THES‘ 3

                                                                                                                            

31293 01570 6207

This is to certify that the

thesis entitled

HEMOSPORID PARASITE COMMUNITIES OF
WATERFOWL (ANATIDAE) FROM THE
KELLOGG BIOLOGICAL STATION AND
DOUGLAS LAKE, MICHIGAN

presented by

RANDALL J. DEJONG

has been accepted towards fulfillment
of the requirements for

MASTERS degree in ZOOLOGY

Major professor

Date /W 18/IL “197

0-7639 MS U is an Aflinnativc Action/Equal Opportunity Institution

 

LIBRARY

Mlchlgan State
University

 

 

 

 

 

 

 

 

 

 

 

PLACE II RETURN BOX to remove thle checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

       

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HEMOSPORID PARASITE COMMUNITIES OF WATERFOWL (ANATIDAE)
FROM THE KELLOGG BIOLOGICAL STATION
AND DOUGLAS LAKE, MICHIGAN

By
Randall John DeJong

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Zoology
and
Interdepartmental Program in Ecology and Evolutionary Biology

1 997

ABSTRACT
HEMOSPORID PARASITE COMMUNITIES OF WATERFOWL (ANATIDAE)
FROM THE KELLOGG BIOLOGICAL STATION
AND DOUGLAS LAKE, MICHIGAN
By

Randall John DeJong

Totals of 2 l 8 and 114 wild waterfowl from the Kellogg Biological Station (KBS)
area in southwestern Michigan were examined for hematozoan parasites in summer and
fall 1995, respectively. Fifty-five common mergansers were examined from Douglas
Lake in northern lower Michigan in July 1995. Haemoproteus nettionis, H. greineri, and
Leucocytozoon simondi occurred in KBS waterfowl, with H. nettionis the most common.
Haemoproteus greineri and L. simondi fiequently and Plasmodium circumflexum
infrequently infected Douglas Lake common mergansers. Overall hematozoan
prevalence was higher in the fall than the summer in KBS waterfowl and was much
higher in Douglas Lake common mergansers than both summer and fall KBS waterfowl.
Intensity of H. nettionis in KBS waterfowl was lower in the fall. Transmission of
hemospon'ds did not occur at KBS and the abundance of potential vectors appeared low.
In contrast, high transmission was evident at Douglas Lake and appeared to be
responsible for differences in component communities between the two locations. The
component communities in KBS waterfowl were isolationist in character while those at
Douglas Lake demonstrated some evidence of species interactions. Common merganser

is reported as a new host record for H. greineri and P. circumflexum.

ACKNOWLEDGMENTS

I thank first of all my major professor, Dr. Patrick M. Muzzall, for his guidance,
patience, and criticism during the development and completion of my thesis research.
His kind advice and investment in me as his graduate student will always be appreciated.
I also wish to thank the members of my graduate committee, Dr. Edward D. Walker, and
Dr. C. Robert Peebles, for serving as counsel during my graduate studies. I especially
thank Dr. Peebles for the natural history trips he led, both on campus and off. My only
regret is that we did not go on more.

I owe a great deal to Dr. Harvey D. Blankespoor, who first introduced me to
parasitology and to research. Furthermore, his character and spirit make him a model for
my life. I thank him also for serving on my graduate committee before his teaching
sabbatical in China forced him to withdraw. Finally, his willingness to let me sample
common mergansers at Douglas Lake added much to the thesis.

Many thanks go to W. C. “Joe” Johnson, wildlife specialist at the Kellogg Bird
Sanctuary, for his interest and involvement in the research. His knowledge of birds was
invaluable and it would not have been possible without him. I thank him also for
providing research space at the Sanctuary. He and Karen Charleston, wildlife assistant at
the Sanctuary, put much effort into helping me capture and sample birds, and in addition
graciously sampled some of the fall birds. I also thank Amy Kletzien and Janae'
Reneaud, the two student interns at the Sanctuary who helped out in numerous ways
throughout the summer. Not only was their workload increased by the sentinels used in

the study, but their help catching mallards on Gull Lake was critical to our success.

iii

Thanks are extended to the following for their expertise: Dr. Ellis Greiner of the
University of Florida confirmed identification of parasite specimens, Dr. Robert Adlard
of the Queensland Museum in Queensland, Australia provided reference specimens from
the International Reference Center for Avian Hematozoa, and Dr. Mary Garvin of Notre
Dame University gave encouragement and feedback on my analyses.

I am grateful for financial support of this Study from the following: the George H.
Lauff Research Award, the Department of Zoology, and the Ecology and Evolutionary
Biology program.

Wild waterfowl were captured under Michigan Department of Natural Resources
Scientific Collector’s Permit 80939 and US. Fish and Wildlife Service Migratory Bird
Permit PRT-804073. Banding was done under the US. Fish and Wildlife Service
Banding Permits of the Kellogg Bird Sanctuary.

Several fellow graduate students are appreciated in many ways. I thank Alex
Hernandez for his friendship and for the many discussions and lunches we shared.
Micaela Szykman’s friendship is also very much appreciated. Erin Boydston enjoyed an
afternoon chasing mallards on Gull Lake. Eric Amoys and Brian Keas endured as my
housemates and I am honored to call them friends.

I thank my parents for their continuing love, encouragement, and prayers.

Jennifer Knox, my treasured fiancee and best friend, receives my deepest thanks
for her support, encouragement, and love.

Finally, I thank my Father in Heaven, the Great Creator, for His marvelous

handiwork in the universe, His care for it and for me, and the gift of wonderment about it

all.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................................................ Vi
LIST OF FIGURES ............................................................................................................... V 11
INTRODUCTION ................................................................................................................. 1
MATERIALS AND METHODS ......................................................................................... 8
Descriptions of Hemosponds .................. 8
Descriptions of Study Locations .................................................................................... 11
The Kellogg Biological Station Area .............................................. 11
Douglas Lake ................................ 14
Field Methods .................................................................................................................. 15
Waterfowl capture and sampling ....................................................................... 15
Sentinel dueklings ............................................................................................... 16
Vector collection ................................................................................................. 16
Laboratory Methods ........................................................................................................ 2 0
Data Analyses .................................................................................................................. 2 1
RESULTS ............................................................................................................................. 23
Waterfowl sampled ......................................................................................................... 23
Hematozoa of KBS summer waterfowl ......................................................................... 23
Hematozoa of KBS fall waterfowl ................................................................................. 26
Hematozoa of Douglas Lake common mergansers ....................................................... 34
Sentinel dueklings ........................................................................................................... 37
Vector collection ............................................................................................................. 37
DISCUSSION ........................................................................................................................ 3 8
Kellogg Biological Station ............................................................................................. 38
Douglas Lake ................................................................................................................... 44
CONCLUSIONS ................................................................................................................... 48
BIBLIOGRAPHY 50

Table I.
Table II.
Table III.
Table IV.
Table V.
Table VI.
Table VII.
Table VIII.
Table IX.

LIST OF TABLES

Number of waterfowl by species, age, and season sampled from KBS ...... 24

Prevalence of hematozoa in KBS summer waterfowl 25

ooooooooooooooooooooooooooooooooooo

Overall hematozoan prevalence by age class in KBS summer waterfowlm25

Mean intensity of hemosporids in KBS summer waterfowl ........................ 27

Species and number of captive waterfowl sampled from the Sanctuary ...... 2 8

Prevalence of hematozoa in KBS fall waterfowl ........................................... 2 9
Mean intensity of hemosporids in KBS fall waterfowl ................................. 3 3
Prevalence of hematozoa in Douglas Lake common mergansers ................ 35

Mean intensity of single and dual infections in common mergansers ......... 36

vi

Figure 1.
Figure 2.
Figure 3.
Figure 4.

LIST OF FIGURES

Diagram of the hemosporid life cycle .............................................................. 4
Map of study locations ...................................................................................... 13
Diagrammatie representation of Bennett nap .................................................. 19

Relationship between age and overall prevalence in KBS waterfowl ........... 32

vii

INTRODUCTION

The structure and organization of parasite communities can be highly complex
and are the product of long and continuing interactions involving both host and parasite
populations over evolutionary time (Esch et al., 1990). In ecological time, the structure
and organization of parasite communities can be influenced quickly by certain factors.
One such factor that can influence community structure is transmission rate; very
different community structures are thought to be produced by low and high transmission
rates. Low transmission rates should produce communities that are unsaturated and in
nonequilibrium and that provide few opportunities for interspecific interactions. Such a
community can be characterized as isolationist (Holmes and Price, 1986). On the other
hand, high transmission rates should lead to saturation and equilibrium, providing more
opporumities for interspecific interactions. Such a community would be characterized as
interactive (Holmes and Price, 1986).

Transmission itself is dependent upon other factors. These can be broadly divided
into abiotic and biotic factors (Holmes and Price, 1986). Abiotic factors include
temperature, humidity, etc. Biotic factors can be divided into host-intrinsic factors, such
as host species, age, sex, and immune response, and parasite-intrinsic factors, such as
parasite species, development rate, and reproductive rate.

Parasite communities are naturally hierarchical. The parasites of individual hosts,
termed infracommunities, collectively comprise all of the parasites in all individuals of a
given host species within an ecosystem, termed the component community (Holmes and
Price, 1986). Likewise, parasite populations are hierarchical and form the basis for the
community hierarchy. Infrapopulations refer to the population of a single parasite species

within an individual host which collectively form the infracommunity. Metapopulations

are all members of a parasite species which infect a given host species within an
ecosystem. Collectively, metapopulations form component communities.

Of the many component communities of birds (intestinal helminths, intestinal
protozoa, ectoparasites, among others), only the hematozoan community uses the
bloodstream for growth and reproduction. Hematozoan communities can include
intracellular hemosporids (Apicomplexa, Hemosporina, Plasmodiidae), filarial nematodes
(N ematoda, Filaroidea, Onchocercidae), and u'ypanosomes (Sacrcomastigophora,
Kinetoplastida). All utilize hematophagous arthropods as vectors. Hemosporid parasites
of birds, i.e., species of Haemoproteus, Leucocytozoon, and Plasmodium have received
much attention by parasitologists and ornithologists worldwide. Hemosporids can be
ecologically important, such as in their role in the decline of native birds in Hawaii (van
Riper III et al., 1986). They are also probably evolutionarily important in birds, and
recently investigators have studied their possible role in the evolution and maintenance of
sexual dichromaticism and dimorphism in birds (the Hamilton-Zuk hypothesis—Hamilton
and Zuk, 1982; see review by Clayton, 1991). Filarial nematodes and trypanosomes are
reported less often and their importance is less understood.

