Loope 1986

Recognizing and Utilizing Vertebrate Tracks in Cross Section: Cenozoic Hoofprints from Nebraska Author(s): David B. Loope Source: PALAIOS, Vol. 1, No. 2 (Apr., 1986), pp. 141-151 Published by: SEPM Society for Sedimentary Geology Stable URL: http://www.jstor.org/stable/3514507 Accessed: 19/10/2008 15:14 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=sepm. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].

SEPM Society for Sedimentary Geology is collaborating with JSTOR to digitize, preserve and extend access to PALAIOS.

http://www.jstor.org

141 141

REPORTS RESEARCH REPORTS RESEARCH

Recognizing and

Cross Cenozoic

Vertebrate

Utilizing

Tracks

in

Section:

Hoofprints from

Nebraska

DAVIDB. LOOPE Departmentof Geology,Universityof Nebraska,Lincoln,NE 68588-0340

PALAIOS,1986, V. 1, p. 141-151 Because vertebratetracksexposedin verticaloutcropsare freas physicallyinduceddeforquentlyoverlookedor misinterpreted mation structures,importantpaleoecologicand sedimentologic informationremainsuntapped.Laminationswithineoliandune sands of the NebraskaSand Hills (Holocene)and ephemeralstreamdepositsof the lowerArikareeGroup(late Oligocene)are commonlybrokenor sharplydownwarpedto form isolated or paired, concave-upstructuresthat varyfrom 4 to 22 cm in diameter.A central ridge divides the lower portion of some structuresinto two distinctlobes.Althoughbedding-plane exposures are rare in bothdeposits,extensivesearch revealedsome structuresin linear alignment,confirminga biogenicoriginfor the deformation.TheHolocenetracks,probablymade by bison, wererepeatedly producedduringthemigrationof thelargeeolian bedforms,suggesting that food and water were available in interduneareas. Smectitegrain coatings made surface sands and encohesive,therebystronglyinfluencingtrackmorphology hancingpreservationpotential.Oligocenetrackswereproduced in veryfine sand by severalspeciesof hoofedvertebrates.Close bedsrevealsthatmostsediment verticalspacingof track-bearing accumulatedin relativelythinpackages.Thegeneralabsenceof heavilytrampledhorizons(as wouldbe expectedalong diastems) may be the result of rapid consolidationof sands by evaporite cementation.Tracksmayhavebeenproducedduringbriefintervals of time immediatelyfollowing depositionand preceding cementation.

INTRODUCTION Vertebratetrackwaysare a well-knownsource of paleontologic andsedimentologicinformation(Sarjeant,1975). Perhaps becausevertebratetracksare so easy to recognizeon beddingplaneexposures, untilrecentlylittleattentionhas been givento theirappearancein verticalsection. Photographsanddrawings publishedby Vander LingenandAndrews(1969), McKee and Bigarella(1972), Lewis and Titheridge(1978), Laury(1980), and Hunter et al. (1984) have, however, documented the Copyright? 1986, The Society of EconomicPaleontologistsand Mineralogists

deformationthat takes place when large animalsmove across soft, laminatedsediment. On the basis of their observationsof Quaternarydepositsin East Africa,LaporteandBehrensmeyer (1980) have recently argued that large vertebrates have the potentialto rework terrestrialsediments to the same extent that benthicinvertebratesreworkmarinestrata. Preservation of tracksrequirescompactiblesubstratesthat are accessibleto vertebrates; rates of tramplingand burial control whether sediments record individualtracks or are totally bioturbated (Laporteand Behrensmeyer, 1980, fig. 4a). The Holocenetracks describedhere are from exposures of duneandinterdunedepositswithinthe NebraskaSandHillsand were probablyproducedby bison. Oligocene hoofprintsare well exposed in outcropsof fluvialsheet-flooddeposits within the GeringFormation(ArikareeGroup)at Scotts BluffNational Monumentin westernmostNebraska;these trackswere made by several differentspecies of cloven-hoofedmammals.The morphologyof both Holocene and Oligocene hoofprintsindicates that they were produced in cohesive sand. Much of surfacesand in the SandHillsis cohesive even when dry, due to clay coatingson grains;such coatingsgreatlyenhancedthe preservationpotentialof the buriedtracks. Individualtracksor pairs of tracks in the Holocene and Oligocenesediments are typicallywidely spaced laterally,but closely spaced vertically. The verticalspacingof track-bearinglaminaeallowsdivisionof stratainto discrete depositionalpackages. Close verticalspacing suggests that sand-drivingwinds deposited relativelythin sedimentpackagesin the NebraskaSandHillsand that tracks were producedby resident, ratherthanmigratorypopulations. Much of the Gering Formationis composed of similarlythin packagesthataccumulatedwithinan ephemeralstreamsystem dominatedby sheet-floodevents. The lackof heavilytrampled zones (whichmight be expected to mark diastems) suggests that the Holocenesedimentaccumulatedsteadily;in contrast, the same patternin the Oligocenerocks couldbe the result of rapidcementationof newly depositedsandby evaporites.The abundanceof vertebrate tracks in these strata shows that, under certain conditions, eolian dune fields and ephemeralstreamfloodplainscan be very favorablesites for preservation of vertebratetracks. 0883-1351/86/0001-0141/$03.00

142

LOOPE

430 ILc

r-

..-

--- cc"l`

~L-

'd

r

41?

?rrS;r*e

.:~

:" I-----

7-

4~,-.----~;

102?

FIGURE1-Map showing outline of the Nebraska Sand Hills and location of Burwell (B), Thedford (T), Valentine (V), and Scotts,Bluff National Monument (SB).

