Ecology, 85(9), 2004, pp. 2467-2477 
? 2004 by the Ecological Society of America 
EXPERIMENTALLY REDUCING NEIGHBOR DENSITY AFFECTS 
REPRODUCTION AND BEHAVIOR OF A MIGRATORY SONGBIRD 
T. SCOTT SILLETT,13 NICHOLAS L. RODENHOUSE,2 AND RICHARD T. HOLMES1 
1
 Department of Biological Sciences, Dartmouth College, Hanover, New Hampshire 03755 USA 
-Department of Biological Sciences, Wellesley College, Wellesley, Massachusetts 02181 USA 
Abstract. Because populations of territorial birds are relatively stable compared to 
those of other animal taxa, they are often considered to be tightly regulated. However, the 
mechanisms that produce density-dependent feedbacks on demographic rates and thus reg- 
ulate these populations are poorly understood, particularly for migratory species. We con- 
ducted a three-year density-reduction experiment to investigate the behavioral mechanisms 
that regulate the abundance of a Nearctic-Neotropical migrant passerine, the Black-throated 
Blue Warbler (Dendroica caerulescens), during the breeding season. We found that the 
number of young fledged per territory, territory size, and the proportion of time males spent 
foraging were significantly greater on territories around which neighbor density was ex- 
perimentally reduced compared to control territories. Territory quality, proportion of nests 
depredated per territory, and male countersinging rates were not statistically different be- 
tween treatments. These results indicate that individuals with more neighbors (i.e., in neigh- 
borhoods with greater conspecific density) have reduced breeding productivity. The results 
also suggest that a crowding mechanism that mediates interactions among territory-holders 
could generate the density dependence needed to regulate local abundance, at least in areas 
of homogeneous, high-quality habitat. The effect of the neighbor-density reduction on 
warbler fecundity and behavior varied with annual fluctuations in weather and food avail- 
ability, and was strongest in 1997, an El Nino year, when conditions for breeding were 
least favorable. This variation in our experimental results among years implies that density 
dependence due to crowding may have its strongest impact on local abundance when 
environmental conditions are relatively poor. 
Key words: Black-throated Blue Warbler; Dendroica caerulescens; density dependence; Hubbard 
Brook Experimental Forest; Nearctic-Neotropical migratory birds; neighbor density; population reg- 
ulation; territorial behavior. 
INTRODUCTION 1995, Krebs 2002). Experimental manipulations are of- 
_,,     . ,      .n r        . ,     . ,  , ten necessary to disentangle the interactions of multiple The identification of regulatory mechanisms and the .... , , .    .        . 
il      ?   ,      .       , . .,.    . , limiting and regulatory processes on population size 
strength of density dependence is critical to under- . f 
standing and managing natural populations (Murdoch 
1994, Hixon et al. 2002, Runge and Johnson 2002). 
However, research on population regulation has fo- 
cused primarily on measuring density dependence, 
whereas the proximate mechanisms by which density 
can affect demographic rates are less well understood 
(Sinclair  1989,  Krebs   1991,  Mylius  and Diekmann 
1995, Ferrer and Donazar  1996,  Rodenhouse et al. 
1997, May  1999, Turchin 1999, Hixon et al. 2002). 
Indeed, direct evidence of regulation and of the action 
of regulatory mechanisms remains elusive, despite sev- . 
,   ,       ,        ? ,_...,. , regulates abundance (Dhondt et al. 1992, Rodenhouse 
eral decades of research. This is due, in part, to the 
(Turchin 1995, den Boer and Reddingius 1996, Moss 
et al. 1996, Rodenhouse et al. 1999). For example, the 
intensity of intraspecific competition may be density 
dependent, but its effect on population size may be 
weak and overwhelmed by a stronger density-indepen- 
dent process, such as climate variation. 
Populations of territorial birds appear to be tightly 
regulated because they are relatively stable compared 
to those of other animal taxa (Hanski and Tiainen 1989, 
Murdoch 1994). Many studies of density dependence 
in territorial birds assume that a crowding mechanism 
scarcity of long-term data (Sinclair 1989, Newton 
1998) and to the rarity of experimental perturbations 
of density (Murdoch 1994, Harrison and Cappuccino 
et al. 1997, 1999, Bonsall et al. 1998). Resource com- 
petition associated with crowding is the basis of den- 
sity-dependent regulation as described by Lack (1954, 
1966) and Fretwell and Lucas (1970). Under a crowd- 
ing   mechanism,   regulation   could   be   accomplished 
Manuscript received 24 April 2003; revised 17 December     through     density-dependent     interactions     between 
2003; accepted 25 January 2004; final version received 20 Feb-     crowded individuals and their natural enemies, such as 
ruary 2004. Corresponding Editor: J. R. Sauer. ,  . , ,,      .      .??,, ,. 
,,,        .   j,        e   -n.      ?     ir     .      t>- AI-*    .    XT        predators  (e.g.,  Martin   1996)  or  disease  organisms 3
 Present address: Smithsonian Migratory Bird Center, Na-      r ?' ' ? 
tional   Zoological   Park,   Washington,   DC.   20008   USA.      (e.g., Hochachka and Dhondt 2000), or through a den- 
E-mail: silletts@si.edu sity-dependent increase in agonistic interactions among 
2467 
2468 T. SCOTT SILLETT ET AL. Ecology, Vol. 85, No. 9 
crowded conspecifics; both processes can decrease de- 
mographic rates. The ultimate causes of agonistic in- 
teractions include competition for limited resources 
such as food (Newton 1998), territory space (McCleery 
and Perrins 1985, Stamps 1990, Mougeot et al. 2003), 
nest sites (Brawn and Balda 1988), and mating oppor- 
tunities (Chuang-Dobbs et al. 2001). Crowding prob- 
ably has the strongest regulatory effect where species 
exist at high densities in relatively homogenous habitat. 
Crowding during the breeding season could amplify 
resource limitation for adult birds and their young by 
reducing territory size, by increasing time spent in ag- 
onistic interactions and thus reducing time spent for- 
aging or provisioning young, or by a combination of 
these factors. The best evidence in passerines for den- 
sity-dependent feedbacks caused by crowding during 
the breeding period comes from species using nest box- 
es (Alatalo and Lundberg 1984, Stenning et al. 1988, 
Torok and Toth 1988, Perrins 1990, Both 1998a) or 
from those confined to islands (e.g., Arcese et al. 1992, 
McCallum et al. 2000). Although these studies indicate 
that density-dependent feedbacks can occur, they do 
not identify the mechanisms by which crowding re- 
duces fecundity. Furthermore, the presence of density- 
dependent negative feedbacks generated by crowding 
during the breeding season has not been determined or 
experimentally tested in open-cup-nesting passerines. 
In this paper, we examine the effect of neighbor den- 
sity on the demography and behavior of a Nearctic- 
Neotropical migrant songbird, the Black-throated Blue 
Warbler (Dendroica caerulescens), and discuss its role 
in population regulation. This research was motivated 
by several lines of evidence that indicated that our 
study population in New Hampshire, USA was regu- 
lated and that a key mechanism involved was crowding- 
induced resource competition during the breeding sea- 
son. First, Black-throated Blue Warbler abundance has 
been relatively stable and has not shown any directional 
trend at our study site since 1969, a period of more 
than 30 years (Holmes and Sherry 2001). Second, pop- 
ulation growth rate and multiple measures of annual 
warbler fecundity (mean number of young fledged per 
territory, mean fledgling mass, and proportion of ter- 
ritories with females attempting second broods, but not 
mean clutch sizes of first- or second-brood nests) were 
strongly and negatively correlated with local popula- 
tion density (Rodenhouse et al. 2003, Sillett and 
Holmes, in press). Third, density-dependent fledging 
success, as illustrated by simulation models parame- 
terized with field data, is sufficient to constrain abun- 
dance within the range observed on the study area (Sil- 
lett and Holmes, in press). Fourth, Black-throated Blue 
Warblers are highly territorial during the breeding sea- 
son (Holmes 1994), and forage almost exclusively on 
their territories. Fifth, annual fecundity is limited, in 
part, by food availability (Rodenhouse and Holmes 
1992, Sillett et al. 2000, Nagy 2002). Sixth, predation 
on Black-throated Blue Warbler nests, although an im- 
portant factor limiting annual fecundity, does not ap- 
pear to be density dependent at Hubbard Brook (Reits- 
ma 1992, Sillett and Holmes, in press). Finally, vari- 
ance in warbler fledging success does not increase with 
population density on our 150-ha study area (Sillett 
and Holmes, in press), suggesting that all individuals 
breeding here are affected equally by density-depen- 
dent processes (see Ferrer and Donazar 1996, Both 
1998a). 
