# Derived subgroup of a group whose all it's Sylow subgroups are cyclic is abelian. [closed]

Let $G$ be a finite group such that all its Sylow subgroups are cyclic. Prove that $G'$ is abelian. (Here $G'$ denotes the derived subgroup.)

## closed as off-topic by Namaste, user61527, Amitesh Datta, M Turgeon, dfeuerSep 28 '13 at 3:35

This question appears to be off-topic. The users who voted to close gave this specific reason:

• "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – Namaste, Community, Amitesh Datta, M Turgeon, dfeuer
If this question can be reworded to fit the rules in the help center, please edit the question.

• This:groupprops.subwiki.org/wiki/… shows that $G$ has a cyclic subgroup $N$ with cyclic quotient (it's metacyclic). Since the derived subgroup must evidently be contained in $N$, it actually follows that $G'$ is cyclic. – Alex Youcis Sep 18 '13 at 6:55
• How advanced results are you allowed to use for this? – Tobias Kildetoft Sep 18 '13 at 6:57
• @TobiasKildetoft: Besides to your question,I think the title should have been more descriptive. – mrs Sep 18 '13 at 7:06

## 2 Answers

First let's prove the solvability of a group all of whose Sylow subgroups are cyclic - we need this to get an induction argument started.

Let $p$ be the smallest prime dividing the order of $G$ and let $P$ be a Sylow $p$-subgroup.. The choice of $p$ implies that the condition of Burnside's theorem is satisfied, that is $P\subset Z(N_G(P))$ (this is equivalent to $N_G(P)=C_G(P)$, and this follows from the fact that $N_G(P)/C_G(P)$ injects homomorphically in $\operatorname{Aut}(P)$, which is an abelian group of order $\varphi(|P|)$ and the fact that $P \subset C_G(P)$. This last inclusion follows from $P$ being abelian and hence $|N_G(P)/C_G(P)$| is not divisible by $p$.) So $G$ has a normal $p$-complement, say $N$. In other words, $G=NP$ and $N \cap P=1$.

The solvability follows easily now by induction on $|G|$, since $N$ is a proper normal subgroup of $G$ and inherits from $G$ all of its Sylow subgroups being cyclic. It follows by induction that $N$ is solvable. But $G/N \cong P$ is a $p$-group whence solvable, so after all, $G$ itself is solvable.

Now we are almost done, but need a small general observation.

Lemma Let $G$ be a group with both $G'/G''$ and $G''$ cyclic. Then $G'$ must be cyclic.

Proof. (sketch) $G/C_G(G'')$ injects homomorphically in $\operatorname{Aut}(G'')$. This last group is abelian and it follows that $G' \subset C_G(G'')$, thus $G'' \subset Z(G') \subset G'$. Since $G'/G''$ is cyclic, this implies $G'/Z(G')$ is cyclic, and it is well-known that this means $G'$ is abelian. But then $G''=1$, so $G'$ must in fact be cyclic. $\Box$

We are now ready to finish the proof. Since $G$ is solvable, $G'$ must be a proper subgroup and by induction on $|G|$ it follows that $G''$ is abelian. Obviously $G'/G''$ is also an abelian group. But both groups have cyclic Sylow subgroups (inherited from $G$ itself) and hence the condition of the previous lemma is satisfied. This concludes the final argument. $\Box$

• Good luck with your exam! And if you consider my answer as THE answer, please tick it as such. – Nicky Hekster Sep 20 '13 at 5:48

This is a non-trivial theorem and even something stronger is true.

Theorem. Suppose that all Sylow subgroups of the finite group $G$ are cyclic. Then both $G'$ and $G/G'$ are cyclic and $\gcd(|G'|,|G/G'|)=1$.

The proof depends heavily on the famous normal $p$-complement theorem of Burnside who showed that if a Sylow $p$-subgroup of a group $G$ is in the center of its normalizer then $G$ has a normal $p$-complement. This implies that if $p$ is the smallest prime dividing the order of a group $G$ and the Sylow $p$-subgroup is cyclic, then $G$ has a normal $p$-complement. It is exactly this result that can be used to prove that a group with only cyclic Sylow subgroups must be solvable. That is a first step.

Another result one needs for the "$\gcd$"-part of the above theorem is the following.

Lemma Let $P$ be a cyclic Sylow $p$-subgroup of $G$. Then $p$ divides at most one of the numbers $|G'|$ and $|G/G'|$.

Details of the proofs can be found for example in Chapter 5 (Transfer) of Marty Isaacs' book Finite Group Theory.

• Thank you very much, please explain more, I quickly need it. – mehranian Sep 19 '13 at 11:51
• @mehranian what do you need it for? – Tobias Kildetoft Sep 19 '13 at 11:59
• I have an exam for PHD and I guess this question will be given – mehranian Sep 19 '13 at 12:51
• @mehranian In that case you really ought to be able to do this yourself given the current amount of information given. – Tobias Kildetoft Sep 19 '13 at 18:09
• OK, I will give another answer with the proofs and I will leave out the gcd-part. – Nicky Hekster Sep 19 '13 at 20:51