Our professor gave us this function to differentiate $$3(x^2+y^2)^2=100xy$$ and I did differentiate it $$\frac{dy}{dx}=\frac{3x^4+3xy^2-25y}{25x-3x^2y+3y^3}$$ But I'm having trouble finding the points that have a vertical or horizontal tangent. I am aware the numerator needs to =0 for the tangent to be horizontal and denominator =0 for tangent to be vertical

I tried using the quadratic formula to get $y=\frac{25\pm\sqrt{25^2-36x^5}}{6x}$ and I know we are supposed to replace y into the original function to get the points, but seeing as the professor doesn't allow us to use any sort of calculator, I have feeling there should be a much simpler way to do this?

Can someone please help me with this question?

  • 2
    $\begingroup$ Your expression for $dy/dx$ is not quite right. It should be $\frac{3x^3+3xy^2-25y}{25x-3x^2y-3y^3}$ $\endgroup$
    – smcc
    Oct 16, 2018 at 18:29
  • $\begingroup$ I might try solving for $x^2+y^2$ instead of $y$ and substituting that back into the original equation. $\endgroup$
    – amd
    Oct 16, 2018 at 19:02

2 Answers 2


Suppose your equation implicitly defines $y$ as a function of $x$:


Differentiating with respect to $x$:

$$6(x^2+y^2)[2x+2yy'(x)]=100[y+xy'(x)] \tag{1}$$

Substitute $y'(x)=0$ into $(1)$:

$$12(x^2+y^2)x=100y \iff 3(x^2+y^2)x=25y$$

This can be used with the original equation to obtain the point(s) where $y'(x)=0$.

The other point(s) can be found by the symmetry of the problem in $x$ and $y$. (You could go through the same working as above but consider the equation as defining $x$ as an implicit function of $y$.)




By the implicit function theorem

$$\begin{align*}y'(x)&=-\frac{F_x}{F_y}\quad \text{ if $F_y\neq 0$}\\ x'(y)&=-\frac{F_y}{F_x} \quad\text{ if $F_x\neq 0$}\end{align*}$$

$$F_x=12x(x^2+y^2)-100y\qquad \text{ and } \qquad F_y=12y(x^2+y^2)-100x$$

The points you require are where $F_x=0$ or $F_y=0$ (but not both).

  • $\begingroup$ The point about symmetry is a good one, but the equation that you’ve derived is the same one that the OP has already solved for $y$. $\endgroup$
    – amd
    Oct 16, 2018 at 18:56
  • $\begingroup$ Not quite, and I had left the equation as it as in order to hint that it is probably better not to solve for $y$ as they have done. $\endgroup$
    – smcc
    Oct 16, 2018 at 19:03
  • $\begingroup$ It’s probably better to make that hint less subtle, then. $\endgroup$
    – amd
    Oct 16, 2018 at 19:04

Hint: You must use both equations $$3(x^2+y^2)^2=100xy$$ $$3x^4+3xy^2-25y=0$$

  • $\begingroup$ So should I just discard my calculations or..? There's gotta be more hints than that. $\endgroup$
    – logan
    Oct 16, 2018 at 18:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .