1. The Lower Quaternary, or Pre-glacial Succession

The vegetation changes deduced from pollen fluctuations, and corroborated by mollusca and foraminifera, in the deposits of the Ludham borehole (West 1961) east of Wroxham, which reaches a depth of -54m. O.D. can be summarised as a fluctuation from mixed woodland to oceanic heath repeated twice. The Ludhamian and Antian are interpreted as temperate stages as they contain the pollen of thermophilous trees, and as forest or woodland, as the A.P. frequency exceeds the N.A.P. The species represented are both deciduous trees, and conifers: some like Quercus, Alnus, Betula, Carpinus, Pinus, are components of our present forests, and some like Pterocarya, Tsuga, Picea, Carya, Eucommia, & a 'Taxodium' type are now exotic. This exotic element appears to be less well represented in Britain than in the equivalent deposits of the continent, and as the Lower Quaternary progressed trees such as Pterocarya and Tsuga were gradually eliminated. The intervening Thurnian and Baventian indicate extensive oceanic heaths with N.A.P. high and mainly composed of the grasses and Ericaceae, particularily Empetrum and Calluna. Both are interpreted as cold phases, but in neither is there any direct evidence of glaciation, although in the Baventian, which seems to have been colder than the Thurnian, there is evidence of periglacial processes, with solifluction, involutions, and ice wedge casts, perhaps signifying permafrost, while the probably contemporaneous deposit, the Weybourne Crag, contains erratics that may have been rafted by, or derived from icebergs.

The highest cold phase in the core is matched in the cliffs at Easton Bavents (ie. the Baventian) and the succession is then traced along the coast of Suffolk and Norfolk, through the Pastonian and Beestonian to link with the Cromerian Interglacial deposits which include the famous Cromer Forest Bed (the first to yield to pollen analysis, Thompson 1949, Duigan 1963). This complex series of deposits was worked out by West and Wilson (Nature 1966) and is often referred to as the Sherringham/Mundesley succession. The upper temperate stage, above the Baventian, in the Ludham borehole correlates with the estuarine sands and silts on the foreshore at Mundesley, and is called the Pastonian Interglacial by West as it contains a temperate flora and fauna. Oak and pine woodland appear to have been present, with Ulmus and Carpinus also represented, and these pass upwards into the so called 'arctic beds' of the next cold phase which West calls the Beestonian (after Beeston Cliffs near Sherringham) which again affords no evidence of glaciation, but like the Baventian contains ice wedge casts indicative of permafrost. The various organic deposits of the Cromer Series above the Beestonian afford the first real detailed picture of an interglacial cycle with all of the biostratigraphical zones represented, although soils and peats exposed in the cliffs at the type site at West Runton (the Cromer Forest Bed sensu stricto) only shows the middle of this temperate stage, the latter part being truncated by a marine transgression in this locality. In the so called 'arctic freshwater bed' of north Norfolk the end of the interglacial and the onset of the next cold, and in this case definitely glacial stage, the Anglian, can be seen, with an absence of thermophilous trees, high N.A.P., together with ice wedge casts and involutions.

Until we reach the Cromerian, then, detail is scanty. What conclusions can be drawn from the available evidence? The main conclusion at this stage is that the process of impoverishment of the British and European forest flora, begun in the Tertiary, is continued through the climatic fluctuations represented by the Lower Quaternary succession, with Tertiary relicts, (ie. species of Mid-Tertiary forests no longer native to Europe) still hanging on in the Ludhamian and Antian, particularily Pterocarya and Tsuga. As forest trees, however, Tsuga appears to have been eliminated during, or by the end of the Beestonian (although negative evidence is inconclusive), while Pterocarya only reappears briefly in the Hoxnian Interglacial (HoIII and HoIV) at the site at Marks Tey. Furthermore by the time we reach the Cromerian only five percent of the flora is not now native to the British Isles and of this percentage one species, Corema intermedia is now extinct.

2. The Middle and Upper Quaternary Interglacials

In the upper part of the Middle Quaternary, and in the Upper Quaternary we encounter indisputable glacial deposits in Britain belonging to the Anglian, Wolstonian, and Devensian glaciations, and between them the Hoxnian and Ipswichian Interglacial stages. Deposits of these temperate stages are more widespread, more complete, and better analysed than earlier deposits and we will spend some time looking at them. Both the Hoxnian and Ipswichian type sites lie in eastern England which preserves considerable records of both of these interglacials. The deposits of Hoxnian age are mainly lacustrine deposits occupying hollows in the Lowestoft till of the Anglian glaciation, or lying in channels believed to have been scoured by ice in Anglian times. The deposits of Ipswichian age are on the other hand, mainly associated with later terrace gravels of existing river systems.

