survey, and then expanded my method in order to include sources from the UKIRT Infrared Deep Sky Survey, which has not yet been fully explored for L dwarfs. In order to search for wide main sequence star companions I have used stars from Hipparcos, Gliese and PPMXL. I then applied a set of selection rules based on photometry and proper motion to find these systems. This search led to the discovery of two new benchmark systems with genuine L dwarfs, that have been spectroscopically confirmed. A large sample of potentially new systems still awaits a complete analysis, and we thus expect more exciting discoveries to stem from the method utilized. A method to identify new late-T and Y dwarfs from the Wide-field Infrared Survey was also been developed, with the aim of using the surveys maximum sensitive limits. The technique relies on objects only detected in the W2-band and through a series of rejection methods 158 sources are retrieved with signal-to-noises ranging from 8-10. I have performed aperture photometry and measured proper motions for a set of candidates that were observed during the follow-up campaign. I then present a new sample of potentially white dwarfs with common proper motions main sequence star companions. The white dwarfs are retrieved from SuperCOSMOS whereas stellar companions are from Tycho. I present spectra for 39 WD candidates and for 6 stars. Temperatures and surface gravities are estimated for the white dwarf objects, whilst main sequence stars abundances are obtained from the fitting of synthetic spectra to the observations. Finally I discuss the implications of these findings and what future work can be done with these benchmark systems.

On the main sequence, roughly 10-20% of A- and B-type stars display a wide range of chemical peculiarities in their photospheres. It is unknown when during stellar evolution these chemical peculiarities develop, and the mechanisms for the formation of some peculiarities are Dot fully understood. In order to provide strong observational constraints, this thesis investigates chemical abundances in Herbig Ae and Be (HAeBe) stars, which are pre-main sequence progenitors of A and B stars. A detailed abundance analysis was performed for 20 HAeBe stars, and 1 dusty young star, using high resolution, high signal-to-noise ratio spectra. The abundance analysis proceeded by directly fitting synthetic spectra to observations, determining effective temperature, surface gravity, microturbulence, and projected rotational velocity self-consistently with chemical abundances. The synthetic spectra were computed with the program ZEEMAN, which was developed further, optimising it for nonmagnetic spectrum synthesis, adding an automatic fitting routine, and parallelising the program. Eleven of the stars are found to display 'Boots chemical peculiarities, one star shows weak Ap/Bp peculiarities, and the remaining 9 stars are found to be chemically normal. The star with weak Ap/Bp peculiarities has a confirmed magnetic field detection in the literature, as do one],. Boo star and one chemically normal star. The other stars have been searched for magnetic fields, but have no confirmed detections. I argue that the large incidence of Boo chemical peculiarities among HAeSe stars provides strong evidence in favour of a selective accretion hypothesis for the formation of A Boo peculiarities. Among the magnetic stars, it appears that Ap/Bp peculiarities can form on the pre-main sequence but, unlike the main sequence, there' are also chemically normal magnetic HAeBe stars.

SdB stars provide a number of challenges in stellar astronomy. Their evolution along the extreme horizontal branch is well-known, but their formation remains a uncertain. Moreover, sdB stars show peculiar photospheric abundances due to diffusion processes and non-radial oscillation have been detected in approximately 10% of sdB stars. Thus, improved knowledge of fundamental parameters of sdB stars through high-quality observations and improved modelling, will allow us to gain a better understanding of these stars, and eventually help address these issues. We have carried out a detailed abundance analysis on a sample of sdB stars, using high quality multiwavelength spectroscopic data from optical telescopes and the HST/STIS and FUSE satellites. The multiwavelength approach allows us verify the consistency of our derived atmospheric parameters, and also allows us to probe elemental abundances of species not readily accessible in the optical region. In addition, we have also used the latest generation of LTE models which implement full metalline blanketing, opacity sampling and updated opacity values from the Opacity Project and IRON Project. Our model grids included OS_m1O models, where abundances were scaled to one-tenth of solar values (as is normally used in sdB studies), and OS_sdB models, where Z < 20 are underabundant by a factor of ten, and Z ~ 20 are overahundant by a factor of ten with respect to solar values. Finally, we have compared our results over the three wavelength regions, to verify the consistency, and we have tri'ed to find any correlations between abundances and sdB properties such as pulsation and evolution.

