NEST-SITE SELECTION OF AMERICAN (HAEMATOPUS PALLIATUS) IN SALT MARSHES

BROOK LAURO AND JOANNA BURGER Departmentof BiologicalSciences, Rutgers University, Piscataway, New Jersey08854 USA

ABSTR•CT.--Wedocumented flexibility of American (Haematopus palliatus) nest-siteselection at marshhabitats in New York, New Jersey,and North Carolina.In New York we examinedthe relationshipof nest-siteselection to tidal flooding and found that chosenest sitesnonrandomly with respectto marshyareas. Three typesof habitat patch siteson marshislands chosen were sand,wrack (tidally washed-updead organicmaterial), and Spartinapatens grass. Nests on sandy siteswere lesssusceptible to tidal flooding than nestson wrack or grassbecause they were higher in elevation.There wasa positivecorrelation betweenthe numberof nestsand the length of sandbeach. Hatching successwas higher for eggsplaced on sand than for eggson grassor wrack. The locationof sandynest sitesused by birdschanged little betweenyears resulting in low turnoverrate. Wrack and grass locations changedbetween years resulting in a high turnover rate. Birdsused different habitatsin different geographicareas. In North Carolina,all nestswere on sand;none were on wrack or grass.In New York and New Jerseynests were placedon sand,wrack, and grass.Comparisons of nest characteristicsshowed that sandsites in North Carolinawere larger and drier than sand sitesin New York and New Jersey.Comparisons between years in New York and New Jerseywere different becauseof the availability of wrack. Received22 October1987, accepted16 September1988.

BIRDSselect breeding habitatsbased on biotic parative examination of habitat selection.The and abiotic factors of the environment. This traditional nesting habitat is describedas sand producesa nonrandom spatial distribution of dune on barrier island, although recently oys- nests.Breeding-habitat selection can be divided tercatchers have nested in marsh habitat (Froh- into three categories(Burger 1985):(1) General ling 1965,Zaradusky 1985). Habitat studieshave habitat use involves choice of a broad habitat been conductedon other marsh-nestingspecies type suchas an oak forest,salt marsh,or prairie; (Beer 1966;Bongiorno 1970; Burger 1974, 1977, (2) territory acquisition involves selection of an 1980;Burger and Lesser1978; Montevecchi 1978; area vigorouslydefended by one or both mem- Howe 1982), but we know of no comparisons bers of a pair; and (3) nest-site selection in- of a salt-marshnesting speciesin different geo- volves the choice of the actual nest location graphical regions. including the substrateon which the nest is Oystercatchersgenerally nest in open habi- placed, cover around the nest, and food avail- tats that are sparselyvegetated (Webster 1941, ability. Legg 1954, Harris 1967, Heppleston 1972, Hart- Flexibility in nestinghabitat is critical to re- wick 1974, Martinez et al. 1983, Summers and productive successin birds becauseindividuals Cooper 1977). Most speciesnest near the shore- within populationsmust adapt to differencesin line, but EuropeanOystercatchers (Haematopus habitat physiognomy.This is particularly true ostralegus)and Mageltanic Oystercatchers(Hae- for the American Oystercatcher(Haematopus matopusleucopodus) also nest in inland habitats palliatus)because habitat physiognomychanges (Baker1973, Buxton 1961, Fatla et al. 1966,Hep- over the breeding range (Not 1984), which ex- pleston1972, Miller and Baker 1980).Thus, some tends from Massachusettsto SouthernArgen- speciesare flexible in choiceof nesting habitat. tina (Hayman et at. 1986). Available descrip- Another indicationof flexibility in nesting-hab- tions of nesting habitat of the American itat selectionis the recent range expansionof Oystercatcher(Baird et al. 1884,Bent 1929,Stone American Oystercatchers(Post 1960, Post and 1967) provide no detailed quantitative or corn- Rayner 1964, Zaradusky 1985), American Black

