Affinities of the Hawaiian Goose Based on Two Types of Mitochondrial Dna Data

Home , Anser (bird), Dusky Canada goose, Giant Canada goose, Nene (bird)

AFFINITIES OF THE HAWAIIAN GOOSE BASED ON TWO TYPES OF MITOCHONDRIAL DNA DATA

THOMAS W. QUINN, • GERALD F. SHIELDS,2 AND ALLAN C. WILSON• •Divisionof Biochemistryand Molecular Biology, University of California,Berkeley, California 94720 USA, and 2Instituteof ArcticBiology, University of Alaska,Fairbanks, Alaska 99775-0180 USA

ABSTRACT.--Wecompared restriction fragments of mitochondrial DNA and sequencesof portionsof the cytochromeb gene of the following: the Hawaiian Goose(Nesochen sandvicensis),five subspeciesof the CanadaGoose (Branta canadensis), the PacificBlack Brant (B. berniclanigricans), and the Emperor Goose(Chen canagica). Both comparisonssupport the association of the Hawaiian Goose and the Canada Goose as sister taxa as well as the association of all four large-bodiedsubspecies of CanadaGeese as separatefrom the one small-bodied subspeciesanalyzed here. Bootstrapand winning-sitestests show theseassociations to be statisticallysignificant. The amountof sequencedivergence implies that Hawaiian andCanada Geesediverged from a common ancestorless than three million years ago and that this divergencepreceded the divergenceof large- and small-bodiedCanada Geese only slightly, if at all. Therefore,we suggesta North American origin for the Hawaiian Goose.Received 2 July 1990,accepted 27 December1990.

THE H^W^nAN Goose (Nesochensandvicensis), Both restriction and sequencingmethods of or Nene as it is called locally,is the only extant characterizingmtDNA yield traits that can be speciesof gooseendemic to Hawaii, yet its an- subjectedto parsimony analysis (e.g. Edwards cestry and geographicorigin have never been and Wilson 1990), allowing branching orders established.Within the present century, the to be testedwithout strict assumptionsof clock- goosesubfamily Anserinae has routinely been like rates of change.We comparedrestriction divided into at least five genera (Anser,Branta, fragmentsof whole mtDNA and DNA sequenc- Chen,Nesochen, and Philacte)(Johnsgard 1978, es of portions of a specificmitochondrial gene Bellrose 1980, Palmer 1976). There is disagree- of the HawaiianGoose with thoseof othergeese. ment as to whether Anser should be split into Geographic considerations influenced our several genera (e.g. Anser,Chen, and Philacte). choice of the other geese. Becauseour intent Moreover, taxonomistshave been equivocalre- was to determine the sister taxon of the Ha- garding the allocation of the Hawaiian Goose. waiian Goose and to estimate when and where Some have emphasizedmorphological differ- it may have arisen, we centered attention on encesfrom the other geeseby placing it in Ne- those geesethat inhabit western North Amer- sochen.Others have emphasizedobvious simi- ica, the Aleutian Islands, and northeastern Asia, laritiesby placingit in Branta.Berger (1972) has and that had not alreadybeen studiedby Shields proposed that Nesochenis closely related to and Wilson (1987a, b). Branta and that it arose specifically from the Canada Goose (Branta canadensis)after one or MATERIALS AND METHODS more migration events from North America. We obtainedapproximately 20 g of hearttissue from This hypothesishas awaited testing with mod- an adult Hawaiian Goose that had died at the Ho- ern methods of phylogenetic analysis. nolulu Zoo. The tissuewas immediatelyminced and Information preserved in the DNA of this maintained in cold (4øC) buffer (mannitol, sucrose, speciesmay be used to infer its relatednessto Tris-EDTA) during subsequenttransport to Fair- other extant speciesof geese. Mitochondrial banks. The combined methods of Shields and Wilson DNA (mtDNA) has been valuable for phylo- (1987a) and Carr and Griffith (1987) were used to ob- genetic reconstructionsbecause of its ease of tain purified mtDNA from this tissueand from hearts isolation, rapid evolutionary change, maternal and kidneys of a single small-bodiedCanada Goose (Taverner's Canada Goose,Branta canadensis taverneri); inheritance and apparent lack of recombination four large-bodiedCanada Geese (the LesserCanada (Wilson et al. 1985). This approach has been Goose,B.c. parvipes;the Dusky Canada Goose,B.c. particularly useful for studiesof phylogeny of occidentalis;the VancouverCanada Goose, B.c. fulva; maternal lineages in waterfowl (Kessler and the WesternCanada Goose, B.c. moffitti);the Pacific Avise 1984; Shields and Wilson 1987a, b; Van BlackBrant, B. bernicla nigricans; and the Emperor Goose, Wagner and Baker 1990;also see Discussion). Chencanagica (also known as Philactecanagica). We

585 The Auk 108:585-593. July 1991 586 QUINN,SHIELDS, AND WILSON [Auk, Vol. 108

TABLE1. Fragmentpatterns of mtDNAs from eight taxa of geese:Nesochen sandvicensis, Branta canadensis taverneri,B.c. parvipes,B.c. occidentalis,B.c. fulva, B.c. moffitti,B. bernicla,and Chencanagica.

