vol. 168, no.
3 the american naturalist september 2006
Notes and Comments
Evaluating Thermoregulation in Reptiles: An Appropriate Null Model
Keith A. Christian,1,* Christopher R. Tracy,1,† and C. Richard Tracy2,‡
1. School of Science, Charles Darwin University, Darwin, Northern
Territory 0909, Australia;
2. Department of Biology, University of Nevada, Reno, Nevada 89557
Submitted November 18, 2005; Accepted May 22, 2006;
Electronically published August 7, 2006
Online enhancement: appendix.
abstract: Established indexes of thermoregulation in ectotherms
compare body temperatures of real animals with a null distribution
of operative temperatures from a physical or mathematical model
with the same size, shape, and color as the actual animal but without
mass. These indexes, however, do not account for thermal inertia or
the effects of inertia when animals move through thermally hetero-
geneous environments. Some recent models have incorporated body
mass, to account for thermal inertia and the physiological control
of warming and cooling rates seen in most reptiles, and other models
have incorporated movement through the environment, but none
includes all pertinent variables explaining body temperature. We pre-
sent a new technique for calculating the distribution of body tem-
peratures available to ectotherms that have thermal inertia, random
movements, and different rates of warming and cooling. The ap-
proach uses a biophysical model of heat exchange in ectotherms and
a model of random interaction with thermal environments over the
course of a day to create a null distribution of body temperatures
that can be used with conventional thermoregulation indexes. This
new technique provides an unbiased method for evaluating ther-
moregulation in large ectotherms that store heat while moving
through complex environments, but it can also generate null models
for ectotherms of all sizes.
Keywords: body size, body temperature, ectotherms, null model, op-
erative temperature, thermal inertia.
* Corresponding author; e-mail: keith.christian@cdu.edu.au.
†
E-mail: chris.tracy@cdu.edu.au.
‡
E-mail: dtracy@biodiversity.unr.edu.
Am. Nat. 2006. Vol. 168, pp. 421–430. 䉷 2006 by The University of Chicago.
0003-0147/2006/16803-41441$15.00. All rights reserved.

The paradigm-shifting article by Cowles and Bogert (1944)
caused biologists to recognize that reptiles employ behavior
to interact with their thermal environments as a means to
regulate their body temperatures. Two decades of field and
laboratory studies subsequent to the article by Cowles and
Bogert gave birth to the field of animal physiological ecology
and to an expanding literature documenting how some rep-
tiles regulate body temperature by behaviorally and physi-
ologically exploiting diverse thermal environments. Nearly
fifty years ago, Heath (1964) published an admonition that
simply measuring body temperature or correlating body
temperature with air temperature was inadequate to char-
acterize complex environments in which thermoregulation
might occur. In the late 1960s, some studies applied the
physics of heat transfer and thermodynamics to add un-
derstanding to observations of reptile interactions with their
thermal environments (Bartlett and Gates 1967; Norris
1967). Subsequently, biophysical models describing the en-
ergy budgets of animals put the thermal biology of reptiles
into an ecological perspective (Porter and Gates 1969; Porter
et al. 1973; Christian and Tracy 1981; Tracy 1982; Wald-
schmidt and Tracy 1983; Stevenson et al. 1985). Contem-
poraneously, the concept of “operative temperature” (Te)
was developed to incorporate both animal and environ-
mental factors into a mechanistic index of the driving force
of the thermal environment on the body temperatures of
animals under a given set of environmental conditions (Bak-
ken and Gates 1975; Bakken 1981; Roughgarden et al. 1981;
Tracy 1982; Bakken et al. 1985).
The Te of an animal incorporates surface areas (areas
exposed to various forms of radiation and convective heat
exchange), but, by definition, it is a “massless” index,
meaning that it is an index of the interaction of an animal
of a particular size, shape, and posture with its thermal
environment, assuming that the animal has no body mass
(and therefore no heat capacity or “thermal inertia” that
can result in a lag in change of body temperature when
subjected to a new thermal environment). Although a
massless index normally can be used for studies of small
animals, its use for larger animals can be misleading be-
cause larger animals tend to have thermal inertia, and their
422 The American Naturalist
body temperature depends on both the instantaneous Te
and also the body temperatures that the animal had in the
recent past, because bodily thermal inertia causes a lag in
body temperatures during warming or cooling (Spotila et
al. 1973; Spotila 1980; Tracy 1982; Stevenson 1985a; Turner
and Tracy 1986). Biophysical models can incorporate ther-
mal inertia and physiological control of heat exchange,
and some models have been developed to calculate body
temperatures for large ectotherms in the warmest and
coolest parts of their thermal environment (Christian et
al. 1983; Stevenson 1985a; Tracy et al. 1986; Christian and
Weavers 1996), to provide context to the body tempera-
tures attained by the animals.