185 The Auk 106: 185-192.April 1989 186 L^UROAND BURGER [Auk,Vol. 106

Oystercatchers(Haematopus bachmani; Eley 1976), long but not very wide. We censusedonly the bay and European Oystercatchers (Buxton 1961, side of the barrier island. Dobbs 1970). We calculated turnover rate (after Erwin et al. 1981) to examinethe stability of nestsites from year to year. Nest sites from 1983 were surveyed in 1984 to note STUDY AREA AND METHODS if siteshad changed.We noted new sitesin 1984.The turnover rate is We observed 31 nests (1983) and 59 nests (1984) in New York at Great South Bay and South Oyster Bay (40ø37'N,73ø24'W) between the Wantagh StatePark- T=•[• + •22' way and Oak Beach.Ninety-one marsh islands lay betweenthe JonesBeach barrier island and Long Is- where S• is the number of sitesoccupied only on the land. The islandsranged in size from < 1 ha to 844 first census,N• is the total number of sites occupied ha; the median was 40.58 ha. We defined a marsh on the first census,S2 is the number of sitesoccupied island as an area surrounded by 30-cm-deepwater at only on the secondcensus, and N2 is the total number low tide and not connectedto either Long Island or of sitesoccupied on the secondcensus. the barrier island. We chose 30 cm becauseyoung We collectedthe following data for nest sitesand chicks were unable to walk to nearby islands at this randomly chosensites on islands:the percentageof depth.The dominantspecies of vegetationin the study vegetation,sand, and wrackwithin 1-m and 5-m radii area were Spartinaalternafiora and S. patens. around the nest, width and length of nest sites In 1984,we examined19 nestsin BarnegatBay, New and distanceto water (creek or bay). A random point Jersey(39ø45'N, 78ø08'W). Barnegat Bay contains259 for each nest was chosenfrom a grid systemof each salt-marsh islands between the barrier island and the nesting island. Measuresof these nest-sitecharacters mainland.The predominatespecies of vegetationwere yielded informationon the degreeof protectionpro- S. alternafioraand S. patensin the wetter areaswith vided from tidal flooding. We also comparedyears Iva frutescensand Baccharishalimifolia in drier areas usingthe samenest characters. All comparisonswere (Burger and Lesser1978). madeusing Kruskal-Wallis Chi-square tests rather than In 1984, we examined 18 nests at Battery Island, analysisof variancebecause the data were not dis- North Carolina (33ø54'N, 78ø01'W), a natural island tributed normally. with depositsof dredgedmaterial. It was dominated We measured nest-site elevation on the sandy by a densegrass community of primarily S.patens and dredge-spoilarea of North Line Island (40ø38'N, a maritime shrub thicket dominated by Xanthoxylum 73ø29'W)because we assumedcompetition for sitesto americanurn,Ilex vomitoria,Quercus virginiana, Juniperus be great due to the high concentrationof breeding virgiana,Myrica cerifera,B. halimifoliaand I. frutescens pairs (9 in 1983;14 in 1984).We useda Nixon level (Parnell and Soots 1979). accurate to 0.03 m to measure elevation (above sea In New York we collected data for each nest site to level) with respectto survey markers. examine habitat-use flexibility and adaptations to We collecteddata on hatchingsuccess in New York marsh habitats. We defined a nest site as the nest in 1983 and 1984. A nest was considered successful scrapeand the surrounding habitat patch. A habitat if at least one chick hatched. We did not determine patch is classifiedby its substrateof sand, wrack, or fledgling successbecause the parentsled the chicks S. patensgrass (see Fig. 1). Sand patchesare created away from the nest after hatching. when dredgedspoil is depositedon the marsh.Wrack We examinedgeographic variation in nest-sitese- patchesconsist of dead organic material that washes lection between New York, New Jersey,and North up on the marshduring tidal flooding. Grasspatches Carolina by comparingthe percentageof vegetation, are on naturally high regions covered by S. patens, sand, and wrack within 1-m and 5-m radii around easily distinguishedfrom S. alternafioraareas. Sandy the nest, length and width of the nest site, and dis- patchesare higher in elevationand lesssusceptible tance to water. to flooding than wrack and grasspatches. Low sites were thosein which nestswere placedon grassand wracksubstrate; high siteswere thosein which nests RESULTS were placed on sand substrate. In New York we useda measuringwheel to survey Adaptationsto marshnesting in New York.--The the dimensionsof all high sites.The area of all sites was calculated from these measurements to determine number of nestson high siteson marshislands if there was a differencein frequencyof nest siteson was significantlygreater with respectto avail- the barrier island comparedto marshislands, and if able sand than the number of nests on high there was a correlationbetween the length of sand sites on barrier islands in 1983 and 1984 (1983: beach and number of nests.We used length for the X2 = 21.6, df = 1, P < 0.005; 1984: X2 = 56.4, df correlation calculationbecause most sandy siteswere = 1, P < 0.005). There was no relationship be- April 1989] OystercatcherNest-site Selection 187

