Journal of Mammalogy, 82(1):1–21, 2001
IMPORTANCE OF BIOGEOGRAPHY AND ONTOGENY OF WOODY
PLANTS IN WINTER HERBIVORY BY MAMMALS
ROBERT K. SWIHART*
AND
JOHN P. BRYANT
Department of Forestry and Natural Resources, Purdue University,
West Lafayette, IN 47907-1159 (RKS)
Institute of Arctic Biology, University of Alaska, Fairbanks, AK 99775 (JPB)
Mammals can influence growth, reproduction, competitive ability, and survival of woody
plants by virtue of selective browsing and gnawing of dormant shoots during winter. Apparently in response to this type of herbivory, plants have evolved chemical and mechanical
deterrents to mammalian herbivores. We report on plant ontogeny and biogeography, which
exert their influence on herbivory at different spatiotemporal scales. To evaluate how plant
ontogeny influences herbivory, we conducted a meta-analysis of 128 studies, encompassing
37 plant and 10 mammal species, in which juvenile and mature growth stages of conspecific
plants were made available to mammals during winter in temperate and northern latitudes.
Mammals ate more of the mature-stage growth in 96% of the studies, and stage-specific
differences in consumption were very large (d1 5 2.16). Plants characterized by rapid
growth rates or low tolerances to resource limitation elicited the greatest degree of stagespecific discrimination by mammals, consistent with existing theories regarding tradeoffs
governing plant growth and defense. The influence of a plant’s growth rate and tolerance
to resource limitation was dependent on climatic regime; plants grown in areas with harsh
winter conditions tended to elicit greater discrimination of juvenile- and mature-stage
growth by mammals than plants grown in more moderate climates. Further evidence for
biogeographical variation in mammalian consumption came from 14 feeding studies, including 6 plant and 6 mammal species, that compared conspecific plants of the juvenile
growth stage either grown or collected at different localities. In 86% of the studies, extent
of herbivory by mammals varied inversely with latitude, and this yielded a moderate effect
(Z1 5 20.46, r1 5 20.53). We discuss potential roles of life history, climate, and historical
association of plants and mammals in shaping these biogeographical patterns.
Key words: biogeography, development, herbivory, latitude, mammals, secondary chemistry, winter, woody plants
In terrestrial ecosystems from temperate
to polar latitudes, mammals are the dominant group of herbivores during winter.
Several orders of mammals depend upon
woody plants as important sources of energy and nutrition in winter, relying on fermentation systems in the digestive tract to
process roots, cambium, shoots, or buds
(Robbins 1993; Van Soest 1982). Of practical concern is the damage incurred in ag-
riculture and forestry as a consequence of
winter herbivory by mammals (Conover et
al. 1995; Gill 1992; Swihart and Conover
1990). Winter herbivory by mammals also
is of interest because mammals can alter
growth, reproduction, competitive ability,
survival, and perhaps other fitness components of woody plants (Buckley et al. 1998;
Healy 1997; Kielland and Bryant 1998;
Tilghman 1989). Ultimately, feeding by
mammals can alter the composition and
successional trajectories of plant commu-
* Correspondent: [email protected]
1
2
JOURNAL OF MAMMALOGY
nities (Bryant and Chapin 1986; Kielland
and Bryant 1998; Pastor and Naiman 1992;
Ritchie et al. 1998).
Browsing mammals during winter tend
to be polyphagous, yet they exhibit clear
preference and avoidance of certain plants,
that is, they are selective generalists (Bryant and Kuropat 1980; Swihart and Yahner
1983). Most descriptive studies of winter
diets of mammalian herbivores have examined variation in consumption of several
species of plants from a single locality. In
those studies relating dietary choices to
chemical characteristics of plants, differential consumption of plants seldom seems to
be based solely on energy or nutrient content (Basey et al. 1990; Rousi 1990). Moreover, mammals often respond negatively to
secondary metabolites when making dietary
choices (Bryant and Kuropat 1980; Palo
and Robbins 1991).
In light of the potential impacts of mammalian herbivory, an important goal of
managers and ecologists is to identify factors that influence dietary choices. These
factors typically function as constraints, either on use of plants by herbivores or on
defense of plants from herbivory. Constraints are manifested as genetic tradeoffs
resulting from physiologic, ecologic, and
evolutionary processes (Herms and Mattson
1992). These processes span an array of
scales in space (e.g., microclimate, geographic range) and time (e.g., bite rate, history of species associations). Mammals
seem capable of exerting considerable selective pressure on woody plants to evolve
defenses to deter winter herbivory. Indeed,
heritable, intraspecific variation in deterrent
capabilities has been documented for several mammal–woody plant interactions
(Chiba and Nagata 1976; Dimock et al.
1976; Rousi 1990; Rousi et al. 1991, 1997;
Silen et al. 1986). We examined the importance of plant ontogeny and biogeography,
which exert their influence on herbivory at
different spatiotemporal scales.
Ontogeny as used here refers to tightly
regulated developmental changes that take
Vol. 82, No. 1
place during a plant’s life. Maturation or
phase change refers to relatively rapid and
predictable ontogenetic changes occurring
early in the life cycle of a woody plant and
characterizing its transition to a sexually reproducing adult (Kozlowski 1971). A similar but slightly broader concept that incorporates physiologic alterations during plant
development has been termed the developmental stream (Kearsley and Whitham
1998). Developmentally based changes in
plant–herbivore interactions have received
relatively little attention from ecologists,
but results in well-studied systems indicate
the need for closer examination. Experiments have demonstrated strong effects of
plant development on insect herbivores of
cotton (Gossypium hirsutum—Karban and
Thaler 1999) and cottonwoods (Populus
sp.—Kearsley and Whitham 1989, 1998).
Other studies have demonstrated strong,
chemically based effects of plant development on winter herbivory of hares (Lepus)
in subarctic taiga (Bryant 1981a, 1981b;
Bryant et al. 1985).
Basic life-history considerations and empirical studies of plant population dynamics
suggest that selection for plant defense
against herbivores is greatest in the juvenile
(pre–sexually reproductive, sensu Kozlowski 1971) stage (Watkinson 1986). Although
the notion is not new that plant developmental stage may influence investment in
defense (Bryant 1981a; Bryant et al.
1983a), the generality and strength of its
effect on mammalian herbivory has not
been examined. We reviewed published
comparisons of stage-specific differences in
consumption of winter-dormant shoots by
mammals, combined these with results
from our unpublished work over the last
several years, and used meta-analysis to test
the null hypothesis that plant ontogeny has
no influence on patterns of consumption of
woody plants. We also tested whether selected physiologic, morphologic, and lifehistory attributes explain additional variation in consumption by mammals.
Even less is known about geographic
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
ecology of mammalian herbivory. Attempts
to generalize findings on winter dietary
choices of mammals from 1 locality to another have met with limited success (Swihart and Yahner 1983; Wolff 1978). In addition to obvious effects due to differences
in relative availability of plants at different
localities, we believe that constraints operating on resident plants and mammals could
differ across large spatial scales. We explored this possibility in 2 ways. First, we
tested whether geographic differences in origins of woody plants, categorized according to classes of winter severity, explained
significant levels of variation in discrimination by mammals when offered shoots
from plants at different stages of ontogeny.
Second, we reviewed published studies of
feeding experiments in which juvenilestage growth was offered from conspecific
plants either growing in different geographic areas or grown in a common environment
from seed collected in different areas. We
then tested for a latitudinal trend in those
data using meta-analysis and proposed possible explanations for patterns that we observed.
MATERIALS
AND
METHODS
Sources of data.—Data for testing patterns of
consumption as related to plant ontogeny were
derived from studies in which mammalian herbivores were provided with a choice of winterdormant juvenile- and mature-stage growth, and
consumption of each type was measured subsequently. Although details varied among studies,
2 general protocols were followed. In some studies, juvenile and mature growth forms were presented in equal (or nearly equal) amounts to
mammals in cafeteria-style feeding trials involving several plant species simultaneously (e.g.,
Klein 1977). In other studies, juvenile and mature growth were presented simultaneously to
mammals in trials involving a single plant species (e.g., Bryant et al. 1985). Of the 2 protocols,
the latter provided a less variable assay of discrimination by mammals as a function of plant
ontogeny because interspecific influences on
consumption were absent. Studies also differed
in whether they relied upon captive or free-rang-
3
ing test subjects. When captive subjects were
used, each usually received its own plant material, and consumption thus reflected per capita
use, subject to the constraints of interpreting results from trials with captive mammals. When
free-ranging mammals were used, .1 individual
could potentially visit any given test station provisioned with juvenile- and mature-stage
growth. Under free-ranging conditions, consumption values did not necessarily reflect per
capita use. Moreover, independence of stations
cannot be assured under free-ranging conditions,
although most studies spaced stations far enough
apart to minimize the likelihood that a single
mammal would visit .1 station during a test period (e.g., Bryant 1981b).
Our review of the literature provided suitable
data from 19 published studies on relative consumption of juvenile- and mature-stage growth
of winter-dormant woody plants by mammals;
those studies included feeding trials involving
82 combinations of mammals and woody plants
(Table 1). We supplemented those data with previously unpublished feeding trials with which
we have been involved over the past 2 decades.
Those trials involved 46 combinations of mammals and woody plants; most of those trials tested plant species singly. Thus, our total database
for examining patterns of consumption related
to plant ontogeny included 128 sets of trials,
conducted with 37 different species of plants
and 10 species of mammals (Table 1).
