Managing Japanese Barberry (Ranunculales: Berberidaceae)
Infestations Reduces Blacklegged Tick (Acari: Ixodidae)
Abundance and Infection Prevalence With Borrelia burgdorferi
SCOTT C. WILLIAMS,1,2 JEFFREY S. WARD,1 THOMAS E. WORTHLEY,3
AND KIRBY C. STAFFORD III4
Environ. Entomol. 38(4): 977Ð984 (2009)
ABSTRACT In many Connecticut forests with an overabundance of white-tailed deer (Odocoileus
virginianus Zimmermann), Japanese barberry (Berberis thunbergii DC) has become the dominant
understory shrub, which may provide a habitat favorable to blacklegged tick (Ixodes scapularis Say)
and white-footed mouse (Peromyscus leucopus RaÞnesque) survival. To determine mouse and larval
tick abundances at three replicate sites over 2 yr, mice were trapped in unmanipulated dense barberry
infestations, areas where barberry was controlled, and areas where barberry was absent. The number
of feeding larval ticks/mouse was recorded. Adult and nymphal ticks were sampled along 200-m
draglines in each treatment, retained, and were tested for Borrelia burgdorferi (Johnson, Schmid,
Hyde, Steigerwalt, and Brenner) presence. Total Þrst-captured mouse counts did not differ between
treatments. Mean number of feeding larval ticks per mouse was highest on mice captured in dense
barberry. Adult tick densities in dense barberry were higher than in both controlled barberry and no
barberry areas. Ticks sampled from full barberry infestations and controlled barberry areas had similar
infection prevalence with B. burgdorferi the Þrst year. In areas where barberry was controlled,
infection prevalence was reduced to equal that of no barberry areas the second year of the study.
Results indicate that managing Japanese barberry will have a positive effect on public health by
reducing the number of B. burgdorferiÐinfected blacklegged ticks that can develop into motile life
stages that commonly feed on humans.
KEY WORDS blacklegged tick, Japanese barberry, Lyme disease, white-footed mouse, white-tailed
Japanese barberry (Berberis thunbergii DC) is a
thorny, perennial shrub native to southern and central
Japan (Ohwi 1965). Japanese barberry was Þrst
planted in North America in the late 1800s (Harrington et al. 2003) and has since escaped from cultivated landscapes. It is now established in 31 states of
the continental United States, the District of Columbia, and Þve Canadian provinces (USDAÐNRCS
2008). Dense barberry stands are associated with a
paucity of desirable tree regeneration and herbaceous
plants (Harrington et al. 2003). Barberry may alter
nitrogen cycling, affecting soil biota (Kourtev et al.
1999, Ehrenfeld et al. 2001), as well as soil structure
and function (Kourtev et al. 2003). A Maine study
reported blacklegged ticks (Ixodes scapularis Say)
were twice as numerous in barberry-infested forests
than in adjacent forests without barberry (Elias et al.
2006). Blacklegged ticks are the major vector for the
agents that cause Lyme disease, human granulocytic
anaplasmosis, and human babesiosis (Magnarelli et al.
2006); thus, barberry infestations may have an indirect, adverse effect on human health.
Throughout the region, especially where whitetailed deer (Odocoileus virginianus Zimmermann)
populations are high, dense barberry stands can develop in the forest understory (Ehrenfeld 1997, Silander and Klepeis 1999).
North America (Martin et al. 1951). It is likely that
increased browsing of palatable species gives less palatable species, such as Japanese barberry, a competitive advantage (Tilghman 1989, Silander and Klepeis
1999, Elias et al. 2006). In addition, white-tailed deer
are the primary host for adult blacklegged ticks,
and their relative abundances are highly correlated
(Daniels et al. 1993, Deblinger et al. 1993, Stafford
1993, Daniels and Fish 1995).
Anecdotal observations of severely browsed forests
in southern Connecticut with deer densities as high as
40 deer/km2 (Williams and Ward 2006) have shown a
dearth of tree saplings and virtually no shrub layer
other than dense Japanese barberry infestations. Such
barberry stands may serve as refugia for small woodland rodents, speciÞcally white-footed mice (Peromyscus leucopus RaÞnesque) (Elias et al. 2006, Prusinski
et al. 2006), providing increased protection from both
avian and terrestrial predators. In addition, barberry is
used as a questing habitat by blacklegged ticks beause
little other suitable vegetation exists in severely
browsed forests. Because white-footed mice are the
primary reservoir for the spirochete Borrelia burgdorferi (Johnson, Schmid, Hyde, Steigerwalt, and Brenner) (Anderson et al. 1987, Magnarelli et al. 2006), the
causal agent of Lyme disease in humans, it is likely that
infection prevalence in blacklegged ticks would be
elevated where high mouse abundances exist.