Ducks, geese, and swans (Anseriformes: Anatidae) are important because
domesticated species are used for food and wild species for recreational bird watching
and hunting. Consequently, the parasites of the Anseriformes have been studied more
than the parasites of any other group of birds except the Galliformes.

The term waterfowl usually applies only to the Anatidae, and not to all aquatic
birds. I follow this convention and use the term waterfowl to refer to the Anatidae only.
Hemosporids are among the most studied of waterfowl parasites (Bennett and Squires-
Parsons, 1992). Leucocytozoon simondi, the only species of Leucocytozoon that occurs in
anatids, causes high mortality in domestic flocks of waterfowl (O’Roke, 1934; Fallis et
al., 1951; Chemin 1956a). Leucocytozoon simondi also can cause mortality in wild

populations. O’Roke (1934) reported losses in wild mallards (Anas platyrhynchos) and

black ducks (A. rubripes) in northern Michigan. Herman et al. (1975) cited the parasite
as a limiting factor of the Canada goose (Branta canadensis) population at the Seney
National Wildlife Refuge in the upper peninsula of Michigan. Of limited or no known
pathogenicity in waterfowl are Plasmodium spp. (see Greiner et al., 1975b; Wobeser,
1981), with 11 species in anatids (Bennett et al., 1993). Waterfowl species of
Haemoproteus, H. nettionis and H. greineri, are also of limited or no known
pathogenicity (see Wobeser, 1981; Bennett et al., 1984). T wanosoma spp. are not
known to cause disease in waterfowl (Herman, 1968; Wobeser, 1981) and the
pathogenicity of filarial nematodes in waterfowl is controversial (W obeser, 1981).

Host specificity of hemosporids varies between genera. Haemoproteus has been
proven to be family specific (Atkinson, 1986), meaning that H. nettionis and H. greineri
only infect species of Anatidae. Leucocytozoon spp. are also believed to be family
specific, based on morphology and vectors (Bennett et al., 1994); Leucocytozoon simondi
only infects species of Anatidae. Plasmodium spp., however, are not completely family-
specific and some species infect more than one bird family (Bennett et al., 1993).

The life cycles of hemosporid genera are similar (Figure 1). All three genera
utilize female true flies (Diptera) as vectors. Haémoproteus, Leucocytozoon, and
Plasmodium use biting midges (Ceratopogonidae) and hippoboscid flies (I-Iippoboscidae),
black flies (Simuliidae), and mosquitoes (Culicidae), respectively. Vectors of
Haemoproteus spp. in waterfowl are believed to be ornithophilic Culicoides spp.
throughout the world, based on the determination of C. downesi as a vector of H.
nettiom's in eastern North America (F allis and Wood, 1957). Sibley and Werner (1984)
reported C. downesi as a vector of H. nettionis in Michigan’s upper peninsula. The
vectors of H. greineri are presumed to be Culicoides spp. as well. Vectors of
Leucocytozoon simondi are several species of ornithophilic Simulium (see O’Roke, 1934).
Vectors of Plasmodium spp. of waterfowl are Culex spp. and Culiseta spp. (see Meyer
and Bennett, 1976; Work et al., 1990).

oocysts

 

ookinetes
SPOROGONY
zygotes
VECTOR
gametes sporozoites
gametocytes
VERTEBRATE HOST
schizogony
GAMETOGONY SCI-IIZOGONY
merozoites

Figure 1. Diagram of the hemosporid life cycle. Redrawn from Atkinson and van Riper III (1991).

Transmission to waterfowl occurs when the vector injects sporozoites into the bird while I
obtaining a blood meal. Sporozoites invade host tissues such as the liver, grow, and
reproduce asexually (schizogony) to produce merozoites.

Although the life cycles of Haemoproteus, Leucocytozoon, and Plasmodium are
similar, they difler in the types of tissue where schizogony occurs. Haemoproteus and
Leucocytozoon undergo schizogony in fixed, non-circulating cells only, and the only
stages of development in circulating blood cells are gametocytes. In the case of
Plasmodium, schizogony occurs in the fixed non-circulating cells as well as in circulating
erythrocytes; therefore both gametocytes and schizonts may be found in blood cells. The
presence of both gametocytes and schizonts is a key difference that separates Plasmodium
fi'om the other two genera (Greiner and Bennett, 1975, Greiner et al., 1975b, Atkinson
and van Riper III, 1991).

Merozoites can reinvade liver cells or invade circulating blood cells. In the case
of Leucocytozoon, some merozoites invade leukocytes and some invade erythrocytes,
whereas Haemoproteus and Plasmodium merozoites invade only erythrocytes. Upon
. invading blood cells, merozoites develop into gametocytes (gametogony).

When another suitable vector ingests gametocytes, the life cycle is continued. In
the midgut of the vector, sexual reproduction occurs when gametes form from the
gametocytes (gametogenesis) and then fuse to form zygotes. Zygotes transform into
elongate ookinetes which penetrate the insect’s midgut epithelium and develop into
oocysts within cells. Asexual reproduction (sporogony) occurs in the oocyst to yield
numerous infective sporozoites. These rupture the epithelial cell and move to the salivary
glands where they remain until the vector’s next blood meal (Atkinson and van Riper III,
1991)

The duration of each hemosporid life stage varies between genera and species;
only estimates are known for some. Of particular importance to the present study,

however, is the length of time required for the waterfowl host to develop a parasitemia

(circulating gametocytes) after the infective bite of a vector, termed the prepatent period.
For H. nettionis, the prepatent period in waterfowl is approximately 14 days (F allis and
Wood, 1957). The prepatent period of H. greineri is not known, but is presumed to be
similar to H. nettionis. For L. simondi, the prepatent period is 7-9 days (Kocan and Clark,
1966a; Desser and Ryckrnan, 1976). Plasmodium spp. have a prepatency period of 14-19
days in ducks and 31 days in geese (Meyer et al., 1974).

In temperate climates, transmission of hemosporids primarily occurs in the spring
and early summer, during the breeding season of birds and emergence period of vectors.
Adult birds with chronic infections suffer a seasonal increase in parasitemia, or spring
relapse, that coincides with egg-laying in female birds (Chernin 1956b) increasing the
likelihood that newly hatched vectors will obtain parasites. The number of bird hosts is
increased when infected vectors obtain subsequent blood meals from hatchlings and
sporozoites are injected (Atkinson and van Riper III, 1991).

Surveys of wild waterfowl in Michigan have been few and limited to the northern
lower peninsula and the upper peninsula (O’Roke, 1931, 1934; Herman et al., 1975). In
addition, several experimental and transmission studies of waterfowl hematozoa were
completed at Douglas Lake in the northern lower peninsula (O’Roke, 1931, 1934;
Chernin, 1956a; Barrow et al., 1968) and at various locations in the upper peninsula
(T arshis, 1976; Desser et al., 1978; Sibley and Werner, 1984). Work has not been done
on waterfowl hematozoa in southern Michigan, except for a report by Kocan (1966) of a
canvasback (Aythya valisineria) infected with a Plasmodium sp. captured at the Kellogg
Biological Station in southwestern Michigan. He believed that the canvasback was
migrating (it was captured in November) and had acquired its infection on its breeding
grounds.

The terms and concepts for parasite communities developed by Holmes and Price
(1986), Esch et al. (1990), and others are based on helrninth communities, but can be

applied to hematozoan communities. Yet only a few investigators have done so. Godfrey

et al. (1990) examined the effects of host-intrinsic factors and collection location on the
haemoproteid community (Haemoproteus spp. only) of mourning doves in western
Texas. Fedynich and Rhodes, Jr. (1995) analyzed the structure and pattern of the
hematozoan community of wintering turkeys in South Carolina. Fedynich et al. (1993)
studied the structure and pattern of the hematozoan community of three species of
waterfowl wintering in south Texas.

The objectives of this study were to characterize the hematozoan community of
waterfowl fi'om the Kellogg Biological Station area in southwestern Michigan by 1)
evaluating the effects of host-intrinsic factors, such as species, age, and sex, 2) evaluating
the effects of season, 3) monitoring transmission in the KBS area, and 4) identifying
vectors which might have a role in transmission. In addition, common mergansers
(Mergus merganser) from Douglas Lake in northern lower Michigan were examined for
blood parasites. The effects of host-intrinsic factors were evaluated in common
merganser data. Comparisons of hematozoan communities are made between the two

locations.

MATERIALS AND METHODS

Descriptions of Hemosporids .

The most diagnostic characters for hemosporids are gametocyte shape and color,
position of the host cell nucleus, and the number of pigment granules in the parasite
cytoplasm. Also, the character of any red bodies is used in distinguishing between
Haemoproteus spp. gametocytes and Plasmodium spp. schizonts.

Haemoproteus nettionis Johnston and Clelland, 1910 is nearly global in
distribution, presumably throughout the range of the Anatidae, and has been reported
fiom at least 47 waterfowl species (Williams and Bennett, 1980). The macrogametocytes
are robustly halteridial, or crescent-shaped with a smooth margin. The host cell nucleus
is laterally displaced, sometimes completely to the periphery. The parasite cytoplasm
stains blue with Giemsa’s stain. Pigment granules are present in the cytoplasm, usually
discrete and yellow-brown in color and average 24 granules per parasite. Numerous
volutin granules, staining pink-red in color, are sometimes present (E. C. Greiner, pers.
comm. ). The parasite nucleus stains pink to rose in color and occupies approximately 5%
of the parasite area. The parasite occupies approximately 80% of the erythrocyte-parasite
complex, hypertrophying the host cell (25%) and atrophying the host cell nucleus (18%)
in area The microgametocytes are similar in appearance to the macrogametocyte, but
displace the host cell nucleus slightly more and the parasite cytoplasm stains a lighter
blue with Giemsa’s stain (Williams and Bennett, 1980; Bennett and Pierce, 1988).

Haemoproteus greineri Bennett, Turner, and Whiteway, 1984 has been reported
from at least seven waterfowl species and is considered endemic to the prairie provinces
of Canada (Bennett et. al., 1984; Fedynich et. al., 1993). The macrogametocytes are

circumnuclear, meaning that they completely encircle the erythrocyte nucleus. The

margin of the parasite is entire, not ameboid. The host cell nucleus is usually slightly
laterally displaced or not at all, although it is rarely displaced to the periphery. The
parasite cytoplasm stains a deep blue with Giemsa’s stain. Yellow-brown pigment
granules are present, but are smaller and more numerous (averaging 38 per parasite) than
for. H. nettionis. Numerous volutin granules, staining pink-red in color, are sometimes
present (E. C. Greiner, pers. comm. ). The parasite nucleus stains deep pink with
Giemsa’s stain and averages about 9% of the parasite area. The parasite occupies 90% of
the erythrocyte-parasite complex, hypertrophying the host cell (20%) and atrophying the
host nucleus (14.5%) in area. There are also “round” forms of the macrogametocyte that
show gross hypertrophy in width of the parasitized erythrocyte. The microgametocyte is
similar in appearance to the macrogametocyte, but averages slightly fewer pigment
granules (35), stains a lighter blue with Giemsa’s stain, and possesses a nucleus which
constitutes a larger portion (41%) of the parasite area (Bennett et al., 1984; Bennett and
Pierce, 1 988).

