“But, perhaps because so familiar, it does not at first strike one that these outlines point to conditions which have now entirely passed away.” Clement Reid, 1887.
These notes were prompted by the review paper “Chalk Landforms of Southern England and Quaternary Landscape Development” Whiteman & Haggart, (2018). One of the conclusions which can be drawn from their review is that the ideas of the Victorian geologist, Clement Reid, on the origins of chalk landscape, were mostly correct but were ignored during the middle years of the 20th century.
The early model of landscape development.
At the end of the 19th century the American geomorphologist William Davis proposed a process of landscape development comprising three phases – uplift, erosion and stability which ended with a low relief erosion surface. This “Davisian Cycle” was used to explain landscape development for most of the 20th century. A local example of its use was the speculation concerning the likely height of the “Wealden Dome” (the Weald-Artois Anticline) before it was eroded. The presumption seems to have been that the uplifting dome retained its shape whilst erosion was somehow put on hold until the uplift was completed, and then the erosion phase started. This was the sequence described by the Davisian Cycle.
The problem of dry valleys in chalk.
Geologists and geomorphologists had long been interested in the paradox of the deep, steep sided, dry valleys in chalk – a permeable rock which should not be capable of sustaining rivers of the size and energy required to erode such valleys.
When looking for clues about the causes of geological features, the “The Principle of Uniformitarianism” (or Uniformity) has been traditionally applied. This has its origins in the late 18th century and is based on the ideas of James Hutton (“The Father of Modern Geology” 1726 – 1797). This principle, firstly, recognises that rocks are continually changing very slowly over a long period of time, rather than being static. Secondly, it relies on the assumption that the processes which formed those rocks followed the laws of physics and chemistry which have been operating throughout time. Therefore, it should be possible to understand the formation of rocks by examining the processes which are forming similar features today. In other words, “The present is the key to the past”.
When Hutton’s Principle was applied to the paradox of dry valleys in chalk, it led to an assessment of the conditions necessary to enable rivers to flow over permeable ground. The conclusion was that the only time rivers achieve this, is when the water table is at or above the base of the valley. This was possible if either,
- the amount of rainfall was greater in the past than it is today,
- tectonic uplift has since raised the valleys above the water table,
- sea level has fallen, lowering the ground water level,
- or a combination of all three.
These conditions could be explained within the framework of the Davisian Cycle, but significantly, it was assumed that weathering and erosion were operating in a cool temperate climate, like the current climate.
The Victorians’ fresh thinking.
Victorian geologists were interested in the dry valleys in chalk paradox. Clement Reid was working for the Geological Survey on the dry valleys in the dip (south-facing) slope of The South Downs, a similar geological environment to north slope of The North Downs.
Reid had difficulty in attributing the shape of these dry valleys to erosion by streams. He was puzzled by the steep slope at the head of these valleys. He could not envisage how water could have issued here, where the slope of the bottom of the valley exceeded the slope of the “plane of saturation” (water table) in the chalk, which he stated “never exceeds, if it reaches, 60 feet in a mile” in contrast to the steepest valley slopes which reached up to “500 feet in the mile”. He noted that even after the heaviest rainfall, no streams flowed from these steep head-slopes.
Reid did acknowledge that those rivers which cut through the South Downs escarpment, and those occupying the lower, flatter, reaches of the larger coombes were controlled by the local water table. They had cut down into the valley floor following the water table in response to uplift phases and left terraces as evidence of each erosional stage. The absence of such terraces in the steep sided coombes suggested to him that those had not been formed by erosion of rivers cutting down in response to the falling water table.
Geologists use the nature of sediments arising from erosion for clues as to what erosional processes were operating, as well as looking at the landforms. Reid began by concentrating on the sediments on the dip slopes of the South Downs as a means of trying to understand the processes which had formed them.
He looked at the widespread sheets of sediment (known locally as “Coombe Rock” ) which extended across flat land, some distance away from the dry valleys in the escarpment. He saw that these deposits did not show any of the characteristics of a sediment laid down by glaciers, by the sea, or by rivers; he found no striated stones, which would have indicated a glacial origin; he found no rolled stones, which would have indicated erosion by the sea or rivers, or any sedimentary bedding structures characteristic of sediments deposited under water. Neither could he find any remains of aquatic flora or fauna which would have been expected if water was present. But he did find the remains of amphibious fauna typical of those living in marshy environments. He also noted that the Coombe Rock was characterised by inclusions of intact chalk, a weak rock which ought not to have survived transportation and deposition by flowing water.
