A Reader's Guide

Euler's Identity

Five fundamental constants in one line: how e, i and π meet on the unit circle, why half a turn lands on −1, and the waves, rotations and number theory that grow from it

In a poll David Wells ran for The Mathematical Intelligencer, readers voted e^(iπ) + 1 = 0 the most beautiful theorem in mathematics, and Richard Feynman called the formula behind it "our jewel". Its appeal is that it seems too good to be true. Five numbers from unrelated corners of mathematics, 0 and 1 from counting, e from growth, i from algebra and π from the circle, are bound together in a single line. The folio shows why it has to be so, and the reason turns out to be a picture: multiplying by i is a quarter turn, e^(iθ) walks around the unit circle, and half a turn lands exactly on −1. Compound interest, taken sideways, bends into that half circle before the reader's eyes. From there the identity spreads into waves, rotations and the roots of unity. Once understood, it stops looking like a coincidence and starts to look inevitable.

Euler's identity · Euler's formula · complex numbers · unit circle · roots of unity · complex logarithmOpen the folio →

The Hyperbolic Plane

Geometry where parallels multiply, triangles are thin and space grows exponentially: the models that make it visible, and the mathematics, physics and data science that live there

"Out of nothing I have created a strange new universe," János Bolyai wrote to his father in 1823, and this folio is a guided tour of it. Drop Euclid's parallel postulate and a perfectly consistent geometry appears in which infinitely many parallels pass through a point, triangles are thin, and space grows exponentially, so fast that in the folio's heptagonal tiling each ring holds about 2.618 times as many tiles as the last. It is the geometry of Escher's Circle Limit prints and of ruffled corals, and, less expectedly, of number theory, special relativity and anti-de Sitter space. Data scientists now use it because it has room for branching trees, holding their distances with little distortion where a flat page soon runs out of space. The folio builds the Poincaré disk from the hyperboloid, compares the models side by side, and lets a strange world become an intuitive one.

hyperbolic geometry · Poincaré disk · non-Euclidean geometry · parallel postulate · hyperbolic tiling · Klein quartic · modular group · anti-de Sitter spaceOpen the folio →

Euler's Polyhedron Formula

Vertices minus edges plus faces always makes two: why the count never fails, what bends it, and how one number grew into topology

Second only to Euler's identity in Wells's poll of the most beautiful theorems, V − E + F = 2 is a truth a child can check and a mathematician can spend a lifetime on. Count the corners of any convex solid, subtract its edges, add its faces, and the answer is always two. The folio shows why the count never fails and what that one number forces on the world: there can be exactly five Platonic solids, every traditional pentagon-and-hexagon football and every fullerene must carry exactly twelve pentagons, and an idealised wind blowing everywhere across the Earth must be calm somewhere. Put a hole through the solid and the answer changes to zero, which is the first hint that the number measures shape itself. What began as a tally of corners became the Euler characteristic, the first quantity known to survive any stretching, and the seed of modern topology.

Euler's polyhedron formula · Euler characteristic · Platonic solids · topology · Gauss–Bonnet theorem · planar graphs · hairy ball theoremOpen the folio →

Infinitely Many Primes

Euclid's four-line proof that the primes never end, and what followed: how many there are, how they are spaced, and the questions about them still unanswered

G. H. Hardy held up Euclid's proof as a model of mathematical beauty, and twenty-three centuries have not dulled it: suppose the primes stop, multiply them all, add one, and the list is shown to be incomplete. Four lines settle a question about infinitely many numbers. The folio honours that argument and then follows where it led, into questions that are anything but settled. The primes thin out yet never stop; twin primes keep appearing, though no one can prove they always will; the Ulam spiral throws up diagonal streaks that hint at hidden order; and primes of the form 4k + 3 lead those of the form 4k + 1 in a race whose lead first changes hands at 26,861. From the record Mersenne prime found in 2024 to the RSA keys that guard the internet, the oldest proof in number theory is still at work.

prime numbers · Euclid's theorem · sieve of Eratosthenes · prime number theorem · twin primes · Ulam spiral · Mersenne primes · Riemann hypothesisOpen the folio →

The Prime Number Theorem

How thinly the primes are spread: Gauss's guess, the 1896 proof through the zeta function, Riemann's zeros behind the error, and what remains open

One by one the primes look random, yet in bulk they obey a law of startling simplicity: near a large number x, about one number in ln x is prime. Gauss guessed it as a teenager, counting primes by the thousand, and it took a century and a new kind of mathematics to prove. The proof of 1896 ran through the Riemann zeta function, and the folio shows the remarkable reason why. Riemann's explicit formula rebuilds the staircase of primes out of waves, one for each zero of zeta, and with the first hundred zeros, computed here from scratch, the staircase rises step by step out of the sum. How closely the primes keep to the law depends on where those zeros lie. That is the Riemann hypothesis, open since 1859 and carrying a million-dollar Clay prize, and the folio makes visible why so much rests on it.

prime number theorem · prime-counting function · logarithmic integral · Riemann zeta function · Riemann hypothesis · explicit formula · prime gapsOpen the folio →

The Irrationality of √2

Why no fraction equals the diagonal of a unit square: the ancient proofs, the best approximations from Babylon to Pell, the digits, and the questions still open