Fragment patternsa Restriction B.c. B. C. enzyme N. sand. B. c. tav. B. c. par. B. c. occ. B. c. ful. moff. bernicla canagica AvaI A A A A A A B C AvaII A A A A A A B C BamHI A A A A A A A B BanI A B C/D/E C C C F G BglII A A A A A A B A BstUI A B C C D C E F ClaI A A A A A A B C EcoRI A A A A A A B C HhaI A B A A A A C D HincII A A A A A A B C HindIII A B A A A A C A HinfI A B B B B B C D NarI A A A / B B B B C D NciI A A A A A A B C NcoI A B B B B B C D PvuII A B A A A A C D SpeI A B B B B B C D StyI A B C C C C D E XbaI A A B B B B A C

completelist of all fragmentsizes is availableupon request. were not concernedabout larger samplesizes because (northern Atlantic and Eurasia,respectively) and past intrasubspecificvariation in CanadaGeese and Brant distributions(never in easternAsia) make it unlikely is extremelylow (Shieldsand Wilson 1987b;Shields that they could have flown to Hawaii. 1990; Van Wagner and Baker 1990). Moreover, the mtDNA diversity among Hawaiian Geesemust have RESTRICTION FRAGMENT ANALYSIS beenreduced drastically when its populationcrashed to only 17 individualsin the 1940s(Kear and Berger We comparedfragment patterns of mtDNA of the 1980).While two other extant geeseof the genusBran- Hawaiian Goose(Table 1) to those of the other taxa ta (the BarnacleGoose, B. leucopsis,and the Red-breast- of geesein severalways. We firstcompared restriction ed Goose,B. ruficollis)could have been included in fragmentsof mtDNA from the Hawaiian Goosein the this study, they are poor candidatesas ancestorsof samegels with thoseof the geeseidentified above. the Hawaiian Goosebecause their breeding ranges Thenwe comparedthese fragment patterns with those

TABLE2. Extent of sequencedifference (percent) for taxa of geesebased on restriction fragment analysisof mtDNA. a Species:Nesochen sandvicensis, Branta canadensis ssp., Branta bernicla, and Chencanagica.

1 2 3 4 5 6 7 8 9 10 11

1. N. sandvicensis -- 1.44 a 1.44 a 1.44 1.60 1.64 1.64 1.64 1.64 5.85 4.87 2. B.c. leucopareia -- 0.06c 0.11c 1.28a 1.31a 1.27 1.00c 1.30c 6.14 5.77 3. B.c. minima -- 0.05 c 1.28 a 1.31 a 1.27 0.80 c 1.10 • 6.14 5.77 4. B.c. taverneri -- 1.28 1.31 1.27 0.90 1.12 c 6.14 5.77 5. B.c. parvipes -- 0.08 0.20 0.16 0.26a 6.02 5.54 6. B.c. occidentalis -- 0.14 0.17 0.26 a 6.05 5.33 7. B.c. fulva -- 0.14 0.26a 6.11 5.32 8. B.c. moffitti -- 0.26c 6.05 5.33 9. B.c. maxima -- &10 b 5.33 10. B. bernicla -- 5.99 11. C. canagica -- • Overall thesevalues tend to be lower than comparableestimates (Van Wagnerand Baker1990), probably because the presentanalysis uses mainly 6-basecutters. bValue computedin Shieldsand Wilson (1987a). cValues computedin Shieldsand Wilson (1987b). a Values averaged. July1991] HawaiianGoose Mitochondrial DNA 587