Leucocytozoon simondi Mathis and Leger, 1910 has a well documented
distribution, especially in North America, and is considered a parasite of Holarctic
anatids. It has been reported from at least 42 waterfowl species (Bennett and Squires-
Parsons, 1992). The macrogametocytes occur in both round and fusiforrn morphs. The
round morph is round to broadly ovoid, but prone to be distorted into a variety of shapes.
The parasite occupies 80-85% of the area of the host cell-parasite complex. The parasite
cytoplasm stains blue with Giemsa’s stain and no pigment granules are present. The
parasite nucleus stains pink with Giemsa’s stain and is round to broadly ovoid,
occasionally narrowly elliptical, and occupies 7% of the area of the parasite. The nucleus
of the host cell is displaced, forming a distinct cap on the host-parasite complex and
covering 33% of the periphery of the parasite and 16% of the area of the host cell-parasite

complex.

10

The fusiform morph of the L. simondi macrogametocyte occupies only 65% of the
host cell-parasite complex, is narrowly ovoid to elliptical in shape, and is never round.
The parasite cytoplasm and nucleus stain similarly to that of the round morph and again
no pigment granules are present. The nucleus of the host cell lies along one side as a
narrow ribbon, and expands into two terminal bulbs. The host nucleus covers 45% of the
periphery of the parasite and occupies 30% of the area of the host cell-parasite complex
(Bennett and Squires-Parsons, 1992).

The microgametocyte also occurs in both round and fusiform morphs. The round
microgametocyte has similar measurements to round macrogametocytes, but the fusiform
microgametocyte averages 5-10% smaller than the fusiform macrogametocyte. Both
morphs of the microgametocyte stain lighter in color with Giemsa’s stain (Bennett and
Squires-Parsons, 1992).

Plasmodium circumflexum Kikuth, 1931 has been described from 43 avian
families and 138 species and is global in distribution (Bennett et al., 1993). The
macrogametocytes are nearly circumnuclear with an irregular margin. The host cell
nucleus is usually laterally displaced. The parasite cytoplasm stains blue with Giemsa’s
stain. Brown pigment granules are numerous and scattered. The microgametocyte is
similar to the macrogametocyte, but stains red in Giemsa’s (Gamham, 1966).

AS stated above, Plasmodium spp. differ fiorn Haemoproteus spp. and
Leucocytozoon spp. in having intraerythrocytic schizogony. Mature erythrocytic
schizonts of P. circumflexum are circumnuclear, do not displace the host cell nucleus, and
the merozoite nuclei can resemble volutin granules, making them very similar in
appearance to H. greineri gametocytes. However, the schizont nuclei are often larger, are
less round, possess a ‘halo’ around them, and are never immediately adjacent to one
another (pers. observ. ). The number of merozoites present in a mature schizont ranges

from six to thirty (Gamham, 1966; Greiner et al., 1975b; E. C. Greiner, pers. comm).

ll

Descriptions of Study Locations

The Kellogg Biological Station Area
The WK. Kellogg Biological Station (KBS) of Michigan State University lies in

southwestern Michigan, at 42°24’N, 85°24’W. This study focused on three locations
within or adjacent to the station: Gull Lake, the Kellogg Bird Sanctuary, and the Kellogg
Farm (Figure 2).

Gull Lake. Gull Lake is an alkaline, oligotrOphic to mesotrophic lake
approximately 822 ha in surface area and up to 35 m deep. The entire lakeshore is
developed and heavily populated, and the lake is extensively used for recreation and
fishing. Mallards are the only waterfowl species to breed on this lake, although other
species use it during the migratory seasons. Typically, about 12 broods of mallard
ducklings are produced each year (W. C. Johnson, pers. comm).

Kellogg Bird Sanctuary. The Kellogg Bird Sanctuary totals 25 ha in area and its
most prominent natural feature is Wintergreen Lake. Wintergreen Lake is small (15 ha),
shallow (6.3 m maximum depth), and alkaline. Resident and migratory waterfowl
pOpulations contribute substantial nutrient input and the lake is considered
hypereutrOphic. Approximately 3-5 wood duck (Aix sponsa), 5-8 mallard, and 10-12
Canada geese broods are produced on Wintergreen Lake each year. Including these
broods, about 50 wood ducks, 200 mallards, and 100 Canada geese use this lake during
the summer. During the migratory season, the lake is used by over 3000 Canada geese,
2000 wood ducks, mallards, and other duck species, and by 50-100 swan. Over the last
20 years, the number of ducks, geese, and swan using the lake has remained constant (W.
C. Johnson, pers. comm): Wintergreen Lake is approximately 300 In from the shore of
Gull Lake and approximately 1500 m away from the biological laboratories of KBS.
Other habitats of Kellogg Bird Sanctuary include smaller ponds used for captive

waterfowl near Wintergreen Lake, and mature beech-maple forest.

12

Figure 2. The Kellogg Biological Station (KBS) area in southwestern Michigan and
Douglas Lake in northern lower Michigan. The KBS area includes Gull Lake,
Wintergreen Lake, and Kellogg Farm. (Figure redrawn based on Tiger Mapping Service,
US. Census Bureau).

 

l3

 

 

z——->
//\

 

Kellogg Farm

GullLake

 

 

Wintergreen Lake
muss
0| 0.1:»I row
oI 1.0 12.0

 

 

Figure 2.

 

 

 

 

 

 

l4

Kellogg Farm The Kellogg Farm is a dairy farm dedicated to research and
education. The two small lagoon ponds associated with the farm are used by 2-3 Canada
goose and 2-3 mallard broods each year (W. C. Johnson, pers. comm). These lagoon
ponds are approximately 1500 m away from Wintergreen Lake and 2000 m from Gull
Lake.

Douglas Lake
Douglas Lake is an alkaline, mesotrophic lake in northern lower Michigan,

45°36’N, 84°42’W, with an approximate surface area of 38 ha and a maximum depth of 9
m (Figure 2). Recreation and fishing use is moderate, and there are about 75 cottages on
the lakeshore. In addition, the University of Michigan Biological Station is located on
the southernmost shore. Approximately 5 broods of mallards, 1-3 broods of wood ducks,
and 1-3 broods of common mergansers are produced on Douglas Lake each year (H. D.

Blankespoor, pers. comm).

15

Field Methods

Waterfowl capture and sampling
Wild waterfowl fi'om the KBS area were captured by drive trap and bait trap and

sampled for blood hematozoa in summer 1995 (June 1-August 24). These birds were.
from all three KBS areas (Gull Lake, Kellogg Bird Sanctuary, and Kellogg Farm). Drive
traps consist of a V-shaped fence into which waterfowl are driven. The point of the V
opens into a small corral from which the waterfowl are removed by hand. A bait trap
consists of a wire cage with openings that allow easy entry but make escape difficult.
These were partially submerged near the shores of Wintergreen Lake and baited with

' corn. Additionally, wild waterfowl from Wintergreen Lake were captured by bait trap
and sampled in fall 1995 (September 9-October 8).

Captive waterfowl belonging to the Kellogg Bird Sanctuary were also sampled
during summer 1995. Many of these were restricted to outdoor pens; others were free-
ranging but clipped wing feathers or previous injuries restricted them from moving
outside the Sanctuary. Common mergansers were captured by drive trap from Douglas
Lake on July 17, 1995.

Upon capture, the species, age, and sex of each bird were determined. Sampled
waterfowl were assigned to one of three age classes. Afier-hatch-year (AHY) birds are .
those that hatched in 1994 or before and have thus completed one or more migrations to
wintering grounds. Hatch-year (HY) birds are those that hatched in 1995 but had attained
flight at the time of sampling. Although HY waterfowl may have hatched in the area of
sampling, it is also possible that they hatched elsewhere. Local (L) birds are those that
hatched in 1995 in the area they were sampled and had not yet attained flight at the time
of sampling.

Blood was drawn via metatarsal vein puncture with a sterile lancet. Two thin

blood smears were made from each bird. Birds without bands were fitted with a US.

16

Fish and Wildlife Service (U SFWS) band and then released. Birds already possessing

bands were recorded and released.

Sentinel ducklings
A total of 186 ducklings was used throughout the summer to assess whether

transmission of waterfowl blood parasites occurred in the KBS area. Sentinels used in
the study were black ducks, mallards, wood ducks, and domestic mallards. These birds
were obtained either from eggs hatched in the laboratory, or as 1-3 day old “orphans”
which were brought to the Sanctuary by the public. All sentinels were raised in the
laboratory until they were 14-17 days old, where exposure to vectors was severely
limited. They were then placed in outdoor pens (breeding pens of the Sanctuary) for the
next 7-14 days, where they were accessible to potential vectors and located approximately
50 m from Wintergreen Lake. The next 20-22 days were spent in cages (also accessible
to potential vectors) partially submerged in the lake, approximately 3 m from the shore.
This exposure regime allowed for at least one group of sentinels to be exposed for nearly
every date between May 30 and September 1.

Following exposure, each group of experimental birds was sampled for blood
parasites. Blood smears were made as for wild waterfowl. Experimental birds, except
domestic mallards, were banded and released after sampling. Eighteen domestic mallards
were handed with non-USFWS bands, placed back into the outdoor pens at the Sanctuary
on July 20, and resampled on August 31. This provided an additional test of whether

transmission of any blood parasites occurred.

Vector collection
Two variations of the Bennett trap (Bennett, 1960), which utilized live ducks as
attractants, were used in collecting waterfowl-feeding flies. The first, similar to the

original trap (Bennett, 1960), consisted of two portions: the holding cage for bait birds,

l7

and the collecting cage for insects (Figure 3). The holding cage was 30 X 30 X 30 cm
and made of l-inch mesh vinyl-coated wire cloth. The top of the holding cage was
hinged, providing access to the inside of the cage. Feed and water dishes were wired to
one side. The collecting cage consisted of a 60 X 60 X 60 cm wood frame covered by
fine nylon screen (250-micron mesh) on five sides, leaving the bottom open. Strips of
sponge rubber were adhered to the bottom of the collecting cage to prevent flies fi'om
escaping. Bait birds were provided with food and water and placed in the holding cages,
and set on a plywood square near the lakeshore at about one hour before srmset. Forty
minutes later, the collecting cage was placed over the bait duck by hand or it was lowered
fi'om above using a rope slung over a tree branch. Engorged and unengorged flies were
collected in an aspirator through a sleeve in the top of the collecting cage and identified

’ later. This process of exposure and collection was repeated at sunset.

