Eastern Illinois UniversityThe Keep
Masters Theses Student Theses & Publications
2000
The Effect of Predator Chemical Cues andConspecific Alarm Signals upon Behavior in Earlyand Late Developmental Stages of American Toad(Bufo americanus) TadpolesCarol Lynette JohnsonEastern Illinois UniversityThis research is a product of the graduate program in Biological Sciences at Eastern Illinois University. Findout more about the program.
This is brought to you for free and open access by the Student Theses & Publications at The Keep. It has been accepted for inclusion in Masters Thesesby an authorized administrator of The Keep. For more information, please contact [email protected].
Recommended CitationJohnson, Carol Lynette, "The Effect of Predator Chemical Cues and Conspecific Alarm Signals upon Behavior in Early and LateDevelopmental Stages of American Toad (Bufo americanus) Tadpoles" (2000). Masters Theses. 1607.https://thekeep.eiu.edu/theses/1607
THESIS/FIELD EXPERIENCE PAPER REPRODUCTION CERTIFICATE
TO: Graduate Degree Candidates (who have written formal theses)
SUBJECT: Permission to Reproduce Theses
The University Library is receiving a number of request from other institutions asking permission to reproduce dissertations for inclusion in their library holdings. Although no copyright laws are involved, we feel that professional courtesy demands that permission be obtained from the author before we allow these to be copied.
PLEASE SIGN ONE OF THE FOLLOWING STATEMENTS:
Booth Library of Eastern Illinois University has my permission to lend my thesis to a reputable college or university for the purpose of copying it for inclusion in that institution's library or research holdings.
Date
I respectfully request Booth library of Eastern Illinois University NOT allow my thesis to be reproduced because:
Author's Signature Date
rt>es1s4 form
The Effect of Predator Chemical Cues and Conspecific Alarm Signals upon Beha,·ior in
Early and Late Developmental Stages of American toad (fu![Q americanus) Tadpoles.
(TITLE)
DATE
DATE
BY
Carol Lynette Johnson
/Ci 7 fi, -
THESIS
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
Masters of Zoology
IN THE GRADUATE SCHOOL, EASTERN ILLINOIS UNIVERSITY CHARLESTON, ILLINOIS
2000
YEAR
I HEREBY RECOMMEND TH.A.T THIS THESIS BE ACCEPTED AS FULFILLING THIS PART OF THE GRADUATE DEGREE CITED ABOVE
Table of Contents
Acknowledgments ........................................ . ...... . .. . ...... ii
Abstract ............ . .. ...... . .. .... . ............. . ......................... .. iii
Introduction .............................................. . .... . .. .... . ...... 1
Methods ..... . ....................... . ................ . ........ . .... .......... 5
Results .................. . .. .. .. . .. ... ... . ..................................... 10
Figures ....... . ... . .... . .. .................. . .................................. 13
Table 1 ..................... . .. . ... . .. ...... . .................................. 22
Discussion ... . ........ . .. .. ... ; ........................................ . ..... 23
Literature Cited ..................... . ...... . ............................... 28
Acknowledgements
I first wish to express gratitude to my committee members: Dr. Kipp Kruse, Dr.
Robert Fischer, and Dr. Charles Pederson for their insight and support throughout the
duration of my work. I also would like to thank Dr. Jeff Laursen for his invaluable
direction in the collection and housing of my numerous tadpoles. Many thanks also to
family, friends, and Ted Watson for all their encouragement, companionship and never
ending support throughout my time at Eastern.
All tadpoles and predatory beetles were collected by permit on the Fox Ridge
State Park, whose staff was extremely helpful; bluegill were collected from the Campus
Pond at Eastern. EIU's biology department provided laboratory space, equipment, and
some financial support. A grant from Eastern Illinois University was instrumental in
funding this thesis.
ii
Abstract
The chemosensory capability and subsequent habitat choice of larval American
toad (Bufo americanus) tadpoles were quantified using choice of refuge with both early
and late developmental stages. Treatments were performed with two tadpole densities to
ascertain the effect of social aggregations upon behavior. Bluegill (1epomis
macrochirus) and predaceous diving beetle larvae (Dytiscus ~.)predators were used to
condition water. Conspecific tadpoles as well as Southern leopard frog (Rana
sphenocephala) tadpoles were used to prepare treatment extracts.
Tadpole density (n= IO and n=20) had no significant effect upon the percentage of
tadpoles seeking cover in any treatment. The percentage of toad tadpoles of early
developmental stages seeking cover was significantly higher when exposed to both early
and late staged conspecific alarm signals as compared to the control, however, the
treatment responses were statistically indistinguishable from one another. Similarly, the
presence of beetle larvae in test chambers elicited a strong fright reaction by tadpoles
when compared to controls. Tadpoles responded more strongly to the presence of beetle
larvae than to chemical cues oflarval beetles. Likewise, the addition of bluegill cues
elicited a significant antipredator response from toad tadpoles.
Late developmental stage toad tadpoles showed no significant increases in the
percentage seeking cover in any treatment; these results are attributable to the high
percentage of larvae seeking cover in control trials. My results suggest that tadpoles may
display antipredator tactics that are stage-specific and geared toward differing suites of
predators.
