Content uploaded by L. Kealoha Freidenburg
Author content
All content in this area was uploaded by L. Kealoha Freidenburg
Content may be subject to copyright.
Oecologia (2003) 134:360–364
DOI 10.1007/s00442-002-1136-9
POPULATION ECOLOGY
M. A. Halverson · D. K. Skelly · J. M. Kiesecker ·
L. K. Freidenburg
Forest mediated light regime linked to amphibian distribution
and performance
Received: 17 May 2002 / Accepted: 5 November 2002 / Published online: 9 January 2003
! Springer-Verlag 2003
Abstract The vegetation in and around the basins of
ephemeral wetlands can greatly affect light environments
for aquatic species such as amphibians. We used hemi-
spherical photographs to quantify the light environment in
terms of the global site factor (GSF), the proportion of
available solar radiation that actually strikes the wetland.
We compared GSF to the distribution and performance of
two amphibian species (Pseudacris crucifer and Rana
sylvatica) within 17 ephemeral wetlands in northeastern
Connecticut, USA. We found that P. crucifer is restricted
to lighter wetlands (GSF >0.34) and that its abundance is
proportional to GSF. By contrast, R. sylvatica is found
across the light gradient and its abundance is unrelated to
GSF. For both species, GSF is a strong predictor of larval
developmental rate. In addition, P. crucifer growth rate is
higher in lighter wetlands. Through thermal effects,
changes in resources, or other influences, light appears
to be an important predictor of the distribution and
performance of amphibians. Because the structure of
canopies can change rapidly, and because amphibians can
be strongly impacted by these changes, vegetation
mediated effects on wetland light environments may be
critical to understanding the dynamics of amphibian
populations within forested biomes.
Keywords Amphibian · Development · Distribution ·
Global site factor · Growth
Introduction
Amphibians resident in small freshwater wetlands have
become a model system for ecologists (Wilbur 1997;
Werner 1998; Morin 1999). Despite this attention,
important aspects of the mechanisms underlying the
distribution and abundance of natural populations remain
poorly understood (Wellborn et al. 1996; Skelly 1997,
2001; Alford 1999; Skelly and Kiesecker 2001). Recently,
long-term surveys have uncovered dramatic changes in
the distribution of amphibian species over time. While
some of these dynamics have occurred in the context of
population declines and species extinction (reviewed by
Alford and Richards 1999), it appears that even healthy
amphibian populations can undergo rapid changes (Sjo-
gren-Gulve 1994; Skelly et al. 1999; Carlson and
Edenhamn 2000). Preliminary evidence suggests that
some observed changes in amphibian distributions could
result from alterations in the vegetation canopy above
small wetlands (Skelly et al. 1999, 2002; Werner and
Glennemeier 1999).
The light environments of nonpermanent wetlands are
apt to be particularly susceptible to the effects of
vegetation. Because they tend to be relatively small, the
crowns of trees and other vegetation can overtop ephem-
eral wetlands. Because they are often shallow, many
nonpermanent wetlands can support rooted vegetation
within their basins. Where succession, fire, timber
harvest, beaver (Castor canadensis), or other disturbances
result in temporal changes in vegetation structure, corre-
sponding changes in light environment may be expected
(Skelly et al. 1999; Skelly and Freidenburg 2000).
Ecologists studying the role of gap formation in terrestrial
forests and other systems have shown that the timing,
size, and spatial distribution of such dynamics can have
an overriding effect on the distribution of species (e.g.,
Paine and Levin 1981; Whitmore 1989; Pacala et al.
1996). The purpose of this study is to evaluate the
relationship between vegetation and amphibians. We
hypothesize that relationships among vegetation, solar
radiation, temperature, and other factors are likely to have
M. A. Halverson (
)
) · D. K. Skelly
School of Forestry and Environmental Studies,
Yale University,
370 Prospect Street, New Haven, CT 06520 USA
e-mail: anders.halverson@yale.edu
Fax: +1-203-4323929
J. M. Kiesecker
Department of Biology,
Pennsylvania State University,
University Park, PA 16802 USA
L. K. Freidenburg
Department of Ecology and Evolutionary Biology,
University of Connecticut,
Storrs, CT 06265 USA
strong effects on the present day distribution and perfor-
mance of amphibians across a light gradient.
