The Auk

A QuarterlyJournal of Ornithology 114 No. 3 July 1997

TheAuk 114(3):315-323+ frontispiece,1997

PHYLOGENETIC PLACEMENT OF MIMIZ UKU G URNEYI (AVES: STRIGIDAE) INFERRED FROM MITOCHONDRIAL DNA

HECTORC. MIRANDA, JR.,TM ROBERTS. KENNEDY,•'3 AND DAVID P. MINDELL2 • Departmentof BiologicalSciences, University of Cincinnati,Cincinnati, Ohio 45221, USA; 2Department of Biologyand Museum of Zoology,University of Michigan, Ann Arbor,Michigan 48109, USA;and 3Frederick and Arney Geier Collections and Research Center, Museumof NaturalHistory and Science, Cincinnati Museum Center, 1720 Gilbert Avenue,Cincinnati, Ohio 45202, USA

ABSTRACT.--Thenucleotide sequences of mitochondrial12S rRNA (12S)and protein-cod- ing cytochromeoxidase subunit I (COI) spanning 1,010 and 1,130 base pairs, respectively, are reportedfor sevenowl speciesrepresenting six genera:Otus rnegalotis (Philippine Scops- ), O. longicornis(Luzon Scops-Owl), Bubo virginianus (Great ), Asiofiarnrneus (Short-earedOwl), Ninoxphilippensis (Philippine Hawk-Owl), Tytoalba (Barn Owl; usedas an outgroup),and the little known Mirnizukugurneyi (Giant Scops-Owl).Separate phylo- geneticanalyses of 12S,COI, and the combined12S and COI data yielded congruentphy- 1ogenies,strongly supporting the phylogeneticaffinity of Mirnizukuwith Otus,as opposed to Bubo.The conservativenature of mitochondrialDNA evolutionin Mirnizuku,in sharpcon- trast to its unique derived morphologicalattributes, suggests that Mirnizukuis a recently evolvedinsular form of Otusthat hasundergone rapid morphologicalevolution towards gi- gantism,rather than being a small derivativeof the much larger eagle-(Bubo) as has beensuggested previously. Received 19 August1996, accepted 5 March 1997.

THEGIANT SCOPS-OWL (Mimizuku gurneyi) is Busse 1973, Marshall 1978, Dickinson et al. a monotypicspecies restricted to the Philip- 1991),whereas others treat Mimizuku as a syn- pine islandsof Mindanao, Dinagat, and Siar- onymof Otus(Delacour and Mayr 1946,Gross- gao.The evolutionaryorigin and phylogenetic man and Hamlet 1964, Burton 1973, Clements placementof thisspecies are unclear Mimizuku 1974, Gruson 1976). Amadon and Bull (1988: gurneyiwas first describedas Pseudoptynxgur- 304) state that Mimizuku "... is sometimes neyiby Tweeddale(1878), congeneric with P. placedin Otus,but may be closerto Bubo..." philippensis(now Bubophilippensis ). Hachisuka Its English name is no less controversial. (1934)removed gurneyi from Pseudoptynxand Mimizukugurneyi is widelyknown as the Giant placed it in the monotypicgenus Mimizuku. Scops-Owl(Delacour and Mayr 1946, Burton Many authorsagree and recognizeM. gurneyi 1973, Clark et al. 1978, Amadon and Bull 1988, asthe solemember of thisdistinct (Peters Sibleyand Monroe 1990),but more recentlyit 1940, dupont 1971, Clark et al. 1978, Eck and has been referred to as the LesserEagle-Owl (Dickinsonet al. 1991,Sibley and Monroe 1993) and the Mindanao Eagle-Owl (Inskipp et al. 4 E-mail: [email protected] 1996).The uncertaintyregarding the phyloge-