Legend has it that the Pythagoreans drowned the man who revealed that the diagonal of a square cannot be measured by any fraction. Whatever the truth of the story, the discovery was a crisis: numbers as the Greeks understood them could not describe even the simplest geometry. The folio proves it twice, once with the classic odd–even argument and once with a picture of squares nesting ever smaller, then follows the long human effort to pin √2 down: a Babylonian tablet, YBC 7289, accurate to less than a millionth some 3,700 years ago; the fractions 7/5, 17/12 and 41/29 that solve Pell's equation; and a million digits computed here. Whether every finite block of decimal digits occurs in them with its expected frequency, the property called normality, remains unknown. A question about the diagonal of a square ended by forcing mathematics to rebuild its idea of number.

square root of 2 · irrational numbers · proof by contradiction · continued fractions · Pell's equation · YBC 7289 · Heron's method · Dedekind cutsOpen the folio →

The Principle of Stationary Action

Physical trajectories make the action stationary: one principle behind mechanics, optics, relativity, field theory and quantum physics

Nature, it seems, does not work out the next step; it compares whole histories. A thrown ball, a ray of light and an orbiting planet all follow the path that makes one quantity, the action, stationary, and from that single demand come Newton's laws, optics, relativity and Maxwell's equations. Richard Feynman never forgot the day his high-school teacher showed it to him. The folio makes the principle concrete: a ball's true path is computed against trial paths, mirrors show that "least" action is often not least at all, and a twin who flies out and back at 0.8c ages 6 years to the stay-at-home's 10. Then comes the deepest picture, Feynman's sum over histories, in which a quantum particle explores every path: of 40,000 arrows, those far from the classical path largely cancel, while those near it, where the action is stationary, add up. The classical world is where the quantum possibilities agree.

principle of stationary action · principle of least action · Lagrangian mechanics · Euler–Lagrange equation · Fermat's principle · Hamilton's principle · path integral · calculus of variationsOpen the folio →

The Euler–Lagrange Equation

Why a whole curve can be the answer to a question: the quickest way down, the equation behind it, least action, symmetry and conservation, and the laws of physics written as one function

In 1696 Johann Bernoulli challenged the best minds of Europe to find the curve down which a bead slides from one point to another in the least time. Newton solved it overnight and sent his answer anonymously; Bernoulli is said to have recognised the lion by its claw. The answer was not a straight line but a cycloid, and the question created a new mathematics in which the unknown is a whole curve. The folio races the slides and lets the cycloid win, then shows its second surprise: beads released from any height on it reach the bottom together. Euler and Lagrange turned such puzzles into one equation, and that equation now writes the laws of physics, from soap films and mirages to Einstein's gravity, and drives the chaos of a double pendulum. Weierstrass added the final twist: sometimes a best curve does not exist at all.

Euler–Lagrange equation · calculus of variations · brachistochrone · cycloid · principle of least action · Lagrangian mechanics · Noether's theorem · Fermat's principleOpen the folio →

Noether's Theorem

Every continuous symmetry brings a conservation law: shifts in time give energy, shifts in space give momentum, rotations give angular momentum

Albert Einstein called Emmy Noether the most significant creative mathematical genius since the higher education of women began, and her theorem of 1918 shows why. Every continuous symmetry of nature's laws brings a conserved quantity: because the laws are the same today as tomorrow, energy is conserved; the same here as there, momentum; the same in every direction, angular momentum. Conservation laws, once separate facts learned from experiment, become consequences of symmetry, and the idea now runs through all of modern physics. The folio proves the theorem in three lines and then breaks the symmetries on purpose, so the reader can watch angular momentum wander when a force stops pointing to the centre and an orbit begin to turn when the inverse-square law is disturbed. It ends where symmetry itself gives way, in the quantum anomaly that lets the neutral pion decay.

Noether's theorem · symmetry and conservation laws · conservation of energy · angular momentum · Runge–Lenz vector · Lagrangian mechanics · symplectic integrators · quantum anomalyOpen the folio →

Bayes' Theorem

How evidence should change belief: the one-line rule behind medical tests, courtrooms, codebreaking, search and machine learning, and the arguments it still provokes

Of all the mathematics in this collection, Bayes' theorem may be the part people use most often without knowing it. It is the rule for changing one's mind: how far a piece of evidence should move a belief, and in which direction. The folio opens on a puzzle that many doctors get wrong. If a disease affects 1 person in 100, and a test catches 90% of cases but also flags 9% of healthy people, a positive result means only about a 9% chance of having the disease. From there it follows the theorem into the courtroom, where confusing two conditional probabilities helped convict Sally Clark; to Bletchley Park, where Alan Turing weighed evidence against Enigma in decibans; and into the searches that found a lost submarine and the wreck of Air France 447. Thomas Bayes never published it. Two and a half centuries later, it runs spam filters, navigation systems and much of machine learning.

Bayes' theorem · conditional probability · Bayesian inference · base rate fallacy · Monty Hall problem · prosecutor's fallacy · Bayesian search theory · Kalman filterOpen the folio →

Maxwell's Equations

Four laws that unite electricity, magnetism and light: how charges and currents shape fields, how fields make waves, and why light travels at 299,792,458 m/s

Seen from ten thousand years in the future, Richard Feynman wrote, the most significant event of the nineteenth century would be Maxwell's discovery of the laws of electrodynamics. Four equations unite electricity, magnetism and light, and when Physics World asked its readers in 2004 to name the greatest equations ever written, Maxwell's were among the favourites. The folio shows the moment of unification. Maxwell adds a single term, the displacement current, and the equations begin to carry waves; their speed follows from two constants measured on a laboratory bench, and it is the speed of light. Light turns out to be electricity and magnetism sustaining each other through empty space. Field lines, Faraday's moving magnet, the spectrum from gamma rays to radio and the antenna that launches a wave follow. Every radio, phone and power station descends from these four lines.

