HOBSON': CLEANING SYMBIOSIS 



ordinary. Rejiorts are widespread (see Feder, 

 1966) of cleanino- recipients assuming- these un- 

 natural-appearing postures. 



Attention-getting postures assumed by fishes 

 being cleaned probably occurred incidentally 

 during the early development of cleaning sym- 

 biosis, when fishes hovering to be cleaned quite 

 naturally stopped moving and passively drifted 

 out of their regular attitudes. As the various 

 cleaning relations evolved, apjiarently this ob- 

 vious cue subsequently assumed a difl'erent role 

 as a signal in different situations. Generally 

 these postures are suggested to be signals be- 

 tween the recipient of cleaning and the cleaner, 

 indicating a readiness to be cleaned. Quite likely 

 this is the primary signal-function in activity 

 involving such cleaners as Labroides phthiro- 

 phagus. where all members of the siiecies are 

 cleaners and where activity is centered around 

 cleaning stations that are well known to other 

 fishes in the area. In this situation a fish in 

 need of cleaning- should be reasonably successful 

 in advertising its condition by assuming the 

 characteristic soliciting posture before a fish 

 recognizable as a cleaner. Losey (1971) showed 

 that various fishes regularly employ this tactic 

 to induce L. phthimiihagiis to clean them. Ob- 

 servations in the Gulf of California demonstrated 

 that the cleaning station itself has played a role 

 in establishing the soliciting- posture as a cue. 

 There I have seen the goatfish MuJloidichthys 

 denfatits hovering head-down at cleaning sta- 

 tions of the butterflyfish Heniochiis nigrirostris, 

 when the resident cleaner was itself temporarily 

 absent. Losey (1971) observed similar behavior 

 among Hawaiian fishes. In such a situation the 

 hovering posture probably alerts the cleaner to 

 fishes that have arrived for cleaning. 



However, in cleaning activity involving the 

 sefiorita, the soliciting- posture usually is assumed 

 only after cleaning has been initiated by the 

 cleaner. The pi'oblem of recognizing an indi- 

 vidual that will clean among the vast majority 

 of seiioritas that do not clean, coupled with the 

 absence of well-defined cleaning stations, would 

 reduce the adaptiveness of the client's soliciting 

 posture as a cue to initiate cleaning. Probably 

 the most eflfective way for a fish to obtain needed 

 cleaning in this situation is to wait until a cleaner 



has identified itself by initiating activity with 

 some fish in the area. Once this has occurred, 

 one can see the value of the posture, when as- 

 sumed by the first fish to be approached, as a 

 cue in alerting other fish that need cleaning to 

 the presence of a cleaner. In eff'ect, then, the 

 fish assuming the soliciting posture advertises 

 the temporary existence of the transient cleaning 

 "station" to other potential recipients of clean- 

 ing. Well-defined cleaning stations like those 

 of Labroides phthirojihagus do not need this sort 

 of advertisement, as their locations are -well 

 known to the fishes that visit them. Nor is it 

 necessary that cleaning individuals of L. phthi- 

 rophagns be pointed out, as all members of that 

 distinctive species are cleaners. Despite this, 

 it is probable that fishes hovering to be cleaned 

 at a Labroides station themselves create a visual 

 cue that tends to attract other fishes. 



There may also be a maladaptive aspect to 

 the postures assumed by fishes that solicit clean- 

 ing. In hovering at an unnatural angle, fins im- 

 mobile and erect, a fish may enhance its chances 

 of being cleaned, but it would also seem likely 

 to draw the attention of predators and to handi- 

 cap itself in evading attack. Perhaps such an 

 increased vulnerability accounts at least in part 

 for the sharp decline in cleaning that occurs 

 with reduced visibility, when predators can ap- 

 proach closer undetected. Increased vulnera- 

 bility may also account for the observation 

 reported earlier (Hobson, 1965c), where poma- 

 dasyids in the Gulf of California abruptly broke 

 away from cleaners when a predator approached. 



THE POSSIBILITY THAT FISHES BEING 

 CLEANED EXPERIENCE STRESS 



Being prodded and picked over by an animal 

 of another species would seem to require a dif- 

 ficult adjustment for a fish. It may well be that 

 fishes experience stress under this circumstance, 

 even when the behavior is well established. Cer- 

 tainly observations have shown that this exper- 

 ience can be uncomfortable, judging from how 

 often fishes being cleaned suddenly bolt forward, 

 and swim away, apparently having been nipped 

 too vigorously by the attending cleaner. Some- 

 times too, a fish approached by a cleaner clearly 



517 



