SHORT COMMUNICATIONS 289

eaters (Meropidae) and cooperative breeding in SAFRIEL,U. N., M. P. HARRIS,M. DE L. BROOKE,AND hot-climate birds. Ibis 114:1-14. C. K. BRITTON. 1984. Survival of breeding oys- GIROUX, J. F. 1985. Nest sites and superclutchesof tercatchersHaematopus ostralegus. J. Anim. Ecol. American Avocetson artificialislands. Can. J. Zool. 53:867-877. 63:1302-1305. SIBLEY,C. G., J. E. AHLQUIST,AND G. L. MONROE, JR. GROVES, S. 1984. Chick growth, sibling rivalry, and 1988. A classification of the living birds of the chick production in American Black Oystercatch- world based on DNA-DNA hybridization studies. ers. Auk 101525-531. Auk 105:408423. LAURO,B., ANDJ. BURGER. 1989. Nest-site selection SMITH,J. N. M. 1990. Summary, p. 593-6 11. In P. of American Oystercatchers(Haematopus pallia- B. Stacey and W. D. Koenig [eds.], Cooperative tus) in salt-marshes.Auk 106:185-192. breedingin birds: long-term studiesof ecologyand MAYR, E. 1939. The sex ratio in wild birds. Am. Nat. behavior. Cambridge Univ. Press,Cambridge, En- 73:156-179. gland. NOL, E. 1985. Sex roles in the American Oystercatch- STACEY,P. B., AND J. D. LIGON. 1987. Territory qual- er. Behaviour 95:232-260. ity and dispersaloptions in the Acorn Woodpeck- NOL, E. 1989. Food supply and reproductive perfor- er, and a challengeto the habitat-saturationmodel mance of the American Oystercatcherin Virginia. of cooperative breeding. Am. Nat. 130:654-676. Condor 91429-435. VAN RHUN, J. G. 1990. Unidirectionality in the phy- NOL, E., A. J. BAKER,AND M. D. CADMAN. 1984. logeny of social organization, with special refer- Clutch initiation dates, clutch size, and egg size ence to birds. Behaviour 115:153-l 73. of the American Oystercatcher in Virginia. Auk WALTERSJ.. AND B. F. WALTERS. 1980. Co-onerative 101:855-867. breedingby SouthernLapwings Vanelluscklensis. RUSSELL,E. M. 1989. Co-operative breeding-a Ibis 122:505-509. Gondwanan perspective. Emu 89:61-62. ZARUDSKY,J. D. 1985. Breeding statusof the Amer- SAFRIEL,U. N. 1981. Social hierarchy among siblings ican Oystercatcher in the town of Hempstead. in broods of the Oystercatcher Haematopus os- Kingbird 35:105-l 13. tralegus.Behav. Ecol. Sociobiol. 9159-63.

The Condor 941289-292 0 The CooperOrnithological Society 1992

SCARCITY OF HAEMATOZOA IN BIRDS BREEDING ON THE ARCTIC TUNDRA OF NORTH AMERICA ’

GORDON F. BENNETT International ReferenceCentre for Avian Haematozoa, Memorial Universityof Newfoundland, St. Johns,’ Newfoundland,AIB 3X9, Canada

ROBERT MONTG~MERIE AND GILLES SEUTIN* Department of Biology, Queens’ University,Kingston, Ontario, K7L 3N6, Canada

