MathOverflow is a question and answer site for professional mathematicians. 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

This question grew out of this post.

Question: Is there a finitely generated, infinite, residually finite group such that every finite index subgroup has $p$-power index for a fixed prime $p$?

The $p$-adic integers $\mathbb Z_p$ give an example of a non-finitely generated such group. This is not entirely obvious, but follows from a result of Jean-Pierre Serre, which states that every finite index subgroup of $\mathbb Z_p$ is closed.

EDIT: André Henriques has pointed out that Rostislav Grigorchuk constucted a finitely generated, infinite, residually finite $2$-group. All finite quotients of this group have to be $2$-groups. Colin Reid asked in a comment whether there is a torsionfree example. So let me take the freedom to extend my question:

Question: Is there a torsionfree example?

share|cite|improve this question
I suppose you want to exclude p-groups from this; if not, any finitely generated infinite residually finite p-group is an example. I wonder if there are any torsion-free examples? – Colin Reid Nov 20 '10 at 13:11
@Andreas: I will remove my comment which you refer to together with my answer there. So you may want to modify your question accordingly. You did not copy my comment precisely. I asked for a non-torsion f.g. example. Of course I know about torsion ones. There are also misprints in the title of your question. – Mark Sapir Nov 20 '10 at 15:41
@Mark: Sorry if I misread your comment. – Andreas Thom Nov 20 '10 at 20:56
up vote 9 down vote accepted

Yes, there are torsion free examples. I do not know who constructed them first, but some examples can be found in papers by Grigorchuk and his co-authors. For example Bartoldi and Grigorchuk proved that a certain Fabrykowski-Gupta group $\Gamma $ has the following properties (see Propositions 6.4 and 6.5 in arXiv:math/9911206):

(a) It is a subgroup of the automorphism group of a rooted tree.

(b) It is virtually torsion free.

(c) It satisfies the following 'congruence property': every finite index subgroup of $\Gamma $ contains a level stabilizer (i.e., the stabilizer of a level of the tree) and the index of every level stabilizer is a power of 3.

By (a) $\Gamma $ is residually finite. By (b) there is a torsion free subgroup $K$ of finite index in $\Gamma $. Since every finite index normal subgroup of $K$ contains a finite index normal subgroup of $\Gamma $, (c) implies that every finite quotient of $K$ is a 3-group.

share|cite|improve this answer

Grigorchuk's group $G$ is an example of what you're looking for:

Every element of $G$ has finite 2-power order, and so every finite quotient of $G$ is a 2-group.

share|cite|improve this answer
Very nice, thanks! – Andreas Thom Nov 20 '10 at 13:10
For the moment, let me not count this as an answer, since the question for a torsionfree example (raised by Colin Reid in a comment) is really a bit more interesting. – Andreas Thom Nov 20 '10 at 13:31

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.