The Hamilton cycle is a cycle passing through every vertex exactly once.
(Dirac, 1952) If is a graph on vertices with and , then contains an Hamilton cycle.
For odd, consider the complete bi-partite graph . The minimum degree is and the graph does not have a Hamiltonian cycle. Why ?
For odd, consider the graph formed by attaching two copies of at a single vertex. What is the minimum degree and does the graph have a Hamiltonian cycle ? Modify the construction for even suitably.
Suppose is a counterexample to the theorem and be such a graph with maximal number of edges i.e., addition of an edge to creates a cycle. Let and hence will contain a Hamilton cycle . Thus is a simple path. Define sets and . Since , and further . Hence and assume that . Then is a Hamiltonian circuit in , contradicting our assumption. ∎
A straightforward modification of the above proof gives the following more general result.
(Ore, 1960) If is a graph on vertices such that for all , then is Hamiltonian.
We will now see a necessary condition for a graph to be Hamiltonian. For a subset , we denote as the graph obtained by removing all the vertices in and edges incident to those vertices. From our results on , we know that has at most components for and connected.
Is there a bound for for connected and ?
Suppose is Hamiltonian. Then has at most connected components for any .
Show the following graph provides a counterexample to the above theorem.
Let be the components of . Assume we start our Hamilton cycle at . ∎
Suppose is bi-partite and Hamiltonian. Then where are the partitions.
From previous theorem, the number of components of (which is ) is at most and so . By symmetry, we get the other inequality. ∎