Could the Blood Parasite Leucocytozoon Deter Mallard Range Expansion? Author(S): Dave Shutler, C

Home , Black fly

Could the Blood Parasite Leucocytozoon Deter Mallard Range Expansion? Author(s): Dave Shutler, C. Davison Ankney, Darrell G. Dennis Source: The Journal of Wildlife Management, Vol. 60, No. 3 (Jul., 1996), pp. 569-580 Published by: Allen Press Stable URL: http://www.jstor.org/stable/3802074 Accessed: 04/12/2008 06:50

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=acg.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].

Allen Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Wildlife Management.

http://www.jstor.org COULDTHE BLOODPARASITE LEUCOCYTOZOON DETER MALLARD RANGEEXPANSION?

DAVESHUTLER,1 Department of Zoology, Universityof WesternOntario, London, Ontario, N6A 5B7, Canada C. DAVISONANKNEY, Department of Zoology, Universityof WesternOntario, London, Ontario, N6A 5B7, Canada DARRELLG. DENNIS,Canadian Wildlife Service, 153 NewboldCourt, London, Ontario, N6E 1Z7, Canada

Abstract: We investigated whether the blood parasite Leucocytozoon simondi could slow mallard (Anas platyrhynchos)population growth in the east that has been associatedwith Americanblack duck (A. rubripes; hereafter black duck) population decline. Susceptibility to parasites was compared among F1 ducklings produced from crossesbetween mallard and black ducks from areas of Leucocytozoonendemicity (Ontario), and between mallards from an area free of Leucocytozoon (Saskatchewan). We produced 6 "types" of ducklings: Ontario black duck x Ontario black duck (OB x OB), Ontario black duck x Ontario mallard (OB x OM), Ontario black duck x Saskatchewan mallard (OB x SM), OM x OM, OM x SM, and SM x SM. We predicted that because of probable coevolution of black ducks and Leucocytozoon, black duck ducklings would have resistance to the parasite. We also predicted that Ontario genes would confer some resistance to ducklings because these ducklings' parents had survived exposure to Leucocytozoon. In contrast, we predicted that mallard and Saskatchewan genes would not confer resistance, i.e., OB x OB ducklings would have greatest resistance to Leucocytozoon, SM x SM ducklings would have least, and remaining duckling types would have intermediate resistance. Of 169 ducklings exposed in 2 years in 3 geographically separate locales, none died, showed noticeable symptoms, or otherwise behaved abnormally. Nonetheless, weekly blood smears indicated that 91% of ducklings became infected, and many developed intense parasitemias. However, infection intensities were not different among the 6 duckling types. In addition, hematocrits were not lowered by intense infections. These results suggest that the effects of Leucocytozoon on wild waterfowl populations have been overestimated, and that Leucocytozoon will not prevent further range expansionof mallards. J. WILDL.MANAGE. 60(3):569-580

Key words: American black ducks, Anas platyrhynchos, Anas rubripes, blood parasites,hybrids, Leuco- cytozoon simondi, mallard, Ontario, Saskatchewan.

Mallards appear to be displacing American Leucocytozoon has the potential to prevent or black ducks in eastern North America through slow mallard encroachment into those parts of competition and/or introgressive hybridization the black duck range in which Leucocytozoon (Rogers and Patterson 1984, Brodsky and occurs. Weatherhead 1984, Ankney et al. 1987, Mer- Leucocytozoon is transmitted by black flies endino et al. 1993). Thus, continued mallard (Simuliidae) (Shewell 1955) and can infect 100% range expansion could lead to local or complete of mallards and black ducks in some populations extinction of black ducks. However, the blood in northeastern North America (O'Roke 1934, parasite Leucocytozoon simondi may be a bar- Nelson and Gashwiler 1941, Chernin 1952, rier to this outcome (Khan and Fallis 1968, Den- Trainer et al. 1962, Bennett et al. 1974, 1975, nis 1974). Leucocytozoon is endemic in much 1991). In contrast, Leucocytozoon is rarely re- of the historic range of black ducks, and the ported from central North America (Savage and parasite reportedly causes substantial duckling Isa 1955, Burgess 1957). Differences in Leuco- mortality (O'Roke 1934, Fallis et al. 1956, 1974; cytozoon prevalence arise because, in the latter Barrow et al. 1968) that is higher among mallard area, black fly vectors have less swift water than black duck ducklings (Khan and Fallis breeding habitat and are often insufficiently nu- 1968). In this paper, we test whether differential merous to spread the parasite (Herman 1968). susceptibility of mallards and black ducks to Historically, mallards occupied western and central North America, whereas black ducks oc- the eastern of the continent Present address:Canadian Wildlife Service, 115 cupied part (Bellrose Perimeter Road, Saskatoon,Saskatchewan, S7N 0X4, 1976). This historical distribution pattern is con- Canada. sistent with current distributions of Leucocy-