Maxwell's equations · electromagnetism · electromagnetic waves · speed of light · Faraday's law of induction · displacement current · electromagnetic spectrum · Gauss's lawOpen the folio →

Einstein's Field Equations

Spacetime geometry and the distribution of matter and energy are coupled: gravity becomes the curvature of spacetime, from Mercury's orbit and GPS clocks to black holes, gravitational waves and the expanding universe

Gravity, Einstein realised, is not a force at all. In John Wheeler's phrase, matter tells spacetime how to curve and spacetime tells matter how to move, and the field equations make that exact. The folio shows how strange the idea is and how practical. It accounts for the slow turning of Mercury's orbit that puzzled astronomers for decades, bends starlight at the Sun's edge, and makes a GPS satellite's clock run 38.5 microseconds a day fast, an error that would ruin navigation if it were not corrected. It predicts black holes, whose shadows the Event Horizon Telescope has since imaged; gravitational waves, first detected by LIGO on 14 September 2015; and a universe that expands. More than a century after Einstein wrote them down, the equations have passed every test they have been set.

Einstein field equations · general relativity · spacetime curvature · black holes · gravitational waves · perihelion precession of Mercury · gravitational lensing · Friedmann equationsOpen the folio →

Schrödinger's Equation

How quantum states evolve: matter waves, tunnelling, energy levels, the hydrogen atom, electrons in crystals and qubits, all from one equation

Electrons do not travel along paths; they spread as waves of possibility, and the equation Schrödinger published in 1926 says exactly how those waves change. Its consequences still defy intuition. A particle meets a barrier it lacks the energy to climb, and part of it appears on the other side: in the folio's computed example, 28.2% gets through where classical particles would pass about 1%. The same tunnelling sets the half-lives of radioactive nuclei, which here span 24 orders of magnitude. The equation explains why atoms have sharp energy levels and hydrogen glows red at 656 nanometres, why some solids conduct and others insulate, and how a qubit is flipped. Yet the equation itself is perfectly deterministic. Chance enters only when a measurement is made, and what happens then is a question physics has not settled.

Schrödinger equation · quantum mechanics · wave function · quantum tunnelling · hydrogen atom · energy levels · band theory of solids · Rabi oscillationOpen the folio →

The Dirac Equation

Quantum mechanics meets special relativity: an equation for the electron that brings spin, the fine structure of hydrogen and antimatter with it

In 1928 Paul Dirac, aged 25, went looking for an equation for the electron that respected Einstein's relativity, and found one so tightly constrained that it delivered things he had not asked for. It predicted the electron's spin and the strength of its magnetism, and it insisted on solutions with negative energy. Dirac refused to throw them away, and they turned out to be antimatter: the positron was found in 1932, a new kind of matter predicted by an equation before anyone had seen it. The folio follows the consequences, from the fine structure of hydrogen and the behaviour of electrons in heavy atoms, where relativity stops being a small correction, to the electrons in graphene, which move as if they had no mass. The equation is carved on Dirac's memorial in Westminster Abbey, a fitting place for a line that knew so much.

Dirac equation · relativistic quantum mechanics · electron spin · antimatter · positron · fine structure · anomalous magnetic moment · grapheneOpen the folio →

Subatomic Matter

From atoms to fundamental particles — what’s real, what we know, and what remains a mystery

An atom is almost entirely empty: its nucleus holds over 99.9% of the mass in a hundred-thousandth of its width. Stranger still, the bare masses of a proton's three valence quarks contribute only about 1% of its mass; most of it comes from the energy of quarks, gluons and their interactions. The folio descends from atoms to fundamental particles, showing electrons as clouds of probability, nuclei held by the strongest force in nature, and the Standard Model's twelve matter particles, force carriers and Higgs boson. It is careful to separate what is known from what is not. The Standard Model has passed test after test across an immense range of experiments, yet the dark matter that outweighs ordinary matter roughly five to one has never been detected directly. The folio maps the territory physics has explained, and marks clearly where the map runs out.

Standard Model · elementary particles · quantum chromodynamics · atomic orbitals · nuclear binding energy · fundamental interactions · dark matter · particle detectorsOpen the folio →

Mass in the Quantum World

From fields and symmetries to inertia, energy, and the mass we observe

Step on a scale and almost none of the number comes from the Higgs field. Mass, this folio explains, is the energy held inside a system, and in ordinary matter most of it is the restless energy of quarks and gluons inside protons and neutrons; the masses of their three valence quarks supply only about 1%. The idea overturns the everyday picture of mass as an amount of stuff. Particles appear as ripples in fields that fill all of space. The Higgs field settles at a value of about 246 GeV and gives the elementary particles their masses, spread so widely that the top quark outweighs the electron some 340,000 times. A seven-step strip then climbs from quantum fields to a 1 kg cylinder, and annihilation and pair creation show mass and energy changing places. What feels most solid about matter turns out to be motion and binding energy.

mass–energy equivalence · invariant mass · Higgs mechanism · quantum field theory · quantum chromodynamics · proton mass · Compton frequency · nuclear binding energyOpen the folio →

The Hierarchy of Cosmic Structure

From planetary systems to superclusters and the cosmic web: a classification by scale