For each set of feeding trials, we also categorized woody plants according to growth rate
(slow, medium, fast), tolerance to resource limitation (principally to limitation of light: low,
medium, high), growth form (shrub ,2 m tall,
shrub .2 m tall, tree), and leaf type (deciduous,
evergreen). Classifications were made after consulting Fowells (1965), Harlow and Harrar
(1969), Loehle (1988), Tutin et al. (1964, 1968),
and Viereck and Little (1972). For species not
covered by those sources (usually shrubs or trees
of little commercial value), levels of categories
were assigned based on our personal knowledge
or after consulting forest ecologists from the region where the species of plant was tested. We
further categorized sets of trials according to the
digestive strategy of the herbivore involved
(foregut or hindgut fermentation). Finally, for
each feeding experiment, we classified the geographic locality from which the plant material
was collected according to severity of winter
Acer rubrum
A. rubrum
Alnus crispa
A. crispa
A. crispa
A. crispa
A. crispa
A. crispa
A. crispa
A. crispa
A. crispa
Alnus incana
Betula alleghaniensis
B. alleghaniensis
B. alleghaniensis
B. alleghaniensis
Betula glandulosad
Betula lenta
B. lenta
B. lenta
B. lenta
B. lenta
Betula papyrifera
B. papyrifera
B. papyrifera
B. papyrifera
Betula pendula
B. pendula
B. pendula
B. pendula
Betula pubescens
Plant
Lepus americanus
Odocoileus virginianus
Clethrionomys rutilus
Dicrostonyx groenlandicus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
Microtus pennsylvanicus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanusc
L. americanus
L. americanus
L. americanus
L. americanus
L. americanusc
O. virginianus
L. americanus
L. americanus
L. americanus
L. americanusc
Alces alces
Lepus timidus
L. timidus
Microtus agrestis
A. alces
Mammal
M-L-D-T-M
M-L-D-T-M
M-H-D-H-V
M-H-D-H-V
M-H-D-H-V
M-H-D-H-V
M-H-D-H-V
M-H-D-H-V
M-H-D-H-V
M-H-D-H-V
M-H-D-H-V
M-H-D-H-V
M-M-D-T-M
M-M-D-T-M
M-M-D-T-M
M-M-D-T-M
M-H-D-L-V
F-M-D-T-M
F-M-D-T-M
F-M-D-T-M
F-M-D-T-M
F-M-D-T-M
F-L-D-T-M
F-L-D-T-M
F-L-D-T-M
F-L-D-T-M
M-M-D-T-M
M-M-D-T-S
M-M-D-T-S
M-M-D-T-M
M-M-D-T-M
Categories of
plant attributes
0.81
0.49
1.91
4.27
2.73
1.56
14.82
NA
NA
3.63
1.52
11.64
0.08
1.16
20.04
3.98
4.18
0.16
0.38
20.12
2.22
10.97
1.14
1.69
0.98
14.38
1.43
6.38
3.65
0.61
0.60
di
0.43
0.35
0.15
0.11
0.03
0.19
0.19
0.48
0.19
0.20
0.21
0.17
0.49
0.39
0.50
0.31
0.10
0.47
0.48
0.52
0.39
0.00
0.44
0.09
0.33
0.05
0.20
0.13
0.18
0.38
0.29
Xj/(Xj
1 Xm)
6
4
10
10
5
25
30
5
10
4
10
30
7
10
8
8
30
7
5
7
8
2
5
7
9
8
13
10
10
12
12
nj
6
4
10
10
5
25
30
5
10
4
10
30
7
10
8
8
30
7
5
7
8
2
5
7
9
8
17
10
10
12
15
nm
Current study
Current studya
Current studyb
Current studyb
Bryant (1981b)
Bryant et al. (1983b)
Current study
Clausen et al. (1986)
Clausen et al. (1986)
Klein (1977)
Current studyb
Current study
Swihart et al. (1994)
Swihart et al. (1994)
Swihart et al. (1994)
Current study
Current study
Swihart et al. (1994)
Swihart et al. (1994)
Swihart et al. (1994)
Current study
Current studya
Swihart et al. (1994)
Swihart et al. (1994)
Swihart et al. (1994)
Current study
Danell et al. (1990)
Bryant et al. (1991)
Bryant et al. (1989)
Danell et al. (1987)
Danell et al. (1990)
Source
TABLE 1.—Feeding experiments used to test hypotheses regarding effects of plant ontogeny on mammalian herbivory during winter. Acronyms
for categories of plant attributes are listed sequentially as follows: growth rate (F 5 fast, M 5 medium, S 5 slow)–tolerance to stress (L 5 low,
M 5 medium, H 5 high)–leaf type (D 5 deciduous, E 5 evergreen)–growth form (T 5 tree, H 5 shrub .2 m, L 5 shrub ,2 m)–severity of
winter temperatures (V 5 very severe, S 5 severe, M 5 moderate, Mi 5 mild). See text for discussion of variables used in meta-analysis. NA 5
not available.
4
JOURNAL OF MAMMALOGY
Vol. 82, No. 1
B. pubescens
B. pubescens
Betula resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera
B. resinifera 3 glandulosa
Juniperus virginiana
Kalmia latifolia
Larix laricinad
Picea glauca
P. glauca
P. glauca
P. glauca
P. glauca
P. glauca
P. glauca
P. glauca
Picea mariana
P. mariana
P. mariana
Pinus radiata
Pinus sylvestris
Populus balsamifera
P. balsamifera
P. balsamifera
Plant
L. timidus
M. agrestis
L. timidus
A. alces
C. rutilus
D. groenlandicus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
M. pennsylvanicus
L. americanus
O. virginianus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
Lepus californicus
L. timidus
C. rutilus
D. groenlandicus
L. americanus
Mammal
M-M-D-T-S
M-M-D-T-M
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-H-V
S-L-E-T-M
S-M-E-H-M
M-H-D-T-V
M-H-E-T-V
M-H-E-T-V
M-H-E-T-V
M-H-E-T-V
M-H-E-T-V
M-H-E-T-V
M-H-E-T-V
M-H-E-T-V
S-H-E-T-V
S-H-E-T-V
S-H-E-T-V
F-M-E-T-Mi
M-M-E-T-S
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
Categories of
plant attributes
1.82
2.76
6.45
NA
1.70
371.90
19.11
339.54
8.22
2.35
8.40
46.71
11.63
1.82
2.50
9.38
2.26
0.07
11.05
NA
7.26
1.92
10.72
16.64
1.95
7.60
1.92
2.51
2.43
1.42
NA
NA
2.37
1.71
4.86
di
TABLE 1.—Continued.
0.29
0.16
0.00
0.07
0.02
0.00
0.08
0.03
0.07
0.01
0.02
0.01
0.12
0.00
0.02
0.05
0.18
0.49
0.38
0.18
0.19
0.24
0.06
0.04
0.10
0.10
0.11
0.39
0.32
0.18
0.14
0.00
0.02
0.03
0.02
Xj/(Xj
1 Xm)
10
10
10
20
10
10
5
30
5
10
10
100
4
8
10
30
7
6
30
5
30
4
3
3
6
3
4
5
30
4
49
5f
10
10
5
nj
10
10
10
20
10
10
5
30
3
10
10
100
8
8
10
30
7
6
30
5
30
4
3
3
8
2
4
5
30
4
49
5f
10
10
5
nm
Bryant et al. (1991)
Danell et al. (1987)
Bryant et al. (1989)
Reichardt et al. (1984)
Current studyb
Current studyb
Bryant (1981b)
Current study
Klein (1977)
Reichardt et al. (1984)
Reichardt et al. (1984)
Reichardt et al. (1984)
Current studye
Current study
Current studyb
Current study
Swihart and Picone (1998)
Current study
Current study
Bryant (1981b)
Current study
Klein (1977)
Sinclair and Smith (1984)
Sinclair and Smith (1984)
Current studye
Current studye
Current studye
Bryant (1981b)
Current study
Klein (1977)
Libby and Hood (1976)
Rousi et al. (1987)
Current studyb
Current studyb
Bryant (1981b)
Source
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
5
P. balsamifera
P. balsamifera
B. balsamifera
B. balsamifera
B. balsamifera
B. balsamifera
B. balsamifera
Populus grandidentata
P. grandidentata
P. grandidentata
P. grandidentata
Populus tremuloides
P. tremuloides
P. tremuloides
P. tremuloides
P. tremuloides
P. tremuloides
P. tremuloides
P. tremuloides
Pseudotsuga menziesii
P. menziesii
Salix alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
S. alaxensis
Salix arbusculoides
Plant
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
M. Pennsylvanicus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
Odocoileus hemionus
O. hemionus columbianus
C. rutilus
D. groenlandicus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. timidus
M. pennsylvanicus
L. americanus
Mammal
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-M
F-L-D-T-M
F-L-D-T-M
F-L-D-T-M
F-L-D-T-V
F-L-D-T-V
F-L-D-T-V
F-L-D-T-M
F-L-D-T-M
F-L-D-T-M
F-L-D-T-M
F-L-D-T-V
F-M-E-T-S
F-M-E-T-Mi
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
F-L-D-H-V
M-M-D-H-V
Categories of
plant attributes
12.42
0.01
11.06
6.28
7.72
2.46
0.75
20.44
20.56
1.04
20.31
7.24
41.20
13.28
0.75
2.68
0.30
2.06
2.41
1.53
NA
12.05
1.92
3.49
21.93
7.60
3.43
20.18
17.29
11.81
0.00
7.09
1.42
4.49
0.27
di
TABLE 1.—Continued.
0.05
0.48
0.04
0.01
0.02
0.16
0.02
0.58
0.58
0.28
0.55
0.03
0.21
0.15
0.34
0.05
0.44
0.17
0.25
0.01
0.34
0.13
0.34
0.10
0.04
0.32
0.02
0.25
0.06
0.11
0.20
0.06
0.50
0.29
0.35
Xj/(Xj
1 Xm)
30
4
6
3
3
5
10
8
5
5
7
5
30
5
7
4
9
5
2
4
8
10
10
10
10
5
9
30
3
3
7
10
10
10
3
nj
30
4
6
3
3
6
10
8
5
5
7
5
30
3
7
4
9
5
3
4
8
10
10
10
10
5
9
30
3
3
8
8
10
10
3
nm
Current study
Klein (1977)
Reichardt et al. (1990)
Sinclair and Smith (1984)
Sinclair and Smith (1984)
Current studye
Current studyb
Swihart et al. (1994)
Swihart et al. (1994)
Swihart et al. (1994)
Swihart et al. (1994)
Bryant (1981b)
Current study
Klein (1977)
Swihart et al. (1994)
Swihart et al. (1994)
Swihart et al. (1994)
Swihart et al. (1994)
Current studye
Dawson et al. (1990)
Silen et al. (1986)
Current studyb
Current studyb
Bryant et al. (1985)
Bryant et al. (1985)
Bryant et al. (1985)
Bryant et al. (1989)
Current study
Sinclair and Smith (1984)
Sinclair and Smith (1984)
Current studye
Current studye
Bryant et al. (1989)
Current studyb
Current studye
Source
6
JOURNAL OF MAMMALOGY
Vol. 82, No. 1
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. americanus
L. timidus
L. timidus
L. timidus
L. americanus
L. americanus
L. americanus
M. agrestis
L. timidus
L. timidus
L. americanus
L. timidus
M. agrestis
L. americanus
L. timidus
L. timidus
L. timidus
L. americanus
M. agrestis
L. americanus
O. virginianus
Mammal
M-M-D-H-V
M-M-D-H-V
M-L-D-H-V
M-L-D-H-V
M-L-D-H-V
M-M-D-H-V
F-L-D-T-S
F-L-D-T-S
F-L-D-T-S
F-L-D-T-S
M-M-D-H-V
M-M-D-H-V
F-L-D-H-V
F-L-D-H-M
F-L-D-H-S
F-L-D-H-S
M-M-D-H-V
F-L-D-T-S
F-L-D-T-M
M-L-D-L-S
M-L-D-L-S
M-L-D-L-S
M-L-D-L-S
M-M-D-L-V
M-M-D-T-M
S-H-E-T-M
S-H-E-T-M
Categories of
plant attributes
8.39
1.66
NA
44.30
17.12
10.22
8.08
2.01
9.12
16.86
11.32
6.02
14.16
1.77
8.26
1.81
12.60
10.49
1.23
1.06
3.20
10.49
5.43
15.84
0.47
3.01
1.56
di
0.07
0.17
0.04
0.05
0.10
0.06
0.22
0.34
0.25
0.18
0.08
0.26
0.09
0.21
0.10
0.41
0.07
0.06
0.32
0.40
0.14
0.11
0.15
0.15
0.43
0.14
0.23
Xj/(Xj
1 Xm)
b
Trials conducted in conjunction with P. Picone.
