Geology

doi: 10.1130/0091-7613(1995)023<0613:CHSEDO>2.3.CO;2 1995;23;613-616Geology

 Stephen G. Wells, Leslie D. McFadden, Jane Poths and Chad T. Olinger landscape evolution in deserts

He surface-exposure dating of stone pavements: Implications for3Cosmogenic   

Email alerting servicescite this article

to receive free e-mail alerts when new articleswww.gsapubs.org/cgi/alertsclick

Subscribe to subscribe to Geologywww.gsapubs.org/subscriptions/click

Permission request to contact GSAhttp://www.geosociety.org/pubs/copyrt.htm#gsaclick

viewpoint. Opinions presented in this publication do not reflect official positions of the Society.positions by scientists worldwide, regardless of their race, citizenship, gender, religion, or politicalarticle's full citation. GSA provides this and other forums for the presentation of diverse opinions and articles on their own or their organization's Web site providing the posting includes a reference to thescience. This file may not be posted to any Web site, but authors may post the abstracts only of their unlimited copies of items in GSA's journals for noncommercial use in classrooms to further education andto use a single figure, a single table, and/or a brief paragraph of text in subsequent works and to make

GSA,employment. Individual scientists are hereby granted permission, without fees or further requests to Copyright not claimed on content prepared wholly by U.S. government employees within scope of their

Notes

Geological Society of America

on June 15, 2014geology.gsapubs.orgDownloaded from on June 15, 2014geology.gsapubs.orgDownloaded from

Cosmogenic 3He surface-exposure dating of stone pavements:Implications for landscape evolution in desertsStephen G. Wells Department of Earth Sciences, University of California, Riverside, California 92521Leslie D. McFadden Department of Earth and Planetary Sciences, University of New Mexico,

Albuquerque, New Mexico 87131Jane Poths

Isotope Chemistry, Los Alamos National Laboratory, Los Alamos, New Mexico 87545Chad T. Olinger

ABSTRACTThe formation of stone pavements, a ubiquitous gravel armor mantling landforms in

arid regions of the world, has been previously attributed to erosion by wind and water oralternating shrinking and swelling of soil horizons, implying that gravel is concentrated atthe land surface in a time-transgressive manner. A newly proposed model for pavementevolution differs from these models in that pavement clasts are continuously maintainedat the land surface in response to deposition and pedogenic modification of windblown dust.In-situ cosmogenic 3He surface-exposure ages on volcanic and alluvial landforms in theMojave Desert of California are used to understand pavement evolution over geologic timescales and to test this new model. These exposure ages are stratigraphically consistent,show internal consistency at each site, and, for stone pavements adjacent to pristine,continuously exposed volcanic bedrock, are indistinguishable at the 1s level. We concludethat stone pavements are born at the surface and that pavements may provide one of thelongest-term records of geologic, hydrologic, and climatic processes operating on desertsurfaces.

INTRODUCTIONSurface-exposure ages of landforms, as

determined from in-situ–produced cosmo-genic isotopes, are currently used to deter-mine the timing and rates of geologic andhydrologic events during the Quaternary(reviewed in Cerling and Craig, 1994). Inthis paper, we apply the method of cosmo-genic 3He exposure dating to elucidate theformation of stone (desert) pavements, oneof the most prevalent features mantling al-most all types of desert surfaces (Cooke etal., 1993; Mabbutt, 1977). Although a stonepavement is only a one- to two-clast-thicklayer of closely packed angular to sub-rounded gravels that armor low-relief sur-faces (Fig. 1), stone pavements control sur-face stability and hydrology, record activitiesof chemical and physical processes on desertsurfaces, serve as a source and storage sitefor archaeological materials, and serve as amappable feature for relative age control(Cooke et al., 1993). An understanding ofthe geomorphic processes that concentrateclasts at the land surface and the time scalesof these processes is important to any land-management application or scientific inter-pretation of pavement attributes.Pavement-forming processes have been

evaluated primarily by experimental obser-vations at time scales of years to decades.Pavement formation has been attributed to(1) wind deflation of fine particles, creatinga coarse-grained lag; (2) surface runoff andlateral transport by creep, resulting in a win-nowing of fine clasts; and (3) shrink-swellprocesses of soils, causing upward migration

