Edinburgh Rock: The Geology of the Lothians: Chapter 5

Silurian of the Pentland Hills

The Pentland Hills rise abruptly south of Edinburgh and extend south-west, as a line of distinctive summits, for some 22 km. These hills are less than 600 m high, yet they are largely uninhabited, consisting of grass- and heather-covered moorlands, with few trees. Rocks are exposed only on steep hillsides, or along the banks of streams and reservoirs. These hills are mainly composed of volcanic and lesser sedimentary rocks of late Silurian and early Devonian age. The sediments were deposited in a semi-arid desert, and are reddened by iron oxide. They have long been referred to as of ‘Old Red Sandstone’ age, and before we deal with the underlying Silurian, upon which they rest unconformably, we should explain the meaning of this term.

What do we mean by ‘Old Red Sandstone’?

The somewhat archaic-sounding name ‘Old Red Sandstone’ is retained from the early days of geology. It had been recognised in the early 19th century that there were three separate red sandstone sequences in the British Isles. Of these, the uppermost comprise the redbeds of the topmost Carboniferous rocks and all the desert sands and marls of the Permian and Trias. This suite of sediments became collectively known as the ‘New Red Sandstone’. An older sequence of sediments was known to underlie the Carboniferous, and to this was given the name ‘Old Red Sandstone’ whose usage was subsequently enshrined in the writings of Hugh Miller (1802–1856), especially his exquisitely written (1841) Travels in the Old Red Sandstone, or New Walks in an Old Field. There was some controversy, however, about the precise age of the Old Red Sandstone, and some geologists were slow to recognise that the marine Devonian of south-west England was time-equivalent to the Old Red Sandstone, in other words that the red sediments below the Carboniferous were Devonian in age. Even as late as 1842 it was still possible for the Rev. David Williams (actually a competent field geologist) to write a paper entitled Plausible reasons and positive proofs that no portion of the ‘Devonian System’ can be of the age of the Old Red Sandstone. Yet it became increasingly recognised that the Old Red Sandstone sediments actually interfingered with the classical marine Devonian in North Devon, and that the Old Red Sandstone of south-west England and Wales was unquestionably Devonian. The term Old Red Sandstone has continued in use, but it is now legitimate to use it also to describe ‘facies’. By this we mean a suite of sediments laid down in a particular kind of depositional environment, and here in Scotland ‘Old Red Sandstone facies’ refers to beds laid down in semi-arid desert conditions. Thus the red-beds of the Middle Silurian Henshaw Formation in the North Esk Inlier are of Old Red Sandstone facies, and as we shall see, such facies continue into the early Carboniferous. We should also point out that the third suite of red subaerially deposited sediments, the Proterozoic Torridonian sediments of the north-west Scottish Highlands, were originally thought to be Devonian, and that this view was initially held by such geological giants as Roderick Murchison. It was only when Cambrian trilobites were discovered in the overlying marine sediments that the true age of the Torridonian beds (incidentally the most extensive and best exposed sedimentary succession in the British Isles) was properly recognised.

Three Silurian inliers

In the Pentland Hills, Silurian rocks are exposed in three areas only, and they are the oldest rocks exposed in the immediate vicinity of Edinburgh. The surface of the land, in early Devonian times, was irregular: a buried landscape where the highest hills were covered by a relatively thin blanket of ‘Old Red Sandstone’ rocks. Where this was removed by later erosion, the tops of the older Silurian hills were once more exposed as inliers, or ‘windows’ surrounded by the younger ‘Old Red Sandstone’ lavas, sandstones, and conglomerates.

The three earlier Silurian ‘windows’ are the easternmost of a chain of inliers which can be traced as far as western Ireland, and eastwards into Norway and Sweden. The two eastern inliers, of Bavelaw and Loganlea–Craigenterrie, consist mainly of poorly fossiliferous siltstones. They are separated by the volcanic mass of Black Hill, considered further in Chapter 7. If these two Silurian areas were all that we had in the local area, they would be of no special significance. But the large, westerly, North Esk Inlier is of compelling interest, and has been the focus of intensive research for the last 150 years. The first point to note about the Silurian inliers is that the beds stand vertical, or nearly so. They were rotated into this position some time later in the Silurian, and over a long period of time were uplifted and eroded before the Lower ‘Old Red Sandstone’ was laid down horizontally on top of them. The Old Red Sandstone has since been tilted slightly, but the contrast between the attitude of the vertical Silurian and the gently dipping ‘Old Red Sandstone’ is very evident. The angular junction between them is an unconformity, as explained in Chapter 2. Yet in the other inliers of the Midland Valley of Scotland the Silurian is only gently folded and passes up into the overlying ‘Old Red Sandstone’ without any apparent break, and certainly without angular unconformity. Why there should be this remarkable difference in the attitude of the Silurian in the Pentlands and the other inliers is still debated. Although the actual unconformity is never seen in the Pentlands, being covered by thick soil and dense vegetation, its location can be fixed within a few metres (NT 156.576).

