1

PERCOLATION AND RANDOM GRAPHS

Remco van der Hofstad

Department of Mathematics and Computer Science
Eindhoven University of Technology
P.O. Box 513, 5600 MB Eindhoven, The Netherlands.
Abstract
In this section, we define percolation and random graph models, and
survey the features of these models.

1.1 Introduction and notation

In this section, we discuss random networks. In Section 1.2, we study perco-
lation, which is obtained by independently removing vertices or edges from a
graph. Percolation is a model of a porous medium, and is a paradigm model of
statistical physics. Think of the bonds in an infinite graph that are not removed
as indicating whether water can flow through this part of the medium. Then,
the interesting question is whether water can percolate, or, alternatively, whether
there is an infinite connected component of bonds that are kept? As it turns out,
the answer to this question depends sensitively on the fraction of bonds that are
kept. When we keep most bonds, then the kept or occupied bonds form most of
the original graph. In particular, an infinite connected component may exist, and
if this happens, we say that the system percolates. On the other hand, when
most bonds are removed or vacant, then the connected components tend to be
small and insignificant. Thus, percolation admits a phase transition. Despite the
simplicity of the model, the results obtained up to date are far from complete,
and many aspects of percolation, particularly of its critical behavior, are ill un-
derstood. In Section 1.2 we shall discuss the basics of percolation, and highlight
some important open questions. The key challenge in percolation is to uncover
the relation between the percolation critical behavior and the properties of the
underlying graph from which we obtain percolation by removing edges.
In Section 1.3, we discuss random graphs. While in percolation, the random
network considered naturally lives on an infinite graph, in random graph theory
one considers random finite graphs. Thus, all random graphs are obtained by
removing edges from the complete graph, or by adding edges to an empty graph.
An important example of a random graph is obtained by independently remov-
ing bonds from a finite graph, which makes it clear that there is a strong link to
percolation. However, also other mechanisms are possible to generate a random
graph. We shall discuss some of the basics of random graph theory, focussing
2 Percolation and random graphs

on the phase transition of the largest connected component and the distances in
random graphs. The random graph models studied here are inspired by applica-
tions, and we shall highlight real-world networks that these random graphs aim
to model to some extent.
The fields that this contribution covers, percolation and random graph the-
ory, have attracted tremendous attention in the past decades, and enormous
progress has been made. It is impossible to cover all material appearing in the
literature, and we believe that one should not aim to do so. Thus, we have strived
to cover the main results which have been proved, as well as recent results in
which we expect that more progress shall be made in the (near?) future, and we
list open problems which we find of interest. We have tried to give particular
attention to results that are of interest to the stochastic geometry community,
thus giving detailed accounts of the recent progress on two-dimensional perco-
lation and percolation on tesselations, as well as on continuum percolation and
random geometric graphs and its relations to telecommunications. We hope that
this contribution gives an idea of the breadth and depth of these fields, as well
as on the challenges for the future.
We now start by introducing some notation. Let G = (V, E) be a graph, where
V is the vertex set and E ⊆ V × V is the edge set. For percolation, the number
of vertices, denoted by |V|, is naturally infinite, while for random graphs, |V| is
naturally finite. A random network is obtained by a certain rule that determines
which subset of the edges E is occupied, the remaining edges being vacant. Let
v ∈ V, and denote by C(v) the set of vertices which can be reached from v by
occupied edges. More precisely, for u ∈ V, we say that u ←→ v when there exists
a path of occupied edges that connects u and v, and we write
C(v) = {u ∈ V : u ←→ v}. (1.1)
The central question in the study of random networks involves the cluster size
distributions, i.e., for percolation whether there exists an infinite connected com-
ponent, and for random graphs what the distribution is of the largest connected
component.