569 570 LEUCOCYTOZOONAND DUCKS * Shutler et al. J. Wildl. Manage. 60(3):1996 tozoon (Khan and Fallis 1968). Because mal- his crew provided guidance in the ways of Leu- lards have not coevolved with Leucocytozoon cocytozoon; the staff of Canadian Wildlife Ser- but are closely related to black ducks genetically vice in London provided logistical assistance (J. (Ankney et al. 1986, Avise et al. 1990), they Robinson, N. North, and J. Sullivan); B. Clark should be more susceptible to the parasite (Ewald of Canadian Wildlife Service in Saskatchewan 1983, 1993; Van Riper et al. 1986, Black 1992, sent us ducklings; A. Toline provided logistical Bennett et al. 1993). assistance in Algonquin Park, and the Ontario If a Leucocytozoon-mediated barrier to mal- Ministry of Natural Resources provided logis- lards still exists, however, it is necessary to ex- tical assistance at Swan Lake (in particular, H. plain continuing mallard range expansion into Anderson, J. Rice, D. Strickland, and M. Wil- areas where the parasite is endemic (Johnsgard ton). Drs. A. M. Fallis and R. A. Khan provided and DiSilvestro 1976, Rusch et al. 1989, Mer- historical details of earlier studies, and com- endino and Ankney 1994). One possibility is that mented on our conclusions. Funding was pro- mallards have acquired genes for Leucocyto- vided by the Natural Sciences and Engineering zoon resistance, either from introgressive hy- Research Council of Canada, the Black Duck bridization with resistant black ducks, or via Joint Venture, the Ontario Federation of An- natural selection that has culled non-resistant glers and Hunters, and the University of West- mallards (collectively, the resistance hypothe- ern Ontario. sis). A second possibility is that mallards are able to survive in habitats within the boreal only METHODS forest where Leucocytozoon is not present (e.g., areas without fast running water [the suscepti- Initial Preparations bility hypothesis]). The null hypothesis is that a In April 1991, we advertised rewards to in- Leucocytozoon-mediated barrier does not and dividuals finding active wild mallard and black did not exist, and that no differences in suscep- duck nests in an area of Leucocytozoon endem- tibility occur between duck species. icity in Ontario bounded approximately by Par- Our objective was to determine whether ry Sound in the west (80?0'W), North Bay in the duckling mortality could limit mallard range north (46?30'N), Ottawa in the east (75?30'W), expansion into areas of Leucocytozoon endem- and Haliburton in the south (45?0'N). Eggs we icity. We tested whether greater parasite resis- collected from these nests were transported to tance occurred among ducklings with a greater Lake St. Clair, an area where vectors of Leu- proportion of genes from presumably coevolved cocytozoon are not found. In addition, we re- black ducks, or from parents recently exposed ceived 40 Saskatchewan mallard ducklings (Can. to Leucocytozoon; i.e., resistance of black ducks Wildl. Serv., Saskatchewan). Eggs were hatched > mallards, and resistance of Ontario birds > in incubators, and we gave each duckling in- resistance of Saskatchewan birds. Our protocol dividually numbered web tags. Ducklings were was to expose ducklings to black fly vectors of sexed by cloacal inspection (Schemnitz 1980) Leucocytozoon in seminatural conditions, and within a day after hatching, and those that sur- then observe ducklings for differences in mor- vived to 4 weeks were given individually num- tality or signs of illness. We first ensured that bered leg bands. Because we needed hybrids for sex, initial date of exposure, and site of exposure our experiments on Leucocytozoon, we used did not obscure infection patterns among duck- these ducks for captive breeding rather than ling species and hybrids (collectively, "types"). experiments. In 1992, we had 107 yearlings to We thank J. Haggeman for help at Lake St. breed for producing ducklings. We prepared Clair including taking care of ducks during our yearlings for breeding by putting them in out- absences. A. Mullie, R. Bon, and G. Alderson door group pens in October so that they would provided field assistance, and colleague D. Hoy- form natural pair bonds (opposite-sex siblings sak shared duties associated with his and our were kept in separate pens). Up to 20 ducks concurrent studies. Team Cage Henge (J. Kom- were assigned to 1 of 6 group pens so that their iniek, P. Rutherford, S. Seiler, A. Mullie, D. pairings would produce ducklings hypotheti- Gagnier, B. and P. Wisenden, D. Hamilton, and cally exhibiting the full range of Leucocytozoon R. Haggeman) helped build duck enclosures; P. resistance (Table 1). In early April, pairs were Schleifenbaum and family accommodated ducks isolated in 2.4- x 1.5- x 0.9-m (8- x 5- x 3-ft) in 1992; S. Desser loaned equipment and with breeding pens with a nest box, ad libitum corn, J. Wildl. Manage. 60(3):1996 LEUCOCYTOZOONAND DUCKS * Shutler et al. 571

Table 1. Arrangementof ducks in grouppens in 1992 at the black ducks, and mallards are relatively rare (K. Lake St. ClairNational Wildlife Area. A similararrangement was used in 1993. Ross, Can. Wildl. Serv., pers. commun.). We numbered duckling "groups" by week of hatch Group (i.e., group I hatched in the first week in Jun, Pen Ma Fa size group II in the second week in Jun, etc.). 1 10 OM (4) 10 OM (4) 20 Experimental ducklings were housed in the 2 9 OM 4 SM 5 OB 18 (7) (4), (2) same cages we had used for breeding. Cages 3 7 OM 7 OM 14 (7) (7) had 2.5- x 2.5-cm mesh hardware cloth that 4 9 SM (6) 4 SM (3), 5 OB (3) 18 5 9 OB (4) 10 OM (4) 19 allowed black flies free access to ducklings. We 6 9 OB (3) 4 SM (4), 5 OB (2) 18 put cages on lake shores where black fly vectors Total 53 (16) 54 (15) 107 are most abundant (Bennett 1960). To substitute

a OM denotes Ontario mallard; OB denotes Ontario black duck; SM for the warmth normally provided by brooding denotes Saskatchewan mallard. No. of different clutches that account hens, we supplied ducklings with a heat lamp for each group is given in parentheses. for the first 7 days that they were outdoors. Ducklings were maintained on an ad libitum commercial duck food, and grit. Birds that had diet, and were provided with a nest box for not formed pairs were assigned mates and sim- shelter. Initially, we kept up to 15 ducklings per ilarly isolated. This procedure was repeated in cage. As ducklings became larger, the maximum 1993 (birds that bred in 1992 were paired with was reduced to 8. Experimental ducklings were different mates), except that we also paired some exposed continuously until the end of July. yearlings produced in 1992. Of the first 6 ducklings from a clutch, 1 of We had unequal numbers of adult types to each sex was assigned to 1 of 2 experimental pair at the outset (Table 1). Furthermore, breed- locales (see below) and a control group. For ing success was uneven (Table 2). Hence, we clutches producing fewer than 3 ducklings of 1 had unequal numbers of each type of duckling sex, we allocated individuals to experimental for our experiments. locales before allocating individuals to the control group. Controls were kept in outdoor cages Methods for Testing Leucocytozoon at Lake St. Clair. For larger clutches, only 2 Susceptibility. ducklings were retained as controls; the re- To mimic conditions that wild ducklings mainder was divided by sex and randomly as- would experience in contracting Leucocyto- signed to our experimental locales. This protocol zoon, we exposed them to vectors as early as avoided results from separate locales that may possible in their development, and kept them have been biased because of differences in sex exposed continuously for several weeks. Duck- ratio or genetic origin of ducklings. lings hatched asynchronously. At each of 5 con- Because black fly population density and Leu- secutive weekly intervals, when they were be- cocytozoon population virulence vary geo- tween 2 and 9 days old, ducklings were trans- graphically (Herman et al. 1975, Desser and ported from Lake St. Clair to our study areas Ryckman 1976, Desser et al. 1978), results from in and around Algonquin Park, Ontario, Can- 1 locale may not be applicable to other locales. ada. This area continues to be a stronghold for Hence, in each year, we divided experimental

Table2. Breedingsuccess of pairtypes, and numberof each type of ducklingexposed or kept as controls.

1992 1993 % of pairs Ducklin s % of pairs Ducklings producing viable expose3 producing viable exposed x M Fa Pairs isolated ducklings (controls) Pairs isolated ducklings (controls) OB x OB 5 40 8 (2) 7 57 12 (3) OB x OM, OM x OB 9 11 6 (1) 8 25 5 (0) OB x SM, SM x OB 7 0 0 (0) 7 57 6 (2) OM x OM 10 80 52 (7) 8 63 33 (8) OM x SM, SM x OM 10 50 17 (2) 7 43 21 (5) SM SM 3 33 7 (1) 4 50 2 (2) Totals 44 39 90 (13) 41 49 79 (20)