Sand Deposit

AverageSolIdaHigh o$• Tide Line S;artlna alte•nafl•r= Fig. I. Profileof marshisland habitat, illustrating the differencesin elevationof the threebasic types of nest-sitelocations: sand, grass, and wrack.

tween number of nests and island size but there nest characterswere significantly different for was a positive correlation for number of nests low sites,but only 17%of characterswere dif- and length of sandy areas(1983: r = 0.69, P < ferent for high sites(Table 4). 0.06; 1984: r = 0.69, P < 0.006). On North Line Island, nests in 1983 were 1.08 We comparednest-site types (high and low) + 0.41 m above sea level and nests in 1984 were for frequencyof occurrence,turnover rate, and 1.01 + 0.18 m abovesea level; randompoints hatching successto determine if there was a were 0.69 + 0.33 and 0.58 + 0.29 m above sea relationshipbetween habitat type and repro- level in 1983 and 1984, respectively.In both ductive success.In both yearsmore birds nested yearsnests were significantlyhigher than ran- on high sitesthan low sites(Table 1). All of the dom points (1983: X2 = 8.76, df = 1, P < 0.003; high sitesused in 1983were usedagain in 1984. 1984:X 2 = 6.83, df = 1, P < 0.01). Thus, oyster- The turnover rate between years was 0.24 be- catchersselected areas least susceptibleto tidal cause of the increased use of habitat as a result flooding. of a population increase. Geographicvariability.--In North Carolina all Thirty-five percent of low sitesused in 1983 nestswere on high sites(Table 1) while in New were used again in 1984. The turnover rate was Jerseyonly 58%of nestswere on high sites.In 0.74 and was related to wrack availability. In New Jersey, 50% of low sites were on wrack, 1983, 78% of low nests were on wrack, but in and 50%were on grass.In New York (1983 and 1984 the amount of wrack decreased and all 1984),birds nestedon high and low sites (see nestswere on S.patens (Table 1). In 1983,young previous section for details). hatchedin only 22%of low nests;in 1984,hatch High sandysites in North Carolinawere larg- occurredin 43%. Young hatched in 60% (1983) er and drier than nestingsites in New York and and 64% (1984) of high nests.Therefore nests New Jersey(Tables 2, 5). In North Carolina most placed on high siteswere more stablein loca- nest characteristicswere significantlygreater tion betweenyears and had a higher reproduc- than New York in 1983 and 1984 (Tables 2, 5). tive successthan nestsplaced on low sites. However, the North Carolina and New York We comparedhigh and low nest sitesto ran- 1983values for the amountof sandand vege- dom sites and found that in 1983 all measured tation within a 5-m radius of the nest were not characteristicsdiffered significantly (Tables 2, significantlydifferent. In New York in 1983 the 3). All randomsites were grassareas while nest population was smaller than in 1984 (Table 1) siteswere mainly on wrack or sand. Nest sites and therefore larger, sandier areaswere used. had greater amounts of sand and wrack, and smalleramounts of live vegetation,in compar- ison to random sites. Accordingly, width and TABLEI. Frequencyof nest sites. length of wrack and sand areaswere greater than those of random sites. Nest sites were far- High Low ther away from a water body than randomsites. Total Sand Wrack Grass In 1984,high sandnesting-site patterns were New York 1983 31 22 7 2 similarto 1983.Low nestingsites had a different New York 1984 59 52 0 7 pattern,however, and were mainly on grassand New Jersey1984 19 11 4 4 North Carolina 1984 18 18 0 0 not wrack (Tables1, 3). Betweenyears, 57% of 188 LAUROAND BURGER [Auk,Vol. 106

TABLE2. Means and standarddeviations of nest characteristicsfor high and low nesting sitesin New York, New Jersey,and North Carolina. Random sites(on islands)were chosenusing a table of random numbers.

