Reference

Glossary: Hardy ↔ modern

Hardy wrote in 1908. Many of his words are still in use with the same meaning; some have changed or fallen out of use. Each entry links to the lesson where the idea first appears.

HardyModernMeaningWhere
'hypergeometric' series Gauss hypergeometric series ₂F₁(α, β; γ; 1) 1+αβ1⋅γ+α(α+1)β(β+1)1⋅2⋅γ(γ+1)+⋯1 + \frac{\alpha\beta}{1\cdot\gamma} + \frac{\alpha(\alpha+1)\beta(\beta+1)}{1\cdot2\cdot\gamma(\gamma+1)} + \cdots; converges iff γ>α+β\gamma > \alpha + \beta. §183
{PP′} (arc) and PP′ (chord) arc length and chord length Hardy writes {PP′}\{PP'\} for the length of the arc between PP and P′P', and PP′PP' for the chord. A curve whose inscribed polygons have bounded length is now called rectifiable. §149
vnv_n, wnw_n (the positive and negative terms) positive and negative parts un+u_n^+, un−u_n^- vn=max⁡(un,0)v_n = \max(u_n, 0), wn=max⁡(−un,0)w_n = \max(-u_n, 0), so that un=vn−wnu_n = v_n - w_n and ∣un∣=vn+wn|u_n| = v_n + w_n. §191
a constant constant function A function attaching the same value yy to every xx. §20
Abel's theorem (§179) Abel–Pringsheim (Olivier) theorem If ∑un\sum u_n converges with positive decreasing terms then nun→0n u_n \to 0. Not the theorem usually called Abel's theorem today (on power series). §179
abscissa, ordinate x-coordinate, y-coordinate For the point P=(a,b)P = (a, b), aa is the abscissa and bb the ordinate. §21
absolutely convergent (infinite integral) absolutely convergent improper integral ∫a∞f\int_a^\infty f with ∫a∞∣f∣\int_a^\infty |f| convergent; if ∫a∞f\int_a^\infty f converges but ∫a∞∣f∣\int_a^\infty|f| does not, it is conditionally convergent. §203
absolutely convergent (series) absolutely convergent ∑un\sum u_n is absolutely convergent if ∑∣un∣\sum |u_n| converges; it then converges, and may be rearranged freely. §191
aggregate set Any collection of numbers or points. §18
aggregate (of real numbers) set Any collection of real numbers; Hardy also says “system”. §80
algebraic number algebraic number A root of an equation a0xn+⋯+an=0a_0x^n + \dots + a_n = 0 with integer coefficients, not all zero. §15
algebraical function of degree m algebraic function A root yy of ym+R1(x)ym−1+⋯+Rm(x)=0y^m + R_1(x)y^{m-1} + \dots + R_m(x) = 0 with rational functions RkR_k. §27
all the intervals δ_v tend to zero the mesh of the partition tends to 0 The length of the longest piece, max⁡vδv\max_v \delta_v, tends to 00; Hardy bounds it by a number δ\delta. §161
alternating series alternating series A series φ0−φ1+φ2−⋯\varphi_0 - \varphi_1 + \varphi_2 - \cdots with all φn>0\varphi_n > 0. If φn\varphi_n decreases to 00 it converges, with error at most the first omitted term. §195
ambiguous sign branch of a many-valued inverse function The ±\pm in formulas such as Dxarcsin⁡x=±1/1−x2D_x \arcsin x = \pm 1/\sqrt{1 - x^2}; the sign depends on which value of the inverse function is meant. With principal values it is ++. §120
amplitude, am z argument, arg z (principal value Arg z) The angle θ\theta with z=∣z∣(cos⁡θ+isin⁡θ)z = |z|(\cos\theta + i\sin\theta); defined up to multiples of 2π2\pi. §44
an infinite number of infinitely many Shorthand for “the class is not finite”: for every nn it has more than nn members. Not a number called infinity. §52
analytical theory of the circular functions rigorous definition of sin, cos, tan without geometry Defining the trigonometric functions by analysis alone; Hardy's choice is arctan⁡x=∫0xdt/(1+t2)\arctan x = \int_0^x dt/(1+t^2), 12π=∫0∞dt/(1+t2)\frac12\pi = \int_0^\infty dt/(1+t^2), tan⁡\tan by inversion, then cos⁡\cos and sin⁡\sin from tan⁡\tan. §224
angular points; centre of gravity; perpendiculars vertices; centroid; altitudes Old names in triangle geometry used in Examples XXI, 10; the altitudes meet at the orthocentre. §46
approximately true (equation) true up to o(‖(δx, δy)‖) The difference of the two sides is small compared with the larger of ∣δx∣|\delta x| and ∣δy∣|\delta y|. §159
arbitrary constant (of integration) constant of integration, + C+\,C The constant that may be added to any integral; on a domain made of several intervals, a separate constant is allowed on each. §131
arc cos x, arc sin x arccos x, arcsin x (all values: 2nπ ± arccos x) Hardy’s inverse cosine is infinitely many-valued: all yy with cos⁡y=x\cos y = x. Today arccos⁡x\arccos x is the principal value in [0,π][0, \pi]. §28
arc sin x, arc tan x, cosec x arcsin x, arctan x, csc x Hardy's spellings of the inverse sine and tangent (many-valued for him; we use principal values) and of the cosecant 1/sin⁡x1/\sin x. §146
arc tan x arctan x (principal value) Defined by arctan⁡x=∫0xdt/(1+t2)\arctan x = \int_0^x dt/(1+t^2), with values in (−12π,12π)(-\tfrac12\pi, \tfrac12\pi); Hardy's notation also allowed the many-valued inverse. §163
area of a plane curve; (ONPP₀) area of a region; ∫0xϕ(t) dt\int_0^x \phi(t)\,dt The area of the region bounded by a curve, the axis and two ordinates. Hardy writes the area of a region by listing its corners in brackets. §148
arg sinh, arg cosh, arg tanh arsinh, arcosh, artanh (sinh⁻¹, …) Inverse hyperbolic functions, e.g. arg sinh⁡s=log⁡{s+s2+1}\operatorname{arg\,sinh} s = \log\{s + \sqrt{s^2 + 1}\}; Hardy's arg cosh is the positive solution of cosh⁡x=c\cosh x = c. §217
Argand's diagram the complex plane The plane in which x+yix + yi is represented by the point (x,y)(x, y). §44
argument (of a function) input variable The number zz at which a function f(z)f(z) is evaluated. Not the angle of zz, which Hardy calls the amplitude. §228
argument (of a point) the complex number attached to a point Hardy says the point PP has the argument zz when PP corresponds to zz; not to be confused with the modern ‘argument’ = his amplitude. §44
arithmetical continuum the real line ℝ The set of all real numbers; the linear continuum is the line whose points correspond to them. §15
as small as we please for every ε > 0 … “We can make y−xy - x as small as we please” means: for every positive δ\delta there are xx, yy with y−x<δy - x < \delta. §4
assume the form ‘0/0’ indeterminate form 0/0 A ratio f(x)/ϕ(x)f(x)/\phi(x) with f,ϕ→0f, \phi \to 0; its limit is found by comparing expansions or by L'Hôpital's rule. §154
axiom of Archimedes Archimedean property For any length BCBC there is a whole number kk with k⋅BC>1k \cdot BC > 1. §3
