Table 1. A description of the variables measured in Serra do Japi.
| Measurement | Description |
| Spider Nest Characteristics | |
| Nest Size | Area of the largest horizontal cross section of the nest basket; A. nigrescens sometimes has a less clearly defined nest basket, so we measured the longest and widest horizontal web cross section(Purcell and Avilés 2007) |
| Prey Capture Web Height | Greatest vertical extent of the loose webbing above the basket (Purcell and Avilés 2007) |
| Habitat-Scale Spatial Measurements | |
| Distance from Forest Edge | Distance from the nest to the nearest forest edge, measured up to 10 m and estimated at longer distances |
| Percent Canopy Cover | Visual estimate of the percent of canopy cover directly above the nest |
| Forest Height | An estimate of the average height of the canopy above the nest |
| Local-Scale Spatial Measurements | |
| Plant Identity (Substrate) | Identity of the plant supporting the nest (classified to family) |
| Height Above Ground | The distance from the lowest part of the nest's basket to the ground below it |
| Plant Diameter at Breast Height (DBH) | The size of the plant supporting the nest, classified in categories: 0: plants shorter than1.4 m; small: up to 10 cm diameter; medium: up to 40 cm; large: greater than 40 cm |
| Nest Position on Plant | Position of the nest on the plant: branch tip, middle of the branch, core of the plant |
| Vegetation sturdiness index | An index based on: plant height, DBH, diameter at 50 cm (knee) height, length and diameter of the longest branch. The index is the first axis of a principal component analysis (see Appendix S1). |
| Temporal Measurement | |
| Spider Instar | Most nests contained a single instar (juvenile 1–4, subadult, adult, egg sac present); when two instars were represented, we assigned the nest to the category representing the most common instar. For analyses, juvenile instars were grouped into one category. |
We determined the spider nest characteristics, habitat- and local-scale spatial measurements (except vegetation sturdiness index) and temporal measurement for each spider nest along each transect. In addition, we measured the habitat- and local-scale spatial (except vegetation sturdiness index) variables at 20 randomly selected points along each transect in 2005. The vegetation sturdiness index was calculated based on measurements taken in 2010 at 20 randomly selected points along two sets of two parallel transects, one of each along the forest edge and the other inside the forest.