Mar Biol (2009) 1561375ndash1387

DOI 101007s00227-009-1178-1

ORIGINAL PAPER

Phylogeography of California and Galaacutepagos sea lions and population structure within the California sea lion

Yolanda Schramm middot S L Mesnick middot J de la Rosa middot D M Palacios middot M S Lowry middot D Aurioles-Gamboa middot H M Snell middot S Escorza-Trevintildeo

Received 23 June 2008 Accepted 4 March 2009 Published online 21 March 2009copy Springer-Verlag 2009

Abstract We investigate the phylogeography of California(Zalophus californianus) and Galaacutepagos (Z wollebaeki) sealions and describe within-population structure for theCalifornia sea lion based on mitochondrial DNA Fiftycontrol-region haplotypes were found 41 from Z californi-anus and 9 from Z wollebaeki with three Wxed diVerencesbetween the two species Ranked population boundariesalong the range of Z californianus were deWned based onthe Monmonier Maximum DiVerence Algorithm resultingin Wve genetically distinct populations two in the PaciWcOcean and three inside the Gulf of California A MinimumSpanning Network showed a strong phylogeographic signal

with two well-deWned clusters Z californianus and Z wol-lebaeki separated by six base-pair diVerences supportingthe existence of two genetically distinct species with anestimated divergence time of raquo08 Ma Results are dis-cussed in the context of the historical geologic and paleoce-anographic events of the last 1 Ma in the eastern PaciWc

Introduction

California sea lions (Zalophus californianus Lesson 1828)are distributed along the mainland and oVshore islands ofthe eastern North PaciWc Ocean from British ColumbiaCanada to central Meacutexico including the Gulf of California(King 1983) The species is occasionally recorded outside

Communicated by MI Taylor

Y Schramm (amp) middot J de la RosaFacultad de Ciencias Marinas Universidad Autoacutenoma de Baja California Km 103 Carretera Tijuana-Ensenada sn 22830 Ensenada Baja California Mexicoe-mail yschrammuabcmx

S L Mesnick middot M S LowryProtected Resources Division Southwest Fisheries Science Center NOAA Fisheries 8604 La Jolla Shores Drive La Jolla CA 92037 USA

D M PalaciosJoint Institute for Marine and Atmospheric Research University of Hawaii 1000 Pope Road Marine Sciences Building Room 312 Honolulu HI 96822 USA

D M PalaciosEnvironmental Research Division Southwest Fisheries Science Center NOAA Fisheries 1352 Lighthouse Avenue PaciWc Grove CA 93950-2097 USA

D Aurioles-GamboaCentro Interdisciplinario de Ciencias Marinas Instituto Politeacutecnico Nacional Ave IPN sn Colonia Playa Palo de Santa Rita 23000 La Paz Baja California Sur Mexico

H M SnellCharles Darwin Research Station Charles Darwin Foundation Puerto AyoraIsla Santa Cruz Galaacutepagos Ecuador

Present AddressH M SnellMuseum of Southwestern Biology Department of Biology University of New Meacutexico Albuquerque NM 87131 USA

S Escorza-TrevintildeoDepartment of Biological Sciences California State University Los Angeles Los Angeles CA 90032-8201 USA

123

1376 Mar Biol (2009) 1561375ndash1387

of its normal range as far as Alaska to the north (Manisc-alco et al 2004) and southern Meacutexico to the south (Gallo-Reynoso and Soloacuterzano-Velasco 1991) There are fourmain breeding rookeries in the United States on San Mig-uel Santa Barbara San Nicolas and San Clemente islands(Lowry et al 1992) In Meacutexico there are 19 main breedingrookeries from the Coronado Islands to Margarita Islandalong the PaciWc coast (Le Boeuf et al 1983) and fromRocas Consag to Los Islotes in the Gulf of California(Aurioles-Gamboa and Zavala-Gonzaacutelez 1994 Fig 1) Formanagement purposes there are three currently recognizedstocks deWned by the geographic location of their reproduc-tive core areas (Lowry et al 1992) The ldquoUnited Statesrdquostock extends northward of the MeacutexicondashUnited States bor-der including Canada and Alaska with a reproductive cen-ter at the Channel Islands in Southern California and anestimated population size of 238000ndash241000 (Lowry andMaravilla-Chaacutevez 2002) The ldquoWestern Baja Californiardquostock extends southward from the MeacutexicondashUnited Statesborder to the tip of the Baja California peninsula with itsreproductive center at islands near Punta Eugenia and atSanta Margarita Island and an estimated population size of75000ndash85000 (Lowry and Maravilla-Chaacutevez 2002) TheldquoGulf of Californiardquo stock has its reproductive center atislands located within the central and northern portions ofthe Gulf of California and has an estimated population sizeof 31393 (Aurioles-Gamboa and Zavala-Gonzaacutelez 1994)

Despite being one of the most common marine mammalsin the eastern North PaciWc little is known about thegenetic relationships among California sea lion rookeriesWhile there is clear evidence of genetic diVerentiation

between geographically isolated rookeries in the PaciWcOcean and the Gulf of California (Maldonado et al 1995Bowen et al 2006) the picture is complicated by the factthat males (at least within PaciWc populations) undertakeextensive seasonal migrations and individuals of both sexesare capable of moving between rookeries (Bartholomew1967 Aurioles-Gamboa et al 1983 M S Lowry unpub-lished data) although the rate of exchange among them isunknown

Galaacutepagos sea lions (Zalophus wollebaeki Sivertsen1953) are endemic and common throughout the GalaacutepagosArchipelago although major rookeries in the central andsouthern islands (Floreana Santa Cruz San CristoacutebalIsabela Santiago Espantildeola Mosquera Santa Feacute andFernandina) represent almost 90 of the population whichis currently estimated at 16000ndash18000 animals (Salazar2002 Salazar and Michuy 2008) Despite their smallergeographic range and lack of an established migrationGalaacutepagos sea lions are capable of long-range movementsVagrant individuals are occasionally reported oV theCentral and South American coasts as far as 1570 km fromtheir population center (Palacios et al 1997 Capella et al2002)

After the Galaacutepagos sea lion was described a close rela-tionship with the California sea lion was assumed How-ever the taxonomic designation has been controversialWhile most researchers support a species-level separationbased on diVerences in cranial morphometrics (Sivertsen1953 1954) social behavior (Eibl-Eibesfeldt 1984) vocal-izations (Cenami Spada et al 1991) and molecular genetics(Wolf et al 2007) a recent taxonomic review of the familyOtariidae based on cranial morphometry supported divisionof the two taxa only at the subspeciWc level (Brunner 2004)In this study we investigate variation in mitochondrialDNA (mtDNA) throughout the entire breeding range ofboth California and Galaacutepagos sea lions and make phyloge-ographic inferences to help explain their current distribu-tion and degree of taxonomic diVerentiation Further weexamine the level of genetic structuring among Californiasea lion rookeries to identify distinct population units thatcan lead to improved management practices

Materials and methods

Samples

A total of 299 tissue samples were collected in CaliforniaUSA (n = 82 from 2 islands from 1996 to 1998) Meacutexico(n = 170 from 5 islands along the PaciWc coast and 6 islandsin the Gulf of California during 1997) and the GalaacutepagosIslands Ecuador (n = 47 from 8 islands in 1998 and 1999)(see Figs 1 2 for sampling sites) Animals were sampled

Fig 1 Geographic location and number in parenthesis of Z califor-nianus samples collected from California and Meacutexico Rocas Consag(no samples) is shown because it is the northernmost rookery in theGulf of California

123

1376 Mar Biol (2009) 1561375ndash1387

of its normal range as far as Alaska to the north (Manisc-alco et al 2004) and southern Meacutexico to the south (Gallo-Reynoso and Soloacuterzano-Velasco 1991) There are fourmain breeding rookeries in the United States on San Mig-uel Santa Barbara San Nicolas and San Clemente islands(Lowry et al 1992) In Meacutexico there are 19 main breedingrookeries from the Coronado Islands to Margarita Islandalong the PaciWc coast (Le Boeuf et al 1983) and fromRocas Consag to Los Islotes in the Gulf of California(Aurioles-Gamboa and Zavala-Gonzaacutelez 1994 Fig 1) Formanagement purposes there are three currently recognizedstocks deWned by the geographic location of their reproduc-tive core areas (Lowry et al 1992) The ldquoUnited Statesrdquostock extends northward of the MeacutexicondashUnited States bor-der including Canada and Alaska with a reproductive cen-ter at the Channel Islands in Southern California and anestimated population size of 238000ndash241000 (Lowry andMaravilla-Chaacutevez 2002) The ldquoWestern Baja Californiardquostock extends southward from the MeacutexicondashUnited Statesborder to the tip of the Baja California peninsula with itsreproductive center at islands near Punta Eugenia and atSanta Margarita Island and an estimated population size of75000ndash85000 (Lowry and Maravilla-Chaacutevez 2002) TheldquoGulf of Californiardquo stock has its reproductive center atislands located within the central and northern portions ofthe Gulf of California and has an estimated population sizeof 31393 (Aurioles-Gamboa and Zavala-Gonzaacutelez 1994)

Despite being one of the most common marine mammalsin the eastern North PaciWc little is known about thegenetic relationships among California sea lion rookeriesWhile there is clear evidence of genetic diVerentiation

between geographically isolated rookeries in the PaciWcOcean and the Gulf of California (Maldonado et al 1995Bowen et al 2006) the picture is complicated by the factthat males (at least within PaciWc populations) undertakeextensive seasonal migrations and individuals of both sexesare capable of moving between rookeries (Bartholomew1967 Aurioles-Gamboa et al 1983 M S Lowry unpub-lished data) although the rate of exchange among them isunknown

Galaacutepagos sea lions (Zalophus wollebaeki Sivertsen1953) are endemic and common throughout the GalaacutepagosArchipelago although major rookeries in the central andsouthern islands (Floreana Santa Cruz San CristoacutebalIsabela Santiago Espantildeola Mosquera Santa Feacute andFernandina) represent almost 90 of the population whichis currently estimated at 16000ndash18000 animals (Salazar2002 Salazar and Michuy 2008) Despite their smallergeographic range and lack of an established migrationGalaacutepagos sea lions are capable of long-range movementsVagrant individuals are occasionally reported oV theCentral and South American coasts as far as 1570 km fromtheir population center (Palacios et al 1997 Capella et al2002)

