| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 13 No. 4: 551-560
© 2002 International Society for Behavioral Ecology
Size-dependent predation by snakes: selective foraging or differential prey vulnerability?
Biological Sciences A08, University of Sydney, Sydney, NSW 2006, Australia
Address correspondence to S.J. Downes, who is now at the Division of Botany and Zoology, Australian National University, Canberra, ACT 0200, Australia. E-mail: sharon.downes{at}anu.edu.au .
Received 29 March 2000; revised 2 October 2001; accepted 25 November 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONDISCUSSIONREFERENCES |
|---|
I staged replicate encounters between unrestrained lizards and snakes in outdoor enclosures to examine size-dependent predation within the common garden skink (Lampropholis guichenoti). Yellow-faced whip snakes (Demansia psammophis) forage widely for active prey and most often consumed large skinks, whereas death adders (Acanthophis antarcticus) ambush active prey and most often consumed small skinks. Small-eyed snakes (Rhinoplocephalus nigrescens) forage widely for inactive prey and consumed both small and large skinks equally often. Differential predation may reflect active choice by the predator, differential prey vulnerability, or both. To test for active choice, I presented foraging snakes with an inert small lizard versus an inert large lizard. They did not actively select lizards of a particular body size. To test for differential prey vulnerability, I quantified variation between small and large lizards in behavior that is important for determining the outcome of predator—prey interactions. Snakes did not differentiate between integumentary chemicals from small and large lizards. Large lizards tend to flee from approaching predators, thereby eliciting attack by the visually oriented whip snakes. Small lizards were more mobile than large lizards and therefore more likely to pass by sedentary death adders. Additionally, small skinks were more effectively lured by this sit-and-wait species and less likely to avoid its first capture attempt. In contrast, overnight retreat site selection (not body size) determined a lizard's chances of being detected by small-eyed snakes. Patterns of size-dependent predation by elapid snakes may arise not because of active choice but as a function of species-specific predator tactics and prey behavior.
Key words: antipredator behavior, body size, foraging mode, lizards, prey selection, prey vulnerability, size-dependent predation, snakes.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONDISCUSSIONREFERENCES |
|---|
Field collections of snakes often reveal nonrandom predation upon prey items of different size (for reviews, see Arnold, 1993
;
Greene, 1983;
Mushinsky, 1987;
Voris and Voris, 1983).
Although there have been few empirical attempts to elucidate the reasons for
these patterns (but see Mushinsky et al.,
1982; Shine, 1991;
Webb and Shine, 1993), one
popular explanation is that snakes actively reject certain prey in some
circumstances (Arnold, 1993).
This notion is based on theories that assume that predators forage efficiently
as a result of natural selection. Efficient foraging may entail maximizing net
energy or nutrient gain or minimizing the time and energy spent acquiring a
set quantity of resources (Pyke,
1984; Sih and Moore,
1990; Stephens and Krebs,
1986). However, application of such models to snake species can be
problematic (reviewed in Forsman,
1996; Shine,
1991). In venomous snakes, for example, the metabolic costs and
time spent in subduing and swallowing a prey item are insignificant compared
to the energy gained through eating it
(Cruz-Neto et al., 1999;
Feder and Arnold, 1982;
Pough and Andrews, 1985) and
the time taken to encounter another food item
(Allan and Flecker, 1988;
Greene, 1984). Hence, probably
the most efficient foraging strategy for these predators is to attempt to
consume any prey item of the appropriate size and type that they encounter.
However, a possible exception occurs when snakes encounter more than one prey
item simultaneously, as might be the case if prey are socially gregarious. In
this situation, even venomous snakes are expected to select and consume the
energetically most profitable prey (e.g.,
Griffiths, 1980;
Onkonburi and Formanowicz,
1997; Werner and Hall,
1974).
Alternative explanations for size-dependent predation in such systems might
be that snakes concentrate their search in the habitat of the most often
consumed prey, that snakes consume the most abundant prey type, or that the
less frequently eaten prey evade capture by snakes (reviewed in
Peckarsky and Penton, 1989).
With respect to size-dependent consumption of lizards by venomous snakes, a
lizard's body size can affect its selection of microhabitats (Asplund,
1968,
1974;
Stamps, 1988) and can affect
behavior such as activity and basking patterns
(Bartholomew and Tucker, 1964;
Carothers, 1983;
Carrascal et al., 1992;
Melville and Swain, 1997),
which may influence its chances of being detected by snakes. Additionally, a
lizard's body size can alter important antisnake behavior, including locomotor
performance (Bauwens et al.,
1995; Huey and Hertz,
1982,
1984), tendency to employ tail
loss (Brown and Ruby, 1977;
Castilla and Bauwens, 1991),
approach and flight distances from predators
(Castilla and Bauwens, 1991;
Smith, 1997), and escape
effectiveness (Ferguson and Talent,
1993). If such processes are important, snakes may be
size-dependent predators not because of active choice but as a function of
species-specific predator behavior and prey defensive attributes (e.g.,
Peckarsky and Penton,
1989).
Most predation events consist of several phases
(Endler, 1986), and successful
evasion attempts may result from efficiency of the prey in any one or more of
these stages (Vermeij, 1982).
So, although body size can differentially affect lizard defense behavior, such
variation may not be pertinent to the outcome of encounters unless it operates
during an important phases of predation. Likewise, because predator avoidance
and antipredator mechanisms are not universally effective
(Edmunds, 1974;
Endler, 1986;
McPeek, 1990), body size may
alter behavior that affects a lizard's chances of being consumed by some
snakes but not others. For example, running speed may determine a lizard's
probability of escaping snake species that actively pursue prey, but may be
inconsequential against species that initiate capture attempts from a
sedentary position. Thus, larger (and faster) lizards may have less chance of
being attacked by pursuing snakes but be equally vulnerable to ambush
predators than are smaller lizards.
In this study, I examined size-dependent predation within a single prey
taxon, the common garden skink (Lampropholis guichenoti: 20-40 mm
snout-vent length [SVL]). These diurnal lizards are distributed along coastal
southeastern Australia (Cogger,
1994) and are important prey items for several small venomous
(elapid) snakes (Shine, 1991).
They are socially gregarious and do not have well-defined home ranges, and it
is rare for lizards to displace conspecifics from basking or shelter sites
(Torr and Shine, 1996). As
predators, I used three elapid snake species that adopt very different
foraging modes to capture garden skinks. Yellow-faced whip snakes
(Demansia psammophis) are visually oriented and search widely for
active prey during the day (Downes and
Shine, 2000; Scanlon,
1998). Eastern small-eyed snakes (Rhinoplocephalus
nigrescens [= Cryptophis nigrescens in earlier literature]) are
chemically oriented and search widely for inactive prey during the night
(Downes and Shine, 2000;
Shine, 1984). Common death
adders (Acanthophis antarcticus) are visually oriented sit-and-wait
predators that attempt to lure active lizards by vigorously moving the distal
end of their tails (Carpenter et al.,
1978; Chizar et al.,
1990). My first aim was to document differences among snake
species in patterns of predation on lizards of different body size. To
simplify interpretation, these experiments were conducted under conditions
where predator and prey densities were controlled and mimicked those found in
natural situations. However, it was not possible to control for prior
experience in either the snakes or lizards (see below). My second aim was to
quantify behavioral components of the predator—prey interactions between
these snakes and lizards and thus identify the mechanisms or processes
underlying the actual patterns of size-dependent predation.