The second Bennett trap variation used was similar to that of Meyer and Bennett
(1976). Four polythene funnels of about 25 cm diameter were inserted in two sides of the
screened frame (Figure 3). The narrow stems of the funnels were removed, leaving
enhance holes of 5 cm in diameter at the base of the funnels. Bait birds provided with
food and water were placed in the holding cage (described above) within these traps at
1800-1900 hours and left until the following morning (approximately 0800 horns).
Engorged and unengorged flies were collected and preserved as above. Both Bennett trap
variations were used on calm, dry nights on the following 23 dates: June 6, 7, 18, 19, 21,
25, 28; July 2, 3, 4, 5, 8, 9, 10 , 21, 23, 24, 27, 29, 31; August 6, 12, 13. Four locations

near the shore of Wintergreen Lake were used, alternating location each collection night.

18

Figure 3. Diagrammatic representation of Bennett trap, consisting of insect trap and bait-
bird holding cage. Two variations of insect trap were used. In the first the insect trap
was placed over the bird from above after 40 minutes exposure. In the second, funnels
were added as entrances for insects and the insect trap placed over the bird for the entire

night.

 

position of
large sleeve
for insect
collection by
aspirator

funnels providing

enhances for insectsg.

(second variation only)

 

 

 

 

 

em sponge rubber
strip

 

holding cage
with water and
feed dishes
wired to one
side

Figure 3.

20

Laboratory Methods

Blood smears were air-dried, fixed in methanol, and stained with Fisher Scientific
LeukoStatTM Stain Kit (Fisher Scientific, Pittsburgh, PA, 15219). The results of
LeukoStat do not differ markedly from Giemsa’s stain with the exception of the
cytoplasm of microgametocytes, which stains red instead of light blue. Therefore,
parasites identified in this study were comparable to the descriptions above. To
determine prevalence of hematozoans, blood smears were first scanned at 200x
magnification so that both smears were examined in' their entirety (most gametocytes and
schizonts are easily seen at this magnification). Then they were examined at 1000X
magnification for 10 minutes each so that a total of 20 minutes was expended to examine
both slides from each bird, allowing scanning of about 250-350 fields of view, resulting
in an estimated 15,000-25,000 erythrocytes scanned. Prevalence is defined as the number
of birds infected with a particular hematozoan species divided by the number of birds
examined, in accordance with Margolis et al. (1982). Overall hematozoan prevalence is
defined as the number of birds infected with any hematozoan species divided by the
number of birds examined.

Quantification of hematozoans generally followed the recommendations of
Godfrey et al. (1987) and Fedynich et a1. (1995). If hematozoans were present, the best
slide was selected (based on smear thickness and staining) and 5,000 erythrocytes were
counted in 50 replicates of 100 erythrocytes each (at 4OOX) to provide an estimate of
parasite intensity within each infected bird. Each count of 100 erythrocytes (one
replicate) was obtained using one or more different fields of view delineated by the large
square of 3 Miller ocular disc. A random number table was used to determine the number
of fields skipped between each field of view examined. If the field of view was
inadequate for examination (e. g., too thick) the smear was advanced to the next suitable

field of view. Intensity is defined as the number of erythrocytes infected by a parasite

21

species divided by 5,000 erythrocytes counted in a particular host, in accordance with
Margolis et al. (1982). Intensities <1/5,000 erythrocytes were arbitrarily assigned a value
of 0.5 so that they could be included in analysis (Godfrey et al., 1990; Fedynich and
Rhodes, Jr., 1995).

Leucocytozoon spp. have rarely been quantified on blood smears (Allan and
Mahrt, 1989; Fedynich and Rhodes, Jr., 1995) because of concern over potential pooling
of this parasite on the smear and because Leucocytozoon spp. infect leukocytes in
addition to erythrocytes (Godfrey et al., 1987; F edynich et al., 1995). However, it is
possible to test if Leucocytozoon gametocytes vary concordantly with erythrocyte
densities (F edynich and Rhodes, Jr., 1995). Based on regression analysis from a
subsample of ten smears from ten common mergansers from Douglas Lake,
Leucocytozoon simondi densities varied concordantly with erythrocyte densities. Thus,
pooling does not appear to be a significant factor and the number of L. simondi

gametocytes per 5000 infected erythrocytes is an accurate measure of intensity.

Data Analyses

Contingency analyses were used to analyze KBS prevalence data and the effects
of host species, age, sex, and season. For this analysis, black ducks and black-duck
mallard hybrid data were removed because of the small number of birds captured.
Douglas Lake prevalences were compared between sexes using contingency analysis.
Comparisons between KBS summer data and Douglas Lake data were made using
contingency analyses of overall hematozoan prevalence and the prevalences of
hematozoan species occurring in both areas.

A regression analysis was performed on waterfowl age in years and overall

hematozoan prevalence. For this analysis, minimum waterfowl age in years was

determined. Local birds were arbitrarily given an age of 0.1 years, HY birds, 0.2 years,
and AHY birds, 1 year unless possessing a band when captured. Ages of birds already
possessing bands at the time of sampling were determined using the banding records of
the Kellogg Bird Sanctuary and the USFWS. If the bird originally received its hand as an
adult, only the minimum age of the bird could be determined.

Because intensity data were not normally distributed, Kruskal-Wallis tests or
Mann-Whitney U tests were used in their analysis. Comparisons were made between
mean intensities in KBS waterfowl for host species, ages, sexes, and seasons. Douglas
Lake mean intensities between waterfowl sexes were compared. Mean intensities of
hematozoan species common to both locations were also compared (using KBS summer
data only). All statistical analyses were performed using SYSTAT For Windows, v.6.0.1
(SPSS, Inc., 1996).

23

RESULTS

Waterfowl sampled

In total, five species of wild waterfowl were sampled from the KBS area: three
black ducks, 77 Canada geese, 220 mallards, two mallard-black duck hybrids, and 30
wood ducks (Table I). The most important differences between the summer and fall
samples are the absence of Canada geese in the fall sample and the natural absence of the
L age class in the fall sample. Fifty-five L common mergansers were sampled fiom

Douglas Lake.

Hematozoa of KBS summer waterfowl

Three hemosporid species, Haemoproteus nettionis, H. greineri, and
Leucocytozoon simondi were identified fiom wild waterfowl (Table 11). Overall
hematozoan prevalence was 5.5 %. Haemoproteus nettiom‘s was the most common
parasite species infecting 4.1 % of waterfowl. Leucocytozoon simondi and H. greineri
infected 1.8 % and 0.5 % respectively. One mallard was infected with both H. nettionis
and L. simondi, and one wood duck was infected with both H. nettionis and H. greineri.
Canada geese were infected with only one hematozoan, H. nettionis.

Low prevalences of individual hemosporid species precluded statistical analysis
of the effects of host-intrinsic factors on prevalence data within each host species.
However, using overall hematozoan prevalence, comparisons between waterfowl species

could be made and no differences were found (after removing black duck data because of

24

aoraoae-ee<u>m< .aaaosmim ._83n.:
32 .w 3200a Lopeosaomu.
32 .vm mama: 83...

 

 

 

 

Na E m a o E : em as 58.
cm a e m o a 2 S _ case 825
N N N o o o o o o can: v.26 £32653:
8N § a R o 2 _ NA 2 R 2232
R o o o o R mm o a. 38m «ago
m N _ _ o _ _ o o seas seam
Eon Bossm :2 E a 393m >5. E S macaw
:3— inseam

 

.mmM 88m BEES common 23 .owe .86on .3 REESE: we c3832 .H 033.

25

Table II. Prevalence of hematozoa in KBS summer waterfowl.

 

Haemoproteus Haemoproteus Leucocytozoon

 

 

Bird species N" All hematozoa nettionis greineri simondi
Black duck l 0.0 % (0)1 0.0 % (0) 0.0 % (0) 0.0 % (0)
Canada goose 77 3.9 % (3) 3.9 % (3) 0.0 % (0) 0.0 % (0)
Mallard ' 119 6.7 % (8) 4.2 % (5) 0.0 % (0) 3.4 % (4)
Wood duck 21 4.8%(1) 4.8%(1) 4.8%(1) 0.0%(0)
Total 218 5.5 %(12) 4.1%(9) 0.5%(1) 1.8%(4)
*Number of birds examined

'l'Prevalence (Number of birds infected)

Table 111. Overall hematozoan prevalence by age class in KBS summer waterfowl.

 

 

 

Waterfowl age class N“ All hematozoa

Lt 123 0 % (0)1

HY 24 8.3 % (2)

AHY 71 14.1 % (10)

All classes 218 5.5 % (12)
‘Number of birds examined

TL=Local, HY=Hatch-year, AHY=After-hatch-year
IPrevalence (Number of birds infected)

26

small sample size, chi-square analysis, X2 = 0.8, P > 0.05). Therefore, data from different
waterfowl species could be combined and comparisons between waterfowl age and sex
classes could be made statistically. No L waterfowl were infected with any blood
parasite (Table III). Overall hematozoan prevalence did not differ significantly between
HY and AHY age classes (X2 = 2.1, P > 0.05) or between sexes (X2 = 0.4, P > 0.05).

Mean intensities of hemosporids are in Table IV. The low prevalences of
hematozoans precluded statistical analysis of the effects of host-intrinsic factors on
intensity data within each host species. Mean intensities of H. nettionis (the only parasite
species to infect multiple host species) did not differ between waterfowl species
(Kruskall-Wallis test, W = 1.9, P > 0.05). When waterfowl species were combined, host-
intrinsic factors were analyzed using intensity data, but no significant differences were
found for mean intensities of any parasite between waterfowl ages or sexes (Mann-
Whitney U-tests, U= 1.0, 6.0, P > 0.05).

Thirty-five captive waterfowl of native and exotic species were sampled for blood
parasites during the summer (Table V). None were found to be infected with any blood

parasites.