315 FRONTISPIECE.Adult Giant Scops-Owl(Mimizuku gurneyi), an uncommonresident of low and middle el- evationforests in the Philippines.Painting by HectorC. Miranda, Jr. 316 MIRANDA,KENNEDY, AND MINDELL [Auk,Vol. 114 netic placement of Mimizuku reflects insuffi- CCATCAACATAAAACCCCC-3') or L6704 with cientmaterials for comparativestudy and the H7216 (Mindell et al. 1991). Letters L and H refer to lack of knowledgeabout its biology.During re- light andheavy strands, and numbers correspond to cent surveysof Mindanao conductedby the approximatestarting positionsin the human mt- DNA sequence(Anderson et al. 1981). Basesfor de- National Museum of the Philippines(NMP) generateprimer sitesare shownin brackets.Double- and the CincinnatiMuseum of NaturalHistory strandedPCR amplificationswere performedusing and Science(CMNH), we obtained frozen tis- Taq DNA polymerase and the reaction buffer sueand a completeskeleton that haveclarified (500mM KC1, 100 mM Tris-HC1, 1.0% Triton X-100) whetherthis uniqueinsular species represents from PromegaCorporation. Reaction mixtures were a small eagle-owlor a large scops-owl.Reso- subjectedto 35 cyclesof denaturationat 94øCfor 30 lutionof thisissue provides clues to thehistory s, annealingfrom 40 to 50øCfor 30 s, and extension and processesunderlying avian speciesdiver- at 72øCfor 1 min. After amplification,15 mL of the sificationin the Philippinesand other archi- double-strandedPCR productwere separatedin low pelagos. melting-pointagarose, cut from the gel,and purified using Magic PCR Preps(Promega). Both the heavy and light strandsof all amplifiedproducts were se- METHODS quencedusing the CircumVentSequencing System (New England Biolabs)with direct incorporationof We obtainedDNA sequencecharacters for the en- 35S-dATE tire mitochondrial(mt) 12S rRNA gene (977 base Manual alignmentof COI sequencesis non-contro- pairs [bp] long in M. gurneyi;1,010 bp long in the versial,with no gapsbeing invoked. 12S rRNA align- alignment,including gaps) from two individuals of ments, however,require numerousinsertions and M. gurneyiand single representativesof Otus mega- deletions.Although phylogeneticanalyses can be lotis, O. longicornis,Bubo virginianus, Asio fiammeus, sensitiveto order of sequenceinput for alignments Ninoxphilippensis, and Tytoalba, the last beingused with numerous insertions and deletions (Mindell for outgroup comparison.We also sequenced1,130 1991),we focusedon well-alignedregions only, and bp of mitochondrialcytochrome oxidase subunit I thesedid not changesignificantly in alignmentbouts (COI) from the sameset of taxaexcept O. longicornis. with alternativeinput orders.CLUSTAL W (Thomp- All Otus,Ninox, and Mimizukutissue samples were son et al. 1994)was used to generatethe alignment obtainedduring the 1992to 1994NMP / CMNH Phil- usingthe defaultparameters. Conserved motifs and ippineBiodiversity Inventory surveys. Voucher spec- other information regarding secondarystructure imens for individuals analyzed are: M. gurneyi (Hickson et al. 1996, Mindell et al. 1997) also were (CMNH Nos. 36488 and 35735), O. longicornis used to improve and adjust the alignmentby eye. (CMNH No. 36491), O. megalotis(CMNH No. 36490), Percentdifference between sequences was initially and N. philippensis(CMNH No. 36492).GenBank ac- calculatedusing p = zd / zt, where zdis the number cessionnumbers for the sequencesare: O. m. everetti of nucleotidedifferences between two sequencesand (12S, U83754; COI, U83779), O. longicornis(12S, zt is the total number of nucleotidescompared. Se- U84751),M. gurneyi(12S, U83756; COI, U83780),B. quence differences corrected for substitution-rate virginianus(12S, U83757; COl, U83777),A. fiammeus differenceswere also calculated using Kimura's (12S, U83758; COl, U83781), N. philippensis(12S, (1980) two-parametermethod in the DNAdist pro- U83760; COI, U83783), and T. alba(12S, U83749; COI gram of PHYLIP (Felsenstein1995). Sites with inser- U91604). tionsand deletionswere excludedfrom sequence-di- DNA was extracted from muscle, liver, or heart tis- vergence estimates. sueusing standard techniques (Knight and Mindell We conductedexhaustive parsimony analyses us- 1993).Mitochondrial DNAs were amplified as a se- ing PAUP 3.1.1 (Swofford 1993) for the CO! and 12S ries of overlappingsegments using the following rRNA data setsindividually and combined.To ex- primer pairs. For 12S rRNA, we used L613/H887 plore the effect of differentialweighting of transi- (Mindell et al. 1991), L821 (5-GCCACACCCCCAC tionsand transversions,we usedthe followingchar- GGGTAC TCAGCAGT-3')/Hl194 (5'-TCGATTATA acterweighting schemes in alternativeanalyses: (1) GAACAGGCTCCTCTAG-3'), L1091 (Kocher et al. all characterchanges equally weighted, (2) transver- 1989) /H1557 (Knight and Mindell 1994),and L1373 sions (TVs) only, and (3) TVs:TIs (transitions) (5'-GAAATGGGCTACATTTTCT-3') / H1621 (5'-CTT weighted2:1. For the 12SrRNA alignments,we con- IT/C][T / C]AGGTGTAAGCT[G / A][G / A]ATGCTT- ductedseparate analyses on all sites(1,010 aligned 3'). For COI, we used L5844 (5'-CCTCTGTAAAAA positions)and on well-aligned regions only (944 GGACTACAGCC-3')/H6290 (5'-TTGCCAGCTAGT alignedpositions), in which sequenceregions with GGGGGTA-3'), L6195 (5'-AATAACATAAGCTTC gaps were excludedto reducehomoplasy. We cal- TGACT-3') / H6762 (5'-GATGTAAAGTAGGCTC culated support indices for each node within the GGGTGTCTAC-3'), and either L6424 (5'-ACCG most-parsimonioustopology by findingthe shortest July1997] PhylogeneticPlacement ofMimizuku gurneyi 317