Trials conducted in conjunction with H. Henttenon.
c Trials conducted using hares in Alaska.
d Tolerance ratings to stress were based on nutrients for B. glandulosa and L. laricina, and to light for all other species.
e Trials conducted in conjunction with J. Cary and C. Krebs.
f Approximate number of hares present.
a
S. arbusculoides
S. arbusculoides
Salix bebbiana
S. bebbiana
S. bebbiana
Salix brachycarpa
Salix caprea
S. caprea
S. caprea
S. caprea
Salix glauca
S. glauca
Salix lasiandra
Salix myrsinifolia phylicifolia
Salix nigricans
S. nigricans
Salix novae-anglieae
Salix pentandra
S. pentandra
Salix phylicifolia
S. phylicifolia
S. phylicifolia
S. phylicifolia
Salix planifolia
Sorbus aucuparia
Tsuga canadensis
T. canadensis
Plant
TABLE 1.—Continued.
nj
30
2
5
30
2
30
9
2
10
3
3
3
30
9
3
2
30
3
10
9
10
3
2
30
20
6
6
nm
30
2
5
30
2
30
9
2
10
3
3
3
30
9
3
2
30
3
10
9
10
3
2
30
20
6
6
Source
Current study
Klein (1977)
Bryant (1981b)
Current study
Klein (1977)
Current study
Bryant et al. (1989)
Bryant et al. (1991)
Bryant et al. (1989)
Tahvanainen et al. (1985)
Sinclair and Smith (1984)
Sinclair and Smith (1984)
Current study
Danell et al. (1987)
Tahvanainen et al. (1985)
Tahvanainen et al. (1985)
Current study
Tahvanainen et al. (1985)
Danell et al. (1987)
Bryant et al. (1989)
Bryant et al. (1989)
Tahvanainen et al. (1985)
Tahvanainen et al. (1985)
Current study
Danell et al. (1987)
Current study
Swihart and Picone (1998)
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
7
8
JOURNAL OF MAMMALOGY
Vol. 82, No. 1
TABLE 2.—Summary of studies reporting winter herbivory by mammals on woody plants originating from .1 geographic locality. Data from $3 localities were required for a study to be included
in the meta-analysis testing overall strength of the correlation between latitude and extent of herbivory, and $4 localities were required for inclusion in the meta-analysis using Fisher’s Z-transformation.
In some studies, it was not possible to calculate r (or Z); for those studies, we merely report authors’
tests of significant differences for consumption of conspecific plants from different localities. NS 5
not significant.
Plant
Betula pendula
B. pendula
B. pendula
B. pendula
B. pendula
Pinus contorta
P. contorta
P. contorta
Pinus ponderosa
P. ponderosa
P. ponderosa
Pinus sylvestris
P. sylvestris
Populus grandidentata
Populus tremuloides
P. tremuloides
P. tremuloides
Mammal
Lepus timidus
L. timidus
L. timidus
L. timidus
Microtus agrestis
M. agrestis
Alces alces
A. alces
Lepus
Odocoileus
Lepus californicus
Alces alces
A. alces
Lepus americanus
L. americanus
L. americanus
L. americanus
ri
Zi
20.54
20.60
20.94
1.74
20.85
21.26
20.86
21.29
NS
20.44
20.47
P , 0.01
P ø 0.10
20.36
20.38
20.62
20.72
0.38
0.40
20.63
20.74
20.51
20.56
20.93
20.99
20.99
20.99
temperatures (very severe, severe, moderate,
mild). Categories reflected a gradient from long,
cold winters (very severe) to short, warm winters (mild). Generally, sites with more severe
winter temperatures occurred at higher latitudes,
with reductions in winter severity associated
with lower latitudes and maritime climates.
Additional information on geographic variation in palatability of woody plants to mammals
was obtained from studies that monitored consumption of a single growth stage (juvenile) collected from conspecific plants at .1 source.
Those studies varied somewhat in terms of design and sampling procedures, but generally
they took 1 of 2 forms. In some feeding trials,
mammals were presented with shoots of plants
collected from different geographic localities
and hence exposed to different environmental
regimes during development (e.g., Bryant et al.
1994). In the remaining trials, mammals chose
among plants of different geographic origins but
grown from seed in a common environment
(e.g., Niemelä et al. 1989). Both captive and
free-ranging mammals were used as test subjects. Our review of the literature provided data
from 9 published studies, including trials involving 6 plant and 6 mammalian species (Table 2).
n
Source
4
4
4
4
2
25
35
16
10
10
17
29
30
3
3
3
3
Rousi et al. (1989)
Rousi et al. (1989)
Rousi et al. (1989)
Rousi et al. (1991)
Rousi (1988)
Hansson (1985)
Hansson (1985)
Hansson (1985)
Squillace and Silen (1962)
Squillace and Silen (1962)
Read (1971)
Niemelä et al. (1989)
Niemelä et al. (1989)
Swihart et al. (1994)
Swihart et al. (1994)
Bryant et al. (1994)
Bryant et al. (1994)
Meta-analysis.—Meta-analysis refers to a set
of statistical methods that enables comparison
and synthesis of results of multiple studies (Gelber and Goldhirsch 1991; Gelber et al. 1992;
Hedges 1992). Ecologic and evolutionary studies have begun to use meta-analytic techniques
(Bender et al. 1998; Côté and Poulin 1995; Gurevitch et al. 1992; Hamilton and Poulin 1997;
Møller and Thornhill 1998), and meta-analysis
can improve rigor of reviews and syntheses
(Arnqvist and Wooster 1995; Gurevitch and
Hedges 1993). Meta-analysis was chosen because data were compiled from multiple sources
that used similar techniques but analyzed data
and reported results differently (Rosenberg et al.
1997). We were interested in whether regularities existed in the direction of differences reported for multiple sets of trials that collectively
indicated a strong underlying process.
Procedures for meta-analysis are covered
elsewhere (Gurevitch and Hedges 1993; Gurevitch et al. 1992; Hedges and Olkin 1985). After
gathering data from multiple studies that address
a common question, meta-analysis tends to advance sequentially (Arnqvist and Wooster 1995).
First, results of studies were transformed to a
common scale, called effect size, which repre-
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
sented the magnitude and sign of the effect of
interest. By necessity, studies of the effect of
plant ontogeny on mammalian herbivory involve a comparison of consumption of 2 growth
stages (j 5 juvenile, m 5 mature). To test whether mammals discriminate between conspecific
juvenile- and mature-stage growth of plants, we
computed an effect size, d (Hedges and Olkin
1985), for each study in which it was possible
to obtain information on mean consumption of
each growth stage, standard deviations about
each mean, and sample sizes (Table 1). The effect size, di, for the ith set of trials was given by
J(Xm 2 Xj)/s, where Xm was the mean measure
of consumption of mature-stage growth, Xj was
the mean consumption of juvenile-stage growth,
and s was the pooled standard deviation of the
juvenile and mature groups (Hamilton and Poulin 1997; Rosenberg et al. 1997). J was used to
correct for bias caused by small samples and
was given by Hedges and Olkin (1985) as J 5
1 2(3/[4k 2 1]), where k 5 nj 1 nm 2 2, and
nj and nm represented the samples for juvenileand mature-stage growth, respectively. Effect
size was considered small if di , 0.2, moderate
if di 5 0.5, large if di . 0.8, and very large if
di . 1.0 (Cohen 1969). The variance in di was
vi 5 [(nj 1 nm)/njnm] 1 [di2/2(nj 1 nm)].
Next, we combined effect sizes from individual sets of trials into an overall effect size. To
compute a mean effect size across all t sets of
trials (i 5 1, . . . , t), we computed d1 5 Swidi/
Sdi, where wi was the weight of the reciprocal
of vi, which gave greater weight to sets of trials
with larger samples and, presumably, more precise results (Gurevitch et al. 1992; Hedges and
Olkin 1985). The variance of d1 was v1 5 1/
Swi. Confidence intervals (95%) for d1 were d1
6 1.96v1.
To test for differences in effect sizes among
sets of trials grouped by plant growth rate, tolerance to limiting resources (hereafter termed
stress tolerance), growth form, leaf type, mammalian digestive strategy, or winter severity, we
used MetaWin (Rosenberg et al. 1997). We expressed the result of each set of trials for this
part of the analysis as a response ratio, RR,
where RR 5 Xj/Xm. In meta-analysis, effect size
typically is computed for ln(RR) rather than RR.
An advantage of ln(RR) as a measure of effect
size was that, when combined with resampling
tests (see below), it did not require knowledge
of intraclass variation in resampling (Rosenberg
9
et al. 1997), which permitted us to increase the
number of studies used in some comparisons
(Table 1). Thus, ln(RR) was used in statistical
analyses. Each set of trials received a nonparametric weighting, wi 5 njnm/(nj 1 nm) (Adams et
al. 1997; Rosenberg et al. 1997). We used
mixed-model analyses because they do not require that all sets of trials within a particular
category (e.g., fast-growing plant species) share
a common, true effect size (Rosenberg et al.
1997). For each analysis, we computed a measure of the variation in mean effect size between
categories (e.g., between sets of trials conducted
using evergreen and deciduous leaf types),
termed QB* (Rosenberg et al. 1997). Larger values of QB* reflect greater differences in mean
effect sizes between categories. Adams et al.
