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

I have tried to show that every metric space $(X,d)$ is homeomorphic to a bounded metric space. My book gives the hint to use a metric $d'(x,y)=\mbox{min}\{1,d(x,y)\}$.

If we can show that $d(x,y) \le c_1 \cdot d'(x,y)$ with $c_1$ some positive constant and $d'(x,y) \le c_2 \cdot d(x,y)$ for $c_2$ some positive constant, then the identity map $i:(X,d) \to (X,d')$ is continuous, and also obviously a bijection, thus showing that $(X,d)$ is homeomorphic to $(X,d')$, where $(X,d')$ is bounded, thus giving the desired result.

Suppose $d(x,y)<1$. Then $d'(x,y)=d(x,y)$. If $d(x,y)\ge 1$, then $d'(x,y) \le d(x,y)$. Thus we can set $c_2 = 1$ and $d'(x,y) \le d(x,y)$ for all $x,y \in X$.

But when $d(x,y)>1$ why won't it always be the case that $c_1$ will depend on what $d(x,y)$ is?

share|cite|improve this question
Note: You are not trying to show that the two metrics are strongly equivalent; you are trying to show that they are topologically equivalent. In fact, the two metrics need not be strongly equivalent, since $d'$ is bounded but $d$ need not be. – Arturo Magidin Jun 10 '12 at 2:29
Requiring $d(x,y)\leq c_1\,d'(x,y)$ is for some $c_1$ and $d'(x,y)\leq c_1\,d(x,y)$ for some $c_2$ would show that $i$ and its inverse are Lipschitz continuous. This is a stronger condition than $i$ and its inverse being continuous (which is what is required). – matt Jun 10 '12 at 2:30
up vote 12 down vote accepted

You’re working too hard: just show that $d$ and $d'$ generate the same open sets. Remember, a set $U$ is $d$-open if and only if for each $x\in U$ there is an $\epsilon_x>0$ such that $B_d(x,\epsilon_x)\subseteq U$. Once you have that $\epsilon_x$ that’s small enough, you can use any smaller positive $\epsilon$ just as well, so you might as well assume that $\epsilon_x<1$. Can you take it from there?

share|cite|improve this answer
If $\varepsilon_x < 1$ then we will also have $B_{d'}(x,\varepsilon_x) \subset U$, so that $U$ is also $d'$-open. Also if we let $\delta_y < 1$ be such that $B_{d'}(y,\delta_y) \subset V$, then $B_d(y,\delta_y) \subset V$ as well, so $V$ is $d$-open. Since all open sets in $(X,d)$ are open in $(X,d')$ and vice versa, $(X,d)$ is topologically equivalent to $(X,d')$, a bounded metric space. This works, right? – Alex Petzke Jun 10 '12 at 20:43
@Alex: Yes, that’s exactly right. – Brian M. Scott Jun 10 '12 at 20:49
Very good, thanks. – Alex Petzke Jun 10 '12 at 21:58

Well, I am just now getting into topology, but I was struggling with this same issue. Are you using Davis? I like the book, but his hint is not very useful for the complication you pointed out. We cannot ensure equivalent topology when we also restrict the points within a d-ball to a distance <1. After talking with a more experienced student I was given a much more useful hint....

Consider for all x,y in X the metric d'(x,y):=d(x,y)/[1 + d(x,y)] . It is not difficult to see that this metric is bounded and now we do not need to put distance constraints on x,y in our d-ball. Using metric d', we can make a triangle inequality argument to show our inclusions.

Sorry for leaving the actual proof out, but I was really grateful for this new hint.

share|cite|improve this answer
This does not provide an answer to the question. To critique or request clarification from an author, leave a comment below their post - you can always comment on your own posts, and once you have sufficient reputation you will be able to comment on any post. – user251257 Aug 19 '15 at 1:59

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.