Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $G$ be group and $G$ has a subgroup $H$ with infinite index.

Is it true that the number of subgroups of $G$ containing $H$ is infinite?

share|cite|improve this question
up vote 5 down vote accepted

No. I'm sure there are far simpler examples, but the first that comes to mind is a Tarski monster. This is an infinite group $G$ in which every nontrivial proper subgroup has order $p$, for some fixed prime $p$. Any such subgroup thus has infinite index, but is not contained in any subgroup except itself and $G$.

share|cite|improve this answer
Tarski monster group is a very interesting. Can you give me some reference about subgroup of this group? – Babak Miraftab Jan 29 '13 at 18:58
What's wrong with the references cited in the Wikipedia article I linked? – Chris Eagle Jan 29 '13 at 19:02
You claimed that it is not contained in any subgroup except itself and $G$. I want to see more detail about this. – Babak Miraftab Jan 29 '13 at 19:05
@Babgen: How many ways can one subgroup of order p contain another subgroup of order p? – hardmath Jan 29 '13 at 19:21

I believe the following should provide a relatively simple example. Let $N$ be the additive group of the rationals, and $H$ the multiplicative group of the nonzero rationals. Let \begin{equation} G = N \rtimes H \end{equation} be the semidirect product, in which $H$ acts on $N$ by multiplication.

Now $H$ is clearly a subgroup of infinite index in $G$. But $H$ is maximal in $G$, that is, there is no subgroup $L$ such that $H < L < G$. In fact, if $H < L$, then $L$ contains a nonzero $a \in N = \mathbf{Q}$, so if $b \in N = \mathbf{Q}$ is nonzero, then applying the element of $H$ that multiplies by $b a^{-1}$ we see that $b = (b a^{-1}) \cdot a \in L$, so that $L$ contains $N$, and thus $L = G$.

share|cite|improve this answer
What's wrong with the rather widespread $\,\Bbb Q\,,\,\Bbb Q^*\,$ ...? – DonAntonio Jan 29 '13 at 19:38
Nothing @DonAntonio, but on the one hand I wanted to have an $H$ as in the original question, and then I wanted to make clear whether I was taking an element in one group or in the other. – Andreas Caranti Jan 29 '13 at 19:47

In general, your result is false, but if $H$ is a normal subgroup, then it is true. Indeed, there is a bijection between the subgroups of $G$ containing $H$ and the subgroups of $G/H$; but $G/H$ has finitely many subgroups iff $G/H$ is finite iff $H$ is of finite index in $G$.

In particular, the result is true for abelian groups.

share|cite|improve this answer
Let me ask, since I'm not the answerer here: why the downvotes? Has the poster said something wrong, in which case a comment pointing out the mistake is called for? Or is the observation correct but so irrelevant to the question that some feel a need to discourage this behavior? Inquiring minds wonder... – hardmath Jan 29 '13 at 18:57
I think some rather bitter serial downvoter is amongst us...Some people rush to downvote whenevever is, or seems to them to be, wrong. – DonAntonio Jan 29 '13 at 19:40

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.