(1997) urged ecologists to use resampling methods when evaluating the significance of main effects in meta-analysis studies, thereby avoiding
problems associated with distributional assumptions of parametric tests. To test the null hypothesis that effect sizes did not differ among
categories, randomization tests consisting of
4,999 iterations were used to derive a null distribution for QB*, from which a P-value was obtained. We also used MetaWin to calculate biascorrected bootstrap confidence intervals (95%)
to facilitate comparisons (Adams et al. 1997;
Dixon 1993; Rosenberg et al. 1997).
Because some of our main effects for plant
ontogeny were correlated traits (e.g., growth rate
and leaf type), we conducted our tests according
to the following criteria. First, each main effect
was tested as described above. Main effects
yielding at least marginally significant differences (P # 0.15) in ln(RR1) among categories were
subsequently subjected to fine-scale analysis by
cross-classifying pairs of categories and repeating the meta-analysis. No category was included
in an analysis if it contained #5 sets of trials.
When samples were sufficient, we also examined cross-classified pairs of categories for individual genera of woody plants (Betula, Populus, Salix).
A slightly different form of meta-analysis was
used to examine whether geographically based
differences in mammalian herbivory existed
among juvenile-stage growth of conspecific
woody plants (Table 2). Based on previous work
(Swihart et al. 1994), we hypothesized that latitudinal gradients in palatability would exist,
with northern plants being less palatable than
10
JOURNAL OF MAMMALOGY
more southern conspecifics. To test that hypothesis, we used procedures described by Côté and
Poulin (1995). For study i, we computed the correlation, ri, between the latitude from which a
plant originated and the extent of herbivory.
Next, we computed an overall correlation coefficient, r1, which weighted each ri by its sample
size, Ni: r1 5 SNiri/SNi. Observed population
variance was computed as vr 5 S[Ni(ri 2 r1)2]/
SNi. Observed variance was partitioned into the
true population variance and variance due to
sampling error. Variance due to sampling error,
ve, was approximated as ve 5 (1 2 r12)2/(N 2 1),
where N is the average sample size across studies. The true population standard deviation, s,
was then estimated as (vr 2 ve)½. If the ri are
normally distributed, then a test of the null hypothesis that r1 5 0 can be made using Z 5 r1/
s. Because the correlation coefficient can exhibit
undesirable statistical properties, particularly
when samples are small, we also conducted a
second meta-analysis on these data after applying Fisher’s Z-transformation (Rosenberg et al.
1997): Zi 5 {ln[(1 1 ri)/(1 2 ri)]}/2. We used
Zi as effect sizes, with weights of Ni 2 3 (Rosenberg et al. 1997), to derive an overall effect
size, Z1.
Meta-analysis has many advantages over
qualitative summaries and ‘‘vote counting’’ procedures (Arnqvist and Wooster 1995; Hedges
and Olkin 1985; Hunter et al. 1982). In particular, meta-analysis is useful because it provides
an estimate of the magnitude of an effect that
takes into account differences in sample size
among studies, and it provides greatly improved
control of type II error rates (Arnqvist and Wooster 1995). However, meta-analysis is not without its limitations. One concern is that the probability of a type I error could increase if studies
with nonsignificant results were published less
frequently than studies with significant results.
A plot of sample size versus effect size failed to
suggest that selective reporting occurred in our
data (Palmer 1999). We doubt that a publication
bias exists because data that formed the basis for
our analyses usually represented only a fraction
of the data reported in any given publication,
and they often were presented ancillary to a paper’s central theme. In addition, 36% of the data
sets in Table 1 represented unpublished work
that clearly had not been subjected to such bias.
A 2nd concern is that researchers might select
subjects for study because they anticipate deriv-
Vol. 82, No. 1
ing significant results (Adams et al. 1997). Research bias did not likely play a role in the majority of studies used in our meta-analyses because choice of species often was governed by
factors, such as economic importance or availability, that are unrelated to hypotheses we tested. Another concern is a possible lack of independence of sets of trials included in a metaanalysis, which could result when multiple tests
are reported from single studies (Arnqvist and
Wooster 1995). We could not rule out that possibility because several studies provided .1 effect size for use in our analyses and some species were represented disproportionately often.
However, all effect sizes were independent in
the sense that they were derived from different
combinations of test subjects, plant species, and
localities.
RESULTS
Plant selection based on ontogeny.—If
mammals do not discriminate between juvenile and mature growth stages of conspecific plants during winter, then each growth
stage should be eaten in equal proportion.
However, relative consumption nearly always favored mature-stage growth; juvenile-stage growth comprised ,50% of the
average total consumption by mammals in
123 of 128 trials (96%). In the majority of
feeding experiments, consumption of juvenile-stage growth comprised ,20% of the
total (Fig. 1). Effect sizes showed a similar
pattern; di was ‘‘very large’’ (Cohen 1969)
for 99 of the 120 experiments (82%) for
which it could be calculated (Fig. 1). The
overall effect size for the studies was d1 5
2.16, with a 95% CI of 2.03–2.29. When
presented with a choice between juvenileand mature-stage winter-dormant growth,
mammals avoid the juvenile stage.
When feeding experiments were categorized by plant growth rate, there was a tendency for differential discrimination to occur between juvenile and mature growth
stages (QB* 5 6.27, P 5 0.062). Overall
response ratios exhibited a gradient, with
the greatest level of discrimination by
mammals occurring between juvenile and
mature growth stages of fast-growing plants
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
FIG. 1.—A) Distributions illustrating the proportion of total consumption of juvenile-stage
growth in trials in which mammalian herbivores
were presented with approximately equal
amounts of juvenile- and mature-stage growth;
note the preponderance of studies for which
,50% of total consumption consisted of juvenile-stage growth. B) Distribution of effect sizes,
d, from Table 1; designations regarding effect
sizes follow Cohen (1969).
(RR1 5 0.11, n 5 72, CI 5 0.07–0.16), an
intermediate amount occurring for plants
with medium growth rates (RR1 5 0.19, n
5 49, CI 5 0.12–0.25), and the least discrimination between ontogenetic stages occurring for slow-growing plants (RR1 5
0.36, n 5 7, CI 5 0.24–0.56). An analogous pattern was seen for experiments involving Betula (QB* 5 12.94, P 5 0.004),
with mammals exhibiting greater discrimination between ontogenetic stages when offered fast-growing species (RR1 5 0.04, n
5 22, CI 5 0.02–0.09) than when offered
species with medium growth rates (RR1 5
11
0.35, n 5 12, CI 5 0.14–0.87). Discrimination of ontogenetic stages by mammals
did not differ as a function of growth rate
when Salix was presented in feeding experiments (QB* 5 2.72, P 5 0.12).
When experiments were categorized according to stress tolerance of the plants involved, a marginal tendency for differential
discrimination occurred (QB* 5 5.18, P 5
0.108). Plants with low tolerance ratings
elicited the greatest discrimination in their
juvenile and mature growth stages when
presented to mammals (RR1 5 0.11, n 5
74, CI 5 0.07–0.16), whereas proportionately twice as much juvenile-stage growth
was consumed by mammals offered plants
with medium (RR1 5 0.21, n 5 30, CI 5
0.10–0.35) or high (RR1 5 0.20, n 5 24,
CI 5 0.14–0.27) tolerance ratings.
Mammals did not discriminate between
ontogenetic stages as a function of either
leaf type (QB* 5 0.04, P 5 0.832) or plant
growth form (QB* 5 1.94, P 5 0.582). Categorizing mammals by type of fermentation
system also failed to reveal any additional
variation in response ratios (QB* 5 0.17, P
5 0.693), although only 9 feeding experiments involved mammals with foregut fermentation (Table 1).
Interactive effects of winter severity and
plant life history.—To examine more closely how mammalian herbivory varies as a
function of plant ontogeny, we partitioned
feeding experiments according to winter severity and plant growth rate. When mammals were presented with juvenile and mature growth stages of fast-growing plants,
they discriminated differentially as a function of winter severity (QB* 5 16.04, P 5
0.001). Fast-growing plants from areas of
moderate winter temperatures elicited less
discrimination between juvenile and mature
growth stages than fast-growing plants from
areas of severe or very severe winter temperatures (Fig. 2). A similar pattern of discrimination was evident when mammals
were presented with plants characterized by
medium growth rates (QB* 5 15.57, P 5
0.003), except that the lowest RR1 occurred
12
JOURNAL OF MAMMALOGY
FIG. 2.—Response ratios (RR1 5 J/M) measuring the discrimination exhibited by mammalian herbivores between juvenile- and maturestage growth of winter-dormant woody plants.
Response ratios are categorized by plant growth
rate and severity of winter temperatures; colder
winters elicit greater discrimination against juvenile-stage growth by mammals. Vertical lines
are 95% bias-corrected bootstrap confidence intervals based on 4,999 samples.
for feeding trials of plants from areas with
severe winter temperatures (Fig. 2). Qualitatively, mammals did not seem to discriminate among ontogenetic stages of slowgrowing plants on the basis of winter severity (moderate, RR1 5 0.41, n 5 5; very
severe, RR1 5 0.43, n 5 3).
We also partitioned feeding experiments
according to winter severity and stress tolerance of plants. Insufficient samples were
available to permit analysis of experiments
on plants from areas with mild or severe
winter temperatures, or for plants characterized by high stress tolerance (Table 1).
Mammals exhibited differential discrimination of juvenile and mature growth stages
Vol. 82, No. 1
FIG. 3.—Response ratios (RR1 5 J/M) measuring discrimination exhibited by mammalian
herbivores between juvenile- and mature-stage
growth of winter-dormant woody plants. Response ratios are categorized by plant tolerance
to limiting resources and severity of winter temperatures; colder winters elicit greater discrimination against juvenile-stage growth by mammals. Vertical lines are 95% bias-corrected bootstrap confidence intervals based on 4,999 samples.
when presented with plants of low (QB* 5
18.14, P 5 0.0002) and medium stress tolerance (QB* 5 11.32, P 5 0.011). For both
tolerance classes of plants, mammals discriminated between ontogenetic stages to
the greatest extent when plants were grown
in areas with very severe winter temperatures (Fig. 3).
