Can someone shed a little bit of light on the problem of volumes of revolution about the $x$-axis and the $y$-axis of the same shapes? Take for example $f(x)=x^2$. If we want to find the volume bounded by the parabola and the $x$-axis with axis of revolution at $y=0$ we would use the standard method of disks and get the volume $\pi/5$ for $x=[0,1]$. Now if we want to find the volume by rotating the curve around the $y$-axis instead, using the cylindrical shells method, we get the volume to be $\pi/2$.


Now, maybe it's just my flawed intuition, but since we are basically rotating the same shape/area by 360$^\circ$ (but in different "directions"), I would guess the two volumes to be the same. Anyone knows some good graphics/animation/resources to help me visualize this problem?

  • $\begingroup$ Note also that for rotating about y, you could delete the whole area to the left of the axis and not change the volume. Rotating about x you cannot. $\endgroup$ – Ross Millikan May 16 '12 at 15:36
  • $\begingroup$ @RossMillikan the area to the left that you speak of, is as if it was "deleted" when integrating for $x=[0,1]$, isn't it? Or am I missing your point? $\endgroup$ – Milosz Wielondek May 16 '12 at 15:45
  • $\begingroup$ my point was that it is another way of seeing the disconnect between area and rotated volume. You could have it or not when rotating about y and get the same volume, but the area changes by a factor 2. $\endgroup$ – Ross Millikan May 16 '12 at 16:00

When you rotate about $y$, most of the area is farther from the axis than if you rotate around $x$. If you think about a small area of size $dxdy$ when rotated around $x$ it sweeps out a volume element $ydxdy$ and when rotated around $y$ is sweeps out a volume $xdxdy$. The methods of discs and shells just do one dimension of the integral for you.

You could think of a rectangle $[0,10] \times [0,0.1]$. If you rotate it around $x$ you get a cylinder of radius $0.1$ and height $10$, for volume $0.1\pi$. If you rotate it around $y$ the radius is $10$ and the height is $0.1$ for a volume of $10\pi$

  • $\begingroup$ Naturally, it does make perfect sense mathematically. But am I the only one to think it's somewhat counter-intuitive? $\endgroup$ – Milosz Wielondek May 16 '12 at 16:25

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.