found by Shieldsand Wilson (1987a,b) for the Aleu- tian CanadaGoose (B.c. leucopareia),the Cackling Can- ada Goose (B.c. minima), Taverner's Canada Goose, the Western Canada Goose, the Giant Canada Goose (B.c. maxima),and the Pacific Black Brant. The 19 enzymesAval, Avail, BamHI, BanI, BglII, BstUI, ClaI, EcoRI, HhaI, HinclI, HindIII, HinfI, NarI, NciI, NcoI, PvuII, SpeI,StyI and XbaI were used in the first seriesof fragment comparisons.Restriction en- donucleasedigestions were carried out under the conditions specifiedby the vendor (New England Bio- labs.). Five of these enzymes (BstUI, HhaI, HindIII, I PvuII, and XbaI) unequivocally differentiate large- bodied from small-bodied Canada Geese (Shields and 4260 Wilson 1987b),and thus they were included here as potential phylogenetically informative enzymes relative to the Hawaiian Goose. The 22 restrictionenzymes used in the secondcom- parison are listed in Shields and Wilson (1987a, b); 12 of thesecoincide with enzymesused in the first 2250 comparison.Divergence values for both data setswere computedfrom the fraction of sharedfragments ac- 1960 cording to equation 20 of Nei and Li (1979). In cases where pairwise comparisonswere not possible,we usedaverage values among small-bodied Canada Geese (first comparison)and among large-bodied Canada Geese(second comparison) to accountfor missingdata (Table 2). For phylogeneticanalysis, we inferred restriction sites from the fragment data for the taxa studied here and used the exhaustive search algo- rithm of the PAUP (PhylogeneticAnalysis Using Par- simony) computerprogram (Swofford 1989). Finally, bootstrap (Felsenstein 1985) and winning-sites tests were applied to the tree arrangementsto test for sig- nificance(Shields and Wilson 1987b,Prager and Wil- son 1988). 600 GENE AMPLIFICATIONSAND DIRECT SEQUENCING

We usedthe methodsof Gyllenstenand Eriich (1988) and Kocheret al. (1989) to amplify and sequencepor- tionsof the cytochromeb genesof the Hawaiian Goose, Taverner's Canada Goose, the Lesser Canada Goose, the Dusky Canada Goose, the Vancouver Canada Goose, the Western Canada Goose, the Pacific Black Brant, and the Emperor Goose.Two primer pairs were Fig. 1. Restriction fragment patterns of mtDNA used. The first pair (L14841 and H15149) were de- (for 6 geese)digested with the restrictionenzyme scribedby Kocher et al. (1989). The secondpair are BanI.Numbers on the left refer to the known lengths L15420 (5'-ATCCCATTCCACCCATACTACTC-3') and of phagelambda DNA fragmentsgenerated with the H15915 (5'-AACTGCAGTCATCTCCGGTTTACAA- enzyme HindIII. The fragment patternsare identified GAC-3'). In each case,L and H refer to the light and in parenthesesas follows: lanes 1-3 (C), lane 4 (A), heavy strands,respectively, and the numbersdescribe lane 5 (G), lane 6 (F). the 3' base position according to the numbering sys- tem of Anderson et al. (1981). Productsof the initial double-strandedamplifications were visualized by reaction conditions used to generate single-stranded ethidium bromide staining of 2% NuSieve agarose template differed by having one primer reducedin gels, cut from the gels, diluted (with heating) in 1 ml concentrationto 0.02/zM, increasingthe total volume of distilled water, and 1/zl of each was used as a source to 50/zl, and completing 32 cyclesof the reaction. This of template for single-strandedamplifications. The product was then dialyzed centrifugally three times 588 Q•JINN,SHIELDS, AND WILSON [Auk, Vol. 108

TABLE3. Diagnosticfragment patternsof mtDNA from eight subspeciesof CanadaGeese and the Nene. "Plus"symbols refer to the presenceof the uncut fragmentand "minus"symbols to its absence.

Small-bodieda Large-bodieda Nature of site change Enzyme (fragment sizes) leuc. min. tav. parv. occ. fulva moji. max. Nene BstUI (FnuDII) 6,700 • 4,000 + 2,700 + + + ..... + HhaI 3,350 • 2,700 + 650 + + + ...... HhaI' 1,170 • 640 + 530 + + + ...... HindIII 6,000 • 3,800 + 2,200 + + + ...... XbaI 16,500 • 9,500 + 7,000 + + + ..... + PvuII 5,800 • 4,700 + 1,100 - - - + + + + + +

