Palaeoenvironment and taphonomy of dinosaur

tracks in the Vectis Formation (Lower

Cretaceous) of the Wessex Sub-basin, southern

England

*Jonathan D. Radley1, *Michael J. Barker

and {Ian C. Harding

* Department of Geology, University of Portsmouth, Burnaby Building, Burnaby Road, Portsmouth,PO1 3QL, UK; 1current address: Postgraduate Research Institute for Sedimentology, The University ofReading, PO Box 227, Whiteknights, Reading RG6 6AB, UK{Department of Geology, University of Southampton, Southampton Oceanography Centre, European Way,Southampton, SO14 3ZH, UK

Revised manuscript accepted 20 June 1997

Reptilian ichnofossils are documented from three levels within the coastal lagoonal Vectis Formation(Wealden Group, Lower Cretaceous) of the Wessex Sub-basin, southern England (coastal exposuresof the Isle of Wight). Footprints attributable to Iguanodon occur in arenaceous, strongly trampled,marginal lagoonal deposits at the base of the formation, indicating relatively intense ornithopodactivity. These were rapidly buried by in¯uxes of terrestrial and lagoonal sediment. Poorly-preservedfootcasts within the upper part of the Barnes High Sandstone Member are tentatively interpreted asundertracks. In the stratigraphically higher Shepherd's Chine Member, footcasts of a small to med-ium-sized theropod and a small ornithopod originally constituted two or more trackways and are pre-served beneath a distinctive, laterally persistent bioclastic limestone bed, characterised by hypichnialDiplocraterion. These suggest relatively low rates of dinosaurian activity on a low salinity, periodicallywetted mud¯at. Trackway preservation in this case is due to storm-induced shoreward water move-ments which generated in¯uxes of distinctive bioclastic lithologies from marginal and offshore lagoo-nal settings. The rapidly-deposited footprint-®lls occasionally contain fully articulated shallowburrowing bivalves. # 1998 Academic Press

KEY WORDS: Vectis Formation; Lower Cretaceous; southern England; dinosaur tracks;palaeoenvironments; taphonomy.

1. Introduction and geological setting

Dinosaur tracks were ®rst recognised in the Lower Cretaceous Wealden strata of

southern England in the mid nineteenth century (Delair, 1989). These generally

occur in alluvial plain facies, both in the Wessex (or `Channel') Sub-basin (Wes-

sex Formation of the Isle of Wight, Barremian; Allen & Wimbledon, 1991) and

the Weald Sub-basin (principally within the Hastings Beds of Sussex, Berriasian-

Valanginian; Allen & Wimbledon, 1991). Most records refer to medium to large

robust three-toed tracks, usually attributed to the pes (hind feet) of Iguanodon, the

common large Wealden ornithopod (Tylor, 1862; Ticehurst, 1928; Delair, 1983).

More recently, theropod footprints and footcasts as well as `four-toed' ankylosaur

or sauropod pes footcasts have been recognised (Woodhams & Hines, 1989;

Parkes, 1993; Radley, 1994a).

The Vectis Formation (Barremian to possibly early Aptian in age: Allen, 1989;

Allen & Wimbledon, 1991) of the Isle of Wight (Figure 1) overlies the Wessex

Formation, recording a phase of warm-temperate or subtropical coastal lagoonal

deposition in the Wessex Sub-basin, prior to the Lower Greensand (Aptian) mar-

Cretaceous Research (1998) 19, 471±487 Article No. cr970107

0195±6671/98/030471 + 17 $30.00/0 # 1998 Academic Press

ine transgression (Allen, 1981, 1989; Stewart, 1981; Stewart et al., 1991). It com-

prises two essentially argillacous units (the Cowleaze Chine and Shepherd's

Chine Members) enclosing the arenaceous Barnes High Sandstone Member

(Figure 2). Sections occur at Sandown Bay on the southeast coast [National Grid

Reference (NGR) SZ 615 852-621 853] and between Barnes High and Ather®eld

Point (SZ 438 805-452 791; Ather®eld section of this account), Shippards Chine

(SZ 376 843-375 845) and at Compton Bay (SZ 372 848-370 850) on the south-

west coast (Figure 1). The Vectis Formation reaches a thickness of nearly 70 m

(White, 1921).

Sedimentological studies (Stewart, 1981; Stewart et al., 1991) indicate that

muddy horizons are of low-energy subaqueous and mud¯at origin whilst the

Barnes High Sandstone and thinner sand bodies represent deltaic in¯uxes into the

lagoon. Thin (generally less than 10 cm), laterally persistent beds of shelly arenac-

eous limestone (`coquinas') occur within the Shepherd's Chine Member

(Figure 2; Table 1) and represent storm deposits. Subsurface data and prove-

nance studies (Stewart et al., 1991; Radley et al., in press) indicate that the faulted

northern margin of the Wessex Sub-basin occurred immediately to the north of

the present-day Sandown and Compton Bay outcrops.

Faunas are dominated by bivalves, gastropods, ostracods, and ®sh remains.

Palaeoecological analyses indicate ¯uctuating oligohaline-low brachyhaline sali-

nities throughout deposition (Stewart, 1981; Stewart et al., 1991). The overall

increase in salinity and storm events recorded upwards through the Shepherd's

Chine Member (Stewart et al., 1991) indicates increasing marine in¯uence prior

to the Aptian transgression.

Research into reptilian ichnofossils has traditionally been of a palaeobiological

nature, providing information on the morphology and behaviour of trackway pro-

ducers. In recent years however, palaeoenvironmental studies have realised the

potential of vertebrate ichnofossils in the understanding of palaeogeographical,

palaeohydrological and sedimentological problems (Lockley, 1986).

Figure 1. Outline map of the Isle of Wight, southern England, showing outcrop of Wealden(Lower Cretaceous) strata (shaded) and localities mentioned in the text.

472 J. D. Radley et al.

Table 1. Characterisation of limestone marker bands in the Shepherd's Chine Member, Vectis For-mation, Ather®eld, Isle of White.

Shelly limestone(bed number) 1 2 3 4

Thickness 1±2 cm, up to 10cm in footcasts

1±2 cm, possiblythickening to 5 cm

at Shepherd's Chine

1±5 cm 2±5 cm

Dominant taxa abundant Filosinaand Viviparus

abundant Filosina abundantPraeexogyra,Procerithium,Paraglauconia

abundant Filosina,rarer Viviparus

Other features strongly pyritic,Diplocraterion sp. on

base, trackwayslocally present on

base, darkcolouration

abundant ®shdebris, redcolouration

locally pinches out,abundant grey-brown muddy

matrix

locally encased in®brous calcite,

shells frequentlypreserved in white

calcite

Figure 2. Generalised stratigraphic column for the Vectis Formation (Wealden Group) at Ather-®eld, Isle of Wight, showing distribution of dinosaur tracks and limestone marker bands (lime-stones not to scale).