We used hemispherical photographs of the canopy
cover to quantify the light regime over 17 temporary
ponds in Connecticut, USA. This technique has been
largely developed by plant ecologists who have long
recognized the importance of solar radiation and the need
for a method with which to analyze it at relatively fine
scales and over long time periods (Evans and Coombe
1959; Anderson 1964). We used 3 years worth of survey
and sampling data to evaluate what effect this factor has
on the distribution, abundance, and performance of two
amphibian species: wood frogs (Rana sylvatica) and
spring peepers (Pseudacris crucifer).
Materials and methods
This study was conducted at the 3,800-ha Yale-Myers Forest in
northeastern Connecticut, USA. Timber harvesting, as well as the
activities of beaver (Castor canadensis), have promoted the
development of a diversity of light environments within wetlands
at the Yale-Myers Forest. The 17 ephemeral wetlands used in this
study range from heavily shaded to relatively open. Maximum
surface areas range from 40 to 5,200 m
2
and maximum depths were
between 25 and 250 cm. They were selected for study during 1998
or earlier because they were known amphibian-breeding sites and
were nonpermanent.
We quantified the light environment at each wetland in terms of
the global site factor (GSF) (Anderson 1964). This ratio estimates
the amount of light that actually strikes a point over the amount of
light that would strike the same point if there were no overhead
obstructions. The GSF changes depending both on the length of
time considered and the time of year. We estimated GSF using
hemispherical photographs of the forest canopy. With such
photographs, it is possible to estimate the amount of diffuse and
direct beam radiation that would strike the photograph point during
a given time interval (Evans and Coombe 1959; Anderson 1964).
This method has been used by plant ecologists for many years, and
estimates of photon flux based on hemispherical photographs have
been shown to closely correspond to average direct sensor
measurements (Chazdon and Field 1987; Becker et al. 1989; Rich
et al. 1993). We analyzed the photographs and estimated GSF with
Hemiview (Delta-T 2001). Necessary parameters for this program
were estimated from the 1960–1990 data in the National Solar
Radiation Database (2001) for Hartford, Connecticut (approxi-
mately 40 km from the study site).
Photograph poin ts w ere laid out in agridpatternwith5-mintervals
bounded by the high water mark of the wetlands in 1999. Two
photographs were taken at each point. “Leaf-on” photographs were
taken during the summer of 1999 and“leaf-off”photographswere
taken during the winters of 1999 and 2000. We used these two sets of
photographs to estimate the average GSF from April to August, the
embryonic and larval periods of our focal species at this site. As leaf
emergence at the study site typically occurs during mid-May, we
based our calculations on leaf-off photographs prior to and leaf-on
photographs after 15 May. The average GSF for a wetland thus
estimated is ref erred to simply as GSF in the remainder of this paper.
We measured natural patterns of distribution, abundance, and
performance of two amphibian species in each of the 17 wetlands:
wood frogs ( Rana sylvatica) and spring peepers (Pseudacris
crucifer). These species are the most common anurans in nonper-
manent wetlands at Yale-Myers Forest. From 1998 to 2000, every
wetland was visited in March or April and at least three more times
during discrete sampling periods between early May and early July
(I: first half of May; II: second half of May; III: first half of June;
IV: first half of July). Sampling intervals usually lasted 2 days or
less and never lasted more than 1 week. During March and April
wetlands were visually inspected for eggs. During May through
July sampling visits, depth was measured, wetlands were visually
inspected for adult and larval amphibians and eggs, and timed
continuous dip-net sampling (between 5 and 15 person minutes
depending on surface area) was conducted throughout the wetland.
Up to 20 amphibian larvae of each of the focal species were
collected and stored in 70% EtOH. The larvae were later measured
(snout-vent length, SVL), and staged (Gosner 1960) under a
dissecting scope.
