Let G be generated by two cycles. One ihas order $n$ and moves the points $\{1,\ldots,n\}$ and the other has order $n-1$ and moves the points $\{1,\ldots,n-1\}$. I found following examples: $$ \begin{array}{ccl} n = 3 &:& S_3 \\ n = 4 &:& S_4 \\ n = 5 &:& S_5,\ C_5 \rtimes C_4\\ n = 6 &:& S_5,\ S_6 \\ n = 7 &:& S_7,\ C_7 \rtimes C_6\\ n = 8 &:& S_8,\ \operatorname{PSL}(3,2) \rtimes C_2\\ n = 9 &:& S_9 \\ n = 10 &:& S_{10} \\ n = 11 &:& S_{11},\ C_{11} \rtimes C_{10} \\ \end{array} $$

Is there a way to predict the structure of these groups for arbitrary $n$?

Edit Changed for $n=6 :S_5$ into $PGL(2,5)$ in accordance with the article mentioned in the answer from @Geoff Robinson.


With the particular choice $\alpha = (12 \ldots n)$ and $\beta = (12\ldots n-1)$ you will always get $\langle \alpha, \beta \rangle = S_{n}$ as one of the possibilities.This because $\langle \alpha, \beta \rangle = \langle \alpha, \alpha \beta^{-1}\rangle = \langle (12 \ldots n), ( n-1 n) \rangle.$ It is a standard fact, proved in many texts, that $S_{n} = \langle (12 \ldots n), (12) \rangle,$ and it is just a matter of relabelling letters to see that $S_{n} = \langle (12 \ldots n), ( n-1 n) \rangle.$ Equivalently, conjugate $\alpha$ and $\alpha \beta^{-1}$ by the permutation $\sigma$ which interchanges $j$ and $n+1-j$ for $1 \leq j \leq n$ ( and fixes $\frac{n+1}{2}$ if $n$ is odd).

Later edit: Since your group contains an odd permutation and is doubly transitive, I think that the possibilities (other than $S_{n}$) are included in the groups listed in Theorem 1.2 of https://arxiv.org/pdf/1209.5169.pdf ( by G.A. Jones). Note that doubly transitive groups are primitive ( though primitive groups are not always doubly transitive). Note that it is extremely rare in the examples given in this paper that the group contains both an $n$-cycle and an $n-1$-cycle.

  • $\begingroup$ I knew that already, but I had to include them to be complete, the main point of the question are the other groups. $\endgroup$ – Marc Bogaerts Apr 17 '17 at 19:31
  • $\begingroup$ OK. The groups you get will always be doubly transitive, and doubly transitive groups are reasonably well understood now ( but a full understanding requires the classification of finite simple groups). $\endgroup$ – Geoff Robinson Apr 17 '17 at 19:38
  • $\begingroup$ The examples I gave are exhaustive for the given $n$. The cases of the form $C_n \rtimes C_{n-1}$ can also be explained away for $n$ prime because of the structure of the automorphism group of $C_n$. Remain the special cases for $n=6$ and $n = 8$ and who knows for other values of $n$. $\endgroup$ – Marc Bogaerts Apr 17 '17 at 19:41
  • $\begingroup$ Indeed, I was inspired by this question. I had an answer for the $S_n$ case but had to delete it because I realized there were other groups falling under this description. $\endgroup$ – Marc Bogaerts Apr 17 '17 at 19:48
  • $\begingroup$ I have edited my answer to include some more information. $\endgroup$ – Geoff Robinson Apr 17 '17 at 19:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.