I explained in the Introduction that it was gradually realized from geological evidence in the Alps that there had been a succession of ice ages, and how it was not just in Britain that seeking the local equivalents of these Alpine ice ages became something of an obsession. They were regarded by many workers as geological universals that were waiting to be identified in every continent of the world, and this often impeded progress rather than helped it, by providing straitjackets of conformity even where things looked very different. What was needed to break out of the straitjacket were much more detailed and continuous archives of past climatic changes than could be found in the fragmentary sequences of the Alps or the Scottish highlands, that could help translate past climatic records across the whole planet. And just as important as finding better records was the need to know what lay behind the ice ages. As with the model of continental drift, which languished in neglect until geological data were discovered which showed that the continents really had moved around, an explanation for the ice ages was proposed in 1924, but it was fifty years before its veracity began to be generally recognized.
A Serbian mathematician called Milutin Milankovitch used very precise calculations of the orbits of the planets to model the position of the Earth in relation to the sun over the last 600,000 years. The Earth’s orbit is not exactly circular, so the time of the year when it is furthest from the sun (and thus receives less heat) changes through time, over many thousands of years. In addition, two different factors affect the tilt of the Earth’s axis of rotation in relation to the sun through time, accentuating or reducing the differences in the seasons. From these data Milankovitch was able to reconstruct the varying patterns of summer warmth in the northern hemisphere and show how the polar ice caps would have waxed and waned in the face of hotter or cooler summers. If all three factors (orbit shape and the two tilt factors) increased the amount of sunlight falling on the Earth, there was an interglacial. If the factors all worked in the opposite direction and sunlight was reduced, the Earth suffered an ice age. Consequently, he was able to identify three periods of particularly low solar warmth that might correspond to the Gunz, Mindel and Riss glacials, as well as a prolonged period of warmth that could have corresponded to the Mindel-Riss or Great Interglacial of Penck and Bruckner. But Milankovitch realized that most of the time the three factors, working on cycles of about 100,000, 40,000 and 20,000 years, would be pulling in different directions, so the reality of climate change was going to be a lot more complex than what had been observed in the Alps.
In the 1950s a ‘Rosetta Stone’ of past climate at last began to be uncovered, not from desert sands but from the ocean floor. The man who began the laborious process of reading the record was an Italian palaeontologist called Cesare Emiliani. Despite the disruption of World War II, he managed to complete his doctoral research on marine microorganisms in Italy, and then moved to Chicago to continue his studies. At that time, the first deep cores were being drilled through the ‘calcareous oozes’ (accumulations of the chalky shells of dead microorganisms) on the tropical ocean floors, with the aim of looking at the environment of the oceans through time – the deeper the ocean floor sediment, the older the record it contained. Although Emiliani was interested in the species of fossil microorganisms preserved in the deep sea cores, as these might give clues about past changes in the oceans, he concentrated on the chemistry of their carbonate shells or tests. In theory, the composition of the tests the creatures built should reflect the kind of ocean in which they had lived, whether warm or cold, fresher or saltier. In particular Emiliani analysed the ratio of two stable isotopes of oxygen – 16oxygen (the normal kind) and 18 oxygen (a rarer and heavier form) – in the tests sampled at 10cm intervals of depth. Using the theory that more 16O was an indication of warm oceans, and more 18O colder oceans, he was astonished to find that the oxygen isotope ratios in the cores fluctuated dramatically in a saw-tooth pattern, and that even the subtropical Pacific and Caribbean oceans had regularly been several degrees colder in the past. What was more, he identified seven cold stages in the Caribbean cores, and fifteen in the Pacific cores, compared with Penck and Bruckner’s four in the Alps. Counting from the top of the cores, Emiliani identified each warm stage by an odd number (our present interglacial is thus Marine Isotope Stage 1, or MIS 1), and each cold stage by an even number (the peak of the last ice age is thus MIS 2).