of clasts to the land surface (Cooke et al.,1993; Thomas, 1989). In all of these hypoth-eses, clasts are concentrated at the land sur-face at significantly different times and, thus,are not necessarily related to the time offormation of the landform. None of thesehypotheses, however, has been quantita-tively evaluated at scales of 103 to 104 yr.A new model of desert-pavement forma-

tion (McFadden et al., 1986, 1987; Wells etal., 1984, 1985) proposes that deposition ofwindblown sediments, not deflation or watererosion, is the major agent of pavement evo-lution. According to this model, the stonepavement remains at the land surface on anaccretionary (i.e., vertically growing) mantleof soil-modified dust (Fig. 1). This hypoth-esis implies (1) that pavement clasts havebeen continuously exposed since the forma-tion of the underlying landform and associ-ated deposits and (2) that the properties ofclasts in pavements (e.g., Wells et al., 1985,1987) reflect the types of surficial processesoperative since the underlying landform ordeposit was created. Thus, we refer to thismodel as pavement formation by being‘‘born at the surface.’’Surface-exposure dating using cosmo-

genic isotopes provides a quantitative ap-proach to understanding pavement evolu-tion beyond the time scale of experimentalobservations. We compare in-situ cosmo-genic 3He exposure ages of clasts from pave-ments with exposure ages of their bedrocksource, late Quaternary basaltic flows. Thismethod tests whether pavement clasts re-main continuously at the surface or are con-

centrated at the surface randomly over time.Young basaltic lava flows provide a uniqueopportunity to test the various models ofpavement formation since they can bereadily dated using cosmogenic 3He (Kurz,1986; Cerling, 1990; Laughlin et al., 1994),which is quantitatively retained in olivineand pyroxene. Such flows have unerodedbedrock highs surrounded by topographi-cally lower stone pavements, thus allowing acomparison of the exposure age of the lavaflow (i.e., time of emplacement of flow) withthe surface-exposure ages of stone-pave-ment clasts derived from the flows.

Figure 1. Stratigraphic section and corre-sponding pavement and soil-profile descrip-tion for accretionary mantle developed on ba-salt flow derived from vent I (CVF-Qv5). SeeWells et al. (1985) and McFadden et al. (1987)for processes of accretionary-mantle forma-tion. For morphological properties of soil pro-file and properties of stone pavement, seeDohrenwend et al. (1984), Wells et al. (1984,1985), and McFadden et al. (1986, 1987); overgeologic time, pavement surfaces are char-acterized by reduction in original construc-tional relief of a landform, concentration ofcoarse interlocking gravel, relatively gravel-free layer, and weakly to moderately devel-oped soil.

Geology; July 1995; v. 23; no. 7; p. 613–616; 3 figures; 1 table. 613

on June 15, 2014geology.gsapubs.orgDownloaded from

Cosmic rays penetrate the upper few me-tres of the Earth’s surface, creating detect-able amounts of rare nuclides (Lal, 1991).The production of cosmogenic nuclides de-clines by a factor of two for every 75 cmbelow the surface in a sediment with a den-sity of 1.58 g/cm3. Previous pavement stud-ies (Mabbutt, 1977) suggest clasts in pave-ments were buried in excess of 50 cm and, insome cases, in excess of 100 cm prior to theirexposure at the surface. If clasts currentlywithin a desert pavement were buried belowthe surface at such depths for a significantamount of time prior to their concentrationat the land surface, a stone pavement devel-oped over a basaltic flow will have (1) anapparent surface-exposure age that is lessthan the age of constructional topographichighs continuously exposed since the lavasolidified and (2) considerable scatter in theages as clasts arrive at the surface at differ-ent times. In contrast, stone pavements withsurface-exposure ages similar to the age ofthe continuously exposed bedrock highs in-dicate that the pavements are born at thesurface.

SAMPLING STRATEGY ANDMETHODOLOGYThis study area is the Cima volcanic field

in the Mojave Desert of eastern Californiathat contains ;40 basaltic scoria cones and.60 associated lava flows, ranging in agefrom late Tertiary to latest Pleistocene(Dohrenwend et al., 1984; Turrin et al.,1985) Stone pavements are present on pedo-genically altered accretionary mantles(Fig. 1) developed over topographic lowsand highs on these lava flows (McFadden etal., 1986; Wells et al., 1985). Three latePleistocene basalt flows (from oldest toyoungest: Qv4, Qv5, and Qv6) with distinct

