摘要
Foundational concepts of trait spaces, including phenotypic plasticity and function of traits, should be expanded and better integrated with ecological theory. This article addresses two areas where plasticity theory can become further integrated with ecological, evolutionary, and developmental thinking. First is the idea that not only trait means within environments and plasticity of trait means across environments is optimized by selection, but that the entire shape of phenotype distributions such as variance or skew should be optimized within and across environments. In order for trait distribution shape to evolve into adaptations, there must be a genetic basis for and selection upon variation in distribution shapes and their plasticities. I present published and new data demonstrating genetic control and selection for higher moments of phenotype distributions; though, plasticity in these values has not yet been tested. Genetic control of phenotype distribution moments is shown for Neurospora crassa ascospore size and shape. Selection on trait distribution moments is shown for Eurosta solidaginis gall size. Second, there is a tradition in modeling plasti- city as an adaptive strategy that pits it as an alternative to ecological specialization or generaliza- tion. However, these strategies need not be considered alternatives. Rather, with environmental fluctuation within generations plasticity may produce additive or non-additive intermediate （gener- alist） phenotypes, or something new altogether. I present published and new data on the snail Physa virgata and fish Gambusia affinis that show plasticity produces partly intermediate （general- ist） and partly unique phenotypic elements in mixed and fluctuating environments. Plasticity can thus be viewed in the context of a broader trait space and as having broader ecological roles than currently is conceived.
