Neighborhoods and Local Structure | Malone Math

The Central Idea

A neighborhood describes the local environment around a point.

If \(X\) is a topological space and \(x\in X\), a set \(N\subseteq X\) is a neighborhood of \(x\) if there exists an open set \(U\) such that

\[ x\in U\subseteq N. \]

Thus a neighborhood need not itself be open.

It only needs to contain an open set around the point.

Open Neighborhoods

If a neighborhood \(N\) is itself open and contains \(x\), then it is an open neighborhood of \(x\).

Thus every open set containing \(x\) is automatically a neighborhood of \(x\).

Open neighborhoods are often the most convenient neighborhoods to use in proofs.

Neighborhoods in Metric Spaces

In a metric space \((X,d)\), open balls provide the standard local building blocks.

The open ball of radius \(r>0\) centered at \(x\) is

\[ B_r(x)=\{y\in X:d(x,y) A set \(N\) is a neighborhood of \(x\) if some open ball around \(x\) lies inside \(N\).

\[ \boxed{ N\text{ is a neighborhood of }x \iff \exists r>0\text{ such that }B_r(x)\subseteq N }. \]

Neighborhoods on the Real Line

In \(\mathbb{R}\) with the usual metric,

\[ B_r(x)=(x-r,x+r). \]

For example,

\[ (2,8) \]

is a neighborhood of

\[ x=5 \]

because some open interval centered at \(5\) lies inside \((2,8)\).

A Neighborhood Need Not Be Open

Consider

\[ N=[0,2] \]

in \(\mathbb{R}\).

The set is not open in \(\mathbb{R}\).

But it is a neighborhood of

\[ x=1 \]

because, for example,

\[ \left(\frac12,\frac32\right) \subseteq [0,2]. \]

A Set Can Fail to Be a Neighborhood of One of Its Points

The set

\[ [0,2] \]

contains the point \(0\).

But it is not a neighborhood of \(0\) in \(\mathbb{R}\), because every open interval around \(0\) contains negative numbers.

Therefore no open interval around \(0\) lies entirely inside \([0,2]\).

Neighborhoods and Interior Points

A point \(x\) is an interior point of a set \(A\) exactly when \(A\) is a neighborhood of \(x\).

\[ \boxed{ x\in\operatorname{int}(A) \iff A\text{ is a neighborhood of }x }. \]

This connects neighborhood language directly to interior.

Open Sets Through Neighborhoods

A set \(U\) is open exactly when it is a neighborhood of each of its points.

\[ \boxed{ U\text{ is open} \iff U\text{ is a neighborhood of every }x\in U }. \]

This gives a purely local characterization of openness.

Boundary Through Neighborhoods

A point \(x\) lies on the boundary of \(A\) if every neighborhood of \(x\) meets both \(A\) and its complement.

\[ \boxed{ x\in\partial A \iff N\cap A\ne\varnothing \text{ and } N\cap A^c\ne\varnothing \text{ for every neighborhood }N\text{ of }x }. \]

Closure Through Neighborhoods

A point \(x\) belongs to the closure of \(A\) if every neighborhood of \(x\) intersects \(A\).

\[ \boxed{ x\in\overline{A} \iff N\cap A\ne\varnothing \text{ for every neighborhood }N\text{ of }x }. \]

This definition works in arbitrary topological spaces.

Exterior Through Neighborhoods

A point \(x\) lies in the exterior of \(A\) when some neighborhood of \(x\) avoids \(A\) completely.

\[ \boxed{ x\in\operatorname{ext}(A) \iff \exists N\text{ neighborhood of }x \text{ with }N\cap A=\varnothing }. \]

Limit Points Through Neighborhoods

A point \(x\) is a limit point of \(A\) if every neighborhood of \(x\) contains a point of \(A\) other than \(x\) itself.

\[ \boxed{ x\text{ limit point of }A \iff (N\setminus\{x\})\cap A\ne\varnothing } \]

for every neighborhood \(N\) of \(x\).