From a dwarf planet 940 km across to the observable universe, some 93 billion light-years wide, the cosmos spans about twenty-one orders of magnitude, and this folio lays the whole ladder out on a single page. Each rung holds a different kind of order: belts and clouds of icy bodies around the Sun; visitors such as 'Oumuamua, arriving from other stars; more than six thousand known exoplanets; stars from blue giants to red dwarfs, the commonest stars of all, though not one is visible to the naked eye; galaxies sorted by shape; and the cosmic web of filaments and voids. Nested arcs give Earth's full cosmic address, ending in Laniakea, a supercluster whose name is Hawaiian for "immeasurable heaven". Gravity has assembled every level up to the clusters of galaxies; beyond them, superclusters are outlined by the flow of galaxies rather than bound by it.

cosmic structure · solar system · exoplanets · stellar classification · stellar remnants · nebulae · Hubble sequence · cosmic webOpen the folio →

Cosmology and Astrophysics

Observations, physical laws, and computational models — from cosmic expansion and dark matter to stars and black holes

All ordinary matter, every star, planet, gas cloud and living thing, makes up about 5% of the Universe. The rest is dark matter, known only through its gravity, and dark energy, which is accelerating the expansion of space and whose nature no one understands. This folio shows how such claims can be made at all: a ladder of distance measurements from stellar parallax to the frozen sound waves of the early universe, the cosmic microwave background glowing at 2.7255 K, galaxies spinning too fast for their visible mass, stars forging the elements of our bodies, and two black holes colliding in the gravitational waves of 2015. Six numbers summarise nearly fourteen billion years of history. And yet two precise ways of measuring the expansion rate disagree, 67.4 against 73.0 km/s per megaparsec, a tension that may be the first sign of new physics.

cosmology · cosmic microwave background · Lambda-CDM model · Friedmann equations · dark matter · dark energy · Hubble tension · gravitational wavesOpen the folio →

Quantum Foundations and Theory

From postulates and superposition to measurement, interpretation, and the theorems that constrain reality

Quantum mechanics has never failed an experimental test, yet a century after it was written physicists still disagree about what it says the world is like. That is the subject of this folio, and it is among the deepest open questions in science. A single particle passes through two slits and interferes with itself. Measurement seems to force a definite outcome, although nothing in the equations says when or how, and Wigner's friend sets two observers' accounts against each other. Decoherence explains why interference between macroscopically distinct states is normally so hard to observe, but not why one result occurs rather than another. Six interpretations, from Copenhagen to many worlds, are set side by side, and the theorems of Bell and Kochen–Specker show which intuitive pictures of reality cannot survive. The folio does not pretend to settle the question; it shows precisely what any answer must face.

quantum mechanics postulates · measurement problem · decoherence · interpretations of quantum mechanics · Bell's theorem · uncertainty principle · Wigner's friend · no-go theoremsOpen the folio →

Quantum Information

From qubits to entanglement and channels — what can be copied, sent, and kept secret, and what remains unsolved

Quantum information obeys rules that would be impossible for ordinary data. An unknown quantum state cannot be copied. Two entangled particles can share correlations stronger than any classical system allows, strong enough to violate Bell's inequality, as loophole-free experiments confirmed in 2015. A qubit's state can even be teleported, destroyed in one place and rebuilt in another, using one shared entangled pair and two classical bits. The folio sets out this new theory of information: states drawn as points in the Bloch ball, entanglement counted in ebits, noise as a deformation of the ball, and a conditional entropy that can turn negative, something the conditional entropy of classical discrete data cannot do. It also maps the limits, from no-cloning to the Holevo bound. In well-designed protocols those prohibitions become guarantees, which is why this theory underlies both quantum cryptography and quantum computing.

quantum information theory · density matrix · quantum entanglement · von Neumann entropy · quantum teleportation · superdense coding · no-cloning theorem · quantum channel capacityOpen the folio →

Quantum Algorithms

From interference and phase estimation to end-to-end resource estimates: what it takes to turn a quantum speed-up into a useful application

A quantum computer does not try every answer at once, whatever the popular story says. It does something subtler: it arranges for the paths that lead to wrong answers to cancel and those that lead to the right one to reinforce. This folio explains that mechanism and shows what it can and cannot buy. Shor's algorithm factors large numbers far faster than any known classical method, and a 2025 resource estimate puts breaking RSA-2048 at under a million physical qubits and under a week. Grover's search gives a square-root speed-up, and for an unstructured search no quantum algorithm can do better. A runtime chart shows why a modest speed-up can take hours to pay off on slow, error-corrected hardware. For anyone weighing the claims made for quantum computing, the folio is a map of where real advantage lies.

quantum algorithms · quantum Fourier transform · phase estimation · Grover's algorithm · Shor's algorithm · Hamiltonian simulation · quantum singular value transformation · BQPOpen the folio →

Quantum Communications

From the limits of light to quantum key distribution, its protocols, attacks, and place beside post-quantum cryptography

Encrypted messages recorded today could be read once a large quantum computer exists, so secrets meant to last for decades are already at risk. This folio shows the two answers being built. Post-quantum cryptography replaces vulnerable mathematics with problems believed hard even for quantum computers; quantum key distribution draws its security from physics, because measuring quantum signals disturbs them in ways that can be detected and bounded. The folio traces the ultimate limits on reading light, the families of QKD protocols and keys carried through 1,002 km of fibre in 2023. It also shows the practical side: losses in fibre and from satellites, and real attacks that blinded detectors or slipped through imperfect hardware, with their countermeasures. Quantum security is powerful, and it still depends on authentication, well-characterised devices and careful statistics.

