Blood parasites in owls with conservation implications for the Spotted Owl (Strix occidentalis).

Abstract

The three subspecies of Spotted Owl (Northern, Strix occidentalis caurina; California, S. o. occidentalis; and Mexican, S. o. lucida) are all threatened by habitat loss and range expansion of the Barred Owl (S. varia). An unaddressed threat is whether Barred Owls could be a source of novel strains of disease such as avian malaria (Plasmodium spp.) or other blood parasites potentially harmful for Spotted Owls. Although Barred Owls commonly harbor Plasmodium infections, these parasites have not been documented in the Spotted Owl. We screened 111 Spotted Owls, 44 Barred Owls, and 387 owls of nine other species for haemosporidian parasites (Leucocytozoon, Plasmodium, and Haemoproteus spp.). California Spotted Owls had the greatest number of simultaneous multi-species infections (44%). Additionally, sequencing results revealed that the Northern and California Spotted Owl subspecies together had the highest number of Leucocytozoon parasite lineages (n = 17) and unique lineages (n = 12). This high level of sequence diversity is significant because only one Leucocytozoon species (L. danilewskyi) has been accepted as valid among all owls, suggesting that L. danilewskyi is a cryptic species. Furthermore, a Plasmodium parasite was documented in a Northern Spotted Owl for the first time. West Coast Barred Owls had a lower prevalence of infection (15%) when compared to sympatric Spotted Owls (S. o. caurina 52%, S. o. occidentalis 79%) and Barred Owls from the historic range (61%). Consequently, Barred Owls on the West Coast may have a competitive advantage over the potentially immune compromised Spotted Owls.