stratigraphic and geographic positions weresampled to evaluate whether the cosmo-genic 3He surface-exposure dates would re-flect the relative chronology of the volcaniceruptions (Fig. 2). The topographicallyhigher areas of these flows were carefullysampled in locations where original volcanictextures (e.g., pahoehoe) were well pre-served and where it appeared that erosionhad been minimal (,10 cm). Such locationsprovided the continuously exposed bedrocksamples in this study, including the flowsfrom vent I (Qv5) and vent H (Qv4) (Fig. 2).We also collected individual large pavementclasts derived from and adjacent to thesetwo flows. In the first case, two samples each(CVF-Qv5-P2 through -P4) were taken fromtwo geographically distinct desert pave-ments separated by a few metres horizon-tally and ;1 m vertically. Bedrock sampleswere taken from a nearby and isolated vol-canic tumuli (CVF-Qv5b-FS6 and -FS7). On

the older lava flow (Qv4), samples were col-lected from a stone pavement (CVF-Qv4-P4), and another from an adjacent bedrockhigh (CVF-Qv4-FS5) (Fig. 3). At a secondsample site on flow Qv4, a sample was col-lected from the distal end of a flow thatburied a stone pavement developed on alate Pleistocene alluvial fan surface (Qf3)(Harden et al., 1991) (Fig. 2). Near the ter-minus of flow Qv4, the sample CVF-Qf3-P1was collected from the exposed portion of astone pavement developed on an alluvial fanthat had been partially buried by flow Qv4.From stratigraphic relations, the sampleCVF-Qf3-P1 from the alluvial fan shouldpredate samples fromQv4, yielding an olderexposure age. Samples CVF-Qv6-CS1 and-CS2 were taken from volcanic bombs onthe scoria cone rim of Black Tank (vent A)volcanic center, and sample CVF-Qv6-FS1was taken from the surface of a basalt flowthat breaches and thus postdates the scoria

Figure 2. Location map of cos-mogenic 3He surface-exposuresamples on volcanic and alluvi-al-fan landforms, southwesternCima volcanic field. Stratigraph-ic relations of landforms shownin upper left corner along withtheir associated sample num-bers (see Table 1 for sampledata).

Figure 3. Cosmogenic 3He sur-face-exposure ages for selectedstratigraphic samples of desertpavement, basalt-flow surfaces,and bombs in scoria cones,southwestern Cima volcanicfield (see Table 1 for details andFig. 2 for locations).

614 GEOLOGY, July 1995

on June 15, 2014geology.gsapubs.orgDownloaded from

cone (Fig. 2). The Qv6 samples are strati-graphically younger than all the other sam-ples (Wells et al., 1994). Thus, our samplingstrategy has a built-in check from strati-graphic relations, allowing us to assess theresolution of the chronologic concurrenceyielded by the surface-exposure dates.The concentrations of cosmogenic 3He

were determined on olivine and pyroxeneseparates by means of a standard, two-stepprocess (Kurz, 1986; Cerling, 1990; Laughlinet al., 1994). Mineral separates were firstcrushed in a vacuum, allowing us to analyzethe magmatic He released from inclusions.We then extracted and analyzed the remain-ing mixture of cosmogenic and magmaticHe, released by a fusion step. The amount ofcosmogenic 3He is determined by the equa-tion

3Hec 5 3Hef 2 4Hef(3He/4He)cr, (1)

where the subscripts c, f, and cr denote cos-mogenic, fusion, and crush, respectively.This deconvolution assumes that only twoHe components are present: cosmogenic3He and magmatic He. The magmatic He ischaracterized by the 3He/4He ratio that re-sults from the crushing step. The 3He/4Heratios released by crushing were averagedfor samples that were duplicates or from agiven flow (e.g., all of the samples for flowQv5, Fig. 2). The averaged (3He/4He)crvalue used for the members of those samplesets has an uncertainty that reflects the scat-ter of the individual values. Concentrationsof cosmogenic He were then corrected by2% to 5% for the sample thickness (Cerlingand Craig, 1994), giving the values listed inTable 1. We calculated the surface-exposureages by using the production rate of Cerling(1990), corrected for altitude, latitude, andthe recalibration of the 14C time scale (Cer-ling and Craig, 1994). Stated errors are 1sand include the uncertainties in these as-sumptions as well as the analytical assump-

tions. We omit the ;20% uncertainty asso-ciated with the absolute calibration of theproduction rate in that this factor will biasall ages equally.