binomial equation the equation z^n = c An equation of the form zn−c=0z^n - c = 0. Proof (B) of the fundamental theorem assumes it can be solved; Littlewood's variant avoids that assumption. App. II
binomial form (of a polynomial) polynomial with binomial coefficients a0xn+(n1)a1xn−1+⋯+ana_0x^n + \binom{n}{1}a_1x^{n-1} + \dots + a_n, written symbolically (a0,a1,…,an)(x,1)n(a_0, a_1, \dots, a_n)(x, 1)^n. §117
bounded above, bounded below, bounded bounded above, bounded below, bounded There is a KK with s≤Ks \le K for all members ss (above), a kk with s≥ks \ge k (below), or both. §80
Cauchy's condensation test Cauchy condensation test For positive decreasing ϕ\phi, ∑ϕ(n)\sum \phi(n) and ∑2nϕ(2n)\sum 2^n\phi(2^n) converge or diverge together. §182
Cauchy's form of the remainder Cauchy remainder Rn=(1−θ)n−1f(n)(a+θh)hn(n−1)!R_n = \frac{(1-\theta)^{n-1}f^{(n)}(a+\theta h)h^n}{(n-1)!}, 0<θ<10 < \theta < 1; it follows from the exact integral remainder by the first mean value theorem. §167
Cauchy's test root test If vn1/n≤r<1v_n^{1/n} \le r < 1 for all large nn then ∑vn\sum v_n converges; usually stated today as lim sup⁡vn1/n<1\limsup v_n^{1/n} < 1. §174
circle of convergence; radius of convergence circle (disc) of convergence; radius of convergence The circle ∣z∣=R|z| = R such that ∑anzn\sum a_n z^n converges absolutely for ∣z∣<R|z| < R and not at all for ∣z∣>R|z| > R; R=1/lim sup⁡∣an∣1/nR = 1/\limsup|a_n|^{1/n}. §200
circle of curvature; radius of curvature; measure of curvature osculating circle; radius of curvature; curvature The circle with contact of order at least 2; radius (1+f′2)3/2/∣f′′∣(1 + f'^2)^{3/2}/|f''|; the curvature is the reciprocal of the radius. §155
circular functions trigonometric functions sin⁡\sin, cos⁡\cos, tan⁡\tan, … and their inverses. §28
circular functions of a complex variable complex trigonometric functions cos⁡\cos, sin⁡\sin, tan⁡\tan, cot⁡\cot, sec⁡\sec, cosec⁡\operatorname{cosec}; for complex ζ\zeta defined by cos⁡ζ=12{exp⁡(iζ)+exp⁡(−iζ)}\cos\zeta = \frac12\{\exp(i\zeta) + \exp(-i\zeta)\}, sin⁡ζ=−12i{exp⁡(iζ)−exp⁡(−iζ)}\sin\zeta = -\frac12 i\{\exp(i\zeta) - \exp(-i\zeta)\}. §238
circular measure (of an angle) radian measure Hardy defines the measure of an angle as twice the area of the sector it cuts from the unit circle, avoiding arc length. §163
Cis θ eiθe^{i\theta} Abbreviation for cos⁡θ+isin⁡θ\cos\theta + i\sin\theta; De Moivre reads (Cis⁡θ)n=Cis⁡nθ(\operatorname{Cis}\theta)^n = \operatorname{Cis} n\theta. §45
class set The collection of all objects with a given property. §52
closed interval (a, b) closed interval [a, b] The set a≤x≤ba \le x \le b. Hardy writes (a,b)(a, b) for both closed and open intervals and says which he means. §122
coaxal system of circles pencil of circles The circles through two points A,BA, B (common point kind, elliptic pencil), or the circles ∣z−b∣=λ∣z−a∣|z - b| = \lambda|z - a| (limiting point kind, hyperbolic pencil, limiting points A,BA, B). §46
common Cartesian geometry the affine (Euclidean) plane R² Geometry in which a point is a pair of reals (x,y)(x, y) and a line is given by ax+by+c=0ax + by + c = 0; it has no infinite elements, and parallel lines do not meet. App. IV
common logarithm, log₁₀ x common (decimal) logarithm, lg x The logarithm to base 1010: log⁡10x=log⁡x/log⁡10=0.4343log⁡x\log_{10} x = \log x / \log 10 = 0.4343\log x. §217
commutative (operations) the operations can be interchanged Two operations are commutative on an object when performing them in either order gives the same result; for two limit operations this is the problem of interchanging limits. App. III
complete equation, completely true equality of sets of values An equation between many-valued expressions in which every value of either side is a value of the other, e.g. Log⁡z1z2=Log⁡z1+Log⁡z2\operatorname{Log} z_1 z_2 = \operatorname{Log} z_1 + \operatorname{Log} z_2. §231
complex function of a real variable complex-valued function on an interval f(x)=ϕ(x)+iψ(x)f(x) = \phi(x) + i\psi(x) with ϕ,ψ\phi, \psi real; its integral is ∫ϕ+i∫ψ\int\phi + i\int\psi. §170
complex number x + yi complex number A pair of real numbers (x,y)(x, y), written x+yix + yi, with (x+yi)(x′+y′i)=(xx′−yy′)+(xy′+yx′)i(x + yi)(x' + y'i) = (xx' - yy') + (xy' + yx')i. §39
components and resultant components and sum of a vector [x,0][x, 0] and [0,y][0, y] are the components of [x,y][x, y], and [x,y][x, y] is their resultant (sum). §36
concave (negative curvature) concave function A function whose graph lies above each of its chords, as log⁡x\log x does (its second derivative −1/x2-1/x^2 is negative). App. I
conditionally convergent (series) conditionally convergent ∑un\sum u_n converges but ∑∣un∣\sum |u_n| diverges. Both its positive and its negative parts then diverge, and rearranging it can change the sum (Riemann). §193
conjugate (z₀) complex conjugate zˉ\bar z The conjugate of x+yix + yi is x−yix - yi, the reflection of zz in the real axis. §44
contact of the nth order contact of order n Two curves y=f(x)y = f(x), y=ϕ(x)y = \phi(x) whose values and first nn derivatives agree at ξ\xi but whose (n+1)(n+1)-th do not; the gap is then of order hn+1h^{n+1}. §155
continuous for x = ξ continuous at ξ ϕ(ξ−0)=ϕ(ξ)=ϕ(ξ+0)\phi(\xi - 0) = \phi(\xi) = \phi(\xi + 0); equivalently, for every ε\varepsilon there is δ\delta with ∣ϕ(x)−ϕ(ξ)∣<ε|\phi(x) - \phi(\xi)| < \varepsilon for ∣x−ξ∣<δ|x - \xi| < \delta. §99
continuous real variable real variable, x ∈ ℝ A symbol xx standing for an unspecified real number; its values are all real numbers. §16
continuous with respect to x and y jointly continuous For every ε\varepsilon some square ∣x−ξ∣,∣y−η∣≤δ|x - \xi|, |y - \eta| \le \delta has ∣ϕ(x,y)−ϕ(ξ,η)∣<ε|\phi(x, y) - \phi(\xi, \eta)| < \varepsilon throughout; stronger than continuity in each variable separately. §108
contour line, contour map level curve, level set The curve f(x,y)=cf(x, y) = c: the projection onto the plane XOYXOY of the section of z=f(x,y)z = f(x, y) by the plane z=cz = c. §33
convergent infinite series; sum of the series convergent series; sum u1+u2+⋯u_1 + u_2 + \cdots converges to ss if its partial sums sn=u1+⋯+uns_n = u_1 + \cdots + u_n tend to ss. §76
convex to the axis of x convex (concave upwards) Every chord of the graph lies above the curve, as for y=x2y = x^2. §23