Biogeographic patterns of mammalian
herbivory.—Categorizing plants according
to severity of winter temperatures where
they were grown revealed strong differential discrimination in consumption of juvenile-stage and mature-stage growth (QB* 5
24.59, P 5 0.0002). Overall response ratios
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
exhibited a gradient with respect to winter
severity. Mammals discriminated greatly
between juvenile and mature growth stages
from areas with very severe (RR1 5 0.09,
n 5 76, CI 5 0.06–0.13) and severe winter
temperatures (RR1 5 0.13, n 5 16, CI 5
0.04–0.26). However, a nearly 4-fold increase in relative consumption of juvenilestage growth occurred in areas experiencing
moderate winter temperatures (RR1 5 0.40,
n 5 34, CI 5 0.25–0.55). Although an insufficient number of experiments (n 5 2)
were conducted in mild localities to warrant
inclusion in the analysis, they exhibited
qualitative agreement with our other results
(RR1 5 0.28). Betula (QB* 5 39.06, P 5
0.0002) and Populus (QB* 5 16.90, P 5
0.0008) followed that pattern, although
only plants from areas with moderate and
very cold winters were subjected to testing.
When mammals were offered Betula from
areas with very severe winter temperatures,
they were more discriminating (RR1 5
0.02, n 5 14, CI 5 0.01–0.04) than when
offered Betula from areas with moderate
winters (RR1 5 0.34, n 5 17, CI 5 0.15–
0.58). Similarly, when mammals were offered Populus from areas with very severe
winter temperatures, they were much more
discriminating (RR1 5 0.06, n 5 14, CI 5
0.03–0.12) than when offered Populus from
areas with moderate winters (RR1 5 0.53,
n 5 8, CI 5 0.23–0.93). Experiments with
Salix were conducted only using plants
from areas with very severe and severe
winter temperatures, and no difference in
discrimination of ontogenetic stages was
noted (QB* 5 0.92, P 5 0.34).
Intensity of browsing was negatively correlated with latitude in 12 (86%) of the 14
studies (Table 2). Overall correlation between browsing intensity and latitude was
20.53, yielding a Z of 21.53 (P 5 0.066).
For Fisher’s Z-transformed correlations, we
obtained a moderate (Cohen 1969) effect
size of 20.46, which differed from the zero
predicted by the null hypothesis of no relationship (CI 5 20.62 to 20.30).
13
DISCUSSION
Plant ontogeny and cost of herbivory.—
In the absence of herbivory, competition
between plants is the major factor determining resource allocation patterns in a given environment (Tilman 1990). Relative
costs of herbivory are magnified in the juvenile stage of ontogeny for woody plants
because a given level of herbivory removes
a proportionately greater amount of a juvenile plant’s biomass and growing points
compared with a larger, reproductively mature conspecific. Herbivory also can have a
greater impact on a plant’s fitness if it occurs in the juvenile stage and subsequently
diminishes (or precludes) lifetime reproductive success via, for example, mortality before maturation, a delay in age at 1st reproduction, or a decline in competitive ability
(Bulmer 1994; Prins and Nell 1990; Roughgarden 1998; Stearns 1992). As a consequence of the potentially greater cost of
herbivory in the prereproductive stage,
plants should invest relatively more in defense of this growth stage. For woody
plants in which the chemical basis of palatability to mammals has been determined,
juvenile-stage growth is more heavily defended than mature-stage growth, usually
by lipid-soluble secondary metabolites of
low molecular weight (Bryant 1981a; Clausen et al. 1986; Reichardt et al. 1984,
1990). In our analysis, mammals strongly
and consistently chose mature-stage growth
over juvenile-stage growth. Mammals discriminate among plants on the basis of plant
chemistry, morphology, and texture and
avoid ingestion of plants or plant parts that
contain high levels of defensive substances
(Palo and Robbins 1991). Our analysis provides overwhelming support for the idea
that palatability of plants to mammals differs as a function of ontogenetic stage in a
manner consistent with an evolutionary response of plants to stage-specific differences in costs of herbivory (Fig. 1). Moreover,
because those experiments were conducted
on winter-dormant woody plants, it seems
14
JOURNAL OF MAMMALOGY
highly unlikely that insects or pathogens
could have caused the response we observed. Rather, mammals likely are the evolutionary force generating these stage-specific differences.
Effects of plant life history.—Mammalian
discrimination between juvenile- and mature-stage growth varied somewhat as a
function of plant growth rate and tolerance
to resource stress. We suspect that these
patterns are a consequence of differences in
responses to resource-rich and resourcepoor environments. In resource-rich environments, competition among plants is intense, and success is dependent on rapid acquisition of resources and on allocation of
these resources to vegetative structures
needed for subsequent acquisition of additional resources (Coley et al. 1985; Herms
and Mattson 1992). As a result, plants
adapted to resource-rich environments typically exhibit fast rates of growth (Herms
and Mattson 1992). In resource-poor environments, competition among plants may
be less intense, and success is dependent on
efficient use and retention of resources
(Chapin 1980; Tilman 1990; Vitousek
1982). Consequently, plants adapted to resource-poor environments exhibit inherently slow rates of growth (Chapin et al. 1989;
Coley et al. 1985). Not surprisingly, plant
growth rate is correlated inversely with tolerance to resource limitation (Shipley and
Keddy 1988).
Plants adapted to resource-rich environments typically are more nutritious and thus
are more valuable to mammalian herbivores
than plants from resource-poor environments (Bryant and Kuropat 1980; Coley et
al. 1985). Herbivory presents an added dilemma for growth-dominated plants in rich
environments because beyond the direct
costs associated with lost tissue and nutrients, reduced competitive ability in the juvenile stage could reduce future reproductive success. Moreover, growth-dominated
plants exhibit considerably greater phenotypic flexibility than stress-tolerant plants
(Herms and Mattson 1992). We suspect that
Vol. 82, No. 1
plants characterized by fast growth rates
and low tolerances to resource limitation
(i.e., plants facing intense competition in resource-rich environments) should invest
relatively more in defense of the juvenile
stage than slow-growing, stress-tolerant
plants. Consistent with this expectation,
when feeding experiments were categorized
by plant growth rate, mammals exhibited
the greatest degree of ontogenetic discrimination for fast growers and the least discrimination for slow growers. Similarly,
when feeding experiments were categorized
by plant tolerance rating, mammals were
most discriminating when presented with
juvenile and mature growth stages of lowtolerance plants and least discriminating
when offered ontogenetic stages of plants
with either medium or high tolerance ratings. Overall differences in discrimination
between ontogenetic stages, as measured
using RR1, indicated a 2- to 3-fold change
for the extreme classes of growth rate and
tolerance rating.
Constraints of winter severity.—Discrimination of ontogenetic stages by mammals
varied as a function of winter severity, even
after correcting for differences in plant
growth rate (Fig. 2) or tolerance rating (Fig.
3). Mammals tended to consume relatively
less juvenile-stage growth when occurring
in areas with severe or very severe winter
temperatures.
Greater discrimination in colder climates
may result from more stringent constraints
for mammals faced with a trade-off between energetic needs and detoxification
capabilities for ingested toxins. Energetic
costs of thermoregulation during winter
typically increase for a mammal of a given
size as temperature declines (Lindstedt and
Boyce 1985; Moen 1973; Robbins 1993).
Although costs can be partially offset by
changes in behavior (e.g., movement, huddling), insulative quality of pelage, or reduced metabolism, it seems likely that in
many instances increased costs also must be
offset by increased food intake. An exception may be larger species that can rely to
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
a greater degree on fat reserves during winter. Nonetheless, sites with colder winter
temperatures in temperate and northern latitudes are associated with shorter growing
seasons; thus, mammalian herbivores from
these areas are forced to rely on dormant
woody plants as a primary food source for
a longer period each year. Finally, herbivores from northern latitudes typically must
rely on a food base that is less diverse than
at more southern latitudes, which can limit
dietary breadth.
Juvenile-stage growth of woody plants
often contains elevated concentrations of
secondary metabolites that can elicit acute,
subacute, or chronic toxicosis (Bryant et al.
1992; Harju 1996a, 1996b; MacArthur et al.
1991; Reichardt et al. 1984). Detoxification
of plant secondary metabolites is essential
to maintenance of acid–base homeostasis in
mammals, but it is a saturable process (Foley et al. 1995). If intake of secondary metabolites is great enough to saturate a mammal’s system for excretion of organic anions, deleterious effects on protein metabolism and sodium balance result (Foley et
al. 1995; Iason and Palo 1991; Illius and
Jessop 1995). Increases in dietary energy
and protein apparently can ameliorate effects of toxins, at least in ruminants (Burritt
et al. 2000; Wang and Provenza 1997).
When faced with few plant species from
which to choose, such nutrient supplementation seems unlikely for herbivores in
northern latitudes.
We suggest that mammals in harsh climates have 2 options for dealing with constraints imposed by their greater rate of
food intake and need to avoid toxicosis.
They must either forage more selectively
with respect to plant secondary metabolites
or evolve more effective means of detoxification. Our results are consistent with the
former prediction, although discrimination
in our analysis also might result from climatic constraints on plants. Additional,
stronger support for the influence of winter
climate on discrimination comes from 3
feeding experiments in which reciprocal
15
feeding trials were conducted, and 3 species
of Betula grown in a moderate climate were
provided to snowshoe hares (Lepus americanus) inhabiting interior Alaska (Table 1).
In all cases, overall level of discrimination
by hares in Alaska was greater than any of
9 sets of trials for the same plant species
provided to hares inhabiting the more moderate climate of Connecticut (Swihart et al.
1994).
Mammals are equipped with genetic,
physiologic, and neurologic capabilities for
learning and remembering quality of potential food items (Provenza 1995, 1996).
Moreover, they prefer the flavor or odor of
nutritious foods (Villabla and Provenza
1996), yet they limit their intake of nutritious but toxic foods and thereby avoid saturating their detoxification system (Pfister
et al. 1997; Wang and Provenza 1996a,
1996b, 1997). Our data do not permit us to
assess whether mammals in harsher climates have developed more effective detoxification mechanisms. However, evidence
exists for Lepus that physiologic capabilities for detoxification vary (Iason and Palo
1991).
Latitudinal trends and evolution of plant
defense.—Although constraints imposed by
geographic variation in winter severity
seem to influence levels of ontogenetically
based discrimination of woody plants by
mammals, this explanation is insufficient to
explain geographic variation in patterns of
consumption of conspecific, juvenile-stage
growth (Table 2). In these studies, mammals at a single locality were either presented with plants grown on site from seed
collected at several different localities, or
offered plants that had been collected as juveniles from several localities. Yet our analysis indicates a moderate effect size for latitude, with mammals consuming proportionately more biomass of plants of southern origin. This effect likely has a
substantial genetic component, because
plants were grown in common environments for 67% of the studies exhibiting inverse relationships between latitude and
16
JOURNAL OF MAMMALOGY
palatability. Moreover, comparison of herbivory by snowshoe hares on juvenile-stage
growth of closely related species (Populus
grandidentata, P. tremuloides) grown in
common environments repeatedly has
shown that hares consume less of P. tremuloides, the species with the more northern geographic range (Swihart et al. 1994).
