An experimental test of the evolutionary consequences of sympatry in Drosophila subquinaria
Abstract
Prezygotic isolation is often stronger between sympatric as opposed to allopatric taxa, but the underlying cause can be difficult to infer from comparative studies alone. Experimental evolution, where evolutionary responses to treatments manipulating the presence/absence of heterospecific individuals are tracked, can provide a powerful complementary approach. We used experimental evolution to investigate a naturally occurring pattern of reproductive character displacement in the mushroom-feeding fly, Drosophila subquinaria. In nature, female D. subquinaria from populations sympatric with the closely related D. recens discriminate more strongly against heterospecific males than do females from allopatric populations. Starting with 16 replicate allopatric populations of D. subquinaria, we manipulated the presence/absence of D. recens during mating (experimental sympatry vs. control) and, when present, we allowed hybrids to live or killed them each generation. Across 12 generations, heterospecific offspring production from no-choice mating trials between D. subquinaria females and D. recens males declined in both experimental sympatry treatments relative to the control, suggesting increased sexual isolation. Male cuticular hydrocarbon profiles also evolved, but only in the hybrids killed treatment. Our results strongly imply that the existing reproductive character displacement in wild D. subquinaria populations was an evolutionary response to selection arising from secondary contact with D. recens.
README: An experimental test of the evolutionary consequences of sympatry in Drosophila subquinaria
[Access this dataset on Dryad] https://doi.org/10.5061/dryad.5x69p8db7
Uploaded data for a multi-generation evolution experiment testing the effect of artificial sympatry between two drosophila species. Data includes measures of mating rates and CHCs (cuticular hydrocarbons, a kind of contact pheromone) at each generation in each population.
Description of the data and file structure
data_cons.csv <- contains conspecific mating rates from no-choice mating trials conducted in each population at each sampling generation (0,6, and 12). Columns are as follows: generation (sampling generation), block (experimental block where populations were placed in block 1 or block 2), blockXgen (factor for statistical methods concatonating block and generation), treatment (experimental treatment; Allo treatment = control, Str reinforcement = Hybrids killed, Wk reinforcement = Hybrids live), popID (population ID), male mated with (indicating conspecific or heterospecific mating rate, N mated (number of successful matings), N not mated (number of failed matings), N total trials (total number of matings), fraction mated (mating rate).
data_hetero.csv <- contains heterospecific mating rates from no-choice mating trials conducted in each population at each sampling generation (0,6, and 12). Columns are as follows: generation (sampling generation), block (experimental block where populations were placed in block 1 or block 2), blockXgen (factor for statistical methods concatonating block and generation), treatment (experimental treatment; Allo treatment = control, Str reinforcement = Hybrids killed, Wk reinforcement = Hybrids live), popID (population ID), male mated with (indicating conspecific or heterospecific mating rate, N mated (number of successful matings), N not mated (number of failed matings), N total trials (total number of matings), fraction mated (mating rate).
dm.csv <- contains male CHC data from each sampling generation. Columns are as follows: sampleID (CHC sample ID), block (experimental block where populations were placed in block 1 or block 2), population (label for population number within a treatment and block), popID (population ID), Popgen (factor for statistical methods concatonating population and generation), blockXgen (factor for statistical methods concatonating block and generation), treatment (experimental treatment; Allo treatment = control, Str reinforcement = Hybrids killed, Wk reinforcement = Hybrids live), generation (sampling generation), sex (male or female), clr1:clr24 (centered log ratio transformed relative CHC concentrations, numbers indicate compound numbers from Curtis 2013, see methods for details)
df.csv <- contains female CHC data from each sampling generation. Columns are as follows: sampleID (CHC sample ID), block (experimental block where populations were placed in block 1 or block 2), population (label for population number within a treatment and block), popID (population ID), Popgen (factor for statistical methods concatonating population and generation), blockXgen (factor for statistical methods concatonating block and generation), treatment (experimental treatment; Allo treatment = control, Str reinforcement = Hybrids killed, Wk reinforcement = Hybrids live), generation (sampling generation), sex (male or female), clr2:clr20 (centered log ratio transformed relative CHC concentrations, numbers indicate compound numbers from Curtis 2013, see methods for details)
Additional data used can be found in Dyer et al. 2014 (https://doi.org/10.1111/evo.12335)