Part V — Chapter 17

Riemann & Fourier

Decomposing functions, reshaping geometry, and the deepest mystery of prime numbers

17.1 Joseph Fourier (1768 – 1830)

Jean-Baptiste Joseph Fourier was born on March 21, 1768 in Auxerre, France. Orphaned by age nine, he was educated by the Benedictines and later became a teacher and administrator. He accompanied Napoleon on the Egyptian expedition of 1798 and served as governor of Lower Egypt before returning to France. Despite his active political and administrative life, Fourier made one of the most consequential contributions in the history of mathematics.

Fourier's Life — Key Dates

1768 — Born in Auxerre, France
1789 — Presents paper on algebraic equations to the Académie Royale
1795 — Appointed professor at the École Polytechnique
1798–1801 — Egyptian expedition with Napoleon
1802 — Appointed Prefect of Isère by Napoleon
1807 — Presents memoir on heat conduction to the French Academy
1822 — Publishes Théorie analytique de la chaleur
1826 — Elected to the Académie Française
1830 — Dies in Paris

Fourier's interest in heat was not purely academic. During his years in Egypt, he had become fascinated by the desert climate and the physics of heat transfer. He reportedly kept his rooms uncomfortably warm for the rest of his life, a habit that some biographers have speculated may have contributed to his early death. His obsession with heat, however, led him to one of the most powerful ideas in all of mathematics.

Fourier's masterwork, the Théorie analytique de la chaleur (Analytical Theory of Heat), grew from his study of heat conduction. He modeled the flow of heat through a solid body with the heat equation.

17.2 The Heat Equation and Its Solution

Fourier derived the partial differential equation governing the conduction of heat in a homogeneous solid. In one spatial dimension, the equation takes the form:

$$\frac{\partial u}{\partial t} = k\,\frac{\partial^2 u}{\partial x^2}$$

where $u(x, t)$ is the temperature at position $x$ and time $t$, and$k$ is the thermal diffusivity of the material. In three dimensions, the equation generalises to:

$$\frac{\partial u}{\partial t} = k\,\nabla^2 u = k\left(\frac{\partial^2 u}{\partial x^2} + \frac{\partial^2 u}{\partial y^2} + \frac{\partial^2 u}{\partial z^2}\right)$$

To solve this equation, Fourier used the method of separation of variables. He assumed a solution of the form $u(x,t) = X(x)\,T(t)$. Substituting into the heat equation and dividing both sides by $X(x)\,T(t)$:

$$\frac{1}{T}\frac{dT}{dt} = k\,\frac{1}{X}\frac{d^2X}{dx^2} = -\lambda$$

Since the left side depends only on $t$ and the right side only on $x$, both must equal a constant $-\lambda$. This yields two ordinary differential equations:

$$T'(t) = -\lambda k\,T(t), \qquad X''(x) = -\lambda\,X(x)$$

Solution for a Rod with Fixed Endpoints

Consider a rod of length $L$ with endpoints held at zero temperature:$u(0,t) = u(L,t) = 0$. The boundary conditions force$\lambda_n = (n\pi/L)^2$ for $n = 1, 2, 3, \ldots$, giving eigenfunctions $X_n(x) = \sin(n\pi x / L)$. The general solution is:

$$u(x,t) = \sum_{n=1}^{\infty} b_n \sin\!\left(\frac{n\pi x}{L}\right) e^{-k(n\pi/L)^2 t}$$

The coefficients $b_n$ are determined by the initial temperature distribution$u(x,0) = f(x)$ via the Fourier sine series:

$$b_n = \frac{2}{L}\int_0^L f(x)\sin\!\left(\frac{n\pi x}{L}\right)dx$$

The exponential decay factor ensures that higher harmonics die out faster, so the temperature distribution smooths out over time — a physical fact encoded beautifully in the mathematics.

17.3 Fourier Series

Fourier's bold claim was that any periodic function can be represented as an infinite sum of sines and cosines — what we now call a Fourier series.