Animals and their maintenance
During November to February each year, female garden skinks lay one or two
clutches of up to six eggs (Qualls and
Shine, 1998). Young hatch from December through March, and under
favorable conditions they mature at around 38 mm SVL by the reproductive
season after their birth (Simbotwe,
1985). This study comprised a series of experiments conducted
during February and March in 1997, 1998, and 1999. At these times, hatchling
garden skinks (hereafter referred to as "small" lizards) and
yearling garden skinks (hereafter referred to as "large" lizards)
were collected by hand from areas around Sydney, New South Wales, Australia.
On average, small lizards were 37% shorter (mean ± SE SVL: 24.20
± 0.20 vs. 37.39 ± 0.21 mm) and 70% lighter (mean ± SE
mass: 0.33 ± 0.01 vs. 1.12 ± 0.02 g) than large lizards.
Agonistic encounters are rare in captive garden skinks, especially among
females (Torr and Shine,
1996). Nonetheless, to reduce the potentially confounding effects
of aggressive interactions among prey, I only used female lizards
(Kaiser and Mushinsky, 1985).
Few animals in my study population have original tails because of the high
incidence of tail loss (Downes and Shine,
2000) ; thus, I used only lizards that had entire tails. Lizards
were not used in more than one experiment unless otherwise stated.
Snakes were captured by hand from coastal southeastern Australia between
September and November in 1996, 1997, and 1998, except the death adders. Only
immature adders (<3 years old) specialize on lizard prey
(Shine, 1980). Because it is
extremely difficult to locate free-living immature adders, I used snakes that
were born in captivity during February 1997 and 1998 from wild-caught breeding
pairs collected near Sydney. These neonatal snakes were raised under
seminatural conditions on a diet of both small and large scincid lizards and
used in experiments as prey-experienced year-ling snakes. I assumed that these
rearing conditions did not substantially affect the foraging behavior of
adders during my experiments, but this assumption was not explicitly tested.
For all snake species, I used each snake only once per experiment, but the
same individuals were used in different experiments throughout the study.
I housed the skinks and snakes at the University of Sydney in separate
rooms at 18°C; the light cycle was the natural cycle of the surrounding
area. Lizards were maintained in plastic cages (220 x 130 x 70 mm)
covered with soil (10 mm depth) and containing wooden shelters and were fed
crickets every third day. I kept lizards used in the predation encounters in
groups of six individuals (three small lizards and three large lizards), but
skinks were housed individually in all other cases. Snakes were maintained
individually in plastic cages (220 x 260 x 70 mm) lined with paper
and containing rock shelters and fed one adult garden skink every 7-10 days.
However, in all experiments involving snakes, I used individuals that were
deprived of food for 8 days. Heating was provided for all reptiles by means of
an underfloor element that maintained a thermal gradient from ambient to
38°C within each cage for 8 h/day, falling to ambient temperatures
overnight. Animals were supplied with water ad libitum. The laboratory
performance trials described below were conducted at ambient temperatures that
approximate the mean body temperatures of active garden skinks in the field,
26.8°C (Shine, 1983).
Estimates of size-dependent predation on lizards
Methods
I first tested whether predation by snakes upon lizards was dependent on
the size of lizards. I staged replicate encounters between one snake versus
one prey group (three small skinks and three large skinks) in outdoor
enclosures (1.8 x 1.8 m, metal walls 1.6 m high). Hereafter these trials
are referred to as "predation encounters." The enclosures were
located within a compound covered with plastic mesh to exclude large
predators. Grasses and weeds in each enclosure were regularly pruned to about
50 mm high. Two sandstone retreat sites (280 x 280 x 15 mm) and
two wooden blocks (200 x 100 mm2 pine, 10 holes [18 x
60 mm] in one face) were supplied such that the same shelter type was
available in diagonally opposing corners.
To determine the order in which the six prey in each group were consumed by
snakes, I used wire microtags (0.25 mm diam, Northwest Marine Technology, USA)
to uniquely code skinks and a hand-held X-ray scope (Lixi, USA) to read the
tags from within the digestive tract of snakes (see
Downes, 2000, for a detailed
account of this procedure). The wire tags were injected under the skin of the
lizard's ventral surface between the forearms. Individuals 1-3 were inserted
with one, two, or three 0.5-mm long tags, and individuals 4-6 were injected
with one, two or three 1-mm long tags. Upon injection, individual codes were
checked using the X-ray scope. Extensive trials suggest that these methods
have little, if any, influence on the behavior and well-being of skinks
(Downes, 2000).
I introduced one prey group into an enclosure and allowed them to acclimate
undisturbed for 3 days. I began the predation encounters on day 4 by
introducing a snake to each enclosure. Small-eyed snakes were introduced
during the morning, and the other predators were released at dusk; this
procedure enabled snakes to explore the new environment for up to 12 h before
foraging commenced. Shortly after dawn on every day during the encounter, I
captured each snake and scanned its body cavity with an X-ray scope to
identify which prey had been eaten and the order in which they were consumed
(see Downes, 2000, for more
information on these procedures). Snakes were returned to their point of
capture immediately after scanning. An encounter was terminated when at least
three of the lizards were eaten (about 3-6 days), but all of the lizards were
usually consumed.
I performed this experiment over three summers (1997-1999) with 23 whip snakes (mean ± SE = 524 ± 53 mm SVL), 22 small-eyed snakes (502 ± 47), and 21 death adders (152 ± 07). During each trial, a maximum of 14 enclosures was used and equal numbers of each snake species were predators, except death adders were used only in 1998 and 1999. All trials began on sunny days with the forecast of similar weather for at least the next 3 days.
Results
I scored the body size of the first prey consumed in each predation
encounter and used these data to examine whether predation by snakes was size
dependent. A higher than expected proportion of small lizards was captured
first by death adders (
2 test: 1 df, = 10.72, p
=.001), but the opposite trend was evident for whip snakes (2
test: 1 df, = 9.78, p =.002;
Figure 1). In contrast, body
size did not influence the chances of a lizard being the first prey item for
small-eyed snakes (2 test: 1 df, = 0.73, p
=.37; Figure 1).
|
Foraging patterns of snakes
Methods
Before examining processes that may underlie size-dependent predation on
skinks, it was necessary to verify the foraging patterns of the different
snake predators. I determined whether snakes consumed lizards during the day
or night by scanning a predator's body cavity shortly before dusk, as well as
early in the morning, during pilot trials conducted for this purpose
(n = 9 snakes for each species). I estimated if whip snakes searched
for active versus inactive lizards by noting the times at which the ship
snakes and garden skinks emerged from their overnight retreat sites during the
first morning of each predation encounter. Garden skinks were considered to be
active before snakes if three or more skinks (of any body size) were observed
out of their retreat site for at least 20 min before their potential predator.
I characterized the foraging patterns of snakes (n = 9 snakes for
each species) by observing them for 5-min blocks during the periods that they
foraged and recorded distance and duration of each move and duration of each
stop.
Results
Whip snakes and death adders always consumed lizards during the day,
whereas small-eyed snakes always consumed lizards during the night. Whip
snakes and small-eyed snakes moved relatively long distances per move (mean
± SE 
3.11 ± 0.36 m) and per unit time ( 0.09 ±
0.01 m/s) and also moved a large proportion of the time ( 0.75 ±
0.03): I classify these snakes as a diurnal and nocturnal widely foraging
predator, respectively. Death adders were never observed to move: I classify
this snake as a diurnal sit-and-wait predator. In the mornings, skinks were
usually active before the whip snake in the same enclosure (this was the case
in 22 vs. 1 enclosures, 2: 1 df, = 19.17, p
<.001). Late in the afternoons, whip snakes usually became inactive before
skinks (this was the case in 19 vs. 4 enclosures, 2: 1 df,
= 9.78, p =.002). Thus, whip snakes probably interacted with
garden skinks that were active rather than sequestered within overnight
retreat sites. Because death adders initiated capture attempts from a
sedentary position, I assumed that they also interacted with garden skinks
that were active. Small-eyed snakes encountered skinks that were sequestered
within overnight retreat sites.
