Specia l i ssue:

4 t h Internat ional Biogeography Society Meet ing

Celebrating the diversity of biogeographical researchSpecial issue: International Biogeography Society, 4th biennialmeeting

David Nogues-Bravo and Carsten Rahbek

D. Nogues-Bravo (dnogues@bio.ku.dk) and C. Rahbek, Center for Macroecology, Evolution and Climate, Dept of Biology, Univ. ofCopenhagen, Universitetsparken 15, DK-2100, Copenhagen, Denmark.

Biogeography aims to understand the temporal and spatial distribution of life on Earth. Biogeographical research is aimednot only at describing where organisms live, at what densities, with whom, and how it all relates to the environmental andgeographical setting but also why this is so. The International Biogeography Society, IBS, is a young and vibrantinternational and interdisciplinary society contributing to the advancement of all studies of the geography of nature,including spatial ecology (). In January 2009, the 4th International Conference of theInternational Biogeography Society took place in Merida on the Yucatan Peninsula, Mexico. Ecography providedfinancial support, acting as the sponsor of the Symposium of Extinction Biogeography and contributing to studenttravel awards. In addition, Ecography was the officially designated journal for publishing some of the many excitingtalks and posters presented at the conference. All of the papers in this special issue of Ecography arose from the IBSconference. They have all been subject to external peer review, subsequent revision, and final editorial decisions ofacceptance/rejection.

The special issue starts with an article by the local organizers(Vazquez-Dominguez and Arita 2010) that provides anintroduction and overview of the biogeographic history ofthe Yucatan Peninsula, the setting of the conference beforeit delves into a series of 22 papers that represents thediversity of what constitutes biogeographical research in the21st century.

The first series of papers focuses on speciation, extinctionand migration as the three key principal forces that drive thedistribution of biological diversity. Understanding when,how and where new species arise is of fundamentalimportance to our basic understanding of biodiversity onEarth. Reconstructing the evolutionary history of the familyOriolidae by generating a molecular phylogeny based onboth nuclear and mitochondrial DNA sequence data,Jnsson et al. (2010) shed new light on how species in thisclade dispersed first from their Australian area of origin toAsia and then onwards to Africa before back-colonisingAsia and the Indonesian archipelago. The hypothesis thatdiversification rates are higher in active than in passivetectonic settings is explored in the paper by Badgley (2010),and Casner and Pyrcz (2010) show that speciation ofbutterflies in tropical mountain regions occurs primarilywithin elevational bands. Using a global database on theworlds amphibians, Hof et al. (2010) find an indication ofhistorical signals in the realized climatic niches of species.

Understanding past and current extinctions and theirspatio-temporal dynamics is of tremendous direct interest,but insight from such studies is also of importance inunderstanding the impacts of contemporary and futureglobal changes in land use and climate on species. SouthernEuropean peninsulas, for example, were traditionally recog-nized as glacial refugia where many species survived duringthe ice ages. In a study using species distribution modelsin a phylogeographical research framework, Vega et al.(2010) challenge this view by showing that it is plausible thatthe pygmy shrew had northern refugia during the LastGlacial Maximum. However, climate change was not theonly factor affecting global or local extinctions during theLate Quaternary. Humans were also a well-known factorcausing the extinction of species, mainly on islands, wherehumans have disrupted key ecosystem functions. To mini-mize the unwarranted effect of disrupted ecosystem func-tioning, Hansen et al. (2010) propose that humans shouldactively replace extinct taxa by introducing analogue taxawith presumed similar ecological functions as the extinctspecies. They illustrate this approach with some taxonsubstitution projects on islands using large tortoises asexamples. Also on islands, the dramatic extinction debtrevealed by Triantis et al. (2010) calls for better manage-ment, including the restoration and expansion of nativeforests. Species living at the top of mountains are like oceanic

Ecography 33: 209211, 2010doi: 10.1111/j.1600-0587.2010.06677.x

# 2010 The Authors. Journal compilation # 2010 Ecography

Subject Editor: Carsten Rahbek. Accepted 17 May 2010

209

IBS

SP

EC

IA

LIS

SU

E

islands in that they are also expected to be highly exposed toextinctions because there appears to be nowhere to migrateupwards when lower altitudes warm up as a consequence ofglobal warming. However, upward migrations of speciestracking climate change may not be the only possiblescenario; some species could go against the flow. Lenoiret al. (2010), focusing on the latter, discuss potentialmechanisms for unexpected downward range shifts ofmountain plant species under climate change. The impactof climate change on species distribution has traditionallybeen attempted by using species distribution modeling, butthe usefulness of this model approach may well be affected bythe uncertainty embedded in the climate models used toforecast future climatic conditions (Real et al. 2010).

Migration of species through evolutionary and ecologicaltime has profoundly shaped biogeographical patterns atdifferent scales, from populations to whole continentalbiotas. Using 240 datasets, Jenkins et al. (2010) show thatin the era of landscape genetics, isolation by distance stillmatters in modern population genetics. Migrations acrossbiogeographic boundaries such us the Great American BioticInterchange, have profoundly shaped current patterns ofbiological diversity in the New World. In a Special Featurewithin this special issue, introduced by Riddle and Hafner(2010), a small series of papers focus on understandingthe timing and the biological consequences of the GreatAmerican Biotic Interchange (Cody et al. 2010, Smithand Klicka 2010), on the vicariance processes in MiddleAmerica (Daza et al. 2010), and on the biogeographicpatterns across the Mexican Transition Zone (Morrone2010). These contributions provide novel results andillustrate fresh research venues to revisit traditional biogeo-graphical questions that are rooted in the research legacyof classical biogeographers such as Alfred Russel Wallace(Riddle and Hafner 2010).

Speciation, extinction and dispersal in interaction withthe dynamics of abiotic and ecological processes are tradi-tionally viewed as what determine current biogeographicalpatterns, including life history traits. Different approaches tostudy body size patterns and their drivers in Pacific islandbirds are explored by Olalla-Tarraga et al. (2010) and Boyerand Jetz (2010), respectively. These studies are followed bya study assessing factors thought to cause patterns in thegeographical distribution of African palm species (Blach-Overgaard et al. 2010), and a study assessing the interspecificrange size variability of butterflies in relation to life historytraits and geographic features of species distributions(Garcia-Barros and Romo Benito 2010).

Not only do the distribution of species and patterns ofdiversity vary in time and space, so do the derived andunderlying distributions of geographical ranges sizes ofspecies assemblages. Krabbe Borregaard and Rahbek(2010) highlight the potential of using range-diversity plotsfor generating and testing hypotheses about how generalecological processes shape the location and size of speciesranges and species richness. The study illustrates that muchis still to be learned concerning the causes of large-scalepatterns of species richness, a theme which is also thefocus of Kreft et al.s (2010) study on the global speciesrichness pattern of ferns and seed plants. They suggestedthat taxon-specific ecological and life-history traits play

an important role in defining global richness gradients.Another classic research area of biogeography is the relation-ship between richness and area. In the last paper of thespecial issue, Guilhaumon et al. (2010) has contributedas a Software note, an R-package that allows users to easilyimplement model selection and parameter estimation toassess uncertainties in speciesarea-relationship models.

This special issue illustrates the current convergence ofdifferent academic fields such as evolutionary biology, eco-logy, phylogeography, and global change biology within abiogeographic framework to explain large scale patternsof biological diversity. The holistic nature of biogeographyconstitutes both a challenge but also an exciting opportu-nity for inter-disciplinary research. In light of the ongoingspecies extinction crisis caused primarily by habitat altera-tion and global changes in land-use with the recent addedfocus on the impact of global changes in climate on bio-logical systems, a diverse research program is as importantand relevant as ever in the history of biogeographicalresearch. We hope that this special issue, presenting andpromoting presentations at the International BiogeographySocietys conference in 2009 as peer-reviewed scientificjournal papers, will contribute to a more thorough under-standing of life on Earth.

References

Badgley, C. 2010. Tectonics, topography, and mammaliandiversity. Ecography 33: 220231.

Blach-Overgaard, A. et al. 2010. Determinants of palm speciesdistributions across Africa: the relative roles of climate,non-climatic environmental factors, and spatial constraints. Ecography 33: 380391.

Boyer, A. G. and Jetz, W. 2010. Cross-species and assemblage-based approaches to Bergmanns rule and the biogeography ofbody size in Pacific island birds. Ecography 33: 369379.

Casner, K. L. and Pyrcz, T. W. 2010. Patterns and timingof diversification in a tropical montane butterfly genusLymanopoda (Nymphalidae, Satyrinae). Ecography 33:251259.

Cody, S. et al. 2010. The Great American Biotic Interchangerevisited. Ecography 33: 326332.

Daza, J. M. et al. 2010. Using regional comparative phylogeo-graphic data from snake lineages to infer historical processes inMiddle America. Ecography 33: 343354.

Garcia-Barros, E. and Romo Benito, H. 2010. The relationshipbetween geographic range size and life history traits: isbiogeographic history uncovered? A test using the Iberianbutterflies. Ecography 33: 392401.

Guilhaumon, F. et al. 2010. mmSAR: an R-package for multi-model speciesarea relationship inference. Ecography 33:420424.

Hansen, D. M. et al. 2010. Ecological history and latentconservation potential: large and giant tortoises as a modelsystem for taxon substitutions. Ecography 33: 272284.

Hof, C. et al. 2010. Phylogenetic signals in the climatic nichesof the worlds amphibians. Ecography 33: 242250.

Jenkins, D. G. et al. 2010. A meta-analysis of isolation by distance:relic or reference standard for landscape genetics? Ecography33: 315320.

Jnsson, K. A. et al. 2010. Phylogeny and biogeography ofOriolidae (Aves: Passeriformes). Ecography 33: 232241.

210

IBS

SP

EC

IA

LIS

SU

E

Krabbe Borregaard, M. and Rahbek, C. 2010. Dispersion fields,diversity fields and null models: uniting range sizes and speciesrichness. Ecography 33: 402407.

