At each site 5-25 soil samples were taken for organism enumeration and moisture content analysis as follows: Sampling bags were prepared with one sterile 'Whirlpak' bag and clean plastic scoop per sample. The location of the sampling was recorded each year so that areas were not re-sampled. Using the plastic scoop, soil was collected to a 10 cm depth. Very large rocks (>20 mm diameter) were excluded from the sample. The soil was shoveled into the 'Whirlpak' bag until three quarters full (about 1.5 kg soil). The soil was mixed well in the bag, then the bag was closed tightly, expelling as much air as possible. The soil samples were stored in a cooler for transportation. On return to the laboratory (within 8 hours of sampling), the soils were stored at +5 degrees C until further processing. In the laboratory, soil samples were handled in a laminar flow hood to prevent contamination. The Whirlpak bags of soil were mixed thoroughly prior to opening. Approximately 200cm3 of soil was placed in a pre-weighed 800mL plastic beaker. Rocks greater than 3-4mm in diameter were removed from the sample. A sub-sample of approximately 50g was removed and placed in a pre-weighed aluminum dish, and weighed on a balance accurate to 0.01g. This sample was dried at 105 degrees C for 24 hours. The sample was removed, placed in a desiccator to cool down, and re-weighed. These data were used to calculate water content of the soil and to express data as numbers of soil organisms per unit dry weight of soil. The remaining soil in the plastic beaker was weighed. Cold tap water was added up to 650 mL. The soil suspension was stirred carefully (star stir or figure of 8) for 30 seconds, using a spatula. Immediately the liquid was poured into wet screens - a stack of 40 mesh on top of a 400 mesh. The screens were rinsed gently with ice cold tap water (from a wash bottle) through the top of the stack, keeping the screens at an angle as the water filtered through. The water was kept on ice at all times. The top screen was removed, and the lower screen rinsed top down, never directly on top of the soil, but at the top of the screen and from behind. The water was allowed to cascade down and carry the particles into the bottom wedge of the angled screen. The side of the screen was tapped gently to filter all the water through. The suspension was rinsed from the front and the back, keeping the screen at an angle and not allowing the water to overflow the edge of the screen. The soil particles were backwashed into a 50mL plastic centrifuge tube, tipping the screen into the funnel above the tube and rinsing the funnel gently. The suspension was centrifuged for five minutes at 1744 RPM. The liquid was decanted, leaving a few mL on top of the soil particles. The tube was filled with sucrose solution (454g sucrose per liter of tap water, kept refrigerated) up to 45mL. This was stirred gently with a spatula until the pellet was broken up and suspended. The suspension was centrifuged for one minute at 1744 RPM, decanted into a wet 500 mesh screen, rinsed well with ice cold tap water and backwashed into a centrifuge tube. Samples were refrigerated at 5 degrees C until counted. Samples were washed into a counting dish and examined under a microscope at x10 or x20 magnification. Rotifers and tardigrades were identified and counted. Nematodes were identified to species and sex, and counted. Total numbers in each sample were recorded on data sheets. All species of nematode, and all rotifers and tardigrades found in the sample were recorded. Data were entered into Excel files, printed, and checked for errors.