See Table 2 for full scientificnames of subspecles. in Centricoh 30 (Amicon) or Ultrafree-mc (Millipore) We estimated(Table 2) the percentnucleotide tubes in a volume of 1 ml or 0.35 ml of distilled water, sequencedifference for all taxaof this studyas respectively.Approximately 50 #1 of resuspended well as the additional subspeciesof Canada templateremained in the Centricoh30 tubes,and the Geesestudied by Shieldsand Wilson(1987a, b). samplesin the Ultrafree-mctubes were resuspended Based on these values, the Hawaiian Goose is in 40 #1d H20. We sequencedby the dideoxymethod more similar to Canada Geese(mean difference, (Sanger et al. 1977) with Sequenase(United States Biochemical)kits and following the manufacturer's 1.6%) than it is to either the Emperor Goose suggestions,starting with 7 #1 of resuspendedtem- (4.9%) or to the Black Brant (5.9%). In fact, the plate,2 #1reaction buffer, and 1 #1of the primer (10 Hawaiian Gooseis only slightly more different #M), which was limiting in the single-strandedam- from subspeciesof CanadaGeese than are the plification.While someoverlapping sequences were large-bodiedsubspecies from the small-bodied obtained from the L14841 and H15149 primers, in subspecies(mean difference, 1.2%). most casessequences were determined in a single As mentioned earlier, six restriction sites un- direction.Trees were built by the parsimonymethod equivocally separatelarge-bodied from small- and testedstatistically as describedabove for restriction sites. bodied CanadaGeese (Shields and Wilson 1987b: table 2 and Shields and Helm-Bychowski 1988: fig. 2). Hawaiian Goose mtDNA does not fall RESULTS exclusivelyinto either one or the other of these FRAGMENT ANALYSIS two groups (Table 3). At two of the diagnostic sites,this gooseis identical to the small-bodied Four fragmentpatterns were observedwhen Canada Geese, whereas at the other four sites mtDNAs from six geesewere digestedwith the the oppositepattern is shown. Thus, based on enzyme BanI (Fig. 1). Fragment patterns ob- sitesthat are phylogeneticallyinformative for servedwhen mtDNAs were digestedwith the subspeciesof CanadaGeese, the Hawaiian Goose 19 enzymesare listed (Table 1). The patterns of cannotbe linked exclusivelyto either the small- the Hawaiian Goose were often identical to those bodied or the large-bodiedtypes. of CanadaGeese. In only four cases(BamHI, The mostparsimonious tree basedon analysis BglII, HindIII, and XbaI) was identity found be- of the restriction sites links the Hawaiian Goose tween the Hawaiian Goose and either the Brant with the CanadaGoose (Fig. 2: upper). Bootstrap or the Emperor Goose. Together, the 19 en- analysis(Felsenstein 1985) showedthat the Ha- zymes produced an average of approximately waiian Goose was always associatedwith the 109 fragmentsper individual (Table 1), which CanadaGoose lineage (Fig. 2: upper). meansthat we monitored, on average,approximately 109 cleavagesites per mtDNA. Because SEQUENCESOF THE CYTOCHROMEb GENE these sitesrange in size from four-baseto six- basepairs, our methodssurveyed approximate- We compared612 nucleotidesof portions of ly 4% of the mitochondrialgenome. The mean the cytochromeb genesfor the eight taxa (Fig. genomesize of the mtDNA of thesegeese (16.7 3). The considerabledistance between primers + 0.3 kb) is similar to those of other birds an- L15420 and H15915 made it difficult to sequence alyzedin the sameway (Quinn and White 1987, acrossthe entire interior region of the gene; Shieldsand Helm-Bychowski1988). hence sequenceresults are presentedin three July1991] HawaiianGoose Mitochondrial DNA 589 parts(Fig. 3: a-c). All large-bodiedCanada Geese Restriction Sites had identical sequencesacross all regions. 100 Canada occ Large-bodied subspecies Among the codons acrossall taxa, there were Canada par 100 13 variable sitesin the first position, 4 in the Canada mof secondposition, and 69 in the third position. Canada ful Changes at 16 of these sites causeamino acid small-bodied replacementsin at leastone of the compared Hawaiian taxa. Brant