a OM denotes Ontario mallard; OB denotes Ontario black duck; SM denotes Saskatchewan mallard. 572 LEUCOCYTOZOONAND DUCKS * Shutler et al. J. Wildl. Manage. 60(3):1996 ducklings between locales that were more than gametocytes occurs in ducklings' blood streams 20 km apart (Lake Sasajewan and Kennisis Lake at about 10-24 days post-infection, and this is in 1992, Lake Sasajewan and Bena Lake in 1993). also when duckling mortality peaks (Desser There was no electricity available for heat lamps 1967). Thus, the most biologically meaningful at Bena Lake, so we kept ducklings outdoors at comparison is among ducklings that have been Lake Sasajewan for 7 days before transporting infected for these many days. However, nega- them to Bena Lake. Because wild hen mallards tive blood smears do not necessarily imply that move their broods between lakes, this manip- a duckling has not been infected, because low ulation mimicked wild ducklings' experience. intensity parasitemias may not be detected in Ducklings were visited at least every 2 days individual blood smears. Thus, we could never to be fed and checked for signs of parasitism. be certain of the initial date on which a duckling After a duckling is infected with Leucocytozoon had become infected. The best we could do was sporozoites, 6 to 10 days elapse before mature to compare infection intensities of ducklings rel- gametocytes of the parasite appear in the blood ative to the interval since they had been ex- stream (Desser 1967). To monitor infection rates, posed. However, black fly populations fluctuate beginning at 10 days post-exposure, we took within and among years so that transmission weekly blood smears from each experimental dynamics vary according to Julian date on which duckling's leg vein (Bennett 1970). A subsample a duckling hatches (and consequently when we of controls was also periodically checked for added it to our experiments). Thus, we also com- infection. Twenty-five microscope fields (rough- pared course of infection relative to Julian date ly 6,000 red blood cells) of smears were exam- at which ducklings were exposed. ined under a microscope using a magnification Genuine Leucocytozoon-negative smears may of 400, and all mature Leucocytozoon game- arise for 2 main reasons. First, ducklings may tocytes were counted. Infection intensities are have a sterile immunity to the parasite; i.e., the expressed as number of parasites per 1,000 red duckling's immune system kills all parasites be- blood cells. Because density of parasites subsides fore they develop into gametocytes. Second, just at night (Roller and Desser 1973), smears were by chance, a duckling may not have been bitten made during the day. Nonetheless, in 1993 we by, or received any Leucocytozoon from, in- recorded the exact time that samples were taken fected black flies. In the first case, differences to confirm that time was not confounding our in sterile immunity are biologically meaningful results. and negatives should be included in compari- In 1992, we sampled ducklings until 8 August. sons. In the second case, negatives may add sig- Hence, ducklings in group I were sampled more nificant but biologically meaningless variation times than subsequent groups. A subsample of to results. Because we could not be certain surviving ducks from 1992 was also sampled in whether causes of negatives were biologically spring 1993 for evidence of relapse, the surge meaningful, we analyzed data twice, once with of parasites in blood that serves to reinfect vector and once without negatives. populations each year (Chernin 1952). Results Infection intensity data that included nega- from 1992 suggested that by the sixth weekly tives were not normally distributed (Fig. 1, top, blood smear, infection intensities were minimal Kolmogorov-Smirnov tests, d's > 0.30, P's < in more that 95% of cases. Hence, each duckling 0.0001). Box-Cox tests (Krebs 1989) indicated was sampled only 6 times in 1993. that transformations would not improve fit to Intensity of Leucocytozoon infection has been normality. Hence, except where sample sizes correlated with degree of anemia in domestic were large, we used non-parametric tests when ducks (Kocan and Clark 1966, Maley and Desser analyzing these data. If negatives were exclud- 1977). To measure this symptom, in 1993 we ed, data could be log-transformed to produce took a microcapillary of blood from each duck- distributions that were closer to normality (Fig. ling at the same time as we took blood smears. 1 bottom, Kolmogorov-Smirnov tests, d's < 0.18, Capillaries were centrifuged and percent packed P's still <0.0001). We assume that parametric red blood cell volume (hematocrit) of each sam- tests were sufficiently robust to deal with re- ple was calculated. Hematocrits of group III, maining departures from normality; at any rate, IV, and V ducklings were not sampled 6 times non-parametric tests gave qualitatively similar because of logistic problems. results. Maximum density of mature Leucocytozoon We judged that a repeated measures analysis J. Wildl. Manage. 60(3):1996 LEUCOCYTOZOONAND DUCKS * Shutler et al. 573 of variance (ANOVA) should not be used on infection data that included negatives because 300- the data were not normally distributed. How- ever, if we excluded negatives, the resulting re- duction in sample sizes prevented statistically . 200- meaningful comparisons. Hence, we included C) negatives but used non-parametric repeated I- measure ANOVA (Friedman tests). Because non- 100- parametric tests are less powerful than parametric tests, we also present separate parametric 1- - i II - results from each interval to be certain 0- *1 I" l sampling <0.01 6 14 22 30 >36 that no differences were inadvertently over- Parasites/ 1,000 red blood cells looked. To derive appropriate statistical cutoff * 1992 probabilities for these separate tests, we used a sequential Bonferroni cutoff criterion wherein 0 1993 the standard 0.05 cutoff probability is divided by the number of tests (Daniel 1978, Holm 1979, 125 Rice 1989). Most statistical analyses were performed on 100- a Macintosh version of SYSTAT (1992). Power calculations for correlations were done by hand u 75 using formulae in Zar (1984). Sample sizes that we report vary because of breakage of hemat- 5050- ocrit tubes during centrifugation, subsampling, .1-L predator-caused mortality (about 5%), and other reasons given above.

RESULTS 0.1 0.5 0.9 1.3 1.7 >2.1 1. Mortality, Infection Intensity, and He- Log(parasites / 1,000 red blood cells) matocrit.-Eighty-three percent of ducklings in 1992 and 100% of ducklings in 1993 tested Fig. 1. Top. Distributionof infectionintensities (all data com- in 1992 and 1993 Bottom. positive at least once for How- bined) (solidbars) (open bars). Log- Leucocytozoon. transformeddata with negatives excluded. Except for cate- ever, no ducklings died as a result of Leucocy- gories with "<" or ">", values on loweraxes are midpoints. tozoon infection (n = 169). Furthermore, we observed no symptoms that could be ascribed to Leucocytozoon in any duckling. In fact, we within the range of infection intensities reported had only 1 duckling that became visibly sick elsewhere (e.g., Desser 1967, Khan and Fallis and died. Although we could not determine its 1968). cause of death, we found no Leucocytozoon in Intensity assessments were highly repeatable this duckling 4 days before or on the day it died. based on a sample of smears that were known No mortality from disease occurred among the to have parasites (n = 98, r, = 0.90, P < 0.001). control population either (n = 32), and we found We also tested whether an individual blood sam- no Leucocytozoon in 28 smears from controls ple reliably reflected parasite loads. Paired (some individuals were sampled as many as 3 smears taken from the same individual within times, some were not sampled). Because we had an hour had similar parasite densities (r, = 0.64, no Leucocytozoon-caused mortality, the re- n = 10, P = 0.03). Hematocrit samples that were mainder of our results focus on infection inten- similarly paired were also significantly corre- sity and hematocrit. lated (r, = 0.59, n = 10, P = 0.04). Finally, in Infection intensities in 972 blood smears from 1993, we tested whether our results could have experimental ducklings ranged from 0 to 67 been biased by time of day when samples were parasites/1,000 blood cells (mean ? SD = 3.15 taken. We found no relation between time of + 6.73) (Fig. 1 top). Taking into account that day and infection intensity (all 1993 data com- these data include ducklings sampled after in- bined, r = -0.01, n = 444, P = 0.78, power = fection intensities had subsided, our mean is 0.96) or time of day and hematocrit (all 1993 574 LEUCOCYTOZOONAND DUCKS * Shutler et al. J. Wildl. Manage. 60(3):1996

