Bonsoir à tous :
Voici la preuve du théorème de dualité de Poincaré ( Le "Poincaré pairing" est non dégénéré ), que je ne comprends pas bien : ( rédigé en anglais )
The proof is by induction on the numberof elements in a good cover of
.
If, then
is diffeomorphic to
, and hence, the assrtion follows from exemple
( Cet exemple se trouve dans les premiers pages de mon bouquin ).
Now, let, suppose that
is a good cover, and suppos that Poincaré duality holds for every oriented
- manifold with a good cover by at most
open sets.
Denote bythe open sets :
and
.
Then, the induction hypothesis asserts that Poincaré duality holds for the manifoldsand
.
We shall prove thatsatisfies Poincaré duality by considering simultaneously the Mayer - Vietoris sequences for
and
associated to the cover
.
Thus we have commuting diagrams :
and
Commutativity of the first square ( 8.23 ) asserts that, for all closed formsand
, we have :
This follows from the définition of. ( See (8.16) ).
Commutativity of the second square ( 8.23 ) asserts that, for all closed forms,
,
andwe have :
This follows from the definition of. ( See ( 8.6 )
Commutativity of the diagram ( 8.24 ) asserts that, for all closed formsand
, we have :
.
To see this, we recall that :
, and
Hereis extended to all of
by setting it equal to zero on
, and
is restricted to
where it still has compact support.
Since, we obtain :
as claimed.
With the commutativity of (8.23) and (8.24) established, we obtain a commuting diagram :
Since the horizontal sequences are exact and the Poincaré duality homomorphismsare isomrphisms for
,
and
, it follows by Five Lemma that
is an isomorphism as well.
This proves the theorem.
{ \bf Questions } :
Pourriez vous m'expliquer les passages suivants :
- for all closed formsand
, we have :
- for all closed forms,
,
andwe have :
Le reste, je l'ai compris.
Merci d'avance.
-----