cross ratio (z₁z₂, z₃z₄) cross ratio (z1−z3)(z2−z4)(z1−z4)(z2−z3)\dfrac{(z_1 - z_3)(z_2 - z_4)}{(z_1 - z_4)(z_2 - z_3)}; real exactly when the four points are concyclic or collinear. §46
curve f(x, y) = 0 implicitly defined curve The set of points satisfying one equation in xx and yy; Hardy’s standard way of describing a plane curve. §32
curvilinear integral line integral, contour integral ∫Cf(z) dz=∫t0t1f(z(t)) z′(t) dt\int_C f(z)\,dz = \int_{t_0}^{t_1} f(z(t))\,z'(t)\,dt along a path z=z(t)z = z(t). §229
cylinder, cone, ruled surface cylinder, cone, ruled surface A cylinder is made of parallel lines through a plane curve, a cone of lines through a vertex, a ruled surface of any moving line (e.g. the hyperboloid of one sheet). §33
D_x differential operator d/dx The operation of forming {ϕ(x+h)−ϕ(x)}/h\{\phi(x + h) - \phi(x)\}/h and letting h→0h \to 0. §116
D_y x dx/dy Hardy's operator notation for the derivative of xx with respect to yy. §213
d'Alembert's test ratio test If vn+1/vn≤r<1v_{n+1}/v_n \le r < 1 for all large nn then ∑vn\sum v_n converges; if vn+1/vn≥1v_{n+1}/v_n \ge 1 for all large nn it diverges. §175
decreases steadily (of the terms of a series) non-increasing un+1≤unu_{n+1} \le u_n for every nn. The hypothesis of Abel's theorem, the integral test and the condensation test. §178
defined for x = 0 0 is in the domain For Hardy, a formula defines ϕ(0)\phi(0) only if it can actually be evaluated at 00: x/xx/x is not defined at 00. §97
degrees of freedom dimension The number of coordinates needed to fix a position: one on a line or curve, two on a plane or surface, three in space. §33
derived function derivative (as a function) The function x↦ϕ′(x)x \mapsto \phi'(x), defined wherever the derivative of ϕ\phi exists. §111
difference taken positively absolute value of the difference ∣ϕ(n)−l∣|\phi(n) - l|, the distance between ϕ(n)\phi(n) and ll. §58
differentiable function of two variables differentiable (Fréchet), with total derivative f(x+h,y+k)−f(x,y)=Ah+Bk+εh+ηkf(x+h, y+k) - f(x, y) = Ah + Bk + \varepsilon h + \eta k with ε,η→0\varepsilon, \eta \to 0; then A=fxA = f_x, B=fyB = f_y. §159
differential coefficient derivative ϕ′(x)=lim⁡h→0{ϕ(x+h)−ϕ(x)}/h\phi'(x) = \lim_{h \to 0} \{\phi(x + h) - \phi(x)\}/h, when this limit exists as a finite number. §111
differential coefficients of all orders smooth, C∞ Having derivatives of every order; not enough for the Taylor series to converge to the function (e−1/x2e^{-1/x^2}). §152
differential dx, dy, dz differential (linear part of the increment) dx=δxdx = \delta x for an independent variable; dz=zx dx+zy dydz = z_x\,dx + z_y\,dy, valid whether or not xx and yy are independent. §160
displacement [x, y] (free) vector in the plane The move carrying (a,b)(a, b) to (a+x,b+y)(a + x, b + y), whatever (a,b)(a, b) is; two displacements are equal when they have the same length, direction and sense. §34
divergent series series diverging to ±∞\pm\infty For Hardy, a series is divergent only when its partial sums tend to +∞+\infty or −∞-\infty. Today “divergent” usually means simply “not convergent”. §76
doubly infinite array (of products) double sequence; Cauchy product The products umvnu_m v_n arranged by rows and columns; summing along the diagonals m+n=km + n = k gives the Cauchy product ∑wk\sum w_k, wk=∑m+n=kumvnw_k = \sum_{m+n=k} u_m v_n. §177
dy/dx Leibniz notation for the derivative A single symbol for lim⁡δy/δx\lim \delta y/\delta x, also written DxyD_x y, y′y' or ϕ′(x)\phi'(x); not a quotient of two numbers dydy and dxdx. §116
e (the number whose logarithm is 1) Euler's number e = 2.71828… The unique xx with log⁡x=1\log x = 1; equal to lim⁡(1+1/n)n\lim (1 + 1/n)^n (§215). §211
equiangular spiral logarithmic spiral A curve r=Cekθr = Ce^{k\theta}, cutting every radius at the same angle; the image under exp⁡\exp of a straight line. §236
Euclidean methods straightedge-and-compass constructions Geometrical constructions with ruler and compasses; they produce exactly the lengths built from rationals by +,−,×,÷+, -, \times, \div and square roots. §13
Euler's constant γ Euler–Mascheroni constant γ γ=lim⁡(1+12+⋯+1n−log⁡n)=0.5772…\gamma = \lim (1 + \frac12 + \dots + \frac1n - \log n) = 0.5772\ldots §218
every algebraical equation has a root fundamental theorem of algebra Every polynomial of degree n≥1n \ge 1 with complex coefficients has a complex root, and hence exactly nn roots counted with multiplicity. App. II
exp x exp x (sum of the exponential series) Hardy's name for ∑xn/n!\sum x^n/n! in the alternative development of §223, kept distinct from the power exe^x until the two are shown to agree; for irrational xx it then serves as the definition of exe^x. §223
exp ζ e^z The one-valued exponential eξ(cos⁡η+isin⁡η)e^\xi(\cos\eta + i\sin\eta). Hardy reserves eζe^\zeta for the many-valued power with base ee, of which exp⁡ζ\exp\zeta is the principal value. §232
explicit algebraical function function expressible by radicals A function built from xx by finitely many rational operations and root extractions, such as x+x+x\sqrt{x} + \sqrt{x + \sqrt{x}}. §26
exponential function e^y exp y For real yy, eye^y is the positive xx with log⁡x=y\log x = y: the inverse of the logarithm. §212
exponential series Maclaurin series of exp 1+x+x22!+x33!+…1 + x + \frac{x^2}{2!} + \frac{x^3}{3!} + \dots, which converges to exe^x for every real xx. §219
exponential theorem exp(x) exp(y) = exp(x + y) for the series The product of the series for exe^x and eye^y is the series for ex+ye^{x+y}: ∑xnn!∑ynn!=∑(x+y)nn!\sum \frac{x^n}{n!}\sum\frac{y^n}{n!} = \sum\frac{(x+y)^n}{n!}. §219
f + λφ = 0 pencil of curves The family of curves f+λϕ=0f + \lambda\phi = 0, all passing through the common points of f=0f = 0 and ϕ=0\phi = 0. §32
f = O(φ) big O ∣f(n)∣≤Kϕ(n)|f(n)| \le K\phi(n) for some constant KK and all sufficiently large nn, where ϕ\phi is positive (Hardy: positive and monotone). O(1)O(1) means bounded. §89
f = o(φ) little o f(n)/ϕ(n)→0f(n)/\phi(n) \to 0, where ϕ\phi is positive (Hardy: positive and monotone). o(1)o(1) means tending to 00. §89
f ~ lφ f∼lϕf \sim l\phi (asymptotic to lϕl\phi) f(n)/ϕ(n)→lf(n)/\phi(n) \to l with l≠0l \ne 0. Modern usage writes f∼gf \sim g only for f/g→1f/g \to 1, which is Hardy’s case f∼lϕf \sim l\phi with g=lϕg = l\phi. §89
f tends to infinity more slowly than φ f ≺ φ, f = o(φ) Both tend to ∞\infty and f(x)/ϕ(x)→0f(x)/\phi(x) \to 0. §210