111
Introduction
Predation is a constant threat throughout the life of an amphibian; therefore
behavioral responses appropriate to the developmental stage may be shaped by natural
selection. The antipredator responses of amphibian larvae have been well studied;
tadpoles are known to respond to predators in a variety of ways ranging from altering life
history strategies (Skelly and Werner 1990, Werner and Anholt 1993) to various
behavioral responses (Petranka et al.1987, Skelly 1994). Behavioral antipredator tactics
include altering activity levels (Lawler 1989, Anholt et al. 1996), seeking refuge (Babbitt
and Tanner 1997), and swarming (Watt et al. 1997). A distinct evolutionary advantage
would exist for tadpole species able to detect predators before contact occurs, thereby
increasing the escape interval as well as the possibility of avoiding a predation episode
entirely. Several species of larval amphibians, including toads (Bufonidae ), use indirect
signals (Petranka and Hayes 1998) to detect and respond to substances associated with
predators including diet and waste metabolites (Hews 1988, Lawler 1989, Wilson and
Lefcort 1993, Laurila et al. 1997, Petranka and Hayes 1998), as well as alarm signals
issuing from the skin of damaged conspecifics (Hews and Blaustein 1985, Petranka and
Hayes 1998, Kiesecker et al. 1999).
Larvae of the American toad, (Bufo americanus) are capable of further
antipredator tactics. They exhibit aposematic coloration which alerts predators to their
toxic and noxious qualities caused by a chemical called bufotoxin, which is only
produced late in the larval period (Brodie et al. 1978) beginning at developmental stage
41 Gosner (1960). This production ofbufotoxin by granular glands in the skin just prior
to the onset of metamorphosis causes toad tadpoles to be unpalatable or even
occasionally toxic (Brodie at al. 1978) to several species of predators, including some
insect predators (e.g., dytiscids) and fish (Brodie et al. 1978, Formanowicz and Brodie
1982),. In earlier developmental stages, however, tadpoles may be fully palatable
(Brodie et al. 1978, Semlitsch and Gavasso 1992) and consequently much more
susceptible to predation.
Toad tadpoles form complex aggregations (Beiswenger 1975, Rodel and
Linsenmair 1997, Watt et al 1997), in which members both recognize and preferentially
associate with siblings (Waldman and Adler 1979, Waldman 1982). This schooling
behavior bas several potential benefits to members, including enhanced predator
vigilance and food location (Beiswenger 1975), along with increased survival due to the
dilution effect (Watt et al. 1997). These groups may also lend protection to palatable
individuals. Used in conjunction with aposematism and kin preference, swarming is
instrumental in increasing inclusive fitness of swarm members (Brodie and Formanowicz
1981, Brodie and Formanowicz 1987) by facilitating avoidance of unpalatable larvae by
predators. Social aggregations are known to be comprised of mainly early stage tadpoles,
whereas later stage tadpoles (Gosner 1960 stage 41 or later) have been demonstrated to
be much less social (Beiswenger 1975).
Bufonids also will modify activity levels, avoid areas containing predators (Skelly
and Werner 1990, Anholt et al. 1996), and utilize cover in escaping predators (Babbitt
and Jordan 1996, Babbitt and Tanner 1997). These tactics, while effective against some
predators, are extremely costly in terms of growth (Kupferberg 1997). Therefore,
tadpoles need to adaptively balance the risk of predation they experience against the
2
percentage of time spent foraging. At the time of metamorphic climax, tadpoles may be
exceptionally vulnerable to predators (Arnold and Wassersug 1978, Crump 1984) as
limbs develop and the tail is reabsorbed which has been shown to severely decrease
swimming velocity beginning at Gosner (1960) developmental stage 42 (Huey 1980).
Because tadpoles cease feeding at stage 42 (Beisenger 1975) they may have no need to
engage in foraging, and therefore may appropriate more time to the avoidance of
predators. Given the differing physiologies, nutitive needs, and swimming capabilities, it
seems possible that toad tadpoles may exhibit antipredator tactics that are stage-specific.
To further explore the role of chemo-sensory capability in mediating behavior
choice in early and late developmental stage American toad tadpoles, several questions
were posed and then translated into laboratory experiments. First, what is the extent of
chemo-sensory capabilities in both early and late stage tadpoles? Second, how do these
chemo-sensory capabilities affect behavior of early and late stage tadpoles?
Experimental treatments were performed using chemical cues of conspecifics,
heterospecifics, and two predators to condition water. Treatments were performed with
two densities of tadpoles (n=IO, n=20) to determine how and if social aggregations
affected behavior, and choice of refuge was used to quantify behavior. Predaceous diving
beetle larvae (Dytiscus §R.) were chosen because they are known to be a highly efficient
predator of toad tadpoles, and have been found to kill and eat more toad tadpoles of all
sizes than other predators (Brodie and Formanowicz (1983). Diving beetle larvae have
also been shown to be primarily non-visual, relying instead upon tactile and chemical
cues before initiating a strike (Formanowicz 1987). In contrast, bluegill sunfish, Lepomis
macrochirus, are visually oriented predators that only occasionally are syntopic with toad
3
tadpoles. Toad tadpoles are distasteful to bluegill (Voris and Bacon 1966), although
naive predators may still engulf tadpoles. Furthermore, tadpoles are known to
behaviorally respond to these fish (Lawler 1989), perhaps due to their fragility.