Based on earlier work employing a cost-benefit ap-
proach to thermoregulation (Huey and Slatkin 1976),
Hertz et al. (1993) drew attention to the fact that the
question “How carefully does an animal thermoregulate?”
is a complex of several distinct and interacting questions.
The answer to that complex question requires not simply
an examination of the pattern of the animal’s Tb but also
a comparison with the thermal options available in the
environment. One part of this complex question is “How
variable is the animal’s Tb (‘precision’ of thermoregula-
tion)?” A second question that must be addressed is “How
closely do Tb’s match the preferred or set point range of
the species (‘accuracy’ of thermoregulation)?” A third
question is “To what extent do thermoregulatory behaviors
actually enhance the accuracy of Tb’s relative to those of
a nonregulating control (the ‘effectiveness’ of thermoreg-
ulation)?” On the basis of these questions, Hertz et al.
(1993) have provided methods to evaluate data bearing on
the quantification of thermoregulation. These methods
represent a conceptual advance and have since become
commonly used in studies of reptilian thermoregulation
(e.g., Bauwens et al. 1996; Blouin-Demers and Weather-
head 2001; Fitzgerald et al. 2003; Ibargüengoytı́a 2005),
despite having some well-discussed limitations to their use
(e.g., Christian and Weavers 1996; Wills and Beaupre
2000). The thermoregulatory indexes of Hertz et al. (1993)
have been modified and extended by various authors
(Christian and Weavers 1996; Brown and Weatherhead
2000; Blouin-Demers and Weatherhead 2001; Blouin-
Demers and Nadeau 2005), but all of theses indexes require
a null model (the body temperatures of a nonregulating
animal) with which to compare empirical data.
Indeed, one of the major contributions of the Hertz et
al. (1993) article (see also Hertz 1992a, 1992b) was to
compare the animal’s Tb to a null model. They suggested
using a random distribution of Te as the null distribution
for comparison. However, they also pointed out that Te
does not accurately reflect the temperatures available for
large animals or for small animals that move quickly
among thermal environments, and so they cautioned
against using their approach with large ectotherms. Chris-
tian and Weavers (1996) addressed this problem by using
biophysical equations to incorporate the effects of thermal
inertia. This resulted in a predicted body temperature for
a stationary large lizard that could then be substituted for
the massless Te in the indexes proposed by Hertz et al.
Seebacher and Shine (2004) pointed out that the effects
of thermal inertia in ectotherms can be substantial when
large ectotherms move about in a thermally heterogeneous
environment. They offered equations and a table of coef-
ficients to correct Te when applying the thermal indexes of
Hertz et al. (1993). Their technique explicitly models animal
movement through the environment at regular intervals.
The approach illustrates that large ectotherms with large
thermal inertia have a reduced range of attainable body
temperatures. While this work draws attention to the prob-
lems associated with thermal inertia, it does not create a
true null model of available body temperatures for large
ectotherms because that would require both random move-
ments (both in terms of time and in terms of direction
relative to thermal environments) and the option that any
random “decision” to move would include not moving.
A second problem in the Seebacher and Shine (2004)
model is that it apparently (as evidenced by their fig. 1b)
does not incorporate physiological (cardiovascular) con-
trol of heat exchange that typically allows reptiles to warm
at a faster rate than they cool, sometimes twice as fast
(Bartholomew 1982; Turner and Tracy 1983; Dzialowski
and O’Connor 1999, 2001, 2004; O’Connor 1999). Thus,
the 5-kg animal (in fig. 1b of Seebacher and Shine 2004)
should warm faster than it cools. Cardiovascular control
of thermoregulation creates efficiencies in thermoregula-
tion that can be critically important to homeostasis in
individuals, especially for large-bodied ectotherms (see
Tracy et al. 1986).
