- Email: [email protected]
Behavioral and hormonal correlates of alternative reproductive strategies in a polygynous lizard: Tests of the relative plasticity and challenge hypotheses
HORMONES
AND
BEHAVIOR
26,
568-585 (1992)
Behavioral and Hormonal Correlates of Alternative Reproductive Strategies in a Polygynous Lizard: Tests of the Refative Plasticity and Challenge Hypotheses CHRISTOPHER W. THOMPSON’ Department of Zoology, Arizona
AND MICHAEL
C. MOORE
Sate Universiry, Ternpe, Arizona 8.52874501
Species with alternative reproductive tactics are good models to investigate the poorly understood question of whether individual variation within sexes results from the same physiological mechanisms that control variation between sexes. We have shown previously that adult male tree lizards, Vrosaurus ornatus, of different throat color morphs express different levels of aggression in the laboratory. Further field results support the suggestion that the two morphs practice alternative reproductive tactics because the two morphs express different levels of aggressive behavior under field conditions and exhibit dramatic and opposite responses to aggressive challenges. However, despite these behavioral differences, the two morpbs do not differ in levels of testosterone or corticosterone either in undisturbed situations or following aggressive challenge. These results are consistent with the relative plasticity hypothesis which proposes that organizational, rather than activation& actions of steroid hormones will be more important in morph differentiation when morphs are fixed in adult life, as they are in tree lizards. These results also support the hypothesis that steroid hormonal levels are insensitive to social modulation in males of species such as U. ornatus without paternal care. 0 1992 Academic Press, 1~.
A major goal of biologists is to explain the physiological bases of individual variation in morphology and behavior. Such variation may be continuous, such as human body weight, or discontinuous, such as human eye color. Attempts to identify physiological bases of such variability have been much more successful in studies of discontinuous than continuous variation (Feder, 1984; Wingfield and Ramenofsky, 1985). Behavioral and morphological variation is usually more discontinuous between sexes than within sexes. Therefore, much of our understanding of physiological mechanisms that control variable expression of behavior and morphology has 1 Current address: Burke Museum and Department Washington, Seattle, WA 98195. 568 0018-506X/92 $4.00 Copyright All ri&s
8 19!72 by Academic Press, Inc. of reproduction in any form reserved.
of Zoology DB-10, University of
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
569
come from studies of sexual dimorphism (Kelley, 1988). However, whether variation in morphology and behavior within sexes results from the same physiological mechanisms that control variation between sexes is poorly understood. This issue can be addressed by studying species that exhibit discontinuous within-sex behavioral and morphological variation that is as great as that between sexes. Such within-sex polymorphisms are usually associated with alternative reproductive strategies such as occur in male bluegill sunfish, Lepomis macrochirw (Gross, 1984) male Ruffs, Phi1omachu.spugnax, (van Rhijn, 1983, 1991), and male Pacific salmon, Orzcorhynchus sp. (Gross, 1984,1985). Species with multiple reproductive phenotypes probably arose evolutionarily by exaggeration of preexisting behavioral variation in their monophenotypic ancestors (West-Eberhard, 1986). Thus, it seems likely that the mechanisms producing within-sex variation in behavior between alternative phenotypes are similar to mechanisms producing within-sex behavioral variation in more typical species. This suggests further that an understanding of the mechanisms of withinsex behavioral differentiation in the former will help elucidate the mechanisms producing within-sex behavioral variation in the latter. The tree lizard, Urosaurus ornatus, is a small (3-6 g) iguanid lizard of the desert southwestern United States and northwestern Mexico that is polymorphic in the color of its extensible throat fan called a dewlap. Within single populations, dewlaps of adult males are completely blue, yellow or orange, or bicolor combinations of a central blue spot surrounded by an orange perimeter (hereafter orange-blue) or yellow perimeter (hereafter yellow-blue). The size of the central blue spot varies among individuals (Thompson and Moore, 1991a). The dewlap is displayed conspicuously during many social interactions suggesting that polymorphism in this social signal may be correlated with polymorphic social behavior. This has been partially confirmed in laboratory studies which have shown that, independent of age (1) the size of the central blue spot is a reliable predictor of dominance between pairs of size-matched males, regardless of dewlap perimeter color, and (2) orange-blue and yellowblue males respond more aggressively to other males than do orange or yellow males (Hover, 1982, 1985; Thompson and Moore, 1991b). The two main purposes of this study were (1) to confirm and extend in the field our previous laboratory results which suggested that free-living male U. ornatus of orange and orange-blue morphs (the two most common morphs in our study population) practice alternative reproductive tactics and (2) to test predictions of the relative plasticity hypothesis regarding the relationship between male behavioral and hormonal levels in species with alternative reproductive tactics (Moore and Thompson, 1990; Moore, 1991). Our data also address components of the challenge principle, a set of hypotheses regarding the relationship between male be-
570
THOMPSON
AND MOORE
havioral and hormonal levels (Wingfield, 1984a,b, 1985, 1988; Wingfield, Ball, Dufty, Hegner, and Ramenofsky, 1987; Wingfield, Hegner, Dufty, and Ball, 1990). Relative Plasticity Hypothesis Moore and Thompson (1990) and Moore (1991) propose an extension of the classical organization-activation model, called the relative plasticity hypothesis, to explain the hormonal control of differentiation of reproductive polymorphisms within sexes. The classical organization-activation model of sexual differentiation states that sex steroid hormones typically act sequentially during life (Arnold and Breedlove, 1985). First, they have permanent actions during early development which organize target tissues that regulate expression of sexually dimorphic characters. Second, they have temporary actions during adult. load which activate target tissues that were organized early in development. The relative plasticity hypothesis extends this model by first distinguishing between alternative reproductive morphs that are fixed or plastic. Ftied morphs are those in which individuals develop one phenotype at or before sexual maturity and do not change it during adulthood. PZastic morphs are those in which individuals may change phenotype, reversibly or irreversibly, after reaching sexual maturity. By analogy to organization and activation, the relative plasticity hypothesis then states that (1) differences in organizational influences of hormones will be more important in determining differences between fixed morphs and (2) differences in activational influences will be more important in determining differences between plastic morphs. Alternatives to the relative plasticity hypothesis are that (1) hormones play no role in the differentiation and expression of alternative male phenotypes and (2) hormones play a role that is not analogous to organization and activation. Both laboratory and field data indicate that dewlap color morph in U. ornatus is fixed (Hover, 1982, 1985; Thompson and Moore, unpublished data). Consistent with the prediction of the relative plasticity hypothesis that activational influences are not important in producing morph differences, previous studies have shown that seasonal patterns of circulating levels of testosterone and corticosterone secretion are virtually identical in adult males of the two morphs (Thompson and Moore, 1989; Moore and Thompson, in preparation). However, it is still possible that activational influences of hormones cause differences between the morphs by different short-term hormonal responses to social stimuli as appears to be the case in dominant and subordinate baboons (Sapolsky, 1987). We tested the prediction of this hypothesis that hormonal levels in the two morphs will differ after social challenge by measuring hormonal levels in free-living male tree lizards of both morphs after a staged aggressive encounter.
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
571
Challenge Principle Wingfield and colleagues (Wingfield, 1984a,b, 1985, 1988; Wingfield et aZ., 1987, 1990) have proposed the challenge principle which states that males increase their testosterone levels in response to (1) aggression directed toward them from conspecific males and (2) sexual solicitation by females. Recently, however, Wingfield et al. (1990) suggested that these responses occur only in species in which males contribute substantially to parental care (hereafter called care-giving males) and not in many poiygynous species in which males contribute little parental care (hereafter cafled non-care-giving males). Wingfield et al. (1990) hypothesized that adult males in non-paternal-care species are less responsive to behavioral cues that affect secretion of testosterone in paterna1-care species because selection has favored their testosterone levels to be “largely genetically determined” at “maximum levets . . . regardless of social situations.” We refer to this new hypothesis as the social insensitivity hypothesis. Moore (1987a) proposed a similar explanation for the lack of hormonal responsiveness to challenge in the non-paternal-care-giving lizard, Sce/oporus jarrovi. We beheve that the apparent Iack of sensitivity of care-giving males can also be explained by an alternative hypothesis that we call the social modulation hypothesis. According to this hypothesis non-care-giving males may be much more sensitive to social cues. Wingfield et at. (1990) rejected a similar hypothesis based on the lack of responsiveness in non-care-giving males. However, we believe this hypothesis is still valid since it is possible that non-care-giving males are so responsive that it is difficult to measure a true baseline level of testosterone. According to this hypothesis, baseline levels of testosterone in breeding non-care-giving males could only be revealed by housing them in complete isolation from acoustic and visual signals from other males. Our review of the examples cited by Wingfield ef ai. (1990) indicates that this has never been done for these species. Therefore, it is possible that all the reported baseline levels actually reflect sociatiy stimulated elevations of testosterone. U. ornutus is an ideal species for discriminating among these hypotheses because it is polygynous, exhibits no male parental care, and engages in intense male-male aggression throughout the breeding season, but does not have maximal testosterone levels during the entire breeding seasou (Thompson and Moore, 1989; Moore and Thompson, in preparation). We conducted our study at a time of year when testosterone levels are reduced to 50% of maximal levels. Therefore, at the time of year this study was conducted, the social insensitivity hypothesis predicts that, in response to maie challenges, orange-blue (and possibly orange) males will not increase their testosterone levels whereas the social modulation hypothesis predicts that testosterone levels should increase.
572
THOMPSON
MATERIALS
AND MOORE
AND METHODS
Study Area and Animals This study was conducted on 57 free-living adult (> 47 mm snout to vent length (SVL)) male U. ornatus from 14 May to 9 June 1988 in Maricopa Co., Arizona within a 5-km radius north and west of the intersection of Arizona Highway 87 and Sycamore Creek-Sugarloaf Mountain Road (33”42’North lll”32’West) at an elevation of 600-700 m (see Thompson and Moore, 1991a for description). As discussed above, this time of year is the middle of the breeding season when male-male aggressive behavior and female sexual solicitation behavior is frequent and intense, but male testosterone levels are well below their maximum levels exhibited in April (Thompson and Moore, 1989; Moore and Thompson, in preparation). Assignment
of Males to Treatment
Groups
Previously uncaptured males of both morphs were subjected to two treatments described below as each male was encountered in the field. In the first treatment, behaviors were recorded during undisturbed observation periods. In the second treatment, behaviors were recorded during and following staged territorial encounters. Males were captured at the end of both types of observations, morph was assessed, and blood samples were collected as described below. Due to practical considerations, only one type of observation was performed on any given day. Types of observations were alternated on successive days. Measurement of Behaviors of Undisturbed Males All observations were made during the daily maximum of activity from 0650 to 1300 hr. After spotting each adult male at a distance greater than 10 m, C.W.T. approached each male (hereafter resident male) to within 4 to 8 m and remained as motionless as possible for 15 min to minimize possible effects of observer presence on each lizard’s behavior. For the subsequent 30 min, the distance moved (estimated to within 0.5 m) and the number of pushup, fullshow, and shudder displays exhibited by each resident male was recorded. Pushups and fullshows are stereotypic territorial displays and shudders are stereotypic courtship displays (Carpenter and Grubitz, 1961; Moore, unpublished data). The number of 1-min intervals during which these males directed behaviors toward other identifiable male and female U. ornatus was also recorded. Resident males were captured within 6 min of the end of each 30-min observation period. Within 2 min after capture, each male was assessed for morph type, had a sample of blood removed as described below, and was toe-clipped, marked with a spot of paint on the dorsal surface (to prevent using them again), and released.