Of the amino acid changes,a disproportion- Emperor ate number (10/14) occur in the transmembrane i I I I domainsof cytochromeb depictedby Howell's 4 2 0 (1989) structural model. Overall, 109/202 (54%) of the amino acids determined fall into that Percent Sequence Difference region. The transmembraneregion is highly variable in amino acid sequencein mammals DNA Sequence relativeto mostother partsof the molecule(Ir- Hawaiian win et al. 1991),and the sameappears to be true lOO in birds. Of the remaining four variable sites small-bodied (all in the outermembrane region), three are at Canada Canadalarge-bodied hypervariable sites as describedin mammals (Irwin et al. 1991).Percent sequence differences Brant for every pair of thesetaxa, as well astransition / Emperor transversionratios for all comparisons,are re- ported (Table 4). Tree for the cytochromeb gene.--Parsimony 8 6 4 2 0 analysiswas performed on the 18 informative siteselucidated by the sequenceanalysis. The Percent Sequence Difference mostparsimonious tree (Fig. 2: lower) associates Fig. 2. Two treesshowing order of branchingand the Hawaiian Goose with the small-bodied Can- geneticdistances among goosemtDNAs. Both the re- adaGoose, and that pair in turn with the large- strictiontree (upper) and the sequencetree (lower) bodiedCanada Geese. Two other treesjust one are basedon a parsimony analysis(PAUP 3.0, Swof- step longer than the most parsimoniousone ford 1989)with the EmperorGoose designated as an maintainthe samegeneral topology except that outgroup.In each case,a single most parsimonious the HawaiianGoose is associatedwith the large- treeresulted from an exhaustivesearch of all possible bodiedgeese in one,but the two typesof Can- branchingorders. Bootstrap values were averagedover ada Geese are closer to each other than to the 5 setsof 500 replicatescalculated with the Swofford program. Only those values >95% are shown (thick Hawaiian Goose in the other. From those three lines). Internal branches with lower values were col- trees,the nextmost parsimonious tree is 10steps lapsedto producepolychotomies. In the precollapsed longer. restriction tree, the Hawaiian Goose branched off be- Statisticaltesting.--Bootstrap analysis (Felsen- fore all CanadaGeese with a bootstrapvalue of 90% stein 1985) showed that of 2,500 trials, the Ha- on the stem leading to the common ancestorof Can- waiian Goosewas always associatedwith the adaGeese. By contrast, in the sequencetree, the large- CanadaGoose lineage, and within that lineage, bodied Canada Goose branched off first with a boot- was sister to the small-bodied Canada Geese strapvalue of 72%on the stemleading to the common 72% of the time. ancestor of the Hawaiian Goose and the small-bodied CanadaGoose. Each node of the tree is plotted with Another test which can be applied to this respectto the distance scale, which is based on values problemis the winning-sitesmethod used by in Tables 2 and 4. The restriction tree is based on Shieldsand Wilson (1987b)and explicitly il- inferred cleavagesites for the entire mtDNA and the lustratedby Prager and Wilson (1988). To test consistency index for informative sites is 0.81. The whether the Hawaiian Goose is closer to the sequencetree is basedon thosepositions shown for Black Brant than it is to the Canada Goose, we that part of the cytochromeb gene (Fig. 3), and the haveincluded the EmperorGoose and a (large- consistencyindex is 0.83. bodied)Canada Goose. This test(Fig. 4) shows that 13 of the 16 informativesites support the 590 QUINN,SHIELDS, AND WILSON [Auk,Vol. 108

(x)

...... fig ...... c ..... c ..... t ...... ß ...... , .... , ..... • .... c ..

(c)

L s • L L • W I L V * • L t I L T ß V G S (3 P V • H • F , BOO CT CTC TCC CAGCTC CT& TTC TG4 &TC CTA GT& C,CC G4C CTC CTC ATC CTA A• T• •A • A• •G C• GTC•A CAC Ch TTC ATC T5790

...... c ..c ... "] ]]• •]] ]] ...... ß...] ]:::: ].• .'• .;]; .... ; . • ...... c.• . .c • • . .c ...... • ......

Fig. 3. Sequencesfor threeparts of the cytochrome• gene.The second•ine showsthe DNA sequenceof a •arge-bodiedgoose (Branta canadensis occidentalis). The otherthree •arge-bodied goose sequences (B.c. parvipes, B.c. fulva, B.c. mo•tti) were a• identica• to this and are not shown. Nucieotide sequencesare numbered accordingto the humanmtDNA sequence(Anderson et aL 1981).Beiow that •ine, dotsindicate identity and "N"indicates unresolvedbases. Among the large-bodiedgeese not shown,bases that could not be resoived inciudepositions 15111-2, 15581-3 (B.c. parvipes) and positions15468, 15736, and 15739(B.c. fulva). We used the human mtDNA geneticcode (Andersonet aL 1981) to trans,atethe DNA sequencefor B.c. occidentalis, and we obtainedthe amino acid sequenceshown (top row). Basesubstitutions causing amino acid changes in othertaxa are shown by underliningthe affectedcodons. Solid circles mark 18phylogenetica•ly informative positions;16 are informative for the test in Figure 4.

Hawaiian Goose-Canada Goose alliance, and DISCUSSION that this supportis statisticallysignificant (P < 0.02). This test shows with statistical confidence Our mitochondrial search for the nearest rel- that the Hawaiian Gooselineage is closerto the atives of the Hawaiian Goose has considered Canada Goose than to either the Pacific Black four other taxa, namely the large- and small- Brantor to the EmperorGoose. Within that lin- bodied subspeciesof CanadaGeese, as well as eage, however, whether the Hawaiian Goose the Brant and Emperor Goose. In addition, split beforeor after the small-and large-bodied Shields and Wilson (1987b) provided relevant CanadaGoose divergence is not statisticallyre- informationabout three other species,all rel- solvable with this data set. atively distant from the Canada Goose in the July1991] HawaiianGoose Mitochondrial DNA 591

