The question is as follows: Prove that in a graph $G$ a set of edges $X$ which is not contained in any spanning tree is a cycle (or possibly an edge disjoint union of cycles).
My thoughts: Proceed by contradiction.
Any help will be appreciated.
The question is as follows: Prove that in a graph $G$ a set of edges $X$ which is not contained in any spanning tree is a cycle (or possibly an edge disjoint union of cycles).
My thoughts: Proceed by contradiction.
Any help will be appreciated.
Your statement seems wrong. The good property is : if $X$ is not contained in any spanning tree, then $X$ contains a cycle.
Notice that if $X$ contains a cycle, $X$ isn't contained in any spanning tree (since a spanning tree is cycle-free), so that we couldn't prove any stronger property.
You can prove the property by contradiction :
First notice that for $X \subset E$ that does not contain a cycle, then $X$ is included in a spanning tree. Just add edges to $X$ as much as you can without creating a cycle, and you get a spanning tree. This is because, if a vertice $k$ is not on one of the edges of $X$, then you can add an edge from $X$ to $k$ without creating a cycle.
If you can't add any edge, this means $X$ is a spanning tree.
(Both Kruskal and Prim algorithm make us of this property).