4
Methods
Collections
American toad (Bufo americanus) tadpoles, Southern leopard frog (Rana
sphenocephala) tadpoles, and predaceous diving beetle larvae (Dytiscus fil2.) were dip
netted from an ephemeral pond in Coles County, Illinois, beginning early May 2000.
Tadpoles were separated by species and housed in aerated plastic aquaria (39x25xl0.5
cm) filled to a depth of 10 cm with tap water treated using a commercial aquarium water
conditioner, and were housed in environmental chambers kept at 22°C, with 14:10
photoperiod (which approximates the natural photoperiod during this time of the year);
water changes were performed every second day. Aquaria were wrapped in black plastic
to reduce startling tadpoles by human presence. Tadpoles were maintained on a diet of
commercial fish pellets fed ad libi:rnm; plastic plants were provided for refuge. Tadpoles
were kept in the laboratory for at least 48 hours prior to testing to allow for acclimation.
Beetle larvae were housed individually in I .SL plastic containers filled with conditioned
water to a depth of 4 cm and containing plastic plants, with their diet consisting of
backswimmers (notonectids). Bluegill sunfish (Lepomis macrochirus) were seined from
a permanent pond on the Eastern Illinois University campus (Coles Co., IL) and were
housed in aerated 40L aquaria. Fish were fed bloodworms and guppies; sufficient live
food was provided to ensure continuous access to prey. All predators were kept at room
temperature near windows under natural photoperiod conditions; chambers were cleaned
every 48 hours. Predators also were held at least 48 hours in the laboratory before use to
ensure no residual chemicals from prey captured in the wild remained in their digestive
5
track. Following conclusion of the study, tadpoles and predators were released at the site
of capture.
Experiments
The experimental protocol was consistent for all treatments tested. Test chambers
were of the same dimensions as holding chambers, and filled with conditioned tap water
to a depth of 4 cm. All experiments were conducted under fluorescent lighting at room
temperature. Forty percent ( 40%) of each chamber contained plant cover as one
contiguous clump, consisting of 10-12 stalks approximately 20 cm long of plastic Elodea
and Anacharis plants. In each experimental chamber, cover was randomly placed in
either the north or south direction of each arena. Toad tadpoles were netted from aquaria
and staged according to Gosner (1960). For experiments in Part I, tadpole stages 25-27
were used and designated as early stage. In Part II, tadpole stages 41-42 were used and
designated as late stage. Toad tadpoles were then introduced into test arenas and allowed
10 minutes for acclimation. Each experiment lasted 15 minutes, during which period the
number of tadpoles under cover was recorded immediately and then every 3 minutes
thereafter (repeated measure). Each of the experimental treatments was performed with
tadpole densities of 10 and 20. For each density 5 replicates were performed; no tadpoles
were used twice (excluding those in experiment #7-see below) so all data are
independent.
Additions of chemical and control stimuli were poured into the center oftest arenas
during each experimental replicate. To prepare extract, tadpoles of similar size and stage
were humanely killed then homogenized for approximately 15 seconds using a blender
6
and 150 ml of deionized water. From this mixture, 10 ml of liquid were collected and
slowly poured into the center of the arena. Fresh extract was used for each trial to
prevent potential degradation of chemical cues, and a total of 12 beetle larvae and 10
bluegill were used in both preparing extracts as well as in certain trials (see below). For
treatments requiring addition of an extract to the test chamber, a control was established
as follows. Five minutes into the acclimation period, 10 ml of conditioned tap water was
poured into the center of the arena, which controlled for the tactile disturbance effect of
adding liquids. At the conclusion of each experiment, arenas were thoroughly washed,
rinsed, and refilled with conditioned water.
Part I: Treatments
To determine how early stage toad tadpoles respond to conspecific alarm signals,
extracts of early and late stage tadpoles were prepared and added to test aquaria. To
determine the response of early stage tadpoles to a heterospecific, an extract ofR
sphenocephala tadpoles was prepared and added to test aquaria.
Tadpole response to diving beetle larvae was ascertained using two methods. First,
chemical cues were collected from a chamber holding a diving beetle larva and added to
the test chamber. Prior to use, beetle larvae were transferred and held individually for a
minimum of 5 hours in a clean chamber with 100 ml of conditioned water to prevent
introduction of conflicting chemical cues from prey. From a chamber containing an
individual beetle larva, 10 ml of water were collected and carefully added to the center of
the test arena. Secondly, to approximate natural conditions more closely, a single, live
beetle larva was placed in an enclosure in each test arena. Enclosures consisted of
styrofoam cups that were thoroughly washed before use and then discarded after each
7
trial. A small washed rock was placed in the bottom of each cup to prevent it from
floating. Each cup was pierced 60-70 times with a sterile wooden toothpick to allow for
water flow. Enclosures were introduced prior to the introduction of tadpoles for
acclimation, and randomly placed in one of 3 locations approximately 5 cm from the
edge of the arena opposite the side containing plants. A control for this treatment was
established using empty enclosures.
A series of experiments were designed to test tadpole response to bluegill using
chemical cues. Ten bluegill were kept for 5 hours in a clean chamber with 1 L of
conditioned tap water prior to use. From this holding chamber, 10 ml of water were
collected and introduced into the test chamber 5 minutes into the acclimation period.
