sign in sign up
<<
Home , Neochen, Orinoco goose, Plumed whistling duck, Presbyornis, Whistling duck

A PARTIAL CLASSIFICATION OF WATERFOWL () BASED ON SINGLE-COPY DNA

CORT S. MADSEN, KEVIN P. MCHUGH, AND SIWO R. DE KLOET Instituteof MolecularBiophysics, The Florida State University, Tallahassee, Florida 32306 USA

ABSTR^CT.--Single-copyDNA-DNA hybridizationwas used to establishphylogenetic re- lationshipsamong 13 speciesof waterfowl (Anatidae) chosenfrom 10 tribes. On the basisof UPGMA clusteringof ATtodistances, we suggestthat the tribes Anatini, Aythyini, Tadornini, Mergini, and Cairinini diverged more recently than the Anserini, Stictonettini,, Dendrocygnini,and Anseranatini.The FreckledDuck (Stictonettanaevosa, Stictonettini) is only distantlyrelated to the other Anatidae.Presumably the lineage divergedvery early. The sheldgeese(tribe Tadornini) and the true geese(Anserini) are only remotely related. The Oxyurini, consideredto be in the subfamilyAnatinae, is remotely related to the other Anatidae. The Dendrocygnini form an isolatedtribe with no closerelationship to the and geese(subfamily ). We found that the (Anhimidae) are distantly related to the Anatidae. A method to estimate missing cells in a data matrix of pairwise distancesis presented. ReceivedI May 1987,accepted 16 February1988.

RECENTLY,nucleic acid sequenceanalysis has up about 10-40% of the total DNA (Eden et al. been usedto estimaterelationships of many or- 1978) but containsonly a small percentageof ganisms. The analysis can include direct se- the total sequencediversity. The remaining60- quencing of individual cloned genes or gene 90%are unique sequencesand representabout products, restriction enzyme analysis of ho- 98% of the total sequencecomplexity (Sibley mologousDNA fragments,and generalanalysis and Ahlquist 1983). of unique DNA by single-copyDNA-DNA hy- The mostconvenient procedurefor sequence bridization. In higher eukaryotes,perhaps only comparisonof total DNA of different 5-10% of the DNA (including gene-codingre- involves measurementof the thermal stability gions and a small fraction of the single-copy of reannealed heteroduplex single-stranded DNA) is under selection (Britten 1986). The ma- DNA. Becauserepetitive DNA sequencesrean- jor proportion of DNA, unique as well as re- heal first, such sequencesmight contribute in petitive, consistsof nucleotide sequencesthat a disproportionateway to the thermal stability are never expressedas proteins, have no known of resultingduplexes (Sibley and Ahlquist 1983). function, and are therefore presumablyfree to For this reason the repetitive DNA fraction is evolve (Britten 1986). usually removed. Nevertheless, repetitive se- Although both single-copyand repeatedDNA quences,because of rapid evolution, can also comparisonshave been used in studiesof phy- provide valuable data on the relationship of logenetic relationships in (Shields and recently evolved species (Gillespie 1977, Ar- Strauss1975; Eden et al. 1978; Sibley and Ahl- nason and Widegren 1986). quist 1982, 1983), the total DNA of higher eu- Although the record of birds is incom- karyotesis composedof many subsetsof DNA plete, that of waterfowl appears better than in that show varying degreesof reiteration (Brit- many other families (Olson and Feduccia1980). ten and Kohne 1968). These include the ex- Paleontological investigations show that the tremes of specific sequences,present as only majorwaterfowl radiation probablytook place one copy per genome through sequencesre- within the last 60 million years (Delacour and peated millions of times. The repeated families Mayr 1945, Olson and Feduccia 1980). These show sequenceas well as copy-number evolu- eventsfall within the sensitivityrange of DNA- tion. Sequencesthat occur only a few times in DNA hybridization proposed by Sibley and one organism may occur many times even in a Ahlquist (1983). closelyrelated species(Zwiebel et al. 1982).De- We propose a phylogeny of the Anatidae pending on reassociationconditions, the re- basedon single-copyDNA relationshipsamong petitive DNA fraction in genomes makes samplesfrom 13 of the 148 speciesof waterfowl

452 The Auk 105:452-459. July 1988 July1988] WaterfowlPhylogeny 453 belonging to 10 traditionally defined tribes. after the addition of ammonium acetate to a final Single-copyDNA sequencerelationships among concentrationof 2.5 M. After briefly washing with these waterfowl and three other avian species 70% ethanol, the DNA was dissolved in TE buffer and (Northern , Chauna chavaria; Spur- the concentrationadjusted to 1.0 mg/ml. winged Plover, Vanellusspinosus; and domestic Isolationof single-copyDNA.--A DNA fraction en- riched for single-copymaterial was isolatedas de- chicken, Gallusgallus) were investigated. scribed by Sibley and Ahlquist (1983). DNA was shearedby sonicationinto fragmentswith an average MATERIALS AND METHODS length of 500 nucleotides.Fragment size was deter- minedby electrophoreticcomparison with calibrated Material camefrom Mergus Waterfowl Farms(Tal- markers. The DNA was denatured by immersion in lahassee, Florida). Species were the White-faced boiling water for 10 min and reannealedat 50øCin Whistling-(WW) (Dendrocygnaviduata), Plumed 0.48 M sodium phosphatebuffer (pFI 7.0) to a Cot of Whistling-Duck (D. eytoni),Bar-headed (BG) 1,000.A single strandedfraction was isolatedby hy- (Anserindicus), Red-breasted Goose (Branta ruficollis), droxyapatitechromatography at 50øC.The fraction RuddyDuck (RD) (Oxyurajamaicensis), wasdialyzed against TE buffer, concentrated with iso- (OG) (Neochenjubata), Hooded Merganser(HM) (Lo- butanol (Maniatis et al. 1982), and precipitated with phodytescucullatus), Wood Duck (WD) (Aix sponsa), ethanol. After washing with 70% ethanol the DNA White-eyed Pochard(WP) (Aythya nyroca),Mallard was dissolved in TE buffer and the concentration ad- (ML) (Anasplatyrhynchos), White-cheeked Pintall (WC) justed to 1.0 mg/ml. (Anasbahamensis), and the nonanatidSpur-winged Radio-labelingofsingle-copy DNA.--Single-copy DNA Plover and domesticchicken. Northern Screamer(SC) was labeledwith 32P at the 5' terminusby incubation materialcame from Mary Dam (HainesCity, Florida), with polynucleotidekinase and gamma-labeledATP 32. the Magpie Goose(MG) (Anseranassemipalmata) and A 5' terminuslabeling kit wasobtained from Bethesda Trumpeter (TS) (Cygnusbuccinator) from Mi- ResearchLaboratories and used accordingto the in- chaelLubbock (Sylva, North Carolina),and the Freck- structionsprovided by the manufacturer.The specific led Duck (FD) (Stictonettanaevosa) from the Wildfowl activity of the radio-labeledDNA preparedby this Trust (Slimbridge,United Kingdom). Gamma-labeled procedureaveraged 5 x 106counts/•g of DNA. The ATP32 was obtained from ICN (Irvine, California). labeled DNA was freed of unincorporated labeled Isolationof DNA.--DNA was isolatedfrom blood ATP by the spin-column technique (Maniatis et al. cells by a procedure modified from Blin and Stafford 1982). The maximum amount of unincorporated la- (1976). Blood was obtained from a wing vein and beled ATP contamination was less than 3%. Terminal collectedin a heparinizedsyringe. After one-folddi- labeling was substitutedfor iodination (Sibley and lution with 0.15 M NaC1 containing 0.05 M EDTA at Ahlquist 1983) for safety reasons. pH 7.6 (buffer A), cells were collectedby centrifu- Hybridizationof single-copytracer DNA with driver gation, resuspendedin 25 volumes of buffer A, and DNA.