The proof is about the statement: if $f$ is an uniformly continuous function and $(x_n)$ is a Cauchy sequence then $f(x_n)$ is a Cauchy sequence.

A sequence $(x_n)$ can be defined as Cauchy as

$$\forall\delta>0,\exists N\in\Bbb N:|x_i-x_j|<\delta\quad\quad\forall i,j>N\tag{1}$$

And a definition of uniformly convergent function on some set $A$ can be

$$\forall\varepsilon>0,\exists\delta>0,\forall x,y\in A: |x-y|<\delta\implies|f(x)-f(y)|<\varepsilon\tag{2}$$

Then if we combine both statement we can have a "proof" that uniformly continuous functions preserves Cauchy sequences. For some $\varepsilon>0$ and some Cauchy sequence $(x_n)$ we have that


I dont see why this last statement can proof the preservation of Cauchy sequences for some uniformly continuous function. My problem is that I cant deduce from $(3)$ that this implies the $\forall\delta>0$ required in the definition of a Cauchy sequence.

The statement $(3)$ just is saying that exist some $\delta>0$ on the domain for a Cauchy sequence $f(x_n)$ in the range (because I can choose statements for all $\varepsilon>0$).

So I get lost somewhere, there is something that Im not seeing, can you help me please? Thank you in advance.

  • $\begingroup$ The confusion comes from deltas being used in two different definitions being used slightly differently. $\endgroup$ – Doug M Jun 2 '16 at 1:36

You want to show that the sequence $(f(x_n))$ is a Cauchy sequence. What is the criteria for that? For every $\epsilon>0$, there exists $N>0$ such that $|f(x_i)-f(x_j)|<\epsilon.$

Now, combine the given definitions, for every $\epsilon>0$, there exists $\delta$ such that $|x-y|<\delta$ implies $|f(x)-f(y)|<\epsilon$. So, choose $N$, this can be done as $x_n$ is a Cauchy sqeuence, such that $|x_i-x_j|<\delta$ for all $i,j>N$. Then, this same $N$ satisfies what you're looking for.

  • $\begingroup$ Ah! I finally understand. Now my next step is to see why I dont wasnt seeing it before. I was focused on epsilon-delta but the important thing is the $N$. The $|x_i-x_j|$ is a Cauchy tail where you can choose eventually any $N$. Thank you very much. $\endgroup$ – Masacroso Jun 2 '16 at 2:01

I don't know if you had follow, but you prove that (3) holds for any $\varepsilon >0$ That any $\varepsilon >0$ is the same that your delta in the definition of a Cauchy sequence. But... I would recommend you to see this prove in a more visual way. Do a drawing.

  • $\begingroup$ I think I understand. I can see it visually but I cant understand the proof. $\endgroup$ – Masacroso Jun 2 '16 at 1:39
  • $\begingroup$ The problem here is that you can't use delta for 2 different things. So you need a new letter. $\endgroup$ – PrinceAndrew Jun 2 '16 at 1:42

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.