Chapter 3 · Introduction

Chapter 3 — Continuity and Differentiability

What it means for a function to have no breaks, why closed bounded intervals are special, how far apart continuity and uniform continuity are, and what derivatives and the mean value theorem add.

For a century after Newton and Leibniz, a "continuous function" meant one given by a single formula, and everybody "knew" that such a function takes every value between any two of its values, has a largest value on a closed interval, and has a tangent almost everywhere. Then the examples arrived. Dirichlet described a function equal to $1$ on the rationals and $0$ on the irrationals — continuous nowhere. Thomae found one that is continuous at every irrational point and at no rational point. Weierstrass wrote down a continuous function with no tangent at any point. Intuition about "unbroken curves" was not going to be enough.

The cure was the $\varepsilon$–$\delta$ language and the completeness of $\mathbb{R}$. With them, Bolzano's intermediate value theorem and Weierstrass's extreme value theorem become theorems about compact sets: closed bounded intervals behave like finite sets in exactly the right way. Compactness also upgrades continuity to uniform continuity, the form needed for integration and approximation. On the way we meet the discontinuities that monotone functions can have (only jumps, and only countably many), and finally the derivative, the mean value theorem — the most used theorem of analysis — and Taylor's theorem. In CSIR NET this chapter is a steady source of both Part B and Part C questions.

ε–δ limits→Continuity→Heine–Borel→EVT & IVT→Uniform continuity→Monotone functions & jumps→Derivatives→Mean value theorems→Taylor
🎯 By the end of this chapter you can…
  • prove and disprove limits with $\varepsilon$–$\delta$ arguments and with sequences, and find all points of continuity of functions defined separately on $\mathbb{Q}$ and $\mathbb{R}\setminus\mathbb{Q}$;
  • state and prove the Heine–Borel theorem in $\mathbb{R}$ and decide quickly whether a given set is compact;
  • use the extreme value and intermediate value theorems to locate roots and fixed points and to decide what continuous maps can do to intervals;
  • decide whether a function is uniformly continuous on a given domain, using Heine–Cantor, Lipschitz and Hölder bounds, continuous extension and sequences;
  • classify discontinuities, and prove that monotone functions have only countably many (jump) discontinuities;
  • apply Rolle's, Lagrange's and Cauchy's mean value theorems, L'Hôpital's rule, Darboux's theorem and Taylor's theorem with remainder.
Teaser — A Function Continuous Exactly at the Irrationals

Thomae's function is $T(p/q)=1/q$ (lowest terms) and $T(x)=0$ for irrational $x$; the dots show it for $q\le60$. Pick a point $x_0$ and a tolerance $\varepsilon$ (the amber band around $T(x_0)$). Only finitely many dots — the red ones — break the band, so at an irrational $x_0$ a green window that avoids them always exists. At a rational $x_0$ the irrationals nearby (value $0$) break the band as soon as $\varepsilon\le T(x_0)$. Press Play to shrink $\varepsilon$.

ε—
T(x₀)—
dots breaking the band—
largest δ—

How this is examined

CSIR NET draws heavily on this chapter. Part B typically asks for the number of points of continuity of a function defined by different formulas on the rationals and irrationals, for the one compact set among four, for the one uniformly continuous function on a given interval, or for a value guaranteed by the intermediate or mean value theorem. Part C, where several options may be correct, prefers statements to be tested against counterexamples: continuous images of open, closed, bounded or compact sets; whether bounded continuous functions are uniformly continuous; what the set of discontinuities of a monotone function can be; whether a differentiable function has a continuous derivative; whether derivatives have the intermediate value property; and estimates from Taylor's theorem. Each section collects the standard counterexamples in its ⚠️ box and drills the question types in its 🎯 Exam corner. GATE MA asks similar questions, often about uniform continuity and the mean value theorem.

Sections in this chapter

✅ Before you start

You will use the supremum and its $\varepsilon$-characterisation, the Archimedean property and the density of $\mathbb{Q}$ and of the irrationals (§1.2), countable sets (§1.1), convergent and Cauchy sequences (§2.1) and the Bolzano–Weierstrass theorem (§2.2). The single most important skill is reading and negating statements with several quantifiers: "for every $\varepsilon\gt0$ there is $\delta\gt0$ such that for all $x$…" — and noticing that uniform continuity is the same sentence with the quantifiers in a different order.