The contact between the two tills is sharp and undulating, and locally a thin sand and gravel horizon, 0.1-0.15 m thick, lies between the two tills. Rae considered at this and all the other sites that there was no firm evidence to support more than one glacial episode, and therefore the multiple tills were deposited by the last ice sheet to cover Orkney.
From the orientations of striae on the bedrock beneath the brown till, the lithological composition of the till ( a relatively high percentage of Lower Eday Sandstone) and clast fabric measurements, Rae concluded that the ice movement associated with the lower till was from between south and south-east. Ice following such a flow pattern would have traversed the Lower Eday Sandstone most of the way across the Deerness peninsula and avoided the Middle Eday Sandstones.
The overlying red till has a relatively higher content of red sandstone clasts similar to the Middle Eday sandstone. Since no onshore outcrop of this lithology appeared to fit with the pattern of striation observed beneath the red till elsewhere in Deerness where it rests directly on bedrock, Rae postulated an offshore outcrop, an interpretation supported by the presence of chalk erratics in the red till but not in the brown till. The red till was therefore associated with ice flow from a more easterly azimuth than the underlying brown till. Shell fragments have been recovered from both till units (Gordon, 1993).
Den Wick is an important section representing one of the key elements of the till stratigraphy of Orkney. The multiple till sequences, of which Den Wick is a particularly good example, are important in several respects. First, their lithological and sedimentological characteristics demonstrate close relationships with the source bedrock traversed by the ice. Second, the superimposition of the different till units indicates changes in ice-flow direction, thus supporting the inferences based on striation patterns. Third, the contacts between the different units a individual sites provide no conclusive evidence for deposition during more than one glacial period. Fourth, the superimposition of multiple till units in consistent fashion (Rae, 1976) and the inferred shifts in the direction of ice flow provide important evidence for the dynamics of the ice-sheet and its driving forces. Further study at both regional and national scales, coupled with ice-sheet modelling, is required to elaborate the origins of these changes and the extent to which they may relate to climatic factors and other variables which determine ice-sheet flow patterns. Multiple till sequences, like that at Den Wick, will be an important source of field evidence to provide constraints on the appropriate mathematical models.