2.1. Hoxnian Interglacial

The type site is in a brick pit at Hoxne in Suffolk, but here the upper part of the sequence, after Ho III, has been eroded by ice of the Wolstonian glacial stage, which has sealed in the interglacial material beneath glacial till. At Marks Tey in Essex, however, the complete interglacial cycle is represented by lacustrine deposits occupying a deep trough cut in the subglacial surface during the Anglian glaciation. This trough subsequently formed a lake basin which filled with sediment during the entire interglacial and part of the ensuing Wolstonian glacial stage. It contains laminated diatomaceous muds, and if the laminations are annual, as may well be the case, then the sediments represent the accumulation over a period of 30,000-50,000 years. The trend of vegetation change is similar at both Hoxne and Marks Tey and is summarised on the handout.

First, at the base of the sequence is the Late Anglian, with high N.A.P. frequencies, with the sea buckthorn Hippophae rhamnoides prominent. This passes into the Pre-Temperate Ho I. characterised first by Betula and then Pinus, and then into the Early Temperate mixed oak forest of Ho II. with the thermophilous trees such as lime (Tilia) and other thermophilous elements particularily Hedera (ivy), Ilex (holly), Taxus (yew), and Viscum album (mistletoe). Apart from yew the main understorey shrubs appear to have been Corylus (hazel) and Alnus (alder). The hazel only begins its expansion well into Ho II. and reaches its maximum for example at Marks Tey in Ho III., a point of contrast with the Flandrian, or Post-glacial period, where as we shall see it expands much earlier. Mixed oak forest continues into Ho III. but is marked by the expansion of the hornbeam (Carpinus), spruce (Picea) and later in the zone the fir (Abies). This probably represents not just climatic change, but also soil deterioration for thermophilous elements are still present: sparse thermophilous tree pollen such as that of lime, and evidence of the presence of the box (Buxus semperviriens) and the vine (Vitis) indicating persistent high summer temperatures. The soil deterioration would both explain the presence of, and be accelerated by species such as the spruce, fir, beech (Fagus), and the wingnut (Pterocarya) which occurs at the end of Ho III., and together with the fir continues into HoIV., the Post-Temperate stage. Here the birch (Betula) and pine (Pinus) expand once more and the spruce (Picea) becomes relatively more important than hitherto. The presence of Pterocarya fraxinifolia (now not native nearer than the Caucasus) with its ecological preference for wet and moist conditions, and which may have been replacing alder which declines in Ho IV. may indicate bog formation. Also in Ho IV. there is a slight expansion of the Ericales, but these decline into the Early Wolstonian (eW) where high N.A.P. frequencies are particularily made up of the grasses and open habitat species such as Artemisia, but although heaths as a vegetation type appear to decline, species such as Empetrum do remain as components of the sub-arctic landscape.

Although there are several other Hoxnian sites in eastern England, I'll only consider one more; Clacton-on-Sea. Here, Reid & Chandler found 110 macrofossils in a buried channel deposit interpreted as a relict of a higher interglacial sealevel. These deposits were subjected to pollen analysis by Pike and Godwin in 1953. During a marine transgression on this coast the vegetation shifted from predominently mixed oak forest to one dominated by the fir, but the main interest of the site is the 'Clactonian Culture' of Lower Palaeolithic Acheulian age (S. Hazzledine Warren). The evidence for this culture consists of flint implements and a yew spear that Godwin refers to as the oldest wooden artifact in the country. This indisputable evidence for human activity (Palaeolithic man was also present at the site at Hoxne) is very important in the interpretation of certain peculiarities in Hoxnian pollen diagrams. Both West and Turner note depressions of the tree pollen curves at the end of Ho II. Mixed oak forest with elm (Ulmus) yew and hazel is abruptly destroyed. Wetter sites appear not to have been so affected as the alder curve is not depressed. At the same time there is a corresponding rise in the birch and pine curves. Subsequently oak regenerates, hazel recovers, and finally the yew. The big question posed is: does the behavior of the tree pollen curves indicate forest clearance by man? Alternatively, perhaps it was accidental firing of the forest. Charcoal does indeed occur at Marks Tey at the right horizon. Pearson (1964) suggests that perhaps here we have the earliest evidence of primitive man practicing a 'slash and burn' economy and corresponding with a very pronounced phase of forest clearance comparable with those of the European early Neolithic, and operating on a regional rather than parochial scale.