Dinosaur tracks in southern England 473

The examples documented and discussed below comprise an in situ footprint

horizon and occurrences of ex situ footcasts from two higher levels. The terminol-

ogy used for reptilian ichnofossils throughout this account is essentially that of

Lockley (1991, appendix B).

2. History of ®nds in the Vectis Formation

The White RockThe basal metre of the Cowleaze Chine Member within the Ather®eld section is a

sandstone unit known as the White Rock (Figure 2). Over the last ®fteen years

this has yielded occasional large iguanodontid pes footcasts, found as loose blocks

on the shore in the vicinity of Cowleaze Chine (SZ 444 801). In situ footprints

have been recently detected (see below and Figure 3a, b).

The Barnes High Sandstone MemberLoose blocks of cross-laminated, bioturbated and shelly sandstone on the shore

between Shepherd's Chine and Cowleaze Chine (SZ 445 798, Figures 1, 2) are

derived from the highest 2 m of the Barnes High Sandstone Member. These have

recently revealed large, poorly-preserved and incomplete iguanodontid pes foot-

casts (see below and Figure 3c) and a small theropod footcast (Dr D. Martill,

pers. comm., 1998).

The Shepherd's Chine MemberMantell (1846, p. 94) recorded ``....markings....supposed to be the imprints of the

feet either of birds or reptiles...'' from shelly limestone slabs, weathered from the

Shepherd's Chine Member in Sandown Bay. Examples of these have not been

traced, either in the ®eld or in available collections. Consequently their authen-

ticity cannot be con®rmed.

In 1984 a schoolboy collector, James Crouch, presented a well-preserved thero-

pod pes footcast to the Museum of Isle of Wight Geology (MIWG 5768;

Figure 3d), preserved on the underside of a slab of shelly limestone. It was discov-

ered on the foreshore near Cowleaze Chine, and has been cited by Ruffell (1988),

Woodhams & Hines (1989), Stewart et al. (1991), brie¯y described by Radley

(1992) and ®gured by Radley (1994a, b).

Since the initial discovery, at least 50 variably preserved theropod pes footcasts

have been further recovered as ex situ limestone slabs, between Cowleaze Chine

and Ather®eld Point, and in the mouth of Shepherd's Chine (SZ 447 798). Many

of these are in private or university collections; however to date (1996), 27 speci-

mens are held in the collections of the Museum of Isle of Wight Geology. This

study is based largely on these specimens.

These are preserved in an identical lithology (see below). Undistorted examples

are all of closely comparable morphology and size to the initial ®nd (Figure 3d, e),

indicating derivation from a single track-bearing layer and generation by one or

more theropods of the same species. In recent years, the majority of recovered

specimens have exhibited considerable wear, suggesting that they were mostly

eroded from their parent bed at approximately the same time as the initial ®nd.

Palaeoecological and sedimentological investigation of Vectis Formation lime-

stones has now successfully located the track-bearing layer (Figure 2), and culmi-

nated in the discovery of a recently eroded footcast-bearing slab (Figure 3e). Insitu material may still exist at this important site, but excavation is presently

474 J. D. Radley et al.

Figure 3. Dinosaur tracks from the Vectis Formation. a, Iguanodon footprint (pes) in loose blockderived from lower part of White Rock, Cowleaze Chine foreshore (NGR SZ 444 801). Note theconglomeratic in®ll, �0.07. b, Iguanodon footprint (pes), middle part of White Rock, CowleazeChine foreshore (NGR SZ 444 801). Print is impressed into pale silt, subsequently ®lled withdarker, carbonaceous sediment, �0.065. c, Iguanodon footcast (pes), upper part of Barnes HighSandstone Member between Cowleaze Chine and Shepherd's Chine (NGR SZ 445 798),adapted from ®eld sketch, �0.05. d, theropod footcast (pes; MIWG 5768), limestone bed 1(Diplocraterion limestone), Shepherd's Chine Member, found loose on Cowleaze Chine foreshore(NGR SZ 444 801), �0.12. e, incomplete theropod footcast (pes; IWCMS: 1995.856), limestonebed 1 (Diplocraterion limestone), Shepherd's Chine Member, recovered from slope to north-westof Shepherd's Chine (NGR SZ 4456 7989), �0.18. f, small ornithopod footcast (pes; M. Greencollection), limestone bed 1 (Diplocraterion limestone), Shepherd's Chine Member, base of cliffbetween Shepherd's Chine and Ather®eld Point (NGR SZ 448 795), �0.26.

Dinosaur tracks in southern England 475

impractical (see below). More recently (1996), Mr M. Green of Ryde, Isle of

Wight, has discovered a small ornithopod footcast, weathered from an outcrop of

the same layer south-east of Shepherd's Chine (SZ 448 795; Figure 3f and

below).

3. Footprints in the White Rock

LithologyThe White Rock rests on oxidised, bioturbated and mudcracked lenticular-

bedded alluvial mudstones and sandstones of the Wessex Formation (the Hypsilo-phodon Bed; Reid & Strahan, 1889). The contact is locally gradational but com-

monly irregular or channelised. In the vicinity of Cowleaze Chine, the lower part

of the White Rock comprises laminated, cross-laminated or burrow-mottled grey

sandstone, intercalated with lenses of fusain-rich carbonaceous sandstone,

organic-rich silty mudstone and poorly-sorted granule-pebble conglomerate. The

mudstones yield poorly preserved (?acid-etched) corbiculid bivalves (Filosina cf.

gregaria Casey) and sporadic ostracods [Theriosynoecum ®ttoni (Mantell)]. The

conglomerates comprise varying proportions of calcrete and fusain clasts, worn

reptilian bone fragments and disarticulated or more rarely articulated unionoid

bivalves, as well as disarticulated Filosina. These variable layers display much sedi-

mentary disturbance (contortion, upturning and pinching of laminae), and grade

up into muddy carbonaceous grey sandstones, penetrated by Beaconites burrows

(Goldring & Pollard, 1995) and rootlet traces. In an unpublished thesis, D. J.

Stewart recognised a similar suite of lithologies constituting the White Rock in

cliff sections running for 700 m to the north-west (see also Stewart, 1981).

Dinosaur tracksIguanodontid footcasts found on the shore at Cowleaze Chine are preserved in

pale laminated sandstone, indicating derivation from adjacent exposures of the

White Rock. One of these is preserved in the collections of the Museum of Isle of

Wight Geology (MIWG 5311).

