Soil lipids originate from both plants and animals as products of deposition, exudation, and decomposition as well as various other sources, such as fungi, bacteria and mesofauna. The composition, concentration and diagenetic fate of soil lipids varies between environments due to differences in sources of organic matter and changes in inorganic mineral components (Miller and Donahue, 1995). For example, it has been shown that a relatively greater lipid content is found in soils with a low microbiological activity, whilst soil pH also affects lipids with strongly acidic environments exhibiting higher lipid contents (Fridland, 1976). Lipids can exist in a number of microenvironments within the soil being either free entities in the soil matrix, chemically bound components of humic material or physically adsorbed to clay particles (Jambu et al., 1978). In this paper we will consider the lipid compositions of soils sampled from the Rothamsted Classical Experiment. The research employed a combination of analytical techniques to investigate both free, solvent extractable components, and those associated with insoluble residues of the vegetation and soils in order to assess the fate of the various lipid classes present in the original biological inputs. The Rothamsted Classical Experiment. This site is the oldest and best-documented agricultural site in the world. Long-term experiments have been in progress for more than 150-years and records detailing histories of use are available. It consists of plots which have supported continuous growth of arable crops such as barley (Hoosfield Spring Barley) and wheat (Broadbalk), as well as continuous grass land (Park Grass Experiment) and other areas which have been allowed to revert back to natural woodland following cultivation (Broadbalk Wilderness and Geescroft Wilderness). Thus, this site provides a unique opportunity to study the effects of changing vegetation and agricultural practices on solvent soluble organic matter and thereby improve our understanding of the fate of soil lipids. The work focused on: (i) characterising the lipid composition of soils with well-defined biological inputs, (ii) assessing the fate of specific lipid classes in a soil profile relative to biomacromolecules, and (iii) determining the factors which control the preservation of soil organic matter. Lipid composition of soils with known recent histories of use. Broadbalk Wilderness is composed of three distinct sub-plots each of which exhibits a different type of overlying vegetation (i.e. woodland, grassland and mixed herband grassland). Analyses were made on total lipid extracts using high temperature gas chromatography (HT-GC) and HTGC mass spectrometry (HT-GC/MS). Samples of vegetation were examined in parallel with underlying soils in an effort to follow the fate of major plant components. A number of different compound classes were identified including, n-alkanes, nalkanols, (hydroxy-) n-alkanoic acids, steroids, triterpenoids, wax esters, mono-, diand triacylglycerols. Lipids from the woodland are dominated by inputs from leaf-derived components. In contrast, the lipid extracts of soils from the grassland and mixed vegetation area are markedly different reflecting the mixed vegetation cover dominated by grass species. Particularly important classes of compounds are the n-alkanols and wax esters, i.e. a very strong predominance exists of the Cz6 alkanol (and wax esters based on this alkanol) in the soils with grass input whereas the alkanols and wax esters of the woodland are dominated by the C24 homologue. These data quite clearly indicate that molecular signals derived from the standing vegetation are reflected in the soils. Fate of specific lipids in a soil prof i le.The main soil horizons from another reverted woodland (Geescroft Wilderness) were studied to determine the fate of