New York 1983 High sites Low sites Characteristics Nest Random Nest Random

% Wrack or sand/1-m circle 57.91 + 29.65 18.52 + 25.98 75.90 + 27.41 12.00 + 28.98 % Vegetation/ 1-m circle 44.19 + 31.95 77.95 + 31.73 24.10 + 7.41 87.70 + 28.85 % Wrack or sand/5-m circle 56.00 + 29.54 19.42 + 23.04 59.20 + 31.67 17.90 + 12.41 % Vegetation/5-m circle 41.71 + 27.23 77.14 + 25.19 41.70 + 32.46 74.20 + 23.75 Distance to water (m) 7.59 + 3.49 2.33 + 3.28 23.04 + 11.38 3.90 + 4.42 Habitat patch width (m) 7.39 + 3.54 0.53 + 0.96 6.63 + 4.02 0.40 + 0.70 Habitat patch length (m) 113.75 + 139.80 1.99 + 3.78 22.69 + 26.09 1.38 + 3.14

In 1983 New York sites were similar to North DISCUSSION Carolina. In North Carolina the amount of sand was not significantlydifferent from New Jersey. Flexibility in nesting-habitat choice is im- The distanceto water, the length, and the width portant to the successof a speciesthat breeds of nestsites were significantlygreater in North over a large geographicregion becauseof the Carolina than in New Jerseyand New York in biological constraintsof food availability, pre- 1983 and 1984. The percentageof vegetation dation, and competition. Abiotic limitations of and sand,the length and width of nestsite, and the physical environment (such as weather, the distanceto water were not significantlydif- available space,and physiognomyof habitat) ferent for high sites in New Jerseyand New alsoplay a critical role in flexibility. Flexibility York in 1983and 1984.One exceptionwas that in nest-site choice can minimize the costs of the percentageof sandwithin a 1-m radiusof tidal flooding or predation,thereby increasing the nestwas greater in New Jerseythan in New reproductivesuccess. This may resultin higher York in 1984. population numbers and expansion into new We comparedlow nestingsites only between habitats. New Jerseyand New York becauseNorth Car- Adaptationsto marshnesting.--In New York olina had no low nests.The length and width more birds nested on marsh islands than barrier of marshwrack were greaterin New Jerseythan islands, perhaps becauseof increased human in New York in both years. In 1984, this effect use of the barrier island and lower predation wasparticularly pronounced in New York be- rates on marshes.Similarly, breeding areas of causeall low-lying nestswere on grassand none African Black Oystercatchers(Haematopus mo- was on wrack. The lack of nests on wrack in quini)are threatenedby human use of beaches New York in 1984 also explainswhy the New (Summersand Cooper1977, Hockey 1983).Sev- Jerseyvalues were significantlydifferent in the eral colonial birds which nested only on barrier amount of wrack and vegetation. islandspresently nest on barrier and marshis-

TABLE 2. Continued.

New York 1984 High sites Low sites

Characteristics Nest Random Nest Random

% Wrack or sand/1-m circle 48.10 + 30.55 11.15 + 20.11 4.00 + 10.58 21.43 + 27.80 % Vegetation/1-m circle 50.62 + 30.40 84.75 + 25.56 99.71 + 0.95 79.00 + 27.76 % Wrack or sand/5-m circle 40.56 + 24.12 11.62 + 16.31 4.57 + 7.72 25.00 + 30.14 % Vegetation/5-mcircle 54.39 + 20.50 82.79 + 25.69 94.00 + 11.65 74.29 + 28.94 Distance to water (m) 12.74 + 14.43 5.91 + 9.30 26.34 + 22.07 3.53 + 4.52 Habitat patch width (m) 10.26 + 13.91 0.29 + 0.53 0.30 + 0.79 0.46 + 0.93 Habitat patch length (m) 77.16 + 107.08 0.00 + 1.65 0.59 + 1.55 0.54 + 1.04 April 1989] OystercatcherNest-site Selection 189

TABLE 2. Continued.

