The timing of the ice margin retreat of the Late Glacial Patagonian Ice Sheet (PIS) in southern Patagonia has been the object of discussion for many years. In order to resolve questions about the complex response of the PIS to past climate change, any geological interpretation and data modelling need evaluation against an absolute chronology. The aim of this project is to investigate the applicability of OSL dating to sediments from southern Patagonia; in particular, we examine the dating potential of K-feldspar IRSL signals. Samples were collected from landforms interpreted as being deposited during deglaciation of the PIS, with an expected age range of 17 and 22 ka, and from recently deposited sediment. We measure small aliquots and single grain distributions using an IR50 SAR protocol with IRSL stimulation at 50 degrees C following a preheat at 250 degrees C (held for 60 s). Uncertainties are assigned to our individual dose estimates based on the over-dispersion (OD) observed in laboratory gamma dose recovery experiments (22% for small aliquots and 18% for single grains). Then the possible effects of incomplete bleaching and differential fading are examined. For our natural samples we observe environmental ODs between 30 and 130% and mean residual doses between similar to 30 and 80 Gy. Minimum age models are used to identify the part of the dose population that is most likely to have been well-bleached and results from these models are compared. The models give ages that are consistent with each other; this may imply that they successfully identified the fully-bleached grains in the distributions, although there are some discrepancies between our small aliquot and single grain data. We observe large fading rates (on average 7.9 +/- 0.6%/decade for large aliquots) but nevertheless a comparison of our fading corrected ages with the expected age range shows that 2 out of 3 ages are consistent with geological interpretation and an established radiocarbon and cosmogenic nuclide chronology. We conclude that these investigations suggest that fading corrections can be based on laboratory average small aliquot/single grain fading rates. The third age is supported by an alternative geological interpretation, and the two ages consistent with the existing chronology imply that in the Strait of the Magellan the hills of the Brunswick peninsula (>70 m.a.s.l) were deglaciated at around 22 ka.