We used logistic regression to determine the relationship
between GSF and the presence-absence distribution of spring
peepers and wood frogs. A species was considered present if eggs
or larvae were ever found in a wetland between 1998 and 2000.
Linear regression was used to relate larval density of each species
to GSF. Density was measured as the number of larvae encountered
per person-minute averaged across the three dip-net sampling
rounds (I, II and III). In addition, we analyzed the relationship
between GSF and larval performance. For each sampling period
during which a species was present in at least three wetlands, we
related GSF to the average body size (mm SVL) and developmental
stage (Gosner stage, Gosner 1960) of the larvae from a given
wetland using linear regression. Results from 2000 are presented
here; results from prior years were comparable. Finally, we used
multiple linear regression to determine whether the relationship
between size and developmental stage varied with GSF. In these
regressions, the response variables were the average sizes of
tadpoles from each developmental stage from each wetland.
Predictor variables were GSF and Gosner stage. A significant
effect of GSF in this regression suggests that tadpoles at a given
Gosner stage vary in size in a manner that is correlated with GSF.
Water temperature was measured in 11 of the study wetlands
using a Hobo temperature logger suspended 10 cm below the surface
at the site of maximum wetland depth. Temperature was recorded
every hour during the month of April. Temperature is probably lower
during this month than at any other time during the larval period and
is thus probably more important at this time as a developmental
constraint. In addition, the embryos and larvae are less developed
during this month and are thus less able to compensate for pond
temperature variation by seeking out different spots within the pond
(L.K. Freidenburg, unpublished data). To estimate the association
between water temperature and light environment we calculated a
regression of GSF against the average April temperature from each
of the wetlands for which we had such data.
Results
Wood frogs were present in each of the 17 wetlands, thus
their distribution was independent of GSF. By contrast,
distribution of spring peepers was strongly correlated with
wetland light environment (logistic regression: df=1,
c
2
=5.83, P=0.016). These frogs were found in most of
the high light wetlands and in none of the darkest
wetlands (GSF<0.34).
The density of wood frog larvae was unrelated to GSF
(Fig. 1a; linear regression: R
2
=0.001, MS=0.06, F
1, 15
=0.02,
P=0.889) while density of spring peeper larvae was
positively related to GSF (Fig. 1b; linear regression:
R
2
=0.44, MS=1.89, F
1, 15
=11.98, P=0.003). The increase
in spring peeper density with increasing GSF persisted
when the analysis was restricted to the eight wetlands
with nonzero densities (linear regression: R
2
=0.60,
MS=1.62, F
1, 6
=9.12, P=0.023).
Larval developmental stage of both species tended to
be more advanced on a given date within lighter wetlands
(Table 1, Fig. 2). The relationship between GSF and
developmental stage was small or insignificant during
361
early sampling and became strong and significant by early
June (Table 1, Fig. 2).
Body size of spring peepers tended to show a pattern
similar to that observed for development: a strong positive
relationship between body size and GSF became evident
as the season progressed (Table 1, Fig.2). This was not the
case for wood frogs. While there was some evidence of a
positive relationship between light and body size for
wood frogs early in the season, this relationship tended to
disappear as the season progressed.
Multiple linear regression showed that wood frog
larvae at a given stage of development tended to be larger
in darker wetlands (1998: slope ="3.038, SE=0.819,
t
82
="3.7071, P<0.001. 1999: slope="1.288, SE=0.587,
t
99
="2.19, P=0.031. 2000: slope ="1.387, SE=0.5062,
t
64
="2.740, P=0.008). There was no significant relation-
ship between GSF and size of spring peepers at a given
stage of development (1998: slope="1.854, SE=1.354,
t
7
="1.3688, P=0.2134. 1999: slope ="414, SE=0.4942,
t
19
="0.838, P=0.4125. 2000: slope="0.351, SE=0.583,
t
29
="0.602, P=0.552).
Among the 11 wetlands in which water temperature
was measured (April 2000), wetlands with higher
GSF tended to be warmer (linear regression: MS=2.26,
F
1, 9
=6.43, R
2
=0.35, P=0.032).