Kreft, H. et al. 2010. Contrasting environmental and regionaleffects on global pteridophyte and seed plant diversity. Ecography 33: 408419.

Lenoir, J. et al. 2010. Going against the flow: potentialmechanisms for unexpected downslope range shifts in awarming climate. Ecography 33: 295303.

Morrone, J. J. 2010. Fundamental biogeographic patterns acrossthe Mexican Transition Zone: an evolutionary approach. Ecography 33: 355361.

Olalla-Tarraga, M. A. et al. 2010. Cross-species and assemblage-based approaches to Bergmanns rule and the biogeographyof body size in Plethodon salamanders of eastern NorthAmerica. Ecography 33: 362368.

Real, R. et al. 2010. Species distribution models in climate changescenarios are still not useful for informing policy planning: an

uncertainty assessment using fuzzy logic. Ecography 33:304314.

Riddle, B. R. and Hafner, D. J. 2010. Integrating pattern withprocess at biogeographic boundaries: the legacy of Wallace. Ecography 33: 321325.

Smith, B. T. and Klicka, J. 2010. The profound influence of theLate Pliocene Panamanian uplift on the exchange, diversifica-tion, and distribution of New World birds. Ecography 33:333342.

Triantis, K. A. et al. 2010. Extinction debt on oceanic islands. Ecography 33: 285294.

Vazquez-Domnguez, E. and Arita, H. T. 2010. The Yucatanpeninsula: biogeographical history 65 million years in themaking. Ecography 33: 212219.

Vega, R. et al. 2010. Northern glacial refugia for the pygmyshrew Sorex minutus in Europe revealed by phylogeographicanalyses and species distribution modelling. Ecography 33:260271.

211

IBS

SP

EC

IA

LIS

SU

E

The Yucatan peninsula: biogeographical history 65 million yearsin the making

Ella Vazquez-Domnguez and Hector T. Arita

E. Vazquez-Domnguez (evazquez@ecologia.unam.mx), Inst. de Ecologa, Universidad Nacional Autonoma de Mexico, Ap. Postal 70-275,Ciudad Universitaria, Mexico DF, 04510, Mexico. H. T. Arita, Centro de Investigaciones en Ecosistemas, Universidad Nacional Autonomade Mexico, Antigua Carretera a Patzcuaro No. 8701, Col. Ex-Hacienda de San Jose de La Huerta, 58190, Morelia, Michoacan, Mexico.

The fourth biennial meeting of the International Biogeo-graphy Society (IBS) in Merida, Yucatan in January 2009represented a double opportunity for Mexican biologists.First, it fostered the integration of the large community ofMexican biogeographers with the activities of the IBS.Second, the meeting allowed us to welcome a large numberof delegates from distant parts of the world who were ableto visit what has been considered an obligate destination fornature lovers and cultural tourists alike: the Yucatanpeninsula.

As Edward O. Wilson pointed out, besides economicpower every country has two additional and importanttypes of wealth: cultural and natural. Cultural richness is anaturally embedded component of the Mexican way of life.It is manifested in the rich legacy of ancient Mesoamericancivilizations, in the remarkable diversity of human groups,indigenous languages and dialects, local customs and food,and also in the seamless integration of modernity withtradition that can be seen in every major city. This culturalwealth is paralleled by an amazing natural richness, bestillustrated by the countrys extremely high biologicaldiversity. Mexico is the only nation in the world containingthe totality of a continental border between two biogeo-graphic realms, the Nearctic and the Neotropical. Themixing of elements of these two regions across a highlyheterogeneous landscape is the perfect recipe for a mega-diverse country like Mexico.

Mexicos cultural and natural richness becomes rapidlyevident to any traveller to the Yucatan peninsula. Considerthe caves of the southern part of the state of Yucatan as anexample. In many of these caves, a casual visitor will noticea multitude of fossil seashells embedded in the walls andceiling. Looking down, she could find small pieces of Mayaceramics interspersed with the sediment, and perhaps even apiece of the tooth of a Pleistocene horse Equus conversidens.These three interesting elements are in fact separated byorders of magnitude of time (Fig. 1): the limestone with theshells is of Oligocene origin, ca 25 million yr old, the horsebecame extinct some 10 000 yr ago, and the piece ofceramic is around 800 yr old. Furthermore, a much larger

and older piece of evidence of a past event might be in frontof the visitor: many of the large sinkholes that punctuate thelandscape of the Yucatan are located along the rim of theChicxulub crater, a 180-km wide scar created by the impactof an enormous asteroid 65 million yr ago that is believed tohave caused the mass extinction event of the end of theCretaceous. Standing in front of this diverse mixing ofelements of various origins, one cannot help being amazedby the particularities of the geologic, evolutionary andcultural history of the Yucatan that have produced thepresent-day diversity of this unique part of Mexico.

In this introduction to the special section of paperspresented at the IBS meeting, we offer a brief overview ofthe biological and cultural features that make the Yucatanpeninsula such a special place. When choosing a Mexicanvenue for the IBS meeting, our first option was Merida, theWhite City, the peaceful and charming capital of thestate of Yucatan. What better place could it be for abiogeography meeting than atop a 180-km wide, 65 millionyr old crater that testifies one of the most spectacular eventsin the history of life on Earth?

65 million years of history

Chicxulub: the dinosaur connection

It can be safely stated that the biological history of theYucatan started, or at least was reset, 65 million yr ago(Fig. 1). The Cretaceous-Tertiary (KT) episode thathappened then is one of the so-called big-five extinctionevents in the history of life on Earth (Raup and Sepkoski1982, Alroy et al. 2008). The KT episode wiped out75% of all animal species, including entire clades such asnon-avian dinosaurs, ammonites, rudists and inoceramidbivalves (Marshall and Ward 1996). Current knowledgestrongly suggests that the KT event was triggered by thecollision of a 10-km asteroid with what is now the northernYucatan peninsula, producing a 100 million megatonexplosion that in an instant obliterated the geological

Ecography 33: 212219, 2010doi: 10.1111/j.1600-0587.2009.06293.x

# 2010 The Authors. Journal compilation # 2010 Ecography

212

IBS

SP

EC

IA

LIS

SU

E

profile of the region and extinguished all life in hundreds ofkilometres around.

When first proposed (Alvarez et al. 1980), the idea of anextraterrestrial object hitting the Earth 65 million yr agowas received with considerable scepticism. The hypothesiswas supported by the strong empirical evidence of a spike iniridium concentration in sediments 65 million yr old,exactly at the K-T transition. Because iridium is anextremely rare element on Earth but occurs in measurableconcentrations in extraterrestrial objects, the most plausibleexplanation for such a spike was a space object collidingwith the Earth, disintegrating in an explosion that dispersediridium-rich sediment all over the world. The Alvarez teamcalculated that the hypothetical asteroid or comet shouldhave measured ca 10 km, and should have produced a crater200 km in diameter. One of the problems with the theorywas that no crater of the right age and size was known at thetime.

Shortly after the Alvarez et al. (1980) paper waspublished, Allan Hildebrand and Stein Boynton developeda theoretical model for an asteroid impact as predicted byAlvarez and collaborators, and called for a search for the

missing crater. Unknown to Hildebrand and Boynton,evidence for a candidate crater fitting the theoreticaldescription had been found in the 1960s and 1970s duringoil exploration drillings financed by the Mexican oilcompany PEMEX. In 1981, Glen Penfield used thePEMEX data to describe an underwater crater north ofYucatan, but his report went unnoticed. Many years later, asPenfield and Hildebrand joined forces, the evidence wasfinally published in a scientific paper proposing an under-ground circular feature 180 km in diameter centred in thecoastal town of Chicxulub as a formal candidate for themissing crater of the KT impactor (Hildebrand et al. 1991).

Today, most scientists have accepted the idea that anextraterrestrial impact caused the KT mass extinction, andthat the Chicxulub crater is indeed the scar of that episode(Fig. 1, 2) (Schulte et al. 2010). Statistical analyses of thefossil record show that besides background extinctionthroughout most of the Cretaceous, there was a clearmass extinction of ammonites coinciding with the KTboundary (Marshall and Ward 1996). There is also physicalevidence of the short- and long-term effects of the collisionin areas adjacent to the peninsula and farther away: the

Figure 1. Time line of major events in the biogeographical history of the Yucatan peninsula. Periods of the Pre-Hispanic era are:Paleoindian, Archaic, Preclassic, Classic, and Postclassic. Note the logarithmic scale.

Figure 2. Yucatan peninsula (states of Yucatan, Campeche and Quintana Roo), with location of the Chicxulub crater (ring of cenotes)and Sierrita de Ticul-Loltun cave, and the limits of the Yucatan Peninsula Biotic Province.

213

IBS

SP

EC

IA

LIS

SU

E

100-million megaton explosion, mega-tsunamis with 300-m waves, wildfires hundreds of kilometres away fromground zero, massive forest destruction, global climatechange with a subsequent reduction of 80% in thephotosynthesis rate, and of course mass extinction of plantand animal species (Kring 2007). Recent analyses ofmaterial extracted from the site have dated the Chicxulubcrater at 0.3 million yr (Myr) before the KT horizon,casting doubt on the idea that the Chicxulub object was thesole detonator of the KT extinction (Keller et al. 2004).These findings, which suggest a more complex series ofevents, including even multiple impacts, have ignited a newround of controversy regarding the Chicxulub site.

A recent study has added a new and surprising twist tothe story (Bottke et al. 2007). Simulation models of thedynamics of the group of asteroids called the Baptistinafamily show that they could have originated with afragmentation of a large asteroid 160 Myr BP, perhapsdue to a collision with another object. According to themodel, the largest piece resulting from the fragmentation isthe present-day asteroid 298-Baptistina, which still residesin the asteroid belt. Smaller pieces were scattered and manyof them entered the inner Solar System. Bottke et al. (2007)speculate that the spectacular Tycho crater in the Moon isthe result of the collision of one of this Baptistina objects109 Myr BP. The Chicxulub crater could have beenproduced by another of the Baptistina objects that endedits 95 Myr pilgrimage with an explosive encounter withEarth 65 Myr BP. If this is correct, Tycho and Chicxulubcould be sister craters produced by an amazing sequenceof improbable events.