Galaxy clusters are the most massive gravitationally bound structures in the Universe. They can be used both as cosmological probes and as laboratories for testing theories of galaxy formation and evolution. This thesis presents a study of the assembly of stellar mass in galaxy cluster cores between redshift z = 1 and the present, focusing principally on brightest cluster galaxies (BeGs) and the intracluster light (ICL). First, the growth of BeGs is examined through the evolution of their half-light radii using rest-frame matched samples of galaxy clusters at redshifts z = 1 and z = 0.23. A critical analysis of the methods available to measure half-light radius is presented, then the size evolution of the BeGs is measured through the parameterisation of their surface brightness profiles. Using a surface brightness profile fit of fixed slope, the half-light radii of BCGs are found on average to increase by 53% in the 5 Gyr between z = 1 and z = 0.23. This result is in stark contrast to recent findings for the size growth of field ellipticals, which are found. to increase in radius by 2-5 times between 0 < z < 2. It is concluded that the dense environments which BCGs occupy has allowed them to rapidly assemble the majority of their stellar masses early in the Universe, making them almost fully mature by z = I, whereas field ellipticals undergo a slower assembly. Secondly, the first ever measurement of the intracluster light in galaxy clusters at z = 1 is presented. The ICL is detected with near-infrared data from HAWK-/ on the Very Large Telescope and is measured using a simple surface brightness thresholding technique. Below a surface brightness threshold of J.J..J = 22 maglarcsec2 the ICL is found to contain -4% of the total cluster light at z = 1, a fraction which must increase substantially when compared with nearby galaxy clusters

This thesis explores the stellar mass growth of galaxies and the bimodality of galaxy morphologies and properties throughout the last 11 billion years of cosmic time. The data is from three deep, near infrared surveys: the GOODS NICMOS Survery (GNS), the Ultra Deep Survey (UDS) and the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey (CANDELS). Firstly, we use a sample of 8298 galaxies observed as part of the GNS using the Hubble Space Telescope (BST) H16o-band. We construct the galaxy stellar mass function both as a function of redshift and stellar mass in the redshift range 1 < z < 3.5. We find that a significant fraction of all massive (111* > 10111110) galaxies are in place up to the highest red shifts we probe, with a decreasing fraction of lower mass galaxies present at all redshifts. This is consistent with the downsizing scenario where the most massive galaxies have built up their stellar mass before low mass galaxies. By examining the stellar mass function divided into blue/red systems, as well as for star forming and nonstar forming galaxies, we find a similar mass down sizing present for these populations. We fit the mass functions with a Schechter function and find that the faint end slope parameter values range from ex = -1.36 to -1. 73, which is significantly steeper than what is found in previous studies in the literature. We compute the stellar mass density by integrating the Schechter functions that we have fit and find our results to be higher than those in the literature. As a result, the steeper mass function better matches the stellar mass added due to star formation, thereby alleviating some of the discord between these two measures of the evolution of galaxy mass. We construct the total galaxy stellar mass function, in the red shift range 0.3 < z < 3.0, using the full UDS and CANDELS UDS data sets. The combination of these two Abstract 3 surveys gives us excellent number statistics from the full UDS (more than 100,000 galaxies over the 0.88 deg2 field). Furthermore, we have the depth to probe the low mass end of the galaxy stellar mass function from the CANDELS UDS (50' depth of H=26.3 in a 1" -aperture). We find similar results to the GNS total mass functions, including similar values of a despite deeper data. Furthermore, we divide the galaxy stellar mass function by environment and colour. We show that the growth of quiescent galaxies could be driven by mass and environment dependent quenching processes out to a redshift of z rv 2. Finally, we examine 1188 massive galaxies with JVI* ~ 1010 M0 between redshifts Z = 1- 3 within the CANDELS UDS. Using this sample we determine how galaxy structure and morphology evolve with time, and investigate the nature of galaxy structure at high redshift. We visually classify our sample into disks, ellipticals and peculiar systems, conecting for redshift effects on these classifications through simulations. We find significant evolution in the fractions of galaxies at a given visual classification as a function of redshift. The peculiar popUlation is dominant at Z > 2 with a substantial spheroid population, and a negligible disk population. We compute the transition redshift, Ztrans , where the combined fraction of spheroidal and disk galaxies is equal to that of the peculiar population, as Ztrans = 1.86 ± 0.62 for galaxies in our stellar mass range. We find that this transition changes as a function of stellar mass, with Hubble-type galaxies becoming dominant at higher redshifts for higher mass galaxies ( Ztrans = 2.22 ± 0.82), than for the lower mass galaxies (Ztrans = 1.73 ± 0.57). Higher mass galaxies become morphologically settled before their lower mass counterparts, a form of morphological downsizing. We furthermore compare our visual classifications with Sersic index, concentration, asymmetry and clumpiness (CAS) parameters, star formation rate and rest frame U - B colour. We find links between the colour of a galaxy, its star formation rate and how extended or peculiar it appears. We discuss the negligible Z > 2 disk fraction based on visual morph~logies and speculate that this is due to disks in formation appearing peculiar through processes such as violent disk instabilities or mergers: To properly define high redshift morphology and structure we suggest that a new and more exact classification scheme is needed.

Massive stars have a profound influence on their surroundings. However, relatively little is known about their formation. The study of massive star formation is hindered by a lack of observational evidence, primarily due to difficulties observing massive stars at early stages in their development.