In the spring of 1996, foreshore exposures immediately west of Cowleaze Chine

exposed hummocky surfaces at several levels in the lower, relatively coarse-

grained and lithologically variable part of the White Rock. These preserve shallow,

broadly rounded depressions, sometimes with surrounding rims of deformed sedi-

ment, and correspond to contortions seen in cross-section in the adjacent cliffs.

Laminated sands and fusain-rich sediments beneath these depressions displayed

compression and some distortion. Several of these structures were recognisable as

large, robust, variably-oriented tridactyl footprints (Figure 3a, b), and match

tracks generally attributed to Iguanodon pes. Examples have been described by

Woodhams & Hines (1989).

Footprint in®lls include both coarse and ®ne-grained lithologies. The former

include coarse pebbly sandstone containing small (up to 5 mm) sub-rounded cal-

crete clasts, fusain fragments and disarticulated bivalve shells (Figure 3a). The

highest observed footprint was in the middle of the bed and was in®lled by grey

carbonaceous silt (Figure 3b).

Track taphonomyTaphonomic studies of ancient and Recent vertebrate tracks and trackways poten-

tially provide important sedimentological data, and enhance our understanding of

476 J. D. Radley et al.

preservational mechanisms (Lockley, 1986; Cohen et al., 1991). Sediment grain-

size and substrate water content are important factors affecting the formation of

prints. Subsequent entry into the sedimentary record depends on relative burial

rates and substrate trampling (Laporte & Behrensmeyer, 1980).

The strongly trampled nature of the lower part of the White Rock (dinoturba-

tion; Dodson et al., 1980) indicates repeated disturbance of marginal lagoonal

substrates and consequential time-averaging of several trackway generations over

an unknown, but probably considerable, time period. The occasional preservation

of sharply-de®ned iguanodontid footprints within these surfaces indicates periodic

rapid burial of freshly-trampled, moist, or slightly damp substrates (Laporte &

Behrensmeyer, 1980), within this essentially destructive regime.

We suggest that coarser, conglomeratic or sandstone ®lls were introduced by

seasonal ¯ood-generated alluvial in¯uxes, and raised rims around some footprints

(see above) may have enhanced their ef®ciency as sediment sinks (Laporte & Beh-

rensmeyer, 1980). The highest preserved footprint (Figure 3b) is in®lled with

®ner sediment, indicating rapid depositional in®ll during lagoonal transgression.

Consequently, track preservation in the White Rock was primarily climatically-

controlled, and enhanced by alluvial in¯ux in the lower part and by lagoonal

transgression in the upper part.

DepositionThe White Rock was interpreted by Stewart (1981) as a low-salinity marginal

lagoonal sand¯at deposit, dissected by narrow ¯uvial channels. Accordingly, the

coarser, cleaner sands in the lower part of the unit at Cowleaze Chine are envi-

saged as low-energy marginal lagoonal sediments, deposited primarily by runoff

from alluvial environments to the west (Stewart et al., 1991), but subject to shore-

line reworking.

Given the strongly seasonal Wealden climate (Allen, 1981), the associated

poorly-sorted lenticular intraformational conglomerates are interpreted as rapidly

deposited, channelised ¯ood sediments, largely comprising admixtures of lithologi-

cal and biogenic ¯oodplain debris. Articulated, closed, unionoid bivalves [includ-

ing Margaritifera (Pseudunio) valdensis (Mantell)] indicate input from alluvial

environments (Tevesz & Carter, 1980; Taylor, 1988). Conversely, Filosina typi®es

subaqueous lagoonal muds in the Vectis Formation and is interpreted as a shallow

infaunal suspension feeder, characteristic of oligohaline-mesohaline salinities (0.5±

18 parts per thousand). The disarticulated nature of these shells in the conglom-

eratic lenses suggests some local input from marginal lagoonal shell pavements.

Fusain-rich layers re¯ect dry-season hinterland wild®re events (Batten, 1974;

Sellwood & Price, 1993) and subsequent wet-season ¯ushing of charcoal-rich

debris into the lagoon. This material may have been periodically deposited on

marginal sand¯ats by seiches. Lenses of ®ner, fossiliferous grey-coloured mud-

stone indicate bodies of standing water on the ¯ats, similarly fed by lagoonal

washover. The ostracod Theriosynoecum ®ttoni was considered by Horne (1995) to

be a probable indicator of freshwater conditions.

Exposures of the lower part of the White Rock in sections 200 m, 300 m and

800 m north-west of Cowleaze Chine reveal ®ner-grained and more uniform ligni-

tic clays containing dispersed, well-preserved Filosina. Additionally, micropalaeon-

tological preparation of sediment from the latter site (below Barnes High; SZ 438

807) has revealed abundant Theriosynoecum ®ttoni, suggesting an onshore-offshore

gradient in this direction.

Dinosaur tracks in southern England 477

At Cowleaze Chine, the ®ner-grained and bioturbated nature of the higher part

of the bed similarly indicates an increase in water depth, linked to lagoonal trans-

gression. This culminated in deposition of the fossiliferous dysaerobic muds of the

Cowleaze Chine Member (Stewart, 1981). Rootlets penetrating the top of the

White Rock indicate temporary plant colonisation during this transgressive phase.

4. Footcasts in the Barnes High Sandstone Member

Lithology and track preservationThe Barnes High Sandstone at Ather®eld (approximately 6 m thick) is a coarsen-

ing-up succession, comprising lenticular and wavy-bedded mudstone/sandstone

units passing up into cross-bedded shelly and bioturbated calcite-cemented and

ferruginous sandstones (Daley & Stewart, 1979). Fallen slabs of sandstone con-

taining abundant disarticulated and fully articulated Filosina sp. and Unio cf. elon-gata Cornuel occur on the shore between Shepherd's Chine and Cowleaze Chine

(Figure 1) and occasionally display distorted or otherwise poorly-preserved robust

tridactyl hypichnial sandstone footcasts (Figure 3c). These are approximately

60 cm long and match Iguanodon pes tracks (Woodhams & Hines, 1989 and

above). Recently, Dr D. Martill (pers. comm., 1998) has additionally recorded a

small theropod footcast. The fossil content indicates that the slabs are derived

from a level within the highest two metres of the member.

DepositionThe Barnes High Sandstone is interpreted as a shallowing-upward river mouth

bar succession and the higher cross-bedded units may have been deposited in a

bar crest setting (Stewart, 1981). The distorted nature of many of the footcasts

suggests that they may be undertracks (Lockley, 1989), thus possibly relating to

dinosaur activity at a higher level within the top of the Barnes High Sandstone or

basal Shepherd's Chine Member.

5. Footcasts in the Shepherd's Chine Member

Description of theropod footcastsMedium-sized footcasts, tridactyl, digitigrade, deepening distally and terminated

by three, slender, clawed digits (ii, iii, iv). Angles between digits are approximately

35�. Mean length is approximately 265 mm, mean width is approximately

245 mm. Overall length of digit iii and distal claw is approximately 190 mm.