T^l•LE4. Percent sequencedifference (above the diagonal) and transition/transversionratios (below the diagonal)among 612 basepairs of cytochromeb genesof geese.a

Goose Hawaiian Small-bodied Large-bodied Brant Emperor Hawaiian -- 1.8 2.1 7.8 9.0 Small-bodied 9/2 -- 1.3 6.9 8.4 Large-bodied 12/1 7/1 -- 6.7 8.8 Brant 38/10 34/8 32/9 -- 10.1 Emperor 47/ 8 45 / 6 47/ 7 56/ 6 -- ' Small-bodied= Taverner'sCanada Goose; large-bodied = LesserCanada Goose, Dusky Canada Goose, Vancouver Canada Goose, and Western CanadaGoose. Since all large-bodiedsubspecies of Canadageese tested had identicalcytochrome b sequences,they are combinedhere under one category. mtDNA tree. Only the Canada Goose had Geese.--Ourfragment analysis, and that of Van mtDNA closelyrelated to the Hawaiian Goose. Wagnerand Baker(1990), extend the work of This link between the Hawaiian Goose and Shieldsand Wilson (1987a,b) to the point where CanadaGeese supports the assertionby Kear there is now restriction fragment information and Berger (1980: 23) that "Ornithologistsbe- on the extents of divergenceamong all 11 of lieve that the Canada Goose (Branta canadensis) the generallyrecognized subspecies of Canada is the closest living relative of the Hawaiian Geese. This includes some subspecificinfor- Goose and, therefore, that its ancestors came mation on 9 of the 11 subspecies.These sub- from North America." speciesfall clearlyinto two sistergroups, large- CanadaGeese winter along the westerncoast- bodied and small-bodied, which share no al regions of the continental United Statesand mtDNA types.This result contrastswith that return to Alaska and arctic Canada in spring basedon molecularstudies of proteinsencoded to breed. Thus, it is not difficult to imagine that by nucleargenes (Van Wagnerand Baker1986, a smallnumber of ancestorsof thesegeese may 1990). The nuclear evidence associatesall the have originallybecome disoriented during mi- subspeciesvery closely,similar to the situation gration and arrived on Hawaii, where they sur- reportedby Getter(1989) in Old Worldflycatch- vived without severe competition from other ers. Further, B.c. hutchinsi,a small-bodied sub- forms. During the past century alone, 27 wa- speciesthat breedsand winters far to the east terfowl speciesfrom North America have been of other small-bodiedsubspecies, is closelyakin seen on the Hawaiian islands, 24 as stragglers (in its nucleargenes) to geographicallyneigh- or chance migrants, and three (including the boring subspecieswith large bodies.This is Cackling Canada Goose) as regular winter res- probablythe resultof introgressionmediated idents (Berger 1972). In addition, two captive- by large-bodiedmales from thoseneighboring reared Aleutian Canada Geese banded and re- subspecies. leasedfrom a recovery program site on Agattu Approximatemolecular time scale.--The se- Island,Alaska (173ø40'E, 52ø30'N), on 6 Septem- quencecomparisons make it possibleto obtain ber 1979 were seen 86 days later on Roi-Namur rough estimatesof the time of the initial mi- (167ø30'E, 09ø20'N) of the Marshall Islands tochondrialradiation among Canada Geese and (Springeret at. 1986).Clearly, CanadaGeese are the time of mitochondriat divergencebetween capable of long-distancemovement over vast the lineagesleading to Hawaiian Geeseand areasof open ocean. CanadaGeese. Because of the tendencyfor tran- Given the short time scalethat we propose sitionsto accumulateso quickly that somesites for the time of origin of the Hawaiian Goose have experiencedmultiple hits, we confineour (seebelow), it would appearthat the founders analysis to transversions,which accumulate of this speciesunderwent relatively rapid mor- roughly 10 times more slowly (Kocher et at. phological changes. Prominent among these 1989).The averagenumber of transversionsby changeswere reduction in the webbing of the which the large- and small-bodiedsubspecies toesand increasedsize and strength of the legs differis 1,while thatby whichthe CanadaGoose as a responseto the relatively dry upland hab- differsfrom the Brantand EmperorGoose is 7.5. itats that the Hawaiian Goose now occupies Assumingstrictly clock-like evolution during a (Miller 1937). periodof 4-5 million years(Shields and Wilson Relationsbetween large- and small-bodied Canada 1987a),the large- and small-bodiedsubspecies 592 QUINN,SHIELDS, AND WILSON [Auk, Vol. 108

