Natural selection can produce local adaptation, but local adaptation can be masked by maladaptive plasticity. Maladaptive plasticity may arise as a result of gene flow producing novel gene combinations that have not been exposed to selection. In the 1980s, populations of the red-shouldered soapberry bug (Jadera haematoloma) were locally adapted to feed on the seeds of a native host plant and an introduced host plant; by 2014, local differentiation in beak length had been lost, likely as a consequence of increased gene flow. In this study, I assess the relative contributions of natural selection and plasticity to beak length on these two hosts. I confirm the earlier hypothesis that the host plant seedpod drives divergent natural selection on beak length. I then demonstrate that the proximate cause of the loss of observable differentiation in beak length is maladaptive plasticity, which masks persistent genetic differences between host-associated populations. Maladaptive plasticity is highest in areas where the two plants co-occur; in combination with historical measures of plasticity in hybrids, this indicates that maladaptive plasticity may be a consequence of ongoing gene flow. Although natural selection produced locally adapted genotypes in soapberry bugs, maladaptive plasticity is masking phenotypic differences between populations in nature.
Data from Experiment 1: Testing natural selection in the lab
Data collected from lab feeding trials of Jadera haematoloma on opened and closed seedpods of C. corindum and K. elegans. This file includes data on adult morphology (beak length, thorax width, wing length, wing morph), family, population, host, feeding trial data (weight change, latency time, feeding time), and reproduction (egg number and mass). Seedpod treatments are abbreviated C and O (closed and open). Columns F1-F16 indicate whether or not an individual was feeding at each time point during the feeding trial; these are ordered chronologically. Some individuals do not have data for all 16 time points, depending on the length of the individual feeding trial. Latency time for individuals that never began feeding are labelled NA. There are some individuals included in this data file from a third host plant (C. microcarpum) that are not included in the manuscript.
Experiment.1.data.csv
Experiment 2: Testing natural selection in the field
Data collected during field feeding trials for J. haematoloma on open and closed seedpods of C. corindum in Key Largo, FL in April 2014. This file includes data on morphology (beak length, thorax width, wing morph), feeding activity (latency time, feeding time), and flight ability. Seedpod treatments are abbreviated C and O (for closed and open pods). The flight.test column indicates whether or not flight behavior was observed when individuals were gently tossed into the air three times. Columns F1-F16 indicate whether or not an individual was observed feeding at each time point during the feeding trial; these are ordered chronologically.
Experiment.2.data.csv
Experiment 3: Measuring the plastic effect of rearing host on adult morphology
Data collected from lab trials for J. haematoloma reared on the hosts K. elegans and C. corindum. This file includes data on morphology (beak length, thorax width, wing length), development time, sex, family, and ancestral location (population, latitude, host plant). Data is also included in this file for individuals from a third ancestral host, C. microcarpum, that is not reported on in the manuscript.
Experiment.3.data(morph,dvpt).csv