DOI: 10.1126/science.1122457, 1778 (2005);310 Science

Roger S. ThorpePopulation Evolution and Island Biogeography

This copy is for your personal, non-commercial use only.

clicking here.colleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others

  here.following the guidelines

can be obtained byPermission to republish or repurpose articles or portions of articles

  ): September 4, 2014 www.sciencemag.org (this information is current as of

The following resources related to this article are available online at

http://www.sciencemag.org/content/310/5755/1778.full.htmlversion of this article at:

including high-resolution figures, can be found in the onlineUpdated information and services,

http://www.sciencemag.org/content/310/5755/1778.full.html#relatedfound at:

can berelated to this article A list of selected additional articles on the Science Web sites

http://www.sciencemag.org/content/310/5755/1778.full.html#ref-list-1, 2 of which can be accessed free:cites 9 articlesThis article

4 article(s) on the ISI Web of Sciencecited by This article has been

http://www.sciencemag.org/content/310/5755/1778.full.html#related-urls1 articles hosted by HighWire Press; see:cited by This article has been

http://www.sciencemag.org/cgi/collection/ecologyEcology

subject collections:This article appears in the following

registered trademark of AAAS. is aScience2005 by the American Association for the Advancement of Science; all rights reserved. The title

CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience

on

Sep

tem

ber

4, 2

014

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

on

Sep

tem

ber

4, 2

014

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

on

Sep

tem

ber

4, 2

014

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

1778

al. extend their interpretation to oceanbasins in which water injected into theoceanic crust plays the same role as mete-oric water in continental aquifers. Theirarticle therefore does justice to the carefulwarning given by Oxburgh and O’Nions(8), who advised that “the systematics ofthe relation between the fluxes of heliumand heat depend on transport processes.”This warning was unfortunately lost in theturmoil of the debate between the advo-cates of whole-mantle versus layered-man-tle convection.

One of the cornerstones of layered-man-tle convection is therefore weakening. Theremaining evidence essentially revolvesaround our understanding of the propertiesand inventories of rare gases. We are stillawaiting incontrovertible data on rare-gassolubility in mantle minerals and melts,which would consolidate the dominantinterpretation of the 3He/4He evidence inoceanic basalts. Helium-3 is a stablenuclide essentially unaffected by deep-seated radioactive processes. Even ifhelium turns out to be particularly incom-patible, enough undegassed mantle mate-rial with high 3He/4He ratios may be con-cealed in the lower mantle as streaks inter-layered with recycled material (11). Inaddition, the argument based on the inven-tory of 40Ar in Earth is crucially dependent

on our knowledge of the terrestrial concen-tration of potassium, an element that isknown to be extremely volatile during plan-etary accretion.

By coincidence, the consensus on theheat/helium imbalance at the surface ofEarth is being challenged at almost thesame moment as new experiments overturnthe well-entrenched idea that high 3He/4Heratios in basalts are the hallmark of primi-tive mantle. 3He is a stable isotope ofhelium, whereas 4He (α particles) is contin-uously produced by the decay of uraniumand thorium. High 3He/4He ratios in hotspotbasalts (e.g., Hawaii) with respect to mid-ocean ridge basalts have been held as primeevidence that the deep mantle never lost itsprimordial gases. Measurement of thehelium solubility in mantle minerals (12)suggests instead that a high 3He/4He mantleratio may not be primordial (3, 4) but rathercorresponds to residues of earlier stages ofmelting (5, 13, 14).

We should not let the heritage of layered-mantle convection models fall into oblivion,however. Numerical models of mantle con-vection repeatedly suggest that radial trans-port is particularly slow across the 660-kmdiscontinuity and display episodic surges oflayered-mantle convection regime (14). Thearticle by Castro et al. nevertheless providesthe first rigorous framework against one of

the strongest arguments used to support therole of the 660-km discontinuity as a con-vection boundary, and the authors should becommended for their work. Further full 2Dand 3D models of large well-characterizedaquifer systems around the world will soonlet us reevaluate the terrestrial heat andhelium fluxes and provide a new perspec-tive on the thermal regime of our planet.

References

1. R. van der Hilst, R. Engdahl, W. Spakman, G. Nolet,Nature 353, 37 (1991).

2. S. P. Grand, J. Geophys. Res. 99, 11591 (1994).3. M. D. Kurz, W. J. Jenkins, S. R. Hart, Nature 297, 43

(1982).4. C. J. Allègre, T. Staudacher, P. Sarda, M. Kurz, Nature303, 762 (1983).

5. N. Coltice, Y. Ricard, Earth Planet. Sci. Lett. 174, 125(1999).

6. C. J. Allègre, A. W. Hofmann, R. K. O’Nions, Geophys.Res. Lett. 23, 3555 (1996).

7. R. K. O’Nions, E. R. Oxburgh, Nature 306, 429 (1983).8. E. R. Oxburgh, R. K. O’Nions, Science 237, 1583 (1987).9. P. E. van Keken, C. Ballentine, D. Porcelli, Earth Planet.

