Each revision of a source file is placed within a branch.  A branch is a parallel line of development and is used to separate different development tasks that operate on the same file at the same time.  By default, if no branch is specified, the HEAD or MAIN branch is used.  Entire branches can be filtered based on branch name.  Multiple branches can clutter a revision graph and reduce the performance of rendering the graph.  Older branches, for example, may not be important anymore and could be filtered out.  The filter can be either to include or exclude the branches that match the filter based on the show/hide toggle.