Under the sea: Cenozoic biogeography

For more than 100 Myr, from the Cretaceous until thePleistocene, numerous marine transgressions submerged thearea of what is now the Yucatan peninsula under warm

tropical waters. During this time limestone strata wereformed with the remains of ancient coral reefs and seashells,including the uppermost Miocene-Pliocene (242 Myr BP)Carrillo Puerto Formation, a 15-m thick deposit of almostpure calcium carbonate that surrounds the shallow portionsof present-day karst systems. Thus, the whole peninsula isbasically a large limestone slab, submerged for millions ofyears, that is slowly emerging from south to north andwhere older deposits are located near the base.

When the Baptistina object collided with Earth at theend of the Cretaceous, what is now the Yucatan peninsulawas a shallow coastal shelf at the southern extreme of NorthAmerica. For millions of years, south of this tip there was awide ocean separating North and South America, produ-cing the independent evolution of early New Worldmammals in splendid isolation (Simpson 1980) untilthe Panamanian land bridge connected the two land massesca 3.12.8 Myr BP (Fig. 3), triggering the great Americanbiotic interchange (GABI), a major mixing of biotas fromSouth and North America that shaped the high-leveltaxonomic composition of modern floras and faunas ofthe New World (Marshall et al. 1982, MacFadden 2006,Webb 2006).

Because the northern portion of the peninsula did notemerge until a few million years ago, the role of the Yucatanpeninsula in the evolution of the terrestrial faunas of theCaribbean region during the Cenozoic was probably minor.Dispersal and vicariant theories have been proposed for thecolonization of the Antilles and posterior in situ evolution(Davalos 2004, Hedges 2006, Ricklefs and Bermingham2008). Both types of hypotheses, however, call for a SouthAmerican origin for the major vertebrate clades in theAntilles, with arrival and isolation times varying fromtheory to theory but pointing to around 35 Myr ago(Iturralde-Vinent and MacPhee 1999). An alternativehypothesis for the origin of some vertebrate groups inthe Antilles is the existence of a connection between theYucatan peninsula and the islands at the beginning of the

Figure 3. Emerged landmasses during the Middle Eocene (4937 Myr BP), when North and South America were not connected (dottedline), and during the Pliocene (3.12.8 Myr BP) after the closure of the Isthmus of Panama (solid line), based on the model of Heinickeand collaborators (Fig. 4; Heinicke et al. 2007). Abbrevations: NA: North America, MA: Middle America, SA: South America, PA: ProtoAntilles, Cu: Cuba, BB: Bahama Bank, Ja: Jamaica, Hi: Hispanola, PR: Puerto Rico, LA: Lesser Antilles.

214

IBS

SP

EC

IA

LIS

SU

E

Cenozoic, with posterior isolation and vicariant evolution, amodel that fits data for cichlid fish (Chakrabarty 2006). Inany of the scenarios, biotic interchange between Yucatanand the Caribbean islands is thought to have been minimalthroughout most of the Cenozoic, accounting for thepresent-day low similarity between the vertebrate faunasof these two areas (Vazquez-Miranda et al. 2007).

Nevertheless, for certain groups the Yucatan peninsulaserved as a bridge for dispersal between Central Americaand the Caribbean islands. For example, the present-daydistribution of eleutherodactyline frogs is best explained bya model that includes several events of dispersal over waterfrom and to South America and from Cuba to Yucatanat different times of the Cenozoic (Heinicke et al. 2007).Similarly, the evolution of mormoopid bats probablyinvolved dispersal over water from the northern Neo-tropics to Central America and from there to the Antilles,most likely through the Yucatan peninsula (Davalos2006).

Interesting examples of evolution in the Yucatan faunacome from invertebrates inhabiting the fresh-water oranchihaline underground bodies of water. Shrimps of thegenus Typhlatya are represented in the peninsula bythree species occurring in fresh-water habitats. However,the divergence of this clade predates the origin of itspresent-day habitat, according to molecular data (Hunteret al. 2008). This result implies that the three species musthave originated in marine habitats (the original mediumof the genus) before the end of the Pliocene, when thefreshwater habitats started to form. Subsequently, thethree fully-formed species could have invaded the newhabitat.

The great American biotic interchange and theconfiguration of modern biotas

The emergence of the Panama isthmus and the subsequentgreat American biotic interchange, peaking at approximately2.8 Myr BP, marked the start of the processes that haveconfigured the modern floras and faunas of the Yucatanpeninsula and other Middle American regions (Fig. 1;MacFadden 2006). Before the closure of the Panamaisthmus, all Mexican mammal faunas were completely NorthAmerican in composition (Ferrusquia-Villafranca 2003,Webb 2006). Today, faunas of Middle America (Mexicoplus Central America) are a rich and complex mixture ofNorth and South American components, a clear evidence ofthe processes associated with the GABI.

North American mammalian faunas north of the Tropicof Cancer still consist mostly of elements of native families,with only a few South American components, such asopossums (Didelphidae), armadillo (Dasypodidae), a hand-ful of phyllostomid and mormoopid bats. In contrast, faunasof tropical Mexico are rich assemblages that include bothNorth and South American elements, some of which haveevolved into idiosyncratic Mesoamerican endemics. Somefamilies of North American origin (e.g. Tapiridae, Felidae,Sciuridae) are represented in the peninsula by speciestypically considered tropical (e.g. Bairds tapir Tapirusbairdii, jaguar Panthera onca and Deppes squirrel Sciurusdeppei). In addition, clades of South American origin

are represented by primates (black howler monkeyAlouatta pigra and Geoffroys spider monkey Ateles geoffroyi),marsupials, bats, cingulata and pilosa (armadillo Dasypusnovemcinctus and northern tamandua Tamandua mexicana)and hystricognath rodents (Central American agouti Dasy-procta punctata, and spotted paca Cuniculus paca).

Because of the geological history of the Yucatanpeninsula, present-day faunas of the northern part of thepeninsula are of recent origin, B2.8 Myr. With very fewexceptions, vertebrate faunas of northern Yucatan aresubsets of the fauna of the base of the peninsula (the Petenand adjacent areas). Bats of the state of Yucatan, forexample, represent a subset that cannot be distinguishedfrom random samples of the fauna of the base of thepeninsula, except that species with high dispersal capabilityare overrepresented in the northern fauna (Arita 1997).This suggests that faunas of the northern part of thepeninsula originated simply by dispersal of species from thesouth. The big-eared climbing rat Ototylomys phyllotis forexample, diverged and dispersed from South Americatoward Middle America coinciding in time with theGABI, and its present range includes the Yucatan peninsula(Gutierrez-Garca and Vazquez-Domnguez unpubl.).

An example involving vicariant and dispersal events isthe evolution of cantil pitvipers of the genus Agkistrodon(Parkinson et al. 2000). Phylogeographical studies showedthat the genus Agkistrodon originally occupied relativelytemperate habitats and evolved toward more tropical ones;the species Agkistrodon bilineatus, present now in theYucatan peninsula, shows a historical initial divergencebetween populations from the eastern and western coasts inMexico, with a posterior dispersal of one population to theYucatan peninsula through subhumid corridors alongnorthern Central America that diverged into a differentsubspecies, Agkistrodon bilineatus russeolus. These examplesshow that despite its apparent simplicity, the process ofconformation of the Yucatan fauna can have manyvariations that depend on the idiosyncratic features of thedifferent clades that are involved (Arita and Vazquez-Domnguez 2003).

As pointed out by Webb (2006), a major problem facedwhen studying the GABI is the lack of fossils of the rightage at the right place. We have fossils either from themiddle Miocene (well before the GABI) or from thePleistocene (after the important processes had happened).In the Yucatan peninsula, fossils come from cave deposits ofrecent origin such as those from the Loltun cave in thesouthern part of the state. Sixty-eight animal species in tenorders, 25 families and 52 genera have been recorded asfossils in the cave, ranging in time between 30 000 and500 yr (Arroyo-Cabrales and Alvarez 2003 and referencestherein). Among the mammals found in the Loltun depositsare seven extinct species, including Pleistocene horses Equusconversidens, saber-toothed cats Smilodon fatalis, wolfs Canisdirus, mastodonts Cuvieronius sp. and camels Hemiuauche-nia sp., together with the bat Desmodus draculae and themarsupial Marmosa lorenzoi (Arroyo-Cabrales and Alvarez2003). These extinctions probably were caused by changesin climate or by the arrival of humans, although there is nodirect evidence for either of these two factors.

215

IBS

SP

EC

IA

LIS

SU

E

The Holocene and Anthropocene

Evidence of human occupation of the Yucatan peninsula inthe Paleoindian period has been found in the Loltun caveand in Belize, where distinctive Clovis points have beenrecovered from deposits in Ladyville, that have been datedat 9000 to 7500 BC (Kelly 1993). Since then, the regionhas been populated by humans without interruption(Fig. 1). During the Classic period (250900 AD), a largeMaya city such as Tikal harboured populations of at least50 000 people, or perhaps even 100 000 according to someestimates, and when the Spaniards arrived in 1519, thehuman population in the peninsula was probably as high asit is today (Fig. 1). Many people, thinking always of thegreat cities of the Classic period, visualize the Maya as avanished human group, without realizing that they stillnumber in the millions all over southern and easternMexico and in Central America.

Just as in other parts of the world, the cycles ofdevelopment and decline of human groups in the Yucatanpeninsula have been closely tied to climate conditions andthe use of natural resources (Diamond 2005). There ismounting evidence of significant climate shifts in thepeninsula associated with global conditions. For example,important sea level changes in the Yucatan peninsula 121kyr ago, during the last interglacial period have beendocumented by analyzing fossil reefs of the peninsula(Blanchon et al. 2009). The study showed changes in sealevel of up to 3 m in ecological time, perhaps as fast as ina few decades. Paleobotanical studies have documenteddramatic changes in the vegetation cover of the peninsula.Only four thousand years ago, the Peten region was warmerand much dryer than it is today, and extensive savannahsexisted in what is now covered with tropical rainforest.Forests began to dominate the landscape only about2500 yr ago. In eastern Middle America, including sitesin the Yucatan peninsula, the period with the densesttropical forests and deepest lakes coincides with the so-called Little Ice Age, 13501850 AD (Lozano-Garca et al.2007).

