Notes on the Nesting of the Common Moorhen and Purple Gallinule in Southwestern Louisiana

Home , Common gallinule, Common moorhen, Moorhen

J. Field Ornithol., 58(1):55-61

NOTES ON THE NESTING OF THE COMMON MOORHEN AND PURPLE GALLINULE IN SOUTHWESTERN LOUISIANA

ROBERT N. HELM • LouisianaCooperative Fish and WildlifeResearch Unit LouisianaState University Baton Rouge,Louisiana 70803 USA

D^vio N. P^SHLE¾ Schoolof Forestry,Wildlife, and Fisheries LouisianaState University Baton Rouge,Louisiana 70803 USA PHILLIP J. ZW^•K LouisianaCooperative Fish and WildlifeResearch Unit LouisianaState University Baton Rouge,Louisiana 70803 USA

Abstract.--Selected aspectsof the breedinL biology of the Common Moorhen and Purple Gallinule were comparedin coastalmarshes and rice fieldsof Louisianaduring two breeding seasons.Both speciesconstructed trial-nests, egg-nests, and elevatedplatforms in marsh and rice fields.Nests were constructedover water out of plant parts availablenear the nest site. The interiors of Common Moorhen nestswere significantlylarger, but external di- mensionswere similar for both species.The mean height above water level of Purple Gallinule nestswas significantlygreater than for nestsof the Common Moorhen. Purple Gallinules had a 78.6% nestingsuccess rate, while 63.3% of CommonMoorhen nestswere successful.Success for both specieswas distributeduniformly overthe rangeof nestheights. Averageclutch sizesfor Common Moorhen were 6.7 in the marsh and 8.6 in rice fields. CommonMoorhen and Purple Gallinule both producedsignificantly larger clutchesin rice fields than in marshes.

NOTAS SOBRE EL ANIDAMIENTO DE GALLINULA CHLOROPUS Y PORPHYRULA MARTINICA EN EL SUROESTE DE LUISIANA, E.U.A. Sinopsis.--Durantc dospcriodos rcproductivos, aspcctos particularcs dc la biologlarcpro- ductira dc G. chloropusy P. martinica,fucron comparadoscn cicnagasy scmbradosdc arroz de Luisiana.Sc cncontr6 quc ambascspccics construycn nidos dc prSctica,nidos para poncr, nidospara rcsguardarsus pichoncs durante la nochcy plataformasclcvadas tanto cn cicna- gascomo cn camposdc arroz. Los nidossc construycron, sobrc cl agua, dc partesdc plantas disponiblcscerca dcl lugar dc anidamicnto.Las dimcnsioncscxtcrnas dc los nidosrcsultaron ser similarescn ambascspccics, aunquc el interior result6 scr significativamcntcmayor cn P. martinica.La altura de los nidossobrc cl nivcl dcl agua result6 scr significativamcntc mayor cn P. martinicaquccn G. chloropus.Sc cncontr6un •xito dc anidamicntodc 78.6% y dc 63.3% cn P. martinicay •. chloropus,rcspcctivamcntc; cstc fuc uniformc para ambas cspccicsa difcrcntcsalturas dc los nidos. La camadapromcdio para G. chloropusresult6 scr dc 6.7 cn cicnagasy 8.8 cn arrozalcs.P. martinicaprodujo un promcdiodc 5.8 hucvos cn las cicnagasy 8.6 cn arrozalcs.Las camadasdc ambascspccics rcsultaron scr significativamcntcmayor cn arrozalcsquccn cicnagas. The Common Moorhen (Gallinula chloropus)and Purple Gallinulc (Porphyrulamartinica) nest sympatricallyin coastalmarshes and rice fields along the Gulf of Mexico. Information relating to the breeding

55 56] R. N. Helmet al. j. FieldOrnithol. Winter 1987 biologyof thesebirds in sympatryis restrictedto unpublishedM.S. theses (Bell 1976, Helm 1982, Reagan 1977) and a referenceto late nestingin southTexas (Cottam and Glazener 1959). We presentresults of a study of nestcharacteristics, clutch size, and hatchingsuccess of thesebirds in coastal marshes and rice fields in southwestern Louisiana.

