Uniqueness of Limits for Sequences and Subsequences in Metric Spaces

February 17, 2026 · Luciano Muratore

Uniqueness of Limits for Sequences and Subsequences in Metric Spaces

Let (X,d)(X, d) be a metric space.

Let (xn)(x_n) be a sequence in XX that converges to xXx \in X, and let (xnk)(x_{n_k}) be a subsequence of (xn)(x_n) that converges to yXy \in X.

We prove that

x=y.x = y.

Proof

Let ε>0\varepsilon > 0 be arbitrary.
We aim to show that

d(x,y)<ε.d(x, y) < \varepsilon.

Step 1 — Triangle Inequalities

By the triangle inequality,

d(x,y)d(x,xn)+d(xn,y).d(x, y) \le d(x, x_n) + d(x_n, y).

Again applying the triangle inequality,

d(xn,y)d(xn,xnk)+d(xnk,y).d(x_n, y) \le d(x_n, x_{n_k}) + d(x_{n_k}, y).

Therefore,

d(x,y)d(x,xn)+d(xn,xnk)+d(xnk,y).d(x, y) \le d(x, x_n) + d(x_n, x_{n_k}) + d(x_{n_k}, y).

Step 2 — Control Each Term

Since xnxx_n \to x, there exists N1NN_1 \in \mathbb{N} such that for all n>N1n > N_1,

d(x,xn)<ε3.d(x, x_n) < \frac{\varepsilon}{3}.

Since xnkyx_{n_k} \to y, there exists N2NN_2 \in \mathbb{N} such that for all k>N2k > N_2,

d(xnk,y)<ε3.d(x_{n_k}, y) < \frac{\varepsilon}{3}.

Because (xn)(x_n) converges, it is a Cauchy sequence. Therefore, there exists N3NN_3 \in \mathbb{N} such that for all n,k>N3n, k > N_3,

d(xn,xnk)<ε3.d(x_n, x_{n_k}) < \frac{\varepsilon}{3}.

Step 3 — Combine the Estimates

Let

N=max{N1,N2,N3}.N = \max\{N_1, N_2, N_3\}.

Then for all n,k>Nn, k > N, we obtain

d(x,y)<ε3+ε3+ε3=ε.d(x, y) < \frac{\varepsilon}{3} + \frac{\varepsilon}{3} + \frac{\varepsilon}{3} = \varepsilon.

Step 4 — Conclude

Since ε>0\varepsilon > 0 was arbitrary, it follows that

d(x,y)=0.d(x, y) = 0.

Hence,

x=y.x = y. \blacksquare

Why This Matters

This result formalizes an intuitive idea:

A subsequence cannot “escape” the limit of a convergent sequence.

It guarantees the uniqueness of limits in metric spaces and is a fundamental tool in:

  • Real Analysis
  • Topology
  • Functional Analysis
  • Compactness arguments

Without uniqueness, convergence theory would collapse.