Casts of hallux and digit v are absent. Possible phalangeal pad casts are occasion-

ally preserved.

DiscussionThe footcasts are attributed to the pes of a small to medium-sized theropod dino-

saur on account of their gracile `bird's foot' appearance and clear traces of sharp

elongate claws (Thulborn & Wade, 1984; Pittman, 1989; Woodhams & Hines,

1989, Figure 3d). English Wealden theropods are known to include coelurosaurs,

carnosaurians and the enigmatic Baryonyx walkeri (Benton & Spencer, 1995);

however, the traces cannot con®dently be attributed to a known taxon, given the

relatively generalised nature of theropod pedal morphology. Using the morpho-

metric ratios of Thulborn (1989), the biometric data suggest that hip height of

the track producer was approximately 125 cms. Data from other bipedal dinosaur

478 J. D. Radley et al.

trackways and estimates of stride (Alonso, 1989; Thulborn, 1989) suggest that

stride length of the trackway producer may have been in the order of 1.5±1.7 m.

Similar footprints or casts have frequently been attributed to the ichnogenus Gral-lator (Lull, 1904; Irby, 1995).

Description of ornithopod footcastSmall footcast, tridactyl, digitigrade, deepening distally and terminated by three

blunt digits (ii, iii, iv). Angles between digits are approximately 50�. Length is

170 mm, width is 150 mm.

DiscussionThe footcast is attributed to the pes of a small ornithopod on account of the pre-

sence of relatively robust digit casts, which do not bear claws (Thulborn, 1990;

Figure 3f). The biometric data indicate a hip height of approximately 80 cm

(Thulborn, 1989).

PreservationThe theropod and small ornithopod footcasts described above occur as hypichnial

casts on tabular grey limestone slabs. The lithologies of the casts and the overlying

tabular layer differ, although mixing often occurs. Sharply de®ned theropod foot-

casts are common (Figure 3d, e), but range into distorted examples. Examples of

shallow, `ghost' casts have also been found. Maximum footcast depth is preserved

distally, and ranges from approximately 20 mm in undistorted examples to

around 75 mm in distorted specimens. The high-®delity of preservation and

nature of the cast medium (see below) indicate that these are primary casts, rather

than remnants of an undertrack (Lockley, 1989). Descriptions of the preserving

lithology are outlined below.

Footcast in®llThis comprises sand-grade bioclastic sediment with a ®ne-grained sparite cement.

It is packed with gastropod shells (Viviparus infracretacicus Huckriede), scattered

quartz grains, clay and much comminuted phosphatic ®sh debris (Figure 4a).

This lithology is referred to herein as Viviparus coquina (see below and Figure 5).

Low angle lamination is occasionally discernible, and traces of elongate shallow

burrow casts (up to 17 mm) occur sporadically on the footcasts, especially in the

vicinity of the shallow heel region. Fish debris is sometimes concentrated as a

basal lag. The gastropods are small (up to 8 mm long), and are ®lled with sparry

bioclast sand or rarely, lithi®ed mud. Both disarticulated and articulated Filosinagregaria occur, but are relatively rare. Articulated examples are in®lled by bioclas-

tic sand, sparry calcite, or rarely by laminated mudstone. Algal borings sometimes

occur, preserved as pyritic in®lls around shell peripheries. Slabs without footcasts

occasionally preserve the Viviparus coquina as a thin (up to 15 mm) planar veneer.

The slabs sometimes retain patches of pale to dark grey silty mudstone in inter-

stices adjacent to toe casts, as intraclasts, or as lenses within the skeletal accumu-

lation. The shells have been neomorphically replaced by pseudopleochroic calcite,

or pyritised.

Overlying tabular layerThis is referred to herein as the burrow-cast coquina (see below and Figure 5).

The base constitutes a slightly irregular to planar pavement of dominantly con-

Dinosaur tracks in southern England 479

vex-up Filosina shells (up to 25 mm long), interspersed with Viviparus and ®sh

debris in a matrix of bioclastic-quartz sand. Rarely, articulated Filosina occur.

Hypichnial Diplocraterion frequently occur below the basal shell pavement

(Figures 4b, 5), preserved as a lithi®ed admixture of clastic grains and comminu-

Figure 4. Features of theropod trackway horizon (limestone bed 1: Diplocraterion limestone), Shep-herd's Chine Member, Vectis Formation. a, sediment in®lling theropod track, representing the®rst depositional phase (IWCMS: 1995.856, `Viviparus coquina', see text). Note the well-pre-served, relatively uncompacted Viviparus shells, �1.3. b, Diplocraterion preserved on the undersideof the `burrow-cast cquina', which represents the second depositional phase (MIWG 6813),�0.26. c, articulated, closed, spar-®lled Filosina preserved in worn, fragmentary theropod footcast(MIWG 7037, `Viviparus coquina', ®rst depositional phase), �0.25.

Figure 5. Diagrammatic representation of the main biostratinomic divisions of the theropod track-way horizon (limestone bed 1: Diplocraterion limestone, Shepherd's Chine Member, Vectis For-mation).

480 J. D. Radley et al.

ted vertebrate debris. They are randomly-oriented, vertical or subvertical, and up

to about 10 mm deep and 15 mm long. Individual burrows sometimes intersect

or merge. Smaller sinuous burrow casts also occur and are tentatively attributed

to a small variety of the ichnogenus Planolites.The main skeletal accumulation comprises 15±30 mm of disarticulated Filosina

shells in a matrix of Viviparus-rich bioclastic sand and mud with a ®ne sparite

cement. The clay mineral content of the rock comprises illite and kaolinite, with

illite-smectite mixed layer minerals and some smectite (Dr I. M. West, pers.

comm., 1996). Extensive pyritisation has occurred in higher parts of the layer.

Filosina shells occur in convex-up, convex-down and rarely, oblique con®gur-

ations and are sometimes stacked (Kidwell et al., 1986). The top surface is a pave-

ment of disarticulated, dominantly convex-up Filosina, interspersed with

Viviparus.

ProvenanceThe footcast-bearing slabs match the lowest of four shelly limestone beds which

are traceable throughout the Ather®eld section (Figure 2; Table 1). This is the U

burrow-bearing layer noted by Wach & Ruffell (1991). Stewart et al. (1991) ®rst

assigned the burrows to the ichnogenus Diplocraterion (composite shell layer of

their account). The bed occurs approximately 17 m above the base of the Shep-

herd's Chine Member (Figure 2; limestone bed 1). It is intermittently exposed

from the cliff top south-east of Cowleaze Chine (SZ 4444 8000) to a low bank

south-east of Shepherd's Chine (SZ 4491 7953) and in the sides of Shepherd's

Chine (SZ 4470 7979-4485 7985). Diplocraterion is abundant on the base of the

bed throughout these sections (Figure 4b; Table 1), along with occasional straight

to curved groove casts. We propose the name Diplocraterion limestone for this dis-

tinctive unit, which is easily distinguishable from the thicker and ®ner grained

Diplocraterion ironstone higher in the sequence (Daley & Stewart, 1979).