Hematozoa of KBS fall waterfowl

The same three hemosporid species were found in waterfowl sampled in the fall
as in summer waterfowl (Table VI). Four dual infections occurred in mallards; two were
infected with H. nettionis and H. greineri and two were infected with H. nettionis and L.
simondi. In addition, one mallard was infected with three parasites, H. nettionis, L.
simondz', and an unidentified microfilaria Fall waterfowl had an overall hematozoan
prevalence of 11.4 %, about 2X that of summer waterfowl (5.5 %), but this difference

was only marginally significant (X2 = 2.9, P = 0.09). Prevalences of the three hemosporid

27

Table IV. Mean intensity of hemosporids from KBS summer waterfowl.

 

 

Waterfowl species Haemoproteus nettionis Haemoproteus greineri Leucocytozoon simondi
Canada goose 4.7122“ —§ —
(2-9)t
31
Mallard 52:1 .6 — 0.91:0.1
(1-10) (0.5-1)
5 4
Wood duck 1.0 0.5 —
(1) (0.5)
l 1
Total 4.6il.l 0.5 0.9101
(1-10) (0.5) (0.5-1)
9 1 4

 

"Mean intensity t standard error of hemosporids/5,000 erythrocytes
t(Range)

INumber of birds infected

§None infected

28

Table V. Species and number of captive waterfowl sampled from the Kellogg Bird
Sanctuary.

 

Species Number sampled

 

Black duck-northem pintail hybrid 2
Black swan 5
Domestic mallard 6
Domestic greylag goose 1
Mallard 1
Mute Swan 4
Trumpeter Swan 16
Total 35

 

29

688%: E3 .«o Lona—52v ooze—32$

cos—Swan 3.5 me 39:52...

 

 

E .x. 3 8V .x. 3. S e. 2 as s 3. G: a. e: 3. .28.

aged 83.5... 3.x. 2. Ass: 5.x. 2. a 3.2% .83

§ .x. a... SO .x. 3. S .x. 98 é a. 3. S as 9% N 252 as £83653:

8 .x. 3 5 a. o.“ 5 .x. a. 5 .x. 2 8: e. 2 2: 3%:

SO .x. as 8 e. eon 8V e. 3. so a. a... .2: a. in N v.26 v.85

ova—506:2 E225“. .thoxm haeSoz «80882— =< ..2 360% Em
coonoioeoaen maocouuesoot Eoceiefioo:

 

438.583 =£ mmv— S 83380: we coco—92m ._> 2me

30

species were comparable; Leucocytozoon simondi, H. nettionis, and H. greineri infected
5.3 %, 4.4 %, and 3.5 % of waterfowl, respectively.

Low prevalences of individual hemosporid species precluded statistical analysis
of the effects of host-intrinsic factors on prevalence data within each host species. Using
prevalences of all hematozoa combined, comparisons between species could be made
statistically and no significant differences were found (X2 = 6.12, P > 0.05). By
combining host species data, statistical comparisons between waterfowl age and sex
classes could be made. Significant differences were not found for overall prevalence
between age or sex classes (X2 = 0.3, 0.6, P > 0.05).

Regression analysis showed a significant linear relationship between waterfowl
<3 years of age (ages 2 3 were removed because of small sample sizes) and overall
hematozoan prevalence (r2 = 0.75, P = 0.00000). A similar significant relationship
existed when ages 2 3 were added to the analysis (r2 = 0.37, P = 0.00000; Figure 4). The
boxed insert in Figure 4 lists the number of birds sampled for each age.

Low prevalences precluded statistical analysis of the effects of host-intrinsic
factors on fall intensity data within each host species (Table VII). Low prevalences also
precluded statistical analysis to test for differences in mean intensity between host
species. When all waterfowl species were combined, no significant differences were
found for parasite mean intensities between waterfowl ages or sexes (U = 0.5, 7.0, P >
0.05). Mean intensity of H. nettionis was significantly higher in the summer than in the
fall (U = 47.0, P = 0.01678). No differences in mean intensities existed between seasons

for H. greineri or L. simondi (U = 0.0, 15.0, P > 0.05).

31

Figure 4. Relationship between age and overall hematozoan prevalence in KBS
waterfowl. Boxed insert in upper left lists the number of waterfowl sampled in each age
group. In the lower right is the regression equation and coefficient.

32

S

3 Q

 

 

3.2.0 n at
CNN... + £8: u A

 

 

 

 

OPCD
v-v-e-

VOIDMPVCONV-Pv-
P

8‘3»

..

I-N
O'O'wNDV'IDONO

 

 

 

l

l

6 Saw?—

 

.o
.2
.ON
0
A
3
.3 W
n.
.m.
.8 n
m
now u
u
m
.8 m
w
9
u
.oe a
m
.8
.8
.2:

33

Table VII. Mean intensity of hemosporids in KBS fall waterfowl.

 

 

 

Waterfowl species Haemoproteus nettionis Haemoproteus greineri Leucocytozoon simondi
Black duck —" — 0.5
(0.5)
l
Mallard 1310.21 4.0:t0.0 0.8:0.1
(0.5-2): (4) (0.5-l)
8§ 2 5
Mallard-black duck — 1.0 —
hybrid (I)
1
Wood duck — 1.0 —
(I)
1
Total l.3:t0.2 2010.7 0.8101
(0.5-2) (1-4) (0.5-1)
8 4 6
*None infected

' tMean i standard error of hemosporids/5,000 erythrocytes

1(Range)
§Number infected

34

Hematozoa of Douglas Lake common mergansers

Three species of hemosporids were identified, H. greineri, L. simondi, and
Plasmodium circumflexum with prevalences of 50.9 %, 47.3 %, and 3.6 %, respectively
(Table VIII). Overall hematozoan prevalence was 74.6 %. Thirteen common mergansers
(23.6 %) were infected with H. grez'neri and L. simondi. Two (3.6 %) dual infections
with H. greineri and P. circumflexum occurred. Prevalences did not differ between sexes
for overall hematozoan prevalence or for prevalences of each parasite species individually
(X2 = 0.005-2.2, P > 0.05). Plasmodium circumflexum prevalences were not analyzed
because only two birds were infected.

Mean intensities of H. greineri and L. simondi were not significantly different (U
= 443.5, P > 0.05; Table IX). However, H. greineri mean intensity in single infections
(H. greineri only) was significantly greater than H. greineri mean intensity in dual
infections with L. simondi (U = 123.5, P = 0.03954). Leucocytozoon simondi mean
intensities did not differ between single and dual infections with H. greineri (U = 80.0, P
> 0.05). One bird infected with L. simondi had an intensity of 78/5 ,000 erythrocytes,
much higher than any other bird. The difference in H. greineri intensity between single
and dual infections with L. simondi remained the same whether this outlying data point
was removed or retained in the analysis. The statistical tests reported above and Table IX
include the outlier.

Mean intensities of either H. greineri or L. simondi did not differ significantly
between host sexes (U = 74.5-95.0, P > 0.05). Plasmodium circumflexum mean intensity
was l.25/5,000 erythrocytes, with a range of 0.5-2, and was excluded from statistical
analysis because of the low number of birds infected.

Compared to overall hematozoan prevalence at KBS, Douglas Lake overall
prevalence was significantly higher (X2 = 149.8, P = 0.00000). When analyzed

35

Table VIII. Prevalence of hematozoans in 55 L Douglas Lake common mergansers.

 

 

Haemoproteus Leucocytozoon Plasmodium
All hematozoa greineri simondi circumflean
74.6 % (41)"' 50.9 % (28) 47.3 % (26) 3.6 % (2)

 

*Prevalence (Number of birds infected)

36

Table IX. Mean intensity of single and dual infections in Douglas Lake common mergansers.

 

 

Infection type Haemoproteus greineri Leucocytozoon simondi

Single 4011.1" 7.4:5 .9
(0.5-13)? (0.5-78)

131 13
Dual . l.5:1:0.4 1.6:1:0.5
(0.5-5) (0.5-6)

l3 13
Total 29:06 4.51:3.0
(0.5-13) (0.5-78)

26 26

 

*Mean 1: standard error of hemosporids/5,000 erythrocytes

1'(Range)
INumber of birds infected

37

individually, parasite species common to both locations (H. greineri and L. simondi) had
significantly higher prevalences at Douglas Lake (X2 = 147.6, H. greineri; X2 = 109.6, L.
simondi; P = 0.00000). Mean intensifies of these two common species did not differ
between locations (U = 4.5, 53.5, P > 0.05).

Sentinel ducklings

None of the 186 ducklings used as sentinels at KBS became infected with any
blood parasite. Of eighteen sentinels resampled on August 31, none were infected with
any blood parasite.

Vector collections

Bennett traps caught only two species of mosquito, Coquillittidia perturbans and
Culex tarsalis, capturing 310 and 89 individuals respectively. Engorged and unengorged
individuals of both species were captured. Neither biting midges or black flies were
captured nor ever seen feeding on KBS waterfowl.

38

DISCUSSION
Kellogg Biological Station

The three hemosporid species found in KBS waterfowl have been previously
reported from northern Michigan waterfowl (O’Roke, 1931, 1934; Herman et al., 1975,
Desser et al., 1978; Sibley and Werner 1984) but not from southern Michigan waterfowl.
Thus, the present study is the first survey of waterfowl at KBS and in southern Michigan.
Sibley and Werner (1984) did not survey wild waterfowl but did report natural
transmission of H. nettionis to domestic ducks in the upper peninsula of Michigan. They
also noted gametocytes which completely encircled the host cell nucleus in addition to
halteridial gametocytes, but identified all of them as H. nettionis, since H. greineri had
not yet been described (described in Bennett et al., 1984). Therefore, it can be suggested
that H. greineri also was transmitted in their study. Bennett et al. (1984) stated that H.
greineri was endemic to the prairie regions of Canada, but my analysis of Sibley and
Wemer’s (1984) results show that the transmission range of this parasite also includes the
upper peninsula of Michigan. In addition, a recent study extended the known
transmission range of H. greineri to include eastern Canada and northeastern United
States (Pung et al., 1997). In the present study, none of the birds infected with H.
grez'neri were local, so it is not clear if southern Michigan is part of the transmission
range of this parasite.

Compared to other surveys of Michigan waterfowl, prevalences of hematozoans
in the KBS area were low. Herman et a1. (1975) reported L. simondi as the causative
agent of Canada geese gosling mortality at the Seney National Wildlife Refuge in the
upper peninsula. They reported prevalence as 80 % in AHY geese just prior to egg-

39

laying and 100 % in goslings each year. They also reported the presence of H. nettiom's,
Plasmodium spp., and trypanosomes, but stated that prevalences were low compared to L.
simondi. O’Roke (1931, 1934) found that 100 % of black ducks and mallards were
infected in the Douglas Lake vicinity and 30 % of black ducks were infected in the
Munuscong Lake vicinity, which lies between Lake Huron and Lake Superior.