The only other extensive Hoxnian pollen diagram from England is from a boring in lake deposits at Nechells, Birmingham (Duigan 1956, Kelly 1964). The vegetation development is similar to Marks Tey, the main difference being that hornbeam is barely represented in Ho IIIa., but a great expansion of spruce took place. Nechells has also yielded an important insect fauna.

Four Irish sites, Gort (Co Galway), Kilbeg (Co Waterford), and Baggotstown (Co Limerick), and Kildromin - nb. the Hoxnian is locally known as the Gortian in Ireland - show a sequence of vegetation development that is generally comparable with English sites, but with relatively higher proportions of pine pollen to that of mixed oak forest trees, and particularily Gort and Kilbeg give evidence of highly oceanic conditions during the Late, and Post-Temperate zones. This trend is marked both by the presence of exotics with southern affinities such as Rhododendron ponticum and Erica scoparia as well as members of the existing 'Lusitanian' element in the British flora such as Dabroecia cantabrica, Erica mackiana & Erica ciliaris well beyond their present limits of distribution. In 1969 Birks and Ranson describe an interglacial peat from Fugla Ness, Shetland Islands, very plausibly referred to the late Hoxnian and yielding seeds of Dabroecia cantabrica, Erica mackiana, & Erica scoparia var. macrosperma with a good deal of pine, some spruce and fir and a wealth of other pollen. This assemblege is taken as providing us with an indication of the northern oceanic facies of the Gortian/Hoxnian vegetation pattern. These records taken together with the presence of Pterocarya & Erica terminalis at Marks Tey all strongly indicate that towards the end of the Hoxnian Interglacial conditions became generally much milder and very oceanic - a point of contrast, as we shall see, with the next temperate stage the Ipswichian Interglacial. Furthermore the picture that emerges of the late Hoxnian is of a largely evergreen forest with pine, fir, spruce, yew, rhododendron, and box accompanied by alder and sparse mixed oak forest elements. Now this is not at all unlike the present native montane forest of the Caucasus and it is interesting that Szafer working on late Tertiary deposits at Kroskienko in Poland also compared his Pliocene forest with those of the Caucasus. This raises the possibility that the 'Lusitanian' element of the Hoxnian (and indeed of the present) might conceivably be relicts from an old Pliocene vegetation type, formerly widespread, but now represented by a few species of very disjunct distribution.

2.2. Ipswichian Interglacial

The principal Ipswichian sites all lie in east and southeast England, but none covers the full time span of the interglacial. Taken together, as the diagram shows, the sequences revealed at these ten sites overlap sufficiently to allow a composite sequence, covering the whole of the interglacial to be established. Starting at the base of the sequence at Bobbits Hole, Ipswich, at Selsey, Sussex, and at Ilford, Essex, the Late Wolstonian is represented by a pollen spectrum dominated by N.A.P. making up 80-90% T.P. and indicating an open late glacial tundra landscape, but some birch, willow and juniper was also present. This tree pollen increases into Ip. Ia. which marks the closing of the canopy of the boreal forest of the Pre-Temperate Zone, a forest dominated by birch and pine. At this early stage, however, there are regional differences between sites covering this part of the sequence. At Selsey, for example, the two trees, birch and pine, are represented equally, whereas at Ipswich birch is more important. By Ip. Ib., however, pine has become dominant and oak and elm appear. The early part of Ip. II., ie. Ip. IIa. seems to have been a transitionary period with elm increasing (though it subsequently declines at Bobbits Hole); oak reaches frequencies of 50% by the end of Ip. IIa. and both the hazel and maple (Acer) appear. In the latter part of the Early Temperate zone, Ip. IIb., pine which had remained important through Ip. IIa. declines and both oak and hazel expand their representation. The maple, although it varies from site to site, finds its maximum extension in this zone. Alder, except at the site at Wretton, is not well developed. The Late Temperate zone of the Ipswichian, Ip. III., is, perhaps, the most characteristic, or diagnostic zone of the whole interglacial. It is marked by a rise to what appears to be dominance in the forest of the hornbeam (Carpinus betulus) with frequences of 60-70% A.P. in some sites. At the same time oak and hazel decline in importance, while there is a reciprocal rise in pine pollen frequencies and the forests appear to become more open as the N.A.P. frequencies increase. The hornbeam declines into the Post Temperate, Ip. IV., and deciduous forest gives way to an open boreal forest of pine and birch. The spruce, which characterises this zone on the continent, is absent from British Ipswichian sites, and there is no 'Picea Zone' as in Denmark. So the pine and birch forest gives way in the Early Devensian (eD) to high frequencies of herbaceous pollen types characteristic of periglacial conditions in southern Britain. This early glacial stage is recorded at the top of the Wretton sequence in Norfolk where a meander cut off pool is filled with organic mud.

