Tourmaline, not hitherto recorded in Lewisian rocks of Sutherland, nor reported as a secondary mineral in Precambrian palaeosols, occurs in a late Proterozoic, sub-Cambrian palaeosol in NW Scotland. From field and petrological criteria, the tourmaline is secondary and peculiarly associated with the alteration profile. Hence, it formed either during weathering or later in the profile’s history. A minor constituent of the palaeoregolith immediately below the unconformity, the tourmaline mostly occurs as sub-millimetre, radiating, acicular aggregates and is schorl in composition. Black, tourmaline+quartz veinlets cross-cut the profile. Stable isotope and fluid inclusion data suggest that the tourmaline formed from surface-derived fluids at c. 110ºC. It is concluded that boron was adsorbed onto precursor illitic clays of the soil profile during the Cambrian marine transgression. Thus the tourmaline formed subsequently after burial under the sedimentary pile, probably during the Caledonian orogeny, when temperatures were high enough to allow its formation at suitable sites of nucleation, e.g. adjacent to biotite. Its peculiarity as a palaeosol phase is due to the combination of a marine transgression over an exposed, permeable illitic soil and underlying regolith, followed by increase in temperature suffcient to allow the tourmaline to form.