Many evolutionary arguments are based on the assumption that quantitative characters are highly evolvable entities that can be rapidly moulded by changing selection pressures. The empirical evaluation of this assumption depends on having an operational measure of evolvability that reflects the ability of a trait to respond to a given external selection pressure. We suggest short-term evolvability be measured as expected proportional response in a trait to a unit strength of directional selection, where strength of selection is defined independently of character variation and in units of the strength of selection on fitness itself. We show that the additive genetic variance scaled by the square of the trait mean, I A , is such a measure. The heritability, h 2 , does not measure evolvability in this sense. Based on a diallel analysis, we use I A to assess the evolvability of floral characters in a population of the neotropical vine Dalechampia scandens (Euphorbiaceae). Although we are able to demonstrate that there is additive genetic variation in a number of floral traits, we also find that most of the traits are not expected to change by more than a fraction of a percent per generation. We provide evidence that the degree of among-population divergence of traits is related to their predicted evolvabilities, but not to their heritabilities.

Intra- and interspecific variation in flower color is a hallmark of angiosperm diversity. The evolutionary forces underlying the variety of flower colors can be nearly as diverse as the colors themselves. In addition to pollinator preferences, non-pollinator agents of selection can have a major influence on the evolution of flower color polymorphisms, especially when the pigments in question are also expressed in vegetative tissues. In such cases, identifying the target(s) of selection starts with determining the biochemical and molecular basis for the flower color variation and examining any pleiotropic effects manifested in vegetative tissues. Herein, we describe a widespread purple-white flower color polymorphism in the mustard Parrya nudicaulis spanning Alaska. The frequency of white-flowered individuals increases with increasing growing-season temperature, consistent with the role of anthocyanin pigments in stress tolerance. White petals fail to produce the stress responsive flavonoid intermediates in the anthocyanin biosynthetic pathway (ABP), suggesting an early pathway blockage. Petal cDNA sequences did not reveal blockages in any of the eight enzyme-coding genes in white-flowered individuals, nor any color differentiating SNPs. A qRT-PCR analysis of white petals identified a 24-fold reduction in chalcone synthase (CHS) at the threshold of the ABP, but no change in CHS expression in leaves and sepals. This arctic species has avoided the deleterious effects associated with the loss of flavonoid intermediates in vegetative tissues by decoupling CHS expression in petals and leaves, yet the correlation of flower color and climate suggests that the loss of flavonoids in the petals alone may affect the tolerance of white-flowered individuals to colder environments.

The short-term evolvability of a character is closely related to its level of additive genetic variation. However, a large component of the variation in any one character may be pleiotropically linked to other characters under the influence of different selective factors. Therefore, the organization of the organism into quasi-independent modules may be an important prerequisite for evolvability. In this paper we propose to study character evolvability in terms of conditional genetic variation. By estimating the amount of genetic variation in a character, y, that is independent of other characters, x, we can assess the evolvability of y when there is stabilizing selection on x. We suggest that systematic use of conditioning may help build a picture of modular organization and quasi-independent evolvability. As an illustration, we use this approach to assess the evolvability of floral characters in Dalechampia scandens (Euphorbiaceae). Although our study population had relatively low levels of genetic variation at the outset, we find evidence that conditioning may lead to substantial further reduction in the genetic variation available for independent adaptation. This provides additional evidence that the D. scandens blossom is constrained in its short-term evolvability.

Summary1. We studied vascular plant and soil-dwelling testate amoeba communities in deglaciated sites across a range of substrate ages in Kenai Fjords, Alaska, USA to test four hypotheses. (i) Patterns of community assembly are similar for vascular plants and testate amoebae. (ii) Vascular plant and testate amoeba communities are more strongly correlated to abiotic variables than to each other, since these communities are not directly linked trophically. (iii) Plant community structure becomes less associated with abiotic condition in succession relative to testate amoebae, as species replacement is believed to be more common for plants than testate amoebae. (iv) Above-and belowground communities become more strongly linked over the succession, due a shift from predominantly allogenic to autogenic forces. 2. We assessed relationships among biotic communities and abiotic site variables across the chronosequence using multiple factor analysis, redundancy analysis (RDA) and a moving-window analysis. 3. The diversity patterns and the communities' response to site and soil variables differed between groups. The composition of both communities was significantly explained by bedrock type and moisture regime. The vascular plant community, however, was more influenced by distance from the glacier. 4. Testate amoeba and vascular plant community patterns were significantly linked to each other and to location and physical conditions. The moving-window RDA indicates the variation explained by the physical and chemical environment tended to slightly decrease through the chronosequence for testate amoebae, while a bell-shape response was evidenced for vascular plants. The variation of the microbial community explained by the plant community was very low in the early stages of the succession and became higher than the variation explained by the environmental variables later in the chronosequence. 5. Synthesis. These results suggest that vascular plants and testate amoebae are as linked or more in ecosystem development than either community is to changes in site condition. Furthermore, the strength of interactions varies along the succession. Thus, ecological links may be more important than macro-scale abiotic site condition is to community development, even between communities without direct trophic interactions.