Given a random cycle, how many conjugates of the cycle can be given? This is, given $\sigma$ how many $\alpha, \beta$ are such that $\alpha\sigma \alpha^{-1}=\beta.$

It is possible to calculate a conjugates, I found posts here that do so: Conjugate permutations in $S_n$ and / or $A_n$.

But is it possible to determine how many are in general? Theorem 5.2 here http://www.math.uconn.edu/~kconrad/blurbs/grouptheory/conjclass.pdf states that

All cycles of the same length in $S_n$ are conjugates

Then to determinate how many conjugates a cycle have, is to know how many conjugates cycles of the same length can be constructed. But how many are there? Aren't $\frac{n!}{r(n-r)!}$ r-cycles in $S_n$? I don't think this gives the number of conjugate cycles, but just r-cycles in general.


Every $r$-cycle is conjugate to every other $r$-cycle. This holds not just for all $r$, but for more complicated cycle structures. So in $S_4$ we have that $(12)(34)$ is conjugate to $(13)(24)$ because they both have 2 $2$-cycles. Thus to count the size of conjugacy classes, you do as you did to just count the number of such permutations based on cycle structure.

  • $\begingroup$ Sometimes the word cycle type is used in place of cycle structure. $\endgroup$ – Yacoub Kureh May 10 '18 at 0:51
  • $\begingroup$ So for each cycle there are $(r-1)$ conjugates to it? Then if there are $\frac{n!}{r(n-r)!}$ cycles there should be $\frac{n!(r-1)}{r(n-r)!}$ conjugates. $\endgroup$ – Cure May 10 '18 at 0:53

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.