Discussion
Incoming solar radiation is apt to be particularly variable
in ephemeral wetlands. Tree crowns often extend entirely
across smaller wetlands and some species such as red
maple (Acer rubrum) and buttonbush (Cephalanthus
occidentalis) can root within shallow basins. As a result,
the light environments in our darkest wetlands are
comparable to those found in the understory of mature
forests in the northeastern United States (Canham et al.
1994; Finzi and Canham 2000).
This gradient in light among wetlands created by
variation in vegetation structure is strongly associated with
the distribution and performance of amphibian species. In
the case of one species, spring peepers, we identified a
light boundary (GSF=0.34) below which the species was
consistently absent. In those wetlands in which spring
Fig. 2 Performance of wood frogs (a, c) and spring peepers (b, d)
during 2000 as a function of light level within a wetland (indexed
as global site factor, GSF). Performance was measured as the
average Gosner (Gosner 1960) developmental stage (a, b), or body
size (c, d). Samples were collected during three sampling periods:
early May (filled circles), mid May (hollow diamonds), and early
June (crosses). Lines of best fit from least squares regression are
presented for each sampling period (Table 1). The standard error
for average body size (mm SVL) and average Gosner stage was
generally less than 5% of the mean
Fig. 1 Density of a wood frog and b spring peeper larvae across 17
wetlands as a function of light level. Light level is indexed as
global site factor (GSF). Larval density is estimated as the number
of tadpoles encountered per person minute while dip-netting. Each
point represents an average calculated across visits during up to
3 years. The bars represent 1 SE
Table 1 Results from least squares regression of average tadpole
size and developmental stage as a function of light environment
(indexed by global site factor, GSF). Responses for wood frogs and
spring peepers were collected during 2000 while sampling 17
wetlands at the Yale-Myers Forest in northeastern Connecticut.
There were three sampling periods: early May (I), late May (II) and
early June (III)
Species Response Period N Slope R
2
P
Spring
Peepers
Gosner stage I 3 "0.20 0.11 0.789
II 6 4.38 0.52 0.107
III 8 13.90 0.78 0.003
Body size I 3 "1.05 0.87 0.239
II 6 5.02 0.57 0.082
III 8 9.94 0.62 0.020
Wood
Frogs
Gosner stage I 15 3.23 0.56 0.001
II 14 5.40 0.55 0.107
III 13 11.48 0.56 0.003
Body size I 15 2.65 0.33 0.025
II 14 3.10 0.12 0.227
III 13 1.93 0.02 0.685
362
peepers were present, their density was proportional to
GSF (Fig. 1b). Despite their reliance on forested terrestrial
environments after metamorphosis (Delzell 1958), spring
peepers appear to be excluded from wetlands under heavy
forest cover during embryonic and larval phases. Given
that closed-canopy forests grow throughout much of the
range of this species, this means spring peepers necessarily
depend on wetlands that are too large to be overtopped by
tree crowns, where shoreline vegetation has recently been
disturbed, or where a higher light environment is main-
tained through some other means.
What prevents spring peepers from using heavily
shaded wetlands? Spring peeper larvae grow and develop
more slowly in darker wetlands. During June, when
spring peepers in the lightest wetlands were nearing
metamorphosis, conspecifics in the darker wetlands were
still in early stages of development (Fig. 2b). We
documented up to three and fourfold differences in body
length during late season. Field transplant experiments
have shown similar decreases in growth and development
in low light environments (Werner and Glennemeier
1999; Skelly et al. 2002). Such reduced performance may
lead to an increased likelihood of larval mortality from
drying events in nonpermanent wetlands (Skelly 1995).
While reduced growth and development could explain
the restricted distribution of spring peepers as well as
decreased density in darker wetlands, amphibians also are
known to be selective in their choice of oviposition sites
(Hopey and Petranka 1994; Resetarits and Wilbur 1989;
Resetarits and Wilbur 1991; Kiesecker and Skelly 2000).