Key words: Haematozoa; tundra; arctic:parasites; American arctic-nesting birds. In a review of hae- longspur;ptarmigan; redpoll;plumage color. matozoanprevalence in North American birds, Greiner et al. (1975) indicate that lessthan 3% ofbirds sampled Despite widespreadinterest in documentingthe from the “arctic barrens” (their region 6) were para- parasites of birds (e.g. Loye and Zuk 1990) there is sitized, but they provide no further information on the little information available on the haematozoa of spe- sample of birds involved in this analysis. cies occurring in arctic regions. Laird (196 1) reported In this paper, we report on the haematozoa found that none of the 149 individuals of 23 bird specieshe in 276 breeding birds of 10 species samnled in the samuled on Prince of Wales Island (72-74”N. 96- courseof field studiesof their behavior and ecologyat 103”W) during one summer harbored haematozoa,but four very different arctic sites. While some of these no other intensive surveys have been done of North specieshave been sampled for haematozoabefore, our samples allow us to compare haematozoa prevalence between habitat types, both within and between sites. I Received 24 June 199 1. Accepted 27 September We also discuss the implications of our findings for 1991. recent comparative analysesof parasite prevalence in 2Present address: Department of Biology, Univer- relation to plumage brightness in birds (see Moller sity of Pennsylvania, Philadelphia, PA 19104. 1990). 290 SHORT COMMUNICATIONS

TABLE 1. Prevalence of Haemoproteus (H), Leucocytozoon (L) and Trypanosoma (T) in tundra-nestingbirds.

Total individuals Site Individuals with Species Examined Infected H L T Sarcpa Lake Lagopus mutt.0 39 Eremophila alpestris 3 Calcarius lapponicus 29 Plectrophenax nivalis 8 Churchill Carduelisjlammea 97 63 53 Carduelis hornemanni 19 12 12 Carduelis sp. 6 Calcarius pictus 19 3 Zonotrichia leucophrys 4 4 : St. Paul Island Aethia pusilla 14 0 Fort Simpson Cygnus buccinator 38 25 18 8 0 Total 276 108 26 80 13 Percent 39.1 9.4 29.0 4.1

METHODS We also report on blood samples collected in June Birds were captured opportunistically in a variety of and July 1987 from Least Auklets (Aethiapusilla) nest- ways (mist nets, noose carpets,noose poles and potter ing on St. Paul Island, Pribilof Islands,Alaska (57”08’N, traps), often in conjunction with other researchon their 170”17’W) and in July and August 1989 from Trum- behavior and ecology. All individuals were banded to peter Swans (Cygnus buccinator) nesting on the Mac- facilitate later identification. Blood smearswere made kenzie River near Fort Simpson, NWT (62”N, 122”W). following the protocols of Bennett (1970). All smears These two samplespermit comparisonwith those from were air-dried in the field and fixed in 100% ethanol related speciesreported elsewhere.St. Paul Island is a within 24 hr. At the International ReferenceCentre for small oceanic island in the Bering Sea, well beyond the Avian Haematozoa, smears were stained with Giem- arctic limit of trees and tall shrubs,whereas Fort Simp- sa’s stain and examined for Leucocytozoon, Trypano- son is at the southern edge of the forest-tundra/open soma and microfilaria at 250 x and for Haemoproteus forest vegetationzone (seeDanks 198l), about 500 km and at 400x magnification. Each slide southwestof the arctic treeline. was examined until approximately 100,000 erythro- At Sarcpa Lake, Churchill and on St. Paul Island, cytes had been scanned.In this paper we mainly report all blood smearswere taken from adult breeding birds; parasiteprevalence, i.e., whether or not each individual at Fort Simpson, they were taken from 25 adult breed- was infected with eachof thesetaxa. Parasiteintensities ers and 13 goslings.Adults of both sexes were repre- (i.e., infestation levels per individual bird) tend to vary sented in all samples except Horned Larks (females seasonallyand among age and sex cohorts (Weather- only). head and Bennett 1991) and are thus not very useful in interspecificcomparisons unless these factors can be RESULTS AND DISCUSSION statistically or experimentally controlled. Sarcpa Lake, Northwest Territories. No haematozoa Most of our samples were collected at Sarcpa Lake, were found in the 79 individual birds of four species Melville Peninsula, Northwest Territories (68”33’N, sampled at Sarcpa Lake (Table 1). Both Rock Ptar- 83”19’W) during June and July 1987-1989 and at migan (Lagopus mutus) and Lapland Longspurs(Cal- Churchill, Manitoba (58”45’N, 94”05’W) during June car&s lapponicus) were sampled during the breeding and July 1989. The habitat at Sarcpa Lake is upland seasonin three different years; nine individual Rock high arctic tundra (see Montgomerie et al. 1983 for Ptarmigan and two individual Lapland Longspurswere details), more than 1,000 km north of the northern sampled in more than one year. Thus, the absenceof limit of treesat this loneitude (Danks 1981). Churchill. haematozoa in breeding birds at this site appearsto be on the other hand, is onthe arctic treeline at the north: a general phenomenon. ern edge of the forest-tundra/open boreal forest vege- Forty-three individuals of the same four speciesalso tation zone (see Fig. 3 in Danks 1981). All birds cap- were sampled by Laird (1961) in a single season on tured at Churchill were breeding within 5 km of the Prince of Wales Island and no haematozoawere found Hudson Bay coast, either on relatively open tundra there either. Horned Larks (Eremophila alpestris) and with scatteredclumps of black spruce (Picea mariana Rock Ptarmigan also nest south of the northern limit (Mill.) BSP) or within the spruce stands themselves of trees where haematozoa have been found in their (see Jehl and Smith 1970 for habitat details). blood - 36% of 50 Rock Ptarmigan from insular New- SHORT COMMUNICATIONS 291