We hypothesized that densities of white-footed
mice and blacklegged ticks and associated B. burgdorferi infection prevalence were higher in wooded areas
with dense Japanese barberry infestations than in adjacent wooded areas with no barberry. We also hypothesized that controlling barberry would result in a
reduction in mouse and tick densities and potentially
B. burgdorferi infection prevalence. Data presented
are from the Þrst 2 yr of a multiyear study.
Materials and Methods
Study Areas. Three replicate study areas were established in geographically separate areas within Connecticut: in south-central Connecticut on South Central Connecticut Regional Water Authority property
in the town of North Branford (Gaillard), in western
Connecticut on Aquarion Water Company property
in the town of Redding (Redding), and in northeastern Connecticut on the University of Connecticut
Forest in Storrs (Storrs) (Fig. 1). All study areas had
remnant stone walls running throughout and were
once agricultural Þelds or pastures; Storrs and Gaillard
were abandoned in the early 1900s and Redding in the
Management was negligible (fuelwood harvests of
declining and subcanopy trees), except at Gaillard
where 70% of eastern hemlock [Tsuga canadensis L.
(Carrie` re)] were removed during a salvage harvest in
the early 1990s. The remaining upper canopy of Gaillard was primarily sugar maple (Acer saccharum
Marsh.) with mixed oak (Quercus spp.), white ash
(Fraxinus americanaL.), American beech (Fagus grandifolia Ehrh.), and scattered yellow poplar (Liriodendron tulipifera L.). Upper canopies of Storrs and Redding were characterized by a predominance of white
ash, red maple (Acer rubrum L.), oak, yellow poplar,
and some black cherry (Prunus serotina Ehrh.). All
study areas had medium to dense stands of mature
Japanese barberry that were excluding desirable forest
regeneration and native herbaceous vegetation.
Fig. 1. Plot locations in Connecticut.
978 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 4
Soil classiÞcations, elevations, and coordinates of
study areas are shown in Table 1. Elevations ranged
from 55 to 355 m above mean sea level. Climatic data
(NOAA 1991) were from Hartford, CT, geographically centered among the study areas, which are
within the northern temperate climate zone. Mean
monthly temperature ranged from 3C in January to
23C in July. There was an average of 176 frost free
days per year. Average annual precipitation was 1,128
mm/yr, evenly distributed over all months.
Plot Design and Japanese Barberry Control. Three
treatment plots were established each at Gaillard,
Storrs, and Redding. These included an intact barberry infestation where barberry was not manipulated
(full barberry), an area where barberry was managed
by a series of control methods (controlled barberry),
and an area where barberry was minimal or absent
(no barberry). The understory of full and controlled
barberry areas also contained Oriental bittersweet
(Celastrus orbiculatus Thunb.), winged euonymus
[Euonymus alatus (Thunb.), Siebold], wine raspberry
(Rubus phoenicolasius Maxim.), northern spicebush
(Lindera benzoin L.), American witchhazel (Hamamelis virginiana L.), and occasional common barberry
(Berberis vulgaris L.), but were dominated by Japanese barberry. Because of severe browsing, little understory vegetation was present in no barberry areas
except for unpalatable species such as northern spicebush, winged euonymus, American witchhazel, and
an occasional Japanese barberry plant. Treatment
plots within the three barberry cover types corresponded to mouse trapping areas, were 45 by 60 m,
and were 250 m from one another at Gaillard, Storrs,
or Redding. Barberry cover was estimated by sampling
100 0.5-m2 areas at 3-m spacing within each treatment
plot. The 0.5-m2 sampling instrument consisted of a 4
by 4 grid with which percent barberry cover was
determined by presence/absence within each cell.