Following use in the prior experiment, the response of each tadpole group to fish
chemical cue conditioning was determined. Each tadpole group was removed from the
test chamber and housed separately in l .5L containers for 48 hours. Every 8 hours, 50 ml
of water from fish aquaria were added. After conditioning the tadpoles to fish stimuli in
this manner, they were then retested for response to fish cues. Controls were performed
in which tadpoles were housed in identical groups and conditioned using conditioned tap
water additions every 8 hours.
Part II: Treatments
Treatments in Part II were performed using protocols similar to those in Part I.
Controls were established as in Part I using late stage tadpoles, and predators were
randomly assigned to replicates.
Late stage tadpoles were exposed to additions of tadpole extract prepared using early
stage tadpoles and late stage tadpoles. Similarly, tadpoles were exposed to predator cues
8
using bluegill cues collected from a lL aquarium housing 10 bluegill, and 10 ml water
samples taken from a larval diving beetle chamber.
Analysis
Data was converted to percentage of tadpoles under cover, then transformed using an
arcsine transformation (Sokal and Rohlf 1981 ). An analysis of variance was used to
compare means, with time as a repeated measure. Computer software (NWA ST ATP AK
1986) was used, with an alpha value set~ priori at 0.05. Figures are composed of
untransformed data, in the form of percentages of tadpoles seeking cover, with means +
one standard error shown.
9
Results
A three-way ANOVA, with main effects (tadpole density and treatment) was
calculated with one repeated measure (time). Results of these analyses revealed that
tadpole density did not contribute a significant amount of variance to the statistical model
in any experiment. Consequently, density was pooled in further analyses, resulting in a
two-way ANOVA with one repeated measure (time), and treatment effect.
Part I: Early Stage
The mean percentage of toad tadpoles seeking cover varied significantly (F=5.12,
P=0.0047) among treatments (Fig. 1). Significantly higher mean percentages of tadpoles
exposed to early and late stage extracts were found seeking cover (mean=47.1±8.2%,
mean=46.8±6.7%, respectively) as compared to tadpoles exposed to control treatments
(mean=21 .6±4.8%), however, the treatment means were statistically indistinguishable
from one another. Similarly, the mean percentages of tadpoles in the control
(mean=21.58±4.81 %) and R. sphenocephala treatments (mean=28.5±7.47%) were not
different. Over time, the percent of tadpoles in these treatments seeking cover increased
significantly (F=8.35, P<0.0001--Fig. 2) as compared to controls.
The mean percentage of toad tadpoles seeking cover varied among Dytiscus ~·
treatments. Tadpoles exposed to caged beetle larvae had a mean percent under cover
(53.95±2.52%) that was significantly higher than the mean percent (43.49±3.70%)
observed for tadpoles exposed to an empty cage (F=9.85, P=0.005--Fig 3). Trials with
beetle larvae cues alone also differed significantly (F=7.24, P=0.014), with the mean
percent of tadpoles in control water conditions under cover being 34.97±3.91 %, while the
10
mean percent of tadpoles seeking cover during exposure to beetle larval cues was
45.52±2.97% (Fig 4). The two control treatments utilized (empty predator cage and
water additions), did not differ significantly from one another (F=3.0l , P>0.05). Of these
two sources of beetle larvae cues, caged beetles had a significantly greater effect on cover
usage (F=13.52, P=0.0017) than did the addition of water from beetle larvae chambers.
Treatments involving bluegill stimuli varied in the percentages of tadpoles
seeking cover (Fig. 5). A significant difference (F=8.97, P=0.007) was found between
the mean percentages of tadpoles seeking cover in water controls (mean=2 l .58±4.81 % )
and fish cue treatments (mean=38.42±3.51%-- Fig 5). A significant difference (F=22.87,
P=0.0001) also resulted between tadpoles conditioned to bluegill ,cues
(mean=25.92±2.85%), and tadpoles treated with conditioned tap water
(mean=42.42±2.98%--Fig 6). Surprisingly, more tadpoles were found seeking cover in
the conditioning control than in the fish treatment. Of the two control treatments used,
tadpoles assigned to the conditioning regime were found under cover a significantly
higher percentage of time (F=l 7.05, P=0.0006) than tadpoles exposed to water additions
(mean = 50.3±3.0%, mean= 36.0±3.8%, respectively). Initially, 47.5% of tadpoles
responded to the addition of bluegill cues by seeking cover, whereas after conditioning,
tadpoles significantly reduced use of cover to 39.7%. The percentage of tadpoles seeking
cover increased significantly (F=4.15, P=0.0013) with time following a 48 hour
conditioning period to bluegill cues (Fig 7).
Part II: Late Stage
There were no significant differences among treatment means (Fig 8) involving
late stage toad tadpoles (F=l.11 , P=0.363). During exposure to extracts of both early and
11
late stage tadpoles, use of cover varied significantly (F=3.89, P=0.0025) over time (Fig
9). Analysis of controls for early versus late stage tadpoles revealed that a greater
percentage of late stage tadpoles sought cover (F= l2.47, P=0.002). The results of
pairwise analysis of variance in early and late stage tadpoles are shown (Table 1).
12
Fig I. Effects of water controls, early [Bufo(e)] stages (25-27) Bufo americanus, late
[Bufo(l)] stages (41-42) B. americanus tadpole extract, and early stages Rana
spenocephala (Rana) extract upon the percentage of early stage B . americanus tadpoles
seeking cover. Means + one standard error are shown.