Seebacher and Shine’s (2004) model also predicts very
large second-order body temperature dynamics (see
Turner 1987 for discussion of second-order body tem-
perature dynamics). Systems with second-order dynamics
(so named because the dynamics are better described by
second-order differential equations) have an initial time
lag (the second-order dynamics) before body temperatures
begin to change in response to a change in thermal en-
vironment. Thereafter, body temperatures change in re-
lation to the heat capacitance of the body (the first-order
dynamics). The first-order dynamics are dominated by the
heat capacitance of the body, resulting in a lag in change
in body temperature with respect to time due to mass-
related thermal inertia (appropriately described in terms
of a “time constant” t). The second-order dynamics, in
contrast, are likely related to establishing a new gradient
of temperatures within the body after a change in thermal
environment. There is some theoretical basis for conclud-

ing that animals can exhibit second-order body temper-
ature dynamics (Turner 1987), but there is a dearth of
empirical evidence that second-order effects are important
in the thermal biology of lizards. This may be because
physiological control of temperature gradients in the body
by cardiovascular mechanisms can rapidly establish new
temperature gradients in the body. Thus, a 6.9-kg land
iguana has been shown to have no second-order lag in
response to an experimentally controlled change in ther-
mal environment (see Tracy et al. 1986), and a 5-kg Var-
anus varius appears not to have second-order temperature
dynamics in response to change in thermal environment
(fig. 3 in Seebacher and Shine 2004). Nevertheless, See-
bacher and Shine’s model (likely incorrectly) predicted
that a 5-kg lizard would respond to a novel thermal en-
vironment so slowly that it would show no discernible
change in body temperature even after 15 min. This con-
cept of second-order dynamics needs to be studied further
(see Turner 1987, 1994a, 1994b, 1994c), but here we accept
the empirical evidence that large second-order body tem-
perature dynamics are not likely to be very important in
predicting body temperatures of lizards in changing ther-
mal environments.
Here, by using a randomly generated null model against
which one can compare the body temperatures achieved
by an animal, we address the problems associated with
data for ectotherms with large body size that move through
a thermally heterogeneous environment. We contend that
our method overcomes the inadequacies associated with
previous approaches and that it offers a robust technique
for evaluating thermoregulation, according to the pre-
scriptions in Hertz et al. (1993). The method is particularly
useful for animals with substantial thermal inertia (large
animals) but also can be used more generally.
A New Null Model Incorporating Size and Movement
In our approach, we first establish the extreme tempera-
tures available in the environment at a given time by cal-
culating the maximum (Te max) and minimum (Te min) body
temperatures from a steady state biophysical model (Porter
and Gates 1969; Porter et al. 1973; Porter and James 1979;
Porter and Tracy 1982; Tracy 1982; Waldschmidt and Tracy
1983). Input data included animal characteristics (e.g., so-
lar absorptivity and surface areas) and microclimate data
from days when lizards were followed with telemetry
(Christian and Weavers 1996). The Te max and Te min represent
the way in which the animal integrates its thermal envi-
ronments in the most hot and most cool environments
available, and the following procedures were used to cal-
culate predicted body temperature (Tb pred), within this
range of thermal environments, of an animal with thermal
inertia using Te max and Te min as driving variables.
Null Model for Ectotherm Thermoregulation 423
We developed a computer program (in True BASIC) to
construct a null distribution of Tb pred. The Tb pred represents
the body temperatures achievable by the animal based on
the animal characteristics, Te, and the body temperature
of the animal in the recent past (as the animal is warming
or cooling). Calculation of Tb pred is accomplished by first
assigning a random starting situation (corresponding to a
site in which the microclimate, at that time, is either
Te max or Te min) and a random length of time for the modeled
animal to remain in that spot. Then, Tb pred’s are calculated
for the animal in that place for each minute that the animal
remains in that place. This is done using Te as the driving
force and the thermal time constant t to incorporate body
mass. Each place is assigned a random proportion of the
distance between Te max and Te min as an indicator of the
animal being in full sun, full shade, or any intermediate
partial shade. The proportion remains constant until the
animal’s next move, but the Te max and Te min are updated
every minute of the day such that the predicted Tb also
may change while an animal is in one place as the daily
conditions change. The time constant t is taken from re-
gression equations generated from data in Dzialowski and
O’Connor (2001). Different time constants are used for
warming (th p 1.96 # mass0.434) and cooling (tc p
2.34 # mass0.362). Alternatively, time constants could be
measured for an individual species. The form of the first-
order equation for the predicted body temperature of the
animal adjusted for thermal inertia and time in a given
place is Tb pred p T0 ⫹ (Te ⫺ T0 ) # e⫺1/t, where T0 is the
temperature at the time the animal first enters the new
place. Next, the program chooses a new random place and
time, calculates the corresponding predicted Tb, and con-
tinues these steps over the course of the entire day (24 h).
The animal could be modeled as remaining above ground
for the full period or as retreating into a refugium, so long
as Te max and Te min are adjusted accordingly. Finally, the
model repeats this process for as many days as desired (we
used 1,000 replicate days) to generate a distribution of
random predicted Tb achievable over the day (i.e., a null
distribution).