This thesis describes the contributions made by the author to the IPHAS survey and its follow-up. The IPHAS survey is currently observing the northern Galactic Plane in Ha, r', and i', and the work presented here focuses primarily on the effort to construct a new, large and well-defined sample of cataclysmic variables (CVs) from the survey. I first investigate the properties of the known population ofCVs in IPHAS. A selection algorithm has been developed to pick out objects from the IPHAS photometry that display a clear Her excess. This has led to the recovery of::: 70 per cent of the known CVs as Ha emitters, roughly independent ofCV type and orbital period. The selection algorithm has been employed further to create a catalogue of Her excess objects, which is presented and analysed. The resulting catalogue contains 4869 Ha emitters. I present and analyse initial follow-up observations that are relevant to the construction ofa new CV sample. A series of long-slit spectroscopic observing runs has so far provided identification spectra of several hundred Ha emitters. A classification algorithm has been developed and used to assign spectral types to these sources. The majority are found to be early-type stars, whereas interacting binaries including CVs and symbiotic stars, make up a relatively small percentage of the emission line ~tar population. The ID spectroscopy taken so far has allowed the discovery of 11 new CV candidates. Additional time-resolved observations for 3 of these confirm their CV classification and provide estimates of their orbital periods. Overall, this thesis shows that IPHAS is capable of discovering a significant sample of CVs that should be largely free of the selection effects that have plagued previous surveys. Once a significant population has been discovered, it will be possible to test theories of binary evolution and CV population synthesis models.

I provide observational constraints on the size and period distribution of the Galactic cataclysmic variable (CV) population, and derive the implications that these constraints have for the theory of CV evolution. The results are based on quantitative modelling applied to three observational CV samples, tw<? of which are newly constructed here. Large differences between the size and other properties of the known sample of CVs and the predictions of the theory of binary star evolution have long been recognized. However, because all existing observational CV samples suffer from strong selection effects, observational biases must be taken into account before it is possible to tell whether there are real inconsistencies. In order to address this, I implement a Monte Carlo method to model selection effects in observed CV samples, and make a new measurement of the space density of CVs. I illustrate the effects of the biases that are introduced by several of the most common observational strategies for identifying CVs. Also, by simulating the selection criteria of the Palomar-Green (PG) Survey, I show that selection effects cannot reconcile the relative sizes of the long- and short-period CV populations predicted by standard CV evolution theory with the observed sample. The selection criteria used to define most CV samples (including the PG sample) discriminate heavily against the discovery of intrinsically faint, short-period systems. The situation can be improved by selecting systems for the presence of emission linesj I have therefore constructed aohomogeneous new sanlple of 17 CVs, selected on the basis of Ha emission from the AAOjUKST SuperCOSMOS Ha Survey (SHS). I present observations of the CVs discovered in this search, and use the sample to constrain the properties of the intrinsic CV population. I show that even very generous allowance for selection effects is not sufficient to reconcile the ratio of short- to long-pe°riod CVs predicted by standard CV evolution theory with the observed sample, confirming theo result based on the PG survey. The most likely implication is that short-period systems evolve faster than predicted by the disrupted magnetic braking model. This would require that an angular momentum loss mechanism besides gravitational radiation acts on CVs with orbital periods below the period gap. To bring the model into agreement with observations, the rate of angular momentum loss below the period gap must be increased by a factor of at least 3, unless the model also overestimates the angular momentum loss rate of long-period CVs. In order to constrain the size of the Galactic CV population, I construct a small, but purely X-ray flux-limited sample of CVs, using the ROBAT North Ecliptic Pole (NEP) survey. The sample includes only 4 systems, 2 of which are new discoveries. Orbital periods are measured for both these systems from time-resolved spectroscopy, and the distances of all the CVs in this sample are estimated. The space density of ~he CV population represented by the sample is 1.1~5:~ x 10-5 pc-3. I also place upper limits on the space density of any population of CVs too faint to be included in the NEP survey-if the overall space density of CVs is as high as 2 x 10-4 pc-3 (as has been predicted theoretically), the vast majority of CVs must have X-ray luminosities below ~ 2 x 1029 ergs-I.

No description available.

We consider the four-derivative modification to the Einstein-Hilbert action of general relativity, without a cosmological constant. Higher derivative terms are interesting because they make the theory renormalisable (but non-unitary) and because they appear generically in quantum gravity theories. We consider the classical, static, spherically symmetric solutions, and try to enumerate all solution families. We find three families in expansions around the origin: one corresponding to the vacuum, another which contains the Schwarzschild family, and another which does not appear in generic theories with other number of derivatives but seems to be the correct description of solutions coupled to positive matter in the four-derivative theory. We find three special families in expansions around a non-zero radius, corresponding to normal horizons, wormholes and exotic horizons. We study many examples of matter-coupled solutions to the theory linearised around flat space, which corroborate our arguments. We are assisted by use of a "no-hair" theorem that certain conditions imply that $R=0$, which is applicable in many cases including asymptotically flat space-times with horizons. The Schwarzschild black hole still exists in the theory, but a second branch of black hole solutions is found that can have both positive and negative mass, and that coincide with the Schwarzschild black holes at a single mass. The space of asymptotically flat solutions is probed numerically by shooting inwards from a weak-field solution at large radius, and the behaviour at small radius is classified into the families of series solutions (most of which make an appearance). The results are inconclusive but show several interesting features for further study.