Pronormality is normalizing join-closed: Difference between revisions

From Groupprops
No edit summary
Line 46: Line 46:
#  Now consider <math>z = yx</math>. We claim that <math>(HK)^z = (HK)^x</math>('''No problem data used'''):. We have <math>H^z = H^{yx} = (H^y)^x = H^x</math> (since <math>H^y = H</math>). We already know that <math>H^x = H^g</math> (step (1)), so <math>H^z = H^g</math>. Similarly, <math>K^z = (K^y)^x = (K^{gx^{-1}})^x = K^{gx^{-1}x} = K^g</math>. Thus, <math>z</math> acts like <math>x</math> on both <math>H</math> and <math>K</math>, and so <math>(HK)^z = (HK)^x</math>.
#  Now consider <math>z = yx</math>. We claim that <math>(HK)^z = (HK)^x</math>('''No problem data used'''):. We have <math>H^z = H^{yx} = (H^y)^x = H^x</math> (since <math>H^y = H</math>). We already know that <math>H^x = H^g</math> (step (1)), so <math>H^z = H^g</math>. Similarly, <math>K^z = (K^y)^x = (K^{gx^{-1}})^x = K^{gx^{-1}x} = K^g</math>. Thus, <math>z</math> acts like <math>x</math> on both <math>H</math> and <math>K</math>, and so <math>(HK)^z = (HK)^x</math>.
# We now check that <math>z \in \langle (HK), (HK)^g \rangle</math> ('''No problem data used'''): For this, observe that <math>x \in \langle H,H^g \rangle \le \langle HK, (HK)^g \rangle</math>. Also, <math>y \in \langle K, K^{gx^{-1}} \rangle</math>. The subgroup <math>K</math> is contained in <math>HK</math>. The subgroup <math>K^g</math> is contained in <math>(HK)^g</math>, and the element <math>x</math> is in <math>\langle HK, (HK)^g \rangle</math> as argued already, so the subgroup <math>K^{gx^{-1}}</math> is also in <math>\langle HK, (HK)^g \rangle</math>. Thus, <math>\langle K, K^{gx^{-1}} \rangle \le \langle HK, (HK)^g \rangle</math>, and hence <math>y \in \langle HK, (HK)^g \rangle</math>. Thus, <math>z = yx</math> is also in the subgroup <math>\langle HK, (HK)^g \rangle</math>.
# We now check that <math>z \in \langle (HK), (HK)^g \rangle</math> ('''No problem data used'''): For this, observe that <math>x \in \langle H,H^g \rangle \le \langle HK, (HK)^g \rangle</math>. Also, <math>y \in \langle K, K^{gx^{-1}} \rangle</math>. The subgroup <math>K</math> is contained in <math>HK</math>. The subgroup <math>K^g</math> is contained in <math>(HK)^g</math>, and the element <math>x</math> is in <math>\langle HK, (HK)^g \rangle</math> as argued already, so the subgroup <math>K^{gx^{-1}}</math> is also in <math>\langle HK, (HK)^g \rangle</math>. Thus, <math>\langle K, K^{gx^{-1}} \rangle \le \langle HK, (HK)^g \rangle</math>, and hence <math>y \in \langle HK, (HK)^g \rangle</math>. Thus, <math>z = yx</math> is also in the subgroup <math>\langle HK, (HK)^g \rangle</math>.
==References==
* {{paperlink|Rose-pronormal}}, Proposition 1.7, Page 450

Revision as of 22:56, 21 February 2009

This article gives the statement, and possibly proof, of a subgroup property (i.e., pronormal subgroup) satisfying a subgroup metaproperty (i.e., normalizing join-closed subgroup property)
View all subgroup metaproperty satisfactions | View all subgroup metaproperty dissatisfactions |Get help on looking up metaproperty (dis)satisfactions for subgroup properties
Get more facts about pronormal subgroup |Get facts that use property satisfaction of pronormal subgroup | Get facts that use property satisfaction of pronormal subgroup|Get more facts about normalizing join-closed subgroup property


Statement

Suppose H,KG are pronormal subgroups such that KNG(H): in other words, K normalizes H. Then the join of subgroups H,K (also the same as HK) is also a pronormal subgroup.

Related facts

Similar facts: statements about normalizing join-closedness

Some facts with very similar proofs:

Other facts about normalizing joins, but with a different kind of proof:

Related facts about pronormality

Join-closedness of some related properties

Proof

(Note: This proof adopts the right-action convention. It can be adapted to the left-action convention).

Given: A group G, pronormal subgroups H,KG such that KNG(H).

To prove: HK is pronormal in G. Specifically, for any gG, our goal is to find a zHK,(HK)g such that (HK)z=(HK)g.

Proof:

  1. (Given data used: H is pronormal in G): First, consider g on H. By pronormality of H, there exists xH,Hg such that Hx=Hg. Thus, Hgx1=H.
  2. (Given data used: K normalizes H): Recall that KNG(H), so Kgx1NG(Hgx1)=NG(H). Thus, K,Kgx1NG(H).
  3. (Given data used: K is pronormal in G): By pronormality of K, there exists yK,Kgx1 such that Ky=Kgx1. In particular, yNG(H), so Hy=H.
  4. Now consider z=yx. We claim that (HK)z=(HK)x(No problem data used):. We have Hz=Hyx=(Hy)x=Hx (since Hy=H). We already know that Hx=Hg (step (1)), so Hz=Hg. Similarly, Kz=(Ky)x=(Kgx1)x=Kgx1x=Kg. Thus, z acts like x on both H and K, and so (HK)z=(HK)x.
  5. We now check that z(HK),(HK)g (No problem data used): For this, observe that xH,HgHK,(HK)g. Also, yK,Kgx1. The subgroup K is contained in HK. The subgroup Kg is contained in (HK)g, and the element x is in HK,(HK)g as argued already, so the subgroup Kgx1 is also in HK,(HK)g. Thus, K,Kgx1HK,(HK)g, and hence yHK,(HK)g. Thus, z=yx is also in the subgroup HK,(HK)g.

References