L I

. J,

The mainpurposes of this paper are: 1) to documentsedi- FIGURE 2-Concave-up deformation structures in large-scale crossmentarystructuresproducedby the hooves of large mammals stratified Holocene dune deposits at Calamus River dam site near in nonmarinesands and sandstones of central and western Burwell, Nebraska. Note that deformation occurs throughout vertical Nebraska;2) to providecriteriaby whichthese featurescan be extent of exposure. differentiatedfromnonbiogenicdeformationstructures;and3) to further demonstrate the utility of vertebrate tracks for winds swept ripples across surfaces with dips well below the sedimentologicaland paleoecologicalinterpretations. angle of repose. HOLOCENEOF THE NEBRASKASAND HILLS Descriptionof Tracks Locationand GeologicSetting The features interpretedhere as vertebrate tracks are condeformationstructuresthat are circularto oval in plan cave-up Occupyingan area of 57,000 square kilometers, the Ne- and from7 to 16 cm in diameter(Figs. 3, 4, 5). Near the range the in braska Sand Hills (Fig. 1) are largest dune field the Western Hemisphere (Smith, 1965). This dune field, now tops of structures, laminaeare abruptlytruncatedor sharply stabilizedby prairievegetation, is composed of simple and folded(Figs. 3, 4a). As in the structuresreportedby Van der LingenandAndrews(1969), concavitydies out graduallydowncompound,transverse to obliquebedformsthat reach heights ward. Althoughdeformationextends verticallyas much as 25 andFryberger,1980). Carbon-14dates up to 100 m (Ahlbrandt individual laminaeare displacedno more than15 cm. Within cm, reported by Ahlbrandtet al. (1983) from sediments directly some individualstructures, a central ridge divides the lower beneath the dunes suggest that the dune field is primarily Holocenein age. The sedimentarystructureswithinthe dunes portioninto two distinctlobes, whichare visible in both crossare visiblein a largenumberof exposures throughoutthe Sand sectionaland planviews (Fig. 4). The interiorsof many strucHills (Ahlbrandtand Fryberger, 1980). Most of the observations for this paperwere made near the eastern marginof the Sand Hills at excavations for the CalamusRiver dam near Burwell, Nebraska, and at blowouts and stream cuts near Thedfordin the central Sand Hills (Fig. 1). Structuresinterpreted here as vertebrate tracks are present throughoutthe Sand Hills; many are recognizablein publishedphotographs from the northern and central Sand Hills (Ahlbrandtand Fryberger, 1980, figs. 9, 10, 11, and 13). At all vertically extensive exposures, it is clear that the tracks are not restrictedto the upperportionsof the dunes, but are distributed throughoutthe thicknessof the eoliandeposit (Fig. 2). All deformationstructuresinterpretedas hoofprintsare developedin thin,inverselygradedlaminaedippingbetween 0 and 24 degrees. These laminaewere clearlydepositedby migrating windripplesand, in the terminologyof Hunter(1977), can be classifiedas subcriticallyclimbingtranslatentstrata. Laminae with very low dips accumulatedin interduneareas or as sand sheets (Frybergeret al., 1979); more steeply dippingstrata FIGURE3-Tracks in horizontal wind-ripple laminae. Interdune or were deposited on leeward slopes of bedforms, where side sand-sheet deposit near Burwell. Pen is 13 cm in length.

.

143

HOOFPRINTS CENOZOIC

N=41 LL

I

2

4

8

6

10

16

14

12

TRACK DIAMETER(CENTIMETERS) AFIGURE......~~~~verti

j"~n iN! ,,Hs*'"

~

..

AtI" :Ar ....-:e?o: ~ ' l'-i:;:':"!",.-~,!

b,,."/.the .." . ' ,:

FIGURE Size-frequency data for Holocene tracks measured from cal exposures, NebraskaSand Hills. Some variationin diameter .. is result of oval plan of structures(Fig. 4b).

Sand Hills, fortuitous exposures and small excavations providedglimpses of short trackways(Fig. 6) that along with the blgobedmorphologyof some structures, confirmedthe biogenic hypothesis. After the structures are recognized as tracks, the distributionof individualdeformationstructuresand other aspects of their morphologyprovideimportantsedimentologicaland paleoecologicalinsights.

ContrastwithNonbiogenicStructures The size-frequencydistributionof the deformationstructures (Fig. 5) stronglysupportsthe biogenicinterpretation:none of structuresare too large or too small to be tracks. When appearing32the closely spaced, however, the SandHillsstructuressuperficially resemble convolute bedding, a structure that Allen (1982) extensive series of more or less regular defines as a "lateray exterally folds developedthroughoutor confinedto the upperpart of a single sedimentationunit." Convolute bedding, however,i is vertical wallstrdand bilobedstlaower 4 (d-A)17FIGo~~~~uRE sFIGURE to root rapid sedimentation, portion.o Ir 4-AuTac due c B dati ENo ~with dee wly ~leTrack th sedimntatonuit.etres wthvetiypical Surfaceconcavitywas infilledby coarse-grained lag deposit.Note typicaly developed in strata that, due to rapidsedientation

raised "rim"at upper margins. Subhorizontallines crossing struc- were originally very loosely packed (Allen, 1982, p. 343). tures are post-depositionalaccumulationsof silt and clay (soil lame- Accordingly, this structure has been used as evidence for rapid Ilae or "dissipationstructures"of Ahlbrandtand Fryberger,1980). B) deposition (Collinson and Thompson, 1982, p. 145). Studies of Plan view of bilobed track near the middle of a three-meter-thick ' crossbed set. ?'*' : . ' , R, _ '.i:

tures are composedof materialthat is texturallydistinctfrom t ._ . the underlyingand surroundingdeformed sediment. Com' monly, this central portion contains very coarse sand andgranules(Fig. 4a); in rare cases, a thin (about5 mm) layer of .:! -* ": silt is preserved on the floor of the concave-upstructure. _, Along individualstratigraphiclevels, concave-upstructures are laterallyisolatedor in pairs. The structuresare commonly closely spacedvertically(Fig. 3) andmay dominatethe aspect of outcrops exposing several meters of strata (Fig. 2); laterally

adjacentbeds may lack such deformation.