SURFACE-EXPOSURE AGES:EVALUATING STRATIGRAPHICRELATIONS AND MODELS OFPAVEMENT FORMATIONThe 14 surface-exposure ages are concor-

dant with the chronology of volcanic erup-tions established in the field by stratigraphicrelations (Table 1; Figs. 2 and 3). Samplesfrom Qf3 (85 6 9 and 80 6 10 ka) are ap-proximately the same age or slightly olderthan Qv4 (74 6 7 and 72 6 7 ka), which, inturn, are older than Qv5 (37 6 6 and 31 67 ka). Flow Qv6 yields the youngest ages(18 6 12 ka and 20 6 10 ka for the volcanicbombs and 13.2 6 3 ka for the lava flow), inagreement with an independent surface-exposure age (J. Stone, 1989, personal com-mun.) and thermoluminescence dates (Wellset al., 1994).The pavement and bedrock data also

show internal agreement at each site wheremultiple samples were analyzed. Stone-pavement samples CVF-Qv5-P2 through-P5 are indistinguishable at the 1s level,ranging from 41 6 6 to 31 6 6 ka. Likewise,the ages of the stone-pavement samples arestatistically identical with the bedrock sam-ple ages of 31 6 7 and 37 6 6 ka (CVF-Qv5-FS6 and -FS7). Bedrock samples fromthe stratigraphically older flow, Qv4, haveconcordant values of cosmogenic 3He ages,72 6 7 and 74 6 7 ka, as well as a slightlyyounger but overlapping value (65 6 9 ka)on its associated stone pavement. Theagreement between surface-exposure agesof well-preserved pahoehoe surfaces on ba-saltic flows and those of their associatedstone pavements developed over accretion-ary mantles indicates that clasts formingstone pavements have been exposed contin-

uously since the emplacement of the basaltflows.At vent I (Fig. 2), two samples (CVF-

Qv5-P4 and -P5) were collected from apavement overlying a 20- and 30-cm-thickaccretionary mantle on a slight bedrockhigh. Two additional samples (CVF-Qv5-P2and -P3) were collected from a pavement ina topographic depression (i.e., created dur-ing emplacement of lava and not by defla-tion) in which the accretionary mantle(Fig. 1) exceeds 50 cm thickness. Both sam-ple sets from pavements in differing geo-morphic positions yield similar ages (Fig. 3,Table 1). These results substantiate themodel that stone pavements remain at theland surface in response to vertically accret-ing eolian fines and their pedogenic alter-ation (McFadden et al., 1987; Wells et al.,1985). These results also demonstrate thatdifferent regions of stone pavements on thesame underlying rock unit yield the samesurface-exposure ages.If clasts comprising a stone pavement ar-

rive at the land surface by shrinking andswelling of soils, wind deflation, or winnow-ing by water, the surface-exposure ages inthese two different geomorphic positionsshould not be the same. Soil-profile proper-ties, including shrink-swell conditions, differbetween topographic lows and highs on ba-salt flows because of the trapping efficiencyof fine sediments in the internally drainedbasins (McFadden et al., 1986). Also, theaccretionary mantle overlying flow Qv5 has2–3-cm-thick, weakly developed Bt horizonsoverlying decimetre-thick CBk horizons(Fig. 1). This degree of soil-profile develop-ment is incapable of creating shrink-swellconditions that would allow clasts to migrate0.5 to 0.7 m upward to the land surface. Thelack of winnowing by water at these sites isdemonstrated by field observations showingthat boundaries of many pavement clasts fittogether like a jigsaw puzzle. Significant al-

GEOLOGY, July 1995 615

on June 15, 2014geology.gsapubs.orgDownloaded from

teration of the surface by running waterwould eliminate the compatible geographicposition with adjoining pavement clasts.On the basis of similar stone-free eolian

accretionary layers and pedogenic features,McFadden et al. (1987) and McDonald(1994) proposed that stone pavements onalluvial fans evolve in a manner similar tothat on the basalt flows. Testing this hypoth-esis by cosmogenic exposure ages is difficultbecause clasts on alluvial landforms havetransport histories with the potential forlong-term exposure (Nishiizumi et al.,1993). In our study, exposure ages of thealluvial-fan (Qf3) pavement are estimatedto be 856 9 and 806 10 ka (QVF-Qf3-P1/1and -P1/2). The only independent age de-terminations on the Qf3 deposits are given,in part, by a series of K-Ar dates (Turrin etal., 1985) that bracket the exposure ages de-rived from the alluvial-fan pavement. No in-dependent isotopic date exists to supportthe surface-exposure age of alluvial fan Qf3.