Addressing a meeting of the Geological Society in 1887, when referring to the typical rolling outlines of the chalk downland, Reid remarked:
“But, perhaps because so familiar, it does not at first strike one that these outlines point to conditions which have now entirely passed away.” (Reid, 1887).
He proposed that those conditions had existed in the Pleistocene (Ice Age) during periods of a Tundra climate and which disappeared with the ice retreat. Permafrost (ground which is below freezing for more than two years) would have been the norm. When such Chalk included moisture, it froze and formed an impermeable barrier, so that any summer rainfall or snow melt would have flowed over, not into, the chalk as brief but vigorous streams. This run-off easily eroded the already frost shattered, weakened, but still frozen chalk. He suggested that the chalk was still frozen when it was re-deposited, which would explain why individual blocks remained intact. He proposed that this was the process which formed the deep, steep sided valleys and was the origin of the coombe deposits.
The phenomenon of torrential flows caused by spring thaws running over frozen ground was widely recognised at that time. Dr Hinde, a delegate to the meeting at which Reid presented his paper, was reported to have responded that in Canada,
“….a greater amount of denudation is effected in 24 hours, at the sudden break-up of frost in spring, than takes place in the whole of the rest of the year.”
In the same year that Clement Reid presented his paper, The Rev. E. Maule-Cole, the vicar of Wetwang, in the East Riding of Yorkshire, made the following remarks occasioned by a particularly cold winter on the Yorkshire Wolds:
“The month of January, 1887, will long be remembered on the Chalk Wolds of East Yorkshire, as presenting one of the most curious sights ever witnessed. A succession of frosts had frozen the bare ground so hard, that no rain could penetrate. What little rain fell was quickly converted into ice. On a sudden came a rapid thaw; and in a few hours the dale bottoms were converted into roaring torrents, in some cases 3 feet deep. The ground was still frozen hard underneath. The melting snow could not penetrate, and so rivers ran in dry places.” (Maule-Cole Rev. E., 1887).
Reid made the point that whilst the processes forming the Tundra landscape are available for inspection in Siberia and North America, those regions have been repeatedly glaciated so that any younger softer rocks, comparable with the chalk, have been removed leaving only the ancient hard bedrock. Consequently, there are no opportunities to study the processes of weathering and erosion on a weak permeable rock (such as chalk) in a Tundra climate. In effect, he was demonstrating that Hutton’s Principle cannot be applied if there is currently nowhere on earth with conditions comparable with those in the past.
Reid’s explanation invoking the effect of ice melt torrents may have explained the steep sides of the coombes but it did not explain their steep head walls. Interest in the paradox of chalk valleys continued into the early 20th century.
Osborne White (1924) suggested that, when referring to the south facing coombes in the dip slope of the South Downs above Brighton,
“…the sloughing-off of the superficial portions of the Chalk….affected the terminal southern slopes, as well as the lateral slopes, of the downland ridges.” (by terminal southern slopes he meant the south facing head-slopes) and also, when referring to a section near Black Rock Brighton, that…
“the bulk of the Coombe Rock there came down the seaward slope of Red Hill in a pasty condition….”
Here he was describing a type of mass movement which occurs in fine-grained soils with a sufficiently high water content to make it behave as a viscous fluid (gelifluction and solifluction). Wet frost-shattered chalk would take on these characteristics. He was also making the point, which Reid had not, that this process would affect the sides and head-slopes equally. So, this supplied the missing explanation of the origin of the steep head-slopes which Reid had recognised but had not otherwise explained.
The 20th century resurrection of the Davisian Cycle.
As the 20th century progressed, the Victorians’ ideas seemed to have been side-lined, and the Davisian model was revised. It was acknowledged that the uplift and erosion processes occurred concurrently, rather than being discrete phases, and it was also recognised that there were several uplift and erosion cycles. Subsequent discussions centred on the timing of these cycles or the, so called, “denudation chronology”.
