Chapter 1

Introduction

1.1 PDE motivations and context

The aim of this is to introduce and motivate partial differential equations
(PDE). The section also places the scope of studies in APM346 within the
vast universe of mathematics. A partial differential equation (PDE) is an
gather involving partial derivatives. This is not so informative so let’s break
it down a bit.

1.1.1 What is a differential equation?

An ordinary differential equation (ODE) is an equation for a function which
depends on one independent variable which involves the independent variable,
the function, and derivatives of the function:
F (t, u(t), u′ (t), u(2) (t), u(3) (t), . . . , u(m) (t)) = 0.
This is an example of an ODE of order m where m is a highest order of
the derivative in the equation. Solving an equation like this on an interval
t ∈ [0, T ] would mean finding a function t 7→ u(t) ∈ R with the property
that u and its derivatives satisfy this equation for all values t ∈ [0, T ].
The problem can be enlarged by replacing the real-valued u by a vector-
valued one u(x, y) = (u1 (t), u2 (t), . . . , uN (t)). In this case we usually talk
about system of ODEs.
Even in this situation, the challenge is to find functions depending
upon exactly one variable which, together with their derivatives, satisfy the
equation.

1
Chapter 1. Introduction 2

What is a partial derivative?

When you have function that depends upon several variables, you can
differentiate with respect to either variable while holding the other variable
constant. This spawns the idea of partial derivatives. As an example,
consider a function depending upon two real variables taking values in the
reals:

u : Rn → R.

As n = 2 we sometimes visualize a function like this by considering its

graph viewed as a surface in R3 given by the collection of points

{(x, y, z) ∈ R3 : z = u(x, y)}.

We can calculate the derivative with respect to x while holding y fixed.

This leads to ux , also expressed as ∂x u, ∂u
∂x
∂
, and ∂x u. Similarly, we can hold
x fixed and differentiate with respect to y.

What is PDE?
A partial differential equation is an equation for a function which depends
on more than one independent variable which involves the independent
variables, the function, and partial derivatives of the function:

F (x, y, u(x, y), ux (x, y), uy (x, y), uxx (x, y), uxy (x, y), uyx (x, y), uyy (x, y))
= 0.

This is an example of a PDE of order 2. Solving an equation like this

would mean finding a function (x, y) → u(x, y) with the property that u
and its derivatives satisfy this equation for all admissible arguments.
Similarly to ODE case this problem can be enlarged by replacing the real-
valued u by a vector-valued one u(x, y) = (u1 (x, y), u2 (x, y), . . . , uN (x, y).
In this case we usually talk about system of PDEs.

Where PDEs are coming from?

PDEs are often referred as Equations of Mathematical Physics (or Mathe-
matical Physics but it is incorrect as Mathematical Physics is now a separate
field of mathematics) because many of PDEs are coming from different
Chapter 1. Introduction 3

domains of physics (acoustics, optics, elasticity, hydro and aerodynamics,

electromagnetism, quantum mechanics, seismology etc).
However PDEs appear in other fields of science as well (like quantum
chemistry, chemical kinetics); some PDEs are coming from economics and
financial mathematics, or computer science.
Many PDEs are originated in other fields of mathematics.

Examples of PDEs
(Some are actually systems)

Simplest first order equation

ux = 0.

Transport equation

ut + cux = 0.

∂¯ equation

¯ := 1 (fx + ify ) = 0,
∂f
2
(∂¯ is known as “di-bar” or Wirtinger derivatives), or as f = u + iv
(
ux − vy = 0,
uy + vx = 0.

Those who took Complex variables know that those are Cauchy-Riemann
equations.

Laplace’s equation (in 2D)

∆u := uxx + uyy = 0

or similarly in the higher dimensions.

Chapter 1. Introduction 4

Heat equation
ut = k∆u;
(The expression ∆ is called the Laplacian (Laplace operator ) and is defined
as ∂x2 + ∂y2 + ∂z2 on R3 ).

Schrödinger equation (quantum mechanics)

ℏ2

iℏ∂t ψ = − ∆ + V ψ.
2m
Here ψ is a complex-valued function.

Wave equation
utt − c2 ∆u = 0;
sometimes □ := c−2 ∂t2 − ∆ is called (d’Alembertian or d’Alembert operator ).
It appears in elasticity, acoustics, electromagnetism and so on.
One-dimensional wave equation
utt − c2 uxx = 0
often is called string equation and describes f.e. a vibrating string.