Neighborhood Systems

The collection of all neighborhoods of a point \(x\) is sometimes denoted

\[ \mathcal{N}(x). \]

This neighborhood system records the entire local topological environment around \(x\).

Much of a topology can be reconstructed from these neighborhood systems.

Basic Properties of Neighborhood Systems

For every point \(x\), neighborhoods satisfy several natural properties.

Every neighborhood of \(x\) contains \(x\).
Any set containing a neighborhood of \(x\) is also a neighborhood of \(x\).
The intersection of finitely many neighborhoods of \(x\) is a neighborhood of \(x\).
Every neighborhood of \(x\) contains an open neighborhood of \(x\).

Intersection of Neighborhoods

If \(N_1\) and \(N_2\) are neighborhoods of \(x\), then

\[ N_1\cap N_2 \]

is also a neighborhood of \(x\).

Each contains some open set around \(x\), and the intersection of two open neighborhoods is again an open neighborhood of \(x\).

Neighborhood Bases

A neighborhood basis at \(x\) is a collection of neighborhoods that is sufficient to describe all neighborhoods of \(x\).

A family

\[ \mathcal{B}_x \]

is a neighborhood basis at \(x\) if every neighborhood of \(x\) contains some member of \(\mathcal{B}_x\).

Open Balls Form a Neighborhood Basis in Metric Spaces

In a metric space, the family

\[ \{B_r(x):r>0\} \]

forms a neighborhood basis at \(x\).

Every neighborhood of \(x\) contains some sufficiently small open ball centered at \(x\).

A Countable Neighborhood Basis in Metric Spaces

It is not necessary to use every positive radius.

The balls

\[ B_{1/n}(x), \qquad n=1,2,3,\ldots \]

already form a neighborhood basis at \(x\).

Given any radius \(r>0\), choose \(n\) large enough that

\[ \frac1n Then

\[ B_{1/n}(x)\subseteq B_r(x). \]

First Countability

A space is first countable if every point has a countable neighborhood basis.

Every metric space is first countable.

First countability helps explain why sequences work so well in metric spaces.

Local Structure

The local structure of a space near \(x\) refers to the topology of sufficiently small neighborhoods of \(x\).

Two spaces can be globally very different while having the same local structure around every point.

Local vs. Global Properties

A local property can be checked in neighborhoods around individual points.

A global property concerns the entire space.

For example:

being locally Euclidean is local;
connectedness is global;
compactness is global;
local connectedness is local;
genus is global.

A Circle Is Locally Like a Line

The circle

\[ S^1 \]

is globally different from the real line.

But near each point, a sufficiently small arc of the circle is homeomorphic to an open interval.

Thus the circle is locally one-dimensional and locally Euclidean.

A Sphere Is Locally Like a Plane

The sphere

\[ S^2 \]

is globally compact and curved.

But every sufficiently small neighborhood on the sphere is homeomorphic to an open disk in

\[ \mathbb{R}^2. \]

This local planar structure makes \(S^2\) a two-dimensional manifold.

A Torus Is Also Locally Planar

The torus has a global handle, unlike the sphere.

Yet near every point, it also looks like an open region of

\[ \mathbb{R}^2. \]

Therefore the sphere and torus have the same local dimension but different global topology.

Local Similarity Does Not Imply Global Homeomorphism

Both \(S^2\) and \(T^2\) are locally homeomorphic to \(\mathbb{R}^2\).

But

\[ \chi(S^2)=2 \]

while

\[ \chi(T^2)=0. \]

Therefore they are not globally homeomorphic.

Local sameness is weaker than global equivalence.

Subspace Neighborhoods

If \(Y\subseteq X\) has the subspace topology, neighborhoods inside \(Y\) are determined relative to \(Y\).

A set \(N\subseteq Y\) is a neighborhood of \(y\in Y\) if it contains a set of the form

\[ U\cap Y, \]

where \(U\) is open in \(X\) and

\[ y\in U\cap Y. \]

Endpoint Neighborhoods in a Subspace

Let

\[ Y=[0,1]\subseteq\mathbb{R}. \]

Relative to \(Y\), a neighborhood of \(0\) can look like

\[ [0,\varepsilon). \]

This set is not open in \(\mathbb{R}\), but it is open in the subspace topology on \(Y\).

