Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am asking for a rigorous proof of the following:

Theorem: Let $R$ be a region and $C$ be a simple closed curve so that $C=\partial R$. If $\gamma:[a,b]\to \mathbb{R}^2$ is parametrization of $C$ then $\gamma$ is either positive or negative orientiated with respect to $R$.

Definition of positive orientation: Define $n(t)=(-y^{\prime}(t),x^{\prime}(t))$ where $\gamma(t)=(x(t),y(t))$. Then $\gamma$ is positive orientiated with respect to $R$ if $\forall t\in [a,b]\ \exists \epsilon>0$ so that $\gamma(t)+\epsilon n(t)\in R$. Negative orientation is defined similarly but with $n(t)=(y^{\prime}(t),-x^{\prime}(t))$.

Note that I request a proof that doesn't utilise the Jordan Curve Theorem and this is the reason the definitions and the theorem are stated this way.

share|cite|improve this question
up vote 2 down vote accepted

In general, the theorem is not true. For example, $\gamma$ need not be differentiable or the derivative might vanish sometimes.

But if $x'$ and $y'$ exist and are continuous and there is no $t$ with $x'(t)=y'(t)=0$, then we can do the following: Consider a point $t_0\in [a,b]$. Wlog. $a<t_0<b$, $\gamma(t_0)=(0,0)$, $h:=x'(t_0)>0$, $y'(t_0)=0$ (apply a suitable translation and rotation to the figure). Then there is $\delta>0$ such that $x'(t)>\frac h2$ for $t_0-\delta< t<t_0+\delta$. As a consequence $x(t)>\frac h4 (t-t_0)^2 $ for $t_0<t<t_0+\delta$ and $x(t)<-\frac h4 (t-t_0)^2$ for $t_0-\delta< t<t_0$.

The compact set $\gamma([a,t_0-\delta])\cup \gamma([t_0+\delta,b])$ has a positive distance $d$ from $(0,0)$ because $C$ is simple.

Because $(0,0)\in\partial R$, there is a point $(\xi,\eta)\in R$ at distance $<\min\{\frac d2,\frac h4\delta^2\}$. There is exactly one $t_1\in(t_0-\delta,t_0+\delta)$ with $x(t)=\xi$ because $x$ is monotone there. If $\eta<y(t_1)$, then we can walk inside $R$: first vertically down to $\left(\xi, \sqrt{\frac {d^2}4-\xi^2}\right)$, then along the circle of radius $\frac d2$ to $(0,-\frac d2)$, then to $(0,u)$ with $0>u>-\frac d2$. Similarly, if $\eta>y(t_1)$, then we find a point $(0,u)\in R$ with $0<u<\frac d2$. This shows: For all $t\in[a,b]$ there is $\epsilon>0$ such that $\gamma(t)+\epsilon n(t)\in R$ or $\gamma(t)-\epsilon n(t)\in R$. Since the two maps $t\mapsto \gamma(t)\pm\epsilon n(t)$ are continuous, the set of points $t$ where $\gamma is locally positive/negative oriented, is open.

The next step should be to show that these sets are also closed.

share|cite|improve this answer
You can omit the last sentence: As $S^1$ is connected one of the two open sets has to be empty. – Christian Blatter Oct 25 '12 at 16:19
Thank you for your answer. – Nameless Oct 25 '12 at 16:57

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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