Estimates of size-dependent prey selection by snakes
Methods
To examine whether snakes actively selected small versus large prey, I
presented individual predators in outdoor enclosures (1.8 x 1.8 m, metal
walls 1.6 m high) with two different-sized inert lizards and scored which prey
was consumed first. The snakes were acclimated to the enclosure for at least 1
day. I presented prey simultaneously because snakes are likely to encounter
more than one lizard at a time (see Discussion). For the trials with whip
snakes (n = 13, mean ± SE = 564 ± 32 mm SVL) and death
adders (n = 14, mean ± SE = 154 ± 13 mm SVL), a large
skink and a small skink were killed using carbon dioxide. The lizards were
then immediately placed onto separate thin, wooden carriages (20 x 3 mm,
the body of the skink entirely covered the wood) that were both attached by
fishing line to the end of a rod (1 m long). For whip snakes, these carriages
(and skinks) were lowered onto an area devoid of vegetation about 150 mm in
front of the head of an actively foraging snake and were simultaneously
slightly moved vertically by raising and lowering the horizontal rod. For
death adders, I lowered the skinks about 50 mm from the head of the snake and
slowly moved them toward the snake's lure. I recorded the first prey
seized.
For the trials with small-eyed snakes (n = 12, mean ± SE =
444 ± 46 mm SVL), I placed a live large and small skink (individually
tagged) in the same hole of one of the wooden retreat sites shortly before
dusk and covered the hole with cardboard until the snake became active. I then
removed the cover and allowed the snake to select one or the other prey. A
previous study on 10 lizard pairs indicated that skinks remain within the same
hole of the retreat site at least until 0700 h the next day
(Downes and Shine, 2000). I
scanned the body cavity of a snake before 0700 h the next morning using the
same procedure outlined above. Opportunistic observations (n = 3)
with the night vision scope suggest that a lizard is unable to escape from the
retreat site upon being discovered by a foraging small-eyed snake. On four
occasions a snake did not select either skink; however, these experiments were
successfully repeated on the next night.
Results
All snake species randomly selected their first prey when I presented them
with an inert small lizard and an inert large lizard at the same time
(2: 1 df, = 0.08, 0.33, 0.29, p =.78,.56,.59
for whip snakes, small-eyed snakes, and adders, respectively:
Figure 1). Thus, these
predators did not actively select among lizards of different body size.
Estimates of size-dependent vulnerability of lizards
Discrimination of chemical cues by snakes
Methods. The responses of snakes to chemical cues of potential
prey items may indicate the differential detection of certain prey (reviewed
in Ford and Burghardt, 1993).
I recorded the chemoreceptive responses of snakes to control and lizard scents
by presenting them with chemical stimuli on the cotton swabs of 300-mm wooden
applicators. Four conditions were presented. (1) Swabs were dipped into
distilled water (neutral control). (2) Swabs were dipped into a 1:1 solution
of a commercial cologne and distilled water (pungency control). (3, 4)
Integumentary chemicals were obtained from small garden skinks and large
garden skinks. Swabs were dipped into distilled water and blotted dry before
they were rolled across the dorsal, lateral, and ventral surfaces of the
lizard, between the neck and abdominal region
(Dial and Schwenk, 1996).
To begin a trial, I slowly approached the snake's home cage and placed the swab 10-20 mm anterior to its snout. I then recorded the number of tongue flicks for 60 s, beginning with the first tongue extrusion. The snakes were tested 2-3 days before a predation encounter. Trials were conducted in a preset, fixed order whereby each treatment was presented first to a randomly selected sample of five individuals of the same species. The next treatment followed the order control, cologne, small garden skink, and large garden skink. I allowed at least 2 h before presenting animals with another swab. Whip snakes (n = 23) and death adders (n = 21) were tested between 0800 and 1630 h, and trials with small-eyed snakes (n = 22) were carried out between 1800 and 2330 h with the aid of a dim red light. The room temperature was 25 ± 0.5°C, and snakes were always alert and stationary and never attacked the swabs.
Results. The chemoreceptive behavior of snakes did not indicate a preference for different-sized lizards: There was no significant variation in the rate of tongue extrusion of snakes to the chemical cues of large versus small skinks (ANOVA with snake species as a factor and scent as a repeated measure: 1,62 df, F = 0.80, p =.37; Figure 2).
|
Discrimination of chemical cues by lizards
Methods. Chemoreception is a common mechanism by which lizards can
detect (and subsequently respond to) the presence of potential snake predators
(reviewed in Schwenk, 1995). I
used the same procedures described above for snakes to determine the responses
of lizards to snake chemical cues (n = 25 large, 25 small). In
addition to the (1) control and (2) cologne swabs, the skinks were presented
with scents from (3) whip snakes, (4) death adders, and (5) small-eyed snakes.
The trials were conducted in a preset, fixed order whereby each treatment was
presented first to a randomly selected sample of five small skinks and five
large skinks. The next treatment followed the order control, cologne, whip
snake, death adder, and small-eyed snake. My scoring system was based on the
assumption that fleeing from the swab during the trial was a stronger response
than any number of tongue flicks (for a detailed justification of this
procedure, see Burghardt,
1969; Cooper and Burghardt,
1990). Each trial was scored as the total number of tongue flicks
in 60 s if the lizard did not run away from the swab. If the reptile fled from
the swab, I used a base unit (the greatest number of tongue flicks by any
lizard in response to any stimulus) + (60-the latency to run in seconds). In
addition, I analyzed the fleeing data separately from the tongue flick data.
These trials were performed between 0800 and 1630 h at 25 ±
0.5°C.
The responses of snakes and lizards to control scents were extremely weak
(
5.62 ± 0.87 flicks/min) compared with the responses of snakes and
lizards toward the other chemical treatments (see Results); thus, I performed
analyses without these data. I was interested in a reptile's capacity to
respond to biologically meaningful chemical cues rather than its ability to
simply detect a novel stimulus. Thus, the response variable was the difference
between the number of tongue flicks toward reptile scent and the number of
tongue flicks toward cologne scent.
Results. A lizard's body size did not affect its ability to
discriminate predator chemical cues: The rate of tongue extrusion toward the
chemical cues of snakes was similar for small and large lizards (ANOVA with
body size as a factor and scent as a repeated measure: 2,96 df, F =
0.13, p =.42; Figure
2). There was no significant difference in the tendency of small
and large lizards to flee from scent stimuli (number of small vs. large
lizards that fled: 26 vs. 22; 2:1 df, = 0.33, p
=.56).
Activity and movement patterns of lizards
Methods. A lizard's activity and movement patterns may influence
its chances of encountering and being detected by visually oriented predators
such as whip snakes and death adders (e.g.,
Formanowicz et al., 1991). I
observed the behavior of skinks within each enclosure prior to releasing the
snake in the predation encounters. To simulate conditions where a lizard
perceives a potential risk of predation, I covered each enclosure with the
scents of predators by allowing a snake to move over the grass (for trials
with whip-snakes and small-eyed snakes) or placing the paper and rocks from a
snake's cage in a randomly chosen location (for trials with death adders). At
0800 h on the third day after introducing the skinks, I began observations
that continued until 1700 h. I scanned each enclosure for 2 min approximately
each hour and noted the number of small and large lizards that were active
(i.e., not under shelter) and whether they were stationary (this included
basking) or moving.