After the Galaacutepagos sea lion was described a close rela-tionship with the California sea lion was assumed How-ever the taxonomic designation has been controversialWhile most researchers support a species-level separationbased on diVerences in cranial morphometrics (Sivertsen1953 1954) social behavior (Eibl-Eibesfeldt 1984) vocal-izations (Cenami Spada et al 1991) and molecular genetics(Wolf et al 2007) a recent taxonomic review of the familyOtariidae based on cranial morphometry supported divisionof the two taxa only at the subspeciWc level (Brunner 2004)In this study we investigate variation in mitochondrialDNA (mtDNA) throughout the entire breeding range ofboth California and Galaacutepagos sea lions and make phyloge-ographic inferences to help explain their current distribu-tion and degree of taxonomic diVerentiation Further weexamine the level of genetic structuring among Californiasea lion rookeries to identify distinct population units thatcan lead to improved management practices

Materials and methods

Samples

A total of 299 tissue samples were collected in CaliforniaUSA (n = 82 from 2 islands from 1996 to 1998) Meacutexico(n = 170 from 5 islands along the PaciWc coast and 6 islandsin the Gulf of California during 1997) and the GalaacutepagosIslands Ecuador (n = 47 from 8 islands in 1998 and 1999)(see Figs 1 2 for sampling sites) Animals were sampled

Fig 1 Geographic location and number in parenthesis of Z califor-nianus samples collected from California and Meacutexico Rocas Consag(no samples) is shown because it is the northernmost rookery in theGulf of California

123

Mar Biol (2009) 1561375ndash1387 1377

by clipping a small piece of skin from the hind limbs Livepups were sampled in Meacutexico while stranded dead pups oryearlings were sampled in California and dead individualsfrom all age categories were sampled in Galaacutepagos Sam-ples of California sea lions were collected from a singlerookery on each island except for San Miguel (several sitesalong a 3-km stretch of coastline) and San Nicolas (twosites) In total we deWned 13 sampling strata for Californiasea lions Galaacutepagos samples were obtained from 17 siteson 9 diVerent islands (Fig 2) Due to the low sample sizeper site we combined these samples into one stratum for allGalaacutepagos rookeries

Samples were either stored in liquid nitrogen in the Weldlater transferred to an ultrafreezer and kept at iexcl70degC or inan aqueous solution of 20 (vv) DMSO saturated withNaCl (Amos and Hoelzel 1991) and kept at iexcl40degC untilDNA extraction

DNA extraction

Samples (40 mg) were digested for 20 h at room tempera-ture in 40 L of Proteinase K (10 mgmL SIGMA or GibcoBRL) 400 L extraction buVer (01 M NaCl 10 mM TrisndashHCl pH 80 1 mM EDTA) and 40 L 10 SDS The DNAwas puriWed by standard phenolndashchloroformndashisoamyl alco-hol (25241) extractions (modiWed from Sambrook et al1989) The precipitate was resuspended in TrisndashEDTAbuVer (10 mM TrisndashHCl pH 80 1 mM EDTA) to an aver-age concentration of 150 ngL The quality of the DNAwas examined via electrophoresis on 1 agarose gels usingraquo150 ng of DNA

AmpliWcation and sequencing

A mtDNA fragment of raquo550 base pairs (bp) from thehypervariable region I of the control region was ampliWedfrom the 299 individuals using the polymerase chain reac-tion (PCR) Two primers developed by the Marine Mam-mal Genetic Group at the Southwest Fisheries ScienceCenter (SWFSC) La Jolla California were used TheTro primer (5-CCTCCCTAAGACTCAAGG-3) annealsbetween the tRNA threonine gene and the tRNA prolinegene (L-strand) and the Dx primer (5-CCTGAAGTAAGAAACCAGATG 3) anneals within the conserveddomain of the control region (H-strand) Reactions wereperformed in 25 L volumes containing raquo20 ng of geno-mic DNA 20 mM TrisndashHCl pH 84 50 mM KCl 3 mMMgCl2 200 M of each dNTP 03 M of each primer and1 U of Taq DNA polymerase (Gibco BRL) The thermalcycling proWle was as follows an initial hot-start of 5 minat 94degC 35 ampliWcation cycles of denaturation for 1 minat 94degC annealing for 1 min at 50degC and extension for15 min with a 2 s increase per cycle at 70degC and a Wnal5 min incubation at 70degC to ensure complete extension ofthe PCR products

Successful ampliWcation products were then cleaned bypuriWcation columns (Concerttrade Rapid PCR PuriWcationSystem Gibco BRL) according to the manufacturersrsquo spec-iWcations Both heavy and light strands were cycle-sequenced using the BigDyereg Terminator Sequencing Stan-dard (Applied Biosystems Inc) Reactions were performedin 12 L volumes containing 60ndash100 ng of double-stranded cleaned PCR product 025 M of one primer and2 L of terminator ready-reaction mix The thermal cyclingproWle included an initial hot-start of 5 min at 95degC fol-lowed by 25 cycles of denaturation for 30 s at 95degC anneal-ing for 15 s at 50degC extension for 4 min at 60degC and a Wnal2 min incubation at 60degC to ensure complete extension ofthe PCR products Sequenced products were puriWed byethanol precipitation and then run on an ABI 377 DNAautomated sequencer

Data analyses

Editing of opposite strands was performed simultaneouslyusing Sequenchertrade version 41 software to produce383 bp-long sequences Initial sequence comparisons andmeasures of variability were performed using MEGA version21 (Kumar et al 2001) Final sequences were comparedwith the complete mitochondrial genome of the harbor seal(Phoca vitulina) from GenBank accession number NC001325 (Arnason and Johnsson 1992) as a reference Hap-lotype (h) and nucleotide () diversity was estimated andTajimarsquos test of neutrality was performed on both groups ofsamples (California and Galaacutepagos) and on each putative

Fig 2 Geographic location of 47 Z wollebaeki samples collectedfrom 17 diVerent sites on nine islands of the Galaacutepagos Archipelago

123

1378 Mar Biol (2009) 1561375ndash1387

population within California samples using Arlequinversion 20 (Schneider et al 2001)

The phylogeographic structure was analyzed by compar-ing phylogenetic relationships among unique haplotypesand the geographic location of each haplotype Phyloge-netic relationships were inferred from a Minimum Span-ning Network (MSN) of all haplotypes The number ofpairwise nucleotide diVerences among haplotypes was usedin Minspnet (ExcoYer and Smouse 1994) to derive theMSN Because there were no shared haplotypes betweenCalifornia and Galaacutepagos samples (see ldquoResultsrdquo) andbecause of the number and type of mutations separatingthese clusters in the MSN were both indicative of species-level diVerentiation the Galaacutepagos samples were excludedfrom analyses of population genetic structure withinCalifornia sea lions

Steller sea lion (Eumetopias jubatus) the sister taxon toZalophus and its closest extant relative in the North PaciWc(Wynen et al 2001 Demeacutereacute et al 2003) was used as acalibration point to estimate the minimum time of geneticdivergence between Z californianus and Z wollebaekiFirst the three most divergent sequences in the E jubatuswere chosen from GenBank and compared to three of themost divergent Z californianus sequences from the presentstudy The mean divergence time between Z californianusand E jubatus was then estimated based on the meanD-loop sequence divergence rate for marine mammals(325 per million years (My) ie the mean between thedivergence rate in cetaceans (05 per My) and that in ele-phant seals (raquo6 per My) Stewart and Baker 1994) Theprocedure was then repeated for the three most divergentsequences of Z californianus and Z wollebaeki to estimatethe minimum mean genetic divergence time between thetwo

To assess the extent of genetic structure within Califor-nia sea lions the Monmonier Maximum DiVerence Algo-rithm (MMDA Manel et al 2003) was used to determinein a ranked order potential boundaries separating putativepopulations First the 13 sampling strata were placed on aDelaunay network (Brassel and Reif 1979) connectingadjacent sampling strata The MMDA was then imple-mented in the Barrier version 22 program (Manni et al2004) using Neirsquos Da genetic distance between mtDNAhaplotypes as a measure of genetic distance among the 13sampling strata The resultant putative populations deWnedby these potential and ranked boundaries were then testedusing an analysis of molecular variance (AMOVAExcoYer et al 1992) implemented in Arlequin version 20(Schneider et al 2001) AMOVA was performed to esti-mate F-statistics and their analogue -statistics For STthe genetic distance between pairs of haplotypes was esti-mated as the proportion of the nucleotide diVerencesbetween them The null distribution of pairwaise FST and

ST values under the hypothesis of panmixia was obtainedby 16000 permutations of haplotypes between populationsguaranteeing less than 1 diVerence with the exact proba-bility in 99 of the cases (Guo and Thomson 1992) TheWnal number of populations was determined as that whichresulted in statistically signiWcant diVerentiation betweenall pairs of adjacent populations when using the largestnumber in their ranked order of potential boundaries

Results

Genetic diversity and neutrality

Fifty haplotypes were found fourty-one speciWc to Z cali-fornianus and nine speciWc to Z wollebaeki (Table 1)Twenty-nine sites were variable with 28 transitions and asingle transversion The 50 diVerent haplotypes weredeposited in the GenBank database under accession num-bers EF512168 to EF512217 Overall haplotype diversityfor California sea lions was h = 08860 sect 00123 The low-est values were found in the ldquoPaciWc Temperaterdquo(h = 06712 sect 00404) and in the Galaacutepagos (h = 07604 sect0521) populations (Table 2 see ldquoPopulation Structurerdquosection for population deWnitions) Overall nucleotidediversity from California sea lions was = 00088 sect00050 The lowest value was found in Galaacutepagos( = 00037 sect 00026) this value was almost half that ofthe nearest lowest values (ldquoPaciWc Temperaterdquo and ldquoSouth-ern Gulfrdquo Table 2) The null hypothesis of neutralitywas not rejected in all cases Tajimarsquos D-statistics werestatistically non-signiWcant (P gt 03 Table 2) No sharedhaplotypes were found between samples collected fromZ californianus and Z wollebaeki (Table 1) A uniquetransversion (site number 214 Table 1) and two transitions(sites 233 and 234 Table 1) represented Wxed diVerencesthat distinguished Galaacutepagos from California samples