foundland harbored haematozoa (Bennett and Inder exposureto biting compared to sympatric species 1972) and 13% of 23 Homed Larks from St. Bride’s occupyingmore shelteredhabitats. A similar argument Newfoundland and False River, Ungava Bay, Quebec was made by Laird (196 1) to explain the low prevalence had either Leucocytozoon or Haemoproteus (unpub- of parasites in several Charadriiforme species when lished data, see Greiner et al. 1975). Thus, the absence compared to sympatrically breeding speciesof other of haematozoain thesespecies on the high arctic tundra avian orders at Lac Kohlmeister, Quebec. However, it is not a species-specificphenomenon, but rather seems should be noted that the Charadriidae in general are to be site-related. seldom infected with blood parasites-only 1.2% of The complete absence of avian haematozoa from individuals in this family, sampled in North America, birds nesting in high arctic regions is almost certainly had haematozoa (Greiner et al. 1975) and it is not due to the absenceof suitable vectors (see also Laird known whether this is due to microhabitat or some 1961, Greiner et al. 1975) especiallythe various spe- parasite resistancein this taxon. cies of omithophilic black flies (Simuliidae; see Danks Fort Simpson, Northwest Territories. Twentv-live 1981). The fact that haematozoa have been found in adult Trumpeter Swans(Cygnus buccinator) were-Sam- both Homed Larks and Rock Ptarmigan nesting south nled near Fort Simoson. of which 19 (76%) were in- of the arctic treeline supportsthis argument. The four fected with haematozoa:’L. simondi, i. net&onis and bird speciessampled at SarcpaLake also winter mainly H. greineri. At this boreal forest site, six (46%) of the in the northern half of North America where there 13 goslingssampled were also infected with L. simondi would be no active vectors for haematozoa in the win- and H. greineri, indicating that suitable vectors for ter. Thus, the only potential vectors for haematozoa thesehaematozoa were presenton the breedinggrounds that they might have encountered on the breeding and that transmission occurred there. grounds are mosquitoes (Culicidae). Mosquitoes are The prevalence (66%; Table 1) of blood parasitesin not particularly common at SarcpaLake, but it is also swans from Fort Simpson is similar to that observed possible that arctic speciesdo not act as vectors for in seven other anatid speciesat Goose Bay, Labrador avian haematozoa. (where 48% of 56 individuals were infected; Bennett Churchill, Manitoba. Haematozoans were found in 1972) another site within the open forest/forest-tundra at least some individuals of all four passerinespecies zone in the boreal forest (Danks 1981). This is in sharp sampled at Churchill (Table 1). Leucocytozoon fringil- contrastto haematozoaprevalence recorded from Snow linarum was found in all four bird species, Trypano- (Chen caerulescens) and Canada Geese (Branta cana- soma avium in all but the longspur,and Haemoproteus densis) in the delta of the McConnell River, NWT macropigmentatus and H. chloris only in the two red- (60”50’N, 94”25’W), where only 3% (21/736) of the polls. Parasite intensities were low (5 10 infected cells geeseexamined were infected (Bennett and MacInnes per 100,000 examined) in all birds except seven in- 1972). The McConnell River delta is on the open tun- dividual redpolls that had up to 500 infected cells per dra about 200 km north of the treeline. This suggests 100,000 examined. that the scarcityof haematozoaat the McConnell River Parasite prevalence was highest (63-100%) in the is attributable to the absenceof vectors there. three