quantum communication · quantum key distribution · BB84 · post-quantum cryptography · Holevo bound · Helstrom bound · twin-field QKD · quantum hackingOpen the folio →

Quantum Networking

From photons and memories to repeaters, protocol stacks, and the path to a quantum internet

A quantum internet would link quantum computers and sensors as today's internet links ordinary ones, but it cannot work the same way. Ordinary networks copy and amplify signals at every step; an unknown quantum state cannot be copied, so many proposed quantum networks, especially over long distances, distribute entanglement between nodes and use teleportation to transfer states. The folio shows why this is so hard and how it might be done. Over a direct pure-loss link no protocol can beat the PLOB bound, and light fades so quickly in fibre that at 1,000 km, where the folio puts the transmission near 10⁻²⁰, a 10 GHz source would deliver one photon every ~300 years. Quantum repeaters, memories, entanglement swapping and purification are the way around, and satellites reached 12,900 km in 2025. It is a portrait of one of the great engineering projects now under way.

quantum networking · quantum internet · quantum repeaters · entanglement swapping · quantum key distribution · quantum memory · entanglement purificationOpen the folio →

Quantum AI

From variational circuits and quantum kernels to neural decoders: where quantum computing and artificial intelligence genuinely help each other

Few phrases attract more hype than "quantum AI", and this folio sets every claim beside its fine print. It separates two real ideas: quantum computers used for machine learning, and machine learning used to build quantum computers. For the first it sets the promise of variational circuits and quantum kernels beside the fine print: barren plateaus where training stalls, the cost of loading data, and classical algorithms that keep catching up. Where the data are themselves quantum, an advantage is proven: a learner with quantum memory can need exponentially fewer experiments. The second direction is already paying off, as neural networks such as AlphaQubit decode errors and others calibrate devices. A mid-2026 scoreboard then grades each claim, from established to doubtful.

quantum machine learning · variational quantum circuits · quantum kernels · barren plateaus · quantum advantage · neural decoders · dequantisationOpen the folio →

Building a Quantum Computer

From qubits and Hamiltonians to error correction, the physical platforms, and the path to fault tolerance

A quantum computer has to do two contradictory things: keep its qubits so well isolated that the world cannot disturb them, and control them precisely enough to compute. This folio shows how that contradiction is being overcome. Typical physical qubits make an error roughly once in a thousand operations, while long computations call for logical error rates nearer one in ten billion. The gap is closed by error correction, which spreads one logical qubit across many physical ones, and in 2024 Google's Willow chip showed errors falling as its code grew larger, the long-awaited sign that the approach works. Six rival technologies, superconducting circuits, trapped ions, neutral atoms, photons, spins in semiconductors and topological qubits, are compared on equal terms. The hard problem is no longer making a qubit but making a million of them work together as one machine.

quantum computer · qubit · quantum error correction · surface code · superconducting qubits · trapped ions · neutral atoms · fault-tolerant quantum computingOpen the folio →

Quantum Hardware Engineering

From millikelvin cryostats and microwave wiring to lasers, materials and control electronics: the machinery behind the qubits

Some of the coldest places in the known universe are inside laboratories. A superconducting qubit must sit near 10 millikelvin, hundreds of times colder than deep space, or stray heat will scramble it. This folio is a cutaway of the machinery that makes that possible: the dilution refrigerator, in which helium-3 crossing into helium-4 absorbs heat; wiring that brings signals in while keeping warmth out; Josephson junctions whose insulating barrier is only 1–2 nm thick; and the lasers, optical tweezers and photonics that drive atom, ion and photon machines. It shows the noise that hardware engineers fight, including cosmic rays that can disturb many qubits at once. Behind every qubit stands a large classical machine built to keep it cold, quiet and controlled.

dilution refrigerator · superconducting qubits · qubit readout · Josephson junction · cryogenic control electronics · trapped ions · photonic quantum computing · quantum noiseOpen the folio →

Information Theory

From entropy and channel capacity to error-correcting codes and multi-user networks: the mathematical limits of communication

In 1948 Claude Shannon published a paper that has been called the Magna Carta of the information age. It gave information a unit, the bit, and proved that every noisy channel has a capacity: below it, messages can be sent with as few errors as anyone likes; above it, they cannot. Every phone, modem, deep-space probe and streaming service lives under that limit. The folio builds the theory from one idea, that information is surprise, and shows entropy setting the limit of compression, the Shannon limit standing as a hard wall for communication, and the same mathematics reaching into physics, statistics, biology and today's language models. Shannon proved that codes approaching capacity must exist without showing how to build them; finding such codes took nearly half a century.

information theory · Shannon entropy · channel capacity · Shannon limit · mutual information · data compression · error-correcting codes · rate–distortion theoryOpen the folio →

Coding Theory

From Hamming distance and finite fields to LDPC, turbo and polar codes: how redundancy defeats noise

Whenever a scratched disc still plays, a smudged QR code still scans or a probe beyond the planets sends home a clean picture, coding theory is at work. Its idea is to add redundancy with structure, spacing valid messages so far apart that a corrupted one can still be recognised and repaired. The folio follows the subject from Hamming's codes, through Reed–Solomon codes built from polynomials over finite fields and the Viterbi algorithm, to the modern codes that finally came close to Shannon's limit: turbo codes in 1993, LDPC codes, and Arıkan's polar codes. One computed picture shows polar coding splitting 1,024 channels into some that are almost perfect and some that are almost useless, and sending data only through the good ones. Nearly half a century separated Shannon's promise from codes that kept it.

