Ecological communities often transition from phylogenetic and functional clustering to overdispersion over succession as judged by space-for-time substitution studies. Such a pattern has been generally attributed to the increase in competitive exclusion of closely related species with similar traits through time, although colonisation and extinction have rarely been examined. Using 44 years of uninterrupted old-field succession in New Jersey, USA, we confirmed that phylogenetic and functional clustering decreased as succession unfolded, but the transition was largely driven by colonisation. Early colonists were closely related and functionally similar to residents, while later colonists became less similar to the species present. Extirpated species were generally more distantly related to residents than by chance, or exhibited random phylogenetic/functional patterns, and their relatedness to residents was not associated with time. These results provide direct evidence that the colonisation of distant relatives, rather than extinction of close relatives, drives phylogenetic and functional overdispersion over succession.

Predicting which nonnative species become invasive is critical for their successful management, and Charles Darwin provided predictions based on species' relatedness. However, Darwin provided two opposing predictions about the relatedness of introduced nonnatives to indigenous species. First, environmental fit is the dominant factor determining invader success; thus, we should expect that invasive species are closely related to local native residents. Alternatively, if competition is important, we should expect successful invaders are distantly related to the native residents. These opposing expectations are referred to as Darwin's naturalization conundrum. The results of studies that examine nonnative species relatedness to natives are largely inconsistent. This inconsistency arises from the fact that studies occur at different spatial and temporal scales, and at different stages of invasion, and so implicitly examine different mechanisms. Further, while species have evolved ecological differences, the mode and tempo of evolution can affect species' differences, complicating the predictions from simple hypotheses. We outline unanswered questions and provide guidelines for collecting the data required to test competing hypotheses.

Darwin's naturalisation conundrum describes the paradox that the relatedness of exotic species to native residents could either promote or hinder their success through opposing mechanisms: niche pre-adaptation or competitive interactions. Previous studies focusing on single snapshots of invasion patterns have provided support to both sides of the conundrum. Here, by examining invasion dynamics of 480 plots over 40 years, we show that exotic species more closely related to native species were more likely to enter, establish and dominate the resident communities, and that native residents more closely related to these successful exotics were more likely to go locally extinct. Therefore, non-random displacement of natives during invasion could weaken or even reverse the negative effects of exotic-native phylogenetic distances on invasion success. The scenario that exotics more closely related to native residents are more successful, but tend to eliminate their closely related natives, may help to reconcile the 150-year-old conundrum.