Signi®cantly, subaerially-weathered fragments of the Diplocraterion limestone on

shale slopes around SZ 4456 7989 have yielded a single, fresh, theropod footcast-

bearing slab (Museum of Isle of Wight Geology specimen IWCMS: 1995.856).

Although incomplete (Figure 3e), it matches the beach ®nds in terms of size and

morphology. Consequently this indicates that part of a trackway probably remains

in situ. Small, disturbed sections of the Diplocraterion limestone occur amongst the

extensive slips near the site of the recent ®nd and may well yield track-bearing

material.

Ornithopod footcasts do not appear to have been recorded from the track-bear-

ing layer prior to the recent ®nd (see above and Figure 3f). This indicates the

existence of a largely unrevealed trackway near the place of discovery.

Depositional environmentThe Diplocraterion limestone (limestone bed 1) occurs within ®ne grey blocky

mudstones and paper shales containing thin ostracod shell beds. The ostracods

are mainly Cypridea cf. valdensis (J. de C. Sowerby) and Theriosynoecum ®ttoni(Mantell) which are low salinity (possibly freshwater) taxa (Horne, 1995). Taxa

indicative of higher salinities occur occasionally at this level (Jones, unpublished

thesis; cited by Stewart et al., 1991) implying ¯uctuating, but nevertheless essen-

tially low salinities. Ex situ theropod footcasts, a sample of the Diplocraterion lime-

stone and enclosing shale yield abundant low salinity dino¯agellate cysts (also see

Batten & Lister, 1988; Harding & Allen, 1995). Assemblages are dominated by

Dinosaur tracks in southern England 481

Vesperopsis fragilis (Harding), with rarer occurrences of taxa such as Loboniella hir-suta Batten & Lister, Corculodinium uniconicum Batten & Lister, and Valensiella(ex Cassiculosphaeridia) parvulum (Batten & Lister). Chlorococcalean algae are

represented by rare Botryococcus sp. and Tetraedron paraincus Batten & Lister.

Additionally, palynological preparations from ex situ footcasts yielded abundant

equidimensional brown wood (co-dominant with Vesperopsis fragilis), whilst the insitu limestone sample was dominated by sheet-like plant tissue material. This indi-

cates a degree of variability in palynological signature, but overall a low salinity.

Thin layers of Viviparus infracretacicus occur in the mudstone immediately

above the limestone, and the limestone itself largely comprises Viviparus infracreta-cicus and Filosina gregaria (see above).

Recent viviparid gastropods typify muddy substrates in standing or slow moving

freshwater environments (Gray, 1988). Lower Cretaceous viviparids are generally

taken to indicate freshwater to lower oligohaline salinities (0±5 parts per thou-

sand, FuÈ rsich & Kauffman, 1984; Morter, 1984). The common co-occurrence of

Viviparus and the probably oligohaline-mesohaline Filosina (see above) in parau-

tochthonous assemblages (Kidwell et al., 1986) throughout the Vectis Formation

indicates comparable tolerances, and in all probability a degree of euryhaline tol-

erance within the oligohaline-mesohaline range.

The ®nely laminated nature and dark grey colouration of the underlying mud-

stones indicate deposition under at least episodically dysaerobic conditions. The

presence of thin ostracod shell beds implies periodic wave-induced subaqueous

winnowing. Shallow (a few mm deep) ostracod-®lled vertical polygonal cracks

were noted 30 cm below the coquina layer in Shepherd's Chine (SZ 4485 7985),

implying periodic emergence and desiccation of muddy substrates. The high ®de-

lity preservation of the footcasts indicates that at the time of trackway formation,

the underlying mud was moist and moderately to ®rmly cohesive (Laporte & Beh-

rensmeyer, 1980; Lockley, 1991) and therefore submerged to a shallow depth or

temporarily emergent. Shallow, poorly-preserved examples suggest variation in

sediment water content across the mud¯at (Lockley, 1986), as does a partial cast

of a probable polygonal desiccation crack on the underside of one limestone slab.

Biostratinomy of Viviparus coquinaAs a subdiscipline of taphonomy, biostratinomic studies chart the post-mortem, pre-

burial histories of skeletal particles (Seilacher, 1973) and potentially provide import-

ant sedimentological data. The high grade preservation of the footcasts indicates a

minimal time lapse between trackway emplacement and deposition of the footprint

®ll. The clean-washed bioclastic in®ll within most gastropods from the footprint ®lls

suggests derivation from an agitated shoreface or strandline setting.

The sharp-based nature of the ®ll and the occasional preservation of low-angle

lamination implies subaqueous deposition as a migrating sheet of skeletal sand.

The occurrence of a silty lens (see above) indicates that this ¯ood event com-

prised at least two depositional pulses with brief intervening quiescence. High res-

olution footcast preservation also implies rapid deposition of the ®ll, with only

localised erosion of the underlying substrate (Laporte & Behrensmeyer, 1980).

Supporting evidence for rapid deposition comes from the presence of articu-

lated bivalves within the skeletal concentration. Within the Vectis Formation, Filo-sina occurs commonly in background mud facies as disarticulated shells in

parautochthonous winnowed pavements. Closed, articulated examples are rela-

tively rare at most levels, and con®rm the susceptibility of shallow infauna to

482 J. D. Radley et al.

reworking (Tanabe & Arimura, 1987; Bromley, 1990). Local concentrations of

articulated Filosina in footprint in®lls (Figure 4c) therefore indicate exhumation of

live bivalves from subaqueous muds and onshore entrainment within the gastro-

pod-rich sediment. Rapid sedimentation of this material resulted in near-instan-

taneous burial in an unfamiliar sedimentary medium. Some of the bivalves occur

in the inferred normal life position, with the commissural plane oriented vertically

or subvertically, and posterior margin directed upwards (Stanley, 1970; FuÈrsich,

1980) whereas others are disoriented. All orientations indicate failed escape

responses, due either to siphonal clogging or toxic effects (Kranz, 1974; Brett &

Seilacher, 1991). Sparry calcite sometimes partially or completely ®lls the remain-

ing voids between valves (Figure 4c), indicating aerobic decay of soft tissues

(Brett & Baird, 1986; Lockley, 1986). In®lls of mud and geopetal bioclast sand in

some examples suggest post-depositional sedimentary in®ltration of perforated or

slightly gaping bivalves.