Phylogeographic structure and divergence time

The MSN showed a strong phylogeographic signal withtwo distinct clusters corresponding to California (haplotypesH1ndashH41) and Galaacutepagos (haplotypes H42ndashH50) sea lions(upper and lower clusters in Fig 3 respectively) HaplotypeH1 was the most common one with the highest number ofconnections followed by H29 and H34 Galaacutepagos haplo-types (H42ndashH50) were grouped together and separated fromCalifornia haplotypes by six mutations (Fig 3) The Galaacutepa-gos haplotype cluster had fewer reticulations than the Cali-fornia cluster (Fig 3) In the latter common haplotypes andhaplotypes with high numbers of connections occurred morefrequently in the PaciWc populations (ldquoPaciWc Temperaterdquoand ldquoPaciWc Subtropicalrdquo Fig 4) Peripheral and private

123

Mar Biol (2009) 1561375ndash1387 1379

haplotypes occurred most frequently in ldquoNorthern Gulfrdquo andldquoCentral Gulfrdquo populations with whole haplotype clusterspresent only in the Gulf of California (Fig 4) Despite theirlower sample sizes Gulf populations showed higher haplo-type diversities (Table 2) Very few haplotypes were sharedbetween all populations (Fig 4)

The three most divergent sequences for E jubatus(GenBank accession numbers AY340888 AY340917 and

AY340937 Baker et al 2005) and for Z californianus(haplotypes 1 29 and 35 from the present study) yielded amean sequence divergence of 99 between the two generaand a mean genetic divergence time of 305 million yearsago (Ma) (using the 325 per My mean D-loop sequencedivergence rate for marine mammals) Based on this esti-mate the three most divergent Z californianus and Z wol-lebaeki sequences (haplotypes 42 47 and 50) yielded a

Table 1 List of 50 haplotypes deWned by 29 variable sites on the basis of 383 bp of the hypervariable region I of mitochondrial DNA of Z cali-fornianus (H01ndashH41) and Z wollebaeki (H42ndashH50)

Site No1 of the complete sequence is equivalent to site No16304 of the harbor seal (Phoca vitulina) sequence by Arnason and Johnsson (1992)GenBank accession number NC 001325 A transversion in site number 214 and two transitions (233 234) are Wxed diVerences that distinguishedZ wollebaeki from Z californianus haplotypes

rebmuN etiS

Haplotype43

73

92

112

119

126

133

143

144

148

212

214

215

220

222

226

232

233

234

238

239

242

244

245

247

253

265

274

341

H01 C T T G T T A G A T T A T C C T A A C T T T C A A T A G TH02 bull bull bull bull bull bull bull bull bull bull bull bull bull bull T bull bull bull bull bull bull bull bull bull bull bull bull bull bull H03 bull bull bull bull bull C bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H04 bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H05 bull bull bull bull bull bull bull bull G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H06 bull bull bull bull bull bull G bull G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H07 bull bull bull bull bull bull G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H08 bull bull bull bull bull bull bull A bull bull bull bull bull bull bull C bull bull bull bull bull bull bull bull bull bull bull bull bull H09 bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull bull bull bull bull bull bull bull bull bull bull H10 bull bull bull bull bull bull bull bull G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull CH11 bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull CH12 bull bull bull bull bull bull bull A bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull CH13 bull bull bull bull bull bull bull A bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H14 bull bull bull bull bull bull G A bull C bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H15 bull bull bull bull bull bull G A bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H16 T bull bull bull bull bull bull A bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull G bull C bull bull bull H17 T bull bull bull bull C bull A bull bull bull bull C bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H18 T bull bull bull bull C bull A bull bull bull bull C bull bull bull G bull bull bull bull bull bull bull bull C bull bull bull H19 T bull C bull bull C bull A bull bull bull bull C bull bull C bull bull bull bull bull bull bull bull bull C bull bull bull H20 T bull bull bull bull bull bull A bull bull bull bull C bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H21 T bull bull bull bull bull G A bull bull bull bull C bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H22 T bull bull bull bull bull G A bull C bull bull C bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H23 T bull bull bull bull bull G A bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull G bull bull A bull H24 T bull bull bull bull bull G A bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull A bull H25 bull bull bull bull bull bull bull A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H26 bull bull bull bull bull C bull A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H27 T bull bull bull bull C bull A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull H28 T bull bull bull bull C bull A bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H29 T bull bull bull bull bull bull A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H30 T bull bull bull C bull bull A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H31 T bull bull bull bull bull G A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H32 T bull bull bull bull C G A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H33 T bull bull bull bull C bull A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull G bull C bull bull bull H34 T bull bull bull bull C bull A G bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H35 bull bull bull bull bull bull G A G bull bull bull C bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull H36 bull bull bull bull bull bull G A G bull bull bull C bull bull bull bull bull bull bull bull bull bull bull bull C bull bull CH37 bull bull bull bull bull bull G A G bull bull bull bull bull bull bull G bull bull bull bull bull bull bull bull C G bull bull H38 bull bull bull bull bull bull G A G bull bull bull bull bull bull bull G bull bull C bull bull bull bull bull C G bull bull H39 bull bull bull bull bull bull G A G bull bull bull C bull bull C bull bull bull bull bull bull bull bull bull C G bull bull H40 bull bull bull A bull bull G A G bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull C bull bull bull H41 bull bull bull A bull bull G A G bull bull bull bull bull bull bull bull bull bull bull bull C bull bull bull C bull bull CH42 bull bull bull bull bull bull G A G bull bull C bull bull bull bull bull G T C bull C bull bull G C bull bull bull H43 bull bull bull bull bull bull G A G bull bull C bull bull bull bull bull G T C C C bull bull G C bull bull bull H44 bull bull bull bull bull bull G bull G bull bull C bull bull bull bull bull G T C bull C bull bull G C G bull bull H45 bull bull bull bull bull bull G bull G bull bull C bull bull bull bull bull G T C bull C bull bull G C bull bull bull H46 bull bull bull bull bull bull G bull G bull bull C bull bull T bull bull G T C bull C bull bull G C bull bull bull H47 bull bull bull bull bull bull G A G bull bull C bull bull T bull bull G T C bull C bull bull G C bull bull bull H48 bull C bull bull bull bull G A G bull bull C bull bull T bull bull G T C bull C bull bull G C bull bull bull H49 bull bull bull bull bull bull G A G bull C C bull bull T bull bull G T C bull C T bull G C bull bull bull H50 bull bull bull bull bull bull G A G bull C C bull T T bull bull G T C bull C T bull G C bull bull bull

123

1380 Mar Biol (2009) 1561375ndash1387

mean divergence of 26 and a minimum mean divergencetime of 08 Ma

Population structure

When the Wrst four potential boundaries produced by theMMDA were used overall AMOVA results for the result-ing Wve putative populations were statistically signiWcantfor both Wxation indexes (FST = 0135 P lt 0001ST = 0135 P lt 0001) Additionally all pairwise compar-isons showed statistically signiWcant diVerences among theWve populations for both FST and ST values (Table 3)Based on the approximate geographic range of each ofthese populations we name them ldquoPaciWc Temperaterdquo(comprising San Miguel San Nicolas and Coronadosislands) ldquoPaciWc Subtropicalrdquo (including Benito CedrosAsuncioacuten and Margarita islands) ldquoSouthern Gulfrdquo (LosIslotes) ldquoCentral Gulfrdquo (comprising San Esteban Island andLos Cantiles rookery on Aacutengel de la Guarda Island) andldquoNorthern Gulfrdquo (including Granito Lobos and San Jorgeislands) (see Fig 1)

Discussion

Genetic diversity and neutrality

All substitutions in California sea lion haplotypes weretransitions in agreement with values reported for controlregion sequences from nine other species of Otariidae allof which showed values for transitions gt90 (Wynen et al2001) but higher than in a previous California sea lionstudy that reported 71 transitions (Maldonado et al1995)

The overall haplotype diversity among Z californianussamples (h = 08860) fell among the values reported for

E jubatus (h = 0927 Bickham et al 1996) Callorhinusursinus (h = 0994 Dickerson et al 2008) and the Guada-lupe fur seal (Arctocephalus townsendi) (h = 0798 Weberet al 2004) Similarly nucleotide diversity for Z californi-anus ( = 00088) was comparable to that reported for otherspecies of sea lions such as E jubatus ( = 0004) Otariabyronia ( = 0008) and Phocartos hookeri ( = 0004)(Wynen et al 2001) The lowest haplotype and nucleotidediversity values were found in the ldquoPaciWc Temperaterdquo andin the Galaacutepagos samples (Table 2) despite the fact thatthese were the only strata where samples were collectedfrom more than one site per island and in more than 1 year

This low genetic diversity may be related to a possiblefounder eVect or to historical events that could havereduced the populations such as strong El Nintildeo eventsdisease epidemics or commercial harvesting For instancethere is archeological evidence that the San Miguel Island(ldquoPaciWc Temperaterdquo) population was considerably smaller(ldquorarerdquo) between 1425 and 1500 AD than it is today(Walker et al 1999) suggesting this rookery may havebeen colonized only in the last few centuries Faunalremains from archaeological sites on San Miguel Island areof particular importance in this respect since they are theonly source of information currently available concerningthe recent history However since faunal remains are aproduct of human subsistence activity they cannot be con-sidered an unbiased sample of the sea mammal populationsliving prehistorically in the vicinity of San Miguel Island(Walker and Craig 1979) In more recent times commercialharvesting in Southern California and the Mexican PaciWcreduced these populations to only about 1500 animals bythe 1920s (Heath 2002) while harvesting in the Gulf ofCalifornia was not as intensive (Lluch-Belda 1969 Zavala-Gonzaacutelez and Mellink 2000) In addition all PaciWc coastrookeries are exposed to dramatic population Xuctuationsassociated with recurring El Nintildeo events (DeLong et al