12 I .I . . tensity infections at each sampling interval 1992 whether or not negatives were included. Fur- thermore, the proportion of each sex that was 10- N=25 infected in each sampling interval was the same 8- (Chi-square tests, all P's > 0.10). Friedman tests on data from 1992 (F = 0.71, n = 44, P = 0.40) = = also 6- and 1993 (F 0.05, n 69, P = 0.82) did not reveal differences between the sexes in (A any -4 infection intensities. In hematocrits of fe- 4- 1993, 0 males and males were similar at all sampling 0 intervals, and a Friedman test also detected no 2- differences (n = 35 for test up to and including 0 45 days post-exposure, F = 0.92, n = 61, P = I! 0 0.34; for test up to and including 31 days post- o0 exposure, F = 0.44, P = 0.51). r-1 3. Site at Which Ducklings Were Ex- V3 posed.-In 1992, duckling parasitemias were u lower and fewer ducklings became infected at Kennisis Lake than at Lake Sasajewan (Table 3). These differences persisted for the first 6 sampling intervals, and reappeared during relapse at 320 days post-exposure. When we excluded negatives, however, parasitemias were similar among locales (Table 3). When negatives are included, a Friedman test on data up to and including 45 days post-exposure revealed that differences were significant overall (F = 108, n = 44, P < 0.001). 10 17 24 31 38 45 In 1993, ducklings had higher intensity in- Days post exposure fections at Bena Lake than at Lake Sasajewan 10 days post-exposure (Table 3) whether or not 2. Infectionintensities at each intervalin 1992 Fig. sampling were included. After no (top) and 1993 (bottom).Upper data points in each graphdo negatives 10 days, sig- not includenegatives whereas lowerdata pointsdo. One stan- nificant differences emerged. At no sampling darderror is shown for each pointand sample sizes are given interval did the of infected above or below associated errorbars. proportion ducklings differ between study sites (Chi-square tests, all P's > 0.10). A Friedman test indicated that in- data, r = 0.03, n = 384, P = 0.51, power = 0.91). fection intensities did not differ overall in 1993 Furthermore, no non-linear patterns were visi- between sites (F = 0.25, n = 69, P = 0.62). ble in plotted data. Hematocrits were the same between sites As reported elsewhere (Desser 1967), para- based on the Bonferroni cutoff probability of sitemias reached a peak within the first 10-17 0.008. A Friedman test confirmed that there days (Fig. 2) and then declined to barely de- were no differences in hematocrits of ducklings tectable levels (also see Tables 3 and 4 after 31 according to the site at which they were exposed days post-exposure). If negatives are included, (F = 0.01, n = 35, P = 0.91). parasitemias peaked later in 1992 than in 1993 4. Date of Exposure.-Comparison of prog- (Fig. 2). Otherwise, patterns of parasitemia were ress of parasitemias among duckling groups in similar in each year. 1992 revealed different infection intensities at Hematocrits ranged from 0.12 to 0.49. How- 38 days post-exposure; higher infection inten- ever, more intense Leucocytozoon infections sities of ducklings in group II accounted for this were not associated with lower hematocrits (all result. This result persisted when negatives were 1993 data, r = 0.05, n = 389, power = 0.84, P not included; ducklings in group II had more = 0.33). An outlier individual whose hematocrit intense infections than their counterparts from was 0.12 survived (the next lowest hematocrit groups I, III, and IV (Tukey's multiple com- recorded was 0.25). parison tests, P's < 0.05). In the remaining post- 2. Sex.-Females and males had similar in- exposure intervals, infection intensities were J. Wildl. Manage. 60(3):1996 LEUCOCYTOZOONAND DUCKS * Shutler et al. 575

Table3. Infectionintensities (parasites/1,000 blood cells) by exposure site for each samplinginterval in 1992-93.

Days Including negatives Not including negatives post- Alternate lakea Lake Sasajewan Alternate lakea Lake Sasajewan exposure f SD n SD n pb ; SD n SD n pb 1992 10 0.4 2.0 43 5.1 9.7 37 <0.001 9.3 2.6 2 8.2 11.3 23 0.38 17 0.9 2.6 46 6.0 6.8 41 <0.001 3.5 4.3 12 6.5 6.9 38 0.07 24 2.2 7.5 40 8.1 6.1 40 <0.001 5.3 11.0 17 6.5 8.1 38 0.20 31 1.6 3.7 35 4.9 6.9 32 <0.001 2.9 4.7 19 4.9 6.9 32 0.08 38 1.4 2.8 32 5.6 12.1 30 <0.001 2.5 3.4 18 5.6 12.1 30 0.15 45 0.9 2.4 24 1.6 1.5 27 <0.001 1.9 3.3 12 1.7 1.5 26 0.42 52 1.0 2.4 14 1.1 0.8 17 0.02 2.4 3.3 6 1.2 0.8 16 0.41 59 0.6 0.8 8 0.7 0.9 9 0.55 1.1 0.8 4 0.9 0.9 7 0.61 320 0.1 0.1 24 0.2 0.2 24 <0.001 0.2 0.1 5 0.3 0.2 16 0.36 1993 10 11.4 14.1 35 4.9 10.0 41 <0.001 12.5 14.3 32 6.1 10.9 33 <0.001 17 4.8 5.9 35 6.4 4.8 42 0.62 4.8 5.9 35 6.4 12.3 42 0.53 24 2.1 3.9 35 3.4 4.7 43 0.02 2.3 4.0 33 3.4 4.7 43 0.07 31 2.0 2.1 35 2.0 2.5 41 0.70 2.0 2.1 34 2.0 2.5 40 0.77 38 3.0 4.1 35 2.5 3.9 37 0.51 3.2 4.1 33 2.6 4.0 35 0.48 45 0.7 0.7 35 1.6 2.1 35 0.02 0.9 0.7 28 1.8 2.1 32 0.04

a Kennisis Lake in 1992, Bena Lake in 1993. b For comparisons that include negatives, probabilities are based on Mann-Whitney U-tests on untransformed data. For comparisons that do not include negatives, probabilities are based on t-tests on log-transformed data. Significant Bonferroni cutoff probability is 0.006 for 1992, and 0.008 for 1993. similar regardless of when a duckling was ex- In 1993, differences among groups in infec- posed. Furthermore, no differences among tion intensities occurred at 24, 38, and 45 days groups persisted when ducklings were sampled post-exposure. When we excluded negatives, the following spring at 320 days post-exposure. significant differences were detected only at 24 A Friedman test also suggested that date of ex- days post-exposure, when groups IV and V had posure had no significant effect on average in- higher infection intensities than groups I, II, and fection intensities from the entire season (F = III (Tukey tests, P's < 0.05). At 38 days post 1.15, n = 53, P = 0.35). exposure, group II had higher infection inten-

Table4. Infectionintensities (parasites/1,000 blood cells) by ducklingtype at each samplinginterval in 1992-93.

