Table 1. Summary of the data and results used to support carbon-limitation hypothesis.
| Data Class | Data | Variable | Factor | Correlation* | Evidence for main hypothesis**/against alternative hypothesis? (MH/AH) | Figure/Table |
| Environmental | Precipitation | Total annual (mm) | Latitude | Negative | MH | Figures 2, S1; Table S1 |
| Duration | ||||||
| Temperature | Annual mean (°C) | Weakly positive | AH | Table S2 | ||
| Soil nutrients | µg/g | None | AH | Table S3 | ||
| Light | % openness | Positive; Controlled | AH | Table S4 | ||
| Organismal | Scale insects | Density per tree | Precipitation | Negative | MH | Figure 3A |
| Ant workers | Density per tree | Negative | MH | Figure 3B | ||
| Ant reproductives | Number per colony | None | MH | Table S5 | ||
| Ant defensive behavior | Patrolling ants (no.) | Negative | MH | Figure S5A | ||
| Ant response | Latency (sec) | Positive | MH | Figures 3C, S5B, S4 | ||
| Ant exclusion | % leaf area | Negative | MH | Figure 3D | ||
| Background herbivory | % leaf area | Positive | MH | Figures S9, S10 | ||
| Organismal (Stoichiometric) | Tree NSCs | Total (mg/g) | Precipitation | Weakly positive | MH | Figures 4, S6; Tables S6, S7 |
| Starch (mg/g) | Positive | MH | ||||
| Sucrose (mg/g) | Negative | MH | ||||
| Tree allometry | Mass (g) | None | AH | Figure S7 | ||
| Leaf area (cm2) | ||||||
| Tree nutrients | Nitrogen to Phosphorus ratio | None | AH | Figure S8 | ||
| Ant isotopes | δ15N | None | AH | Figure S3A | ||
| Ant nutrients | % Carbon, % Nitrogen | None | AH | Figure S3B,C | ||
| Theoretical | Optimality models | Tree carbon pool | Precipitation | Positive | MH | Figures 5, S10; Text S1 |
| Tree carbon invested in ants | Negative | MH | ||||
| Ant carbon invested in workers | Negative | MH |
*
Bold indicates significance at p<0.05 level.
**
Mechanistic hypothesis illustrated in Figure 1 schematic.