1. PF Contest - Win "Conquering the Physics GRE" book! Click Here to Enter
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Isomorphism question

  1. May 25, 2010 #1
    1. The problem statement, all variables and given/known data
    Let H and K be normal Subgroups of a group G s.t H intersect K = {e}. Show that G is isomorphic to a subgroup of G/H + G/K.

    2. Relevant equations
    G/H+G/K= direct product of G/H and G/K.

    3. The attempt at a solution

    Lets define are mapping f:G to G/H+G/K by f(g)=(gH,GK). Need to show that this is a homomorphism. Obviously f is a well defined function (Do I need any more explanation here?). Operation Preserving- Let x,y be elements of G. Then f(xy)= (xyH,xyK)= (xHyH,xKyK)= (xH,XK),(yH,yK)=f(x)f(y) ( I need help justifying my step on this)
    Thus f is a homomorphism. Since kerf ={g element of G| (gH,gK)=(H,K)}, then ker f={e}. Thus f is injective (one to one).
    So By 1st isomorphism theorem, G/Ker f is isomorphic to f(G). Since f(G) is a subgroup of G/H+G/K (since f is one to one), then G is isomorphic to a subgroup of G/H + G/K (since G/Kerf= G because kerf={e}, is this right?)
    My question is for clarification, does this seem like I have proven that G is isomorphic to a subgroup of G/H + G/K, or I have I proven that G is isomorphic to G/H + G/K ?
  2. jcsd
  3. May 26, 2010 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Your proof looks good. A general rule of thumb for well-definedness is that you only need to be concerned when your domain involves picking representative elements of cosets (and then whether the function is independent of representative). In this case you're not, only the codomain involves cosets so you're OK.

    For the homomorphism part, where are you stuck? Let's look at it slowly

    f(xy)= (xyH,xyK) This is how f is defined

    = (xHyH,xKyK) This is by definition of multiplication by cosets

    = (xH,XK),(yH,yK) This is by the definition of multiplication for the product of two groups

    =f(x)f(y) this is by the definition of f again

    Then once you have that kerf is trivial, G/kerf = G is isomorphic to the image of f by the first isomorphism theorem. As to whether G is isomorphic to G/H + G/K, this will only be true if the image of f is all of G/H + G/K. Is this always true?
  4. May 26, 2010 #3
    Thanks for the help, I just needed some clarification.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook