Bolzano-Weierstrass Theorem

Imagine you are watching a pencil drop point by point on a number line or in the plane ✏️. You may not know exactly where the next point lands, but if the points stay trapped inside a bounded region, something remarkable happens: there is always a subsequence that clusters around a single point. That is the heart of the Bolzano-Weierstrass theorem.

In this lesson, students, you will learn how to:

explain the key ideas behind the Bolzano-Weierstrass theorem,
use it to reason about sequences in $\mathbb{R}$,
connect it to Cauchy sequences and completeness,
and recognize why it is one of the central ideas in Real Analysis.

We will focus on intuition first, then on the precise mathematical statement, and finally on examples that show how the theorem works in practice.

What the theorem says

The Bolzano-Weierstrass theorem is a compactness result for sequences in $\mathbb{R}$.

In its most common form for Real Analysis, it says:

$$\text{Every bounded sequence in } \mathbb{R} \text{ has a convergent subsequence.}$$

Let’s unpack this carefully.

A sequence is a list of numbers $\{x_n\}_{n=1}^{\infty}$.
The sequence is bounded if there exists a number $M>0$ such that $|x_n|\le M$ for all $n$.
A subsequence is formed by choosing terms from the original sequence in increasing order, such as $x_{n_1}, x_{n_2}, x_{n_3}, \dots$ with $n_1<n_2<n_3<\cdots$.
A subsequence converges if its terms approach a limit in $\mathbb{R}$.

So the theorem says that even if the whole sequence does not converge, a bounded sequence must contain a part that does converge. 📌

For example, the sequence $x_n = (-1)^n$ is bounded because $|x_n|\le 1$ for all $n$, but it does not converge. Still, it has convergent subsequences:

$$x_{2n}=1 \to 1, \qquad x_{2n-1}=-1 \to -1.$$

This shows the theorem in action.

Why boundedness matters

Boundedness means the sequence cannot escape to infinity. That restriction is essential. Without it, convergence of a subsequence is no longer guaranteed.

Consider $x_n=n$. This sequence is not bounded, and it has no convergent subsequence in $\mathbb{R}$, because every subsequence still increases without bound. Its terms do not stay inside any fixed interval, so there is no place for them to “settle down.”

On the other hand, if a sequence is bounded, its terms are trapped inside some interval $[a,b]$. That does not force the whole sequence to converge, but it does force enough repetition or clustering that some subsequence converges.

A useful way to picture this is to imagine throwing infinitely many darts into a small region. Even if the darts do not land in a neat pattern, they cannot all avoid each other forever. A cluster point must appear.

This idea also connects to intervals. If a bounded sequence lies in $[a,b]$, then one can repeatedly split the interval into two halves and choose the half containing infinitely many terms. Repeating this process creates nested intervals whose lengths shrink to $0$. The chosen terms inside those intervals produce a convergent subsequence.

A simple proof idea in $\mathbb{R}$

students, the full proof is often built using interval subdivision. Here is the main idea in a student-friendly way.

Start with a bounded sequence $\{x_n\}$ inside an interval $[a,b]$.

Split $[a,b]$ into two halves.
At least one half must contain infinitely many terms of the sequence.
Keep that half and split it again.
Continue forever.

This gives a nested family of closed intervals

$$I_1 \supset I_2 \supset I_3 \supset \cdots$$

with lengths shrinking to $0$.

Because the intervals are nested and their lengths go to $0$, they have exactly one common point $L$.

Now choose one term of the sequence from each interval, making sure the indices increase. This produces a subsequence $\{x_{n_k}\}$.

Since the interval lengths shrink to $0$ and each selected term stays inside the corresponding interval, the subsequence must approach the common point $L$. Therefore,

$$x_{n_k} \to L.$$

This argument shows why boundedness leads to a convergent subsequence. The sequence is not necessarily convergent itself, but it cannot avoid producing convergent behavior somewhere inside it.

Examples and counterexamples

Let’s work through a few examples to build intuition.

Example 1: Alternating sequence

Take $x_n = (-1)^n$.

This sequence is bounded, but it jumps between $1$ and $-1$. It does not converge. However, it has two obvious convergent subsequences:

$$x_{2n}=1 \to 1, \qquad x_{2n-1}=-1 \to -1.$$

This is a classic example of Bolzano-Weierstrass. The sequence itself is chaotic in the sense that it does not settle, but boundedness still forces convergent subsequences.

Example 2: Sine values

Consider $x_n=\sin n$.