13
... G.>
50 > 0 u ... 40
G.> mg c ::> 30 Ul G.> -0 a. 20 mg «S I-
~ 10
Control Bufo (e) Bufo (I) Rana
Treatment
Fig 2. The effects of time upon the percentage of early stages (25-27) Bufo americanus
tadpoles seeking cover in treatments: water controls (CTRL), extracts of early stages B.
americanus [B(e)], late stages (41-42) B. americanus [B(l)], and early stages Southern
leopard frog (Rana). Means + one standard error are shown.
14
70
... 60 Q)
> 0 () 50 ... Q)
"'C c:
::> 40 rn Q) -0 c. 30 "'C ca I-~ 0 20
--- ........... / /
I ........... / · . . ...........
I ...........
/ ..... · · ........... ....:.../
/ -.......
/ ,,,,.....,.. ..............
/
0 Min 3 Min 6 Min 9 Min 12 Min 15 Min
Time (minutes)
- CTRL B(e)
- - B(I) - Rana
Fig 3. Effects of controls with empty predator cages (C. cage) and enclosed predaceous
diving beetle larvae (Dytiscus) upon the percentage of early stage (25-27) Bufo
americanus tadpoles seeking cover. Means+ one standard error are shown.
15
60
... 50 Q)
> 0
(.) ... 40 Q) "'O c
::::> en 30 Q) -0 Q.
"'O 20 co ..... ~ 10
C. Cage Dytiscus
Treatment
Fig 4. The effects of water controls [Ctrl (w)] and diving beetle larvae (Dytiscus m.)
chemical cues [Dyts(w)] upon the percentage of early stages (25-27) Bufo americanus
tadpoles seeking cover. Means + one standard error are shown.
16
50
45
... 40 Cl)
> 0 35 0 ... Cl) 30 "C c
:::> 25 en Cl)
20 -0 Q. "C 15 ra I-~ 0
10
5
0 Ctrl (w) Oyts (w)
Treatment
Fig 5. Effects of water controls (Control) and Lepomis rnacrochirus chemical cues (Fish
I) upon the percentage of early stages (25-27) Bufo americanus tadpoles seeking cover.
Means + one standard error are shown.
17
50
45 ... 40 Cl)
> 0 35 0 ... Cl) 30 "CS c
:::> 25 ,,, Cl)
20 -0 c.
"CS 15 ca ~
';le. 10
5
0 Control Fish I
Treatment
Fig 6. Percentage of early stages (25-27) Bufo americanus tadpoles seeking cover
following 48 hour conditioning treatments to controls [Ctrl(c)], and Lepomis macrochirus
stimuli [Fish(c)]. Means+ one standard error are shown.
18
60
... 50 Cl) .r
> 0
CJ 40 ... Cl)
"'C c
::> 30 0 Cl) -0 c. 20 "'C ca ~
~ 0 10
Ctrl (c) Fish (c)
Treatment
r
Fig 7. Effects of time upon the percentage of late stages (41-42) Bufo americanus
tadpoles seeking cover following a 48 hour conditioning period to water controls
[Ctrl(c)], and Lej>omis macrochirus stimuli [Fish(c)]. Means+ one standard error are
shown.
19
.. CD > 0 0 .. CD ,, c
:::> fl) CD -0 Q. ,, ca I-~ 0
55
50
45
40
35
30
25
20
. . .
..
0 Min. 3 Min. 6 Min. 9 Min. 12 Min. 15 Min.
Time (minutes)
- Ctrt (c) . • . Fish (c)
Fig 8. Effects of various treatments upon the percentage of late stages (41-42) Bufo
americanus tadpoles seeking cover. Means+ one standard error are shown.
20
70 ... G> 60 > 0 0 ... 50 G>
"C c: ::> 40 ,,, G> - 30 0 Q.
"C ca 20 I-~ 0
10
0 Control Bufo (e) Bufo (I) Dyl Cues Fish
Treatment
Fig 9. Effects of time upon the percentage of late stages (41-42) Bufo americanus
tadpoles seeking cover during exposure to controls, extract of early stages (25-27) B.
americanus tadpoles, and extract of late stages B. americanus tadpoles.
21
60
.... G> > 0 50 0 .... G>
"C c
:::> 40 ti) G> -0 c.
"C 30 ca I-~ 0
20
. .
.· /9 ·--....._.....-...
. . . . . .
.
. .
0 Min. 3 Min. 6 Min. 9 Min. 12 Min. 15 Min.
Time (minutes) -Ctrl ... Bufo (e)
- - Bufo (I)
Eb
Early ext
Lb
Early ext
Eb
Late ext
Lb
Late ext
Eb
Fish cue
Lb
Fish cue
Eb
Dyt. cue
Lb
Oyt. cue
Table 1. Results of pairwise ANOV A's of early (E) stages (25-27) American toad, Bufo americanus, tadpoles (b) compared to late stages (41-42) toad tadpoles.
0 3 6 9 12 15 St.