The next development in unravelling the marine record came from two geologists, Nick Shackleton at Cambridge and the American John Imbrie. They suggested that factors other than local water temperature lay behind the oxygen isotope fluctuations, and these were global. In their view, the main reason behind past oxygen isotope variation in the oceans was the size of the ice caps. Water that evaporates from the oceans is isotopically ‘light’ – that is, it has proportionately more 16O. Ice caps are made up of atmospheric water that has fallen as rain, hail or snow, or has been directly frozen from the air by low temperatures, like frost. Thus if ice caps grow, they store up more and more light water, and the ocean waters left behind get ‘heavier’ (more 18O). If ice caps melt, the 18O-rich oceans get diluted with more 16O again. So a low proportion of 18O atoms in the marine core fossils indicates a time of small ice caps and high sea levels (an interglacial), while a high proportion of 18O must indicate large ice caps and a relatively low sea level (a glacial).
To be fair to Emiliani, he was aware of ice cap influence as well, but he considered it much less significant than the local water temperature in its effects on the chemical composition of the micro-fossils. But with this breakthrough in interpretation, it was at last possible to get a global view of past climatic changes and test Milankovitch’s theories. To do this, there had to be a time scale for the isotope changes in the cores, and hence for the warm and cold stages. Unfortunately, because of their particular chemical composition, it is very difficult to date ocean floor cores using the radioactive decay clocks that work well on land deposits, such as in radiocarbon, uranium-series or argon-argon dating.
Nevertheless, in the 1960s it proved possible to date ancient coral terraces by uranium-series decay that had apparently been laid down in the high sea levels of the last interglacial, suggesting that Emiliani’s MIS 5 was about 100,000 years old. Even more useful, there is one clear dating signal that is worldwide and recorded in the cores. The Earth has a liquid iron core and, as it spins, a natural dynamo is produced, giving the planet an electrical and magnetic field, with a North and South Pole. For reasons that are still poorly understood, but are thought to be related to instability in the liquid core, the Earth’s magnetic field fluctuates, and periodically it completely switches from a situation like today – when a compass needle points to the north – to the opposite condition, where a compass needle would instead point to the south. These events are called palaeomagnetic reversals, and they have happened many times in the history of the Earth. When sediments are laid down, their metallic content can preserve a snapshot of the Earth’s magnetic field at the time, and where the poles were. The last major reversal was about 780,000 years ago, and that switch is recorded in many deep-sea core records that were being laid down at the time. As we said earlier, Emiliani numbered the isotope stages, starting from the present-day warm stage as 1. The palaeomagnetic reversal occurs in the cores close to the transition between MIS 19–20, and in itself this is a revelation, since it means that while Penck and Bruckner and their followers recognized four cold stages and four warm, the deep-sea records have over twice as many in the last 800,000 years, and many more before that time. Given this advance in dating the cores, Shackleton, Imbrie and colleagues were able to show in the 1970s that Milankovitch’s theoretical astronomical model and the data from the ocean cores fitted together extremely well – the Earth’s orbital changes truly were the pacemakers of the ice ages. And this in turn made it possible to estimate the dates of the various marine stages much more accurately.
Despite the general progress that was being made in understanding
the ice ages during the 1960s and 1970s, a fierce debate was developing in Britain at that time about how to build up an accurate local picture of the ice age sequence. On the one hand, there were those who relied on vegetation changes recorded from pollen in ancient lakebeds, such as the one at Hoxne. In this view, enshrined in a detailed 1973 report by the British Geological Society, there were four interglacials – the Cromerian, Hoxnian, Ipswichian and Holocene (our present one) – with three glaciations in between, the Anglian, Wolstonian and Devensian. On the other hand there were palaeontologists such as my late colleague at the Natural History Museum, Tony Sutcliffe, who believed that the sequence of fossil mammals could give a better picture of the changing climates of Britain than could the pollen record. For example, based on pollen data, the Ipswichian interglacial included sites that contained hippos and elephants on the one hand, and horses and mammoths on the other. The pollen people argued that this merely reflected change within the interglacial. It would have been relatively cold at its beginning and end (when horse and mammoth could have lived), and warm in the middle (with the hippos and elephants). Sutcliffe said that only the hippo-elephant fauna was truly Ipswichian, while the sites with horses and mammoths were in a different geological position, and represented an earlier interglacial, recorded at sites in Essex such as Ilford and Aveley.