Biostratinomic features of the overlying burrow-cast coquina (see below) imply

that the ¯ood event described above was due to a barometrically and wind

induced shoreward water swell, during the onset of an onshore-directed storm

(Aigner, 1985; Brett & Seilacher, 1991). Occasional groove casts indicate tool

marks created by onshore movement of small objects (probably biogenic debris)

across the mud¯ats during this phase.

Biostratinomy of burrow-cast coquinaThe basal layer of convex-up Filosina (Figure 5) indicates renewed bioclastic

input, from an offshore subaqueous Filosina-rich substrate, followed by wave or

current reworking (Futterer, 1978; Kidwell & Bosence, 1991). This material

accumulated as a uniform sheet, well beyond the vicinity of the trackways. The

presence of obruted Filosina unsuccessfully escaping from the underlying Vivi-parus coquina into this higher layer indicates at best only a short intermission

between the two events.

At least some of the gastropod shells present in the higher layer (Figure 5) must

be reworked products of pre-existing skeletal accumulations, deposited on mud-

¯ats around the trackways during the earlier phase. The abundance of disarticu-

lated and variably-oriented, sometimes stacked Filosina shells in the main part of

the burrow-cast coquina indicates continued bioclast input from Filosina pave-

ments and rapid subaqueous storm deposition of this material (Kidwell &

Bosence, 1991; Banerjee & Kidwell, 1991). The resulting storm sheet sealed the

deeper footprint in®lls and armoured the surrounding mud¯ats, but locally

reworked Viviparus shells from shallower prints.

The Diplocraterion burrows are interpreted as the dwelling structures of soft-

bodied suspension or deposit feeders. They are absent from the deeper, distal

parts of the footcasts, but may be sporadically developed on the shallow heel

regions and attain normal background abundance between digits. This suggests

that the burrowers colonised the freshly ¯ooded mud¯ats in the brief interval

between the in¯uxes of Viviparus and burrow-cast coquinas. Such widespread

colonisation must have been rapid and hindered only by local accumulations of

poorly-penetrable Viviparus sand such as that ®lling the ¯ooded theropod tracks.

Bioturbation was curtailed by the ensuing in¯ux of Filosina-rich sediment com-

prising the burrow-cast coquina. The apparent absence of endichnial Diplocrater-ion within the burrow-cast coquina suggests that the burrowers did not survive

this latter depositional event.

Dinosaur tracks in southern England 483

The shell pavement capping the deposit re¯ects decreased sediment supply and

winnowing of the recently deposited coquina as the storm abated (Seilacher &

Aigner, 1991). Thin, parautochthonous Viviparus and ostracod shell beds in

shales capping the coquina re¯ect recolonisation of submerged substrates by

opportunistic feeders and periodic winnowing during and after terminal mud fall-

out. Pyrite within the burrow-cast coquina re¯ects early diagenetic reduction of

rapidly buried organic material.

6. Discussion and conclusions

These occurrences of vertebrate ichnofossils provide further evidence for optimum

preservational conditions in wetted areas, marginal to long-standing water bodies

(Laporte & Behrensmeyer, 1980; Scrivner & Bottjer, 1986; Cohen et al., 1991).

Lithologically variable sediments comprising the lower part of the White Rock

were deposited in a marginal lagoonal environment by an interplay of climatically-

controlled terrestrial runoff and occasional lagoonal in¯uxes. Iguanodon footprints

and extensive dinosaur-induced trampling indicate at least periodic subaerial

exposure of damp substrates. Prints are in®lled by both conglomeratic and ®ner-

grained ®lls, and high resolution preservation indicates that they escaped erosion

or further trampling, by rapid burial. Iguanodon and theropod footcasts on fallen

blocks of the Barnes High Sandstone may possibly be transmitted undertracks.

Molluscs have inhabited non-marine environments since the mid-Palaeozoic

period (Gray, 1988) but it was only in post-Triassic times that they became

important sedimentary constituents in low salinity regimes and potentially signi®-

cant contributors to trackway preservation. Theropod and ornithopod trackways

in the Shepherd's Chine Member owe their preservation to molluscan pro-

ductivity within the Vectis lagoon, and a combination of background and event

biostratinomic processes. In contrast to the underlying alluvial facies (Wessex For-

mation), dinosaur remains are rare in the Vectis Formation, and generally

restricted to partial Iguanodon skeletons (Hooley, 1925; Clarke & Barker, 1993;

Benton & Spencer, 1995). Consequently the chance preservation of these track-

ways provides unique evidence of potentially widespread theropod and ornitho-

pod activity on low salinity marginal lagoonal mud¯ats. Live reptilian prey,

stranded or drowned vertebrate carcasses or live aquatic prey may have attracted

theropods; additionally, open shoreline corridors may have been advantageous in

providing unobstructed all-round views.

The preservation of footcasts in the Shepherd's Chine Member re¯ects the

dynamics of a powerful storm. Initial surges inundated a marginal lagoonal mud-

¯at and had the dual effect of introducing Viviparus-rich shell sand from a rela-

tively high-energy area and exhuming shallow-burrowing bivalves (Filosina) from

®ner substrates. The fresh trackways acted as a series of taphonomic `traps' for

this material (Laporte & Behrensmeyer, 1980) entombing the bivalves.

The small size and relatively compact, unornamented morphology of the gastro-

pod shells (Figure 4a) enhanced the high resolution footcast preservation. These

and associated burrow casts were subsequently armoured by a sheet of Filosina-

rich sediment (the burrow-cast coquina), which also effectively sealed the fate of

rapidly-buried Filosina and opportunistic soft-bodied burrowers. Mud fallout

resumed in standing water as storm conditions waned, burying and preserving the

shelly accumulation.

484 J. D. Radley et al.

Preservation of primary trackways in the White Rock and Shepherd's Chine

Member was due to rapid burial soon after formation. The co-occurrence of well-

preserved Iguanodon footprints and indeterminate trampling features in sand¯at

facies at the base of the formation (White Rock) indicates relatively high rates of

trampling (Figure 6) in an area supporting large dinosaur populations. By con-

trast, the higher, spatially separated occurrences of theropod and ornithopod

trackways in storm-in¯uenced mud¯at facies (Shepherd's Chine Member) indi-

cates a low productivity environment, occasionally visited by opportunistic preda-

tors or scavengers (Figure 6). To our knowledge these are the youngest

occurrences of dinosaur ichnofossils in the Cretaceous deposits of Great Britain.