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
573
TABLE 1 Behavioral Intensity Scale Used in the Painted Dewlap Experiment to Rank Aggressive Responses of Males to One Another during Pairwise Contests for a Limiting Resource Score 0
1 2 3 4 5 6
Description No response Pushups without approaching opponent Fullshows without approaching opponent Fullshows while approaching opponent Faceoff (fullshow displayed in close proximity and parallel to opponentj Charge Bite
Measurement of Behaviors during Staged Encounters
All observations were made during the daily activity maximum from 0625 to 1225 hr. Staged encounters were performed using tethered males as described by Ruby (1976, 1978) and Moore (1987b). To minimize the effects of handling stress on their behavior (Moore, Thompson, and Marler, 1991), intruder males were captured away from the main study area between 0610 and 0930 hr and used only on that day for no more than three tethered presentations. By visual comparison, an attempt was made to match the SVL of intruders within 1 mm of the SVL of the resident. If, after capture, the resident male was found to differ in SVL from the intruder by more than 1.5 mm, the data from these staged encounters and subsequent postencounter observations were omitted from our analyses. Dewlap colors of intruders were not matched with those of resident males. The percentage blue (Defined below) of dewlaps of male intruders presented to orange males was 21.9 -+ 3.8% (means +- 1 SE, limits O45%) and did not differ significantly from the percentage blue (22.3 -I3.5%, limits 5-50%) presented to orange-blue males (two-sample t test using separate variances, t = 0.077, df = 24.7, P = 0.940). Intruders were tethered with about 50 cm of waxed dental floss to the end of a 2.0m pole and lowered onto the territory of the resident about 1 m away from the resident. Intruders were restrained from running off the resident’s territory and from retreating into crevices, but otherwise had free movement. The presentation was continued until (1) the resident gave a maximum response on the behavioral intensity scale (see Table l), (2) the resident ceased to direct behaviors toward the intruder and moved away from the intruder (e.g.., foraging), or (3) 15 min had passed, whichever came first. The maximum intensity of aggressive behavior displayed by
574
THOMPSON
AND MOORE
the resident was recorded on a ranked scale (see Table 1) modified Moore (1987b).
from
Measurement of Behaviors Following Staged Encounters
For 30 min following each staged encounter, the frequency and duration of territorial and courtship behaviors exhibited by resident males was recorded as described above for undisturbed observations. Resident males then were captured and treated as described above. Assessment of Morph Type
As in previous studies (Hover, 1982, 1985; Thompson and Moore, 1991a,b), the extent of blue color in each male’s unextended dewlap was estimated as a percentage of the total surface area of the unextended dewlap within 1 to 2 min after capture by noosing. Males having a distinct central blue spot (always greater than 10% blue) in their dewlaps were assigned to the orange-blue morph and males with no blue or only scattered blue scales (always 10% or less blue) in their dewlaps were assigned to the orange morph. Blood Samples and Radioimmunoassays
Blood samples were collected from the orbital sinus as described by Moore (1986). Blood samples were returned to the laboratory and centrifuged within 8 hr. The resulting plasma samples were stored at -20°C until assayed. Testosterone, corticosterone, progesterone, and estradiol were separated from each other and from plasma lipids by chromatography and then measured by radioimmunoassay as described by Moore (1986). Body Size of Tested Males
A two-way ANOVA indicated that SVL did not differ significantly (1) between orange and orange-blue males used for undisturbed observations, (2) between orange and orange-blue males presented with intruder males, or (3) within or between orange and orange-blue males in comparisons between undisturbed and postencounter observations (P > 0.9). Therefore, differences in behavioral frequencies and intensities reported below cannot be attributed to differences in mean SVL among treatment groups of males. Statistics
Behavioral frequency data were square-root transformed prior to analysis to normalize their distribution (see Moore, 1987a). A value of 0.5 was added to all behavioral frequency data prior to transformation to eliminate data values of zero. No significant skewness or kurtosis of the transformed data was detected by calculation of g, and g,, respectively, of the transformed data. Kolmogorov-Smirnov and Lilliefors tests also
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
575
70-r 13ORANGEWLES,n=lZ IORANGE-BLUEtiS,n=16
rn ii 60-50 -3 40-34F2 30-8 20-t.i a IO-OC
1 BEHAVIOR
3
2
INTENSITY
4
5
- AGGRESSTVE
6 RESPONSE
FIG. 1. Aggressive response of free-living resident orange and orange-blue males to tethered intruder males. See Table 1 for list of behaviors ranked by aggressive intensity.
indicated that the transformed data did not deviate significantly from normality (Sokal and Rohlf, 1981; Wilkinson, 1990). Hormonal data were log transformed prior to analysis to reduce heteroscedasticity of variances between treatment groups (undisturbed versus postencounter hormonal levels). Transformed hormonal data were not heteroscedastic using the statistically sensitive Smith’s test (Van Valen, 1978). Transformed hormonal and behavioral frequency data were analyzed by two-way analysis of variance (ANOVA) with the observation type (undisturbed or postencounter) as treatments and dewlap color (orange or orange-blue) as factors. Nonsignificant between-factor interaction terms are not reported. Post hoc multiple pairwise comparisons were made using Tukey’s Honestly Significant Difference method (Neter, Wasserman, and Kutner, 1990; Sokal and Rohlf, 1981). The level of significance is defined as P < 0.05. All statistical analyses were performed using SYSTAT (Wilkinson, 1990). RESULTS Observations during Staged Encounters
Orange-blue males exhibit a higher median intensity of aggression (see Table 1 for list of behaviors ranked by aggressive intensity) in response to intruder males than do orange males (Mann-Whitney U test, V = 44.5, df = 1, P = 0.012, Fig. 1) and in particular were much more likely to bite intruders than orange males. Observations before and after Staged Encounters Distance moved. Distance moved differs significantly between orange and orange-blue males (F = 20.206, df = 1, Y < 0.001, Figs. 2a and 2b), but not between undisturbed versus postencounter observations (F = 0.048, df = 1, P = 0.828, compare Figs. 2a and 2b). Post hoc com-
576
THOMPSON
8 70- UORANGE,n=Q 8 60-
IORANGE-BLUFi,n=20
& w 60E 40-
*** I;I
AND MOORE
a
20
15
1
10
SEUDDERS FULISIiOWB PU; WP8
-5; 703 60-
2 E g
DISTANCE mvED
WV
DORANGE,n=12 IORANGE-BLUE,n=l6
***
***
b
-20
%
07
h. SHUDDERSFDUSEOWSPUSRUPS
DISlWCEi MOVED
,O
FIG. 2. Frequency (means + 1 SE) of aggressive displays by free-living resident orange and orange-blue males during 30-min periods (a) of undisturbed observation, and (b) following termination of staged encounters with tethered intruder males.***P < 0.001.
parisons indicate that orange-blue males move significantly greater distances than orange males both before (P < 0.001) and after (P < 0.001) staged encounters. Frequency of pushups. Pushup frequency differs significantly between orange and orange-blue males (F = 19.804, df = 1, P < 0.001, Figs. 2a and 2b) and undisturbed versus postencounter observations (F = 8.014, df = 1, P = 0.007, compare Figs. 2a and 2b). Post hoc comparisons indicate that pushup frequency does not differ between orange and orange-blue males during undisturbed observations (P = 0.484). However, following staged encounters, orange males decrease their pushup frequency (P = 0.010) whereas orange-blue males do not (P = 0.707) relative to their respective pushup frequencies during undisturbed observations. As a result, orange-blue males exhibit a greater pushup frequency than orange males following staged encounters (P < 0.001). This differential response to staged encounters by orange and orange-blue males is reflected by a significant between-factor interaction (F = 5.629, df = 1, P = 0.021).
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
577
Frequency offullshows. Fullshow frequency differs significantly between orange and orange-blue males (F = 13.811, df = 1, P < 0.001, Figs. 2a and 2b), but not between undisturbed versus postencounter observations (F = 0.579, df = 1, P = 0.450, compare Figs. 2a and Zb). Post hoc comparisons confirm the overall result that fullshow frequency does not differ between orange and orange-blue males during undisturbed observations (P = 0.790). However, post hoc comparisons reveal that following staged encounters, orange males decrease (P = 0.044) whereas orange-blue males increase (P < 0.001) their frequency of fullshows relative to their respective fullshow frequencies during undisturbed observations. As a result, orange-blue males exhibit a greater fullshow frequency than orange males following staged encounters (P < 0.001). This differential response to staged encounters by orange and orangeblue males is reflected by a significant between-factor interaction (F = 25.62, cff = 1, P < 0.001). Frequency of shudders. Shudder frequency tends to be greater in orange-blue than orange males, but this difference was not significant (F = 3.171, df = 1, P = 0.081, Figs. 2a and 2b). Shudder frequency also did not differ between undisturbed versus postencounter observations (F = 0,011, df = 1, P = 0.915, compare Figs. 2a and 2b). Male-male and male-female interactions. During undisturbed observations, orange and orange-blue males did not differ in the proportion of time spent interacting with other males (Mann-Whitney U test, U = 70.5, df = 1, P = 0.243, Fig. 3a). In addition, orange-blue males tended to spend a greater proportion of time interacting with females than did orange males (Mann-Whitney U test, U = 50.5, df = 1, P = 0.053, Fig. 3a), although the difference is not quite significant. Following staged encounters, orange-blue males spent a greater proportion of time interacting with males (Mann-Whitney U test, U = 2.00, df = 1, P < 0.001, Fig. 3b) and a smaller proportion of time interacting with females (MannWhitney U test, U = 38.00, df = 1, P = 0.005, Fig. 3b) than did orange males. Circulating Blood Plasma Adrenal and Sex Steroid Levels Despite marked differences in behavior, orange and orange-blue males did not exhibit significantly different testosterone or corticosterone levels during undisturbed observations or following staged encounters (testosterone, F = 2.476, df = 1, P = 0.122, Fig. 4a; corticosterone, F = 2.750, df = 1, P = 0.103, Fig. 4b). In addition, neither orange nor orange-blue males exhibited significantly different testosterone or corticosterone levels following staged encounters compared to their respective levels during undisturbed observations (testosterone, F = 0.138, df = 1, P = 0.712, Fig. 4a; corticosterone, F = 2.346, df = 1, P = 0.132, Fig. 4b). We also found no significant differences in mean progesterone levels
578
THOMPSON
AND MOORE
OORANGE,n=Q IORANGE-ELI&n=20
a
OORANGE,n=12 IORANGE-BLUFa,n=16
b
40-
8 35-
8 E
30-
6 a
25-
gE
***
**
20.
