Some Sampling Variations in Soil Fungal Numbers

SUMMARY: Fungal counts (as measured by direct plate counts) from soil samples taken from the same field have frequently shown variations of the order of 1:13 for some New Zealand pastures. The main source of error of mean counts is the variation between cores in the field, much less being due to subsampling and plating methods in the laboratory. For an arbitrary standard of accuracy, based on differences observed between soils at various stages of development, about twenty-five cores per field are necessary even if the field appears homogeneous. If the cores are to be bulked before plating thirty-two should be taken, preferably in eight groups of four cores. Little advantage accrues from taking more than three plates per sample for incubating and counting. Wide variations are occasionally obtained between batches on the mechanical shaker, but this could not be attributed to the effect of time of shaking.