New Jersey1984 N. Carolina 1984 High sites Low sites High sites Characteristics Nest Nest Nest

% Wrack or sand/1-m circle 67.92 _+ 33.33 69.17 _+ 18.00 79.28 _+ 16.20 % Vegetation/1-m circle 29.85 _+ 31.01 30.50 _+ 18.38 23.61 _+ 23.75 % Wrack or sand/5-m circle 42.23 _+ 38.73 37.00 _+ 40.59 67.11 + 22.56 % Vegetation/5-m circle 51.46 _+ 37.55 59.50 _+ 39.07 27.94 _+ 16.89 Distance to water (m) 10.06 _+ 6.94 18.73 _+ 7.12 21.45 _+ 8.98 Habitat patchwidth (m) 9.59 _+ 10.06 13.65 _+ 5.80 33.00 _+ 15.39 Habitat patch length (m) 47.72 _+ 53.09 74.98 _+ 27.99 339.44 _+ 295.52

lands, a difference attributed to increased hu- found. We suggestthat sandnests are selected man use of beaches(Burger 1980, Erwin 1980, more frequently and have a lower turnover rate Rodgersand Burger 1981).Predation on barrier becausethey were moresuccessful. Several oth- islandsfrom mammalian predatorssuch as rac- er marsh-nestingspecies exposed to tidal flood- coons (Procyonlotor), foxes (Vulpesvulpes) and ing nest on high sandysites on marsh islands. fetal cats(Felis catus) may influence breeding Herring Gulls (Larusargentatus) and Willets (Ca- birds to move onto marsh habitats (Krunk 1964, toptrophorussemipalmatus) are two examples Summersand Cooper 1977, Greenwood 1982, (Burger 1980, Howe 1982). Moller 1983, Pierce 1986). In New York pre- Low nests had lower hatching successcom- dation on marsh islands seemed to be low. Pre- pared to high nests,but they showedmoderate dation by a rat (Rattussp.) and by a gull (Larus success.Thus, it isadvantageous to neston wrack sp.) were noted only once. and grasswhen sandsites are limited. The ad- Oystercatchersnonrandomly chosenest sites vantagesto nesting on wrack are that it floats with respect to physiognomic features of the to provide a raft for nests, it is higher in ele- marsh.Nest siteswere generally patchesof sand vation than the surroundingarea, and it is far- or wrack which had more substrate and less ther from water than sand nests and random vegetation,and they were farther from a water sites. Conversely, wrack is lower in elevation body (creekor bay) than randomly selectedsites than sand locations and its availability and lo- on nesting islands.On North Line Island, sand cation varies between years which results in nest sites were higher in elevation than ran- high turnover rates. The few grassnests were domly chosensites around the breeding area. in high areascovered with S. patensthat pro- Nesting at higher elevation is critical to marsh- vided refuge from tidal flooding. breeding birds becausenests can be destroyed Geographicalcomparisons.--American Oyster- by tidal flooding. Therefore, sand nestsshould catchersbreed in two types of general habitat flood lessand have a higher successrate, as we on the east coast of the : sand dunes

TABLE3. Comparisonof nest sitesand random sitesin New York for different nest characteristics.