Fourier Series Representation

A periodic function $f(x)$ with period $2\pi$ can be expressed as:

$$f(x) = \frac{a_0}{2} + \sum_{n=1}^{\infty}\left(a_n \cos(nx) + b_n \sin(nx)\right)$$

where the Fourier coefficients are:

$$a_0 = \frac{1}{\pi}\int_{-\pi}^{\pi} f(x)\,dx$$

$$a_n = \frac{1}{\pi}\int_{-\pi}^{\pi} f(x)\cos(nx)\,dx \qquad (n \geq 1)$$

$$b_n = \frac{1}{\pi}\int_{-\pi}^{\pi} f(x)\sin(nx)\,dx \qquad (n \geq 1)$$

Derivation of the Fourier Coefficients

The key idea behind the derivation is orthogonality. The trigonometric functions satisfy the following orthogonality relations over $[-\pi, \pi]$:

$$\int_{-\pi}^{\pi} \cos(mx)\cos(nx)\,dx = \begin{cases} 0 & m \neq n \\ \pi & m = n \neq 0 \\ 2\pi & m = n = 0 \end{cases}$$

$$\int_{-\pi}^{\pi} \sin(mx)\sin(nx)\,dx = \begin{cases} 0 & m \neq n \\ \pi & m = n \end{cases}$$

$$\int_{-\pi}^{\pi} \cos(mx)\sin(nx)\,dx = 0 \quad \text{for all } m, n$$

To find $a_m$, multiply both sides of the Fourier series by $\cos(mx)$ and integrate over $[-\pi, \pi]$. By orthogonality, every term on the right vanishes except the one with $n = m$, yielding $\pi \cdot a_m$. The formula for $b_m$ is obtained similarly by multiplying by $\sin(mx)$.

Convergence and the Dirichlet Conditions

Fourier's claim provoked fierce debate. Lagrange objected that a sum of smooth sines and cosines could never reproduce a discontinuous function. This controversy forced mathematicians to confront fundamental questions: What exactly is a function? When does an infinite series converge?

Dirichlet Conditions for Convergence

The Fourier series of $f(x)$ converges to $f(x)$ at every point where$f$ is continuous, provided:

$f$ is periodic with period $2\pi$
$f$ has at most a finite number of maxima and minima in each period
$f$ has at most a finite number of discontinuities in each period, all of which are finite jumps
$\int_{-\pi}^{\pi} |f(x)|\,dx$ converges (i.e., $f$ is absolutely integrable)

At a jump discontinuity $x_0$, the Fourier series converges to the average of the left and right limits: $\frac{1}{2}[f(x_0^-) + f(x_0^+)]$.

Dirichlet proved these conditions in 1829. The question of which functions have pointwise convergent Fourier series turned out to be extraordinarily deep. Lennart Carleson proved in 1966 that the Fourier series of any $L^2$ function converges almost everywhere — a result so difficult that it earned him the Abel Prize in 2006.

Fourier Series of a Square Wave

Consider the square wave: $f(x) = 1$ for $0 < x < \pi$ and$f(x) = -1$ for $-\pi < x < 0$. Since $f$ is an odd function,$a_n = 0$ for all $n$. Computing $b_n$:

$$b_n = \frac{1}{\pi}\int_{-\pi}^{\pi} f(x)\sin(nx)\,dx = \frac{2}{\pi}\int_0^{\pi}\sin(nx)\,dx = \frac{2}{n\pi}\Big[-\cos(nx)\Big]_0^{\pi}$$

$$= \frac{2}{n\pi}(1 - \cos(n\pi)) = \frac{2}{n\pi}(1 - (-1)^n)$$

This gives $b_n = \frac{4}{n\pi}$ when $n$ is odd, and $b_n = 0$ when $n$ is even:

$$f(x) = \frac{4}{\pi}\left(\sin x + \frac{\sin 3x}{3} + \frac{\sin 5x}{5} + \frac{\sin 7x}{7} + \cdots\right)$$

Setting $x = \pi/2$ yields the beautiful Leibniz series: $\frac{\pi}{4} = 1 - \frac{1}{3} + \frac{1}{5} - \frac{1}{7} + \cdots$

Fourier Series of a Sawtooth Wave

The sawtooth wave $f(x) = x$ for $-\pi < x < \pi$ (extended periodically) is another classic example. Since $f$ is odd, again $a_n = 0$. For $b_n$, integration by parts gives:

$$b_n = \frac{1}{\pi}\int_{-\pi}^{\pi} x\sin(nx)\,dx = \frac{2}{\pi}\int_0^{\pi} x\sin(nx)\,dx = \frac{2(-1)^{n+1}}{n}$$

Therefore:

$$f(x) = 2\left(\sin x - \frac{\sin 2x}{2} + \frac{\sin 3x}{3} - \frac{\sin 4x}{4} + \cdots\right)$$

At the discontinuity $x = \pi$, the partial sums overshoot by about 9% — this is the famous Gibbs phenomenon, discovered by Wilbraham (1848) and rediscovered by Gibbs (1899). The overshoot does not diminish as more terms are added; it merely becomes narrower.

