Effects of novel predator on prey behaviour
For the proportion of melomys emerging in open field tests during the 0­–10 min period, there was a significant interaction between quoll presence and trapping session (χ2 (5) = 4.386,P = 0.04; Fig. 4). There was no interaction between quoll presence and trapping session for the proportion of melomys emerging and interacting with the novel object during 10–20 min period (χ2 (5) = 2.567, P = 0.109; Fig. 4). The model without this interaction, however, revealed a significant effect of quoll presence, with fewer melomys emerging from hiding and interacting with the novel object during the 10–20 min period of open field tests from sites where quolls were present than from sites where quolls were absent (χ2 (5) = -4.696, P < 0.001; Fig. 4).