Maternal effects can maintain fitness across environmental conditions regularly experienced by a population. However, the consequences of maternal effects for ecological resilience in novel environments remain poorly understood. We produced seeds by breeding field-collected individuals of two closely related, but ecologically distinct, species of Sicilian daisy (Senecio, Asteraceae) in a quantitative genetics design. Using siblings of the same families, we quantified seedling root morphology in the laboratory, and planted seeds at four elevations that included the native elevation of each species. Seedling survival of both species was high at their native elevations, but declined at the novel elevation furthest from their native range, demonstrating adaptive divergence. Seedling development rate changed with elevation similarly for both species, and aligned with selection for faster development at lower elevation and slower development at higher elevation. Selection on paternal (sire) variance in development rate was consistent across elevation, and opposed phenotypic selection in the novel elevations. By contrast, selection on maternal variance in development rate changed with elevation and aligned with the direction of selection at both the native and novel elevations, suggesting that maternal effects underlying plasticity in development rate can increase fitness in novel environments. Selection on root morphology was only strong for the high elevation species, and occurred in the direction of the native phenotype for maternal variance, but not for sire variance. Together, these results suggest that maternal variation in development time plasticity and root architecture can increase ecological resilience in novel habitats, despite theory suggesting that they may reduce fitness in such situations.

BackgroundAcute cannabis use stimulates appetite, while general population studies suggest that chronic use is associated with reduced risk of obesity and other cardiometabolic risk factors.MethodsIn this study, we investigated changes in body mass index (BMI), fasting blood glucose and lipids, and rates of metabolic syndrome risk factors in cannabis users vs. non-users in 109 minimally treated patients with first-episode schizophrenia, schizophreniform or schizo-affective disorder who were treated according to a standardized treatment regime with depot antipsychotic medication over 12 months. Participants underwent repeated urine toxicology tests for cannabis and those testing positive at any time during the study (n=40), were compared with those who tested negative at all time points (n=69).ResultsThere was a significant group*time interaction effect (p=0.002) with the cannabis negative group showing a greater increase in BMI than the cannabis positive group, after adjusting for age, sex, methamphetamine use and modal dose of antipsychotic. There were no group*time interaction effects for fasting blood glucose or lipids. Post hoc tests indicated significant increases in fasting blood glucose and triglycerides and a decrease in high-density lipoprotein cholesterol for the cannabis negative group, with no significant changes in the cannabis positive group. Rates of metabolic syndrome did not differ significantly between groups. However, more cannabis negative patients had elevated waist-circumference at endpoint (p=0.003).DiscussionAlthough other indirect effects such as dietary neglect and smoking may be contributory and could explain our findings, it may be that chronic cannabis use directly suppresses appetite, thereby preventing weight gain in users.