Compared to other North American studies of waterfowl hematozoa, the overall
prevalences of hematozoans in the KBS area in the summer (5.5 %) and fall (11.4 %)
were low. Although a wide range of overall hematozoan prevalences has been reported
for different localities in North America (0-100 %), similarly low prevalences have been
reported from only a few localities during the summer and early fall. Burgess (1957)
reported only one hematozoan species, L. simondi, infecting <1 % of waterfowl from
several areas in Saskatchewan and Manitoba. An overall hematozoan prevalence of 3 %
was reported in snow and Canada geese from the Northwest Territories (Bennett and
MacInnes, 1972). Blood parasites were not found in waterfowl from eastern
Newfoundland (Bennett and Inder, 1972; Bennett et al., 1991). Bennett et al. (1982)
reported an overall prevalence of 9 % fiom one region in Saskatchewan. Thul et al.
(1980) reported an overall prevalence of <1 % in wood ducks from southeastern states
and O’Dell and Robbins (1994) reported an overall prevalence of 10 % in wood ducks in
Missouri. Most authors attribute low prevalences during the summer and early fall
months to an absence or low abundance of vectors. In eastern Newfoundland, the
absence of vectors is attributed to the steady easterly winds that blow off the Atlantic
Ocean (Bennett and Inder, 1972; Bennett et al., 1991). Other authors do not explain the
absence Or low abundance of vectors.

Other North American studies have reported higher prevalences (>25 %) in
waterfowl during the summer and early fall in the following regions: northeastern U.S.
(Nelson and Gashwiler, 1941; O’Meara, 1956; Herman et al., 1971; Bennett et al., 1974;
Thul et al., 1980), maritime Canadian provinces (Bennett et al., 1975; Bennett et al.,

40

1991), Alberta and Saskatchewan (Williams et al., 1977; Bennett et al., 1982), Northwest
Tenitories (Williams et al., 1977).

A few other studies have reported an absence (Michot et al., 1995) or low
prevalence ($16 %; Levine and Hanson, 1953; Bradshaw and Trainer, 1966; Kocan et al.,
1979; Fedynich et al., 1993) in migrating waterfowl during the late fall and in wintering
waterfowl. However, low prevalences in late fall and winter are expected because
infections can go undetected due to low intensities (F edynich et al., 1993) whereas
infections are expected to be more readily detected during the summer and early fall
because intensities are higher.

In the avian blood parasite literature, many authors compare survey results to the
summary regional data in Greiner et al. (1975a), who arbitrarily divided North America
into seven topographical regions which varied in overall prevalence of avian hematozoa
as well as in the relative fi'equencies of the three hemosporid genera. The Great Lakes
area was included in a region which had a moderately high overall prevalence (38 %;
ranked third amongst the seven regions) and prevalences of Leucocytozoon spp.,
Haemoproteus spp., and Plasmodium spp. were 23 %, 18 %, and 4 %, respectively.
Prevalences at KBS were low comparatively, but the summary regional data of Greiner et
al. (1975a) is for all bird families combined, and therefore is of only limited benefit in
comparison to data from the Anatidae only.

Greiner et al. (1975a) also summarized prevalence data by family and species, and
reported that the overall hematozoan prevalence for north american Anatidae was 31.9 %
with prevalences of L. simondi, Haemoproteus spp., and Plasmodium spp. 15.5 %, 19.5
%, and” 2.9 %, respectively. The overall prevalence for mallards was 27 % with
prevalences of L. simondi, Haemoproteus spp., and Plasmodium spp. 11.3 %, 15.6 %, and
1.7 %, respectively. For wood ducks, the overall prevalence was 55.4 %, with
prevalences of L. simondi, Haemoproteus spp., and Plasmodium spp. 21.8 %, 49.9 %, and
5.3 %, respectively. For Canada geese, the overall prevalence was 6.5 %, with

41

prevalences of L. simondi, Haemoproteus spp., and Plasmodium spp. 4.0 %, 2.1 %, and
0.2 %, respectively. Except for Canada geese, KBS prevalences are low compared to
these summary data. However, the Canada geese data in Greiner et al. (1975a) may be
biased by location and do not include any data from the Michigan upper peninsula, where
Canada geese have very high prevalences of hematozoa (Herman et al., 1975).

Species richness values of the KBS hemosporid community in both summer (3)
and fall (3) were moderate compared to other studies of North American waterfowl. The
number of hemosporid species reported ranges from zero to six and averages 2.6 (my own
analysis of 22 of the above Studies). '

Intensities of hemosporids in KBS summer waterfowl were low (Table IV), an
expected result because low intensities are indicative of chronic infections (Allan and
Mahrt, 1989) and 10 of the 12 birds infected were adults. It is possible that these infected
birds had earlier suffered spring relapses, but sampling occurred late enough in the
summer that intensities were probably returning to chronic levels. The difference
between seasons in the mean intensity of the most common parasite, H. nettionis (Tables
IV and VII) is expected since fall intensities should be close to the chronic infection state
and summer intensities should be somewhere between the extremely high intensities
caused by spring relapse and the low intensities of chronic infection. The lack of a
seasonal difference for H. greineri or L. simondi mean intensities may be due either to the
small number of birds infected or that birds infected with these parasites were captured
toward the end of the srunmer period.

Most authors only give rough estimates of hemosporid intensity in wild
waterfowl, without quantifying intensity per number of counted erythrocytes. For
example, many investigators estimate intensity by counting the number of parasites per
minutesexpended examining the smear. These kinds of estimates are not regarded as
accurate by some researchers (Godfi'ey et al., 1987; Fedynich et al., 1995), but broad

comparisons can be made to the present study. O’Roke (1934) and Herman et al. (1975)

42

noted high intensities of L. simondi, much higher than mean intensities in KBS
waterfowl, during the breeding season. Estimates given for summer waterfowl sampled
after breeding are similar to the H. nettionis mean intensity for KBS summer waterfowl
(Herman et al., 1971; Williams et al., 1977; O’Dell and Robbins, 1994). Estimates given
for wintering waterfowl (Trainer et al., 1962; Polcyn and Johnson, 1968; Kocan and
Knisley, 1970) are lower than H. nettionis mean intensity for KBS summer waterfowl and
are similar to or slightly lower than that for KBS fall waterfowl. Only Fedynich et al.
(1993) quantified intensity of waterfowl hemosporids using methods similar to mine, and
they report a H. nettionis mean intensity of waterfowl wintering in Texas (1.6iO.1, mean
i standard error of gametocytes/5,000 erythrocytes) which is similar to H. nettionis mean
intensities in KBS fall waterfowl (l .2:l:0.3), but lower than in KBS summer waterfowl.

The component community of hematozoa in KBS waterfowl can be characterized
as depauperate, having moderate species richness but low prevalences and intensities.
Therefore, the metapopulations of each parasite species appear to be small. This is a
somewhat surprising finding given the large, high-density waterfowl populations and the
abundance and variety of aquatic habitats suitable for vector larvae in the KBS area. The
low prevalences and intensities and the low number of multiple infections suggest that
species interactions are few or non-existent. Thus the component community of
hemosporids in KBS waterfowl can also be characterized as isolationist. Similarly,
Fedynich et 'a1. (1993) classified the hemosporid community of waterfowl wintering in
south Texas as isolationist based on low prevalences, low intensities, and low number of
multiple infections.

Parasite transmission is critical to the structure of metapopulations and component
communities. One expects low transmission rates in an isolationist community, because
they lead to unsaturated, nonequilibrium communities that provide few opportunities for
interspecific interactions (Holmes and Price, 1986). There is strong evidence from my

study to indicate that transmission does not occur in the KBS area. First, hematozoans

43

were absent from 123 wild L waterfowl sampled. Secondly, 35 captive waterfowl, which
are year round residents of the Kellogg Bird Sanctuary, were hematozoan flee. Finally,
none of the 186 sentinel ducks exposed over the course of the summer became infected
With any hematozoan. The five week exposure period should have allowed ample time
for the development of hemosporids since the longest time required for patency in ducks
is 14-19 days for Plasmodium spp. (Meyer et al., 1974).

Transmission appears not to occur because of a lack of vectors. While Culex
tarsalis is known to be a vector of P. relictum (see Work et al., 1990), no Plasmodium
spp. were found in KBS waterfowl. Coquillittidia perturbans is not known to vector any
avian blood parasite.

It is not surprising that black flies were never seen or captured feeding on
waterfowl because their larval habitat requirements are restricted to swift, oxygenated
streams which are uncommon in the KBS area. In fact, Kocan and Clark (1966a, 1966b)
housed their experimental waterfowl at KBS because L. simondi was assumed not to be
transnritted there. I did find black fly larvae near the outlet of Wintergreen Lake, but
these were present only at the beginning of the summer. The site at which I found them
was an unnatural one, where water was rushing over a recently installed water level
control gate, which had not existed in previous years (W. C. Johnson, pers. comm).
Although I did not attempt to identify these larvae, they were likely Simulium decorum, a
species that commonly colonizes lake outlets (Crosskey, 1990; E. D. Walker, pers.
comm).

It is surprising that no ceratopogonids were ever seen or captured feeding on
waterfowl, since their larval habitats are very diverse. While waterfowl-feeding species
were not present, it is possible that other ceratopogonids were present at KBS. No other
surveys have been done of hematophagous insects in the KBS area (B. D. Walker, pers.

comm).

The only host-intrinsic factor for which any significant differences were found in
KBS waterfowl was age. This was due to the lack of transmission in the KBS area
Infected AHY waterfowl breeding in the KBS area probably did not obtain hemosporid
infections in the KBS area, but instead may have obtained them during migration or on
wintering areas. It is known that some of the KBS waterfowl winter as far south as
Missouri and Tennessee (Kellogg Bird Sanctuary banding records) where transmission
may occur in the early fall or begin in the spring before the birds leave on their northward
migrations. The high vagility of waterfowl probably plays a role in the structure of their
hematozoan communities.

The significant relationship between overall prevalence and waterfowl age in
years was not unexpected (Figure 4). It is known that birds can retain hematozoan
infections over many years and it is believed that eradication of the parasite by host
immune response is uncommon (Allan and Mahrt, 1989). The gradual increase in
prevalence is probably the result of highly vagile birds visiting areas of transmission from
time to time.

Although it was not statistically significant, season had an effect on the overall
prevalence of hematozoa, with fall overall prevalence being 2X that in summer (Tables I]
and VI). Many of the waterfowl sampled in the fall were migrants from outside the KBS
area, because they did not possess bands when captured. However, because none of the
captured birds possessed bands, it was not possible to tell from how far away their
breeding sites were. It is believed, however, that most of them were from other areas in
southern Michigan (W. C. Johnson, pers. comm). The increase in prevalence in fall
waterfowl then, may indicate that hematozoan transmission occurs in Other areas in

southern Michigan.