1983 1984

High sites Low sites High sites Low sites Characteristics X2 P X2 P X• P X• P % Wrack or sand/1-m circle 14.04 0.0002 11.29 0.0008 40.70 0.0001 2.45 NS % Vegetation/1-mcircle 8.57 0.0034 10.77 0.0010 29.86 0.0001 6.83 0.0090 % Wrack or sand/5-m circle 15.86 0.0001 8.70 0.0032 36.62 0.0001 1.35 NS % Vegetation/5-m circle 14.40 0.0001 4.49 0.0340 35.11 0.0001 3.49 NS Distance to water (m) 17.57 0.0001 11.59 0.0007 24.01 0.0001 5.61 0.0178 Habitat island width (m) 29.30 0.0001 11.54 0.0007 74.40 0.0001 1.08 NS Habitat island length (m) 27.19 0.0001 11.03 0.0009 70.62 0.0001 0.77 NS 190 LAURO AND BURGER [Auk, Vol. 106

TABLE4. Comparisonof years1983 and 1984in New York for nest characteristics.

High sites Low sites Characteristics X2 P X2 P % Wrack or sand/1-m circle 1.48 NS 9.18 0.0072 % Vegetation/1-m circle 1.11 NS 8.61 0.0037 % Wrack or sand/5-m circle 4.54 0.0336 10.17 0.0016 % Vegetation/ 5-m circle 2.96 NS 7.95 0.0048 Distance to water (m) 3.10 NS 0.12 NS Habitat island width (m) 0.24 NS 0.40 NS Habitat island length (m) 0.11 NS 0.01 NS

along barrier islands (Bent 1929, Nol 1984) and the surroundingarea and on loosesubstrate such salt-marshislands (Zaradusky 1985, this study). as wrack and sand. In New York, birds rarely nestedin sanddune The importanceof flexibilityto re-expansion.- habitats.In North Carolinaand New Jerseyoys- When nesting habitat is limited, birds can fore- tercatcherscommonly nest in sand dune habi- go breeding, adapt to new habitats,expand into tats(Bent 1929,Parnell pers.comm., pers. obs.). similar habitats elsewhere, or they can adapt to Choiceof marsh-nestinghabitat by the Amer- a new habitat in other areas. Flexibility is an ican Oystercatcherwas flexible geographically. important aspectof adapting to new environ- In North Carolina all nests were on sparsely ments during a range expansion. vegetatedsand habitat. High sand nest sitesin In the past 20 years, oystercatchershave re- North Carolina had more sand, were wider and expandednorth into New Jersey,New York, longer, and were farther from the water than and Massachusetts(Post 1960, Postand Raynor in New York and New Jersey.Nests were on 1964,Zaradusky 1985). This may have been pos- elevated, drier habitat. In North Carolina oys- sible in part becauseof the adaptationto marsh tercatchersnested on sparselyvegetated sand nesting. Oystercatcherswere reported nesting areas (Soots and Parnell, 1975). Parnell (pers. on marshesin New Jersey in 1965 (Frohling comm.) noted that oystercatchersdid not nest 1965). Frohling suggestedthat marsh nesting directly on marsh wrack or grassin this area. may be an important adaptationto oystercatch- In New York and New Jerseyoystercatchers er populations becausehuman beach use de- nested on all three types of habitat: the most creasedavailable nestingareas and forcedbirds common was sand, followed by marsh wrack to nest in other habitats. Our data support his and Spartinagrass. The birds nesting on marsh suggestion.Thus, nesting on sand, as well as islandsin New York and New Jerseyused hab- wrack and grasson marsh islands, may be an itat differently than in North Carolina. Al- important aspectof the expansionand resulting though nest charactersdiffered in all locations, increase in oystercatcher populations in the oystercatchersnested at higher elevationsthan Northeast.

TABLE5. Comparisonsof nestcharacteristics between New York (NY), New Jersey(NJ), and North Carolina (NC) for high and low sites.

