Chapter 9 — Topology
Continuity without distance: open sets and bases, closure and density, continuous maps and homeomorphisms, new spaces from old by subspaces, products and gluing, and the separation, connectedness and compactness properties that tell spaces apart.
In 1736 Leonhard Euler showed that nobody could walk through the city of Königsberg crossing each of its seven bridges exactly once, and he remarked that the answer depended on no length, angle or area — only on which land masses the bridges joined. A century later Johann Listing in Göttingen named the study of such properties Topologie, and in 1858 he and August Möbius independently described the one-sided strip that now carries Möbius's name. Bernhard Riemann's surfaces and Georg Cantor's point sets made it clear that analysis needs a precise notion of "nearness" more general than distance, and Henri Poincaré's Analysis Situs (1895) turned the qualitative study of shape into a subject of its own. The definition of a topological space came last: Maurice Fréchet introduced metric spaces in 1906, Felix Hausdorff defined spaces by systems of neighbourhoods in 1914, Kazimierz Kuratowski axiomatised the closure operation in 1922, and by the 1940s the short list of axioms for open sets that we use today had become standard.
The pay-off of that abstraction is that one argument now works everywhere: on the real line, in function spaces, on spheres and tori, on finite sets of pixels, and on spaces that have no natural distance at all. Topology is also unexpectedly practical. The positions of a robot arm with two rotating joints form a torus $S^1\times S^1$, and motion-planning algorithms work on that space rather than in the room; topological data analysis detects loops and voids in clouds of data points, from protein structures to brain-activity recordings; and the 2016 Nobel Prize in Physics (Thouless, Haldane and Kosterlitz) was awarded for topological phases of matter, whose properties survive any small deformation precisely because they are topological invariants. This chapter develops point-set topology to the depth that CSIR NET requires: Section 9.1 builds topological spaces, bases, closure, density, countability and continuity; Section 9.2 constructs subspaces, products and quotient spaces; Section 9.3 compares spaces by how well their points and closed sets can be separated; and Section 9.4 studies connectedness and compactness, the two properties that every continuous image inherits.
- decide whether a family of sets is a topology, a basis or a subbasis, and compare topologies such as the discrete, indiscrete, cofinite, cocountable, standard and lower limit topologies on $\mathbb{R}$;
- compute interiors, closures, boundaries, limit points and dense subsets in each of these spaces, and know when sequences are enough to detect closure;
- decide whether a space is separable, first countable or second countable, with the Sorgenfrey line and the cofinite line as the standard test cases;
- prove or disprove that a map between two spaces is continuous, open, closed or a homeomorphism;
- work with subspace, product (finite and arbitrary, product versus box) and quotient topologies, and recognise circles, cylinders, Möbius strips, tori, Klein bottles and projective planes as quotients of a square;
- state and use the separation axioms $T_0$, $T_1$, $T_2$, regularity and normality, with the standard examples and non-examples;
- prove that connectedness and compactness survive continuous images and products, and use them (and the other invariants of the chapter) to show that two spaces are not homeomorphic.
How this is examined
Topology questions in CSIR NET are almost always about examples, so the best preparation is to know a small zoo of spaces very well. Part B typically asks for one thing: the closure or interior of a set in a stated topology (cofinite, cocountable, lower limit, or a topology given by a basis), the number of topologies or of open sets satisfying some condition on a finite set, which of four spaces is separable, second countable or Hausdorff, or which space a given quotient is homeomorphic to. Part C, where more than one option may be correct, presents a space or a map and asks which properties hold — "continuous / open / a homeomorphism", "connected / compact / Hausdorff / normal", "first countable but not second countable", "the product (box) topology makes this map continuous" — or asks which pairs of spaces are homeomorphic, to be settled by an invariant such as compactness, connectedness, the number of components after removing a point, or countability. Almost every false option is a standard counterexample: the Sorgenfrey line, the cofinite and cocountable topologies, the topologist's sine curve, the box topology, $[0,1)\to S^1$, or $\mathbb{R}/\mathbb{Q}$. Each section's ⚠️ box collects the ones you need. GATE MA asks similar questions, often in multiple-select form.
Sections in this chapter
Topological Spaces, Bases and Continuity
Topologies (discrete, indiscrete, cofinite, cocountable, standard, lower limit); bases and subbases; interior, closure, boundary, limit points and dense sets; separable, first and second countable spaces; continuous maps and homeomorphisms.
Subspace, Product and Quotient Topologies
The subspace topology and the pasting lemma; finite and arbitrary products, product versus box topology, projections; quotient maps, identification spaces, the circle, cylinder, Möbius strip, torus, Klein bottle and projective plane.
Separation Axioms
$T_0$, $T_1$, Hausdorff, regular and normal spaces; examples and non-examples including the cofinite topology and the Sorgenfrey line; metric spaces are normal; Urysohn's lemma; uniqueness of limits.
Connectedness and Compactness
Connected and path-connected spaces, components, products and continuous images; compact spaces, closed subsets, Tychonoff's theorem, compact Hausdorff spaces, local compactness and the one-point compactification; invariants that distinguish spaces.
This chapter generalises metric spaces, and it is much easier if you have met them first: open and closed sets, interior and closure, dense sets, continuity and homeomorphisms in a metric space (the sister course CSIR NET Unit 1 — Analysis and Linear Algebra, §7.1 Metric Spaces and Completeness), compactness in metric spaces and the Heine–Borel theorem (Unit 1, §7.2 Compactness in Metric Spaces) and connectedness (Unit 1, §7.3 Connectedness). Every metric example reappears here as a special case, and we use those sections' results freely, but each definition is restated in the general setting. From set theory you need images and preimages, equivalence relations and partitions, and countable versus uncountable sets (Unit 1, §1.1 Sets, Functions and Countable Sets). The construction of quotient spaces parallels that of quotient groups in §6.2: an equivalence relation, its classes, and a "universal property" that tells you when a map out of the quotient is well defined and continuous. Read the sections in order: §9.2 uses bases and continuity from §9.1, and §§9.3–9.4 use all three constructions of §9.2.