Let $M$ be prime closed compact orientable 3-manifold with infinite fundamental group $G=\pi_1(M)$. Can there be finite order elements in $G$ ? What are the possibilities for $H_1(M)$ - can it be any abelian finitely generated group ?

If we have finite cover $N\to M$. In this case there is monomorphism $\pi_1(N)\to\pi_1(M)$. Is there also monomorphism $H_1(N)\to H_1(M)$ ?

From the "virtually Haken conjecture" I conclude that there exists finite cover $N$ which is Haken. It means it contains incompressible surface. It means $H_2(N)\neq0$.

If $M$ is closed compact orientable 3-manifold with finite fundamental group then $M$ is spherical. It means it is quotient of $S_3$ by finite subgroup of $SO_4$. This subgroup is fundamental group of $M$.


1 Answer 1


It cannot be $S^1\times S^2$, so in fact it is irreducible. So $\pi_1$ infinite implies its universal cover is contractible and moreover is $\mathbb R^3$ [Geometrization Theorem]. To see it's universal cover is contractible, observe that irreducibility implies $\pi_2$ is zero. So infact it's universal cover has $\pi_2$ zero. Now $\pi_1$ infinite implies it's universal cover is non-compact, so $H_3=0$ and thus by Hurewich theorem $\pi_3=0$. And with similar arguments it's all homotopy groups are zero. So it is contractible. And Thurston-Parelman's geometrization says that it is infact diffeomorphic with $\mathbb R^3$

Now if it has a finite order element in $\pi_1$ then infact we can have a finite free group action on $\mathbb R^3$ (WHY?). But it is not possible. So contradiction.

  • $\begingroup$ Thank you. How do you deduce that universal cover is contractible ? Then $H_1$ is just sum of few copies of $Z$... or maybe not . I am not familiar with infinite group properties. $\endgroup$
    – mmm
    Commented Jun 30, 2018 at 13:44
  • $\begingroup$ Thurston's-Parelman's Geometrization conjecture(theorem). Since you sited vertual Haken, I thought you must be familiar with this as well $\endgroup$ Commented Jun 30, 2018 at 17:34
  • $\begingroup$ @MarekMitros see the edit $\endgroup$ Commented Jun 30, 2018 at 21:31
  • $\begingroup$ Thank you for more details. I know about geometrization conjecture. I am trying to understand how decomposition of prime manifold along incompressible tori help classify 3-manifolds. Can we construct list of all prime manifolds this way ? I should post new question about it. $\endgroup$
    – mmm
    Commented Jul 1, 2018 at 19:36
  • $\begingroup$ @MarekMitros what do you mean by 'this way' here in your last sentence? $\endgroup$ Commented Jul 1, 2018 at 20:57

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .