Palaeogeography, Palaeoclimatology, Palaeoecology 309 (2011) 327–332

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

j ourna l homepage: www.e lsev ie r.com/ locate /pa laeo

Atmospheric paleo-CO2 estimates based on Taxodium distichum (Cupressaceae)fossils from the Miocene and Pliocene of Eastern North America

Debra Z. Stults a, Friederike Wagner-Cremer b, Brian J. Axsmith c,⁎a Department of Marine Sciences, LSCB 25, University of South Alabama, Mobile, AL 36688, USAb Palaeoecology, Laboratory of Palaeobotany and Palynology, Institute of Environmental Biology, Utrecht University, Budapestlaan 4, 3584 CD, Utrecht, The Netherlandsc Biology Department, LSCB 124, University of South Alabama, Mobile, AL 36688, USA

⁎ Corresponding author. Tel.: +1 251 460 7528; fax:E-mail address: baxsmith@jaguar1.usouthal.edu (B.J

0031-0182/$ – see front matter © 2011 Elsevier B.V. Aldoi:10.1016/j.palaeo.2011.06.017

a b s t r a c t

a r t i c l e i n f o

Article history:Received 4 April 2011Received in revised form 23 June 2011Accepted 25 June 2011Available online 2 July 2011

Keywords:Brandywine FormationCitronelle FormationMiocenePaleo-CO2

Pliocene

Neogene atmospheric paleo-CO2 estimates based on fossils of the extant cupressaceous conifer speciesTaxodium distichum from the Brandywine Formation of Maryland and the Citronelle Formation of southernAlabama are presented. These are important as the first such estimates from eastern North Americanpaleofloras, and provide new evidence from a time for which the role of CO2 in climate change is controversial.Comparisons of the stomatal density (SD) of the fossil leaf cuticles to a calibration curve constructed frommodern leaves of the same species collected over the last century of anthropogenic CO2 increase producesMiocene and Pliocene atmospheric paleo-CO2 mean estimates of 360 and 351 ppmv, respectively. Althoughthe temporal resolution of the fossil sites is low, these results are in agreement with multiple independentproxies that indicate near modern CO2 levels during this interval, and demonstrate the utility of T. distichumleaves as instruments for stomatal frequency analysis.

+1 251 414 8220.. Axsmith).

l rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

The use of fossil plant cuticle micromorphology as an atmosphericpaleo-CO2 proxy has become an important tool for understanding thepattern and causes of climate change throughout Phanerozoic history(Royer et al., 2001). Most of the original research demonstrating astomatal response to changing atmospheric CO2 in living plants(Woodward, 1987), as well as most fossil studies, involves C3

angiosperms; however, some gymnosperms have also been shownto be useful and exhibit certain advantages. For example, Ginkgo andMetasequoia fossils used to estimate Paleocene – Miocene paleo-CO2

appear identical to extant species (Royer et al., 2001), and these andmany other extant gymnosperm species probably evolved duringtimes of elevated CO2 (Haworth et al., 2010).

In this paper, we demonstrate that megafossil remains from theMioceneBrandywine FormationofMaryland and thePlioceneCitronelleFormation of southern Alabama – including leafy shoots, isolated leaves,pollen cones, isolated seed cone scales, and seeds – are assignable to theextant cupressaceous conifer species Taxodiumdistichum (Bald Cypress)(Fig. 1). Multiple organs support this determination, but the uniquearrangement of stomatal bands on the leaves is the most compellingfeature that allies these fossils with T. distichum as opposed to the otherextant Taxodium species and the similar extinct formT. dubium. Basedonthis finding, we provide the first estimates of atmospheric paleo-CO2

based on comparisons of the fossil leaf cuticles to a stomatal responsecurve derived from freshly collected and historical herbarium speci-mens of this species. This is noteworthy, as the late Neogene was animportant time of climate changewithmajor biotic responses includingthe C3 to C4 plant transition (Cerling et al., 1997). Furthermore, theCitronelle Formation flora existed during, or immediately following, thePliocenewarm interval,whichhas becomeamajor research focusdue toits relevance as a model for future climate change responses (Dowsettand Caballero, 2010; Stults et al., 2010).

1.1. Brief review of fossil and extant Taxodium

Fossils of Taxodium are first recognized from the Late Cretaceous inEurope and North America, and become widespread in these regionsduring the Paleogene and Neogene (Knobloch and Mai, 1986;Aulenback and LePage, 1998). Historically, the European specieshave been assigned to several extinct species; however, in anextensive review, Kunzmann et al. (2009) have determined that allof the Neogene records should be considered as variants of the singleextinct species T. dubium. Most of the Neogene records from NorthAmerica are also attributed to this species (e.g., Huggins, 1985). Sincethe Neogene, the distribution of Taxodium has greatly contracted. Onlythree extant species are recognized: T. distichum and T. ascendans(sometimes recognized as a variety of T. distichum) from the lowerAtlantic and Gulf Coastal Plain extending north to southern Illinoisalong the Mississippi Valley; and T. mucronatum from southernmostTexas, Mexico and Guatemala (Little, 1971).