In controlled barberry plots, initial control was accomplished by mechanical cutting and shredding of
the above-ground portion of the plant and was completed in March 2007. We used a hydraulically driven
rotary wood shredder (model BH74FM, Bull Hog;
Fecon, Lebanon, OH) mounted to a compact track
loader (model T300; Bobcat, West Fargo, ND) for
initial control. Barberry clumps missed by the wood
shredder (adjacent to trees, stone walls, or large
rocks) were cut with a brush saw. Follow-up methods
used to control resprouting ramets (individual stems)
were directed ßame with a propane torch (model BP
223 C Weed Dragon; Flame Engineering, LaCrosse,
KN) and foliar applications of glyphosate and triclopyr. Follow-up control methods were completed in
late June 2007. Total area where barberry was controlled averaged 4,500 m2 each at Gaillard, Storrs, and
Redding. More details on treatment speciÞcs can be
found in Ward et al. (2009).
Mouse Trapping and Larval Ticks. Mice were
trapped using folding Sherman live traps (H. B. Sherman Traps, Tallahassee, FL) from July to September
2007 and July to August 2008. Twenty traps were set
in permanent grids with 15-m spacing at each of the
three treatment plots (n 60) at each replicate study
area and baited with peanut butter. Traps were set at
each replicate study area on Þve different trapping
events in 2007 and three different trapping events in
2008, resulting in a total of 480 trap nights/treatment
(n 1440 trap nights). Captured mice were temporarily sedated using the inhalant anesthetic isoßuorane. Each mouse received a uniquely numbered ear
tag (National Band and Tag, Newport, KY), and the
number of larval ticks feeding on mice was recorded
Sedated mice were allowed to recover from the
effects of isoßuorane and were released into the plot
from which they were originally captured. Mouse capture and handling protocols were approved by the
Wildlife Division of the Connecticut Department of
Environmental Protection and The Connecticut Agricultural Experiment StationÕs Institutional Animal
Care and Use Committee in accordance with the
American Society of Mammalogists guidelines for the
use of wild animals in research (Gannon et al. 2007).
Based on pelage and morphological characteristics, it
was assumed that all captured mice were white-footed
mice rather than deer mice (Peromyscus maniculatus
Wagner). Although deer mice are difÞcult to distinguish from white-footed mice based on appearance,
the known range of deer mice in Connecticut is restricted to the northwestern portion of the state (DeGraaf and Rudis 1986), which was outside our study
Population estimates were originally derived using
the Jolly-Seber model (Jolly 1965, Seber 1965), but
resulted in unacceptably high error because of inconsistent recaptures across treatments. Therefore, we
used counts of Þrst-captured mice to compare estimated population densities between treatments, which
have been shown to be good indices of estimating intraspeciÞc population size using our trapping protocols
(Slade and Blair 2000).
One-way analysis of variance (ANOVA) was used
to determine differences between treatments for Þrstcaptured mice for both years. One-way ANOVA was
also used to test difference of feeding larval ticks/
mouse between treatments for all captured mice,
which included recaptures for both years. Tukey honestly signiÞcantly different (HSD) test was used to
maintain levels at P 0.05 for multiple comparison
tests of differences between treatments.
Adult Tick Sampling. Each treatment was sampled
(with removal) 27 times (9 times at each of the three
replicate sites) for adult blacklegged ticks from November 2007 to November 2008 using standard ßagging techniques when adults were active (Stafford
Table 1. Description of study area locations and soil types
area Soil classiÞcation Elevation
(m) Latitude Longitude
Redding Canton and Charlton 150 41.28 73.37
85 41.37 72.77
Storrs Paxton and Montauk 190 41.82 72.25
August 2009 WILLIAMS ET AL.: BARBERRY CONTROL REDUCES B. burgdorferi PREVALENCE 979
2007). A 1-m2 white canvas cloth attached to a dowel
was used to ßag vegetation or the forest ßoor over
established transects totaling 200 m in each treatment.
Flags were checked for ticks every 15 m. Gathered
ticks were relocated to a laboratory, stored in a hydrator, and incubated at 10C. Samples were pooled to
determine the effect of Japanese barberry control on
adult tick abundances immediately after (fall 2007/
spring 2008) and 1 yr after management (fall 2008).
Nymphal ticks also were retained, but sample sizes
were too small to conduct meaningful density estimates. One-way ANOVA was used to determine differences in adult tick counts between treatments for
each sampling interval. Tukey HSD was used to maintain levels at P 0.05 for multiple comparison tests
of differences between treatments.
Borrelia burgdorferi Testing and Health Risk. Gathered tick midguts were dissected under a stereo
microscope and contents were smeared on 12-well
glass microscope slides (30-103HTC; Thermo Fisher,
Portsmouth, NH). B. burgdorferi spirochetes were
identiÞed in midgut contents by using indirect ßuorescent antibody (IFA) staining methods with monoclonal antibody H5332, which is speciÞc for outer
surface protein A of B. burgdorferi (Magnarelli et al.