waiian Goose.Tissues were providedby ThomasRothe of the AlaskaDepartment of Fish and Game(Canada CanadaHawaiian L • 13 • BrantEmperor Geese), Jim Sedingerof the Universityof Alaska(Em- peror Geese),James Hawkings of the CanadianWild- life Service and Robert Bromley of RenewableRe- sources,Northwest Territories (Black Brant). Judy Reed Hawaiian 3 Canada L isolated the mtDNAs and performed the fragment analysis.We thank ThomasD. Kocher, Janet R. Kor- Emperor• • Brantnegay and Svante P/i•ibowho designed and made primers available to us. Matt Hare provided advice on the phylogeneticanalyses. A portion of this work was funded by a grant from the Alaska Department of Fish and Game, the U.S. Forest Service and the U.S. HawaiianBrant • 0 •CanadaEmperor L Fish and Wildlife Serviceto Shields.Polymerase chain reactionand DNA sequencingwork were funded by Fig. 4. Winning-sites test applied to the cyto- an A. P. Sloan fellowship to Quinn. We thank Allan chromeb genesof four goosetaxa. Baker, Scott Edwards, Shannon Hackett, and Ellen Prager for valuable commentswhich improved this of Canada Geese had a common mitochondrial paper. ancestor0.6 million years ago. Similarly, we

estimatean apparent time of commonancestry LITERATURE CITED of 0.9 million yearsfor the Hawaiian Gooseand Canada Goose. However, there are large sto- AMERICAN ORNITHOLOGISTS' UNION. 1983. Checklist chasticerrors in such estimates,which prevent of North American birds, 6th ed. Washington, us from ruling out times as great as 3 million D.C., Am. Ornithol. Union. yearsfor thesedivergences. Nevertheless, these ANDERSON,S., A. T. BANKIER,B. G. BARRELL,M. H. L. DEBRUIJN, A. R. COULSON,J. DROUIN, I. C. EPERON, figures are well accommodated within the D. P. NIERLICH, B. A. ROE, F. SANGER, P. H. known geologicalrecord for Hawaii (McDou- SCHREmR,A. J. H. SMrrH, R. STADEN,& I. G. YOUNG. gall 1964, Macdonald and Abbott 1970). 1981. Sequenceand organizationof the human Taxonomicrecommendations.--Although it may mitochondrialgenome. Nature 290: 457-465. be tempting to reclassifythe Hawaiian Goose BELLROSE,F. C. 1980. Ducks, geeseand swans of as a member of the genus Branta,we feel that North America.Harrisburg, Pennsylvania, Stack- suchan action would be premature. In the first pole Books. place, the mitochondrial trees pertain only to BERGER,A. J. 1972. Hawaiian birdlife. Honolulu, the maternal lineage. When dealing with lin- Univ. Hawaii Press. eages that have diverged within the past few CARR,S. M., & O. M. GRIFFITH.1987. Rapid isolation of animal mitochondrial DNA in a small fixed- million years,the possibilityexists that nuclear angle rotor at ultrahigh speed.Biochemical Ge- gene comparisonscould reveal a more complex netics 25: 385-390. picture (cf. Shields and Wilson 1987b, Gelter EDWARDS,S. V., & A. C. WILSON. 1990. Phylogenet- 1989,Van Wagner and Baker 1990). Second,we ically informative length polymorphismand se- do not wish to overlook the magnitude of the quencevariability in mitochondrialDNA of Aus- phenotypic differencesbetween the Hawaiian tralian songbirds (Pomatostomus).Genetics 126: Gooseand Canada Geese.If quantitative mor- 695-711. phometric comparisonsof the sort done by Van FELSENSTEIN,J. 1985. Confidencelimits on phylog- Wagnerand Baker(1990) on CanadaGeese were enies:an approachusing the bootstrap.Evolution 39: 783-791. extendedto other geese,the extent of this phe- GELTER,H. P. 1989. Genetic and behavioral differ- notypic difference could be evaluated objec- entiation associatedwith speciationin the fly- tively and quantitatively.Such an evaluationis catchers,Ficedula hypoleuca and F. albicollis.Ph.D. advisable before deciding whether cladistic dissertation,Sweden, Uppsala Univ. considerationsshould take precedencein the GYLLENSTEN,U. B., & H. A. ERLICH. 1988. Generation classification of the Hawaiian Goose. of single-strandedDNA by the polymerasechain reactionand its applicationto the direct sequenc- ACKNOWLEDGMENTS ing of the HLA-DQA locus.Proc. Natl. Acad. Sci. USA 85: 7652-7656. We thank Rebecca Cann and Leonard Freed of the HOWELL,N. 1989. Evolutionaryconservation of pro- University of Hawaii, Manoa, for tissuesof the Ha- tein regionsin the protonmotivecytochrome b July1991] HawaiianGoose Mitochondrial DNA 593