0-r
HbLE-MALE MALE-ENCOUNTERTYPE FIG. 3. Percentage (means -t 1 SE) of time that free-living orange and orange-blue resident malts interacted with other male and female Urosaurus orna~.~ during 3O-min pexiods (a) of undisturbed observation, and (b) following termination of staged encounters with tethered intruder males.**P < 0.01 and ***P < 0.001.
between orange and orange-blue males (1) during undisturbed observations or following staged encounters (orange, 0.63 2 0.09 rig/ml; orange-blue, 1.98 2 0.52 rig/ml, F = 0.832, df = 1, P = 0.366), or (2) following staged encounters compared to their respective levels during undisturbed observations (F = 0.840, df = 1, P = 0.364). Estradiol levels were undetectable in all samples. Detectability limits for each sample for estradiol varied from 13 to 18 pg/ml depending on the the volume of plasma used (25-80 ~1) and the percentage recovery (46-81%).
DISCUSSION Our results confirm and extend our previous laboratory results which suggested that orange and orange-blue males pursue alternative reproductive strategies. Our results also support the relative plasticity hypothesis (Moore and Thompson, 1990; Moore, 1991) and the social insensitivity hypothesis (Wingfield et al., 1990), but are inconsistent with the social modulation hypothesis.
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
IXORANGE I ORANGE-BLUE
579
a
T
OBSERVA!l'ION
3
0
g
6
I! s fi
4
3 *
ENCOUNTER
1 ORANGE-BLUE
2 0
UND-
AFPER
OBSERVATION ENCOUNTER FIG. 4. Circulating blood plasma levels of (a)testosterone and (b) corticosterone during undisturbed observations and following staged encounters with tethered intruder males.
Alternative Reproductive Tactics
Our laboratory results suggested that orange males were less aggressive than orange-blue males (Thompson and Moore, 1991b; see also Hover, 1982, 1985). The results of our field studies reported here confirm and amplify these findings and suggest more precisely how the strategies of the two morphs differ. The data indicate that (1) during undisturbed observations, orange-blue males spend more time patrolling their territories and interacting with females than do orange males, (2) during staged encounters orange-blue males respond more aggressively than orange males, and (3) following staged encounters with intruder males, orangeblue males increase their rate of fullshows and the proportion of time spent interacting with males whereas orange males decrease their rate of fullshows and pushups. In addition, orange-blue males also decrease the proportion of time spent interacting with females following staged encounters, probably as a consequence of increasing the proportion of time spent interacting with males. The differential response of orange and orange-blue males to the same stimulus (staged encounters) suggests that orange-blue males are increasing their conspicuousness after an encounter whereas orange males are decreasing their conspicuousness. Additional
580
THOMPSON
AND
MOORE
field data (Thompson, Oliver, and Moore, unpublished data) indicate that orange-blue males reside for much longer periods of time on our study areas than do orange males. The majority of orange males spend only a day or two on the study area. Taken together, these differences in behavior suggest that orange-blue males pursue a strategy of aggressive defense of a territory containing females and perhaps resources important to females. The facts that orange males are less aggressive, move nomadically through the habitat, and behave cryptically following encounters are consistent with the interpretation that these males pursue either a satellite strategy (a permanent but submissive association with a territorial male that results in occasional matings or inheritance of the territory on the death of the owner), a sneaker strategy (wandering in search of unguarded females with which to mate), or both. The nomadic nature of some orange males suggests the latter, whereas the presence of some aggressiveness in orange males is more consistent with the former. Further studies will be necessary to clarify these possibilities. Implications for the Relative Plasticity Hypothesis The relative plasticity hypothesis predicts that fixed morphs will not exhibit differences in their adult hormonal levels. As discussed above, both laboratory and field data indicate that orange and orange-blue males represent fixed alternative morphs. Neither testosterone nor corticosterone levels differed between orange and orange-blue males during undisturbed observations or following staged encounters with other males (Figs. 4a and 4b) or throughout the course of the breeding season (Thompson and Moore, 1989; Moore and Thompson, in preparation). In addition, neither orange nor orange-blue males exhibited a significant change in testosterone or corticosterone levels following staged encounters with other males relative to their levels prior to these encounters (i.e., during undisturbed observations) (Figs. 4a and 4b). These results are consistent with the hypothesis that differences in behavior between orange and orange-blue males are not due to activational differences in testosterone or corticosterone levels and, therefore, support the relative plasticity hypothesis. This hypothesis is further supported by other studies showing that adult males do not change dewlap color when they are castrated or when they receive large testosterone implants (Moore and Thompson, in preparation). As a consequence, these results suggest that differences in behavior between orange and orange-blue males arise due to different organizational actions of testosterone or corticosterone on neural substrates that influence adult male sexual and aggressive behavior associated with reproduction. This suggestion is supported by recent findings that males castrated at hatching develop only orange dewiaps as adults whereas males given testosterone at hatching develop largely into orange-blue males (Hews, Moore, and Knapp, unpublished data).
RELATIVE
PLASTICITY
AND CHALLENGE
Implications for the Social Insensitivity Hypotheses
HYPOTHESES
581
and Social Modulation
Our results support the prediction of the social insensitivity hypothesis that orange-blue (and possibly orange) males should not have higher testosterone levels following staged encounters than during undisturbed observations. Even though our studies were conducted at a time of year when testosterone levels are below the annual maximum, orange and orange-blue males exhibit no increase in testosterone (or corticosteronej levels in response to staged encounters, despite showing dramatic and opposite changes in behavior. In turn, our results do not support the social modulation hypothesis which makes the converse prediction. Two methodological concerns that could account for our failure to detect any change in hormonal levels by orange and orange-blue males in response to staged encounters can be discounted. First, is it possible that 45 min (Iti-min male-male encounter period followed by 30-min observation period prior to capture and bleeding) is insufficient time for testosterone levels to change? This seems unlikely because changes in testosterone levels have been observed within 45 min in many taxonomically diverse vertebrates, including &her iguanid lizards, in response to malemale interactions (e.g., Balthazart, 1983; Wingfield, 1985; Sapolsky, 1986; Wingfield et al., 1987; Greenberg and Crews, 1990) as well as to sexual arousal by females (review by Wingfield and Marler, 1988) and artitlcial and natural stressors (e.g., Wingfield, Smith and Farner, 1982; Moore et al., 1991 and references cited therein). Second, is it possible that staged encounters were not of sufficient intensity or duration to elicit a hormonal response? Wingfield et al. (1987) state that in Song sparrows, Melospiza melodia, challenges must last at least 10 min to cause an increase in testosterone levels. In this study, staged encounters were continued for 15 min or until the resident male exhibited a maximal response, i.e., charged and bit the intruder. These encounters were as intense or more intense than any naturally occurring encounters that we have observed. We cannot completely discount the additional possibilities that the hormonal response was too short in duration or too low in magnitude to be detected by our methods (cf. Moore, 1987b, 1988), but methods similar to ours have been sufficient to detect hormonal responses in a wide variety of vertebrates (Wingfield et al., 1990). Data from Moore (1987a) and Brown (1990) also support the social insensitivity hypothesis in another polygynous iguanid lizard, Sceloporus jarrovi, that exhibits no parental care and engages in intense competition over territories. Despite a marked increase in aggressive reproductive behaviors in response to male challenges, territorial males of this species do not increase testosterone following a challenge either during the breeding season when testosterone levels are high (Moore, 1987a), or during
582
THOMPSON
AND MOORE
the nonbreeding season when males are territorial and have testosterone levels that are much lower (Brown, 1990). However, these results do not constitute as strong a test of the social insensitivity hypothesis as the data reported here because lower intensity male-male competition in S. jurrovi in the nonbreeding season could account for the lack of response at that time (Brown, 1990). In U. ornatus, levels of competition were still high when we conducted our study. We are aware of only two studies of species in which males are polygynous, non-care-giving, and intensely territorial that are inconsistent with the social insensitivity hypothesis. First, Cardwell and Liley (1991) found that androgen levels in male stoplight parrotfish, Sparisoma viride, are very responsive to social stimuli. For example, males increase their levels of testosterone and 11-ketotestosterone, the major androgen in teleost fish, six- to seven-fold in the early stages of territory establishment compared to nonterritorial males. Second; in pair-wise laboratory encounters between male green anoles, Anolis carolinensis, Greenberg and Crews (1990) found that the males that dominated encounters increased their testosterone levels about five-fold compared to subordinate and control males. These data suggest that the social insensitivity hypothesis is either not universally applicable or that it needs to be modified to incorporate other factors that will account for cases like these. Further studies are necessary to examine these possibilities. U. ornatus differ from several of the non-paternal-care-giving bird species that have been studied in that adult males do not have maximal testosterone levels throughout the entire breeding season. Nevertheless, testosterone appears to be necessary for the expression of aggressive reproductive behavior in male U. ornatus because free-living castrated males cannot maintain their territories (Knapp and Moore, unpublished data). Thus, we believe that a minimum level of testosterone is necessary to activate neural substrates so that males will exhibit the full repertoire of aggressive reproductive behaviors in response to appropriate social stimuli such as conspecific males and females (Arnold and Breedlove, 1985). However, short-term changes in the frequency and intensity of aggressive reproductive behaviors following male-male interactions probably are not modulated by short-term increases in testosterone levels above this threshold. Rather these behavioral changes are probably modulated by mechanisms entirely within the central nervous system or by other rapidly responding systems such as the adrenal catecholamines (Moore, 1987a; Moore and Marler, 1987). Why do paternal care-giving species use sex steroids to modulate aggressive behavior whereas many species without paternal care apparently do not? In care-giving males, time and energy demands of parental care require that males reduce territorial aggression during periods of parental care (Silverin, 1980; Moore, 1984; Wingfield, 1985; Wingfield and Ra-
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
583
menofsky, 1985; Wingfield and Moore, 1987, Wingfield et al., 1987). Wingfield et al. (1990) propose that increases in testosterone directly cause increases in aggressive behavior. In contrast, we propose, similar to a previous suggestion by Moore (1987a), that care-giving males possess an unknown physiological mechanism that inhibits full expression of aggressive reproductive behavior during periods of parental care and that this mechanism is disinhibited by an increase in testosterone levels that occurs in response to intense challenges. In other words, increased testosterone levels do not cause the rapid increase in aggressive behavior observed during and immediately following an initial challenge, but rather result from an initial challenge and act to increase the rapidity and intensity of future responses to challenges by reducing the effectiveness of the inhibiting system (Wingtield et al., 1987). In contrast, in non-care-giving males such an inhibiting mechanism would serve no function and, thus, would not be adaptive. A good test of this hypothesis would be to compare the effect on testosterone levels of challenges to territorial males in a polygynous species, such as red-winged blackbirds, Agelaius phoenicem, in which male parental care is expressed in some populations but not others (Beletsky and Orians, 1990). Similarly, it is also possible that orange morph male U. ornatus possess an inhibiting mechanism similar to caregiving male birds that reduces their response to encounters. In this case, however, the effectiveness of the mechanism is not modulated by testosterone because the reduction in aggressiveness is a permanent characteristic of this morph. Further comparisons are needed to clarify this hypothesis. ACKNOWLEDGMENTS This research was supported by Presidential Young Investigator Award DCB-8451641 from the National Science Foundation to M. C. Moore and by an Arizona State University Graduate College Research Assistantship to C. W. Thompson. We thank D. Crews, D. K. Hews, R. Knapp, I. T. Moore, and S. Rohwer for helpful criticisms of the manuscript, and R. Knapp for helping us critically review the literature upon which Wmgfield et al. (1590) based their social insensitivity model.
