The Condor 86:301-304 0 The CooperOrnithological Society 1984

NEST SITE SELECTION BY MASKED AND BLUE-FOOTED ON ISLA ESPAfiOLA, GALAPAGOS

DAVID CAMERON DUFFY

ABSTRACT.-On Punta Suarez, Isla Espanola, GalapagosIslands, Masked Boo- bies (Sula dactylatra) nested at the edge of a cliff while Blue-footed Boobies (Sula nebouxiz)nested farther inland. Elsewhere in the Galapagos, Blue-footed Boobies nested near cliff edgesat seven of eight sites where Masked Boobies were absent. Masked Boobies appeared to have difficulty taking flight except from the cliff edge, while Blue-footed Boobies were able to take off from the interior. Neither speciesappeared dominant in interspecific interactions or could displacethe other from occupied nest sites. Blue-footed Boobies may avoid nesting near cliff edges where Masked Boobies are presentin to avoid more frequent conflicts caused by a restricted habitat. Any such competition which occurs seemsto involve only a small percentage of either speciesand seemsunlikely to limit the numbers of breeding or the total population.

Boobies (Ma spp.) nest in habitats that range made my observations at the southern part of from tree-tops to cliffs, although most of the the point, along about 100 m of cliff edge and speciesnest on level ground (Nelson 1978). I extending about 50 m inland. The terrain is investigated how two of these , the level, with guano-covered rocks at the top of Masked or White (S. dactylatra) and Blue- a 7-m cliff giving way to low herbaceousvege- footed (S. nebouxiz) boobies, share nesting tation and scattered rocks farther inland. Ex- habitat in the Galapagos Islands. cept for a few l-m bushes, the vegetation is Nelson (1978892) suggestedthat “With a not tall enough to obstruct the perambulations few local exceptions, such as the crater on of boobies (cf. fig. 220b in Nelson 1978:522 Daphne Major, there are no habitats suitable which showshabitat similar to the study area). for the blue-foot, but not for the brown [S. The site was not visible from tourist trails and leucogaster]or white boobies.” He felt that remained undisturbed except for my visits. either no “important” competition takes place I also visited 10 other nesting areas or the is dominant over the during 1980-l 98 1. Isla Isabela-Tagus Cove, Blue-footed Booby. He mentioned Punta Beagle Crater, Islas Marielas in Elizabeth Bay, Suarez, Isla Espafiola, GalapagosIslands as an Caleta Iguana, and Punta Vincente Rota; Isla example of a place where habitat partitioning Espafiola- Punta Cevallos; Isla Seymour occurs. The Masked Boobies there occupy a Norte; Isla Floreana-Punta ; Isla “strip facing the open sea” and the Blue-footed Daphne Major; and Isla Fernandina-Cabo Boobies nest farther inland. I decided to in- Douglas.Blue-footed Boobiesnested at all these vestigate the following questions in the Gala- sitesbut Masked Boobies nested only at Punta pagos: Cevallos and Daphne Major. 1. Do the two speciesactually differ in the distance between their nests and the sea at METHODS Punta Suarez? Approximately 90 Masked and 40 Blue-footed 2. Is the Masked Booby dominant over the boobies were present at the study area on Pun- Blue-footed Booby? ta Suarez during 2-6 May 198 1. Most Masked 3. Do the two speciesdiffer in their behavior Boobies were either feeding large young or at- while taking flight, a factor that might affect choice of nesting site? 4. Does the Blue-footed Booby nest closer to the sea in the absenceof the Masked Booby TABLE 1. Distance from nests to cliff edge for Blue- footed and Masked boobies at Punta Suarez. at other locations in the archipelago?

Distance Blue-footedBooby Masked Booby STUDY AREAS (m) (n = 20) (n = 40)

Isla Espafiola is the southeastern-mostof the O-2 m 0 40 Galapagos Islands. Punta Suarez, its western 3-5 m 10 22.5 point, is a major nesting area for Waved Al- 6-10 m 20 25 batross (Diomedea irrorata), Masked, and 1 l-20 m 20 7.5 Blue-footed boobies (Nelson 1968, 1978). I >20m 50 5

t3011 302 DAVID CAMERON DUPFY

TABLE 2. Colonies visited, nesting habitats used, and population estimates for Blue-footed and Masked boobies (colony size estimatesare numbers of adult individuals; cliff nestsare those within 2 m of a cl@ inland nestsare those farther than 2 m from cliffs).