STUDY AREA AND METHODS Nesting observationswere made on three Louisiana sitesduring the 1977 and 1978 breedingseasons. Two siteswere locatedon or near the RockefellerWildlife Refuge, Cameron Parish, in "coastaldeep fresh marsh" (Shaw and Fredine 1956). The third site was in rice fields on the Louisiana State University Rice Experiment Stationin Acadia Par- ish, approximately70 km northeastof the Rockefellermarsh sites. Rice field nestswere found each year in thorough searchesconducted in early June and mid-July. There was no evidencethat any nestinghad been completedbefore the first search.Nests in the marsh siteswere locatedinitially in roadsidecanals during weekly searchesfrom April throughAugust. The composition,size and placementof all nestswere recorded.Active nestswere checkedat five-day intervalsand the number of eggsnoted until clutcheswere completed(i.e., when a nest with an adult bird in attendancedid not containadditional eggs on two successive visits). Date of initiation of laying of incomplete clutcheswas derived by subtractingone day for each egg presenton the day of discovery.For clutchescompleted before discovery, date of initiation was estimatedby subtracting26 d from hatching date. These estimateswere based on observations(Krauth 1972) that the birds lay one egg per day, begin incubationon the fifth day, and incubatefor approximately21 d before hatching begins. Nests were considered successful if there was evidence that at least one egg hatched.When eggsdisappeared, either the presenceof numerous small eggshell fragmentsamid the nestingmaterial (Oney 1954) or ready separationof the vitellinemembrane from the shell (Rearden1951) pro- videdevidence that eggshad hatchedand had not beenlost to predation.

RESULTS Both Purple Gallinule and CommonMoorhen constructedtrial-nests, egg-nests(often with associatedramps), brood-nests,and elevatedplatforms in marsh and rice fields. Both sexesparticipated in construction and maintenance of these structures. The interiors of Common Moorhen egg-nestswere larger than those of Purple Gallinule, in terms of both mean bowl depth (4.5 cm vs. 3.6 cm, t = 3.6, df = 13, P (0.05) and bowl diameter (12.1 cm rs. 10.8 cm, t = 3.4, df = 13, P (0.05). External nest diameter (• = 24.1 cm, Common Moorhen; • = 21.7 cm, Purple Gallinule) and external nest height(• -- 12.0 cm, CommonMoorhen; • = 10.9 cm, Purple Gallinule) did not differ betweenspecies (t = 1.07, P ) 0.05; t = 0.63, P ) 0.05). Vol.58, No. • NestingofCommon Moorhen and Purple Gallinule [5 7

All nestsexcept for 3 Common Moorhen nestswere built over water. There were no differences(t = 0.19, P > 0.05) in the depth of water over which nests were built (• = 61.2, Common Moorhen; • = 62.9, Purple Gallinule). However, Purple Gallinule nestswere locatedunder significantlydenser cover than were Common Moorhen nests (Mann- Whitney U-test, Z -- -4.29, P < 0.05). Heights of nestsabove water were groupedinto 10 cm intervalsfor analysis.The meanheight category for Purple Gallinule nests(40-50 cm) was greater (x2= 76.6, df = 9, P < 0.05) than for that of the Common Moorhen (20-30 cm). Egg nestsof both specieswere constructedeither on floating or in emergent vegetation.Floating nests, found only in marsh sites, were constructedon water hyacinth (Eichhorniacrassipes), alligator weed (Al- ternantheraphiloxeroides), and pennywort (Hydrocotylespp.). Stems of bulltongue(Sagittaria lancifolia) and southernbulrush (Scirpuscaliforni- cus) were incorporatedinto their bases.These nestswere often located near the centers of canals with little overhead cover. On three occasions, floatingnests drifted approximately75 m during strongwinds and were then abandoned.The disappearanceof six otherswas accompaniedby a wide swath through the vegetationindicative of alligator (Alligatormis- sissippiensis)predation. Nestsin emergentmarsh vegetationwere typically anchoredin south- ern bulrush, bulltongue or giant cutgrass(Zizaniopsis miliacea). Leaf bladeswere bent over in a criss-crosspattern to form the base, and leaf portionswere molded to form the nest cup. Growing leaveswere bent over nestsduring incubation,thereby increasingoverhead cover. Exten- sive ramps were commonlyassociated with higher marsh and maturing rice field nests.Nests in lower emergentstended to be susceptibleto flooding. Nests were built out of readily available plant material. In rice fields, nests were built out of rice foliage. In the marsh, 78% of Common Moorhen nests (n = 97) and 93% of Purple Gallinule nests (n = 28) were built from plants found within 1 m of the nest site. In only one instance,a floatingCommon Moorhen nestbuilt on water hyacinth,was a nest constructedentirely of plant parts not found within 1 m. One unusual Common Moorhen nest was built in a black willow (Salix nigra) overhanginga roadsidecanal, and two others were con- structedatop abandonedBoat-tailed Grackle (Quiscalusmajor) nests. Three casesof egg-dumpingby other speciesinto Common Moorhen nestsincluded one by a Least Bittern (Ixobrychusexilis) into a marsh nestsubsequently lost to predation,and two by FulvousWhistling-Ducks (Dendrocygnabicolor) into neststhat were abandonedshortly thereafter. In the marsh, Common Moorhen initiated clutch productionfrom the first week of April until early August, a total of 122 d in 1977 and 131 d in 1978. Marsh-nestingPurple Gallinules began laying in May and ceasedclutch-initiation in early August, a total of 95 d in 1977 and 72 d in 1978. Neither speciescould begin nestingin rice fields until mid- May, when the rice had grown to a height of 80-90 cm. Harvest began 58] R. N. Helmet al. J.Field WinterOrnithol. 1987