As with the Hoxnian, several of the species, and their behaviour give valuable clues as to the nature of the interglacial climate. The maple of Ip. II. has been identified from macrofossils as Acer monospessulanum a tree now with a southern and central European distribution, which, although it can now grow in Britain does not set viable seed. Several water plants including the water chestnut Trapa natans occur in British Ipswichian sites well to the north of their present ranges. These facts taken with the dominance of the hornbeam in Ip. III. suggest that the climatic optimum of the Interglacial was perhaps 2 to 3 C warmer than the Flandrian or Postglacial, and probably more continental.

3. The Implications of Variations in Vegetation Response to the Temperate Stages

We have now looked at all of the temperate stages of the Quaternary, with the exception of the last, or present interglacial, the Flandrian (or as the optimists would call it the Postglacial), and it will be useful to consider the differences that exist in the vegetation response between these stages, and to discuss the significance of these differences. Reference should be made at this stage to the relevant handout.

The first diagram summarises these differences for the Pastonian, Cromerian, Hoxnian, Ipswichian, and Flandrian and shows that many of the differences occur between Zones II & III, ie. the Early & Late Temperate zones, although other zones also show some variation. This significance of the Zone II/III boundary is emphasised in the second diagram which shows the behaviour of certain forest genera astride this boundary both in the pre-glacial succession and particularily from the Pastonian onwards. Some are not present in all of the interglacial stages, eg. lime, which is present in the Cromerian for a brief period either side of the zone boundary. It is present again in the Hoxnian for a more extended period, but appears to have been absent in the Ipswichian, but reappears in the Flandrian. Other trees, like both spruce and hemlock occur in the earlier temperate stages, with the spruce being absent from the last two temperate stages, and the hemlock being even more restricted, not occuring in an interglacial site later than the Pastonian. Fir on the other hand is like the lime, little in evidence earlier on, but occuring in the Cromerian and Hoxnian. In the final diagram another difference is encountered in the time of arrival, the time of maximum extension, and the relative importance of the hazel. It arrives earlier, expands progressively more rapidly, and attains relatively greater values, from the Ludhamian through to the Flandrian:- why? Indeed why is it that all of these forest genera differ either in presence or in relative expression between the Interglacial stages?

The interplay of factors that may be responsible for such observed differences between interglacials is probably very complex, but certainly one important factor is probably variations in migration rates displayed by these genera from interglacial to interglacial. These variations in migration rates, and probably also in the migration routes followed, will have combined with the timing of sealevel change, thereby affecting the moment at which the English Channel, Southern North Sea, and Irish Sea became effective barriers to immigration. The behaviour of the lime and the spruce can probably be viewed in these terms, although, and particularily in the case of the lime, variations in the rate of climatic amelioration may also have been important. The difference in relative expression of genera within the temperate stages is probably explained by several factors such as variations in the rates of succession and in soil maturation - affecting the time at which conditions became right for the expansion of particular genera, or variations in competitive interactions - affecting again the time at which conditions, but this time biotic conditions rather than environmental conditions, became right for expansion. It would also appear that the differant interglacials differed in their climates, especially in the nature of the climatic optimum - remember that the evidence suggests that the Hoxnian was both milder and more oceanic than the Ipswichian, which in turn was apparently more continental, and hence experienced an optimum with greater climatic extremes. Finally there may also have been more subtle influences at work, such as evolutionary change in the physiological tolerance of the species concerned throughout this timespan. Joy Deacon suggests that this might well have been the case with the hazel and in her paper in the New Phytologist in 1974 she argues that such changes in tolerance allowed the hazel to survive in refuges progressively closer to the British Isles during the glacial stages. This in turn allowed the hazel to return progressively earlier in the succeeding interglacials until in the Flandrian it arrived well before the trees of mixed oak forest and expanded early, flowered freely, and produced copious pollen in the absence of competition from a closed forest canopy.