REFERENCES Arnold, A. P., and Breedlove, S. M. (1985). Organizational and activational effects of sex steroids on brain and behavior: A reanalysis. Horm. Behav. 19, 469-498. Balthazart, J. (1983). Hormonal correlates of behavior. In D. S. Farner, J. R. King, and K. C. Parkes (Eds.), Avian Biology, Vol. 7, pp. 221-365. Academic Press, New York. Beletsky, L. D., and Orians, G. H. (1990). Male parental care in a population of redwinged blackbirds, 1983-1988. Can. J. Zool. 68, 606-609. Brown, M. R. (1990). Hormonal and Environmental Znteractions Regulating Post-Challenge Aggressive Responses of Territorial Male Lizards. Unpublished M. S. thesis, Arizona State University, Tempe, AZ. Cardwell, J. R., and Liley, N. R. (1991). Androgen control of social status in males of a
584
THOMPSON
AND MOORE
wild population of stoplight parrotfish, Sparisoma viride (Scaridae). Horm. Behav. 25, 1-18. Carpenter, C. C., and Grubitz G., III. (1961). Time-motion study of a lizard. Ecology 42, 199-200. Feder, H. H. (1984). Hormones and sexual behavior. Annu. Rev. Psychol. 34, 165-200. Greenberg, N., and Crews, D. (1990). Endocrine and behavioral responses to aggression and social dominance in the green anole lizard, Anolis carolinensis, Gen. Comp. Endocrinol. 77, 246-255. Gross, M. (1984). Sunfish, salmon, and the evolution of alternative reproductive strategies in fishes. In G. W. Potts and R. J. Wooton (Eds.), Fish Reproduction: Strategies and Tactics, pp. 55-75. Academic Press, New York. Gross, M. R. (1985). Disruptive selection for alternative life histories in salmon. Nature (London) 3l3, 47-48. Hover, E. L. (1982). Behavioral Correlates of a Throat Color Polymorphism in a Lizard, Urosaurus ornatus. Unpublished Ph.D. Dissertation, University of Michigan, Ann Arbor, MI. Hover, E. L. (1985). Differences in aggressive behavior between two throat color morphs in a lizard, Urosaurus ornatus. Copeia 1985,933-940. Kelley, D. B. (1988). Sexually dimorphic behaviors. Annu. Rev. Neurosci. 11, 225-251. Moore, M. C. (1984). Changes in territorial defense produced by changes in circulating levels of testosterone: A possible hormonal basis for mate-guarding behavior in whitecrowned sparrows. Behaviour 88, 215-226. Moore, M. C. (1986). Elevated testosterone levels during nonbreeding season territoriality in a fall-breeding lizard, Sceloporus jarrovi. Z. Comp. Physiol. 158A, 159-163. Moore, M. C. (1987a). Circulating steroid hormones during rapid aggressive responses of territorial male mountain spiny lizards, Sceloporus jarrovi. Horm. Behav. 21, 511-521. Moore, M. C. (1987b). Castration affects territorial and sexual behaviour of free-living male lizards, Sceloporus jarrovi. Anim. Behav. 35, 1193-1199. Moore, M. C. (1988). Testosterone control of territorial behavior: tonic-release implants fully restore seasonal and short-term aggressive responses in free-living castrated lizards. Gen. Comp. Endocrinol. 70, 450-459. Moore, M. C. (1991). Application of the organization-activation theory to alternative male reproductive strategies: A review. Horm. Behav. 25, 154-179. Moore, M. C., and Marler, C. A. (1987). Effects of testosterone manipulations on nonbreeding season territorial aggression in free-living male lizards,Sceloporu.s jarrovi. Gen. Comp. Endocrinol. 65, 225-232. Moore, M. C., and Thompson, C. W. (1990). Field endocrinology of reptiles: Hormonal control of alternative reproductive tactics. In A. Epple, C. G. Scanes, and M. H. Stetson (Eds.), Progress in Comparative Endocrinology, pp. 685-690. A. R. Liss, New York. Moore, M. C., Thompson, C. W., and Marler, C. A. (1991). Reciprocal changes in corticosterone and testosterone levels following acute and chronic handling stress in tree lizards, Urosaurus ornatus. Gen. Comp. Endocrinol. 81, 217-226. Neter, J., Wasserman, W., and Kutner, M. H. (1990). Applied Linear Statistical Models, 3rd ed. R. D. Irwin, Boston, MA. Ruby, D. E. (1976). The Behavioral Ecology of the Viviparous Lizard Sceloporus jarrovi. Unpublished Ph.D. Dissertation, University of Michigan, Ann Arbor, MI. Ruby, D. E. (1978). Seasonal changes in territorial behavior of the iguanid lizard Sceloporus jarrovi. Copeia 1978, 430-438. Sapolsky, R. (1986). Stress-induced elevation of testosterone concentrations in high ranking baboons: Role of catecholamines. Endocrinology ll8, 1630-1635.
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
585
Sapolsky, R. M. (1987). Stress, social status, and reproductive physiology in free-living baboons, ln D. Crews (Ed.), Psychobiology of Reproductive Behavior, an Evolutionary Perspective, pp. 291-322. Prentice-Hall, Englewood Cliffs, NJ. Silverin, B. (1980). Effect of long acting testosterone treatment on free-living pied flycatcbers, Ficedula
hypo1euca.
Anim.
Behav.
28, 906-912.
Sokal, R. R., and Rohlf, F. J. (1981). Biometry, 2nd ed. Freeman, San Francisco, CA. Thompson, C. W., and Moore, M. C. (1989). Interactions among hormones, dominance and color in tree lizards. Am. Zool. 29, 447A. Thompson, C. W., and Moore, M. C. (1991a). Syntopic occurrence of multiple dewlap color morphs in male tree lizards, Urosaurus ornatw. Copeia 1991,493-503. Thompson, C. W., and Moore, M. C. (1991b). Throat colour reliably signals status in male tree lizards, Urosaurus ornatus. Anim. Behav. 42, 745-753. van Rhijn, J. G. (1983). On the maintenance and origin of alternative strategies in the ruff Philomachus
pugnax.
Ibis
125, 482-498.
van Rbijn, J. (1991). The Rufl, Individuality in a Gregarious Wading Bird. Academic Press, New York. Van Valen, L. (1978). The statistics of variation. Evof. Theory 4, 33-43. West-Eberhard, M. J. (1986). Alternative adaptations, speciation and phylogeny (a review). PNAS 83, 1388-1392. Wilkinson, L. (1990). SYSTAT: The System for Statisrics. Systat, Inc.. Evanston, IL. Win@eld, J. C. (1984a). Environmental and endocrine control of reproduction in the song sparrow, Melospiza melodia. I. Temporal organization of the breeding cycle. Gen. Camp. Endocrinol. 56, 406-416. Wingfield, J. C. (1984b). Environmental and endocrine control of reproduction in the song sparrow, Melospiza mefodia. II. Agnostic interactions as environmental information stimulating secretion of testosterone. Gen. Comp. Endocrinol. 56, 417-424. Wing&Id, J. C. (1985). Short-term changes in plasma levels of hormones during establishment and defense of a breeding territory in male song sparrows, Melospiza melodia. Horm. Behav. 19, 174-187. Wingfield, J. C. (1988). The challenge hypothesis: interrelationships of testosterone and behavior. In H. Ouellet (Ed.), Actu XIX Internationaks Ornithologici, Vol. 2, pp. 1685-1691. Univ. of Ottawa Press, Ottawa, Ontario, Canada. Wingfield, J. C., Ball, G. F., Dufty, A. M., Hegner, R. E., and Ramenofsky, M. (1987). Testosterone and aggression in birds. Am. Sci. 75, 602-608. Wingfield, J. C., Hegner, R. E., Dufty A. M., Jr., and Ball, G. F. (1990). The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. Am. Nat l36, 829-846. Wingfield, J. C., and Marler, P. (1988). Endocrine basis of conunumcation: Reproduction and aggression. In E. Knobil and J. D. Neill (Eds.), The Physiology of Reproduction, pp. 1647-1677. Raven Press, New York. Wingfield, J. C., and Moore, M. C. (1987). Hormonal, social and environmental factors in the reproductive biology of free-living male birds.In D. Crews (Ed.), Psychobiology of Reproductive Behavior: An Evolutionary Perspective, pp. 149-175. Prentice-Hall, Englewood Cliis, NJ. Wingfield, J. C., and Ramenofsky, M. (1985). Testosterone and aggressive behaviour during the reproductive cycle of male birds. In R. Gilles and J. Balthazart (Eds.), Neurobiology, pp. 92-104. Springer-Verlag, Berlin. Wingfield, .I, C., Smith, J. P., and Farner, D. S. (1982). Endocrine responses of whitecrowned sparrows to environmental stress. Condor 84, 399-409.