Oscillating rod or plate (elasticity) Equation of vibrating rod (with

one spatial variable)
utt + Kuxxxx = 0
or vibrating plate (with two spatial variables)
utt + K∆2 u = 0.

Maxwell equations (electromagnetism) in vacuum


E t − c∇ × H = 0,

H t + c∇ × E = 0,

∇ · E = ∇ · H = 0.

Here E and H are vectors of electric and magnetic intensities, so the first
two lines are actually 6 × 6 system. The third line means two more equations,
and we have 8 × 6 system. Such systems are called overdetermined.
Chapter 1. Introduction 5

Dirac equations (relativistic quantum mechanics)

X
iℏ∂t ψ = βmc2 −


icℏαk ∂xk ψ
1≤k≤3

with Dirac 4 × 4-matrices α1 , α2 , α3 , β. Here ψ is a complex 4-vector, so in

fact we have 4 × 4 system.

Elasticity equations (homogeneous and isotropic)

utt = λ∆u + µ∇(∇ · u).

homogeneous means “the same in all places” (an opposite is called

inhomogeneous) and isotropic means “the same in all directions” (an opposite
is called anisotropic).

Navier-Stokes equation (hydrodynamics for incompressible liquid)

(
ρv t + (v · ∇)ρv − ν∆v = −∇p,
∇ · v = 0,

where ρ is a (constant) density, v is a velocity and p is the pressure; when

viscosity ν = 0 we get Euler equation
(
ρv t + (v · ∇)ρv = −∇p,
∇ · v = 0.

Both of them are 4 × 4 systems.

Yang-Mills equation (elementary particles theory)

∂xj Fjk + [Aj , Fjk ] = 0,
Fjk := ∂xj Ak − ∂xk Aj + [Aj , Ak ],

where Ak are traceless skew-Hermitian matrices. Their matrix elements are

unknown functions. This is a 2nd order system.
Chapter 1. Introduction 6

Einstein equation for general relativity

Gµν + Λgµν = κTµν ,

where Gµν = Rµν − 12 gµν is the Einstein tensor, gµν is the metric tensor
(unknown functions), Rµν is the Ricci curvature tensor, and R is the scalar
curvature, Tµν is the stress–energy tensor, Λ is the cosmological constant
and κ is the Einstein gravitational constant. Components of Ricci curvature
tensor are expressed through the components of the metric tensor, their first
and second derivatives. This is a 2nd order system.

Black-Scholes equation Black-Scholes Equation (Financial mathematics)

is a partial differential equation (PDE) governing the price evolution of a
European call or European put under the Black–Scholes model. Broadly
speaking, the term may refer to a similar PDE that can be derived for a
variety of options, or more generally, derivatives:

∂V 1 ∂ 2V ∂V
+ σ 2 S 2 2 + rS − rV = 0
∂t 2 ∂S ∂S
where V is the price of the option as a function of stock price S and time t,
r is the risk-free interest rate, and σ is the volatility of the stock.
Do not ask me what this means!
Remark 1.1.1. (a) Some of these examples are actually not single PDEs but
the systems of PDEs.

(b) In all these examples there are spatial variables x, y, z and often time
variable t but it is not necessarily so in all PDEs. Equations, not including
time, are called stationary (an opposite is called nonstationary).

(c) Equations could be of different order with respect to different variables

and it is important. However if not specified the order of equation is the
highest order of the derivatives invoked.

(d) In this class we will deal mainly with the wave equation, heat equation
and Laplace equation in their simplest forms.
Chapter 1. Introduction 7

1.2 Initial and boundary value problems

1.2.1 Problems for PDEs
We know that solutions of ODEs typically depend on one or several constants.
For PDEs situation is more complicated. Consider simplest equations

ux = 0, (1.2.1)
vxx = 0 (1.2.2)
wxy = 0 (1.2.3)