TABLE1. Percentdifferences that are presumedtransitions (above diagonal) and percentsequence diver- gences(below diagonal) for pairwise comparisons of mitochondria112SrRNA sequences based on observed sequencedifferences (left of slash)and Kimura's (1980) two-parameter model (right of slash).

I 2 3 4 5 6 7 I Mimizukugurneyi -- 71.0 71.2 65.2 61.7 58.5 61.4 20tus megalotis 6.3/6.8 -- 72.4 64.8 64.1 63.3 63.0 30tus longicornis 11.2/11.1 8.0/8.7 -- 68.6 66.5 62.3 59.5 4 Bubovirginianus 16.4/16.1 14.1/17.1 16.1/20.0 -- 62.5 66.7 58.8 5 Asioflammeus 17.7/21.8 16.1/19.6 16.2/19.6 15.1/18.1 -- 61.8 59.9 6 Ninoxphilippensis 16.6/20.0 15.5/18.3 16.0/19.2 16.8/20.1 16.8/20.2 -- 58.3 7 Tytoalba 17.3/20.7 16.6/19.8 15.5/18.5 16.9/19.9 16.9/21.7 16.5/19.5 -- tree in whichthe particularnode is not found,and mative(23%). Presumed transitions for pairsof denotingon branches the number of additionalsteps speciesranged from 58.3 to 72.4% of all required(Bremer 1988, Kallersj•5 et al. 1992).In ad- changeswith an averageof 63.9%across all dition, bootstrapresampling (Felsenstein 1985) was comparisons(Table 1). Percenttransitional dif- donewith 2,000replications using the branch-and- ferences were based on observed values. These boundsearch option. relativelyhigh percentagessuggested that the 12S rRNA transition substitutions retained RESULTS phylogeneticinformation, and that they were 12S RNA.--Of the 1,010 bp examined,55 notyet saturatedwith change(Mindell and Ho- gapswere invoked. Within the regionswithout neycutt 1990, Mindell et al. 1991). Excluding gaps, 325 positionswere variableand 233 of siteswith gaps(where alignments may be un- those positionswere phylogeneticallyinfor- reliable),the mean observedpairwise percent divergenceacross all comparisonswas 15.0%, ranging from 6.3 to 17.7%. Sequencediver- 12S rRNA gencebetween O. megalotisand M. gurneyiwas 6.3%,lower than between the two Otusspecies at 8.0%. Percentsequence divergences between n.• Otuslongicornis M. gurneyiand the two Otusspecies were less than that found between Mimizuku and B. vir- 3-60 12-99I Otus megalotis ginianus at 16.4%. Transformed divergence --t I Mimizukugurneyimeasuresusing the Kimura 2-parameter model increasedthe averageby 3.2%, but distances I3-65 [ Bubovirginianus betweenM. gurneyi and O. megalotis,O. longi- Asio flammeus cornis,and B. virginianusat 6.8%, 11.1%, and Ninox philippensis 16.1%,respectively, did not changedrastically (seeTable 1). As a precautionagainst mislabel- Tyro alba ing or contaminationof samples,500 bp were sequencedfrom a secondMimizuku individual, F•G.1. Themost-parsimonious phylogenetic tree and no difference was found between the two demonstrating placement of Mimizuku gurneyi, basedon all alternativeanalyses of the mitochondri- individuals. al 12S rRNA. An analysisusing transversionsonly Using Tyto alba as an outgroup,the parsi- yieldedthe same topology as above, except Asiofiam- mony analyseswere unequivocalin support- meuswas basal, rather than sister, to Bubovirginianus. ing an Otus/Mimizukusister relationship (Fig. Tytoalba was designatedas the outgroup.The first 1). Parsimonyanalyses of the 12S rRNA data number above the branch is the support index set based on alternativecharacter weighting (Bremmer1988, Kallersj•5 et al. 1992)denoting the numberof additionalsteps needed to breakthe node, approachesconsistently yielded the same sister and the secondnumber denotes bootstrap support relationship.All of the most-parsimonious basedon 2,000replications. These numerical calcu- trees were identical,except for the transver- lationsare basedon 12SrRNA well-alignedregions sion-onlyanalysis for well-aligned regionsin only. whichA. fiammeusis basalrather than sisterto 318 MIRANDA,KENNEDY, AND MINDELL [Auk, Vol. 114

TABLE2. MitochondrialCOl percenttransitional differences (above diagonal) based on first, second,and third codonpositions (left of slash)and first and second codon positions only (right of slash).Values below diagonalare percentdivergences for pairwisecomparisons based on observeddifferences (left of slash) and Kimura's(1980) two-parameter model (right of slash).

