The discussion of links between genes codifying enzymes and enzyme activities of soil involves two different philosophical approaches since the classical soil enzymology is based on the holistic approach whereas the molecular approach is based on a detailed knowledge of enzymes involved in the measured enzyme activity. The latter approach requires the classification of the target enzyme. Soil enzymologists are dealing with enzyme activities which are the result of different enzymes catalyzing the same reaction and this is not always taken into consideration. Another drawback about the present soil enzymology is the fact that the extensive bibliography before the arrival of electronic searches is frequently ignored with the risk “to reinvent the wheel”. Increases in chitinase activity of soil by sludge or sludge and lime treatment were associated with a lower diversity in chitinase enzymes whereas casein-hydrolysing activity was mainly associated with sub genes encoding subtilisin, mainly present in soil bacteria. Laccase-encoding genes of basidiomycetes paralleled laccase activity of the soil extract, are assumed to represent the respective soil enzyme activity. Future research should involve to set up reliable soil assays for determining the activity of important enzymes such as nuclease, phytase, laccase, lignin peroxidase, etc. Since most of the microbial species inhabiting soil are unknown also some of the enzymes contributing to the measured enzyme activities are unknown. The degenerated primers may help if they cover a sufficiently long target region without loosing their specificity as it has been recently shown in the amplification of dioxygenase encoding genes.