Since $-1\le \sin n\le 1$, the sequence is bounded. It does not converge, but by Bolzano-Weierstrass it has a convergent subsequence. In fact, there are many such subsequences, because the values of $\sin n$ keep moving around inside $[-1,1]$.

Example 3: Unbounded sequence

Take $x_n=n^2$.

This sequence is not bounded, so Bolzano-Weierstrass does not apply. Indeed, no subsequence can converge in $\mathbb{R}$ because every subsequence still tends to infinity.

These examples show the boundary of the theorem: boundedness is the key hypothesis.

Connection to Cauchy sequences and completeness

Bolzano-Weierstrass fits into the larger topic of completeness in an important way.

A sequence $\{x_n\}$ is Cauchy if for every $\varepsilon>0$ there exists $N$ such that whenever $m,n\ge N$,

$$|x_n-x_m|<\varepsilon.$$

Cauchy sequences are sequences whose terms eventually become arbitrarily close to each other. In a complete space like $\mathbb{R}$, every Cauchy sequence converges.

How does Bolzano-Weierstrass relate?

Every convergent sequence is bounded.
Every Cauchy sequence is bounded.
By Bolzano-Weierstrass, every bounded sequence has a convergent subsequence.
In $\mathbb{R}$, this helps support the completeness picture: boundedness leads to subsequences with limits, and Cauchy behavior guarantees a single limit for the whole sequence.

Here is the connection in a practical way:

If a sequence is Cauchy, then it is bounded, so Bolzano-Weierstrass gives a convergent subsequence. Suppose $x_{n_k}\to L$. Because the whole sequence is Cauchy, the terms of the sequence eventually stay close to each other. That forces the full sequence to approach the same limit $L$.

This logic is one of the ways completeness is used in analysis: once a sequence has enough internal stability, the real numbers provide a limit for it.

Why the theorem matters in Real Analysis

Bolzano-Weierstrass is more than a fact about sequences. It is a tool used throughout analysis because it turns boundedness into convergence somewhere inside the sequence. This is useful when exact convergence is hard to prove.

For example, if you are studying a complicated process that produces a bounded sequence of measurements, the theorem guarantees that some subsequence of measurements becomes stable. That is often enough to prove existence results.

It also appears in proofs involving:

optimization, where maximizing or minimizing sequences are bounded,
compactness arguments, where every sequence has a convergent subsequence,
and the study of continuous functions on closed intervals, where bounds and limits work together.

The theorem also has a geometric version in $\mathbb{R}^n$: every bounded sequence in $\mathbb{R}^n$ has a convergent subsequence. The intuition is the same, but the proof uses the fact that each coordinate sequence is bounded and therefore has a convergent subsequence.

Common misunderstandings

A few points often cause confusion:

A bounded sequence does not have to converge.
The theorem guarantees a convergent subsequence, not necessarily the whole sequence.
The limit of the subsequence does not have to be a term from the original sequence.
Unbounded sequences are outside the theorem’s guarantee.

A helpful mental check is this: if the sequence is trapped in a box, it must have a clustering behavior. If it can run off to infinity, that guarantee disappears.

Conclusion

Bolzano-Weierstrass is a cornerstone theorem in Real Analysis. It tells us that bounded sequences in $\mathbb{R}$ cannot stay completely scattered forever; they must contain a convergent subsequence. This idea is closely linked to the completeness of $\mathbb{R}$ and to the behavior of Cauchy sequences.

students, the key takeaway is simple but powerful: boundedness creates compactness-like behavior, and that behavior produces convergence somewhere inside the sequence. This is one of the main reasons the real numbers are such a well-behaved number system in analysis. ✅

Study Notes

The Bolzano-Weierstrass theorem says that every bounded sequence in $\mathbb{R}$ has a convergent subsequence.
A bounded sequence stays inside some interval $[a,b]$ or satisfies $|x_n|\le M$ for some $M>0$.
A subsequence is formed by choosing terms with increasing indices.
The theorem does not say the whole sequence converges.
Example: $x_n=(-1)^n$ is bounded and has convergent subsequences $x_{2n}\to 1$ and $x_{2n-1}\to -1$.
Example: $x_n=n$ is unbounded, so the theorem does not apply.
The proof idea uses nested intervals whose lengths shrink to $0$.
Cauchy sequences in $\mathbb{R}$ are bounded, and in a complete space like $\mathbb{R}$ they converge.
Bolzano-Weierstrass helps connect boundedness, subsequences, and completeness in Real Analysis.