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

What conditions must be satisfied for $f(x)g(y)dx + h(x)p(y)dy = 0$ to be exact?

I found it is exact when $f(x)g'(y) = h'(x)p(y)$ but can I assume that $h(x) = \int f(x)\mathrm{d}x + c$ and $g(y) = \int p(y)\mathrm{d}y + c$?

share|cite|improve this question

Let the OP is exact. So we have $f(x)g'(y)=h'(x)p(y)$ and so there is a function $F$ such that: $$F_x=f(x)g(y),~~F_y=h(x)p(y)$$ Let's assume the $p(y)\neq0$ for a while. We have $$F(x,y)=g(y)\int f(x)dx+K(y)$$ and then $$F_y=g'(y)\int f(x)dx+K'(y)=h(x)p(y)$$ so $$h(x)=\frac{g'(y)}{p(y)}\int f(x)dx+\frac{K'(y)}{p(y)}$$ for a suitable function $K(y)$.

share|cite|improve this answer

Almost, but not quite. Rather, under mild assumptions on your functions, you have that $$ h^\prime(x) = C f(x), \quad g^\prime(y) = C p(y) $$ for some constant $C \in \mathbb{R}$, which need not be equal to $1$.

Suppose that your equation is indeed exact, so that, as you observed, $f(x)g^\prime(y) = h^\prime(x)p(y)$. Then, whenever $f(x) \neq 0$ and $p(y) \neq 0$, $$\frac{h^\prime(x)}{f(x)} = \frac{g^\prime(y)}{p(y)}.$$

Now, suppose for simplicity that all your functions are $C^1$ and that, indeed, $f(x) \neq 0$ and $p(y) \neq 0$ for all but finitely many $x$ and $y$. Then, on the complement of a finite number of points $(x,y)$, we can define $$ F(x,y) := \frac{h^\prime(x)}{f(x)} = \frac{g^\prime(y)}{p(y)}, $$ which therefore has $\tfrac{\partial F}{\partial x} = 0$ and $\tfrac{\partial F}{\partial y} = 0$. Thus, $F(x,y) = C$ for some constant $C$, and hence $$ h^\prime(x) = C f(x), \quad g^\prime(y) = C p(y) $$ for all but finitely many $x$ and $y$; by continuity of $f$, $g^\prime$, $h$, and $p$, these equalities are therefore true for all $x$ and $y$. Note that $C$ need not be $1$, as you guessed. If you have seen any PDE, this is the same basic argument as that underpinning separation of variables.

share|cite|improve this answer

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.