Figure legends
Figure 1 Schematic plan of the experimental design used to test the effect of an invader on the diversity of disturbed communities. Microcosms (n=48) containing five bacterial species were subject to a pulse disturbance using a 1% transfer either every 2, 4, 8 or 16 days. On days 4, 8 and 12 an invader (Pseudomonas aeruginosa ) was added to half (n=6) of the microcosms in each disturbance regime. On day 16 all microcosms were frozen.
Figure 2 Resident diversity (effective number of species) of a five-species community after 16 days in one of four pulse regimes (2, 4, 8 or 16 days), and either invaded by Pseudomonas aeruginosa (red dotted line and triangles) or not (black solid line and circles). See Figure S1 for the same analysis using Gini-Simpson’s index. The best fit model (Eq. 1) and the standard error are shown as the fitted lines and their envelopes. Individual points correspond to each microcosm, and their enclosed number shows the number of resident species present,i.e. , species richness.
Figure 3 The (A) proportion and (B) the density of P. aeruginosa in a stable community of five species of bacteria exposed to different disturbance frequencies from every 16 days (low) to every 2 days (high) for 16 days. P. aeruginosa was added on days 4, 8 and 12. Triangles show individual microcosms, the dashed line the model fit and its envelope the standard error.
Figure 4 . Effects of disturbance frequency and invasion on (A) resident density and (B) total density (resident + invader) of a five-species community either invaded by P. aeruginosa (red triangles) or not (black circles). Triangles and circles show individual microcosms.
Figure 5 Density (CFU per microcosm) of five bacterial species after 16 days in one of four pulse disturbance regimes (every 2, 4, 8 or 16 days), and either invaded by Pseudomonas aeruginosa (red triangles) or not (black circles).