Fig. 1. Representative Taxodium distichum fossils from the Citronelle Formation (A–D) and Brandywine Formation (E–F). A) Leafy shoots attached to stem. Scale bar=10 mm.B) Isolated deciduous shoot. Scale bar=5 mm. C) Isolated leaves. Scale bar=1 mm. D) Complete male shoot system with attached pollen cones. Scale bar=10 mm. E) Isolatedovulate cone scales. Scale bar=2 mm. F) Isolated seeds. Scale bar=1 mm.

328 D.Z. Stults et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 309 (2011) 327–332

T. distichum prefers subhumid to humid climates with abundantprecipitation and an extended growing season with mean annualtemperatures of 12–25 °C, and mean annual precipitation of approxi-mately 1000 mm (Little, 1971; Thompson et al., 1999). It growsprincipally below 30m above sea level, which is relevant to its utility inCO2 proxy studies as potentially confounding altitudinal effects onstomatal numbers are mitigated (Woodward and Bazzaz, 1988). It is along-lived tree (often greater than 1000 years), and can form extensive,pure stands inwet soils andperiodicallyflooded sites (Conner et al., 1986;Conner, 1988; Farjon, 2005). It is noteworthy in the context of this studythat the fossil localities occur within the current range of T. distichum.

T. distichum is deciduous with dimorphic shoots. The long shoots areindeterminate,while its short shoots are determinate and shed seasonally(Watson, in Flora of North America, 1993+). Leaves are narrow withtapered apices (approximately 1.5 mm wide, 5–19 mm long), linear tolanceolate, and flattened. Leaves are spirally arranged on the shoot andtwisted at the base so that the entire shoot occupies one plane (Godfrey,1988; Vidakovic, 1991; Watson, Flora of North America, 1993+).Abundant stomata are randomly oriented in two broad bands on theabaxial leaf surface, one on each side of the midvein. Fewer stomata arepresent in the narrower bands of the adaxial surface (Plate 2A). Manyworkers have described the stomata as occurring mostly in individualrows within the bands, although additional stomata do occur in groupsbetween rows (Bertrand, 1874; Mahlert, 1885; Coulter, 1889; Florin,1931; Cross, 1940; Farjon, 2005). Our observations of abundant extantand fossil material indicate that consistent rows of stomata areexceedingly difficult to recognize, as should be expected based on Cross's(1940) report of substantial intercalary growth during leaf ontogeny.When dealing practically with fossils of T. distichum from both localitiesdescribedhere, it is evenmoredifficult to recognize rows (Fig. 2B–C). Thishas important implications for the counting strategy and stomatal countsemployed in this study as detailed below.

1.2. Conifers in stomatal paleo-CO2 proxy research: The case forTaxodium stomatal density

In many angiosperm paleo-CO2 based studies, stomatal index(SI=percentage of stomatal density relative to stomatal density plusepidermal cell density) is often utilized in preference to stomataldensity (SD=total no. stomata/area), as it is usually less affected byvariables like water availability and leaf position potentially affectingleaf expansion and cell size (Salisbury, 1927; Ticha, 1982; Poole et al.,1996; Royer, 2001). SI has also been shown to be a robust proxy forchange in atmospheric CO2 concentration in the extant gymnospermsGingko and Metasequoia (e.g., Royer, 2003; Haworth et al., 2010). Inmore recent conifer based studies, true stomatal density per length(TSDL) has been the preferredmetric in species with stomatal rows orbands (e.g., Kouwenberg et al., 2003), but was not successfullyemployed in this study due mainly to sudden increases in the

bandwidths in samples from the most recent years (years 2002 andyounger). The suggestion that the number of stomatal rows could beused as a proxy for bandwidth (Kouwenberg et al., 2003) was nottaken due to the difficulty in accurately discerning stomatal rows,especially in the fossil specimens of T. distichum (Fig. 2B-C).

In this study, SD alone is used for several additional reasons. Coniferleaf development itself mitigates cell size issues, as stomatalmaturationgenerally occurs later than in angiosperms, and there is typically a fixedratio of stomatal/epidermal cell complexes (Esau, 1977; Croxdale, 2000;Kouwenberg et al., 2003). Epidermal cell size variation due to wateravailability and differential light exposure that can confound SD-basedstudies is also a relativelyminor consideration for T. distichum due to itsoccurrence in swamp habitats with very moist to submerged soils, andthe relatively open crown of this species (Thompson et al., 1999). Mostimportantly, SD results with several conifer species have shownsignificant responses to atmospheric CO2 change (Kouwenberg et al.,2003; Lammertsma et al., 2011), and a response is demonstrated herebased on historical herbarium specimen records (Fig. 3). Finally, SD hadto be used in this study, as it is not possible to consistently delimit theboundaries of the non-guard cell epidermal cells on the fossil Taxodiumleaves from both localities due to preservational factors and the delicatenature of the cuticles.