Local Structure at Boundary Points

Interior points of a two-dimensional manifold with boundary have neighborhoods resembling open disks.

Boundary points have neighborhoods resembling half-disks.

This difference is local and allows the boundary to be recognized intrinsically.

Local Connectedness

A space is locally connected if every point has a neighborhood basis consisting of connected open sets.

Informally, one can zoom in around any point and find arbitrarily small connected neighborhoods.

Local Path Connectedness

A space is locally path connected if every point has a neighborhood basis consisting of path connected open sets.

This is stronger than local connectedness.

\[ \boxed{ \text{locally path connected} \Longrightarrow \text{locally connected} }. \]

Euclidean Spaces Are Locally Path Connected

Every point in

\[ \mathbb{R}^n \]

has arbitrarily small open balls.

Open balls are convex.

Convex sets are path connected.

Therefore,

\[ \boxed{\mathbb{R}^n\text{ is locally path connected}}. \]

Open Subsets of Euclidean Space

\[ U\subseteq\mathbb{R}^n \]

is open, every point of \(U\) has a sufficiently small open ball contained in \(U\).

Hence every open subset of Euclidean space is locally path connected.

Why Local Path Connectedness Matters

In locally path connected spaces, connected components and path components coincide.

This explains why connectedness and path connectedness are often interchangeable in familiar geometric settings.

The distinction appears more sharply in spaces with unusual local structure.

The Topologist's Sine Curve Revisited

The closure of the graph

\[ y=\sin\left(\frac1x\right), \qquad 0 together with the vertical limit segment at \(x=0\), is connected but not path connected.

Its behavior near the vertical segment is an important example of pathological local structure.

Local Compactness

A space is locally compact if each point has a neighborhood whose closure is compact, or equivalently under common Hausdorff conventions, each point has a compact neighborhood.

The exact formulation may vary slightly across texts, especially without separation assumptions.

Euclidean Space Is Locally Compact

Around any point

\[ x\in\mathbb{R}^n, \]

choose an open ball

\[ B_r(x). \]

Its closure is the closed ball

\[ \overline{B_r(x)}. \]

Closed bounded subsets of \(\mathbb{R}^n\) are compact by Heine-Borel.

Therefore \(\mathbb{R}^n\) is locally compact.

Locally Compact Does Not Mean Compact

The real line

\[ \mathbb{R} \]

is locally compact because every point has compact closed intervals nearby.

But \(\mathbb{R}\) itself is not compact.

Thus,

\[ \boxed{ \text{locally compact} \centernot\Longrightarrow \text{compact} }. \]

Compact Spaces Are Locally Compact in Common Hausdorff Settings

In a compact Hausdorff space, each point has neighborhoods with compact closure.

Thus compact Hausdorff spaces provide natural examples of locally compact spaces.

Again, local compactness concerns what happens near each point rather than the size of the whole space.

Local Finiteness

A family of subsets

\[ \{A_\alpha\} \]

is locally finite if every point has a neighborhood intersecting only finitely many members of the family.

Local finiteness is useful when assembling global objects from many local pieces.

Why Local Finiteness Matters

A construction may involve infinitely many sets globally while remaining finite in every local neighborhood.

This balance between global infinity and local finiteness is important in:

partitions of unity;
manifold theory;
triangulations;
sheaf theory;
analysis on spaces.

Local Homeomorphisms

A map

\[ f:X\to Y \]

is a local homeomorphism if every point \(x\in X\) has a neighborhood \(U\) such that

\[ f|_U:U\to f(U) \]

is a homeomorphism onto an open subset of \(Y\).

Local Homeomorphism Does Not Mean Global Homeomorphism

A map can look like a homeomorphism near every point while failing to be one-to-one globally.

Covering maps provide important examples.

