Cleaning time vs Size and Family

Time Labroides dimidiatus Spends Cleaning Depending on the Size and Family of the Client

Antoine Baud, Mélody Busuttil, Lauréline Humair, Melvyn Staiff

Introduction

Background

Cooperation between species is a very interesting topic. In humans for instance, we developed a complex language system to create an event that could lead to cooperation. How can several species ‘communicate’ well enough to start a cooperation event? Here, we are interested in the mutualistic relationship.

We use the term ‘mutualistic’ when individuals interact and both have an immediate benefit out of this interaction. This is the case for the interaction between the Bluestreak Cleaner Wrasse (Labroides dimidiatus) and its clients, other reef fish. The cleaner wrasse will offer a cleaning session to the fish in need. The cleaner will eat the parasites off the client’s skin.

Mutualistic interactions between cleaner fish and coral reef fish remain one of the most interesting interspecific interaction, especially when considering the variation in cleaning quality: How does a cleaner decides how it should clean?

This question has already peaked the interest of several researchers and we now know some influencing factors: the parasite load of the client (Grutter, 1995; Soares et al., 2008); client residency (Bshary & Schaffer, 2002); hormones - Arginine, Vasotocin (Soares et al. 2012); reputation/loyalty management (Grutter, 1995; Binning et al., 2017).

However, studies have failed to show a correlation between the client species and the duration of the cleaning interaction (Gorlick, 1980; Grutter, 1995). These studies focused on species, not taking into account the prevalence of families, the number of species may alter the results.

Another study showed that some fish genera are preferred by cleaners or spend more time with cleaners because of lineage characteristics and not because of body characteristics (Grutter & Poulin, 1998).

In this study, we focused on the relationship of the client service quality by comparing the time of cleaning inspection with the client fish size and family.

Questions & Hypothesis

Is the duration of a cleaner-client interactions related to the client size and/or the client family?

In this study we want to find whether a cleaner takes more time (prolongs interaction) on bigger clients and if the cleaning time could depend of the client family. The hypothesis is therefore that the cleaner will provide better quality (represented here by interaction duration standardised with client size) cleaning on a bigger client with a specific family.

Response: cleaning time (time in seconds of interaction, continuous)

Explanatory factor: client size (categorical: small, medium, large)

Random factor: client family (categorical)

Random factor: cleaner identity (each individual cleaner can behave differently)

Predictions:

The cleaner will take more time on bigger fish

Interactions with more than one cleaner on the same client were excluded from the analysis, focusing the model on single cleaner-client interactions.

Material and Methods

Study site & Subject

We collected data in April 2017 at Kilo 14 Safaga Road, Abu Sauatir, Hamawein, Red Sea, Egypt (26°12’12.812” N, 34°13’3.608” E). Cleaning stations varied between 1 to 10 meters. Cleaner observed at each site was bluestreak cleaner wrasse (Labroides dimidiatus). The client species varied and we paid attention to its size and family.

Figure 1: Bluestreak cleaner wrasse during cleaning session.

Data collection

Our first dive was to locate the different cleaning stations among the reef. Then each diver observe at each dive a different cleaner for about 10 to 20 minutes. So 2 to 3 cleaner fishes were filmed at each dive by each divers. We recorded each interaction in the cleaning station in second from first to last contact. In total 18 different bluestreak cleaner wrasses were recorded. The duration of each inspection was scored not the frequency of inspection (Bshary R. 2002). We paid attention to differentiate each cleaner fish regarding to the filmed cleaning station. We looked for client species helped by the book of Ewald Lieske and Robert F. Myers; Coral Reef Guide Red Sea. We then define the client size as:

Small (S) : Equal or shorter than the cleaner
Medium (M) : Up to 3 times the cleaner size
Large (L) : More than 3 times the cleaner size

Statistics

We used a mixed-effect model to test whether the time of an interaction (in seconds) could be explained by the size of a client (in three categories: small, medium and large). The duration of interactions were normalized with a log transformation and standardized with a standard deviation of 1 around a mean of 0. We also added the client family and the cleaner identity as random factors, because each cleaner may have its own individual behaviour. We performed a model comparison to verify whether the random factors are necessary in the model. We compared the models using the Akaike’s An Information Criterion (AIC). The data was analyzed with R software version 3.3.3.