Large-scale clinal variation in palatability
could result from evolutionary responses of
plants to $2 selective forces associated
with cold climates. Although numerous exceptions exist, soil temperatures in northern
latitudes generally are colder than in southern latitudes of the same altitude. Colder
soils reduce the rate at which nutrients are
mineralized from organic matter and hence
made available for uptake by plants (Chapin and Shaver 1985). Consequently, coldsoil plants growing in otherwise comparable conditions could become more nutrient
limited (Swihart et al. 1994). Nutrient limitation constrains rate of plant growth and
thus simultaneously serves to increase cost
of a unit of herbivory (Bryant et al. 1983a;
Coley et al. 1985) and diminish opportunity
costs associated with investing in defenses
at the expense of growth (Chapin 1989;
Herms and Mattson 1992). Greater levels of
defense against herbivory could evolve,
then, as a consequence of nutrient limitation
associated with cold soils. Although harsh
winter conditions also can lead to physiologic changes in plant responses to dessication and winter hardening, these changes
do not explain patterns of herbivory that we
observed (Swihart et al. 1994).
Growing seasons are shorter at more
northern latitudes, which forces northern
herbivorous mammals to rely on dormant
plants as sources of food for longer periods
of time than their southern counterparts. For
an identical complement of mammals, then,
plants in northern latitudes will experience
more herbivore-days of browsing pressure
during winter. Greater productivity in the
more southern latitudes of the temperate
zone could ameliorate this selective force
by yielding greater densities of herbivores.
Vol. 82, No. 1
However, the lower species richness and
productivity of woody plant communities
typifying taiga and subarctic regions presumably would elevate selective pressure
exerted on any given plant. Moreover, in
northern latitudes with long winters, populations of arvicoline rodents and lagomorphs associated with specialized predators and relatively unfragmented habitat
seem more prone to cyclical dynamics of
high amplitude (Bjørnstad et al. 1995; Hanski and Korpimaki 1995; Keith 1990; Keith
et al. 1993). During peak years, herbivory
can cause widespread mortality of seedlings
and saplings (Wolff 1980). Thus, some evidence exists that latitudinal variation in
plant defense could arise as a direct consequence of latitudinal variation in intensity
of winter herbivory.
Geographic variation in defense.—Effect
sizes for latitude were not large, and other
factors unrelated to latitude clearly influence geographic variation in mammalian
herbivory and plant defense. We were unable to examine effects of altitude in our
analysis, but we suspect that plants adapted
to montane environments exhibit defensive
characteristics similar to plants grown in
northern latitudes. Indirect evidence is consistent with this notion. Hansson (1985) rated damage by field voles (Microtus agrestis) in Sweden to lodgepole pine (Pinus
contorta) collected from a variety of localities in western North America. For plants
from 25 localities, our rank correlation of
latitude and damage was 20.44 (Table 2).
Hansson (1985) noted that northern provenances were less damaged than southern
ones, with 1 dramatic exception. Voles only
weakly damaged the most southern provenance, which came from a high-altitude locality. Removal of this locality from the
analysis strengthened the latitudinal correlation by 41%, to 20.62.
Geographic variation in the historical intensity of herbivory also may be unrelated
to latitude. Bryant et al. (1989) documented
dramatically different responses of hares to
congeneric woody plants; plants from an
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
area with no history of browsing mammals
were least defended chemically and consumed in greatest quantities by hares, plants
from an area with a long history of browsing mammals but hare populations that did
not cycle were intermediate in defense and
palatability, and plants from areas with a
long history of browsing mammals and
populations of hares with high-amplitude
cycles were least palatable and exhibited
the greatest levels of chemical defense. Intraspecifically, only 2 studies in our analysis failed to show an inverse relation between extent of mammalian herbivory and
latitude (Table 2). In 1 of these, Read
(1971) recorded browsing by black-tailed
jackrabbits (Lepus californicus) in Nebraska on ponderosa pine (Pinus ponderosa)
seedlings originating from 17 localities
throughout the western United States. He
documented a significant difference in herbivory as a function of longitude of origin;
plants with origins west of the Continental
Divide were browsed more severely. Interestingly, the intermountain area west of the
Divide has no history of intense herbivory
by large mammalian herbivores, in contrast
to land east of the Divide (Mack and
Thompson 1982). Thus, even though
Read’s (1971) study did not yield an inverse
latitudinal gradient in herbivory, its findings
are consistent with mammalian herbivory
as a force influencing geographic patterns
of defense in woody plants.
CONCLUSIONS
We reviewed the literature and documented patterns in the consumption of winter-dormant woody plants by mammalian
herbivores that reflected small- and largescale variation in mammal–plant interactions. Only by examination of a broader
spectrum of mammalian herbivores and
woody plants will the generality of our
findings be determined. In particular, comparative studies of ruminants are needed.
Based on available information, at any particular locality mammals perceive significant intraspecific and intraplant differences
17
in palatability. These differences are ontogenetically based, they often are large compared with many of the interspecific differences that we and others have observed for
plants of the same ontogenetic stage, and
they seem to have evolved principally as
defensive responses of plants to mammalian
herbivory during winter.
The degree to which our findings generalize to other seasons, biomes, and taxa remains to be seen. Few studies have examined effects of plant developmental stage on
herbivorous birds in northern latitudes
(Muller 1995; Ryala 1966; Svoboda and
Gullion 1972), but results thus far have
been consistent with our findings. Insect
herbivores also seem sensitive to developmental differences in plants during the
growing season, but their responses vary interspecifically and the number of studies is
small (Karban and Thaler 1999; Kearsley
and Whitham 1989, 1998; Waltz and Whitham 1997).
Large-scale spatial variation also is evident and seems to be driven largely by winter severity. Physiologic and ecologic constraints imposed by harsh environments in
northern latitudes seem to have resulted in
evolution of plants with better defenses and
mammalian herbivores with more discriminating palates than their counterparts from
more moderate latitudes. Moreover, these
selective forces have generated a latitudinal
pattern that seems to differ from patterns of
geographic variation reported for insect
(Coley and Aide 1991; Levin 1976) and
aquatic (but see Steinberg 1986; Targett et
al. 1992) systems. Thus, any general theory
formulated to predict large-scale patterns in
plant defense against herbivory will have to
reconcile these disparate findings.
ACKNOWLEDGMENTS
We thank F. S. Chapin III, W. J. Foley, S. H.
Jenkins, L. K. Page, F. D. Provenza, K. Raffa,
K. Schwaegerle, H. P. Weeks, Jr., and an anonymous reviewer for helpful comments on the
manuscript. K. Danell and L. Ericson provided
rankings of life-history categories for several
18
JOURNAL OF MAMMALOGY
species of European plants. Support was provided by Purdue University and the Institute of Arctic Biology, University of Alaska. This is Purdue
University Agricultural Research Programs
manuscript 15811.
LITERATURE CITED
ADAMS, D. C., J. GUREVITCH, AND M. S. ROSENBERG.
1997. Resampling tests for meta-analysis of ecological data. Ecology 78:1277–1283.
ARNQVIST, G., AND D. WOOSTER. 1995. Meta-analysis:
synthesizing research findings in ecology and evolution. Trends in Ecology and Evolution 10:236–
240.
BASEY, J. M., S. H. JENKINS, AND G. C. MILLER. 1990.
Food selection by beavers in relation to inducible
defenses of Populus tremuloides. Oikos 59:57–62.
BENDER, D. J., T. A. CONTRERAS, AND L. FAHRIG. 1998.
Habitat loss and population decline: a meta-analysis
of the patch size effect. Ecology 79:517–533.
BJøRNSTAD, O. N., W. FALCK, AND N. C. STENSETH.
1995. A geographic gradient in small rodent density
fluctuations: a statistical modelling approach. Proceedings of the Royal Society of London B 262:
127–133.
BRYANT, J. P. 1981a. Phytochemical deterrence of
snowshoe hare browsing by adventitious shoots of
four Alaskan trees. Science 313:889–890.
BRYANT, J. P. 1981b. The regulation of snowshoe hare
feeding behavior during winter by plant antiherbivore chemistry. Pp. 720–731 in Proceedings of the
1st International Lagomorph Conference (K. Meyers, ed.). University of Guelph, Guelph, Ontario,
Canada.
BRYANT, J. P., AND F. S. CHAPIN III. 1986. Browsing–
woody plant interactions during boreal forest plant
succession. Pp. 313–325 in Forest ecosystems in the
Alaskan taiga (K. Van Cleve, F. S. Chapin III, P. W.
Flanagan, L. A. Viereck, and C. T. Dyrness, eds.).
Springer-Verlag, New York.
BRYANT, J. P., F. S. CHAPIN III, AND D. R. KLEIN. 1983a.
Carbon/nutrient balance of boreal plants in relation
to vertebrate herbivory. Oikos 40:357–368.
BRYANT, J. P., K. DANELL, F. D. PROVENZA, P. B. REICHARDT, AND T. P. CLAUSEN. 1991. Effects of mammal browsing upon the chemistry of deciduous
woody plants. Pp. 135–154 in Phytochemical induction by herbivores (D. W. Tallamy and M. J. Raup,
eds.). John Wiley & Sons, New York.
BRYANT, J. P., AND P. KUROPAT. 1980. Selection of winter forage by subarctic browsing vertebrates: the role
of plant chemistry. Annual Review of Ecology and
Systematics 11:261–285.
BRYANT, J. P., F. D. PROVENZA, P. B. REICHARDT, AND
T. P. CLAUSEN. 1992. Woody plant–mammal interactions. Pp. 343–370 in Herbivores: their interactions with secondary plant metabolites (G. A. Rosenthal and M. Berenbaum, eds.). 2nd ed. Academic
Press, New York.
BRYANT, J. P., R. K. SWIHART, P. B. REICHARDT, AND
L. NEWTON. 1994. Biogeography of woody plant
chemical defense against snowshoe hare browsing:
Vol. 82, No. 1
comparison of Alaska and eastern North America.
Oikos 70:385–394.
BRYANT, J. P., J. TAHVANAINEN, M. SULKINOJA, R. JULKUNEN-TIITTO, P. B. REICHARDT, AND T. GREEN. 1989.