speciesthat primarily occupy treed habitats and St.Paul Island, Alaska. None ofthe 14 Least Auklets nest mainly within the treeline: the White-crowned (Aethia pusilla) sampled harbored any haematozoa Sparrow (Zonotrichia leucophrys) and the two redpolls (Table 1). This is consistentwith the findina of Greiner (Carduelisflammea and C. hornemanni; Table 1). The et al. (1975) who recordedno haematozoain four other prevalenceof haematozoain thesespecies at Churchill speciesof alcids from the north temperate zone. We is similar to levels reported for other speciesfrom three can again attribute the absenceof haematozoa in the localities within the boreal forest zone: Lac Kohlmeis- Alcidae, in general, to the absenceof vectors in their ter, Quebec (60% of 129 birds of 29 species; Laird nesting habitat on windy, treelessoceanic islands or at 1961); Churchill Falls, Labrador (6 1O/a of 140 birds of sea. 2 1 species;Bennett 1972); and insular Newfoundland (48% of 209 birds of 3 tetraonid species;Bennett and CONCLUSIONS Inder 1972). Given the abundanceof potential vectors We conclude from this study that blood parasitesare in treed habitats at Churchill (pers. obser.), it is likely virtually absent in bird species nesting on the open that haematozoa were picked up on the breeding tundra. Although the evidencepresented here indicates grounds. that this pattern is due to the absenceof suitablevectors Smith’s Longspurs(Calcarius pictus), which nest on on the treelesstundra, this hypothesiswill need to be the ground in more open forest-tundra habitats (Jehl tested critically by actually sampling vectors (see Ben- 1968). had a significantly lower nrevalence of blood nett 1960 for suitable methods). It is possible, for ex- parasites( 16%) ihan each of the other three passerines ample, that tundra-nestingbirds are either avoided by sampled at Churchill (Table 1; G-tests with William’s suitable vectors or are more resistant to parasite in- correction, G,, = 9.0-15.9, P < 0.01, df = 1 in each fection than forest-dwelling species. This seems un- case).Since the redpolls and White-crowned Sparrows likely becausetwo speciesthat we studied (Rock Ptar- mainly nest in or among black spruce trees (Jehl and migan and Homed Lark) have haematozoain the boreal Smith 1970) this differencein parasiteprevalence sug- forest zone but not on the arctic tundra. Similarly, at geststhat even when vectors for a parasitesare present Churchill, there was a striking difference in parasite in the generalvicinity, microhabitatcharacteristics may prevalence across speciesbetween habitats often sep- be important in parasite transmission. Thus, the con- arated by only a few hundred meters. Thus, neither a stantly windy conditions in the more open tundra sites species-specificdifference in parasite resistancenor a inhabited by Smith’s Longspursprobably reducestheir difference in behavior due to climate appear to 292 SHORT COMMUNICATIONS be able to account for the scarcity of haematozoa in Kelly Fund of the Province of Quebec Society for the tundra-nesting birds. Protection of Birds (both to GS). Combined with previously published work, the re- sultsof our studvsuavest that birds nestingon the arctic LITERATURE CITED tundra and on smali&eanic islandstypically lack hae- BENNETT,G. F. 1960. On some omithophilic blood- matozoa. Among passerinebird species,in particular, suckingDiptera in Algonquin Park, Ontario, Can- only those populations nesting on the arctic tundra ada. Can. J. Zool. 38:377-389. have been found consistentlywithout blood parasites. BENNE~, G. F. 1970. Simple techniquesfor making Our resultsalso have implications for analysesof the avian blood smears. Can. J. ZooI.