Min Min. Min. Min. Min. Min. Mean Error F cal
34 45 35 51 56 53 46 5.22
0.721
37 32 49 39 45 34 41 2.61
40 35 52 53 48 55 47 4.37
22 39 41 40 42 47 39 3.16 2.256
31 31 35 37 48 43 34 3.51 0.636
24 34 42 37 32 37 34 3.98
38 31 34 36 39 30 35 2.98 0.633
42 34 29 42 37 38 37 3.21
P ca1
0.401
0.150
0.426
0.437
Discussion
Lefcort (1998) found that group size affected behavior in larvae ofBufo terrestris
under laboratory conditions. Similarly, Graves et al. (1993) with B. cognatus, and Hokit
and Blaustein (1995) found further that larger groups of Rana cascadae tadpoles moved
significantly more than did smaller groups in response to predator cues. Group size may
be an especially important mediator of behavior in bufonids because of their complex
aggregative behaviors (Beiswenger 1975, Hokit and Blaustein 1995). In contrast to these
studies, however, I found that group size did not have any significant effect upon cover
use by B. americanus tadpoles regardless of the developmental stage. The possibility
exists that the numbers of tadpoles I used were too small to have a significant impact
upon tadpole behavior, although Rodel and Linsenmair (1997), using Phrynornantis
rnicrops, found density effects beginning with a sample size of n= lO tadpoles; group
sizes of 10 and larger preferred open waters. My findings do however support the results
of Hews and Blaustein (1985) who found that alarm substances tended to break up
swarms of tadpoles; it seems possible that this effect may also inhibit swarm formation in
the presence of predatory cues. Waldman and Adler (1979), Hokit and Blaustein (1995),
and Watt et al (1997), demonstrated that tadpoles are more likely to form aggregations
when with siblings than when with non-siblings. Since I collected tadpoles randomly it
was likely that the coefficient of relatedness among them was low, therefore inhibiting
the formation of swarms. After collection, tadpoles were reared with conspecifics, which
has also been demonstrated to affect the intensity with which tadpoles respond to
predator cues (Bridges and Gutzke 1997).
23
Bufonids are known to recognize conspecifics through release of alarm signals
during predatory events (Hrbacek 1950, Adams and Claeson 1998), and to exhibit
avoidance tactics (Petranka 1989) upon detection of those signals. This early warning
system allows tadpoles to modify their behavior in advance of an encounter with a
predator and has been shown to significantly reduce predator success (Hews 1988). I
found that early stage toad tadpoles sought cover in signilicantly higher percentages
when exposed to extracts of both early and late stage conspecifics, which is congruent
with previous studies of other Bufonids (Petranka 1989, Petranka and Hayes 1998).
Although both cue types elicited an increase in the percentage of tadpoles seeking cover,
no apparent differences existed between extracts of early or late stages. This may be an
adaptive generalized antipredator response of early stage tadpoles to all stages of
conspecific alarm signals. These results suggest that no alterations to alarm signals occur
prior to metamorphosis, however, Belden et al. (2000) found an apparent alteration in
signals between larvae and juvenile Bufo boreas.
The rising percentages of tadpoles seeking cover over the 15 minute time period
in several of the experiments might be explained by slow diffusion rates of cues through
the test chamber. However, chemical cues are known to rapidly degrade, as
demonstrated by Petranka (1989), who showed that B. americanus tadpoles failed to
respond to alarm cues after 8 minutes in open water, thereby imposing a limit to
experimental duration.
Though they are often syntopic, American toad tadpoles exhibited no signilicant
response to R. sphenocephala larvae extract. This could be evolutionarily advantageous
because the two species are subject to different predators at least during the latter
24
development stages, a possibility attributable to the great size of R. sphenocephala larvae.
Rana sphenocephala larvae attain a much greater size than toad tadpoles, and owing to
the fact that many tadpole predators are size-dependant (e.g. diving beetle larvae-
Formanowicz 1986, Babbitt and Tanner 1998), ignoring alarm signals from
heterospecifics would be adaptive if no inherent threat existed.
Exposure to beetle larvae in enclosures as well as chemical cues from water taken
from larval beetle chambers resulted in increased use of available cover by toad tadpoles
as compared to controls. These treatments were designed to simulate two separate
situations in nature. By adding water, a situation existed similar to when a diving beetle
larva inhabits and then vacates an area; the cues slowly dissipate and degrade.
Interestingly, the number of tadpoles seeking cover with beetle larvae maintained in
enclosures was approximately 16% greater than when just chemical cues of larval beetles
were added. This may have been a response to the heightened risk of predation posed
when an actual predator was present. Anuran larvae are known to perceive and respond
differentially to varying levels of predation (Horat and Semlitsch 1994, Anholt et al.
1996, Laurila et al. 1997). It is possible that cues emanating from the caged larvae were
more potent than the chemical cues present in the water sample, or perhaps the tadpoles
may have sensed movement within the cage and responded with increased cover use.
Use of chemical cues from bluegill resulted in higher numbers of tadpoles seeking
cover as compared to the control. While bufonids have been shown to be unpalatable to
fish (Voris and Bacon 1966, Kruse and Stone 1984, Lefcort 1998), they may still respond
to fish cues if they operate under a generalized antipredator defense system. Lawler
(1989) demonstrated that refuge use increased by tadpoles in response to the presence of
25
fish; furthermore, Sernlitsch and Gavasso (1992) also showed that tadpoles decreased
swimming time in the presence of fish (but see Kats et al. 1988). This strategy may be
adaptive for small, vulnerable larvae that may be injured or killed by naive predators
even if they were to be expelled, uneaten (Brodie et al. 1978). It is interesting to note
that following a 48 hour conditioning period the percentage of tadpoles seeking cover
decreased; this effect has been previously shown (Semlitsch and Ryer 1992). Tadpoles
are also known to respond to novel stimuli (Manteifel 1995), and in a geographic range
where toad tadpoles rarely encounter fis~ this is also a plausible explanation. During the
conditioning period, tadpoles either became habituated to the novel stimulus, or, through
lack of predatory attempts, failed to respond to fish cues. Quite surprisingly and
inexplicably, tadpoles that were "conditioned" with conditioned tap water sought cover at
a rate similar to tadpoles assigned to the fish cue treatment. It is unlikely that this
resulted from the simple addition of liquid because tadpoles conditioned with fish water
actually decreased cover usage.