;'

-

... '

-"

-

S

'

-

F

Interpretation In contrast to the easily recognized vertebrate trackways appearingon bedding planes (Sarjeant, 1975), most tracks appearingin cross sectionhave probablybeen eitherignoredor FIGURE 6-Part of shorttrackway in large-scalecross-stratified sand, misinterpreted (Lewis and Titheridge, 1978). Tracks in vertical

outcropscan closely resemble other types of deformation.In

I

Calamusdam site, Burwell,NE. Note lack of root traces in laminae belowtracks.Machetehandleis 15 cm long.

144

moder wind-rippledeposits indicatethat the stratacontaining the SandHillsdeformationstructureshad low initialporosities and were deposited relatively slowly. Because of strong differencesin grainpackingand porosity, the differenteolian stratificationtypes vary in their susceptibilityto deformation (Bagnold,1941; Hunter, 1981). Avalancheor grainflowstrata arelooselypacked;the large-scaledeformationcommonlyfound in ancient eolian sandstones is commonly concentrated in deposits of this type, because their high porosity allows liquefaction.In contrast, strata dominatedby climbing-ripple deposits-apparently due to their low initialporosities-are rarelyinvolvedinthiskindof deformation(Doe andDott, 1980). In lightof the evidence frommodem andancienteoliansands, the abundanceof relativelylarge-scaledeformationstructures in the wind-rippledeposits of the Sand Hills initiallyseemed incongruous.It was this paradoxthatled to the hypothesisthat the deformationstructureswithinthe typicallystablestrataare biogenic. AhlbrandtandFryberger(1980) explainedsome deformation structures within eolian strata of the Sand Hills (which are identicalto the ones interpretedhere as tracks)as the result of compressionat the base of slipfacedeposits. Several workers have reportedobservationsof this type of deformation.From the Coorongregion of southernAustralia,Brown (1969) describedarcuatefolds up to 200 m long and 0.5 m in amplitude that developedwithinlagoonalmuds that were overriddenby dunes 15-30 m high. McKee et al. (1971) experimentally producedsmall-scalewarps, folds, andoverthrustsin avalanching sand, but noted that "no contorted structures caused by tensionalor compressionalstresses normallyoccur in saltated deposits.... " There are, furthermore,fundamentaldifferences in formbetween the productsof lateralcompressionand the SandHillsdeformationstructures.The circularto ovalplan, lack of directionalasymmetry,and steep marginsof the Sand Hills structuresgive them a "punched-in" appearance,clearly indicatingthat they were formed by vertically, rather than laterallydirectedstress. The Substrate Trackmorphologyindicatesthat trackswere made in cohesive sand. Coarse-grainedsedimentsoverlyingtruncatedlaminae (Fig. 4a) are lag deposits that filled vertically walled depressions.Silts preservedin similarpositionsrepresentdust thatwas trappedandprotectedby concavities.Tracksmadein moist, well-sortedsand (wetted sand of McKee et al., 1971) have steep walls, whichcan be maintainedas long as the sand remainsmoist (Lewis andTitheridge,1978, fig. ld). Were the Sand Hills tracks formed in moist sand and buried before drying?Observationsof moder cattle tracksin the SandHills suggest an alternativehypothesis. Unlike beach sand, dune sand fromthe SandHillscontainsas muchas 4% silt and clay (Ahlbrandtand Fryberger, 1980, p. 21). The clay fractionis smectite, occurringas thin, detrital coatings on sand grains (Fig. 7; Ahlbrandtand Fryberger, 1982, fig. 21d). Such grain coatings are deposited by water moving throughthe vadose zone; they cansurviveconsiderableeoliantransport(Walkeret al., 1978; Walker,1979). In the Sand Hills duringAugust, 1985, vertical-sidedcattle trackswere abundantin wind-rippled,cohesive surface sands

LOOPE

FIGURE 7-SEM image of sand grains with detritalclay coatings and bridges. Sample is a "crumb"of recently deposited, but cohesive surface sand collected fromthe side of a vertical-sidedcow track in the central Sand Hills.

with a moisturecontent less than 1%. "Crumbs"of this cohesive sand collected in the field were found to retain their structureeven after 24 hours of oven dryingat 40?C. Using samples of loose sand from the Sand Hills and distilledwater, cohesive sand "crumbs"identicalto those collected in the field can be producedin the laboratorywith a single wetting/drying cycle (Fig. 8). In the field, cohesive surfacesands are relativelyresistantto wind erosion. Tracks made within such sands have a much higherpreservationpotentialthan tracks producedin mobile, cohesionlesssands that have not been moistened(Figs. 9, 10). This observationmay explain the prevalence of tracks with verticalsides in Holocenestrataof the SandHillsand suggests that clay-coateddune sandis an especiallysuitablemediumfor the preservationof tracks. In their study of the PermianLyons Sandstone, Walkerand Harms (1972) hypothesizedthat thin layers of clay deposited duringcalm periods between sanddrivingwindsmayhave allowedthe preservationof smalltracks and raindropimprintsmade in dry sand. Walker(1979) later showed the importanceof graincoatings composed of detrital clays to the reddening of eolian dune sands. With further

CENOZOIC HOOFPRINTS

145 I

FIGURE 8-Crumbs of cohesive sand (left) producedin the laboratory when loose surficialsand fromthe Sand Hillswas placed in a paper cup, saturated with distilled water, and dried in an oven at 40?C. Cohesionless sample on right was ultrasonicallycleaned and wet sieved to removeclays priorto identicaltreatment.Originalsample is a moderatelysorted, nearly symmetricalfine sand (Mz = 2.21; ua=0.662; Skl=.088) containing 1.3% clay, 10 YR7/3 (Munsell ColorChart).

FIGURE 9-Bison tracks in clay-coated eolian sand containing less than 1% water (by weight), FortNiobraraNationalWildlifeRefuge, near Valentine.Knifeis 15 cm long.

Holoceneage for the deposits; the largest cloven-hoofedmammals of the Holocene-bison-are capableof producingall the tracks so far observed. Paleoecologicalconsiderationsalso suggest that these tracks were made by bison. Archaeologicas well as paleontologic petrographicwork andlaboratoryexperiments,it may be pos- evidencefromseveral GreatPlainsandMiddle RockyMountain sible to assess the role (if any) of detritalclay coatingsin the sites links bison to semiarid in which environments grassland preservationof the delicate surface traces foundin Paleozoic sandy sediments have been subjected to of eolian episodes eoliansandstones. transport.Not only have bisonbones been recovered fromthe Sand Hills (Ahlbrandtand Fryberger, 1980) and from dune Track-Makersand Paleoenvironments The size and morphologyof the tracks, the probableage of sands at many other localities, but at the 10,000-year-old in the Sand Hills, the known mammalianfossil record, and pa- Caspersite east-centralWyoming,paleo-Indiansused paraleoecologicalarguments strongly suggest that bison are responsiblefor the tracks. The tracksof modem bison (Fig. 10) are nearlyidenticalin size andformto those preservedwithin + , E4a g S F F; r V. .. the SandHills. Untilrecently, students of the Sand Hills had placedmajor dune formationin either the early or late Wisconsin(Pleistor. '<*.. , t v f ,.b . y cene). Fluvialdeposits fromthe east-centralSandHills, which AK~~~~~~~~~~~~~~~~-l w are now knownto underlieas muchas 40 m of dunesand,have, _ ~ v r from 8410 10 radiocarbon dates to however, yielded ranging 3000 yrs. B.P., indicatingthat the dune field is primarilyof ii~GHolocene age (Ahlbrandtet al., 1983). This view has been al. on of et the basis radiocarbon (1985), challengedby Wright t h d dates and pollen from five interdune lake deposits in the P d t S? r. northernandwestern SandHills. At the Calamusdam site, all 44.' / f P O i S1. $ / ijP*pr? trackslie above a peat depositthathas been radiocarbon-dated x / 4 e r .~~~~~~~~~~~~~~~~~~~~~~~. at 7260 ? 90 yrs. B.P. (J. Swinehart,unpublisheddata;Beta' ^> r a' "i 11621); most tracks are above an organic-richinterdunedePL I Pi. .ti" '* rsr. posit dated at 3450 + 110 yrs. B.P. (ibid, Beta-11622). A ' L, diverse assemblage of large herbivores, includingcamels, and sloths roamed the Great Plains mammoths,horses, during C C the Pleistocene;of these, only camels producetracksthat are 's ''?.'t at all similarin size and shape to those describedhere. The youngest radiocarbondate for a North Americanfossil site FIGURE 10-Loose sand coveringa deposit of cohesive sand, which containingcamel bones is 8240 + 960 yrs. B.P. (Mead and contains numerousvertical-sided cattle tracks. CentralSand Hills Meltzer, 1984). All track observationsare consistent with a near Thedford.Shovel in backgroundfor scale. AW

ii

'

;'

A

( X

X

3'??2,

146

LOOPE

50 ft-15m

-U

0-0 c

-

C t C

V 0 c 0O

V

0 LU

vv9

9 /

E

9 T (/ )

0

I-

COt

- o

cn: 0.

a

vv T :--.V jL I

.' .

:.

I

'

I Massive sand

I Massive

I II siltstone

Concretions

_*1

sand

I

:

C-rv

*

.|Horizontallystratified Silicic

ash

sand

boliceoliandunes as naturaltraps for bison procurement(Frison, 1974). The abundanceof tracks in these sediments seems difficult to reconcile with the lack of root traces below track-bearing laminae. Were the tracks produced during transient movements of nonresidentherds?The wide spacingof tracks along beddingplanes and the close vertical spacingof track-bearing laminae argue against infrequent mass-migration events. McKee and Bigarella(1972) observed that small roots penetratingdunesandfollowlamination,therebyleavingfew traces. Trackabundanceand distributionsuggest that semipermanent water-perhaps located in interdune areas-and vegetation were availableto resident individualsor herds while the dunes were activelymigrating. OLIGOCENEOF SCOTTSBLUFF NATIONAL MONUMENT Stratigraphicand SedimentologicSetting Nonmarinerocks of mid-Tertiaryage are well exposed and easily accessible at Scotts BluffNationalMonumentin westernmost Nebraska (Fig. 1). These rocks contain abundant volcaniclasticmaterial,which, together with epiclasticdebris shed eastwardfrom the Rocky Mountains,blanketedthe surface of the GreatPlains(Stanley,1976; Swinehartet al., 1985). At Scotts Bluff,siltstones of the BruleFormation(WhiteRiver Group) are unconformablyoverlain by 27 m of very-finegrained,horizontallystratifiedsandstones of the Gering Formation(ArikareeGroup)(Fig. 11). Structuresinterpretedhere as vertebratetracks are restricted to the lower two-thirdsof the GeringFormation.The Geringis overlainby about65 m of large-scalecrossbeddedto massive sandstones of similartexture (MonroeCreek-Harrisonunit,Fig. 11). A volcanicash bed near the base of the Gering at Scotts Bluff has yielded a radiometricdate of 25.6 m.y. (Everndenet al., 1964), placing these basal Arikaree rocks within the Oligocene Series (Harlandet al., 1982). Figure 12 shows the locationof features describedand illustratedin this paper. At Scotts Bluff, the Geringis horizontallybedded; the only channelsobserved duringthis study are a few tens of centimeters deep. Small-scale, steeply dipping cross-lamination (Fig. 13), andplanarlaminationexhibitingpartinglineation,are widespreadthroughoutthis stratigraphicintervaland are clear evidence of fluvialdeposition(Stanleyand Fagerstrom, 1974). Earlydiageneticfeaturesin fluvialsedimentscommonlyprovide importantclues to depositionalprocesses andpaleoclimate (Collinson, 1978). Rosettes of calcite-cemented sand composed of discoids up to 8 cm in diameterare present at four separate stratigraphicintervals within the Gering (Figs. 11, 14). The morphologyof the discoidsandrosettes is identicalto that of modem gypsum sand crystals described by Cody (1979). Afterburial,as the gypsumwas dissolvedby less saline groundwater,replacementof gypsum by calcite was probably facilitatedby the commonness of the calciumion. The thick volcanicash bed near the middleof the Gering(Fig. 11) andthe FIGURE 11-Stratigraphic section and interpretationof depositional processes; Scotts Bluff National Monument (from Swinehart and Loope, in press). g=evidence of gypsumcrystallization.