Biogeographic evidence for the evolution of chemical defense against mammal browsing by boreal
birch and willow. The American Naturalist 134:20–
34.
BRYANT, J. P., G. D. WIELAND, T. P. CLAUSEN, AND P.
KUROPAT. 1985. Interactions of snowshoes hares and
feltleaf willow (Salix alaxensis) in Alaska. Ecology
66:1564–1573.
BRYANT, J. P., G. D. WIELAND, P. B. REICHARDT, V. E.
LEWIS, AND M. C. MCCARTHY. 1983b. Pinosylvin
methyl ether, a snowshoe hare antifeedant isolated
from green alder (Alnus crispa) resin. Science 222:
1023–1025.
BUCKLEY, D. S., T. L. SHARIK, AND J. G. ISEBRANDS.
1998. Regeneration of northern red oak: positive and
negative effects of competitor removal. Ecology 79:
65–78.
BULMER, M. 1994. Theoretical evolutionary ecology.
Sinauer Associates, Inc., Publishers, Sunderland,
Massachusetts.
BURRITT, E. A., R. E. BANNER, AND F. D. PROVENZA.
2000. Sagebrush ingestion by lambs: effects of experience and macronutrients. Journal of Range Management 53:91–96.
CHAPIN, F. S., III. 1980. The mineral nutrition of wild
plants. Annual Review of Ecology and Systematics
11:233–260.
CHAPIN, F. S., III. 1989. The cost of tundra plant structures: evaluation of concepts and currencies. The
American Naturalist 133:1–19.
CHAPIN, F. S., III, R. H. GROVES, AND L. T. EVANS.
1989. Physiological determinants of growth rate in
response to phosphorous supply in wild and cultivated Hordeum species. Oecologia 79:96–105.
CHAPIN, F. S., III, AND G. R. SHAVER. 1985. Arctic. Pp.
16–40 in Physiological ecology of North American
plant communities (B. F. Chabot and H. A. Mooney,
eds.). Chapman and Hall, New York.
CHIBA, S., AND Y. NAGATA. 1976. Studies on the breeding of Larix species. Experimental estimations of
vole resistance and heritability among hybrids, backcrosses and triple hybrids. Oji Institute of Forest
Tree Improvement Technical Note 3:33–44 (in Japanese, English summary).
CLAUSEN, T. P., J. P. BRYANT, AND P. B. REICHARDT.
1986. Defense of winter-dormant green alder against
snowshoe hares. Journal of Chemical Ecology 12:
2117–2131.
COHEN, J. 1969. Statistical power analysis for the behavioral sciences. Academic Press, New York.
COLEY, P. D., AND T. M. AIDE. 1991. A comparison of
herbivory and plant defenses in temperate and tropical broad-leaved trees. Pp. 25–49 in Plant–animal
interactions: evolutionary ecology in tropical and
temperate regions (P. W. Price, T. M. Lewinsohn, G.
W. Fernandes, and W. W. Benson, eds.). John Wiley
& Sons, New York.
COLEY, P. D., J. P. BRYANT, AND F. S. CHAPIN III. 1985.
Resource availability and plant antiherbivore defense. Science 230:895–899.
CONOVER, M. R., W. C. PITT, K. K. KESSLER, T. J.
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
DUBOW, AND W. A. SANBORN. 1995. Review of human injuries, illnesses, and economic losses caused
by wildlife in the United States. Wildlife Society
Bulletin 23:407–414.
CÔTÉ, I. M., AND R. POULIN. 1995. Parasitism and
group size in social animals: a meta-analysis. Behavioral Ecology 6:159–165.
DANELL, K., R. BERGSTROM, AND K. DIRKE. 1990.
Moose browsing on juvenile and adult birches (Betula pendula and Betula pubescens): test of a hypothesis on chemical defence. Proceedings of the International Union of Game Biologists 16:400–407.
DANELL, K., T. ELMQVIST, L. ERICSON, AND A. SALMONSON. 1987. Are there general patterns in barkeating voles on different shoot types from woody
plants? Oikos 50:396–402.
DAWSON, R. J., H. M. ARMLEDER, AND M. J. WATERHOUSE. 1990. Preferences of mule deer for Douglasfir foliage from different-sized trees. The Journal of
Wildlife Management 54:378–382.
DIMOCK, E. J., II, R. R. SILEN, AND V. E. ALLEN. 1976.
Genetic resistance in Douglas-fir to damage by
snowshoe hare and black-tailed deer. Forest Science
22:106–121.
DIXON, P. M. 1993. The bootstrap and jacknife: describing the precision of ecological indices. Pp.
290–318 in Design and analysis of ecological experiments (S. M. Scheiner and J. Gurevitch, eds.).
Chapman and Hall, New York.
FOLEY, W. J., S. MCLEAN, AND S. J. CORK. 1995. Consequences of biotransformation of plant secondary
metabolites on acid–base metabolism in mammals—
a final common pathway? Journal of Chemical Ecology 21:721–743.
FOWELLS, H. A. 1965. Silvics of forest trees of the
United States. United States Department of Agriculture Handbook 271:1–762.
GELBER, R. D., A. S. COATES, AND A. GOLDHIRSCH.
1992. Meta-analysis. The fashion of summing-up
evidence. 2. Interpretations and uses. Annals of Oncology 3:683–691.
GELBER, R. D., AND A. GOLDHIRSCH. 1991. Meta-analysis. The fashion of summing-up evidence. 1. Rationale and conduct. Annals of Oncology 2:461–468.
GILL, R. M. A. 1992. A review of damage by mammals
in north temperate forests: 3. Impacts on trees and
forests. Forestry 65:363–388.
GUREVITCH, J., AND L. V. HEDGES. 1993. Meta-analysis:
combining the results of independent experiments.
Pp. 378–398 in Design and analysis of ecological
experiments (S. M. Scheiner and J. Gurevitch, eds.).
Chapman and Hall, New York.
GUREVITCH, J., L. L. MORROW, A. WALLACE, AND J. S.
WALSH. 1992. A meta-analysis of field experiments
on competition. The American Naturalist 140:539–
572.
HAMILTON, W. J., AND R. POULIN. 1997. The Hamilton
and Zuk hypothesis revisited: a meta-analytical approach. Behaviour 134:299–320.
HANSKI, I., AND E. KORPIMAKI. 1995. Microtine rodent
dynamics in northern Europe: parameterized models
for the predator–prey interaction. Ecology 76:840–
850.
HANSSON, L. 1985. Damage by wildlife, especially
small rodents, to North American Pinus contorta
19
provenances introduced into Sweden. Canadian
Journal of Forestry Research 15:1167–1171.
HARJU, A. 1996a. Effects of birch (Betula pendula)
bark and food protein level on root voles (Microtus
oeconomus): I. Food consumption, growth, and mortality. Journal of Chemical Ecology 22:709–718.
HARJU, A. 1996b. Effects of birch (Betula pendula)
bark and food protein level on root voles (Microtus
oeconomus): II. Detoxification capacity. Journal of
Chemical Ecology 22:718–728.
HARLOW, W. M., AND E. S. HARRAR. 1969. Textbook
of dendrology. 5th ed. McGraw-Hill, New York.
HEALY, W. M. 1997. Influence of deer on the structure
and composition of oak forests in central Massachusetts. Pp. 249–266 in The science of overabundance:
deer ecology and population management (W. J.
McShea, H. B. Underwood, and J. H. Rappole, eds.).
Smithsonian Institute Press, Washington, D.C.
HEDGES, L. V. 1992. Meta-analysis. Journal of Educational Statistics 17:277–278.
HEDGES, L. V., AND I. OLKIN. 1985. Statistical methods
for meta-analysis. Academic Press, New York.
HERMS, D. A., AND W. J. MATTSON. 1992. The dilemma
of plants: to grow or defend. Quarterly Review of
Biology 67:283–335.
HUNTER, J. E., F. L. SCHMIDT, AND G. B. JACKSON.
1982. Meta-analysis. Cumulating research findings
across studies. Sage Publications, Beverly Hills, California.
IASON, G., AND R. T. PALO. 1991. Effects of birch phenolics on a grazing and a browsing mammal: a comparison of hares. Journal of Chemical Ecology 17:
1733–1743.
ILLIUS, A. W., AND N. S. JESSOP. 1995. Modeling metabolic costs of allelochemical ingestion by foraging
herbivores. Journal of Chemical Ecology 21:693–
719.
KARBAN, R., AND J. S. THALER. 1999. Plant phase
change and resistance to herbivory. Ecology 80:
510–517.
KEARSLEY, M. J. C., AND T. G. WHITHAM. 1989. Developmental changes in resistance to herbivory: implications for individuals and populations. Ecology
70:1040–1047.
KEARSLEY, M. J. C., AND T. G. WHITHAM. 1998. The
developmental stream of cottonwoods affects ramet
growth and resistance to galling aphids. Ecology 79:
178–191.
KEITH, L. B. 1990. Dynamics of snowshoe hare populations. Pp. 119–196 in Current mammalogy (H.
H. Genoways, ed.). Plenum Press, New York 2:1–
577.
KEITH, L. B., S. E. M. BLOOMER, AND T. WILLEBRAND.
1993. Dynamics of a snowshoe hare population in
fragmented habitat. Canadian Journal of Zoology
71:1385–1392.
KIELLAND, K., AND J. P. BRYANT. 1998. Moose herbivory in taiga: effects on biogeochemistry and vegetation dynamics in primary succession. Oikos 82:
377–383.
KLEIN, D. R. 1977. Winter food preferences of snowshoe hares (Lepus americanus) in Alaska. Proceedings of the International Congress of Game Biologists 13:266–275.
20
JOURNAL OF MAMMALOGY
KOZLOWSKI, T. T. 1971. Growth and development of
trees. Academic Press, New York 1:1–443.
LEVIN, D. A. 1976. The chemical defenses of plants to
pathogens and herbivores. Annual Review of Ecology and Systematics 7:121–159.
LIBBY, W. J., AND J. V. HOOD. 1976. Juvenility in
hedged radiata pine. Acta Horticultura 56:91–98.
LINDSTEDT, S. L., AND M. S. BOYCE. 1985. Seasonality,
fasting endurance, and body size in mammals. The
American Naturalist 125:873–878.
LOEHLE, C. 1988. Tree life history strategies: the role
of defenses. Canadian Journal of Forestry Research
18:209–222.
MACARTHUR, C., A. E. HAGERMAN, AND C. T. ROBBINS.