1994). Fluorescein isothiocyanate-conjugated goat
anti-mouse immunoglobulins (KPL, Gaithersburg,
MD) were diluted 1:40 in phosphate-buffered saline
solution and used as a second antibody. Procedural
details followed established protocols (Anderson et al.
1991; Magnarelli et al. 1994, 1997).
Borrelia burgdorferi infection prevalence data were
pooled for adult and nymphal ticks as both stages were
found in treatment plots, commonly feed on humans,
and are capable of transferring spirochetes. When B.
burgdorferi infection prevalence differed between
treatments, the following procedures (Neter et al.
1982, pp. 325Ð329) were used to determine which
To assess relative risk to public health, the estimated
density of infected ticks per hectare was determined
for each of the three treatments by taking the product
of infection prevalence (%) and relative tick density
(including nymphs for fall 2008) for each sampling
interval. One-way ANOVA was used to determine
differences in relative density of infected ticks between treatments for each sampling interval. Tukey
HSD was used to maintain levels at P 0.05 for
multiple comparison tests of differences between
Japanese Barberry Control. Virtually all cut and
shredded Japanese barberry clumps produced new
ramets after initial mechanical control. Although the
initial control did not kill all barberry genets (the
entire plant), mechanical control was successful in
reducing the size of barberry clumps. All follow-up
control methods resulted in both increased mortality
of barberry genets and smaller clump sizes (Ward et
al. 2009). The control methods were successful in
reducing Japanese barberry cover from 62% pretreatment to 3% post-treatment in controlled barberry areas. Japanese barberry cover estimates for treatments
for both years are shown in Table 2. For a more
detailed account of results of Japanese barberry management, see Ward et al. (2009).
Mouse Trapping and Larval Ticks on Mice. In 2007,
there was minimal disturbance of traps; a few were
triggered prematurely and moved several meters from
their original locations, likely by raccoons (Procyon
lotor L.). For 2007, we assumed equal disturbance
across treatments and study areas. However, major
trap disturbance occurred at all three treatments at
Redding in 2008, and as a result, we were unable to
obtain any usable data for mouse abundances or for
larval ticks feeding on mice.
A total of 269 captures of 149 Þrst-captured mice
occurred in 2007. Counts of Þrst-captured mice did
not differ between treatments (F 0.01; df 2,3;
P 0.98): full barberry (n 51), controlled barberry (n 49), and no barberry (n 49). In 2008,
there were 133 captures of 84 Þrst-captured mice.
Counts of Þrst-captured mice did not differ between
treatments (F 0.24; df 2, 3; P 0.80): full
barberry (n 27), controlled barberry (n 22),
and no barberry (n 24). However, mean feeding
larval ticks per captured mouse did differ between
treatments and were greatest in full barberry in both
2007 (F 3.37; df 2,265; P 0.04) and 2008 (F
3.10; df 2,128; P 0.05; Fig. 2). It was assumed that
all larvae feeding on mice were blacklegged ticks
Table 2. Percent Japanese barberry cover at each replicate
site for no barberry (No Barb.), controlled barberry pretreatment
(Pre-treat), controlled barberry post-treatment (Post-treat), and
full barberry (Full Barb.) by study area by year
Study Area No Barb. Controlled Barberry Full Barb.
Redding 2007 5.2 52.0 2.6 63.5
Redding 2008 5.3 Ñ 1.1 61.0
Gaillard 2007 0.6 63.7 7.2 44.5
Gaillard 2008 1.4 Ñ 4.9 45.9
Storrs 2007 1.9 70.8 0.3 23.9
Storrs 2008 0.8 Ñ 2.8 25.8
Mean 2.9 62.2 3.2 44.1
980 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 4
because the only other tick species we encountered
throughout the entire study were a fewadult dog ticks
(Dermacentor variabilis Say) (n 8).
Adult Tick Sampling. A total of 525 adult blacklegged ticks (245 males, 280 females) were collected
throughout all treatment areas: 23 (15 males, 8 females) in no barberry, 76 (21 males, 55 females) in
controlled barberry, and 221 (103 males, 118 females) in full barberry in fall 2007/spring 2008 and
26 (11 males, 15 females) in no barberry, 35 (18
males, 17 females) in controlled barberry, and 144
(77 males, 67 females) in full barberry in fall 2008.