Acknowledgements

We thank Roland Goldring (University of Reading), for constructive comments

on an early draft of this paper. Ian West (University of Southampton) provided

details of clay mineralogy and Owen Sutcliffe (University of Bristol) provided use-

ful advice on invertebrate ichnology. JDR thanks Steve Hutt (Museum of Isle of

Wight Geology) and Mick Green (Ryde, Isle of Wight) for assistance in the ®eld

and allowing access to collections in their care. Dave Martill (University of Ports-

mouth) provided unpublished data concerning a theropod track.

References

Aigner, T. 1985. Storm depositional systems, 174 pp. (Springer-Verlag, Berlin).Allen, P. 1981. Pursuit of Wealden models. Journal of the Geological Society, London 138, 375±405.Allen, P. 1989. Wealden research-ways ahead. Proceedings of the Geologists' Association 100, 529±564.Allen, P. & Wimbledon, W. A. 1991. Correlation of NW European Purbeck-Wealden (nonmarine

Lower Cretaceous) as seen from the English type-areas. Cretaceous Research 12, 511±526.Alonso, R. N. 1989. Late Cretaceous dinosaur trackways in northern Argentina. In Dinosaur tracks and

traces (eds Gillette, D. D. & Lockley, M. G.), pp. 223±228 (Cambridge University Press, Cam-bridge).

Figure 6. Model relating track preservation in the White Rock and Shepherd's Chine Member torates of trampling and deposition.

Dinosaur tracks in southern England 485

Banerjee, I. & Kidwell, S. M. 1991. Signi®cance of molluscan shell beds in sequence stratigraphy; anexample from the Lower Cretaceous Mannville Group of Canada. Sedimentology 38, 913±934.

Batten, D. J. 1974. Wealden palaeoecology from the distribution of plant fossils. Proceedings of the Geol-ogists' Association 85, 433±457.

Batten, D. J. & Lister, J. K. 1988. Early Cretaceous dino¯agellate cysts and chlorococcalean algaefrom freshwater and low salinity palynofacies in the English Wealden. Cretaceous Research 9, 337±367.

Benton, M. J. & Spencer, P. S. 1995. Fossil reptiles of Great Britain, 345 pp. (Chapman & Hall,London).

Brett, C. E. & Baird, G. C. 1986. Comparative taphonomy; a key to paleoenvironmental interpretationbased on fossil preservation. Palaios 1, 207±227.

Brett, C. E. & Seilacher, A. 1991. Fossil LagerstaÈtten; a taphonomic consequence of event sedimen-tation. In Cycles and events in stratigraphy (eds Einsele, G., Ricken, W. & Seilacher, A.), pp. 283±297 (Springer-Verlag, Berlin).

Bromley, R. G. 1990. Trace fossils: biology and taphonomy, 280 pp. (Unwin Hyman, London).Clarke, J. B. & Barker, M. J. 1993. Diagenesis in Iguanodon bones from the Wealden Group, Isle of

Wight, southern England. Kaupia-DarmstaÈdter BeitraÈge zur Naturegeschichte 2, 57±65.Cohen, A., Lockley, M., Halfpenny, J. & Michel, A. E. 1991. Modern vertebrate track taphonomy at

Lake Manyara, Tanzania. Palaios 6, 371±389.Daley, B. & Stewart, D. J. 1979. Week-end ®eld meeting: the Wealden Group in the Isle of Wight.

Proceedings of the Geologists' Association 90, 51±54.Delair, J. B. 1983. Cretaceous dinosaur footprints from the Isle of Wight: a brief history. Proceedings of

the Isle of Wight Natural History and Archaeological Society 7, 609±615.Delair, J. B. 1989. A history of dinosaur footprint discoveries in the British Wealden. In Dinosaur tracks

and traces (eds Gillette, D. D. & Lockley, M. G.), pp. 19±25 (Cambridge University Press, Cam-bridge).

Dodson, P., Behrensmeyer, A. K., Bakker, R. T. & McIntosh, J. S. 1980. Taphonomy and paleoecol-ogy of the dinosaur beds of the Jurassic Morrison Formation. Paleobiology 6, 208±232.

FuÈrsich, F. T. 1980. Preserved life positions of some Jurassic bivalves. PalaÈontologische Zeitschrift 54,289±300.

FuÈrsich, F. T. & Kauffman, E. G. 1984. Palaeoecology of marginal marine sedimentary cycles in theAlbian Bear River Formation of south-western Wyoming. Palaeontology 27, 501±536.

Futterer, E. 1978. Hydrodynamic behaviour of biogenic particles. Neues Jahrbuch fuÈr Geologie undPalaÈontologie, Abhandlungen 157, 37±42.

Goldring, R. & Pollard, J. E. 1995. A re-evaluation of Ophiomorpha burrows in the Wealden Group(Lower Cretaceous) of southern England. Cretaceous Research 16, 665±680.

Gray, J. 1988. Evolution of the freshwater ecosystem; the fossil record. Palaeogeography, Palaeoclimatol-ogy, Palaeoecology 62, 1±214.

Harding, I. C. & Allen, R. M. 1995. Dinocysts and palaeoenvironmental interpretation of non-marinesediments: an example from the Wealden of the Isle of Wight (Lower Cretaceous, southern Eng-land). Cretaceous Research 16, 727±743.

Hooley, R. W. 1925. On the skeleton of Iguanodon ather®eldensis sp. nov., from the Wealden Shales ofAther®eld (Isle of Wight). Quarterly Journal of the Geological Society of London 81, 1±61.

Horne, D. J. 1995. A revised ostracod biostratigraphy for the Purbeck-Wealden of England. CretaceousResearch 16, 639±663.

Irby, G. V. 1995. Posterolateral markings on dinosaur tracks, Cameron Dinosaur Tracksite, Lower Jur-assic Moenave Formation, northeastern Arizona. Journal of Paleontology 69, 779±784.

Kidwell, S. M. & Bosence, D. W. J. 1991. Taphonomy and time-averaging of marine shelly faunas. InTaphonomy: releasing the data locked in the fossil record (eds Allison, P. A. & Briggs, D. E. G. ), pp.115±209 (Plenum Press, New York).

Kidwell, S. M., FuÈ rsich, F. T. & Aigner, T. 1986. Conceptual framework for the analysis and classi®-cation of fossil concentrations. Palaios 1, 228±238.

Kranz, P. M. 1974. The anastrophic burial of bivalves and its palaeoecological signi®cance. Journal ofGeology 82, 237±265.

Laporte, L. F. & Behrensmeyer, A. K. 1980. Tracks and substrate reworking by terrestrial vertebratesin Quaternary sediments of Kenya. Journal of Sedimentary Petrology 50, 1337±1346.

Lockley, M. G. 1986. The paleobiological and paleoenvironmental importance of dinsaur footprints.Palaios 1, 37±47.