2. Materials and methods

2.1. Geological setting and fossil preparation

The Brandywine Formation was considered Pleistocene (Clark, 1915)until regional palynological investigations indicated a Late Miocene(~11.2 to 6.5 Ma) age (Pazzaglia et al., 1997). The discovery of fossilplants – including the T. distichum fossils used in this study – from atemporary exposure approximately 20 km southeast of Washington D.C.also support this older agedeterminationdue to thepresenceof abundanttaxa now restricted to Asia (McCartan et al., 1990). The BrandywineFormation was part of a braided stream system typical of those that stillcharacterize the Salisbury Embayment (Ward & Powars, 1989), and theflora accumulated in afluvial uplandenvironment, probablywithin a sub-channel intermittently reconnected to a main channel during floodingevents (McCartan et al., 1990). Forty-nine plant taxa were originallyidentified from the Brandywine flora, including a reference to Taxodiumc.f. “T. distichum.” Based on a qualitative consideration of the plantassemblage, climate during the time of deposition was warm temperatewith periods of extended rainfall (McCartan et al., 1990).

The Citronelle Formation extends along the northeastern Gulf ofMexico Coastal Plain from the Florida panhandle to eastern Texas. Theage of the formation has been a source of considerable controversy, butmore recent studies convincingly reaffirm the original Pliocenedesignation of Matson (1916), and indicate that deposition occurredsometime between 3.4 and 2.7 Ma (Otvos, 1997, 1998). The Citronelle

Table 1Modern Taxodium distichum samples. Atmospheric CO2 values from Mauna Loa andSiple Station ice core data (Neftel et al., 1994; Keeling et al., 2009).

Yearcollected

Collection locality CO2

2008 Fakahatchee Strand Preserve, FL; Threemile creek, Mobile, AL 3862004 “ 3782003 “ 3762002 “ 3731998 “ 3671988 USAM herbarium, Auburn, AL 3511977 USAM herbarium, Baldwin County, AL, shore of Tensaw River 3341974 USAM herbarium, Baldwin County, AL, shore of D'Olive Bay 3301970 USAM herbarium, Mobile County, AL, Chickasawbogue Creek 3261964 (LA)herbarium, Bossier Parish, Barksdale AFB, LA 3191953 (LA)herbarium, Lincoln Parish, Bank of Sugar Creek, LA 3131943 (LA)herbarium, St. Landry Parish, Little Teshe Bayou, LA 3081932 (LA)herbarium, St. Tammany Parish, Lake Ponchartrain, LA 3061909 (LA)herbarium, East Baton Rouge Parish, University Lake, LA 299

329D.Z. Stults et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 309 (2011) 327–332

depositional environment is mostly considered a braided river system,but considerable fine-grained flood plain clays and estuarine facies alsooccur (Otvos, 1997). The climate during deposition has been describedas similar to that of the region today – humid, sub-tropical with highannual precipitation totals (Berry, 1916; Stults, 2003), but quantitativeclimate estimates based on the flora are in progress. Nearly all of theplant fossils have been recovered from occasional thin, relativelyunoxidized clay lenses. Eighteen plant taxa, including T. distichum,were originally identified (Berry, 1916). Recent study of the CitronelleFormationhas documented additional taxa, including thefirst records ofmany species still occurring in the region, and regionally extinct formssuch as Begonia and Pterocarya (Stults and Axsmith, 2009, 2011).

Fossils from the Brandywine Formation locality and all CitronelleFormation sites occur in sandy clays and siltstones. Megafossils can bediscerned directly on the matrix surfaces with little or no preparation(Fig. 1A–B, D). As part of the original McCartan et al. (1990) study, someof the plant-bearingmatrixwasmacerated and screened formesofossils.Specimens of T. distichum cone scales, seeds, and leaves were picked outof the residue for use in this study (Fig. 1E–F). T. distichum fossils occur atfive Citronelle Formation sites in Alabama. Themost articulatedmaterialcomes from the Scarborough School locality (30° 43.74′N, 88° 8.57′W);unfortunately, those cuticles couldnot be cleared sufficiently for stomatalanalysis. Matrix from the Red Bluff locality (30° 20.61′ N, 87° 29.12′W),however, was successfully broken down in water and screened. Conescales, seeds, and leaves were recovered. Individual fossil leaves ofT. distichum from the Brandywine Formation and Citronelle Formation

440

420

400

380

360

340

320

300

280

260100 120 140 160 180 200 220 240 260 280 300

SD

CO

2 (p

pmv)

CO2 (ppmv)

SD

350

300

250

200

150

100280 320 360 400

Fig. 3. Modern training set for Taxodium distichum. Thick black line: Model for CO2

estimates based on linear regression through log-transformed stomatal density andmodern CO2 data for T. distichum (CO2=103.093− (0.2463⁎(LOG SDfossil))); r2=0.46; thinlines indicate root mean square error (RMSE) of 21.9 ppmv CO2. Inset: stomatal densityresponse from herbarium specimens to the historical CO2 range from 299 ppmv in 1909to 385 ppmv in 2008. Error bars are±1 standard deviation.

were placed in chloral hydrate and periodically checked for adequateclearingof themesophyll. The treated leaveswere then rinsedwithwaterseveral times, stained with saffranin, and mounted in glycerin jelly onglass slides for light microscopic study (Fig. 2B–C).

