dependent status signalling honest in an ... - Wiley Online Library

2 downloads 0 Views 95KB Size Report
Abstract Australian painted dragon lizards Ctenophorus pictus occur in three head colours (red, orange and yellow) that differ in their level of aggression (reds ...
Austral Ecology (2009) 34, 636–640

Too big for his boots: Are social costs keeping condition-dependent status signalling honest in an Australian lizard? MO HEALEY* AND MATS OLSSON School of Biological Sciences, University of Wollongong, NSW 2522, Australia (Email: [email protected])

Abstract Australian painted dragon lizards Ctenophorus pictus occur in three head colours (red, orange and yellow) that differ in their level of aggression (reds being most aggressive), hormone profile (reds having higher testosterone levels) and in their frequency in our study population over time. They are also polymorphic in bib colour; some males have a bright yellow area under the chin, while others lack this coloured area entirely. We show that red males with a bib are in better body condition than red males that lack a bib.This contrasts sharply to yellow males, in which males with a bib are in poorer condition than yellow males that lack a bib. Our analysis also shows that following exposure to a high percentage of red (more aggressive) neighbours, all males suffer a reduction in body condition, and importantly, males with a bib (regardless of their head colour) suffer a more severe loss of body condition than males that lack a bib. Finally, this condition loss is significantly higher for yellow bibbed males than for red bibbed males, suggesting that the cost of sporting a bib may be higher for them. Orange males showed a non-significant difference in condition between bib morphs. Our analysis also shows that bibbed yellow males (the morph with lower body condition), but no other morph category, declined significantly in their frequency between 2 years. Key words: body condition, colour polymorphism, multiple signal.

INTRODUCTION Male Australian painted dragon lizards Ctenophorus pictus occur in three different head colours (red, orange and yellow).They differ in dominance, with red males winning contests more frequently than yellow males (Healey et al. 2007) and they also differ in plasma levels of testosterone, with red males building up higher levels throughout a day of territory patrolling than yellow males (Olsson et al. 2007a). Yellow males have larger relative testis size and sire three times more offspring than red males in sperm competition experiments (Olsson et al. 2007c), suggesting that the different colour morphs may have different reproductive tactics. Orange morphs have so far not been behaviourally characterized because of their previously low frequency in the wild. Some males (regardless of head colour) also have a yellow bib under the chin, while others do not and instead are ‘whitish blue’ (skin-coloured) in the corresponding area. This bib area is conspicuous during bouts of head bobbing displays that males engage in to deter rivals and may therefore be perceived by them as

*Corresponding author. Accepted for publication September 2008.

© 2009 The Authors Journal compilation © 2009 Ecological Society of Australia

a badge of status, serving as an indicator of social status and aggressiveness (Bradbury & Vehrencamp 1998). Signals of quality or status require costs that keep them honest in order to persist in the population (Zahavi 1975). Badges of status, however, appear to be an inconsistent facet of signalling as they are considered to have very low production costs, and, hence, may be open to invasion by a cheating mutant strategy (Rohwer 1975; Dawkins & Krebs 1978; Mc Graw et al. 2003). In birds, for example, studies have found that melanin-based badges of status may be independent of body condition and of the nutritional state and overall health of the individual (Gonzalez et al. 1999; Mc Graw et al. 2003) and some badges may be composed of pigment-free feathers, which are relatively cheap to produce (Pärt & Qvarnström 1997). As yet, we do not know the underlying physiological mechanism behind bib production in the painted dragon, although high-performance liquid chromatography and acidified pyridine analysis showed that yellow head colour is determined by integumental deposition of carotenoids, and red head colour is also determined by deposition of another (yet to be identified) pigment, most likely a pteridine (Olsson et al. 2007b). The bib remains vivid when the animal is cold, unlike head colour that fades to almost black, and the bib is not doi:10.1111/j.1442-9993.2009.01968.x