Duckling type a pb when Days OB x OB OB x OM OB x SM OM x OM OM x SM SM x SM negatives post- expo- Not sure f nl n2 f nl n2 * nl n2 i nl n2 x nl n2 i nl n2 Incl. incl. 1992 10 3.0 6 3 7.1 62 1.8 45 9 2.0 14 7 4.6 7 4 0.18 0.84 17 5.2 7 4 0.9 64 4.0 51 31 1.7 16 7 1.7 7 4 0.66 0.60 24 1.0 6 3 6.3 65 4.4 44 33 5.5 17 10 0.8 7 4 0.34 0.44 31 1.0 6 4 1.6 54 3.9 37 29 3.6 14 11 0.8 5 3 0.69 0.74 38 0.8 4 3 1.0 64 2.5 36 28 1.3 9 26 14.3 7 6 0.10 <0.001 45 1.9 4 4 1.5 64 1.0 29 19 1.4 6 6 2.2 6 5 0.20 0.52 52 0.0 1 0 0.6 60 1.3 20 15 0.8 4 3 0.40 0.15 320 0.1 5 2 0.4 20 0.1 27 10 0.2 7 4 0.2 7 4 0.23 0.72 1993 10 7.3 12 11 0.8 5 5 7.9 5 4 6.0 32 23 11.2 20 20 26.2 2 2 <0.001 0.03 17 6.2 12 12 2.2 5 5 2.6 5 5 8.1 32 32 3.2 21 21 7.1 2 2 0.10 0.18 24 2.4 12 12 1.8 5 5 1.8 5 5 4.2 33 32 1.5 21 20 3.3 2 2 0.31 0.28 31 1.5 12 12 0.6 5 4 1.0 5 5 2.8 31 30 1.7 21 21 1.5 2 2 0.19 0.16 38 1.2 12 11 2.1 5 4 4.3 5 4 2.9 28 27 3.4 21 21 0.7 1 1 0.04 0.25 45 0.8 12 10 0.8 5 4 0.6 5 4 1.1 26 23 1.7 21 18 1.8 1 1 0.53 0.38

a No OB x SM ducklings were produced in 1992. Negatives are included in means. nl denotes sample sizes including negatives, n2 denotes sample sizes without negatives. b P values are based on Kruskal-Wallis tests when negatives are included, and ANOVAs when negatives are not included. Significant Bonferroni cutoff probability is 0.006 for 1992, and 0.008 for 1993. Power for tests that do not include negatives ranged from 0.31-0.52 for cases where Ps > 0.05. 576 LEUCOCYTOZOONAND DUCKS * Shutler et al. J. Wildl. Manage. 60(3):1996

Table5. Hematocritsby ducklingtype at each samplinginterval in 1993.

Duckling typea b when pb when Days OB x OB OB x OM OB x SM OM x OM OM x SM SM x SM negatives post expo- Not sure i nl n2 i n2 i nl n2 X nl n2 i nl n2 i nl n2 Incl. incl. 10 0.34 12 11 0.34 3 3 0.35 5 5 0.35 32 23 0.33 20 20 0.34 1 1 0.29 0.25 17 0.38 12 12 0.35 5 5 0.41 5 5 0.38 31 31 0.39 21 21 0.36 2 2 0.16 0.16 24 0.38 11 11 0.38 4 4 0.41 5 5 0.40 27 26 0.38 20 19 0.39 1 1 0.12 0.14 31 0.38 12 12 0.36 5 4 0.37 5 5 0.39 28 27 0.37 19 19 0.31 1 1 0.21 0.32 38 0.38 9 8 0.36 5 4 0.37 4 3 0.38 21 20 0.36 19 19 0.29 1 0 0.33 0.26 45 0.39 3 3 0.36 4 3 0.39 3 3 0.40 14 12 0.37 19 18 0.66 0.71