Full text

Page 1

Blood Parasites in Owls with Conservation Implications
for the Spotted Owl (Strix occidentalis)
Heather D. Ishak1*, John P. Dumbacher1,2, Nancy L. Anderson3, John J. Keane4, Gediminas Valkiu ¯nas5,
Susan M. Haig6, Lisa A. Tell7, Ravinder N. M. Sehgal1
1Department of Biology, San Francisco State University, San Francisco, California, United States of America, 2Department of Ornithology and Mammalogy, California
Academy of Sciences, San Francisco, California, United States of America, 3Lindsay Wildlife Museum, Walnut Creek, California, United States of America, 4Sierra Nevada
Research Center, USDA Forest Service Pacific Southwest Research Station, Davis, California, United States of America, 5Institute of Ecology, Vilnius University, Vilnius,
Lithuania, 6USGS Forest and Rangeland Ecosystem Science Center, Corvallis, Oregon, United States of America, 7Department of Medicine and Epidemiology, School of
Veterinary Medicine, University of California Davis, Davis, California, United States of America
Abstract
The three subspecies of Spotted Owl (Northern, Strix occidentalis caurina; California, S. o. occidentalis; and Mexican, S. o.
lucida) are all threatened by habitat loss and range expansion of the Barred Owl (S. varia). An unaddressed threat is whether
Barred Owls could be a source of novel strains of disease such as avian malaria (Plasmodium spp.) or other blood parasites
potentially harmful for Spotted Owls. Although Barred Owls commonly harbor Plasmodium infections, these parasites have
not been documented in the Spotted Owl. We screened 111 Spotted Owls, 44 Barred Owls, and 387 owls of nine other
species for haemosporidian parasites (Leucocytozoon, Plasmodium, and Haemoproteus spp.). California Spotted Owls had the
greatest number of simultaneous multi-species infections (44%). Additionally, sequencing results revealed that the Northern
and California Spotted Owl subspecies together had the highest number of Leucocytozoon parasite lineages (n=17) and
unique lineages (n=12). This high level of sequence diversity is significant because only one Leucocytozoon species (L.
danilewskyi) has been accepted as valid among all owls, suggesting that L. danilewskyi is a cryptic species. Furthermore, a
Plasmodium parasite was documented in a Northern Spotted Owl for the first time. West Coast Barred Owls had a lower
prevalence of infection (15%) when compared to sympatric Spotted Owls (S. o. caurina 52%, S. o. occidentalis 79%) and
Barred Owls from the historic range (61%). Consequently, Barred Owls on the West Coast may have a competitive advantage
over the potentially immune compromised Spotted Owls.
Citation: Ishak HD, Dumbacher JP, Anderson NL, Keane JJ, Valkiu ¯nas G, et al. (2008) Blood Parasites in Owls with Conservation Implications for the Spotted Owl
(Strix occidentalis). PLoS ONE 3(5): e2304. doi:10.1371/journal.pone.0002304
Editor: Matthew Baylis, University of Liverpool, United Kingdom
Received February 25, 2008; Accepted April 20, 2008; Published May 28, 2008
Copyright: ? 2008 Ishak et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Ravinder N. M. Sehgal, Heather Ishak, and this research were supported in part by an NIGMS MORE Institutional Research and Academic Career
Development Award to UC Davis and San Francisco State University, grant number K12GM00679.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: heatherishak@gmail.com
Introduction
Emerging infectious diseases (EIDs) affecting wildlife appear to
be increasing in number and have led to localized decreases in
population sizes and species extinctions [1–4]. Wildlife diseases are
especially relevant today due to the potential of zoonotics such as
West Nile Virus, HIV, and H5N1 avian influenza [5–7]. EIDs are
facilitated by the movement of vectors and pathogens due to
environmental alterations caused by climate change and human
impacts (e.g. deforestation or the introduction of invasive species)
[8]. Diseases are also pertinent to conservation biology because
endangered or threatened species may be pushed to extinction due
to direct mortality or indirectly by a reduced reproductive success
[9,10].
The Spotted Owl (Strix occidentalis) is threatened not only by
habitat loss, but also an invasive owl species. All three subspecies of
the Spotted Owl (Northern, S. o. caurina; California, S. o. occidentalis;
and Mexican, S. o. lucida) have been the focus of intense
conservation efforts. This study focuses on the Northern and
California Spotted Owls since they are the two subspecies
currently most impacted by the Barred Owl (Strix varia) range
expansion [11–15].
Over the past fifty years, Barred Owls have expanded their
historic range in the eastern United States by crossing through the
southwestern region of Canada and moving down the Pacific
Coast of Canada and the U.S. [11]. Barred Owls began their
range expansion in British Columbia in 1943, and by 2007, were a
prime reason for near extirpation of Spotted Owls in the province
[16,17]. In the United States, Barred Owls were first recorded in
Washington in 1965, in Oregon in 1974 [18], and in California in
1981 [16]. Presently, they continue to move south, increasing in
population size throughout the Spotted Owl range [11,19]. Barred
Owls pose a threat to Spotted Owls because they are more
aggressive than Spotted Owls and compete for food and nesting
resources [11,20]. Expansion of Barred Owls may have additional
adverse effects on Spotted Owl populations if they introduce novel
infectious diseases.
Over two decades of intense research on Spotted Owls has
made them one of most studied birds in the world. However, little
is known about their blood parasites or disease threat [15,21–23].
In one of only two studies of blood parasites in Spotted Owls,
Gutie ´rrez (1989) showed that every Spotted Owl tested had at least
one blood parasite (Leucocytozoon or Haemoproteus spp.) and 79% had
simultaneous multi-species infections [19, Table 1]. Interestingly,
PLoS ONE | www.plosone.org1May 2008 | Volume 3 | Issue 5 | e2304