In recent years, several studies have shown a strongcorrelation between changes in climate and the demise ofthe Classic Maya city-states. These studies are made possiblenot only by modern techniques that allow the reconstruc-tion of past climates, but also by the precise calendar (thelong count) that the Classic Mayas used to recordimportant events (Sharer and Traxler 2005). In everyimportant city, steles were erected every 19 yr and 265 dto mark the start of a new katun (period of 7200 days). Inthe year 790, at least 45 such monuments were built,but 100 yr later only a dozen were produced, and on15 January, 909, a sole stele was carved in the city ofTonina, in the highlands of Chiapas. This decline in theelaboration of monuments testifies the fall of each of themajor Maya city-states of the Classic period: Palenque andYaxchilan, in the Usumacinta river basin were abandonedfirst, at the beginning of the 9th century. Then, cities ofpresent-day Belize and Guatemala followed suit anddisintegrated by 860. Finally, the mega-metropolis of thePeten, such as Tikal and Calakmul were deserted before910. This sequence is important because it points to the fact

that there was not a single collapse of the Classic Maya, buta series of events that took almost 100 yr to develop.

In 1995, a study of stable oxygen isotopes (d18O) fromsediments in a lake in Quintana Roo showed the existenceof important dry spells coinciding with the end of theClassic period (Hodell et al. 1995), suggesting the idea thata megadrought could have triggered the fall of the cities.Years later, a study of the sediments of the Cariaco basin inSouth America, which allow the estimation of year by yearrainfall patterns, demonstrated the existence not of a singleevent, but a series of extremely dry periods correspondingwith the end of the Classic (Haug et al. 2003). Even more,the driest years (dated at 760, 810, 860 and 910 AD)coincided with the sequence of abandonment of the mainMaya sites. This result points to severe drought as one ofmany possible causes of the collapse of Classic Maya city-states. New data on d18O from the northern peninsulashows that the 15th century abandonment of somePostclassic sites, such as Mayapan, also coincides with aparticularly harsh dry spell (Hodell et al. 2007), and recentevidence from several camps corroborates the megadroughttheory (Pringle 2009). Of course, the so-called collapse ofthe Classic Mayas was a very long and complex process thatinvolved other environmental, social, political and religiousfactors as well.

On 24 March, 1519, a new type of biogeographicalprocess took place in the coast of Tabasco. Sixteen horsesthat arrived with the army of Hernan Cortes became thefirst animals introduced by Europeans into continentalNorth America (Fig. 1). It is ironic that horses, which hadevolved in North America only to become extinct there atthe end of the Pleistocene (MacFadden 2006), gave a smallband of a few hundred Spaniards the leverage to vanquishthe powerful Aztec empire. After the Pleistocene extinc-tions, the native Middle American fauna lacked largemammals suitable for human use, so big domesticatedanimals were totally unknown to Mesoamerican Indians.Horses caused a tremendous impression on natives, becom-ing one of the most powerful weapons of the conquistadors.They were also the first in a long list of plants and animalsintroduced by Europeans that changed the structure andfunctioning of many ecosystems.

In the other direction, many native crops of the Yucatanpeninsula were exported to the rest of the world. Two plantsin particular played important roles in the configuration ofthe modern landscape of the Yucatan. The tapped sap of thesapodilla tree Manilkara zapota was used since pre-Hispanictimes to produce a gum (the chicle) that could be chewed.From the 1870s, when the chewing gum was introduced tothe United States until the mid 1940s, the increasingdemand for natural sapodilla gum was so big that it fosteredthe exploration of the forests of southern Yucatan in searchof more trees to be exploited. These explorations con-tributed to the finding of many Maya ruins (Sharer andTraxler 2005), and catapulted the economic developmentof the whole area. After the invention of artificial substitutesfor chicle in the 1940s, the demand for the natural productplummeted. Today, chicle is harvested only for specializedmarkets, mostly in Asia, that still prefer chewing gum basedon natural products.

216

IBS

SP

EC

IA

LIS

SU

E

The other plant that drove Yucatans economy for a longtime was henequen Agave fourcroydes. During the 19thcentury, demand for henequen or sisal fibers soared. At onepoint, up to 85% of the fiber used worldwide came fromYucatan, making the state one of the more affluent regionsof Mexico by the 1880s. It was at this time that the hugehaciendas in Yucatans countryside flourished and theluxurious houses in Merida were built. Unfortunately forthe environment, extensive plantations of the agave plantruined hundreds of thousands of hectares, in an area wherepoor soil makes very difficult the cultivation of other crops.After World War I, the development of artificial fibres andthe competition from countries that had smuggled hene-quen plants out of Yucatan and were producing their ownfibre marked the end of the green gold boom in thepeninsula.

The chicle tree and the henequen plant are only twoexamples of the complicated processes involving the use ofnatural resources in the Yucatan. Todays Yucatan environ-ments are the result of millions of years of evolution, butalso of the direct interaction with humans within the past10 000 yr. Even within Biosphere reserves, such as in theCalakmul area in the southern part of the peninsula, thelandscape is a complex matrix of natural and human-modified environments whose intermingling determines thedynamics of the rich biological diversity of the region(Vester et al. 2007).

Present-day Yucatan

The geographic setting

The ecological features and biogeography of the present-dayYucatan peninsula show the indelible mark of its 65 millionyr history, as well as the evident effect of modern humanactivity. Politically, the peninsula comprises the entireterritory of the states of Campeche, Yucatan and QuintanaRoo. From a geomorphological point of view it also includesBelize, the Peten area of Guatemala and small portions of theMexican states of Chiapas and Tabasco (Fig. 2). Thelimestone bedrock determines a terrain that is typical karst,dominated by a low and relatively flat plain of porouslimestone with little soil. The highest point in the north isonly 250 m (750 ft) in the Sierrita de Ticul. Surface water, inthe form of small lakes and rivers, is confined to the southernpart of the peninsula. In the north, all water reservoirs areunderground, where there is a complex freshwatersaltwaterinterface (Escolero et al. 2007). The karst is also characterizedby a large number of caves and cenotes (water-filledsinkholes) such as those at the rim of the Chicxulub crater(Perry et al. 1995, Schulte et al. 2010), that provide uniquehabitats for plants and animals (Arita 1996, MacSwiney et al.2009, Vazquez-Domnguez et al. 2009).

Most of the region is warm and subhumid, but climatefollows a pattern from dry in the north-northwest of thepeninsula to very humid in the south-southeast. Tempera-ture and rainfall vary from high mean annual temperatures(268C) and low annual rainfall (500 mm) in the northwestto lower temperatures and more abundant rainfall inthe southeast (14002000 mm; Orellana et al. 2003).

Throughout most of the peninsula there is a very welldefined rainy season from June to October, although winterrains are not uncommon in the south. Proximity to theTropic of Cancer and the influence on the region of theAtlantic Bermuda-Azores anticyclone create both a highatmospheric activity and a strong north to south gradient ofatmospheric pressure. This, together with the effect of tradewinds and the influence of tropical perturbations allow theformation of hurricanes, a defining climatic feature of thewhole Caribbean (Orellana et al. 2003). Because of itsposition, the Yucatan is hit harder and with higherfrequency by hurricanes on the east coast, contributing tothe east-to-west gradient of humidity that determines thephysiognomy, phenology and structure of the vegetation ofthe peninsula. Hence, vegetation also follows a SE-NWgradient, from tropical rainforests in the Peten to tropicalscrubland in the extreme NW portion of the peninsula.Extensive areas between these two extremes were originallycovered with deciduous or semideciduous tropical forests(Carnevali et al. 2003).

Diversity patterns and conservation

The definition of biogeographic provinces is commonlybased on the homogeneous distribution of the biota of aregion, compared to that of adjacent areas. Early surveys ofthe Yucatan demonstrated a Neotropical affinity for its floraand fauna, evidenced by their composition. In particular,the Yucatan fauna is similar to that of other tropical dryzones, but it differs due to the presence of elements fromthe more humid areas of the south. This distinctiveness hasprompted most scholars to consider the Yucatan peninsula abiotic province on its own, the Yucatan Peninsula BioticProvince (Fig. 2; Goldman and Moore 1945) or a regionwith two provinces, the Peten province in the south and theYucatan province in the north. The former scheme hasreceived much more support from recent analyses ofgeological and physiographic features, and of the distribu-tion of plants, birds and mammals (Fa and Morales 1993,Morrone 2005).

Vascular plants in the Yucatan peninsula are very diverse,reaching 26003000 species. The six most commonfamilies represent ca 41% of the total flora of the region,including Fabaceae, Poaceae and Orchidaceae. However,species richness in the Yucatan is lower than that ofcomparable Neotropical regions of similar size. This factresults basically from the fairly recent origin of thepeninsula, its relatively dry flat terrain, and its lack ofsuperficial water, all of which preclude the presence of themany different microclimates and local heterogeneity thatare typical of other Neotropical zones (Carnevali et al.2003). Approximately 7% of the flora is endemic to theYucatan peninsula, with some very distinctive, conspicuousand even dominant species, such as Acacia gaumeri(Fabaceae) and Myrmecophila christinae (Orchidaceae).Four genera are restricted to the province: Golmanella,Harleya, Plagiolophus and Asemnantha. Endemic speciesfollow a particular distribution pattern, geographicallydivided in three parts: a northern belt, with speciessuch as Ipomea sororia (Convolvulaceae) and Mammilaria

217

IBS

SP

EC

IA

LIS

SU

E

heyderi spp. gaumeri (Cactaceae); a southern zone withdominant species like Golmanella sarmentosa (Celastraceae),Epidendrum martinezii (Malvaceae) and Maytenus schippii(Orchidaceae); and a widespread component includingspecies like Hampea trilobata (Malvaceae) or Acacia gaumeri(Carnevali et al. 2003).

