To successfully colonize new habitats, organisms not only need to gain access to it, they also need to cope with the selective pressures imposed by the local biotic and abiotic conditions. The number of immigrants, the preadaptation to the local habitat and the presence of competitors are important factors determining the success of colonization. Here, using two experimental set-ups, we studied the effect of interspecific competition in combination with propagule pressure and preadaptation on the colonization success of new habitats. Our model system consisted of tomato plants (the novel habitat), the two-spotted spider mite (Tetranychus urticae) as our focal species and the red spider mite (Tetranychus evansi) as a competitor. Our results show that propagule pressure and preadaptation positively affect colonization success. More successful populations reach larger final population sizes either by having higher per capita growth rates (due to preadaptation effects) or by starting a population with a larger number of individuals. Although populations are more successful colonizing non-competitive environments than competitive ones, propagule pressure and preadaptation counteract the negative effects of competition, promoting colonization success. Our study shows the importance of propagule pressure and preadaptation for successful colonization of new habitats by providing the ability to cope with both the exigencies of new environments and the community context.

The two-spotted spider mite Tetranychus urticae Koch, 1836 (Acari, Tetranychidae) is a generalist herbivore that feeds on a wide variety of host plants (Gotoh et al. 1993). Because of its small body size (female size about 0.4mm length), high fecundity (1-12 eggs/day) and short developmental time (11-28 days; Nacimiento de Vasconcelos et al. 2008), T. urticae is an ideal model organism for microcosm experiments on adaptation (Gould 1979, Egas & Sabelis 2001, Magalhaes et al. 2007, Kant et al. 2008, Bonte et al. 2010, Alzate et al. 2017, 2019). Moreover, its biology has been thoroughly described and its genomics are well-known (Grbić et al. 2011). All populations used in this study were derived from the London strain, originally collected from the Vineland region in Ontario, Canada (Grbić et al. 2011). We used 3 populations of T. urticae that differ in their level of adaptation to tomato (the novel environment used in this study): non-adapted, medium adapted and highly adapted. The non-adapted population was reared on bean plants (Phaseolus vulgaris variety “prelude”) for more than 6 years. Both the medium adapted and the highly adapted populations were derived from the non-adapted population, but medium adapted populations were reared on tomato plants (Solanum lycopersicum variety ‘moneymaker’) for about 20 generations (populations receiving 2 female mites per week in Alzate et al. 2017), and the highly adapted population was reared on tomato plants for more than 100 generations, prior our experiment. These populations differed in their fitness, measured as fecundity (number of eggs as a proxy of fitness) on tomato plants, suggesting differences in their adaptation to the tomato host plant (Alzate et al. 2017).

As a competitor, we used the red spider mite Tetranychus evansi Baker and Pritchard, 1960 (Acari, Tetranychidae), which is a specialist herbivore of (mainly) Solanaceae (incl. tomato). Adult females are easily distinguishable from T. urticae as they show a characteristic red coloration and are slightly larger (0.5 to 0.6mm length). Fecundity ranges from 10 to 14 eggs per day (Navajas et al. 2013) and development time can vary from 6.3 to 13.5 days, depending on the environmental temperature and host (Bonato 1999).

Experiments

Using two experiments (see below), we examined how total population abundance and per capita growth rate was affected by interspecific competition and either propagule pressure or preadaptation. Before each experiment we minimized epigenetic effects (juvenile and maternal effects) by collecting individual females from each population (non-adapted, medium adapted and highly adapted) and rearing them separately in a common garden for 2 generations (Magalhães et al. 2011). The common garden consisted of a 5cm diameter bean leaf disk (per female) on cotton wool soaked in distilled water. All individuals derived from a single female are therefore considered an iso-female line and each line was used as a replicate for the experiments performed in this study.

Propagule pressure and competition – In the first experiment we used the highly adapted population to study the effect of propagule pressure and competition on colonization success (H1, H3). We tested three levels of propagule pressure (3, 5, 10 individual adult female mites), and the presence or absence of competition with T. evansi (3 individuals). Per iso-female line, we placed adult female mites (3, 5 or 10) on either a complete (four weeks old) tomato plant with or without competition. In total we tested six treatment combinations with eight replicates (8 iso-female lines) for treatments with propagule pressure of 3 individuals and five replicates (5 iso-female lines) for treatments with propagule pressure of 5 and 10 individuals (Fig. S1 in Supplementary material).

Preadaptation and competition – In the second experiment we evaluated the effect of preadaptation and competition on colonization success to new environments (H2, H3). We placed 3 adult females from each adaptation treatment (and iso-female line) either on a complete (four weeks old) tomato plant without cohabitants (no competition treatment) or on a complete tomato plant together with 3 females of T. evansi (competition treatment). We tested three preadaptation and two competition levels, for a total of six treatment combinations. We used 8 replicates (iso-female lines) for treatments with non-adapted and highly adapted populations and 12 replicates for the treatment with medium adapted populations. Medium adapted populations have more replicates because we collected females from four independent populations, whereas for the non-adapted and highly adapted treatments, females came from a single population (Fig. S2 in Supplementary material).

For both experiments, plants were maintained in a climate regulated room at 25 ± 0.5ºC with a 16/8h light/dark regime for 15 days. Because fecundity varies with female age (Sabelis & van der Meer 1986) we chose adult female mites of similar age (1-2 days after emerging from the last quiescent stage) to be placed on tomato plants. In the competitive environments, we ensured that females of T. urticae were placed on the same leaves as females of T. evansi for competition to be effective. Population sizes of T. evansi and T. urticae, and growth rate of T. urticae were recorded after 15 days (one generation). To estimate population sizes, we counted all adult female mites present on each complete tomato plant. Juveniles and males were not included in the counting because their small size made their detection with the naked eye difficult. Complete population sizes are therefore larger than the ones presented in this study, when accounting for juveniles and males (S2 in De Roissart et al. 2015). Per capita growth rate was calculated by first subtracting the initial number of females from the final number of females after one generation, then dividing this by the initial number of females.

In addition to the competition experiments using T. evansi and T. urticae populations with different propagule pressures and preadaptations, we simultaneously studied 5 populations of T. evansi without competition, in which 3 adult females of T. evansi were placed on tomato plants.

Description of the data is added as a tab in the excel file