Wiki
Wiki

Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.

Updated


Claim. For k,ℓ≥2k,\ell\ge2 let Qk,ℓQ_{k,\ell} be the tree obtained from a path P4P_4 on four vertices by appending a star K1,k−2K_{1,k-2} at one end and a star K1,ℓ−2K_{1,\ell-2} at the other; its classes have kk and ℓ\ell vertices. Lemma 4.1 (p. 252) of S. A. Burr and P. Erdős, Extremal Ramsey theory for graphs, Utilitas Math. 9 (1976), 247--258, which writes the tree as Sk,ℓS_{k,\ell}, states that

r(Qk,ℓ)=max⁡(2k−1, k+2ℓ−1)(k≥ℓ≥2).r(Q_{k,\ell})=\max(2k-1,\,k+2\ell-1)\qquad(k\ge\ell\ge2).

At classes 2k2k and kk this is r(Q2k,k)=4k−1r(Q_{2k,k})=4k-1 for every k≥2k\ge2: the equality of Problem 549 holds for the tree made of a path on four vertices with stars on 2k−12k-1 and k−1k-1 vertices at its ends. The lower bound is Burr's 1974 coloring bound; the upper bound takes a red K1,k−1∪K1,ℓK_{1,k-1}\cup K_{1,\ell}, which exists by a result of Rosta, r(K1,k−1∪K1,ℓ)=max⁡(2k−1,k+2ℓ−1)r(K_{1,k-1}\cup K_{1,\ell})=\max(2k-1,k+2\ell-1), quoted in the paper from Burr's 1974 survey, and completes it to a monochromatic Qk,ℓQ_{k,\ell} by a short case analysis. The statement is recorded on the result page Lemma 4.1 of the library home burr_1976_extremal_ramsey_theory_graphs; Erdős, Faudree, Rousseau and Schelp (1982, p. 284) credit the result to this paper.

Covers. For every k≥2k\ge2, the tree with classes 2k2k and kk formed from P4P_4 by the stars K1,2k−2K_{1,2k-2} and K1,k−2K_{1,k-2} at its ends has R(T)=4k−1R(T)=4k-1. Every other tree with these classes is outside it; the problem's equality fails for the double stars, as the full claim pages record.

Depends on. Nothing in this wiki; the upper bound uses Rosta's result on r(K1,k−1∪K1,ℓ)r(K_{1,k-1}\cup K_{1,\ell}), unpublished and quoted through Burr's 1974 survey (Lecture Notes in Math. 406), which this corpus does not hold.

Acceptance. Refereed: the paper is a journal publication in Utilitas Mathematica, volume 9 (1976), received 5 November 1974, the refereed evidence; the record carries no month, so this page is dated to the first day of the year. The site's curator lists this family in the commentary, but the label DISPROVED credits the disproof, not this case, so reviewed is not listed.