Page 2

neither study by Greiner (n=1 individual) or Gutie ´rrez (n=105)
recorded Plasmodium infections in the Spotted Owl [19,22].
At least nine studies have analyzed blood parasites in one or
more Barred Owls (n=64; Table 1) [24–32]. From these studies,
43 of 64 (67%) Barred Owls examined were infected with at least
one blood parasite. There were seven documented cases of
Plasmodium infections (Table 1) including discovery of a new
taxonomically distinct Plasmodium species (subgenus Novyella) [25].
Telford et al. (1997), found Barred Owls had longer and wider
asexually dividing cells (a.k.a. schizonts) of P. forresteri than seven
other raptor species living in Florida, suggesting that they might
have a morphologically distinct Plasmodium strain [28].
Nine morphologically distinct haemosporidian blood parasites
(order Haemosporidia) have been recorded in owls: Haemoproteus
noctuae, H. syrnii, Plasmodium subpraecox, P. fallax, P. forresteri, P.
gundersi, P. hexamerium, P. elongatum, and one species of Leucocytozoon,
L. danilewskyi (=L. ziemanni) [33]. While the concept of a species is
currently in flux, there is evidence of cryptic speciation of
haemosporidian blood parasites from genetic sequencing [34–
36]. There is also evidence for genetic variation in the cytochrome
b gene within microscopically-defined morphospecies [37,38].
Plasmodium, Haemoproteus, and Leucocytozoon spp. are spread to avian
hosts via insect vectors, mosquitoes (Culicidae), biting midges
(Ceratopogonidae), hippoboscid flies (Hippoboscidae), and black
flies (Simuliidae), respectively [33]. In general, Plasmodium spp. are
thought to be more pathogenic than Haemoproteus or Leucocytozoon
because they display a lower degree of host specificity, and cause a
more severe blood pathology [33,39–41]. However, numerous
species of Leucocytozoon and some hemoproteids also cause disease
in birds [33,42–44]; thus different species of haemosporidian
parasites can have differing effects on avian hosts.
Blood parasites are indicators of immune quality in birds, and
parasite prevalence data can be used to reveal information about
individual and population fitness [45–48]. Many species of blood
parasites are generally thought to be harmless because they appear
in otherwise healthy looking birds [45]. However, research has
shown that blood parasites can have negative fitness impacts on
the host [47–53]. Parasites can be pathogenic during energy-
demanding or stressful phases of a host’s life such as the first year
[49,50], migration [51,52], breeding [48,53,54], and years of low
food abundance [54,55]. Research has also shown that parasitic
infections can negatively impact reproductive success by delaying
arrival to the breeding grounds [56], reducing clutch sizes
[47,54,57], reducing nest defense behavior [58,59], increasing
probability of clutch desertion [60], reducing hatching success
[47,60], reducing fledging success [47], and siring nestlings with
poorer body condition [57]. From a global perspective, cumulative
effects of blood parasites on individuals can have serious
consequences on host populations [45]. Blood parasites can also
be extremely virulent when introduced to an immunologically
naı ¨ve species [4,33,43,44].
We surveyed haemosporidian parasites from the blood of twelve
western North American owl species using PCR and DNA
sequencing techniques. We examined the phylogenetic relation-
ships, host specificities, and distributions among hosts with the
intention to determine whether Barred Owls may be the source of
novel parasites to Spotted Owl populations on the West Coast of
North America. We also studied blood samples from Europe and
Africa to help elucidate phylogeographic relationships of haemo-
sporidian parasites in owls.
Results
Prevalence of Blood Parasites
Five hundred forty two individuals (317 belonging to the
Strigidae, and 225 belonging to the Tytonidae) from twelve owl
species were tested for Plasmodium, Haemoproteus, and Leucocytozoon
spp. parasites (Table 2) using PCR techniques. In the Strigidae
family, the overall prevalence of infection was 62% (n=197), while
24% (n=54) of the Tytonidae family had at least one blood
parasite infection. Prevalences for the three haemosporidian
parasites varied within and between families.
PCR and microscopy techniques indicated blood parasite
prevalences ranged from zero in Barn Owls from Denmark
(n=45), to 100% in the African owls. The African owls (n=3) had
at least three blood parasite species identified by blood smear
analysis. One individual Sjo ¨stedt’s Owlet (Glaucidium sjostedti) was
infected with five morphologically distinct haemosporidians
(Leucocytozoon danilewskyi, Haemoproteus noctuae, Haemoproteus syrnii,
Plasmodium (subgenus Haemamoeba) sp., and Plasmodium (subgenus
Giovannolaia) sp.).
Prevalences also varied between Spotted Owl subspecies in this
study. The California Spotted Owl was significantly more likely to
be infected with a blood parasites (p,0.007; x2=7.36) and also
has significantly more multiple infections (p=0.002; x2=9.18)
than the Northern Spotted Owl. When compared to owls
belonging to the same taxonomic family (Strigidae total numbers
from Table 2), the California Spotted Owl and Northern Spotted
Owls had significantly more Haemoproteus infections (S. o. occidentalis
p=0.0001, x2=30.83; S. o. caurina p=0.0001, x2=15.36), and the
California Spotted Owl had significantly less Plasmodium infections
(p=0.05; x2=3.77). Even though the Northern Spotted Owl
prevalence was also low (n=1), it was not significantly different
from the other Strigidae owls. However, finding one individual
infected with a Plasmodium infection is significant because it is the
first documentation of Spotted Owls with this parasite.
Northern Spotted Owl samples were further tested for change in
blood parasite prevalence between ten-year time spans (1994–
1996 to 2004–2005) but no statistical differences were found
Table 1. Previous research on Spotted and Barred Owl
haematozoa
Speciesn# infectedLHP Citation
Strix occidentalis1111024
S. o. occidentalis76 7671 670 21
S. o. caurina22 22 21110 21
S. o. lucida77430 21
Total106 106 97 820
Strix varia 543 n/an/a3 28
Strix varia28190 19*3 31
Strix varia21 191920 29
Strix varia51010 26
Strix varia43323 24
Strix varia32212 27**
Strix varia1111130**
Strix varia11 n/an/a1 25
Strix varia10000 32
Total64 42 22246
L=Leucocytozoon, H=Haemoproteus, P=Plasmodium
*This number does not account for multiple Haemoproteus species.
**Multiple infections were found.
doi:10.1371/journal.pone.0002304.t001
Blood Parasites in Spotted Owl
PLoS ONE | www.plosone.org2May 2008 | Volume 3 | Issue 5 | e2304