Parseval's Theorem

The total “energy” of a function equals the sum of the energies of its Fourier components:

$$\frac{1}{\pi}\int_{-\pi}^{\pi} |f(x)|^2\,dx = \frac{a_0^2}{2} + \sum_{n=1}^{\infty}(a_n^2 + b_n^2)$$

This is the infinite-dimensional analogue of the Pythagorean theorem. Applying it to the sawtooth wave yields $\frac{\pi^2}{6} = 1 + \frac{1}{4} + \frac{1}{9} + \frac{1}{16} + \cdots$, recovering Euler's famous Basel sum.

17.4 Bernhard Riemann (1826 – 1866)

Georg Friedrich Bernhard Riemann was born on September 17, 1826 in Breselenz, a village in the Kingdom of Hanover. The son of a Lutheran pastor, Riemann was shy and sickly throughout his life, but possessed a mathematical vision of extraordinary depth and originality. He studied at Göttingen under Gauss and at Berlin under Dirichlet and Jacobi.

Riemann's Short but Monumental Life

1826 — Born in Breselenz, Hanover
1846 — Enters Göttingen; studies under Gauss
1847–1849 — Studies at Berlin under Dirichlet, Jacobi, and Eisenstein
1851 — Doctoral thesis on complex analysis (Riemann surfaces)
1854 — Habilitation lecture: “On the Hypotheses which Lie at the Foundations of Geometry”
1857 — Paper on abelian functions
1859 — Eight-page paper on the zeta function and prime distribution
1862 — Marries Elise Koch; contracts tuberculosis
1866 — Dies of tuberculosis in Selasca, Italy, at age 39

Despite publishing fewer than twenty papers in his short life, Riemann transformed virtually every area of mathematics he touched. His doctoral thesis introduced Riemann surfaces, which gave complex analysis a geometric foundation. His habilitation lecture created Riemannian geometry. His eight-page 1859 paper on the distribution of primes opened up one of the deepest mysteries in mathematics. Each of these works alone would have secured his place among the greatest mathematicians.

Riemann's approach to mathematics was profoundly geometric and intuitive, in contrast to the algebraic and computational style of many of his contemporaries. He preferred to understand the deep structural reasons behind mathematical truths rather than verify them through lengthy calculations. This visionary style meant that many of his ideas were far ahead of his time — some took decades to be made fully rigorous by later mathematicians.

17.5 The Riemann Integral

The need to rigorously define integration arose directly from the debates about Fourier series. When can we integrate a function term by term? Riemann provided the first rigorous definition of the integral for a broad class of functions in his 1854 habilitation thesis.

The Riemann Integral

Let $f$ be a bounded function on $[a, b]$. A partition$P$ of $[a, b]$ is a finite set $a = x_0 < x_1 < \cdots < x_n = b$. Define $\Delta x_i = x_i - x_{i-1}$ and the mesh $\|P\| = \max_i \Delta x_i$.

The upper and lower Riemann sums are:

$$U(f, P) = \sum_{i=1}^{n} M_i \Delta x_i, \qquad L(f, P) = \sum_{i=1}^{n} m_i \Delta x_i$$

where $M_i = \sup_{x \in [x_{i-1}, x_i]} f(x)$ and $m_i = \inf_{x \in [x_{i-1}, x_i]} f(x)$.

The function $f$ is Riemann integrable if $\inf_P U(f,P) = \sup_P L(f,P)$, and the common value is the integral:

$$\int_a^b f(x)\,dx = \lim_{\|P\| \to 0} \sum_{i=1}^{n} f(c_i)\Delta x_i$$

for any choice of sample points $c_i \in [x_{i-1}, x_i]$.