Thus local equivalence is not the same as global equivalence.

Example: The Exponential Map onto the Circle

Consider

\[ p:\mathbb{R}\to S^1 \]

defined by

\[ p(t)= \bigl( \cos(2\pi t), \sin(2\pi t) \bigr). \]

Locally, \(p\) behaves like a homeomorphism.

Globally, it is not injective because

\[ p(t)=p(t+n) \]

for every integer \(n\).

Charts

A chart on a manifold is a homeomorphism from an open neighborhood of the manifold to an open subset of Euclidean space.

If \(M\) is an \(n\)-dimensional manifold, a chart has the form

\[ \varphi:U\to V\subseteq\mathbb{R}^n. \]

Charts turn local topology into coordinates.

Atlases

An atlas is a collection of charts whose domains cover a manifold.

No single coordinate system may describe the entire space.

Instead, many overlapping local coordinate systems are assembled into a global description.

The Sphere Needs Multiple Charts

The sphere

\[ S^2 \]

cannot be covered by one chart homeomorphic to an open subset of \(\mathbb{R}^2\).

But multiple overlapping charts can cover it.

This is another example of local Euclidean structure with non-Euclidean global topology.

Local Dimension

Manifolds are classified locally by dimension.

An \(n\)-dimensional manifold locally resembles

\[ \mathbb{R}^n. \]

Thus:

curves are locally one-dimensional;
surfaces are locally two-dimensional;
ordinary physical space is locally three-dimensional.

Local Dimension Is a Topological Property

Euclidean spaces of different dimensions are not locally homeomorphic.

In particular,

\[ \mathbb{R} \not\cong_{\text{loc}} \mathbb{R}^2. \]

The rigorous theory behind this includes the invariance of domain and invariance of dimension theorems.

Local Separation Behavior

The way neighborhoods respond to removing a point can reveal local structure.

In a small interval around a point of \(\mathbb{R}\), removing the center splits the neighborhood into two pieces.

In a small disk around a point of \(\mathbb{R}^2\), removing the center leaves a path connected punctured disk.

This difference helps reveal dimension.

Neighborhoods and Continuity

Continuity can be expressed entirely through neighborhoods.

A function

\[ f:X\to Y \]

is continuous at \(x\in X\) if for every neighborhood \(V\) of \(f(x)\), there exists a neighborhood \(U\) of \(x\) such that

\[ f(U)\subseteq V. \]

Neighborhood Form of Continuity

The neighborhood definition can be written:

\[ \boxed{ \forall V\in\mathcal{N}(f(x)), \quad \exists U\in\mathcal{N}(x) \text{ such that } f(U)\subseteq V }. \]

This generalizes the familiar \(\varepsilon\)-\(\delta\) idea from analysis.

Metric Continuity as Neighborhood Control

In metric spaces, the usual statement

\[ d_X(x,y)<\delta \Longrightarrow d_Y(f(x),f(y))<\varepsilon \]

says that a sufficiently small neighborhood of \(x\) maps inside a desired neighborhood of \(f(x)\).

Thus \(\varepsilon\)-\(\delta\) continuity is a metric version of neighborhood continuity.

Neighborhoods and Convergence

A sequence

\[ x_n \]

converges to \(x\) if eventually every term lies in every neighborhood of \(x\).

Symbolically,

\[ \boxed{ x_n\to x \iff \forall N\in\mathcal{N}(x), \exists N_0 \text{ such that } n\ge N_0 \Longrightarrow x_n\in N }. \]

Why Neighborhoods Generalize Distance

In a metric space, neighborhoods can be described using distances.

In a general topological space, there may be no distance function at all.

Neighborhoods preserve the local ideas needed for:

continuity;
convergence;
interior;
closure;
boundary;
local structure.

Topological Indistinguishability

In spaces without strong separation properties, two distinct points may have very similar neighborhood systems.

If two points belong to exactly the same open sets, topology may be unable to distinguish them by neighborhood information alone.

Separation axioms control this behavior.