There was no room in the pollen-based scheme for such an inter-glacial, but Sutcliffe argued that there was unrecognized complexity in the British record, and in reality it was more like the picture emerging from the marine isotope records. AHOB members Andy Currant (Tony’s successor as the Museum’s expert on Pleistocene mammals), Danielle Schreve and Roger Jacobi have taken this approach even further by developing Mammal Assemblage-Zones (MAZ) that typify periods of time and have provided an entirely new framework of which AHOB is making great use.
The River Thames has played an important, if indirect, part in helping to settle some of the big questions about the sequence of British ice ages. This is because of the significant change of course in its early history, when it was pushed southwards to its present position by a huge ice cap, the largest to cover southern Britain. Once the Thames was in its new position, it started to accumulate masses of new sediments that have given us many clues about the ancient human occupation of Britain. And there is a key signal in the marine cores of a particularly pronounced cold stage (MIS 12) that occurred about 450,000 years ago, and which was followed by a particularly marked warm stage about 400,000 years ago (MIS 11). That cold stage almost certainly corresponds to the Anglian ice advance in Britain and the diversion of the Thames. The succeeding warm stage is then the Hoxnian interglacial, in which we can now place the famous Swanscombe skull, as we will see in the next chapter. But what about the warm period about 500,000 years ago, before the Anglian ice swept down most of Britain? This stage, MIS 13, was apparently not quite as warm or prolonged as the Hoxnian, and as we have seen it was thought until recently that people never made it to Britain at that time, and Swanscombe was our oldest evidence of human occupation.
Opinions started to change after 1969, when a site at Westbury in the Mendip Hills of Somerset began to be exposed during limestone quarrying. A small earthy fissure appeared in the top of the north face of the quarry after blasting, and fossil animal bones began to tumble out of it, including those of bears and rhinos.
At first it was thought that the fossils indicated a late Pleistocene age, perhaps only about 100,000 years ago, but gradually some very unusual species were noted, ones that were rare or previously completely unknown in Britain. These included a primitive form of cave bear, a primitive rhinoceros, a dhole (wild dog), a jaguar, and a scimitar-toothed cat, suggesting that Westbury must be older than Hoxnian sites like Swanscombe. Yet the vole clock indicated that Westbury could not be as old as the Cromerian interglacial deposits of East Anglia. As we have seen, the vole clock is based on the evolutionary transition between the primitive species Mimomys savini and its descendant Arvicola terrestris cantiana. Cromerian and pre-Cromerian sites in Britain and Europe have Mimomys, while Hoxnian and later sites have Arvicola fossils. Mike Bishop, then a curator of geology at University College London, studied the Westbury fossils for his doctoral thesis and argued that the combination of typical Cromerian mammals with Arvicola meant that they probably represented a hitherto unrecognized interglacial stage in the British Pleistocene, one that lay between the Cromerian (estimated at about 600,000 years old) and the Hoxnian (about 400,000 years). Moreover Bishop had been given, and had himself collected, what looked like stone tools from the site. If these really were stone tools, and they were the same age as the fossil mammal bones, they were older than any others yet dated in Britain. Bishop published these in the journal Nature in 1975, and their age was estimated at about 500,000 years, but scientific opinion was divided about whether they really were artefacts, and whether they really were as old as Bishop claimed.