2.2. Extant leaf preparation and stomatal counting strategy

The extant T. distichum leaves used for calibration were obtainedfrom natural populations in Alabama, Florida, and Louisiana, andinclude freshly collected material and herbarium specimens datingback over a century (Table 1). Individual leaves were removed fromthe middle portion of deciduous branches and placed in 2% sodiumhypochlorite, which gently disintegrated the mesophyll tissue. Thetreated leaves were then rinsed with water several times, placed inPetri dishes, and carefully sliced open longitudinally to flatten out thecuticular envelopes. Any remaining mesophyll was removed with asmall brush or fine needle. Cleared cuticles were stained withsaffranin, and mounted in glycerin jelly on glass slides (Fig. 2A).

A total of 177 extant leaves were analyzed, and 1–3 fields/leaf(depending upon the integrity of the cuticle) were measured. Stomataldensity (SD) was measured using the 400× magnification of a NikonEclipse 400 microscope. Each counting field was 0.0375 mm2 in areafollowing established procedures (Beerling, 1999; Poole and Kürschner,1999). Only areas within stomatal bands on either side of the mid-veinand from the midsection of the needle were counted. Raw stomataldensity counts per field were converted to SD. Counts on either side ofthe mid-vein of twenty-four fossil leaves were counted for theBrandywine Formation and thirty were counted from the CitronelleFormation. These raw counts were also converted to SD with errorscalculated to 1 standard deviation.

3. Results and discussion

3.1. Taxodium distichum in the southeastern US Neogene

T. distichum was tentatively identified from the Brandywine Flora(McCartan et al., 1990), but no figures were presented. The materialexamined here verifies the original identification. None of theBrandywine fossils are articulated, but the recovered ovulate conescales, seeds, and leaves are all indistinguishable from those of theextant species. Berry (1916) identified articulated shoots, small piecesof axis bearing pollen cones, isolated seeds, and ovulate cone scalesfrom the Lamberts location near the type section of the CitronelleFormation in Alabama and provided figures. We have discoveredadditional material from the original localities and several new sites,including articulated branches of up to two orders, complete pollencone-bearing branches, seeds and ovulate cone scales. All of thismaterial is indistinguishable from the extant species (Figs. 1–2).

The ovulate cone scales from the Brandywine Formation andCitronelle Formation localities only rarely have prominent surfaceembellishments (i.e., excessive spines or thorns) (Fig. 1E), whichdistinguishes them from most T. dubium fossils – especially the olderspecimens (Kunzmann et al., 2009). The pollen cones are born onshort, crowded branches (Fig. 1D), which eliminates affinity with theextant species T. mucronatum (Schulz et al., 2005). However, the mostconvincing evidence for considering these fossils as T. distichum is theleaf cuticle morphology. In an exhaustive analysis of fossil and modernTaxodium leaves, Kunzmannet al. (2009) demonstrate that the stomatalarrangement on the leaves is consistent and species specific. All specieshave pairs of stomatal bands, but in T. dubium the adaxial surface istypically devoid of stomata, while in T. mucronatum there are equalnumbers of stomata on both leaf surfaces. In T. distichum, and theBrandywine and Citronelle fossils described here, the adaxial surfacetypically has narrow bands with relatively few stomata, and the abaxialsurface has wider bands with numerous stomata (Fig. 2).

Fig. 2. Representative Taxodium distichum leaf cuticles. A) Extant cuticular envelope spread out in one plane. Brackets indicate broad abaxial stomatal bands. (Scale bar=250 μm).B) Adaxial surface of fossil cuticle from the Brandywine Formation showing narrow stomatal band (bracket). Scale bar=100 μm. C) Abaxial surface of leaf from C showing broadstomatal band (bracket). Scale bar=100 μm.

330 D.Z. Stults et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 309 (2011) 327–332

The cuticular evidence conclusively demonstrates that the Bran-dywine and Citronelle fossils are representative of the extant speciesT. distichum, and the pollen and seed cone evidence is consistent withthis determination. The occurrence of the geologically young fossillocalities within the range of the extant species, along with evidenceof a similar habitat and associated taxa, provides additional support.Consequently, the paleo-CO2 estimates presented below should beconsidered particularly robust, as they are inferred from a moderncalibration data set derived from a living conspecific.

3.2. Stomatal density (SD) inference model and paleo-CO2 estimate

Analysis of recently collected and herbarium samples of T. distichumshows a negative correlation between SD and historical atmosphericCO2 (Fig. 3 inset). The response interval covers the period from 1909 to2008 and an approximately 87 ppmv increase in atmospheric CO2 from299 ppmv to 385 ppmv (Neftel et al., 1994; Keeling et al., 2009). Toaccount for the potential non-linear response of the SD to changingatmospheric CO2 and facilitate direct palaeo-CO2 estimates from fossilSD, both SD and CO2 values are log-transformed before fitting a linearresponse curve through the data sets (Wagner et al., 2004, 2005).Applying the statistical model (Fig. 3) to the SD measurements of theT. distichum leaves from the Brandywine Formation (mean SDmm-2=150±33) and Citronelle Formation (mean SDmm-2=166±42) esti-mates mean atmospheric pCO2 levels of 360 ppmv (1 std. dev. 343–383 ppmv) and 351 ppmv (1 std. dev. 333–378 ppmv), respectively.Both results are similar to modern values. The course temporalresolution of the fossil sites – especially the Brandywine Formation –

precludes precise comparisons, but these results are in agreement withother proxy data as discussed below.