HOOFPRINTS CENOZOIC

147

N

21b

0 I

I J Scale

FIGURE 14-Calcite-cemented sand, pseudomorphousafter gypsum FIGURE 12-Approximate map patternof contact between BruleFor- "desertroses," middlepartof GeringFormation.Knifehandle is 9 cm mation (WhiteRiverGroup)and GeringFormation(ArikareeGroup); from Scottsbluff South Quadrangle,4400-foot contour. Numbers long. show locations of text figures.

uppermostone meter of the Brule Formationcontainabundant discoids, 0.4 to 8 cm in diameter,that are composed of calcitespar. Withinthis finer-grainedmaterial,gypsumcrystal growthpushed the matrixaside to form relativelypure crystals; after dissolutionof gypsum, voids were filledby calcite. Gypsumcrystals of similarsize and morphologyto those recorded in the strata of the study area are today confinedto inlandand coastal sabkhas, where the groundwatertable approaches the land surface and undergoes evaporation(Watson, 1983). According to Watson, such materials are re-

3 Clr_ZI LY L:

I

?r----cikCi?T

cJ 6LLCYlk

;,?

,

P.

9_?;rF?T.L=,c-i,

L 7C?fi CIPh

.xz S

I

-lj

FIGURE 13-Subaqueous climbing-rippledeposits and concave-up deformationstructures, near middle of GeringFormation.Head of hammeris 18.5 cm long.

stricted to environmentswhere annualrainfallis less than 200 mm/yr and where there is a monthly excess of evaporation over precipitationthroughoutthe year. Evidence of growthof evaporiteswithinthe relativelypermeablefluvialsands of the study area strongly suggests that the strata were deposited by ephemeralstreams. Several lines of evidence indicate that a portion of the Arikaree Groupin eastern Wyomingand western Nebraska accumulatedwithinan eolian dune field (Bart, 1977; Stanley, 1980; Hunt, 1985; Swinehartand Loope, in press). Withinthe Monroe Creek-Harrisoninterval at Scotts Bluff, starting at about 10 m above the highest occurrence of tracks, inverse gradingand wind-rippleforesets (Hunter, 1977) occur in horizontallybedded sands and withinwedge planarcrossbed sets up to 1.7 m thick. Where the crossbed sets are overlainby horizontalwind-rippledeposits, the interveningboundingsurfaces commonlyhave an irregular,"corrugated"appearance with localrelief up to 10 cm (Fig. 15). Analogoussurfaces are commonin moder interduneareas where wind erosion has etched moist, wet, or evaporite-cemented,crossbeddeddune sands into strong relief (McKee, 1966, pl. VII, c and d; Fryberger et al. 1983, p. 298). Trenches dug in moder interdunesreveal crossbeds with wavy or "corrugated"upper boundingsurfaces, overlainby flat-beddedinterdunedeposits (Frybergeret al., 1983, fig. 23a; Simpsonand Loope, 1985). Stanley(1980) has noted that physicaland biogenicstructures of the mid-Tertiaryeolianstrataof the GreatPlainsbear many similaritiesto those withinthe Holocenesands in the Nebraska Sand Hills. Traces of invertebratesand plantroots are especiallyclosely analogous,suggestingto Stanleythat habitatsand climaticconditionswere very similar.The absence, however, of vertebrate tracks in the eolian strata of the Arikaree Group-both at Scotts Bluff and at the Bear Creek locality describedby Bart (1977)-indicates that, unlikethe Holocene

148

LOOPE

25N=104

20u 15-

O0

u-

5-

2

4 TRACK DIAMETER (CENTIMETERS)

FIGURE 17-Size-frequency data for deformationstructures inter15-Eolian interdunedepositsoverlyingirregular FIGURE contact pretedas tracks in GeringFormation,Scotts Bluff. Note that largest withcross-stratified dunesands, MonroeCreek-Harrison interval, tracks are largerthan those of Holocene bison (Fig. 5).

summitto museumtrail. Differential winderosionof lightlycementedordampcross-strata tookplaceininterdune areaadjacentto stoss side of dune.Notrackswereobservedin these eolianstrata.

SandHills, the mid-Tertiarydunefields either didnot harbora populationof large vertebrates, or lackedconditonsfavorable for the preservationof their tracks. Descriptionand Interpretationof Tracks Concave-updeformationstructures that closely resemble the Sand Hills (Holocene) bison tracks are common in the GeringFormation(Figs. 13, 16, 19). Deformationsinthe Gering vary in apparentdiameterfrom4 to 22 cm (Fig. 17). Because these rocks are poorlyindurated,bedding-planeexposures are small. On the undersidesof overhangingledges, however, the circularto oval plan of the deformationstructures is easily observed. An extensive search of such exposures revealed several distinct trackwayscomposed of three to five aligned tracks (Fig. 18). In vertical outcrops, isolated, paired, and

I,

_r-

.

;

closely spaced deformationstructures are visible. Some of these structures have vertically walled infillingsand bilobed lower portions (Fig. 19). As in the Sand Hills, tracks were producedin relativelyfirmbut compactiblesand. None of the tracks so far observed suggest that the substratewas "quick" at the time of deformation.Accordingto Simons et al. (1961), horizontallylaminatedsands deposited duringupper-flow-regime conditionsare firm relative to sediments deposited by on the lee side of ripplesor dunes. For most tracks, avalanching due to uniformityof grain size and to disruptioncaused by burrowingand evaporite diagenesis, it is difficultto discern whether the parent sediment was originallyparallelor crosslaminated.Therefore, no attempt has been made to compare trackdepth in the two types of strata. In contrastto the Holocenehoofprints,the Oligocenetracks appearin sedimentsthatalso show some evidence of physically induced deformation:convolute laminationappears in some exposures (Fig. 20). In nearlyallcases, however, biogenicand

M

'-

FIGURE16-Large,

X

'

'-, ',ts

closely spaced tracks at the tops of numerous, thin sediment packages, near base of Gering Formation.