Blue-footed Booby Masked Booby Habitat Habitat Colony size - Colony s,ze Colony Cliff Inland (individuals) Cliff Inland (individuals)

Tagus Cove X 102 0 BeagleCr. X X 103 Cal. Iguana X X 102 i Is. Marielas X 102 Vincente Rota X X 103 8 Pta. Cevallos X 102-103 X X 10’ Seymour N. X X 10’ 0 Pta. Cormorant X 102 0 Daphne Major X X 103 X X 102 Cabo Douglas X X lo’-104 0 Punta Suarez X 10’ X X 10’

tending nest sites in pairs. Pair-maintenance through the nestingareas or loafing away from behavior (e.g., Sky-pointing, Parading, Mutual any apparent nest site. Jabbing, Allopreening, and Symbolic Nest The other colonies were visited only briefly, Building, cf. Nelson 1978) was common. Most during the course of other field work. I made Blue-footed Boobies had eggs,although pairs rough estimatesof colony size. If I sawboobies in all stages of breeding were present, from of either speciesat nest siteswithin 2 m of cliff courting pairs to thosewith large downy young. edgesor steep (> 30 degrees)drop offs, I con- To determine if the distribution of nests of sidered these as cliff-edge nesters. Since boo- the two speciesdiffered at Punta Suarez, I re- bies frequently roost on cliff ledges, I did not corded the distance to the windward seacliff count birds but only nestswith eggsor young. for 20 Blue-footed and 40 Masked booby pairs. Although I considered cliff nesting to occur Distances were divided into five classes:O-2 within a colony if even a single pair nested m; 3-5 m; 6-10 m; 1 l-20 m; and >20 m. To within 2 m of an edge, in all casesmore than see if the speciesdiffered in the locations used one pair was involved. for take-offs, I spent 247 min on four after- noons estimating the distancebetween the cliff RESULTS edge and the locations where the two species At Punta Suarez, over 70% of Masked Booby took off within the study colony. Take-offs were pairs occupied sites within 10 m of the cliff divided into cliff take-offs that occurred within edge (Table 1); 32% of all pairs nested within 2 m of the cliff edge, or interior take-offs that 2 m. In contrast, 70% of all Blue-footed Boo- occurred more than 2 m from the edge. For bies nested more than 10 m from the cliff and birds that did not take off from the nest sites, none were within 2 m of the cliff edge. The I estimated the distance walked before taking distributions differed significantly (P < 0.001; flight. During the four observation periods, the 5 x 2 contingency table; x2 = 19.36; Siegel wind was from the southeast,striking the cliff 1956). face directly at an estimated speed of 20-30 In 21 take-offs by Blue-footed Boobies, all kph. occurred more than 2 m from the cliff. In 28 I also watched for interspecific interactions take-offs by Masked Boobies, 24 occurred in order to see if they might determine nesting within 2 m of the cliff edge. The difference is locations of the two species.I considered Yes highly significant (P < 0.001; 2 x 2 contin- Head-shaking, Jabbing, and Wing-flailing dis- gencytable with correction for continuity; x2 = plays (Nelson 1978:551) as interspecific 3 1.93). Of the 24 Masked Boobies that walked aggressionwhen these behaviors appeared to to the cliff edge before taking off, 16 traveled be directed at the other species.Outcomes of a mean distanceof 1.9 m (SD = 1.3). Four that these encounters were divided into two cate- took off inland walked a mean distance of 7.5 gories, based on whether site-holders retained m (SD = 2.9). Two of these birds had great or lost possessionof their sites. Site-holders difficulty in remaining airborne and they did were singlebirds or pairs that had eggsor young, not gain more than 0.5 m in altitude until they or were pairs associatedwith nest depressions. reached the updraft at the cliff edge. Birds without sites were those wandering Significantly more Masked Boobies took off BOOBY NEST SITES 303

TABLE 3. Measurements of Blue-footed and Masked boobies (data from Nelson 1978:322, 5 11).

Blue-footed Booby Masked Booby Male F~lllale Male F~lll&Z

Wing length (mm): mean 432 457 451 484 range 406-438 432-470 445-457 457-495 sample 9 8 6 5 Mass (g): mean 1,283 1,801 1,627 1,881 range l,lOO-1,580 1,450-2,230 1,220-1,970 1,470-2,350 sample 23 28 48 37 Ratio of mass to wing length 2.97 3.94 3.61 3.89

