SUMMARYAllocation trade-offs of limited resources are thought to ensure the honesty of sexual signals and are often studied using controlled immune challenges. One such trade-off between immunity and ornaments is that involving carotenoids. Phytohemagglutinin (PHA)-induced immune response is a widely used immune challenge, yet more details on the underlying physiological mechanisms and potential costs are needed. We investigated the temporal dynamics of PHA-induced immune response and associated changes in blood carotenoids, body mass and a carotenoid-based coloured signal. We found variation in individual response patterns to PHA after peak swelling was reached, with birds showing either a rapid or a slow subsequent decrease in swelling, suggesting variation in the duration of the immune response and/or inflammation. Body mass did not affect immune response. Plasma carotenoids followed a transient decrease closely matching the dynamics of the swelling. The peak of the immune response was negatively related to initial plasma carotenoid levels and positively correlated to the relative decrease in plasma carotenoids. Individual variation in duration of the swelling could be partly explained by plasma carotenoids; high initial carotenoid levels were associated with a slower decrease of the swelling. These contradictory effects of carotenoids suggest a complex role in the immune response. Bill colour was positively correlated to initial plasma carotenoid concentration but it did not predict or change as a consequence of immune response to PHA. Bill colour thus reflects medium-or long-term quality rather than immediate quality. Taking into account the dynamics of the immune response and that of associated physiological parameters would thus yield new insights into our interpretation of variation in PHA response.

Conservation biologists and palaeontologists are increasingly investigating the phylogenetic distribution of extinctions and its evolutionary consequences. However, the dearth of palaeontological studies on that subject and the lack of methodological consensus hamper our understanding of that major evolutionary phenomenon. Here we address this issue by (i) reviewing the approaches used to quantify the phylogenetic selectivity of extinctions and extinction risks; (ii) investigating with a high-resolution dataset whether extinctions and survivals were phylogenetically clustered among early Pliensbachian (Early Jurassic) ammonites; (iii) exploring the phylogenetic and temporal maintenance of this signal. We found that ammonite extinctions were significantly clumped phylogenetically, a pattern that prevailed throughout the 6.6 Myr-long early Pliensbachian interval. Such a phylogenetic conservatism did not alter – or may even have promoted – the evolutionary success of this major cephalopod clade. However, the comparison of phylogenetic autocorrelation among studies remains problematic because the notion of phylogenetic conservatism is scale-dependent and the intensity of the signal is sensitive to temporal resolution. We recommend a combined use of Moran's I, Pearson's ϕ and Fritz and Purvis' D statistics because they highlight different facets of the phylogenetic pattern of extinctions and/or survivals.