The results of a pilot experiment in which the canopy was
removed from a heavily shaded wetland at the Yale-
Myers Forest are consistent with such an effect. Spring
peeper larvae were absent from this wetland during the
3 years prior to canopy removal and have been present
during each of 3 years since (D.K. Skelly, unpublished
data). While darker wetlands may act as demographic
sinks for spring peepers, it also is possible that adults
avoid placing their offspring in these environments.
Unlike spring peepers, wood frogs breed in wetlands
across the light gradient and their larval density was not
significantly related to GSF (Fig. 1a). Neither was there any
indication that light was related to wood frog growth patterns
(Fig. 2c). While wood frogs seem able to maintain relatively
dense populations of rapidly growing larvae even in heavily
shaded wetlands, their development rate was correlated with
GSF (Fig. 2a). Thus, for both species there is a performance
cost associated with living in shaded wetlands.
Solar radiation could affect amphibian performance
through a variety of mechanisms. Light affects the
composition and abundance of the primary producers in
aquatic ecosystems (Feminella et al. 1989; Wetzel 2001).
Periphyton is an important food source for tadpoles
(Kupferberg et al. 1994; Hill et al. 1995) and can be less
abundant and less diverse in shaded wetlands (Skelly et
al. 2002). In addition, there is evidence that spring
peepers and wood frogs are less able to assimilate organic
matter available in benthic substrates from shaded
wetlands (Skelly and Golon, unpublished data).
The performance patterns we observed also are
consistent with the effects of temperature on ectotherms.
In this study, darker wetlands tended to be colder.
Controlled temperature laboratory experiments show that
both growth and development rates of amphibian larvae
can be limited by temperature (e.g., Smith-Gill and
Berven 1979; Berven and Gill 1983). Even when provided
abundant, high quality food resources, modest decreases
in ambient temperature can cause large reductions in
growth and development of amphibians (Newman 1998).
These reductions are typically asymmetric; for a given
decrease in temperature, larvae will suffer greater reduc-
tions in development rate versus growth rate (Ray 1960;
Smith-Gill and Berven 1979). Consequently, larvae
sampled from colder environments tend to be larger at a
given developmental stage compared with conspecifics
from warmer environments. The asymmetry of tempera-
ture effects is widely reported in ecotherms but the
mechanisms remain debated (Atkinson 1994, 1995; Van
der Have and de Jong 1996). Whatever the reason, the
existence of this asymmetric pattern among wood frogs is
strongly suggestive of an impact of temperature on
growth and development patterns. Its absence among
spring peepers suggests that this species may respond
differently to temperature change, that it is not temper-
ature limited in its performance, or that its restriction to
relatively lighter wetlands precluded observation of the
asymmetric temperature effect.
The links established here between wetland light
environment and amphibian performance and distribution
are particularly meaningful in the context of vegetation
dynamics. The forest cover on the Yale-Myers Forest and
central New England has changed dramatically over the
last 300 years. After settlement reduced the forest cover in
some parts of New England to as low as 20% by the middle
of the eighteenth century, many farms were abandoned and
forests allowed to regenerate. Today, as in pre-settlement
times, forest covers more than 90% of the Yale-Myers
Forest and central New England (Meyer and Plusnin 1945;
Foster 1992; P.M.S. Ashton, unpublished data). Distur-
bances from such things as timber harvest, beaver and fire
can also cause the canopy above wetlands to change
rapidly (Skelly et al. 1999; Skelly and Freidenburg 2000).
When allowed to regrow, forest can encroach and overtop
a small wetland within a few decades or less (Skelly et al.
1999). This study suggests amphibians may undergo
changes in distribution, abundance and performance as a
result of such changes in vegetation.
The ability of wood frogs to persist in spite of canopy
closure could be related to their ability to maintain
relatively rapid growth and development across different
light environments. There also is some evidence that this
species may evolve rapidly in response to thermal
changes that accompany change in canopy (Skelly and
Freidenburg 2000). Regardless of origin, the persistence
of wood frogs across the light gradient means that this
species is able to utilize a much higher fraction of
ephemeral wetlands in our study area than spring peepers.