--Hybrids were formed from a mixture com- lysedby the additionof 0.3%sodium dodecyl sulfate. posed of 1 part (=200 ng) radio-labeled single-copy An equal volume (with respectto the total lysate)of DNA and 1,000parts (=200 •g) sheared,whole DNA redistilled phenol, equilibrated with 0.1 M Tris-HCL (Sibleyand Ahlquist 1983).Hybrids were denatured buffer pH 9.0, was added, and the mixture was shaken by immersion in boiling water for 10 min, precipi- in a gyrotory waterbathovernight at room tempera- tated with ethanol, and redissolved in 0.5 M sodium ture (25øC).After the addition of one-half volume of phosphate(pH 7.0). Reannealingwas carried out at chloroform,the phaseswere separatedby centrifu- 60øC until a Cot of 8,000 was reached. The formation gation, the aqueous phase collected, and the crude of double-strandedhybrid DNA was analyzed by DNA precipitatedby the addition of an equal volume thermal elution from hydroxyapatite.Thermal elu- of absolute ethanol. The DNA was washed with 70% tion columns were submergedwithin an insulated ethanol and dissolvedin TE (0.01 M Tris-HCL buffer column box connectedto a Haacke temperature-con- pH 7.6, 0.001 M sodium EDTA) (5 ml/ml original trolled (ñ0.1øC) circulating waterbath. After appli- blood). After completesolution was obtained,0.5 ml cationof the samplesto the column (200 •g of DNA/ of 1.0 M Tris-HCL buffer pH 8.0 was added followed ml of hydroxyapatite)at 55øCin 0.03 M sodiumphos- by 100 •g of pancreaticribonuclease. The mixture was phate, the column was washed with 40 ml of 0.12 M incubated for 30 min at 37øC,5 mg of pronasewas sodium phosphate (pH 7.0) to remove the unbound added,followed immediately by 0.2ml of 10%sodium tracer. The last fraction of the wash (5 ml) was counted dodecyl sulfate, and the solution was incubated for 2 in a liquid-scintillationcounter, and radioactivitywas h at 60øC.The mixturewas againdeproteinized by determined to be no higher than the normal back- shaking overnight with an equal volume of phenol ground,indicating that the unboundtracer was elut- saturatedwith 0.1 M Tris-HCL buffer pH 9.0, the ed from the column.At eachof the eight temperatures phaseswere separatedby centrifugation,and the DNA (60-95øCin 5øCincrements) the single-strandedDNA was precipitated with two volumes of cold ethanol was eluted with 10 ml of 0.12 M sodiumphosphate 454 MADSEN,McHocH, AND DE KLOET [Auk,Vol. 105

(pH 7.0) and counted in a liquid-scintillation counter. The averagenumber of sampleschromatographed si- multaneouslywas 6, with a maximum number of 10. We used Tmto contrast homoduplex thermal disso- 2 ciation distribution with those of heteroduplexes 8C 4 (Sibley and Ahlquist 1983, Sheldon 1987). Figure 1 õ \ givesexamples of thermal elution curvesfrom which r• ATtovalues were extrapolated.Each 1ø difference for hi r• AT• values indicates a 1% divergence in nucleotide Z sequences(Bonner et al. 1973). All other statistics • 80 (percentagehybridization [%H], normalized percent- age hybridization [NPH], and T5oH)were calculated as describedby Sheldon (1987). To calibrate AT• and T•oH values with time, one z • must assume that DNA evolves at a constant rate (Sib- ley and Ahlquist 1983). In view of recent evidence suggestinga nonconstantrate (Britten 1986,Sheldon 1987),we use ATtovalues only to estimatethe branch- ing order. 2G Constructionof the datamatrix and phylogenetic tree.-- A 13 x 13 matrix was constructed (Table 1) from the averageATto values for eachpairwise comparison. The 23 missing AT• values required for constructing a completematrix were estimatedusing the following 70 80 90 rationale. First, a distancefunction must satisfy the triangle inequality: T,•C d, < d,k+ di•. IOO Second, some distance functions satisfy a stronger rule called the "ultrametric inequality" (Hartigan 1975):

d.•< max(d,k,djk).