with u = u(x, y), v = v(x, y) and w = w(x, y). Equation (1.2.1) could be
treaded as an ODE with respect to x and its solution is a constant but this
is not a genuine constant as it is constant only with respect to x and can
depend on other variables; so u(x, y) = ϕ(y).
Meanwhile, for solution of (1.2.2) we have vx = ϕ(y) where ϕ is an
arbitrary function of one variable and it could be considered as ODE with
respect to x again; then v(x, y) = ϕ(y)x + ψ(y) where ψ(y) is another
arbitrary function of one variable.
Finally, for solution of (1.2.3) we have wy = ϕ(y) where ϕ is an arbitrary
function of one variable and it could beR considered as ODE with respect to
y; then (w − g(y))y = 0 where g(y) = ϕ(y) dy, and therefore w − g(y) =
f (x) =⇒ w(x, y) = f (x) + g(y) where f, g are arbitrary functions of one
variable.
Considering these equations again but assuming that u = u(x, y, z),
v = v(x, y, z) we arrive to u = ϕ(y, z), v = ϕ(y, z)x + ψ(y, z) and w =
f (x, z) + g(y, z) where f, g are arbitrary functions of two variables.
Solutions to PDEs typically depend not on several arbitrary constants
but on one or several arbitrary functions of n − 1 variables. For more
complicated equations this dependence could be much more complicated and
implicit. To select a right solutions we need to use some extra conditions.
The sets of such conditions are called Problems. Typical problems are

(a) IVP (Initial Value Problem): one of variables is interpreted as time t

and conditions are imposed at some moment; f.e. u|t=t0 = u0 ;
(b) BVP (Boundary Value Problem) conditions are imposed on the bound-
ary of the spatial domain Ω: f.e. u|∂Ω = ϕ where ∂Ω is a boundary of
Ω and ϕ is defined on ∂Ω;
Chapter 1. Introduction 8

(c) IVBP (Initial-Boundary Value Problem a.k.a. mixed problem): one of

variables is interpreted as time t and some conditions are imposed at
some moment but other conditions are imposed on the boundary of
the spatial domain.

Remark 1.2.1. In the course of ODEs students usually consider IVP only.
F.e. for the second-order equation like

uxx + a1 ux + a2 u = f (x)

such problem is u|x=x0 = u0 , ux |x=x0 = u1 . However one could consider

BVPs like

(α1 ux + β1 u)|x=x1 = ϕ1 ,
(α2 ux + β2 u)|x=x2 = ϕ2 ,

where solutions are sought on the interval [x1 , x2 ]. Such are covered in
advanced chapters of some of ODE textbooks (but not covered by a typical
ODE class). We will need to cover such problems later in this class.

1.2.2 Notion of “well-posedness”

We want that our PDE (or the system of PDEs) together with all these
conditions satisfied the following requirements:

(a) Solutions must exist for all right-hand expressions (in equations and
conditions)–existence;

(b) Solution must be unique–uniqueness;

(c) Solution must depend on these right-hand expressions continuously,

which means that small changes in the right-hand expressions lead to
small changes in the solution.

Such problems are called well-posed. PDEs are usually studied together
with the problems which are well-posed for these PDEs. Different types of
PDEs admit different problems.
Sometimes however one needs to consider ill-posed problems. In particu-
lar, inverse problems of PDEs are almost always ill-posed.
Chapter 1. Introduction 9

1.3 Classification of equations

1.3.1 Linear and nonlinear equations
Equations of the form
Lu = f (x) (1.3.1)
where Lu is a partial differential expression linear with respect to unknown
function u is called linear equation (or linear system). This equation is
linear homogeneous equation if f = 0 and linear inhomogeneous equation
otherwise. For example,
Lu := a11 uxx + 2a12 uxy + a22 uyy + a1 ux + a2 uy + au = f (x) (1.3.2)
is linear; if all coefficients ajk , aj , a are constant, we call it linear equation
with constant coefficients; otherwise we talk about variable coefficients.
Otherwise equation is called nonlinear. However there is a more subtle
classification of such equations. Equations of the type (1.3.1), where the right-
hand expression f depends on the solution and its lower-order derivatives,
are called semilinear, equations where both coefficients and right-hand
expression depend on the solution and its lower-order derivatives are called
quasilinear. For example,
Lu := a11 (x, y)uxx + 2a12 (x, y)uxy + a22 (x, y)uyy = f (x, y, u, ux , uy )
(1.3.3)
is semilinear, and

Lu := a11 (x, y, u, ux , uy )uxx + 2a12 (x, y, u, ux , uy )uxy +

a22 (x, y, u, ux , uy )uyy = f (x, y, u, ux , uy ) (1.3.4)
is quasilinear, while
F (x, y, u, ux , uy , uxx , uxy , uyx ) = 0 (1.3.5)
is general nonlinear.