3.3. Comparisons with other proxies

3.3.1. Late MioceneMany significant geological and paleoecological events occurred

during the Miocene that are particularly relevant to our understandingof modern ecosystems, such as the intensification of orogenies onseveral continents, the expansion of grasslands, and theMessinianCrisis(Hsu et al., 1973; Catlos et al., 1997; Cerling et al., 1997;Retallack, 2001).There is currently considerable controversy regarding the paleoclimatesignal from this time due to discordance among proxy methods andcurrent concepts of the CO2 – climate relationship. These factors, inaddition to the nearly five million-year uncertainty in the age of theBrandywine Formation, make precise comparisons of the Taxodiumpaleo-CO2 value difficult.

The pCO2 estimate of 360 ppmv reported here is close to currentlevels. This is in very good agreement with the estimates derived fromQuercus patraea (Sessile or Durmast Oak) SI-based studies from theTortonian (approximately 8–10 Ma)of theRhineEmbayment (Kürschneret al., 1996), and higher to slightly higher than the Late Miocene CO2

increase reportedby somestudiesbasedonalkenonedata, δ11 Bofmarinecarbonate, seawater pH, and marine carbonate mineralogy (Pagani et al.,1999; Pearson and Palmer, 2000; Demicco et al., 2003). This result alsoagrees with recent paleoclimate modeling experiments of the LateMiocene (11 to 7 Ma) that indicate probable CO2 concentrations between360 and 460 ppmv (Micheels et al., 2009).

331D.Z. Stults et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 309 (2011) 327–332

3.3.2. Late PliocenePliocene climate was mainly characterized by gradual global

cooling culminating in the onset of significant Northern Hemisphereglaciation. However, a period of sustained global warmth and climaticstability occurred during the Piacenzian Stage between ca. 3.28 and2.97 Ma. During this time, maximum sea level was 10 to 35 m abovepresent levels, and Antarctic ice was reduced (Zachos et al., 2001;Dwyer and Chandler, 2009; Williams et al., 2009). This warm intervalhas been intensively studied as a possible model for future projectedclimate; however, the cause is debated, with most researchersimplicating increased atmospheric CO2, orographic effects from LateCenozoic uplift, closure of the Isthmus of Panama leading to changesin meridional heat transport, or some combination of these factors(Crowley, 1996).

The Taxodium leaf-based paleo-CO2 estimate of 351 ppmv reportedhere is consistent with most other proxies. For example, Raymo et al.(1996) provided δ13C POM (particulate organic matter) derivedestimates ranging from 329 to 435 ppmv for periods between 2.91and 3.28 Ma. The Taxodium-based estimate is also close to SI data fromQuercuspetraea fossils indicatingCO2 levels duringperiodsof themiddleto Late Pliocene (approximately 3.2– 2.5 Ma) of about 340–370 ppmv(Kürschner et al., 1996). Recent high resolution boron and alkenoneestimates indicate a paleo-CO2 level of about 330–400 ppmv for thePliocene warm interval, with a rapid drawdown to 275–285 ppmvoccurring between 3.2 and 2.8 Ma (Pagani et al., 2009; Seki et al., 2010).Assuming the accuracy of the Taxodium-based estimate, these resultsindicate that theCitronelleflora existed prior to theCO2 drawdown. Thiswould constrain the age of the Citronelle Formation closer to the olderend of Otvos's (1998) original 2.7 – 3.4 Ma estimate – at least for thefossil plant-bearing exposures at Red Bluff.

4. Conclusion

The findings here clearly demonstrate that the cupressaceousconifer species T. distichum has been part of the flora of southeasternNorth America since at least the Late Miocene. Stomatal densitymeasurements of modern leaves of this species collected over the lastcentury show a negative correlation to atmospheric CO2 increaseduring that interval, which allows for mean estimates of Miocene andPliocene paleo-CO2 from the fossil leaves of 360 and 351 ppmv,respectively. Although the temporal resolution of the fossil sites maybe considered poor, these results are in agreement with multipleindependent proxies that indicate near modern paleo-CO2 levelsduring this interval, and demonstrate the utility of this species as aNeogene paleo-CO2 proxy. The consistency between CO2 estimatesfrom the North American gymnosperm Taxodium distichum and theEuropean angiosperm Quercus petraea demonstrates the integrity ofthe stomatal frequency inferred CO2 data.