1 2 3 4 5 6

I Mimizuku gurneyi -- 87.1/63.9 93.6/77.8 91.1/72.2 86.1/59.1 91.6/72.3 20tus megalotis 10.5/11.4 -- 91.8/67.9 90.0/60.7 83.8/47.2 91.0/62.5 3 Bubovirginianus 15.9/18.2 14.7/16.7 -- 96.0/73.4 89.9/58.6 96.4/80.0 4 Asiofiammeus 15.4/ 17.6 16.1/18.7 13.3/14.8 -- 89.2/62.5 95.8/76.5 5 Ninox philippensis 17.2/20.0 17.1/20.1 16.3/18.4 15.6/17.9 -- 89.5/57.7 6 Tyroalba 18.3/21.6 18.3/21.8 16.4/19.2 16.6/19.2 19.1/23.0 --

B. virginianus.All trees that were generated, gencebetween Mimizuku and Bubowas 18.2% based on all regionsor well-alignedregions a value higher than that betweenOtus and only,showed O. longicornis basal to the O.mega- Mimizukuor Buboand Asio.The COI genealso lotis/M. gurneyi clade. The support index showedbias in basecomposition, with guanine showedan additional 12 steps required to deficiencyin the secondand third codonposi- breakthe O. megalotis/M.gurneyi clade and 11 tions,and relativelylow thyminefrequency at additionalsteps to breakthe entire O. longicor- the third codonposition, as observedin other nis/ O. megalotis/ Mimizuku clade. Bootstrap mt protein-coding genes (Lanyon and Hall valuesfor all brancheswere greaterthan 50%, 1994).The meanCOI percentbase composition with the O. megalotisand M. gurneyiclade for the sevenowl specieswas A: 25.4,C: 30.8, showinga 99%bootstrap value. G: 16.6, and T: 24.7. Cytochromeoxidase I.--COI showed325 vari- Analyseswith equalweighting for all char- ablesites among five ingrouptaxa. There were acterchanges, and with TVs:TIsweighted 2:1 no indels (insertions/deletions)observed. For- (for first, second,and third codon positions ty-fivechanges (13.9%) were at the first codon combined)yielded a treeincongruent with the positions,28 (8.6%)were at the secondcodon 12Stopology in whichBubo is basalto a Ninox/ positions,and 252 (77.4%)were at third codon Asioclade. Transversion parsimony for all co- positions,a ratio of roughly1.5:1:9. Although don positionsdiffered, with Asio basal to a the substitution rate difference between the Bubo/Ninoxclade. However, all topologiesgen- first two codonsand the third codonpositions erated under various weighting approaches is relativelyhigh, the observedpatterns are basedon the first,second, and third codonpo- consistentwith studiesof otherprotein-coding sitionssupported the Otus/Mimizukusister re- mitochondrialgenes in (Kocheret al. lationship.To reducethe amountof homoplasy 1989, Edwards and Wilson 1990, Lanyon and due to multiplesubstitutions, third codonpo- Hall 1994).High percenttransition values for sitionswere given an a prioriweighting of zero. pairwise estimates were observed, ranging Exhaustiveparsimony analysis of all characters from 83.8%between O. megalotisand N. philip- in the first and secondcodon positions resulted pensisto 96.4%between T. albaand B. virgini- in a singlemost-parsimonious tree that united anus(Table 2). Thiswas due mainly to therapid Mimizukuwith Otus(Fig. 2A). Parsimonyanal- nucleotidechange at the third codonposition. yses also were conductedusing: (1) only the Substitutionbias at the third codonposition first codonposition, (2) onlythe secondcodon has beenshown to contributeto phylogenetic position,and (3) transversionsonly at the first misinformation due to saturation(Mindell and andsecond codon positions. Each of thesesup- Thacker1996). Percent transitional changes for portedthe sisterrelationship of Otusand Mim- both first and secondcodon positions were izukurelative to Bubo.Transversion parsimony lower,ranging from 47.2% (Ninoxand Otus)to of codonposition 1 and 2 only resultedin two 80.0% (Ninox and Bubo).Overall percent se- shortesttrees, both supporting the Otus/Mim- quencedivergence based on the Kimura two- izuku clade, with Bubo basal to this branch. A parametermodel (Table2) betweenstrigiform consensustree (Fig. 2A) yielded an unresolved speciesranged from 11.4%(Otus and Mimizu- Bubo/Asio/Ninox branch. Bremer and boot- ku) to 23.0% (Tytoand Ninox).Sequence diver- strap support indices for the two resolved July 1997] PhylogeneticPlacement of Mimizuku gurneyi 319

