A class of descendant \(p\)-groups of order \(p^{9}\) and Higman's PORC conjecture (Q324241): Difference between revisions

The number of (isomorphism classes of) groups of order \(p^{4}\), where \(p\) is a prime, is \(15\) if \(p\) is odd, \(14\) for \(p = 2\). The number of groups of order \(p^{5}\) depends, for \(p > 3\), on \(\gcd(p-1, 12)\). The number of groups of order \(p^{6}\) depends, for \(p > 3\), on \(\gcd(p-1,60)\). \textit{G. Higman} has conjectured [Proc. Lond. Math. Soc., III. Ser.~10, 566--582 (1960; Zbl 0201.36502)] that the number \(f_{n}(p)\) of groups of order \(p^{n}\) is polynomial on residual classes (PORC), that is, for a fixed \(n\) there are a positive integer \(N\), and finitely many polynomials \(g_{i}(x)\), \(0 \leq i < N\), such that \(f_{n}(p) = g_{i}(p)\) if \(p \equiv i \pmod{N}\). Higman's conjecture [loc. cit.] has been shown to be true up to \(n \leq 7\). \textit{M. R. Vaughan-Lee} has shown [``Groups of order \(p^8\) and exponent \(p\)'', Int. J. Group Theory 4, No. 4, 25--42 (2015)] that the number of groups of order \(p^{8}\) and exponent \(p\) is PORC. When classifying groups of order \(p^{n}\) (see \textit{E. A. O'Brien}'s paper [J. Symb. Comput. 9, No. 5--6, 677--698 (1990; Zbl 0736.20001)]), one starts from a classification of the groups of order \(p^{k}\), for \(k < n\). For each of these groups, one builds its immediate descendants. Here, a finite \(p\)-group \(G\) of \(p\)-class \(c > 1\) is said to be an immediate descendant of \(G/G_{c}\), where \(G_{c}\) is the \(c\)th term of the lower \(p\)-central series of \(G\). The groups of order \(p^{n}\) will be all these immediate descendants, plus the elementary abelian group. The main result of this paper is that there is a group of order \(p^{8}\), exponent \(p\), and nilpotence class \(2\), such that the number of its immediate descendants of order \(p^{9}\) and exponent \(p\) is not PORC. An example of this behaviour with a group of order \(p^{9}\) and its descendants of order \(p^{10}\) had been constructed by \textit{M.P.F. du Sautoy} and \textit{M. R. Vaughan-Lee} [J. Algebra 361, 287--312 (2012; Zbl 1267.20025)]. The author notes the total number of groups of order \(p^{9}\) and \(p^{10}\) may still be PORC. But it is surely surprising and interesting to find that in the natural partition in terms of immediate descendants, one of the summands is not PORC.