This data collection pertains to the feeding and body condition benefits of assisted territory defense exhibited among long-term pair bonds in two species of butterflyfishes (f. Chaetodontidae), Chaetodon lunulatus and C. baronessa. Four datasets are included in this collection: 1) swimming coordination and competitor agonism between male and female partners within pairs, 2) differences in agonism towards competitors, feeding bites, and access of coral prey between paired and solitary individuals, 3) differences in intra-pair relations, and per capita agonism towards competitors, and feeding bites between enduring and new pairs, and 4) differences in liver hepatocyte vacuolation between solitary, newly paired, and enduringly paired individuals.

The full methodology is available in the publication shown in the Related Publications link below.

Summary of methods for coordination and competitor agonism between male and female partners (dataset 1): Pair bonded individuals were haphazardly encountered on the reef, individuals were observed from a distance of 2-4-m and allowed 3-minutes to acclimate to observer presence. Thereafter, in situ observations were conducted for 6-minutes, to record pair coordination and agonism towards neighboring butterflyfish. Studies were conducted on snorkel between 08:30-17:30hr. Pair coordination was identified as the focal fish being positioned within a 2-m distance from its partner whilst being faced within a 315-45° angle relative the faced position of its partner (designated as 0°), and sampled once every ten-seconds. For each partner, rates of agonistic behavior were quantified as the total number of agonistic acts (staring, head down, tail-up display, chasing, fleeing, and encircling) towards other butterflyfish.

Summary of methods for differences in competitor agonism, bite rates, and access of coral prey between paired and solitary individuals (dataset 2): To test whether one or both sexes benefit from pairing by reduced competitor agonism, increased feeding rates or increased access to preferred coral prey, I compared these variables between naturally occurring paired and solitary individuals of both sexes. Pairs and singletons were haphazardly encountered, underwent 3-min acclimation to observer, and underwent a 6-min observation to record: i) total feeding bite rate, determined by the number of bites taken on any coral ii) total feeding bites on preferred coral types (data only collected for C. baronessa, whose preferred coral food is Acropora hyacinthus, A. florida, and Pocillopora damicornis, and iii) rates of agonism. Given that rates of agonism may be affected by the local densities of competitors (independent of levels of agonism exhibited by focal individuals), the number of agonistic acts recorded during replicate observations was standardized to total butterflyfish neighbors present. Variation in per capita food availability within territories of paired versus solitary conspecifics was estimated for both species. The territory boundary for each pair or individual was estimated by marking the exact position of the focal individual(s) at regular individual through the course of the observation. The outermost limits of where each individual or pair were located were then considered to roughly correspond with their territorial limits. To estimate percent coral cover within territories, I ran 2-4 replicate 2-metre point-intercept transects within the boundary, run from haphazardly selected starting points. In cases where estimated territory size was small, it was only possible to run a maximum of 2 replicate transects. For each replicate transect, I recorded the substrate (focussing on availability of different coral species) underlying each of 20 points per transect. Since food availability is theoretically shared equally between paired individuals, per capita availability of overall coral prey and preferred coral prey was determined by dividing total cover of these categories by two prior to analysis. After, individuals and sacrificed for sex determination and body condition analysis.

Summary of methods for differences in intra-pair relations, and per capita competitor agonism and feeding strikes between enduring and new pairs (dataset 3): I used a partner removal-replacement experiment to examine whether pair bond endurance reduces territory defense or increases feeding of paired individuals by promoting cooperative territory defense or reducing intra-pair conflict. Naturally occurring pair bonds of C. lunulatus (n = 9) and C. baronessa (n = 10) were identified among individuals on 3 adjacent sheltered reefs using methods previously described. Prior to experimentation, one individual from each pair was haphazardly chosen as the focal individual for the experiment. Unique body markings were used to identify the focal individual and its partners throughout the experiment. Behavioral expression of the focal individual while with its original partner was measured throughout an 8-min observation, for 5 consecutive days. Prior to each observation, the focal individual and its partner acclimated to observer presence (as described above). During each observation, time spent coordinatedly swimming with partner, agonism towards partner, agonism per competitor, and feeding bites of the focal individual were recorded using the methods previously described. Immediately following observations, the partner of the focal individual was removed via spearing and sacrificed in an ice slurry for sex determination and body condition analysis. Within 24 hours of experimental partner removal, all focal individuals had naturally re-paired with a new partner with whom they remain paired for the remainder of the study. The same observations were repeated for a further 7 consecutive days (in the case of C. lunulatus) or 8 days (in the case of C. baronessa). After experimentation, the focal individual and its new partner were collected and sacrificed in order to determine the sex of both individuals and body condition of the focal individual’s new partner.

Summary of methods for differences in liver hepatocyte vacuolation between solitary, newly paired, and enduringly paired individuals (dataset 4): To assess changes in energy reserves in association with pairing and partner endurance, I compared liver hepatocyte vacuole density between individuals who naturally occurred in solitude (liver specimens acquired from in situ observation study), who were in new pair bonds (C. lunulatus: 5 day old partnerships; C. baronessa: 7 day old partnerships), and who were in naturally occurring enduring pair bonds (liver specimens for the latter two conditions were acquired from individuals from partner removal experiment). Whole livers were dissected and fixed in 4% phosphate-buffered formalin (PBF). Fixed liver tissues were then dehydrated and embedded in paraffin. Tissues were sectioned at 5µm, mounted onto glass slides, and stained using Mayer's hematoxylin and eosin to emphasize hepatocyte vacuoles. Hepatocyte vacuole density was then quantified using a Weibel eyepiece to record the proportion of points (out of 121) that intersected with hepatocyte vacuoles when viewed at X 40 magnification. Three estimates of hepatocyte vacuolation were taken for each of 3 haphazardly chosen cross sections, totaling 9 replicate estimates per fish liver; from which the mean proportion of vacuoles in liver hepatic tissues of each fish was calculated.

This research complied with permits and ethics standards granted to JPN from James Cook University.