JOURNAL OF QUATERNARY SCIENCE (2008) 23(1) 85–92 Copyright # 2007 John Wiley & Sons, Ltd. Published online 9 August 2007 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jqs.1118 Correspondence Comment: Middle Pleistocene sedimentation at Pakefield, Suffolk, England y J. R. LEE, J. ROSE, I. CANDY and R. W. BARENDREGT (2006) Journal of Quaternary Science 21: 155–179 Introduction In their recent article, Lee et al. (2006) present a reappraisal of the succession exposed in the cliff sections at Pakefield, Suffolk. They differentiate a series of 11, apparently discrete, lithostratigraphical units (denoted: Pa A–K) which they inter- pret as representing a series of sea-level variations, and fluvial sedimentation, soil development and glaciation episodes dur- ing the early and later Middle Pleistocene. They equate the events they recognise to the marine isotope stratigraphy and suggest that the depositional episodes span Marine Isotope Stages (MIS) 20–12. The succession at this site is significant since interglacial deposits at the base of the cliffs include the earliest evidence of human occupation yet identified in north- west Europe (Parfitt et al., 2005). These deposits (Pa A–C), included within the Cromer Forest-bed Formation, are not the subject of this response, since they underlie a marked sequence-boundary unconformity the surface beneath which is clearly identifiably modified by pedogenesis and possibly periglacial ground-ice structural modification (Fig. 1). This note concerns units Pa D–I which, according to Lee et al. (2006: table 6), represent respectively a complex succession of mar- ine, fluvial, glaciofluvial and marine sedimentation that have been claimed to represent not only changing global eustatic sea-level but also changing climate and indeed age. Detailed examination of the sections at Pakefield has been undertaken by several generations of workers, originally by Blake (1890) who was the first to identify a twofold division of the cliff succession into preglacial fluvial and marine depos- its and an overlying glacial succession. Before publication of the Lee et al. (2006) article, all subsequent workers substan- tiated this scheme, assigning the interglacial and associated sediments at the cliff base to the Cromer Forest-bed Formation (West, 1980; Stuart and Lister, 2000, 2001), whilst the overly- ing sands and till were correlated to the Anglian-age Corton Sands Member (North Sea Drift Formation) of Scandinavian and North-Sea floor origin, and the Lowestoft Formation till (Mitchell et al., 1973; Perrin et al., 1979; Pointon, 1978; West, 1980; Hopson and Bridge, 1987; Ehlers and Gibbard, 1991; Stuart and Lister, 2000; Lewis, 1999) deposited by British ice, respectively. On the basis of our observations, we believe that the Lee et al. (2006) reinterpretations of the Pakefield sequence, which form part of a larger programme of re-evalua- tion of the critical East Anglian glacial and preglacial sequence of coastal Norfolk and northern Suffolk, are questionable and open to interpretations that differ substantially from those pro- posed by them. The evidence considered here includes the lithology and sedimentary sequence, palaeontology, chronol- ogy and regional stratigraphical implications. Depositional succession and environment The units Pa D–J are described by Lee et al. (2006: fig. 4) based on sedimentary logs of nine profiles. Contrary to their descrip- tions, we observe three broad-scale subdivisions within the Corton Member at this locality (Fig. 1). The lowest unit (Pa D, E and G) comprises a suite of sands and pebbly sands that rest unconformably on the Cromer Forest-bed deposits beneath (Lee et al.’s units Pa A–C). In places, a gravel bed with indistinct cross-bedded lamination occurs at the base but grades upwards into trough to tabular cross-bedded sands. This basal unit is disposed into intersecting channel-fill sequences of trough–cross-bedded pebbly sands rich in chalk clasts. These basal units show predominant palaeocurrents towards the east to southeast contrary to those noted by Lee et al. (2006). This lowest unit is not present throughout the sections but is localised at sites such as 3, 8 and 7, or represented by only the basal gravel lag horizon. Elsewhere the lowest unit is conformably overlain by a sub- stantial unit of sands that forms the main body of the profile throughout and locally reaches thicknesses of  10 m. These sands (units Pa D, H and I of Lee et al. (2006)) comprise at the base a thick section of metre-scale beds formed by planar and trough–cross-bedded medium sands (dunes) mostly char- acterised by tangential foresets. This section is overlain by laterally continuous decimetre-scale tabular beds charac- terised by small-scale trough–cross bedding including climbing ripples. A distinct alternating pattern of planar and ripple-drift bedding structures is found throughout. Loading, dewatering and slump structures also occur in discrete beds in the upper interval. Again, palaeocurrent directions indicate a predomi- nant flow towards the south and east. This upper interval also comprises, at the very top, narrow interbeds including horizontally bedded sand and silt passing upwards locally into thinly interbedded sand and silt drapes. In turn, this grades upwards into finely laminated silt and clay rhythmites, the individual laminae being only a few millimetres thick in the uppermost part of the sequence. This uppermost sequence, not described by Lee et al. (although unit Pa I includes elements of it) again rests conformably on that beneath with no apparent hiatus intervening, overall displaying a general thinning and fining-upward trend for the entire Corton Member succession. y This article was published online 9 August 2007. An error was subsequently identified and corrected by an erratum notice that was published online on 6 December 2007; DOI 10.1002/jqs. 1160. This printed version incorporates the amendment identified by the erratum notice.