AND
BEHAVIOR
26,
568-585 (1992)
Behavioral and Hormonal Correlates of Alternative Reproductive Strategies in a Polygynous Lizard: Tests of the Refative Plasticity and Challenge Hypotheses CHRISTOPHER W. THOMPSON’ Department of Zoology, Arizona
AND MICHAEL
C. MOORE
Sate Universiry, Ternpe, Arizona 8.52874501
Species with alternative reproductive tactics are good models to investigate the poorly understood question of whether individual variation within sexes results from the same physiological mechanisms that control variation between sexes. We have shown previously that adult male tree lizards, Vrosaurus ornatus, of different throat color morphs express different levels of aggression in the laboratory. Further field results support the suggestion that the two morphs practice alternative reproductive tactics because the two morphs express different levels of aggressive behavior under field conditions and exhibit dramatic and opposite responses to aggressive challenges. However, despite these behavioral differences, the two morpbs do not differ in levels of testosterone or corticosterone either in undisturbed situations or following aggressive challenge. These results are consistent with the relative plasticity hypothesis which proposes that organizational, rather than activation& actions of steroid hormones will be more important in morph differentiation when morphs are fixed in adult life, as they are in tree lizards. These results also support the hypothesis that steroid hormonal levels are insensitive to social modulation in males of species such as U. ornatus without paternal care. 0 1992 Academic Press, 1~.
A major goal of biologists is to explain the physiological bases of individual variation in morphology and behavior. Such variation may be continuous, such as human body weight, or discontinuous, such as human eye color. Attempts to identify physiological bases of such variability have been much more successful in studies of discontinuous than continuous variation (Feder, 1984; Wingfield and Ramenofsky, 1985). Behavioral and morphological variation is usually more discontinuous between sexes than within sexes. Therefore, much of our understanding of physiological mechanisms that control variable expression of behavior and morphology has 1 Current address: Burke Museum and Department Washington, Seattle, WA 98195. 568 0018-506X/92 $4.00 Copyright All ri&s
8 19!72 by Academic Press, Inc. of reproduction in any form reserved.
of Zoology DB-10, University of
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
569
come from studies of sexual dimorphism (Kelley, 1988). However, whether variation in morphology and behavior within sexes results from the same physiological mechanisms that control variation between sexes is poorly understood. This issue can be addressed by studying species that exhibit discontinuous within-sex behavioral and morphological variation that is as great as that between sexes. Such within-sex polymorphisms are usually associated with alternative reproductive strategies such as occur in male bluegill sunfish, Lepomis macrochirw (Gross, 1984) male Ruffs, Phi1omachu.spugnax, (van Rhijn, 1983, 1991), and male Pacific salmon, Orzcorhynchus sp. (Gross, 1984,1985). Species with multiple reproductive phenotypes probably arose evolutionarily by exaggeration of preexisting behavioral variation in their monophenotypic ancestors (West-Eberhard, 1986). Thus, it seems likely that the mechanisms producing within-sex variation in behavior between alternative phenotypes are similar to mechanisms producing within-sex behavioral variation in more typical species. This suggests further that an understanding of the mechanisms of withinsex behavioral differentiation in the former will help elucidate the mechanisms producing within-sex behavioral variation in the latter. The tree lizard, Urosaurus ornatus, is a small (3-6 g) iguanid lizard of the desert southwestern United States and northwestern Mexico that is polymorphic in the color of its extensible throat fan called a dewlap. Within single populations, dewlaps of adult males are completely blue, yellow or orange, or bicolor combinations of a central blue spot surrounded by an orange perimeter (hereafter orange-blue) or yellow perimeter (hereafter yellow-blue). The size of the central blue spot varies among individuals (Thompson and Moore, 1991a). The dewlap is displayed conspicuously during many social interactions suggesting that polymorphism in this social signal may be correlated with polymorphic social behavior. This has been partially confirmed in laboratory studies which have shown that, independent of age (1) the size of the central blue spot is a reliable predictor of dominance between pairs of size-matched males, regardless of dewlap perimeter color, and (2) orange-blue and yellowblue males respond more aggressively to other males than do orange or yellow males (Hover, 1982, 1985; Thompson and Moore, 1991b). The two main purposes of this study were (1) to confirm and extend in the field our previous laboratory results which suggested that free-living male U. ornatus of orange and orange-blue morphs (the two most common morphs in our study population) practice alternative reproductive tactics and (2) to test predictions of the relative plasticity hypothesis regarding the relationship between male behavioral and hormonal levels in species with alternative reproductive tactics (Moore and Thompson, 1990; Moore, 1991). Our data also address components of the challenge principle, a set of hypotheses regarding the relationship between male be-
570
THOMPSON
AND MOORE
havioral and hormonal levels (Wingfield, 1984a,b, 1985, 1988; Wingfield, Ball, Dufty, Hegner, and Ramenofsky, 1987; Wingfield, Hegner, Dufty, and Ball, 1990). Relative Plasticity Hypothesis Moore and Thompson (1990) and Moore (1991) propose an extension of the classical organization-activation model, called the relative plasticity hypothesis, to explain the hormonal control of differentiation of reproductive polymorphisms within sexes. The classical organization-activation model of sexual differentiation states that sex steroid hormones typically act sequentially during life (Arnold and Breedlove, 1985). First, they have permanent actions during early development which organize target tissues that regulate expression of sexually dimorphic characters. Second, they have temporary actions during adult. load which activate target tissues that were organized early in development. The relative plasticity hypothesis extends this model by first distinguishing between alternative reproductive morphs that are fixed or plastic. Ftied morphs are those in which individuals develop one phenotype at or before sexual maturity and do not change it during adulthood. PZastic morphs are those in which individuals may change phenotype, reversibly or irreversibly, after reaching sexual maturity. By analogy to organization and activation, the relative plasticity hypothesis then states that (1) differences in organizational influences of hormones will be more important in determining differences between fixed morphs and (2) differences in activational influences will be more important in determining differences between plastic morphs. Alternatives to the relative plasticity hypothesis are that (1) hormones play no role in the differentiation and expression of alternative male phenotypes and (2) hormones play a role that is not analogous to organization and activation. Both laboratory and field data indicate that dewlap color morph in U. ornatus is fixed (Hover, 1982, 1985; Thompson and Moore, unpublished data). Consistent with the prediction of the relative plasticity hypothesis that activational influences are not important in producing morph differences, previous studies have shown that seasonal patterns of circulating levels of testosterone and corticosterone secretion are virtually identical in adult males of the two morphs (Thompson and Moore, 1989; Moore and Thompson, in preparation). However, it is still possible that activational influences of hormones cause differences between the morphs by different short-term hormonal responses to social stimuli as appears to be the case in dominant and subordinate baboons (Sapolsky, 1987). We tested the prediction of this hypothesis that hormonal levels in the two morphs will differ after social challenge by measuring hormonal levels in free-living male tree lizards of both morphs after a staged aggressive encounter.
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
571
Challenge Principle Wingfield and colleagues (Wingfield, 1984a,b, 1985, 1988; Wingfield et aZ., 1987, 1990) have proposed the challenge principle which states that males increase their testosterone levels in response to (1) aggression directed toward them from conspecific males and (2) sexual solicitation by females. Recently, however, Wingfield et al. (1990) suggested that these responses occur only in species in which males contribute substantially to parental care (hereafter called care-giving males) and not in many poiygynous species in which males contribute little parental care (hereafter cafled non-care-giving males). Wingfield et al. (1990) hypothesized that adult males in non-paternal-care species are less responsive to behavioral cues that affect secretion of testosterone in paterna1-care species because selection has favored their testosterone levels to be “largely genetically determined” at “maximum levets . . . regardless of social situations.” We refer to this new hypothesis as the social insensitivity hypothesis. Moore (1987a) proposed a similar explanation for the lack of hormonal responsiveness to challenge in the non-paternal-care-giving lizard, Sce/oporus jarrovi. We beheve that the apparent Iack of sensitivity of care-giving males can also be explained by an alternative hypothesis that we call the social modulation hypothesis. According to this hypothesis non-care-giving males may be much more sensitive to social cues. Wingfield et at. (1990) rejected a similar hypothesis based on the lack of responsiveness in non-care-giving males. However, we believe this hypothesis is still valid since it is possible that non-care-giving males are so responsive that it is difficult to measure a true baseline level of testosterone. According to this hypothesis, baseline levels of testosterone in breeding non-care-giving males could only be revealed by housing them in complete isolation from acoustic and visual signals from other males. Our review of the examples cited by Wingfield ef ai. (1990) indicates that this has never been done for these species. Therefore, it is possible that all the reported baseline levels actually reflect sociatiy stimulated elevations of testosterone. U. ornutus is an ideal species for discriminating among these hypotheses because it is polygynous, exhibits no male parental care, and engages in intense male-male aggression throughout the breeding season, but does not have maximal testosterone levels during the entire breeding seasou (Thompson and Moore, 1989; Moore and Thompson, in preparation). We conducted our study at a time of year when testosterone levels are reduced to 50% of maximal levels. Therefore, at the time of year this study was conducted, the social insensitivity hypothesis predicts that, in response to maie challenges, orange-blue (and possibly orange) males will not increase their testosterone levels whereas the social modulation hypothesis predicts that testosterone levels should increase.
572
THOMPSON
MATERIALS
AND MOORE
AND METHODS
Study Area and Animals This study was conducted on 57 free-living adult (> 47 mm snout to vent length (SVL)) male U. ornatus from 14 May to 9 June 1988 in Maricopa Co., Arizona within a 5-km radius north and west of the intersection of Arizona Highway 87 and Sycamore Creek-Sugarloaf Mountain Road (33”42’North lll”32’West) at an elevation of 600-700 m (see Thompson and Moore, 1991a for description). As discussed above, this time of year is the middle of the breeding season when male-male aggressive behavior and female sexual solicitation behavior is frequent and intense, but male testosterone levels are well below their maximum levels exhibited in April (Thompson and Moore, 1989; Moore and Thompson, in preparation). Assignment
of Males to Treatment
Groups
Previously uncaptured males of both morphs were subjected to two treatments described below as each male was encountered in the field. In the first treatment, behaviors were recorded during undisturbed observation periods. In the second treatment, behaviors were recorded during and following staged territorial encounters. Males were captured at the end of both types of observations, morph was assessed, and blood samples were collected as described below. Due to practical considerations, only one type of observation was performed on any given day. Types of observations were alternated on successive days. Measurement of Behaviors of Undisturbed Males All observations were made during the daily maximum of activity from 0650 to 1300 hr. After spotting each adult male at a distance greater than 10 m, C.W.T. approached each male (hereafter resident male) to within 4 to 8 m and remained as motionless as possible for 15 min to minimize possible effects of observer presence on each lizard’s behavior. For the subsequent 30 min, the distance moved (estimated to within 0.5 m) and the number of pushup, fullshow, and shudder displays exhibited by each resident male was recorded. Pushups and fullshows are stereotypic territorial displays and shudders are stereotypic courtship displays (Carpenter and Grubitz, 1961; Moore, unpublished data). The number of 1-min intervals during which these males directed behaviors toward other identifiable male and female U. ornatus was also recorded. Resident males were captured within 6 min of the end of each 30-min observation period. Within 2 min after capture, each male was assessed for morph type, had a sample of blood removed as described below, and was toe-clipped, marked with a spot of paint on the dorsal surface (to prevent using them again), and released.