S TAT U S S I G N A L L I N G I N A L I Z A R D

visible from above, which is likely to lower the cost from the perspective of predation pressure. Theoretical models of badge evolution, in particular those in relation to honesty, suggest that badges are more correlated with aggressive motivation than with some correlate of fighting ability (Maynard Smith & Harper 1988). Maynard Smith and Harper’s model also suggested that large badges (signalling high levels of aggression vs. small badges signalling less aggression) would only be evolutionarily stable if the costs of escalated fights, relative to the benefits of winning, increased rapidly with increasing badge size and that animals with large badges were often challenged by others with large badges. Deceitful individuals that display a large badge that is inconsistently greater than their motivation to win, and/or their fighting ability, would suffer higher costs (matching their degree of deceit) than an aggressive opponent that was signalling honestly (Maynard Smith & Harper 1988; Bradbury & Vehrencamp 1998). Evidence is now growing for this theory from the ornithological literature (Rohwer & Rohwer 1978; Johnstone & Norris 1993; Mc Graw et al. 2003) and studies of eusocial insects (Tibbetts & Dale 2004). In the present study, we examine these ‘social costs’ of bearing a badge under the assumption that C. pictus morphs of more or less aggressive strategies may be under different selection pressures depending on morph-specific costs of carrying a badge of status in the wild. We have shown elsewhere that red painted dragon males dominate yellow males in staged contests, despite their strong morphological similarities, suggesting that it is disparities in their levels of aggression and motivation to fight that swing the cost and benefit ratio to the red male’s favour (possibly mediated via higher testosterone levels; Olsson et al. 2007a). We examined variation in body condition (estimated as residuals from mass snout–vent length (SVL) regressions) among free-ranging male C. pictus of different head colours and bib categories in order to assess links between the social environment (levels of competition), morph-specific bib expression (present or absent) and condition. Our proxy for the level of competition was the percentage of red neighbours. As this is the more aggressive morph, we would expect that being in an environment that has a higher density of red males would lead to a potential increase in aggressive encounters, or at least ‘fear’ of rivals. METHODS Field protocol The Australian painted dragon C. pictus is a small (adult SVL 65–95 mm, mass 8–16 g), diurnal lizard of © 2009 The Authors Journal compilation © 2009 Ecological Society of Australia

637

typically sandy habitats and low vegetation, with a range covering central and western New South Wales to Western Australia. It is short-lived (only approx. 10% live to a second year) and strongly territorial (Olsson et al. 2007b). We studied a desert population at Yathong Nature Reserve, New South Wales, Australia (145°35′E; 32°35′S) every day that the weather permitted lizard activities during the mating season (August–January) in 2000 and 2005. We caught males by noosing from a slow-moving (and then stationary) vehicle. Noosed lizards were weighed and measured and we estimated body condition as the residuals from mass SVL regressions. Morph colouration has no spectral UV reflectance component and is perceived by the human eye as discrete morphs as opposed to continuous variation in colour (Olsson et al. 2007b, 2008). We therefore used this red/yellow classification for determining morphs in this study. Lizards were scored for head and bib colouration, and had a number painted on their backs before being released at the place of capture. Every time a lizard was observed in the field, a GPS (Global Positioning System) reading was taken of its location, so that we could map territories and assess neighbouring colour morphs.

Statistical methods We based our choice of main effects and covariates on already published data on main determinants of reproductive tactics in this species (Olsson et al. 2007c), while extending these analyses to include the effects of bibs in the present study. We analysed generalized linear models with respect to three main effects in Proc GLM (SAS Institute): head colour (red, orange, yellow), bib (present, absent), year (2000, 2005), the interaction effect between head and bib colour, and two covariates; day, and an index of social competition (% red neighbours), with body condition as the dependent variable. We then conducted morph-specific analyses to assess quantitative effects of competition on male condition.

RESULTS Our analysis showed significant effects on male body condition of year, day of the year and the social environment a male experienced, expressed as percentage of red neighbours (Table 1). Head colour and the presence or absence of a bib were not significant main effects on body condition, but their interaction effect was statistically significant (P < 0.0001; Table 1), suggesting that bib colour is differently related to body condition depending on head colour. Morph-specific analyses (Table 2) showed that red males with bibs were in better condition than red males lacking bibs doi:10.1111/j.1442-9993.2009.01968.x

638

M. HEALEY AND M. OLSSON

Table 1. Effects of head colour, bib, their interaction, year, percentage of red neighbours and day on male body condition (anova, Proc GLM, SAS) Source

d.f.

Type III SS

MS

F

P

2 1 2 1 1 1

3.2 0.3 24.0 7.8 5.8 28.5

1.6 0.3 12.0 7.8 5.8 28.5

1.6 0.3 12.0 7.8 5.7 28.4

NS NS