A second point concerns the depositional setting of the Silurian, on the south- eastern margin of the Laurentian continent with the Iapetus Ocean to the south. The Laurentian continental shelf consisted of a series of fault-bounded but interconnected basins, rather than being uniform. The inliers of the Pentland Hills were deposited in one of these; they all received much sediment from the north, during rapid subsidence of the basin floor. But there was another important factor, in the form of an overall marine regression. For the continental shelf was gradually rising along all of its length from Ireland to Sweden and, as it rose, the sea retreated to the south. This had an inevitable, and primary effect on processes of sedimentation, and we shall shortly study its consequences. (By contrast, a substantial marine transgression took place contemporaneously on the Avalonian continent, and the sea flooded from Wales as far as the English Midlands).

The Silurian is divided into four Series. In ascending order these are the Llandovery, Wenlock, Ludlow, and Pridoli. The first three are based upon type localities in Wales and the Welsh Borders. The type locality of the Pridoli Series, on the other hand, is in the Czech Republic, where it is represented by marine limestones. It is not known whether any of the higher Silurian sandstones in the Midland Valley belong to the Pridoli, since they are unfossiliferous red-beds.

In the North Esk Inlier we have a sequence of over a thousand metres of Silurian sedimentary rocks (Figs. 5.1–3). From the palaeoenvironmental and palaeoecological point of view, these sediments and their fossils are perhaps as instructive as any in the world. And indeed the beds are in many places highly fossiliferous, with brachiopods, trilobites, corals, bivalves, crinoids and many other kinds of fossils, and they are very well preserved and undistorted. Most of the fossil groups present in the Pentland Hills originally had calcareous shells or skeletons, but the shell material itself is seldom preserved. Instead, the fossils are preserved as moulds, for the calcareous shells have been dissolved away over time through percolating and slightly acid ground water. This led to the formation of both internal and external moulds of the same shell, which show respectively the outer and inner surfaces. This mode of preservation is quite common in the Lower Palaeozoic of the British Isles, and it is very useful for the palaeontologist, since details of both surfaces of a shell are clearly visible. Moreover, when calcareous shells are preserved intact, they often have much adherent matrix, which has to be carefully removed in order to show the details; this problem does not arise with mouldic material. The internal and external moulds are effectively ‘negatives’, but it is a simple matter to produce a ‘positive’, i.e. a precise replica of the orginal surface, by applying rubber latex, usually in several thin layers, to the surface of the mould, and pulling it off when it has set. The result is an exact, if flexible, copy of the original surface. The specimens illustrated in Figure 5.5 are all internal moulds, except for Fig. 5-f, which is a latex replica.

Since the beds in the North Esk Inlier are vertical, it is possible to walk northwestwards over their upturned edges, and thereby facilitate a kind of time travel, from the oldest rocks to the youngest. We have mentioned that the Silurian is divided (chronostratigraphically) into four Series. The lowest two of these, the Llandovery and Wenlock, are represented in the Pentland Hills, and locally, the Silurian succession is further divided into five Formations, which can be traced all the way across the inlier, striking approximately NE–SW. The first four of these Formations belong to the Llandovery Series, but the base is not seen and we do not know how far down into the Llandovery they extend. The base of the fifth, or Henshaw Formation, approximately corresponds to the base of the Wenlock, but whether the higher beds of this non-marine Formation extend upwards into the Ludlow or Pridoli remains unknown. We should briefly consider each of these Formations in turn.

We begin with the oldest, the Reservoir Formation, which remains the hardest to interpret. It is very thick, consisting of mudstones and siltstones. There are very few fossils or sedimentary structures, and we still do not really know the depth of water in which it was deposited. It was probably laid down in deeper waters than was the overlying Deerhope Formation, in which fossils become rather more abundant. Near the base of the latter, along the Gutterford Burn, are some layers with broken shells. These may have been deposited in a storm, hence the smashed shells, or they may have resulted from debris flows, which travelled chaotically down a submarine slope. Here also is a bed rich in the remains of eurypterids or water scorpions, discovered in the late 19th century, which have been the subject of many publications. There are some greenish layers with graptolites, and dendroids, a kind of graptolite rooted to the sea floor rather than being planktonic. Higher in the sequence, in the middle of the Deerhope Formation, are several layers in which the remains of fossil corals are abundant, along with brachiopods and bivalves (Fig. 5.4). These corals are small and of various kinds; they all belong to extinct groups. They are always found in the more sandy horizons within the sequence, and probably the sea floor was too soft and sloppy for any organisms to be able to settle, and it was only when the surface was stabilised by sand coming in that they were able to do so. The corals and other organisms were able to flourish for some length of time until they were smothered by an influx of suspended sediment. It was only after this had settled and more sand was deposited that they were able to thrive once more. After some further time, however, the rate of sediment influx became too great for any living creatures to inhabit the sea floor; this increased rate of sediment influx was related to the next phase in the history of the area.

