Diaspore and shoot size as drivers of local, regional and global bryophyte distributions


  • Editor: Martin Sykes

Correspondence: Risto Virtanen, Department of Biology, P.O. Box 3000, FI-90014, University of Oulu, Finland.

E-mail: risto.virtanen@oulu.fi



Ecological theory provides divergent views about how patterns of bryophyte occurrence are related from local to global scales. Here, I test the hypotheses that, based on the high dispersal capacity of bryophytes, patterns of occurrence are similar from local to global scales and independent of variation in diaspore and size traits, or alternatively that the patterns are dissimilar and depend on traits that are important for dispersal and competition.




The occurrence patterns of 28 bryophyte species and their relationship to diaspore and shoot size were analysed for three study systems: (1) a local metacommunity on erratic calcareous boulders; (2) a regional study system in the biogeographical provinces of the Nordic countries; and (3) based on data available from the Global Biodiversity Information Facility (GBIF).


Contrary to expectations, bryophyte occurrence was not similar among local or global study systems, but regional occurrences matched both local and global occurrences. Contrary to neutral theoretical assumptions, the occurrences depended on traits. In the local metacommunity, there was a tendency towards a negative relationship between occurrence and diaspore size: small and large species tended to be rare, whereas species of intermediate size tended to be widespread. In the regional study system, species with large diaspores tended to be widespread, with frequency of occurrence positively correlated with shoot length.

Main conclusions

In contrast to patterns detected for micro-organisms, local and global frequencies of occurrence for bryophytes are not necessarily similar. However, occurrences at the regional scale may be useful for predicting both local and global occurrences. Links between species traits important for dispersal and competition on the one hand, and occurrences on the other, supports the utility of these traits for the analysis of distributions and questions the adequacy of assumptions of functional equivalence.