A COI B 12S and COI combined

__1-66 -- Mimizukugurneyi 2-62

Bubo virginianus -- 1.$•AsioOtus flammeusmegalotis ulf6••s Ninox philippensis

Tytoalba FIG.2. (A) COl consensusof two shortesttrees resulting from transversion parsimony analysis of all first and secondcodon position characters. (B) Single most-parsimonious tree found under all weightingschemes for 12Sand COl combined,with and withoutexclusion of ambiguousalignment regions in 12Sand thesat- uratedthird codonposition in COl. Thefirst numbers above branches represents the support index (Bremer 1988,Kallersj6 et al. 1992),and the secondnumber denotes bootstrap support with 2,000replications. brancheswere low (Fig.2A), duein part to the DISCUSSION smallnumber of changesat the firstand second codon positions. We presentevidence that Mimizukuis more 12S rRNA and COI combined.--The combined closelyrelated to Otusthan to Bubo,assuming data set comprised2,140 nucleotide bases. Par- the monophyly of Bubo.All analytical ap- simony and transversion analyses of all data proachesfor the 12SrRNA (i.e.all generegions and all data excludingambiguous 12S align- andthe well-alignedregions only, all character ment regionsyield a singlemost-parsimonious changesweighted equally, transversions only, tree (Fig. 2B)that wasentirely congruent with or transversions:transitionsweighted 2:1) sup- the 12Sgene tree (Fig. 1). This topologyis ro- port the singleshortest tree presentedin Fig- bustas indicated by the lackof sensitivityto al- ure 1 with Mimizuku sister to Otus. This also is ternativeweighting approaches, and inclusion concordantwith the COI analyses,the com- or exclusionof ambiguousalignment regions in bineddata set analyses, and the grosscompar- 12Sand the third codonposition in COI. isonsof skullmorphology (Fig. 3). Combining Skullmorphology.--The recent availability of a datafrom differentregions of the mitochondri- M. gurneyiskeleton enabled gross comparison al genomemaximizes the amountof character of its skullwith thoseof O.megalotis and B. vir- evidencein parsimonyanalyses (Cracraft and ginianus.We evaluatedthe same characters Mindell 1989, Kluge 1989, deQueiroz et al. usedby Ford(1967): (1) medio-lateralslope of 1995).The apparentconsistency of morpholog- cranial side, (2) foreheadsize and shape,(3) ical and molecular evidence increases our con- cranialwidth relativeto height,and (4) supra- fidencein the mitochondrialDNA analyses. orbitalprocess shape. The O. megalotisskull ex- Our evidenceindicates that Mimizukugur- hibitsa high and steepforehead, the supraor- neyiis a large scops-owlthat is intermediatein bital processis long and pointed,and the cra- size between scops-owlsand eagle-owls.Ac- nium is high relativeto width (Fig. 3). The cordingly,this speciesshould be knownby the Mimizuku skull also exhibits these features Englishname "Giant Scops-Owl."The plum- (Fig. 3). In contrastto Otusand Mimizuku,the age patternof the breastand abdomendiffers Bubo skull has a low, thicker forehead and a from that of most scops-owlsin being less small,knob-like supraorbital process, and the cryptic, with plain black streaksalong the cranialheight is low relativeto the width (Fig. shaftsof the ventralfeathers (see Frontispiece). 3). Thestructural similarities of themajor skull Field observationsby Miranda and Kennedy featuresin Otusand Mimizuku support the mo- suggestthat Mimizuku'scall (a loud, single- lecularevidence presented here, although a de- note"whaaack") is nottypical of scops-owlvo- tailedphylogenetic analysis of skeletalcharac- calizations.Despite these derived characters, ters is warranted. the mitochondrial12S sequence divergence of 320 MIRANDA,KENNEDY, AND MINDELL [Auk, Vol. 114

&

FIG.3. Lateral (A) and dorsal(B) view of skullsof (1) Otusmegalotis (CMNH No. 36489),(2) Mimizuku gurneyi(CMNH No. 35735),and (3) Bubovirginianus (CMNH No. 34913).Dorsal view is shown20% smaller than lateral view.