Properties of the Riemann Integral

If $f$ and $g$ are Riemann integrable on $[a,b]$, then:

Linearity: $\int_a^b [\alpha f + \beta g]\,dx = \alpha\int_a^b f\,dx + \beta\int_a^b g\,dx$
Additivity: $\int_a^b f\,dx = \int_a^c f\,dx + \int_c^b f\,dx$ for $a < c < b$
Monotonicity: If $f(x) \leq g(x)$ on $[a,b]$, then $\int_a^b f\,dx \leq \int_a^b g\,dx$
Product: $fg$ is also Riemann integrable
Absolute value: $|f|$ is Riemann integrable and $\left|\int_a^b f\,dx\right| \leq \int_a^b |f|\,dx$

Riemann showed that a bounded function on $[a, b]$ is Riemann integrable if and only if its set of discontinuities has “measure zero” (the precise formulation came later, from Lebesgue in 1902). Continuous functions and monotone functions are always Riemann integrable.

However, the Riemann integral has limitations. The characteristic function of the rationals (Dirichlet's function, which equals 1 on the rationals and 0 on the irrationals) is not Riemann integrable, and pointwise limits of Riemann-integrable functions need not be Riemann integrable. These deficiencies motivated Henri Lebesgue to develop his more powerful integral (1901), which partitions the range of a function rather than its domain, and can integrate a much broader class of functions.

17.6 Riemann Surfaces and Complex Analysis

Riemann's 1851 doctoral thesis, supervised by Gauss, introduced the concept of a Riemann surface. The problem he addressed was fundamental: multi-valued functions like $\sqrt{z}$ or $\log z$ do not fit naturally into the framework of single-valued complex analysis. Riemann's solution was to construct a new surface on which these functions become single-valued.

Riemann Surface

A Riemann surface is a one-dimensional complex manifold — a topological surface equipped with a complex-analytic structure. Locally, it looks like an open subset of the complex plane, but globally it may have a much more interesting topology (handles, branch points, etc.).

The Riemann Surface of the Square Root

The function $w = \sqrt{z}$ is double-valued: for each $z \neq 0$, there are two values of $w$. Riemann's construction takes two copies of the complex plane, cuts each along the negative real axis, and glues them together crosswise — the top edge of the cut in one sheet is joined to the bottom edge in the other. The result is a single connected surface on which $\sqrt{z}$ is a well-defined, single-valued, analytic function. The point $z = 0$ is a branch point where the two sheets meet.

More generally, an algebraic equation $P(z, w) = 0$ defines a Riemann surface. The genus of this surface — roughly, the number of “holes” or “handles” — is a fundamental topological invariant that determines many properties of the algebraic curve. Riemann proved his celebrated Riemann–Roch theorem, which relates the number of meromorphic functions on a surface to its genus:

$$\ell(D) - \ell(K - D) = \deg(D) - g + 1$$

where $D$ is a divisor, $K$ is the canonical divisor, $g$ is the genus, and $\ell(D)$ is the dimension of the space of meromorphic functions associated to$D$. This theorem became one of the cornerstones of algebraic geometry.

17.7 Riemannian Geometry

On June 10, 1854, Riemann delivered his habilitation lecture “On the Hypotheses which Lie at the Foundations of Geometry” (“Über die Hypothesen, welche der Geometrie zu Grunde liegen”) to the faculty at Göttingen. Gauss, then 77 years old, was in the audience and was deeply impressed. This single lecture created the field of Riemannian geometryand laid the mathematical foundation for Einstein's general theory of relativity sixty years later.

The Metric Tensor

Riemann proposed that the geometry of a space is determined by a metric tensor$g_{ij}$, which specifies how to measure infinitesimal distances:

$$ds^2 = \sum_{i,j} g_{ij}\,dx^i\,dx^j$$

For example, the metric of Euclidean space in Cartesian coordinates is$ds^2 = dx^2 + dy^2 + dz^2$ (where $g_{ij} = \delta_{ij}$), while the metric on a sphere of radius $R$ is$ds^2 = R^2(d\theta^2 + \sin^2\!\theta\,d\varphi^2)$.

