English

Notes on Spreads of Degrees in Graphs

Combinatorics 2018-06-22 v1

Abstract

Perhaps the very first elementary exercise one encounters in graph theory is the result that any graph on at least two vertices must have at least two vertices with the same degree. There are various ways in which this result can be non-trivially generalised. For example, one can interpret this result as saying that in any graph GG on at least two vertices there is a set BB of at least two vertices such that the difference between the largest and the smallest degrees (in GG) of the vertices of BB is zero. In this vein we make the following definition. For any BV(G)B\subset V(G), let the spread sp(B)sp(B) of BB be defined to be the difference between the largest and the smallest of the degrees of the vertices in BB. For any k0k\geq 0, let sp(G,k)sp(G,k) be the largest cardinality of a set of vertices BB such that sp(B)ksp(B)\leq k. Therefore the first elementary result in graph theory says that, for any graph GG on at least two vertices, sp(G,0)2sp(G,0)\geq 2. In this paper we first give a proof of a result of Erd\" os, Chen, Rousseau and Schelp which generalises the above to sp(G,k)k+2sp(G,k)\geq k+2 for any graph on at least k+2k+2 vertices. Our proof is short and elementary and does not use the famous Erd\" os-Gallai Theorem on vertex degrees. We then develop lower bounds for sp(G,k)sp(G,k) in terms of the order of GG and its minimum, maximum and average degree. We then use these results to give lower bounds on sp(G,k)sp(G,k) for trees and maximal outerplanar graphs, most of which we show to be sharp.

Keywords

Cite

@article{arxiv.1806.08303,
  title  = {Notes on Spreads of Degrees in Graphs},
  author = {Yair Caro and Josef Lauri and Christina Zarb},
  journal= {arXiv preprint arXiv:1806.08303},
  year   = {2018}
}