Vertebrate assemblages in the Yucatan peninsula are richin species, but less diverse than comparable regions incentral and western Mexico (Lee 1980, Arita 1997,Zambrano et al. 2006). Nonetheless, high phylogeneticand taxonomic diversity characterise the region whencompared with some areas of Central and tropical SouthAmerica and Africa (Schipper et al. 2008). Few endemics,low richness of restricted species and higher representationof wide ranging species are also noticeable patterns (Aritaand Rodriguez 2002, Arita and Vazquez-Domnguez 2003,Schipper et al. 2008). Finally, another defining feature ofthe peninsula is its low beta diversity when compared withother regions in Mexico (Arita and Rodriguez 2002). This isa consequence of the everyone is everywhere distributionpattern, shown most clearly by mammals, which in turn isthe result of the peninsulas simple topography, lack ofgeographical barriers and low habitat heterogeneity.

Simpsons peninsula effect a decrease in speciesdiversity from the base to the tip of peninsulas is clearlyobserved in the Yucatan peninsula (Simpson 1964). This ismore evident for mammals, which vary in number fromaround 130 species in the base to 90 in the tip; for frogs,with 22 species in the base and nine in the north (Lee1980), and for bats, with 85 species present in the southand only 31 in the north (Arita 1997). Likewise, the florafollows a conspicuous diversity pattern along the SE-NWrain gradient; the humid communities in the south havingmore species than their northern counterparts. An exceptionis seen for snakes and lizards, which are less diverse at thecentre of the peninsula, and increase their richnesss towardsthe tip (Lee 1980).

Endemics include ca 20 reptiles, seven birds and 10mammals. Richness of endemic amphibian and reptilianspecies follows an inverse peninsula pattern in whichmore endemic species occur in the north than in the south.In contrast, there is no distinctive gradient of richness ofendemic mammals and birds, most of which are widelydistributed within the peninsula (Arita and Vazquez-Domnguez 2003). Many endemic mammals, for instance,are distributed all over the peninsula and sometimesmarginally to the piedmont of the highlands of Chiapasand Guatemala. Species showing this pattern include theYucatan yellow bat Rhogeessa aeneus, the Yucatan squirrelSciurus yucatanensis, Hatts vesper rat Otonyctomys hatti, andthe Yucatan black howler monkey Alouatta pigra.

All these particular biogeographic traits make theYucatan an important place for conservation strategiesdespite the moderate absolute species richness of the region.For example, parts of the Yucatan have been identified aspriority areas for the conservation of trees of the tropicaldeciduous forest (Cue-Bar et al. 2006) and carnivores(Valenzuela-Galvan and Vazquez 2008). Likewise, theYucatan peninsula is a hotspot for endemic helminthparasites of freshwater fishes (Aguilar-Aguilar et al. 2008).

Conclusion

Biogeography is by necessity a historical science, in the sensethat present-day patterns of diversity and distribution ofspecies cannot be understood without considering thegeological and evolutionary history of the region. In todaysYucatan, patterns are the result of a wide variety of processesthat have shaped the environments of the peninsula atdifferent time scales, from 65 million yr to a few decades.Moreover, a full understanding of those processes isnecessary to face the present and future conservationchallenges posed by the complex intermixing of naturaland social elements that characterize the peninsula. AsDanish philosopher Soren Kierkegaard once wrote, life canonly be understood backwards but it must be livedforwards.

References

Aguilar-Aguilar, R. et al. 2008. Richness and endemism ofhelminth parasites of freshwater fishes in Mexico. Biol. J.Linn. Soc. 94: 435444.

Alroy, J. et al. 2008. Phanerozoic trends in the global diversity ofmarine invertebrates. Science 321: 97100.

Alvarez, L. W. et al. 1980. Extraterrestrial cause for theCretaceous-Tertiary extinction. Experimental results and the-oretical interpretation. Science 208: 10951108.

Arita, H. T. 1996. The conservation of cave-roosting bats inYucatan, Mexico. Biol. Conserv. 76: 177185.

Arita, H. T. 1997. Species composition and morphologicalstructure of the bat fauna of Yucatan, Mexico. J. Anim.Ecol. 66: 8397.

Arita, H. T. and Rodriguez, P. 2002. Geographic range, turnoverrate and the scaling of species diversity. Ecography 25: 541550.

Arita, H. T. and Vazquez-Domnguez, E. 2003. Fauna y laconformacion de la provincia biotica yucateca: Biogeografa ymacroecologa. In: Colunga-Garca Marn, P. and Larque-Saavedra, A. (eds), Naturaleza y sociedad en el area maya,pasado, presente y futuro. Academia Mexicana de Ciencias,CICY, pp. 6080.

Arroyo-Cabrales, J. and .Alvarez, T. 2003. A preliminary report ofthe late Quaternary mammal fauna from Lotun Cave,Yucatan, Mexico. In: Schubert, B. W. et al. (eds), Ice Agecave faunas of North America. Indiana Univ. Press and DenverMuseum of Nature and Science, pp. 262272.

Blanchon, P. et al. 2009. Rapid sea-level rise and reef back-stepping at the close of the last interglacial highstand. Nature458: 881885.

Bottke, W. F. et al. 2007. An asteroid breakup 160 Myr ago as theprobable source of the K/T impactor. Nature 449: 4853.

Carnevali, G. et al. 2003. Flora y vegetacion de la pennsula deYucatan. In: Colunga-Garca Marn, P. and Larque-Saavedra, A. (eds), Naturaleza y sociedad en el area maya,pasado, presente y futuro. Academia Mexicana de Ciencias,CICY, pp. 5368.

Chakrabarty, P. 2006. Systematics and historical biogeographyof Greater Antillean Cichlidae. Mol. Phylogenet. Evol. 39:619627.

Cue-Bar, E. M. et al. 2006. Identifying priority areas forconservation in Mexican tropical deciduous forest based ontree species. Interciencia 31: 712719.

Davalos, L. M. 2004. Phylogeny and biogeography of Caribbeanmammals. Biol. J. Linn. Soc. 81: 373394.

218

IBS

SP

EC

IA

LIS

SU

E

Davalos, L. M. 2006. The geography and diversification in themormoopids (Chiroptera: Mormoopidae). Biol. J. Linn.Soc. 88: 101118.

Diamond, J. M. 2005. Collapse: how societies choose to fail orsucceed. Viking.

Escolero, O. et al. 2007. Dynamic of the freshwatersaltwaterinterface in the karstic aquifer under extraordinary rechargeaction: the Merida Yucatan case study. Environ. Geol. 51:719723.

Fa, J. E. and Morales, L. E. 1993. Patterns of mammalian diversityin Mexico. In: Ramamoorthy, R. et al. (eds), Biologicaldiversity of Mexico: origins and distribution. Oxford Univ.Press, pp. 319361.

Ferrusquia-Villafranca, I. 2003. Mexicos middle Miocene mam-malian assemblages: an overview. In: Flynn, L. (ed.),Vertebrate fossils and their context: contributions in honorof Richard H. Tedford. American Museum of NaturalHistory, pp. 321347.

Goldman, E. A. and Moore, R. T. 1945. The biotic provinces ofMexico. J. Mammal. 26: 347360.

Haug, G. H. et al. 2003. Climate and the collapse of Mayacivilization. Science 299: 17311735.

Hedges, S. B. 2006. Paleogeography of the Antilles and origin ofWest Indian terrestrial vertebrates. Ann. Mo. Bot. Gard. 93:231244.

Heinicke, M. P. et al. 2007. Major Caribbean and CentralAmerican frog faunas originated by ancient oceanic dispersal. Proc. Nat. Acad. Sci. USA 104: 1009210097.

Hildebrand, A. R. et al. 1991. Chicxulub crater: a possibleCretaceous/Tertiary boundary impact crater on the YucatanPeninsula, Mexico. Geology 19: 867871.

Hodell, D. A. et al. 1995. Possible role of climate in the collapse ofClassic Maya civilization. Nature 375: 391394.

Hodell, D. A. et al. 2007. Climate and cultural history ofthe northeastern Yucatan Peninsula, Quintana Roo, Mexico. Clim. Change 83: 215240.

Hunter, R. L. et al. 2008. Phylogeny and historical biogeographyof the cave-adapted shrimp genus Typhlatya (Atyidae) in theCaribbean Sea and western Atlantic. J. Biogeogr. 35: 6575.

Iturralde-Vinent, M. A. and MacPhee, R. D. E. 1999. Paleogeo-graphy of the Caribbean region: implications for Cenozoicbiogeography. Bull. Am. Mus. Nat. Hist. 238: 195.

Keller, G. et al. 2004. Chicxulub impact predates the K-Tboundary mass extinction. Proc. Nat. Acad. Sci. USA 101:37533758.

Kelly, T. C. 1993. Preceramic projectile-point typology in Belize. Ancient Mesoamerica 4: 205227.

Kring, D. A. 2007. The Chicxulub impact event and itsenvironmental consequences at the Cretaceous-Tertiaryboundary. Paleogeogr. Paleoclimatol. Paleoecol. 255: 421.

Lee, J. C. 1980. An ecogeographic analysis of the herpetofauna ofthe Yucatan Peninsula. Univ. Kansas Misc. Publ. 67: 175.

Lozano-Garca, M. S. et al. 2007. Tracing the effects of the LittleIce Age in the tropical lowlands of eastern Mesoamerica. Proc. Nat. Acad. Sci. USA 104: 1620016203.

MacFadden, B. J. 2006. Extinct mammalian biodiversity ofthe ancient New World tropics. Trends Ecol. Evol. 21:157165.

MacSwiney, M. C. et al. 2009. Insectivorous bat activity atcenotes in the Yucatan Peninsula, Mexico. Acta Chiropt. 11:139147.

