Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

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

I am struggling with the idea that all compact subsets of a metric space are closed after reading chapter 2 of Rudin's Principles of Mathematical Analysis.

The reason I am confused is that it seems unreasonable that a compact subset must be closed. If we were to take an open subset $K$ of a compact subset $Y$ in a metric space $X$, where $K$ is open relative to $X$, then every finite subcover of $Y$ would also be a finite subcover of $K$. But then, $K$ would be both compact and open in $X$, which contradicts the theorem that all compact subsets in $X$ are closed.

Could someone please explain to me why my reasoning is incorrect and how it is that compact subsets of metric spaces must be closed?

Thank you,


share|cite|improve this question
In addition to what the answers have pointed out, it seems you are making the popular mistake of thinking that open and closed are opposites. They are not. A set can be open, closed, both, or neither. – dfeuer Dec 8 '13 at 7:20
up vote 10 down vote accepted

The problem with your reasoning is that $K$ may have an open cover $\mathscr{U}$ that does not arise from an open cover of $Y$, so that compactness of $Y$ doesn’t tell you anything about $\mathscr{U}$.

It’s easiest, I think, to prove the contrapositive: if $A\subseteq X$ is not closed, then $A$ is not compact. Let $\langle X,d\rangle$ be a metric space, and suppose that $A\subseteq X$ is not closed. Then there is a point $p\in(\operatorname{cl}A)\setminus A$. For $n\in\Bbb Z^+$ let $$U_n=\left\{x\in X:d(x,p)>\frac1n\right\}\;,$$ and show that $\{U_n:n\in\Bbb Z^+\}$ is an open cover of $A$ with no finite subcover.

share|cite|improve this answer
Thank you for your explanation! I understand it now. – Evan Phibbs Dec 8 '13 at 20:00
@Evan: You’re welcome! – Brian M. Scott Dec 8 '13 at 20:06

Let $F$ be our compact set. Then take $x\in X\setminus F$. Because {$x$} and $F$ are compact ,then they have a distance,say $d$.And thus the open disk $D(x,\frac {d}{2})\subset X\setminus F$. So $X\setminus F$ is open and thus $F$ is closed.

share|cite|improve this answer

Alternate proof if you know that compactness is equivalent to sequential compactness: suppose $Y$ is compact, and $x$ is a limit point of $Y$. Let $(x_n) \subset Y$ with $x_n \to x \in X$ and for all $n$, $x_n \neq x$. $(x_n)$ has a convergent subsequence converging to some $y \in Y$ because $Y$ is compact. But since the whole sequence $(x_n)$ is convergent and converges to $x$, $x=y$. So $x_n \to y \in Y$. $Y$ contains all its limit points. $Y$ is closed.

share|cite|improve this answer

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.