Table 2.
Outputs and significance of the final linear mixed models testing the differences in mean δ 13C and trophic position (TP) between the species across the mesocosm experiment, where mesocosm was the random effect on the intercept
Final model structure (and result): | |||
---|---|---|---|
δ 13C ~ species × experimental treatment (AIC = 141.8; log likelihood = −64.9; P < 0.01) | |||
Trophic position ~ species × experimental treatment (AIC = −178.9; log likelihood = 95.4; P < 0.01) | |||
Pairwise comparison | Mean difference in δ 13C | Mean difference in TP | |
Allopatric | Allopatric Squalius cephalus | 2.12 ± 0.36, P < 0.01* | 0.13 ± 0.03, P < 0.01* |
B. barbus | Sympatric with S. cephalus | 0.85 ± 0.36, P > 0.1 | 0.11 ± 0.03, P = 0.01* |
Allopatric | Sympatric with B. barbus | 0.36 ± 0.36, P > 0.1 | 0.02 ± 0.03, P > 0.1 |
S. cephalus | |||
B. barbus in sympatry with S. cephalus | 1.63 ± 0.23, P < 0.01* | 0.004 ± 0.02, P > 0.1 |
Mean differences are from estimated marginal means (* = difference significant at P < 0.05).