The Indo-Pacific cleaner wrasse, Labroides dimidiatus, is one of the most studied cleaner fish species and has been a model system for work on the evolution of among-species cooperation (Soares, 2017). This cleaner has a lateral black stripe on a conspicuously vivid blue background, together with some yellow in the front, resembling the coloration of several other cleaner species in the same family (Arnal et al., 2006). These color patterns are used by client fishes to identify cleaners (Stummer et al., 2004; Cheney et al., 2009), and their blue color within the pattern attracts more client fish to cleaning stations (Cheney et al., 2009). Moreover, at least in juvenile L. dimidiatus, their color pattern appears to deter predation (Cheney, 2013). Therefore, the vivid colors of cleaner fish are an interspecific communication signal of cleaning for prospective clients (Vaughan et al., 2017).

In non-cleaner species, vivid colors often signal among-individual differences in some aspects of quality. For example, colorful signals are often sexually dichromatic and used in mate choice and may indicate aspects of individual health or condition such as parasite infection level (Johnstone, 1996; Maan et al., 2008; Megía-Palma et al., 2016). In aposematic species, differences in color can also indicate differences in toxicity among individuals (reviewed in Summers et al., 2015). Similarly to sexual and aposematic color signals, among-individual differences in the vivid colors of cleaner fish might also provide cues to aspects of individual quality that are useful for either conspecifics or to interspecific clients. For example, clients use this blue color to distinguish cleaner fish from non-cleaning species (Cheney et al., 2009), and recent laboratory experiments suggested that blue color saturation in L. dimidiatus is condition-dependent, thus providing cues on the quality of individuals or the quality of the cleaning service they provide (Trigo et al., 2020). These cues could inform clients decisions when evaluating individual cleaners, which in nature would allow choosing a better cleaning service. But natural variation in color within a wild population has never been studied for any cleaner fish species. Therefore, there is a lack of insight from the wild on which individual traits color differences may signal.

In order to gain further understanding on the communication role of cleaner fish color variation, specifically their blue color saturation, here we asked if among-individual differences in coloration are explained by sex, body size or degree of parasitism in L. dimidiatus. This cleaner species is protogynous hermaphrodite, first breeding as females and some eventually changing sex to become male harem owners (Robertson, 1972, 1974). As in other members of the genus Labroides, males are on average larger than females (Robertson, 1972) but, although detailed studies of sex differences in color are lacking, the species is described as sexually monochromatic, that is, females can display the same color patterns as males (Pandian, 2010). Since sex and size are correlated in this protogynous hermaphrodite species, we measured color in males and females, tested for sex differences in color, and whether those differences could be a consequence of, or independent from, body size. Additionally, these cleaner fish also naturally carry ecto and endoparasites (Narvaez et al., 2021), which they obtain from their environment or are transmitted from their clients (e.g., Jones et al., 2004; Narvaez et al., 2022). Because parasite load can be a major challenge to the condition of cleaner fish we also asked if this is reflected in their color signal variation.

We found that male cleaner wrasses (L. dimidiatus) are heavier, longer and had a non-significant tendency to show more saturated blue color than females. Endo- and ectoparasite loads did not predict blue color saturation. The saturation of blue color was positively correlated with size but, when using a model with multiple predictors (sex, size, size-corrected mass, parasite loads) to control for the effects of covariates, the trend for sex differences in color was no longer detected, and the association of color with size was also slightly weakened. Therefore, sex differences in color may appear in the cleaner wrasse as a consequence of sex differences in size and an association between size and blue color saturation. The remaining color traits studied here did not differ between the sexes nor were associated with morphology or parasite loads. These latter results should be interpreted with caution, because repeatability was not strong for some of those other color measurements (namely for yellow saturation and brightness of black).