Marshall, C. R. and Ward, P. D. 1996. Sudden and gradualmolluscan extinctions in the latest Cretaceous of westernEuropean Tethys. Science 274: 13601363.

Marshall, L. G. et al. 1982. Mammalian evolution and the GreatAmerican Interchange. Science 215: 13511357.

Morrone, J. J. 2005. Hacia una sntesis biogeografica de Mexico. Rev. Mex. Biodivers. 76: 207252.

Orellana, R. et al. 2003. Presente, pasado y futuro de los climas dela pennsula de Yucatan. In: Colunga-Garca Marn, P. andLarque-Saavedra, A. (eds), Naturaleza y sociedad en el areamaya, pasado, presente y futuro. Academia Mexicana deCiencias, CICY, pp. 3752.

Parkinson, C. L. et al. 2000. Phylogeography of the pitviper cladeAgkistrodon: historical ecology, species status, and conservationof cantils. Mol. Ecol. 9: 411420.

Perry, E. et al. 1995. Ring of cenotes (sinkholes), northwestYucatan, Mexico its hydrogeologic characteristics andpossible association with the Chicxulub impact crater. Geology 23: 1720.

Pringle, H. 2009. A new look at the Mayas end. Science 324:454456.

Raup, D. M. and Sepkoski, J. J. 1982. Mass extinctions in themarine fossil record. Science 215: 15011503.

Ricklefs, R. E. and Bermingham, E. 2008. The West Indies as alaboratory of biogeography and evolution. Phil. Trans. R.Soc. B 363: 23932413.

Schipper, J. et al. 2008. The status of the Worlds land and marinemammals: diversity, threat, and knowledge. Science 322:225230.

Schulte, P. et al. 2010. The Chicxulub asteroid impact and massextinction at the CretaceousPaleogene Boundary. Science327: 12141218.

Sharer, R. and Traxler, L. 2005. The ancient Maya, 6th ed. Stanford Univ. Press.

Simpson, G. G. 1964. Species density of North American recentmammals. Syst. Zool. 12: 5773.

Simpson, G. G. 1980. Splendid isolation: the curious history ofSouth American mammals. Yale Univ. Press.

Valenzuela-Galvan, D. and Vazquez, L. B. 2008. Prioritizing areasfor conservation of Mexican carnivores considering naturalprotected areas and human population density. Anim.Conserv. 11: 215223.

Vazquez-Domnguez, E. et al. 2009. Contrasting genetic structurein two codistributed freshwater fish species inhabiting highlyseasonal systems. Rev. Mex. Biodivers. 80: 181192.

Vazquez-Miranda, H. et al. 2007. Biogeographical patterns of theavifaunas of the Caribbean Basin Islands: a parsimonyperspective. Cladistics 23: 180200.

Vester, H. F. M. et al. 2007. Land change in the southern Yucatanand Calakmul Biosphere Reserve: effects on habitat andbiodiversity. Ecol. Appl. 17: 9891003.

Webb, S. D. 2006. The great American biotic interchange:patterns and processes. Ann. Mo. Bot. Gard 93: 245257.

Zambrano, L. et al. 2006. Fish community structure in freshwaterkarstic wetlands of the Yucatan peninsula, Mexico. Ichthyol.Explor. Freshwaters 17: 193206.

219

IBS

SP

EC

IA

LIS

SU

E

Tectonics, topography, and mammalian diversity

Catherine Badgley

C. Badgley (cbadgley@umich.edu), Dept of Ecology and Evolutionary Biology and Museum of Paleontology, Univ. of Michigan, Ann Arbor,MI 48109, USA.

Terrestrial vertebrates show striking changes in species richness across topographic gradients. For mammals, nearly twiceas many species per unit area occur in topographically complex regions as in adjacent lowlands. The geological context ofthis pervasive biogeographic pattern suggests that tectonic processes have a first-order impact on regional diversity.

I evaluate ecological, evolutionary, and historical influences of tectonics and topography on the regional diversity ofterrestrial mammals, focusing on the hypothesis that diversification rates are higher in active versus passive tectonicsettings. Ten predictions follow from this hypothesis. 1) The timing of peaks in speciation should be congruent with thetimescale for tectonic episodes. 2) The rates of speciation and genetic differentiation of populations should be greater forspecies inhabiting topographically complex regions than spatially continuous landscapes. 3) If topographic complexity perse promotes diversification, then a cluster of young divergences should occur for montane species compared to lowlandrelatives. 4) Endemism in tectonically active regions should reflect origination within the region rather than rangereduction from larger areas. 5) Extinction rates should differ for lineages in tectonically active regions compared toadjacent lowlands. 6) The relationship between local and regional species richness should differ between topographicsettings because of higher beta diversity in topographically complex regions. 7) Species originating in topographicallycomplex regions should colonize adjacent lowlands more often than the reverse pattern. 8) North-south mountain rangesshould have higher regional species richness than east-west mountain ranges. 9) Areas with multiple mountain rangesshould have higher regional species richness than comparable areas with single mountain ranges. 10) Global climatechanges should affect diversification in tectonically active regions. Research addressing these topics places elevationaldiversity gradients into a geohistorical context and integrates data from modern biotas and the fossil record.

One of the most striking patterns in biogeography is thehigh species richness of terrestrial vertebrates in topogra-phically complex regions compared to adjacent regions oflow relief. The pattern involves both the accumulation andspatial turnover of species along steep elevational andenvironmental gradients, resulting in high regional speciesrichness. The association between complex topography andhigh species richness has been documented for many groupsof terrestrial vertebrates (mammals, birds, amphibians) andvascular plants, as well as for different continental regions(Qian and Ricklefs 2008). The geological context of thisdiversity pattern suggests that tectonic and associatederosional processes, which create gradients in topographiccomplexity, have a first-order impact on regional diversity.

In this paper, I evaluate the potential causes of elevatedrichness of terrestrial mammals in areas of high versus lowtopographic complexity resulting from different tectonichistories, with data from modern and fossil mammalianfaunas and lineages. I review ecological, evolutionary, andhistorical processes that could determine the major featuresof the general pattern. In particular, I focus on thehypothesis that evolutionary processes affecting speciation,extinction, dispersal, and adaptation differ in active versus

passive tectonic settings. Several predictions that followfrom this hypothesis are evaluated in light of current dataand suggest directions for new research.

Background

Mountainous regions are well known to harbor greaterspecies richness than adjacent lowland areas for manygroups and regions, resulting in spectacular diversity hot-spots for terrestrial vertebrates (Humboldt 1805, Simpson1964, Rahbek and Graves 2001, Sechrest et al. 2002,Grenyer et al. 2006, Wiens et al. 2007, Thomas et al.2008). Increasing topographic complexity creates newhabitat, enlarges environmental gradients, establishes bar-riers to dispersal, and isolates populations, potentiallycontributing to adaptation to new environmental condi-tions and speciation in excess of extinction for terrestrialorganisms. For freshwater fishes, topographic complexityreduces habitat area and connectedness and results inelevated extinction rates (Smith et al. pers. comm.).Although this idea has old roots (Simpson 1964, Cracraft1985, Moritz et al. 2000, Brown 2001), it has received far

Ecography 33: 220231, 2010doi: 10.1111/j.1600-0587.2010.06282.x

# 2010 The Author. Journal compilation # 2010 Ecography

Subject Editor: Douglas A. Kelt. Accepted 29 April 2010

220

IBS

SP

EC

IA

LIS

SU

E

less attention than latitude and the associated environmen-tal and historical correlates in relation to taxonomic andecological diversity; and predictions associated with theinfluence of topography remain to be adequately tested.

For mammals, the association of high species richnesswith regions of complex topography has been documentedon several continents. In the first continent-wide analysis ofmammalian biogeography, Simpson (1964) highlightedlatitudinal and longitudinal gradients in species density ofextant North American mammals. The longitudinal gra-dient follows topographic complexity (Fig. 1). Simpson(1964, p. 69) noted that where there are latitudinalgradients, these are additive with topographic gradients,the two accounting for most of the pattern. Kerr andPacker (1997) documented the predictive power of topo-graphy, interpreted as a surrogate for habitat heterogeneity,

on species density of North American mammals. Badgleyand Fox (2000) showed that climatic and topographicvariables predict most of the variation in both speciesdensity and ecological structure of North Americanmammals. South American mammals also show a stronglongitudinal gradient: species density across the Andes istwice as great as in the Amazon Basin at the same latitude(Patterson et al. 2005, InfoNatura 2007). In Europe, spatialclustering of mammalian species density and ecologicalstructure reflects climate and physiography (Heikinheimoet al. 2007). In equatorial Africa, mammalian faunas(excluding bats) of the East African Rift system harbor106122 species, whereas faunas from the Congo Basincontain 5678 species (Badgley unpubl.). Most Australianmammal species inhabit the coastal mountains and dis-sected plateaus of eastern and northeastern Australiacompared to the vast lowland desert interior (data fromStrahan 1995). Although the documentation of Asianmammals is uneven, regional compilations show exception-ally high species richness and extinction risk in the easternHimalayas and mountainous peninsulas of southeasternAsia compared to adjacent lowlands or high plateaus(Sechrest et al. 2002, Schipper et al. 2008).

