Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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

Given a function $ f:(a,b) \to E $ where $E$ is a Banach space, then clearly when $f$ is differentiable, the derivative $ f^\prime(x) $ can be seen as the tangent vector to the curve in $E$ graphed by $f$, at the point $f(x)$.

But the book I use says that value of the Frechet derivative of $f$ acting on $1$ i.e. $ f^\prime(x)(1) $ or $ Df(x)(1) $ is also the same as above definition.

I am confused. The infinitesimal "h" here is $1$. But isnt "h" supposed to be as small as possible and moreover it maybe that $ |b-a| < 1 $ ? How is this interpretation brought about?

share|cite|improve this question
$h$ is not small, it's just some direction. The thing getting small is the parameter $s$ in $\lim_{s \rightarrow 0}\frac{1}{s}[f(x+sh)-f(x)]$. – Nick Alger Sep 3 '12 at 6:54
in that case...still $ s $ shrinking to $ 0 $ is same as saying $ sh =t $ getting small right??The $ "h" $ in the question is actually the $ "t" $ here. So in general $ "t" $ is supposed to be a vector in the neighbourhood of the zero vector so that $ x+t $ is in the open domain right?domain here is $ (a,b) $ I was asking how $ 1 $ crept into the scheme of things?? – Vishesh Sep 3 '12 at 7:11
How much would the function change if you walked 1 unit in the direction of $h$, assuming the function was replaced with it's best linear approximation at point $x$. – Nick Alger Sep 3 '12 at 7:52
up vote 1 down vote accepted

When you start doing multivariable calculus you learn that the derivative of a function $f$ at a point $p$ should not be viewed as a number (or a vector) but as a linear map $Df(p):\ T_p\to T_{f(p)}$. Then of course one has to answer the question how this fits with the well-known derivative from one-variable calculus. The latter is defined by $$f'(p)=\lim_{h\to 0}{f(p+h)-f(p)\over h}$$ and is a number or a vector depending on the type of $f$. The equation defining $f'(p)$ may also be written in the denominator-free form $$f(p+h)-f(p)= h f'(p)+o(|h|)\qquad(h\to 0)\ ,$$ from which we immediately deduce that the Fréchet derivative of $f$ at $p$ is the linear map $df(p):\ h\mapsto h f'(p)$.

Now to your question: When the Fréchet derivative $df(p)$ (a linear map) is given to us, how can we extract from it the "old" derivative $f'(p)$? The answer is: Apply $df(p)$ to the "special" vector $h:=1\in{\mathbb R}^1$, and you get $df(p)(1)=1 f'(p)=f'(p)$.

share|cite|improve this answer
Oh i get it now, thanks.I was trying to make sense of $ h = 1 $ in $ f(p+h) $.I never looked at it as a mapping. – Vishesh Sep 3 '12 at 8:36

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.