CENOZOIC HOOFPRINTS

149

I

FIGURE18-Overhanging ledge revealing four aligned, bilobed tracks, middle partof GeringFormation.

FIGURE 19-Cross-sectional view of small, bilobed track made in cohesive sand by small entelodont or camel; middle part of Gering Formation.Coin is 1.9 cm in diameter.

physicallyinducedstructurescan be confidentlydifferentiated on the basis of scale, continuity,or three-dimensionalgeometry. A diverse assemblageof large mammalsis knownfromlate Oligocenerocks of the Great Plains. The distinct polymodal natureof the size-frequencydistribution(Fig. 17) suggests that the tracksof severalspecies are preservedin the Geringsands. The moder ungulateorders, PerissodactylaandArtiodactyla, first appeared in the Eocene (Romer, 1966). By the late Oligocene, diversificationof artiodactylsubgroups was well under way. Representative taxa of six artiodactyland four perissodactylfamiliesare knownin Geringsedimentsof Wildcat Ridge, southeast of Scotts Bluff (Swisher, 1982). A single fortuitousbedding-planeexposure (Fig. 21a) indicatesthat at least some of the largest trackswere producedby a two-toed animal. Morphologicallysimilar tracks from Oligocene and Miocene rocks have been described by Robertson and Sternberg (1942), Chaffee (1943), Bjork (1976), and Demathieu et al. (1984). Of the two-toed members of the Gering faunal assemblage, only entelodonts were large enough to makethese tracks(R. M. Hunt,personalcommunication). The small,bilobedtracks(Figs. 18, 19) were probablyproducedby smallerartiodactylssuch as camels. It is difficultto dividea horizontallybedded sequence composed of very-well-sortedsediment into distinct depositional packages. Traces of invertebrates provide clues in marine sequences (Howard,1978); tracksof terrestrialvertebratesin the Geringcan be utilizedin a similarway. The trackdistribution at Scotts Bluff mirrors that of the Sand Hills: laterally scattered tracks are typicallyseen at closely spaced vertical intervals.If tracks are assumed to markboundariesbetween depositionalevents, the distributionof tracks indicates that streamfloodsdepositedpackagesof stratabetween a few centimeters to over three meters in thickness. Thoroughlytrampled horizons(Fig. 16), as mightbe expected alongdiastems, are relativelyrare. Why are the tracks that markthe tops of sediment packages so widely spaced? One possibilityis that time intervals between depositionalevents were very brief

(LaporteandBehrensmeyer,1980, fig. 4a). Anotheris that the populationdensity of large vertebrates was very low. A third possibilityis that evaporiticsurface crusts formed after each pulse of sedimentation.Early evaporite diagenesis may have quicklymade the sandy substrate less compactibleand thus unsuitablefor preservationof tracks. If this was the case, only a very smallpercentageof the vertebrateactivitythatoccurred here maybe recordedby the tracks.The preservationpotential of tracks or other surficialtraces in sandy fluvialsequences would seem to be quite low because of their vulnerabilityto scour. Earlyevaporitecementationmay partiallyexplainboth the preservationof tracksand, conversely, the rarityandsmall scale of channelsin these ancientfluvialstrata. CONCLUSIONS 1. HoloceneeolianandOligocenefluvialsedimentsof Nebraska contain abundanttracks of large vertebrates. The most

c n_.r

i -?-?pr

-s-.-- ka . 71- - CJii;'.?i-regFI.

b j-a-l;?

;r;? inh

FIGURE 20-Convolute lamination(nonbiogenic),middlepartof Gering Formation.Note consistent asymmetryof folds.

LOOPE

150

,

'Wu 4I-?

ap

..

4 .:"

, .

i.-'O

.

I

. 41 All-~C4 "YW'T r 40L

have enhancedthe preservationpotentialof individualtracks by preventingscour and intense trampling. ACKNOWLEDGMENTS J. B. Swinehartprovidedtimely field trips and lively discussions. R. M. Hunt and M. R. Voorhies freely shared their knowledgeof Cenozoicfaunasandenvironments.I thankthese three Universityof Nebraskacolleagues for their enthusiasm and editorialassistance. DonaldBairdof PrincetonUniversity reviewedthe manuscriptandprovidedadditionalreferences on mammaltracks. Larry Cast of the Bureau of Reclamation helped with access to the Calamusdam site. Mike Leite's darkroomskills improvedmany of the photographs;Kit Lee helpedwith the SEM. Acknowledgmentis made to the Donors of the PetroleumResearch Fund, administeredby the American ChemicalSociety, for supportof this research. REFERENCES AHLBRANDT,T. S., and FRYBERGER,S. G., 1980, Eolian deposits in the

FIGURE21-Tracks of large entelodonts. A) Small, overhanging ledge revealing two large tracks (12 and 16 cm in length). Both tracks are cloven in front; large track was made by front foot. B) Cross section of asymmetric pair of tracks. Deeper track on right probably made by front foot. Middle part of Gering Formation.

obvious manifestationof the tracks is the downwarpingof laminaeseen in verticalsection. 2. Although tracks may superficiallyresemble nonbiogenic soft-sedimentdeformationstructures, their isolationalong bedding planes, circularto oval plan, and restricted size distributionaid in their recognition.Convolutebeddingof physical origin is generally restricted to initiallyporous, rapidlydepositedsediments. 3. TrackswithinHolocenestrataof the SandHillswere made in clay-coatedsands deposited by migratingwind ripples. Resident bison that were sustainedby food and water resources made the tracks while the dune field was active. 4. Oligocenerocks at Scotts Bluffare preservedat the tops of thin packages of fluvial deposits. Abundantevidence of evaporite precipitationsuggests deposition by ephemeral streams. Size-frequencyplots indicate that tracks were made by several differenttypes of ungulates;entelodonts producedthe largesttracks.Cementationby evaporitesmay

