Recently, dramatic environmental changes due to urbanization have been shown to impact the evolution of various organisms. However, much remains still unknown about how urban environments influence evolutionary processes and outcomes. Because many studies on the impacts of urbanization on evolution have applied dichotomic classification between urban vs rural environments, the urban area has been typically treated as a homogenous environment. Thus, evolutionary consequences of heterogeneity of urban environments due to various types of compositions (e.g., roads, parks, and green area) and their configurations are overlooked. This environmental heterogeneity can lead to different selection pressures on populations found even within a single urban region. Our previous study showed that the heterogeneity of urban landscape shaped the evolution of defense traits of white clover. Here, we hypothesize that the presence of remaining natural green spaces from urban to rural areas (hereafter “remnant forests”) contribute significantly to create the heterogeneity of the urban landscape, and therefore are critical for the evolution of plant defense traits.

In this study, we aimed to clarify the evolutionally effect of the presence of remnant forests on white clover through the following approaches: 1) quantifying the spatial heterogeneity in urban environments and in the evolution of white clover in multiple cities, respectively, and examining the relationship between the two. In particular, we focus on the spatial distribution of remnant forests as a factor that generates heterogeneity; 2) empirically examining the differences in adaptations of white clover of different genotypes through transplant experiments in urban and rural areas. Especially, we considered frequency‐dependent selection as a foundation for genetic polymorphism and local adaptation.

First, we collected white clovers from 122 populations in Sapporo, 40 populations in Hakodate, 41 populations in Asahikawa, and 30 populations in Kushiro. Among the four cities with different urban structures such as the arrangement of remnant forests, we compared the urban heterogeneity and the spatial variation of the frequencies of antiherbivore defense trait of white clover. Differences in spatial heterogeneity in the urban environments would influence the degree of urban evolution. Second, we examined the survival rate and biomass as fitness measures in the transplanting experiment. The results suggest a relationship between benefits of defensive traits and negative impacts of abundance of surrounding defensive genotypes is likely to determine local optimal frequency of defensive traits. We will discuss how urban heterogeneity shape the evolution of white clover.