The Hausdorff Condition Is Local Separation Between Points

A space is Hausdorff if distinct points can be placed in disjoint open neighborhoods.

\[ x\ne y, \]

then there exist neighborhoods \(U\) of \(x\) and \(V\) of \(y\) with

\[ U\cap V=\varnothing. \]

This condition ensures that distinct points are topologically well separated.

Common Error: Thinking Every Neighborhood Is Open

A neighborhood only needs to contain an open neighborhood.

For example,

\[ [0,2] \]

is a neighborhood of \(1\) in \(\mathbb{R}\), even though it is not open.

Common Error: Thinking Every Set Containing a Point Is a Neighborhood

A set must contain some open region around the point.

The singleton

\[ \{0\} \]

contains \(0\) but is not a neighborhood of \(0\) in the usual topology on \(\mathbb{R}\).

Common Error: Forgetting the Ambient Topology

A neighborhood is always relative to a topology.

The set

\[ [0,\varepsilon) \]

is not a neighborhood of \(0\) in \(\mathbb{R}\), but it is an open neighborhood of \(0\) in the subspace

\[ [0,1]. \]

Common Error: Confusing Local and Global Structure

Two spaces can look the same around every point and still differ globally.

The sphere and torus are both locally two-dimensional manifolds.

But their genus and Euler characteristic differ.

Common Error: Thinking Locally Compact Means Compact

The real line is locally compact but not compact.

Local compactness requires compact control near each point, not over the entire space.

Common Error: Thinking Locally Connected Means Connected

A space may be locally connected around every point while having several disconnected components globally.

For example,

\[ (-2,-1)\cup(1,2) \]

is locally connected but disconnected.

Common Error: Thinking Local Homeomorphism Means Homeomorphism

A map may be a homeomorphism in small neighborhoods while failing to be globally one-to-one.

The covering map

\[ \mathbb{R}\to S^1 \]

gives a standard example.

A Reliable Procedure for Neighborhood Questions

Identify the ambient topological space.
Identify the point \(x\).
Ask whether the proposed set contains an open set around \(x\).
In metric spaces, try to fit an open ball \(B_r(x)\) inside it.
In subspaces, use relative open sets of the form \(U\cap Y\).
Do not assume the neighborhood itself must be open.

A Reliable Procedure for Local Structure

Choose a point \(x\).
Examine arbitrarily small neighborhoods of \(x\).
Identify familiar spaces to which those neighborhoods are homeomorphic.
Check whether the same local description works at every point.
Separate local conclusions from global ones.
Use local invariants such as dimension, orientability near points, or separation behavior when appropriate.

Example: Is \([0,2]\) a Neighborhood of \(1\)?

Yes.

The open interval

\[ \left(\frac12,\frac32\right) \]

contains \(1\) and lies inside \([0,2]\).

Therefore,

\[ \boxed{[0,2]\text{ is a neighborhood of }1}. \]

Example: Is \([0,2]\) a Neighborhood of \(0\) in \(\mathbb{R}\)?

No.

Every open interval around \(0\) contains negative numbers, so no open interval around \(0\) lies inside \([0,2]\).

\[ \boxed{[0,2]\text{ is not a neighborhood of }0\text{ in }\mathbb{R}}. \]

Example: Is \([0,2]\) a Neighborhood of \(0\) in \([0,5]\)?

Let the ambient space be

\[ Y=[0,5]. \]

Relative to \(Y\),

\[ [0,1) = Y\cap(-1,1) \]

is open and contains \(0\).

Since

\[ [0,1)\subseteq[0,2], \]

we have

\[ \boxed{[0,2]\text{ is a neighborhood of }0\text{ in }[0,5]}. \]

Example: Local Structure of the Circle

Choose any point

\[ p\in S^1. \]

A sufficiently small open arc around \(p\) is homeomorphic to an open interval.

Therefore \(S^1\) is locally Euclidean of dimension \(1\).

Example: Local Structure of the Sphere

Choose any point

\[ p\in S^2. \]

A sufficiently small neighborhood around \(p\) is homeomorphic to an open disk in \(\mathbb{R}^2\).