f(a + h) = … + (hⁿ/n!){f⁽ⁿ⁾(a) + η}, η → 0 Taylor's theorem with Peano (or Young) remainder, o(hⁿ) If f(n)(a)f^{(n)}(a) exists, then f(a+h)=∑k≤nhkk!f(k)(a)+o(hn)f(a+h) = \sum_{k \le n} \frac{h^k}{k!} f^{(k)}(a) + o(h^n) as h→0h \to 0. §151
f/φ → l whenever f′/φ′ → l (§154 (3)) L'Hôpital's rule If f(0)=ϕ(0)=0f(0) = \phi(0) = 0 and f′(x)/ϕ′(x)→lf'(x)/\phi'(x) \to l, then f(x)/ϕ(x)→lf(x)/\phi(x) \to l; sufficient, not necessary. §154
f′ₓ, f′ᵧ; D_x f; ∂f/∂x f_x, ∂f/∂x Notations for the partial derivatives; Hardy uses ∂f/∂x\partial f/\partial x to distinguish Dxf(x,y)D_x f(x, y) from ddxf{x,ψ(x)}\frac{d}{dx} f\{x, \psi(x)\}. §156
factor theorem (for complex numbers) no zero divisors A product of two complex numbers is 00 only if one of the factors is 00. §40
field of definition domain The set of values of the variable for which a function is defined; for a function of nn it is the positive integers. §50
field of variation domain of a variable The set whose members are the values of a variable, e.g. the reals or the positive integers. §16
finite integral (proper) Riemann integral An ordinary integral over a bounded interval of a continuous (or bounded integrable) function, as in Ch. VII. §184
first mean value theorem (for integrals) mean value theorem for integrals For continuous ff, ∫abf=(b−a)f(ξ)\int_a^b f = (b - a)f(\xi) for some ξ\xi in [a,b][a, b]: the average height is a value of ff. §165
for large values of n (for all sufficiently large n) eventually; for all n≥Nn \ge N There is an n0n_0 such that the property holds for every n≥n0n \ge n_0; equivalently, it fails for only finitely many nn. §54
formula of reduction reduction formula A relation expressing an integral with integer index nn in terms of the same integral with smaller indices, e.g. nIn=sin⁡xcos⁡n−1x+(n−1)In−2nI_n = \sin x\cos^{n-1}x + (n-1)I_{n-2} for In=∫cos⁡nx dxI_n = \int\cos^n x\,dx. §142
function (y is a function of x) relation, or multivalued function; a modern function is single-valued with a stated domain A relation under which to some values of xx, at any rate, there correspond one or more values of yy. No formula, full domain or uniqueness is required. §20
function of a function composite function F{f(x)}F\{f(x)\}, today written (F∘f)(x)(F \circ f)(x). §114
function of n (of a positive integral variable) sequence A rule giving a number ϕ(n)\phi(n) for each n=1,2,3,…n = 1, 2, 3, \dots; today written (an)(a_n). §50
function of the complex variable z holomorphic (analytic) function Hardy restricts the phrase to functions defined directly from zz (polynomials, rational and algebraical functions, log, exp, circular functions) and does not give the general definition; today: a function with a complex derivative. §227
function of two variables function of two variables z=f(x,y)z = f(x, y): values of zz correspond to pairs (x,y)(x, y); its graph is a surface in space. §31
functional equation functional equation An equation whose unknown is a function, such as f(xy)=f(x)+f(y)f(xy) = f(x) + f(y); its differentiable solutions are Clog⁡xC\log x. §206
fundamental theorem of the integral calculus fundamental theorem of calculus For continuous ff, ddx∫axf(t) dt=f(x)\frac{d}{dx}\int_a^x f(t)\,dt = f(x). §165
general bilinear transformation Möbius (linear fractional) transformation w=(az+b)/(cz+d)w = (az + b)/(cz + d) with ad−bc≠0ad - bc \ne 0; it maps circles and lines to circles and lines. §46
general mean value theorem; Taylor's theorem Taylor's theorem with Lagrange remainder f(a+h)=∑k<nhkk!f(k)(a)+hnn!f(n)(a+θnh)f(a + h) = \sum_{k<n} \frac{h^k}{k!} f^{(k)}(a) + \frac{h^n}{n!} f^{(n)}(a + \theta_n h), 0<θn<10 < \theta_n < 1. §150
general power a^x a^x = exp(x ln a), a > 0 For a>0a > 0 and real xx, ax=exlog⁡aa^x = e^{x\log a}; it agrees with roots of powers for rational xx and is always positive. §214
general power a^ζ complex power (multivalued) aζ=exp⁡(ζLog⁡a)a^\zeta = \exp(\zeta\operatorname{Log} a); infinitely many values unless ζ\zeta is real and rational. §235
general principle of convergence Cauchy’s criterion; “every Cauchy sequence converges” ϕ(n)\phi(n) converges if and only if for every δ>0\delta > 0 there is n0n_0 with ∣ϕ(n2)−ϕ(n1)∣<δ|\phi(n_2) - \phi(n_1)| < \delta for all n2>n1≥n0n_2 > n_1 \ge n_0. §83
general theorem of the means (G) weighted AM–GM inequality a1q1⋯amqm≤q1a1+⋯+qmama_1^{q_1}\cdots a_m^{q_m} \le q_1 a_1 + \dots + q_m a_m for positive aa's and positive real weights with ∑q=1\sum q = 1; equality only when all the aa's are equal. App. I
geometrical series geometric series 1+r+r2+⋯1 + r + r^2 + \cdots, which converges to 1/(1−r)1/(1-r) exactly when ∣r∣<1|r| < 1. §78
graphical solution of an equation solving by intersecting graphs The roots of f(x)=ϕ(x)f(x) = \phi(x) are the abscissae of the intersections of y=f(x)y = f(x) and y=ϕ(x)y = \phi(x). §30
h → 0 by positive values one-sided limit h → 0⁺ Hardy also writes h→+0h \to +0. The two one-sided limits of the chord slope are the slopes of the right and left half-tangents; a derivative exists when they are finite and equal. §111
imaginary (number) complex number (today: purely imaginary means yiyi) Hardy's established but ‘not happy’ name for complex numbers in general. §43
implicit function implicit function A function y=ϕ(x)y = \phi(x) with f(x,ϕ(x))=0f(x, \phi(x)) = 0 identically, defined by an equation rather than a formula. §109
incomplete symbol (∞) notation defined only in context A symbol with no meaning on its own, only inside phrases that are defined as wholes: ∞\infty in “ϕ(n)→∞\phi(n) \to \infty” or “∑1∞\sum_1^\infty”. App. IV
increases steadily non-decreasing (increasing) ϕ(x2)≥ϕ(x1)\phi(x_2) \ge \phi(x_1) whenever x2>x1x_2 > x_1. “Steadily” has nothing to do with speed. §92
increases steadily (function of n) non-decreasing (monotonically increasing) ϕ(n+1)≥ϕ(n)\phi(n+1) \ge \phi(n) for every nn; equal consecutive values are allowed. With >> it is strictly increasing. §69
increment increment, change The changes δx=h\delta x = h and δy=ϕ(x+h)−ϕ(x)\delta y = \phi(x + h) - \phi(x); the chord slope is δy/δx\delta y / \delta x. §116
inferior number eventual lower bound ξ\xi is inferior for ϕ(n)\phi(n) if ϕ(n)≥ξ\phi(n) \ge \xi for all sufficiently large nn. §82
infinite integral (of the first kind) improper integral over an infinite range ∫a∞ϕ=lim⁡X→∞∫aXϕ\int_a^\infty \phi = \lim_{X\to\infty}\int_a^X \phi when the limit exists; otherwise the integral diverges or oscillates. §184