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
573
TABLE 1 Behavioral Intensity Scale Used in the Painted Dewlap Experiment to Rank Aggressive Responses of Males to One Another during Pairwise Contests for a Limiting Resource Score 0
1 2 3 4 5 6
Description No response Pushups without approaching opponent Fullshows without approaching opponent Fullshows while approaching opponent Faceoff (fullshow displayed in close proximity and parallel to opponentj Charge Bite
Measurement of Behaviors during Staged Encounters
All observations were made during the daily activity maximum from 0625 to 1225 hr. Staged encounters were performed using tethered males as described by Ruby (1976, 1978) and Moore (1987b). To minimize the effects of handling stress on their behavior (Moore, Thompson, and Marler, 1991), intruder males were captured away from the main study area between 0610 and 0930 hr and used only on that day for no more than three tethered presentations. By visual comparison, an attempt was made to match the SVL of intruders within 1 mm of the SVL of the resident. If, after capture, the resident male was found to differ in SVL from the intruder by more than 1.5 mm, the data from these staged encounters and subsequent postencounter observations were omitted from our analyses. Dewlap colors of intruders were not matched with those of resident males. The percentage blue (Defined below) of dewlaps of male intruders presented to orange males was 21.9 -+ 3.8% (means +- 1 SE, limits O45%) and did not differ significantly from the percentage blue (22.3 -I3.5%, limits 5-50%) presented to orange-blue males (two-sample t test using separate variances, t = 0.077, df = 24.7, P = 0.940). Intruders were tethered with about 50 cm of waxed dental floss to the end of a 2.0m pole and lowered onto the territory of the resident about 1 m away from the resident. Intruders were restrained from running off the resident’s territory and from retreating into crevices, but otherwise had free movement. The presentation was continued until (1) the resident gave a maximum response on the behavioral intensity scale (see Table l), (2) the resident ceased to direct behaviors toward the intruder and moved away from the intruder (e.g.., foraging), or (3) 15 min had passed, whichever came first. The maximum intensity of aggressive behavior displayed by
574
THOMPSON
AND MOORE
the resident was recorded on a ranked scale (see Table 1) modified Moore (1987b).
from
Measurement of Behaviors Following Staged Encounters
For 30 min following each staged encounter, the frequency and duration of territorial and courtship behaviors exhibited by resident males was recorded as described above for undisturbed observations. Resident males then were captured and treated as described above. Assessment of Morph Type
As in previous studies (Hover, 1982, 1985; Thompson and Moore, 1991a,b), the extent of blue color in each male’s unextended dewlap was estimated as a percentage of the total surface area of the unextended dewlap within 1 to 2 min after capture by noosing. Males having a distinct central blue spot (always greater than 10% blue) in their dewlaps were assigned to the orange-blue morph and males with no blue or only scattered blue scales (always 10% or less blue) in their dewlaps were assigned to the orange morph. Blood Samples and Radioimmunoassays
Blood samples were collected from the orbital sinus as described by Moore (1986). Blood samples were returned to the laboratory and centrifuged within 8 hr. The resulting plasma samples were stored at -20°C until assayed. Testosterone, corticosterone, progesterone, and estradiol were separated from each other and from plasma lipids by chromatography and then measured by radioimmunoassay as described by Moore (1986). Body Size of Tested Males
A two-way ANOVA indicated that SVL did not differ significantly (1) between orange and orange-blue males used for undisturbed observations, (2) between orange and orange-blue males presented with intruder males, or (3) within or between orange and orange-blue males in comparisons between undisturbed and postencounter observations (P > 0.9). Therefore, differences in behavioral frequencies and intensities reported below cannot be attributed to differences in mean SVL among treatment groups of males. Statistics
Behavioral frequency data were square-root transformed prior to analysis to normalize their distribution (see Moore, 1987a). A value of 0.5 was added to all behavioral frequency data prior to transformation to eliminate data values of zero. No significant skewness or kurtosis of the transformed data was detected by calculation of g, and g,, respectively, of the transformed data. Kolmogorov-Smirnov and Lilliefors tests also
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
575
70-r 13ORANGEWLES,n=lZ IORANGE-BLUEtiS,n=16
rn ii 60-50 -3 40-34F2 30-8 20-t.i a IO-OC
1 BEHAVIOR
3
2
INTENSITY
4
5
- AGGRESSTVE
6 RESPONSE
FIG. 1. Aggressive response of free-living resident orange and orange-blue males to tethered intruder males. See Table 1 for list of behaviors ranked by aggressive intensity.
indicated that the transformed data did not deviate significantly from normality (Sokal and Rohlf, 1981; Wilkinson, 1990). Hormonal data were log transformed prior to analysis to reduce heteroscedasticity of variances between treatment groups (undisturbed versus postencounter hormonal levels). Transformed hormonal data were not heteroscedastic using the statistically sensitive Smith’s test (Van Valen, 1978). Transformed hormonal and behavioral frequency data were analyzed by two-way analysis of variance (ANOVA) with the observation type (undisturbed or postencounter) as treatments and dewlap color (orange or orange-blue) as factors. Nonsignificant between-factor interaction terms are not reported. Post hoc multiple pairwise comparisons were made using Tukey’s Honestly Significant Difference method (Neter, Wasserman, and Kutner, 1990; Sokal and Rohlf, 1981). The level of significance is defined as P < 0.05. All statistical analyses were performed using SYSTAT (Wilkinson, 1990). RESULTS Observations during Staged Encounters
Orange-blue males exhibit a higher median intensity of aggression (see Table 1 for list of behaviors ranked by aggressive intensity) in response to intruder males than do orange males (Mann-Whitney U test, V = 44.5, df = 1, P = 0.012, Fig. 1) and in particular were much more likely to bite intruders than orange males. Observations before and after Staged Encounters Distance moved. Distance moved differs significantly between orange and orange-blue males (F = 20.206, df = 1, Y < 0.001, Figs. 2a and 2b), but not between undisturbed versus postencounter observations (F = 0.048, df = 1, P = 0.828, compare Figs. 2a and 2b). Post hoc com-
576
THOMPSON
8 70- UORANGE,n=Q 8 60-
IORANGE-BLUFi,n=20
& w 60E 40-
*** I;I
AND MOORE
a
20
15
1
10
SEUDDERS FULISIiOWB PU; WP8
-5; 703 60-
2 E g
DISTANCE mvED
WV
DORANGE,n=12 IORANGE-BLUE,n=l6
***
***
b
-20
%
07
h. SHUDDERSFDUSEOWSPUSRUPS
DISlWCEi MOVED
,O
FIG. 2. Frequency (means + 1 SE) of aggressive displays by free-living resident orange and orange-blue males during 30-min periods (a) of undisturbed observation, and (b) following termination of staged encounters with tethered intruder males.***P < 0.001.
parisons indicate that orange-blue males move significantly greater distances than orange males both before (P < 0.001) and after (P < 0.001) staged encounters. Frequency of pushups. Pushup frequency differs significantly between orange and orange-blue males (F = 19.804, df = 1, P < 0.001, Figs. 2a and 2b) and undisturbed versus postencounter observations (F = 8.014, df = 1, P = 0.007, compare Figs. 2a and 2b). Post hoc comparisons indicate that pushup frequency does not differ between orange and orange-blue males during undisturbed observations (P = 0.484). However, following staged encounters, orange males decrease their pushup frequency (P = 0.010) whereas orange-blue males do not (P = 0.707) relative to their respective pushup frequencies during undisturbed observations. As a result, orange-blue males exhibit a greater pushup frequency than orange males following staged encounters (P < 0.001). This differential response to staged encounters by orange and orange-blue males is reflected by a significant between-factor interaction (F = 5.629, df = 1, P = 0.021).
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
577
Frequency offullshows. Fullshow frequency differs significantly between orange and orange-blue males (F = 13.811, df = 1, P < 0.001, Figs. 2a and 2b), but not between undisturbed versus postencounter observations (F = 0.579, df = 1, P = 0.450, compare Figs. 2a and Zb). Post hoc comparisons confirm the overall result that fullshow frequency does not differ between orange and orange-blue males during undisturbed observations (P = 0.790). However, post hoc comparisons reveal that following staged encounters, orange males decrease (P = 0.044) whereas orange-blue males increase (P < 0.001) their frequency of fullshows relative to their respective fullshow frequencies during undisturbed observations. As a result, orange-blue males exhibit a greater fullshow frequency than orange males following staged encounters (P < 0.001). This differential response to staged encounters by orange and orangeblue males is reflected by a significant between-factor interaction (F = 25.62, cff = 1, P < 0.001). Frequency of shudders. Shudder frequency tends to be greater in orange-blue than orange males, but this difference was not significant (F = 3.171, df = 1, P = 0.081, Figs. 2a and 2b). Shudder frequency also did not differ between undisturbed versus postencounter observations (F = 0,011, df = 1, P = 0.915, compare Figs. 2a and 2b). Male-male and male-female interactions. During undisturbed observations, orange and orange-blue males did not differ in the proportion of time spent interacting with other males (Mann-Whitney U test, U = 70.5, df = 1, P = 0.243, Fig. 3a). In addition, orange-blue males tended to spend a greater proportion of time interacting with females than did orange males (Mann-Whitney U test, U = 50.5, df = 1, P = 0.053, Fig. 3a), although the difference is not quite significant. Following staged encounters, orange-blue males spent a greater proportion of time interacting with males (Mann-Whitney U test, U = 2.00, df = 1, P < 0.001, Fig. 3b) and a smaller proportion of time interacting with females (MannWhitney U test, U = 38.00, df = 1, P = 0.005, Fig. 3b) than did orange males. Circulating Blood Plasma Adrenal and Sex Steroid Levels Despite marked differences in behavior, orange and orange-blue males did not exhibit significantly different testosterone or corticosterone levels during undisturbed observations or following staged encounters (testosterone, F = 2.476, df = 1, P = 0.122, Fig. 4a; corticosterone, F = 2.750, df = 1, P = 0.103, Fig. 4b). In addition, neither orange nor orange-blue males exhibited significantly different testosterone or corticosterone levels following staged encounters compared to their respective levels during undisturbed observations (testosterone, F = 0.138, df = 1, P = 0.712, Fig. 4a; corticosterone, F = 2.346, df = 1, P = 0.132, Fig. 4b). We also found no significant differences in mean progesterone levels
578
THOMPSON
AND MOORE
OORANGE,n=Q IORANGE-ELI&n=20
a
OORANGE,n=12 IORANGE-BLUFa,n=16
b
40-
8 35-
8 E
30-
6 a
25-
gE
***
**
20.