Table 2 Measures of genetic diversity (sectSD) and the results of Tajimarsquos neutrality test by population and totals for Z californianus and Z wol-lebaeki

ldquoPaciWc Temperaterdquo includes San Miguel San Nicolas and Coronados islands ldquoPaciWc Subtropicalrdquo includes Benito Cedros Asuncioacuten andMargarita islands ldquoSouthern Gulfrdquo includes Islotes ldquoCentral Gulfrdquo includes Esteban Island and Cantiles rookery on Aacutengel de la Guarda IslandldquoNorthern Gulfrdquo includes Granito Lobos and San Jorge islands

PS = Polymorphic sites

Population No of samples

No of haplotypes (PS)

Haplotype diversity (h) ()

Nucleotide diversity () ()

Tajimarsquos D statistic

PaciWc Temperate 94 10 (9) 6712 sect 404 064 sect 039 09647 P = 0854

PaciWc Subtropical 67 11 (10) 8173 sect 247 081 sect 047 13157 P = 0910

Southern Gulf 16 9 (6) 8833 sect 612 062 sect 040 10730 P = 0873

Central Gulf 27 12 (12) 8746 sect 513 082 sect 049 00099 P = 0556

Northern Gulf 48 19 (16) 9309 sect 168 109 sect 061 04770 P = 0725

Total (Z californianus) 252 41 (21) 8860 sect 123 088 sect 050 iexcl00621 P = 0543

Z wollebaeki 47 9 (8) 7604 sect 521 037 sect 026 iexcl06008 P = 0314

123

1380 Mar Biol (2009) 1561375ndash1387

mean divergence of 26 and a minimum mean divergencetime of 08 Ma

Population structure

When the Wrst four potential boundaries produced by theMMDA were used overall AMOVA results for the result-ing Wve putative populations were statistically signiWcantfor both Wxation indexes (FST = 0135 P lt 0001ST = 0135 P lt 0001) Additionally all pairwise compar-isons showed statistically signiWcant diVerences among theWve populations for both FST and ST values (Table 3)Based on the approximate geographic range of each ofthese populations we name them ldquoPaciWc Temperaterdquo(comprising San Miguel San Nicolas and Coronadosislands) ldquoPaciWc Subtropicalrdquo (including Benito CedrosAsuncioacuten and Margarita islands) ldquoSouthern Gulfrdquo (LosIslotes) ldquoCentral Gulfrdquo (comprising San Esteban Island andLos Cantiles rookery on Aacutengel de la Guarda Island) andldquoNorthern Gulfrdquo (including Granito Lobos and San Jorgeislands) (see Fig 1)

Discussion

Genetic diversity and neutrality

All substitutions in California sea lion haplotypes weretransitions in agreement with values reported for controlregion sequences from nine other species of Otariidae allof which showed values for transitions gt90 (Wynen et al2001) but higher than in a previous California sea lionstudy that reported 71 transitions (Maldonado et al1995)

The overall haplotype diversity among Z californianussamples (h = 08860) fell among the values reported for

E jubatus (h = 0927 Bickham et al 1996) Callorhinusursinus (h = 0994 Dickerson et al 2008) and the Guada-lupe fur seal (Arctocephalus townsendi) (h = 0798 Weberet al 2004) Similarly nucleotide diversity for Z californi-anus ( = 00088) was comparable to that reported for otherspecies of sea lions such as E jubatus ( = 0004) Otariabyronia ( = 0008) and Phocartos hookeri ( = 0004)(Wynen et al 2001) The lowest haplotype and nucleotidediversity values were found in the ldquoPaciWc Temperaterdquo andin the Galaacutepagos samples (Table 2) despite the fact thatthese were the only strata where samples were collectedfrom more than one site per island and in more than 1 year

This low genetic diversity may be related to a possiblefounder eVect or to historical events that could havereduced the populations such as strong El Nintildeo eventsdisease epidemics or commercial harvesting For instancethere is archeological evidence that the San Miguel Island(ldquoPaciWc Temperaterdquo) population was considerably smaller(ldquorarerdquo) between 1425 and 1500 AD than it is today(Walker et al 1999) suggesting this rookery may havebeen colonized only in the last few centuries Faunalremains from archaeological sites on San Miguel Island areof particular importance in this respect since they are theonly source of information currently available concerningthe recent history However since faunal remains are aproduct of human subsistence activity they cannot be con-sidered an unbiased sample of the sea mammal populationsliving prehistorically in the vicinity of San Miguel Island(Walker and Craig 1979) In more recent times commercialharvesting in Southern California and the Mexican PaciWcreduced these populations to only about 1500 animals bythe 1920s (Heath 2002) while harvesting in the Gulf ofCalifornia was not as intensive (Lluch-Belda 1969 Zavala-Gonzaacutelez and Mellink 2000) In addition all PaciWc coastrookeries are exposed to dramatic population Xuctuationsassociated with recurring El Nintildeo events (DeLong et al

Table 2 Measures of genetic diversity (sectSD) and the results of Tajimarsquos neutrality test by population and totals for Z californianus and Z wol-lebaeki

ldquoPaciWc Temperaterdquo includes San Miguel San Nicolas and Coronados islands ldquoPaciWc Subtropicalrdquo includes Benito Cedros Asuncioacuten andMargarita islands ldquoSouthern Gulfrdquo includes Islotes ldquoCentral Gulfrdquo includes Esteban Island and Cantiles rookery on Aacutengel de la Guarda IslandldquoNorthern Gulfrdquo includes Granito Lobos and San Jorge islands

PS = Polymorphic sites

Population No of samples

No of haplotypes (PS)

Haplotype diversity (h) ()

Nucleotide diversity () ()

Tajimarsquos D statistic

PaciWc Temperate 94 10 (9) 6712 sect 404 064 sect 039 09647 P = 0854

PaciWc Subtropical 67 11 (10) 8173 sect 247 081 sect 047 13157 P = 0910

Southern Gulf 16 9 (6) 8833 sect 612 062 sect 040 10730 P = 0873

Central Gulf 27 12 (12) 8746 sect 513 082 sect 049 00099 P = 0556

Northern Gulf 48 19 (16) 9309 sect 168 109 sect 061 04770 P = 0725

Total (Z californianus) 252 41 (21) 8860 sect 123 088 sect 050 iexcl00621 P = 0543

Z wollebaeki 47 9 (8) 7604 sect 521 037 sect 026 iexcl06008 P = 0314

123

Mar Biol (2009) 1561375ndash1387 1381

1991 Boness et al 1991 Francis and Heath 1991 Morriset al 1999) in contrast to the Gulf of California wherepopulations are relatively protected (Aurioles-Gamboa andLe Boeuf 1991 Hernaacutendez-Camacho et al 2008a) Forexample in the islands of the Southern California Bightduring the 1982ndash1983 El Nintildeo births decreased 30 at SanMiguel Island 43 at San Nicolas Island 62 at SanClemente Island and 71 at Santa Barbara Island (DeLonget al 1991) The eVects of El Nintildeo are also pervasive inGalaacutepagos where population declines gt30 and pup mor-talities gt90 have been documented during the 1982ndash1983(Trillmich and Limberger 1985 Trillmich and Dellinger1991) and 1997ndash1998 (Salazar and Bustamante 2003)strong events

Phylogeography of Zalophus

A strong phylogeographic signal with two well-deWnedclusters separated by six mutations in the MSN supportedthe existence of two species Z californianus and Z wol-lebaeki Our estimated mean genetic divergence timebetween Eumetopias and Zalophus of 305 Ma fell withinthe 95 conWdence interval (157ndash306 Ma) for the mini-mum divergence time of 225 Ma reported by Harlin-Cog-nato et al (2006) for these two taxa Based on this valuethe estimated time of genetic divergence between Z cali-fornianus and Z wollebaeki is 08 Ma which is three timeslower than the 23 sect 05 Ma recently estimated by Wolfet al (2007) This discrepancy may be due to diVerences inthe calibration value used or to the diVerent number of WxeddiVerences detected in each study (nine in Wolf et al 2007versus three in the present study see Hey 1991) which inturn may be due to the larger and more geographically rep-resentative sample size evaluated in our study In particularthe closest haplotypes to Z wollebaeki (H40 and H41) werefound only in samples from the ldquoNorthern Gulfrdquo and ldquoCen-tral Gulfrdquo populations which were not represented in Wolfet al (2007) We emphasize however that our estimate isonly an approximation that suggests a more recent time ofdivergence the inexact mutation rate used in our study and

Fig 3 Minimum Spanning Network for 41 haplotypes of Z californi-anus (252 samples) and 9 haplotypes (H42ndashH50) of Z wollebaeki (47samples) Each circle represents a haplotype inside are its number andfrequency (in parenthesis) The small circles represent one individualThe transverse marks between haplotypes indicate the number of muta-tions between them Straight lines represent direct relations and curvedlines are alternative relations

Fig 4 Geographic representation of the Minimum Spanning Net-works for the Wve populations of Z californianus The ldquoPaciWc Tem-peraterdquo population includes San Miguel San Nicolas and Coronadosislands ldquoPaciWc Subtropicalrdquo comprises Benito Cedros Asuncioacutenand Santa Margarita islands ldquoSouthern Gulfrdquo is represented by LosIslotes ldquoCentral Gulfrdquo includes San Esteban Island and Los Cantilesrookery on Aacutengel de la Guarda Island ldquoNorthern Gulfrdquo comprisesGranito Lobos and San Jorge islands The arrangement of haplotypescorresponds to that in Fig 3 with Wlled circles representing thehaplotypes found in that population

123

1382 Mar Biol (2009) 1561375ndash1387

the very limited representation of Zalophus in the fossilrecord (see Demeacutereacute et al 2003 Ho et al 2005) do not allowa more constrained value Additionally the need for a goodfossil dating will be necessary to more accurately trace theevolutionary history of pinnipeds