infinite integral of the second kind improper integral of an unbounded function If ϕ→∞\phi \to \infty as x→a+0x \to a+0, ∫aAϕ=lim⁡ε→+0∫a+εAϕ\int_a^A\phi = \lim_{\varepsilon\to+0}\int_{a+\varepsilon}^A\phi when the limit exists. §187
infinity (of a function) infinite discontinuity; pole A point where ϕ\phi tends to a limit or to ±∞\pm\infty from each side, and to ±∞\pm\infty from at least one side, like 1/x1/x at 00. §100
integral function; indefinite integral antiderivative; integral with variable upper limit A function FF with F′=fF' = f; for continuous ff one is ∫axf(t) dt\int_a^x f(t)\,dt, and then ∫abf=F(b)−F(a)\int_a^b f = F(b) - F(a). §162
integral part [x] floor ⌊x⌋ The greatest integer not exceeding xx; x−[x]x - [x] is the fractional part. §29
integral, integral function antiderivative, primitive (indefinite integral) A function ϕ\phi with ϕ′(x)=ψ(x)\phi'(x) = \psi(x); written ϕ(x)=∫ψ(x) dx\phi(x) = \int \psi(x)\,dx. Today "the integral" usually means the definite integral. §130
integration antidifferentiation (finding an indefinite integral) The operation of forming an integral ϕ\phi of a given ψ\psi, that is, solving ϕ′=ψ\phi' = \psi. §130
integration by rationalisation rationalising substitution Choosing a new variable tt so that xx and the algebraic function yy are both rational in tt, making the integrand a rational function of tt. §136
intermediate number level crossed infinitely often ξ\xi is intermediate if ϕ(n)>ξ\phi(n) > \xi for infinitely many nn and ϕ(n)<ξ\phi(n) < \xi for infinitely many nn. §82
interval (β, γ) closed interval [β, γ] In Ch. I Hardy's (β,γ)(\beta, \gamma) means all xx with β≤x≤γ\beta \le x \le \gamma. §17
inverse function inverse function For x=F(y)x = F(y) with FF continuous and strictly monotone, the unique continuous y=ϕ(x)y = \phi(x) with F(ϕ(x))=xF(\phi(x)) = x. §110
irrational number irrational number A real number whose upper class has no least member, so that it corresponds to no rational. §8
k and k′ = k/(k − 1) conjugate (dual) exponents p, q with 1/p + 1/q = 1 For k>1k > 1, the index k′=k/(k−1)k' = k/(k-1) is the one with 1/k+1/k′=11/k + 1/k' = 1; it pairs with kk in Hölder's inequality. App. I
Lagrange's form of the remainder Lagrange remainder Rn=hnn!f(n)(a+θnh)R_n = \frac{h^n}{n!} f^{(n)}(a + \theta_n h), the error after nn terms of Taylor's series. §152
level curves level curves (of |f|) The curves ∣f(z)∣=k|f(z)| = k in the plane of zz. §240
limit operation limiting process (limit, derivative, integral, infinite sum) Any operation that involves passing to a limit, such as differentiation, integration or summing an infinite series. App. III
limits of indetermination limit superior and limit inferior (lim sup, lim inf) For a bounded ϕ(n)\phi(n): Λ=lim⁡‾ ϕ(n)\Lambda = \overline{\lim}\,\phi(n) is the lower bound of the superior numbers, λ=lim⁡‾ ϕ(n)\lambda = \underline{\lim}\,\phi(n) the upper bound of the inferior numbers. Equivalently, the limits of the upper and lower bounds of the tails. §82
limits of indetermination as x → a lim sup and lim inf of φ at a Λ=lim⁡δ→+0sup⁡0<∣x−a∣<δϕ(x)\Lambda = \lim_{\delta\to+0}\sup_{0<|x-a|<\delta}\phi(x) and λ\lambda likewise with inf⁡\inf; ϕ(x)→l\phi(x) \to l exactly when λ=Λ=l\lambda = \Lambda = l. §96
limits of integration; range of integration limits of integration; interval of integration In ∫abf(x) dx\int_a^b f(x)\,dx, aa and bb are the lower and upper limits and [a,b][a, b] the range of integration; f(x)f(x) is the subject of integration (the integrand). §162
line at infinity; points at infinity the line z = 0 of the projective plane In homogeneous Cartesian geometry, the special line z=0z = 0; its points (x:y:0)(x : y : 0) correspond to directions, and parallel lines meet there. App. IV
locus; the equation represents the locus solution set (curve) of an equation The set of points whose coordinates satisfy a given equation or condition; the equation represents the locus when a point lies on it if and only if its coordinates satisfy the equation. §21
log x (the logarithm) natural logarithm, ln x For x>0x > 0, log⁡x=∫1xdt/t\log x = \int_1^x dt/t. Hardy's log⁡\log is always to base ee; calculators write ln⁡\ln. §205
Log z (general logarithm) the multivalued logarithm, log z ∫Cdz/z\int_C dz/z along a path from 11 to zz avoiding 00; its values are log⁡∣z∣+i(ϕ+2kπ)\log|z| + i(\phi + 2k\pi). Note that many modern books use Log for the principal value instead. §230
log z (principal value) principal branch of the logarithm, Log z The value of Log⁡z\operatorname{Log} z with imaginary part in (−π,π](-\pi, \pi]; the real logarithm when z>0z > 0. §231
log₂ x, log₃ x (iterated logarithms) log log x, log log log x (not base-2 logarithm) In §218 Hardy writes log⁡2x=log⁡log⁡x\log_2 x = \log\log x, log⁡3x=log⁡log⁡log⁡x\log_3 x = \log\log\log x; elsewhere log⁡ax\log_a x means the logarithm to base aa. §218
logarithmic function, log x natural logarithm, ln x Introduced in §131 as the integral of 1/x1/x for x>0x > 0 with log⁡1=0\log 1 = 0; its existence is assumed until Chapter VII, its properties proved in Chapter IX. §131
logarithmic series Mercator series for ln(1 + x) x−x22+x33−…x - \frac{x^2}{2} + \frac{x^3}{3} - \dots, equal to log⁡(1+x)\log(1+x) exactly when −1<x≤1-1 < x \le 1. §220
lower bound (of a set or sequence) greatest lower bound, infimum (inf) The number mm that no member is below, while every number greater than mm exceeds some member. §80
lower class, upper class (L, R) lower set and its complement The two classes of a section. Hardy takes LL to have no greatest member, so a rational number goes into the upper class. §6
Maclaurin's (or Cauchy's) integral test integral test For positive, continuous, decreasing ϕ\phi, ∑ϕ(n)\sum \phi(n) and ∫1∞ϕ(x) dx\int_1^\infty \phi(x)\,dx converge or diverge together. §180
magnification dilation (homothety) about 0 The transformation w=ρzw = \rho z with ρ>0\rho > 0. §46
many-valued function multivalued function (branches, Riemann surface) A rule assigning several, possibly infinitely many, values to each point, such as Log⁡z\operatorname{Log} z or aζa^\zeta. §230
Math. Trip. Cambridge Mathematical Tripos examination Marks an example set in the Tripos examination of the year given. Misc.