Nebraska Sand Hills, in Geologic and Paleontologic Studies of the Nebraska Sand Hills: U.S. Geol. Surv. Prof. Paper 1120, p. 1-24. AHLBRANDT,T. S., and FRYBERGER,S. G., 1982, Eolian deposits, in Scholle, P. A., and Spearing, D., eds., Sandstone Depositional Environments: Amer. Assoc. Petroleum Geologists Memoir No. 33, p. 11-47. AHLBRANDT,T. S., SWINEHART, J. B., and MARONEY,D. G., 1983, The dynamicHolocene dune fields of the Great Plains and Rocky Mountain basins, U.S.A., in Brookfield,M. E., and Ahlbrandt,T. S., eds., Eolian Sediments and Processes: Developments in Sedimentology, v. 38, p. 379-406. ALLEN,J. R. L., 1982, Sedimentary Structures, Their Character and Physical Basis: Developments in Sedimentology, v. 30B, 662 p. R. A., 1941, The Physics of Blown Sand and Desert Dunes: BAGNOLD, London, Methuen and Co., Ltd., 265 p. BART, H. A., 1977, Sedimentology of cross-stratified sandstones in Arikaree Group, Miocene, southeastern Wyoming:SedimentaryGeology, v. 19, p. 165-184. P. R., 1976, Mammaliantracks from the Brule Formationof South BJORK, Dakota: Proc. South Dakota Acad Sci., v. 55, p. 154-158. BLATT,H., MIDDLETON, G., and MURRAY, R., 1980, Originof Sedimentary Rocks (2nd ed.): Englewood Cliffs, N.J., Prentice-Hall, 782 p. R. G., 1969, Modern deformationalstructures in sediments of the BROWN, Coorong Lagoon, South Australia, in Brown, D. A., ed., Proceedings of the Specialist's Meeting: Geol. Soc. Australia, Spec. Publ. No. 2, p. 237-242. CHAFFEE, R. G., 1943, Mammalfootprintsfrom the White River Oligocene: Notulae Naturae (Acad. Nat. Sci. Philadelphia),No. 116, 13 p. CODY,R. D., 1979, Lenticular gypsum: Occurrences in nature, and experimentaldeterminationsof effects of soluble green plant material on its formation:Jour. Sed. Petrology, v. 49, p. 1015-1028. COLLINSON, J. D., 1978, Alluvial sediments, in Reading, H. G., ed., Sedimentary Environments and Facies: Oxford, Blackwell Scientific Publications,p. 15-60. COLLINSON, J. D., and THOMPSON,D. B., 1982, Sedimentary Structures: Winchester, Mass., Allen and Unwin, 194 p. G., GINSBURG, L., and TRUC,G., 1984, Etude paleontoloDEMATHIEU, gique, ichnologique et paleoecologicique du gisement oligocene de Saigon (bassin d'Apt, Vaucluse): Bulletin du Museum National d'Histoire Naturelle, series 4, v. 6, section C, no. 2, p. 153-183. DOE,T. W., and DOTT,R. H., JR., 1980, Genetic significanceof deformed crossbedding-with examples from the Navajo and Weber sandstones of Utah: Jour. Sed. Petrology, v. 50, p. 793-812. D. E., CURTIS, G. H., and JAMES,G. T., 1964, J. F., SAVAGE, EVERNDEN, Potassium-argon dates and the Cenozoic mammalianchronology of North America: Amer. Jour. Sci., v. 262, p. 145-198.

