Post-invasion changes in reproductive and dispersal traits, and their correlations, in three invasive mustards

Background/Question/Methods

One potential explanation for invasion success of many exotics is the occurrence of evolutionary changes affecting ecologically relevant traits in the introduced range. Reproductive traits have important consequences for population dynamics and thus for invasion potential. Reproductive success in the new range may depend on the ability of introduced genotypes to adapt their flowering schedules to novel biotic and abiotic conditions and to optimize dispersal in the new environment. However genetic correlations between phenological and fitness traits entail trade-offs that may constrain local adaptation.

We tested for both shifts in means and correlations between reproductive traits in invasive and native populations of three species of yellow mustards (Brassicaceae): Brassica nigra, Sinapis arvensis and B. rapa. The study species are native to Eurasia and invasive in N America; they differ in flowering phenology and plant size, from late-flowering big-sized B. nigra to early-flowering small B. rapa. In a common garden in California we grew seeds from 4 to 11 populations per species and range (native vs. invasive), totaling 53 populations. We asked (i) whether invasive genotypes of yellow mustards show shifts in reproductive and dispersal traits and (ii) whether trait correlations differ across invasive and native populations.

Results/Conclusions

Differences in trait means and correlations between invasive and native populations were species-specific, but shared some common responses. All three species had altered flowering times with invasion, even when latitude of origin was incorporated as a covariate. Invasive populations of later-flowering B. nigra and S. arvensis flowered earlier than native populations, while early-flowering B. rapa flowered later. In all three species, invasive populations produced fewer fruits than native populations, though often total seed number did not differ. B. nigrainvaders uniquely had significant increased height at dispersal.

We also detected across-population differences in trait correlations, but no evidence for trade-offs within populations. Early-flowering plants of native populations of B. nigra are lower in height, but early-flowering plants of invasive populations do not differ in height from later-flowering ones. We did not find any relationship between flowering phenology and seed production in native populations of S. arvensisbut early-flowering plants produced more seeds than late-flowering plants in invasive populations.  

Our results suggest that evolutionary changes in the new range can affect trait means but also patterns of allocation and trade-offs that might result in phenological niche shift and improved dispersal strategies of the invasive populations in the new environmental context.