maximum (minimum) strict local maximum (minimum) ϕ(ξ)>ϕ(x)\phi(\xi) > \phi(x) (resp. <<) for all x≠ξx \ne \xi in some interval around ξ\xi. Hardy's maxima are strict and local. §123
mean value theorem of the second order Taylor's theorem with n = 2 f(a+h)=f(a)+hf′(a)+12h2f′′(a+θ2h)f(a + h) = f(a) + hf'(a) + \tfrac12 h^2 f''(a + \theta_2 h) with 0<θ2<10 < \theta_2 < 1. §150
mixed quadratic surd element a + √b of a quadratic field a±ba \pm \sqrt b with aa rational and b\sqrt b a pure quadratic surd. §13
mode of subdivision; points of division partition of [a, b] A finite set of points a=x0<x1<⋯<xn=ba = x_0 < x_1 < \dots < x_n = b cutting [a,b][a, b] into pieces; Hardy writes δv\delta_v both for the vv-th piece and for its length. §161
modulus |z| modulus, absolute value ∣x+yi∣=x2+y2|x + yi| = \sqrt{x^2 + y^2}, the distance from 00 to the point zz. §44
modulus |α| absolute value Whichever of α\alpha and −α-\alpha is not negative. §9
multiplication of series Cauchy product The series ∑wn\sum w_n with wn=u0vn+u1vn−1+⋯+unv0w_n = u_0v_n + u_1v_{n-1} + \cdots + u_nv_0; its sum is stst when both series converge absolutely. §202
mutatis mutandis with the necessary changes Latin: “the things to be changed having been changed”. §31
n tends to infinity n→∞n \to \infty nn runs through 1,2,3,…1, 2, 3, \dots, increasing beyond every bound. The symbol ∞\infty means nothing on its own, only inside phrases defined as a whole. §55
Napierian logarithm natural logarithm, ln⁡x\ln x The logarithm to base ee; §75 obtains it as lim⁡n(xn−1)\lim n(\sqrt[n]{x} - 1) before logarithms are defined. §75
neighbourhood of (a, b) neighbourhood Some square centred at (a,b)(a, b) (for one variable, some interval round aa); a condition holds “near” a point if it holds throughout some neighbourhood. §109
no solution fundamentally distinct from log x every (differentiable) solution is a constant multiple of log x Hardy's phrase for uniqueness up to a constant factor, under the assumption that the solution has a derivative. §206
nth roots of unity, ω_n roots of unity, ζn=e2πi/n\zeta_n = e^{2\pi i/n} The nn solutions 1,ωn,…,ωnn−11, \omega_n, \dots, \omega_n^{n-1} of zn=1z^n = 1, where ωn=cos⁡2πn+isin⁡2πn\omega_n = \cos\frac{2\pi}{n} + i\sin\frac{2\pi}{n}. §48
numerically less less in absolute value ∣vn∣<∣un∣|v_n| < |u_n|. §194
numerically less than less in absolute value “rr is numerically less than 11” means ∣r∣<1|r| < 1. §88
of the mth order of smallness φ(x) ~ l xᵐ (vanishes to order m) As x→0x \to 0: ϕ(x)/xm\phi(x)/x^m tends to a limit l≠0l \ne 0. “Of the mmth order of greatness”: xmϕ(x)→l≠0x^m\phi(x) \to l \ne 0. §98
one-valued function (single-valued) function A rule giving exactly one value ϕ(x)\phi(x) for each xx. In Chapter V x\sqrt{x} always means the positive root. §90
one-valued, many-valued single-valued; multivalued (split into branches) A function is one-valued at xx if exactly one yy corresponds to xx, many-valued if several do. Today one splits a many-valued rule such as y2=xy^2 = x into branches ±x\pm\sqrt x. §20
open interval (a, b) open interval (a, b) The set a<x<ba < x < b: the closed interval minus its end points. §122
ordinate function value (sample) f(x) The height f(x)f(x) of the graph at a point; “Simpson’s rule with five ordinates” uses five values of the integrand. §169
ordinate, abscissa y-coordinate, x-coordinate The ordinate of a point is its yy-coordinate, the abscissa its xx-coordinate; "the ordinate x=ξx = \xi" is the vertical line at ξ\xi. §148
oscillates diverges without tending to ±∞\pm\infty A sequence oscillates if it tends neither to a limit, nor to +∞+\infty, nor to −∞-\infty. §62
oscillates finitely, oscillates infinitely bounded divergent; unbounded divergent (not to ±∞\pm\infty) An oscillating sequence oscillates finitely if ∣ϕ(n)∣<K|\phi(n)| < K for some KK and every nn, and infinitely otherwise. §62
oscillating series divergent series (not tending to ±∞\pm\infty) A series whose partial sums neither converge nor tend to ±∞\pm\infty, such as 1−1+1−⋯1 - 1 + 1 - \cdots; finitely or infinitely, as the partial sums are bounded or not. §76
oscillation of φ(x) in (a, b) oscillation (osc) on [a, b] O(a,b)=M(a,b)−m(a,b)O(a, b) = M(a, b) - m(a, b), the difference between the upper and lower bounds of ϕ\phi on the interval. §104
oscillatory discontinuity essential discontinuity (second kind) Any discontinuity that is neither simple nor an infinity, e.g. sin⁡(1/x)\sin(1/x) at 00. §100
partial differential coefficient partial derivative fx=lim⁡h→0{f(x+h,y)−f(x,y)}/hf_x = \lim_{h\to0} \{f(x+h, y) - f(x, y)\}/h: the ordinary derivative with the other variables held fixed. §156
partial integration; integration by parts integration by parts ∫abfϕ′=[fϕ]ab−∫abf′ϕ\int_a^b f\phi' = [f\phi]_a^b - \int_a^b f'\phi for f′,ϕ′f', \phi' continuous. §166
path of integration contour, path The curve, traversed in a given direction, along which a curvilinear integral is taken. §229
point of accumulation accumulation point, limit point ξ\xi is one for SS if every interval (ξ−δ,ξ+δ)(\xi - \delta, \xi + \delta) contains a point of SS other than ξ\xi. §18
point of application starting point (of a bound vector) The original position of the displaced particle; ignored when displacements are compared. §34
point of inflexion inflection point A point where the tangent has contact of order at least 2 with the curve, i.e. f′′(ξ)=0f''(\xi) = 0. §155
polar coordinates (r essentially positive) polar coordinates x=rcos⁡θx = r\cos\theta, y=rsin⁡θy = r\sin\theta with r>0r > 0 and 0≤θ<2π0 \le \theta < 2\pi in Hardy; many modern texts allow negative rr, which can change what a polar equation represents. §22
positive (of the terms of a series) non-negative In Ch. VIII Hardy's 'positive' includes zero: un≥0u_n \ge 0. Only the ratio tests need un>0u_n > 0. §172
power series in z power series A series ∑anzn\sum a_n z^n with fixed coefficients ana_n and complex variable zz. §198
principal value principal branch The distinguished value of a many-valued function (Log, powers, inverse circular functions) singled out by taking the principal amplitude. §231
principal value of the amplitude principal argument Arg z The value of the amplitude with −π<θ≤π-\pi < \theta \le \pi. §44
principal value of the nth root principal nth root For a=ρ(cos⁡ϕ+isin⁡ϕ)a = \rho(\cos\phi + i\sin\phi), −π<ϕ≤π-\pi < \phi \le \pi: the root ρn (cos⁡ϕn+isin⁡ϕn)\sqrt[n]{\rho}\,(\cos\frac{\phi}{n} + i\sin\frac{\phi}{n}). §48
