To breed or not to breed—food competition and other factors involved in female breeding decisions in the pair-living nocturnal fork-marked lemur (<Emphasis Type="Italic">Phaner furcifer</Emphasis>)

Differential access to food resources is thought to be the main determinant of differences in female reproductive success but is poorly studied in both pair-living and nocturnal primates. The modes of food competition within and between families were investigated following the principles proposed by the ecological model using 3 years of field data from seven fork-marked lemur (Phaner furcifer) families. The major food resources were identified from year-round feeding observations and the strength and mode of competition were inferred from differences in physical condition. The most important food resource of fork-marked lemurs were tree exudates which occurred in small, defendable food patches, characterized by fast depletion and rapid renewal. These characteristics led to strong within-group contest and scramble competition, which were found to yield a positive dominance effect and a negative group-size effect on female net energy gain. Differential physical condition, however, did not translate directly into differential reproductive success. Low female fertility was best predicted by large family size associated with delayed dispersal by previous offspring. Although there is no obvious benefit from full-grown offspring in their territory, adults tolerate delayed natal dispersal, probably because dispersal poses extraordinary costs for the offspring. These costs are likely to accrue from decreased foraging efficiency in unfamiliar habitats because exudate feeding requires very rigid feeding itineraries. In conclusion, the presented evidence for group-size effects on reproductive success in pair-living females opens a new area for research on the costs and benefits of delayed dispersal and female reproductive decisions.This revised version was published in September 2003 with corrections to the Authors Present address.An erratum to this article can be found at Communicated by E.H.M. Sterck     

Do changes in risk-taking affect habitat shifts of sticklebacks?

The distribution of individuals is often the outcome of conflicting demands, such as between predator avoidance and reproduction. A factor that has seldom been considered in studies on habitat choice is time-dependent changes in risk-taking. We investigated the distribution of threespine sticklebacks, Gasterosteus aculeatus, over two breeding seasons and found it to change with time towards shallower areas with a more open habitat structure. Shallow and structurally less complex habitats were probably favorable due to a higher reproductive rate, but costly due to an increased risk of predation. Contrary to expectation, changing predation pressure was not a predictor of the shift in habitat use and, thus, not the proximate cue. Instead date was the main predictor. This suggests that increased risk-taking in relation to predation contributed to the habitat shift. The possibility was supported by a laboratory experiment that showed sticklebacks to take larger risks and prefer more predator-exposed areas at the end of the season than at the start of the season. These results demonstrate that temporal changes in risk-taking occur and can influence habitat choice, which points to the importance of considering risk-taking, in addition to predation pressure, when studying the effect of predators on distribution.Communicated by J. Krause     

Mate choice in the brown thornbill (<Emphasis Type="Italic">Acanthiza pusilla</Emphasis>): are settlement decisions,divorce and extrapair mating complementary strategies?

In socially monogamous birds, females may express mate preferences when they first select a breeding partner, through divorce and subsequent breeding dispersal to a new partner and through extrapair mating. We examined settlement patterns, divorce and breeding dispersal in a sedentary Australian passerine, the brown thornbill (Acanthiza pusilla), in relation to two traits known to influence extrapair paternity (male age and male size). Settlement decisions, divorce and territory switching behaviour were all female strategies that reduced their likelihood of breeding with 1-year-old males. Females preferred to settle in territories with 2+ -year-old males, were more likely to divorce 1-year-old males, and only switched territories if they had an opportunity to form a new pair bond with an old male. In contrast, female settlement and divorce decisions were not influenced by male size. Female thornbills obtain a direct benefit from preferring older males as social mates because breeding success improves with male age in brown thornbills. Nevertheless, divorce rates in this species were low (14% of pair bonds were terminated by divorce), and individuals rarely switched territories following the death of a mate. Both of these mating strategies appeared to be primarily constrained by the distance adults moved to initiate a new pair bond (1–2 territories) and by the limited availability of unpaired older males in the immediate neighbourhood.Communicated by M. Webster     

Population genetic structure of the milkfish,<Emphasis Type="Italic"> Chanos chanos</Emphasis>, based on PCR-RFLP analysis of the mitochondrial control region

The milkfish, Chanos chanos (Forsskål, 1775) is a pelagic, monotypic gonorhynchiform widely distributed in the tropical Indo-Pacific. This study evaluates temporal variability of milkfish samples from the Philippine archipelago, and spatial variability at two geographic scales based on restriction fragment length polymorphism (RFLP) analysis of a portion of the mitochondrial control region. High levels of genetic diversity characterize the milkfish control region (mean h=0.908, =1.59%), with 74 haplotypes detected among the 367 fish analyzed. For temporal analysis of Philippine samples, milkfish were collected over 2 years from three sites (inter-annual variation), and sampled twice within a year during different seasons at four sites (intra-annual variation). No significant temporal variability was detected between or within years. Significant spatial differentiation among the Philippine samples was observed (F_ST=0.006, P<0.05), with two northeastern samples, Claveria and Dingalan, found to be genetically distinct. However, an hierarchical analysis of molecular variance (AMOVA), where samples were grouped into four geographic regions, revealed very low levels of genetic partitioning, with less than 1% of the total variation attributed to between-region differences, and lack of genetic structure. Nonetheless, the existence of putative northeastern Philippine populations is not discounted. Strong genetic structure across broad geographical scales was revealed by AMOVA, with 11% of the molecular variance based on haplotype frequencies allocated between three distinct groups: Indian Ocean, west Pacific (Philippines) and north central Pacific (Hawaii) The broad-scale genetic structure points to limited gene flow among disjunct Indo-Pacific populations.Communicated by T. Ikeda, Hakodate