a Means include negatives. nl denotes samples when negatives are included, n2 denotes samples when negatives are excluded. b P values are based on ANOVAs. Types for which n = 1 were not included in tests. Significant Bonferroni cutoff probability is 0.008. Power ranged from 0.16-0.64. sities than group III (Tukey test, P = 0.02). How- no significant differences were detected (Tukey ever, mean infection intensities were the same tests on data excluding negatives, all P's > 0.07). among groups in 1993 (Friedman test excluding A Friedman test also indicated a relation be- group V because of insufficient data, F = 2.12, tween duckling type and Leucocytozoon sus- n = 69, P = 0.11). ceptibility (pure Saskatchewan mallards exclud- Hematocrits were similar among groups ex- ed because of insufficient data, F = 3.19, n = cept at 17 days post-exposure, when group II 68, P = 0.02). Although these results suggested had higher hematocrits than groups I and III, some support for the susceptibility hypothesis, and group IV had higher hematocrits than group the data (Table 4) indicate no clear relation III (Tukey tests, P's < 0.05) whether or not between predicted susceptibility and infection negatives were included. We restricted our re- intensity. Because of small sample sizes for some peated measures test to the first 4 post-exposure types, we repeated a Friedman test on pure intervals to include all groups, and found no Ontario black duck, pure Ontario mallard, and differences in hematocrits (Friedman test ex- Ontario x Saskatchewan mallard ducklings, but cluding group V, F = 1.66, n = 61, P = 0.19). again found no significant differences (up to and 5. Duckling Type.-Our principal objective including 45 days post-exposure, F = 0.81, n = was to measure differences in Leucocytozoon 58, P = 0.45). susceptibility among different duckling types. Hematocrits did not differ among duckling In 1992, no significant differences were ob- types at any post-exposure interval in 1993, served in infection intensities except at 38 days whether or not negatives were included (Table post-exposure if negatives were excluded (Table 5). Despite this, hematocrits differed overall 4); in this single instance, pure Saskatchewan among duckling types (Friedman test excluding mallard ducklings had the highest intensity in- Saskatchewan mallards, F = 4.15, n = 35, P = fections of duckling types. Differences in infec- 0.01). If we included only the 3 duckling types tion intensities among duckling types were not used in the infection intensity analysis above, significant overall (up to and including 45 days differences among types were not as extreme post-exposure, Friedman test, F = 0.23, n = 44, (Friedman test up to 38 days post-exposure, F P = 0.92). Because of small sample sizes for some = 2.62, n = 43, P = 0.09). types, we repeated these tests using only pure DISCUSSION Ontario mallard ducklings and Ontario x Sas- katchewan mallard ducklings, but still found no Substantial mortality from Leucocytozoon significant differences (Friedman test up to and frequently has been reported among waterfowl including 45 days post-exposure, F = 0.10, n = (see Desser and Bennett 1993 for a review). 31, P = 0.76). However, many of these reports are based on In 1993, infection intensities differed among mortality in domestic stocks of mallards (e.g., duckling types at 10 days post-exposure, but this Chernin 1952, Fallis et al. 1956). Many other difference was not significant if negatives were reports are based on experimental injection of excluded (Table 4). Pure Saskatchewan mallards Leucocytozoon into ducks (e.g., Anderson et al. and Ontario mallard x Saskatchewan mallard 1962, Desser 1967). Hence, results from these ducklings had higher intensity infections, but studies should not be extrapolated to wild pop- J. Wildl. Manage. 60(3):1996 LEUCOCYTOZOONAND DUCKS * Shutler et al. 577 ulations naturally exposed to Leucocytozoon. In wild populations (Bennett et al. 1993). Further- fact, although many species of birds can carry more, number of broods used in their study was Leucocytozoon (Desser and Bennett 1993, Ben- not known (Fallis and Khan, pers. commun.), nett et al. 1993), only rarely are there reports so it is possible that their results were confound- of mortality from Leucocytozoon in wild birds ed by a few highly related individuals. If either (Khan and Fallis 1968, Bennett et al. 1993). It of these was the case, the Khan and Fallis (1968) could be argued that ad libitum diets enhanced results might be ascribed to susceptible genetic the survival of our ducklings. However, domes- stocks. A third possibility is that hybridization tic ducks on ad libitum diets suffer substantial events between mallards and black ducks in the mortality (references above). Hence diet is not last 30 years have integrated genes for Leuco- a complete explanation for the absence of mor- cytozoon resistance into the mallard gene pool, tality we observed relative to the 31% mortality including Saskatchewan. This homogenization Khan and Fallis observed in their wild duck- of the mallard gene pool may have occurred lings. In fact, most observations of parasites in because of substantial movements of males from wild birds suggest that absence of mortality is their natal areas to other breeding areas. Indeed, to be expected (Bennett et al. 1993). modern black duck populations appear geneti- That any mortality of wild ducks was ob- cally indistinguishable from mallards (Ankney served in the Khan and Fallis study is thus sur- et al. 1986, Avise et al. 1990). Further investi- prising, and requires some explanation. One gation would be necessary to distinguish among possibility is that rearing conditions for their these possibilities. more ducklings were different from ours; for example Even though we observed no mortality, Khan and Fallis (1968) did not use brood lamps subtle differences in susceptibility to Leucocy- when keeping ducklings outdoors (A. Murray tozoon could possibly affect persistence of mal- Fallis, Caledon, Ont., and Russ Khan, Dep. Biol., lard populations in northeastern North America. Memorial Univ. Newf., pers. commun. 1995). For example, more intense infections could slow Another possibility is that Leucocytozoon pop- growth rates, weaken individuals so that they ulations were more virulent in the years of their were more susceptible to predators, decrease study. Parasite virulence can increase substan- their ability to invest in secondary sex characters and tially when host density is sufficient to maintain that are used to attract mates (Hamilton transmission rates (Herre 1993). At the time of Zuk 1982), etc. Hence, we also compared symp- the Khan and Fallis study, research on Leuco- toms of infection among ducklings, including of cytozoon had been ongoing at Lake Sasajewan hematocrits and infection intensity. Intensity for 20 years, and captive ducks had been main- Leucocytozoon infection has been correlated tained at much higher densities than occurs in with degree of anemia in domestic ducks (Kocan this area in the wild. Thus, it is possible that and Clark 1966, Maley and Desser 1977). How- Leucocytozoon had been selected for increased ever, we found no relation between infection virulence during this interval. The explanation intensity and hematocrit, despite the fact that for the differences in mortality between this and our ducklings had infection intensities similar in the previous study is uncertain. to those observed in domestic ducklings pre- Another aspect of Khan and Fallis (1968) re- vious studies. Although we did detect differ- port is more significant. They reported that 40% ences among duckling types in hematocrit, be- of 25 mallard and 18% of 17 black duck duck- cause hematocrit was not affected by Leuco- lings died from Leucocytozoon. However, this cytozoon, this result is of no importance to our difference, although statistically significant, is experiments. It would appear that wild popu- based on a total of 42 ducklings. There are nu- lations of ducks are able to maintain normal merous possible explanations for this difference blood chemistry even when 7% of their eryth- other than differential susceptibility of mallards rocytes are occupied by Leucocytozoon. The and black ducks. For instance, the differences ability to maintain normal hematocrit also sug- they observed may have been the result of a gests that Leucocytozoon is more benign to wild type I statistical error. Second, their ducklings ducks than has been suggested by research based came from game farms, and the number of gen- primarily on domestic ducks. erations the population had been in captivity We observed some differences in Leucocy- was unknown (Fallis and Khan, pers. commun.). tozoon prevalence among sites. In particular, This history could easily have made these birds Kennisis Lake ducklings took longer to obtain more susceptible to parasites than completely infections than their counterparts at Lake Sa- 578 LEUcocYTOZOONAND DUCKS * Shutler et al. J. Wildl. Manage. 60(3):1996