Presently, Neogene fossil floras from eastern North American areexceedingly rare with generally low temporal resolution. However,the relatively recent discovery of fossil plants from the BrandywineFormation (McCartan et al., 1990), and new plant localities from theCitronelle Formation (Stults et al., 2010) suggests that this situation isa result of poor prospecting rather than an actual dearth of suitablematerial. It is hoped that the research presented here will stimulatemore systematic research into Neogene floras of eastern NorthAmerica, as T. distichum is likely to be found at more localities. Thiswill allow for greater stratigraphic resolution of southeasternNeogene deposits, and a better understanding of paleo-CO2 and itsrole in the evolution and possible future of the southeastern NorthAmerican flora.

Acknowledgments

Thanks to ScottWingand the staff of theNationalMuseumofNaturalHistory in Washington D.C. for facilitating access to the Brandywine

Formation fossils. Discussions with Bruce Tiffney regarding theBrandywine flora were indispensable. This research was supported byNSF grant EAR-0642032 (to BJA).

References

Aulenback, K.R., LePage, B.A., 1998. Taxodium wallisii sp. nov: first occurrence ofTaxodium from the upper Cretaceous. International Journal of Plant Sciences 159,367–390.

Beerling, D.J., 1999. Stomatal density and index: theory and application. In: Jones, T.P.,Rowe, N.P. (Eds.), Fossil Plants and Spores: Modern Techniques. Geological Society,London, UK, pp. 251–256.

Bertrand, C.E., 1874. Anatomie comparée des tiges et des feuilles les Gnétacées et lesConiféres. Ann. Sci. Nat. Bot. Ser V T 20.

Berry, E., 1916. The flora of the Citronelle Formation. USGS Professional Paper 98-L.Catlos, E.J., Harrison, T.M., Grove, M., Lovera, O.M., Yin, A., Kohn, M.J., Ryersen, F.J.,

LeFort, P., Upreti, B.N., 1997. Further evidence for Late Miocene reactivation of theMain Central Thrust (Nepal Himalaya) and the significance of the MCT-1. EosTransactions AGU 78 Fall meeting supplement, p. F651.

Cerling, T.E., Harris, J.M., MacFadden, B.J., Leakey, M.G., Quade, J., Eisenmann, V.,Ehleringer, J.R., 1997. Global vegetation through the Miocene and Pliocene. Nature389, 153–158.

Clark, W.B., 1915. The Brandywine Formation of the Middle Atlantic Coastal Plain.American Journal of Science 40, 499–506.

Conner, W.H., 1988. Natural and artificial regeneration of baldcypress [Taxodiumdistichum (L.) Rich.] in the Barataria and Lake Verret basins of Louisiana. Ph.D.Dissertation, Louisiana State University: Baton Rouge, Louisiana.

Conner, W.H., Toliver, J.R., Sklar, F.H., 1986. Natural regeneration of baldcypress[Taxodium distichum (L.) Rich.] in a Louisiana swamp. Forest Ecology andManagement 14, 305–317.

Coulter, S., 1889. Histology of the leaf of Taxodium. Botanical Gazette 14 (76–81), 101–107.Cross, G.I., 1940. Development of the foliage leaves of Taxodium distichum. American

Journal of Botany 27, 471–482.Crowley, T.J., 1996. Remembrance of things past: greenhouse lessons from the geologic

record. Consequences, the Nature and Implications of Environmental Change, 2.Croxdale, J.L., 2000. Stomatal patterning in angiosperms. American Journal of Botany 87,

1069–1080.Demicco, R.V., Lowenstein, T.K., Hardie, L.A., 2003. Atmospheric pCO2 since 60 Ma from

records of seawater pH, calcium, and primary carbonate mineralogy. Geology 31,793–796.

Dowsett, H.J., Caballero Gill, R.P., 2010. Pliocene climate. Stratigraphy 7, 106–110.Dwyer, G.S., Chandler, M.A., 2009. Mid-Pliocene sea level and continental ice volume

based on coupled benthic Mg/Ca palaeotemperatures and oxygen isotopes.Philosophical Transactions of the Royal Society Series A 367, 157–168.

Esau, K., 1977. Anatomy of seed plants. JohnWiley and Sons, New York, New York, USA.Farjon, A., 2005. A monograph of Cupressaceae and Sciadopitys. Royal Botanical

Gardens, KEW. Richmond, Surrey UK.Florin, R., 1931. Untersuchungen zur stammesgeschichte der coniferales und

cordaitales. Kungliga Svenska Vetenskapsakademiens Handlingar 10 (1), 1–588.Godfrey, R.K., 1988. Trees, Shrubs, and Woody vines of Northern Florida and Adjacent

Georgia and Alabama. The University of Georgia Press, Athens, Ga.Haworth, M., Heath, J., McElwain, J.C., 2010. Differences in the response sensitivity of

stomatal index to atmospheric CO2 among four genera of Cupressaceae conifers.Annals of Botany 105, 411–418.

Hsu, K.J., Ryan, W.B.F., Cita, M.B., 1973. Late Miocene desiccation of the MediterraneanSea. Nature 242, 240–244.