We present the results of a three-year density-re- 
duction experiment designed to test if crowding is an 
important mechanism regulating Black-throated Blue 
Warbler abundance. At the start of our experiment, we 
hypothesized that the number of neighbors (i.e., local 
neighbor density) would affect reproductive output and 
breeding behavior of Black-throated Blue Warblers. We 
predicted that: (1) territories with a reduced number of 
neighbors would fledge more young than control ter- 
ritories; (2) territories would be larger in the reduced- 
density treatment; (3) nest predation rates would not 
differ between treatments; (4) levels of male-male in- 
teractions would be lower at reduced density, enabling 
these males to spend more time foraging; and (5) pa- 
rental feeding rates of nestlings would be higher in the 
reduced-density treatment. As a corollary of prediction 
5, we predicted that fledgling mass would be greater 
in the reduced-density treatment. We analyze and dis- 
cuss differences in warbler fecundity, behavior, and 
territory quality between control territories and terri- 
tories around which the density of neighboring con- 
specifics was experimentally reduced. We consider 
these results in relation to annual variation in weather 
and food availability during the three years of our 
study. 
METHODS 
Study site and species 
Field research was conducted from May to August 
1997-1999 in the 3160-ha Hubbard Brook Experimen- 
tal Forest in Woodstock, New Hampshire, USA. The 
forest was extensively logged in the early 1900s. Our 
gridded, 150-ha study site extended from 520 to 610 
m above sea level on one south-facing hillside. This 
site represented high-quality, high-density breeding 
habitat for Black-throated Blue Warblers (Holmes et 
al. 1996), with each warbler pair having 4-6 neigh- 
boring conspecihc pairs. The relatively homogeneous 
vegetation on the study site consisted of a 20-25 m 
tall canopy of American beech (Fagus grandifolia), 
sugar maple {Acer saccharum), and yellow birch (Bet- 
ula alleghaniensis), and a thick understory dominated 
by hobblebush {Viburnum alnifolium), striped maple 
(A. pensylvanicum), and beech saplings. Further details 
on characteristics of the study site can be found in 
Holmes et al. (1996) and Holmes and Sherry (2001). 
The Black-throated Blue Warbler is a common breed- 
er  in   hardwood  forests   of  northern   New  England 
September 2004 BEHAVIORALLY MEDIATED DENSITY DEPENDENCE 2469 
TABLE 1. Interannual variation in climate, clutch completion dates, and food abundance for 
Black-throated Blue Warblers at Hubbard Brook Experimental Forest, New Hampshire, USA, 
1997?1999. Except where noted, values are given as means ? 1 SE. 
Year 
Mean daily 
temperature, 
May (?C)f 
Total rainfall, Mean clutch        Mean caterpillar 
June (mm)t       completion date$     biomass (mg)? 
1997 6.5 ? 1.6 79.5 170.4 ? 3.7 8.4 ? 2.0 
1998 13.3 ? 2.2 325.0 154.1 ? 3.2 6.0 ? 1.1 
1999 12.8 ? 2.5 78.6 154.3 ? 1.8 13.5 ? 2.5 
Long-term mean 11.5 ? O.5|| 89.7 ? 6.4|| 158.3 ? 1.7<H 11.5 ? 1.4<H 
t Climate data obtained by scientists of the Hubbard Brook Ecosystem Study [Online, URL: 
(http://www.hubbardbrook.org)]. The Hubbard Brook Experimental Forest is operated and 
maintained by the Northeastern Research Station, U.S. Dept. of Agriculture, Newtown Square, 
Pennsylvania, USA. 
$ Julian date for first clutches, where day 1 = 1 January. 
? Mean dry biomass of caterpillars/200 American beech and sugar maple leaves recorded at 
40 points; biomass data were collected during four biweekly surveys (once every two weeks) 
from late May to late July (see Rodenhouse et al. 1992, Jones et al. 2003) and were summed 
for each point. 
|| 50-year mean. 
II Mean for 1986-1999. 
(Holmes 1994). This sexually dichromatic species is 
insectivorous and males defend exclusive, non-over- 
lapping territories during the breeding season. At Hub- 
bard Brook, Black-throated Blue Warblers are multiple- 
brooded and socially monogamous, although 5-10% of 
males are bigamous in an average year (Holmes et al. 
1992; T. S. Sillett, N. L. Rodenhouse, and R. T. Holmes, 
unpublished data). Females build nests in understory 
shrubs and incubate eggs; mean and modal clutch size 
is 4 (Holmes 1994). Both parents feed nestlings and 
fledglings. Adults have strong fidelity to breeding ter- 
ritories between years (Holmes 1994), but natal dis- 
persal is high, judging from the low return rate of fledg- 
lings banded on the study plot (T. S. Sillett, N. L. Ro- 
denhouse, and R. T. Holmes, unpublished data). Nev- 
ertheless, reproductive success of warblers breeding on 
our site each year is significantly and positively cor- 
related with the number of yearling recruits in the fol- 
lowing spring (Sillett et al. 2000), indicating that our 
study population is representative of warbler popula- 
tion dynamics occurring at a larger, regional scale (see 
also Jones et al. 2003). 
Weather and warbler food availability on our study 
site differed dramatically between the 1997 through 
1999 breeding seasons (Table 1), and this natural en- 
vironmental variation formed an important backdrop 
for our research. In 1997, an El Nino year, conditions 
for Black-throated Blue Warblers were least favorable 
(Sillett et al. 2000, Sillett and Holmes, in press). Mean 
daily temperature in May was lower in 1997 than in 
either 1998 or 1999, or in any year since 1986, and 
this resulted in a later mean date of first clutch com- 
pletion (Table 1). Mean food availability in 1997 was 
also lower than average. In June 1998, record-high rain- 
fall was recorded in many parts of New Hampshire 
(Bell et al. 1999). At Hubbard Brook, approximately 
four times as much rain fell in June 1998, the height 
of the Black-throated Blue Warbler breeding season, 
compared to either 1997, 1999, or the 50-year mean 
(Table 1). Mean food availability in 1998 was the low- 
est of the three years. In 1999, La Nina conditions 
predominated, food was most abundant, and the weath- 
er was most benign (Table 1). 
Experimental design and field methods 
Our experimental design built upon prior research at 
Hubbard Brook, which demonstrated that the territory 
space occupied by a Black-throated Blue Warbler pair 
remained empty for an entire breeding season if both 
the female and male were permanently removed after 
the end of spring migration (Marra and Holmes 1997). 