coding theory · error-correcting codes · Hamming code · Reed–Solomon code · Viterbi algorithm · LDPC codes · turbo codes · polar codesOpen the folio →

Digital Signal Processing

From sampling and the Fourier transform to filters, the FFT and spectral estimation: how signals become numbers and back again

Every song streamed, photograph taken and call made passes through the mathematics in this folio. A continuous signal becomes numbers by sampling and rounding: the sampling theorem shows that ideal samples of a signal with no frequency at or above half the sampling rate determine it exactly, while the rounding adds a small quantisation error. The folio makes the counter-intuitive visible, as a 7 kHz tone sampled at 10 kHz becomes indistinguishable from a 3 kHz one. It then shows the tools that run the digital world: the Fourier transform, which turns a signal into its frequencies, filters, spectrograms, and the fast Fourier transform, which cuts a million-point calculation from about a trillion multiplications to about ten million. Gauss had the idea in 1805, and whether it can be done faster still is unknown.

digital signal processing · sampling theorem · aliasing · discrete Fourier transform · fast Fourier transform · digital filters · spectral leakage · delta-sigma modulationOpen the folio →

Cryptography

From symmetric ciphers and hash functions to public keys, signatures and post-quantum algorithms: the mathematics of trust

Every secure website, card payment and private message rests on a few mathematical problems that are easy to pose and believed to be hard to solve. This folio shows how that trust is built. Since Kerckhoffs in 1883 a cipher has been expected to stay secure even when everything but its key is public, and modern designs such as AES and SHA-3 are published, scrutinised and unbroken in practice. Public-key cryptography, introduced in 1976, lets two strangers agree on a secret in full view of an eavesdropper, an idea that once sounded impossible. Quantum computers now threaten it, and the folio sets out the response, the post-quantum standards of 2024, alongside zero-knowledge proofs and real failures such as Heartbleed. At the bottom lies an open question: proving that one-way functions exist would also establish that P ≠ NP.

cryptography · AES · hash functions · public-key cryptography · RSA · elliptic-curve cryptography · post-quantum cryptography · zero-knowledge proofsOpen the folio →

Graph Theory

From Euler’s bridges to spectral methods and random networks: the mathematics of connection

In 1736 Euler proved that no walk could cross each of Königsberg's seven bridges exactly once, and in doing so invented a mathematics of pure connection. The folio shows how far that idea has travelled: the same reasoning finds the shortest route in a map app and lays the cheapest network of cables. It also holds one of the deepest mysteries in computing. Deciding whether a walk can use every edge once is easy; deciding whether one can visit every vertex once is NP-complete, and no one knows whether such problems can ever be solved efficiently. Planarity, maximum flows and the spectral methods behind modern clustering follow. Even small questions resist: how large a party must be to guarantee five mutual friends or five mutual strangers, the Ramsey number R(5,5), is known only to lie between 43 and 46.

graph theory · Seven Bridges of Königsberg · Eulerian path · shortest path · spanning tree · planar graph · max-flow min-cut theorem · spectral graph theoryOpen the folio →

Network Science

From small worlds and hubs to communities, cascades and epidemics: the structure and dynamics of complex networks

Any two people seem to be linked by a short chain of acquaintances, and a 2012 study of a Facebook graph of 721 million users put the average distance at 4.74. Network science explains why the world is so small, and why the same patterns appear in friendships, the internet, power grids, brains and epidemics. The folio shows a few random shortcuts collapsing the distances in a clustered network, hubs emerging as the well-connected attract still more links, communities found by algorithm, and networks that shrug off random failures yet fall apart when their hubs are attacked. It shows why diseases spread so readily through networks with hubs, and why most people's friends have more friends than they do. Whether something spreads, breaks or falls into step depends on the wiring as much as on the parts.

network science · small-world network · scale-free network · preferential attachment · community detection · network robustness · epidemic threshold · Kuramoto modelOpen the folio →

6G: The Mathematical Frontier

From integrated sensing and near-field arrays to AI-native control and semantic communication: the mathematics the next generation may force us to develop

Each mobile generation so far has been defined by speed. The sixth is meant to be different: a network that senses its surroundings, learns and controls machines as well as carrying data, and the folio argues that this union is where genuinely new mathematics will be needed. It sets out the targets of the ITU's IMT-2030 framework and then the open problems behind them: how one signal can serve both communication and radar-like sensing, how vast antenna arrays can focus energy on a spot a few metres away, how a network can optimise itself safely, how to transmit meaning rather than bits, and how reliable a very short message can be. It is a look at research questions that are still open before the technology arrives. Communication, sensing and control each have mature theories; their joint limits are still unknown.

6G · IMT-2030 · integrated sensing and communication · near-field MIMO · cell-free massive MIMO · semantic communication · finite blocklength · OTFS modulationOpen the folio →

Regions of the Brain

From lobes and gyri to deep nuclei, pathways and networks: where each region of the brain sits, what it does, and what its name means

Neuroanatomy has a reputation for impenetrable names, and this folio dissolves it. Most names are a position plus a landmark, so once about twenty word parts are known, "dorsolateral prefrontal cortex" decodes itself, and the hippocampus turns out to be named for a seahorse. On that foundation the folio maps the brain region by region, from the lobes and deep nuclei to the limbic circuits of memory and emotion and the networks active when the mind is at rest. It is also the one folio that works almost like an app: a search built into the page finds any term printed on the plate, shows a magnified preview of every match, and jumps straight to it, with a miniature map of the whole folio marking where each one lies. The mysteries are real too. The cerebellum, about a tenth of the brain's volume, holds roughly 69 of its 86 billion neurons, and although its part in thought and emotion is well attested, exactly what it contributes is still being worked out.