4. Vegetation Response in the Glacial Stages

The evidence that we have of the vegetation associated with the earlier glacial stages of the Quaternary is rather limited, and much of it is related to the late and early glacial deposits which have been investigated in conjunction with the interglacial sequences. Tne Beestonian, for example, reveals macroscopic remains of northern plants like the dwarf birch, dwarf willows, saxafrages, mountain sorrel, while in the Anglian we perhaps have some evidence of an interstadial in the Corton Sands between the Cromer Till and Lowestoft Till - the Corton Interstadial. So we will concentrate on the events of the last glacial stage, the Devensian, where numerous and prolific plant deposits have been recovered, often associated with abundant faunal remains, correlated with the geomorphologcal and geological history of the glaciation, and coordinated by the use of radiocarbon dating which fortunately just about extends back to the end of the lpswichian.

Three main sites shed light on the conditions of the Early Devensian, about 70,000 - 50,000 years BP; Sidgewick Avenue, Cambridge; Wretton, Norfolk (which we have already considered when looking at the end of the Ipswichian Interglacial); Chelford, Cheshire. The mixed assemblege of 60 or 70 vascuIar plants recovered from Sidgewick Avenue (70,000 years BP) indicate a variety of habitats in agreement with the site conditions that suggest braided streams shifting over terrace gravels and sands solidified by permafrost. Locally the gravels appear to have been Ieached carrying acidophilc species, while shallow pools on the terrace surface became saline through intense evaporation under the cold dry climate. The main plant elements found include ruderals, northern and montane plants, a southern element, numerous aquatics, and a significant group of halophytes.

Between Congleton and Chelford in Cheshire exposed in a sand pit is a thick suite of illuvial sands (Chelford Sands) containing fossil ice wedges and overlain, and intermittently underlain by glacial till. These sands appear to have been laid down under periglacial conditions and according to Peter Worsley probably originated as a permafrost alluvial fan emanating from the Pennines, with the nearest modern equivalent being those of the Mackenzie Delta in the North West Territories of Canada. In these sands there is a stratum of laminated detrital organic muds which reveal strong dorninance of arboreal pollen, together with peat and macrofossils, needles, cones, and even tree stumps and trunks. The main species are birch, pine and spruce and the forest type indicated is thought to be similar to that growing in northern Finland today under a cool continental climate with a mean annual temperature of c. +2oC. This conclusion is corroborated by the beetle fauna. This, then, is the Chelford Interstadial dated between c. 59,000 and 65,000 years BP and probably correlated with the Børup Interstadial of Denmark.

When we move into the Middle Devensian there is still some controversy as to whether there are any further periods of sufficient climatic amelioration to warrent the term interstadial before the well documented interstadials of the Late Devensian. At several sites in the Midlands and particularity at Upton Warren in Worcestershire Professor Shotton of Birmingham University refers to the Upton Warren Interstadial inspite of the lack of evidence for trees but based on the coleopteran fauna which is taken to indicate warm conditions. Similar evidence is also derived from the third organic lens in the new Devensian type site at Four Ashes in Staffordshire. The absence of trees is sometimes ascribed to a very high grazing intensity by a large herbivorous mammalian fauna for which evidence certainly exits at Upton Warren and Fladbury. with mammoth, woolly rhinoceros, horse, reindeer, and bison. Ecologically, however, this explanation is not convincing and Godwin argues that although the postulated climate of the Middle Devensian might have provided summer temperatures (July 12-16°C) that would have allowed tree growth, other strongly adverse features existed such as very cold winters (January -15°C), severe wind exposure, severe spring thaw and flooding, and. highly unstable soils and likens the environment in many of these sites throughout the Middle Devensian to that of Siberian river floodplains and in part to the great glacial outwash complexes of braided channels such as those of the Canterbury Plains of South Island New Zealand, thereby accounting for the great variety of edaphic and ecological conditions that the variety of floral elements from British sites of this age display. He would, therefore, maintain that the Middle Devensian sites collectively give us a view of the British flora that persisted through the glacial stage beyond the ice front. In the Late Devensian, however, indisputable interstadial conditions occur and provide us with a very detailed picture of such a climatic 'oscillation and the floral and faunal response.