Emeleus, C.H. & Gyopari, M.C. 1992. British Tertiary Volcanic Province, Geological Conservation Review Series No. 4. JNCC, Peterborough, ISBN 0 412 47980 X. The original source material for these web pages has been made available by the JNCC under the Open Government Licence 3.0. Full details in the JNCC Open Data Policy
East Canna and Sanday
Highlights
The inter-lava fluvial sediments and pyroclastics exposed here are the best developed in the British Tertiary Volcanic Province, providing an essential link in a chain of sites in the study of the Tertiary volcanic history of the region. Derived clasts suggest that the Rum Complex is appreciably older than the Skye Cuillin centre.
Introduction
Agglomerates, fluvial conglomerates and other sediments closely associated with the volcanic succession of Canna, occur within this site and are the best developed and most extensive examples within the BTVP. These formations were the first such deposits within the Province to be described in detail. Geikie (1897) concluded that the conglomerates had filled a river channel cut into the lavas. The islands were later mapped by Harker (1908) who accepted the fluvial origin. Allwright (1980) has more recently interpreted coarse conglomerates on Compass Hill in terms of a nearby source of coarse pyroclastic debris which was possibly connected with a volcano which fed some of the lava flows.
Description
The spectacular sea cliffs, stacks and intertidal areas of south-eastern Sanday, Eilean a'Bhaird and eastern Canna contain good exposures of the interlava clastic sediments. In and near to the stacks of Dùn Mòr (
The impressive hundred-metre-high cliffs and the coastal sections below Compass Hill
The sea stacks of Alman
Eilean a' Bhaird
Interpretation
The value of the Canna–Sanday site lies in the evidence that it provides for the development of extensive, vigorous river systems during the period of Palaeocene lava accumulation. Geikie (1897) placed particular emphasis on the palaeogeographical implications of the clasts in the fluviatile sediments; the occurrence of gneisses, schists and Torridonian clasts indicates erosion in the Palaeocene of a landmass containing lithol-ogies similar to present-day western Scotland. The abundance of Torridonian fragments in the pyroclastic-derived deposits of Compass Hill might alternatively indicate a Torridonian basement to the Tertiary lavas, although it is surprising that no Mesozoic sedimentary fragments have been recognized, since Canna overlies a Mesozoic basin (Binns et al., 1974). The position of the Eilean a'Bhaird flow within the Canna sequence is problematic. Allwright argues that it must be fairly low in the stratigraphic succession of lavas; however, it could be one of the latest lavas in the area, since it overlies conglomerates filling a valley eroded in lavas and no examples of a lava of similar composition are known to be overlain by other lavas on Canna or Sanday.
The similarity between some of the Canna–Sanday lavas and those of north-west Rum (Emeleus, 1985), and the occurrence of clasts apparently derived from the acid rocks in the Rum complex, establishes a close connection with Rum. The site is a vital link in the chain of sites from Muck and Eigg, through Rum, to southwest Skye and Ardnamurchan which enables continuous, relative dating of igneous rocks across the British Tertiary Volcanic Province to be attempted, thus aiding assessment of the evolution of the Province (Meighan et al., 1981; Dagley and Mussett, 1986).
Conclusions
The intra-lava sediments in the volcanic succession of east Canna and Sanday provide unequivocal evidence that a river system draining areas of gneiss, schist and probably Torridonian sediments, was established during periods when active effusion of basaltic and intermediate lavas was occurring. The rivers also drained Rum and probably extended to Skye, the distinctive pebbles laid down in the associated fluviatile deposits make it possible to conclude that the Skye Cuillin Central Complex almost certainly post-dated the Rum Central Complex.