Constitutive and induced bioassay defense
We used the same 12 pairs of non-native species and their native congeners for assessment of constitutive and induced defenses. We sowed the seeds of each species separately into 96-cell trays and then transplanted similar sized plants of each species individually into pots (15 cm diameter, 18 cm height) filled with 50% potting media (Pindstrup, Denmark) and 50% topsoil collected from the common garden. Greenhouse conditions were as described above.
When plants had 20 leaves, we started the herbivory treatment and bioassay. As the non-native plant species were mainly attacked by Lepidoptera in the field survey (Figure S2A) and common garden experiment (Figure S2B), thus we applied Spodoptera littoralis(Lepidoptera: Noctuidae), a species that feeds on more than 40 plant families (Kempel et al., 2011), as the generalist herbivore. We obtained eggs of S. littoralis from Keyun Biological Control Co., Ltd., China and fed them on bean-based artificial diet. The newly emergedS. littoralis larvae from the next generation were used to quantify the constitutive defense of pairs of non-native species and their native congeners. We measured larval weight gain on leaves harvested from healthy undamaged plants. We clipped one fully expanded leaf from the top of plants and placed it on a Petri dish (9 cm diameter) with moist filter paper. Then, we added a newly emerged pre–weighed larva (weight1) into the Petri dish. We replaced each leaf with a fresh one from another healthy plant every day. After four days we re-weighed larva and calculated weight gain for 10 replicates.
To quantify induced defenses of pairs of non-native species and native congeners, we measured larval weight gain on leaves that were harvested from plants previously damaged by herbivores. The initial damage level was at roughly 10%, which was consistent with the percentage of leaf damage in common garden experiment (11.5±1.2 %). We then added 2-6 second instar larvae to the two leaves in the middle of each plant, depending on the size of the different plant species, and covered them with a mesh bag (0.8 mm openings). After two days we removed larvae. Plants were grown 4-8 days to produce new leaves and then we followed the same protocol as above with 10 replicates.