1991. Physiological strategies of mammalian herbivores against plant defenses. Pp. 103–114 in Plant
chemical defenses against mammalian herbivory (R.
T. Palo and C. T. Robbins, eds.). CRC Press, Boca
Raton, Florida.
MACK, R. N., AND J. N. THOMPSON. 1982. Evolution in
steppe with few large, hooved mammals. The American Naturalist 119:757–773.
MOEN, A. N. 1973. Wildlife ecology: an analytical approach. W. H. Freeman and Company, San Francisco, California.
MøLLER, A. P., AND R. THORNHILL. 1998. Bilateral symmetry and sexual selection: a meta-analysis. The
American Naturalist 151:174–192.
MULLER, F. P. 1995. Herbivore–plant–soil interactions
in the boreal forest: selective winter feeding by
spruce grouse. M.S. thesis, University of British Columbia, Vancouver, British Columbia, Canada.
NIEMELÄ, P., M. HAGMAN, AND K. LEHTILÄ. 1989. Relationship between Pinus sylvestris L. origin and
browsing preference by moose in Finland. Scandinavian Journal of Forestry Research 4:239–246.
PALMER, A. R. 1999. Detecting publication bias in
meta-analyses: a case study of fluctuating asymmetry and sexual selection. The American Naturalist
154:220–233.
PALO, R. T., AND C. T. ROBBINS (EDS.). 1991. Plant
defenses against mammalian herbivory. CRC Press,
Boca Raton, Florida.
PASTOR, J., AND R. J. NAIMAN. 1992. Selective foraging
and ecosystem processes in boreal forests. The
American Naturalist 139:690–705.
PFISTER, J. A., F. D. PROVENZA, G. D. MANNERS, D. R.
GARDNER, AND M. H. RALPHS. 1997. Tall larkspur
ingestion: can cattle regulate intake below toxic levels? Journal of Chemical Ecology 23:759–777.
PRINS, A. H., AND H. W. NELL. 1990. Positive and negative effects of herbivory on the population dynamics of Senecio jacobaea L. and Cynoglossum officinale L. Oecologia 83:325–332.
PROVENZA, F. D. 1995. Tracking variable environments:
there is more than one kind of memory. Journal of
Chemical Ecology 21:911–923.
PROVENZA, F. D. 1996. Acquired aversions as the basis
for varied diets of ruminants foraging on rangelands.
Journal of Animal Science 74:2010–2020.
READ, R. A. 1971. Browsing preference by jackrabbits
in a ponderosa pine provenance plantation. United
States Department of Agriculture Forest Service Research Note RM-186:1–4.
REICHARDT, P. B., J. P. BRYANT, T. P. CLAUSEN, AND G.
Vol. 82, No. 1
WIELAND. 1984. Defense of winter-dormant Alaska
paper birch against snowshoe hare. Oecologia 65:
58–69.
REICHARDT, P. B., J. P. BRYANT, B. R. MATTES, T. P.
CLAUSEN, F. S. CHAPIN III, AND M. MEYER. 1990.
The winter chemical defense of balsam poplar
against snowshoe hares. Journal of Chemical Ecology 16:1941–1959.
RITCHIE, M. E., D. TILMAN, AND J. M. H. KNOPS. 1998.
Herbivore effects on plant and nitrogen dynamics in
oak savanna. Ecology 79:165–177.
ROBBINS, C. T. 1993. Wildlife feeding and nutrition.
2nd ed. Academic Press, New York.
ROSENBERG, M. S., D. C. ADAMS, AND J. GUREVITCH.
1997. MetaWin. Statistical software for meta-analysis with resampling tests. Sinauer Associates, Inc.,
Publishers, Sunderland, Massachusetts.
ROUGHGARDEN, J. 1998. Primer of ecological theory.
Prentice Hall, Upper Saddle River, New Jersey.
ROUSI, M. 1988. Resistance breeding against voles in
birch: possibilities for increasing resistance by provenance transfers. European and Mediterranean Plant
Protection Organization Bulletin 18:257–263.
ROUSI, M. 1990. Breeding forest trees for resistance to
mammalian herbivores—a study based on European
white birch. Acta Forestalia Fennica 210:1–20.
ROUSI, M., J. HAGGMAN, AND J. P. BRYANT. 1987. The
effect of bark phenols upon mountain hare barking
of winter-dormant Scots pine. Holarctic Ecology 10:
60–64.
ROUSI, M., J. TAHVANAINEN, H. HENTTONEN, D. A.
HERMS, AND I. UOTILA. 1997. Clonal variation in susceptibility of white birches (Betula spp.) to mammalian and insect herbivores. Forest Science 43:
396–402.
ROUSI, M., J. TAHVANAINEN, AND I. UOTILA. 1989. Inter- and intraspecific variation in the resistance of
winter-dormant birch (Betula spp.) against browsing
by the mountain hare. Holarctic Ecology 12:187–
192.
ROUSI, M., J. TAHVANAINEN, AND I. UOTILA. 1991. A
mechanism of resistance to hare browsing in winterdormant European white birch (Betula pendula).
The American Naturalist 137:64–82.
RYALA, P. 1966. Riekon jakiirunan talvisesta kasviravinnon valinnasta ja puissa ruokalilusta. Suomen
Riista 19:79–93.
SHIPLEY, B., AND P. A. KEDDY. 1988. The relationship
between relative growth rate and sensitivity to nutrient stress in twenty-eight species of emergent
macrophytes. The Journal of Ecology 76:1101–
1110.
SILEN, R. R., W. K. RANDALL, AND N. L. MANDEL.
1986. Estimates of genetic parameters for deer
browsing of Douglas-fir. Forest Science 32:178–184.
SINCLAIR, A. R. E., AND J. N. M. SMITH. 1984. Do plant
secondary compounds determine feeding preferences of snowshoe hares? Oecologia 61:403–410.
SQUILLACE, A. E., AND R. R. SILEN. 1962. Racial variation in ponderosa pine. Forest Science Monograph
2:1–27.
STEARNS, S. C. 1992. The evolution of life histories.
Oxford University Press, Oxford, United Kingdom.
STEINBERG, P. D. 1986. Chemical defenses and the sus-
February 2001
SWIHART AND BRYANT—WINTER HERBIVORY
ceptibility of tropical marine brown algae to herbivores. Oecologia 69:628–630.
SVOBODA, F. J., AND G. W. GULLION. 1972. Preferential
use of aspen by ruffed grouse in northern Minnesota.
The Journal of Wildlife Management 36:1166–1180.
SWIHART, R. K., J. P. BRYANT, AND L. NEWTON. 1994.
Latitudinal patterns in consumption of woody plants
by snowshoe hares in the eastern United States. Oikos 70:427–434.
SWIHART, R. K., AND M. R. CONOVER. 1990. Reducing
deer damage to yews and apple trees: testing Big
Game Repellentt, Ro-Pelt, and soap as repellents.
Wildlife Society Bulletin 18:156–162.
SWIHART, R. K., AND P. M. PICONE. 1998. Selection of
mature growth stages of coniferous browse in temperate forests by white-tailed deer (Odocoileus virginianus). The American Midland Naturalist 139:
269–274.
SWIHART, R. K., AND R. H. YAHNER. 1983. Browse
preferences of jackrabbits and cottontails for species
used in shelterbelt plantings. Journal of Forestry 81:
92–94.
TAHVANAINEN, J., E. HELLE, R. JULKUNEN-TIITTO, AND
A. LAVOLA. 1985. Phenolic compounds of willow
bark as deterrents against feeding by mountain hare.
Oecologia 65:319–323.
TARGETT, N. M., L. D. COEN, A. A. BOETTCHER, AND
C. E. TANNER. 1992. Biogeographic comparisons of
marine algal polyphenolics: evidence against a latitudinal trend. Oecologia 89:464–470.
TILGHMAN, N. G. 1989. Impacts of white-tailed deer
on forest regeneration in northwestern Pennsylvania.
The Journal of Wildlife Management 53:524–532.
TILMAN, D. 1990. Constraints and tradeoffs: toward a
predictive theory of competition and succession. Oikos 58:3–15.
TUTIN, T. G., V. H. HEYWOOD, N. A. BURGES, D. M.
MOORE, D. H. VALENTINE, S. M. WALTERS, AND D.
A. WEBB (EDS.). 1968. Flora Europaea. Cambridge
University Press, Cambridge, United Kingdom 2:1–
455.
TUTIN, T. G., V. H. HEYWOOD, N. A. BURGES, D. H.
VALENTINE, S. M. WALTERS, AND D. A. WEBB (EDS.).
21
1964. Flora Europaea. Cambridge University Press,
Cambridge, United Kingdom 1:1–464.
VAN SOEST, P. 1982. Nutritional ecology of the ruminant. O & B Books, Corvallis, Oregon.
VIERECK, L. A., AND E. L. LITTLE, JR. 1972. Alaska
trees and shrubs. United States Department of Agriculture Handbook Number 410, Washington, D.C.
VILLABLA, J. J., AND F. D. PROVENZA. 1996. Preference
for flavored wheat straw by lambs conditioned with
intraruminal administrations of sodium propionate.
Journal of Animal Science 74:2362–2368.
VITOUSEK, P. 1982. Nutrient cycling and nutrient use
efficiency. The American Naturalist 119:553–572.
WALTZ, A. M., AND T. G. WHITHAM. 1997. Plant development affects arthropod communities: opposing
impacts of species removals. Ecology 78:2133–
2144.
WANG, J., AND F. D. PROVENZA. 1996a. Food preference
and acceptance of novel foods by lambs depend on
the composition of the basal diet. Journal of Animal
Science 74:2349–2354.
WANG, J., AND F. D. PROVENZA. 1996b. Food deprivation affects preference of sheep for foods varying in
nutrients and a toxin. Journal of Chemical Ecology
22:2021–2031.
WANG, J., AND F. D. PROVENZA. 1997. Dynamics of
preference by sheep offered foods varying in flavors,
nutrients, and a toxin. Journal of Chemical Ecology
23:275–288.
WATKINSON, A. R. 1986. Plant population dynamics.
Pp. 137–184 in Plant ecology (M. J. Crawley, ed.).
Blackwell Scientific, Oxford, United Kingdom.
WOLFF, J. O. 1978. Food habits of snowshoe hares in
interior Alaska. The Journal of Wildlife Management 42:148–153.
WOLFF, J. O. 1980. The role of habitat patchiness in
the population dynamics of snowshoe hares. Ecological Monographs 50:111–130.
Submitted 30 September 1999. Accepted 15 April
2000.
Associate Editor was Joseph A. Cook.