M. gurneyifrom O. megalotisis relatively low zukuis moreclosely related to the largereagle- (6.8%) and is lower than the observed se- owls, is that Mimizukuevolved smaller body quence-divergenceestimates between O. mega- size, perhapsin responseto competitionwith lotisand O.longicornis (8.7%). The derived phe- the PhilippineEagle-Owl (B. philippensis).Our notypicfeatures of Mimizuku,in contrastto its data and phylogeneticanalyses support the conservative mitochondrial DNA evolution, firstscenario (toward gigantism) over the latter suggestthe occurrenceof rapid differentiation. (toward size reduction).There is an activede- Discordantrates of morphologicaland molec- bateas to the ecologicalforces that drive shifts ular evolution also have been documented in in body size among insular reptiles (Miles and other avian species(Shields and Wilson 1987; Dunham 1996, Schoener 1970, P•rflliams 1972, Avise and Zink 1988; Zink et al. 1991, 1995; Roughgarden1992). In insularforms, random Zink and Dittmann 1993) and other vertebrate geneticdrift that accompaniesa founderevent lineages (King and Wilson 1975, Baverstock can play a subtlebut major role in rapid spe- and Adams 1987, Schwaner and Sarre 1988, ciation. Radtkey 1996). This notion is supportedby Our study raisesthe issueof whichcriteria field investigationsshowing that speciation are appropriateand reasonablefor delimiting and evolutionarychanges in morphologyin genera.In the past, it was commonto place a birds can be rapid (e.g. Grant 1992). speciesin a separategenus if it exhibiteda Mimizukugurneyi is an uncommonresident greater degree of morphologicaldivergence of low and middle elevation forests. On Min- than that observedamong other related spe- danao, its habitat overlapswith that of the des. But, should a separategenus be recog- morewidespread Otus megalotis, which occurs nized solelyon thebasis of apparentphenotyp- in forestedareas and open woodlands.It is ic uniqueness,regardless of how recentlythe possibleto devisea scenariowhereinMimizuku taxon has evolved in evolutionarytime? Or, representa lineageof colonizersthat under- shouldgenera reflect phylogeny and relative went rapid morphologicalchanges towards gi- ageregardless of degreeof morphologicaldif- gantism.The alternativescenario, that Mimi- ferentiation? Taxonomists will encounter these July1997] PhylogeneticPlacement ofMimizuku gurneyi 321 questionswith increasingfrequency as greater BREMER,K. 1988. The limits of amino-acidsequence emphasisis placed on phylogeneticperspec- datain angiospermphylogenetic reconstruction. tivesand on historicalcontexts of biodiversity. Evolution 42:795-803. BURTON,J. A. (Ed.). 1973. Owls of the world. E. E Dutton and Company,New York. ACKNOWLEDGMENTS CLARK, g. J., D.G. SMITH, AND L. H. KELSO. 1978. Workingbibliography of owls of the world. Na- We are grateful to numerouscolleagues, particu- tional Wildlife Federation Scientific and Techni- larly J. Brownand R. Fernandez,who helpedobtain cal SeriesNo. 1, Washington,D.C. specimensof Mimizuku gurneyi. We owe special thanks to E C. Gonzales of the National Museum of CLEMENTS,J. E 1974. Birds of the world: A checklist. Two Continents, New York. the Philippines,and the many Philippine institu- tions and agenciesthat assistedduring our field CRACRAFT,J., AND D. P. MINDELL.1989. The early work. The NMP/CMNH PhilippineBiodiversity In- historyof modernbirds: A comparisonof mo- ventorywas supportedby a grant from the JohnD. lecularand morphologicalevidence. Pages 389- and Catherine T. MacArthur Foundation to P. C. Gon- 403 