Page 3

(Table 3). Furthermore, blood parasite prevalences did not differ
within the Northern Spotted Owls when compared between states
(WA vs. OR) (Table 3).
Barred Owl results were divided into two subgroups to examine
potential geographic differences (CA, OR, WA vs MN, WI, TX;
Table 2). Barred Owls in the historic eastern range had
significantly higher prevalences
significantly more Plasmodium infections (p=0.05; x2=4.70) than
the Barred Owls on the west coast.
The blood parasite prevalence also differed between western
Barred Owls and Spotted Owls. Spotted Owls had a higher
prevalence of blood parasites (52% in Northern subspecies, 79% in
California subspecies) than Barred Owls on the West Coast (15%).
These sympatric Barred Owls had significantly lower prevalences
of Leucocytozoon (p=0.001; x2=28.75) and Haemoproteus infections
(p=0.001; x2=25.80), and fewer cases of multiple infections
(p=0.001; x2=12.36) and
(p=0.001; x2=20.56) when compared to all Spotted Owls.
Barred Owls, however, did have significantly higher Plasmodium
spp. prevalences (p=0.05; x2=4.17), but the differences in
Plasmodium prevalence were more pronounced when comparing
all Spotted Owls with all Barred Owls sampled (p=0.001;
x2=14.38) (see Table 2).
DNA Sequencing and Host-Parasite Relationships
Leucocytozoon.
The Leucocytozoon dataset consisted of 623
base pairs of the mitochondrial cytochrome b gene sequenced from
181 owls among eleven species. There were 38 lineages found
(including two outgroups) with 219 variable characters and 124
parsimony informative characters. The maximum likelihood
heuristic search for the Leucocytozoon analysis yielded 34 trees
which varied only at the branch tips (Figure 1).
The likelihood tree revealed two Leucocytozoon clades (1 & 2) with
relatively high sequence divergence (uncorrected p=8.05%). The
presence of two distinct clades suggests that L. danilewskyi is a
cryptic species. A side by side blood smear analysis of Leucocytozoon
from blood parasites within Boreal Owls from both clades showed
identical morphologies. However, the Barn Owls, which were only
found in the first clade, only had the round morph gametocytes of
Leucocytozoon present in the blood smears. Typically, elongated and
round morph gametocytes develop in Leucocytozoon danilewskyi
infections, and they are frequently present simultaneously. In
summary, although the morphology of L. danilewskyi appears
identical in all blood smears, the mtDNA reveals that there are
many different haplotypes/lineages and high genetic divergence,
perhaps justifying reclassifying L. danilewskyi into two or more
species or subspecies.
The Strigidae family appears to be susceptible to parasites from
both clades of Leucocytozoon spp (clade 1 & 2, Figure 1). The
Tytonidae (Barn Owls) only had parasite strains from clade 1. This
Table 2. Prevalence of haematozoa infections in Strigiformes
Family Scientific NameCommon Name*Code Locationn I (%)MI (%)L (%)H (%)P (%)
StrigidaeAegolius funereus Boreal OwlBOOW Lithuania542402
Asio otusLong-earred Owl LEOWCA, Lithuania 2823 (82)10 (36)22 (79)10 (36)1 (4)
Athene cunicularia Burrowing OwlBUOW CA710100
Bubo virginianusGreat-horned OwlGHOW CA54 34 (63)4 (7) 34 (63)6 (11)3 (6)
Glaucidium sjostedtiSjostedti Owln/a Cameroon222220
Megascops kennicottii Western Screech OwlWESOCA 5345 (85) 10 (25)42 (79)0 (0)15 (28)
Otus scopsScops Owln/a Khazakstan200100
Strix alucoTawny Owl n/a Lithuania200000
Strix o. caurina Spotted Owl, Northern SPOWnoCA, OR, WA 63 33 (52) 10 (16) 25 (40)16 (25)1 (2)
Strix o. occidentalisSpotted Owl, California SPOWcaCA4838 (79) 21 (44)29 (60)30 (63) 0 (0)
Strix varia** Barred OwlBDOW MN, WI, TX18 11 (61)2 (11)1 (6) 6 (33) 6 (33)
Strix varia** Barred Owl BDOWCA, OR, WA264 (15)1 (4) 2 (8)1 (4)2 (8)
S. o. and S. v. hybridSpotted/Barred hybridn/aOR 810200
Strix woodfordiiAfrican Wood Owln/aCameroon 111110
TOTAL 317197 (62) 66 (21)166 (52)72 (23) 30 (9)
TytonidaeTyto alba pratincolaBarn OwlBNOWCA 180 54 (30)3 (2) 54 (30) 3 (2)0 (0)
Tyto alba guttataGerman Barn Owln/aDenmark 450 (0)0 (0)0 (0)0 (0)0 (0)
TOTAL 225 54 (24)3 (1)54 (24)3 (1) 0 (0)
I=Infected, MI=multiple infections, L=Leucocytozoon, H=Haemoproteus, P=Plasmodium
*American Ornithologist’s Union (AOU) name code
**Strix varia was split into two groups to due differing prevalences on the West Coast vs their historic range.
doi:10.1371/journal.pone.0002304.t002
Table 3. Prevalence of blood parasites in Northern Spotted
Owls collected over ten years
Year CollectedLocationn# infected (%) L (%)H (%)
1994–1996 Washington16 7 (43.8)7 (43.8)2 (12.5)
1994–1996 Oregon6 3 (50) 3 (50)1 (16.7)
Total22 10 (45.5) 10 (45.5) 3 (13.6)
2005 Washington73 (42.8) 2 (28.6)1 (14.3)
2004–2005Oregon21 13 (61.9) 6 (28.6)8 (38.1)
Total 28 16 (57)8 (28.6)9 (32.1)
L=Leucocytozoon, H=Haemoproteus
doi:10.1371/journal.pone.0002304.t003
Blood Parasites in Spotted Owl
PLoS ONE | www.plosone.org3 May 2008 | Volume 3 | Issue 5 | e2304