0-r
HbLE-MALE MALE-ENCOUNTERTYPE FIG. 3. Percentage (means -t 1 SE) of time that free-living orange and orange-blue resident malts interacted with other male and female Urosaurus orna~.~ during 3O-min pexiods (a) of undisturbed observation, and (b) following termination of staged encounters with tethered intruder males.**P < 0.01 and ***P < 0.001.
between orange and orange-blue males (1) during undisturbed observations or following staged encounters (orange, 0.63 2 0.09 rig/ml; orange-blue, 1.98 2 0.52 rig/ml, F = 0.832, df = 1, P = 0.366), or (2) following staged encounters compared to their respective levels during undisturbed observations (F = 0.840, df = 1, P = 0.364). Estradiol levels were undetectable in all samples. Detectability limits for each sample for estradiol varied from 13 to 18 pg/ml depending on the the volume of plasma used (25-80 ~1) and the percentage recovery (46-81%).
DISCUSSION Our results confirm and extend our previous laboratory results which suggested that orange and orange-blue males pursue alternative reproductive strategies. Our results also support the relative plasticity hypothesis (Moore and Thompson, 1990; Moore, 1991) and the social insensitivity hypothesis (Wingfield et al., 1990), but are inconsistent with the social modulation hypothesis.
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
IXORANGE I ORANGE-BLUE
579
a
T
OBSERVA!l'ION
3
0
g
6
I! s fi
4
3 *
ENCOUNTER
1 ORANGE-BLUE
2 0
UND-
AFPER
OBSERVATION ENCOUNTER FIG. 4. Circulating blood plasma levels of (a)testosterone and (b) corticosterone during undisturbed observations and following staged encounters with tethered intruder males.
Alternative Reproductive Tactics
Our laboratory results suggested that orange males were less aggressive than orange-blue males (Thompson and Moore, 1991b; see also Hover, 1982, 1985). The results of our field studies reported here confirm and amplify these findings and suggest more precisely how the strategies of the two morphs differ. The data indicate that (1) during undisturbed observations, orange-blue males spend more time patrolling their territories and interacting with females than do orange males, (2) during staged encounters orange-blue males respond more aggressively than orange males, and (3) following staged encounters with intruder males, orangeblue males increase their rate of fullshows and the proportion of time spent interacting with males whereas orange males decrease their rate of fullshows and pushups. In addition, orange-blue males also decrease the proportion of time spent interacting with females following staged encounters, probably as a consequence of increasing the proportion of time spent interacting with males. The differential response of orange and orange-blue males to the same stimulus (staged encounters) suggests that orange-blue males are increasing their conspicuousness after an encounter whereas orange males are decreasing their conspicuousness. Additional
580
THOMPSON
AND
MOORE
field data (Thompson, Oliver, and Moore, unpublished data) indicate that orange-blue males reside for much longer periods of time on our study areas than do orange males. The majority of orange males spend only a day or two on the study area. Taken together, these differences in behavior suggest that orange-blue males pursue a strategy of aggressive defense of a territory containing females and perhaps resources important to females. The facts that orange males are less aggressive, move nomadically through the habitat, and behave cryptically following encounters are consistent with the interpretation that these males pursue either a satellite strategy (a permanent but submissive association with a territorial male that results in occasional matings or inheritance of the territory on the death of the owner), a sneaker strategy (wandering in search of unguarded females with which to mate), or both. The nomadic nature of some orange males suggests the latter, whereas the presence of some aggressiveness in orange males is more consistent with the former. Further studies will be necessary to clarify these possibilities. Implications for the Relative Plasticity Hypothesis The relative plasticity hypothesis predicts that fixed morphs will not exhibit differences in their adult hormonal levels. As discussed above, both laboratory and field data indicate that orange and orange-blue males represent fixed alternative morphs. Neither testosterone nor corticosterone levels differed between orange and orange-blue males during undisturbed observations or following staged encounters with other males (Figs. 4a and 4b) or throughout the course of the breeding season (Thompson and Moore, 1989; Moore and Thompson, in preparation). In addition, neither orange nor orange-blue males exhibited a significant change in testosterone or corticosterone levels following staged encounters with other males relative to their levels prior to these encounters (i.e., during undisturbed observations) (Figs. 4a and 4b). These results are consistent with the hypothesis that differences in behavior between orange and orange-blue males are not due to activational differences in testosterone or corticosterone levels and, therefore, support the relative plasticity hypothesis. This hypothesis is further supported by other studies showing that adult males do not change dewlap color when they are castrated or when they receive large testosterone implants (Moore and Thompson, in preparation). As a consequence, these results suggest that differences in behavior between orange and orange-blue males arise due to different organizational actions of testosterone or corticosterone on neural substrates that influence adult male sexual and aggressive behavior associated with reproduction. This suggestion is supported by recent findings that males castrated at hatching develop only orange dewiaps as adults whereas males given testosterone at hatching develop largely into orange-blue males (Hews, Moore, and Knapp, unpublished data).
RELATIVE
PLASTICITY
AND CHALLENGE
Implications for the Social Insensitivity Hypotheses
HYPOTHESES
581
and Social Modulation
Our results support the prediction of the social insensitivity hypothesis that orange-blue (and possibly orange) males should not have higher testosterone levels following staged encounters than during undisturbed observations. Even though our studies were conducted at a time of year when testosterone levels are below the annual maximum, orange and orange-blue males exhibit no increase in testosterone (or corticosteronej levels in response to staged encounters, despite showing dramatic and opposite changes in behavior. In turn, our results do not support the social modulation hypothesis which makes the converse prediction. Two methodological concerns that could account for our failure to detect any change in hormonal levels by orange and orange-blue males in response to staged encounters can be discounted. First, is it possible that 45 min (Iti-min male-male encounter period followed by 30-min observation period prior to capture and bleeding) is insufficient time for testosterone levels to change? This seems unlikely because changes in testosterone levels have been observed within 45 min in many taxonomically diverse vertebrates, including &her iguanid lizards, in response to malemale interactions (e.g., Balthazart, 1983; Wingfield, 1985; Sapolsky, 1986; Wingfield et al., 1987; Greenberg and Crews, 1990) as well as to sexual arousal by females (review by Wingfield and Marler, 1988) and artitlcial and natural stressors (e.g., Wingfield, Smith and Farner, 1982; Moore et al., 1991 and references cited therein). Second, is it possible that staged encounters were not of sufficient intensity or duration to elicit a hormonal response? Wingfield et al. (1987) state that in Song sparrows, Melospiza melodia, challenges must last at least 10 min to cause an increase in testosterone levels. In this study, staged encounters were continued for 15 min or until the resident male exhibited a maximal response, i.e., charged and bit the intruder. These encounters were as intense or more intense than any naturally occurring encounters that we have observed. We cannot completely discount the additional possibilities that the hormonal response was too short in duration or too low in magnitude to be detected by our methods (cf. Moore, 1987b, 1988), but methods similar to ours have been sufficient to detect hormonal responses in a wide variety of vertebrates (Wingfield et al., 1990). Data from Moore (1987a) and Brown (1990) also support the social insensitivity hypothesis in another polygynous iguanid lizard, Sceloporus jarrovi, that exhibits no parental care and engages in intense competition over territories. Despite a marked increase in aggressive reproductive behaviors in response to male challenges, territorial males of this species do not increase testosterone following a challenge either during the breeding season when testosterone levels are high (Moore, 1987a), or during
582
THOMPSON
AND MOORE
the nonbreeding season when males are territorial and have testosterone levels that are much lower (Brown, 1990). However, these results do not constitute as strong a test of the social insensitivity hypothesis as the data reported here because lower intensity male-male competition in S. jurrovi in the nonbreeding season could account for the lack of response at that time (Brown, 1990). In U. ornatus, levels of competition were still high when we conducted our study. We are aware of only two studies of species in which males are polygynous, non-care-giving, and intensely territorial that are inconsistent with the social insensitivity hypothesis. First, Cardwell and Liley (1991) found that androgen levels in male stoplight parrotfish, Sparisoma viride, are very responsive to social stimuli. For example, males increase their levels of testosterone and 11-ketotestosterone, the major androgen in teleost fish, six- to seven-fold in the early stages of territory establishment compared to nonterritorial males. Second; in pair-wise laboratory encounters between male green anoles, Anolis carolinensis, Greenberg and Crews (1990) found that the males that dominated encounters increased their testosterone levels about five-fold compared to subordinate and control males. These data suggest that the social insensitivity hypothesis is either not universally applicable or that it needs to be modified to incorporate other factors that will account for cases like these. Further studies are necessary to examine these possibilities. U. ornatus differ from several of the non-paternal-care-giving bird species that have been studied in that adult males do not have maximal testosterone levels throughout the entire breeding season. Nevertheless, testosterone appears to be necessary for the expression of aggressive reproductive behavior in male U. ornatus because free-living castrated males cannot maintain their territories (Knapp and Moore, unpublished data). Thus, we believe that a minimum level of testosterone is necessary to activate neural substrates so that males will exhibit the full repertoire of aggressive reproductive behaviors in response to appropriate social stimuli such as conspecific males and females (Arnold and Breedlove, 1985). However, short-term changes in the frequency and intensity of aggressive reproductive behaviors following male-male interactions probably are not modulated by short-term increases in testosterone levels above this threshold. Rather these behavioral changes are probably modulated by mechanisms entirely within the central nervous system or by other rapidly responding systems such as the adrenal catecholamines (Moore, 1987a; Moore and Marler, 1987). Why do paternal care-giving species use sex steroids to modulate aggressive behavior whereas many species without paternal care apparently do not? In care-giving males, time and energy demands of parental care require that males reduce territorial aggression during periods of parental care (Silverin, 1980; Moore, 1984; Wingfield, 1985; Wingfield and Ra-
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
583
menofsky, 1985; Wingfield and Moore, 1987, Wingfield et al., 1987). Wingfield et al. (1990) propose that increases in testosterone directly cause increases in aggressive behavior. In contrast, we propose, similar to a previous suggestion by Moore (1987a), that care-giving males possess an unknown physiological mechanism that inhibits full expression of aggressive reproductive behavior during periods of parental care and that this mechanism is disinhibited by an increase in testosterone levels that occurs in response to intense challenges. In other words, increased testosterone levels do not cause the rapid increase in aggressive behavior observed during and immediately following an initial challenge, but rather result from an initial challenge and act to increase the rapidity and intensity of future responses to challenges by reducing the effectiveness of the inhibiting system (Wingtield et al., 1987). In contrast, in non-care-giving males such an inhibiting mechanism would serve no function and, thus, would not be adaptive. A good test of this hypothesis would be to compare the effect on testosterone levels of challenges to territorial males in a polygynous species, such as red-winged blackbirds, Agelaius phoenicem, in which male parental care is expressed in some populations but not others (Beletsky and Orians, 1990). Similarly, it is also possible that orange morph male U. ornatus possess an inhibiting mechanism similar to caregiving male birds that reduces their response to encounters. In this case, however, the effectiveness of the mechanism is not modulated by testosterone because the reduction in aggressiveness is a permanent characteristic of this morph. Further comparisons are needed to clarify this hypothesis. ACKNOWLEDGMENTS This research was supported by Presidential Young Investigator Award DCB-8451641 from the National Science Foundation to M. C. Moore and by an Arizona State University Graduate College Research Assistantship to C. W. Thompson. We thank D. Crews, D. K. Hews, R. Knapp, I. T. Moore, and S. Rohwer for helpful criticisms of the manuscript, and R. Knapp for helping us critically review the literature upon which Wmgfield et al. (1590) based their social insensitivity model.