The term d̄b was defined by Hertz et al. (1993) as the
mean deviation of an animal’s field-active body temper-
atures from its set point (or preferred) range. This index
is an indication of the accuracy of thermoregulation, but
on its own, this index cannot be used to determine whether
an animal is thermoregulating because it lacks information
about the availability of thermal opportunities in the en-
vironment, as Hertz et al. (1993) noted. The term d̄e was
defined as the mean deviation of Te from the animal’s set
point range and is thus a measure of the thermal quality
of the animal’s environment (Hertz et al. 1993). For large
ectotherms, it is appropriate to substitute the mean achiev-
able body temperature (Tb pred) for mean Te to calculate
424 The American Naturalist

Figure 1: Examples of predicted body temperatures as a function of time of day for lizards differing in body size, as calculated from a model that
assumes random movements through the thermal environment. A total of 1,000 replicates (days) of the model were used to determine the null
distribution, which was then used in thermoregulatory indexes. The shaded area represents the middle 950 replicates, or the 95% confidence interval
of the distribution; lines are representative individual replicates for lizards of each size.

d̄e. The Tb pred calculated for each hour of each iteration of
the model was compared to the set point range to yield a
d¯ b, and then a d¯ e (mean) was calculated for 1,000 iterations
for comparison with hourly values of d̄b.
The simplest use of these indexes is a direct comparison:
when d¯ b ! d¯ e, it suggests that the animal is regulating its
body temperature because the selected body temperatures
are closer to the set point range than would be the case
if the animal were randomly selecting body temperatures
in its environment (as represented by the null model; Hertz
et al. 1993). We used a one-tailed paired t-test to compare
d¯ b with d¯ e.
The “effectiveness” of thermoregulation is defined by
Hertz et al. (1993) as E p 1 ⫺ (d¯ b /d¯ e ), where a value for
E of 0 indicates a random selection of thermal environ-
ments and a value for E of 1 indicates careful thermoreg-
ulation (Hertz et al. 1993). Others have argued that a better
index of effectiveness of thermoregulation is simply the
difference, d¯ e ⫺ d¯ b, with higher values indicative of more
effective thermoregulation (Blouin-Demers and Weath-
erhead 2001; Blouin-Demers and Nadeau 2005). The “ex-
ploitation” of the thermal environment (Ex) is defined as
the amount of time an animal has a body temperature
within its set point range divided by the amount of time
during the day in which it is possible for the animal to
achieve its set point range (Christian and Weavers 1996).
Example
To illustrate the methods developed in this article, we have
reanalyzed data shown in figures 1A and 1B of Christian
and Weavers (1996) on body temperatures of two species

Figure 2: Mean predicted Tb as a function of time of day, calculated for lizards ranging from 10 to 1,000 kg. The means were calculated from 1,000
iterations (replicate days) of the model for each body size. Error bars (95% confidence interval) are shown for 0.01–1,000-kg animals.
of varanid lizards. These lizards include the floodplain
monitor Varanus panoptes and the sand monitor Varanus
gouldii, studied during the wet season in northern Aus-
tralia. The two species are sympatric, and the data for these
lizards and their environments were collected simulta-
neously. In this study, individuals of each species of ap-
proximately 1.5 kg body mass were selected (Christian and
Weavers 1996), and, except where specifically stated oth-
erwise, a mass of 1.5 kg was used in the models below.
During the period of data collection, the thermal envi-
ronment was such that lizards could attain body temper-
atures within the set point range during most of the day-
light hours (Christian and Weavers 1996). Despite the
similarities between the two species (body mass, same
study site, and simultaneous data collection), the patterns
of body temperatures selected by the species were strikingly
different (fig. 1 in Christian and Weavers 1996). For com-
parison, we additionally used the methods suggested by
Seebacher and Shine (2004; using equations from their
table 1) for the two varanid lizards, assuming regular shut-
tling movements between thermal patches every 5 min,
every 15 min, and every 30 min (namely, three separate
simulations).
Our model calculated predicted body temperatures
(Tb pred’s) that fluctuate widely, as would be expected for a
null distribution of random “thermoregulation” (fig. 1).
Null Model for Ectotherm Thermoregulation 425
There was no obvious pattern in the mean Tb pred for lizards
ranging from 200 to 5,000 g, except that the body tem-
peratures of the 5,000-g lizard seemed to lag behind body
temperatures of the smaller lizards (fig. 2).