The Riemann Curvature Tensor

From the metric, one constructs the Christoffel symbols$\Gamma^k_{ij}$, which describe how coordinate basis vectors change from point to point:

$$\Gamma^k_{ij} = \frac{1}{2}g^{kl}\left(\frac{\partial g_{il}}{\partial x^j} + \frac{\partial g_{jl}}{\partial x^i} - \frac{\partial g_{ij}}{\partial x^l}\right)$$

The Riemann curvature tensor $R^i{}_{jkl}$ measures how parallel transport around a small loop fails to return a vector to its original direction:

$$R^i{}_{jkl} = \frac{\partial \Gamma^i_{jl}}{\partial x^k} - \frac{\partial \Gamma^i_{jk}}{\partial x^l} + \Gamma^i_{mk}\Gamma^m_{jl} - \Gamma^i_{ml}\Gamma^m_{jk}$$

Contracting the curvature tensor gives the Ricci tensor $R_{jl} = R^i{}_{jil}$and the scalar curvature $R = g^{jl}R_{jl}$.

Connection to Non-Euclidean Geometry

Riemann's framework elegantly subsumes the non-Euclidean geometries discovered by Lobachevsky and Bolyai. In Riemann's language, these are simply manifolds with constant curvature: Euclidean geometry has $K = 0$, the geometry of the sphere (elliptic geometry) has $K > 0$, and hyperbolic geometry has $K < 0$. But Riemann went far beyond this trichotomy — his framework allows curvature to vary from point to point and works in any number of dimensions.

The Hyperbolic Plane

The Poincaré upper half-plane model of hyperbolic geometry uses the metric:

$$ds^2 = \frac{dx^2 + dy^2}{y^2}, \qquad y > 0$$

This metric has constant Gaussian curvature $K = -1$. Geodesics (the shortest paths) are vertical lines and semicircles centred on the $x$-axis. The angle sum of a triangle is always less than $\pi$, with the deficit proportional to the area.

The Gauss–Bonnet Theorem

For a compact 2-dimensional Riemannian manifold $M$ with Gaussian curvature $K$and Euler characteristic $\chi(M)$:

$$\int_M K\,dA = 2\pi\,\chi(M)$$

This remarkable formula connects the local geometry (curvature at each point) to the global topology (the Euler characteristic). For a sphere, $\chi = 2$, giving$\int K\,dA = 4\pi$. For a torus, $\chi = 0$, so the total curvature is zero — the positive curvature on the outside exactly cancels the negative curvature on the inside.

Riemann's framework was revolutionary because it allowed geometry in any number of dimensions and did not require an ambient space — the geometry is defined entirely intrinsically by the metric tensor. Einstein used this framework in his 1915 general theory of relativity, where the metric tensor describes the geometry of spacetime and the Einstein field equations relate the curvature to the distribution of matter and energy:

$$R_{\mu\nu} - \frac{1}{2}Rg_{\mu\nu} + \Lambda g_{\mu\nu} = \frac{8\pi G}{c^4}T_{\mu\nu}$$

Here $R_{\mu\nu}$ is the Ricci curvature tensor, $R$ is the scalar curvature,$g_{\mu\nu}$ is the metric tensor, $\Lambda$ is the cosmological constant,$G$ is Newton's gravitational constant, and $T_{\mu\nu}$ is the stress-energy tensor describing the distribution of matter and energy. The left side encodes the curvature of spacetime; the right side encodes what produces that curvature. As John Wheeler memorably summarised: “Spacetime tells matter how to move; matter tells spacetime how to curve.”

17.8 The Riemann Zeta Function

In 1859, Riemann published his only paper on number theory: “Über die Anzahl der Primzahlen unter einer gegebenen Grösse” (“On the Number of Primes Less Than a Given Magnitude”). In just eight pages, he laid out a program for understanding the distribution of prime numbers that continues to drive research today.

The Riemann Zeta Function

For $\text{Re}(s) > 1$, the zeta function is defined by the absolutely convergent series:

$$\zeta(s) = \sum_{n=1}^{\infty} \frac{1}{n^s} = \frac{1}{1^s} + \frac{1}{2^s} + \frac{1}{3^s} + \frac{1}{4^s} + \cdots$$

For real values, this includes the harmonic series $\zeta(1) = \infty$ and Euler's celebrated values: $\zeta(2) = \frac{\pi^2}{6}$, $\zeta(4) = \frac{\pi^4}{90}$, and more generally $\zeta(2k) = \frac{(-1)^{k+1}(2\pi)^{2k}B_{2k}}{2(2k)!}$ where$B_{2k}$ are the Bernoulli numbers.