brain anatomy · cerebral cortex · lobes of the brain · limbic system · basal ganglia · brainstem · neuroanatomical terminology · resting-state networksOpen the folio →

Computational Neuroscience

From ion channels and spikes to synapses, neural codes and learning rules: the mathematics of how neurons represent the world and learn from it

A human brain runs on about 20 watts, yet its slow and noisy cells learn, recognise and decide in ways machines still struggle to match. Computational neuroscience asks how, and this folio follows the answer from the bottom up, computing every figure from the models themselves: the Hodgkin–Huxley equations of the nerve impulse, which earned a Nobel Prize; neurons as small electrical circuits; a population of cells voting on the direction of a movement; a network that restores a damaged memory; synapses that strengthen when one cell fires just before another; and dopamine signalling the gap between expected and actual reward. Several of these ideas became foundations of machine learning, and the folio's closing table traces that lineage. Brains inspired artificial neural networks; now those networks are helping to explain brains in return.

computational neuroscience · Hodgkin–Huxley model · integrate-and-fire neuron · cable theory · neural coding · Hopfield network · synaptic plasticity · reinforcement learningOpen the folio →

How Cells Find Their Role

From one fertilised egg to hundreds of cell types: the gradients that tell a cell where it is, the gene switches that decide what it does, and the memory that keeps it that way

Nearly every nucleated cell in the body carries the same genome, yet one becomes a neuron and another a liver cell, and each remembers what it is through every division that follows. How a single fertilised egg gives rise to hundreds of cell types, each in the right place, is one of the great questions of biology, and this folio shows how much of the answer is now known. Cells read their position from gradients of signalling molecules, as in Wolpert's French flag; neighbours talk each other into different fates; genes that repress each other act as switches with two stable states; Turing's chemistry paints spots and stripes; and a molecular clock lays down the segments of the body, about 315 pairs of them in the corn snake. Much remains unexplained. The other fates are often not erased but closed off, which is why a skin cell can be reprogrammed into a stem cell.

cell fate · developmental biology · Waddington landscape · morphogen gradient · Notch signalling · Turing patterns · Hox genes · epigeneticsOpen the folio →

Head and Neck

From the skull and muscles of facial expression to the larynx, thyroid and cervical spine: the head and neck in six views

The neck is one of the most crowded passages in the body. In a few centimetres it carries the airway and the foodway, the spinal cord, major vessels, and the muscles that turn the head, while the face above it holds the muscles behind every expression. The plate gives six views: the skull and face from the front and the side, the muscles at the back of the neck, a midline section through nose, mouth and throat, a typical cervical vertebra and the larynx. The section repays a close look, showing where air and food share a single passage and how the larynx protects the airway at every swallow.

head and neck anatomy · skull · muscles of facial expression · cervical vertebrae · larynx · thyroid gland · sternocleidomastoid · pharynxOpen the folio →

The Spine

From the atlas to the coccyx: the vertebral column and its curves, the typical vertebra and the intervertebral disc

The spine must be rigid enough to carry the body, flexible enough to bend and twist, and strong enough to protect the spinal cord, all at once. It manages with an elegant design: a stack of separate vertebrae cushioned by discs and arranged in four alternating curves that work like a spring. The plate counts the column from the atlas at the top to the coccyx, shows where the spinal cord ends, near L1–L2 in adults, and the cauda equina of nerve roots below, then looks closer at a lumbar vertebra, a disc with its tough outer ring and soft core, and a nerve root leaving the canal. It is a clear map of a structure most people only think about when it hurts.

vertebral column · spine anatomy · cervical lordosis · thoracic kyphosis · lumbar vertebra · intervertebral disc · cauda equina · spinal cordOpen the folio →

The Shoulder

From the clavicle and scapula to the rotator cuff and glenohumeral joint: the anatomy of the body’s most mobile joint

No joint in the body moves through a wider range than the shoulder, and the price of that freedom is instability: a large ball rests against a shallow socket and is held there less by bone than by muscle. The plate shows how the arrangement works, from the front and from behind, then isolates the rotator cuff, the muscles that keep the ball centred, and opens the joint itself to reveal the labrum, the capsule and the bursa beneath the acromion. A final view shows the bones alone. For anyone who throws, swims, lifts or has felt a shoulder complain, it shows what is happening beneath the skin.

shoulder anatomy · rotator cuff · glenohumeral joint · scapula · clavicle · deltoid · glenoid labrum · subacromial bursaOpen the folio →

The Arms

From shoulder to hand: anterior and posterior muscles, elbow and forearm bones, the deeper brachialis and their roles in movement

Turning a key or a doorknob relies on a quiet piece of engineering: the forearm has two bones, and one rolls around the other to turn the palm up or down. The plate shows the right arm from the front and from behind, from the deltoid and biceps to the long forearm muscles whose tendons run into the hand. It then sets out the elbow and forearm bones and a deeper dissection of brachialis, the muscle that does much of the work of bending the elbow whichever way the palm faces. A table of actions ties each muscle to a movement, turning a list of names into an understanding of how the arm works.

upper limb anatomy · biceps brachii · triceps brachii · brachialis · radius and ulna · forearm muscles · pronation and supination · elbowOpen the folio →

The Hands

Palmar and dorsal anatomy: intrinsic muscle groups, flexor and extensor tendons, the bones of the hand and their roles in movement

