ABSTRACT
The Adaptive Significance of Cleaning A function for cleaning was fi rst suggested by the discovery of parasites in the gut of cleaner wrasses by Randall (1955, 1958). Soon aft er, a cleanerremoval experiment by Limbaugh (1961) from two Bahamian reefs resulted in dramatic immigration of fish from cleaning goby-free areas as well as a marked increase in disease in those individuals that remained. Th is apparently crucial role played by cleaning gobies in maintaining the health of their clients, combined with evidence that cleaners actually feed on ectoparasites, implied that cleaning was a co-operative interaction occurring between cleaner and client. Cleaning symbioses were viewed from then on as textbook examples of mutualistic interactions (Trivers, 1971; McFarland, 1985; Begon et al., 1990; Th ompson, 1994). It has taken a surprisingly long time for researchers to verify this longheld perception. However, it is now agreed that the interactions between cleaner wrasses and their clients are largely mutualistic, on the basis of the signifi cant decrease in client parasite loads eff ected by cleaner wrasses (e.g., Grutter, 1999) and the large eff ects of on client fi sh density and community structure resulting from cleaner removal (Bshary, 2003; Grutter et al., 2003).
By contrast, the evidence has not been as strongly forthcoming for cleaning gobies. Cleaning gobies do remove ectoparasites from the fish they inspect (Fig. 4.3.3). Th e identifi able diet of coral-dwelling Elacatinus prochilos, for example, is composed largely of larval gnathiid isopods (Whiteman and Côté, 2002; Fig. 4.3.2B), the commonest ectoparasites found in the stomach contents of cleanerfi sh worldwide. Client fi sh tend to visit cleaning stations more in the morning (Côté and Molloy, 2003; Sikkel et al., 2004), which coincides with periods of peak ectoparasite infestation (Chambers and Sikkel, 2002; Sikkel et al., 2006). Th ere is also evidence that reef fi sh with access to cleaning gobies (E. evelynae and E. prochilos) have lower parasites loads. For example, Barbadian longfin damselfish Stegastes diencaeus that visit these cleaning gobies often, by virtue of having a cleaning station within their territories, have signifi cantly fewer ectoparasites than damselfi sh that live further away from cleaners and hence visit them less frequently (Cheney and Côté, 2001). Th is diff erence is not simply due to diff erences in damselfi sh quality because a similar eff ect was observed at the end of a 6-month experiment in which half of the damselfi sh ‘lost’ the cleaning gobies that were originally in their territories (Cheney and Côté, 2003a). However, in both cases, the diff erence in ectoparasite load was small-approximately 1 gnathiid per damselfish-and such a small reduction did not result in a detectable decline in damselfish breeding output in Barbados (Cheney and Côté, 2003a). Th e apparently small benefi t of being cleaned by gobies may be biased by the fact that these studies have all been conducted in Barbados, where ectoparasite abundance is generally very low. In other parts of the Caribbean (e.g., in Puerto Rico) where there are more ectoparasites, the benefi t of being cleaned by gobies can be twice as high (Cheney and Côté, 2005), but this still amounts to only two 1-mm long gnathiids per fi sh (Fig. 4.3.2B). Nevertheless, if gnathiids are an important source of skin irritation or vectors for blood diseases (e.g., Davies et al., 1994), then the removal of a few parasites may have an appreciable impact on host fi tness and health. Geographic variation in benefi t of being cleaned by gobies should be accompanied by variation in willingness of clients to incur costs of seeking cleaners. Th e costs of being cleaned, for clients, include travelling time to cleaning stations, risk of predation during travel, and resources lost from temporarily abandoned territories. Most of these are linked to travel distance (Cheney and Côté, 2001). Th us, in areas where ectoparasites are abundant and the scope of a benefi t of being cleaned is high, clients should be more willing to travel long distances to be cleaned. Conversely, when the
