Contraction of the canonical symplectic form by vertical vectors

Join the discussion
Registration is free. Start your own thread to ask a follow-up.
4 replies · 4K views
mma
Messages
266
Reaction score
5
The canonical symplectic form on [itex]T^*M[/itex] is the exterior derivative of the tautological 1-form:
[tex]\omega=d\alpha[/tex]​
where [itex]\alpha_p(X):=p(d\pi(X))[/itex] is the tautological 1-form.

Let [itex]Y \in T_pT^*M[/itex] a vertical vector, that is [itex]d\pi(Y)=0[/itex].

It's trivial to prove using canonical coordinates that for all [itex]X \in T_pT^*M[/itex]
[tex]\omega(X,Y) = y(d\pi(X))[/tex]​
where [itex]y \in T_{\pi(p)}^*M[/itex] such that for any differentiable function [itex]f: T^*M \to \mathbb R[/itex] [tex]Y(f)=\left. \frac{df(p+ty)}{dt}\right|_{t=0}[/tex].

But how can it be proved in a coordinate-free manner?
 
Last edited:
Physics news on Phys.org
Erratum.

mma said:
The canonical symplectic form on [itex]T^*M[/itex] is the exterior derivative of the tautological 1-form:
[tex]\omega=d\alpha[/tex]​

should be

The canonical symplectic form on [itex]T^*M[/itex] is the negative of the exterior derivative of the tautological 1-form:
[tex]\omega=-d\alpha[/tex]​


Any idea?
 
Your question is giving me a headache, but have you looked in the book by Anna Canna Da Silva (first or second chapter I think)? There, she proves many properties of the tautological 1-form in coordinate free form.
 
quasar987 said:
Your question is giving me a headache, but have you looked in the book by Anna Canna Da Silva (first or second chapter I think)? There, she proves many properties of the tautological 1-form in coordinate free form.


Yes, I know (and like) da Silva's book, but I didn't find this in it. I give a better chance to the book of Libermann and Marle. Thanks anyway.
 
I didn't know about this book until now. Thanks.