3
$\begingroup$

I was told yesterday that in a system of Sylow $p$-subgroups of a finite group $G$, if, $\{S_1,S_2, \cdots, S_n\}$ make up the system, it can happen that, say, the intersection of $S_1$ and $S_2$ has order $p^k$, while the intersection of $S_3$ and $S_4$ has size $p^j$, with $k \neq j$.

1) My first question is if some examples are available of this. I think I can't find examples in the alternating groups and symmetric groups, due to multiple transitivity.

2a) If $j = k$ in the above, must the intersections $S_1 \cap S_2$ and $S_3 \cap S_4$ be isomorphic?

2b) If those two intersections in 2a) above are isomorphic, then must there be an inner automorphism mapping $S_1$ to $S_3$ and $S_2$ to $S_4$? That is, must Sylow systems have as much conjugacy as "possible", in addition to simple transitivity?

Thanks.

$\endgroup$
  • 1
    $\begingroup$ You should be able to find counterexamples for all claims by choosing the right groups $G$ and $H$ and looking at $G\times H$. The Sylow subgroups of $G\times H$ are the direct products of the Sylows of $G$ with those of $H$, so just pick the latter ones for your needs. For 1) and 2b) you can take $p=2$ and $G=H=S_3$. For 2a) $p=2$, $G=Z_5\rtimes Aut(Z_5)$ and $H=A_5$ (or another group with 2-Sylows isomorphic to $Z_2\times Z_2$ and two 2-Sylows intersecting trivially). $\endgroup$ – j.p. Apr 14 '15 at 6:18
  • 1
    $\begingroup$ In the alternating group $A_7$ the $3$-Sylow subgroups $S_1=S_3=\langle (123), (456)\rangle$, $S_2=\langle (123), (457)\rangle$ and $S_4=\langle (124), (356)\rangle$ answer 1). For 2) the examples would become biggish, but should exist. $\endgroup$ – j.p. Apr 14 '15 at 7:51
4
$\begingroup$

Let $C_3$ be the cyclic group of order $3$ and $D_8$ the dihedral group of order $8$, and let $G$ be the regular wreath product $C_3 \wr D_8$, which has order $3^8 \cdot 8$.

There are $3^7 = 2187$ Sylow $2$-subgroups of $G$, and all subgroups of $P$ arise as intersections of two of these. Since $P$ has non-isomorphic subgroups of order $4$, this answers Questions 1 and 2a) in the negative.

I found examples of Sylow $2$-subgroups $P_1,P_2,P_3$ with $|P_1 \cap P_2| = |P_1 \cap P_3| = 2$, such that $P_2$ and $P_3$ are not conuugate in $N_G(P_1)$, so the answer to 2b) is also no.

$\endgroup$

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.