The historical process by which Zalophus sea lionsarrived in the Galaacutepagos Islands and became a distinct spe-cies remains highly conjectural The basal otariinae evolvedin the temperate eastern North PaciWc in the middle Miocene(before 11 Ma) and it is generally assumed that they dis-persed into the Southern Hemisphere (in one or multiplewaves) in the late Plioceneearly Pleistocene (raquo2ndash3 Ma) viaa cool-water pathway known as the East PaciWc Corridorconnecting the California and Peruacute currents (Davies 1958Repenning et al 1979 Demeacutereacute et al 2003) This dispersalevent which is consistent with cooling and very high levelsof biological productivity in the tropics at the time (Lawrenceet al 2006) likely culminated in the origin of the threegenera of present-day Southern Hemisphere otariinae(Otaria Phocarctos Neophoca) In an analogous mannerwe suggest that the establishment of Z wollebaeki as a distinctspecies in Galaacutepagos occurred later in the Pleistocene

A determining factor for the colonization of the Galaacutepagosby a large predator like Zalophus must have been theavailability of suitable foraging habitat and adequate preybase Unlike the extensive and relatively more stable conti-nental shelves which support ample Wsh and cephalopodbiomasses the Galaacutepagos are the product of hotspot volca-nism with a complex history of emergence and submer-gence The age of the present-day Galaacutepagos has beenestimated at between 03 Ma for Fernandina Island in thewest and 63 Ma for San Cristoacutebal Island in the east (Geist1996) although now-drowned but once-emergent volca-noes have been dated at 5ndash14 Ma (Christie et al 1992

Werner et al 1999) This process would have provided thestepping stones for the persistence and evolution of theunique Galaacutepagos terrestrial biota (Rassmann 1997 Grehan2001 Beheregaray et al 2004) but would not have beenconducive to the development of a marine ecosystem capa-ble of supporting a large biomass of epipelagic and demer-sal prey until a conWguration similar to the present-dayArchipelago was reached Further while cool upwellingand high oceanic productivity characterized the glacialperiods of the Pleistocene (18ndash001 Ma) (Lawrence et al2006 Lea et al 2006 Koutavas and Sachs 2008) sea levelwas 100ndash125 m below present such that most of the shal-low shelves connecting the islands where most foraging bysea lions takes place today (Villegas-Amtmann et al 2008)were exposed (see Fig 2 in Geist 1996) For these reasonsthe successful colonization and establishment of a distinctZalophus form in Galaacutepagos may have only occurred in themiddlendashlate Pleistocene Indeed speciation in other central-place foraging marine predators of Galaacutepagos appears tohave occurred very recently (raquo05 Ma) (Browne et al1997 Akst et al 2002 Friesen et al 2002)

The phylogeographic signal from the presenceabsenceof haplotypes in the MSN (Fig 4) also reveals informationabout the origin and history of Z californianus populationsAll regions presented haplotype H1 as well as haplotypesderived from it indicating that this haplotype was presentin the ancestral population (Crandall and Templeton 1993ExcoYer and Smouse 1994) In contrast haplotypes H29and H34 (second and third most common haplotypes) onlyoccurred in the ldquoPaciWc Temperaterdquo and ldquoPaciWc Subtropi-calrdquo populations revealing a strong separation betweenpopulations in the PaciWc Ocean and in the Gulf of CaliforniaHowever the ldquoNorthern Gulfrdquo and ldquoCentral Gulfrdquopopulations were characterized by whole clusters of private

Table 3 Pairwise FST and ST values (lower matrix) for comparisons among Wve populations ldquoPaciWc Temperaterdquo includes San Miguel SanNicolas and Coronados islands

ldquoPaciWc Subtropicalrdquo includes Benito Cedros Asuncioacuten and Margarita islands ldquoSouthern Gulfrdquo includes Islotes ldquoCentral Gulfrdquo includes EstebanIsland and Cantiles rookery on Aacutengel de la Guarda Island ldquoNorthern Gulfrdquo includes Granito Lobos and San Jorge islands

Corresponding P values (upper matrix) were calculated from 16000 random permutation tests The null hypothesis of panmixia (no structure) wasrejected in all cases at P lt 005

PaciWc Temperate PaciWc Subtropical Southern Gulf Central Gulf Northern Gulf

PaciWc Temperate FST lt0001 lt0001 lt0001 lt0001

ST lt0001 0015 lt0001 lt0001

PaciWc Subtropical 01329 0002 lt0001 lt0001

01063 0003 lt0001 lt0001

Southern Gulf 01956 00912 lt0001 0001

01072 01443 lt0001 0003

Central Gulf 02418 01475 01089 0032

02228 02462 02191 0024

Northern Gulf 01500 00954 00619 00243

01020 01398 01002 00446

123

Mar Biol (2009) 1561375ndash1387 1383

haplotypes derived from H29 and H34 including haplo-types H40 and H41 which link Z californianus withZ wollebaeki Furthermore haplotype diversity was highestin all three Gulf populations

The Gulf of California in its current conWguration hasbeen in existence since raquo37 Ma (Jacobs et al 2004) andtherefore a possible interpretation for the patterns of haplo-type diversity in our MSN is that the Gulf was initiallycolonized by eastern North PaciWc animals which laterdispersed to Galaacutepagos An alternative and more plausibleinterpretation involves a more complex scenario in whichNorth PaciWc marine fauna colonized and evolved in isola-tion inside the Gulf during times of inhospitable conditionsin the outer PaciWc coast (eg Bernardi et al 2003 Jacobset al 2004) resulting from strong oscillations in upwellingand sea level such as occurred around the middle Pleisto-cene transition peaking raquo09 Ma (Clark et al 2006 Law-rence et al 2006) Once conditions became favorable againin the late Pleistocene animals from the Gulf recolonizedthe PaciWc giving rise to the present-day populations(Jacobs et al 2004) Under this ldquorefugiumrdquo hypothesis forNorth PaciWc marine fauna Zalophus populations from theGulf of California would have been the source for both theGalaacutepagos and PaciWc populations An analogous vicariantprocess occurring in insular and continental refugia southof the North American and Eurasian ice sheets during thePlio-Pleistocene has been invoked by Harlin-Cognato et al(2006) to explain the phylogeography of Steller sea lions

Haplotypes in the clusters derived from H29 and H34would have evolved in the Gulf of California during thetime when the species was restricted to this area Theabsence of H29 and H34 in the Gulf could be explained bylower frequencies in the Gulf or by their disappearanceafter migration had occurred back to the PaciWc The lowerhaplotype diversity found in the PaciWc would further sup-port a more recent origin of PaciWc populations This ldquorefu-giumrdquo hypothesis is also supported by a recentmorphological study across the breeding range of Zalophusreporting a cline in the presence of double or triple roots inthe postcanines andor the presence of a sixth postcaninesuch that the highest frequencies occur in Galaacutepagosanimals followed by those in the Gulf of California andWnally those in the PaciWc (Aurioles-Gamboa et al 2000)

Population structure within California sea lions

The pattern of genetic variation found in this study not onlyconWrms previous results regarding the genetic isolation ofsea lions in the Gulf of California (Maldonado et al 1995Bowen et al 2006) but it also provides evidence for latitu-dinal structuring in the PaciWc populations Further thehigh degree of genetic diVerentiation among the Wve puta-tive populations identiWed (FST = 0024ndash0242 ST =

0045ndash0246 Table 3) is up to Wve times higher than thatreported for Steller sea lions (FST = 005 Bickham et al1996) in most pairwise comparisons This population-leveldiVerentiation is in general agreement with the studies ofpopulation structure among California sea lions based oncranial morphometrics (Zavaleta-Lizaacuterraga 2003) feedinghabits and trophic level (Garciacutea-Rodriacuteguez and Aurioles-Gamboa 2004 Porras-Peters et al 2008) heavy metal con-centrations (Elorriaga-Verplancken and Aurioles-Gamboa2008) diseases (Szteren 2006) and population trends(Gonzaacutelez-Suaacuterez et al 2006 Szteren et al 2006) althoughsome of the boundaries vary among the various studiesDespite the well-known capability of individual Californiasea lions for long-distance travel our population structureresults are consistent with the strong philopatric behaviordisplayed by the species not only in reproductive females(Riedman 1990) but also in males (Hernaacutendez-Camachoet al 2008a) Additional evidence for a low reproductiveexchange among rookeries is the large diVerence in chlori-nated hydrocarbon contents in the blubber of California sealions from Southern California USA and Ensenada BajaCalifornia Meacutexico (Kannan et al 2004 Del Toro et al2006) separated by only 350 km

Ecological studies of Z californianus (Garciacutea-Rodriacuteguezand Aurioles-Gamboa 2004 Espinosa de los Reyes 2007Porras-Peters et al 2008) and Z wollebaeki (Wolf et al2008) suggest that inter-population diVerentiation may berelated to diVerences in feeding habits ldquoNorthern GulfrdquoldquoCentral Gulfrdquo and Galaacutepagos populations have foragingdistances from the rookeries of middot20 km (Garciacutea-Rodriacuteguezand Aurioles-Gamboa 2004 Kooyman and Trillmich 1986Villegas-Amtmann et al 2008) In contrast foraging dis-tances in San Miguel Island (ldquoPaciWc Temperaterdquo popula-tion) have been reported at 70ndash100 km (Antonelis et al1990 Melin and DeLong 1999) It is likely that these diVer-ences are driven by prey distribution and abundance indiVerent oceanographic regimes

The boundaries among Z californianus populations areconsistent with the major oceanographic patterns in theregion The ldquoPaciWc Temperaterdquo population is containedwithin a recirculation cell of the California Current knownas the Southern California Eddy and is separated from themore open upwelling-dominated coast of northern BajaCalifornia by the Ensenada Front (Hickey 1998 Santamariacuteadel Aacutengel et al 2002) About halfway down the peninsulaand within the range of the ldquoPaciWc Subtropicalrdquo populationthe prominent headlands of Punta Eugenia and Cape SanLaacutezaro induce Xow instabilities that result in dynamic eddiesand jets forming at these locations (Hickey 1998 Espinosa-Carreoacuten et al 2004) Inside the Gulf of California fouroceanographic regimes can be distinguished (1) the shallownorthern Gulf (2) the tidally energetic islands and sillsregion (3) the central deeper gulf and (4) the mouth region

