Bars have a complex three-dimensional shape. In particular their inner part is vertically much thicker than the parts further out. Viewed edge-on, the thick part of the bar is what is commonly known as a boxy-, peanut- or X- bulge and viewed face-on
it is referred to as a barlens. These components are due to disc and bar instabilities and are composed of disc material. I review here their formation, evolution and dynamics, using simulations, orbital structure theory and comparisons to observations.
Barred galaxies have interesting morphological features whose presence and properties set constraints on galactic evolution. Here we examine barlenses, i.e. lens-like components whose extent along the bar major axis is shorter than that of the bar an
d whose outline is oval or circular. We identify and analyse barlenses in $N$-body plus SPH simulations, compare them extensively with those from the NIRS0S (Near-IR S0 galaxy survey) and the S$^4$G samples (Spitzer Survey of Stellar Structure in Galaxies) and find very good agreement. We observe barlenses in our simulations from different viewing angles. This reveals that barlenses are the vertically thick part of the bar seen face-on, i.e. a barlens seen edge-on is a boxy/peanut/X bulge. In morphological studies, and in the absence of kinematics or photometry, a barlens, or part of it, may be mistaken for a classical bulge. Thus the true importance of classical bulges, both in numbers and mass, is smaller than currently assumed, which has implications for galaxy formation studies. Finally, using the shape of the isodensity curves, we propose a rule of thumb for measuring the barlens extent along the bar major axis of moderately inclined galaxies, thus providing an estimate of which part of the bar is thicker.
`Conspiracy between the dark and the baryonic mater prohibits an unambiguous decomposition of disc galaxy rotation curves into the corresponding components. Several methods have been proposed to counter this difficulty, but their results are widely d
iscrepant. In this paper, I revisit one of these methods, which relies on the relation between the halo density and the decrease of the bar pattern speed. The latter is routinely characterised by the ratio ${cal R}$ of the corotation radius $R_{CR}$ to the bar length $L_b$, ${cal R}=R_{CR}/L_b$. I use a set of $N$-body+SPH simulations, including sub-grid physics, whose initial conditions cover a range of gas fractions and halo shapes. The models, by construction, have roughly the same azimuthally averaged circular velocity curve and halo density and they are all submaximal, i.e. according to previous works they are expected to have all roughly the same ${cal R}$ value, well outside the fast bar range (1.2 $pm$ 0.2). Contrary to these expectations, however, these simulations end up having widely different ${cal R}$ values, either within the fast bar range, or well outside it. This shows that the ${cal R}$ value can not constrain the halo density, nor determine whether galactic discs are maximal or submaximal. I argue that this is true even for early type discs (S0s and Sas).
We follow the formation and evolution of bars in N-body simulations of disc galaxies with gas and/or a triaxial halo. We find that both the relative gas fraction and the halo shape play a major role in the formation and evolution of the bar. In gas-r
ich simulations, the disc stays near-axisymmetric much longer than in gas-poor ones, and, when the bar starts growing, it does so at a much slower rate. Due to these two effects combined, large-scale bars form much later in gas-rich than in gas-poor discs. This can explain the observation that bars are in place earlier in massive red disc galaxies than in blue spirals. We also find that the morphological characteristics in the bar region are strongly influenced by the gas fraction. In particular, the bar at the end of the simulation is much weaker in gas-rich cases. In no case did we witness bar destruction. Halo triaxiality has a dual influence on bar strength. In the very early stages of the simulation it induces bar formation to start earlier. On the other hand, during the later, secular evolution phase, triaxial haloes lead to considerably less increase of the bar strength than spherical ones. The shape of the halo evolves considerably with time. The inner halo parts may become more elongated, or more spherical, depending on the bar strength. The main body of initially triaxial haloes evolves towards sphericity, but in initially strongly triaxial cases it stops well short of becoming spherical. Part of the angular momentum absorbed by the halo generates considerable rotation of the halo particles that stay located relatively near the disc for long periods of time. Another part generates halo bulk rotation, which, contrary to that of the bar, increases with time but stays small.
After presenting three ways of defining a bulge component in disc galaxies, we introduce the various types of bulges, namely the classical bulges, the boxy/peanut bulges and the disc-like bulges. We then discuss three specific topics linked to bulge
formation and evolution, namely the coupled time evolution of the bar, buckling and peanut strengths; the effect of velocity anisotropy on peanut formation; and bulge formation via bar destruction.