TABLE1. Nestingsuccess of CommonMoorhen and Purple Gallinule in coastalmarsh and rice fields of southwestern Louisiana, 1977 and 1978 combined.

Marsh Rice fields

Unsuccessful Successful Unsuccessful Successful

Common Moorhen 48 (45%) 59 (55%) 10 (20%) 41 (80%) PurpleGallinule 15 (32%) 32 (68%) 10 (14%) 60 (86%)

on 3 August 1977 and 31 July 1978 and limited nestingin rice fieldsto approximately70 d. Clutch sizewas determinedfor 100 CommonMoorhen and 92 Purple Gallinule nests.Egg numberswere determinedfor completedclutches only.No differenceswere notedbetween marsh sites in 1977 (Common Moorhen, t = 0.65, df = 47, P > 0.05; Purple Gallinule, t = 1.05, df = 22, P > 0.05) or between marsh sites in 1977 and 1978 (Common Moorhen, t = 0.85, df = 57, P > 0.05; Purple Gallinule, t = 0.48, df = 29, P > 0.05), sodata were combined.Average clutch sizes for Common Moorhen were 6.7 in the marsh (n -- 59) and 8.8 in rice fields (n = 41). The Purple Gallinuleslaid an averageof 5.8 eggsin the marsh(n = 32) and 8.6 in rice fields (n = 60). Both speciesproduced larger clutchesin rice fields than in marshes(Common Moorhen, t = 5.3, df = 98, P < 0.05; Purple Gallinule, t = 7.9, df = 90, P < 0.05). A regressionof clutch size by 10-d periodsover time for each speciesin each habitat revealedthat no trends exist (P > 0.05) (Common Moorhen in marsh, F = 0.19; in rice, F = 0.87; Purple Gallinule in marsh, F = 0.56; in rice, F = 0.06). Over the two years, Purple Gallinule had a successrate of 78.6% comparedto 63.3% for Common Moorhen (Table 1). There was no relationshipfor either speciesbetween success rate and height above water level (Common Moorhen, t = 0.71, P > 0.05; Purple Gallinule, t = 0.43, P > 0.05). Of 25 unsuccessfulPurple Gallinule nests, 20 (80%) were lost to predators,one (4%) was abandonedafter floatingaway, and four (16%) were abandoned for unknown reasons. Of 58 unsuccessful Common Moorhen nests,37 (64%) were lost to predators,10 (17%) destroyedby weather or high water, two (3%) were abandonedafter floating away, and nine (16%) were vacatedfor unknown reasons.

DISCUSSION We notedsimilarities and differencesbetween selected aspects of Com- mon Moorhen and Purple Gallinule breedingbiology along the upper Gulf Coast. No differenceswere noted in nest composition;both species usedparts of thoseplants growing near the site.Bowl depthand diameter of CommonMoorhen nestswere greater,perhaps reflecting their greater weight (Ripley 1977), but there were no differencesin external nest Vol.58, No. ! NestingofCommon Moorhen and Purple Gallinule [59

TABLE 2. Clutch sizesreported for the Purple Gallinule and CommonMoorhen.