If a distance function satisfies the ultrametric in- equality, then a matrix of suchpairwise distanceshas an exact representation as a tree. Conversely,any tree has an exactrepresentation in terms of a matrix of distancesthat satisfy the ultra- metricinequality (Hartigan 1975).We estimatedmiss- ing values from the ultrametric inequality. For ex- ample, if d23was missing,then d2, < max(d2k,dsk) for • 40 z

Hooded Merganser (1), Orinoco Goose( ju- bata;2), Mallard (Anasplatyrhynchos; 3), Wood Duck (Aix sponsa;4), Ruddy Duck (Oxyurajamaicensis; 5), and White-faced Whistling-Duck (Dendrocygnavidua- ta; 6). Bottom: Thermal dissociationcurves in which 60 70 80 90 the tracer DNA of the White-facedWhistling-Duck T, oC washybridized with the driver DNAs of White-faced Whistling-Duck (1), Mallard (2), Hooded Merganser Fig. 1. Top: Thermal dissociationcurves in which (3), Orinoco Goose (4), White-eyed Pochard (Aythya the tracer DNA of the Hooded Merganser(Lophodytes nyroca;5), Wood Duck (6), Ruddy Duck (7), and Spur- cucullatus)was hybridized with the driver DNAs of winged Plover (Vanellusspinosus; 8). July1988] WaterfowlPhylogeny 455 456 MADSEN,McHuGH, AND DE Kco•r [Auk, Vol. 105

TABLE2. ReciprocalATto values. Anosplatvrhvncho• (ML) • Anasbohomen$i$ (WC) I Ap•? ny•oc=(WP) Mean a nb Range AJx sponsa (WD) Anasplatyrhynchos 0.7 12 1.1 Neoche•n]ubato (OG) Anas bahamensis 0.1 4 0.3 I Lophodytescucullatus (HM) Aythya nyroca 0.5 7 0.8 I AnserCygnusindlcusbuccinotor (BG)(TS) Aix sponsa 0.3 9 0.7 I Stictonettanoevoso (FD) Neochenjubata 0.4 8 1.7 Lophodytescucullatus 0.3 8 0.8 L O•yuraDendrocy•nojamolcensisvlduata (RD)(WW) Anser indicus 1.3 7 2.7 I Anseran0s$emip•lmata (MG) Cygnusbuccinator 1.0 6 1.6 I Chaunachovaria (SC) Stictonetta naevosa 1.0 5 0.6 Oxyurajamaicensis 0.5 9 0.9 12.0 I0.0 8.0 6.0 4.0 2 0 0.0 Dendrocygnaviduata 0.7 11 1.7 % DIVERGENCE Anseranassemipalmata 1.0 6 2.4 Fig. 2. Dendrogram generatedby UPGMA based Chauna chavaria 1.1 6 2.8 on DNA-DNA hybridization (ATto)of 13 speciesof • Average distance between average of a labeled Anasplatyrhynchos waterfowlrepresenting 10 tribesin 3 subfamilies.See distancefrom anotherspecies and averageof the reciprocaltest. Table 1 for the distance matrix and abbreviations. bNumber of speciesfor which reciprocaltests were madewith Anas platyrhynchos.