Huggins, L.M., 1985. Cuticular analysis and preliminary comparisons of some Mioceneconifers from Clarkia, Idaho. In: Smiley, J.C. (Ed.), Late Cenozoic history of thePacific Northwest. American Association for the Advancement of Science, SanFrancisco, pp. 21–126.

Keeling, R.F., Piper, S.C., Bollenbacher, A.F., Walker, J.S., 2009. Atmospheric CO2 recordsfrom sites in the SIO air sampling network. Trends: A Compendium of Data onGlobal Change. Carbon Dioxide Information Analysis Center, Oak Ridge NationalLaboratory, U.S. Department of Energy, Oak Ridge, Tenn., U.S.A.

Knobloch, E., Mai, D.H., 1986. Monographie der Fruchte und Samen in der Kreide vonMitteleurope. Rozpr. Ustr. Ust. Geol. 47, 1–219.

Kouwenberg, L.L.R., McElwain, J.C., Kurschner, W.M., Wagner, F., Beerling, D.J., Mayle,F.E., Visscher, H., 2003. Stomatal frequency adjustment of four conifer species tohistorical changes in atmospheric CO2. American Journal of Botany 90, 610–619.

Kunzmann, L., Kvacek, Z., Mai, D.H., Walther, H., 2009. The genus Taxodium(Cupressaceae) in the Palaeogene and Neogene of Central Europe. Review ofPalaeobotany and Palynology 153, 153–183.

Kürschner, W.M., van der Burgh, J., Visscher, H., Dilcher, D.L., 1996. Oak leaves asbiosensors of late Neogene and early Pleistocene paleoatmospheric CO2 concen-trations. Marine Micropaleontology 27, 299–312.

Lammertsma, E.I., de Boer, H.J., Dekker, S.C., Dilcher, D.L., Lotter, A.F., Wagner-Cremer,F., 2011. Reduced maximum stomatal conductance in Florida vegetation underanthropogenic CO2 increase. Proceedings of the National Academy of Science 108,4035–4040.

Little, Elbert L., Jr., 1971.AtlasofUnitedStates trees.Vol. 1. Conifers and importanthardwoods.U.S. Department of Agriculture. Miscellaneous Publication 1146.Washington DC. 9p.313maps.

Mahlert, A., 1885. Belträge zur Kenntnis der Anatomie der Laudbblätter der Coniferen,mit besonderer Berucksichtigung des Spaltöffnungs-Apparates. Bot. Centralblatt24, 243–249.

332 D.Z. Stults et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 309 (2011) 327–332

Matson, G.C., 1916. The Pliocene Citronelle Formation of the Gulf Coastal Plain. USGSProfessional Paper 98-L.

McCartan, L., Tiffney, B.H., Wolfe, J.A., Agar, T.A., Wing, S.L., Sirkin, L.A., Ward, L.W.,Brooks, J., 1990. Late Tertiary floral assemblage from upland gravel deposits of thesouthern Maryland Coastal Plain. Geology 19, 311–314.

Micheels, A., Bruch, A., Mosbrugger, V., 2009. Miocene climate modeling sensitivityexperiments for different CO2 concentrations. Palaeontologia Electronica 12, 20.

Neftel, A., Friedli, H., Moor, E., Lotscher, H., Oescher, H., Siegenthaler, U., Stauffer, B.,1994. Historical CO2 record from the Siple Station ice core. Trends: A Compendiumof Data on Global Change. Carbon Dioxide Information Analysis Center, Oak RidgeNational Laboratory, U.S. Department of Energy, Oak Ridge, Tenn., U.S.A.

Otvos, E.G., 1997. Northeastern Gulf Coastal Plain revisited Neogene and Quaternaryunits and events – old and new concepts. Gulf Coast Association of GeologicalSocieties Annual Meeting, New Orleans. New Orleans Geological Society Guide-book, Louisiana – Northwest Florida Field Trip. New Orleans Geological Society,pp. 20–36.

Otvos, E.G., 1998. Citronelle Formation, Northeastern Gulf Coastal Plain: Pliocenestratigraphic framework and age issues. Gulf Coast Association of GeologicalSocieties Transactions 48, 321–333.

Pagani, M., Freeman, K.H., Arthur, M.A., 1999. Late Miocene atmospheric CO2

concentrations and the expansion of C4 grasses. Science 285, 876–879.Pagani, M., Liu, Z., LaRiviera, J., Ravello, A.C., 2009. High climate sensitivity to

atmospheric carbon dioxide for the past 5 million years. Nature Geoscience 3,27–30.

Pazzaglia, F.J., Robinson, R.A.J., Traverse, A., 1997. Palynology of the Bryn MawrFormation (Miocene): insights on the age and genesis of Middle Atlantic marginfluvial deposits. Sedimentary Geology 108, 19–44.

Pearson, P.N., Palmer, M.R., 2000. Atmospheric carbon dioxide concentrations over thepast 60 million years. Nature 406, 695–699.