in The hierarchyof life (B. Fernholm,K. Bre- zales and R. S. K. Molecular laboratory work was mer, and H. J6rnvall,Eds. ). ElsevierScience Pub- lishers, Amsterdam. supported by National ScienceFoundation grants DEB-9019669 and 9496343 to D. E M. and a Univer- DELACOUR,J., AND E. MAYR. 1946. Birds of the Phil- sity of CincinnatiWieman/Wendell SummerFellow- ippines.Macmillan, New York. shipgrant to H. C. M. B. Collierkindly preparedthe DEQUEIROZ,g., M. J. DONOGHUE,AND J. KIM. 1995. photographsof the owl skulls.H. C. M. thanksT. Separateversus combined analysis of phyloge- Kane,B. Kinkle, S. Rogstad,and S. Pelikanfor their netic evidence. Annual Review of Ecology and advise and logistical support. We are indebted to J. Systematics26:657-681. Marks, D. Miles, M. Wink, R. Zink, and an anony- DICKINSON, E. C., g. S. KENNEDY, AND K. C. PARKES. mous reviewer for their valuable comments on the 1991.The birds of thePhilippines: An annotated manuscript.We also thank Mrs. EugeneFarny, and check-list.British Ornithologists' Union, Check- Joeand JanHerron for financialsupport of Philip- list No. 12. pine research,and Outdoor Adventuresfor assis- DUPONT,J.E. 1971. Philippinebirds. DelawareMu- tancewith field equipment.This paper is Contribu- seum of Natural History, Greenville. tion No. 15 of the NMP/CMNH Philippine Biodi- ECK, S., AND H. BUSSE.1973. Eulen, die resentenund versity Inventory. fossilen Formen Aves, Strigidae. Die Neue' Brehm-Bucherei.A. Ziemsen,Wittenberg Luth- LITERATURE CITED erstadts, Germany. EDWARDS,S. V., ANDA. C. WILSON.1990. Phyloge- AMADON, D., AND J. BULL. 1988. Hawks and owls of neticallyinformative length polymorphism and the world: A distributional and taxonomic list sequencevariability in mitochondrialDNA of (with the genusOtus by J. T. Marshalland B. E Australian songbirds (Pomatostomus).Genetics King). Proceedingsof the WesternFoundation of 126:695-711. VertebrateZoology 3:295-357. FELSENSTEIN,J. 1985. Confidencelimits on phylog- ANDERSON, S., A. T. BANKIER, B. G. BARRELL,M. H. enies:An approachusing the bootstrap.Evolu- L. BRUIJN,A. R. COULSON,J. DROUIN, I. C. EPE- tion 39:783-791. RON, D. P. NIERLICH, B. A. ROE, •. SANGER,P. H. FELSENSTEIN,J. 1995. PHYLIP (PhylogenyInference SCHREIRER,A. J. H. SMITH, g. STADEN,AND I. G. Package),version 3.57c. Department of Genetics, YOUNG.1981. Sequenceand organizationof the Universityof Washington,Seattle. human mitochondrialgenome. Nature 290:457- FORD,N. 1967. A systematicstudy of the owlsbased 465. on comparativeosteology. Ph.D. dissertation, AVISE,J. C., ANDg.M. ZINK. 1988. Moleculargenetic Universityof Michigan,Ann Arbor. divergencebetween avian sibling species:King GRANT,1v. R. 1992. Hybridization of species. and Clapperrails, Long-billedand Short-billed Science 256:193-197. dowitchers,Boat-tailed and Great-tailedgrack- GROSSMAN,M. L., AND J. HAMLET. 1964. Birds of les, and Tufted and Black-crested titmice. Auk prey of the world. BonanzaBooks, New York. 105:516-528. GRUSON,G. S. 1976. Checklist of the world's birds. SAVERSTOCK,P. g., AND g. ADAMS. 1987. Compar- Quadrangle,New York. ative rates of molecular, chromosomal and mor- HACHISUKA,M. 1934. Birds of the Philippine Is- phologicalevolution in someAustralian verte- lands,vol.2. H. E & G. Witherby,London. brates.Pages 175-188 in Ratesof evolution(K. S. HICKSON, R. E., C. SIMON, A. COOPER,G. S. SPICER,J. W. Campbelland M. E Day,Eds.). Allen and Un- SULLIMAN, AND D. PENNY. 1996. Conserved se- win, London. quencemotifs, alignment and secondarystruc- 322 MIRANDA,KENNEDY, AND MINDELL [Auk,Vol. 114

