Cycle lengths and minimum degree of graphs (1508.07912v1)

Abstract: There has been extensive research on cycle lengths in graphs with large minimum degree. In this paper, we obtain several new and tight results in this area. Let $G$ be a graph with minimum degree at least $k+1$. We prove that if $G$ is bipartite, then there are $k$ cycles in $G$ whose lengths form an arithmetic progression with common difference two. For general graph $G$, we show that $G$ contains $\lfloor k/2\rfloor$ cycles with consecutive even lengths and $k-3$ cycles whose lengths form an arithmetic progression with common difference one or two. In addition, if $G$ is 2-connected and non-bipartite, then $G$ contains $\lfloor k/2\rfloor$ cycles with consecutive odd lengths. Thomassen (1983) made two conjectures on cycle lengths modulo a fixed integer $k$: (1) every graph with minimum degree at least $k+1$ contains cycles of all even lengths modulo $k$; (2) every 2-connected non-bipartite graph with minimum degree at least $k+1$ contains cycles of all lengths modulo $k$. These two conjectures, if true, are best possible. Our results confirm both conjectures when $k$ is even. And when $k$ is odd, we show that minimum degree at least $k+4$ suffices. This improves all previous results in this direction. Moreover, our results derive new upper bounds of the chromatic number in terms of the longest sequence of cycles with consecutive (even or odd) lengths.