Lockley, M. G. 1989. Summary and prospectus. In Dinosaur tracks and traces (eds Gillette, D. D. &Lockley, M. G.), pp. 441±447 (Cambridge University Press, Cambridge).

Lockley, M. G. 1991. Tracking dinosaurs: a new look at an ancient world, 238 pp. (Cambridge UniversityPress, Cambridge).

Lull, R. S. 1904. Fossil footprints of the Jura-Trias of North America. Memoirs of the Boston Society forNatural History 5, 461±562.

486 J. D. Radley et al.

Mantell, G. A. 1846. Notes on the Wealden strata of the Isle of Wight, with an account of the bonesof Iguanodons and other reptiles discovered at Brook Point and Sandown Bay. Quarterly Journal ofthe Geological Society of London 2, 91±95.

Morter, A. A. 1984. Purbeck-Wealden beds Mollusca and their relationship to ostracod biostratigra-phy, stratigraphical correlation and palaeoecology in the Weald and adjacent areas. Proceedings ofthe Geologists' Association 95, 217±234.

Parkes, A. S. 1993. Dinosaur footprints in the Wealden at Fairlight, East Sussex. Proceedings of theGeologists' Association 104, 15±21.

Pittman, J. G. 1989. Stratigraphy, lithology, depositional environment, and track type of dinosaurtrack-bearing beds of the Gulf Coastal Plain. In Dinosaur tracks and traces (eds Gillette, D. D. &Lockley, M. G.), pp. 135±153 (Cambridge University Press, Cambridge).

Radley, J. D. 1992. Geological records for 1991±1992. Proceedings of the Isle of Wight Natural Historyand Archaeological Society 12, 133±137.

Radley, J. D. 1994a. Field Meeting, 24±5 April, 1993: the Lower Cretaceous of the Isle of Wight. Pro-ceedings of the Geologists' Association, 105, 145±152.

Radley, J. D. 1994b. Tracking a carnivorous dinosaur on the Isle of Wight. Earth Heritage 2, 15.Radley, J. D., Gale, A. S. & Barker, M. J. (in press). Derived Jurassic fossils from the Vectis Formation

(Lower Cretaceous) of the Isle of Wight, southern England. Proceedings of the Geologists' Association109.

Reid, C. & Strahan, A. 1889. The geology of the Isle of Wight, second edition. Memoir of the GeologicalSurvey of England and Wales, 349 pp. (HMSO, London).

Ruffell, A. H. 1988. Palaeoecology and event stratigraphy of the Wealden-Lower Greensand transitionin the Isle of Wight. Proceedings of the Geologists' Association 99, 133±140.

Scrivner, P. J. & Bottjer, D. J. 1986. Neogene avian and mammalian tracks from Death ValleyNational Monument, California: their context, classi®cation and preservation. Palaeogeography,Palaeoclimatology, Palaeoecology 57, 285±331.

Seilacher, A. 1973. Biostratinomy: the sedimentology of biologically standardized particles. In Evolvingconcepts in sedimentology (ed. Ginsburg, R. N.), pp. 159±177 (John Hopkins University Press, Balti-more).

Seilacher, A. & Aigner, T. 1991. Storm deposition at the bed, facies, and basin scale: the geologicalperspective. In Cycles and events in stratigraphy (eds Einsele, G., Ricken, W. & Seilacher, A.), pp.249±267 (Springer-Verlag, Berlin).

Sellwood, B. W. & Price, G. D. 1993. Sedimentary facies as indicators of Mesozoic palaeoclimate. Phi-losophical Transactions of the Royal Society of London, Series B 341, 225±233.

Stanley, S. M. 1970. Relation of shell form to life habits of the Bivalvia (Mollusca). Memoir of the Geo-logical Society of America 125, 296 pp. (Geological Society of America, Boulder).

Stewart, D. J. 1981. A ®eld guide to the Wealden Group of the Hastings area and the Isle of Wight. InField guides to modern and ancient ¯uvial systems in Britain and Spain (ed. Elliott, T.), Proceedingsof the Second Conference on Fluvial SedimentsÐModern and ancient ¯uvial systems: sedimentol-ogy and processes, pp. 3.1±3.32 (University of Keele).

Stewart, D. J., Ruffell, A., Wach, G. & Goldring, R. 1991. Lagoonal sedimentation and ¯uctuatingsalinities in the Vectis Formation (Wealden Group, Lower Cretaceous) of the Isle of Wight,southern England. Sedimentary Geology 72, 117±134.

Tanabe, K. & Arimura, E. 1987. Ecology of four infaunal bivalve species in the Recent intertidal zone,Shikoku, Japan. Palaeogeography, Palaeoclimatology, Palaeoecology 60, 219±230.

Taylor, D. W. 1988. Aspects of freshwater mollusc ecological biogeography. Palaeogeography, Palaeocli-matology, Palaeoecology 62, 511±576.

Tevesz, M. J. S. & Carter, J. G. 1980. Environmental relationships of shell form and structure ofunionacean bivalves. In Skeletal growth of aquatic organisms (eds Rhoads, D. C. & Lutz, R. A.),pp. 295±322 (Plenum Press, New York).

Thulborn, R. A. 1989. The gaits of dinosaurs. In Dinosaur tracks and traces (eds Gillette, D. D. &Lockley, M. G.), pp. 39±50 (Cambridge University Press, Cambridge).

Thulborn, R. A. 1990. Dinosaur tracks, 410 pp. (Chapman & Hall, London).Thulborn, R. A. & Wade, M. 1984. Dinosaur trackways in the Winton Formation (mid-Cretaceous) of

Queensland. Memoirs of the Queensland Museum 21, 413±517.Ticehurst, N. F. 1928. Iguanodon footprints at Bulverhythe. The Hastings Naturalist 4, 15±19.Tylor, A. 1862. On the footprint of an Iguanodon lately found at Hastings. Quarterly Journal of the Geo-

logical Society of London 18, 247±253.Wach, G. & Ruffell, A. H. 1991. Sedimentology and sequence stratigraphy of a Lower Cretaceous tide and

storm-dominated clastic succession, Isle of Wight and S. E. England, 100 pp. (13th Congress, Inter-national Association of Sedimentologists, Nottingham, England).

White, H. J. O. 1921. A short account of the geology of the Isle of Wight. Memoir of the Geological Sur-vey of the United Kingdom, 201 pp. (HMSO, London).

Woodhams, K. E. & Hines, J. S. 1989. Dinosaur footprints from the Lower Cretaceous of East Sussex,England. In Dinosaur tracks and traces (eds Gillette, D. D. & Lockley, M. G.), pp. 301±307 (Cam-bridge University Press, Cambridge).

Dinosaur tracks in southern England 487