1
$\begingroup$

The problem is stated as this:

"Prove that if $$G:=G_1\times G_2 \times \dots\times G_s \cong H_1 \times H_2 \times \dots\times H_t=:H,$$ where each $G_i$ and $H_i$ is finite simple group, then $s=t$ and there exists $\sigma\in S_t$ such that $G_j \cong H_{\sigma(j)}$ for $j=1,\dots,t$."

This comes after first being introduced to concept of solvable groups and composition series. We are given the Jordan-Hölder Theorem as follows:

Any two composition series for a finite group have the same length. Further, there exists a one-to-one correspondence between composition factors of the two composition series under which corresponding composition factors are isomorphic.

As far as I understand, each $G_i$ being simple implies that the only composition series for $G$ is $$G \trianglelefteq e_G$$ where $e_G=(e_{G_1},\dots,e_{G_s})$, and similarly for $H$. Is this correct? If it is, I am unsure how to apply the Jordan-Hölder Theorem in this instance, since $G$ and $H$ are merely isomorphic.

Can anyone offer any hints or a solution? Thank you for help.

$\endgroup$
3
$\begingroup$

HINT: $$G_1 \le G_1 \times G_2 \le G_1 \times G_2 \times G_3 \le \dots \le G_1 \times \cdots \times G_n$$ is a composition series for $G$. You can see this noting that $G_1 \times \cdots \times G_{k-1} $ is an epimorphic image of $G_1 \times \cdots \times G_k$ simply by the projection, and the kernel is $1 \times \cdots \times 1 \times G_k \cong G_k$.

Now, do the same thing with the $H_i$s and use Jordan-Hölder.

$\endgroup$
  • $\begingroup$ When you say $G_1 \le G_1\times G_2\times\dots$, do you mean $G_1\times { e_{G_2} }\times {e_{G_3}}\times \dots \times {e_{G_s}}\le G_1\times G_2\times {e_{G_3}}\times \dots \times {e_{G_s}}\le \dots $ ? Because $G_1$ is not a subgroup of $G_1 \times G_2$, is it? Also, can I apply Jordan-Hölder across two groups which are isomorphic? Thanks for the help.! $\endgroup$ – user279685 Oct 12 '15 at 22:24
  • $\begingroup$ Everybody thinks of $G$ as a subgroup of $G \times H$, although you are correct because the subgroup is actually $\{(g,1_H): g \in G \}$. Strangely enough, your question about isomorphic groups having isomorphic composition factors was asked very recently: math.stackexchange.com/questions/1477230 $\endgroup$ – Derek Holt Oct 13 '15 at 10:57

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.