Wooldridge and Linton, (Wooldridge & Linton, 1939), working within the framework of Davisian Cycles, identified three fragmented plateaus in the North Downs and the London Basin which they interpreted as the remains of three erosion surfaces. Each successive surface showed a decreasing degree of tilt which they interpreted as evidence of progressive uplift of the developing Weald-Artois anticline (reviewed in Jones, 1999). They also noted that oldest surface was distorted whereas the younger surfaces were more-or-less undisturbed, from which they concluded that significant uplift and disturbance must have occurred after the formation of the earliest surface but before the younger surfaces were formed. They called this event the “Alpine Storm”, a mountain building episode dating from around 23 million years ago. This provided the framework for interpreting landscape development for most of the middle years of the 20th century.
The mid-century interest in melt water floods.
During the first half of the 20th century, evidence for the existence of meltwater floods was recognised in river deposits. Members of a Geologists’ Association visit, during Whitsun 1925, looked at River Terrace Deposits exposed in gravel pits the Stour valley, at Sturry, near Canterbury. They revealed:
“…masses of Woolwich Clay [older] rested at high angles on even-bedded, laminated brick-earth [younger] and at the point of contact the laminae were broken, bent or twisted.” (Dewey et al., 1925).
The authors used this strange occurrence to demonstrate that the normally malleable clay must have been ripped up from the riverbed in a solid, frozen, condition to have survived transportation intact, and that a high energy water flow was involved.
The Geological Survey Memoir for Canterbury and Folkestone (Smart et al 1966) includes the following (unusually imaginative) description of what one of these meltwater torrents might have looked like:
“…the picture of conditions…is thought to be one of large quantities of rapidly flowing water moving, besides their load of chalk, sand and gravel, blocks of floating and rolling ice charged with debris. In most cases debris must have choked the water courses and become piled into aggradations of debris, with the higher levels channelling into the lower and overstepping them to rest upon solid strata.” (Smart et al., 1966).
This description of conditions was referring to the Great Stour valley but similar floods on a smaller scale would have occurred in the tributary valleys, for example the lower part of the Elham valley, as evidenced by the River Terrace Deposits, which are shown on the geological maps, on the upper sides of the valley from Barham downstream and mentioned in Smart et al. (1966).
These “torrent bedded” deposits could be several metres thick, so rather than being interpreted as the result of seasonal thaws it was concluded that they were the result of large-scale thawing at the end of each glaciation and, because they comprised deposits of various ages, they indicated that there were several glacial retreats.
The significance of this was that it placed the formation of chalk downland landscape to within the Ice Age, before the last ice melt and the formation of the deep dry valleys described by Read and Osborne White. There seems to have been a tendency to attribute the formation of all Chalk downland valleys to glacial meltwater torrents, even those steep sided coombes which the Victorians had difficulty in attributing to flowing water.
The recognition of the role of climate.
Gibbard & Lewin, (2003), and reviewed in Whiteman & Haggart, (2018) looked at the effects of weathering on rocks as the climate cooled from warm temperate to tundra at the onset of the Ice Age. They concluded that it was the onset of a colder and unstable climate, rather than tectonic uplift with the resulting rejuvenation of rivers, which was the instigator of valley erosion in Chalk, although it is agreed that uplift contributed.
Chemical weathering in warm climates produces fine grained residues above a deeply buried surface, etched into the bedrock. Evidence for the presence of chemical weathering, prior to the Ice Age was found by examination of sediments which were newly accessible from boreholes in the North Sea and which originated from erosion of the Weald – Artois anticline. These sediments were fine grained, and in quantities consistent with products of chemical weathering. There was no evidence of the coarser sediments which would have resulted from erosion in a cool temperate climate, which were the conditions assumed to exist by Wooldridge & Linton, (1939).
There was a re-evaluation of the Wooldridge and Linton erosion surfaces. The high-level surface that had been attributed to marine erosion by Wooldridge and Linton, (1939) was re-interpreted as one of those deeply buried chemically etched surfaces, formed during a warmer pre-Ice Age climate, which had been exposed by removal of the overlying weathering debris.
The Clay-with-flints Formation found on the surface of the chalk had been recognised for some time as a deposit which was older than the recent drift deposits, probably pre – Ice Age. The re-assessment led to the understanding that it was a relict tropical soil, a chemical weathering deposit.