Area Mean n Range Source Purple Gallinule Panama 4 1 4 Grossand Van Tyne 1929 Louisiana 4.5 12 3-6 Bell 1976 South Texas 6.5 87 4-12 Cottam and Glazener 1959 Louisiana 6.7 38 4-10 Causeyet al. 1968 Louisiana Marsh 5.8 32 3-8 this study Rice field 8.6 60 2-13 this study Common Moorhen Iowa 7.1 13 5-10 Fredrickson 1971 Ohio 8.0 55 3-15 Brackneyand Bookhout1982 Louisiana 8.1 11 5-15 Bell 1976 Wisconsin 8.1 18 5-12 Krauth 1972 Louisiana 8.6 35 6-16 Causeyet al. 1968 South Texas 9.1 142 4-17 Cottam and Glazener 1959 Pennsylvania 10.0 2 10 Miller 1910 Pennsylaniaand New Jersey 10.0 26 6-14 Harlow 1918 Louisiana Marsh 6.7 59 3-10 this study Rice field 8.8 41 5-10 this study

dimensions.Both speciesnested most often over water and water depth under nestsdid not differ between them. Yet, Purple Gallinules built nestshigher abovethe water surfaceunder thicker coverand had greater nesting success. Clutch sizesreported for the Common Moorhen in the New World (:• = 8.8, recalculatedfrom originaldata; range 3-17) are similar to those that we observedin rice fields (Table 2). Reportedclutch sizes (:• = 6.4; range3-12) for Purple Gallinule fall betweenour marshand rice field averages(Table 2). Interspecificdifferences in clutch size within each habitat were less than intraspecificdifferences between habitats. Common Moorhens laid a mean of 0.9 more eggsin marsh and 0.2 more eggsin rice fieldsthan Purple Gallinules. These differencesare well below thosereported by other researchers:Common Moorhens average2.6 more eggsthan Purple Gallinules in southernTexas (Cottam and Glazener 1959), and 3.6 more eggsfrom a small sampleof nestsin an earlier Louisianastudy (Bell 1976). Possibleexplanations for the significanthabitat differencesin clutch size that we observedinclude: (1) a samplingartifact, (2) divergentevo- lution in isolatedpopulations, or (3) differencesin resourcesavailable for eggproduction and/or the successfulrearing of chicks. The differencescould be spuriousif large mean marsh clutchesrep- resentan averagingof very largeearly clutcheswith smallerlater clutches 60] R. N. Helmet al. j. FieldOrnithol. Winter 1987 that are not different in size from clutches laid at the same time in rice where early nestingis precludedby a lack of vegetation.Indeed, observations of juvenile Common Moorhens associatedwith adults nesting late in the seasonin the marsh is evidencethat secondbroods are pro- ducedin that habitat. However, a regressionanalysis of each speciesin each habitat revealedno changeover time. Therefore, we think the re- suitsare not due to a samplingartifact. Noordwijk et al. (1981) commentedthat differencesof a few eggsin clutchsize of Great Tits (Parusmajor) couldevolve in a matter of de- cades.Rice and marsh populationsof both specieswould have to exhibit very little interbreedingfor the differencesto be geneticallybased. How- ever, there are no apparent geographicbarriers acrossthe 70 km of apparentlysuitable habitat that separatesthe studysites. Also, banding data show that there is at least occasionalmixing of birds betweenmarsh and rice productionareas. Thus, significantgenetic divergence does not appear likely. There may be differencesin the types of resourcesavailable between the sites.Both speciesare omnivorous(Reagan 1977), but they may be more insectivorousin rice fieldswhere plant parts they normally consume (Howell 1932, Miller 1946, Simpson1939) are generallynot available (Harmon et al. 1960). A comparativestudy of food habits in the two habitatscould help determinethe correlationof available resourceswith eggproduction and/or the successfulrearing of chicks. ACKNOWLEDGMENTS

We thank R. B. Hamilton, P. Mason, D. P. Pashley,and J. V. Remsen,Jr. for reading and improvingthis manuscript.The field study was developedand conductedunder the directionof J. D. Newsom and P. J. Zwank. Data were collectedand initially analyzed by R. N. Helm. Further analysisand assistancewith manuscriptpreparation were by D. N. Pashleywhile being supportedby a LouisianaState University Alumni Federation Fellowship.Contribution of LouisianaCooperative Fish and Wildlife ResearchUnit: Lou- isiana State University and Agricultural Center, Louisiana Department of Wildlife and Fisheries,U.S. Fish and Wildlife Service,and Wildlife Management Institute, cooperating.