and their possibleroles in redoxcatalysis. J. Mol. QUINN,T. W., & B. N. WHITE. 1987. Analysisof DNA Evol. 29: 157-169. sequencevariation. Pp. 163-198 in Avian genet- IRWIN, D. M., T. D. KOCHER,& A. C. WILSON. 1991. ics: a population and ecological approach (F. Evolutionof the cytochromeb geneof mammals. Cooke and P. A. Buckley,Eds.). New York, Ac- J. Mol. Evol. 32: 128-144. ademic Press. JOHNSGARD,P. A. 1978. Ducksßgeese and swansof SANGER,F., S. NICKLEN,& A. R. COULSON. 1977. DNA the world. Lincoln, Univ. Nebraska Press. sequencingwith chain-terminating inhibitors. KEAR,J., & A. J. BERGER.1980. The Hawaiian Goose: Proc. Natl. Acad. Sci. USA 74: 5463-5467. an experimentin conservation.Vermillion, South SHIELDS,G.F. 1990. Analysisof mitochondrial DNA Dakotaß Buteo Books. of the Pacific Black Brant (Brantabernicla nigri- KESSLER,L. G., & J. C. AvISE. 1984. Systematicrela- cans).Auk 107: 620-623. tionships among waterfowl (Anatidae) inferred ß & K. M. HELM-BYCHOWSKI. 1988. Mitochon- from restriction endonucleaseanalysis of mito- drial DNA of birds. Pp. 273-295 in Current or- chondrial DNA. Syst.Zool. 33: 370-380. nithology,vol. 5 (R. F. Johnston,Ed.). New Yorkß KOCHER, T. D., W. K. THOMAS, A. MEYERßS. V. ED- Plenum Press. WARDS, S. P'AX•O, F. X. VILLABLANCA, & A. C. ß & A. C. WILSON. 1987a. Calibration of mi- WILSON.1989. DynamicsofmitochondrialDNA tochondrialDNA evolution in geese.J. Mol Evol. evolution in animals:amplification and sequenc- 24: 212-217. ing with conservedprimers. Proc. Natl. Acad. Sci. --, & 1987b. Subspeciesof the Canada USA 86: 6196-6200. Goose (Branta canadensis)have distinct mitochon- MACDONALD, G. A., & A. T. ABBOTT. 1970. Volcanoes drial DNAs. Evolution 41: 662-666. in the sea:the geologyof Hawaii. HonolulußUniv. SPRINGER,P. F., F. B. LEE, W. L. SCHIPPER,& D. R. Hawaii Press. YPARRAGUIRRE.1986. Captive-reared Aleutian McDoUGALL,I. 1964. Potassium-argonages from la- CanadaGeese migrate to the Marshall Islands. vas of the Hawaiian Islands. Geol. Soc. Am. Bull. 'Elepaio 46: 153-154. 75: 117-128. SWOFFORDßD.L. 1989. PAUP: phylogeneticanalysis MILLER, A.H. 1937. Structural modifications in the using parsimony,version 3.0 g. Champaign,Il- Hawaiian Goose (Nesochensandvicensis), a study linois: Illinois Natl. Hist. Surv. in adaptive evolution. Univ. Calif. Publ. Zool., VAN WAGNER,C. E., & A. J. BAKER. 1986. Genetic 42: 1-80. differentiation in populationsof Canada Geese NEI, M., & W. H. LI. 1979. Mathematical model for (Brantacanadensis). Can. J. Zool. 64: 940-947. studying geneticvariation in terms of restriction --, & •. 1990. Association between mito- endonucleases. Proc. Natl. Acad. Sci. USA 76: chondrial DNA and morphologicalevolution in 5269-5273. Canada Geese. J. Mol. Evol. 31: 373-382. PALMER,R. S. 1976. Handbook of North American WILSON, A. C., R. L. CANN, S. M. CARR, M. GEORGEß birds. Vol. 2, Waterfowl (Part 1). New Haven, U. B. GYLLENSTEN,K. M. HELM-BYCHOWSKI, R. G. Yale Univ. Press. HIGUCHI, S. R. PALUMBI, E. M. PRAGER,R. D. SAGE, PRAGER,E. M., & A. C. WILSON. 1988. Ancient origin &M. STONEKING. 1985. Mitochondrial DNA and of lactalbuminfrom lysozyme:analysis of DNA two perspectiveson evolutionary genetics.Biol. and amino acid sequences.J. Mol. Evol. 27: 326- J. Linn. Soc. 26: 375-400. 335.