In Part II, late stage tadpoles showed no significant responses to predators or
conspecific alarm signals as compared to the controls. It was found, however, that in
treatments involving use of early and late stage B. americanus extracts, the percentages of
tadpoles seeking cover increased by 8.8% over time, however, this was statistically
insignificant. Late developmental stage toad tadpoles in control trials were found seeking
cover in significantly higher percentages when compared to control trials with early
developmental stage toad tadpoles. This increased use of cover by late stage tadpoles
could explain the lack of statistical significance in further analyses of early versus late
developmental stage tadpoles. Perhaps this increased use of cover is a generalized
26
response to predators by tadpoles at a highly vulnerable time in life. At Gosner (1960)
stage 42, tadpoles cease feeding (Beiswenger 1975), and production ofbufotoxin surges
(Brodie et al. 1978, Formanowicz and Brodie 1982), which may then become the primary
antipredator defense mechanism in this toad species.
Behavioral disparities could likely exist between early and late developmental
stage toad tadpoles due to a variety of factors such as differing suites of predators,
habitat, escape tactics, biochemistry, and chemical detection capability. Studies focusing
on these behavioral changes between juveniles and pre-metamorphic individuals are few,
but see (Bridges and Gutzke 1997, Belden et al. 2000). In this study, American toad
tadpoles responded adaptively to conspecific alarm signals and predator cues, at least in
early stages of development. However, this was not found to be the case with tadpoles
nearing metamorphosis. This study is one of the first to demonstrate that toad tadpoles
approaching metamorphic climax exhibit behaviors that are different than that of tadpoles
in earlier developmental stages.
27
Literature Cited
Adams M. and S. Claeson. 1998. Field response of tadpoles to conspecific and
heterospecific alarm. Ethology 104:955-961.
Anholt B., D. Skelly, and E. Werner. 1996. Factors modifying antipredator behavior in
larval toads. Herpetologica 52:301-313.
Arnold S. and R. Wassersug. 1978. Differential predation on metamorphic anurans by
Garter snake (Thamnophis): social behavior as a possible defense. Ecology 59:367-
373.
Babbitt K. and G. Tanner. 1998. Effects of cover and predator size on survival and
development of Rana utricularia tadpoles. Oecologia 114:258-262.
Babbitt K. and G. Tanner. 1997. Effects of cover and predator identity on predation
of Hyla squirella tadpoles. J. Herp. 31: 128-130.
Babbitt K. and F. Jordan. 1996. Predation on Bufo terrestris tadpoles: effects of cover
and predator identity. Copeia 1996:485-488.
Beiswenger R. 1975. Structure and function in aggregations of tadpoles of the
American toad, Bufo americanus. Herpetologica 31 :222-233.
Belden L., E. Wildy, A. Hate~ and A. Blaustein. 2000. Juvenile western toads, Bufo
boreas, avoid chemical cues of snakes fed juvenile, but not larval, conspecifics.
Anim. Behav. 59:871-875.
Bridges C. and W. Gutzke. 1997. Effects of environmental history, sibship, and age
on predator-avoidance responses of tadpoles. Can. J. Zool. 75:87-93.
28
Brodie E. Jr., D. Formanowicz Jr. , and E. Brodie III. 1978. The development of
noxiousness ofBufo americanus tadpoles to aquatic insect predators. Herpetologica
34:302-306.
Brodie E. Jr. and D. Formanowicz Jr. 1981. Larvae of the predaceous diving beetle
Dytiscus verticalis acquire an avoidance response to skin secretions of the newt
Notophthalrnus viridescens. Herpetologica 37: 172-176.
Brodie E. Jr. and D. Formanowicz Jr. 1983. Prey size preference of predators:
differential vulnerability of larval anurans. Herpetologica 39:67-75.
Brodie E. Jr., and D. Formanowicz Jr. 1987. Antipredator mechanisms of larval anurans:
protection of palatable individuals. Herpetologica 43:369-373.
Crump M. 1984. Ontogenetic changes in vulnerability to predation in tadpoles of
Hyla pseudopurna. Herpetologica 40:265-271.
Formanowicz D. Jr. and E. Brodie Jr. 1982. Relative palatabilities of members of a
larval amphibian community. Copeia 1982:91-97.
Formanowicz D. Jr. 1986. Anuran tadpole/aquatic insect predator-prey interactions:
tadpole size and predator capture success. Herpetologica 42:367-373.
Formanowicz D. Jr. 1987. Foraging tactic ofDytiscus verticalis larvae (Coleoptera:
Dytiscidae): prey detectio~ reactive distance and predator size. J. Kansas
Ent. Soc. 60:92-99.