Results. Body size did not affect the time that lizards spent in shelter, but it did influence their patterns of movement while active. The number of active lizards in each enclosure during the day was not significantly different for large and small lizards (mean ± SE number of active small vs. large lizards/enclosure/h: 1.46 ± 0.05 vs. 1.42 ± 0.04; ANOVA with snake scent and body size as factors; 1,126 df, F = 0.04, p =.83). However, relative to large skinks, a significantly greater proportion of small skinks remained stationary while they were away from their retreat site (mean ± SE % active small vs. large lizards that were stationary: 0.80 ± 0.02 vs. 0.88 ± 0.01; using the same ANOVA model: 1,126 df, F = 9.83, p =.002).
Differential probability of lizards eliciting luring by snakes
Methods. Prey-derived cues stimulate the tail movements of death
adders (Carpenter et al.,
1978, Chizar et al.,
1990) and these snakes may more often attempt to lure lizards of a
particular body size. I tested this idea by recording the distance at which a
death adder began to lure lizards of different body size. Half of the 16
snakes used in the experiment were presented with small lizards in the first
instance, and the other half were first presented with large lizards. These
trials were performed in a large, wooden arena (600 x 650 x 1800
mm) in a room at 25 ± 0.5°C. At one end of an arena, I positioned a
card (150 x 150 mm) on a slightly larger platform (170 x 170 mm)
and heaped sand on the former to a depth of 10 mm. I attached fishing line to
the platform and fed the other end through a hole in the floor at the opposite
end of the arena and passed it to one side. A snake was confined to the card
using an upturned plastic container and left undisturbed for 20 min. I then
rotated the settled snake (by moving the card) so that its head faced the
opposite end of the arena and left it undisturbed for a further 10 min. I
raised a partition located in front of the snake and used fishing line to pull
(from outside the box) a frozen lizard across the width (and back again) of
the opposite end of the arena. I assumed the snake had detected the skink if
luring was stimulated (snakes almost never lure without a prey stimulus;
Chizar et al., 1990, Downes,
unpublished data). If luring was not stimulated, I moved the platform (and
snake) 100 mm closer to the lizard, replaced the partition and left the snake
undisturbed for a further 10 min. This procedure was repeated until the snake
attempted to lure the skink.
Results. A lizard's body size influenced its probability of eliciting a luring response by death adders. Skinks with larger bodies were lured from longer distances than skinks with shorter bodies (mean ± SE detection distance of small vs. large lizards: 130.7 ± 2.28 vs. 136.7 ± 3.03 cm, respectively) but this result was only marginally significant (ANOVA with body size as a repeated measure: 1,14 df, F = 4.85, p =.045).
Retreat site selection by lizards and snakes
Methods. A diurnal lizard's choice of overnight retreat site may
affect its chances of being detected by nocturnal predators such as small-eyed
snakes (Downes and Shine,
2000). After the snakes were introduced to each enclosure during
the predation encounters, I waited for the skinks to become inactive (around
1700 h) and then recorded the overnight retreat site of each individual
(lizards were returned to their retreat after identification). During the
first morning of the encounter I recorded the overnight retreat site of the
snakes before I captured them for scanning. I also scored whether a lizard's
choice of retreat site affected its chances of being the first prey item for
snakes in the predation encounters. This test was only performed for trials
when small-eyes snaked were the predator because whip snakes and death adders
do not forage for lizards that occupy retreat sites.
Results. The retreat site selection behavior of lizards was not
significantly influenced by body size (log-linear model of independence
between body size, snake scent, and retreat site, 12 df, G = 7.78, p
=.802). Lizards usually selected wooden blocks as overnight shelters (mean
± SE number of skinks/enclosure: 0.7 ± 0.1 rock vs. 3.6 ±
0.2 wood vs. 1.7 ± 0.2 grass; 2: 2 df, = 142.24,
p <.001). Whip snakes preferred rocks as retreat sites (20 rock
vs. 1 wood vs. 2 grass; 2: 2 df, = 30.09, p
<.001), whereas small-eyed snakes and death adders preferred grass (2, 0
rock vs. 2, 1 wood vs. 18, 20 grass for small-eyed snakes and adders,
respectively; 2: 2 df, = 23.39, 36.29, p
<.001 in both cases). Snakes and lizards never shared the same retreat
site.
Skinks that sheltered in grass were significantly more likely to be the
first prey for small-eyed snakes than were those that spent the night in
wooden or rock retreat sites (% lizards in each retreat site consumed first:
14 wooden, 30 grass, 0 rock; 2: 2 df, = 44.21,
p =.002).
Flight responses of lizards
Methods. Lizards often become motionless upon detecting danger but
flee toward hiding places when approached by a predator (e.g.,
Dill and Houtman, 1989). This
situation is likely to occur when garden skinks are faced with whip snakes. I
scored whether a lizard's body size affected its tendency to flee from an
approaching snake predator. The predator was a model plastic snake (400 mm
long), and the trials were performed in a large, wooden arena (600 x 650
x 1800 mm) in a room at 25 ± 0.5°C. I attached fishing line
to the snake's neck and fed the other end through a small hole in the floor at
the opposite end of the arena and passed it to one side. A transparent holding
container (50 x 170 x 250 mm) was placed immediately behind the
hole, and the entire arena base was lined with sand. A lizard was allowed to
acclimate inside the chamber for 10 min. I then commenced the trial if the
lizard was at the front of the chamber and facing the snake; otherwise, I
herded the lizard into this position but did not disturb it for another 5 min.
I then raised a partition from in front of the model and swung open the front
of the holding chamber. The model was left stationary for 5 s. I then moved
the model 50 mm closer to the skink and left it stationary for 5 s. This
procedure was repeated until the skink fled to the back of the container or
the snake touched the skink.
Results. Body size significantly affected the flight responses of
skinks: Large lizards were significantly more likely to flee from the
approaching model snake than were small lizards (number of small vs. large
lizards that fled: 16 large vs. 8 small; 2: 1 df, =
5.33, p =.021).
Locomotor performance of lizards
Methods. A lizard's locomotor performance can substantially
influence its chances of being consumed by some snake predators
(Downes and Shine, 2000). I
estimated the locomotor performance of the lizards that were used in chemical
detection trials (n = 25 small, 25 large). Each lizard was tested
over a range of body temperatures (16, 21.5, 27 and 32.5°C) in a
randomized order. I also estimated the locomotor performance at 25°C of
lizards used in the predation encounters. To begin a trial, a lizard was
transferred directly from its container to the holding area of a raceway (40
mm wide), where-upon it was released and allowed to run 1 m. If necessary, the
lizard was chased with an artist's paintbrush. Photocells located at 250-mm
intervals along the runway recorded the cumulative time taken for lizards to
cross each successive infrared beam, and readings were corrected and expressed
as meters per second. Each lizard was run twice with at least 30 min
separating runs. From these data, I calculated the mean sprint speed over 1 m
(average of the runs) and mean burst speed (average of the fastest speed over
any 250-mm segment).