1.3.2 Elliptic, hyperbolic and parabolic equations

General Consider second order equation (1.3.2):
X
Lu := aij uxi xj + l.o.t. = f (x) (1.3.6)
1≤i,j≤n
Chapter 1. Introduction 10

where l.o.t. means lower order terms (that is, terms with u and its lower
order derivatives) with aij = aji . Let us change variables x = x(x′ ). Then
the matrix of principal coefficients
 
a11 . . . a1n
 . .
. . ... 

A= ..
 
an1 . . . ann

in the new coordinate system becomes A′ = Q∗ AQ where Q = T ∗ −1 and

∂xi
T = ∂x ′ is a Jacobi matrix. The proof easily follows from the
j i,j=1,...,n
chain rule (Calculus II).
Therefore if the principal coefficients are real and constant, by a linear
change of variables matrix of the principal coefficients could be reduced
to the diagonal form, where diagonal elements could be either 1, or −1 or
0. Multiplying equation by −1 if needed we can assume that there are at
least as many 1 as −1. In particular, for n = 2 the principal part becomes
either uxx + uyy , or uxx − uyy , or uxx and such equations are called elliptic,
hyperbolic, and parabolic respectively (there will be always second derivative
since otherwise it would be the first order equation).
This terminology comes from the curves of the second order conical
sections: if a11 a22 − a212 > 0 equation

a11 ξ 2 + 2a12 ξη + a22 η 2 + a1 ξ + a2 η = c

generically defines an ellipse, if a11 a22 − a212 < 0 this equation generically
defines a hyperbole and if a11 a22 − a212 = 0 it defines a parabole.
Let us consider equations in different dimensions:

2D If we consider only 2-nd order equations with constant real coefficients

then in appropriate coordinates they will look like either

uxx + uyy + l.o.t = f (1.3.7)

or
uxx − uyy + l.o.t. = f, (1.3.8)

where l.o.t. mean lower order terms, and we call such equations elliptic and
hyperbolic respectively.
Chapter 1. Introduction 11

What to do if one of the 2-nd derivatives is missing? We get parabolic

equations
uxx − cuy + l.o.t. = f. (1.3.9)
with c ̸= 0 (we do not consider cuy as a lower order term here) and IVP
u|y=0 = g is well-posed in the direction of y > 0 if c > 0 and in direction
y < 0 if c < 0. We can dismiss c = 0 as not-interesting.
However this classification leaves out very important Schrödinger equa-
tion
uxx + icuy = 0 (1.3.10)
with real c ̸= 0. For it IVP u|y=0 = g is well-posed in both directions y 0 and
y < 0 but it lacks many properties of parabolic equations (like maximum
principle or mollification; still it has interesting properties on its own).

3D Again, if we consider only 2-nd order equations with constant real

coefficients, then in appropriate coordinates they will look like either
uxx + uyy + uzz + l.o.t = f (1.3.11)
or
uxx + uyy − uzz + l.o.t. = f, (1.3.12)
and we call such equations elliptic and hyperbolic respectively.
Also we get parabolic equations like
uxx + uyy − cuz + l.o.t. = f. (1.3.13)
What about
uxx − uyy − cuz + l.o.t. = f ? (1.3.14)
Algebraist-formalist would call it parabolic-hyperbolic but since this equation
exhibits no interesting analytic properties (unless one considers lack of such
properties interesting; in particular, IVP is ill-posed in both directions) it
would be a perversion.
Yes, there will be Schrödinger equation
uxx + uyy + icuz = 0 (1.3.15)
with real c ̸= 0 but uxx − uyy + icuz = 0 would also have IVP u|z=0 = g
well-posed in both directions.
Chapter 1. Introduction 12

4D Here we would get also elliptic

uxx + uyy + uzz + utt + l.o.t. = f, (1.3.16)
hyperbolic
uxx + uyy + uzz − utt + l.o.t. = f, (1.3.17)
but also ultrahyperbolic
uxx + uyy − uzz − utt + l.o.t. = f, (1.3.18)
which exhibits some interesting analytic properties but these equations are
way less important than elliptic, hyperbolic or parabolic.
Parabolic and Schrödinger will be here as well.
Remark 1.3.1. (a) The notions of elliptic, hyperbolic or parabolic equations
are generalized to higher dimensions (trivially) and to higher-order equations,
but most of the randomly written equations do not belong to any of these
types and there is no reason to classify them.
(b) There is no complete classifications of PDEs and cannot be because any
reasonable classification should not be based on how equation looks like but
on the reasonable analytic properties it exhibits (which IVP or BVP are
well-posed etc).

Equations of the variable type To make things even more complicated

there are equations changing types from point to point, f.e. Tricomi equation
uxx + xuyy = 0 (1.3.19)
which is elliptic as x > 0 and hyperbolic as x < 0 and at x = 0 has a
“parabolic degeneration”. It is a toy-model describing stationary transsonic
flow of gas. These equations are called equations of the variable type (a.k.a.
mixed type equations).
Our purpose was not to give exact definitions but to explain a situation.

1.3.3 Scope of this Textbook

- We mostly consider linear PDE problems.
- We mostly consider well-posed problems.
- We mostly consider problems with constant coefficients.
- We do not consider numerical methods.
Chapter 1. Introduction 13

1.4 Origin of some equations

1.4.1 Wave equation
Example 1.4.1. Consider a string as a curve y = u(x, t) (so it’s shape depends
on time t) with a tension T and with a linear density ρ. We assume that
|ux | ≪ 1.
Observe that at point x the part of the string to the left from x pulls it
up with a force −F (x) := −T ux .

x1 x2

Indeed, the force T is directed along the curve and the slope of angle θ
between
p the tangent to the curve and the horizontal line is ux ; so sin(θ) =
ux / 1 + u2x which under our assumption we can replace by ux .
Example 1.4.1 (continued). On the other hand, at point x the part of the
string to the right from x pulls it up with a force F (x) := T ux . Therefore
the total y-component of the force applied to the segment of the string
between J = [x1 , x2 ] equals
Z Z
F (x2 ) − F (x1 ) = ∂x F (x) dx = T uxx dx.
J J
R
According to Newton’s law it must be equal to J ρutt dx where ρdx is
the mass and utt is the acceleration of the infinitesimal segment [x, x + dx]:
Z Z
ρutt dx = T uxx dx.
J J

Since this equality holds for any segment J, the integrands coincide:

ρutt = T uxx . (1.4.1)

Example 1.4.2. Consider a membrane as a surface z = u(x, y, t) with a

tension T and with a surface density ρ. We assume that |ux |, |uy | ≪ 1.
Chapter 1. Introduction 14

Consider a domain D on the plane, its boundary L and a small segment

of the length ds of this boundary. Then the outer domain pulls this segment
up with the force −T n · ∇u ds where n is the inner unit normal to this
segment.
Indeed, the total force is T ds but it pulls along the surface and the slope
of the surface in the direction of n is ≈ n · ∇u.
Therefore the total z-component of force applied to D due to Gauss
formula in dimension 2 (A.1.1) equals
Z ZZ
− T n · ∇u ds = ∇ · (T ∇u) dxdy.
L D
RR
According to Newton’s law it must be equal to D ρutt dxdy where ρdxdy
is the mass and utt is the acceleration of the element of the area:
ZZ ZZ
ρutt dxdy = T ∆u dx
D D

because ∇ · (T ∇u) = T ∇ · ∇u = T ∆u. Since this equality holds for any

domain, the integrands coincide:
ρutt = T ∆u. (1.4.2)
Example 1.4.3. Consider a gas and let v be its velocity and ρ its density.
Then
ρv t + ρ(v · ∇)v = −∇p, (1.4.3)
ρt + ∇ · (ρv) = 0 (1.4.4)
dv
where p is the pressure. Indeed, in (1.4.3) the left-hand expression is ρ
dt
(the mass per unit of the volume multiplied by acceleration) and the right
hand expression is the force of the pressure; no other forces are considered.
problem Further, (1.4.4) is continuity equation which means the mass
conservation since the flow of the mass through the surface element dS in
the direction of the normal n for time dt equals ρn · v.
Remark 1.4.1. According to the chain rule
du ∂u dx
= + (∇u) · = ut + (∇u) · v
dt ∂t dt
is a derivative of u along trajectory which does not coincide with the partial
derivative ut ; v · ∇u is called convection term. However in the linearization
with |v| ≪ 1 it is negligible.
Chapter 1. Introduction 15