from the cliff edge than would be expected length, male Blue-footed Boobies are lighter from the distribution of nests(P < 0.001; 2 x than female Blue-footed Boobiesand both sexes 2 contingency test with correction for conti- of the Masked Booby, which appear similar nuity; nest data reduced to two classes:less (Table 3). Hence, female Blue-footed Boobies than and more than 2 m; x2 = 12.39). should have the same difficulty with take-offs I saw few interspecific interactions. In nine as do Masked Boobies. Female Blue-footed conflicts at nest sites,all were “standoffs” with Boobies may have wider wings, giving them neither speciesbeing displaced. When at nest greater lift at lower speeds;they appear, how- sites, Masked Boobies displaced non-nesting ever, to have more difficulty becoming air- Blue-footed Boobiesin all five cases(P = 0.03 1; borne than do males (Nelson 1978). binomial distribution with P = 0.5).In the re- Although difficulty in taking off inland may verse situation, Blue-footed Boobies on nests explain why Masked Boobies prefer to nest displaced non-nesting Masked Boobies twice close to the cliff, it does not explain why Blue- and “standoffs” occurred twice. Nest-site footed Boobies do not nest on the cliff edges holders of both specieswere never displaced at Punta Suarez. This speciesuses ledges and by intruders. I saw no conflicts between the cliff edges elsewhere in the Galapagos where speciesaway from nest sites at Punta Suarez. the Masked Booby does not nest, suggesting Only Blue-footed Boobies nested at 8 of the that the latter preempts cliff nesting at Punta 11 colonies visited, including Punta Suarez. At Suarez. seven of these eight colonies where Masked Aggressive conflicts were rare between the Boobies were absent: Blue-footed Boobies species.I saw no indication that Masked Boo- nestedwithin 2 m of cliff edges(Table 2). Blue- bies were dominant over Blue-footed Boobies footed Boobies did not nest near cliffs at two at nest sites, as suggestedby Nelson (1978: colonies where both specieswere present, and 892). Territory holders could not be displaced only a few did so at the third colony, on Isla by the other species:site tenure, rather than Daphne Major. Nests on this island were sit- species,determined the outcome of aggressive uated on the upper crater floor, where Masked encounters. Blue-footed Boobies may simply Boobies do not nest (T. Price, pers. comm.). avoid the higher densities at cliff edgeswhere Masked Boobies are present, rather than being DISCUSSION actively displaced. Why did Masked Boobies nest at cliff edges On Punta Suarez, only small percentagesof while Blue-footed Boobies nested farther in- the two specieswere exposed to each other, land? The distribution of take-offs by Masked primarily in the zone 3-20 m from the cliff. Boobies stronglyfavored the cliff edge. Masked Within this area of overlap, competition Boobies nestinginland often walked to the cliff seemed not to be aggressive but rather a edge. Those that took off inland had difficulty “scramble” (Miller 1967) or perhaps totipal- staying aloft. At wind speedsless than 20-30 mate waddle, to occupy nest sites. Any inter- kph, take-offs would have been even more dif- specific competition for space that occurs is ficult. Blue-footed Boobies generally took off likely to have little effect on total breeding from their nest sites, even when well inland. I numbers. The situation described here agrees surmise that the difference in choice of take- with suggestionsby Ashmole (1963) and Nel- off locations was related to differencesin wing son (1978) that nestingspace rarely limits pop- shape and wing loading. In relation to wing ulations of tropical . 304 DAVID CAMERON DUFFY

ACKNOWLEDGMENTS MILLER,R. S. 1967. Pattern and processin competition. I thank C. Rechten and C. Vasquez for sharingtheir camp Adv. Ecol. Res. 4:1-79. on Espaiiola and J. B. Nelson, T. Price, W. R. Siegfried, NELSON,J. B. 1968. Galapagos:islands of birds. Long- and two reviewers for information and comments on the mans, London. manuscript. El Servicio National de Parque Galhpagos NELSON, J. B. 1978. The : gannetsand boobies. allowed me to work in the islands. The personnel of the Oxford Univ. Press, Oxford. Darwin Station provided logistic support. The crew and SIEGEL, S. 1956. Nonparametric statisticsfor the behav- scientistson BeagleIV made possibleseveral difficult land- ioral sciences.McGraw-Hill, New York. ings. This study was part of the monitoring pro- gram of the Darwin Station and was supported by the Charles Darwin ResearchStation, Isla Santa Cruz, Ga- World Wildlife Fund. This is contribution number 358 of lfpagos Islands, Ecuador. Present address:Percy Fitz- the Charles Darwin Foundation. Patrick InstituteofAf?ican Ornitholom. Universitvof Cane Town. Rondebosih?700, South Afrsa. Received 5 Mgy LITERATURE CITED 1983. Final acceptance17 January 1984. ASHMOLE,N. P. 1963. The regulation of numbers of tropical seabirds.Ibis 103b:458-473.