Results. Body size substantially influenced a lizard's running performance at all test temperatures: Small lizards ran significantly slower than large lizards (ANOVA with body size as a factor and temperature as a repeated measure: 1,144 df, F = 11.01, 5.24, p =.002,.027 over 1 m and 0.25 m respectively; Figure 3).
|
I classified the prey within each enclosure as fast (ranked first against
the other two lizards of the same type), average (ranked second), or slow
(ranked third) and scored the status of the first individual consumed by each
snake during the predation encounters. The fastest large individuals were
generally less vulnerable to whip snakes than were slower sprinters of the
same prey type (2: 2 df, = 5.19, 7.73, p
=.07,.02 over 1 m and 0.25 m respectively;
Figure 4). However, there was
no significant interaction between a snake's first prey item and a lizard's
rank speed during predation encounters with adders (2: 2 df,
< 3.78, p .156 over 1 m and 0.25 m) or small-eyed snakes
(2: 2 df, = 0.67, p =.72 over 1 m and 0.25 m;
Figure 4).
|
Attractiveness of caudal lures to lizards
Methods. One possible explanation for size-dependent predation by
death adders is that a lizard's body size affects its tendency to be attracted
to the lure of this snake. I recorded staged laboratory encounters on
videotape to measure the efficiency of death adders (n = 16) in
luring small versus large garden skinks. The encounters were staged in
transparent plastic containers (220 x 300 x 100 mm) lined with
river sand (10 mm depth) and sparse leaf litter was provided at one end. I
encouraged a death adder to select an ambushing site at this sheltered end by
restricting it to this area (using a division) and leaving it undisturbed for
30 min. The division was removed after this time, and a lizard entered the
unsheltered end of the cage through an opening in the side of arena. All
snakes began luring immediately after a skink entered the cage, thereby
controlling for potential differences in the tendency of adders to lure
lizards of different sizes. The variable of interest was therefore the time
that it took a lizard to be lured to the snake. Half of the snakes were
presented with large individuals in the first instance. From the videotapes, I
noted the time that the snake first began to lure the skink (begin lure) and
first struck at the skink (first strike). From these times, I determined the
luring time of the skink (first strike—begin lure). I also noted whether
the skink was captured by the first strike of the snake predator.
Results. A lizard's body size affected its tendency to be
attracted to the lure of death adders. The time between onset of luring and
first strike attempt by adders was significantly shorter for small lizards
than for large lizards (mean ± SE luring time for small and large
lizards: 75.33 ± 13.33 vs. 257.0 ± 50.17; ANOVA with body size
as a repeated measure: 1,14 df, F = 8.50, p =.01).
Additionally, a larger number of small lizards (14) were captured by the first
strike of the adder compared to large lizards (10), but this pattern was not
statistically significant (2: 1 df, = 3.33, p
=.17).
Frequency of tail loss by lizards
Methods. Tail loss is often used by garden skinks to evade capture
by snakes (Downes and Shine,
2000). I scored the proportion of small and large lizards used
throughout the study that had regenerated tails. The X-ray scope used to
identify individual prey inside snakes also produced an image of the outline
of a skink that indicated the presence or absence of a tail
(Downes, 2000). I could thus
determine the number of consumed small and large lizards that were tailless
inside the stomachs of snakes during the predation encounters. I did not
compare the proportion of uneaten small and large lizards that were tailless
at the termination of the predation encounters because too few (5) lizards
of each prey type remained.
Results. In my study population, a significantly greater
proportion of large lizards had regenerated tails compared with small lizards
(% large vs. small lizards with regenerated tails = 75 vs. 35, n =
256 vs. 259; 2: 1 df, = 84.20, p <.001).
During the predation encounters with small-eyed snakes, I never found a
recently autotomized lizard within a predator. During the predation encounters
with adders, a similar proportion of consumed large and small lizards were
tailless (% tailless lizards: 7/58 large vs. 2/61 small; 2: 1
df, = 2.82, p =.17). During the predation encounters with whip
snakes, large skinks were more likely to use tail loss before capture than
were small skinks (% tailless lizards: 14/66 large vs. 2/64 small;
2: 1 df, = 7.76, p =.007). I could not assess
at what point during the predation encounters the lizards employed
autotomy.
Time to handle lizards
Methods. An attack is successful only if the predator can kill and
consume the captured prey. I recorded staged laboratory encounters on
videotape to measure the time it took snakes to kill and consume small and
large skinks. These experiments were performed at 25 ± 0.5°C in
glass terraria (400 x 250 x 100 mm) devoid of shelter to permit
the rapid capture of prey. Skinks were introduced to terraria containing snake
predators. Half of the snakes of each species were presented with small
individuals in the first instance, and the remaining snakes were first
presented with large lizards. From the videotapes, I timed the period between
the first seizure of the skink and the swallowing of this prey to the base of
its tail (handling time). All species of snake maintained a firm bite during
envenomation.
Results. The difference in handling time of large and small lizards was much greater for death adders than for whip snakes and small-eyed snakes (ANOVA with snake species as the factor and body size as a repeated measure: species x body size, 2,63 df, F = 94.01, p <.001). I therefore analyzed the data for whip snakes and small-eyed snakes separately from the data for adders. Whip snakes and small-eyed snakes took significantly longer to kill and consume large lizards than small lizards (mean ± SE time to handle small vs. large lizards: 68.8 ± 4.8 vs. 103.1 ± 4.7 s; using the previous ANOVA model: 1,43 df, F = 150.53, p <.001). A similar result was obtained for adders (mean ± SE time to handle small vs. large lizards: 137.8 ± 11.6 vs. 305.6 ± 37.0 s; ANOVA with first lizard type as a factor and body size as a repeated measure, 1,20 df, F = 138.69, p <.001).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONDISCUSSIONREFERENCES |
|---|
The ideal data for examining size-selective predation would be based on individuals of considerably different body size but the same age and experience with predators. It may be possible to collect such data by experimentally manipulating the phenotypes of lizards (e.g., Downes and Shine, 1999
;
Sinervo et al., 1992;
Sorci and Clobert, 1997);
however, any such interference will introduce other potentially confounding
variables. Instead, I used lizards that probably differed in age (and in
potential exposure to predators) by less than 1 year. Nevertheless, I found
that large, free-living lizards were more likely to have regenerated tails
than small lizards, thus indicating that yearlings had experienced more direct
encounters with predators than hatchlings (tail breaks are unlikely to result
from intraspecific agonistic encounters;
Torr and Shine, 1996).
Therefore, some of the behavioral differences between large and small lizards
documented in this study obviously warrant interpretation with respect to the
possible effects of age and predatory experience (see below). Henceforth, my
use of the term "body size" refers to the joint effects of body
size, age, and potential experience with predators.
Field collections of snakes reveal nonrandom predation on lizard prey of
different body sizes (see Introduction). My study extends these findings by
providing empirical evidence of differential patterns of size-dependent
predation among several elapid snake predators under conditions where predator
and prey densities are controlled. I found that whip snakes most often
consumed large lizards, death adders most often consumed small lizards, and
small-eyed snakes consumed large and small lizards equally often. Differential
predation may be due to active choice behavior of the predator or differential
vulnerability of the prey, or both
(Onkonburi and Formanowicz,
1997; Pastorok,
1981; Schwarzkopf and Shine,
1992). Results of additional experiments demonstrate that patterns
of size-selective predation in this system are determined by differential prey
vulnerability rather than by active predator choice. Two lines of
complementary evidence support this conclusion. First, none of the snake
predators actively selected between lizards of different body size. Second,
defensive behavior differs between lizards of different body size and appears
to influence their chances of being consumed by different snake predators.