Page 4

first Leucocytozoon clade has eight haplotypes (lineages) with
relatively little variation (uncorrected p=1.2%). The second
Strigidae clade has 28 lineages with an average of 3.7% variation.
Host specificity for Leucocytozoon spp. appears to differ among
owl species. The EcoSim rarefaction species richness test on Barn,
Western Screech, Spotted, and Great-horned Owls showed that
lineage diversity from 30 randomly selected owls for 1000
replications resulted in a low mean diversity for Western Screech
(2.65) and Barn Owls (1.99) and a high mean lineage diversity for
Spotted Owls (14.99) and Great-horned (14) (Figure 2). Barn Owls
and Western Screech Owls appear to have very specific
Leucocytozoon strains each with one dominant lineage and one or
two less common lineages. In contrast, Spotted Owls and Great-
horned Owls showed low specificity and were infected with many
different parasite lineages. In addition, we found that Spotted and
Great-horned Owls had many unique lineages defined as
haplotypes that were not found in any other owl species (Spotted
Owls=12 unique out of 17 total lineages; Great-horned Owls=10
unique out of 14 total lineages).
Plasmodium and Haemoproteus spp.
sequenced from 106 individuals among 10 owl species. There
were 21 unique lineages (10 Plasmodium and 11 Haemoproteus) found
with 139 variable sites and 100 parsimony-informative sites.
The maximum likelihood heuristic search yielded one tree
433 base pairs were
for Plasmodium and Haemoproteus spp. from ten species of owl
(Figure 3).
In Figure 3, there appears to be four distinct lineages or possible
species of Plasmodium present in these owls. One of the Plasmodium
species from a Barred Owl can be identified as P. elongatum because
it is an identical match to a sequence identified from GenBank
Figure 1. Maximum likelihood tree (1) of Leucocytozoon from eleven owl species. Maximum likelihood bootstrap values are shown above
the branches (100 replicates) and uncorrected ‘p’ distances are shown below. For explanation of bird abbreviation codes, see Appendix 1. Following
each bird abbreviation code is the number of individuals (n) with that particular lineage and the sampling location.
doi:10.1371/journal.pone.0002304.g001
Figure 2. The mean species diversity of Leucocytozoon
lineages randomly drawn from four owl species (n=30).
doi:10.1371/journal.pone.0002304.g002
Blood Parasites in Spotted Owl
PLoS ONE | www.plosone.org4 May 2008 | Volume 3 | Issue 5 | e2304