problem of functional interpolation extending a sequence to a function of a real variable Finding a (natural) function F(x)F(x) with F(n)=ϕ(n)F(n) = \phi(n) for n=1,2,3,…n = 1, 2, 3, \dots; never unique, since F(x)+csin⁡πxF(x) + c\sin\pi x also works. §51
proper part proper subset A class included in, but not identical with, another; α<β\alpha < \beta means the lower class of α\alpha is a proper part of that of β\beta. §9
pure quadratic surd irrational square root of a rational ±a\pm\sqrt a with aa a positive rational that is not the square of a rational. §13
R z, I z Re z, Im z The real part xx and the imaginary part yy of z=x+yiz = x + yi. §44
R(x, y), rational function of x and y rational function of two variables A quotient of two polynomials in xx and yy; with yy algebraic in xx it covers integrands such as ax2+2bx+c\sqrt{ax^2+2bx+c} combined rationally with xx. §135
rational function rational function A quotient P(x)/Q(x)P(x)/Q(x) of polynomials; “rational” refers to how xx enters, not to the coefficients. §24
rational number (in lowest terms) fraction p/q in lowest terms p/qp/q with integers pp, qq, q>0q > 0 and no common factor; every rational number has exactly one such form. §1
rational points of a line points with rational coordinate The point ArA_r at signed distance rr from the origin A0A_0, the unit being A0A1A_0A_1. §2
real homogeneous geometry real projective plane with homogeneous coordinates Geometry in which a point is a class (x:y:z)(x : y : z) of proportional nonzero triads of reals and a line is given by ax+by+cz=0ax + by + cz = 0. App. IV
real number real number (as a Dedekind cut) A section of all the rationals in which both classes exist and the lower class has no greatest member. §8
recurring decimal (pure, mixed) repeating decimal A decimal that repeats a fixed block of digits for ever (from the start: pure; after some initial digits: mixed). Exactly the rational numbers have terminating or recurring decimals. §78
repeated factor multiple root A factor (x−α)m(x - \alpha)^m with m≥2m \ge 2; α\alpha is then a root of both ϕ\phi and ϕ′\phi'. §118
scale of infinity asymptotic scale; growth hierarchy A sequence of functions tending to ∞\infty, each tending to ∞\infty more slowly than the next: ⋯≺log⁡log⁡x≺log⁡x≺xα≺ex≺eex≺…\dots \prec \log\log x \prec \log x \prec x^\alpha \prec e^x \prec e^{e^x} \prec \dots §210
second differential coefficient second derivative The derivative of ϕ′\phi', written ϕ′′(x)\phi''(x), Dx2yD_x^2 y or d2y/dx2d^2y/dx^2; similarly the nnth derivative ϕ(n)(x)\phi^{(n)}(x). §121
section (of the rational numbers) Dedekind cut A division of the rationals into non-empty classes LL and RR with every member of LL less than every member of RR. §6
sense (of a displacement) orientation, which way along the line Whether a displacement runs from PP to QQ or from QQ to PP along a given line. §34
series for the inverse tangent Gregory's (Gregory–Leibniz) series x−x33+x55−⋯=arctan⁡xx - \frac{x^3}{3} + \frac{x^5}{5} - \dots = \arctan x (principal value) for −1≤x≤1-1 \le x \le 1; at x=1x = 1 it gives 14π\frac14\pi. §221
series of positive and negative terms real series (no sign restriction) A series whose terms may have either sign. Only when both signs occur infinitely often can it oscillate rather than converge or diverge. §190
set of intervals (satisfying the Heine–Borel conditions) open cover A family of intervals such that every interior point of [a,b][a, b] is inside one of them and each end point is an end of one of them. §106
similar in the same sense directly similar (orientation-preserving similarity) Two triangles similar with corresponding angles equal in sign as well as in size, so one is a turned and scaled copy of the other, not a mirror image. §38
similar surds rationally dependent square roots Two pure quadratic surds that are rational multiples of the same surd, such as 8=22\sqrt8 = 2\sqrt2 and 25/2\sqrt{25/2}. §14
simple discontinuity discontinuity of the first kind (removable or jump) Both one-sided limits exist, but continuity fails: either they are equal and the value is missing or different, or they are unequal. §100
simple series (of numbers) sequence; a listing Numbers arranged as first, second, third, …; Hardy lists all positive rationals this way, so they are countable. §1
Simpson's rule Simpson's rule (a Newton–Cotes formula) ∫abf≈16(b−a){f(a)+4f(a+b2)+f(b)}\int_a^b f \approx \tfrac16(b-a)\{f(a) + 4f(\tfrac{a+b}2) + f(b)\}, with error −12880(b−a)5f(4)(ξ)-\tfrac{1}{2880}(b-a)^5 f^{(4)}(\xi). §169
standard form (of a rational function) lowest terms P/QP/Q with all common factors of PP and QQ removed; removing a factor can change the function at its zeros. §24
standard forms (of the integral calculus) table of basic integrals The integrals read off directly from known derivatives, such as ∫xmdx=xm+1/(m+1)\int x^m dx = x^{m+1}/(m+1), ∫dx/x=log⁡∣x∣\int dx/x = \log|x|, ∫dx/(1+x2)=arctan⁡x\int dx/(1+x^2) = \arctan x. §131
steadily increasing in the stricter sense strictly increasing ϕ(x2)>ϕ(x1)\phi(x_2) > \phi(x_1) whenever x2>x1x_2 > x_1. §95
strictly increasing at a point increasing at a point ϕ(x)<ϕ(x0)\phi(x) < \phi(x_0) for xx just below x0x_0 and ϕ(x)>ϕ(x0)\phi(x) > \phi(x_0) for xx just above; implied by ϕ′(x0)>0\phi'(x_0) > 0, but weaker than increasing on an interval. §122
subject of integration integrand The function under the integral sign. §142
superior number eventual upper bound ξ\xi is superior for ϕ(n)\phi(n) if ϕ(n)≤ξ\phi(n) \le \xi for all sufficiently large nn. §82
surface; curve (in space) surface; space curve In space, the locus of one equation f(x,y,z)=0f(x, y, z) = 0 is a surface, of two equations a curve. §33
Taylor's series; Maclaurin's series Taylor series (at 0: Maclaurin series) ∑nhnn!f(n)(a)\sum_n \frac{h^n}{n!} f^{(n)}(a); it represents f(a+h)f(a + h) exactly when Rn→0R_n \to 0. §152
tends steadily to zero decreases monotonically to 0 (φn↓0\varphi_n \downarrow 0) φn+1≤φn\varphi_{n+1} \le \varphi_n for all nn and φn→0\varphi_n \to 0. §195
tends to +∞ diverges to infinity For every Δ\Delta there is n0(Δ)n_0(\Delta) with ϕ(n)>Δ\phi(n) > \Delta for all n≥n0n \ge n_0. §60
tends to infinity more slowly than any power of x log x = o(x^α) for every α > 0 f(x)/xα→0f(x)/x^\alpha \to 0 for every α>0\alpha > 0 although f(x)→∞f(x) \to \infty; the logarithm has no finite order of greatness. §207