123

1384 Mar Biol (2009) 1561375ndash1387

(Kahru et al 2004) While we did not sample the smallbreeding rookeries found in their third region (San PedroMaacutertir San Pedro Nolasco and Faralloacuten de San Ignacio) ourldquoNorthern Gulfrdquo ldquoCentral Gulfrdquo and ldquoSouthern Gulfrdquo pop-ulations correspond well with the Wrst second and fourthoceanographic regions of Kahru et al (2004) respectively

The behavior of adult females is a strong force in theprocess of population isolation In California sea lionsfemales return to their natal beach to give birth and nursetheir young Lactation may last for a year or more (Petersonand Bartholomew 1967 Newsome et al 2006) and afemale may give birth to a pup each year almost withoutinterruption from 4 to 12 years of age (Hernaacutendez-Camachoet al 2008b) This long and nearly continuous period ofmaternal investment is the driver for the observed pattern ofphilopatry in adult females who needing to forage andreturn expeditiously to nurse their pups develop speciWcfeeding habits and strategies adapted to local conditions asmany recent ecological studies have shown (Garciacutea-Rodriacute-guez and Aurioles-Gamboa 2004 Espinosa de los Reyes2007 Porras-Peters et al 2008 Wolf et al 2008) The vari-ation in the oceanographic regimes noted above correlatewell with the sea lion population clusters identiWed in thisstudy Thus it is likely that the oceanographic regionswithin the eastern North PaciWc support diVerent feedinggrounds to which females with dispersal constrained by theneed to return to the rookery have adapted for a periodlong enough to create the present genetic structure

Management implications

It is important to consider the vulnerability of the diVerentZalophus population for management purposes Specialattention is warranted for sea lions in the Gulf of Californiabecause of their lower abundance (31393 Aurioles-Gamboaand Zavala-Gonzaacutelez 1994) and greater populationstructure when compared to PaciWc populations (75000ndash85000 in Meacutexico and 238000ndash241000 in the UnitedStates Lowry and Maravilla-Chaacutevez 2002) Also ldquoNorth-ern Gulfrdquo and ldquoCentral Gulfrdquo populations may be especiallyvulnerable because of higher reported frequencies of temp-oro-mandibular osteoarthritis and mandibular osteomielitiscompared to colonies along the western Baja Californiacoast (Aurioles-Gamboa et al 2009) The high frequency oftooth erosion in sea lions of the Gulf of California (Labradaet al 2007) may be linked to these diseases Leptospirosisa sea lion disease caused by the bacteria Leptospira interro-gans results in early births abortions and kidney problemsand occurs in highest frequencies in the northern Gulf ofCalifornia (Acevedo-Whitehouse et al 2003) Along thePaciWc coast the highest susceptibility to hookworm(Uncinaria spp) infection has been shown to be an impor-tant cause of pup mortality at San Miguel Island (ldquoPaciWc

Temperaterdquo population) (Acevedo-Whitehouse et al 2006)which is one of the two largest California sea lion breedingrookeries (the other being San Nicolas Island) and producesnearly 42 of pups in the United States population (Lowryand Maravilla-Chaacutevez 2002) These results indicate thatmore studies about diseases in natural populations areneeded in order to ascertain the extent to which selectivemortality occurs and what the consequences are withrespect to the maintenance of genetic variation of the popu-lation (Acevedo-Whitehouse et al 2006)

The strong intraspeciWc structure within the California sealion with Wve distinct populations in the eastern NorthPaciWc probably reXects the adaptability of the species tolocal and regional environmental conditions These popula-tions may be considered diVerent ldquoManagement Unitsrdquo afundamental concept for proper short-term management andthe logical unit for population monitoring and demographicstudies (Moritz 1994) The support for Z wollebaeki as a sep-arate species is also timely given the current concerns overthe impacts of climate variability widespread disease andhuman interactions on a rapidly declining population (Salazar2002 Wolf et al 2007 Salazar and Michuy 2008) EVortstoward strengthened conservation strategies for this speciesshould be a priority within local and regional plans

Acknowledgments Tissue samples were collected in Meacutexico underpermit No DOO7508106-97 from the Instituto Nacional de Ecologiacuteain California under permit No 1026 from the US Department ofCommerce and in Galaacutepagos Ecuador under permit No PC-009-99from the Galaacutepagos National Park (and permit No 017-00 for sampleexport) Additional sea lion samples from San Miguel Island werekindly provided by S Melin (US National Marine Fisheries Service)and from Galaacutepagos by SK Salazar (Charles Darwin ResearchStation) Work in Meacutexico was facilitated by A Zavala and O Maravillaand carried out on Mexican Navy ships G Heckel L Inclaacuten andML Anoge participated during the cruise We acknowledge theinvaluable logistical support provided by the Charles Darwin ResearchStation in Galaacutepagos (through P Robayo) Thanks to the welcomeprovided by A Dizon and staV at the SWFSC Marine MammalGenetics Laboratory and to C Le Duc K Robertson and J Hyde fortheir assistance in the lab YS had grants from the Mexican NationalScience Foundation (CONACyT) Alstom Power (Rosarito Meacutexico)and the Universidad Autoacutenoma de Baja California DMP wassupported by award No N00014-05-1-0045 from the US OYce ofNaval Research National Oceanographic Partnership ProgramSupplemental funding for Galaacutepagos sample export was provided bythe Protected Resources Division of the SWFSC (through RLBrownell Jr) Earlier drafts of the manuscript beneWted from commentsby G Heckel We thank to two anonymous reviewers for their valuablesuggestions and comments The experiments comply with the currentlaws of the USA Meacutexico and Ecuador

References

Acevedo-Whitehouse K de la Cueva H Gulland FMD Aurioles-Gamboa D Arellano-Carbajal F Suaacuterez-Guemes F (2003)Evidence of Leptospira interrogans infection in California sealion pups from the Gulf of California J Wildl Dis 39145ndash151

123

Mar Biol (2009) 1561375ndash1387 1385

Acevedo-Whitehouse K Spraker TR Lyons E Melin SR Gulland FDelong RL Amos W (2006) Contrasting eVects of heterozygosityon survival and hookworm resistance in California sea lion pupsMol Ecol 151973ndash1982 doi101111j1365-294X200602903x

Akst EP Boersma PD Fleischer RC (2002) A comparison of geneticdiversity between the Galaacutepagos penguin and the Magellanicpenguin Cons Gen 3375ndash383 doi101023A1020555303124

Amos B Hoelzel AR (1991) Long-term preservation of whale skin forDNA analysis In Hoelzel AR (ed) The genetic ecology of whalesand dolphins Special publication no 13 of the IWC Cambridgepp 99ndash103

Antonelis GA Stewart BS Perryman WS (1990) Foraging character-istics of female northern fur seals (Callorhinus ursinus) andCalifornia sea lions (Zalophus californianus) Can J Zool 68150ndash158 doi101139z90-022

Arnason U Johnsson E (1992) The complete mitochondrial DNAsequence of the harbor seal Phoca vitulina J Mol Evol 34493ndash505 doi101007BF00160463

Aurioles-Gamboa D Diacuteaz-Guzmaacuten C Le Boeuf BJ Casper D (2009)Temporomandibular arthritis and osteomielitis in California sealions (Zalophus califorinianus) J Zoo Wildl Med (in press)

Aurioles-Gamboa D Le Boeuf BJ (1991) EVects of the El Nintildeo 1982ndash1983 on California Sea Lions in Meacutexico In Trillmich F Ono KA(eds) Pinnipeds and El Nintildeo responses to environmental stressEcological studies 88 Springer New York pp 112ndash118

Aurioles-Gamboa D Zavala-Gonzaacutelez A (1994) Algunos factoresecoloacutegicos que determinan la distribucioacuten y abundancia del lobomarino Zalophus californianus en el Golfo de California CiencMar 20535ndash553

Aurioles-Gamboa D Sinsel F Fox C Alvarado E Maravilla-ChaacutevezO (1983) Winter migration of subadult male California sea lionsZalophus californianus in the southern part of Baja CaliforniaJ Mammal 64513ndash518 doi1023071380369

Aurioles-Gamboa D Castillo S Contreras AI Barnes LG (2000)Patron latitudinal en caracteres dentarios del lobo marino deCalifornia en Norteameacuterica (Zalophus californianus californi-anus) In XXV Reunioacuten Internacional para el Estudio de losMamiacuteferos Marinos La Paz Baja California Sur Meacutexico

Baker AR Loughlin TR Burkanov V Matson CW Trujillo RGCalkins DG WickliVe JK Bickham JW (2005) Variation ofmitochondrial control region sequences of Steller sea lions thethree-stock hypothesis J Mammal 861075ndash1084 doi10164404-MAMM-A-113R11

Bartholomew GA (1967) Seal and sea lion populations of the ChannelIslands In Philbrick RN (ed) Proceedings of the symposium onthe biology of the Calif Islands Santa Barbara Botanic GardenSanta Barbara CA pp 229ndash243

Beheregaray LB Havill N Gibbs J Fritts T Powell JR Caccone G(2004) Giant tortoises are not so slow rapid diversiWcation andbiogeographic consensus in the Galaacutepagos Proc Natl Acad SciUSA 1016514ndash6519 doi101073pnas0400393101

Bernardi G Findley L Rocha-Olivares A (2003) Vicariance and dis-persal across Baja California in disjunct marine Wsh populationsEvol Int J Org Evol 571599ndash1609

Bickham JW Patton JC Loughlin TR (1996) High variability forcontrol-region sequences in a marine mammal implications forconservation and biogeography of Steller sea lion (Eumetopiasjubatus) J Mammal 7795ndash108 doi1023071382712

Boness DJ Oftedal OT Ono KA (1991) The eVect of El Nintildeo on pupdevelopment in the California Sea Lion (Zalophus californianus)I Early postnatal growth In Trillmich F Ono KA (eds) Pinni-peds and El Nintildeo responses to environmental stress Ecologicalstudies 88 Springer New York pp 173ndash179

Bowen L Aldridge BM DeLong R Melin S Godiacutenez C Zavala AGulland F Lowenstine L Stott JL Johnson ML (2006)MHC gene conWguration variation in geographically disparate

populations of California sea lions (Zalophus californianus) MolEcol 15529ndash533 doi101111j1365-294X200502612x

