An Unsuccessful Archaeological & Geological Outing to Covesea

MFC Paper 22 by SM Ross and D Milne 1977

The caves in the sandstone cliffs on the southern shore of the Moray Firth between Hopeman and Covesea have long been known to have yielded evidence of former occupation by man. Archaeologists, both amateur and professional, have conducted "digs" there since early last century, but unfortunately not all have left records of their labours.

Some 2.5 km east of Hopeman and 400m east of the "90 Steps" Quarry at NJ 173 706 can be seen the eroded rear section of a former cave, filled with debris from roof-falls on top of which is a layer of horizontally stratified sand. The site is adjacent to the famous Sculptor's Cave which Miss Benton (1) showed was occupied by man from the Bronze Age to Post-Roman times.

The sand deposit appeared to be the remains of what had at one time been a substantial sheltered and accessible ledge and when viewed through binoculars from the beach some 6.30m below, thin darkened layers containing shells could be seen in the sand. The site was therefore considered worthy of closer inspection

Because of the unstable nature of the debris it was considered unsafe to climb on to the top of the pile to reach the sand and this was examined with difficulty and at arms length by using an extending aluminium ladder. The top surface had a slimy covering of guano and algae and extended across the back of the cave, varying in width from 1 to 2m and was approximately 0.5m thick, reaching to the lower parts of the roof. Dark horizontal banding in the deposit indicated former surface levels. Some of the bands contained thin layers of charcoal in which many mussel and cockle shells were embedded. A few fragments of animal bone were also exposed and small rib and pelvic bones were recognised. There were a few small burrows or nesting holes in the sand and some of the debris might have been brought in by birds or other predators.

These finds looked encouraging but a more detailed examination has been thwarted by heavy seas during the very high tides of late November and early December which reduced and undermined the rock pile to such an extent that even the use of the ladder is now out of the question, and we may never know if the hand of man ever set foot on the edge.

The floor of the cave is a sloping rock platform 2.45m about the present beach level and about 1.5Om above high water mark. The cave is 9.64m wide and the roof slopes down from 5.00m above the floor on the east side to 3.85m on the west along the dip of the sandstone. The debris in the cave is in three distinct layers:—

  1. On top 0.53m of horizontally stratified sand which reaches to the lower parts of the roof.
  2. Below this 1.05m of rubble consisting of more angular sandstone blocks with small stones and sand.
  3. A lower layer of 2.28m of large, partially rounded sandstone boulders, fitting together with scarcely any matrix of smaller material.

The lower layer is unusual in that the boulders, though partially rounded, show faceted faces where they are in contact with each other and with the walls and floor of the cave, with the result that they form a very close fitting pile. The cave walls and floor are of extremely friable, well-bedded yellow sandstone with a very sparse ferruginous cement, while the majority of the boulders are of the much harder paler type with siliceous cement. While it can be argued that these harder rocks could easily rub away the softer rock of the walls and floor of the cave, the interfaces between the boulders are smooth and match so well that there could be little or no jostling possible and chatter marks are rare.

In 1968 geologist David Shelley (2) gave examples of piles of fitting boulders from New Zealand beaches and drew attention to the fact that although the phenomenon was not described in any standard works on coastal processes, it could be shown to be of common occurrence once recognised.

He considered that while mechanical grinding, pressure solution and the solution of silicate minerals by micro-organisms were well established processes, these would be of doubtful influence in such cases. He pointed out that in the splash zone, water falling on the exterior surfaces of the boulders would either evaporate or run off quickly, but water penetrating the interfaces would be trapped there because of surface tension and slow evaporation. This tendency to continual dampness would allow the rock pores to be thoroughly penetrated by water and the eventual crystallization of salt (or ice) along the interface would have a selective erosional effect. Further correspondence confirmed that the phenomena was indeed of widespread occurrence but acceptance of his theory was not universal.

While the lowest boulder layer in the cave would appear to be of the type described by Shelley, though not so well developed, the middle section is a later deposit of randomly orientated blocks in which interspaces are filled with fragments of friable sandstone and sand. The presence of the friable material at this level rules out wave action as the depositing force for this layer and probably also for the stratified sand layer on top. Water deposited sand at this level would be interpreted as dating from the time of the post-glacial high sea level of some 5.0m which occurred about 4500 years ago, but a windblown origin of a later date seems much more probable, even though no satisfactory cross section could be cut in the layer.


(1) BENTON S. 1931. The Excavation of the Sculptor's Cave, Covesea. Proc. Soc. Antiq. Scot. pp 177-216

(2) SHELLEY D. 1968 Fitting Boulders, the Result of an Important Shore Process. Nature, Lond, 220, 1020.

Posted in .