Euler Product Formula

The connection between the zeta function and prime numbers is given by Euler's product formula:

$$\zeta(s) = \prod_{p \text{ prime}} \frac{1}{1 - p^{-s}} = \frac{1}{1-2^{-s}} \cdot \frac{1}{1-3^{-s}} \cdot \frac{1}{1-5^{-s}} \cdot \frac{1}{1-7^{-s}} \cdots$$

Proof: Each factor $\frac{1}{1-p^{-s}} = 1 + p^{-s} + p^{-2s} + \cdots$ is a geometric series. Multiplying all these together and using the fundamental theorem of arithmetic (unique factorization):

$$\prod_p \sum_{k=0}^{\infty} p^{-ks} = \sum_{n=1}^{\infty} n^{-s}$$

since every positive integer has a unique prime factorization. This identity encodes the fundamental theorem of arithmetic in analytic form. $\blacksquare$

Analytic Continuation and the Functional Equation

Riemann's great contribution was to extend $\zeta(s)$ to the entire complex plane via analytic continuation. The first step uses the alternating series (the Dirichlet eta function):

$$\eta(s) = \sum_{n=1}^{\infty}\frac{(-1)^{n-1}}{n^s} = (1 - 2^{1-s})\,\zeta(s)$$

The eta function converges for $\text{Re}(s) > 0$, extending $\zeta$ to the right half-plane (with a pole at $s = 1$). To extend to the full complex plane, Riemann used the functional equation:

$$\zeta(s) = 2^s \pi^{s-1} \sin\!\left(\frac{\pi s}{2}\right) \Gamma(1-s)\,\zeta(1-s)$$

An equivalent and more symmetric form uses the xi function$\xi(s) = \frac{1}{2}s(s-1)\pi^{-s/2}\Gamma(s/2)\zeta(s)$, which satisfies$\xi(s) = \xi(1-s)$ — a perfect symmetry about the line $\text{Re}(s) = 1/2$.

From the functional equation, we see that $\zeta(-2n) = 0$ for all positive integers$n$ (the trivial zeros, arising from the zeros of$\sin(\pi s/2)$). All other zeros lie in the critical strip$0 < \text{Re}(s) < 1$.

17.9 The Riemann Hypothesis

Riemann made the following conjecture, which is now one of the most famous unsolved problems in all of mathematics:

The Riemann Hypothesis (1859)

All non-trivial zeros of $\zeta(s)$ have real part equal to $\frac{1}{2}$. That is, if $\zeta(\rho) = 0$ and $\rho$ is not a negative even integer, then$\rho = \frac{1}{2} + it$ for some real number $t$.

The connection to prime numbers is profound. The prime counting function $\pi(x)$ (the number of primes less than or equal to $x$) is related to the zeta function through Riemann's explicit formula, which expresses$\pi(x)$ as a sum over the zeros of $\zeta$. The best known approximation is the logarithmic integral:

$$\pi(x) \approx \text{Li}(x) = \int_2^x \frac{dt}{\ln t}$$

The Riemann Hypothesis, if true, would imply the strongest possible error bound for this approximation:

$$|\pi(x) - \text{Li}(x)| = O(\sqrt{x}\,\ln x)$$

What the Zeros Tell Us About Primes

Riemann's explicit formula relates $\pi(x)$ to the zeros $\rho$ of $\zeta$:

$$\pi(x) = \text{Li}(x) - \sum_{\rho} \text{Li}(x^{\rho}) - \ln 2 + \int_x^{\infty}\frac{dt}{t(t^2-1)\ln t}$$

Each zero $\rho = \beta + i\gamma$ contributes an oscillatory correction to $\pi(x)$. If all $\beta = 1/2$ (the Riemann Hypothesis), these corrections are as small as possible, meaning primes are distributed as “regularly” as they can be.

Why the Riemann Hypothesis Matters

The Riemann Hypothesis is not merely a curiosity about one particular function. Its truth or falsehood has far-reaching consequences across mathematics:

It is one of the seven Millennium Prize Problems (Clay Mathematics Institute, 2000), carrying a $1,000,000 prize
Over 10 trillion non-trivial zeros have been computed, and all lie on the critical line — but this does not constitute a proof
Hundreds of theorems in number theory, analysis, and combinatorics have been proved assuming the Riemann Hypothesis
It implies sharp bounds on the gaps between consecutive primes: if $p_n$ is the $n$-th prime, then $p_{n+1} - p_n = O(\sqrt{p_n}\,\ln p_n)$
It is deeply connected to random matrix theory — the statistical distribution of the zeros matches the eigenvalues of random Hermitian matrices (the Montgomery–Odlyzko law)
It would establish the best possible error term in the prime number theorem and tighten results in analytic number theory
It has implications for the distribution of prime numbers in arithmetic progressions via the Generalised Riemann Hypothesis
The hypothesis remains neither proved nor disproved after more than 165 years

David Hilbert, when asked what he would do if he were to awaken after sleeping for five hundred years, reportedly answered that his first question would be whether the Riemann Hypothesis had been proved. The problem continues to resist all attacks, and many mathematicians consider it the single most important open problem in mathematics.

