Stellar disks are well approximated by exponential surface brightness distributions. With flat rotation curves the specific angular momentum increases with radius and it is not unreasonable to expect this property to have a maximum. But is this a sharp and well-defined maximum and where does it result from? Relatively sharp truncations in stellar disks have been known for edge-on galaxies and seems to occur in the majority of cases. Claims have been made for similar features to be observable in moderately or hardly inclined systems. In this review I will address observational questions on the correspondence of such observations in edge-on versus more face-on systems and on the relation to the HI-gas disks as these extend beyond the observable optical extent. Also I will review the observational clues concerning the origin of truncations in stellar disks.

Using imaging data for two large samples of face-on disk galaxies and analyzing the radial stellar light distribution, we have studied the shape of their outer disks. We have identified three basic classes of surface brightness profiles depending on the apparent break features or lack of them: 1) The well known Freeman (1970) Type I that has an exponential profile, with no break. 2) Freeman Type II, a broken exponential with a “downbending break”, now including the so-called “truncations”. 3) A completely new class, called Type III (or “antitruncation”), also described by a broken exponential but with an upbending profile, shallower beyond the break. We are now trying to understand why we have this diversity of outer disk structures and how this is going to improve our picture of galaxy formation and evolution. Various physical explanations have been put forward but we still lack a fully consistent picture on how the stars are formed and possibly redistributed within the galaxy later on. Once the nature of the galactic stellar disk is determined it will be fully justified to apply these breaks as a fundamental disk characteristic. For example, by using the break as a size indicator comparing galaxies at various redshifts, as already done today.

In recent years we have started to appreciate that the outer banks of galaxies contain valuable information about the formation process of galaxies. In hierarchical galaxy formation the stellar halos and thick disks of galaxies are formed by accretion of minor satellites, predominantly in the earlier assembly phases. The size, metallicity, and amount of substructure in current day halos are therefore directly related to issues like the small scale properties of the primordial power spectrum of density fluctuations and the suppression of star formation in small dark matter halos after reionization. We show initial results from our ongoing HST GHOSTS survey of the resolved stellar envelopes of 14 nearby, massive disk galaxies. We detect for the first time the very small halo of NGC4244, a low mass edge-on galaxy. We find that massive galaxies have very extended halos, with equivalent surface brightnesses of 28-29 V-mag arcsec^-2 at 20-30 kpc from the disk. The old RGB stars of the thick disk in the NGC891 and NGC4244 edge-on galaxies truncate at the same radius as the young thin disk stars, providing insights into the formation of both disk truncations and thick disks. We furthermore present the stellar populations of a very low surface brightness stream around M83, showing it had no significant star formation in the last 300 Myr.

The past years have seen a significant shift in our perception of the outermost regions of galaxy disks. Whereas it had been known from radio observations that galaxies are often embedded in extended gaseous disks, these outer disks have now been recognized to be home to stellar populations as well as ongoing star formation displaying a wide variety of interesting features. The outer disks harbor constraints on a galaxy's formation and evolutionary history, and bear the signatures of its growth and interactions. We have been conducting a search for such signatures, using deep optical spectroscopy to observe gas in the outer disks of spiral galaxies at up to twice the radius of the stellar disk, and I will present results on kinematics (including cases of anomalous kinematics indicative of possible interactions), metallicity, and star formation. I will also outline our plans for future observations that will shed light on the relationship between star formation and stellar content on the outer disk and on the role of outer-disk warps as a possible signature of accretion.

I will present some recent spectroscopic results on the velocity structure of outer disks and deep LBT photometry. I will then discuss what might be concluded from the current observational situation regarding the growth of disks and attempt to place these results in the more general context of galaxy structure.

In 2005 the Galaxy Evolution Explorer (GALEX) satellite discovered extended UV (XUV) emission in a number of nearby disk galaxies. We now know this XUV emission is due to the presence of massive young stars at large galactocentric distances. Due to the low gas densities where the XUV emission is located this phenomenon provides an excellent test to determine whether a threshold of star formation actually exists and what is the law of star formation above it. In this talk I will summarize our current knowledge on the XUV phenomenon and on its implications on the star formation threshold and law and on the star formation history and chemical evolution of the outer edges of galaxy disks.

In CDM, disk galaxies form from the inside out. Thus the outer regions of disk galaxies may provide important constraints on galaxy formation theory. Observations reveal that the light profiles of disk galaxies are rarely exponential down to the last measured point. In the majority of galaxies, the profiles exhibit a break to a steeper profile, which had earlier been termed as “disk truncations”. I present early results from collisionless simulations and new simulations including star formation which provide new insights into the formation mechanisms of breaks. The simulations need to reach high mass resolution since the outer regions often have very low densities. We are able to form breaks which agree well with those observed and which lead to new predictions.