Brassel KE Reif D (1979) A procedure to generate Thiessen polygonsGeogr Anal 11289ndash303

Browne RA Anderson DJ Houser JN Cruz F Glasgow KJ HodgesCN Massey G (1997) Genetic diversity and divergence of endan-gered Galaacutepagos and Hawaiian petrel populations Condor99812ndash815 doi1023071370494

Brunner S (2004) Fur seals and sea lions (Otariidae) identiWcation ofspecies and taxonomic review Syst Biodivers 1339ndash439doi101017S147720000300121X

Capella JJ Floacuterez-Gonzaacutelez L Falk-Fernaacutendez P Palacios DM (2002)Regular appearance of otariid pinnipeds along the ColombianPaciWc coast Aquat Mamm 2867ndash72

Cenami Spada E Hanggi EB Schusterman RJ (1991) Variation invocalizations and individual recognition in two subspecies ofCalifornia sea lions In Abstracts of the 9th Biennial conferenceon the biology of marine mammals Chicago Illinois

Christie DM Duncan RA McBirney AR Richards MA White WMHarp KS Fox CG (1992) Drowned islands downstream from theGalapagos hotspot imply extended speciation times Nature355246ndash248 doi101038355246a0

Clark PU Archer D Pollard D Blum JD Rial JA Brovkin V Mix ACPisias NG Roy M (2006) The middle Pleistocene transitioncharacteristics mechanisms and implications for long-termchanges in atmospheric pCO2 Quat Sci Rev 253150ndash153184doi101016jquascirev200607008

Crandall KA Templeton AR (1993) Empirical tests of some predic-tions from coalescent theory with applications to intraspeciWcphylogeny reconstruction Genetics 134959ndash969

Davies JL (1958) The Pinnipedia an essay in zoogeography GeogrRev 48474ndash493 doi102307211670

Del Toro L Heckel G Camacho-Ibar VF Schramm Y (2006) Califor-nia sea lion (Zalophus californianus californianus) have lowerchlorinated hydrocarbon contents in Baja California Meacutexico thanin Southern California USA Environ Pollut 14283ndash92doi101016jenvpol200509019

DeLong RL Antonelis GA Oliver CW Stewart BS Lowry MCYochem PK (1991) EVects of the 1982ndash83 El Nintildeo on severalpopulation parameters and diet of California Sea Lions on theCalifornia Channel Islands In Trillmich F Ono KA (eds) Pinni-peds and El Nintildeo responses to environmental stress Ecologicalstudies 88 Springer New York pp 166ndash172

Demeacutereacute TA Berta A Adam PJ (2003) Pinnipedimorph evolutionarybiogeography Bull Am Mus Nat Hist 27932ndash76 doi1012060003-0090(2003)279lt0032Cgt20CO2

Dickerson BR Ream RR Vignieri SN Bentzen P Antonelis GA(2008) Population Structure as Revealed by mtDNA and Micro-satellites in Northern Fur Seals Callorhinus ursinus In AlaskaMarine Science Symposium Anchorage AK

Eibl-Eibesfeldt I (1984) The Galapagos seals Part 1 Natural history ofthe Galapagos sea lion (Zalophus californianus wollebaekiSiverstsen) In Perry R (ed) Key environments Galapagos Perg-amon Press Oxford pp 207ndash214

Elorriaga-Verplancken F Aurioles-Gamboa D (2008) Trace metalconcentrations in the hair of Zalophus californianus pups andtheir relation to feeding habits Biol Trace Elem Res 126148ndash164 doi101007s12011-008-8186-8

Espinosa de los Reyes MG (2007) Variabilidad espacial de la dieta dellobo marino de California (Zalophus californianus californianusLesson 1828) MSc thesis CICESE Ensenada BC Meacutexico

Espinosa-Carreoacuten TL Strub PT Beier E Ocampo-Torres F Glaxiola-Castro G (2004) Seasonal and interannual variability of satellite-derived chlorophyll pigment surface height and temperature oVBaja California J Geophys Res 109(C03039) doi1010292003JC002105

123

1386 Mar Biol (2009) 1561375ndash1387

ExcoYer L Smouse P (1994) Using allele frequencies and geographicsubdivision to reconstruct gene genealogies within a speciesMolecular variance parsimony Genetics 136343ndash359

ExcoYer L Smouse P Quattro J (1992) Analysis of molecular vari-ance inferred from metric distances among DNA haplotypesapplication to human mitochondrial DNA restriction data Genet-ics 131479ndash491

Francis JM Heath CB (1991) Population abundance pup mortalityand copulation frequency in the California sea lion in relation tothe 1983 El Nintildeo on San Nicolas Island In Trillmich F Ono KA(eds) Pinnipeds and El Nintildeo responses to environmental stressEcological studies 88 Springer New York pp 119ndash128

Friesen VL Anderson DJ Steeves TE Jones H Schreiber EA (2002)Molecular support for species status of the Nazca booby (Sulagranti) Auk 119(3)820ndash826 doi1016420004-8038(2002)119[0820MSFSSO]20CO2

Gallo-Reynoso JP Soloacuterzano-Velasco JL (1991) Two new sightings ofCalifornia sea lions on the southern coast of Meacutexico Mar MammSci 796 doi101111j1748-76921991tb00557x

Garciacutea-Rodriacuteguez FJ Aurioles-Gamboa D (2004) Spatial and temporalvariation in the diet of the California sea lion (Zalophus californi-anus) in the Gulf of California Meacutexico Fish Bull (WashingtonDC) 10247ndash62

Geist D (1996) On the emergence and submergence of the GalaacutepagosIslands Not Galap 565ndash9

Gonzaacutelez-Suaacuterez M McCluney K Aurioles-Gamboa D Gerber LR(2006) Incorporating uncertainty in spatial structure for viabilitypredictions a case study of California sea lions Anim Conserv9219ndash227 doi101111j1469-1795200600022x

Grehan J (2001) Biogeography and evolution of the Galapagos inte-gration of the biological and geological evidence Biol J Linn SocLond 74(3)267ndash287 doi101006bijl20010576

Guo S Thomson E (1992) Performing the exact test of HardyndashWein-berg proportion for multiple alleles Biometrics 48361ndash372doi1023072532296

Harlin-Cognato A Bickham JW Loughlin TR Honeycutt RL (2006)Glacial refugia and the phylogeography of Stellerrsquos sea lion(Eumetopias jubatus) in the North PaciWc J Evol Biol 19955ndash969 doi101111j1420-9101200501052x

Heath CB (2002) California Galapagos and Japanese Sea Lions InPerrin WF Wuumlrsig B Thewissen JGM (eds) Enciclopedia ofMarine Mammals Academic Press San Diego pp 180ndash186

Hernaacutendez-Camacho C Aurioles-Gamboa D Laake J Gerber L(2008a) Survival rates of the California sea lion Zalophus califor-nianus in Meacutexico J Mammal 891059ndash1066 doi10164407-MAMM-A-4041

Hernaacutendez-Camacho C Aurioles-Gamboa D Gerber L (2008b) Age-speciWc birth rates of California sea lions (Zalophus californi-anus) in the Gulf of California Meacutexico Mar Mamm Sci 24664ndash676 doi101111j1748-7692200800199x

Hey J (1991) The structure of genealogies and the distribution of WxeddiVerences between DNA sequence samples from natural popula-tions Genetics 128831ndash840

Hickey BM (1998) Coastal oceanography of western North Americafrom the tip of Baja California to Vancouver Island coastalsegment (8E) In Robinson AR Brink KH (eds) The sea vol 11pp 345ndash393 Wiley New York pp 345ndash393

Ho SYW Phillips MJ Cooper A Drummond AJ (2005) Time depen-dency of molecular rate estimates and systematic overestimationof recent divergence times Mol Biol Evol 221561ndash1568doi101093molbevmsi145

Jacobs DK Haney TA Louie KD (2004) Genes diversity and geo-logic process on the PaciWc coast Annu Rev Earth Planet Sci32601ndash652 doi101146annurevearth32092203122436

Kahru M Marinone SG Lluch-Cota SE Pareacutes-Sierra A Mitchell BG(2004) Ocean-color variability in the Gulf of California scales

from days to ENSO Deep Sea Res Part II Top Stud Oceanogr51139ndash146 doi101016jdsr2200304001

Kannan K Kajiwara N Le Boeuf BJ Tanabe S (2004) Organochlorinepesticides and polychlorinated biphenyls in California sea lionsEnviron Pollut 131425ndash434 doi101016jenvpol200403004

King JE (1983) Seals of the World National History British MuseumCornell University Press New York

Kooyman GL Trillmich F (1986) Diving behavior of Galapagos sealions In Gentry RL Kooyman GL (eds) Fur sealsmdashmaternalstrategies on land and at sea Princeton University Press Princetonpp 209ndash220

Koutavas A Sachs JP (2008) Northern timing of deglaciation in theeastern equatorial PaciWc from alkenone paleothermometry Pale-oceanography 23PA4205 doi1010292008PA001593

Kumar S Tamura K Jakobsen IB Nei M (2001) Mega version 21Molecular Evolutionary Genetics Analysis software ArizonaState University Tempe

Labrada MV Aurioles-Gamboa D Castro-Gonzaacutelez MI (2007) Rela-tion of dental wear to the concentrations of essential minerals inteeth of the California sea lion Zalophus californianus californi-anus Biol Trace Elem Res 114107ndash126

Lawrence KT Liu Z Herbert TD (2006) Evolution of the eastern trop-ical PaciWc through Plio-Pleistocene glaciation Science 32179ndash83 doi101126science1120395

Le Boeuf BJ Aurioles-Gamboa D Condit R Fox C Gisiner RRomero R Sincel F (1983) Size and distribution of the Californiasea lion in Meacutexico Proc Calif Acad Sci 4377ndash85

Lea DW Pak DK Belanger CL Spero HJ Hall MA Shackleton NJ(2006) Paleoclimate history of Galaacutepagos surface waters over thelast 135000 yr Quat Sci Rev 251152ndash1167 doi101016jquascirev200511010

