The paper deals with the grammatical category of number in Czech. The basic semantic opposition of singularity and plurality is proposed to be enriched with a (recently introduced) distinction between a simple quantitative meaning and a pair/group meaning.
After presenting the current representation of the category of number in the multi-layered annotation scenario of the Prague Dependency Treebank 2.0, the introduction of the new distinction in the annotation is discussed. Finally, we study an empirical distribution of preferences of Czech nouns for plural forms in a larger corpus.