17.10 Karl Weierstrass — Rigorous Foundations

Karl Weierstrass (1815–1897), often called the “father of modern analysis,” spent years as a provincial schoolteacher before his mathematical genius was recognized. He perfected the $\varepsilon$-$\delta$ method that Cauchy had initiated and set the standard of rigour that defines modern analysis.

Weierstrass is perhaps most famous for his construction of a function that is continuous everywhere but differentiable nowhere, which shattered the widespread belief that continuous functions must be differentiable except perhaps at isolated points.

The Weierstrass Function (1872)

The function defined by:

$$f(x) = \sum_{n=0}^{\infty} a^n \cos(b^n \pi x)$$

where $0 < a < 1$, $b$ is a positive odd integer, and $ab > 1 + \frac{3\pi}{2}$, is continuous everywhere on $\mathbb{R}$ but differentiable at no point.

This result was deeply disturbing to many mathematicians. Hermite reportedly wrote: “I turn with terror and horror from this lamentable plague of functions which have no derivatives.” Yet Weierstrass's example showed that mathematical intuition, however powerful, cannot substitute for rigorous proof.

The Bolzano–Weierstrass Theorem

Every bounded sequence of real numbers has a convergent subsequence. Equivalently, every bounded infinite set in $\mathbb{R}^n$ has a limit point. This theorem is one of the pillars of real analysis and is equivalent (in appropriate settings) to the completeness of the real numbers.

Among Weierstrass's other contributions: the Weierstrass approximation theorem (every continuous function on a closed interval can be uniformly approximated by polynomials), the Weierstrass M-test for uniform convergence of series, the Weierstrass product theorem for entire functions, and a rigorous construction of the real numbers. He also made fundamental contributions to the theory of elliptic functions, introducing the Weierstrass $\wp$-function:

$$\wp(z) = \frac{1}{z^2} + \sum_{\omega \in \Lambda \setminus \{0\}} \left(\frac{1}{(z-\omega)^2} - \frac{1}{\omega^2}\right)$$

which parameterizes elliptic curves and satisfies the differential equation$(\wp')^2 = 4\wp^3 - g_2\wp - g_3$.

17.11 Legacy — Analysis, Geometry, and Number Theory Unified

The combined legacy of Fourier, Riemann, and Weierstrass represents a transformation in the scope and depth of mathematics. Fourier's series showed that almost any function can be decomposed into simple harmonic components — an idea that underpins modern signal processing, quantum mechanics, medical imaging (MRI and CT scans), data compression (JPEG, MP3), and telecommunications.

Riemann revealed that geometry, analysis, and number theory are not separate subjects but deeply interconnected. His zeta function uses analysis (complex function theory) to study number theory (the distribution of primes). His geometry uses analysis (calculus on manifolds) to generalise geometry to arbitrary dimensions. His surfaces use topology to illuminate complex analysis. His integral definition clarified what Fourier's series actually mean.

Weierstrass provided the rigorous foundations that made all of this mathematically secure. His insistence on $\varepsilon$-$\delta$ proofs, his pathological counterexamples, and his demand for precision transformed mathematics from a subject that relied on geometric intuition into one that rested on logical proof. This standard of rigour is now the hallmark of modern mathematics.

Together, they brought mathematics to the threshold of the modern era. The questions they raised — about convergence, about the nature of functions, about the foundations of geometry, about the distribution of primes — continue to drive mathematical research in the 21st century. The Riemann Hypothesis, in particular, stands as perhaps the greatest open challenge in all of mathematics, a testament to the depth and difficulty of the questions that Riemann posed in a single eight-page paper more than a century and a half ago.

← Chapter 16: Galois & Abel Chapter 18: Cantor & Dedekind →