Measured body temperatures for V. gouldii during the
activity period of the day were always within the set point
range (central 50% of Tb’s selected in a thermal gradient;
Christian and Weavers 1996) for this species (fig. 3),
whereas measured body temperatures for V. panoptes were
never within its set point range (fig. 4). Nevertheless, the
Te’s and the mean Tb pred’s indicate that both species should
have been able to bring their body temperature to within
the set point range. Thus, although these two species are
of similar size, inhabit the same environment, and were
measured simultaneously, they interact with their thermal
environments very differently, and this is consistent with
observations of the behavior and activity patterns of these
two species (Christian et al. 1995; Christian and Weavers
1996). The body temperatures predicted from the models
of Seebacher and Shine (2004) were far from the set point
range for either species (figs. 3, 4), and this indicates a
problem in their approach to correcting for thermal
inertia.
The indexes of thermoregulation for V. gouldii indicate
that it is an excellent thermoregulator (table 1), whereas
the indexes indicate that V. panoptes is a much less precise
426 The American Naturalist

Figure 3: Temperatures during a typical day in the wet season in tropical Australia for Varanus gouldii. The temperatures include mean Tb (as
measured by telemetry), mean predicted Tb (as determined by the model in the appendix in the online edition of the American Naturalist), Temin,
and Temax, and three “available Tb” lines determined using the techniques recommended by Seebacher and Shine (2004), assuming that the animals
shuttle between warm and cool thermal environments every 5, 15, and 30 min.

thermoregulator. The former species actively selects a sub-
set of the thermal conditions available in its environment,
whereas the latter species moves through the thermal en-
vironment with little regard to its set point range of body
temperatures.
Discussion
The new techniques presented here are an easy supplement
to the prescriptions in Hertz et al. (1993) for studies of
thermoregulation of large ectotherms. They represent an
approach for creating a sophisticated null model that can
be used in conjunction with the indexes of Hertz et al.
(1993) and Christian and Weavers (1996). Users need an
array of Te values (either calculated or estimated from
physical models; Tracy 1982; Bakken 1992), Tb’s of real
animals, and the ability to generate the null distribution
(see appendix in the online edition of the American Nat-
uralist). Although we have applied these methods to mean
values of body temperature among animals (to provide a
direct parallel with the data in fig. 1 of Christian and
Weavers 1996), other questions can be addressed by ap-
plying the techniques and indexes to data collected from
each individual in a study (Hertz et al. 1993). The model
easily could be modified to reflect different spatial and
temporal distributions of thermal patches at a particular
 Null Model for Ectotherm Thermoregulation 427

Figure 4: Temperatures during a typical day in the wet season in tropical Australia for Varanus panoptes. The temperatures include mean Tb (as
measured by telemetry), mean predicted Tb (as determined by the model in the appendix), Temin , Temax , and three “available Tb” lines determined
using the techniques recommended by Seebacher and Shine (2004), assuming, in separate simulations, that the animals shuttle between thermal
environments every 5, 15, and 30 min.

study site by using weighted proportional distributions
between Te max and Te min.
The Te integrates and reflects the thermal environment
of an organism of a particular size, shape, posture, and
color. As such, Te is the “driving force” for heat exchange
between an organism and its environment. For example,
when an ectotherm has a body temperature of 35⬚C in an
environment with Te p 40⬚C, the organism will typically
tend to warm toward 40⬚C. However, large ectotherms,
with large thermal inertia, respond to both their current
Te and the Te from the immediate past. For example, if
an ectotherm with a body temperature of 35⬚C moves from
an environment with Te of 40⬚C to an environment with
a Te of 30⬚C, the core Tb of that organism (as opposed to
surface temperature) might continue to rise due to a lag-
ging response to its immediate past thermal environment.
This is in spite of the fact that the animal is currently in
a cooling environment. Thus, it is simply not enough to
know Te for large ectotherms to predict how the animal’s
Tb will respond in particular environments because the
thermal response of a large ectotherm is complexly related
to present and past Te’s.