Where the mammalian fossil record permits comparisonof coeval assemblages from nearby lowlands and uplands,species richness is greater in the upland assemblage. In anearly Cenozoic example, Gunnell and Bartels (2001)compared well sampled, Middle Eocene mammalian faunasfrom basin-center and basin-margin areas of the GreenRiver Basin in Wyoming, USA. The basin-margin site atthe edge of actively rising mountains contained substantiallygreater species richness of mammals and other vertebrategroups than did coeval assemblages from the basin center.Presumably, the higher richness reflected the ecotonebetween lowland and upland habitats. Several groups ofmammals showed evidence of speciation in situ (ancestor-descendant pairs or sister species with an ancestral species inan older interval). A late Cenozoic example contrasts fossilassemblages from the Miocene of Pakistan and south-western China. In floodplain sediments from the Hima-layan foreland basin of Pakistan, the greatest speciesrichness of mammalian assemblages over a 12-million yrperiod was 70 species documented at 10 Ma (million yearsago, unpubl., Barry et al. 2002). In contrast, a late Miocenefossil locality at the same latitude from a montane valley inYunnan Province (southwestern China) preserved ca 120species (Badgley et al. 1988, Z. Qiu pers. comm.). Bothlocalities have mammals indicative of mesic forests andare documented by1000 fossil specimens. Even morecompelling are increases in species richness followingtectonic episodes that altered topographic gradients orgeographic barriers. Barnosky and Carrasco (2002) andKohn and Fremd (2008) documented increases in regionalspecies or generic richness of mammals from the montanewestern United States in step with Middle Miocene tecto-nic extension, volcanism, and rifting. These modern andhistorical examples document high mammal richness intectonically active regions compared to nearby lowlandenvironments and motivate this inquiry into the differentprocesses that have shaped this widespread pattern.

0

500

1000

1500

2000

2500

130 120 110 100 90 80 70

Elev

atio

n (m

)

W Longitude

405060708090

100110120

130 120 110 100 90 80 70

Spec

ies

richn

ess

W Longitude

405060708090

100110120

0 500 1000 1500 2000 2500

Spec

ies

richn

ess

Elevation (m)

West East(A)

(B)

(C)

r = 0.85

Figure 1. A transect across North America at 408N demonstratesthe strong correlation between elevation and mammalian speciesdensity. Data from equal-area quadrats in Badgley and Fox (2000).(A) Average elevation of quadrat in meters above sea level. (B)Species density (including bats) per quadrat. (C) Scatterplot andPearson correlation coefficient between elevation and speciesdensity.

221

IBS

SP

EC

IA

LIS

SU

E

Evolutionary consequences of differenttectonic regimes

The focus of this paper is the hypothesis that tectonicallyactive areas, such as active mountain belts and rift valleys,are engines of mammalian diversity by increasing diversi-fication rates relative to those in tectonically quiescentregions. I am not concerned with species richness orturnover within elevational gradients of individual moun-tains or mountain ranges per se, but with processes thatgenerated differences in species richness, taxonomic com-position, and ecological diversity between major geologicprovinces. The spatial scale of mammalian geographicranges, ecoregions, and life zones is the scale over whichthe evolutionary component of diversity gradients is mostevident (Ricklefs and Schluter 1993). Tectonic activityconverts landscapes of low elevation, low relief, and spatiallycontinuous climate and vegetation into landscapes of highrelief with steep climatic gradients and fragmented vegeta-tion zones. These transformations occur over thousands tomillions of years and are driven by plate subduction, rifting,and mantle hotspots (Windley 1995). Complex topographymay result from deformation at convergent (e.g. Andes,Himalayas) or divergent (e.g. East African Rift, Rio GrandeRift) plate boundaries. Changes in elevation and relief alterclimatic and atmospheric conditions, habitat area, andconnectivity among habitats. For terrestrial mammals, theoperational properties are continuity and area of habitats,proximity and connectivity among areas of similar habitat,strength of environmental and resource gradients, andtopographic stability of habitats (Coblentz and Riitters2004).

Topography interacts with climate at many spatial scales,from the microhabitats of hillsides to entire mountainranges and massive plateaus. Changes in temperature,atmospheric pressure, and humidity with elevation mimiclatitudinal climatic changes, such that alpine tundra andhigh-elevation deserts have climatic, edaphic, and floralparallels with arctic tundra and polar deserts (Barbour andBillings 1999). The altitudinal temperature gradientchanges with latitude and global temperature. Today, forexample, the snowline occurs at 4500 m at the equator, risestoward the Tropics of Cancer and Capricorn, then falls withincreasing latitude, reaching sea level near the poles.Snowlines fell during the last glaciation, by up to 1000 min the tropics (Porter 2001), rose during the currentinterglacial, and are rising further from modern globalwarming (IPCC 2007). Both climate and physiographyinfluence the effectiveness of high mountains or plateaus asbarriers to dispersal. Janzens (1967) assertion that moun-tain passes are higher in the tropics refers to the higherextension of habitable life zones in tropical versus temperateregions, lower seasonal temperature variations throughoutthe tropics, and different thermal tolerances of tropicalversus temperate and boreal organisms. The predictionsabout thermal tolerances have been borne out better forectotherms than for endotherms (Ghalambor et al. 2006).Nonetheless, for birds (Rahbek and Graves 2001) andmammals (McCain 2005), climatic conditions effectivelysort species along elevational gradients. Thus, topographycannot be fully separated from climate and climatic historyin its ecological or evolutionary influences on diversity.

The processes that increase regional (and local) speciesrichness are intensified in montane regions. The environ-mental gradients have the potential to accommodate manykinds of species in superjacent life zones (Merriam 1894,Lomolino et al. 2006). Immigration occurs from adjacentlife zones as well as within zones. During Quaternary glacialcycles, areas of montane vegetation expanded during coolintervals, creating low-elevation connections for dispersal(Brown 1978, Grayson 1993, Thompson and Anderson2000). Geographic isolation of habitats on individualmountains or mountain ranges increases opportunities forgenetic differentiation and speciation, as well as extinction(Cracraft 1985, Brown 2001). Steep environmental gradi-ents present strong selection gradients, creating circum-stances for disruptive selection to act on contiguouspopulations (Endler 1977, Moritz et al. 2000). Theseconditions should affect small mammals more than largeones, because of size-related differences in home-range areaand resource requirements in relation to habitat area andbarriers. Surveys show that small mammals indeed drive themajor gradients in species richness (Patterson et al. 1998,Badgley and Fox 2000, Lomolino 2001, McCain 2005).

In contrast, tectonically quiescent areas, such as theancient shields and passive margins of continents, slowlylose habitats through erosion. Such areas exhibit spatiallycontinuous habitats and low climatic heterogeneity thatpromote geographically extensive populations with highgene flow. Although such areas can potentially accommo-date many species, immigrants from different habitats arefar away. The low frequency of strong barriers facilitateshigh gene flow over large areas, with less opportunity forlineage differentiation or speciation in mammals. Selectiongradients related to environmental conditions are weak.During Quaternary glacial cycles, bioclimatic zones werelatitudinally compressed during glacial advances and ex-panded during interglacials (Wright et al. 1993, Williamset al. 2004). Compression of climatic zones during glacialperiods would have further enhanced gene flow within andamong habitats. This contrast in physiography and Qua-ternary history should lead to differences in the phylogeo-graphic structure of sister taxa occupying regions of highversus low topographic complexity.

The elevational gradient in North America

A North American example highlights the contrast intaxonomic and ecological diversity from different tectoniccontexts. North America (including Central America)consists of tectonically active western and southern regionsand a tectonically passive eastern region. The mountainranges, basins, and plateaus of the tectonically active regionhave been created over the last 100 myr (million years),with current tectonic activity concentrated along thewestern margin of North America and under mantlehotspots, such as the Yellowstone Hotspot (Burchfielet al. 1992, Pierce and Morgan 1992). For a given latitude,species richness per unit area is twice as great in most areasof the tectonically active west as in the tectonically quiescenteast (Fig. 1AB), and richness is strongly correlated withelevation (Fig. 1C). (Active and passive regions occupycomparable areas.) Likewise, certain trophic groups, such as

222

IBS

SP

EC

IA

LIS

SU

E

herbivores and granivores, show high montane speciesrichness, whereas others, notably carnivores and omnivores,do not (Badgley and Fox 2000). Taxonomically, rodentsand bats dominate the latitudinal richness gradient, whereasrodents dominate the longitudinal gradient.

Rodents constitute just over half of extant NorthAmerican mammal species (Hall 1981, Wilson and Reeder2005). Sixty-two percent of rodent geographic ranges liewithin the active region compared to 12% within thepassive region, while 26% have ranges overlapping bothregions (Table 1). Four families cricetids (muroidsincluding voles, deermice, and packrats), geomyids (pocketgophers), heteromyids (pocket and kangaroo mice), andsciurids (squirrels) dominate North American rodentdiversity. All four families have more than twice as manyspecies occurring only in the active region compared to thepassive region, with about one-fourth or fewer speciesoccurring in both regions. This consistent geographicpattern suggests a difference in macroevolutionary processesbetween the active and passive regions.

At a finer scale, the mammals of Colorado illustratespecies richness and turnover along the boundary betweenactive and passive regions (Fig. 2). A strong elevationalgradient spans this boundary. Based on detailed documen-tation of species ranges throughout the state (Fitzgeraldet al. 1994), I recorded the eastern and western range limitsof all species within 1-degree bands of longitude (ca 100km wide) for mammals inhabiting the northern half ofColorado (Fig. 2A). Since the Rocky Mountains run north-south, the longitudinal range boundary is also an elevationalboundary. Twenty-six species have ranges principally in theGreat Plains (eastern Colorado), with some ranges extend-ing into the Rocky Mountain front (central Colorado); 69species have ranges within the Rocky Mountains and theplateaus of western Colorado, with some ranges extendingfurther west; 30 species occur throughout Colorado (Fig.2A). Species richness rises from the Great Plains, peaks atthe Rocky Mountain front, and declines slightly furtherwest (Fig. 2BC). The number of range boundaries, ameasure of spatial turnover, shows a unimodal pattern, withthe highest value at the Rocky Mountain front (Fig. 2D).Notably, the number of range boundaries west of themountain front is not significantly greater than the number

east of the mountain front, signifying that the sustainedincrease in species richness across western Colorado resultsfrom the accumulation of species (mainly from the west)more than from spatial turnover.

These patterns as well as the high beta diversity ofmammals (also, birds and amphibians) in mountainousareas of North and South America (McKnight et al. 2007,Melo et al. 2009, Qian et al. 2009) provide strong evidencefor a general macroecological pattern. But it is necessary toevaluate evolutionary processes in the context of tectonicand environmental history for a more fundamental under-standing of the origins of this gradient.