-48585-586. _ relation between blood parasitesand plumage bright- BENNETT.G. F. 1972. Blood narasitesof some birds ness in birds, For the past decade, there has been in- from Labrador. Can. J. Zool. 50:353-356. tense interest in the hypothesis (Hamilton and Zuk BENNETT,G. F., AND J. G. INDER. 1972. Blood par- 1982) that female birds might use male plumage color asites of game birds from insular Newfoundland. as an honest signal of parasite resistance. Hamilton Can. J. Zool. 50:705-706. and Zuk (1982) predicted, and found, a positive rela- BENNETT,G. F., AND C. D. MACINNES. 1972. Blood tion between blood parasite prevalence and plumage parasitesof geeseof the McConnell River, N. W. brightness across North American passerines.Subse- T. Can. J. Zool. 50:1-4. quent workershave found evidenceboth for and against DANKS,H. V. 1981. Arctic .Ent. Sot. Can., this prediction (see Moller 1990 for review). Our study Ottawa. indicatesthat tundra-nestingspecies should not be used GREINER,E. C., G. F. BENNETT,E. M. WHITE AND R. in any comparative tests of the Hamliton-Zuk hypoth- F. COOMBS. 1975. Distribution of the avian hae- esis becauseinter- and intraspecificvariation in plum- matozoa of North America. Can. J. Zool. 53:1762- age color in thesespecies cannot be related to variation in parasite prevalenceor intensity. Thus, plumagedif- HAMILTON,W. D., ANDM. ZUK. 1982. Heritable true ferences between male Horned Larks (dull), Lapland fitness and bright birds: a role for parasites?Sci- Longspurs(brighter) and Snow Buntings (brightest)as ence 218:384-387. well as intraspecific variation in the plumage of both JEHL,J. R. 1968. The breeding biology of Smith’s male and female Lapland Longspurs (Montgomerie, longspur.Wilson Bull. 80:123-149. unpublisheddata) cannot be explained by haematozoa JEHL, J. R., AND B. A. SMITH. 1970. Birds of the infestations. Churchill region, Manitoba. Manitoba Mus. Man Our analysesalso suggestthat not only the presence and Nature, Winnipeg. of vectors,but also their relative abundancein different LAIRD, M. 1961. A lack of avian and mammalian microhabitats (see Greiner et al. 1975) may influence haematozoa in the antarctic and Canadian arctic. parasite prevalence both within and among species. Can. J. Zool. 39:209-213. Vector abundance has so far not been taken into ac- LOYE,J. E., AND M. ZUK [eds.]. 1990. Ecology, be- count explicitly in tests of the Hamilton-Zuk hypoth- havior and evolution of bird-parasite interactions. esis and could potentially account for some of the ap- Oxford Univ. Press, Oxford, England. parent correlations with plumage brightness. MILLER, A. P. 1990. Parasitesand sexual selection: We thank Jim Briskie,Karen Holder, Ian Jones,Adele current statusof the Hamilton-Zuk hypothesis.J. Mullie and Len Shandruckfor providing blood smears Evol. Biol. 3:3 19-328. from their study species.This study was funded by the MONTGOMERIE,R. D., R. V. C&TAR, R. L. Mc- Canadian Wildlife Service (to Len Shandruckworking LAUGHLIN,AND B. LYON. 1983. Birds of Sarcpa at Fort Simpson) and grants from the Natural Sciences Lake, Melville Peninsula, Northwest Territories: and Eneineerine Research Council (to GFB and RM). Breedingphenologies, densities and biogeography. the Sci&tific Tiaining Grants Program of Indian and Arctic 36165-75. Northern Affairs Canada (to RM and GS), the National WEATHERHEAD,P. J., ANDG. F. BENNETT.199 1. Ecol- Geographic Society (to Ian Jones working on St. Paul ogy of red-winged blackbird by hae- Island), the Frank M. Chapman Fund and the A. B. matozoa. Can. J. Zool. 69:2352-2359.