Thus, by removing already settled pairs of warblers in 
late May and early June, we were able to create two 
neighbor-density treatments: a reduced-density treat- 
ment in which we removed all conspecifics whose ter- 
ritories abutted those of focal pairs, and a control con- 
sisting of focal territories with a normal, high density 
of neighbors. Control (n = 9 in 1997; n = 12 in 1998; 
n = 11 in 1999) and reduced-density territories (n = 
4/yr) were randomly selected from those present on the 
150-ha study plot. Warbler density was not manipulated 
within a 64-ha section of the plot to avoid compro- 
mising an ongoing, long-term population study (see 
Holmes et al. 1992, 1996, Holmes and Sherry 2001, 
Sillett and Holmes 2002, in press). Control territories 
were separated from areas where density was reduced 
by at least one and usually several intervening, occu- 
pied territories. Focal territories for the reduced-density 
treatment were chosen after a preliminary 3-5 day sur- 
vey to map locations of all countersinging males. In 
this treatment, conspecific pairs on territories adjoining 
those of focal pairs were permanently removed by shot- 
gun under approval from the Institutional Animal Care 
and Use Committee, Dartmouth College (protocol 
number 96-09-07) and from appropriate federal and 
state agencies. Removals were conducted during nest- 
2470 T. SCOTT SILLETT ET AL. Ecology, Vol. 85, No. 9 
building and laying, with a few pairs being removed 
during early incubation, and were completed by early 
June. After removals were completed, adults were cap- 
tured in mist nets, individually marked with a unique 
combination of one numbered aluminum leg band and 
two colored plastic leg bands, and aged as either sec- 
ond-year (SY) or after second-year (ASY) using plum- 
age characters. However, we did not consider female 
age in our analyses because some could not be aged 
with certainty. Both male age classes were present in 
each treatment each year (for 1997 control, 5 ASY, 4 
SY; for 1997 reduced density, 1 ASY, 3 SY; for 1998 
control, 8 ASY, 4 SY; for 1998 reduced density; 3 ASY, 
1 SY; for 1999 control, 7 ASY, 4 SY; for 1999 reduced 
density, 3 ASY, 1 SY). All males in the reduced-density 
treatment, and all but one (in 1999) of the control 
males, were mated. No unmated females were detected 
in the study area. 
To test our general hypothesis, that local neighbor 
density would affect warbler reproductive output and 
breeding behavior, and specifically, predictions 1-3, we 
intensively monitored focal territories, found and doc- 
umented fates of nests, and measured territory sizes. 
Nests were usually found during nest-building and were 
checked every two days until Hedging or failure. Cause 
of failure (e.g., predator, abandoned by female) was 
determined from the condition of failed nests (see 
Holmes et al. 1992, 1996). Parental identity was ver- 
ified by observing leg band combinations when birds 
visited nests, and nest success was confirmed by noting 
parents feeding fledglings. Adults were monitored after 
a nest fledged or failed to ensure that further renesting 
and double-brooding attempts were discovered. Move- 
ments of male warblers were mapped every 2-3 days 
from mid-May through late June each year. Males were 
followed closely without disturbing them for 15-20 
min periods, and their positions were recorded on graph 
paper based on their movements within the 25-m plot 
grid. These field maps were transferred to weekly sum- 
mary maps that were used to construct a final territory 
map for each male in each breeding season. Territories 
were delineated with the minimum convex polygon 
method (Ford and Myers 1981). Three-factor, fixed- 
effect ANOVAs were used to test if the number of 
young fledged per territory, territory size, and the pro- 
portion of nests depredated per territory differed be- 
tween the two treatments. Independent variables were 
year, male age, neighbor-density treatment, and all in- 
teraction terms. 
Predictions 4 and 5 were tested by quantifying and 
analyzing singing and foraging behavior of monoga- 
mous males between early June and early July, 1997 
and 1999. No observations were made on males in the 
reduced-density treatment until at least two days after 
removals were completed. Behavioral data were ana- 
lyzed with split-plot, univariate repeated-measures 
ANOVAs (Milliken and Johnson 1992). Fixed effects 
were year, nest stage, male age class, neighbor-density 
treatment, and all two- and three-factor interactions 
between these variables. Individual male nested within 
year, male age class, and treatment (i.e., male [year, 
male age, treatment]) was included in models as a ran- 
dom effect. Dependent variables will be described. 
Models were fit using the Restricted Maximum Like- 
lihood (REML) method (SAS Institute 1996). 
Male song rate.?Male countersinging rate was used 
to measure male-male interactions because fighting 
and chasing were rarely observed after establishment 
of territory boundaries in mid-May, before we started 
removals. A countersong was defined as a song from 
a focal male that was followed immediately by a song 
of one or more neighboring males; at least two sets of 
songs had to be exchanged between males before the 
behavior was recorded as countersinging. Song data 
were collected while mapping the territorial movement 
of a subset of males. Countersinging rates were mea- 
sured during the nest-building-egg-laying, incubation, 
and nestling stages of first broods. Song data were not 
collected for males after first broods fledged. For anal- 
ysis, data for each male were summarized by calcu- 
lating the number of countersongs per minute per stage. 
Male foraging time.?Time budgets were quantified 
for males encountered opportunistically during the 
nest-building-egg-laying, incubation, nestling, and 
fledgling stages of first broods. Females were too cryp- 
tic to observe regularly. Once a male was located and 
identified, he was observed for 30 seconds before his 
behaviors were dictated onto microcassette. This delay 
minimized biasing estimated time budgets toward con- 
spicuous activities (e.g., acrobatic foraging maneu- 
vers). Behavioral data were recorded continuously until 
an individual was either lost from sight, observed for 
10 consecutive minutes, or perched for five consecutive 
minutes. Behavior categories noted were: foraging, 
perching, patrolling (moving without foraging), preen- 
ing, and mate guarding. The latter two behaviors were 
rarely observed, and we were principally interested in 
the time that males spent foraging. Black-throated Blue 
Warblers are active foragers that move quickly and cap- 
ture prey by gleaning, hovering, and snatching them 
from foliage and twigs (Robinson and Holmes 1982, 
Holmes 1994), making the classification of a behavioral 
sequence as "foraging" unambiguous. A maximum of 
20 minutes of data were collected per male per day. 
Microcassette tapes were transcribed, and the duration 
of each observation bout and the time each male for- 
aged per bout were recorded. After discarding obser- 
vation bouts <2 minutes long, we summed foraging 
time and total observation time per male per nest stage 
and then calculated the proportion of time each male 
spent foraging per stage. Only summed observation 
periods a 5 minutes per male per nest stage were in- 
cluded in data analyses. 
Nestling provisioning rate and fledgling mass.?Pa- 
rental feeding rates and fledgling mass were quantified 
for first brooding nests each year. Nests were video- 
September 2004 BEHAVIORALLY MEDIATED DENSITY DEPENDENCE 2471 
taped for continuous 2.5-h periods during the morning 
on day 7 of the nestling period. Previous research in- 
dicated that the nestling provisioning rate of Black- 
throated Blue Warblers increased daily until fledging, 
but did not vary significantly with time of day or with 
parental age (Goodbred and Holmes 1996). Cameras 
were placed 10-15 m from nests and concealed in veg- 
etation. Nestlings were counted immediately prior to 
filming. Nests were not filmed in rain or when vege- 
tation was dripping wet. When transcribing video tapes, 
we discarded the first 15 minutes of data to ensure that 
parents were acclimated to the camera's presence. The 
number of food deliveries per nestling per hour was 
analyzed separately by sex because male and female 
provisioning rates were not significantly correlated (r 
= 0.25, P = 0.29). Food biomass delivered per adult 
visit could not be reliably determined from video tapes. 
Nestling mass on day 6 of the eight-day nestling period, 
the last day to safely handle young without causing 
premature fledging, was used as an estimate of fledgling 
mass. Mean deliveries per nestling per hour and mean 
fledgling mass were analyzed with two-factor, fixed- 
effect ANOVAs with year, neighbor-density treatment, 
and year X treatment as independent variables. Data 
from territories with bigamous males (control, n = 1 
in 1997; reduced-density, n = 3 in 1998) were excluded 
from statistical analyses because these males typically 
provision nestlings less than do monogamous males (T. 
S. Sillett, N. L. Rodenhouse, and R. T. Holmes, un- 
published data) 
To verify that any differences found between treat- 
ments were not due to differences in territory quality, 
i.e., a site-dependent mechanism (Rodenhouse et al. 