45

Douglas Lake

Two of the hemosporid species identified from Douglas Lake common
mergansers, H. greineri and L. simondi, were also found in KBS summer waterfowl, but
prevalences were much higher at Douglas Lake. Overall hematozoan prevalence was also
much higher. The high prevalence of L. simondi is consistent with the high prevalences
previously found at Douglas Lake (O’Roke 1931, 1934) and in other areas of the northern
lower and upper peninsulas of Michigan (O’Roke, 1934; Herman et al., 1975).
Prevalences in these parts of Michigan are higher than in most parts of North America,
but are consistent with the range of prevalences (58-79 %) reported from the northeastern
United States and eastern Canada (Nelson and Gashwiler, 1941; O’Meara, 1956; Bennett
et al., 1971; Bennett et al., 1974; Thul et al., 1980; Bennett et al., 1991).

Haemoproteus greineri has not been reported in waterfowl from Douglas Lake
before and common merganser is a new host record for H. greineri. The presence of H.
greineri in L birds is consistent with my analysis of Sibley and Werner’s (1984) results
and shows that the transmission range of this parasite is not limited to the prairie
provinces of Canada. Plasmodium circumflexum occurred at a low prevalence at Douglas
Lake, but did not occur in KBS waterfowl. Common merganser is a new host record for
P. circumflexum and this is the first record of any Plasmodium spp. in common
merganser. Greiner et al. (197 5a) include summary data on the hematozoa of common
mergansers and report overall hematozoan prevalence as 41.2 % (14 of 34 infected), with
prevalences of L. simondi and Haemoproteus spp. 8.8 % and 35.3 %, respectively.

Since only L (approximately six weeks old) common mergansers were captured,
the high prevalences of H. greineri and L. simondz‘ indicate that transmission of these
species is high, a striked contrast to the KBS area. High transmission rates should lead to
more saturated, equilibrium communities with more opportunities for interspecific

interactions (Holmes and Price, 1986).

46

Although species richness of hemosporids was moderate and not higher than for
KBS waterfowl, there is evidence that the hemosporid community of Douglas Lake
common mergansers is more interactive. Mean intensity of H. greineri in single
infections was significantly greater than H. greineri mean intensity in dual infections with
L. simondi (Table IX), suggesting that there is some interaction between H. greineri and
L. simondi. The converse relationship did not exist for L. simondi mean intensities. It is
possible that L. simondi has a competitive effect on H. greineri. This effect could be due
to direct competition, such as competition for host cells, at either the blood stage or the
tissue stage of the parasites. On the other hand, the effect may not be the result of direct
competition, but instead may be immuno-mediated. Perhaps L. simondi evokes a host
immune response that is detrimental to H. greineri. Another important factor may be
which parasite infects the host first, which may depend upon vector emergence periods.
Other investigators have noted the potential for interaction between hemosporid species
(e.g., Fedynich et al., 1993), but none have reported evidence of interactions. Because of
the small sample size in the present study, further investigation is needed to confirm a
competitive effect.

High rates of transmission of L. simondi to both captive (O’Roke, 1931, 1934;
Chernin, 1956a; Barrow et al., 1968) and wild waterfowl (O’Roke, 1931, 1934) have
been previously documented at Douglas Lake. O’Roke (1931, 1934) found 100 % of
black ducks and mallards (including L and AHY) from Douglas Lake to be infected with
L. simondi. O’Roke (1931, 1934) believed that wild black duck broods on Douglas Lake
were significantly reduced in brood size by mortality caused by L. simondi. More
recently, based on sampling one brood of L mallards every other year since 1980, H. D.
Blankespoor (pers. comm.) believes the prevalence of L. simondi in mallards on Douglas
Lake is close to 100 % and has noticed that brood sizes appear to be reduced by L.
simondi. Before the present study, it was not known whether L common mergansers

from Douglas Lake were infected with L. simondi, but it had been noted that L common

47

mergansers do not suffer the high rate of mortality seen in mallards (H. D. Blankespoor,
pers. comm). It is not known why L. simondi might be less pathogenic in common
mergansers. Diet may be a factor. Common mergansers eat fish almost exclusively, a
higher protein and iron source than the invertebrate and vegetation diet of mallards
(Bellrose, 1980). Iron may be an especially important mineral in combating anemia,
which can be a cause of death in mallards infected with L. simondi (Kocan and Clark,
1966b). Another possibility is that common mergansers are less frequently attacked by
black flies than are mallards, and therefore do not acquire as many infective sporozoites.

High transmission of L. simondi and H. greineri indicates abundance of Simulium
black flies and Culicoides biting midges, respectively. Simulium rugglesi is the vector of
L. simondi at Douglas Lake (O’Roke, 1931, 1934; Barrow et al., 1968). Barrow et al.
(1968) also reported Culicoides spp. feeding on waterfowl at Douglas Lake, but did not
identify which species were present. Although it was low, transmission of P.

circumflexum indicates the presence of Culex spp. or Culiseta spp. vectors.

48

CONCLUSIONS

Three hemosporid species were found in KBS waterfowl: Haemoproteus nettionis
was the most common; H. greineri and Leucocytozoon simondi were less common.
Despite large, high density populations of waterfowl, the hemosporid community of KBS
waterfowl has moderate species richness, has low prevalences and intensities, and can be
characterized as isolationist. Lack of transmission in the KBS area, due to the absence or
. low abundance of vectors, is responsible for this isolationist character. Overall
hematozoan prevalence at KBS was only 5.5 % in summer waterfowl and 11.4 % in fall
waterfowl. The higher prevalence in fall birds may indicate that transmission occurs in
other areas of southern Michigan, since many birds sampled in the fall were migratory
and had been breeding in other areas.

In contrast, transmission appears to be high at Douglas Lake in northern lower
Michigan, resulting in a much higher overall hematozoan prevalence (74.6 %) and more
multiple infections. Interactions between hemosporid species would be more likely in a
community with such patterns. Mean intensities in Douglas Lake common mergansers
were low and similar to those in KBS waterfowl. A possible interaction between H.
greineri and L. simondi is suggested by a significantly lower mean intensity of H.
grez'neri in single infections than in double infections with L. simondi.

Haemoproteus greineri and L. simondi were common, and Plasmodium
circumflexum was uncommon in Douglas Lake common mergansers. The transmission
of H. greineri to L common mergansers at Douglas Lake indicates that the range of H.
greineri, previously thought to be restricted to the prairie provinces of Canada, includes
northern Michigan. Common merganser is a new host record for H greineri and P.

circumflexum.

49

Breeding waterfowl in northern lower Michigan are similar to those in the
northeastern United States and eastern Canada in having high hematozoan prevalences.
Hematozoan prevalences in southern Michigan waterfowl are lower than in many areas of

North America.

50

BIBLIOGRAPHY

Allan, R. A., and J. L. Mahrt. 1989. Influence of transmission period on primary and

relapse patterns of infection of Leucocytozoon spp. and Haemoproteus mansoni.
The American Midland Naturalist 121: 341-349.

Atkinson, C. T. 1986. Host specificity and morphometric variation of Haemoproteus
meleagridis Levine, 1961 (Protozoa: Haemosporina) in gallinaceous birds.
Canadian Journal of Zoology 64: 2634-2638.

Atkinson, C. T., and C. van Riper III. 1991. Pathogenicity and epizootiology of avian
haematozoa: Plasmodium, Leucocytozoon, and Hacmoproteus. In Bird-parasite
interactions: Ecology, evolution, and behaviour, J. E. Loye and M. Zuk (eds).
Oxford University Press, Oxford, England p. 19-48.

Barrow, Jr., J. H., N. Kelker, and H. Miller. 1968. The transmission of Leucocytozoon
simondi to birds by Simulium rugglesi in northern Michigan. The American
Midland Naturalist 79: 197-204.

Bellrose, F. C. 1980. Ducks, geese, and swans of North America. Stackpole Books,
Harrisburg, Pennsylvania, 540 p.

Bennett, G. F. 1960. On some ornithophilic blood-sucking Diptera in Alogonquin Park,
Ontario, Canada. Canadian Journal of Zoology 38: 377-389.

Bennett, G. F. and J. G. Inder. 1972. Blood parasites of game birds from insular
Newfoundland. Canadian Journal of Zoology 50: 705-706.

Bennett, G. F., and C. D. MacInnes. 1972. Blood parasites of geese of the McConnell
River, N.W.T. Canadian Journal of Zoology 50: 1-4.

Bennett, G. F ., and M. A. Pierce. 1988. Morphological form in the avian
Haemoproteidae and an annotated checklist of the genus Haemoproteus Kruse,
1890. Journal of Natural History 22: 1683-1696.

Bennett, G. F., and D. Squires-Parsons. 1992. The leucocytozoids of the avian families
Anatidae and Emberizidae 5.1., with descriptions of three new Leucocytozoon
species. Canadian Journal of Zoology 70: 2007-2014.

Bennett, G. F., M. A. Bishop, and M. A. Pierce. 1993. Checklist of the avian species of
Plasmodium Marchiafava & Celli, 1885 (Apicomplexa) and their distribution by
avian family and Wallacean life zones. Systematic Parasitology 26: 171-179. .

Bennett, G. F., W. Blandin, H. W. Heusmann, and A. G. Campbell. 1974. Hematozoa of
the Anatidae of the Atlantic Flyway. 1. Massachusetts. Journal of Wildlife '
Diseases 10: 442-451.

51

Bennett, G. F., D. J. Nieman, B. Turner, E. Kuyt, M. Whiteway, and E. C. Greiner. 1982.
Blood parasites of prairie anatids and their implication in waterfowl management
in Alberta and Saskatchewan. Journal of Wildlife Diseases 18: 287-296.

Bennett, G. F., M. A. Pierce, and R. A. Earle. 1994. An annotated checklist of the valid
avian species of Haemoproteus, Leucocytzoon (Apicomplexa: Haemosporida) and
Hepatozoon (Apicomplexa: Haemogregarinidae). Systematic Parasitology 29:
61-73.

 

Bennett, G. F., A. D. Smith, W. Whitman, and M. Cameron. 1975. Hematozoa of the
Anatidae of the Atlantic F lyway. II. The Maritime Provinces of Canada. Journal
of Wildlife Diseases 11: 280-289.

Bennett, G. F., V. D. Stotts, and M. C. Bateman. 1991. Blood parasites of black ducks
and anatids from Labrador and insular Newfoundland. Canadian Journal of
Zoology 69: 1405-1407.