copyDNA betweenthe Mallard and the Magpie all k and hence is bounded above by the minimum of the maxima (Meeter pets. comm.). This estimate, Goose(Anseranas) or the screamer(Chauna). Hy- the largest distance consistentwith the ultrametric brids between single-copyDNA of the Mallard inequality, was used for the missingdistances. In the and the total DNA of the chicken or the Spur- estimate of the ATtofor WC and WP (Table 1), 1.7 was winged Plover (Vaneflus)had NPH values of the smallest maximum value of all distances between 40-50%. For unknown reasons NPH values other taxacompared individually with WC and WP. among closely related speciesoften exhibit a To verify the accuracyof this method, a data matrix high degree of variance (Sheldon 1987). We was constructedusing only speciesfor which com- found a high degree of NPH variance(10-20%) plete data were available (Anasplatyrhynchos, Aythya for both distantly related and closely related nyroca,A ix sponsa,Neochen jubata, Lophodytes cucullatus, species.For this reasonwe used the T• statistic Oxyurajamaicensis, and Dendrocygnaviduata ). Five cells in lieu of T•oH. The T• statistic does not incor- within this 7 x 7 complete matrix were chosen ran- domly, and estimatedvalues were comparedwith ac- porate NPH valueswhen calculated;therefore, tual values. The mean difference between measured any variance in NPH will not obscure the and estimated values was 0.26 (SD = 0.11). The dif- branching pattern. ference between measured and estimated values Speciesin the subfamily (Mallard ranged from 0.1 to 0.4. [Anatini], White-eyed Pochard [Aythyini], The distancematrix was clusteredby the unweight- Hooded Merganser [Mergini], Wood Duck ed pair-group method using arithmetic averages [Cairini], and Orinoco Goose[Tadornini]) ap- (UPGMA; Sneath and Sokal 1973). parentlydiverged more recentlythan the other taxa (Fig. 2). These birds exhibited AT• values RESULTS smaller than 3.2. In addition, the White-eyed Delta Tmvalues for a pair of taxa varied from Pochardwas more closelyrelated to the Mallard 10%to 30% of the averagevalue (Table 1). In (ATto = 1.7) than to the Wood Duck, Hooded almostall individual experiments,however, the Merganser,or Orinoco Goose.The latter two order of the ATmwas the same. Discrepancies specieswere consistentlymore closely related occurred in some reciprocal values (Table 2). (AT• = 2.0) to each other than to other species. Many of the higher reciprocalATto values were The White-cheeked Pintail (Anatini) was also influencedby the one or two aberrant values more closelyrelated to the Mallard (AT• = 1.1) relating to a particularspecies or an abnormally than to the other speciestested. low homoduplexT• value (seebelow and Dis- The remaining speciesshowed large AT• val- cussion).The NPH values ranged from 90% for ues with respect to each other and to the five specieswith a AT• of less than 3.0, to 75-90% Anatinae species.The Magpie Goose (Anser- for those with a AT• of less than 8.0 and more anatinae) was the most distant species,with an than 3.0, to about 70% for the hybrids of single- averageATm value of 9.5 with respectto all other July1988] WaterfowlPhylogeny 457 waterfowl. With respect to the Mallard, the Brush (1976), and others (Frith 1955, 1964; Kear White-faced Whistling-Duck (Dendrocygnini) 1967;Sibley and Ahlquist 1972;Jacob and Gla- showed a very early divergence (Z•Tm= 7.8), ser 1975; Bottjer 1983; Kesslerand Avise 1984; followed by the Ruddy Duck (Oxyurini; Z•Tm= Livezey 1986)who usedanatomical, behavioral, 7.5), (Stictonettini; z•Tm= 6.5), and other chemicalor immunologicaldata. The and Bar-headed Goose (Anserini; Z•Tm= 6.0). branchingpattern (Fig. 2) amongcertain species Discrepanciesamong reciprocalvalues were (e.g. Lophodytescucullatus, Neochen jubata, and higher when the number of hybridizations for Aix sponsa)may not represent the true phylo- each comparisonwas limited. Yet, when the genetic pattern becauseof the high degree of Bar-headedGoose and White-faced Whistling- variance and the closeness of Z•Tmvalues. Yet, Duck were compared (n = 6), a large discrep- for the speciestested two main periods of ra- ancy in reciprocal values remained. When the diation have occurred.We believe that the Mag- Bar-headedGoose tracer DNA was hybridized pie Goose (Anseranas)lineage diverged very to the White-facedWhistling-Duck driver DNA, early (von Boetticher 1940) and the whistling- the averageZ•Tm was 5.9 (n = 2, SD = 0.14). The (Dendrocygnini) and the stifftails(Oxy- reciprocalexperiment was repeated twice and urini) diverged somewhatmore recently from had an average Z•Tmof 7.7 (SD = 0.49). The the main lineage. Specificresults with respect experimentwas repeated with a new tracerDNA to the earlydivergence of the RuddyDuck (Oxy- for each species.The Z•Tmvalues were within ura) are interesting because in virtually all 0.3ø of the previousvalues. The overall Z•Tmav- schemes for the classification of waterfowl this eragesfor the tracer DNAs of the Bar-headed tribe has been included in the subfamily Ana- Goose and White-faced Whistling-Duck were tinae. Delacourand Mayr (1945) consideredthe 5.9 (n = 3, SD = 0.10) and 7.7 (n = 3, SD = 0.38). Oxyurini a predabbler; Johnsgard(1978) con- A similar condition existed in the Red-breasted sideredthe Oxyurini alsoto be in the Anatinae Goose and the Plumed Whistling-Duck. The and to have divergedafter the .Feath- Trumpeter Swan was more closely related to er-proteindata (Brush 1976) and morphological the Bar-headed Goose and the Red-breasted characteristics(Livezey 1986) have supported a Goose (Z•Tm= 4.0) than any of the waterfowl possiblelink between the Oxyurini and the tested (ATmS> 5.8). The AT m value (4.0) indi- seaducks(Mergini). The large Z•Tmvalue be- catesthat this swan and gooselineage diverged tween members of these taxa is inconsistent with relatively long ago. this interpretation. protein and mor- Speciesfrom the Anatidae were com- phologicalsimilarities may be the result of con- pared with the screamers(Anhimidae), Spur- vergent evolution due to diving in both tribes. winged Plover (Charadriidae), and domestic The Freckled Duck gave large Z•Tmvalues chicken () (data not shown). These (>6.0) comparedwith all other waterfowl. This data (including reciprocalvalues) indicated that specieswas thought to be an aberrant member ducks,geese, and Anserariassemipalmata are more of the tribe Anatini (Delacour and Mayr 1945). closelyrelated to screamers(average ZXTm = 11.3) Our data support the hypothesisadvanced by than to Spur-winged Plovers (ZXTm= 14.4) or Frith (1964) and Brush (1976) that Stictonettais chickens(average ZXTm = 14.1). only distantly related to the other Anatidae and that its lineage must have diverged early from DISCUSSION the main lineage leading to other waterfowl. We found that different values for the Z•TmS We found that a single-copyDNA similarity between a pair of specieswere generally stable, based on relationships among different taxa of although never as stable as reported by other waterfowl was in good agreementwith results investigators(Sibley and Ahlquist 1983, Shel- based on other criteria. The tribes in the don 1987). A number of factorscould produce subfamily Anatinae apparently radiated more variance in Z•Tmvalues. One is the effect of ho- recently than the Anserinae, Oxyurini, Den- moduplex T• varianceon calibrationsof hetero- drocygnini, Stictonettini, and Anseranatinae, duplex Z•Tmvalues. The majority of homoduplex which