I was listening to an audiobook of Einstein when they started discussing spherical geometry and how Pi was no longer the ratio of a circle's circumference to its diameter, so I set out to find the answer. From what I gathered, it is a variable that depends on an angle $\theta$ between the pole and the circle in question, since there are infinite parallel circles on a sphere.

So the circumference I calculated through Euclidean methods would be $2\pi$ times the length of a side of a triangle opposite $\theta$. If $r$ is the radius of the sphere, then we can find this two dimensional circles's radius as $$x=r\sin\theta.$$ So the circumference, $C_{\theta}$, then is $2\pi\cdot{r}\sin\theta$

Now we need the radius, which is the arc length from the pole, $\frac{\pi}{2}$ to $\sin\theta$. Using parametrization of $$\mathbf{x}(t)=(r\cos{t},r\sin{t}),$$ The arclength would just be $$\rho_{\theta}=\int_{\sin{\theta}}^{\frac{\pi}{2}}||\mathbf{x}(t)||dt=\int_{\sin{\theta}}^{\frac{\pi}{2}}rdt=r\left(\frac{\pi}{2}-\sin\theta\right)$$ Multiplying by 2 gives us our diameter So I think that this is the function in $\theta$ that gives us the ratio of a circle's circumference to its diameter; $$R(\theta)=\frac{2\pi\cdot{r}\sin\theta}{2r\left(\frac{\pi}{2}-\sin\theta\right)}=\frac{\pi\sin\theta}{\left(\frac{\pi}{2}-\sin\theta\right)}$$ Is this a correct approach?

Edit; I can see my lower bound for my arclength integral was completely off, as it was 2 in the morning here when I was figuring this out.... The upper bound would be $\frac{\pi}{2}$ and my lower bound would be $\frac{\pi}{2}-\theta$. Thus, if i would have integrated properly, $$\int_{\frac{\pi}{2}-\theta}^{\frac{\pi}{2}}||\mathbf{x}(t)||dt=\int_{\frac{\pi}{2}-\theta}^{\frac{\pi}{2}}rdt=r\left[\frac{\pi}{2}-\left(\frac{\pi}{2}-\theta\right)\right]=r\theta$$ This would give the correct answer that was given below of $$R(\theta)=\frac{\pi\sin\theta}{\theta}$$


If you slice the sphere by the plane that intersects this circle, you see a disk of radius $r \sin \theta$. So the circumference of the circle is $2 \pi r \sin( \theta)$. The "radius", i.e. arc length from the pole, you see by slicing the sphere by a plane through the poles: it is an arc of angle $\theta$ on a circle of radius $r$, so $r \theta$. Thus the ratio circumference/radius is $2 \pi \sin(\theta)/\theta$.

  • $\begingroup$ I just realized I wanted the circumference to diameter, since I wanted a spherical "pi" value. So just $\frac{\pi\sin\theta}{\theta}$. Is my arclength argument wrong or is your way just a better simplified formula? $\endgroup$ – Eleven-Eleven Aug 30 '13 at 12:25
  • $\begingroup$ I see where I made my error in calculation in my integral!! Thanks @RobertIsrael. $\endgroup$ – Eleven-Eleven Aug 30 '13 at 15:10

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.