Mean tick densities were consistently greater in full
barberry areas for both sampling intervals: fall 2007/
spring 2008 (F 13.67; df 2,42; P 0.01) and fall
2008 (F 7.39; df 2,33; P 0.01). Full barberry
tick densities differed signiÞcantly from both controlled and no barberry treatments for both sampling intervals (Fig. 3).
Borrelia burgdorferi Prevalence and Health Risk. A
total of 342 (132 males, 155 females, 55 nymphs)
blacklegged ticks were tested for B. burgdorferi
presence: 108 (47 males, 61 females) from the fall
2007/spring 2008 sampling interval and 234 (85
males, 94 females, 55 nymphs) from fall 2008. For fall
2007/spring 2008, infection prevalence was equal
for ticks from full and no barberry areas (44%) and
was lower in ticks gathered from no barberry areas
(10%), but did not differ from full barberry (z
2.04; df 2; P 0.10) or controlled barberry treatments (z 1.97; df 2; P 0.10), likely because of
low sample size in no barberry (n 10). After the
second growing season postbarberry control (fall
2008), infection prevalence in controlled barberry
(45%) was lower than full barberry (63%) but did
not differ (z 2.21; df 2; P 0.08). Controlled
barberry tick infection prevalence did not differ
from no barberry areas either (39%) (z 0.51; df
2; P 0.90). Full barberry areas did have signiÞcantly higher infection prevalence than ticks in no
barberry areas (z 2.60; df 2; P 0.03).
Estimated B. burgdorferi infected tick densities differed between treatments for both fall 2007/spring
2008 (F 15.91; df 2,42; P 0.001) and fall 2008 (F
12.22; df 2,33; P 0.001). There were an estimated
324 infected ticks/ha in full barberry in fall 2007/
spring 2008, which differed from both controlled barberry (111 infected ticks/ha; P 0.002) and no barberry (8 infected ticks/ha; P 0.001). The estimated
number of infected ticks/ha did not differ between
controlled barberry and no barberry (P 0.18; Fig. 4).
In fall 2008, there were an estimated 496 infected
ticks/ha in full barberry, which differed from both
controlled barberry (137 infected ticks/ha; P 0.001)
and no barberry (89 infected ticks/ha; P 0.001).
Fig. 2. Mean number of feeding larval ticks/mouse for
Gaillard, Redding, and Storrs in 2007 and for Gaillard and
Storrs in 2008. Means with the same letter within the same
series are not signiÞcantly different. Error bars represent
Fig. 3. Estimated adult blacklegged tick density (ticks/
ha) based on 200-m2 sampling areas. Means with the same
letter within the same series are not signiÞcantly different.
Error bars represent SEM.
Fig. 4. Estimated number of blacklegged ticks infected
with B. burgdorferi based on percent infection prevalence
and estimated tick density for each treatment. Means with
the same letter within the same series are not signiÞcantly
different. Error bars represent SE.
August 2009 WILLIAMS ET AL.: BARBERRY CONTROL REDUCES B. burgdorferi PREVALENCE 981
Estimated infected ticks per hectare did not differ
between controlled barberry and no barberry (P
0.86; Fig. 4).
The white-tailed deer density in portions of Connecticut is exceedingly high; an aerial survey conducted in 2003 indicated densities exceeding 40 deer/
km2 in portions of the Gaillard study area (Williams
and Ward 2006). Continual browsing by white-tailed
deer compromises the health and competitiveness of
native shrubs, thus creating conditions favorable for
Japanese barberry establishment (Tilghman 1989, Silander and Klepeis 1999). Ginsberg and Ewing (1989)
reported that ticks were common in the shrub layer
that provides physically higher questing habitat for
ticks to Þnd suitable hosts. Elias et al. (2006) found that
in the 0.5- to 1.5-m height class, exotic-invasive shrub
stem density was twice that of native shrubs in southern Maine. In some Connecticut forests, the only remaining suitable questing habitat for ticks is Japanese
Questing adult blacklegged tick abundances were
greatest in areas dominated by Japanese barberry
(Fig. 3). This relationship was also noted in southern
Maine, at the northern extent of the blacklegged
tickÕs range (Lubelczyk et al. 2004). Elias et al.
(2006) reported a positive relationship between
questing nymphal and adult blacklegged ticks and
exotic-invasive shrubs, including Japanese barberry.