NY 1983 vs. NJ 1984 NY 1984 vs. NJ 1984 High sites Low sites High sites Low sites Characteristics X2 P X2 P X2 P X2 P % Wrack or sand/1-m circle 1.63 NS 1.19 NS 4.94 0.0262 9.99 0.0016 % Vegetation/l-m circle 2.38 NS 1.18 NS 5.35 0.0207 9.25 0.00234 % Wrack or sand/5-m circle 2.07 NS 1.43 NS 0.01 NS 0.87 NS % Vegetation/5-mcircle 0.72 NS 0.85 NS 0.01 NS 3.28 NS Distance to water 0.01 NS 0.05 NS 0.56 NS 0.18 NS Habitat island width (m) 0.02 NS 5.20 0.0225 0.23 NS 10.02 0.0015 Habitat island length (m) 1.67 NS 6.53 0.0106 0.62 NS 9.99 0.0016 April 1989] OystercatcherNest-site Selection 191

ACKNOWLEDGMENTS DOBBS,A. 1970. Extensionof the breeding range of the oystercatcherto the midlandsßBritish Birds Specialthanks to JohnZaradusky and Mark Shields 63: 428. for their assistancein locatingnest sites and for pro- ELEY,T. J.,JR. 1976. Extensionof the breedingrange viding useful field information.We are grateful to of the BlackOystercatcher in AlaskaßCondor 78: Anthony J. Lauro for his assistancein the collection 115. of elevation data. Finally, we appreciatecomments ERWIN,R.M. 1980. Breedinghabitat use by coloni- on the manuscriptfrom EricaNol and William Boar- ally nesting waterbirdsin two mid-Atlantic U.S. man. This projectwas funded by the Leathem-Stein- regions under different regimes of human dis- etz fund of RutgersUniversity. turbance. Biol. Conserv. 18: 39-51. --, J. GALLI,& J. BURGER.1981. Colony site dy- LITERATURE CITED namics and habitat use in Atlantic coast seabirds. Auk 98: 550-561. BAKER,A.J. 1973. Distribution and numbers of New FALLA, R. A., R. B. SIBSON, & E.G. TURBOTT. 1966. A Zealand oystercatchers.Notornis 20: 128-144. field guide to the birds of New Zealand and out- BAIRD,S. F., T. M. BREWER,& R. RIDGEWAY. 1884. The lying islands.London, Collins. waterbirds of , vol. 1. Boston, Lit- FROHLING,R.C. 1965. American Oystercatcherand tle, Brown and Co. BlackSkimmer nesting on saltmarsh. Wilson Bull. BEERßC. G. 1966. Adaptationsof nesting habitat in 77: 193-194. the reproductivebehaviour of the Black-billed GREENWOOD,R.J. 1982. Nocturnal activity and for- Gull Larus bulleft. Ibis 108: 392-410. aging of prairie raccoons(Procyon lotor) in North BENTßA.C. 1929. Life histories of North American Dakota. Am. Midi. Nat. 107: 238-243. shorebirds,part 2. U.S. Natl. Mus. Bull. 146. HARRIS,M.P. 1967. The biology of Oystercatchers BONGIORNO,S.F. 1970. Nest site selectionby adult (Haematopusostralegus) on Skokholm Island, S. Laughing Gulls (Larusatricilla). Anim. Behav. 18: Wales. Ibis 109: 180-193. 434-444. HARTWICK,E.B. 1974. Breedingecology of the Black BURGER,J. 1974. Breedingadaptations of Franklin's Oystercatcher(Haematopus bachmani Audubon). Gull (Laruspipixcan) to a marsh habitat. Anim. Syesis7: 83-92. Behav. 22: 521-567. HAYMAN, P., J. MARCHANT, & T. PRATER. 1986. Shore- 1977. Nesting behaviourof Herring Gulls: : an identificationguide to the waders of the invasion into Spartinasalt marsh areas of New world. Boston,Houghton Mifflin Co. Jersey.Condor 79: 162-169. HEPPLESTON,P. B. 1972. The comparativebreeding 1980. Nesting adaptationof Herring Gull ecologyof Oystercatchers(Haematopus ostralegus) (Larusargentatus) to salt marshesand storm tides. in inland and coastal habitatsß J. Anim. Ecol. 4.' Biol. Behav. 5: 147-162. 23-51. ß 1985. Habitatselection in temperatemarsh- HOCKEYßP. A.R. 1983. The distribution,population nesting birds. Pp. 253-281 in Habitat selectionin sizeßmovements, and conservation of the African birds (M. Cody, Ed.). London, AcademicPress. (Haematopus moquini). Biol. ß& F. LESSER.1978. Selectionof colony sites Conserv. 25: 233-262. and nest sitesby Common Terns (Sternahirundo) HOWE,M.A. 1982. Socialorganization in a nesting in OceanCounty, New Jersey.Ibis 120:433-449. population of Eastern Willets (Catoptrophorus BUXTON,E.J. 1961. The inland breedingof the oys- semipalmatus).Auk 99: 88-102. tercatcherin GreatBritainß 1958-1959. Bird Study KRUNK,A. 1964. Predatorand anti-predatorbehav- 8: 194-209. ior of the Black-headed Gull (Larus ridibundus). Behav.Suppl. 11: 1-29. LEGG,K. 1954. Nesting and feeding of the Black TABLE 5. Extended. Oystercatchernear Monterey, California. Condor 56: 359-360. NY 1983 vs. NC NY 1984 vs. NC NJ vs. NC MARTINEZ, A., A. MOTIS, E. MATHEU, & F. LLIMONA. High sites High sites High sites 1983. Data on the breeding biology of Oyster- catchers(Haematopus ostralegus) in the EbroDelta. X2 P X2 P X2 P Ardea 71: 229-234. 5.17 0.0230 15.43 0.0001 0.25 NS MILLERßE. H., & A. J. BAKER.1980. Displaysof the 0.37 0.0380 12.04 0.0005 0.07 NS (Haematopus leucopo- 1.12 NS 13.58 0.0002 2.38 NS dus). Wilson Bull. 92: 149-168. 2.19 NS 17.82 0.0001 1.75 NS 24.07 0.0001 16.72 0.0001 6.39 0.0105 MOLLER,A. P. 1983. Damageby rats (Rattusnorvegi- 28.07 0.0001 36.43 0.0001 16.78 0.0001 cus)to breeding birds on Danish Islands. Biol. 9.87 0.0017 18.97 0.0001 14.75 0.0001 Conserv. 25: 5-18. MONTEVECCHI,W. A. 1978. Nest site selection and 192 LAUROAND BURGER [Auk,Vol. 106

