Stepwise regression modelling and canonical correspondence analysis were used to analyse data on soil properties and the abundance of soil mesofauna collected from a woodland typical of the Borders of Scotland. The pattern of relationships revealed by stepwise regression models was different for each month, and the models compiled on the overall dataset were generally weaker than those compiled for separate months. Functional relationships among different microarthropods revealed by stepwise regression modelling are summarised in a structural model of their statistical associations. Interpretation of specific relationships revealed is given and implications for dynamic simulation models are discussed. Canonical correspondence analysis revealed that both microbial feeding nematodes (MF) and plant feeding nematodes (PF) appear to prefer a high level of bacteria and moisture, glomalin and organic matter in the soil. Close scrutiny, however, reveals that microbial feeding nematodes have a particularly high affinity to the sites with a high level of bacteria and organic matter, whilst plant feeding nematodes appear to be more associated with moisture and glomalin. Folsomia candida was abundant in sites with a higher pH level (pH ranged between 3.1 and 4.9), but was not abundant in sites with high ergosterol or a high bacteria, moisture, glomalin and organic matter level. However, other Collembola (mainly represented by Folsomia quadrioculata) appeared to be associated with high levels of ectomycorrhizal fungi. As F. candida is known to feed on fungal food sources, the results suggest that the relatively high local abundances of this collembolan might have caused local declines in ectomycorrhizal fungi, reflected, in turn, in the increase in pH. In addition, environmental plasticity of this species might have allowed them to expand into areas with low fungal density by utilising alternative food sources. The fact that F. candida was a dominant microarthropod in the majority of the samples collected in this research also supports this point. However, for those samples where F. candida were less abundant, overcompensatory fungal growth due to grazing by mites and other Collembola was implicated. Overall, our results suggest that both direct negative and indirect positive effects of the microarthropod community on specific fungal groups appear to take place. The differential effect of specific mesofaunal groups on other soil biota justifies their detailed representation in dynamic simulation models of soil ecosystems.