Interrelationships between ecology, behaviour and life history in the protogynous hermaphrodite Pseudolabrus celidotus (Pisces: Labridae)

Abstract

The protogynous hermaphrodite Pseudolabrus celidotus was studied for four years in shallow temperate reef habitats of north-eastern New Zealand. The purpose was to determine the major factors influencing its abundance and sex structure, which required an examination of the factors operating within and between habitats, and within and between generations. Significant differences were found in juvenile recruitment and breeding densities between habitats. Within these habitats, recruitment was consistent between years and breeding densities were relatively stable over time. Experimental adult removals suggested that recruitment and the growth and survival of juveniles was independent of adult density. Breeding densities increased as a function of recruitment, however, a combination of density dependent growth and mortality within the juvenile cohort(s) led to a ceiling breeding density beyond a certain level of input. Recruitment was positively related to the biomass of algae in a habitat and the density of prey. Density dependent growth and mortality were most likely to be due to intra-cohort competition for food over the immature size range. This was supported by juvenile exclusion and removal experiments, which suggested that prey numbers were sensitive to juvenile feeding. The distribution of individuals within habitats changed as a function of fish size. Juveniles were influenced by the distribution of shelter, whereas adult distribution appeared to relate more closely to food availability.? Juveniles formed loose schools associated with patches of algae, feeding mainly on small algal-dwelling gammarid amphipods. Growth and decreased dependence on shelter were accompanied by a switch to feeding on rock flat substrata. Corresponding changes in diet and an increase in home range size were also observed. Interspecific territoriality by the blennioid Forsterygion varium had a considerable influence on the feeding localities of intermediate sized P. celidotus. During the spawning season females made short excursions away from their normal feeding ranges to spawn at specific sites - relatively deep bare rock areas on the seaward side of boulders. Male removal experiments indicated that features of the site rather than the quality of the male defending the site, determined whether spawning occurred there. It is suggested that choice of spawning sites is influenced by the distribution of planktivorous fish and the distribution of shelter sites. Both initial phase (IP) and terminal phase (TP) males tended to be found in preferred spawning areas rather than where females were most abundant. Although the territorial system was highly variable, the preferred spawning sites were defended by TP males all year round. Those defending the highest quality sites accounted for the large majority of the spawnings observed. A relatively large number of nonbreeding TP males, which rapidly occupied favourable sites when residents were removed, indicated intense competition between males for these sites. A series of field experiments suggested that most of the life history events in P. celidotus were influenced by social factors, which controlled the relative number of individuals in each sexual category within a habitat. Mature females appeared to cause a delay in the maturation of juvenile females, apparently through an increase in inter-female aggression at the onset of the spawning season. Continuous TP male-removals increased the rates of sex colour change by normal protogynous hermaphrodites and colour change by IP males. TP males were sensitive to local breeding conditions. Removal of TP males appeared to increase the rate of TP male immigration. When TP male density was increased, territorial behaviour was abandoned and males gradually emigrated. Emigration of these males was also observed from an area where all IP fish were removed. Low and high juvenile recruitment populations exhibited marked differences in behaviour and sex/colour change schedules. A positive correlation between the proportion of the initial phase that was male and breeding density suggested environmental plasticity in the choice between the two life history strategies and a tendency toward secondary gonochorism with increased density.