Interpretation of the results of common principal components analyses

Common principal components (CPC) analysis is a new tool for the comparison of phenotypic and genetic variance-covariance matrices. CPC was developed as a method of data summarization, but frequently biologists would like to use the method to detect analogous patterns of trait correlation in multiple populations or species. To investigate the properties of CPC, we simulated data that reflect a set of causal factors. The CPC method performs as expected from a statistical point of view, but often gives results that are contrary to biological intuition. In general, CPC tends to underestimate the degree of structure that matrices share. Differences of trait variances and covariances due to a difference in a single causal factor in two otherwise identically structured datasets often cause CPC to declare the two datasets unrelated. Conversely, CPC could identify datasets as having the same structure when causal factors are different. Reordering of vectors before analysis can aid in the detection of patterns. We urge caution in the biological interpretation of CPC analysis results.     

HIERARCHICAL COMPARISON OF GENETIC VARIANCE-COVARIANCE MATRICES. I. USING THE FLURY HIERARCHY

The comparison of additive genetic variance-covariance matrices (G-matrices) is an increasingly popular exercise in evolutionary biology because the evolution of the G-matrix is central to the issue of persistence of genetic constraints and to the use of dynamic models in an evolutionary time frame. The comparison of G-matrices is a nontrivial statistical problem because family structure induces nonindependence among the elements in each matrix. Past solutions to the problem of G-matrix comparison have dealt with this problem, with varying success, but have tested a single null hypothesis (matrix equality or matrix dissimilarity). Because matrices can differ in many ways, several hypotheses are of interest in matrix comparisons. Flury (1988) has provided an approach to matrix comparison in which a variety of hypotheses are tested, including the two extreme hypotheses prevalent in the evolutionary literature. The hypotheses are arranged in a hierarchy and involve comparisons of both the principal components (eigenvectors) and eigenvalues of the matrix. We adapt Flury's hierarchy of tests to the problem of comparing G-matrices by using randomization testing to account for nonindependence induced by family structure. Software has been developed for carrying out this analysis for both genetic and phenotypic data. The method is illustrated with a garter snake test case.     

Influences of microhabitat use and foraging mode similarities on intra- and interspecific aggressive interactions in a size-structured stream fish assemblage

Aggressive interactions, foraging behaviour and microhabitat use were observed among four sympatric stream fishes inhabiting the water column: ayu (Plecoglossus altivelis), white-spotted charr (Salvelinus leucomaenis), masu salmon (Oncorhynchus masou) and Japanese dace (Tribolodon hakonensis), each species being categorised into five body-size classes (species-size groups; SSG's). Aggressive interactions were observed between most pairs of SSG's, an almost linear dominance order being apparent throughout the three-month study period. Ayu were relatively subordinate in June, but became the second most dominant in July and the most dominant in August, as a consequence of a reversal in dominance order with salmon. In contrast, smaller-sized dace, which continually suffered from intra- and interspecific aggression, occupied the most subordinate ranks throughout the study period. Intensive aggression was observed among various SSG's, exhibiting same microhabitat propensity throughout the three months. The direction and frequency of aggressive interactions varied month by month due to a reversal in dominance order between ayu and masu salmon, and/or changes in density, body size and resource use of the component members. Opponent selectivity was higher within SSG's, where resource use was assumed to be highly overlapping, rather than among SSG's throughout the study period. Correlation analysis indicated that opponent selectivity in aggressive interactions among SSG's was positively correlated with similarity in microhabitat selectivity in June, but not in other months or with that in foraging habits, suggesting that intensive aggressive behaviour reflected overlapping habitat use among assemblage members during a certain period.     

Temporal Hierarchies of Variation and the Maintenance of Diversity

Abstract A general model shows how the long-term growth rate of a population can be partitioned into components representing various mechanisms of maintenance of species diversity. One component summarises the effects of fluctuation-independent mechanisms, which include classical resource partitioning and frequency-dependent herbivory. Two other components represent fluctuation-dependent mechanisms, the storage effect and relative nonlinearity of competition.
The general model shows how a community will track an equilibrium set by fluctuation-independent mechanisms and the environmental state when community dynamics are faster than the rate of environmental change. Fluctuation-dependent mechanisms can be important for diversity maintenance with or without such tracking, but on long timescales their effects are indistinguishable from those of fluctuation-independent mechanisms.
These considerations lead to a hierarchical view of mechanisms of diversity maintenance where the effects of different timescales are partitioned or merged depending on the timescale of observation. These issues are illustrated with model examples involving various combinations of resource partitioning, fluctuations in recruitment rates, variation in the timing of germination, and seasonality. The very long timescales associated with climate change contain many complexities but nevertheless many ideas applicable to shorter timescales may be useful in a modified form.     