If you look at a map of the world you will see that there are many places where an offshore bar, or chain of barrier islands, is stretched along the coast. Between the barrier and the coastline proper is a lagoon, which may be fully marine, brackish or even freshwater, if the connection between the lagoon and the sea is closed. The Baltic coastlines of Poland and Lithuania, or the Venice lagoon are good examples. These are large offshore barrier systems, but there are many smaller ones. It has been estimated that some 13% of the world’s coastlines have such barrier systems. They form by wave action where the slope of the sea floor is about 10%, and usually build up by the drift of sand some distance out from the shore. Spurn Head, on the north-east coast of the Humber estuary, is a British example of such a barrier forming at present. In the Pentland Hills, the Cock Rig Formation, which overlies the Deerhope Formation, is a fossil version of the same thing. The rocks in this formation are hard white sandstones, and they show a sequence of informative structures which makes their interpretation possible. The contact between the two formations is transitional; the upper beds of the Deerhope Formation become increasingly sandy towards the top, though regularly bedded. The lower part of the Cock Rig Formation, on the other hand, consists of irregularly lenticular packets of sandstone, often cross-bedded. These were deposited in a high-energy environment of breaking waves, just offshore from the barrier system, and above fair-weather wave base. In such a turbulent situation sedimentary packages were quickly deposited and just as rapidly sculpted by wave erosion.

In some places pebbly or conglomeratic horizons are to be seen, often with herringbone cross stratification, in other words where successive cross-sets face in different directions. Such herringbone structures were formed in tidal conditions, in alternate ebb and flow tides, and here they represent the relics of tidal channels, cutting through barrier islands, where the currents were strong enough to shift the pebbles. These channels formed rapidly, probably during storms, and became blocked soon afterwards. All the evidence is consistent with deposition just offshore, in an area of strong wave action. Now there comes a significant change in sediment type. For the upper part of the Cock Rig Formation consists of vast, flat, continuous sand-sheets, sometimes with ripple-marked upper surfaces. All these were beach deposits, forming successive long strands, either just above high- tide level or exposed during low tides. So why do we interpret these as part of an offshore barrier system, and not just an ordinary shoreline?

It is because the overlying sediments of the Wether Law Linn Formation are clearly marine, with abundant invertebrate fossils of known salt-water types. If it were not so we would find sediments of terrestrial origin above the white tabular sand sheets. Instead these are directly overlain by brown sandstones with broken shells, deposited in a still high-energy situation where waves broke over the barrier, but on its lee side. These grade up into fine mudstones with abundant trilobites, brachiopods, and other invertebrates, a rich, fully marine fauna that flourished in the quiet waters shoreward of the barrier. We can follow this upwards for a few metres, with no great change. Then there comes a 10 cm layer of clay, which has a dramatic effect upon the fauna, and especially on the brachiopods. The rich and diverse assemblage of brachiopods, with some twenty species, below the clay is immediately replaced by a different assemblage above it. This upper assemblage is much less diverse, and there are only two common brachiopods, Eoplectodonta penkillensis and Visbyella visbyensis, with just a few stragglers from the earlier fauna. These two species occur in great numbers, and seem to have colonised the sea floor very soon after the deposition of the clay. They were opportunistic species, colonising vacant ecospace once it had become available. Once they had established themselves, they stayed there. But why should this clay have had such a dramatic effect on the sea floor life of the time? It is because it is a bentonite, orginally a volcanic ash. Somewhere there must have been a volcano, erupting clouds of ash which rained down and settled on the sea floor. In the open sea such ash-falls would not necessarily have had a great influence upon the resident fauna. Yet in the more confined space of the lagoon, the blanketing effect was much more severe – a kind of Pompeii situation. We do not know where the volcano actually was. Although other evidence of Silurian volcanoes is largely lacking in Britain, many ash-bands in Silurian sediments elsewhere testify to their former presence. In Southern Scotland such bentonites interbedded with the graptolite shales may constitute as much as 30% of the total thickness of the total sediment. The volcanoes, however, which were probably ranged along the leading edge of the advancing Avalonian continent, have long since been covered by Upper Palaeozoic sediments.

The Eoplectodonta penkillensis – Visbyella visbyensis assemblage seems to have continued for some time, but higher in the sequence there is clear evidence of increasing ecological instability. Visbyella all but disappears, and the proportion of juvenile Eoplectodonta specimens becomes very much higher. These small brachiopods were killed early in life, as a result of adverse environmental conditions, and the main factor was probably fluctuating salinity. The advantages of living in a coastal lagoon are an abundant food supply and protection from rough waters.