To test whether snakes actively selected certain prey, I presented
individual predators in outdoor enclosures with two different-sized, inert
lizards and scored which prey was consumed first. This procedure controlled
for potential variation between small and large lizards in behavior that may
prevent a snake from obtaining its preferred prey (e.g., encounter rates,
captures per attack). Garden skinks are socially gregarious
(Torr and Shine, 1996);
indeed, they were usually found close to (<100 mm from) at least one other
lizard over the day (in 313 of 413 cases) and night (in 110 of 146 cases)
during predation encounters. Thus, I examined possible selection (or
rejection) by snakes between two lizards encountered at the same time, rather
than selection (or rejection) of lizards encountered at different times. I
expected that the snakes would primarily consume the energetically most
profitable (larger) prey (e.g., Griffiths,
1980; Onkonburi and
Formanowicz, 1997), but found that they did not bias their
attacks. One possible explanation for this lack of prey choice is that snakes
were not able to distinguish the large lizards from the small lizards. For
instance, small-eyed snakes probably use chemoreception to recognize prey, but
they did not discriminate among the scents of large and small lizards.
Similarly, foraging whip snakes orient toward moving prey but often move past
motionless lizards (Downes, personal observation); this behavior suggests that
they may have poor close-range vision. Regardless of the mechanisms, this
result suggests that size-dependent predation during the direct encounter
trials was not caused by active selection by snakes.
Small and large lizards differed in several aspects of defensive behavior that are likely to influence their vulnerability to predation. Vulnerability to predation is apparently influenced by both behavioral characteristics of the prey and foraging tactics of the predators. Because the three snake species used in this study differ profoundly in foraging behavior, I separately discuss size-dependent vulnerability of lizard prey to each of these predators.
Whip snakes are strictly diurnal predators that forage widely for active
prey. Movement typically alerts whip snakes to the whereabouts of prey, and
garden skinks become motionless upon detecting potential danger. When
approached by whip snakes, these lizards flee toward hiding places and thereby
elicit attack from this visually oriented species (Downes, personal
observation; Downes and Shine,
2000; Scanlon,
1998). A lizard's body size did not affect its rate of tongue
extrusion or tendency to flee in response to the scents of whip snakes; thus
small and large lizards were presumably equally likely to detect the presence
of this predator via chemoreception. Active large lizards were less mobile
than active small lizards, and such behavioral variation presumably decreased
the chances of large lizards being detected by whip snakes (e.g.,
Formanowicz et al., 1991;
Skelly, 1994). Instead,
size-specific variation in the flight responses of lizards probably resulted
in higher capture rates of large lizards, with skinks in this size class being
pursued more often because they fled sooner from approaching snakes
(Downes and Shine, 2000;
Kanou and Shimozawa, 1983).
Garden skinks rely heavily on running to escape predators
(Qualls and Shine, 1998).
However, size selectivity by whip snakes was not causally related to a
lizard's absolute locomotor performance: Larger lizards were substantially
faster than smaller conspecifics but were most susceptible. In this case,
inciting attack focuses the attention of the snake on a certain size class of
prey regardless of their running speed. A lizard's running speed did, however,
affect its vulnerability within the size class selected by whip snakes: Faster
large lizards were significantly less vulnerable than slower large lizards
(see also Wassersug and Sperry,
1977; Watkins,
1996). Thus, snakes that simultaneously pursued several large
lizards may have most easily captured the slowest of the group.
Death adders are sit-and-wait predators that initiate capture attempts from
a sedentary position. Prey-derived cues stimulate the tail movements that
function to lure potential prey (Chizar et
al., 1990); the success of this strategy relies on the snakes
being cryptic before attack. Presumably, garden skinks rarely see death adders
but use chemical cues to estimate the snakes' location. A lizard's body size
affected the adder's tendency to use caudal luring. Large skinks elicited
snakes to lure at a distance that was larger than that for small lizards.
Although this result was only marginally significant (p =.045), it
suggests that the onset of tail movements by death adders is differentially
stimulated by cues from large versus small lizards (see
Chizar et al., 1990). These
cues may be visual (i.e., larger skinks may be easier for adders to observe
from a distance) and/or motivational (i.e., snakes may expend more effort to
attract skinks that are more profitable). In any case, these data do not
support the notion that the increased consumption of small lizards by death
adders may have resulted from snakes attempting to lure only small lizards.
Instead, active small lizards were highly mobile relative to active large
lizards, and thus probably more likely to be detected by an ambushing snake
(e.g., Clorec, 1976;
Cresswell, 1996;
Huey and Pianka, 1981). Small
lizards also tended to be more readily lured by tail movements than large
lizards, and small lizards were less likely to escape the first attempted
strike by a death adder. Such variation in the behavior of small and large
lizards may reflect size-related differences in the diets, foraging behavior,
predatory experience, and predator avoidance mechanisms of lizards. For
instance, small lizards may be relatively mobile because they allocate more
time to foraging than large lizards (e.g.,
Melville and Swain, 1997), and
perhaps their attraction to tail movements is greater because this stimulus
resembles prey items that large lizards do not include in their diet (e.g.,
Ballinger et al., 1977;
Best and Pfaffenberger, 1987).
Additionally, small lizards may have lower reaction times and be less
maneuverable than large lizards, thereby making it easier for snakes to
capture them (e.g., Webb, 1976
and references within).
Small-eyed snakes are nocturnal predators that widely forage for inactive
lizards. They rely heavily on chemoreception to locate potential prey (e.g.,
Kubie and Halpern, 1975;
Figure 2a) and actively search
within crevices and tufts of vegetation and under logs
(Downes and Shine, 1998).
Garden skinks are sequestered within retreat sites when captured by this
species and probably rarely evade capture
(Downes and Shine, 2000). A
lizard's body size did not affect its selection of overnight retreat site (cf.
Asplund, 1968,
1974;
Stamps, 1988). Body size
therefore was not an important determinant of a lizard's probability of being
captured by this species (e.g., Christian
et al., 1984). Instead, a lizard's chance of being consumed was
determined primarily by its selection of overnight shelter sites, probably
because the snakes were more likely to detect lizards in some retreat sites
than others (e.g., Pastorok,
1981; Peckarsky and Penton,
1989). For example, small-eyed snakes may have concentrated their
search in grass microhabitats because they usually selected these areas as
diurnal retreat sites. Chemoreception probably affected the detection rates of
skinks by small-eyed snakes (Downes,
1999; Schwenk,
1995), but I do not know how this factor might have varied among
retreat sites.
Different-sized garden skinks did not shift their behavior (activity level,
movement patterns, retreat site selection) to accord with foraging tactics
that were likely to be used by potential predators. Such modifications might
occur if small or large prey use antipredator tactics that are costly to other
behavioral processes and alter vulnerability to some predators but not to
others (e.g., Downes and Shine,
1998; Griffiths et al.,
1998; Pastorok,
1981). For instance, small lizards may be relatively mobile
because they allocate more time to foraging than do large lizards (e.g.,
Melville and Swain, 1997), and
this behavior contributed to their heightened vulnerability to death adders
but not to small-eyed snakes. Thus, I might have expected small lizards to be
relatively mobile under risk from small-eyed snakes but relatively immobile
when faced with death adders (e.g., Downes
and Shine, 1998; Griffiths et
al., 1998; McPeek,
1990). However, at least 13 species of lizard-eating elapid snakes
are distributed sympatrically with garden skinks
(Cogger, 1994), and these
predators use various foraging tactics and are patchily distributed. Hence,
different-sized garden skinks may not alter behavior to efficiently evade
specific size-selective predators because such overspecialization may incur
too high a cost in terms of the ability to avoid a wide range of snake
predators (e.g., Pastorok,
1981).
