logo

Walks, Trails, Paths, and Cycles in Graph Theory 📂Graph Theory

Walks, Trails, Paths, and Cycles in Graph Theory

Definition 1

Let there be a given graph GG.

  1. A finite sequence of edges is called a walk and is denoted as follows: v0v1,v1v2,,vm1vmv0v1v2vm1vm v_{0} v_{1} , v_{1} v_{2} , \cdots , v_{m-1} v_{m} \\ v_{0} \rightarrow v_{1} \rightarrow v_{2} \rightarrow \cdots \rightarrow v_{m-1} \rightarrow v_{m} Here, v0v_{0} is called the initial vertex, vmv_{m} is called the final vertex, and mm is called the length.
  2. If all edges in a walk are different, it is called a trail.
  3. If all vertices in a walk are different, it is called a path.
  4. If the initial and final vertices of a walk are the same, it is called closed.
  5. A closed path is called a cycle.

These concepts can be similarly defined for directed graphs as well, and can also be defined for infinite graphs as follows:

Let GG be an infinite graph. In the definitions below, walks can be replaced with trails, paths, cycles. 6. A (finite) walk is exactly the same as a walk in a conventional graph. 7. A one-way infinite walk is defined as follows: v0v1v2 v_{0} \to v_{1} \to v_{2} \to \cdots 8. A two-way infinite walk is defined as follows: v2v1v0v1v2 \cdots \to v_{-2} \to v_{-1} \to v_{0} \to v_{1} \to v_{2} \to \cdots

Explanation

In the definition of a path, the initial and final vertices are exceptions. In other words, if v0=vmv_{0} = v_{m}, it can be a path, hence a ‘closed path’ can exist. Meanwhile, in the definitions of a path and a trail, if all vertices are different, then all edges must also be different, so a path is also a trail.

A cycle is, simply put, a ’loop shape’ found in a graph, and it’s a concept widely used throughout graph theory. For example, the following figure shows that there are a total of three cycles in the graph:

20200213\_155101.png

Furthermore, the following theorem is known about cycles.

Theorem

If every vertex of a finite graph GG has a degree of 22 or more, then GG contains a cycle.

Proof

Since GG would obviously contain a cycle if it has loops or multiple edges, we assume that GG is a simple graph.

Fix a random vertex v0V(G)v_{0} \in V(G) and consider the following path: v0v1v2 v_{0} \to v_{1} \to v_{2} \to \cdots As the degree of every vertex is 22 or more, according to the assumption, it’s guaranteed that we can continue to select any vertex adjacent to the previous vertex viv_{i}, excluding vi1v_{i-1}. However, since GG is a finite graph, we will inevitably select a vertex vkv_{k} that is already part of the path. Thus, the path vkvkv_{k} \to \cdots \to v_{k}, whose initial and final vertices are vkv_{k}, is confirmed to exist as a cycle in GG.


  1. Wilson. (1970). Introduction to Graph Theory: p27. ↩︎