151

HOOFPRINTS CENOZOIC

G. C., ed., 1974, The CasperSite: A HellGapBisonKillon the FRISON, HighPlains:New York,AcademicPress, 266 p.

O. J., 1954, A Field Guideto AnimalTracks:Cambridge,The MURIE, RiversidePress, 374 p.

FRYBERGER,S. G., AHLBRANDT,T. S., and ANDREWS,S., 1979, Origin,

G. F., 1942, Fossil mammaltracks in G. M., and STERNBERG, ROBERTSON,

G., RICHMOND,B. HUNTER,R. E., CLIFTON,H. E., HALL,N. T., CSASZAR,

H. M., and DIFFENDAL,R. V. L., MCGRAW, SWINEHART, J. B., SOUDERS,

Graham of low-angleeolian"sandsheet" featuresandsignificance County,Kansas:Trans.KansasAcad.Sci., v. 45, p. 258-260. sedimentary A. S., 1966, VertebratePaleontology(3rd ed.): Univ. Chicago andvicinity,Colorado: ROMER, deposits,GreatSandDunesNationalMonument Press, 468 p. Jour.Sed. Petrology,v. 49, p. 733-746. W.A. S., 1975,Fossiltracksandimpressionsof vertebrates,in A. M., and CLISHAM,T. J., 1983, Eoliandune, SARJEANT, S. G., AL-SARI, FRYBERGER, sabkhasedimentsof anoffshore Frey, R. W., ed., The Studyof TraceFossils:New York,Springerinterdune,sandsheet, andsiliciclastic sandsea, Dhahranarea, SaudiArabia:Amer.Assoc. PeVerlag,p. 283-324. prograding M. L., 1961, Flume E. V., and ALBERTSON, D. B., RICHARDSON, troleumGeologistsBull.,v. 67, p. 280-312. SIMONS, P. G., PICKTON,C. A. G., studiesusingmediumsand(0.45 mm):U.S. Geol.Surv.Water-Supply HARLAND,W. B., Cox, A. V., LLEWELLYN, R., 1982, A Geologic Time Scale: CamPaper1490-A,p. A1-A76. SMITH,A. G., and WALTERS, interdunedeposits, D. B., 1985,Amalgamated E. L., andLOOPE, SIMPSON, bridgeUniversityPress, 131 p. WhiteSands,New Mexico:Jour.Sed. Petrology,v. 55, p. 361-365. andtrace fossils, in Basan,P. B., HOWARD, J. D., 1978, Sedimentology andchronologyin centraland H. T. U., 1965, Dunemorphology ed., TraceFossilConcepts:Soc. Econ.Paleontologists Mineralogists, SMITH, westernNebraska:Jour.Geology,v. 73, p. 557-578. ShortCourseNotes 5, p. 13-47. K. 0., 1976, Sandstonepetrofaciesin the CenozoicHighPlains HUNT,R. M., JR., 1985, Faunalsuccession,lithofacies,anddepositional STANLEY, andNebraska:Geol.Soc. Amer.Bull.,v. inArikareerocks(lowerMiocene)of the Hartville environments Table, sequence,easternWyoming in Martin,J. E., ed., FossiliferousCenozoicDe87, p. 297-309. Nebraska-Wyoming, of eolianfacies of the mid-Tertiary K. 0., 1980, Bioturbation posits of SouthwesternSouth Dakotaand NorthwesternNebraska: STANLEY, ArikareeGroup,easternWyoming[abstr.]:Geol. Soc. Amer.Abstr. RapidCity,S. D., Museumof Geology,SouthDakotaSchoolof Mines w. Progr.,v. 12, p. 256-257. andTechnology,Dakoterra,v. 2, pt. 2, p. 155-204. K. 0., and FAGERSTROM, in smalleoliandunes: STANLEY, R. E., 1977, Basictypes of stratification J. A., 1974, Miocene invertebrate trace HUNTER, fossilsfroma braidedriverenvironment,westernNebraska,U.S.A.: v. 24, p. 361-387. Sedimentology, R. E., 1981, Stratification Palaeogeog.Palaeoclim.Palaeoecol.,v. 15, p. 63-82. HUNTER, styles in eoliansandstones:some to Jurassicexamplesfromthe westerninteriorU.S.A., SWINEHART, J. B., and LOOPE,D. B., Late Cenozoic geology along the Pennsylvanian summitto museumhikingtrail,Scotts BluffNationalMonument,Nein Ethridge,F. G., and Flores, R. M., eds., Recent and Ancient braska:Geol.Soc. Amer.-Decade of NorthAmericanGeology,CenModelsfor Exploration: Soc. NonmarineDepositionalEnvironments: Section(inpress). tennialFieldGuide,North-Central Econ.Paleontologists Mineralogists Spec. Publ.No. 31, p. 315-329. of western Nebraska,in F., JR., 1985, Cenozoicpaleogeography of Flores, R. M., andKaplan,S. S., eds., CenozoicPaleogeography West-Central UnitedStates:Soc. Econ.Paleontologists Mineralogists Symposium Section,RockyMountain Paleogeography RockyMountain A. K., 1980, Tracks and substrate L. F., and BEHRENSMEYER, 3, p. 209-229. LAPORTE, of the eastern and biostratigraphy C. C., 1982, Stratigraphy sedimentsof Ken- SWISHER, reworkingby terrestrialvertebratesin Quaternary portionof WildcatRidge,western Nebraska[M.S. thesis]: Lincoln, ya:Jour.Sed. Petrology,v. 50, p. 1337-1346. Univ.Nebraska,172 p. mammothof a late Quaternary R. L., 1980, Paleoenvironment LAURY, P. B., 1969, Hoof-printstructures G. J., and ANDREWS, bearingsinkholedeposit,HotSprings,SouthDakota:Geol.Soc. Amer. VANDERLINGEN, in beachsand:Jour.Sed. Petrology,v. 39, p. 350-357. Bull.,v. 91, p. 465-475. T. R., 1979, Red colorin dunesand,in McKee,E. D., ed., A D. G., 1978, Small scale sedimentary WALKER, LEWIS,D. W., and TITHERIDGE, structuresresultingfromfoot impressionsin dunesands:Jour.Sed. Studyof GlobalSandSeas: U.S. Geol. Surv. Prof. Paper 1052, p. 61-81. Petrology,v. 48, p. 835-838. T. R., andHARMS, E. D., 1966, Structuresof dunesat WhiteSandsNationalMon- WALKER, J. C., 1972, Eolianoriginof flagstonebeds, MCKEE, withstructuresof dunesfrom Lyons Sandstone(Permian),type area, BoulderCounty,Colorado: ument,New Mexico(anda comparison Mountain v. 7, p. 3-69. otherselectedareas):Sedimentology, Geologist,v. 9, p. 279-288. structuresin WALKER,T. R., WAUGH, B., and CRONE, A. T., 1978, Diagenesis in J. J., 1972, Deformational McKEE,E. D., andBIGARELLA, of Cenozoicage, southwesternUnitedStates Brazilian coastaldunes:Jour.Sed. Petrology,v. 42, p. 670-681. first-cycledesertalluvium andnorthwesternMexico:Geol. Soc. Amer.Bull.,v. 89, p. 19-32. S., 1971, Deformation McKEE,E. D., DOUGLASS, J. R., and RITTENHOUSE, of lee-sidelaminaein eoliandunes:Geol. Soc. Amer.Bull.,v. 82, p. WATSON, A., 1983, Gypsumcrusts, in Goudie,A. S., andPye, K., eds., Chemical SedimentsandGeomorphology: London,AcademicPress, p. 359-378. 133-161. D. J., 1984, NorthAmericanlate Quaternary MEAD,J. I., andMELTZER, extinctionsandthe radiocarbon J. C., and GRUGER,J., 1985, Pollen record,in Martin,P. S., andKlein,R. WRIGHT,H. E., JR., ALMENDINGER, G., eds., QuaternaryExtinctions:Tucson, Univ. ArizonaPress, p. diagramfromthe NebraskaSandhillsandthe age of the dunes:Qua440-450. ternaryResearch,v. 24, p. 115-120. J. L., 1984, PlioceneandPleistocenecoastalandshelf M., andCHIN, andassociatedbeds, northwestern depositsof the MercedFormation SanJose, Soc. Econ.PaleontoloSanFranciscoPeninsula,California: No. 3, p. 1-29. gists Mineralogists, MidyearMeeting,Guidebook

I