Finding an association between blue color saturation and body size or mass suggests that the conspicuous blue color of L. dimidiatus could signal for physical condition. This agrees with previous experimental results, showing that an oxidative stress treatment was able to reduce blue color saturation and the quality of the cleaning service in these species (Trigo et al., 2020). Such a condition-dependent signal could be used both in intra- and inter-specific contexts. In an inter-specific context, this condition-dependent signal could be used by potential clients when choosing cleaners. In fact, clients are known to exert choice between alternative individual cleaners and cleaning stations (Bshary and Schäffer, 2002). Furthermore, cleaners also advertise their cleaning stations to clients with “dances” (Potts, 1973) in which they swim in an oscillating manner (Youngbluth, 1967; Losey, 1971), with their conspicuous black stripe and strikingly prominent blue tail during these dances. Stummer et al. (2004) used models of L. dimidiatus to show that, in models of normal body size, a larger black stripe increased the number of clients approaching, but not their interest at close distance. They concluded that the black stripe attracts clients at distance, but at close range, the client interest may be based on other cues such as behavioral (e.g., cheating behavior) or sensorial (e.g., chemical cues). Another example of choice exerted by clients, occurs when these eavesdrop on cleaners’ interaction with other clients in order to choose whether to wait and visit that cleaner or search for another (Pinto et al., 2011). Clients benefit from choosing among individual cleaners, because cleaning service can differ significantly: cleaners may attend to clients rapidly or make them wait, they may or may not provide tactile stimulation, and they can also cheat by feeding on healthy body tissue instead of ectoparasites (Bshary, 2002; Bshary and Côté, 2008). Whether or not clients evaluate among-individual differences in blue color saturation to help with their choice remains to be studied. If the color of the blue tail indicates aspects of quality, then clients may use this information when making their choices.

In an intra-specific context, condition-dependent color may perhaps help to assess dominant status among cleaners, which is important for their social behavior. For instance, dominant cleaner males are generally older individuals that live and clean in pairs, choosing for this the largest female of their harem, but patrolling all other females regularly (Robertson, 1972). When the male disappears, usually the largest (and more dominant) female of that harem, or from a nearby territory, changes sex and takes over the harem (Sakai et al., 2001). We can assume that cleaner wrasses compete to increase social status and hence to gain access to better territories. In the case of males also to gain access to female harems, and in the case females also to clean with the dominant male as a pair (since pairs usually inspect larger clients; Bshary et al., 2008) and eventually to gain priority in transitioning sex to a harem-holding male. Since dominance increases with size (Sakai et al., 2001), and having found that blue color saturation covaries with size, we hypothesize that cleaner wrasses may use the blue color of conspecifics to help evaluate social dominance during these various types of social interactions. This too, however, remains to be studied.

The absence of sex or size differences regarding other colors (yellow or black) is not surprising considering the importance of blue guild coloration to distinguish cleaner fish species from non-cleaners (Cheney et al., 2009). While cleaners are more likely to have a combination of blue and yellow colors and there is evolutionary correlation between displaying black stripes and cleaning behavior (Arnal et al., 2006), blue color is still the source of highest chromatic contrast against most microhabitat backgrounds, as evaluated by the visual systems of most client species (Lettieri et al., 2009). Finally, we expected higher parasitism levels affected coloration, since some studies in other fish species have reported negative associations between parasite infection and aspects of individual quality (e.g., size of sand bowers in Copadichromis spp., Taylor et al., 1998, or territory size and red coloration in a diverse lineage of Lake Victoria cichlids, Pundamilia nyererei, Maan et al., 2008). However, we did not find associations between parasite infestation and cleaners, which could be due to parasite loads in the wild being low to affect coloration, while cleaners not being exempted from parasite infection (Narvaez et al., 2021, 2022), or to these parasites not being detrimental.

In this study we found that, unlike commonly assumed, the cleaner wrasse L. dimidiatus may not be entirely sexually monochromatic (Pandian, 2010) but show size-dependency in both their sex and their blue color. Since the cleaner wrasse is able to change sex bidirectionally, depending on sex-ratio and social-environmental factors, perhaps the association between blue color saturation and size makes the blue color a useful signal of sex, or even a sexually-selected trait. Simultaneously, since blue color saturation indicates size, it can be a more general signal of individual quality or condition, useful also for interspecific (cleaning) interactions with clients. Such a condition-dependent signal of quality might be used in various types of inter- and intraspecific interactions, including choice of cleaning stations, competition for resources or individual dominance within harems, thus suggesting several hypotheses for future research.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

PN performed the material preparation and data collection. ST performed the analysis and wrote the first draft of the manuscript. All authors commented on previous versions of the manuscript, read and approved the final manuscript, and contributed to the study conception and design.

PN has been funded by the James Cook University Postgraduate Research Scholarship (JCUPRS). ST, GC, and MCS were supported by Portuguese National Funds through the Fundação para a Ciência e Tecnologia (grant numbers PTDC/BIA-ECO/32210/2017, PTDC/BIA-COM/2644/2020, DL57/2016/CP1440/CT0011, DL57/2013/CP1440/CT0019), and FEDER funds under COMPETE 2020 (CEEC Individual–2021.01458.CEECIND/CP1668/CT0003). Work co-funded by the project NORTE-01-0246-FEDER-000063, supported by Norte Portugal Regional Operational Programme (NORTE2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.