Suppose we have a $n-1$ dimensional manifold $M \subset \mathbb{R}^n$ and a non-vanishing $n-1$ form $\omega$ on $M$. This implies the existence of a normal unit vector field on $M$.

The proof of this goes as follows:

"Let $x\in M$, there exists a neighborhood $U$ of $x$ in $\mathbb{R}^n$, a submersion $f:U\rightarrow \mathbb{R}$ such that $M\cap U=f^{-1}(0)$, $T_xM=\{u\in T_x\mathbb{R}^n:df_x(u)=0\}$. Write $df_x=(\partial f_1(x),...,\partial f_n(x))$. You can identify $(\partial f_1(x),...,\partial f_n(x))$ with a vector $u_x$ of $T_x\mathbb{R}^n$ such that $Vect(u_x)$ is a supplementary space to $T_xM$. Let $\Omega$ be the canonical volume form $dx_1\wedge...\wedge dx_n$, $(\partial f_1(x)dx_1+...+\partial f_n(x)dx_n)\wedge \omega =c\Omega$, if $c>0$, define $n(x)={1\over{\|u_x\|}}u_x$ if $c<0$, define $n(x)=-{1\over{\|u_x\|}}u_x$. Remark that $n(x)$ is well defined in a neighborhood of $x$ since it does not depend of the choice of $f$. In fact $u_x$ is a unit vector orthogonal to $T_xM$ relatively to the usual scalar product and it continuously depends of $x$."

Anyways, I have a few questions about the proof The first one is: Why is the wedge "$(\partial f_1(x)dx_1+...+\partial f_n(x)dx_n)\wedge \omega = c\Omega $" important, how is fundamentally linked to $n(x)$, I don't see the importance, and okay $\omega$ is non vanishing and there is an $i$ such that $\partial f_i(x)$ is not zero, but that's no reason to say that the wedge product is not zero, so $c$ could still be zero.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.