Let $f_n, f : D \to \mathbb{R}$ with $A(\neq \phi) \subset \mathbb{R}$

Here the $f_n$ are the sequence of the continuous functions.

I already knew the fact "$f_n$ is uniformly converge to $ f$ on $D$ $\Rightarrow$ $f$ is continuous on $D$"

But There is a lecturer's claim that

Say $f_n$ is continuous on $A$

"$f_n$ is uniformly converge to $ f$ on $A$ $\not\Rightarrow$ $f$ is continuous on $A$" (statement (*))

He suggested the counter-example as like the below.

$f_n, f : \mathbb{R} \to \mathbb{R}$ with $A(\neq \phi) \subset \mathbb{R}$

Here the $D = \mathbb{R}, A = \mathbb{Q}^c$ and

Since the rational number set,$\mathbb{Q}$ is countable so $\mathbb{Q} = \{x_1, x_2, .... \}$

Then Defining each functions like the below

$f_n(x) = \begin{cases}1 & \text{$x \in \{ x_1, x_2, ...,x_n\}$} \\ 0 & \text{$o.w.$}\end{cases}$

$f(x) = \begin{cases} 1 & \text{$x \in \mathbb{Q} $ } \\ 0 & \text{$x \in \mathbb{Q}^c$} \end{cases}$

From here, my question begins.

First question) I put in to the sequence $x_{n+1} \in \mathbb{Q}$ to $\Vert f_n(x) - f(x) \Vert$

Then $\Vert f_n(x_{n+1}) - f(x_{n+1}) \Vert$ = $\Vert 0-1 \Vert = 1 \geq \epsilon(= {1 \over 2}) $

So my conclusion is this counterexample is totally wrong, since $f_n$ is not uniformly converge to $f$. What do you think about that? Is my thought wrong?

Second question) Like the lecture's thought, I guess the statement (*) is false. But I can't find any counterexample. If the lecture's counterexample is false, Please give me a counterexample.

p.s.) If the statement is true, Why does the statement(*) hold?



1 Answer 1


There seems to be no connection between $D$ and $A$ in $(*)$. So $(*)$ is correct. For one example take $D=[0, \frac 12 ], A=[0,1]$, $f_n(x)=x^{n}, f(x)=0$ for $ x<1$, $f(1)=1$. Then $f_n \to f$ uniformly on $D$ and $f$ is not continuous on $A$.

However, the example given by the lecturer is wrong since $f_n$ does not converge uniformly to $f$ on $D=\mathbb R$.

Answer for the revised question: If $f_n \to f$ uniformly on $A$ and each $f_n$ is continuous on $A$ then the restriction of $f$ to $A$ is continuous. But you cannot say that $f$ is continuous at points of $A$ as in the example by your lecturer.

  • $\begingroup$ Thanks for reply, Dear Mr,@Murthy. But I should have been asked "$f_n$ is uniformly converge to $ f$ on $A$" in statement (*) not "$f_n$ is uniformly converge to $ f$ on $D$" Sorry for typo. I edited my question. $\endgroup$ Jan 28, 2020 at 8:25
  • $\begingroup$ @se-hyuckyang I have edited my answer. $\endgroup$ Jan 28, 2020 at 8:25
  • $\begingroup$ Ah yes! Thank you. :) $\endgroup$ Jan 28, 2020 at 8:27
  • $\begingroup$ What about the case if we add the condition $A \subset D$? Though adding more condition, his claim and counterexample are still not hold, right? $\endgroup$ Jan 28, 2020 at 14:31
  • $\begingroup$ @se-hyuckyang Actually $f_n(x)=f(x)=0$ for all $x$ in $A$ in his example. So $f_n \to f$ uniformly on $A$. $\endgroup$ Jan 28, 2020 at 23:15

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.