ture for the third domain of 12S rRNA. MINDELL,D. P., M.D. SORENSON,C. J. HUDDLESTON, Molecular Biologyand Evolution13:150-169. H. C. MIRANDA, JR.,A. KNIGHT, S. J. SAWCHUK, INSKIPP, T., N. LINDSEY, AND W. DUCKWORTH. 1996. ANDt. YURI. 1997. Phylogeneticrelationships An annotated checklist of the birds of the Ori- amongand within selectavian ordersbased on ental region. ChandlersPrinter Limited, East mitochondrialDNA. Pages211-245 in Avian Sussex,United Kingdom. molecularevolution and systematics(D. P. Min- KALLERSJO,M., J. S. FARRIS, A. G. KLUGE, AND C. dell, Ed.). Academic Press,New York. BULT. 1992. Skewnessand permutations.Cla- MINDELL, D. P., AND C. E. THACKER. 1996. Rates of distics 8:275-287. molecular evolution: Phylogeneticissues and KIMURA,M. 1980. A simple methodfor estimating applications. Annual Review of Ecology and evolutionaryrates of basesubstitutions through Systematics27:279-303. comparativestudies of nucleotidesequences. PETERS,J. L. 1940. Checklist of birds of the world, Journal of Molecular Evolution 16:111-120. vol. 4. Harvard University Press, Cambridge, KING, M. C., AND A. C. WILSON. 1975. Evolution at Massachusetts. two levelsin humansand chimpanzees.Science RADTKEY,R. R. 1996. Adaptive radiation of day- 188:107-116. geckos(Phelsuma) in the Seychellesarchipelago: KLUGE,A.G. 1989. A concernfor evidenceand a phy- A phylogeneticanalysis. Evolution 50:604-623. logenetichypothesis for relationshipsamong Epi- ROUGHGARDEN,J. 1992. Commentson the paperby crates(Boidae, Serpentes). Systematic Zoology 38: Losos: Character displacementversus taxon 7-25. loop. Copeia 1992:288-295. KNIGHT, A., AND D. P. MINDELL. 1993. Substitution SCHOENER,t. W. 1970. Size patternsin WestIndian bias,weighting of DNA sequenceevolution, and Anolis lizard. II. Correlations with the sizes of the phylogeneticposition of Fea'sviper. System- particular sympatric species-displacementand atic Biology42:18-31. convergence.American Naturalist 104:155-174. KNIGHT,A., ANDD. P. MINDELL.1994. On the phy- SCHWANER,t. D., ANDS. D. SARRE.1988. Body size logeneticrelationship of Colubrinae,Elapidae, of tiger snakesin SouthernAustralia, with par- and Viperidaeand the evolutionof front-fanged ticularreference to Notechisaterserventyi. Journal venom systemsin snakes.Copeia 1994:1-9. of Herpetology22:24-33. KOCHER,T. D., W. K. THOMAS, A. MEYER, S. V. ED- SHIELDS,G. E, ANDA. C. WILSON.1987. Subspecies WARDS,S. P;t;•BO, E X. VILLABLANCA,AND A. C. of the Canada Goose (Brantacanadensis) have WILSON. 1989. Dynamics of mitochondrial distinct mitochondrial DNA's. Evolution 4:662- DNA evolution in : Amplification and 666. sequencingwith conservedprimers. Proceed- SIBLEY,C. G., AND B. L. MONROE,JR. 1990. Distri- ings of the NationalAcademy of SciencesUSA bution and taxonomyof birds of the world. Yale 86:6196-6200. University Press,New Haven, Connecticut. LANYON, S.C., AND J. G. HALL. 1994. Reexamination SIBLEY,C. G., ANDB. L. MONROE,JR. 1993. A sup- of barbetmonophyly using mitochondrial-DNA plement to distribution and taxonomyof birds sequencedata. Auk 111:389-397. of the world. YaleUniversity Press, New Haven, MARSHALL,J. t., JR. 1978. Systematicsof smaller Connecticut. Asiannight birds based on voice. Ornithological SWOFFORD,D. L. 1993. PAUP.Phylogenetic analysis MonographsNo. 25. using parsimony,version 3.1.1. Illinois Natural MILES,D. B.,AND A. E. DUNHAM.1996. The paradox HistorySurvey, Champaign. of the phylogeny:character displacement of THOMPSON,J. D., D. G. HIGGINS,AND t. J.GIBSON. 1994. analysesof body size in island Anolis.Evolution ClustalW: Improvingthe sensitivityof progres- 50:594-603. sive multiple sequencealignment through se- MINDELL,D. P. 1991. Aligning DNA sequences:Ho- quenceweighting, position-specific gap penalties mologyand phylogeneticweighting. Pages 73- and weightmatrix choice. Nucleic Acids Research 89 in Phylogeneticanalysis of DNA sequence 22:4673-4680. (M. M. Miyamoto and J. Cracraft, Eds.). Oxford TWEEDDALE,A. H. 1878. Contributions to the orni- University Press,Oxford. thologyof the Philippines,no. XI. On the collec- MINDELL, D. P., C. W. DICK, AND R. J. BAKER. 1991. tionsmade by Mr. A. H. Everettat Zamboanga, Phylogeneticrelationships among megabats, in the island of Mindanao.Proceedings of the microbats,and primates.Proceedings of the Na- ZoologicalSociety of London 1878:936-954. tional Academy of SciencesUSA 88:10322- WILLIAMS,E. E. 1972. The origin of faunas.Evolu- 10326. tion of lizard congenersin a complexisland fau- MINDELL, D. P., AND R. L. HONEYCUTT. 1990. Ribo- na: A trial analysis.Evolutionary Biology 6:47- somal RNA in vertebrates:Evolution and phy- 88. logeneticapplications. Annual Review of Ecol- ZINK, R. M., AND D. L. DITTMANN. 1993. Gene flow, ogy and Systematics21:541-566. refugia,and evolution of geographicvariation in July1997] PhylogeneticPlacement ofMimizuku gurneyi 323

the SongSparrow (Melospiza melodia). Evolution ZINK, R. M., S. ROHWER,A. V. ANDREEV,AND D. L. 47:717-729. DITTMANN.1995. Trans-Beringiacomparisons ZINK, R. M., D. L. DITTMANN, AND W. L. ROOTES. of mitochondrial DNA differentiation in birds. 1991. Mitochondrial DNA variation and the Condor 97:639-649. phylogenyof Zonotrichia.Auk 108:578-584. Associate Editor: R. M. Zink