its survival value among Laughing Gulls. Behav. SOOTS,R. F., JR.,& J. F. PAP,NELL. 1975. Ecological Ecol. Sociobiol. 4: 143-161. successionof breedingbirds in relation to plant NOL, E. 1984. Reproductivestrategies in the Oys- successionon dredge islands in North Carolina tercatcher (Aves: Haematopodidae). Ph.D. dis- estuaries.UNC Sea Grant Prog. Publ., UNC-SG- sertation, Toronto, Univ. Toronto. 715-27. PARNELL,J. F., & R. F. SOOTSJR. 1979. Atlas of colonial STONE,W. 1967. Bird studiesat Old Cape May: an waterbirds of North Carolina estuaries. UNC Sea ornithologyof coastalNew Jersey,vol. 1. New Grant Prog. Publ., UNC-SG-78-10. York, Dover. PIERCE,R. 1986. Differencesin susceptibilityto pre- SUMMERS,R. W., & J. COOPER.1977. The population, dation during nesting between Pied and Black ecologyand conservationof the Black Oyster- stilts (Hirnantopusspp.). Auk 103: 273-280. catcher (Haernatopusrnoquini). Ostrich 48: 28-40. Post, P. 1960. The AmericanOystercatcher in New WEBSTER,J. D. 1941. The breedingof the BlackOys- York. Kingbird 11: 3-6. tercatcher. Wilson Bull. 53: 141-156. Post, P. W., & G. S. RAYNOR.1964. Recent range ZARADUSK¾,J.D. 1985. Breedingstatus of the Amer- expansion of the American Oystercatcherinto ican Oystercatcherin the town of Hempstead. New York. Wilson Bull. 76: 339-346. Kingbird 35: 105-113. RODGERS,J. A., & J. BURGER.1981. Concluding re- marks: symposium on human disturbanceand colonial waterbirds. Colonial Waterbird 4: 69-70.