© The Author(s) 2019. The management of the remainder of Europe’s once extensive forests is hampered by a poor understanding of the character of the vegetation and drivers of change before the onset of clearance for farming. Pollen data indicate a closed-canopy, mixed-deciduous forest, contrasting with the assertion that large herbivores would have maintained a mosaic of open grassland, regenerating scrub and forested groves. Coprophilous fungal spores from sedimentary sequences are increasingly used as a proxy for past herbivore impact on vegetation, but the method faces methodological and taphonomical issues. Using pollen trap data from a long-running experiment in Chillingham Wild Cattle Park, UK, we investigate the first steps in the mechanisms connecting herbivore density to the incorporation of fungal spores in sediments and assess the effects of environmental variables on this relationship. Herbivore utilisation levels correlate with dung fungal spore abundance. Chillingham is densely populated by large herbivores, but dung fungal spore influx is low. Herbivores may thus be present on the landscape but go undetected. The absence of dung fungal spores is therefore less informative than their presence. Dung fungal spores likely enter the sediment record through a different pathway from wind-borne pollen and thus dung fungal abundance is better expressed as influx rates than as percentage of total pollen. Landscape openness, vegetation type and site wetness do not distort the impact of utilisation levels on dung fungal spore representation. However, dung fungal spore influx varies markedly between seasons and years. Spores travel, leading to a background level of spore deposition across the landscape, and at times a depletion of spores, especially under wet weather conditions. Animal behaviour, as well as husbandry practices, can lead to the accumulation of dung, and thus fungal spores, in specific locations on the landscape that do not directly reflect grazing pressure.