The main goal of this thesis work is to find the links between star formation in galaxy bulges (mainly in the form of star-forming nuclear rings) and other properties of the host galaxy, such as the bar strength or the presence of sigma-drops. To do so we study galaxy non-axisymmetries and their side effects, such as dust lanes in bars. At the end of this work we also include a Chapter on a highly inclined spiral galaxy, namely NGC 4244, and a Chapter on a preliminary study on GALFIT decomposition of infrared images of disc galaxies with outer rings.

In the first Chapter we describe a study of localized minima of the stellar velocity dispersion in the central regions of galaxies, which are known as sigma-drops. In this Chapter we present a sample of 20 sigma-drop galaxies matched with a control sample of galaxies without sigma-drops in order to search for correlations between sigma-drops and the properties, primarily morphological, of the nuclear zones and the discs of their host galaxies. For these two samples we study the dust and H_alpha distribution on 0.1 arcsec scales, using Hubble Space Telescope (HST) imaging, in the circumnuclear zones, searching for differences and trying to establish a link between the bulge kinematics and its morphology. We have also considered the CO and HI emission distribution of the galaxies, and their luminosity profiles. We classify the two samples following both morphological parameters and the stellar luminosity profiles and we find a larger fraction of nuclear dust spirals and star-forming nuclear rings in the sigma-drop sample. We also find that the fraction of Seyfert galaxies in the sigma-drop sample is bigger than that of LINERs and that the reverse is true for the control sample. The conclusions of Chapter 1 are that sigma-drops are very probably due to inflow-induced star formation in a dynamically cool disc, or in a gas ring (shock focused by an inner Lindblad resonance), above a certain critical density level. The same mechanism that feeds the nuclear ring or the nuclear disc is probably responsible for the higher rate of Seyfert galaxies among the sigma-drop hosts. Some of the results of this Chapter have been published in Comerón et al. (2008a).

In Chapter 2 we study four Ultra-Compact Nuclear Rings (UCNRs) found in the HST images used to prepare the samples for Chapter 1. UCNRs are nuclear rings which we define to be less than 200 pc in radius. The point of this Chapter is to find evidence to know whether they have been formed (like their larger counterparts) by an inflow of gaseous material that has been driven to central regions of the galaxy after the loss of angular momentum due to a non-axisymmetric structure (e.g. a bar) and has been trapped in X_2 orbits related to Inner Lindblad Resonances (ILRs), or whether they have been created by another physical process. From archival HST UV and H_alpha images and from dust structure maps of the circumnuclear regions, we analyse the morphology of the star formation and dust of the UCNRs, which are 30-130 pc in radius, that are found in NGC 2985, NGC 4579, and NGC 4800. We find no difference, except for the size, between UCNRs and "classical" nuclear rings. Some of the results of this Chapter have been published in Comerón et al. (2008b).

In the third Chapter we present the most complete atlas of nuclear rings to date, which has been named Atlas of Images of NUclear Rings (AINUR). We include 113 rings found in 107 galaxies, six of which are elliptical galaxies, five are highly inclined disc galaxies, 18 are unbarred disc galaxies, and 78 are barred disc galaxies.  Our aim is to explore possible relationships between the size and morphology of the star-forming nuclear rings and various galactic parameters such as the galaxy size and the galaxy bar properties. We also aim to establish whether UCNRs are a distinct population of nuclear rings, with a distinct origin, or if they are merely the low-end tail of the nuclear ring size distribution. We produce colour index and structure maps, as well as H_alpha and Paschen_alpha continuum-subtracted images from HST archival data. We derive ellipticity profiles from $H$-band Two-Micron All-Sky Survey (2MASS) images in order to detect bars and find their metric parameters. We measure the non-axisymmetric torque parameter, Q_g, and search for correlations between bar and ring metric parameters, and Q_g. We find that nuclear rings span a range from a few tens of parsecs to a few kiloparsecs in radius. We find that dust nuclear rings are found in elliptical and S0 galaxies. We find that star-forming nuclear rings occur in 20+-2% of disc galaxies with types between T=-3 and T=7, with a peak in the distribution between T=2 and T=3 and without strong preference for barred galaxies. The ellipticities of rings found in disc galaxies range from 0 to 0.4, assuming that nuclear rings lie in the galactic plane. For barred galaxies, the maximum radius that a nuclear ring can reach is a quarter of the bar semi-major axis. We found a nearly random distribution of position angle offsets between nuclear rings and bars. There is some evidence that nuclear ring ellipticity is limited by bar ellipticity. We confirm that the maximum relative size of a star-forming nuclear ring is inversely proportional to the non-axisymmetric torque parameter, Q_g ("stronger bars host smaller rings") and that the origin of nuclear rings, even the ones in non-barred hosts, is closely linked to the existence of dynamical resonances. Ultra-compact nuclear rings constitute the low-radius portion of the nuclear ring size distribution. We discuss implications for the lifetimes of nuclear rings and for their origin and evolution. Some of the results of this Chapter have been published in Comerón et al. (2010).

In Chapter 4 we test the theoretical prediction that the straightest dust lanes in bars are found in strongly barred galaxies, or more specifically, that the degree of curvature of the dust lanes is inversely proportional to the strength of the bar. The test used archival images of barred galaxies for which a reliable bar non-axisymmetric torque parameter (Q_b; also called bar strength) and the radius at which Q_b has been measured (r(Q_b)) have been published in the literature. Our results confirm the theoretical prediction but show a large spread that cannot be accounted for by measurement errors. We simulate 238 galaxies with different bar and bulge parameters in order to investigate the origin of the spread in the dust lane curvature versus Q_b relation. From these simulations, we conclude that the spread is greatly reduced when describing the bar strength as a linear combination of the bar parameter Q_b and the ratio of the major and minor axis of the bar, $a/b$. Thus, we conclude that the dust lane curvature is predominantly determined by the parameters of the bar. Some of the results of this Chapter have been published in Comerón et al. (2009).

In Chapter 5 we present an in-depth study of NGC 4244. NGC 4244 is a late-type nearly edge-on galaxy 4.4 Mpc away. It has been reported as the only nearby edge-on galaxy without a confirmed thick disk. If thick disks are indeed, as has been proposed, products of internal evolution, all old enough undisturbed disc galaxies should have a thick disk. Finding such a galaxy without a thick disc would be a serious challenge to a range of galaxy formation models. Using data from the Spitzer Survey of Stellar Structure in Galaxies (S4G) we have found that this galaxy has a tenuous thick disk. We have also found that the asymmetries between the light profiles on either side of the mid-plane of NGC 4244 can be explained by a combination of the galaxy not being perfectly edge-on and a certain degree of opacity of the thin disk. We argue that the low density of the thick disk is a consequence of the lack of secular evolution in NGC 4244.

In Chapter 6 we study galaxies with outer rings using S4G images. We have matched a sample of eleven galaxies with outer rings with another sample of eleven galaxies without an outer ring. We used GALFIT to decompose the galaxies into their main components in order to find differences between the structural parameters of ringed and non-ringed galaxies. We find that outer rings tend to be hosted in galaxies with more massive, wider, and longer, but not stronger, bars.

We list the main conclusions of this thesis in Chapter 7 (in English) and in Chapter 8 (in Spanish).