An Evolutionary Approach to Understanding the Biology of Invasions: Local Adaptation and General-Purpose Genotypes in the Weed Verbascum thapsus

The role of evolution in the invasion of non-native species has important implications for conservation, weed science, risk assessment, and policy. In this paper we first discuss why an evolutionary perspective can be helpful and outline a range of potentially useful approaches from population biology and ecological genetics. As a case study, we then ask how adaptation and genetic structure may promote or constrain the expansion of an invasive weed, Verbascum thapsus , into high elevations in the Sierra Nevada of California. We used growth-chamber and common-garden experiments to assay a range of morphological and physiological traits that could influence plant fitness at high elevation. There was a significant relationship between elevation and leaf size and reflectance above 800 m, although we found no relationship between elevation and most other traits ( growth rate in warm or cool temperatures, freezing tolerance when grown in warm or cool temperatures, leaf number, rosette morphology, plant height ). We did see marked genetic differentiation among populations, possibly reflecting founder effects and a history of genetic drift. When we partitioned the phenotypic variance, there was almost no variation among maternal families within populations, limiting the potential for selection to act. The majority of the variance for all traits was among individuals within families, suggesting that environmental conditions strongly influenced the phenotype. Overall, the increasing success of V. thapsus at high elevations appears to conform more to Baker’s concept of a general-purpose genotype than to invasion by rapid adaptation. Una Aproximación Evolutiva para Entender la Biología de las Invasiones: Adaptación Local y Genotipos de Propósito General en la Hierba Verbascum thapsus Resumen: El papel de la evolución en la invasión de especies no-nativas tiene implicaciones importantes para la conservación, la ciencia de las malezas, la evaluación de riesgos y la política. En este manuscrito discutimos en primer lugar el porqué una perspectiva evolutiva puede ser útil y delineamos un espectro de estrategias de biología poblacional y genética ecológica que pueden ser empleados. Como caso de estudio, posteriormente nos preguntamos como la adaptación y la estructura genética puede promover o restringir la expansión de una hierba invasora, Verbascum thapsus , en sitios de altas elevaciones de la Sierra Nevada de California. Usamos experimentos en cámara de crecimiento y jardín común para evaluar un rango de características morfológicas y fisiológicas que pueden influenciar sobre la adaptabilidad de las plantas a elevaciones altas. Hubo una relación significativa entre la elevación y el tamaño de las hojas y la reflectancia arriba de 800 m, sin embargo, no encontramos relación alguna entre la elevación y la mayoría de las otras características ( la tasa de crecimiento a temperaturas calientes y frías, la tolerancia al congelamiento en plantas cultivadas a temperaturas calientes y frías, el número de hojas, la morfología de la roseta, la altura de la planta ). Observamos una diferenciación genética marcada entre poblaciones, lo que refleja posible‡ email [email protected] Paper submitted January 16, 2002; revised manuscript accepted September 23, 2002. 60 Evolution and Invasion Parker et al. Conservation Biology Volume 17, No. 1, February 2003 Introduction Biological invaders are now widely recognized as one of our most pressing conservation threats (Vitousek et al. 1996; Parker et al. 1999; Mack et al. 2000). Yet only a small percentage of introduced species succeed in establishing themselves, and only a few of those become the widespread, high-density pest species recognized for their conservation impact (Williamson & Fitter 1996; Smith et al. 1999). Is there some distinguishing feature of these widespread, high-impact invasions? This is a key conservation question, not only because we need to predict which species will be successful invaders, but also because we need to quantify and understand how sources of uncertainty limit our ability to make predictions (Ruesink et al. 1995). Evolution may play an important role in reducing our ability to predict whether, where, and when an introduced species will invade. Traditional ecological approaches estimate potential distributions based on the characteristics of a species’ native range using climatematching models such as CLIMEX (McFadyen & Skarratt 1996; Curnutt 2000; Holt & Boose 2000). Such approaches project the range over which a species should be invasive if its physiological tolerances do not change over time. But there is no guarantee that these tolerances, range limits, and habitat specificities will not change. We currently have no general consensus on how often adaptation of such characteristics plays a role in invasions or even what proportion of invaders show evolutionary changes in phenotype from their ancestors in the native range. Many before us have developed evolutionary hypotheses about the success of colonizing species ( Darwin 1859; Carson 1965; Lewontin 1965; Mayr 1965; Baker 1974; Brown & Marshall 1981; Parsons 1983; Barrett & Richardson 1985; Mack 1985; Barrett et al. 1989; Barrett 1992; Baker 1995), yet despite its importance there is a notable absence of an evolutionary perspective in much of the modern literature on invasion biology. To illustrate this point, we performed a five-year BIOSIS database search (1997–2001), yielding 99 papers on the biology of invasive species (key phrases: invasion biology , invasion ecology , or exotic species ). Only 5% of the studies used evolutionary ecology approaches, or even referred to evolutionary keywords such as genetic , fitness, or adaptation . Given that evolutionary ecology studies are rare in the invasion literature, here we explore what could be gained from taking an evolutionary perspective. To begin, we present two contrasting (but not mutually exclusive) views of the mechanism by which an introduced plant spreads throughout a landscape, and we then outline what can be learned from an ecological genetics approach to invasion biology. We then present a case study in which we used this evolutionary framework to develop hypotheses about how adaptation and genetic structure may promote or constrain the successful expansion of Verbascum thapsus L. (Scrophulariaceae), common mullein, into high-elevation sites in the Sierra Nevada of California. Invasion by Adaptation versus the General-Purpose Genotype A common feature in the spread of non-native species is an initial time lag when the species are present but not invasive (Cousens & Mortimer 1995; Kowarik 1995; Mack et al. 2000). Evolution could play an important role in such time lags, because natural selection should act powerfully on organisms to overcome limits to selfsustained population growth. Recent reviews emphasize the potential importance of rapid evolution in ecological processes, and non-native species provide many of the examples studied to date (Thompson 1998; Sakai et al. 2001). Widely successful invaders may expand their range into a broad array of sites by undergoing local adaptation. For those species, high levels of genetic variance within populations allow a more rapid response to selection; therefore, the conditions that favor invasion by adaptation include high outcrossing rates, high numbers of founders in new populations, and the creation of novel genotypes through gene flow among independent introduction foci (Fig. 1). Evolution may increase not only an invader’s distribution, but also its impact. Local adaptation could lead to increased absolute fitness in the new habitat, resulting in higher population densities and increased dominance of the landscape. Analogous to time lags in invasion success, time lags in impact could be widespread and common, but we would be able to detect them only with long-term experimental studies that explicitly incorporate genetics. Many researchers have found evidence for local adapmente los efectos de fundadores y una historia de deriva génetica. Cuando dividimos la varianza fenotípica, casi no hubo variación ninguna entre las familias maternas dentro de las poblaciones, limitando el potencial para la selección natural. La mayor parte de la varianza de todas las características ocurrió entre individuos dentro de las familias, sugiriendo que las condiciones ambientales influyen fuertemente sobre el fenotipo. En general, el éxito creciente de V. thapsus a elevadas alturas parece amoldarse más al concepto de Baker de un genotipo de propósito general que al concepto de invasión por adaptación rápida.