1997), we measured and compared the quality of all 
focal territories in both treatments. Previous research 
indicated that high-quality territories for Black-throat- 
ed Blue Warblers at Hubbard Brook would have a high 
density of hobblebush and other vegetation in the shrub 
layer, ample arthropod prey, and low predator abun- 
dance (Rodenhouse and Holmes 1992, Steele 1993, 
Holmes et al. 1996). Territory quality data were ana- 
lyzed with a two-factor, fixed-effect MANOVA with 
year, neighbor-density treatment, and year X treatment 
as independent variables. Dependent variables for the 
territory model will be described. Model sphericity 
(i.e., the assumption of an equal covariance structure 
between responses for all possible pairs of response 
variables) was assessed with the Mauchly criterion and, 
when necessary, the Geisser-Greenhouse (hereafter G- 
G) correction was used to estimate degrees of freedom 
for F tests (Muller and Barton 1989). 
Vegetation.?Understory leaf density was measured 
at five randomly selected, 394-m2 circular plots per 
territory; plots were &50 m apart. In each plot, 11.2- 
m transects were delineated in the understory in car- 
dinal directions. At the distal end of each transect, all 
leaves of hobblebush and those of the two other dom- 
inant understory species, American beech and striped 
maple, that intersected a ground-level, 9-m2 plane (de- 
marcated by two 3-m vertical poles set 3 m apart) were 
counted. Leaf density data were summed for each ter- 
ritory site. 
Prey abundance.?Availability of the warbler's pri- 
mary prey in summer, lepidopteran larvae and spiders, 
was quantified per territory on four biweekly surveys 
beginning in late May each year. Black-throated Blue 
Warblers typically forage for arthropods in the under- 
story (Robinson and Holmes 1982), and prey biomass 
is not strongly stratified by height on our study plot 
(Holmes and Schultz 1988). A survey entailed non- 
destructive, visual inspections of 16 50-leaf samples 
from hobblebush, American beech, and striped maple 
in the forest understory. Surveys were conducted in 25 
m radius circles centered at nests. Note that these sur- 
veys only indexed the actual arthropod populations be- 
cause we did not adjust for detection probability. 
Lengths of all caterpillars and spiders detected were 
recorded; lengths were converted to biomass using 
length-mass regressions (Rodenhouse 1986). The bi- 
weekly biomass data were summed for each territory 
to produce annual estimates of food availability (see 
Rodenhouse et al. 2003). 
Predator abundance.?Abundances of eastern chip- 
munks Tamias striatus, eastern red squirrels Tamia- 
sciurus hudsonicus, and Blue Jays Cyanocitta cristata, 
the warblers' principal nest predators at Hubbard Brook 
(Holmes 1994), were surveyed with 5-min point counts 
conducted during the same four biweekly periods used 
to estimate food biomass. Like the prey abundance sur- 
veys previously described, predator surveys were not 
adjusted for detection probability and thus only indexed 
predator populations. Four counts were performed in 
each period and were then averaged to estimate the 
number of predators counted per survey per territory 
per year. 
All statistics were calculated with the IMP computer 
program (SAS Institute 2003). Model residuals were 
examined to verify that model assumptions were met, 
and only the male countersong data had to be trans- 
formed (square-root). Results are presented as untrans- 
formed, least-squares means ? 1 SE from the relevant 
model effect tests. 
RESULTS 
The neighbor-density manipulation clearly influ- 
enced Black-throated Blue Warbler reproductive suc- 
cess and behavior. However, some effects varied among 
the three years of the experiment and were counter to 
our predictions. We first present results of the density- 
reduction experiment on warbler reproductive success 
and territory size, and then the differences in warbler 
behavior and territory quality between control and re- 
duced-density territories. 
Fledging success.?As predicted, birds on reduced- 
density territories fledged significantly more young per 
year than did those on control territories (treatment F132 
2472 T. SCOTT SILLETT ET AL. Ecology, Vol. 85, No. 9 
O 
T3 
CD 
O) 
?o 
CD 
CO 
CD   0.5 
A) Young fledged 
(4) 
^Control , ^uced 
(4) 
(11) 
m i 
CD 
N 
03 
O 
,CD 
B) Territory size 
(4) 
o 
CD 04 
O 
c 0 3 
u 
o 0.2 
o 
o 0.1 
n 
u 
C) Nest predation 
(4) 
(12)       (4) 
(10) 
1997 1998 1999 
FIG. 1. Annual variation (1997-1999) in the effects of the 
density-reduction experiment on Black-throated Blue Warbler 
(A) Hedging success, (B) territory size, and (C) nest predation 
rates. Bars show least-squares means (?1 SE) from treatment 
X year effect tests (see Methods). Numbers of territories per 
treatment X year combination are given in parentheses. 
= 4.78, P = 0.04; Fig. 1A). This treatment effect did 
not vary with male age (treatment X age FU2 = 0.24, 
P = 0.63), with year (treatment X year F232 = 0.61, 
P = 0.55), or with the three-way interaction among 
independent variables (treatment X age X year F232 = 
0.11, P = 0.90). Birds in the reduced-density treatment 
therefore fledged more young than did control birds, 
regardless of male age or year, although the treatment 
effect was largest in 1997 and smallest in 1999 (Fig. 
1A). In both treatments, ASY males fledged, on av- 
erage, more young per year than did SY males (age 
F132 = 5.30, P = 0.03). Relative to controls, a greater 
proportion of reduced-density territories fledged mul- 
tiple broods due to double brooding and bigamy, with 
the largest differences in 1997 and 1998 (reduced den- 
sity, three of four territories each year; control in 1997, 
two of nine territories, in 1998, six of 12 territories, in 
1999, eight of 11 territories). Mean clutch sizes for 
both first and second brood nests per territory did not 
differ between neighbor-density treatments in any year 
(T. S. Sillett, N. L. Rodenhouse, and R. T. Holmes, 
unpublished data). 
Territory size.?Reduced-density territories were 
significantly larger than control territories (treatment 
F132 = 42.40, P < 0.0001; Fig. IB). This relationship 
did not vary by male age, by year, or by any interactions 
of independent variables (F < 1.28, P > 0.29). Mean 
territory size was smaller in each successive year of 
the experiment (year FU2 = 7.06, P = 0.003), probably 
because the number of warblers breeding on our study 
area increased from 1997 to 1999 (see Holmes and 
Sherry 2001). 
Nest predation.?As predicted, the proportion of 
nests depredated did not significantly differ between 
reduced density and control territories (treatment F, 31 
= 0.03, P = 0.87; Fig. 1C). Nest predation also did 
not significantly differ by year (F231 = 0.09, P = 0.91), 
by male age (F, 31 = 2.03, P = 0.16), or by any com- 
binations of model effects (interaction F tests < 0.80, 
P > 0.43). Variation in the proportion of nests dep- 
redated among territories, particularly in the reduced- 
density treatment, was high (Fig. 1C). 
Male countersong rate.?Contrary to our prediction, 
the neighbor-density manipulation did not have a sig- 
nificant effect on countersinging rates (treatment F, 20 
= 1.30, P = 0.27), although reduced-density males 
tended to countersing less than control males when 
females were building nests and laying eggs (Fig. 2A). 
Countersinging behavior did differ between ASY and 
SY males (age F, 20 = 7.29, P = 0.01), with older birds 
countersinging at a higher mean rate (0.57 ? 0.08 
songs/min) than yearlings (0.28 ? 0.07 songs/min), 
regardless of nest stage, year, or neighbor-density treat- 
ment (interaction F tests < 2.23, P > 0.13). Coun- 
tersinging rate was also significantly different among 
stages of the nesting cycle (stage F2 4.63, P 
0.02), being highest when females were incubating 
(Fig. 2A). Mean countersinging rates were lower in 
1999 (0.21 ? 0.07 songs/min) than in 1997 (0.65 ? 