diverged much earlier. Our analysisgen- Tmvalues fell close to the ideal 85øCproposed erally agreed with the schemesadvanced by by Sheldon (1987), but a few were 1-2 ø below; Delacour and Mayr (1945), Johnsgard(1961), none were higher. A correction factor was not 458 MADSEN,McHUGH, AND DE KLOET [Auk,Vol. 105 employed (as done by Sheldon 1987) because give different results.The reiteratedsequences the causesof variance(e.g. tracer DNA impurity in the driver may be partly responsiblefor the or improper labeling) in the homoduplex Tm discrepancies.In experimentswhere single-copy might equally affect the heteroduplex Tin.De- instead of total DNA is used as the driver, the cisions about which values, if any, should be differencebetween the reciprocalvalues is re- corrected were difficult because the number of duced by approximately 50% (Madsen and de hybridizations was low. The small number of Kloet in prep.). experimentsfor each comparisonand mechan- Despite the inconsistencies,the Anserini lin- ical error may contribute to the increasedvari- eageapparently diverged relatively early in wa- ance. Producing exact elution rates, column terfowl evolution,although probably later than temperatures,and fraction volumes is difficult the Dendrocygnini. Although these tribes are when performing DNA-DNA hybridization ex- classifiedin the same subfamily (Anserinae), periments. This was evident from the observed they are not as closely related to one another higher degreeof interexperimentalvs. intraex- asgeese are to swans.We believe that the Orino- perimental variance. co Goose (Tadornini) and the Bar-headed Goose Variance in reciprocalvalues was usually no are only remotely related, and that their mor- greater than the degree of variance exhibited phologicaland behavioralsimilarity is the re- for interexperimentalATto values. The most se- sultof convergentevolution (Delacour and Mayr rious discrepancieswere between the geeseand 1945, Johnsgard1961). the whistling-ducks. We have no explanation for thisanomalous result. Nonreciprocity in ATto ACKNOWLEDGMENTS values has been considered by other investi- gators as an indication of "experimental qual- We thank Frances C. James for technical and edi- ity" (Sheldon 1987). The consistentdifference torial assistanceand Duane A. Meeter for help with in reciprocal values (1.8) for these two species, statisticalanalysis. We are indebtedalso to Lawrence however, indicates that some factor other than G. Abele for technicalhelp. Alan H. Brush,Anthony experimental error was involved. The rather Bledsoe,and Jon Ahlquist provided useful criticisms large variance precludes the resolution of in- of the manuscript.We thank Mike Lubbock (Sylva, North Carolina), Mary Dam (Haines City, Florida), equalities of DNA evolution rates. and the Wildfowl Trust(Slimbridge, Gloucestershire, The DNA of severalspecies of waterfowl, in- U.K.) for cooperationin obtainingblood samples. cluding the Bar-headedGoose, contains many

"sequencefamilies" that consistof considerable LITERATURE CITED amountsof repetitive DNA (McHugh, Madsen, and de Kloet in prep.). The reiteration frequen- ARNASON,U., & B. WIDEGREN.1986. Pinniped phy- ciesof the sequencesin these families can differ logeny enlightenedby molecularhybridization considerablyeven among closely related species. using highly reiterated DNA. Mol. Biol. Evol. 3: 356-365. Certain sequencesthat are barely detectablein one speciesare found in multiple copiesin oth- BLIN,N., & D. W. STAFFORD.1976. A generalmethod for the isolationof high molecularweight DNA ers(sometimes accounting for 5-10%of the total from eucaryotes.Nucleic Acid Res.3: 2303-2308. DNA). This phenomenon of sequence differ- BONNER,T. I., P. J. BRENNER,B. R. NEUFIELD,& R. J. encesamong closelyrelated speciesalso occurs BRITTEN. 