Methodological papers advise to collect moss surface samples either at the beginning or at the end of the flowering season. In reality, such collections occur often within the flowering season for purposes of vegetation description or because of accessibility of remote study areas. Here we test whether the season of moss surface sample collection has an influence on pollen spectra. Ten moss/litter samples were collected in different habitats of the natural reserve ‘Spasskaya Gora’ (Perm region, Russia) in July and September from the same locations exactly. The results show that pollen assemblages reflect well the present dominant vegetation of Pinus-Betula-forests and grasslands and demonstrate differences between open and forested areas as well as between abandoned fodder meadows and semi-natural hay meadows in use. Paired t-tests and Wilcoxon tests demonstrate that the dominant pollen taxa Pinus diploxylon-type, Betula and Poaceae as well as the rarefied number of pollen taxa do not differ significantly between seasons, while values of Artemisia are significantly higher in September. A significant decrease in pollen concentration in September indicates the importance of washing-out of pollen by precipitation. PCA results demonstrate the similarity of the pollen spectra between the seasons. Based on our results, we conclude that the season of surface sample collection does not significantly influence the pollen assemblages and appropriate field studies can be carried out in summer during the flowering season. However, we strongly recommend to collect bulk samples of mosses with basal parts and/or litter with surface soils in order to ensure representation of the average pollen signal for the previous years and to dilute local extremes in the pollen deposition.