Seebacher and Shine (2004) laudably drew attention to
the problems associated with a large ectotherm moving
through a complex environment, but their model does not
yield a “null” distribution of body temperatures because
that model assumes behavioral thermoregulation by lizards
that is not random; indeed, it assumes mechanical move-
ment between environments with regular periodicity. In
addition, their model does not incorporate physiological
428 The American Naturalist
Table 1: Thermoregulatory indexes for two species of var-
anid lizards from the wet season in tropical Australia
Indexes Varanus gouldii Varanus panoptes
d̄b (⬚C) .0 2.6
d̄e (⬚C) 1.9 1.8
d¯ b ! d¯ e t p 5.9, t p 3.5,
df p 7, P p .0003 df p 7, P p .99
d¯ e ⫺ d¯ b 1.9 ⫺.8
E 1.0 ⫺.4
Ex 1.0 .0
Note: Calculated, in part, from original data of Christian and Weavers
(1996), these indexes demonstrate that V. gouldii thermoregulate very
carefully but V. panoptes do not. The field data were collected from the
same study site simultaneously for the two species. A paired t-test of
the d¯ e ⫺ d¯ b index (Blouin-Demers and Weatherhead 2001) revealed that
the index was significantly greater for V. gouldii than for V. panoptes
(t p 13.7, df p 7, P ! .0001), indicating that V. gouldii is the more
effective thermoregulator. One-tailed paired t-tests were used to deter-
mine whether d¯ b ! d¯ e within each species. E p effectiveness of ther-
moregulation; Ex p exploitation of the thermal environment. Indexes
are defined in the text.
control of warming and cooling, and this may also explain
the unusually high second-order body temperature dy-
namics predicted by their model. These modeling short-
comings doubtlessly contributed to their underestimate of
the abilities of animals with body masses around 5 kg to
change temperature.
The biological conclusions drawn here about thermo-
regulation of two varanid lizards are the same as the con-
clusions drawn by Christian and Weavers (1996). Obser-
vations of these two species reveal dramatic differences in
behaviors of the two species (K. A. Christian, personal
observation), and the conclusions drawn from the ther-
moregulatory indexes in Christian and Weavers (1996) and
here (figs. 3, 4) are consistent with those observations.
However, the methods of Seebacher and Shine (2004) fail
to identify these species differences. “Available Tb’s” mod-
eled according to Seebacher and Shine (2004; fig. 3) are
almost always lower than the actual Tb’s of Varanus gouldii.
Thus, the “correction” techniques proposed by Seebacher
and Shine (2004) result in unrealistic estimates of body
temperatures of large ectotherms.
Large body mass results in less variance around the
means of Tb pred (figs. 1, 2), and the body temperatures of
larger lizards lag behind those of smaller individuals during
the initial warming phase (fig. 2). However, the extensively
overlapping mean Tb pred’s (fig. 2) indicate that random
movements of animals up to around 10 kg can largely
erase the effects of body size. Although body size is one
important factor in determining how an animal interacts
with its thermal environment, it is only one of many im-
portant factors (e.g., Stevenson 1985b; Tracy et al. 1986).
Behavioral and physiological characteristics (such as
warming faster than cooling) can neutralize the effects of
thermal inertia (Stevenson 1985a, 1985b). Thus, effects of
thermal inertia for small to medium-sized ectotherms (up
to approximately 10 kg), although real and measurable,
should not be overstated in the broader context of the
range of possibilities due to behavior, physiology, and an-
imal-environment interactions. Animals on the order of
100 kg or larger, however, have substantial thermal inertia
(fig. 2).
The tools we provide here generate null models of ther-
moregulation in ectotherms by using properties of animals
and environments and the physics of heat exchange be-
tween animals and their environments. These tools make
it possible to investigate the thermal relations of large ec-
totherms that move about in complex thermal environ-
ments. However, the technique provides a mechanism for
generating null models for ectotherms of all sizes. Even if
Te’s are measured with physical models, unless Te’s are
spatially normally distributed, it would require a prohib-
itively large number of models to adequately create a null
model representative of the thermal heterogeneity in the
environment. Thus, the techniques presented here are ap-
propriate in studies of thermoregulation regardless of body
size.
Acknowledgments
We thank the Australian Research Council (ARC grant
DP0559093 to K.A.C.), the National Science Foundation
(grant INT-0202663 to Christopher R. Tracy), Charles Dar-
win University, and the Biological Resources Research Cen-
ter at the University of Nevada, Reno, for financial support.
Literature Cited
Bakken, G. S. 1981. How many equivalent black-body temperatures
are there? Journal of Thermal Biology 6:59–60.
———. 1992. Measurement and application of operative and stan-
dard operative temperatures in ecology. American Zoologist 32:
194–216.
Bakken, G. S., and D. M. Gates. 1975. Heat transfer analysis of
animals: some implications for field ecology, physiology and evo-
lution. Pages 255–290 in D. M. Gates and R. B. Schmerl, eds.
Perspectives of biophysical ecology. Springer, New York.