With 27 principal bones and a thumb that can meet each fingertip, the human hand is one of the most versatile tools in nature, and much of its strength does not lie in the hand at all. The most powerful finger muscles sit in the forearm and pull through long tendons that pass beneath fibrous bands at the wrist. The plate shows the right hand from the palm and from the back, with the muscles of the thumb and little finger, the flexor and extensor tendons and the interossei, beside a skeleton counting carpals, metacarpals and phalanges. Its notes explain which muscles position the thumb and which spread the fingers, the movements behind every grip and gesture.

hand anatomy · bones of the hand · carpals · metacarpals · phalanges · flexor tendons · thenar muscles · dorsal interosseiOpen the folio →

The Abdomen

The abdominal wall: layered muscles, connective tissues, posterior support and the attachments that link the rib cage to the pelvis

In front of and beside the lumbar spine, nothing bony spans the gap between the ribs and the pelvis, so the abdominal wall does its work with layers of muscle and connective tissue: three muscular sheets whose fibres run in different directions, like plywood, joined by tough aponeuroses, with rectus abdominis in front. The plate opens the wall layer by layer, from rectus abdominis and its tendinous intersections to the external oblique, internal oblique and transversus abdominis, then removes the organs to show the muscles of the back wall, psoas major and quadratus lumborum. A table gives each muscle its role, including the partnership of opposite obliques that turns the trunk.

abdominal wall anatomy · rectus abdominis · external oblique · internal oblique · transversus abdominis · linea alba · psoas major · quadratus lumborumOpen the folio →

The Hip

From the pelvis and proximal femur to the labrum, ligaments and deep gluteal muscles: the ball-and-socket joint that carries the body

Walking puts several times the body's weight through the hip, and the joint is built for it: a deep ball-and-socket in which the head of the femur sits almost enclosed, its rim deepened by the labrum and bound by some of the strongest ligaments in the body. The plate shows the right hip from the front, and from behind, where the deep rotator muscles and the sciatic nerve lie beneath gluteus maximus. It then opens the joint in two sections and gives the femur and the pelvic bones on their own. The socket itself is a meeting point: ilium, ischium and pubis join there, three bones forming one cup.

hip joint anatomy · acetabulum · femoral head · acetabular labrum · proximal femur · gluteal muscles · piriformis · iliofemoral ligamentOpen the folio →

The Legs

From the thigh to the ankle: anterior and posterior muscles, the deeper quadriceps and calf, and their roles in movement

Walking looks effortless, but each step asks the leg to straighten under load, swing clear of the ground and land again, and its muscles are organised front and back to do exactly that. The plate shows the right leg from the front and from behind: the quadriceps converging on the kneecap, the hamstrings, the long muscles of the shin, and the two calf muscles, gastrocnemius and soleus, which share the Achilles tendon, the thickest and strongest tendon in the body. Two deeper dissections lift away a layer each, and a table of actions ties every muscle to its movement, from straightening the knee to turning the foot outward.

lower limb anatomy · quadriceps femoris · hamstrings · gastrocnemius · soleus · Achilles tendon · tibialis anterior · thigh musclesOpen the folio →

The Knee

From the patella and menisci to the cruciate and collateral ligaments: the hinge between femur and tibia, front, back, side and top

The knee is the largest joint in the body and, mechanically, an improbable one: rounded femoral condyles resting on a nearly flat tibial plateau, held together far more by ligaments and cartilage than by the fit of its bones. The plate shows the right knee from the front, from behind, in section and from above, with the patella and its tendons, the collateral ligaments at the sides, the two menisci that cushion and steady the joint, and the anterior and posterior cruciate ligaments crossing at its centre. These are the structures behind the most familiar sports injuries, and the views make clear why each one matters.

knee joint anatomy · anterior cruciate ligament · posterior cruciate ligament · menisci · collateral ligaments · patella · tibial plateauOpen the folio →

The Foot and Ankle

From the tarsal bones to the ankle ligaments, tendons and plantar fascia: the right foot and ankle in seven views

The whole weight of the body reaches the foot through a single bone, the talus, to which no muscle attaches; it is held in place by the bony fork of the ankle and by ligaments on both sides. The plate shows the right foot and ankle in seven views, from above, below, inside and outside: the tendons that pass the ankle, the ligaments most often sprained, and the plantar aponeurosis that supports the sole. Smaller panels give the ankle joint from the front, a section through the talus and the bones of the foot from above, the frame on which every step is balanced.

ankle anatomy · foot anatomy · talus · ankle ligaments · plantar fascia · tarsal bones · deltoid ligament · anterior talofibular ligamentOpen the folio →

The Feet

Dorsal and plantar anatomy: toe extensor tendons, the superficial sole muscles, the bones of the foot and their roles in support and movement

Twenty-six bones, arranged in arches rather than laid flat, let the foot serve both as a flexible platform that adapts to uneven ground and as a rigid lever for pushing off, and it switches between the two with every step. The plate shows the right foot from above and from below: the long extensor tendons fanning out to the toes, the small extensor digitorum brevis on top of the foot, and the first layer of muscles in the sole, beside a skeleton counting its tarsals, metatarsals and phalanges. A closing strip gives each muscle its role. The foot has one bone fewer than the hand, seven tarsals where the wrist has eight carpals.

foot anatomy · bones of the foot · tarsals · metatarsals · phalanges · arches of the foot · abductor hallucis · extensor digitorum brevisOpen the folio →