After seeing the following equation in a lecture about tensor analysis, I became confused.

$$ \frac{d\phi}{ds}=\frac{\partial \phi}{\partial x^m}\frac{dx^m}{ds} $$

What exactly is the difference between $d$ and $\partial$?

  • 3
    $\begingroup$ Total derivative vs partial derivative. $\endgroup$ – user2468 Aug 16 '12 at 18:28
  • $\begingroup$ @Jen, maybe elaborate a bit more in an answer? :) $\endgroup$ – J. M. isn't a mathematician Aug 16 '12 at 18:33
  • $\begingroup$ I know, one is the partial and the other one is a total derivative. But isn't $\frac{\partial f}{\partial x}$ the same as $\frac{df}{dx}$? $\endgroup$ – iblue Aug 16 '12 at 18:35
  • $\begingroup$ @iblue, the first one treats all the other independent variables as if they were constants. The second one doesn't. $\endgroup$ – J. M. isn't a mathematician Aug 16 '12 at 18:40
  • $\begingroup$ @iblue Read what Jennifer linked to, it will help. $\endgroup$ – Pedro Tamaroff Aug 16 '12 at 18:40

As mentioned $d$ means total and $\partial$ partial derivative and are not the same. Total derivative also counts $x$ dependencies in other variables. For instance: $$ f(x,v) = x^2 + v(x) \\ \frac{\partial f}{\partial x} = 2x \\ \frac{\partial f}{\partial v} = 1 \\ \frac{d f}{d x} = 2x + \frac{\partial v(x)}{\partial x} $$

Your formula most probably uses Einstein notation and is only a shorter way to write $$ \frac{d\phi}{ds}= \sum_m \frac{\partial \phi}{\partial x^m}\frac{dx^m}{ds} $$

| cite | improve this answer | |
  • $\begingroup$ Why it's partial dirivative of $v$ when you take derivative to $f$ ? is it suppose to be $\frac{df}{dx} = 2x + \frac{dv(x)}{dx}$ ? $\endgroup$ – peng yu Apr 20 at 14:58

The difference is whether the rest of the variables of $f$ are considered constants or variables in $x.$ Former is partial, latter is total.

  • $\frac{\partial f}{\partial x}$ is the partial derivative: $f$ is differentiated w.r.t. to $x$ while all other variables are considered constants in $x.$

  • $\frac{d f}{d x}$ the is total derivative: $f$ is differentiated w.r.t. to $x$ while nothing is assumed about the other variables; they are considered variables in $x.$ (some variables might be, in fact, constants in $x.$)

| cite | improve this answer | |

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.