0.06 songs/min; year FU20 = 16.30, P = 0.0006), ir- 
respective of treatment or nest stage (interaction F tests 
< 1.17, P > 0.31). Countersinging behavior did not 
vary significantly among individual males (F2031 = 
0.19, P > 0.99). 
Male foraging behavior.?The proportion of time 
that males foraged varied with neighbor-density treat- 
ment, year, and nest stage. Consistent with our predic- 
tion, males in the reduced-density treatment spent a 
significantly greater proportion of their time foraging 
than did control males (treatment F, 16 = 7.50, P = 
0.01), and this difference tended to be greatest when 
females were building nests, laying eggs, and incu- 
bating (Fig. 2B; treatment X stage F318 = 2.06, P = 
0.14). Foraging time was most similar between treat- 
ments when males were feeding nestlings and depen- 
dent fledglings (Fig. 2B). The effect of the neighbor- 
September 2004 BEHAVIORALLY MEDIATED DENSITY DEPENDENCE 2473 
en 
g) 0.6 
O 
I   0.4 
o 
o 
6   0.2 
1.0 
CD 
E   0.8 
o 
c   0.6 
8.0.4 
O 
?
   0.2 
0 
A) Song rate 
(18)    ^ 
(13) 
(6) 
Q Control 
D Redu?ed 
? density 
(15) 
(8) 
no data 
B) Foraging 
(5) 
(7) i 
(7) 
(11) 
1 
(6) 
(7) 
(7) 
Build-lay     Incubation      Nestling       Fledgling 
Nest stage 
FIG. 2. (A) Countersinging rates and (B) proportion of 
time spent foraging for male Black-throated Blue Warblers 
in the two neighbor-density treatments (control and reduced 
density). Least-squares means (?1 SE) from treatment X nest- 
stage effect tests (see Methods) are given for the nest-building 
and egg-laying, incubation, nestling-feeding, and fledgling- 
feeding stages in the warbler reproductive cycle. Numbers of 
males per treatment X nest stage combination are given in 
parentheses. 
density manipulation on foraging behavior did not dif- 
fer significantly with year (treatment X year F, 16 = 
0.37, P = 0.55) or with age class, although the diver- 
gence in the mean proportion of time that individuals 
foraged tended to be greater between ASY males (re- 
duced density, 0.75 ? 0.09; control, 0.49 ? 0.05) than 
between SY males (reduced density, 0.60 ? 0.05; con- 
trol, 0.51 ? 0.05; treatment X age F, 16 = 2.23, P = 
0.15). On average, all males foraged proportionally 
more in 1997 (0.70 ? 0.05) than in 1999 (0.47 ? 0.05; 
year F, 16 = 11.41, P = 0.004), regardless of their age 
class (year X age F, 16 = 0.0005, P = 0.98) or nesting 
stage (year X stage F3 1.15, P = 0.35). Foraging 
time differed strongly among stages of the nesting cycle 
(stage F318 = 6.32, P = 0.004), independent of male 
age (stage X age F318 = 1.03, P = 0.40). Males foraged 
least when females were building and laying, and most 
when they were feeding nestlings and dependent fledg- 
lings (Fig. 2B). No three-way interactions among mod- 
el effects were significant (interaction F tests < 1.05, 
P > 0.32), and foraging behavior did not vary signif- 
icantly among individual males (F1618  =  0.36, P = 
0.98). 
Nestling provisioning rates and fledgling mass.?Pa- 
rental provisioning rates and mean fledgling mass were 
not conclusively different between neighbor-density 
treatments in every year. Female food deliveries per 
nestling per hour differed by neighbor-density treat- 
ment and by year (treatment X year F2 9.63, F 
0.02; Fig. 3A). Reduced-density females fed nestlings 
at a higher rate than did control females in 1997 (or- 
thogonal contrast, f21 = 3.36, P = 0.003), but not in 
1998 (f21 = -0.99, P = 0.33) or 1999 (f21 = 0.04, P 
= 0.97). Male provisioning rates tended to be greater 
in the reduced-density treatment (treatment F, 21 = 
2.71, P = 0.11; Fig. 3B), but did not differ significantly 
between years (year F221 = 0.03, P = 0.97). Our data 
also did not reveal any significant annual differences 
en 
c 
V> 
w 
c 
in 
CD 
| 
CD 
cn 
c 
V> 
CD 
c 
in 
CD 
| 
CD 
A) Female provisioning 
(4) 
(6) 
| | Control 
(4)    I 
(D (8) 
Reduced 
density 
(4) 
i 
B) Male provisioning 
(4) (1) 
(4) 
i 
(6) 
(4) 
J 
\?l      -r 
II 
cn 
cn 
C) Fledgling mass 
(4) 
(D 
(4) 
(6) 
(8) 
(3) 
1997 1998 1999 
FIG. 3. Annual variation (1997-1999) in nestling provi- 
sioning rates by (A) females and (B) males, and (C) annual 
variation in fledgling mass for first-brood nests in the two 
neighbor-density treatments. Bars indicate least-squares 
means (?1 SE) from treatment X year-effect tests (see Meth- 
ods). Numbers of territories per treatment X year combination 
are given in parentheses. 
2474 T. SCOTT SILLETT ET AL. Ecology, Vol. 85, No. 9 
in the treatment effect for males (treatment X year F221 
= 0.74, P = 0.49). However, like the females, male 
deliveries per nestling per hour were most dissimilar 
between neighbor-density treatments in 1997 (Fig. 3B). 
The treatment effect on fledgling mass was marginally 
significant in the direction predicted (treatment F, 20 = 
2.63, P = 0.10; Fig. 3C). Fledgling mass did not differ 
significantly by year (F2 1.31, P = 0.27). Although 
annual differences among control and reduced-density 
territories were not statistically significant (treatment 
X year F220 = 0.89, P = 0.43), fledgling mass appeared 
to be most divergent between treatments in 1997 (Fig. 
3C). 
Territory quality.?Based on MANOVA, nest pred- 
ator abundance (control vs. reduced density, 1.03 ? 
0.10 vs. 0.98 ? 0.17 predators/5-min survey; mean ?), 
mean food biomass (104.31 ? 14.03 vs. 102.97 ? 
23.27 mg dry biomass of caterpillars and spiders/2400 
leaves), and number of deciduous leaves in the shrub 
layer (564.83 ? 18.67 vs. 592.29 ? 30.26 leaves/180 
m2 sample), respectively, did not differ significantly 
between control and reduced-density territories (G-G 
^1134310 = 0.39, P = 0.56). Territory quality differed 
among years (G-G F2274310 = 20.39, P < 0.0001) due 
to annual variation in food and predator abundance. 
However, no statistically significant treatment X year 
interaction was detected (G-G F2 0.83, P 
0.45), indicating that territory quality among control 
and reduced-density territories was similar within sea- 
sons. 
DISCUSSION 
Our experiment demonstrated that local neighbor 
density affects Black-throated Blue Warbler reproduc- 
tive output. To our knowledge, only six other published 
studies of passerines have experimentally manipulated 
intraspecific density to investigate density-dependent 
fecundity: Tompa (1967), Alatalo and Lundberg 
(1984), Torok and Toth (1988), Dhondt et al. (1992), 
Both (1998B), and Both and Visser (2000). All six were 
conducted on European cavity-nesting species that bred 
in nest boxes, and with the exception of Tompa (1967), 
all found a negative effect of density on some variable 
related to reproductive success, such as clutch size, 
fledgling survival, or territory size. However, only two 
of these studies identified the mechanisms responsible 
for density dependence. Dhondt et al. (1992) concluded 
that site preemption coupled with heterogeneity in ter- 
ritory quality (i.e., a site-dependent mechanism; Ro- 
denhouse et al. 1997) regulated their study population. 
In contrast, Both and Visser (2000) concluded that 
crowding and resource competition reduced mean ter- 
ritory size, which was, in turn, positively related to 
fledging success and adult fitness. As we will discuss, 
our results are generally consistent with a crowding 
regulatory mechanism based on intraspecific resource 
competition. 
