I'm trying to compute the line integral $$\int_{C}(y-z)\,dx+(z-x)\,dy+(x-y)\,dz$$ where the curve $C$ is the intersection of the cylinder $x^2+y^2=1$ and the plane $x−z=1$. I know how to do this directly by parametrising with $\vec{r}(t)=(\cos t, \sin t, -1+\cos t)$ for $t\in [0,2\pi]$ and then substituting in for $dx$, $dy$ and $dz$, although the solution I've seen says to look at $\int_{C}\vec{F}\cdot d\vec{r}$ but how would I do this and simlilarly how would I use Stoke's theorem if it's not written as $\int_{C}\vec{F}\cdot d\vec{r}$?

From the solution, it looks like they've chosen $\vec{F}=(y-z,z-x,x-y)$ which is obviously taken from the original integrand, but I don't see why this should be true? Wouldn't I have to look at the following integral, work out the right hand side and try to match things up? $$\int_{C}(y-z)\,dx+(z-x)\,dy+(x-y)\,dz=\int_{C}\vec{F}\cdot d\vec{r}$$

Edit: Never mind, I figured it out. I didn't realise $d\vec{r}=(dx,dy,dz)$ as I'd only seen the definition as $\int_{C}\vec{F}\cdot d\vec{r}=\int_{a}^{b}\vec{F}(\vec{r}(t))\cdot \vec{r}'(t)\,dt$. My bad!


You can use Stokes' theorem.

The curve $C$ bounds an elliptical disc $E$. The normal $n$ to the plane of $E$ is $(-1,0,1)$, up to sign.

Since ${\rm curl}(F)=(-2,-2,-2)$ is constant Stokes' theorem gives $$\int_C F\cdot dr=\int_E{\rm curl}(F)\cdot n\>{\rm d}\omega=\pm(-2,-2,-2)\cdot(-1,0,1)\>{\rm area}(E)=0\ ,$$ so that we don't have to bother about the orientation.


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.