My data demonstrate differential patterns of size-dependent predation among several elapid snakes and suggest that such patterns do not arise because of active prey selection (or rejection). Instead, variation in the defensive behavior of small and large lizards resulted in nonrandom predation, but only when the foraging tactics of the predator and the defense tactics of the prey operated during the same phase of predation. Behavioral processes that differentially affected the chances that small and large lizards would be consumed by snakes therefore varied considerably among predators using different foraging strategies. Future studies of size-dependent predation should be designed not only to test predictions of differential predation based on the premise that evolution should favor energetically efficient predation, but also to test the importance of prey vulnerability as a proximate mechanism of nonrandom prey consumption.
|
ACKNOWLEDGEMENTS |
|---|
This study would not have been possible without the X-ray scope loaned by M. Guinea. P. Birch, M. Elphick, M. Fitzgerald, P. German, H. Giragossyian, Northwest Marine Technology, R. Shine, M. Whicker, and P. Whitaker provided equipment or valuable field assistance. I thank D. Bauwens, P. Borges, M. Elgar, M. Fitzgerald, A. Forsman, K. Handasyde, M. Olsson, R. Shine, T. Underwood, J. Webb, and P. Whitaker for inspiration, statistical guidance, and/or discussion and comment on the manuscript. This project was approved by the University of Sydney Animal Care and Ethics Committee and conducted under Permit B1646 and Radioactive Materials License 21096. It was funded by the American Society of Ichthyologists and Herpetologists, Animal Behavior Society, Australian Federation of University Women, Australian Geographic Society, Ecological Society of Australia, Linnean Society of NSW, Royal Society of New South Wales, Royal Zoological Society of NSW, and Society for the Study of Amphibians and Reptiles (grants to S.D.). The enclosures used in this study were constructed with funds from the Australian Research Council (grant to Rick Shine).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONDISCUSSIONREFERENCES |
|---|
Allan J, Flecker A, 1988. Prey preference in stoneflies: a comparative analysis of prey vulnerability. Oecologia 76: 496-503.
Arnold S, 1993. Foraging theory and prey-size-predator-size relations in snakes. In: Snakes. Ecology and behaviour (Seigel R, Collins J, eds). New York: McGraw-Hill; 87-115.
Asplund KK, 1968. Evolution of body size and habitat selection in whiptail lizards (PhD dissertation). Los Angeles: University of California.
Asplund KK, 1974. Body size and habitat utilization in whiptail lizards (Cnemidophorus). Copeia 1974: 695-703.
Ballinger RE, Newlin ME, Newlin SJ, 1977. Age-specific shift in the diet of the crevice spiny lizard Sceloporus poinsetti in southwestern New Mexico. Am Midl Nat 97: 482-484.
Bartholomew GA, Tucker VA, 1964. Size, body temperature, thermal conductance, oxygen consumption, and heart rate in Australian varanid lizards. Physiol Zool 37: 341-354.
Bauwens D, Garland T Jr, Castilla A, Van Damme R, 1995. Evolution of sprint speed in lacertid lizards: morphological, physiological, and behavioral covariation. Evolution 49: 848-863.
Best TL, Pfaffenberger GS, 1987. Age and sexual variation in the diet of collared lizards (Crotaphytus collaris). Southwest Nat 32: 415-426.
Brown KK, Ruby DE, 1977. Sex-associated variation in the frequencies of tail autotomy in Sceloporus jarrovi (Sauria: Iguanidae) at different elevations. Herpetologica 33: 380-387.
Burghardt G, 1969. Comparative prey-attack studies in newborn snakes of the genus Thamnophis. Behaviour 33: 77-114.[Web of Science]
Carothers JH, 1983. Size-related activity patterns in an herbivorous lizard. Oecologia 57: 103-106.
Carpenter CC, Murphy JB, Carpenter GC, 1978. Tail luring in the death adder, Acanthophis antarcticus (Reptilia, Serpentes, Elapidae). J Herpetol 12: 574-577.
Carrascal LM, López P, Martín J, Salvador A, 1992. Basking and antipredator behavior in a high altitude lizard: implications of heat-exchange rate. Ethology 92: 143-154.
Castilla AM, Bauwens D, 1991. Observations on the natural history, present status, and conservation of the insular lizard Podarcis hispanica atrata on the Columbretes Archipelago, Spain. Biol Conserv 58: 69-84.
Chizar D, Boyer D, Lee R, Murphy J, Radcliff C, 1990. Caudal luring in the southern death adder, Acanthopis antarcticus. J Herpetol 24: 253-260.
Christian KA, Tracy CR, Porter WP, 1984. Physiological and ecological consequences of sleeping-site selection by the Galapagos land iguana (Conolophus pallidus). Ecology 65: 752-758.
Clorec A, 1976. Interactions between different receptors involved in prey capture in Ranatra linearis. Biol Behav 1: 251-266.
Cogger H, 1994. Reptiles and amphibians of Australia. Sydney: Reed Books.
Cooper WE Jr, Burghardt G, 1990. A comparative analysis of scoring methods for chemical discrimination of prey by squamate reptiles. J Chem Ecol 16: 45-65.
Cresswell W, 1996. Surprise as a winter hunting strategy in sparrow hawks Accipiter nisus, peregrines Falco peregrinus and merlins F. columbarius. Ibis 138: 684-692.
Cruz-Neto A, Andade D, Abe A, 1999. Energetic cost of predation: aerobic metabolism during prey ingestion by juvenile rattlesnakes, Crotalus durissus. J Herpetol 33: 229-234.
Dial BE, Schwenk K, 1996. Olfaction and predator detection in Coleonyx brevis (Squamata: Eublepharidae), with comments on the functional significance of buccal pulsing in geckos. J Exp Zool 276: 415-424.[Web of Science]
Dill LM, Houtman R, 1989. The influence of distance to refuge on flight-initiation distance in the gray squirrel (Sciurus carolinensis). Can J Zool 67: 232-235.
Downes SJ, 1999. Prey odour influences retreat-site selection by naive snakes. J Herpetol 33: 156-159.
Downes SJ, 2000. The use of wire microtags to identify small individual prey in snakes. Amphib-Reptilia 21: 126-131.
Downes SJ, Shine R, 1998. Sedentary snakes and gullible geckos: predator-prey coevolution in nocturnal rock-dwelling reptiles. Anim Behav 55: 1373-1385.[Web of Science][Medline]
Downes SJ, Shine R, 1999. Do incubation-induced changes to a lizard's phenotype affect its vulnerability to predators? Oecologia 120: 9-18.
Downes SJ, Shine R, 2000. Why does tail loss increase a lizard's later vulnerability to snakes? Ecology 82: 1293-1303.
Edmunds M, 1974. Defense in animals: a survey of anti-predator defenses. London: Longman.
Endler J, 1986. Defence against predators. In: Predator-prey relationships (Feder ME, Lauder GV, eds). Chicago: University of Chicago Press; 109-134.
Feder M, Arnold S, 1982. Anaerobic metabolism and behavior during predatory encounters between snakes (Thamnophis elegans) and salamanders (Plethodon jordani). Oecologia 53: 93-97.