Density-dependent fecundity is predicted to be a key 
outcome if bird territories become compressed at high 
densities and conspecifics compete primarily for food, 
rather than for nest sites (e.g., tree cavities or nest 
boxes) during the breeding season (Both and Visser 
2003). Territories in the reduced-density treatment, on 
average, were 70% larger than control territories. Den- 
sity of conspecific competitors has been shown to in- 
fluence territory size in several bird species (reviewed 
by Newton 1998). Furthermore, in studies in which 
food levels and territory sizes were quantified in each 
year, a negative correlation was found between con- 
specific density and territory size, irrespective of 
whether food availability was relatively high or low 
per unit area (Krebs 1971, Klomp 1980, Smith et al. 
1980, McCleery and Perrins 1985, Arvidsson and 
Klaesson 1986, Arcese and Smith 1988, Stamps 1990; 
T S. Sillett, N. L. Rodenhouse, and R. T Holmes, un- 
published data). In other words, an increase in con- 
specific density can reduce the size of territories, even 
when food is scarce. Because territory size can be de- 
creased by supplementing food after territory estab- 
lishment (Boutin 1990, Newton 1998), it seems that 
individuals typically defend only the area necessary to 
meet their food requirements. Therefore, a reduction 
in territory size as a result of crowding could yield 
territories that are food limited. 
The differences between neighbor-density treatments 
in male foraging behavior, in adult provisioning rates 
of nestlings, and in fledgling mass suggest that food 
limitation was greater for control birds, at least in some 
years of our study. Compared to males in the reduced- 
density treatment, control males spent less time for- 
aging in both 1997 and 1999, and thus devoted more 
time to other activities like territory defense and mate 
guarding. Although we did not quantify prey-capture 
rates of foraging males, this result suggests that crowd- 
ing at high densities may further reduce food avail- 
ability by limiting the time that males spend foraging. 
A density-dependent increase in adult interactions and 
a concomitant increase in food limitation could also be 
regulatory if they reduce parental care of young. Our 
results supported this possibility in 1997, when adults 
in the reduced-density treatment tended to provision 
nestlings at a higher rate (Fig. 3A, B) and fledge heavier 
young (Fig. 3C) relative to controls. However, results 
from 1998 were inconclusive due to small sample sizes, 
and provisioning rates and fledgling mass did not differ 
between treatments in 1999. We are unaware of any 
other published experimental studies that measure 
whether parental care or foraging behavior varies with 
population density. Additional research is therefore 
needed to understand how crowding-induced compe- 
tition for limited food acts to regulate populations of 
territorial passerines during the breeding season. 
Our experiment indicated that the proportion of nests 
depredated per territory was not influenced by the num- 
ber of conspecific neighbors, although variance among 
September 2004 BEHAVIORALLY MEDIATED DENSITY DEPENDENCE 2475 
territories was high, especially in the reduced-density 
treatment (see Results). Nevertheless, these results, 
along with those from an artificial nest experiment con- 
ducted in the shrub layer at Hubbard Brook (Reitsma 
1992), and from our long-term demographic research 
(Sillett and Holmes, in press), indicate that predation 
on Black-throated Blue Warbler nests is independent 
of conspecific density and thus does not exert a strong 
regulatory influence on this species in areas of high- 
quality, homogeneous habitat. However, other studies 
have demonstrated density-dependent rates of nest pre- 
dation (Martin 1996, Newton 1998), and nest predators 
do appear to play a role in the site-dependent regulation 
of warbler abundance at the broader spatial scale of the 
entire Hubbard Brook valley (Rodenhouse et al. 2003). 
The neighbor-density manipulation did not reveal 
density-dependent variation in countersinging rates, 
our best index of direct agonistic behavior. This is prob- 
ably due to the fact that aggression between males 
peaks during territory establishment in mid-to-late 
May, before we initiated removals. Despite this, the 
differences between density treatments in annual fe- 
cundity, in territory size, and in adult foraging and 
nestling provisioning behavior argue that either the real 
or perceived presence of neighbors did have a regu- 
latory effect on our study population. Density-depen- 
dent aggression has been documented in many bird 
species, and can affect fecundity (Watson 1970, Wil- 
liams et al. 1994) and regulate population size (Moss 
et al. 1994, Mougeot et al. 2003). Intraspecific inter- 
actions were not quantified in any of the six experi- 
mental manipulations of breeding passerine density 
previously cited. Thus, further experimental work is 
needed to determine how aggression generated by 
crowding might regulate passerine populations. 
A regulatory process such as crowding should have 
the strongest effect on abundance in years when weath- 
er and resource levels are most limiting, and the weak- 
est effect when weather is benign and resources are 
abundant (Newton 1998). Much of the annual variation 
in our experimental results probably can be attributed 
to the striking annual variation in weather and food 
availability during this study. The neighbor-density 
manipulation had the largest impact on warbler repro- 
ductive success and behavior in 1997, when environ- 
mental conditions for Black-throated Blue Warblers 
were poorest (Table 1). The late start to breeding in 
1997, coupled with lower than average food availabil- 
ity, probably amplified crowding effects. Indeed, pa- 
rental provisioning rates and fledgling mass from first- 
brood nests were lower on control territories than on 
territories in the reduced-density treatment in 1997. 
Three of four females in our reduced-density treatment 
double-brooded in this year, compared to only one of 
nine control females. The extreme amount of rain in 
June 1998 caused many Black-throated Blue Warbler 
nests to fail throughout the Hubbard Brook valley, and 
some individuals abandoned their breeding territories 
(T S. Sillett, N. L. Rodenhouse, and R. T Holmes, 
unpublished data). Several birds, especially unmated 
yearling males, appeared during June 1998 in areas 
where warbler pairs previously had been removed 
(those that established territories adjacent to experi- 
mental pairs were immediately removed). One apparent 
consequence of this late influx of new birds was that 
three of the four experimental males became bigamous 
in June 1998, and one had three females nesting on his 
territory. Thus, the high reproductive success on re- 
duced-density territories in 1998 was due to bigamy, 
not to double-brooding. Female Black-throated Blue 
Warblers paired to bigamous males usually do not dou- 
ble-brood (T S. Sillett, N. L. Rodenhouse, and R. T 
Holmes, unpublished data). Environmental conditions 
were most favorable in the 1999 breeding season (Table 
1). Males in both neighbor-density treatments spent less 
time foraging in 1999 than in 1997. Furthermore, num- 
ber of young fledged per territory, nestling provisioning 
rates, and fledgling mass differed least between density 
treatments in 1999. Thus, the variation in our experi- 
mental results among the three years suggests that: (1) 
the regulatory potential of a crowding mechanism can 
be affected by annual variation in weather and food 
abundance, and (2) crowding may have its strongest 
affect on local warbler abundance in years when en- 
vironmental conditions are relatively poor. 
In conclusion, results presented in this paper, in com- 
bination with our other recent work (Rodenhouse et al. 
2003, Sillett and Holmes, in press), indicate that the 
Black-throated Blue Warbler population at Hubbard 
Brook is regulated, in part, by breeding season events. 
In high-quality, relatively homogeneous habitat, a neg- 
ative feedback on fecundity generated by crowding- 
induced resource competition appears to be an impor- 
tant regulatory mechanism. The strength of this be- 
haviorally mediated density dependence seems to be 
sufficient to maintain abundance within the levels ob- 
served locally for at least the last three decades (Sillett 
and Holmes, in press). At a landscape scale, a site- 
dependent mechanism also appears to regulate abun- 
dance via territory-level differences in food availabil- 
ity, vegetation density, and nest predation (Rodenhouse 
et al. 2003). Therefore, multiple regulatory mecha- 
nisms are likely to operate on the Black-throated Blue 
Warbler population sampled at Hubbard Brook, but at 
different spatial scales. This finding emphasizes the 
importance of studying populations in a representative 
range of habitats and environmental conditions in order 
to understand the processes involved in regulation. 