The distribution of spring peepers appears be at least
363
partly determined by the pattern of recent disturbance
around smaller wetlands (e.g., Paine and Levin 1981;
Pacala et al. 1996; Batzer et al. 2000).
Acknowledgements Thanks to C. Apse, C. Burns, J. Golon, G.
Jones, J. Morton, E. Sagor, R. Schiff, S. Price, M. Urban and J.
Virdin for help with field work. P.M.S. Ashton generously provided
the use of his camera. H. zu Dohna, E. Palkovacs, C. Burns and M.
Urban provided helpful comments on previous drafts of the
manuscript. This research has been supported by the National
Institutes of Health and the National Science Foundation (Ecology
of Infectious Diseases Grant 1R01ES11067–01) and by a gift to
Yale University from Mrs. E.S. Dwyer.
References
Alford RA (1999) Ecology: resource use, competition, and
predation. In: McDiarmid RW, Altig R (eds) Tadpoles: the
biology of anuran larvae. University of Chicago Press, Chicago,
USA, pp 240–278
Alford RA, Richards SJ (1999) Global amphibian declines: a
problem in applied ecology. Annu Rev Ecol Syst 30:133–166
Anderson MC (1964) Studies of the woodland light climate. 1. The
photographic computation of light conditions. J Ecol 52:27–41
Atkinson D (1994) Temperature and organism size a biological law
for ectotherms? Adv Ecol Res 25:1–58
Atkinson D (1995) Effects of temperature on the size of aquatic
ectotherms: exceptions to the general rule. J Thermal Biol
20:61–74
Batzer DP, Jackson CR, Mosner M (2000) Influences of riparian
logging on plants and invertebrates in small depressional
wetlands of Georgia, USA. Hydrobiologia 441:123–132
Becker P, Erhart DW, Smith AP (1989) Analysis of forest light
environments.1. Computerized estimation of solar-radiation
from hemispherical canopy photographs. Agric For Meteorol
44:217–232
Berven KA, Gill DE (1983) Interpreting geographic variation in life
history traits. Am Zool 23:85–97
Canham CD, Finzi AC, Pacala SW, Burbank DH (1994) Causes
and consequences of resource heterogeneity in forests inter-
specific variation in light transmission by canopy trees. Can J
For Res 24:337–349
Carlson A, Edenhamn P (2000) Extinction dynamics and the
regional persistence of a tree frog metapopulation. Proc R Soc
Lond Ser B Biol Sci 267:1311–1313
Chazdon RL, Field CB (1987) Photographic estimation of photo-
synthetically active radiation: evaluation of a computerized
technique. Oecologia 73:525–532
Delta-T (2001) http://www.delta-t.co.uk
Delzell DE (1958) Spatial movement and growth of Hyla crucifer.
Dissertation, University of Michigan, Ann Arbor, Mich.
Evans GC, Coombe DE (1959) Hemispherical and woodland
canopy photography and the light climate. J Ecol 47:103–113
Feminella JW, Power ME, Resh VH (1989) Periphyton responses to
invertebrate grazing and riparian canopy in three northern
California coastal streams. Freshw Biol 22:445–457
Finzi AC, Canham CD (2000) Sapling growth in response to light
and nitrogen availability in a southern New England forest. For
Ecol Manage 131:153–165
Foster DR (1992) Vegetation dynamics and land use history in
central New England, USA. J Ecol 80:753–772
Gosner K (1960) A simplified table for staging anuran embryos and
larvae with notes on identification. Herpetologica 16:183–190
Hill WR, Ryon MG, Schilling EM (1995) Light limitation in a
stream ecosystem: responses by primary producers and con-
sumers. Ecology 76:1297–1309
Hopey ME, Petranka JW (1994) Restriction of wood frogs to fish-
free habitats; how important is adult choice. Copeia
1994:1023–1025
Kiesecker JM, Skelly DK (2000) Choice of oviposition site by gray
treefrogs, Hyla versicolor: the role of potential parasitic
infection. Ecology 76:1297–1309
Kupferberg SJ, Marks JC, Power ME (1994) Effects of variation in
natural and algal detrital diets on larval anuran (Hyla regilla)
life history traits. Copeia 1994:446–457
Meyer WH, Plusnin BA (1945) The Yale forest in Tolland and
Windham counties. Yale School For Bull 55:1–154
Morin PJ (1999) Community ecology. Blackwell, Malden, Mass.