Ferguson G, Talent L, 1993. Life-history traits of the lizard Sceloporus undulatus from two populations raised in a common laboratory environment. Oecologia 93: 88-94.[Web of Science]
Ford N, Burghardt G, 1993. Perceptual mechanisms and the behavioural ecology of snakes. In: Snakes. Ecology and behaviour (Seigel R, Collins J, eds). New York: McGraw-Hill; 117-164.
Formanowicz DR Jr, Brodie E Jr, Bradley P, 1991. Behavioral compensation for tail loss in the ground skink, Scincella lateralis. Anim Behav 40: 782-784.
Forsman A, 1996. Body size and net energy gain in gape-limited predators: a model. J Herpetol 30: 307-319.
Greene HW, 1983. Dietary correlates of the origin and radiation of snakes. Am Zool 23: 431-441.
Greene HW, 1984. Feeding behavior and diet of the eastern coral snake, Micrurus fulvius. In: Vertebrate ecology and systematics—a tribute to Henry S. Fitch (Seigel RA, Hunt LE, Knight JL, Malaret L, Zuchslag NL, eds). Lawrence, Kansas: Museum of Natural History; 147-162.
Griffiths D, 1980. Foraging costs and relative prey size. Am Nat 116: 743-752.
Griffiths R, Schley L, Sharp P, Dennis J, Roman A. 1998. Behavioral responses of Mallorcan midwife toad tadpoles to natural and un-natural snake predators. Anim Behav 55: 207-214.[Web of Science][Medline]
Huey RB, Hertz PE, 1982. Effects of body size and
slope on sprint speed of a lizard (Stello (Agama) stellio). J
Exp Biol 97:
401-409.
Huey RB, Hertz PE, 1984. Effects of body size and
slope on acceleration of a lizard (Stellio stellio). J Exp
Biol 110:
113-123.
Huey RB, Pianka E, 1981. Ecological consequences of foraging mode. Ecology 62: 991-999.[Web of Science]
Kaiser BW, Mushinsky HR, 1985. Tail loss and dominance in captive adult male Anolis sagrei. J Herpetol 28: 342-346.
Kanou M, Shimozawa T, 1983. The elicitation of the
predatory labial strike of dragonfly larvae in response to a purely mechanical
stimulus. J Exp Biol 107:
391-404.
Kubie J, Halpern M, 1975. Laboratory observations of trailing behavior in garter snakes. J Comp Physiol Psychol 89: 667-674.[Web of Science][Medline]
McPeek M, 1990. Behavioral differences between Enallagma species (Odonata) influencing differential vulnerability to predators. Ecology 71: 1714-1726.
Melville J, Swain R, 1997. Daily and seasonal activity patterns in two species of high altitude skink, Niveoscincus microlepidotus and N. metallicus, from Tasmania. J Herpetol 31: 29-37.
Mushinsky HR, 1987. Foraging ecology. In: Snakes. Ecology and evolutionary biology (Seigel RA, Ford NB, Novak SS, eds). New York: Macmillan; 302-334.
Mushinsky HR, Hebrard JJ, Vodopich DS, 1982. Ontogeny of water snake foraging ecology. Ecology 631624-1629.
Onkonburi J, Formanowicz D Jr, 1997. Prey choice by predators: effect of prey vulnerability. Ethol Ecol Evol 9: 19-25.
Pastorok R, 1981. Prey vulnerability and size selection by Chaoborus larvae. Ecology 63: 1311-1324.
Peckarsky BL, Penton MA, 1989. Mechanisms of prey selection by stream-dwelling stoneflies. Ecology 70: 1203-1218.
Pough F, Andrews R, 1985. Energy costs of subduing and swallowing prey for a lizard. Ecology 66: 1525-1533.
Pyke GH, 1984. Optimal foraging theory: a critical review. Annu Rev Ecol Syst 15: 523-575.[Web of Science]
Qualls F, Shine R, 1998. Geographic variation in `costs of reproduction' in the scincid lizard Lampropholis guichenoti. Funct Ecol 11: 757-763.
Scanlon J, 1998. Prey-scaring by visual pursuit predators: a new use for tail-waving in snakes. Herpetofauna 28: 5-10.
Schwarzkopf L, Shine R, 1992. Costs of reproduction in lizards: escape tactics and susceptibility. Behav Ecol Sociobiol 31: 17-25.[Web of Science]
Schwenk K, 1995. Of tongues and noses: chemoreception in lizards and snakes. Trends Ecol Evol 10: 7-12.
Shine R, 1980. Ecology of the Australian death adder Acanthophis antarcticus (Elapidae): evidence for convergence with the Viperidae. Herpetologica 36: 281-289.
Shine R, 1983. Reptilian viviparity in cold climates: testing the assumptions of an evolutionary hypothesis. Oecologia 57: 397-405.
Shine R, 1984. Reproductive biology and food habitats of the Australian elapid snakes of the genus Cryptophis. J Herpetol 18: 33-39.
Shine R, 1991. Why do larger snakes eat larger prey items? Funct Ecol 5: 493-502.
Sih A., Moore R, 1990. Interacting effects of predator and prey behavior in determining diets. In: Behavioural mechanisms of food selection (Hughes R, ed). Heidelberg: Springer-Verlag; 771-796.
Simbotwe MP, 1985. Sexual dimorphism and reproduction of Lampropholis guichenoti (Lacertilia: Scincidae). In: The biology of Australasian frogs and reptiles (Grigg G, Shine R, Ehmann H, eds). Sydney: Royal Zoological Society of New South Wales; 11-16.
Sinervo B, Doughty P, Huey R, Zamudio K, 1992.
Allometric engineering: a causal analysis of natural selection on offspring
size. Science 258:
1927-1930.
Skelly DK, 1994. Activity level and the susceptibility of anuran larvae to predation. Anim Behav 47: 465-468.
Smith D, 1997. Ecological factors influencing the
anti-predator behaviors of the ground skink, Scincella lateralis.
Behav Ecol 8:
622-629.
Sorci G, Clobert J, 1997. Environmental maternal effects on locomotor performance in the common lizard Lacerta vivipara. Evol Ecol 11: 531-541.
Stamps JA, 1988. The effect of body size on habitat and territory choice in juvenile lizards. Herpetologica 44: 369-376.
Stephens DW, Krebs JR, 1986. Foraging theory. Princeton, New Jersey: Princeton University Press.
Torr G, Shine R, 1996. Patterns of dominance in the small scincid lizard Lampropholis guichenoti. J Herpetol 30: 230-237.
Vermeij G, 1982. Unsuccessful predation and evolution. Am Nat 120: 701-720.
Voris HK, Voris HH, 1983. Feeding strategies in marine snakes: an analysis of evolutionary, morphological, behavioral and ecological relationships. Am Zool 23: 411-425.
Wassersug RJ, Sperry DG, 1977. The relationship of locomotion to differential predation on Pseudacris triseriata (Anura: Hylidae). Ecology 58: 830-839.
Watkins T, 1996. Predator-mediated selection on burst swimming performance in tadpoles of the pacific tree frog, Pseudacris regilla. Physiol Zool 69: 154-167.
Webb J, Shine R, 1993. Prey-size selection, gape limitation and predator vulnerability in Australian blindsnakes (Typhlopidae). Anim Behav 45: 1117-1126.
Webb PW, 1976. The effect of size on the fast-start
performance if rainbow trout Salmo gairdneri, and a consideration of
piscivorous predator-prey interactions. J Exp Biol
65: 157-177.
Werner EE, Hall DJ, 1974. Optimal foraging and the size selection of prey by the blue-gill sunfish (Lepomis macrochirus). Ecology 55: 1042-1052.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||