Bakken, G. S., W. R. Santee, and D. J. Erskine. 1985. Operative and
standard operative temperatures: tools for thermal energetics stud-
ies. American Zoologist 25:933–943.
Bartholomew, G. A. 1982. Physiological control of body temperature.
Pages 167–211 in C. Gans and F. H. Pough, eds. Biology of the
Reptilia. Vol. 12. Physiology C: physiological ecology. Academic
Press, New York.
Bartlett, P. N., and D. M. Gates. 1967. The energy budget of a lizard
on a tree trunk. Ecology 48:316–322.
Bauwens, D., P. E. Hertz, and A. M. Castilla. 1996. Thermoregulation
in a lacertid lizard: the relative contributions of distinct behavioral
mechanisms. Ecology 77:1818–1830.

Blouin-Demers, G., and P. Nadeau. 2005. The cost-benefit model of
thermoregulation does not predict lizard thermoregulatory be-
havior. Ecology 86:560–566.
Blouin-Demers, G., and P. J. Weatherhead. 2001. Thermal ecology
of black rat snakes (Elaphe obsoleta) in a thermally challenging
environment. Ecology 82:3025–3043.
Brown, G. P., and P. J. Weatherhead. 2000. Thermal ecology and
sexual size dimorphism in northern water snakes, Nerodia sipedon.
Ecological Monographs 70:311–330.
Christian, K. A., and C. R. Tracy. 1981. The effect of the thermal
environment on the ability of hatchling Galapagos land iguanas
to avoid predation during dispersal. Oecologia (Berlin) 49:218–
223.
Christian, K. A., and B. W. Weavers. 1996. Thermoregulation of
monitor lizards in Australia: an evaluation of methods in thermal
biology. Ecological Monographs 66:139–157.
Christian, K. A., C. R. Tracy, and W. P. Porter. 1983. Seasonal shifts
in body temperature and use of microhabitats by the Galapagos
land iguana. Ecology 64:463–468.
Christian, K. A., L. Corbett, B. Green, and B. Weavers. 1995. Seasonal
activity and energetics of two species of varanid lizards in tropical
Australia. Oecologia (Berlin) 103:349–357.
Cowles, R. B., and C. M. Bogert. 1944. A preliminary study of the
thermal requirements of desert reptiles. Bulletin of the American
Museum of Natural History 82:265–296.
Dzialowski, E. M., and M. P. O’Connor. 1999. Utility of blood flow
to the appendages in physiological control of heat exchange in
reptiles. Journal of Thermal Biology 24:21–32.
———. 2001. Thermal time constant estimation in warming and
cooling ectotherms. Journal of Thermal Biology 26:231–245.
———. 2004. Importance of the limbs in the physiological control
of heat exchange in Iguana iguana and Sceloporus undulatus. Jour-
nal of Thermal Biology 29:299–305.
Fitzgerald, M., R. Shine, and F. Lemckert. 2003. A reluctant helio-
therm: thermal ecology of the arboreal snake, Hoplocephalus ste-
phensi (Elapidae) in dense forest. Journal of Thermal Biology 28:
515–524.
Heath, J. E. 1964. Reptilian thermoregulation: evaluation of field
studies. Science 146:784–785.
Hertz, P. E. 1992a. Evaluating thermal resource partitioning by sym-
patric lizards Anolis cooki and A. cristatellus: a field test using null
hypotheses. Oecologia (Berlin) 90:127–136.
———. 1992b. Temperature regulation in Puerto Rican Anolis liz-
ards: a field test using null hypotheses. Ecology 73:1405–1417.
Hertz, P. E., R. B. Huey, and R. D. Stevenson. 1993. Evaluating
temperature regulation by field-active ectotherms: the fallacy of
the inappropriate question. American Naturalist 142:796–818.
Huey, R. B., and M. Slatkin. 1976. Costs and benefits of lizard ther-
moregulation. Quarterly Review of Biology 51:363–384.
Ibargüengoytı́a, N. R. 2005. Field, selected body temperature and
thermal tolerance of the syntopic lizards Phymaturus patagonicus
and Liolamus elongatus (Iguania: Liolaemidae). Journal of Arid
Environments 62:435–448.
Norris, K. S. 1967. Color adaptation in desert reptiles and its thermal
relationships. Pages 162–229 in W. W. Milstead, ed. Lizard ecology:
a symposium. University of Missouri Press, Columbia.
O’Connor, M. P. 1999. Physiological and ecological implications of
a simple model of heating and cooling in reptiles. Journal of Ther-
mal Biology 24:113–136.