CONCLUSIONSThe similarity of cosmogenic 3He surface-

exposure ages for relatively uneroded basal-tic-flow surfaces and adjacent stone pave-ments substantiates that individual clastswithin stone pavements have been continu-ously exposed at the land surface since theemplacement of the underlying volcanicflow (i.e., up to 105 yr) and that clasts are notconcentrated at the surface randomly overgeologic time (Fig. 3; Table 1). Rather,stone pavements remain at the surface be-cause of vertical inflation caused by deposi-tion of windblown dust and its subsequentpedogenic modification. Gravel-free layersunderlying stone pavements (Fig. 1) serve asa criterion for distinguishing born-at-the-surface pavements from those resultingfrom wind and water erosion (Williams andZimbelman, 1994).Pavement clasts prevent erosion of eolian

accretionary mantles for hundreds of thou-sands of years and, consequently, preserve along-term record of dust influx and soil for-mation. Over geologic time, the propertiesof pavements change systematically (seeFig. 1; Wells et al., 1985, 1987), reflectingthe history of surficial processes operativesince the formation of the landform and itsunderlying deposit. On pavement surfaces,the degree of soil-profile development, theamount of varnish cover, and Mn accumu-lations in varnish have been used in discrim-inating among deposits of Holocene andPleistocene age (McFadden et al., 1989; Re-neau, 1993). These properties are stronglydependent upon dust influx and relativelyindependent of the depositional characterof the landform, such as lithology (Mc-Donald, 1994). The genetic bond among

dust influx, soil formation, and pavementevolution explains why some pavementproperties provide consistent age estimatesfor the landform and/or deposits that theyoverlie. On basaltic lava flows, the durationof geologic, hydrologic, and climatic activi-ties operating on the pavement can be eval-uated because of the long exposure histo-ries. Exposure ages for pavements mantlingfluvial, lacustrine, or colluvial deposits willmost likely contain an inherited exposureowing to the transport path (Nishiizumi etal., 1993). For those pavement clasts thathave been continually exposed on alluviallandforms, theexposureagesmaynotbe con-sistent with the deposition and/or construc-tion of the underlying landform.

ACKNOWLEDGMENTSSupported by National Science Foundation

grant EAR-9205696, Los Alamos National Lab-oratory, the University of California GraniteMountain Reserve, the California State Univer-sity Desert Studies Center, and the GeologicalSociety of America Gladys Cole Memorial Re-search Award (toWells). We thank E.McDonald,K. Anderson, R. Fleming, J. McAuliffe, C. Har-rington, R. Fulton, and many others for valuablediscussions and insights; K. Anderson for assistingin the field and in preparing this manuscript; andJ. Knott, K. Kendrick, T. Williamson, and A.Gillespie for comments on the manuscript.

REFERENCES CITEDCerling, T. W., 1990, Dating geomorphologic sur-

faces using cosmogenic He-3: QuaternaryResearch, v. 33, p. 148–156.

Cerling, T. W., and Craig, H., 1994, Geomorphol-ogy and in-situ cosmogenic isotopes: AnnualReview of Earth and Planetary Sciences,v. 22, p. 273–317.

Cooke, R., Warren, A., and Goudie, A., 1993,Desert geomorphology: London, UCL Press,526 p.

Dohrenwend, J. C., McFadden, L. D., Turrin,B. D., and Wells, S. G., 1984, K-Ar dating ofthe Cima volcanic field, eastern MojaveDesert, California: Late Cenozoic volcanichistory and landscape evolution: Geology,v. 12, p. 163–167.

Harden, J. W., Taylor, E. M., Hill, C., Mark,R. K., McFadden, L. D., Reheis, M. C., Sow-ers, J. M., andWells, S. G., 1991, Rates of soildevelopment from four soil chronosequencesin the southern Great Basin: Quaternary Re-search, v. 35, p. 383–399.

Kurz, M. D., 1986, Cosmogenic helium in terres-trial igneous rocks: Nature, v. 320, p. 435–439.

Kurz, M. D., Colodner, D., Trull, T. W., Moore,R. B., and O’Brien, K., 1990, Cosmic ray ex-posure dating with in situ produced cosmo-genic 3He: Results from Hawaiian lava flows:Earth and Planetary Science Letters, v. 104,p. 424–439.

