I raised a related question but hope to get some answer using the nonvanishing 2 form definition.

Let P be the real projective plane obtained by identifying antipodal points on the unit sphere of $R^3$.

How to prove that P is nonorientable by showing that any 2 form on P will vanish somewhere?

My idea is to consider the closed curve $a(t)=(\cos t,\sin t, 0)$ , $0 \leq t \leq \pi$

This curve is closed in P and the tangent vectors $a'(0)$ and $a'(\pi)$ are identical.

However, for a vector field V on $a(t)$ defined by $V(a(t))=(0,0,1)$, the tangent vectors at $a(0)$ and $a(\pi)$ differ by a sign. Also V and $\alpha'(t)$ are always linearly independent.

For any 2 form $u$ on P,

$u(a'(0), V(a(0))=-u(a'(\pi), V(a(\pi))$ Therefore $u$ must vanish somewhere on the curve.

Is the constant vector field V continuous on the curve? Are my arguments right?


If $\omega $ is a nonzero two form on $P$ then define $\pi^\ast \omega$ on $S^2$ where $\pi : S^2\rightarrow P$. Note that $$ \pi^\ast \omega(e_1,e_2)=\omega (d\pi e_1,d\pi e_2)\neq 0$$ where $\{ e_1,e_2\}$ is indepedent.

This implies that $\pi^\ast \theta$ is nonzero form on $S^2$. And note that $$ (\pi^\ast \omega)_p (e_1,e_2) = (\pi^\ast \omega)_{-p}(-e_1,-e_2)\neq 0 $$

Let $N$ is unit normal on $S^2$ and $\theta = (N,\ )$

Then $(\pi^\ast \omega)\wedge\theta$ is a nonzero three form. If $\{ (1,0,0),e,N \}$ is an orthnormal frame along $c(t)=( 0,\sin\ t,\cos\ t),\ 0\leq t\leq \pi$, and if $e(0)=(0,1,0)$, then $$ (\pi^\ast \omega \wedge \theta )_{c(t)} ((1,0,0),e,N)\neq 0 $$

So $(\pi^\ast \omega)_{c(0)} ((1,0,0),(0,1,0))$ and $(\pi^\ast \omega )_{c(\pi)} ((1,0,0),(0,-1,0))$ have same sign. So $$e_1=(1,0,0),\ e_2=(0,1,0),\ p=(0,0,1),\ (\pi^\ast \omega)_p (e_1,e_2) (\pi^\ast \omega)_{-p}(e_1,-e_2)>0 $$ It is a contradiction.

  • $\begingroup$ I do not understand the last 4 lines $\endgroup$ – noot Oct 15 '13 at 16:38
  • $\begingroup$ Note that $\pi^\ast \omega\wedge \theta$ is nonzero three form since $\pi^\ast \omega$ is a nonzero two from on tangent space of sphere and $\theta$ is dual of $N$. And $\{ (1,0,0),e, N\}$ is linearly independent. $\endgroup$ – HK Lee Oct 15 '13 at 16:43
  • $\begingroup$ can you just use 2 form? i need to prove it with 2 form. $\endgroup$ – noot Oct 15 '13 at 16:47
  • $\begingroup$ how to derive your last equation $\endgroup$ – noot Oct 15 '13 at 16:50

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.