Predictions and preliminary tests

A number of predictions pertaining to evolutionary andecological processes follow from the hypothesis that tectonichistory is a driver of mammalian diversification. Ten ofthese are given below with representative data from theneontological and paleontological literature, when available.When neither data nor literature are available, I providesuggestions for appropriate tests.

Timing of speciation

The timescale for peaks in mammalian speciation shouldbe consistent with the timescale for tectonic changes intopography. Peaks in origination should correspond topeaks of tectonic activity if the increase in topographiccomplexity forms barriers, fragments populations, andstrengthens selection gradients. Alternatively, if montaneregions are simply accommodating species that haveimmigrated from lowland as well as montane regions,then no correspondence between tectonic history andorigination rates should occur, thereby falsifying theprediction. Data required to test this prediction includewell resolved phylogenies with robust estimates of diver-gence times for the lineages under consideration, fossils ofthe focal lineage to document the timing of origination, andgeochronological data about the timing of tectonism. Sincesmall mammals drive the major features of the diversitytrends, the ideal analyses would involve fossil rodents orinsectivores from paired tectonically active and passiveregions (Finarelli and Badgley 2010). (Fossil bats are tooscarce for such comparisons.) Two studies of mid-Cenozoicmammals demonstrate the potential for comparing origi-nation rates in different tectonic settings (Barnosky andCarrasco 2002, Kohn and Fremd 2008), although the latteranalyzed changes in generic richness.

Speciation and genetic differentiation

The speciation rate should be greater for mammals intopographically complex regions than for sister taxa inspatially continuous landscapes. Speciation rates that areeither systematically higher or statistically similar inspatially continuous landscapes relative to topographicallycomplex regions would falsify the prediction. Fragmenta-tion and isolation of habitats by physiographic barriers andstrong resource gradients should promote speciation in

Table 1. Number of extant North American rodent species withgeographic range entirely within the tectonically active region, thepassive region, or overlapping both regions. Geographic-range datafrom Hall (1981), Patterson et al. (2005), Wilson and Reeder (2005),and InfoNatura (2007).

Active region Passive region Both regions

All rodents (n387) 241 47 99Aplodontidae (n1) 1 Castoridae (n1) 1Caviidae (n1) 1Cricetidae (n191) 117 21 53Dasyproctidae (n2) 1 1Dipodidae (n4) 1 1 2Echimyidae (n3) 1 2Erethizontidae (n3) 1 2Geomyidae (n37) 21 10 6Heteromyidae (n56) 39 5 12Sciuridae (n87) 60 9 18

223

IBS

SP

EC

IA

LIS

SU

E

topographically complex regions, especially for species withsmall home range sizes or stenotopic habitat preferences.Sister species in continuous landscapes should experiencehigher rates of dispersal and gene flow as well as weakresource gradients, thereby lowering the likelihood ofreproductive isolation and allopatric speciation. Datarelevant for testing this prediction include phylogeneticand phylogeographic studies of mammals with sister taxa inboth montane and lowland landscapes.

Two examples involving North American rodents sup-port this prediction. Demastes et al. (2002) conducted aphylogenetic analysis of pocket gophers (Geomyidae) fromthe Mexican Plateau and the Trans-Mexican Volcanic Belt(TMVB). The authors sampled mtDNA from 38 localitiesacross the geographic ranges of Cratogeomys and Pappoge-omys. Their maximum-likelihood tree based on cytochromeb showed a different pattern of cladogenesis on the broad,stable Mexican Plateau compared to the young mountains

108.5 107.5 106.5 105.5 104.5 103.5 102.5405060708090

100

Spec

ies

richn

ess

Met

res

Degrees longitude

010203040

No.

rang

e bo

unda

ries

593

111

15

9

13

121

14

2

11

21

222

7

41N

40N

39N

40N30

39N30

109(A)

(B)

(C)

(D)

108 107 106 105 104 103 102

108.5 107.5 106.5 105.5 104.5 103.5 102.5

30

3000

2000

1000

Figure 2. Transect across the northern half of Colorado (USA), spanning the boundary between tectonically active and passive regions ofNorth America. (A) The longitudinal extent of mammal species ranges is indicated by horizontal lines across a topographic base map(from GeoMapApp, ). Numbers next to lines signify the number of species with that longitudinal rangedistribution. Dark lines indicate geographic ranges that are entirely within the mountainous region or extend further west; light linesindicate ranges on the high plains or extending to the east. (B) Topographic profile across northern Colorado at ca 408N (fromGeoMapApp). (C) Species richness of mammals based on presence of geographic ranges across 1-degree bands of longitude (ca 100 kmwide). (D) Number of range boundaries in each 1-degree band of longitude as a measure of spatial turnover. Geographic-range data fromFitzgerald et al. (1994).

224

IBS

SP

EC

IA

LIS

SU

E

and deep valleys of the TMVB. From divergence estimatedat 3.2 Ma into two lineages of Cratogeomys, one branch gaverise to two species now occupying the northern part of theMexican Plateau and one occupying the southeastern partof the Mexican Plateau. The other branch split into fiveclades (that do not correspond neatly to currently recog-nized species) that live above 2000 m in the TMVB; thedistribution of these clades corresponds to geographicsubdivisions of the TMVB by volcanic plateaus and rivers.Estimated divergence times imply that since 2.6 Ma, twoclades arose in the northern Mexican Plateau and five cladesarose in the TMVB, during an interval of volcanic activityin the TMVB and glacial-interglacial climatic changes.Based on extant species, cladogenesis was about twice asgreat in the volcanically active TMVB as in the stableMexican Plateau.

The second example involves Tamias amoenus (yellow-pine chipmunk), which occupies montane conifer forests innorthwestern North America, and T. striatus (easternchipmunk) from eastern North America. Phylogeographicanalysis based on sequence variation in cytochrome brevealed 12 clades in T. amoenus (Demboski and Sullivan2003). The distribution of haplotypes corresponds closelywith different mountain ranges in several distinct geologicalprovinces. The degree of sequence divergence (4.57.4%)implies differentiation before the mid-Quaternary. Incontrast, the phylogeographic structure of Tamias striatushas no concordance with landscape features but indicatespopulation expansion since the last glacial maximum frommultiple refugia in the eastern United States (Rowe et al.2006). Thus, the geographic structure and temporal depthof mtDNA variation differ in congeners that occupytopographically complex versus simple landscapes.

Species ages

Two scenarios need evaluation. First, if montane regionscontinuously promote speciation, then higher diversificationrates should result in a cluster of young divergencescompared to lowland relatives. Alternatively, if tectonicactivity per se stimulates speciation, then peaks in origina-tion should coincide with the timing of tectonism, irrespec-tive of age. These alternatives could be evaluated with ahigh-quality fossil record that spans intervals of tectonism.Geographic differences in the timing of divergence of extantspecies could also test these scenarios. Moritz et al. (2000)document younger species-level divergence ages for Andeansmall mammals compared to their Amazonian relatives. Forexample, in murid rodents, the mean genetic distancebetween sister taxa in the Amazon is twice as large as themean genetic distance between sister taxa in the Andes,implying that Andean rodent faunas contain youngerspecies. This example supports the first scenario in whichmontane regions systematically promote speciation, sincedivergences occurred over millions of years in both regions.If younger species consistently occur in tectonically quies-cent versus active areas, or if speciation rate does not changein response to an episode of tectonic activity, then thegeneral hypothesis is falsified.

Endemism

Endemism in tectonically active regions should reflectcladogenesis within the montane region rather than con-traction of geographic range(s) from a much larger region.Endemism can result from alternate historical trajectories origination in a region and persistence in that geographicregion alone, origination in one region and relocation to anentirely different area, or reduction of geographic range toproduce a relict distribution (Lomolino et al. 2006).Endemism resulting from the second or third processeswould not support the hypothesis. The geographic isolationof the areas under consideration has a substantial impacton geographic-range expansion, immigration, and ende-mism. In their comparison of vertebrate faunas from twomontane ecosystems of comparable area, YellowstoneNational Park in the northern Rocky Mountains and twonational parks in northern Patagonia, Barnosky et al.(2001) noted lower mammalian species richness and higherlevels of endemism due to origination in situ in Patagoniacompared to Yellowstone. They attributed both propertiesto the isolation of Patagonia from other temperate sources.Well resolved phylogenies can demonstrate whether geo-graphically clustered endemics are also closest relatives,supporting the scenario of endemism reflecting originationand persistence. In addition, data from the fossil recorddocumenting mammalian faunal composition from bothmontane and adjacent lowland areas over time, as well asindicators of paleoenvironmental history, are needed toevaluate the patterns of endemism in tectonically activeregions. Areas with a fossil record of montane and adjacentlowland species include the western U.S. (Carrasco et al.2005, Janis et al. 2008) and Ethiopia (Yalden and Largen1992), each with many endemic mammals today.

Extinction rates

How mammalian extinction rates compare between topo-graphically complex regions and lowlands is an openquestion, since topographically complex regions haveopposing influences on extinction rates. Three scenariosare possible. First, extinction rates could be higher intopographically complex regions than in lowlands. Diver-sification could be still higher in the former if per-lineagespeciation rates exceed extinction rates. This scenarioimplies that the average persistence time of mammalianlineages would be shorter and faunal turnover higher inmontane regions than in lowlands. The smaller, oftenfragmented geographic ranges and large distances betweenareas of suitable habitat in topographically complex regionsshould elevate extinction risk for non-volant mammalpopulations (Brown 2001). Isolation of low- to mid-elevation vegetation zones by high elevation or barrenstretches in extensive, linear mountain ranges or ofalpine regions on smaller, separated mountain ranges(such as in the Great Basin) should reduce dispersal andincrease extinction risk, as is presently occurring forOchotona princeps (pikas) of the western United States(Grayson 2005). Second, mammalian extinction rates couldbe similar between topographically complex regions andtheir adjacent lowlands. Under this scenario, montane

225

IBS

SP

EC

IA

LIS

SU

E

diversification would be driven by speciation and immigra-tion. Third, extinct