Here is the full mixed-effect model:

interaction time ~ client size + 1|cleaner identity + 1|client family

Results

The model comparison (Table 1) reveals that the best model (with lowest AIC) is the full model (AIC=372.6377). This suggests that the random factors have an effect on the interaction time. Table 2 shows a summary of the mixed-effect model. Even though the random factors have an effect on the response variable, there is a significant difference in interaction duration between the small (S) and large (L) clients (p-value=8.32e-05). The interactions with medium clients have a large range of durations, similar to both the large and the small clients (Fig. 2).

Table 1. Model comparison showing different models and their respective AICs.

Table 2. Summary of the fixed factor from the full mixed-effect model.

Interaction time according to client size

Figure 2: Boxplot representing the differences in

the duration of a cleaning interaction according to

the client size category (L = large, M = medium, S = small).

We found that the body size is related to the duration of inspection, however we found no relationship between the client fish family and the interaction. These results follow those found by Grutter (1995) suggesting that body size is related to the ectoparasite load. Cleaners spent more time with larger clients. This may be a strategic choice from the cleaner choosing to inspect larger fish in order to obtain more food from client fish (Grutter, Glover & Bshary, 2005).

In a future study, it could be interesting to focus on the correlation between the time of inspection and the shape of clients or parasite composition. Shapes and parasites may vary between client fish families. During the data collection we observed pairs of cooperating cleaner fish. Studies showed that this kind of cooperating increase the quality of the service (Bshary et al., 2008). We have to keep in mind that the divers presence could influence the interaction duration. Indeed, it is possible that we frightened some clients or cleaners with our presence.

There is a clear difference in interaction time between small and large clients. This study used only the client length, not the various shapes of the clients. The shape affects the surface area on which cleaning interactions take place. However, continuous data on the client size rather than categories (L, M, S) could reveal a stronger relationship.

Binning S.A., Rey O., Wismer S., et al., (2017). Reputation management promotes strategic adjustment of service quality in cleaner wrasse. Scientific Reports, 7, 8425.

Bshary R. & Schaffer D. (2002) Choosy reef fish select cleaner fish that provide high quality service. Animal Behaviour, 63: 557-564.

Bshary R., Grutter A.S., Willener A.S.T. & Leimar O. (2008) Pairs of cooperating cleaner fish provide better service quality than singletons. Nature, 455: doi:10.1038.

Gorlick D.L. (1980) Ingestion of host fish surface mucus by the Hawaiian cleaning wrasse, Labroides phthirophagus (Labridae), and its effect on host species preference. Copeia, 1980:863-868.

Grutter A.S. (1995) Relationship between cleaning rates and ectoparasite loads in coral reef fishes. Mar. Ecol. Prog. Ser. 118: 51-58.

Grutter A.S. & Poulin R. (1998) Cleaning of Coral Reef Fishes by the Wrasse Labroides dimidiatus: Influence of Client Body Size and Phylogeny. Copeia, 1998(1): 120-127.

Grutter A.S., Glover S. & Bshary R. (2005) Does client size affect cleaner fish choice of client? An empirical test using client fish models. Journal of Fish Biology, 66: 1748-1752.

Lieske E. & Myers R.F. (2004) Coral Reef Guide Red Sea. Collins, ISBN 10: 0007159862.

Soares M.C, Bshary R. & Côté I.M (2008) Does cleaner fish service quality depend on client value or choice options? Animal Behaviour, 76(1): 123-130.

Soares M.C., Bshary R., Mendonça R., Grutter A.S., Oliveira R.F., (2012). Arginine Vasotocin Regulation of Interspecific Cooperative Behaviour in a Cleaner Fish. PLoS ONE, 7(7), e39583.

Discussion

Conclusion

References