sajewan. The Kennisis Lake site was more ex- infected themselves. Before ducklings hatch, posed to wind than the Lake Sasajewan site, and duck populations in our study area are at low wind may have reduced black fly attacks on densities,and the chance of a black fly obtaining ducklings. In addition, Kennisis Lake is sur- Leucocytozoon is fairly low. Once ducklings rounded by several recent cottage develop- hatch, host density increases dramatically,and ments, whereas the area around Lake Sasajewan a greater proportionof black flies picks up par- has remained essentially unaltered for several asites. Thus, there is a lag time between first decades. Hence, human alterations of Kennisis black fly emergence and peak in Leucocytozoon Lake habitat may have reduced the black fly transmission(Desser, Univ. Toronto,pers. com- population relative to that of Lake Sasajewan. mun.). Hence, on average,ducklings would have If black fly populations were denser at Lake more time to develop without Leucocytozoon Sasajewan, higher intensity infections found here if they hatched earlier in the year. However, among infected ducklings suggest either that countervailing selection may select for laying Leucocytozoon was more virulent at Lake Sa- dates that are more closely associatedwith food sajewan, or that more black fly bites result in availability. higher parasitemias than do fewer bites. Our Our most importanttests involved comparing efforts to census black fly population densities susceptibility of duckling types to Leucocyto- were unsuccessful, so we have no firm data on zoon. Bennett et al. (1993) pointed out that birds the possibility of differences in black fly densities introduced to unfamiliar areas often suffer among our study sites. However, differences in greatermortality than endemic populations.We infection rate and black fly biting rate have been observed significantly greater infection inten- observed between 2 sites in Algonquin Park sities in 2 mallard types in 1993 relative to the (Khan, pers. commun.). Hence, as might be ex- singleblack duck type, and this is consistentwith pected, certain locales appeared to be safer ha- Bennett et al. observations. However, differ- vens from Leucocytozoon than others, and if ences were not significant after 10 days post- the parasite is important, we might expect ducks exposure.Furthermore, trends were not the same to make use of these areas. in 1992. Consequently,we conclude that duck- We also observed, in the second year of the ling types differ little or not at all in their sus- study, that ducklings on Bena Lake had higher ceptibility to Leucocytozoon. Thus, we have no intensity infections 10 days post exposure than support for either the resistance or the suscep- those at Lake Sasajewan. Because infections take tibility hypotheses;therefore we accept the null 6 to 10 days before they produce mature ga- hypothesis. metocytes in the blood stream, Bena Lake duck- There are at least 2 other hypothesesthat we lings probably acquired their infections in the can consider with our data. One part of our week the ducklings spent at Lake Sasajewan. protocolwas to create hybridsbetween 2 species Ducklings at Bena Lake were sampled 4 days of ducks. The biological species concept (Mayr after they had been moved from Lake Sasaje- 1963)and Bartonand Hewitt (1985) hybridzone wan, and this stress may have played a role in model both predict that hybrids will be more temporarily increasing their parasitemias (Ap- susceptibleto parasitesthan pure parentalstocks. plegate 1970). Stress from moving between ponds Support for this prediction has been found in when led by a hen is probably less pronounced studies of various animal and plant taxa (Sage than what we caused, and probably has a min- et al. 1986, Dupont and Crivelli 1988, Whitham imal effect on parasitemias, so the result we 1989, Mouliaet al. 1991, Bert et al. 1993). How- obtained may not be of significance to wild pop- ever, the opposite patternof "hybridvigor" has ulations. also been observed (Boecklen and Spellenberg Time of year at which ducklings were ex- 1990), and equivalent susceptibility has been posed influenced when they became infected, observed between parental and hybrid stocks especially in 1993. If duckling mortality from (Heaney and Timm 1985). Our data fall into Leucocytozoon is important, hen mallards could the latter category, because 2 types of hybrid time laying of their clutches to coincide with ducklings we used were no more susceptibleto seasons that have, on average, reduced Leuco- Leucocytozoon than pure types. Generaliza- cytozoon transmission. When black flies first tions about how hybridizationaffects suscepti- emerge, they carry no Leucocytozoon. They bility to disease and fitness in general may be need to first feed on infected ducks to become premature(Arnold and Hodges 1995), and may J. Wildl. Manage. 60(3):1996 LEUCOCYTOZOON AND DUCKS * Shutler et al. 579 be hampered by species and hybrid definitions CAMBPELL. 1974. Hematozoa of the Anatidae that do not to all situations. of the Atlantic Flyway. 1. Massachusetts. J. Wildl. apply Dis. 10:442-451. , M. A. PIERCE,AND R. W. ASHFORD. 1993. MANAGEMENTIMPLICATIONS Avian haematozoa: mortality and pathogenicity. Nat. Hist. 27:993-1001. have been viewed as a J. Leucocytozoon may , A. D. SMITH,W. WHITMAN,AND M. CAMER- "passive" management tool for preventing mal- ON. 1975. Hematozoa of the Anatidae of the lard incursion into black duck habitat. However, Atlantic flyway. II. The Maritime provinces of our results suggest that Leucocytozoon will nei- Canada. J. Wildl. Dis. 11:280-289. , V. D. STOTTS,AND M. C. BATEMAN. 1991. ther be a barrier to additional gene flow between Blood parasites of black ducks and other anatids mallards and black ducks, nor will it prevent from Labrador and insular Newfoundland. Can. mallards from occupying areas currently oc- J. Zool. 69:1405-1407. cupied solely by black ducks. This implies that BERT, T. M., D. M. HESSELMAN,W. S. ARNOLD,W. the future of the black duck will continue to be S. MOORE,H. CRUZ-LOPEZ,AND D. C. MARELLI. 1993. High frequency of gonadal neoplasia in a source of controversy and concern (Ankney et a hard clam (Mercenaria spp.) hybrid zone. Mar. al. 1987, 1989; Conroy et al. 1989, Rusch et al. Biol. 117:97-104. 1989, Merendino et al. 1993). BLACK, F. L. 1992. Why did they die? Science 258: 1739-1740. LITERATURECITED BOECKLEN, W. J., AND R. SPELLENBERG. 1990. Structure of herbivore communities in two oak ANDERSON,J. R., D. O. TRAINER,AND C. R. DE- (Quercus spp.) hybrid zones. Oecologia 85:92- FOLIART. 1962. Natural and experimental 100. transmission of the waterfowl parasite, Leuco- BRODSKY,L. M., AND P. J. WEATHERHEAD. 1984. cytozoon simondi Mathis and Leger, in Wiscon- Behavioral and ecological factors contributing to sin. Zoo. Res. 1:155-164. American black duck-mallard hybridization. J. ANKNEY,C. D., D. G. DENNIS, AND R. C. BAILEY. Wildl. Manage. 48:846-852. 1987. Increasing mallards, decreasing American BURGESS, G. D. 1957. Occurrence of Leucocyto- black ducks: coincidence or cause and effect? J. zoon simondi Mathis and Leger in wild water- Wildl. Manage. 51:523-529. fowl in Saskatchewan and Manitoba. J. Wildl. ,AND . 1989. Increasing mal- Manage. 21:99-100. lards, decreasing American black ducks: no ev- CHERNIN, E. 1952. The epizootiology of Leuco- idence for cause and effect: a reply. J. Wildl. cytozoon simondi infections in domestic ducks Manage. 53:1072-1075. in northern Michigan. Am. J. Hyg. 56:39-57. , L. N. WISHARD,AND J. E. SEEB. CONROY,M. J., G. R. BARNES,R. W. BETHKE,AND 1986. Low genic variation between black ducks T. D. NUDDS. 1989. Increasing mallards, de- and mallards. Auk 103:701-709. creasing American black ducks-no evidence for APPLEGATE, J. E. 1970. Spring release of Plasmo- cause and effect: a comment. J. Wildl. Manage. dium relictum infections in an experimental field 51:523-529. population of English sparrows (Passer domes- DANIEL, W. W. 1978. Applied nonparametric sta- ticus). J. Wild. Dis. 6:37-42. tistics. Houghton Mifflin, Boston, Mass. ARNOLD,M. L., AND S. A. HODGES. 1995. Are nat- DENNIS,D. G. 1974. Waterfowl observations during ural hybrids fit or unfit relative to their parents? the nesting season in Precambrian and clay belt Trends Ecol. Evol. 10:67-71. areas of north-central Ontario. Can. Wildl. Serv. AVISE, J. C., C. D. ANKNEY, AND W. S. NELSON. Rep. Ser. 29:53-56. 1990. Mitochondrial gene trees and the evolu- DESSER, S. S. 1967. Schizogony and gametogony of tionary relationship of mallards and black ducks. Leucocytozoon simondi and associated reactions Evolution 44:1109-1119. in its avian host. J. Protozool. 14:244-254. BARROW,J. H., N. KELKER,AND H. MILLER. 1968. , AND G. F. BENNETT. 1993. The genera The transmission of Leucocytozoon simondi to Leucocytozoon, Haemoproteus and Hepatocys- birds by Simulium rugglesi in northern Michi- tis. Pages 273-305 in Kreier, J. P., ed. Parasitic gan. Am. Midl. Nat. 79:197-204. protozoa. Vol. 4. Acad. Press, London, U.K. BARTON,N. H., AND G. M. HEWITT. 1985. Analysis , AND A. K. RYCKMAN. 1976. The develop- of hybrid zones. Annu. Rev. Ecol. Syst. 16:113- ment and pathogenesis of Leucocytozoon si- 148. mondi in Canada and domestic geese in Algon- BELLROSE, F. G. 1976. Ducks, geese, and swans of quin Park, Ontario. Can. J. Zool. 54:634-643. North America. Stackpole Books, Harrisburg, , J. STUHT, AND A. M. FALLIS. 1978. Leu- Penn. 544pp. cocytozoonosis in Canada geese in upper Mich- BENNETT, G. F. 1960. On some ornithophilic blood- igan I. Strain differences among geese from dif- sucking diptera in Algonquin Park, Ontario, ferent localities. J. Wildl. Dis. 14:124-131. Canada. Can. J. Zool. 38:377-389. DUPONT, F.. AND A. J. CRIVELLI. 1988. Do parasites 1970. Simple techniques for making avian confer a disadvantage to hybrids? Oecologia 75: blood smears. Can. J. Zool. 48:585-586. 587-592. , W. BLANDIN,H. W. HEUSMANN,AND A. G. EWALD, P. W. 1983. Host-parasite relations, vec- 580 LEUCOCYTOZOONAND DUCKS * Shutler et al. J. Wildl. Manage. 60(3):1996

