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

Is it possible to find a specific example of two fiber bundles with the same base, group, fiber and homeomorphic total spaces but these bundles are not equivalent/isomorphic, if so

should I find a bundle map F between two bundles, inducing identity on the common base but F does not preserve fibers? (I don't know what it means, got mixed up) or

should I define the action of group on fibers differently?

share|cite|improve this question
for each (h,j) in Z+Z there corresponds an S^3 bundle over S^4 and when h+j=1 the total spaces are all homeomorphic to S^7 and the group is SO(4). I try to understand where the difference come from? – wqr Jun 29 '12 at 23:33
Why do you ask wether such bundles exist if you already know examples of this behavior? – Olivier Bégassat Jun 29 '12 at 23:38

EDIT: This answer is wrong: see Dan Ramras's comment.

The circle (thought of as the set of norm one elements in the complex plane) is a fiber bundle over itself in many ways, via the maps $z \mapsto z^k$ where $k \neq 0$ is an integer. These are all non-isomorphic as fiber bundles (which follows from the computation that the fundamental group of $S^1$ is $\mathbb{Z}$), but the fiber has $|k|$ points, so the bundles corresponding to $\pm k$ have the same fiber, total space, and base.

share|cite|improve this answer
(The group here is just $\mathbb{Z}/k$.) – user29743 Jun 29 '12 at 23:44
These bundles are of different group, though. – Mariano Suárez-Alvarez Jun 30 '12 at 3:20
I don't understand - I am fixing $k$ and looking at the two bundles corresponding to $\pm k$. Surely both these bundles have structure group $\mathbb{Z}/k$, right? – user29743 Jun 30 '12 at 17:23
But the map $z\mapsto z^{-1}$ is a bundle isomorphism between the bundles for $k$ and $-k$. Covering spaces of $S^1$, as you say, are classified by subgroups of $\pi_1 (S^1) = \mathbb{Z}$, by sending a covering space to the image of its fundamental group under the projection map. Both of these covering spaces correspond to the subgroup $k\mathbb{Z} < \mathbb{Z}$. – Dan Ramras Jul 3 '12 at 3:58
hard to argue with that... – user29743 Jul 4 '12 at 18:31

In the paper "K-theory doesn't exist," (J. Pure Appl. Alg., 1978 vol 12) Akin explains that if $p: P\rightarrow B$ is a non-trivial principal $G$-bundle, the map $P\times G\rightarrow B$ (project to the first factor and then apply $p$) can be made into a principal $G\times G$-bundle in two different ways, by specifying two different actions of $G\times G$ on $P\times G$.

In general, these are not isomorphic as principal bundles (Akin shows that if they were always isomorphic, the $K$-theory of $B$ would be trivial, which is the origin of the paper's title).

share|cite|improve this answer

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.