Remark 1.4.2. Consider any domain V with a border Σ. The flow of the
gas inwards for time dt equals
ZZ ZZZ
ρv · n dSdt = − ∇ · (ρv) dxdydz × dt
Σ V

again due to Gauss formula (A.1.2). This equals to the increment of the
mass in V
ZZZ ZZZ
∂t ρ dxdydz × dt = ρt dxdydz × dt.
V V

Σ
ρv

V ν

Remark 1.4.3 (continued). Therefore

ZZZ ZZZ
− ∇ · (ρv) dxdydz = ρt dxdydz
V V

Since this equality holds for any domain V we can drop integral and
arrive to
ρt + ∇ · (ρv) = 0. (4)
Example 1.4.3 (continued). We need to add p = p(ρ, T ) where T is the
temperature, but we assume T is constant. Assuming that v, ρ − ρ0 and
their first derivatives are small (ρ0 = const) we arrive instead to
ρ0 v t = −p′ (ρ0 )∇ρ, (1.4.5)
ρt + ρ0 ∇ · v = 0 (1.4.6)
and then applying ∇· to (1.4.5) and ∂t to (1.4.6) we arrive to
ρtt = c2 ∆ρ (1.4.7)
p
with c = p′ (ρ0 ) is the speed of sound.
Chapter 1. Introduction 16

1.4.2 Diffusion equation

Example 1.4.4 (Diffusion Equation). Let u be a concentration of perfume in
the still air. Consider
RRRsome volume V , then the quantity of the perfume in
V at time t equals V
u dxdydz and its increment for time dt equals
ZZZ
ut dxdydz × dt.
V

On the other hand, the law of diffusion states that the flow of perfume
through the surface element dS in the direction of the normal n for time dt
equals −k∇u · n dSdt where k is a diffusion coefficient and therefore the
flow of the perfume into V from outside for time dt equals
ZZ ZZZ
− k∇u · n, dS × dt = ∇ · (k∇u) dxdydz × dt
Σ V

due to Gauss formula (A.1.1.2).

Therefore if there are neither sources nor sinks (negative sources) in V
these two expression must be equal
ZZZ ZZZ
ut dxdydz = ∇ · (k∇u) dxdydz
V V

where we divided by dt. Since these equalities must hold for any volume the
integrands must coincide and we arrive to continuity equation:

ut = ∇ · (k∇u). (1.4.8)

If k is constant we get

ut = k∆u. (1.4.9)

Example 1.4.5 (Heat Equation). Consider heat propagation. Let T be a

temperature.
RRR Then the heat energy contained in the volume V equals
V
Q(T ) dxdydz where Q(T ) is a heat energy density. On the other hand,
the heat flow (the flow of the heat energy) through the surface element dS
in the direction of the normal n for time dt equals −k∇T · n dSdt where k
is a thermoconductivity coefficient. Applying the same arguments as above
we arrive to

Qt = ∇ · (k∇T ). (1.4.10)
Chapter 1. Introduction 17

which we rewrite as

cTt = ∇ · (k∇T ). (1.4.11)

where c = ∂Q
∂T
is a thermocapacity coefficient.
If both c and k are constant we get

cTt = k∆T. (1.4.12)

In the real life c and k depend on T . Further, Q(T ) has jumps at phase
transition temperature. For example to melt an ice to a water (both at 0◦ )
requires a lot of heat and to boil the water to a vapour (both at 100◦ ) also
requires a lot of heat.

1.4.3 Laplace equation

Example 1.4.6. Considering all examples above and assuming that unknown
function does not depend on t (and thus replacing corresponding derivatives
by 0), we arrive to the corresponding stationary equations the simplest of
which is Laplace equation

∆u = 0. (1.4.13)

Example 1.4.7. In the theory of complex variables one studies holomorphic

(analytic) complex-valued function f (z) satisfying a Cauchy-Riemann equa-
tion ∂z̄ f = 0. Here z = x + iy, f = u(x, y) + iv(x, y) with real-valued
u = u(x, y) and v = v(x, y) and ∂z̄ = 12 (∂x + i∂y ); then this equation could
be rewritten as

∂x u − ∂y v = 0, (1.4.14)
∂x v + ∂y u = 0, (1.4.15)

which imply that both u, v satisfy Laplace equation (1.4.13).

Indeed, differentiating the first equation by x and the second by y and
adding we get ∆u = 0, and differentiating the second equation by x and the
first one by y and subructing we get ∆v = 0.