ACKNOWLEDGMENTS 
This research was funded by the U.S. National Science 
Foundation through grants awarded to R. T. Holmes (DEB 
96-29488) and to N. L. Rodenhouse (DEB 96-33522), by a 
Sigma Xi Grant-in-Aid-of-Research to T. S. Sillett, and by 
the Cramer Fund of Dartmouth College. We are grateful to 
the many people who helped with fieldwork, especially J. 
Barg-Jones, H. Chuang, R. Dobbs, E. Mahar, P. Marra, L. 
2476 T. SCOTT SILLETT ET AL. Ecology, Vol. 85, No. 9 
Nagy, J. Osenkowski, M. Powers, M. Quinn, M. Webster, and 
B. Wright. This paper benefited from the advice and com- 
ments of M. Ayres, D. Bolger, A. Dhondt, P. Doran, M. Kery, 
P. Marra, M. McPeek, S. Morrison, L. Nagy, J. Sauer, M. 
Webster, and two anonymous reviewers. We thank the Hub- 
bard Brook Experimental Forest of the U.S. Department of 
Agriculture Northeastern Research Station for their cooper- 
ation. 
LITERATURE CITED 
Alatalo, R. V., and A. Lundberg. 1984. Density dependence 
in breeding success of the Pied Flycatcher (Ficedula hy- 
poleucd). Journal of Animal Ecology 53:969?977. 
Arcese, P., and J. N. M. Smith. 1988. Effects of population 
density and supplemental food on reproduction in Song 
Sparrows. Journal of Animal Ecology 57:119?136. 
Arcese, P., J. N. M. Smith, W M. Hochachka, C. M. Rogers, 
and D. Ludwig. 1992. Stability, regulation, and the deter- 
mination of abundance in an insular Song Sparrow popu- 
lation. Ecology 73:805-822. 
Arvidsson, B., and P. Klaesson. 1986. Territory size in a 
Willow Warbler Phylloscopus trochilus population in 
mountain birch forest in Swedish Lapland. Ornis Scandi- 
navica 17:24-30. 
Bell, G. 0., M. S. Halpert, C. F. Ropelewski, V. E. Kousky, 
A. V. Douglas, R. C. Schnell, and M. E. Gelman. 1999. 
Climate assessment for 1998. Bulletin of the American Me- 
teorological Society 80:S1?S48. 
Bonsall, M. B., T H. Jones, and J. N. Perry. 1998. Deter- 
minants of dynamics: population size, stability and persis- 
tence. Trends in Ecology and Evolution 13:174?176. 
Both, C. 1998a. Density dependence of clutch size: habitat 
heterogeneity or individual adjustment? Journal of Animal 
Ecology 67:659-666. 
Both, C. 1998b. Experimental evidence for density depen- 
dence of reproduction in Great Tits. Journal of Animal 
Ecology 67:667-674. 
Both, C, and M. E. Visser. 2000. Breeding territory size 
affects fitness: an experimental study on competition at the 
individual level. Journal of Animal Ecology 69:1021-1030. 
Both, C, and M. E. Visser. 2003. Density dependence, ter- 
ritoriality, and divisibility of resources: from optimality 
models to population processes. American Naturalist 161: 
326-336. 
Boutin, S. 1990. Food supplementation experiments with ter- 
restrial vertebrates: patterns, problems, and the future. Ca- 
nadian Journal of Zoology 68:203?220. 
Brawn, J. D., and R. P. Balda. 1988. Population biology of 
cavity nesters in northern Arizona: do nest sites limit breed- 
ing densities? Condor 90:61?71. 
Chuang-Dobbs, H. C, M. S. Webster, andR. T Holmes. 2001. 
The effectiveness of mate guarding by male Black-throated 
Blue Warblers. Behavioral Ecology 12:541-546. 
den Boer, P. J., and J. Reddingius. 1996. Regulation and 
stabilization paradigms in population ecology. Chapman 
and Hall, New York, New York, USA. 
Dhondt, A. A., B. Kempenaers, and E Adriaensen. 1992. 
Density-dependent clutch size caused by habitat hetero- 
geneity. Journal of Animal Ecology 61:643?648. 
Ferrer, M., and J. A. Donazar. 1996. Density-dependent fe- 
cundity by habitat heterogeneity in an increasing popula- 
tion of Spanish Imperial Eagles. Ecology 77:69?74. 
Ford, R. G, and J. P. Myers. 1981. An evaluation and com- 
parison of techniques for estimating home range and ter- 
ritory size. Studies in Avian Biology 6:461-465. 
Fretwell, S. D., and H. L. Lucas, Jr. 1970. On territorial 
behavior and other factors influencing habitat distribution 
in birds. Acta Biotheoretica 19:1?36. 
Goodbred, C. O., and R. T. Holmes. 1996. Factors affecting 
food provisioning of nestling Black-throated Blue War- 
blers. Wilson Bulletin 108:467-479. 
Hanski, I., and J. Tiainen. 1989. Bird ecology and Taylor's 
variance?mean regression. Annales Zoologica Fennici 26: 
213-217. 
Harrison, S., and N. Cappuccino. 1995. Using density-ma- 
nipulation experiments to study population regulation. Pag- 
es 131?147 in N. Cappuccino and P. W. Price, editors. Pop- 
ulation dynamics: new approaches and synthesis. Academic 
Press, San Diego, California, USA. 
Hixon, M. A., S. W Pacala, and S. A. Sandin. 2002. Pop- 
ulation regulation: historical context and contemporary 
challenges of open vs. closed systems. Ecology 83:1490? 
1508. 
Hochachka, W. M., and A. A. Dhondt. 2000. Density-de- 
pendent decline of host abundance resulting from a new 
infectious disease. Proceedings of the National Academy 
of Sciences (USA) 97:5303-5306. 
Holmes, R. T 1994. Black-throated Blue Warbler (Dendroica 
caerulescens). Pages 1?24 in A. Poole and F Gill, editors. 
The birds of North America. Academy of Natural Sciences, 
Philadelphia, and American Ornithologists' Union, Wash- 
ington, DC, USA. 
Holmes, R. T, P. P. Marra, and T W. Sherry. 1996. Habitat- 
specific demography of breeding Black-throated Blue War- 
blers (Dendroica caerulescens): implications for population 
dynamics. Journal of Animal Ecology 65:183?195. 
Holmes, R. T, and J. C. Schultz. 1988. Food availability for 
forest birds: effects of prey distribution and abundance on 
bird foraging. Canadian Journal of Zoology 66:720?728. 
Holmes, R. T, and T. W Sherry. 2001. Thirty-year bird pop- 
ulation trends in an unfragmented temperate deciduous for- 
est: importance of habitat change. Auk 118:589-610. 
Holmes, R. T, T W Sherry, P. P. Marra, and K. E. Petit. 1992. 
Multiple brooding and productivity of a Neotropical mi- 
grant, the Black-throated Blue Warbler (Dendroica caeru- 
lescens), in an unfragmented temperate forest. Auk 109: 
321-333. 
Jones, J., P. J. Doran, and R. T Holmes. 2003. Climate and 
food synchronize regional forest bird abundances. Ecology 
84:3024-3032. 
Klomp, H. 1980. Fluctuations and stability in Great Tit pop- 
ulations. Ardea 68:205-224. 
Krebs, C. J. 1991. The experimental paradigm and long-term 
population studies. Ibis 133(supplement I):3?8. 
Krebs, C. J. 2002. Beyond population regulation and limi- 
tation. Wildlife Research 29:1-10. 
Krebs, J. R. 1971. Territory and breeding density in the Great 
Tit, Parus major L. Ecology 52:2?22. 
Lack, D. L. 1954. The natural regulation of animal numbers. 
Clarendon Press, Oxford, UK. 
