Climate change pushes species toward higher latitudes and altitudes, but the proximate drivers of range expansions vary, and it is unclear whether evolution facilitates climate change–induced range changes. In a temporally replicated field experiment, we translocated wall brown butterflies (Lasiommata megera) descending from range interior and range margin populations to sites at 1) the range interior, 2) the range margin, and 3) beyond the current northern range edge. Thereby, we tested for local adaptation in seasonal timing and winter survival and evaluated to what extent local adaptation influences the ongoing, climate-driven range expansion. Almost all individuals from all populations entered diapause at an appropriate time, despite previously identified among-population variation in diapause induction thresholds. Caterpillars of northern descent, however, grew faster than those from southern populations at all field sites. This may be a countergradient adaptation to compensate for the short, northern growing seasons, but we found no evidence for prewinter body mass affecting winter survival. In fact, winter survival was low overall—extremely so at the beyond range site—regardless of population origin, indicating that the primary constraint to range expansion is an inability to adapt to winter conditions. Hence, although range-expanding wall browns show clear local evolution of two traits related to seasonal timing, these putative local adaptations likely do not contribute to range expansion, which is instead limited by winter survival. To predict future range changes, it will be important to distinguish between the traits that evolve during range expansion and those that set the range limit.

Climate change pushes species polewards and upwards – as temperatures rise, species move to areas that were previously too cold for them. During range expansions, species encounter unfamiliar environmental conditions, which may require evolutionary adaptation, but expanding populations may often be hampered by their genetic and demographic properties. Whether range-expanding populations can adapt may greatly affect species distributions, but the question is largely unexplored for native species expanding in response to climate change. 

In seasonal environments, organisms must endure harsh conditions and synchronise growth and reproduction with the presence of food and mates. To time their life cycles, many animals and plants use seasonal changes in daylength. Insects typically overwinter in diapause (dormancy with paused development and suppressed metabolism), which is induced by short days well before winter. But across-latitude differences in daylength pose challenges for latitudinal range expansions. I focus on whether traits related to seasonal timing and winter survival have evolved during range expansion of the wall brown butterfly (Lasiommata megera) in Sweden.

In Chapter I, I confirmed that the wall brown has, in 2000–2020, expanded northwards along the eastern and western coasts of Sweden, and in Chapter II, I demonstrated that these parallel expansions have proceeded independently from the south, in isolation from each other. Laboratory experiments in Chapter I revealed that caterpillars from northern populations have evolved to correctly interpret their local daylength cues. This rapid evolution, repeated along two range expansions, indicates that latitudinal differences in daylength may seldom hinder insect range expansions. In Chapter II, I found that northern range margin populations have lower genetic variation than southern ones but are unlikely to have received much locally maladaptive gene flow from the south. Further, a genomic scan suggested that the parallel phenotypic changes have evolved through non-parallel genetic changes. In Chapter III, a laboratory experiment showed lack of local adaptation to different winters, in contrast to the rapid evolution of diapause timing. Overall winter survival was low in the coldest treatment, indicating that winter temperatures limit the range.

In Chapter IV, I studied diapause induction, growth rate, and winter survival in a field setting. Almost all individuals entered diapause, with only minimal impact from the among-population differences found in Chapter I. These evolved differences could stem from natural selection on earlier parts of the generation, which experience longer days than our experiment captured. Further, individuals of northern descent grew faster than those from the south. This could help them grow large enough before winter, yet pre-winter mass did not affect winter survival. This time, natural selection may favour high growth rates in late-hatching individuals with less time to grow before winter. Like in Chapter III, there was no evidence for evolution of improved winter survival, and survival dropped markedly when transplanting individuals outside of the current range.

Despite rapid evolution in two traits, cold winters limit the wall brown’s expansion. To predict range expansions, we must pinpoint their drivers, study trait evolution relevant to these drivers, and recognize that traits that are crucial in different seasons may vary in evolvability.

