diff --git a/src/traversals/dfs.jl b/src/traversals/dfs.jl
index c5fefe9..44033fc 100644
--- a/src/traversals/dfs.jl
+++ b/src/traversals/dfs.jl
@@ -25,36 +25,3 @@ end
 function dfs_parents(graph::AbstractNamedGraph, vertex; kwargs...)
   return _dfs_parents(graph, vertex; kwargs...)
 end
-
-#Given a graph, split it into its connected components, construct a spanning tree over each of them
-# and take the union (adding in edges between the trees to recover a connected graph)
-function spanning_forest(
-  g::AbstractNamedGraph;
-  tree_constructor=g ->
-    undirected_graph(bfs_tree(g, first(keys(sort(eccentricities(g); rev=true))))),
-)
-  component_vertices = connected_components(g)
-  return reduce(union, (tree_constructor(g[vs]) for vs in component_vertices))
-end
-
-#Given an undirected graph g with vertex set V, build a set of forests (each with vertex set V) which covers all edges in g
-# (see https://en.wikipedia.org/wiki/Arboricity) We do not find the minimum but our tests show this algorithm performs well
-function build_forest_cover(
-  g::AbstractNamedGraph;
-  tree_constructor=g ->
-    undirected_graph(bfs_tree(g, first(keys(sort(eccentricities(g); rev=true))))),
-)
-  @assert !NamedGraphs.is_directed(g)
-  edges_collected = edgetype(g)[]
-  remaining_edges = edges(g)
-  forests = NamedGraph[]
-  while !isempty(remaining_edges)
-    g_reduced = rem_edges(g, edges_collected)
-    g_reduced_spanning_forest = spanning_forest(g_reduced; tree_constructor)
-    push!(edges_collected, edges(g_reduced_spanning_forest)...)
-    push!(forests, g_reduced_spanning_forest)
-    setdiff!(remaining_edges, edges(g_reduced_spanning_forest))
-  end
-
-  return forests
-end