Reviving earlier ideas.
The Devil’s Kneading Trough coombe. Kerney et al., (1964)
The Devil’s Kneading Trough coombe is in the scarp slope of the downs above the village of Brook (Photo). It has unusually regular sides, even by coombe standards, and its head is double ended giving it a fish tail shape. (The abrupt changes of slope are thought to have been accentuated by Iron Age ploughing adjacent to its top edge and along its base).
A very comprehensive study by Kerney et al., (1964) noted (as Reid had done) that the remains of flora and fauna in the sediments of the valley bottom were predominantly land based, albeit marshy, with very few representatives from streams, indicating that processes other than flowing water were involved in the formation of the coombe. (They also noted the presence of artefacts from the Neolithic, Bronze and Iron Ages and suggested that the abrupt changes of slope were enhanced by Iron Age ploughing adjacent to the top and along the base of the coombe).
They found by dating the sediments eroded from the coombe and deposited in front of it, that most of it dated from a 500-year period between 8,800 and 8,300 BCE comprising what they called Zone III deposits. Revised by later analysis to 9,700 BCE. for completion of the erosion phase.
They also estimated that the Zone III deposit comprised about one third of the total volume of chalk eroded (Kerney et al., 1964, p185) (and this was likely to be an underestimate as it did not take into consideration the unmeasurable volume of chalk which would have been carried away in solution and suspension). This implies that a volume equivalent to at least half of the volume previously removed was eroded during the last erosion phase.
They attributed the rapid erosion to a humid climate with temperature hovering around freezing which was conducive to mass movement of frost shattered chalk. This example shows that it is the alternate freezing and thawing of chalk, rather than a constant frozen condition, which breaks the chalk down to a material which is susceptible to mass movement by solifluction or gelifluction. It also shows how quickly this type of erosion can remove large volumes of chalk.
The relative contribution of erosion by water or by mass movement to the formation of chalk coombes seems to be the subject of current discussions. But it is accepted that the weathering occurred in “conditions which have now entirely passed away” and that chalk downland is a recently formed tundra landscape clad in temperate vegetation.
Andrew Coleman
Rev. 20/11/2025
References:
Dewey, H., Wooldridge, S. W., Cornes, H. W., & Brown, E. E. S. (1925). The geology of the Canterbury District. Proceedings of the Geologists’ Association, 36(3), 257–290. https://doi.org/10.1016/S0016-7878(25)80010-5
Gibbard, P. L., & Lewin, & J. (2003). The history of the major rivers of southern Britain during the Tertiary. In Journal of the Geological Society (Vol. 160). http://jgs.lyellcollection.org/
Jones, D. K. C. (a). (1999). Evolving Models of the Tertiary evolutionary geomorphology of southern England, with special reference to the Chalklands. In Ulift Erosion and Stability: Perspectives on Long-term Landscape development. (Vol. 162, pp. 1–23). Geological Society, London Special Publications.
Kerney, M. P., Brown, E. H., & Chandler, T. J. (1964). The Late-Glacial and Post-Glacial History of the Chalk Escarpment near Brook, Kent. Philosophical Transactions Of The Royal Society of London., 248(745), 135–204.
Maule-Cole Rev. E. (1887). Note on Dry Valleys in Chalk. Proceedings of the Yorkshire Geological and Polytechnic Society., 9, 343–346.
Osborne White, H. J. (1924). The geology of the country near Brighton & Worthing.
Reid, C. (1887). On the origin of Dry Chalk Valleys and of Coombe Rock. Quarterly Journal of the Geological Society, 43, 364–373.
Smart, J. G. O., Bisson, G., & Worssam, B. C. (1966). Geology of the Country around Canterbury and Folkestone. Her Majesty’s Stationery Office.
Whiteman, C. A., & Haggart, B. A. (2018). Chalk Landforms of Southern England and Quaternary Landscape Development. Proceedings of the Geologists’ Association. https://doi.org/10.1016/j.pgeola.2018.05.002
Wooldridge, S. W., & Linton, D. L. (1939). Structure, Surface and Drainage in South-east England. Institute of British Geographers.
Geomusings 