Table 1.
Classification and key morphological characteristics of the study species.
Table 2.
Experimental design of richness and species combination treatments, with the number of replicates as indicated.
Figure 1.
Net oxygen production (A) and total biovolume (B) of phytoplankton monocultures and polycultures.
Each box displays the experiment-long median (line within box), 25th and 75th percentiles (box boundaries), 10th and 90th percentiles (lower and upper error bars) and 5th and 95% percentiles (dots). Medians and percentiles were calculated using all data collected throughout the experiment (n = 54 for polycultures, 108 for monocultures). Net oxygen production in the monocultures was centered on zero due to the consumption of oxygen after rotifer addition (see Figure 3A).
Figure 2.
Relative yield of net oxygen production (A) and total biovolume (B).
The main plots compare species combinations whereas the insets compare levels of species richness. Bars represent experiment-long means (± SE) calculated for each replicate and sampling day. Letters above each bar indicate the results of the post-hoc Tukey tests associated with ANOVA. Species combinations are listed with each plot, using the following abbreviations: D – Dunaliella, N – Nannochloris, R – Rhodomonas, Ch – Chaetoceros, Co – Coscinodiscus, M – Melosira and all – all species in the six-species combination.
Figure 3.
Oxygen production, phytoplankton biovolume and rotifer abundance through time in the different richness treatments.
The top figure (A) shows the mean (± SE) net oxygen production of the different richness treatments. The bottom figures show total phytoplankton biovolume (symbols) and rotifer abundance (bars) of the one- (B), two- (C), four- (D) and six-species (e) treatments before rotifer addition (first three points) and after rotifer addition (last three points). Points and bars in b-e represent means ± standard errors (n = nine for polycultures, 18 for monocultures). The arrow indicates the point of rotifer addition at an initial abundance of one rotifer per mL. Rotifer abundance is shown on a log scale.
Table 3.
Results of mixed model ANOVA comparing the effects of species richness and combination on oxygen production, total phytoplankton biovolume, relative yield (RY), and temporal variability (CV).
Figure 4.
Mean (± SE) phytoplankton biovolume (symbols) and rotifer abundance (bars) of species combinations through time.
The arrows indicate the point of rotifer addition at an initial abundance of one rotifer per mL. Species combinations are listed with each plot, using the following abbreviations: D – Dunaliella, N – Nannochloris, R – Rhodomonas, Ch – Chaetoceros, Co – Coscinodiscus, M – Melosira, all – all species in the six-species combination. All primary y-axes are identical with the exception of the y-axis in the row of plots second from the top. Rotifer abundance is shown on a log scale.
Figure 5.
Temporal CV of net oxygen production (A) and total biovolume (B).
The main plots compare species combinations whereas the insets compare levels of species richness. Bars represent experiment-long means (± SE) calculated for each replicate and sampling day. Note that the temporal CV was calculated for single culture tubes, such that the mean values for richness treatments and species combinations are not influenced by the number of replicates per treatment. Letters above each bar indicate the results of the post-hoc Tukey tests associated with ANOVA. Species combinations are listed with each plot, using the following abbreviations: D – Dunaliella, N – Nannochloris, R – Rhodomonas, Ch – Chaetoceros, Co – Coscinodiscus, M – Melosira and all – all species in the six-species combination.