Lal, D., 1991, Cosmic ray labeling of erosion sur-faces: In-situ nuclide production rates anderosion models: Earth and Planetary ScienceLetters, v. 104, p. 424–439.

Laughlin, A.W., Poths, J., Healey, H. A., Reneau,S., and WoldeGabriel, G., 1994, Dating ofQuaternary basalts using cosmogenic 3Heand 14C methods with implications for excess40Ar: Geology, v. 22, p. 135–138.

Mabbutt, J. A., 1977, Desert landforms: Cam-

bridge, Massachusetts Institute of Technol-ogy Press, 340 p.

McDonald, E., 1994, The relative influences ofclimate change, desert dust, and lithologiccontrol on soil-geomorphic processes and hy-drology of calcic soils formed on Quaternaryalluvial-fan deposits in the Mojave Desert,California [Ph.D. thesis]: Albuquerque, Uni-versity of New Mexico, 383 p.

McFadden, L. D., Wells, S. G., and Dohrenwend,J. C., 1986, Influences of Quaternary climaticchanges on processes of soil development ondesert loess deposits of the Cima volcanicfield: Catena, v. 13, p. 361–389.

McFadden, L. D., Wells, S. G., and Jercinovich,M. J., 1987, Influence of eolian and pedo-genic processes on the origin and evolu-tion of desert pavements: Geology, v. 15,p. 504–508.

McFadden, L. D., Wells, S. G., and Ritter, J. B.,1989, Use of multiparameter relative-agemethods for age determination and correla-tion of alluvial-fan surfaces on a desert pied-mont, eastern Mojave Desert, California:Quaternary Research, v. 32, p. 276–290.

Nishiizumi, K., Kohl, C. P., Arnold, J. R., Dorn,R. I., Klein, J., Fink, D., Middleton, R., andLal, D., 1993, Role of in situ cosmogenic nu-clides 10Be and 26Al in the study of diversegeomorphic processes: Earth Surface Pro-cesses and Landforms, v. 18, p. 407–425.

Reneau, S. L., 1993, Mn accumulation in desertrock varnish: Quaternary Research, v. 40,p. 309–317.

Thomas, D. S. G., 1989, Arid zone geomorphol-ogy: New York, Halsted Press, 372 p.

Turrin, B. D., Dohrenwend, J. C., Drake, R. E.,and Curtis, G. H., 1985, K-Ar ages fromCima volcanic field, eastern Mojave Desert,California: Isochron/West, v. 44, p. 9–16.

Wells, S. G., Dohrenwend, J. C., McFadden,L. D., Turrin, B. D., andMahrer, K. D., 1984,Types and rates of late Cenozoic geomorphicprocesses on lava flows of the Cima volcanicfield, eastern Mojave Desrt, California, inDohrenwend, J. C., ed., Surficial geology ofthe eastern Mojave Desert of California:Reno, Nevada, University of Nevada, p. 69–87.

Wells, S. G., Dohrenwend, J. C., McFadden,L. D., and Turrin, B. D., 1985, Late Cenozoiclandscape evolution on lava flow surfaces ofthe Cima volcanic field, Mojave Desert, Cal-ifornia: Geological Society of America Bul-letin, v. 96, p. 1518–1529.

Wells, S. G., McFadden, L. D., and Dohrenwend,J. C., 1987, Influences of late Quaternary cli-matic changes on geomorphic and pedogenicprocesses on a desert piedmont, eastern Mo-jave Desert, California: Quaternary Re-search, v. 27, p. 130–146.

Wells, S. G., McFadden, L. D., Geissman, J.,Olinger, C. T., and Renault, C. E., 1994,Quaternary geology of polycyclic volcanos:Black Tank volcanic center, Cima volcanicfield, California, in McGill, S. F., and Ross,T. M., ed., Geological investigations of anactive margin: Redlands, California, SanBernardino County Museum Association,p. 195–200.

Williams, S. H., and Zimbelman, J. R., 1994,Desert pavement evolution: An example ofthe role of sheetflood: Journal of Geology,v. 102, p. 243–248.

Manuscript received September 26, 1994Revised manuscript received April 12, 1995Manuscript accepted April 24, 1995

616 GEOLOGY, July 1995Printed in U.S.A.

on June 15, 2014geology.gsapubs.orgDownloaded from