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

In the book I am using, the author defines differentials in the following way. Given smooth manifolds $M,N$ and a smooth mapping $\psi:M\to N$ define the differential $d\psi_m$ at a point $m\in M$ as the following linear mapping: ($M_n$ denotes the tangent space at n, defined here as the space of linear derivations on germs of $m$) $$ d\psi_m:M_m\to N_{\psi(m)} $$ with $d\psi_m(v)g:=v(g\circ\psi)$ for germs $g$ on $\psi(m)$.

If $N=\Bbb R,\,\,\,\,\,d\psi(v)=v(\psi)\frac{d}{dx}|_{\psi(m)}$ which is "identified" with the dual of $d\psi_m$ evaluated at the dual vector of $\frac{d}{dx}|_{\psi(m)}$, namely $d\psi_m(v)=v(f)$ . Why can we identify them?

share|cite|improve this question
up vote 3 down vote accepted

In general, the differential maps each tangent space of $M$ linearly into the appropriate tangent space of $N$. But when $N=\mathbb R$, each tangent space $N_{\psi(m)}$ is canonically isomorphic (as a vector space) to $\mathbb R$, via the map $a\frac{\partial}{\partial x}_{\big| x=\psi(m)} \mapsto a$. When two objects are canonically isomorphic to one another, we can think of them as being the same, as long as we consider only structures covered by the notion of isomorphism at hand. Therefore, we can think of $d\psi_m$ as a linear map from $M_m$ to $\mathbb R$.

On the other hand, we already have a linear map from $M_m$ to $\mathbb R$ — the evaluation of tangent vectors on $\psi$, namely $v\mapsto v(\psi)$. Looking at the definition of $d\psi_m$ again, we realize it's the same map. So... much ado about nothing, really.

share|cite|improve this answer
+1 Nice answer. – Jesse Madnick May 1 '13 at 21:49
Just to build off of this: Ultimately, you're asking why we can identify $v(\psi) \left.\frac{d}{dx}\right.|_{\psi(m)} \in \mathbb{R}_{\psi(m)}$ with $v(\psi) \in \mathbb{R}$. Well, $\mathbb{R}_{\psi(m)}$ is a $1$-dimensional vector space with basis $\{\frac{d}{dx}|_{\psi(m)}\}$ that is canonically isomorphic to $\mathbb{R}$ via the isomorphism described in this answer: $a \left.\frac{d}{dx}\right.|_{\psi(m)} \mapsto a$. – Jesse Madnick May 1 '13 at 21:50
The answer seems so trivial now. Thanks – Nicolas Bourbaki May 2 '13 at 17:33

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.