Bennett, G. F., B. Turner, and M. Whiteway. 1984. Avian Haemoproteidae. l8.
Haemoproteus greineri, a new species of haemoproteid from the waterfowl family
Anatidae. Canadian Journal of Zoology 62: 2290-2292.

Bradshaw, J. E., and D. 0. Trainer. 1966. Some infections diseases of waterfowl in the
Mississippi flyway. Journal of Wildlife Management 30: 570-576.

Burgess, G. D. 1957. Occurrence of Leucocytozoon simondi M. & L. in wild waterfowl
in Saskatchewan and Manitoba. Journal of Wildlife Management 21: 99-100.

Chernin, E. 1956a. The epizootiologyl of Leucocytozoon simondi infections in domestic
ducks in northern Michigan. The American Journal of Hygiene 56: 39-57.

Chernin, E. 1956b. The relapse phenomenon in the Leucocytozoon simondi infection of
the domestic duck. The American Journal of Hygiene 56: 101-118.

Clayton, D. H. 1991. The influence of parasites on host sexual selection. Parasitology
Today 7: 329-334.

Crosskey, R. W. 1990. The natural history of blackflies. John Wiley and Sons Ltd.,
West Sussex, England, 711 p.

Desser, S. S., and A. K. Ryclonan. 1976. The development and pathogenesis of
Leucocytozoon simondi in Canada and domestic geese in Algonquin Park,
Ontario. Canadian Journal of Zoology 54: 634-643.

Desser, S. S., J. Stuht, and A. M. Fallis. 1978. Leucocytozoonosis in Canada geese in
upper Michigan 1. Strain differences among geese from different localities.
Journal of Wildlife Diseases 14: 124-131.

Esch, G. W., A. W. Shostak, D. J. Marcogliese, and T. M. Goater. 1990. Patterns and
processes in helminth parasite communities: an overview. In Parasite

communities: patterns and processes, G. W. Esch, A. 0. Bush, and J. M. Aho
(eds), Chapman and Hall, New York, 335 p.

52

Fallis, A. M., and D. M. Wood. 1957. Biting midges (Diptera: Ceratopogonidae) as
intermediate hosts for Haemoproteus of ducks. Canadian Joumal of Zoology 35:
425-435.

Fallis, A. M., D. M. Davies, and M. A. Vickers. 1951. Life history of Leucocytozoon
simondi Mathis and Leger in natural and experimental infections and blood
changes produced in the avian host. Canadian Journal of Zoology 29: 305-328.

Fedynich, A. M. and O. E. Rhodes, Jr. 1995. Hemosporid (Apicomplexa, Hematozoa,
Hemosporida) community structure and pattern in wintering wild turkeys. Journal
of Wildlife Diseases 31. 404-409.

Fedynich, A. M., D. B. Pence, and R. D. Godfrey, Jr. 1993. Hemosporids
(Apicomplexa, Hematozoa, Hemosporida) of Anatids from the Southern High
Plains of Texas. Journal of the Helminthological Society of Washington 60: 35-
38.

Fedynich, A. M., D. B. Pence, and R. D. Godfrey, Jr. 1995. Letter to the
editor. . .Hematozoa in thin blood smears. Journal of Wildlife Diseases 31: 43 5-
438.

Garnham, P. C. C. 1966. Malaria parasites and other haemosporidia. Blackwell
Scientific Publications, Oxford, 1114 p.

Godfrey, R. D., Jr., A. M. Fedynich, and D. B. Pence. 1987. Quantification of
hematozoa in blood smears. Journal of Wildlife Diseases 23: 558-565.

Godfrey, Jr., R D., D. B. Pence, and A. M. Fedynich. 1990. Effects of host and spatial
factors on a haemoproteid community in mornning doves fi'om western Texas.
Journal of Wildlife Diseases 26: 435-441.

Greiner, E. C., and G. F. Bennett. 1975. A color pictorial guide to some species of
Haemoproteus, Leucocytozoon, and Trypanosoma. Journal of Wildlife Diseases,
supplement on microfiche, 66.

Greiner, E. C., G. F. Bennett, E. M. White, and R. F. Coombs. 1975a. Distribution of the
avian hematozoa of North America Canadian Journal of Zoology 53: 1762-1787.

Greiner, E. C., G. F. Bennett, M. Laird, and C. M. Herman. 1975b. Avian taxonomic
keys and color pictorial guide species of Plasmodium. Journal of Wildlife
Diseases, supplement on microfiche, 68.

Hamilton, W. D., and M. Zuk. 1982. Heritable true fitness and bright birds: A role for
parasites? Science 218: 384-387.

Herman, C. M. 1968. Blood parasites of North American waterfowl. Transaction of the
North American Wildlife Conference 33: 348-358.

Herman, C. M., J. H. Barrow, Jr., and I. B. Tarshis. 1975. Leucocytozoonosis in Canada
geese at the Seney National Wildlife Refuge. Journal of Wildlife Diseases 11:
404-41 1.

Herman, C. M., J. O. Knisley, Jr., and G. D. Knipling. 1971. Blood parasites of wood
ducks. Journal of Wildlife Management 35: 119-122.

53

Holmes, J. C., and P. W. Price. 1986. Communities of parasites. In Community ecology:
pattern and process. D. J. Anderson and J. Kikkawa (eds.), Blackwell Scientific
Co., Oxford.

Kocan, A. A., M. G. Shaw, and P. M. Morgan. 1979. Some parasitic and infectious
diseases in waterfowl in Oklahoma. Journal of Wildlife Diseases 15: 137-141.

Kocan, RM. 1966. The canvasback duck (Aythya valisineria); a new host record for
Plasmodium. Bulletin of the Wildlife Disease Association 4: 86-87 .

Kocan, R. M., and D. T. Clark. 1966a. Prepatent period and parasitenria in
Leucocytozoon simondi infections resulting from short exposures to sporozoites.
Journal of Parasitology 52: 962-966.

Kocan, R. M., and D. T. Clark. 1966b. Anemia in ducks infected with Leucocytozoon
simondi. Journal of Protozoology 13:465-468.

Kocan, R. M., and J. O. Knisley. 1970. Incidence of malaria in a wintering population of
canvasbacks (Aythya valisineria) on Chesapeake Bay. Journal of Wildlife
Diseases 6: 441-442.

Levine N. D., and H. C. Hanson. 1953. Blood parasites of the Canada goose, Branta
canadensis interior. Journal of Wildlife Management 17: 185-195.

Margolis, L., G. W. Esch, J. C. Holmes, A. M. Kuris, and G. A. Schad. 1982. The use of
ecological terms in parasitology (report of an ad hoc committee of the American
Society of Parasitologists). Journal of Parasitology 68: 131-133.

Meyer, C. L., G. F. Bennett, and C. M. Herman. 1974. Mosquito transmission of
Plasmodium (Giovannolia) circumflexum Kikuth, 1931 to waterfowl in the
Tantramar marshes, New Brunswick. Journal of Parasitology 60: 905-906.

Meyer, C. L., and G. F. Bennett. 1976. Observations on the sporogony of Plasmodium
circumflexum Kikuth and Plasmodium polare Manwell in New Brunswick.
Canadian Journal of Zoology 54: 133-141.

Michot, T. C., M. C. Garvin, and E. H. Weidner. 1995. Survey for blood parasites in
Redheads (Aythya americana) at the Chandeleur Islands, Louisiana Journal of
Wildlife Diseases 31: 90-92.

Nelson, E. C., and J. S. Gashwiler. 1941. Blood parasites of some Maine waterfowl.
Journal of Wildlife Management 5: 199-205.

O’Dell, J. P., and L. W. Robbins. 1994. Hematozoa of wood ducks (Aix sponsa) in
Missouri. Journal of Wildlife Diseases 30: 36-39.

O’Meara, D. C. 1956. Blood parasites of some Maine waterfowl. Journal of Wildlife
Management 20: 207-209.

O’Roke, E. C. 1931. A destructive waterfowl parasite. Transactions of the American
Game Conference 18: 248-251.

54

O’Roke, E. C. 1934. A malaria-like disease of ducks caused by Leucocytozoon anatis
Wickware. University of Michigan School of Forestry and Conservation Bulletin
No. 4. 43 p.

Polcyn, G. M., and A. D. Johnson. 1968. Hematozoa of the mallard duck Anas
platyrhynchos L. in South Dakota. Bulletin of the Wildlife Disease Association 4:
11.

Pung, O. J., N. E. Maxwell, E. C. Greiner, J. R. Robinette, and J. E. Thul. 1997.
Haemoproteus greineri in wood ducks from the Atlantic F lyway. Journal of
Wildlife Diseases 33: 355-358. '

Sibley, L. D., and J. K. Werner. 1984. Susceptibility of pekin and muscovy Ducks to
Haemoproteus nettionis. Journal of Wildlife Diseases 20: 108-113.

Tarshis, I. B. 1976. Further laboratory studies on the rearing and feeding of Cnephia
ornithophilia (Diptera: Simuliidae) and the transmission of Leucocytozoon
simondi by this black fly. Journal of Medical Entomology 13: 337-341.

Thul, J. E., D. J. Forrester, and E. C. Greiner. 1980. Hematozoa of wood ducks (Aix
sponsa) in the Atlantic Flyway. Journal of Wildlife Diseases 16: 383-390.

Trainer, D. 0., C. S. Schildt, R. A. Hunt, and L. R. Jahn. 1962. Prevalence of
Leucocytozoon simondi among some Wisconsin waterfowl. Journal of Wildlife
Management 26: 138-143.

van Riper III, C., S. G. van Riper, M. L. Goff, and M. Laird. 1986. The epizootiology
and ecological significance of malaria in Hawaiian land birds. Ecology 56: 327-
344. '

Williams, N. A., and G. F. Bennett. 1980. Avian Haemoproteidae. 13. The
haemoproteids of the ducks and geese (Anatidae). Canadian Journal of Zoology
58: 88-93.

Williams, N. A., B. K. Calverley, and J. L. Mahrt. 1977. Blood parasites of mallard and
pintail ducks from central Alberta and the Mackenzie Delta, Northwest
Territories. Journal of Wildlife Diseases 13: 226-229.

Wobeser, G. A. 1981. Diseases of wild waterfowl. Plenum Press, New York, 300 p.

Work, T. M., R. K. Washino, and C. van Riper III. 1990. Comparative susceptibility of
C ulex tarsalis, Anopholesfranciscanus, and C uliseta inornata (Diptera: .
Culicidae) to Plasmodium relictum (Haemosporidia: Plasmodiidae). Journal of
Medical Entomology 27: 68-71.

"lllllllllllllllllllf