Results 1 to 9 of 9

Math Help - Cycle type of non-disjoint permutation

  1. #1
    Newbie
    Joined
    May 2009
    Posts
    9

    Cycle type of non-disjoint permutation

    What is the cycle type of the permutation (1265)(1365)(2346) \in S6 ? I feel kind of stupid, but I have no clue.
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Senior Member TheAbstractionist's Avatar
    Joined
    Apr 2009
    Posts
    328
    Thanks
    1
    Quote Originally Posted by knarQ View Post
    What is the cycle type of the permutation (1265)(1365)(2346) \in S6 ? I feel kind of stupid, but I have no clue.
    (1\,2\,6\,5)(1\,3\,6\,5)(2\,3\,4\,6)\ =\ (1\,3\,4)(2\,5). (Note that the permutation fixes 6.) Do you know how to work this out?
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Newbie
    Joined
    May 2009
    Posts
    9
    I'm afraid I don't.
    Follow Math Help Forum on Facebook and Google+

  4. #4
    Junior Member
    Joined
    May 2009
    Posts
    27
    Quote Originally Posted by TheAbstractionist View Post
    (1\,2\,6\,5)(1\,3\,6\,5)(2\,3\,4\,6)\ =\ (1\,3\,4)(2\,5). (Note that the permutation fixes 6.) Do you know how to work this out?
    EDIT: This is wrong: I multiplied from left to right rather than from right to left.
    Correct me if I'm wrong, but shouldn't it be
     (1~3~2~5~4~6)

    as we have:

    1\mapsto 2\mapsto 2 \mapsto 3

    3\mapsto 3\mapsto 6 \mapsto 2

    2\mapsto 6\mapsto 5 \mapsto 5

    5\mapsto 1\mapsto 3 \mapsto 4

    4\mapsto 4\mapsto 4 \mapsto 6

    6\mapsto 5\mapsto 1 \mapsto 1

    ?
    Last edited by amitface; June 4th 2009 at 08:30 AM.
    Follow Math Help Forum on Facebook and Google+

  5. #5
    Newbie
    Joined
    May 2009
    Posts
    9
    I still don't get it. I know that the cycle type for a disjoint permutation, say (1\,2)(3\,4\,5\,6) is [2^1\,4^1], but how does it work for this non-disjoint one?
    Follow Math Help Forum on Facebook and Google+

  6. #6
    Junior Member
    Joined
    May 2009
    Posts
    27
    EDIT: This is incorrect.

    You can write the non-disjoint one as a disjoint one using the method that I described above.

    So assuming I didn't make any serious mistakes, we'd have:

    (1\,2\,6\,5)(1\,3\,6\,5)(2\,3\,4\,6)\ =(1~3~2~5~4~6),

    and you can find the cycle type of that permutation.
    Follow Math Help Forum on Facebook and Google+

  7. #7
    Junior Member
    Joined
    May 2009
    Posts
    27
    I apologize. I multiplied incorrectly. The Abstractionist had it right. You should still be able to find the cycle type from his answer though.
    Follow Math Help Forum on Facebook and Google+

  8. #8
    Newbie
    Joined
    May 2009
    Posts
    9
    Quote Originally Posted by TheAbstractionist View Post
    (1\,2\,6\,5)(1\,3\,6\,5)(2\,3\,4\,6)\ =\ (1\,3\,4)(2\,5). (Note that the permutation fixes 6.) Do you know how to work this out?
    Ok, so the cycle type would be [3^1\,2^1\,1^1] ?

    Next i would like to write that same permutation (1\,2\,6\,5)(1\,3\,6\,5)(2\,3\,4\,6) as a product of transpositions. My Biggs book explains this quite poorly so any help is appreciated.
    Follow Math Help Forum on Facebook and Google+

  9. #9
    Super Member Gamma's Avatar
    Joined
    Dec 2008
    From
    Iowa City, IA
    Posts
    517
    All you need to do is multiply these cycles out, and I think that is the problem, I am guessing you are not used to cycle notation. It can be a little confusing at first, but in the end it is waaay faster. The thing to remember is that each individual thing inside of parentheses is a permutation, which is a function that simply rearranges the set of n objects. The operation in S_n is function composition. As with normal functions for some reason we compose from right to left g \circ f (x)=g(f(x)). You do f first, and then g which seems backwards as we read left to right.

    Anyway, the point is within each set of () you read left to right, but you do the permutations from right to left.
    (1 2 6 5)(1 3 6 5)(2 3 4 6)

    Start with 1 and see where it goes. (2 3 4 6) does not affect 1, so carry on to the next one we see (1 3 6 5) 1 \rightarrow 3 because 3 follows the 1. Now check to see what (1 2 6 5 ) does to 3... nothing, so this whole set of three cycles takes 1 to 3.

    1\rightarrow 3

    Now you check to see what happens to 3.
    (2 3 4 6) takes 3 to 4.
    (1 3 6 5) does not affect 4
    (1 2 6 5) does not affect 4

    Thus we get overall
    3 \rightarrow 4

    Now check on 4
    4\rightarrow 6 \rightarrow 5 \rightarrow 1
    This completes the cycle, so write ( 1 3 5 ) as the Abstractionist pointed out.

    Now the smallest number not included in this cycle is 2, check and see what happens to 2.
    2 \rightarrow 3\rightarrow 6\rightarrow 5
    Now see what happens to 5.
    5 \rightarrow 5 \rightarrow 1\rightarrow 2
    This completes the cycle so write (2 5)

    The only one missing is 6, so check on this.
    6\rightarrow 1 \rightarrow 6 so it is fixed as he points out.

    So you are left with the final disjoint cycle decomposition theAbstractionist gave you:

    (1 3 4) (2 5) (6)

    It can be shown this disjoint cycle decomposition is unique up to reordering (disjoint cycles commute) and up to cycling within each cycle [(1 2 3) = (2 3 1)= (3 1 2)]. So you will see this cycle type is perserved. In fact the cycle type is even perserved under conjugation.
    Last edited by Gamma; June 4th 2009 at 11:07 AM. Reason: Left out the punchline
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Permutation with non-disjoint cycles
    Posted in the Number Theory Forum
    Replies: 6
    Last Post: March 27th 2011, 10:08 AM
  2. Cycle of generalized eigenvectors disjoint
    Posted in the Advanced Algebra Forum
    Replies: 2
    Last Post: December 3rd 2009, 11:48 PM
  3. Subgroups involving disjoint cycle notation.
    Posted in the Advanced Algebra Forum
    Replies: 8
    Last Post: January 1st 2009, 08:18 AM
  4. prove permutation is a cycle
    Posted in the Advanced Algebra Forum
    Replies: 1
    Last Post: September 15th 2008, 06:12 PM
  5. Replies: 2
    Last Post: December 9th 2007, 02:33 PM

Search Tags


/mathhelpforum @mathhelpforum