An ever-present danger, however, is the prospect of salinity changes. In such a partially enclosed environment, salinity may decrease considerably following a tropical downpour. Conversely, evaporation during a dry season may significantly raise it. On the whole, brachiopod physiology is not adapted to cope with such changes. Eoplectodonta seems to have struggled for a while, and then disappeared; it is never found above this level. Instead we find a very diverse assemblage of gastropods (marine snails), bivalves (clams), and ostracodes (small, bivalved, bean- shaped crustaceans). And it is a fact that many modern representatives of these same groups are much more tolerant of salinity fluctuations than are brachiopods. The horizons with these fossils carry a more diverse fauna than anywhere else in the North Esk Inlier, with bryozoans, sponges, and many other fossils, despite the virtual absence of brachiopods and trilobites. Clearly it was a very tranquil water environment, judging by the number of large intact sponges, and ostracodes and bivalves with both valves present. Presumably this environment represents the inner and probably deepest part of the lagoon. Yet above this level, fossils become very sparse. It may be that the sea floor had become soft or that the environment had changed in other, currently unknown ways.

Fossils are next encountered in abundance close to the top of the Wether Law Linn Formation, and these are all fully marine. But it is an entirely different fauna from that of lower levels. A single brachiopod species, Pentlandella pentlandica is dominant, there are also large ostracodes, of the species Entomozoe tuberosa, as well as small ostracodes endemic to the Pentland Hills named Craspedobolbina (Mitrobeyrichia) impendens. These tiny ostracodes have the longest name of any of the fossils in the Pentland Hills. There are also elongated phosphatic-shelled jellyfish known as conulariids, and also some trilobites and bivalves unknown from older beds. In addition we find many straight-shelled cephalopods, like uncoiled living Nautilus, which may have floated for long distances before becoming stranded along the shoreline. For this fauna was deposited in a very shallow-water environment, but still protected, judging by the number of unbroken and intact shells. Directly above this come the first red-beds; wind-blown marls and desert sands. In places there are conglomerates, deposited by flash floods in a desert- fluviatile environment. These beds belong to the Henshaw Formation, some 800 m thick, and all are terrestrial in origin, except for a brief interlude when a marginal sea returned, depositing crinoid columnals and fish scales. In this part of the Midland Valley, these red-beds signify the onset of semi-arid desert conditions, typical of the overlying Old Red Sandstone, but in mid-Silurian times rather than in the early Devonian, as elsewhere in the British Isles.

It is interesting to note the difference between the offshore barrier system and protected lagoon in the North Esk Inlier with environments of equivalent age but further south-west, towards Girvan. For here, coarse beach sands and micro-conglomerates replete with smashed shells testify to a high-energy open shoreface, turbulent and wave-dominated. There was no barrier system, but as in the North Esk Inlier, these fossiliferous beds are succeeded by desert sandstones. The contrast could hardly be more striking.

In summary, in the North Esk Inlier we have a magnificent example of an offshore barrier system impounding a sheltered shoreward lagoon. This lagoon was probably several kilometres across and lapped against a desert shoreline. It lasted for thousands or tens of thousands of years. And as sea-levels slowly fell the whole system migrated southwards, yet still retaining its form for a long time. Accordingly, it is possible to travel across the present land surface from SE to NW encountering the five formations in turn, from relatively deep-water sediments, through the barrier system, the lagoon and finally the desert. We do not know how long this Silurian desert persisted, for the North Esk Inlier is truncated at its northwestern end by a major fault beyond which no Silurian rocks are exposed.

The other two Silurian inliers are very sparsely fossiliferous and it is not yet known where they fit into the sequence. According to one view they are lower in the sequence than the sediments of the North Esk Inlier, but evidence from microfossils suggests that they may actually be of the same age. If so, then they represent a deeper water equivalent of the North Esk Inlier sequence, but until more evidence is adduced, nothing much can be said at present. The rocks in the two inliers are rather monotonous grey beds, which yield few fossils. But they are set in a very pleasant countryside, with interesting glacial features, and it is well worth examining them on a walk from Bavelaw to Glencorse, in an area of striking natural beauty.

Our Silurian inliers give information only about a small part of a large continental shelf north of the Iapetus Ocean, as it was before its final collision with the Avalonian continent, which approached obliquely from the south. The collision occurred some time in the late Silurian, when the Silurian sediments of the shelf were compressed and tilted into their present vertical attitude. They were elevated into a new upland area that was then weathered into the irregular landscape upon which the lavas and desert-fluviatile sediments that constitute the ‘Lower Old Red Sandstone’ were deposited. It would be some 80 Myr later, in the early Carboniferous, before the sea returned to this part of what is now the Midland Valley.