Predation of tree seeds can be a major factor structuring plant communities. We present a three year study on tree seed survival on experimental dishes in an old-growth forest in central Europe in Austria. We addressed species specifi c, spatial and temporal aspects of post-dispersal seed predation. Seeds of Norway spruce Picea abies , European beech Fagus sylvatica , and silver fi r Abies alba were exposed on dishes in diff erent types of exclosures which allowed access only to specifi c guilds of seed predators. Removal experiments were carried out in two old-growth forests and a managed forest (macro-sites), including micro-sites with and without cover of ground vegetation. We conducted the experiment in three consecutive years with a mast year of beech and spruce before the fi rst year of the study. Th e seed removal experiments were combined with live trapping of small mammals being potential seed predators. Our experiments showed a distinctly diff erent impact of diff erent predator guilds on seed survival on the dishes with highest removal rates of seeds from dishes accessible for small mammals. We observed diff ering preferences of small mammals for the diff erent tree species. Seed survival in different macro-and micro-habitats were highly variable with lower seed survival in old growth forests. In contrast to our assumption, and in contrast to the satiation hypothesis which assumes higher seed survival in and directly after mast years, seed survival was lower in the year following the mast year of beech when a population peak of small mammals occurred and higher in intermast periods when subsequently small mammal population crashed. Th is suggests a higher importance of sporadic masting shortly after mast years in intermast periods for establishment of forest trees provided that pollination effi ciency is high enough in such years. Combined with the high seed mortality observed after the mast year, this corroborates the important role of seed predation for forest dynamics. An altered synchrony or asynchrony of masting of diff erent tree species and changed masting frequencies through climate change may thus lead to strong and non-linear eff ects on forest dynamics.

Seed hoarding animals may distinctly impact individual fitness of plants as well as plant community dynamics in terms of differential seed consumption and seed dispersal. Addressing methodological challenges and constraints in understanding the role of seed dispersal mutualisms, we tested different seed tagging methods for relatively small seeds. We selected two study areas with dense ground vegetation, irregular terrain, and seed predator guilds with comparatively high dispersal and seed removal potential. We tested three different tagging methods, offering untagged and tagged seeds on experimental dishes: (1) wire threads with plastic flags fixed with solvent-free glue, (2) wire threads with plastic flags twisted around a beechnut, (3) and radio-transmitters fixed with solvent-free glue. Seed predators did not show any preferences for tagged or untagged seeds or for a certain tagging method at any point of time. Instead, both tagged seeds and untagged control seeds were removed from the dishes to a high degree (up to 100%) after 25-35 days. In both study areas, seeds with radio-transmitters experienced longer transport distances with a maximum of nearly 60 m and distances also differed between study areas. In one study area, higher portions of radio-tagged seeds could be recovered than for the flag-tagged seeds. Significantly more seeds tagged with radio-transmitters were cached than expected. Given the advantages and limitations of each approach, a combination of flag-tagging and radiotelemetry (not on the same seed) may be the best solution for tracking seed fate in dense vegetation or irregular terrain.

Herbivory of invertebrates and rodents on tree seedlings in Northern temperate forests has been largely underestimated in the past and rarely addressed in studies. There is evidence that masting‐cycles of forest trees and subsequent rodents outbreaks become more frequent, leading to enhanced browsing pressure on forest regeneration. Consequently, studies exploring tree seedling mortality caused by different herbivores are required. We conducted a study on tree seedling losses in the Wilderness Area Dürrenstein, Austria on two old‐growth and one managed forest site. We conducted an exclosure experiment in two subsequent study years, using transplanted two‐month‐old tree seedlings of Picea abies, Abies alba, and Fagus sylvatica. Exclosures allowed access for invertebrates, invertebrates/rodents, and all potential herbivores. We also installed total exclosures, recording seedling losses due to climatic factors or pathogens. We calculated ZINB regressions assuming that seedling losses and seedling survival are distinct processes. Our results did not reveal fully consistent trends for both study years. The factor “treatment”, however, was significant in the ZINBs in both study years. Herbivory was a driving factor for tree regeneration in the studied forests depending on specific conditions in different years or tree species. In 2005, losses in the total exclosures and herbivory of rodents and invertebrates in the other treatments reached comparable extent. Loss rates differed not only in terms of forest management, but also between the two old‐growth forests. This, in connection with different herbivory of different tree species (higher losses of A. alba), strongly suggests the necessity of considering different herbivore guilds as drivers of tree species composition in late successional mountain forests.