In their simulation study, Garcia-Costoya et al. (2023) conclude that evolutionary constraints might aid populations facing climate change. However, we are concerned that this conclusion is largely a consequence of the simulated temperature variation being too small, and, most importantly, that uneven limitations to standing variation disadvantage unconstrained populations. In their simulation study, Garcia-Costoya et al. (2023) conclude that evolutionary constraints might aid populations facing climate change. However, we are concerned that this conclusion is largely a consequence of the simulated temperature variation being too small, and, most importantly, that uneven limitations to standing variation disadvantage unconstrained populations.image

1. Numerous species shift or expand their ranges poleward in response to climate change. Even when expanding species follow their climatic niches, expanding range margin populations are likely to face unfamiliar environmental conditions and thus natural selection for local adaptation.
2. The wall brown butterfly (Lasiommata megera) has expanded northward in Sweden in the years 2000–2020, most likely as a result of climate change, and has previously been shown to have evolved local adaptations to northern daylength conditions. This evolution has occurred despite hypothesised genetic constraints to adaptation at range margins.
3. We studied local adaptation to winter conditions in four of the previously-studied L. megera populations, using a common garden laboratory experiment with a warm and short, an intermediate, and a cold and long winter treatment. We compared the winter and post-winter survival of caterpillars from two southern core range and two northern range margin populations in Sweden.
4. During the experiment, we measured metabolic rates of a subset of diapausing caterpillars to test whether populations differ in metabolic suppression during diapause. Further, we measured supercooling points, which reflect lower lethal temperature in L. megera, of the same subset of caterpillars. We also compared supercooling points between L. megera and three closely related species with more northern distributions.
5. Few individuals survived the coldest treatment all the way to successful adult emergence, so L. megera seems susceptible to cold winters. Individuals of northern descent did not survive cold winters any better than individuals from southern populations. Similarly, there were no signs of local adaptation in metabolic rates or supercooling points. The comparison among species did not reveal any clear relationship between geographical distribution and supercooling point.
6. Although northern winters probably exert strong selection on L. megera, we provide comprehensive evidence for the lack of local adaptation to winter conditions. This contrasts with the previous finding of quickly evolved local adaptation in diapause timing, highlighting the need to consider how traits associated with different seasons differ in how they may evolve and facilitate climate change-induced range expansions.

Climate change allows species to expand polewards, but non-changing environmental features may limit expansions. Daylength is unaffected by climate and drives life cycle timing in many animals and plants. Because daylength varies over latitudes, poleward-expanding populations must adapt to new daylength conditions. We studied local adaptation to daylength in the butterfly Lasiommata megera, which is expanding northwards along several routes in Europe. Using common garden laboratory experiments with controlled daylengths, we compared diapause induction between populations from the southern-Swedish core range and recently established marginal populations from two independent expansion fronts in Sweden. Caterpillars from the northern populations entered diapause in clearly longer daylengths than those from southern populations, with the exception of caterpillars from one geographically isolated population. The northern populations have repeatedly and rapidly adapted to their local daylengths, indicating that the common use of daylength as seasonal cue need not strongly limit climate-induced insect range expansions.

Understanding how social groups function requires studies on how individuals move across the landscape and interact with each other. Ant supercolonies are extreme cooperative units that may consist of thousands of interconnected nests, and their individuals cooperate over large spatial scales. However, the inner structure of suggested supercolonial (or unicolonial) societies has rarely been extensively studied using both genetic and behavioral analyses. We describe a dense supercolony-like aggregation of more than 1,300 nests of the ant Formica (Coptoformica) pressilabris. We performed aggression assays and found that, while aggression levels were generally low, there was some aggression within the assumed supercolony. The occurrence of aggression increased with distance from the focal nest, in accordance with the genetically viscous population structure we observe by using 10 DNA microsatellite markers. However, the aggressive interactions do not follow any clear pattern that would allow specifying colony borders within the area. The genetic data indicate limited gene flow within and away from the supercolony. Our results show that a Formica supercolony is not necessarily a single unit but can be a more fluid mosaic of aggressive and amicable interactions instead, highlighting the need to study internest interactions in detail when describing supercolonies.