Lack, 0. L. 1966. Are bird populations regulated? New Sci- 
entist 31:98-99. 
Marra, P. P., and R. T Holmes. 1997. Avian removal exper- 
iments: do they test for habitat saturation or female avail- 
ability? Ecology 78:947-952. 
Martin, T. E. 1996. Fitness costs of resource overlap among 
coexisting bird species. Nature 380:338?340. 
May, R. M. 1999. Unanswered questions in ecology. Phil- 
osophical Transactions of the Royal Society of London Se- 
ries B, Biological Sciences 354:1951-1959. 
McCallum, H., J. Kikkawa, and C. Catterall. 2000. Density 
dependence in an island population of silvereyes. Ecology 
Letters 3:95-100. 
McCleery, R. H., and C. M. Perrins. 1985. Territory size, 
reproductive success and population dynamics in the Great 
Tit. Pages 353-373 in R. M. Sibley and R. H. Smith, ed- 
itors. Behavioural ecology: ecological consequences of 
adaptive behaviour. Blackwell, Oxford, UK. 
Milliken, G. A., and 0. E. Johnson. 1992. Analysis of messy 
data. Volume 1. Designed experiments. Van Nostrand Rein- 
hold, New York, New York, USA. 
September 2004 BEHAVIORALLY MEDIATED DENSITY DEPENDENCE 2477 
Moss, R., R. Parr, and X. Lambin. 1994. Effects of testos- 
terone on breeding density, breeding success and survival 
of Red Grouse. Proceedings of the Royal Society of London 
Series B, Biological Sciences 258:175-180. 
Moss, R., A. Watson, and R. Parr. 1996. Experimental pre- 
vention of a population cycle in Red Grouse. Ecology 77: 
1512-1530. 
Mougeot, E, S. M. Redpath, E Leckie, and P. J. Hudson. 
2003. The effect of aggressiveness on the population dy- 
namics of a territorial bird. Nature 421:737?739. 
Muller, K. E., and C. N. Barton. 1989. Approximate power 
for repeated-measures ANOVA lacking sphericity. Journal 
of the American Statistical Association 84:549?555. 
Murdoch, W. W. 1994. Population regulation in theory and 
practice. Ecology 75:271?287. 
Mylius, S. D., and O. Diekmann. 1995. On evolutionary 
stable life histories, optimization and the need to be specific 
about density dependence. Oikos 74:218?224. 
Nagy, L. R. 2002. Causes and consequences of individual 
variation in reproductive output in a forest-dwelling Neo- 
tropical migrant songbird. Dissertation. Dartmouth Col- 
lege, Hanover, New Hampshire, USA. 
Newton, I. 1998. Population limitation in birds. Academic 
Press, London, UK. 
Perrins, C. M. 1990. Factors affecting clutch size in Great 
and Blue Tits. Pages 121-130 in J. Blondel, A. Cosier, J- 
D. Lebreton, and R. H. McCleery, editors. Population bi- 
ology of passerine birds. Springer-Verlag, Berlin, Germany. 
Reitsma, L. 1992. Is nest predation density dependent? A 
test using artificial nests. Canadian Journal of Zoology 70: 
2498-2500. 
Robinson, S. K., and R. T. Holmes. 1982. Foraging behavior 
of forest birds: the relationships among search tactics, diet, 
and habitat structure. Ecology 63:1918?1931. 
Rodenhouse, N. L. 1986. Food limitation for forest passer- 
ines: effects of natural and experimental food reductions. 
Dartmouth College, Hanover, New Hampshire, USA. 
Rodenhouse, N. L., and R. T Holmes. 1992. Results of ex- 
perimental and natural food reductions for breeding Black- 
throated Blue Warblers. Ecology 73:357-372. 
Rodenhouse, N. L., T W Sherry, and R. T Holmes. 1997. 
Site-dependent regulation of population size: a new syn- 
thesis. Ecology 78:2025-2042. 
Rodenhouse, N. L., T W Sherry, and R. T Holmes. 1999. 
Multiple mechanisms of population regulation: contribu- 
tions of site dependence, crowding, and age structure. Pag- 
es 2939?2952 in Proceedings of the 22nd International Or- 
nithological Congress, Johannesburg, South Africa. 
Rodenhouse, N. L., T S. Sillett, and R. T Holmes. 2003. 
Regulation of a migratory bird population by multiple den- 
sity-dependence mechanisms. Proceedings of the Royal So- 
ciety of London Series B, Biological Sciences 270:2105? 
2110. 
Runge, M. C, and F. A. Johnson. 2002. The importance of 
functional form in optimal control solutions of problems 
in population dynamics. Ecology 83:1357-1371. 
SAS Institute. 1996. SAS/STAT software, changes and en- 
hancements through version 6.11, the MIXED Procedure. 
SAS Institute, Cary, North Carolina, USA. 
SAS Institute. 2003. IMP Version 5.0.1.2. SAS Institute, 
Cary, North Carolina, USA. 
Sillett, T. S., and R. T. Holmes. 2002. Variation in survi- 
vorship of a migratory songbird throughout its annual cy- 
cle. Journal of Animal Ecology 71:296?308. 
Sillett, T S., and R. T Holmes. In press. Long-term demo- 
graphic trends, limiting factors, and the strength of density 
dependence in a breeding population of a migratory song- 
bird. In R. S. Greenberg and P. P. Marra, editors. Birds of 
two worlds: the ecology and evolution of temperate?trop- 
ical migration. Johns Hopkins University Press, Baltimore, 
Maryland, USA. 
Sillett, T S., R. T Holmes, and T W. Sherry. 2000. Impacts 
of a global climate cycle on population dynamics of a mi- 
gratory songbird. Science 288:2040?2042. 
Sinclair, A. R. E. 1989. Population regulation in animals. 
Pages 197?241 in J. Cherrett, editor. Ecological concepts: 
the contribution of ecology to an understanding of the nat- 
ural world. 29th Symposium of the British Ecological So- 
ciety. Blackwell, Oxford, UK. 
Smith, J. N. M., R. D. Montgomerie, M. J. Taitt, and Y. Yom- 
Tov. 1980. A winter feeding experiment on an island Song 
Sparrow population. Oecologia 47:164-170. 
Stamps, J. A. 1990. The effect of contender pressure on ter- 
ritory size and overlap in seasonally territorial species. 
American Naturalist 135:614-632. 
Steele, B. B. 1993. Selection of foraging and nesting sites 
by Black-throated Blue Warblers: their relative influence 
on habitat choice. Condor 95:568?579. 
Stenning, M. J., P. H. Harvey, and B. Campbell. 1988. 
Searching for density dependence in a population of Pied 
Flycatchers, Ficedula hypoleuca Pallas. Journal of Animal 
Ecology 57:307-318. 
Tompa, F. S. 1967. Reproductive success in relation to breed- 
ing density in Pied Flycatchers, Ficedula hypoleuca. Acta 
Zoologica Fennica 118:1?28. 
Torok, J., and L. Toth. 1988. Density dependence in repro- 
duction of the Collared Flycatcher (Ficedula albicollis) at 
high population levels. Journal of Animal Ecology 57:251? 
258. 
Turchin, P. 1995. Population regulation: old arguments and 
a new synthesis. Pages 19?40 in N. Cappuccino and P. W 
Price, editors. Population dynamics: new approaches and 
synthesis. Academic Press, San Diego, California, USA. 
Turchin, P. 1999. Population regulation: a synthetic view. 
Oikos 84:153-159. 
Watson, A. 1970. Dominance, spacing behaviour and ag- 
gression in relation to population limitation in vertebrates. 
Pages 167?220 in A. Watson, editor. Animal populations 
in relation to their food resources. Blackwell, Oxford, UK. 
Williams, D. A., M. F. Lawton, and R. O. Lawton. 1994. 
Population growth, range expansion, and competition in 
the cooperatively breeding Brown Jay, Cyanocorax morio. 
Animal Behaviour 48:309-322.