National Solar Radiation Database (2001) http://rredc.nrel.gov/
solar/old_data/nsrdb
Newman RA (1998) Ecological constraints on amphibian meta-
morphosis: interactions of temperature and larval density with
responses to changing food level. Oecologia 115:9–16
Pacala SW, Canham CD, Saponara J, Silander JA, Kobe RK,
Ribbens E (1996) Forest models defined by field measurements:
estimation, error analysis, and dynamics. Ecol Monogr 66:1–43
Paine RT, Levin SA (1981) Intertidal landscapes: disturbance and
the dynamics of pattern. Ecol Monogr 51:145–178
Ray C (1960) The application of Bergmann’s and Allen’s rules to
the poikilotherms. J Morphol 106:85–108
Resetarits WJ, Wilbur HM (1989) Choice of oviposition site by
Hyla chrysoscelis: role of predators and competitors. Ecology
70:220–228
Resetarits WJ, Wilbur HM (1991) Calling site choice by Hyla
chrysoscelis: effect of predators, competitors, and oviposition
sites. Ecology 72:778–786
Rich PM, Clark DB, Clark DA, Oberbauer SF (1993) Long-term
study of solar radiation regimes in a tropical wet forest using
quantum sensors and hemispherical photography. Agric For
Meteorol 65:107–127
Sjogren-Gulve P (1994) Distribution and extinction patterns within
a northern metapopulation of the pool frog, Rana lessonae.
Ecology 75:1357–1367
Skelly DK (1995) A behavioral tradeoff and its consequences for the
distribution of Pseudacris tree frog larvae. Ecology 76:150–164
Skelly DK (1997) Tadpole communities. Am Sci 85:36–45
Skelly DK (2001) Distributions of pond-breeding anurans: an
overview of mechanisms. Isr J Zool 47:313–332
Skelly DK, Freidenburg LK (2000) Effects of beaver on the thermal
biology of an amphibian. Ecol Lett 3:483–486
Skelly DK, Kiesecker JM (2001) Design and outcome in ecological
experiments: manipulations of larval anurans. Oikos 94:198–208
Skelly DK, Werner EE, Cortwright SA (1999). Long-term distri-
butional dynamics of a Michigan amphibian assemblage.
Ecology 80:2326–2337
Skelly DK, Freidenburg LK, Kiesecker JM (2002) Forest canopy
and the performance of larval amphibians. Ecology 83:983–992
Smith-Gill SJ, Berven KA (1979) Predicting amphibian metamor-
phosis. Am Nat 13:563–585
Van der Have TM, deJong G (1996) Adult size in ectotherms:
temperature effects on growth and differentiation. J Theor Biol
183:329–340
Wellborn GA, Skelly DK, Werner EE (1996) Mechanisms creating
community structure across a freshwater habitat gradient. Annu
Rev Ecol Syst 27:337–363
Werner EE (1998) Ecological experiments and a research program
in community ecology. In: Resetarits WJ Jr, Bernardo J (eds)
Experimental ecology: issues and perspectives. Oxford Uni-
versity Press, New York, pp 3–26
Werner EE, Glennemeier KS (1999) Influence of forest canopy
cover on breeding pond distributions of several amphibian
species. Copeia 1999:1–12
Wetzel RG (2001) Limnology: lake and river ecosystems, 3rd edn.
Academic Press, New York
Whitmore TC (1989) Canopy gaps and the two major groups of
forest trees. Ecology 70:536–538
Wilbur HM (1997) Experimental ecology of food webs: complex
systems in temporary ponds. Ecology 78:2279–2302
364