Comparative riverscape genomics of the rainbow darter (Etheostoma caeruleum) in glaciated and unglaciated environments

Periodic glaciation during the Quaternary period shaped the contemporary riverscape and distribution of freshwater fishes in the Mississippi River drainage of central North America. The rainbow darter (Etheostoma caeruleum) is a member of this ichthyofauna and has a disjunct distribution in glaciated and unglaciated environments west of the Mississippi River. Based on glacial history of the region, there are different expectations on the observed spatial genetic structure of populations in these environments. The aim of this study was to utilize genome‐wide SNP data to compare the population genomic structure of the rainbow darter in river networks with disparate glacial histories; the Volga River in the glaciated upper Mississippi River basin and the Meramec River in the unglaciated Ozark Plateau. Individuals were sampled from localities within each river system at distances dictated by the organismal life history and habitat preferences. Riverscape analyses were performed on three datasets: total combined localities of both rivers and one for each river independently. The results revealed a lasting influence of historic glaciation on the population genomic structure of rainbow darter populations. There was evidence of population expansion into the glaciated northern region following glacial retreat. The population genetic signature within the Volga River did not fit expectations of the stream hierarchy model, but revealed a pattern of repeated colonization and extirpation due to cyclic glaciation. The population within the unglaciated Meramec River adhered to the stream hierarchy model, with a directional order of genetic diversity based on the life history and habitat preferences of the species. These results demonstrate the importance of considering the geologic and climatic history of a region as well as the life history of an organism when interpreting spatial genetic patterns.     

Importance of Body Size in Determining Dominance Hierarchies among Diverse Tropical Frugivores1

Most studies examining dominance hierarchies have focused at the intraspecific level. While some examples of interspecific hierarchies have been noted, these have usually been limited to a few species in the same taxonomic group that utilize resources in similar ways. Here, we examine evidence for dominance interference competition among vertebrates comprising a diverse frugivore community, including 19 species of birds, squirrels, and primates in a mature Central African rainforest. A total of 38 fruiting trees from 18 species were observed for 2058 h to record dominance interactions between foraging vertebrates. We show that interference competition occurs within and between taxonomically diverse species of vertebrates at fruiting trees. The resulting cross‐taxonomic dominance hierarchy includes larger vertebrates, such as primates and hornbills, as well as smaller ones, such as squirrels and parrots. Within this hierarchy, the dominance rank of each species is highly correlated with body mass, and is shown to significantly affect the number of fruits removed from a given tree. Because a majority of tropical tree species depend on vertebrates to disperse their seeds, and particular vertebrates may preferentially disperse the seeds of specific tree species, results may have important conservation implications for the maintenance of tree diversity in regions where populations of larger frugivores have been depressed or extirpated.     

Short-term dynamics of an acacia ant community in Laikipia,Kenya

In monospecific stands of Acacia drepanolobium in Laikipia, Kenya, virtually all but the smallest trees are occupied by one of four species of ants. Although trees are a limiting resource, all four ant species are maintained in this system. Three separate lines of evidence confirm a linear dominance hierarchy among these four ants: (1) experimentally staged conflicts, (2) natural transitions among 1773 tagged trees over a 6-month period, and (3) the average sizes of trees occupied by ants of different species. Short-term dynamics during a drying period reveal that many smaller trees (<1 m) occupied by dominant ants were subsequently abandoned, and that abandoned trees had grown more slowly than those that were not abandoned. Height growth increments over 6 months were generally independent of ant occupant, but increased with tree height. Among taller trees (>1 m), changes in ant occupation congruent with the dominance hierarchy (i.e., transitions from more subordinate ant species to more dominant ant species) occurred on trees that grew faster than average. In contrast, the (less frequent) changes in ant occupation ”against” the direction of the dominance hierarchy occurred on trees that grew more slowly than average. Observed correlations between tree vigor and takeover direction suggest that colony growth of dominant ant species is either favored in more productive microhabitats, or that such colonies differentially seek out healthier trees for conquest. Colonies of dominant species may differentially abandon more slowly growing trees during (dry) periods of retrenchment, or suffer higher mortality on these trees. Subordinate ant species appear to move onto these abandoned trees and, to a lesser extent, colonize new recruits in the sapling class. These data reveal that within a simple linear dominance hierarchy, short-term variations exist that may reveal underlying mechanisms associated with coexistence. Received: 24 June 1999 / Accepted: 3 December 1999