We found a greater abundance of larval blacklegged
ticks feeding on white-footed mice in barberry-dominated understories compared with areas where barberry was virtually absent. These collective results
indicate that dense Japanese barberry infestations can
be favorable habitat for all three life stages of the
By controlling Japanese barberry, we reduced the
number of questing adult blacklegged ticks to densities statistically equal that of areas with no barberry for
both years. This result was likely caused by the mechanical removal of the aboveground portion of barberry, effectively eliminated questing habitat, which
corroborates Wilson (1986). The abundance of larval
blacklegged ticks feeding on mice differed initially
only between full barberry and no barberry treatment
areas. In fall 2008 where barberry was controlled,
feeding larval tick abundances were signiÞcantly
lower than full barberry areas. We suspect this was a
result of fewer available egg-laying adult females and
that this trend will continue as the 2-yr life cycle of the
blacklegged tick progresses.
Prusinski et al. (2006) emphasized the importance
of the relationship between preferred small mammal
habitat (food resources and protection from predation) and ideal tick habitat (questing areas, host availability, suitable microclimate). The dense growth
form of Japanese barberry provides increased protection from predation, its abundant fruits provide a potential food source, and it may provide other yet unrecognized ecosystem services beneÞtting mouse
populations (Seagle 2008). Japanese barberry leafs out
1 mo before overstory trees and most native shrubs
(Silander and Klepeis 1999, Xu et al. 2007). This timing
directly corresponds to the peak of spring activity
of adult blacklegged ticks in Connecticut (Stafford
2007). We believe that the dense canopy of Japanese
barberry infestations retains humidity better than surrounding areas, thus creating a favorable microsite for
blacklegged tick survival (Rodgers et al. 2007). As a
result, adult ticks may be concentrated in barberry
infestations in early spring during barberry leaf-out,
when limited forest overstory canopy allows for increased sunlight penetration, creating less than favorable conditions for tick survival by lowering soil moisture and relative humidity. Japanese barberry also
provides excellent questing habitat as its height (upwards of 3 m) is ideal for blacklegged ticks to attach
to white-tailed deer, which readily incorporate barberry infestations in their home ranges.
Prusinski et al. (2006) found that small mammals
had higherB. burgdorferiinfection prevalence in areas
where shrub density was higher. We originally hypothesized that a higher percentage of blacklegged
ticks found in barberry infestations would be infected
with B. burgdorferi because increased white-footed
mouse densities would serve as a larger reservoir.
Although mouse densities were similar across treatments, infection prevalence was consistently higher in
barberry infestations than in areas without barberry.
This, combined with higher abundances of ticks found
in barberry infestations, resulted in signiÞcantly more
questing ticks infected with B. burgdorferi (Fig. 4),
which poses a considerable threat to public health. By
controlling Japanese barberry, we reduced adult
blacklegged tick densities, B. burgdorferi infection
prevalence, and resulting estimated number of B.
burgdorferiÐinfected ticks to equal that of areas where
barberry was absent (Figs. 3 and 4).
It is clear that dense Japanese barberry infestations provide highly favorable habitats for small
mammals, deer, and tick development and survival.
Removal of Japanese barberry will signiÞcantly decrease the abundance of ticks, their infection prevalence with B. burgdorferi, and the environmental
risk of Lyme disease. It is also clear that Japanese
barberry infestations pose an indirect threat to public health. We suggest that municipalities, forest
managers, and private landowners with Japanese
barberry infestations on their properties take immediate management action in the interest of preserving both forest and public health.
We thank Aquarion Water Company, CT ChapterÐThe
Nature Conservancy, South Central Connecticut Regional
Water Authority, and Weed-It-Now ProgramÐThe Nature
Conservancy for Þnancial and technical assistance and providing study sites. The Connecticut Department of Environmental Protection-Division of Forestry provided personnel
assistance. J. P. Barsky, R. M. Cecarelli, R. J. Hannan, Jr., M. R.
Short, G. M. Picard, E. A. Kiesewetter, D. V. Tompkins, R. A.
982 ENVIRONMENTAL ENTOMOLOGY Vol. 38, no. 4
Wilcox, T. M. Blevins, H. Stuber, C. Ariori, K. Drennan, F.
Pacyna, and J. Bravo assisted with plot establishment, treatments, data collection, and tick IFAs. Monoclonal antibody
H5332 was provided by A. Barbour of the Department of
Microbiology and Molecular Genetics at the University of
California, IrvineÕs School of Medicine.
Anderson, B. E., J. E. Dawson, D. C. Jones, and K. H. Wilson.
