2
$\begingroup$

I am ttying to prove that every bounded, infinite set has at least one limit point through the concept of open covers and without using Bolzano Weierstrass. I have no idea where to start. So far I've only assumed that a set A is bounded and infinite, so for some real number M, the set A is entirely contained in the closed and bounded set [-M, M]. By Heine Bornel I know that each open cover for [-M, M] has a finite subcover, but honestly I don't know where to go from there. Am I in the right direction?

$\endgroup$
  • $\begingroup$ What you are trying to prove is Bolzano-Weiertrass. And yes, your starting point is correct : assume that there is no limit point, and try to construct an open cover which cannot have a finite subcover. $\endgroup$ – Captain Lama Mar 29 '16 at 14:18
7
$\begingroup$

Suppose $S$ is an infinite set contained in the interval $I=[-M,M]$. If $S$ has no limit points in interval $I$, then for every $x\in [-M,M]$ there is $r_x>0$ such that the ball $B(x,r_x)$ contains only finitely many points of $S$.

Notice that the family $\{B(x,r_x):x\in [-M,M]\}$ is an open cover of $[-M,M]$, so there are points $x_1,\ldots,x_n$ such that

$$[-M,M]\subseteq \bigcup_{i=1}^n B(x_i,r_{x_i}).$$

Now, by piggeonhole principle, $S\cap B(x_i,r_{x_i})$ must be infinite for some $i=1,\ldots,n$. This contradicts the choice of $r_{x_i}$.

$\endgroup$
1
$\begingroup$

You can suppose the interval is $[0,1]$. If $A$ has no limit point, then for each $x\in[0,1]$ you can find its open neighborhood $U_x$ such that $U_x\cap A = \{x\}$. These from an open cover, so from compactness there are $U_{x_1}, \ldots, U_{x_n}$ covering $[0,1]$. This means $A = \{x_0, \ldots, x_n\}$, contradiction.

$\endgroup$

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.