REFERENCES Arnold, A. P., and Breedlove, S. M. (1985). Organizational and activational effects of sex steroids on brain and behavior: A reanalysis. Horm. Behav. 19, 469-498. Balthazart, J. (1983). Hormonal correlates of behavior. In D. S. Farner, J. R. King, and K. C. Parkes (Eds.), Avian Biology, Vol. 7, pp. 221-365. Academic Press, New York. Beletsky, L. D., and Orians, G. H. (1990). Male parental care in a population of redwinged blackbirds, 1983-1988. Can. J. Zool. 68, 606-609. Brown, M. R. (1990). Hormonal and Environmental Znteractions Regulating Post-Challenge Aggressive Responses of Territorial Male Lizards. Unpublished M. S. thesis, Arizona State University, Tempe, AZ. Cardwell, J. R., and Liley, N. R. (1991). Androgen control of social status in males of a
584
THOMPSON
AND MOORE
wild population of stoplight parrotfish, Sparisoma viride (Scaridae). Horm. Behav. 25, 1-18. Carpenter, C. C., and Grubitz G., III. (1961). Time-motion study of a lizard. Ecology 42, 199-200. Feder, H. H. (1984). Hormones and sexual behavior. Annu. Rev. Psychol. 34, 165-200. Greenberg, N., and Crews, D. (1990). Endocrine and behavioral responses to aggression and social dominance in the green anole lizard, Anolis carolinensis, Gen. Comp. Endocrinol. 77, 246-255. Gross, M. (1984). Sunfish, salmon, and the evolution of alternative reproductive strategies in fishes. In G. W. Potts and R. J. Wooton (Eds.), Fish Reproduction: Strategies and Tactics, pp. 55-75. Academic Press, New York. Gross, M. R. (1985). Disruptive selection for alternative life histories in salmon. Nature (London) 3l3, 47-48. Hover, E. L. (1982). Behavioral Correlates of a Throat Color Polymorphism in a Lizard, Urosaurus ornatus. Unpublished Ph.D. Dissertation, University of Michigan, Ann Arbor, MI. Hover, E. L. (1985). Differences in aggressive behavior between two throat color morphs in a lizard, Urosaurus ornatus. Copeia 1985,933-940. Kelley, D. B. (1988). Sexually dimorphic behaviors. Annu. Rev. Neurosci. 11, 225-251. Moore, M. C. (1984). Changes in territorial defense produced by changes in circulating levels of testosterone: A possible hormonal basis for mate-guarding behavior in whitecrowned sparrows. Behaviour 88, 215-226. Moore, M. C. (1986). Elevated testosterone levels during nonbreeding season territoriality in a fall-breeding lizard, Sceloporus jarrovi. Z. Comp. Physiol. 158A, 159-163. Moore, M. C. (1987a). Circulating steroid hormones during rapid aggressive responses of territorial male mountain spiny lizards, Sceloporus jarrovi. Horm. Behav. 21, 511-521. Moore, M. C. (1987b). Castration affects territorial and sexual behaviour of free-living male lizards, Sceloporus jarrovi. Anim. Behav. 35, 1193-1199. Moore, M. C. (1988). Testosterone control of territorial behavior: tonic-release implants fully restore seasonal and short-term aggressive responses in free-living castrated lizards. Gen. Comp. Endocrinol. 70, 450-459. Moore, M. C. (1991). Application of the organization-activation theory to alternative male reproductive strategies: A review. Horm. Behav. 25, 154-179. Moore, M. C., and Marler, C. A. (1987). Effects of testosterone manipulations on nonbreeding season territorial aggression in free-living male lizards,Sceloporu.s jarrovi. Gen. Comp. Endocrinol. 65, 225-232. Moore, M. C., and Thompson, C. W. (1990). Field endocrinology of reptiles: Hormonal control of alternative reproductive tactics. In A. Epple, C. G. Scanes, and M. H. Stetson (Eds.), Progress in Comparative Endocrinology, pp. 685-690. A. R. Liss, New York. Moore, M. C., Thompson, C. W., and Marler, C. A. (1991). Reciprocal changes in corticosterone and testosterone levels following acute and chronic handling stress in tree lizards, Urosaurus ornatus. Gen. Comp. Endocrinol. 81, 217-226. Neter, J., Wasserman, W., and Kutner, M. H. (1990). Applied Linear Statistical Models, 3rd ed. R. D. Irwin, Boston, MA. Ruby, D. E. (1976). The Behavioral Ecology of the Viviparous Lizard Sceloporus jarrovi. Unpublished Ph.D. Dissertation, University of Michigan, Ann Arbor, MI. Ruby, D. E. (1978). Seasonal changes in territorial behavior of the iguanid lizard Sceloporus jarrovi. Copeia 1978, 430-438. Sapolsky, R. (1986). Stress-induced elevation of testosterone concentrations in high ranking baboons: Role of catecholamines. Endocrinology ll8, 1630-1635.
RELATIVE
PLASTICITY
AND CHALLENGE
HYPOTHESES
585
Sapolsky, R. M. (1987). Stress, social status, and reproductive physiology in free-living baboons, ln D. Crews (Ed.), Psychobiology of Reproductive Behavior, an Evolutionary Perspective, pp. 291-322. Prentice-Hall, Englewood Cliffs, NJ. Silverin, B. (1980). Effect of long acting testosterone treatment on free-living pied flycatcbers, Ficedula
hypo1euca.
Anim.
Behav.
28, 906-912.
Sokal, R. R., and Rohlf, F. J. (1981). Biometry, 2nd ed. Freeman, San Francisco, CA. Thompson, C. W., and Moore, M. C. (1989). Interactions among hormones, dominance and color in tree lizards. Am. Zool. 29, 447A. Thompson, C. W., and Moore, M. C. (1991a). Syntopic occurrence of multiple dewlap color morphs in male tree lizards, Urosaurus ornatw. Copeia 1991,493-503. Thompson, C. W., and Moore, M. C. (1991b). Throat colour reliably signals status in male tree lizards, Urosaurus ornatus. Anim. Behav. 42, 745-753. van Rhijn, J. G. (1983). On the maintenance and origin of alternative strategies in the ruff Philomachus
pugnax.
Ibis
125, 482-498.
van Rbijn, J. (1991). The Rufl, Individuality in a Gregarious Wading Bird. Academic Press, New York. Van Valen, L. (1978). The statistics of variation. Evof. Theory 4, 33-43. West-Eberhard, M. J. (1986). Alternative adaptations, speciation and phylogeny (a review). PNAS 83, 1388-1392. Wilkinson, L. (1990). SYSTAT: The System for Statisrics. Systat, Inc.. Evanston, IL. Win@eld, J. C. (1984a). Environmental and endocrine control of reproduction in the song sparrow, Melospiza melodia. I. Temporal organization of the breeding cycle. Gen. Camp. Endocrinol. 56, 406-416. Wingfield, J. C. (1984b). Environmental and endocrine control of reproduction in the song sparrow, Melospiza mefodia. II. Agnostic interactions as environmental information stimulating secretion of testosterone. Gen. Comp. Endocrinol. 56, 417-424. Wing&Id, J. C. (1985). Short-term changes in plasma levels of hormones during establishment and defense of a breeding territory in male song sparrows, Melospiza melodia. Horm. Behav. 19, 174-187. Wingfield, J. C. (1988). The challenge hypothesis: interrelationships of testosterone and behavior. In H. Ouellet (Ed.), Actu XIX Internationaks Ornithologici, Vol. 2, pp. 1685-1691. Univ. of Ottawa Press, Ottawa, Ontario, Canada. Wingfield, J. C., Ball, G. F., Dufty, A. M., Hegner, R. E., and Ramenofsky, M. (1987). Testosterone and aggression in birds. Am. Sci. 75, 602-608. Wingfield, J. C., Hegner, R. E., Dufty A. M., Jr., and Ball, G. F. (1990). The “challenge hypothesis”: Theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. Am. Nat l36, 829-846. Wingfield, J. C., and Marler, P. (1988). Endocrine basis of conunumcation: Reproduction and aggression. In E. Knobil and J. D. Neill (Eds.), The Physiology of Reproduction, pp. 1647-1677. Raven Press, New York. Wingfield, J. C., and Moore, M. C. (1987). Hormonal, social and environmental factors in the reproductive biology of free-living male birds.In D. Crews (Ed.), Psychobiology of Reproductive Behavior: An Evolutionary Perspective, pp. 149-175. Prentice-Hall, Englewood Cliis, NJ. Wingfield, J. C., and Ramenofsky, M. (1985). Testosterone and aggressive behaviour during the reproductive cycle of male birds. In R. Gilles and J. Balthazart (Eds.), Neurobiology, pp. 92-104. Springer-Verlag, Berlin. Wingfield, .I, C., Smith, J. P., and Farner, D. S. (1982). Endocrine responses of whitecrowned sparrows to environmental stress. Condor 84, 399-409.