1991. Ehrlichia chaffeensis, a new species associated with
human ehrlichiosis. J. Clin. Microbiol. 29: 2838 Ð2842.
Anderson, J. F., R. C. Johnson, L. A. Magnarelli, F. W. Hyde,
and J. E. Myers. 1987. Prevalence of Borrelia burgdorferi
andBabesia microtiin mice on islands inhabited by whitetailed deer. Appl. Environ. Microb. 53: 892Ð 894.
Daniels, T. J., and D. Fish. 1995. Effect of deer exclusion on
the abundance of immature Ixodes scapularis (Acari: Ixodidae) parasitizing small and medium-sized mammals.
J. Med. Entomol. 32: 5Ð11.
Daniels, T. J., D. Fish, and I. Schwartz. 1993. Reduced abundance of Ixodes scapularis (Acari: Ixodidae) and Lyme
disease risk by deer exclusion. J. Med. Entomol. 30: 1043Ð
Deblinger, R. D., M. L. Wilson, D. W. Rimmer, and A.
Spielman. 1993. Reduced abundance of immature Ixodes dammini (Acari: Ixodidae) following incremental
removal of deer. J. Med. Entomol. 30: 144 Ð150.
DeGraaf, R. M., and D. D. Rudis. 1986. England wildlife:
habitat, natural history, and distribution. Gen. Tech. Rep.
NE-108. U.S. Department of Agriculture, Forest Service,
Northeastern Forest Experimental Station, Broomall, PA.
Ehrenfeld, J. G. 1997. Invasion of deciduous forest preserves in the New York metropolitan region by Japanese
barberry (Berberis thunbergii DC). J. Torr. Botan. Soc.
124: 210 Ð215.
Ehrenfeld, J. G., P. Kourtev, and W. Huang. 2001. Changes
in soil functions following invasions of exotic understory
plants in deciduous forests. Ecol. Appl. 11: 1287Ð1300.
Elias, S. P., C. B. Lubelczyk, P. W. Rand, E. H. LaCombe,
M. S. Holman, and R. P. Smith, Jr. 2006. Deer browse
resistant exotic-invasive understory: an indicator of elevated human risk of exposure to Ixodes scapularis (Acari:
Ixodidae) in southern coastal Maine woodlands. J. Med.
Entomol. 43: 1142Ð1152.
Gannon, W. L., R. S. Sikes, and The Animal Care and Use
Committee of the American Society of Mammalogists.
2007. Guidelines of the American Society of Mammalogists for the use of wild animals in research. J. Mammal.
88: 809 Ð 823.
Ginsberg, H. S., and C. P. Ewing. 1989. Habitat distribution
of Ixodes dammini (Acari: Ixodidae) and Lyme disease
spirochetes on Fire Island, New York. J. Med. Entomol.
Harrington, R. A., R. Kujawski, and H.D.P. Ryan. 2003. Invasive plants and the green industry. J. Arbor. 29: 42Ð 48.
Jolly, G. M. 1965. Explicit estimates from capture-recapture
data with both death and immigrationÐstochastic model.
Biometrika 52: 225Ð247.
Kourtev, P., J. G. Ehrenfeld, and M. Haggblom. 2003. Experimental analysis of the effect of exotic and native plant
species on structure and function of soil microbial communities. Soil Biol. Biochem. 35: 895Ð905.
Kourtev, P., W. Z. Huang, and J. G. Ehrenfeld. 1999. Differences in earthworm densities and nitrogen dynamics
in soils under exotic and native plant species. Biol. Invasions 1: 237Ð245.
Lubelczyk, C. B., S. P. Elias, P. W. Rand, M. S. Holman, E. H.
LaCombe, and J.R.P. Smith. 2004. Habitat associations
of Ixodes scapularis (Acari: Ixodidae) in Maine. Environ.
Entomol. 33: 900 Ð906.
Magnarelli, L. A., J. F. Anderson, and K. C. Stafford III. 1994.
Detection of Borrelia burgdorferi in urine of Peromyscus
leucopus by inhibition enzyme-linked immunosorbent assay. J. Clin. Microbiol. 32: 777Ð782.
Magnarelli, L. A., J. F. Anderson, K. C. Stafford III, and J. S.
Dumler. 1997. Antibodies to multiple tickborne pathogens of babesiosis, ehrlichiosis, and Lyme borreliosis in
white-footed mice. J. Wildlife Dis. 33: 466 Ð 473.
