The level of real projective spaces (Q911096): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
Set profile property. |
||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank |
Latest revision as of 02:34, 5 March 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | The level of real projective spaces |
scientific article |
Statements
The level of real projective spaces (English)
0 references
1989
0 references
The real projective space \({\mathbb{R}}{\mathbb{P}}^{2m-1}\) has a natural involution (\({\mathbb{Z}}/2\)-action) induced by multiplication by \(i=\sqrt{- 1}\) on \({\mathbb{C}}^ m\). The main result of this fine paper is the computation of the level \(s(m)=s({\mathbb{R}}{\mathbb{P}}^{2m-1})=\min \{n:\) there exists a \({\mathbb{Z}}/2\)-equivariant map f: \({\mathbb{R}}{\mathbb{P}}^{2m-1}\to S^{n-1}\}.\) The result reads as follows: For \(m\geq 2\) we have \(s(m)=m+1\) if \(m\equiv 0,2\), \(s(m)=m+2\) if \(m\equiv 1,3,4,5,7\) mod 8, \(s(m)=m+3\) if \(m\equiv 6.\) This significantly improves the estimate \(m+1\leq s(m)\leq (3m+1)\) obtained in a joint paper of the author and the reviewer [ibid. 62, 286-291 (1987; Zbl 0634.10020)]. Concerning a lower estimate for s(m) the slightly stronger new bound is obtained by studying suitable n-dimensional vector bundles and using a theorem of \textit{M. C. Crabb} [\({\mathbb{Z}}/2\)-homotopy theory (1980; Zbl 0443.55001)] which says that such a bundle has a non-trivial section if and only if its cohomotopy Euler class vanishes. Finally a theorem of \textit{J. Adams} [Stable homotopy theory (1964; Zbl 0126.390)] on the \(E_ 2\)-term in the Adams spectral sequence leads to an upper bound for s(m) which coincides with the lower bound except for \(m\equiv 4 mod 8.\) Clearly one would like to construct explicit equivariant maps f: \({\mathbb{R}}{\mathbb{P}}^{2m-1}\to S^{s(m)-1}\), hopefully ``quadratic'' maps. Also the ``periodicity mod 8'' has to be explained. These questions remain open for further study.
0 references
real projective space
0 references
involution
0 references
level
0 references
vector bundles
0 references
cohomotopy Euler class
0 references
Adams spectral sequence
0 references