Meanwhile something awful happened to the Westbury site. A conflict had been developing between the quarrying company and investigators like Bishop. The quarrying company had seen the earthy fissure enlarge with every blast of the rock face, from an initial width of about 25 metres in 1969 to about 70 metres in 1973. From their point of view, the loose deposits, while interesting, were an inconvenient and potentially dangerous obstruction to their need to quarry limestone. To Mike Bishop and an enthusiastic group of local supporters the site was of international significance and should be preserved, not quarried. The quarry operators took matters into their own hands in 1974 when they carried out a massive blasting operation, probably with the intention of completely removing the fossiliferous site. To their chagrin, instead of the intrusive deposits disintegrating to the quarry floor, the exposure grew even wider – to about 120 metres. It was now abundantly clear that the ‘fissure’ was in fact a huge cave chamber, filled with sediments to a depth of more than 30 metres, and the blasting had exposed a cross-section of its whole length.
In the bottom of the cave were yellow and white sands and silts, and near the top were red, brown and yellow earthy deposits containing most of the bones. Bishop was now persona non grata at the quarry, but the operators were aware that their work (and reputation) was potentially compromised. In 1975 they began negotiations with the Natural History Museum to undertake a full investigation of the cave, and in 1976 I and colleagues from the Museum began excavations that were to dominate our lives for the next eight years, and keep us busy on research and publications for a further fifteen. Blasting had removed much of the layered strata that Bishop had been studying, and parts of the cave were now dangerously unstable. We had to learn mountaineering techniques and at times excavated by dangling on ropes more than 30 metres above the quarry floor. The fact that the quarry itself was perched on the edge of the Mendips more than 200 metres above the Somerset levels, with distant vistas towards Glastonbury in one direction and the Bristol Channel in the other, sometimes made us feel we were parachuting or hang-gliding, particularly when strong westerly winds were blowing!
Our investigations showed that Bishop had got the overall picture of Westbury Cave right, even if our larger-scale excavations showed that he had got some of the details wrong. We were able to show that the cave deposits were much more complex than even he had thought, since there were actually two intersecting cave chambers, each with their own sequence. Each chamber consisted of many more strata than he had been able to recognize, covering a longer time period and showing much more variation through time. The lowest deposits were probably over 800,000 years old, with sparse remains of mammals, including the primitive vole Mimomys. As we excavated higher and higher up the cave we found evidence of environmental fluctuations in the mammals, with at least two warm stages, and a very cold stage at the top of the cave sequence at both ends, which probably represented the Anglian glaciation. The advanced vole Arvicola was definitely present in all the higher deposits, along with a rich collection of Cromerian fossil mammals, as Mike had claimed. We al
so recovered more possible tools made of flint and chert. Though none were as sophisticated as the hand-axe tools from Swanscombe, most experts are now convinced that their shape is the work of humans rather than Nature. Moreover, from near the top of the sequence at the eastern end, we found even clearer evidence that humans had been at Westbury at least 500,000 years ago, in the form of cut marks made by stone tools on the leg bone of a red deer. So Mike Bishop’s pioneering views were vindicated: humans really were in Britain during a warm stage just before the Anglian ice age. But it would take another discovery, from a site in West Sussex, to show us who these people were.
The Boxgrove site is located in a quarry near Chichester, and for many years flint handaxes like those known from Swanscombe were being found above marine sands that were preserved and quarried there. The marine sands were evidence that the sea once reached 10 kilometres (6 miles) further north of its present shoreline near Bognor Regis, at a time when the Isle of Wight was still joined to the mainland and a huge river, ancestral to the Solent, flowed by. The sea cut a massive cliff 30 kilometres (20 miles) long and 100 metres high (comparable to Beachy Head today) into the chalk hills of what is now the South Downs, and early investigators assumed from the height of the ancient sea and the excellent quality of the handaxes that the site was probably about the same age as Swanscombe. However, archaeological investigations there gradually developed into a huge project, involving over forty specialists and dozens of excavators, and radically changed perceptions of the site and its age.
Homo Britannicus Page 7