Lesson RP (1828) Phoque In Bory de Sanint-Vicent JBGM (ed)Dictionaire Classique drsquoHistoire Naturelle Paris Rey et GravierParis

Lluch-Belda D (1969) El lobo marino de California Zalophus califor-nianus (Lesson 1828) Allen 1880 observaciones sobre suecologiacutea y explotacioacuten Instituto Mexicano de Recursos NaturalesRenovables Meacutexico

Lowry MS Maravilla-Chaacutevez MO (2002) Abundancia de lobos mari-nos de California (Zalophus californianus) en Baja CaliforniaMeacutexico y Estados Unidos de America durante julio y agosto del2000 In XXVII Reunioacuten Internacional para el estudios de losMamiacuteferos Marinos Veracruz Ver Meacutexico

Lowry MS Boveng P DeLong RJ Oliver CW Stewart BS De Anda-Delgado H Barlow J (1992) Status of the California sea lion(Zalophus californianus californianus) population in 1992NOAA-TM-NMFS-SWFSC-LJ-92ndash32 La Jolla pp 1ndash24

Maldonado JE Orta-Daacutevila F Stewart BS GeVen E Wayne RK(1995) IntraspeciWc genetic diVerentiation in California sea lions(Zalophus californianus) from Southern California and the Gulfof California Mar Mamm Sci 1146ndash58 doi101111j1748-76921995tb00273x

Manel S Schwartz MK Luikart G Taberlet P (2003) Landscapegenetics combining landscape ecology and population geneticsTrends Ecol Evol 18189ndash197 doi101016S0169-5347(03)00008-9

Maniscalco JM Wynne K Pitcher KW Hanson MB Melin SR Atkin-son S (2004) The occurrence of California sea lions in AlaskaAquat Mamm 30427ndash433 doi101578AM3032004427

Manni F Gueacuterard E Heyer E (2004) Geographic patterns of (geneticmorphologic linguistic) variation how barriers can be detectedby ldquoMonmonierrsquos algorithmrdquo Am J Hum Biol 76173ndash190doi101353hub20040034

Melin SR DeLong RL (1999) At-sea distribution and diving behaviorof California sea lion females from San Miguel Island CaliforniaIn Browne DR Mitchell KL Chaney HW (eds) Proceeding of

123

Mar Biol (2009) 1561375ndash1387 1387

the Wfth California islands symposium Santa Barbara Museum ofNatural History Santa Barbara pp 402ndash407

Moritz C (1994) DeWning ldquoEvolutionarily SigniWcant Unitsrdquo forconservation Trends Ecol Evol 9373ndash375 doi1010160169-5347(94)90057-4

Morris PA Oliver GW Elliott J Melin SR DeLong RL (1999) ElNintildeo 1998 and changes in California sea lion Zalophus californi-anus use of Antildeo Nuevo island In Abstracts of the 13th BiennialConference on the Biology of Marine Mammals Wailea Hawaii

Newsome SD Etnier MA Koch PL Aurioles-Gamboa D (2006)Using Carbon and Nitrogen isotopes to investigate reproductivestrategies in Northeast PaciWc otariids Mar Mamm Sci 22556ndash572 doi101111j1748-7692200600043x

Palacios DM Feacutelix F Floacuterez-Gonzaacutelez L Capella JJ Chiluiza DHaase BJM (1997) Sightings of Galaacutepagos sea lions (Zalophuscalifornianus wollebaeki) on the coasts of Colombia and EcuadorMammalia 61114ndash116

Peterson RS Bartholomew GA (1967) The natural history andbehavior of the California sea lion Special publication no 1 TheAmerican society of mammalogists pp 1ndash79

Porras-Peters H Aurioles-Gamboa D Cruz-Escalona VH Koch PL(2008) Position breadth and trophic overlap of sea lions (Zalo-phus californianus) in the Gulf of California Meacutexico MarMamm Sci 24554ndash576 doi101111j1748-7692200800197x

Rassmann K (1997) Evolutionary age of the Galapagos iguanas pre-dates the age of the present Galapagos Islands Mol PhylogenetEvol 7158ndash172 doi101006mpev19960386

Repenning CA Ray CE Grigourescou D (1979) Pinniped biogeogra-phy In Gray J Boucot AJ (eds) Historical biogeography platetectonics and the changing environment Proceedings of the 37thAnnual Biology Colloquium and selected papers Oregon StateUniversity Corvallis Oregon pp 357-369

Riedman M (1990) The Pinnipeds Seals sea lions and walrusesUniversity of California Press BerkeleyLos Angeles

Salazar S (2002) Lobo Marino y Lobo Peletero In Danulat E EdgarGJ (eds) Reserva Marina de Galaacutepagos Linea Base de la biodiv-ersidad Fund Charles DarwinSer Par Nac Galap Santa CruzGalaacutepagos Ecuador pp 267ndash290

Salazar S Bustamante RH (2003) EVects of the 1997ndash98 El Nintildeo onpopulation size and diet of the Galaacutepagos sea lion (Zalophuswollebaeki) Not Galap 6240ndash45

Salazar S Michuy V (2008) Estado poblacional y conservacioacuten de lospinniacutepedos de las islas Galaacutepagos XIII Reunioacuten de Trabajo deEspecialiacutestas en Mamiacuteferos Acuaacuteticos de Ameacuterica del Sur y 7degCongreso SOLAMAC Montevideo Uruguay p 196

Sambrook J Fritsch EF Maniatis T (1989) Molecular cloning a labo-ratory manual Cold Spring Harbor Laboratory Press Cold SpringHarbor

Santamariacutea del Aacutengel E Millaacuten-Nuntildeez R Gonzaacutelez-Silvera AMuumlller-Karger F (2002) The color signature of the EnsenadaFront and its seasonal and interannual variability CCOFI Rep43155ndash161

Schneider S Roessli D ExcoYer L (2001) Arlequin version 2 A soft-ware for population genetic data analysis Genetic and BiometryLaboratory University of Geneva Switzerland

Sivertsen E (1953) A new species of sea lion Zalophus wollebaekifrom the Galapagos Islands Det Kong Nor Videnskaps SelskForh 261ndash3

Sivertsen E (1954) A survey of the eared seals (family Otariidae) withremarks on the Antarctic seals collected by MK ldquoNorwegianrdquo in1928ndash1929 Det Nor VidenskapsmdashAkad Oslo

Stewart DT Baker AJ (1994) Patterns of sequence variation in themitochondrial D-loop region of shrews Mol Biol Evol 119ndash21

Szteren D (2006) Regionalizacioacuten ecoloacutegica de las colonias reproduc-tivas de Zalophus c californianus en el Golfo de CaliforniaMeacutexico PhD thesis CICIMAR-IPN La Paz BCS Meacutexico

Szteren D Aurioles D Gerber LR (2006) Population status and trendsof the California sea lion (Zalophus californianus californianus)in the Gulf of California Meacutexico In Trites AW Atkinson SKDeMaster DP Fritz LW Gelatt TS Rea LD Wynne KM (eds)Sea Lions of the World Alaska Sea Grant College Program Low-ell WakeWeld Fisheries Symposium Series Rhode Island pp 369ndash384

Trillmich F Dellinger T (1991) The eVects of El Nintildeo on GalapagosPinnipeds In Trillmich F Ono KA (eds) Pinnipeds and El NintildeoResponses to environmental stress Ecological Studies 88Springer New York pp 66ndash74

Trillmich F Limberger D (1985) Drastic eVects of El Nintildeo on Galapa-gos pinnipeds Oecologia 6719ndash22 doi101007BF00378445

Villegas-Amtmann S Costa DP Tremblay Y Salazar S Aurioles-Gamboa D (2008) Multiple foraging strategies in a marine apexpredator the Galapagos sea lion Zalophus wollebaeki Mar EcolProg Ser 363209ndash299 doi103354meps07457

Walker PL Craig S (1979) Archaeological evidence concerning theprehistoric occurrence of sea mammals at Point Bennet SanMiguel Island Calif Fish Game 6550ndash54

Walker PL Kennett DJ Jones TL DeLong R (1999) Archaeologicalinvestigations at the Point Bennett pinniped rookery on SanMiguel Island In Brown DR Mitchell KC Chaney HW (eds)Proceedings of the Wfth California Islands symposium USDepartment of the Interior Minerals Management Service PaciWcOCS Region pp 628-632

Weber DS Stewart BS Lehman N (2004) Genetic consequences of asevere population bottleneck in the Guadalupe fur seal (Arcto-cephalus townsendi) J Hered 95144ndash153 doi101093jheredesh018

Werner R Hoernle K van den Bogaard P Ranero C von Huene RKorich D (1999) Drowned 14-my-old Galaacutepagos ArchipelagooV the coast of Costa Rica Implications for tectonic and evolu-tionary models Geology 27499ndash502 doi011300091-7613(1999)027lt0499DMYOGPgt23CO2

Wolf JBW Tautz D Trillmich F (2007) Galaacutepagos and California sealions are separate species genetic analysis of the genus Zalophusand its implications for conservation management Front Zool420 doi1011861742-9994-4-20

Wolf JBW Harrod C Brunner S Salazar S Trillmich F Tautz D(2008) Tracing early stages of species diVerentiation ecologicalmorphological and genetic divergence of Galaacutepagos sea lionpopulations BMC Evol Biol 8150 doi1011861471-2148-8-150

Wynen LP Goldsworthy SD Insley SJ Adams M Bickham JW Fran-cis J Gallo-Reynoso JP Hoelzel AR Majluf P White RWGSlade R (2001) Phylogenetic relationships within the eared seals(Otariidae Carnivora) Implications for the historical biogeographyof the Family Mol Phylogenet Evol 21270ndash284 doi101006mpev20011012

Zavala-Gonzaacutelez A Mellink E (2000) Historical exploitation of theCalifornia sea lion Zalophus californianus in Meacutexico Mar FishRev 6235ndash40

Zavaleta-Lizaacuterraga L (2003) Variaciones geograacuteWcas en morfometriacuteacraneal en machos adultos de lobo marino de California (Zalo-phus californianus) en Meacutexico MSc thesis CICIMAR-IPNMeacutexico

123