Page 5

(DQ659588). The other three possible species could only be
identified to the genus Plasmodium due to the poor quality of
available blood smears.
Sequence data identified one Northern Spotted Owl with a
Plasmodium parasite. The sequence differs by only one base change
from a parasite isolated from one invasive Barred Owl but also
other native California owls (3 Great-horned Owls, and 13
Western Screech Owls).
In the Haemoproteus clade, one dominant lineage (n=55) was
isolated from four owl species (Spotted Owl, Barred Owl, Great-
horned Owl, and African Wood Owl) and from North America
and Africa. Additionally, Long-eared Owls from North America
and Europe shared an identical Haemoproteus strain. The global
commonality of blood parasites within these owls provides
evidence that these blood parasites have no distinct phylogeo-
graphic pattern. Also, Barn Owls appear to have a unique
Haemoproteus lineage although the prevalence of this parasite in the
population remains low (1.3%).
Discussion
Prevalence of Blood Parasite
The prevalence of blood parasites varied highly among owl
species. The lowest prevalences were documented in Barn Owls
from Denmark. Four previous studies from northern Europe also
detected no blood parasites in Barn Owls (n=26) [61–66]. This
suggests that the habitats of Barn Owls in Denmark and other
Northern European countries are less suitable for blood parasite
vectors [67]. The African owls, in contrast, came from forested
habitats with a diverse vector fauna resulting in high prevalences
and a high number of co-infections; up to five different
haemosporidian species in one individual [68].
The number of Spotted Owls found with a blood parasite (S . o.
caurina=52% and S. o. occidentalis 79%) in this study, although
high, is significantly lower than the 100% prevalence seen in
Gutie ´rrez’s (1989) study (p=0.001; x2=59.39). The Northern
Spotted Owl prevalence of all haemosporidians over a ten year
time span (Table 3) shows that the blood parasite prevalences
compared between the years 1994–1996 and 2004–2005 were not
statistically different. Therefore, the differences between this study
and Gutie ´rrez’s (1989) results may not represent a decline in blood
parasite infections over time, but may have been due to other
reasons such as differing blood parasite detection methods. This
study is the first to use PCR based detection of blood parasites in
Spotted Owls, which has been proven to yield much greater
sensitivity than older microscopy techniques [69]. Yet, a recent
study found although PCR assays were sufficient in finding single
parasite infections, it may underestimate cases with multiple
parasite infections [46].
Another difference between our study and Gutierrez’s is the
sampling locations. The 1989 study examined California Spotted
Owls from four different locations (Sierra Nevada, San Jacinto
Mountains, San Bernadino Mountains, and Palomar Mountains),
and the Northern Spotted Owl samples were from northwestern
California [19]. The California Spotted Owls in this study came
from one Sierra Nevada location, and the Northern Spotted Owls
came from California locations as well as Oregon and Washing-
ton. Other factors that could contribute to the differing prevalence
rates include time of year the owls were sampled, annual
variations, age of host [70], and environmental conditions [71].
Figure 3. Maximum likelihood tree for Plasmodium and Haemoproteus parasites in ten owl species. Maximum likelihood bootstrap values
are shown above the branches (1000 replicates) and uncorrected ‘p’ distances are shown below. For explanation of bird abbreviation codes see
Table 2. Following each bird abbreviation is the number of individuals (n) with that particular lineage and the sampling location.
doi:10.1371/journal.pone.0002304.g003
Blood Parasites in Spotted Owl
PLoS ONE | www.plosone.org5May 2008 | Volume 3 | Issue 5 | e2304