tends to l by positive values (y → +0) right-hand limit (y → 0⁺) The limit taken over 0<y<δ0 < y < \delta only; likewise “by negative values”, y→−0y \to -0, is the left-hand limit. §93
tends to the limit l converges to l For every δ>0\delta > 0 there is n0(δ)n_0(\delta) with ∣ϕ(n)−l∣<δ|\phi(n) - l| < \delta for all n≥n0n \ge n_0. §58
terminated decimal terminating decimal A decimal with finitely many figures after the point, N/10kN/10^k. §1
the continuum is closed ℝ is complete Every section of the real numbers has a real number at the join (Dedekind's theorem). §17
the identity of §196 Abel summation, summation by parts ∑k=0nakφk=∑k=0n−1sk(φk−φk+1)+snφn\sum_{k=0}^n a_k\varphi_k = \sum_{k=0}^{n-1} s_k(\varphi_k - \varphi_{k+1}) + s_n\varphi_n, where sk=a0+⋯+aks_k = a_0 + \cdots + a_k; the discrete form of integration by parts. §196
the number l of the Heine–Borel proof Lebesgue number of a cover A length l>0l > 0 such that every sub-interval of [a,b][a, b] of length ll lies inside one interval of the cover. §106
the positive value (of a power) principal (positive real) value For a>0a > 0, ap/qa^{p/q} denotes the positive qq-th root of apa^p; Hardy says so explicitly because roots are many-valued. §214
the series Σn^(−s) p-series; Riemann zeta function ζ(s) ∑n−s\sum n^{-s} converges exactly when s>1s > 1; its sum as a function of ss is ζ(s)\zeta(s). §181
the sets (a) and (b) are proportional the vectors a and b are parallel (positive multiples) am/bma_m / b_m is the same for every mm. This is the equality condition in Minkowski's inequality; in Hölder's inequality the condition is that amka_m^k and bmk′b_m^{k'} are proportional. App. I
the sums s = Σ m_v δ_v and S = Σ M_v δ_v lower and upper (Darboux) sums Sums of piece lengths times the lower bound mvm_v (respectively upper bound MvM_v) of ff on each piece; every lower sum is at most every upper sum. §161
the terms taken in any order rearrangement (unconditional convergence) A series ∑uσ(n)\sum u_{\sigma(n)} with σ\sigma a bijection of the positive integers: every term used exactly once, each at a finite place. §176
theorem of the mean mean value theorem ϕ(b)−ϕ(a)=(b−a)ϕ′(ξ)\phi(b) - \phi(a) = (b - a)\phi'(\xi) for some ξ\xi between aa and bb. §126
theorem of the total differential coefficient chain rule (several variables) If fx,fyf_x, f_y are continuous, ddtf{ϕ(t),ψ(t)}=fxϕ′(t)+fyψ′(t)\frac{d}{dt} f\{\phi(t), \psi(t)\} = f_x \phi'(t) + f_y \psi'(t). §157
there is inequality unless … equality holds only if … Hardy's way of stating the case of equality in an inequality: the inequality is strict except under the stated condition. App. I
trace the curve; draw the graph sketch the graph Sketch a graph showing its essential features: intersections with the axes, symmetry, asymptotes, where it rises and falls, behaviour for large ∣x∣|x|. §23
transcendental function transcendental function A function that is not algebraical, e.g. sin⁡x\sin x, arccos⁡x\arccos x. §28
transcendental number transcendental number An irrational number that is not algebraic, such as π\pi. §15
turning value, extremal value local extremum A maximum or minimum value of a function in Hardy's (local) sense. §123
two equal roots a double root (root of multiplicity two) The convention that az2+2bz+c=0az^2 + 2bz + c = 0 with b2=acb^2 = ac has the root −b/a-b/a counted twice. §43
ultimately (of constant sign) eventually For all sufficiently large nn. §183
upper bound (of a set or sequence) least upper bound, supremum (sup) The number MM that no member exceeds, while every number less than MM is exceeded by some member. Hardy’s “upper bound” is always the least one; a modern “upper bound” is any KK with s≤Ks \le K for all ss. §80
upper bound of φ(x) in (a, b) supremum (least upper bound) The number M(a,b)M(a, b) that no value of ϕ\phi on the interval exceeds, while every smaller number is exceeded by some value; when attained, the maximum. §103
viz. namely Abbreviation of Latin *videlicet*, “namely”. §20
winding round the origin winding number The number of times a closed path turns round 00, counter-clockwise positive; equal to 12πi∮dz/z\frac{1}{2\pi i}\oint dz/z. §231
x tends to ∞ continuously x → ∞ through all real values xx passes through every real number beyond some point, as opposed to n→∞n \to \infty through the integers. §90
y → ½π − 0 y → (π/2)⁻ (one-sided limit from below) yy tends to 12π\frac12\pi through values less than 12π\frac12\pi; with +0+0, through values greater than it. §224
y = 1 for rational x, 0 for irrational x Dirichlet’s function Defined everywhere and one-valued, but its graph is two “lines” each missing infinitely many points; discontinuous at every point. §29
δ-intervals of ξ neighbourhoods of ξ on which the oscillation is less than ε Intervals round ξ\xi in which the oscillation of ϕ\phi is less than the given tolerance; used in the proof of uniform continuity. §107
Δ(γ), the increment of am Z 2π × the winding number of f(γ) about 0 The total change of a continuously chosen arg⁡f(z)\arg f(z) as zz goes once round the closed path γ\gamma; for a simple closed path described anticlockwise it is 2π2\pi times the number of roots of ff inside γ\gamma. App. II
θ (in the mean value theorem) intermediate point as a fraction of the interval Writing ξ=a+θ(b−a)\xi = a + \theta(b - a) with 0<θ<10 < \theta < 1, so that ϕ(a+h)−ϕ(a)=hϕ′(a+θh)\phi(a + h) - \phi(a) = h\phi'(a + \theta h). §126
θₙ (in Taylor's theorem) the unknown intermediate point ξ = a + θₙh A number in (0,1)(0, 1), depending on nn, hh and ff, locating the point at which f(n)f^{(n)} is evaluated in the remainder. §150
σ = Σ f_v δ_v Riemann sum A sum of piece lengths times values of ff at arbitrary points of the pieces; for continuous ff it tends to ∫abf\int_a^b f as the longest piece tends to 00. §161
φ(a + 0), φ(a − 0) φ(a⁺), φ(a⁻) Abbreviations for lim⁡x→a+0ϕ(x)\lim_{x \to a+0}\phi(x) and lim⁡x→a−0ϕ(x)\lim_{x \to a-0}\phi(x); not values of ϕ\phi at any point. §95
φ(x) tends to l as x tends to a the limit of φ at a is l For every ε>0\varepsilon > 0 there is δ>0\delta > 0 with ∣ϕ(x)−l∣<ε|\phi(x) - l| < \varepsilon whenever 0<∣x−a∣<δ0 < |x - a| < \delta. Hardy writes δ\delta for our ε\varepsilon and ϵ\epsilon for our δ\delta. §94
φₙ(x), the limit of a function of n and x pointwise limit of a sequence of functions ϕ(x)=lim⁡n→∞ϕn(x)\phi(x) = \lim_{n\to\infty} \phi_n(x), defined at those xx where the limit exists. §79