Poole, I., Kürschner, W.M., 1999. Stomatal density and index: the practice. In: Jones, T.P.,Rowe, N.P. (Eds.), Fossil Plants and Spores: Modern Techniques. Geological Society,London, pp. 257–260.

Poole, I., Weyers, J.D.B., Lawson, T., Raven, J.A., 1996. Variations in stomatal density andindex: implications for palaeoclimatic reconstructions. Plant, Cell & Environment19, 705–712.

Raymo, M.E., Gran, B., Horowist, M., Rau, G.H., 1996. Mid-Pliocene warmth: strongergreenhouse and stronger conveyor. Marine Micropaleontology 27, 313–326.

Retallack, G.J., 2001. Cenozoic expansion of grasslands and climatic cooling. Journal ofGeology 109, 407–426.

Royer, D.L., 2001. Stomatal density and stomatal index as indicators of paleoatmosphericCO2 concentrations. Review of Palaeobotany and Palynology 114, 1–28.

Royer, D.L., Berner, R.A., Beerling, D.J., 2001. Phanerozoic atmospheric CO2 change:evaluating geochemical and paleobiological approaches. Earth-Science Reviews 54,349–392.

Royer, D.L., 2003. Estimating latest Cretaceous and Tertiary atmospheric CO2 fromstomatal indices. In: Wing, S.L., Gingerick, P.D., Schmitz, B., Thomas, E. (Eds.),Causes and Consequences of Globally Warm Climates in the Early Paleogene, 369.Geological Society of America, Boulder, Colorado, pp. 79–93. Special Paper.

Salisbury, E.J., 1927. On the causes and ecological significance of stomatal frequency,with special reference to the woodland flora. Philosophical Transactions of theRoyal Society of London Series B 216, 1–65.

Schulz, C., Knopf, P., Stützel, T., 2005. Identification key to the Cypress family(Cupressaceae). Feddes Repertorium 116, 96–146.

Seki, O., Foster, G.L., Schmidt, D.N., Mackensen, A., Kawamura, K., Pancost, R.D., 2010.Alkenone and boron-based Pliocene pCO2 records. Earth and Planetary ScienceLetters 292, 201–211.

Stults, D.Z., 2003. Paleoecological analysis of the Citronelle Formation from thepaleoflora of three sites in southwest Alabama. M.S. thesis. University of SouthAlabama: Mobile, Alabama.

Stults, D.Z., Axsmith, B.J., 2009. Betulaceae from the Pliocene and Pleistocene ofsouthwest Alabama, southeastern United States. Review of Palaeobotany andPalynology 155, 25–31.

Stults, D.Z., Axsmith, B.J., 2011. First macrofossil record of Begonia (Begoniaceae).American Journal of Botany 98, 150–153.

Stults, D.Z., Axsmith, B.J., Liu, Y.-S., 2010. Evidence of white pine (Pinus subgenusStrobus) dominance from the Pliocene Northeastern Gulf of Mexico Coastal Plain.Palaeogeography, Palaeoclimatology, Palaeoecology 287, 95–100.

Thompson, R.S., Anderson, K.H., Bartlein, P.J., 1999. Atlas of relations between climateparameters and distribution of important trees and shrubs in North America –introduction and conifers. USGS Professional Paper 1650-A.

Ticha, I., 1982. Photosynthetic characteristics during ontogenesis of leaves. 7. Stomatadensity and sizes. Photosynthetica 16, 375–471.

Vidakovic, M., 1991. Conifers morphology and variationRevised English Edition.Graficki Zavod Hrvatske.

Wagner, F., Kouwenberg, L.L.R., van Hoof, T.B., Visscher, H., 2004. Reproducibility ofHolocene atmospheric CO2 records based on stomatal frequency. QuaternaryScience Reviews 25, 1947–1954.

Wagner, F., Dilcher, D.L., Visscher, H., 2005. Stomatal frequency responses in ahardwood swamp vegetation from Florida during 60 years continuous CO2

increase. American Journal of Botany 92, 690–695.Ward, L.W., Powars, D.S., 1989. Tertiary stratigraphy and paleontology, Chesapeake Bay

region, Virginia and Maryland. 28th International Geological Congress Field TripGuidebook T-216. American Geophysical Union, Washington DC.

Watson, F.D., Taxodium. In: Flora of North America Editorial committee, eds. 1993+.Flora of North America North of Mexico. 14+ vols. New York and Oxford. Vol. 2.

Williams, M., Haywood, A.M., Harper, E.M., Johnson, A.L.A., Knowles, T., Leng, M.J., Lunt,D.J., Okamura, B., Taylor, P.D., Zalasiewicz, J., 2009. Pliocene climate and seasonalityin North Atlantic shelf seas. Philosophical Transactions of the Royal Society Series A367, 85–108.

Woodward, F.I., 1987. Stomatal numbers are sensitive to increases in CO2 frompreindustrial levels. Nature 327, 617–618.

Woodward, F.I., Bazzaz, F.A., 1988. The responses of stomatal density to CO2 partialpressure. Journal of Experimental Botany 39, 1771–1781.

Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, andaberrations in global climate 65 Ma to Present. Science 292, 686–693.