Null Model for Ectotherm Thermoregulation 429
Porter, W. P., and D. M. Gates. 1969. Thermodynamic equilibria of
animals with environment. Ecological Monographs 39:227–244.
Porter, W. P., and F. C. James. 1979. Behavioral implications of mech-
anistic ecology. II. The African rainbow lizard Agama agama.
Copeia 1979:594–619.
Porter, W. P., and C. R. Tracy. 1982. Biophysical analysis of energetics,
time-space utilization, and distributional limits of lizards. Pages
55–83 in R. Huey, E. Pianka, and T. Schoener, eds. Ecology of
lizards. Harvard University Press, Cambridge, MA.
Porter, W. P., J. W. Mitchell, W. A. Beckman, and C. B. DeWitt. 1973.
Behavioral implications of mechanistic ecology. Oecologia (Berlin)
13:1–54.
Roughgarden, J., W. P. Porter, and D. Heckel. 1981. Resource, par-
titioning of space and its relationship to body temperature in Anolis
lizard populations. Oecologia (Berlin) 50:256–264.
Seebacher, F., and R. Shine. 2004. Evaluating thermoregulation in
reptiles: the fallacy of the inappropriately applied method. Phys-
iological and Biochemical Zoology 77:688–695.
Spotila, J. R. 1980. Constraints of body size and environment on the
temperature regulation of dinosaurs. Pages 233–252 in R. D. K.
Thomas and E. C. Olson, eds. A cold look at the warm-blooded
dinosaurs. Westview, Boulder, CO.
Spotila, J. R., P. W. Lommen, G. S. Bakken, and D. M. Gates. 1973.
A mathematical model for body temperatures of large reptiles:
implications for dinosaur ecology. American Naturalist 107:391–
404.
Stevenson, R. D. 1985a. Body size and limits to the daily range of
body temperature in terrestrial ectotherms. American Naturalist
125:102–117.
———. 1985b. The relative importance of behavioral and physio-
logical adjustments controlling body temperature in terrestrial ec-
totherms. American Naturalist 126:362–386.
Stevenson, R. D., C. R. Peterson, and J. S. Tsuji. 1985. The thermal
dependence of locomotion, tongue flicking, digestion and oxygen
consumption in the wandering garter snake. Physiological Zoology
58:46–57.
Tracy, C. R. 1982. Biophysical modeling of reptilian physiology and
ecology. Pages 275–321 in C. Gans and F. H. Pough, eds. Biology
of the Reptilia. Vol. 12. Physiology C: physiological ecology. Ac-
ademic Press, New York.
Tracy, C. R., J. S. Turner, and R. B. Huey. 1986. A biophysical analysis
of possible thermoregulatory adaptations in sailed pelycosaurs.
Pages 195–206 in N. J. Hotton III, P. D. MacLean, J. J. Roth, and
E. C. Roth, eds. The ecology and biology of mammal-like reptiles.
Smithsonian Institution, Washington, DC.
Turner, J. S. 1987. On the transient temperatures of ectotherms.
Journal of Thermal Biology 12:207–214.
———. 1994a. Thermal impedance of a contact-incubated bird’s
egg. Journal of Thermal Biology 19:237–243.
———. 1994b. Time and energy in the intermittent incubation of
birds’ eggs. Israel Journal of Zoology 40:519–540.
———. 1994c. Transient-state thermal properties of contact-
incubated chicken eggs. Physiological Zoology 67:1426–1447.
Turner, J. S., and C. R. Tracy. 1983. Blood flow to appendages and
the control of heat exchange in the American alligator. Physio-
logical Zoology 56:195–200.
———. 1986. Body size, homeothermy and the control of heat ex-
change in mammal-like reptiles. Pages 185–194 in N. J. Hotton
III, P. D. McLean, J. J. Roth, and E. C. Roth, eds. The ecology and
430 The American Naturalist

biology of mammal-like reptiles. Smithsonian Institution, Wash- for detecting evidence of thermoregulation in timber rattlesnakes
ington, DC. (Crotalus horridus) from northwest Arkansas. Physiological and
Waldschmidt, S., and C. R. Tracy. 1983. Interactions between a lizard Biochemical Zoology 73:325–334.
and its thermal environment: implications for sprint performance
and space utilization in the lizard Uta stansburiana. Ecology 64:
476–484. Associate Editor: Raymond B. Huey
Wills, C. A., and S. J. Beaupre. 2000. An application of randomization Editor: Jonathan B. Losos

Floodplain monitor lizard, Varanus panoptes (photograph by Keith Christian).