Sci. Lett. 188, 421 (2001).10. M. C. Castro, D. Patriarche, P. Goblet, Earth Planet. Sci.

Lett. 237, 893 (2005).11. M. Boyet, M. O. Garcia, R. Pick, F. Albarède, Geophys.

Res. Lett. 32, 10.1029/2004GL021873 (2005).12. S. W. Parman, M. D. Kurz, S. R. Hart, T. L. Grove, Nature437, 1140 (2005).

13. D. Graham, J. Lupton, F. Albarède, M. Condomines,Nature 347, 545 (1990).

14. S. Xie, P. J. Tackley, Phys. Earth Planet. Inter. 146, 417(2004).

10.1126/science.1120194

In most areas of empirical science,including biology, we take an experimen-tal approach for granted. However, in

evolutionary and related studies, large-scale natural or field experiments are rarebecause the large spatial and temporalscales of evolutionary and biogeographicalprocesses render experimentation problem-atic. Classically, biogeography is aboutlarge-scale pattern: How many species arethere on an island? Is the number related toextent of isolation, island size, or complex-ity of vegetation? For example, why arethere so many tree lizard (anole) species oneach of the Greater Antilles but only one ortwo on each of the Lesser Antilles? Is it sim-ply island size that is accountable, eventhough the small islands are environmen-

tally heterogenous and complex (1)? This isnot readily subject to direct experimenta-tion. However, these large-scale biogeo-graphic patterns are mediated by small-scale population-level processes. Theseinclude ecological processes such as com-petition between species and habitat usage,and evolutionary processes such as adapta-tion by natural selection, ancestor-descen-dant relationships, and speciation (splittinginto distinct species, which do not inter-breed). Exceptionally, it may be possible tomanipulate these fundamental populationprocesses experimentally to gain insightinto biogeographic patterns, although evenpopulation-level experimentation is diffi-cult. For example, experimental introduc-tion of small Caribbean anoles onto islands,and experimental translocation betweenlarge enclosures within islands, haverevealed much about the evolutionary andecological population processes underlying

their biogeography. Large-scale transloca-tion of the small tree lizard Anolis oculatus(Dominica, Lesser Antilles) has demon-strated the rapid effect of natural selectionon a wide range of genetically controlledtraits in response to wet or dry habitats (2,3), thus explaining the nature and cause ofthe geographic variation. In addition,experimental introduction of Anolis sagreionto Bahamanian islands has shown howpredators can alter the behavior, nicheusage, and their selection for prey (4).These studies also suggested how intro-duced predators may render their prey morevulnerable to extinction by catastrophessuch as hurricanes (5).

On page 1807 in this issue, Schoener etal. (6) show how the survival of residentanoles on islands with introduced predatorlizards depends on vegetation height. Onislands without the introduced predator,anoles survive better in habitats withshorter vegetation, but on islands with theintroduced predator, anoles survive betterin habitats with taller vegetation. Island sizeon its own did not appear to have a signifi-cant effect. Hence, the authors take theimportant step of linking a populationprocess (survival) to a key feature of islandbiogeography (vegetation type), and thisdirect demonstration, using a field experi-

E C O L O G Y

Population Evolution

and Island Biogeography Roger S.Thorpe

The author is in the School of Biological Sciences,University of Wales, Bangor, Gwynedd LL57 2UW, UK.E-mail: r.s.thorpe@bangor.ac.uk

P E R S P E C T I V E S

16 DECEMBER 2005 VOL 310 SCIENCE www.sciencemag.orgPublished by AAAS

1779

ment, is novel. This relationship is impor-tant because island biogeographic patternsare determined by such populationprocesses. Although population phenomenaunderlie central aspects of island biogeo-graphic theory, these related disciplines areoften pursued without extensive cross-ref-erence. For instance, the species-area rela-tionship (number of species per island com-pared to size of island) in island biogeogra-phy is usually considered in isolation of thepopulation processes that drive speciation.Comparative island biogeographic analysisof the distribution and number of species ofnative Caribbean island anoles concludesthat “within-island” speciation is rare ornonexistent on small islands (7). However,this proposition has been made in theabsence of population-based genetic evi-dence, and thus biogeographic pattern issuggesting population process, rather thanpopulation process informing biogeo-graphic theory. Moreover, what preciselydoes “within-island” mean? It has becomean axiom of evolutionary ecology that thehistorical component should be taken intoaccount. This, quite rightly, means that thehistory of an organism (phylogeny) shouldbe considered. But in island biogeography,should we not also be considering the geo-logical history? We may anthropocentri-cally regard an island that exists now as a“real” single entity, yet it may have been a