Therefore \(S^2\) is locally Euclidean of dimension \(2\).

Practice 1: Neighborhood in \(\mathbb{R}\)

\[ [1,5] \]

a neighborhood of \(3\) in \(\mathbb{R}\)?

Practice 2: Endpoint Neighborhood

\[ [1,5] \]

a neighborhood of \(1\) in \(\mathbb{R}\)?

Practice 3: Open Neighborhood

\[ (2,6) \]

an open neighborhood of \(4\)?

Practice 4: Interior Connection

Complete:

\[ x\in\operatorname{int}(A) \iff \boxed{\phantom{A\text{ is a neighborhood of }x}}. \]

Practice 5: Closure Connection

Complete:

\[ x\in\overline{A} \iff \boxed{\phantom{\text{every neighborhood of }x\text{ meets }A}}. \]

Practice 6: Boundary Connection

What must every neighborhood of a boundary point \(x\in\partial A\) meet?

Practice 7: Local Compactness

\[ \mathbb{R} \]

locally compact?

Practice 8: Local vs. Global

The sphere and torus are both locally homeomorphic to what familiar space?

Practice 9: Local Path Connectedness

Why is

\[ \mathbb{R}^n \]

locally path connected?

Practice 10: Subspace Neighborhood

In the subspace

\[ Y=[0,1], \]

\[ [0,\tfrac12) \]

a neighborhood of \(0\)?

Practice Answers

Practice 1

Yes. For example,

\[ (2,4)\subseteq[1,5] \]

contains \(3\). Therefore,

\[ \boxed{[1,5]\text{ is a neighborhood of }3}. \]

Practice 2

No. Every open interval around \(1\) contains points less than \(1\), so no open neighborhood of \(1\) lies inside \([1,5]\).

Practice 3

Yes. The set is open and contains \(4\):

\[ \boxed{(2,6)\text{ is an open neighborhood of }4}. \]

Practice 4

\[ \boxed{A\text{ is a neighborhood of }x}. \]

Practice 5

\[ \boxed{\text{every neighborhood of }x\text{ intersects }A}. \]

Practice 6

Every neighborhood must meet both

\[ A \]

and

\[ A^c. \]

Practice 7

Yes. Every point has a neighborhood such as an open interval whose closure is a closed bounded interval and therefore compact.

Practice 8

Both are locally homeomorphic to

\[ \boxed{\mathbb{R}^2}. \]

Practice 9

Every point has arbitrarily small open balls, and open balls are convex and therefore path connected.

Practice 10

Yes. It is open in the subspace topology because

\[ [0,\tfrac12) = [0,1]\cap(-1,\tfrac12). \]

Therefore it is an open neighborhood of \(0\) in \(Y\).

What to Remember

A neighborhood of \(x\) is any set containing an open set around \(x\).
A neighborhood need not itself be open.
In metric spaces, open balls form neighborhood bases.
A point lies in \(\operatorname{int}(A)\) exactly when \(A\) is a neighborhood of that point.
Closure, boundary, exterior, and limit points can all be defined through neighborhoods.
Local structure describes what sufficiently small neighborhoods look like.
Two spaces may be locally alike but globally different.
Local connectedness, local path connectedness, and local compactness are distinct local properties.
Manifolds are spaces that are locally homeomorphic to Euclidean space.
Neighborhood language generalizes metric ideas such as \(\varepsilon\)-\(\delta\) continuity.

The guiding principle is:

\[ \boxed{ \text{global topology is built from local neighborhoods} }. \]

Where We Go Next

Neighborhoods provide the local language that connects nearly every major topic in topology. They define interior and closure, express continuity, describe convergence, and reveal whether a space locally resembles a line, plane, or higher-dimensional Euclidean space.

From here, the natural next steps include local connectedness, local compactness, manifolds, quotient spaces, covering spaces, homotopy, and continuity on general topological spaces. These subjects show how local information can be assembled into global structure.