The Condor86:304 0 The CooperOrnithological Society 1984 RECENT PUBLICATIONS

The Marine Ecology of Birds in the Ross Sea, Antarcti- A Field Guide to the Grand Canyon.-Stephen Whitney. ca.-David G. Ainley, Edmund F. O’Connor, and Robert 1982. Quill, New York. 320 p. Paper cover. $12.50. This J. Boekelheide. 1984. Ornithological MonographsNo. 32, easily-carriedvolume offersa little of everything about the American Ornithologists’ Union, Washington, DC. 97 p. natural history of the Grand Canyon. At least for birds Papercover. $9.00 ($8.00 to AOU members).Source: Frank and their habitats, its treatment is not very successful.The R. Moore, Assistant to the Treasurer, Department of Bi- five pagesof text on birds are lessthan illuminating, while ology, University of Southern Mississippi, Southern Sta- the 15 plates of illustrations are uneven, with mostly tion Box 5018, Hattiesburg, MS 39406; orders must be stiff and generalizedposes. By contrast, the plates of but- prepaid and include a $0.50 handling charge.This mono- terfliesand herptilesare convincingand accurate.Provides graph attempts “to explain the factors that determine sea- 87 plates, 17 figures, index.-J. Tate. bird occurrenceand distribution in the Ross Sea during summer.” Data from seabird censuseson six cruisesare correlated with oceanographicmeasurements and dietary Myology of the Purple-throated Carib (Eulampisjugula- information from collected specimens. Text and charts ris) and Other Hummingbirds (Aves: Trochilidae).- describe the environment and give the findings for 21 Richard L. Zusi and Gregory Dean Bentz. 1984. Smithson- species(Emperor Penguin to Arctic Tern). These results ian Contributions to Zoology No. 385, Smithsonian In- are analyzedand discussedwith regardto the birds’ feeding stitution Press, Washington, DC. 70 p. Paper cover. This behavior, trophic relations, community composition and report describes and illustrates the gross morphology of biomass,and the factors affecting seabird occurrence.The all the muscles attaching on the axial, appendicular, or study reachesimportant conclusionsabout prey selection hyoid skeletons,and the integument. The condition in the and the importance of food abundanceas a limiting factor principalspecies is comparedwith thosein six other species, for high latitude Antarctic bird populations in summer, chosen to represent the major taxonomic subdivisions ideas that may well apply elsewhere. Robert Cushman within the . These findingsare discussedwith regard Murphy would be proud to seehow far his light hascarried. to systematic studies within the Trochilidae and phylo- References,appendixes. genetic relationships of that group to other birds. The authorshave indicated the wealth of structuraladaptations Nature Through Tropical Windows.--Alexander F. Skutch, that are yet to be examined in hummingbirds, and their illustrated by Dana Gardner. 1983. University of Califor- paper is a solid place to begin. Line drawings, references, nia Press, Berkeley. 374 p. $19.95. “It took Kropotkin to index. take the edge off Darwin,” Alexander Skutch has said, voicing profound dissatisfactionwith a vision of nature red in tooth and claw. In his latest book Skutch invests Migration: Paths Through Time and Space.-R. Robin his always painstaking, always valuable accountsof nat- Baker. 1983. Hodder and Stoughton [England]. 248 p. ural history with a senseof his perplexity over the relative $15.95 paperback, $32.50 hardcover. Available: Holmes importance of competition and cooperation in evolution. and Meie; Publishers, Inc., IUB Building, 30 Irving PI., The dissonancebetween graceful descriptionsof what he New York. NY 10003. This is a seauel to. and offsminc! seesas cooperationand reflectionson current views of the of, Baker’s The Evolutionay EcologyofAnimalMigr;tion selfish nature of social behavior make the book worth (1978; noticed in Condor 8 1:12 1). In this book, ethologists thinking about. Lyric writing and elegant illustrations re- are taken to task for their insensitive dissection of pro- flect forty years of unhurried days spent watching the dis- grammed reflex responses,while population ecologistsare appearing tropical world. Although the writing evokes criticized for their calculated descriptions of dispersion, shadesof eighteenthand nineteenth century naturalists,it and population regulation mechanisms. To re- would be a mistake to dismiss Skutch as an isolated ec- place these concepts,we are offered the “individual”: less centric, a relic of Natural Theology. In this book, which than a population, but more than a set of measurable addresses-more explicitly than earlier works-his con- responsesto stimuli. Migration theory, accordingto Baker, cernsabout allowing the predictionsof population genetics is the description of the life path of an integrated, quite to dominate our explanations of social behavior, Skutch anthropomorphic organism making its way through life, is frequently successfulat revealing how highly contrived acrossgreat distances,while remaining true only to itself. some of our currently fashionable selfish-geneideas may Illustrative material identical to the earlier book. Glossary, be.-M. F. Lawton. referencesand author indices.- J. Tate.