Gosner K. 1960. A simplified table for staging anuran embryos and larvae with notes
on identification. Herpetologica 16: 183-190.
Graves B., C. Summers, and K. Olmstead. 1993. Sensory mediation of aggregation
among postmetarnorphic Bufo cognatus. J. Herp. 27:315-319.
29
Hews D. and A. BJaustein. 1985. An investigation of the alarm response in Bufo boreas
and Rana cascadae tadpoles. Behav. Neural Biol. 43:47-57.
Hews D. 1988. A.Jann response in larval western toads, Bufo boreas: release of larval
chemicals by a natural predator and its effect on predator capture efficiency. Anim.
Behav. 36:125-133.
Hokit D. and A. Blaustein. 1995. Predator avoidance and alarm-response behavior in
kin-discriminating tadpoles (Rana cascadae). Ethology 101 :280-290.
Horat P, and R. Semlitsch. 1994. Effects of predation risk and hunger on the
behaviour of two species of tadpoles. Behav. Ecol. Socio. 34:393-401 .
Hrbacek J. 1950. On the flight reaction of tadpoles of the common toad caused by
chemical substances. Experientia 6: I 00-101 .
Huey R. 1980. Sprint velocity of tadpoles (Bufo boreas) through metamorphosis.
Copeia 1980:537-540.
Kats L, J. Petranka, and A. Sih. 1988. Antipredator defenses and the persistence of
ampluoian larvae with fishes. Ecology 69: 1865-1870.
Kiesecker J. , D. Chivers, A. Marco, C. Quilchanos, M. Anderson, and A. Blaustein.
1999. Identification of a disturbance signal in larval red-legged frogs, Rana aurora.
Anim. Behav. 57:1295-1300.
Kruse K. and B. Stone. 1984. Large-mouth bass (Micropterus salmoides) learn to
avoid feeding on toad (Bufo) tadpoles. Anim. Behav. 32: 1035-1039.
Kupferberg S. 1997. The role oflarval diet in anuran metamorphosis. Amer. Zool.
37: 146-159.
30
Laurila A., J. Kujasalo, and E. Ranta. 1997. Different antipredator behaviour in two
anuran tadpoles: effects of predator diet. Bebav. Ecol. and Socio. 40:329-336.
Lawler S. 1989. Behavioral responses to predators and predation risk in four species of
larval anurans. Arum. Behav. 38:1039-1047.
Lefcort H. 1996. Adaptive, chemically mediated fright response in tadpoles of the
Southern frog, Rana utricularia. Copeia 1996:455-459.
Lefcort H. 1998. Chemically mediated fright response in Southern toad CBufo terrestris)
tadpoles. Copeia 1998:445-450.
Manteifel Y. 1995. Chemically-mediated avoidance of predators by Rana temporaria
tadpoles. J. Herp. 29:461-463.
NWA STATPAK. 1986. Version4.l. Northwest Analytical, Inc. Portland, Oregon.
Petranka J. , L. Kats, and A. Sib. 1987. Predator-prey interaction among fish and larval
amphibians: use of chemical cues to detect predatory fish. Anim. Behav. 35:420-425.
Petranka J. 1989. Response of toad tadpoles to conflicting chemical stimuli: predator
avoidance versus "optimal" foraging. Herpetologica 45: 283-292.
Petranka J. and L. Hayes. 1998. Chemically mediated avoidance of a predatory odonate
(Anax junius) by American toad (Bufo americanus) and wood frog~ ~lvatica)
tadpoles. Behav. Ecol. and Socio. 42:263-271.
Rodel M. and K Linsenmair. 1997. Predator-induced swarms in the tadpoles of an
African savanna frog, Phrynomantis microps. Ethology 103:902-914.
31
Semlitsch R. and S. Gavasso. 1992. Behavioral responses ofBufo bufo and Bufo
calamita tadpoles to chemical cues of vertebrate and invertebrate predators. Ethol.
Ecol. and Evol. 4:165-173.
Scmlitsch R. and H. Reyer. 1992. Modification of anti-predator behaviour in tadpoles
by environmental conditioning. Ecology 61 :353-360.
Skelly D. and E. Werner. 1990. Behavioral and life-rustorical responses oflarval
American toads to an odonate predator. Ecology 71 :2313-2322.
Skelly D. 1994. Activity level and the susceptibility of anuran larvae to predation.
Anim. Behav. 47:465-468.
Sokal R. and F. Rohlf. 1981. Biometry (2nd ed.) W. H. Freeman and Co., San Francisco.
859 pp.
Voris H. and J. Bacon. 1966. Differential predation on tadpoles. Copeia 1966:594-598.
Waldman B. and K. Adler. 1979. Toad tadpoles associate preferentially with siblings.
Nature 282:611-613.
Waldman B. 1982. Sibling association among schooling toad tadpoles: field evidence
and implications. Anim. Behav. 30:700-713.
Watt P., S. Nottingham, and S. Young. 1997. Toad tadpole aggregation behaviour:
evidence for a predator avoidance function. Anim. Behav. 54:865-872.
Werner E. and B. Anholt. 1993. Ecological consequences of the trade-off between
growth and mortality rates mediated by foraging activity. Arn. Nat. 142:242-272.
Wilson D. and H. Lefcort. 1993. The effect of predator diet on the alann response of
red-legged frog, Rana aurora, tadpoles. Anim. Behav. 46:1017-1019.
32