1973. Reduction in the rate of DNA in Drosophila(Zweibel et al. 1982). We have not reassociationby sequencedivergence. J. Mol. Biol. determined the actual degree of sequencedi- 81: 123-135. vergenceamong the membersof suchfamilies BOTTJER,P.D. 1983. Systematicrelationships among of reiteratedDNA in any species,but in some the Anatidae: an immunologicalstudy, with a casessuch sequences may have diverged farther history of anatid classification,and a systemof than in others. This would createa disparity in classification. Ph.D. dissertation, New Haven, Connecticut, Yale Univ. the actualsequence complexity for species'tracer BRITTEN,R. J. 1986. Rates of DNA sequenceevolu- DNA. The conditions (0.5 M sodium phosphate tion differ between taxonomicgroups. Science at 60øC)for hybridization in general usagefor 231: 1393-1398. taxonomic studies are considered to be of low ß & D. E. KOHNE. 1968. Repeated sequences stringency.Especially for heterologoushybrids, in DNA. Science 161: 528-540. reciprocal tracer and driver combinations may BRUSH,A.H. 1976. Waterfowl feather proteins:anal- July1988] WaterfowlPhylogeny 459

ysis of use in taxonomicstudies. J. Zool. London Molecular cloning, a laboratory manual. New 179: 467-498. York, Cold Spring Harbor Lab. Press. DELACOUR,J., & E. MAYR. 1945. The family Anatidae. OLSON,$. L., & A. FEDUCCIA.1980. Presbyornisand Wilson Bull. 57: 109-110. the origin of the (Aves: Charadrio- EDEN, F. C., J. i•. HENDRICK,& S.S. GOTLIEB. 1978. morphae). SmithsonianContrib. Zool. No. 323: Hornology of single copy and repeatedsequences 1-24. in chicken, duck, JapaneseQuail and Ostrich SHELDON,F.H. 1987. Ratesof single-copyDNA evo- DNA. Biochemistry17: 5113-5121. lution in herons. Mol. Biol. Evol. 4: 56-69. FRITH,H.J. 1955. The downy ducklingsof the Pink- SHIELDS,G. $., •a: N. A. STRAUSS. 1975. DNA-DNA eared and White-eyed ducks. Emu 55: 310-312. hybridization studiesof birds. Evolution 29: 159- 1964. Taxonomic status of Stictonetta naevosa 166. (Gould). Nature 202: 1352-1353. $1BLEY,G. C., & J. E. AHLQUIST.1972. A comparative GILLESPIE,D. 1977. Newly evolved repeated DNA study of the white proteinsof nonpasserine sequencesin primates. Science 196: 889-891. birds. PeabodyMus. Nat. Hist. Bull. 39: 1-276. HARTIGAN,J.A. 1975. Clustering algorithims. New --, & --. 1982. The relationships of the York, John Wiley & Sons. Hawaiian honeycreepers(Drepaninini) as indi- JACOB,J., & A. GLASER.1975. Chemotaxonomyof catedby DNA-DNA hybridization. Auk 99: 130- Anseriformes. Blochem. Syst. Ecol. 2: 215-220. 140. JOHNSGARD,1•. A. 1961. The taxonomyof the Anati- --, & . 1983. Phylogenyand classification dae--a behavioural analysis. Ibis 103a: 71-85. of birds based on data of DNA-DNA hybridiza- 1978. Ducks,geese and swansof the world. tion. Pp. 245-292 in Current , vol. ! Lincoln, Univ. Nebraska Press. (R. F. Johnston, Ed.). New York, Plenum Press. KFAR,J. 1967. Notes on the eggsand downy young $NEATH, P. H. A., •a: R. R. $OKAL. 1973. Numerical of Thalassornis leuconotus. Ostrich 38: 227-229. . San Francisco,W. H. Freeman. KESSLER,L. G., & J. C. AVISE. 1984. Systematicrela- YONBOETTICHER, H. !940. Bemerkungenzur Syste- tionshipsamong waterfowl (Anatidae) inferred matik der Anatiden. Verh. Ornithol. Ges.Bayern from restriction endonucleaseanalysis of mito- 22: 160-!65. chondrialDNA. Syst.Zool. 33: 370-380. ZWIEBEL,L. J., V. H. COHN, D. R. WRIGHT, •a: G. P. LIVEZEY,B.C. 1986. A phylogeneticanalysis of re- MOORE. 1982. Evolution of single-copy DNA cent anseriform genera using morphological and the ADH genein sevendrosophilids. J. Mol. characteristics. Auk 103: 737-754. Evol. 19: 62-71. MANIATIS, T., E. F. FRITSCH,& J. SAMBROOK. 1982.