tors, and the evolution of disease severity. Annu. MOULIA,C., J. P. AUSSEL,F. BONHOMME,P. BOURSOT, Rev. Ecol. Syst. 14:465-485. J. T. NIELSEN,AND F. RENAUD. 1991. Wormy 1993. Evolution of infectious disease. Ox- mice in a hybrid zone: a genetic control of sus- ford Univ. Press, Oxford, U.K. 256pp. ceptibility to parasite infection. J. Evol. Biol. FALLIS,A. M., R. C. ANDERSON,AND G. F. BENNETT. 4:679-687. 1956. Further observations on the transmission NELSON, E. C., AND J. S. GASHWILER. 1941. Blood and development of Leucocytozoon simondi. parasites of some Maine waterfowl. J. Wildl. Can. J. Zool. 34:389-404. Manage. 5:199-205. , S. S. DESSER, AND R. A. KHAN. 1974. On O'ROKE,E. C. 1934. A malaria-like disease of ducks species of Leucocytozoon. Adv. Parasitol. 12:1- caused by Leucocytozoon anatis Wickware. Univ. 67. Michigan School For. and Conserv. Bull. 4:1-44. HAMILTON,W. D., AND M. ZUK. 1982. Heritable RICE, W. R. 1989. Analyzing tables of statistical true fitness and bright birds: a role for parasites? tests. Evolution 43:223-225. Science 218:384-387. ROGERS,J. P., AND J. H. PATTERSON. 1984. The HEANEY, L. R. AND R. M. TIMM. 1985. Morphology, black duck population and its management. Trans. genetics, and ecology of pocket gophers (genus North Amer. Wildl. Natur. Resour. Conf. 49: Geomys) in a narrow hybrid zone. Biol. J. Linn. 527-534. Soc. 25:301-316. ROLLER, N. F., AND S. S. DESSER. 1973. Diurnal HERMAN, C. M. 1968. Blood parasites of North periodicity in peripheral parasitemias in duck- American waterfowl. Trans. North Am. Wildl. lings (Anas boschas) infected with Leucocyto- Nat. Resour. Conf. 33:348-359. zoon simondi Mathis and Leger. Can. J. Zool. ,J. BARROW JR., AND I. B. TARSHIS. 1975. 51:1-9. Leucocytozoonosis in Canada geese at the Seney RUSCH, D. H., C. D. ANKNEY, H. BOYD, J. R. National Wildlife Refuge. J. Wildl. Dis. 11:404- LONGCORE, F. MONTALBANO, J. K. RINGELMAN, 411. AND V. D. STOTTS. 1989. Population ecology HERRE, E. A. 1993. Population structure and the and harvest of the American black duck: a re- evolution of virulence in nematode parasites of view. Wild. Soc. Bull. 176:379-406. fig wasps. Science 259:1442-1445. SAGE, R. D., D. HEYNEMAN,K.-C. LIM, AND A. C. HOLM, S. 1979. A simple sequentially rejective mul- WILSON. 1986. Wormy mice in a hybrid zone. tiple test procedure. Scand. J. Stat. 6:65-70. Nature 324:60-63. JOHNSGARD, P. A., AND R. DISILVESTRO. 1976. Sev- SAVAGE, A., AND J. M. ISA. 1955. An occurrence of enty-five years of changes in mallard-black duck Leucocytozoonsimondi in Manitoba.Can. J. Mi- ratios in eastern North America. Am. Birds 30: crobiol. 1:291-292. 905-908. SCHEMNITZ, S. D., EDITOR. 1980. Wildlife man- KHAN,R. A., AND A. M. FALLIS. 1968. Comparison agement techniques manual. Fourth ed. The of infections with Leucocytozoon simondi in Wildl. Soc., Washington, D. C. 686pp. black ducks (Anas rubripes), mallards (Anas SHEWELL, G. E. 1955. Identity of the black fly that platyrhynchos),and white Pekins(Anas bochas). attacks ducklings and goslings in Canada (Dip- Can. J. Zool. 46:773-780. tera: Simuliidae). Can. Entomol. 87:345-349. KOCAN, R. M., AND D. T. CLARK. 1966. Anemia in SYSTAT. 1992. SYSTAT: statistics. Version 5. SYS- ducks infected with Leucocytozoon simondi. J. TAT Inc. Evanston, Ill., 724pp. Protozool. 13:465-468. TRAINER, D. O., C. S. SCHILDT, R. A. HUNT, AND L. KREBS, C. J. 1989. Ecological methodology. Harper R. JAHN. 1962. Prevalence of Leucocytozoon and Row, New York, N.Y., 654pp. simondi among some Wisconsin waterfowl. J. MALEY, G. J. M., AND S. S. DESSER. 1977. Anemia Wildl. Manage. 26:137-143. in Leucocytozoon simondi infections. I. Quan- VAN RIPER,C. V. III, S. G. VAN RIPER,M. L. GOFF, tification of anemia, gametocytemia, and osmotic AND M. LAIRD. 1986. The epizootiology and fragility of erythrocytes in naturally infected Pe- ecological significance of malaria in Hawaiian kin ducklings. Can. J. Zool. 55:355-358. land birds. Ecol. Monogr. 56:327-344. MAYR, E. 1963. Animal species and evolution. Har- WHITHAM, T. G. 1989. Plant hybrid zones as sinks vard Univ. Press, Cambridge, Mass. 797pp. for pests. Science 244:1490-1493. MERENDINO, M. T., AND C. D. ANKNEY. 1994. Hab- ZAR, J. H. 1984. Biostatistical analysis. Second ed. itat use by Mallards and American Black ducks Prentice-Hall, Englewood Cliffs, N.J. 718pp. breeding in central Ontario. Condor 96:411-421. , AND D. G. DENNIS. 1993. Increas- Received 10 April 1995. ing mallards, decreasing American black ducks: Accepted 23 January 1996. more evidence for cause and effect. J. Wildl. Associate Editor: Murphy. Manage. 57:199-208.