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

Is the following strong form of Hausdorff equivalent to usual Hausdorff?

$X$ is strong Hausdorff if given distinct elements $x,y$ in $X$ there are open sets $U,V \subseteq X$ with $x \subseteq U$, $y \subseteq V$ and $\overline{U} \cap \overline{V} = \emptyset$.

I think that it is not equivalent, but haven't been able to prove this.

share|cite|improve this question
up vote 9 down vote accepted

No, this property is called being Urysohn (or sometimes completely Hausdorff), see Wikipedia. There we also find an example (from Steen and Seebach's book) of a Hausdorff but not Urysohn space: the relatively prime integer topology: online explanation.

share|cite|improve this answer

As Henno Brandsma's answer shows, this property is not equivalent to Hausdorffness. Another example (of my own construction, though I highly doubt it is original) is as follows:

Let $X = ( \mathbb{N} \times \mathbb{Z} ) \cup \{ -\infty , +\infty \}$. For notational convenience, by $\mathbb{Z}^{>0}$ I will denote the positive integers, and by $\mathbb{Z}^{<0}$ I will denote the negative integers.

We topologise $X$ as follows:

  • each $\langle i,n \rangle$ with $n \neq 0$ is isolated.
  • the basic open neighbourhoods of $\langle i,0 \rangle$ are of the form $\{ \langle i,0 \rangle \} \cup \{ \langle i,n \rangle : |n| \geq k \}$ for $k > 0$.
  • the basic open neighbourhoods of $-\infty$ are of the form $\{ -\infty \} \cup A$ where $A \subseteq \mathbb{N} \times \mathbb{Z}^{<0}$ is such that $\{ i \in \mathbb{N} : \{ n \in \mathbb{Z}^{<0} : \langle i,n \rangle \notin A\text{ is infinite} \} \}$ is finite.
  • the basic open neighbourhoods of $+\infty$ are of the form $\{ -\infty \} \cup A$ where $A \subseteq \mathbb{N} \times \mathbb{Z}^{>0}$ is such that $\{ i \in \mathbb{N} : \{ n \in \mathbb{Z}^{>0} : \langle i,n \rangle \notin A\text{ is infinite} \} \}$ is finite.

(The basic idea is that the subspaces $( \mathbb{N} \times \mathbb{Z}^{<0} ) \cup \{ - \infty \}$ and $( \mathbb{N} \times \mathbb{Z}^{>0} ) \cup \{ +\infty \}$ are copies of the Arens-Fort space, and each $\langle i,0\rangle$ is a limit point of the sections $\{ i \} \times \mathbb{Z}^{<0}$ and $\{ i \} \times \mathbb{Z}^{>0}$.)

It is fairly easy to show that $X$ is Hausdorff.

However, if $U$ and $V$ are (basic) open neighbourhoods of $-\infty$, $+\infty$, respectively, then there must be an $i \in \mathbb{N}$ such that $\{ n \in \mathbb{Z}^{<0} : \langle i,n \rangle \in U \}$ and $\{ n \in \mathbb{Z}^{>0} : \langle i,n \rangle \in V \}$ are both infinite, and so $\langle i,0 \rangle \in \overline{U} \cap \overline{V}$.

share|cite|improve this answer

Here's yet another example based on the examples here. For each integer $n$, let $I_n$ denote the open interval $(n,n+1)$.

Let $X = \mathbb{R} \cup \{p_0,p_1\}$, where $p_0$ and $p_1$ are distinct points not belonging to $\mathbb{R}$. Topologize $X$ by declaring $U \subseteq X$ to be open if and only if the following hold:

  • $U \cap \mathbb{R}$ is open with respect to the standard topology on $\mathbb{R}$
  • If $p_0 \in U$, then $U$ contains all but finitely many of $I_0,I_2,I_4,\ldots$
  • If $p_1 \in U$, then $U$ contains all but finitely many of $I_1,I_3,I_5,\ldots$

This space is Hausdorff. However, if $U$ is an open neighbourhood of either $p_0$ or $q_0$, then $\overline U$ contains all but finitely many of the positive integers. Thus, neighbourhoods of $p$ and $q$ can never have disjoint closures.

share|cite|improve this answer
You can also replace $\mathbb{R}$ with $\mathbb{Q}$ and the $I_n$ by their rational counterparts to get an example $X$ that is also countable and totally disconnected (having only singleton connnected subsets). This is what is done under the link. – Mike F Aug 11 '14 at 18:00

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.