Wiki
Wiki

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

Updated


Statement

Setting (pp. 1, 4). a≺ba\prec b means b≡1(moda)b\equiv1\pmod a, and H(p)H(p) is the length of the longest prime chain p1≺⋯≺pk=pp_1\prec\cdots\prec p_k=p.

Passage in Section 1.3 (p. 4; the print gives it no label). Define the chain 2=q1≺q2≺⋯2=q_1\prec q_2\prec\cdots by letting qj+1q_{j+1} be the smallest prime congruent to 11 modulo qjq_j. Linnik's theorem gives qj+1≤qjLq_{j+1}\le q_j^L for some constant LL, hence H(qj)≥log⁡log⁡qjlog⁡LH(q_j)\ge\frac{\log\log q_j}{\log L}. The paper adds that it is conjectured that qj+1≤qj(log⁡qj)Cq_{j+1}\le q_j(\log q_j)^C for some fixed CC, and that this would imply the much stronger bound H(qj)≫log⁡qjlog⁡log⁡qjH(q_j)\gg\frac{\log q_j}{\log\log q_j}.

The paper uses the chain only to exhibit large values of H(p)H(p); it states no theorem about it and does not name the source of the conjecture.

Proof pointer

The Linnik bound is a citation (Linnik's theorem on the least prime in an arithmetic progression); the deduction log⁡qj≤Lj−1log⁡2\log q_j\le L^{j-1}\log2 is immediate by induction. The conjectured bound is not proved.

Read depth

Claims checked: the remark was read clause by clause on the print (p. 4). Nothing here is independently reviewed.

Dependencies

None in the corpus. External input: Linnik's theorem.

Source. Kevin Ford, Sergei V. Konyagin and Florian Luca, Prime chains and Pratt trees, Geom. Funct. Anal. 20 (2010), no. 5, 1231--1258, doi:10.1007/s00039-010-0089-0, arXiv:0904.0473; page numbers are those of the arXiv version 4 named on the source card.

Bears on

  • Problem 695: the chain q1<q2<⋯q_1<q_2<\cdots is a sequence of the kind the problem considers. The paper does not mention the problem. Reading the remark against its second question (whether some such sequence has pk≤exp⁡(k(log⁡k)1+o(1))p_k\le\exp(k(\log k)^{1+o(1)})): the conjectured qj+1≤qj(log⁡qj)Cq_{j+1}\le q_j(\log q_j)^C gives log⁡qj+1≤log⁡qj+Clog⁡log⁡qj\log q_{j+1}\le\log q_j+C\log\log q_j, so by induction log⁡qk=O(klog⁡k)\log q_k=O(k\log k) and the chain would answer that question yes. The conjecture is unproved, and the unconditional Linnik bound gives only log⁡qk≤Lk−1log⁡2\log q_k\le L^{k-1}\log2, far from the required growth. The remark does not bear on the first question.