single entity for only a brief portion of itshistory. For example, the island ofMartinique in the Lesser Antilles may oncehave comprised up to five separate islands,some more than 20 million years old, whichhave spent only a fraction of their existenceas part of the single island of Martinique(8). Biogeographic patterning suggests thateach Lesser Antillean island has one anolespecies (per clade), implying that eachisland may have been dominated byallopatric speciation via island coloniza-tion. In other words, each island is naturallyinhabited by a single anole species by virtueof having evolved into a new species aftercolonization of a new island because it wasgeographically, and hence genetically,isolated from its ancestor. This widelyaccepted view is perhaps immune to chal-lenge because it is virtually impossible todirectly test the reproductive isolation ofallopatric species. However, it may be pos-sible to test one aspect of this theory byusing population-level analysis. Theallopatric speciation hypothesis predictsthat the endemic forms on ancient islandsshould act as “good” species and not freelyinterbreed on contact with a differentspecies. Such secondary contact existsbetween the anoles from the precursorislands of Martinique (see the f igure).However, population-based genetic studiesof nuclear gene flow do not support this pre-

diction of no interbreeding on secondarycontact of these populations (9). Thereforeit may be more appropriate for population-based studies to inform biogeographic the-ory, than for biogeographic pattern to sug-gest the nature of the population process.

Field experiments on aspects of popula-tion biology that are relevant to biogeogra-phy are replete with challenges. These prob-lems include maintaining conditions that areas natural as possible such as natural densi-ties of organisms, maintaining discreteexperimental units such as small islets orenclosures, having a size and number ofreplicate units that are large enough to givestatistical significance but are still logisti-cally manageable, and having effects thatare intense enough to yield results in thetime frame of the project and variables withsuff icient variance to elicit a response.Island anoles in general, and A. sagrei in theislands of the Bahamas in particular, providea model system that overcomes many ofthese problems. The small size of anoles,their high population density, numerousstudies providing background informationon their evolutionary ecology, and theirpresence on small discrete islands withvarying vegetation types make island anolesa valuable model system for both experi-mental (1–6) and comparative (1, 8–10)studies in evolution and biogeography.

Perhaps the larger challenge will berelating population ecology and populationevolution to broader scale spatial variationin the environment. Schoener et al. haveshown that demographics are certainly sen-sitive to changes in species compositionand that this is habitat specif ic. Anolesadapt closely to the environmental and cli-matic zones that exist on even small islands(1–3, 8). Can similar experimental studiesbe used to provide insight into the speedand limits of adaptation to climate changeof both individual species and whole com-munities?

References and Notes1. J. B. Losos, R. S. Thorpe, in Adapti ve Speciation ,

U. Dieckmann, M. Doebeli, J. A. J. Metz, D. Tautz, Eds.(Cambridge Univ. Press, Cambridge, UK, 2005), pp.322–344.

2. A. Malhotra, R. S.Thorpe, Nature 353, 347 (1991).3. R. S. Thorpe, J. T. Reardon, A. Malhotra, Am. Nat. 165,

495 (2005).4. J. B. Losos,T. W. Schoener, D. A. Spiller, Nature 432, 505

(2004).5. T. W. Schoener, D. A. Spiller, J. B. Losos, Nature 412, 183

(2001).6. T. W. Schoener, J. B. Losos, D. A. Spiller, Science 310,

1807 (2005).7. J. B. Losos, D. Schluter, Nature 408, 847 (2000).8. R. S.Thorpe,A. G. Stenson, Mol. Ecol. 12, 117 (2003).9. R. Ogden, R. S. Thorpe, Proc. Natl. Acad. Sci. U.S.A. 99,

13612 (2002).10. R. Calsbeek,T. B. Smith, Nature 426, 552 (2003).11. I thank the Biotechnology and Biological Sciences

Research Council for support.

10.1126/science.1122457

No evidence for the geographic speciation predicted by biogeographic pattern. Broad biogeo-graphic pattern suggests geographic speciation in Lesser Antillean anoles, yet on Martinique, whereonce separate island species now meet in secondary contact (red line), population-based geneticstudies show no evidence of their reproductive isolation (9). However, there is strong parallel adap-tation to the habitat zonation and some evidence of reduced interbreeding among these habitattypes (ecotypes) (9). Xeric coastal ecotypes of anoles are brown and striped (lower photographs;arrows indicate locality of origin).Where lineages meet, in the montane rainforest, they show paral-lel evolution of green montane ecotypes (upper photographs), irrespective of their independentisland history and deep molecular phylogenetic divergence.

P E R S P E C T I V E S

www.sciencemag.org SCIENCE VOL 310 16 DECEMBER 2005Published by AAAS