Sex change in the endemic Hawaiian labrid Thalassoma duperrey (Quoy and Gaimard) : a behavioral and ecological analysis

Typescript.Bibliography: leaves 166-171.Microfiche.xiv, 171 leaves, bound ill. 29 cmA field and laboratory study was conducted in Kaneohe Bay, Oahu, Hawaii, to determine the factors that cause sex change in the fish Thalassoma duperrey, a protogynous, diandric hermaphrodite. Unlike other sequential hermaphrodites that have been studied thus far, this species is non-haremic and non-schooling. By nature of its unique social organization, it could have a special sex-change mechanism. The home ranges of T. duperrey overlap extensively and are distributed randomly or uniformly over the reef. Home-range size and location do not differ between the sexes. Size-related aggressive dominance controls feeding interactions. Sexual maturity is reached at about 60 mm standard length, less than 1 year after fertilization. Gonad development (gonosomatic index, percent gonad weight of body weight) in both males and females peaks at about 120 mm, nearly 2 years later. Terminal-phase coloration typically appears soon thereafter, accompanied by as much as a 10-fold drop in gonosomatic index. This drop is correlated with a change in spawning mode in males from group to pair spawning, which apparently requires less sperm production. Sex change in females occurs at a mean size of 125 mm, 2 years after sexual maturation. Most females probably change sex in late spring or early summer, after the height of the reproductive season. Transitional gonad stages are similar in structure to those reported in other wrasses. A major change in the reproductive tract is the formation of the sperm duct within and between the urogenital papilla and the oviduct. Anatomical and behavioral evidence points to the existence of three cyclic components in the timing of reproduction: annual, semilunar, and diel. On an annual basis, both percent spawnable ova and number of females spawning are greatest in the winter (December). Over the lunar cycle, the same parameters are greatest during the new and full moon and least during the quarter moons. On a daily basis, they are greatest at the time of high tide during daylight hours. The adaptive significance of having a winter spawning maximum is unknown. Spawning on high spring tides, however, may be an anti-egg-predator or dispersal strategy, or both. Reproductive behavior includes the occurrence of lekking, looping, circling, pectoral fluttering, quivering, and group and pair spawning. Spawning is promiscuous though some females spawn with the same TP male repeatedly. All fish are able to choose spawning partners and initial-phase fish frequently migrate as far as half the reef perimeter to do so. Females spawn at most once a day, but large males pair spawn with as many as 20 or more females per day. This relationship supports the size-advantage model for the evolution of sequential hermaphroditism. A comparison of several population parameters between a small and a large patch reef showed that only fish size differed between reefs (the large reef had larger individuals). This comparison did not support the hypothesis that local population size determines the sexual composition of populations on isolated reefs. An interspecific comparison based on population size and density did support the hypothesis. Both population size and fish density were directly related to the percent of primary males present and inversely related to the percent of terminal-phase males present. To determine the proximate causes of individual sex change, experiments were conducted in outdoor pens or tanks at the Hawaii Institute of Marine Biology. Adult wrasses were subjected to various social regimes using size, sex, coloration (phase), density, and type of isolation from conspecifics as independent variables. Single isolated females did not change sex while the larger individuals of paired females did, regardless of absolute size. These results were independent of the amount of space available to the fish. Females paired with smaller initial-phase males also changed sex, but those paired with smaller heterospecific wrasses did not. A tactile barrier separating two females did not prevent sex change in the larger fish. However, a tactile-visual barrier placed between two females significantly reduced the percentage of sex changes. Sex change did not occur when only olfactory cues were available but did occur when only visual cues were available. The presence of an initial-phase or terminal-phase male did not suppress sex change in larger female wrasses. In the case of female threesomes, only the largest female changed sex, demonstrating inhibition of sex change in the intermediate-sized fish. These results show that sex change is socially controlled in T. duperrey. It is dependent upon stimulation from smaller conspecifics and is inhibited in the presence of larger conspecific females. The stimulation-inhibition process suggests that, in nature, the ratio of the number of smaller fish to larger fish would actually control the initiation of sex change in candidate females. The ratio of the number of behavioral interactions with smaller fish to that with larger fish would be the equivalent behavioral mechanism. Either mechanism would function to predict the number of females available as mates for potential sex changers. The precise role of behavior in this mechanism will require further experimentation