LITERATURE CITED

BELL,G. 1976. Ecologicalobservations of Common and Purple gallinuleson Lacassine National Wildlife Refuge. M.S. thesis,Univ. of SouthwesternLouisiana, Lafayette, Louisiana. BRACKNEY,t. W., AND T. t. BOOKHOUT. 1982. Population ecologyof Common Galli- nules in southwesternLake Erie marshes.Ohio J. Sci. 82:229-237. CAUSE¾,M. K., F. L. BONNER,AND J. B. GRAVES. 1968. Dieldrin residuesin the galli- nulesPorphyrula martinica L. and Gallinulachloropus L. and its effect on clutchsize and hatchability.Bull. Environ. Contam. Toxicol. 3:274-283. COTTAM,C., AND W. C. GLAZENER.1959. Late nestingof water birds in southTexas. Trans. N. Am. Wildl. Nat. Resour. Conf. 24:382-395. FREDRICKSON,L. H. 1971. Common Gallinule breedingbiology and development.Auk 88:914-919. GRoss,A. O., ANDJ. VAN TYNE. 1929. The Purple Gallinule of Barro ColoradoIsland (Canal Zone). Auk 46:431-466. HARLOW,R.C. 1918. Notes on the breedingbirds of Pennsylvaniaand New Jersey.Auk 35:18-29. Vol.58, No. 1 Nestingof Common Moorhen and Purple Gallinule [61

HARMON, B. G., C. H. THOMAS,AND L. L. GLASGOW. 1960. Waterfowl foodsin Loui- siana rice fields. Trans. N. Am. Wildl. Nat. Resour. Conf. 25:153-161. HELM, R.N. 1982. Chronologicalnesting study of Commonand Purple gallinulesin the marshlands and rice fields of southwest Louisiana. M.S. thesis, Louisiana State Univ., Baton Rouge, Louisiana. HOWELL,A. H. 1932. Florida bird life. Florida Dept. of Game and Fresh Water Fish, Tallahassee, Florida. KRAUTH,S. 1972. The breedingbiology of the CommonGallinule. M.S. thesis,Univ. of Wisconsin, Oshkosh, Wisconsin. MILLER, R. F. 1910. Notes on the Florida Gallinule (Gallinulagaleata) in Philadelphia County, Pa. Auk 27:181-184. 1946. The Florida Gallinule, breedingbirds of the Philadelphiaregion (part 3). Ca•sinia36:1-16. NOORDWIJK,A. J. VAN, J. H. VAN BALEN,AND W. SCHARLOO.1981. Genetic and environmentalvariation in clutch size of the Great Tit. Neth. J. Zool. 31:342-372. ONE¾,J. 1954. Final report:Clapper Rail surveyand investigationstudy. Georgia Game and Fish Comm., Atlanta, Georgia. REAGAN,W.W. 1977. Resourcepartitioning in the North Americangallinules in south- ern Texas. M.S. thesis,Utah State Univ., Logan, Utah. RE^RDEN,J.D. 1951. Identificationof waterfowlnest predators. J. Wildl. Manage. 15: 386-395. Rn'LE¾,S. D. 1977. Rails of the world. M. F. Fenelay Publ. Ltd., Toronto, Ontario. SHAW,S. P., AND C. G. FREDINE. 1956. Wetlands of the United States,their extent and their value to waterfowl and other wildlife. U.S. Fish and Wildl. Serv. Circ. 39. SIMPSON,T. W. 1939. The feeding habits of the Coot, Florida Gallinule, and Least Bittern on Reelfoot Lake. J. Tenn. Acad. Sci. 14:110-115. Received31 March 1986; accepted10 Sept. 1986.

• Current address:Louisiana Department of Wildlife and Fisheries, Baton Rouge,Louisiana 70895 USA.