Magnarelli, L. A., K. C. Stafford III, J. W. IJdo, and E. Fikrig.
2006. Antibodies to whole-cell or recombinant antigens
of Borrelia burgdorferi, Anaplasma phagocytophilum, and
Babesia microti in white-footed mice. J. Wildlife Dis. 42:
Martin, A., H. S. Zim, and A. L. Nelson. 1951. American
wildlife and plants: a guide to wildlife food habits. Dover
Publications, New York.
National Oceanic and Atmospheric Administration [NOAA].
1991. Local climatological data 1990. National Oceanic and
Atmospheric Administration, Hartford, Connecticut.
Neter, J., W. Wasserman, and G. A. Whitmore. 1982. Applied statistics, 2nd ed. Allyn and Bacon, Boston, MA.
Ohwi, J. 1965. Flora of Japan. Smithsonian Institution,
Prusinski, M. A., H. Chen, J. M. Drobnack, S. J. Kogut, R. G.
Means, J. J. Howard, J. Oliver, G. Lukacik, P. B. Backenson, and D. J. White. 2006. Habitat structure associated
with Borrelia burgdorferi prevalence in small mammals in
New York State. Environ. Entomol. 35: 308 Ð319.
Rodgers, S. E., C. P. Zolnik, and T. N. Mather. 2007. Duration of exposure to suboptimal atmospheric moisture affects nymphal blacklegged tick survival. J. Med. Entomol.
Rutberg, A. T., R. E. Naugle, L. A. Thiele, and I.K.M. Liu.
2004. Effects of immunocontraception on a suburban
population of white-tailed deer (Odocoileus virginianus).
Biol. Conserv. 116: 243Ð250.
Seagle, S. W. 2008. Ecosystem ecology of the golden mouse,
pp. 81Ð97.In G.W. Barrett and G. A. Feldhamer (eds.), The
golden mouse: ecology and conservation. Springer, New
Seber, G.A.F. 1965. A note on the multiple-recapture census. Biometrika 52: 249 Ð259.
Silander, J. A., and D. M. Klepeis. 1999. The invasion ecology of Japanese barberry (Berberis thunbergii) in the
New England landscape. Biol. Invasions 1: 189 Ð201.
Slade, N. A., and S. M. Blair. 2000. An empirical test of using
counts of individuals captured as indicies of population
size. J. Mammal. 81: 1035Ð1045.
Stafford, K. C., III. 1993. Reduced abundance of Ixodes
scapularis (Acari: Ixodidae) with exclusion of deer by
electric fencing. J. Med. Entomol. 30: 986 Ð996.
Stafford, K. C., III. 2007. Tick management handbook: an
integrated guide for homeowners, pest control operators,
and public health ofÞcials for the prevention of tickassociated diseases. Connecticut Agricultural Experiment Station, New Haven, CT.
Tilghman, N. G. 1989. Impacts of white-tailed deer on forest
regeneration in northwestern Pennsylvania. J. Wildlife
Manag. 53: 524 Ð532.
[USDA–Natural Resources Conservation Service] U.S. Department of Agriculture. 2008. The PLANTS database.
August 2009 WILLIAMS ET AL.: BARBERRY CONTROL REDUCES B. burgdorferi PREVALENCE 983
Ward, J. S., T. E. Worthley, and S. C. Williams. 2009. Controlling Japanese barberry (Berberis thunbergiiDC)in southern
New England, USA. Forest Ecol. Manag. 257: 561Ð566.
Williams, S. C., and J. S. Ward. 2006. Exotic seed dispersal
by white-tailed deer in southern Connecticut. Nat. Area.
J. 26: 383Ð390.
Williams, S. C., J. S. Ward, and U. Ramakrishnan. 2008.
Endozoochory by white-tailed deer (Odocoileus virginianus) across a suburban/woodland interface. Forest
Ecol. Manag. 255: 940 Ð947.
Wilson, M. L. 1986. Reduced abundance of adult Ixodes
dammini (Acari: Ixodidae) following destruction of vegetation. J. Econ. Entomol. 79: 693Ð 696.
Xu, C. Y., K. L. Griffin, and W.S.F. Schuster. 2007. Leaf phenology and seasonal variation of photosynthesis of invasive
Berberis thunbergii (Japanese barberry) and two co-occurring native understory shrubs in a northeastern United
States deciduous forest. Oecologia (Berl.) 154: 11Ð21.
Received 27 October 2008; accepted 6 April 2009