1.2 Percolation
In this section, we discuss percolation. For more details, see the books (Grimmett,
1999), (Hughes, 1996), (Kesten, 1982) and (Bollobás and Riordan, 2006b). For
an expository account of recent progress with a long list of open problems, we
refer to (Kesten, 2002). There is a nice account of the history of percolation in
(Hughes, 1996, Section 1.1.1), the foundation of percolation as a mathematical
discipline being generally ascribed to (Broadbent and Hammersley, 1957).

Introduction of the model.

In this section, G = (V, E) shall denote an infinite graph. We shall assume that
G = (V, E) is transitive, i.e., the neighborhoods of all points are the same. More
precisely, transitivity means that for every x, y ∈ V, there exists a bijection
 Percolation 3

φ : V → V for which φ(x) = y and {φ(u), φ(v)} ∈ E precisely when {u, v} ∈ E.

Such a bijection φ : V → V is called an automorphism. In particular, transitivity
of a graph G implies that each vertex has the same degree. We shall denote the
degree of G by r. We sometimes assume the weaker condition of quasi-transitivity,
which means that there is a finite set of vertices such that for each vertex v, there
is a graph automorphism of the graph which maps v to one of the vertices in
the finite set. We note that if a graph is quasi-transitive and each vertex has a
bounded degree, then the degrees are uniformly bounded.
Bond percolation is obtained by independently removing each of the bonds or
edges with a fixed probability 1 − p. Thus, each edge is occupied with probability
p, and vacant otherwise, and the edge statuses are independent. We shall restrict
to the setting where the probability that a edge is occupied is fixed. In the
literature, also the setting is studied where E = V × V and, for b ∈ E, the
probability that b is occupied depends on b in a translation invariant way. For
simplicity, we refrain from this generality. The resulting probability measure is
denoted by Pp , and Ep denotes expectation w.r.t. Pp .
We define the percolation function p 7→ θ(p) by

θ(p) = Pp (|C(v)| = ∞), (1.2)

where v ∈ V is an arbitrary vertex. By transitivity, the above probability does

not depend on the choice of v. We shall therefore often write C = C(o) where
o ∈ V is an appropriately chosen origin. When G is quasi-transitive, then θv (p) =
Pp (|C(v)| = ∞) possibly depends on v ∈ V, but these is a finite connection of
vertices V0 such that for every v there exists a v0 ∈ V0 such that θv (p) = θv0 (p).
When θ(p) = 0, then the probability that o is inside an infinite connected
component is 0, so that there will be no infinite connected component a.s. When
θ(p) > 0, on the other hand, then, by ergodicity, the proportion of vertices in
infinite connected components equals θ(p) > 0, and we say that the system
percolates. For quasi-transitive graphs, if θv (p) = 0 for some v, then, in fact,
θv (p) = 0 for all v.
We define the critical value by

pc = pc (G) = inf{p : θ(p) > 0}. (1.3)

For quasi-transitive graphs, it might appear that this definition depends on the
choice of v for which θ(p) = θv (p). However, since θv (p) = 0 for some v implies
that θv (p) = 0 for all v, in fact, pc (G) is independent of the choice of v. The above
critical value is sometimes written as pc (G) = pH (G) in honor of Hammersley,
who defined it in (Hammersley, 1961).
An important question is whether the critical value is non-trivial, i.e., whether
pc ∈ (0, 1). We shall study this question in detail below. When θ(p) > 0, then
a natural question is how many infinite connected components there can be.
Denote this number of infinite connected components by N . We shall now prove
that, for any infinite graph, N ∈ {0, 1, ∞} a.s., and that N is constant a.s.
4 Percolation and random graphs

Indeed, by ergodicity arguments and the fact that N is a translation invariant

random variable, N = k almost surely for some k ∈ {0, 1, 2, . . .}∪{∞}. Moreover,
since N is a.s. constant, it is not changed when we change the status of a finite
number of edges. Indeed, for B ⊆ E, let NB (0) denote the number of infinite
components when all edges in B are declared vacant, and NB (1) the number of
infinite components when all edges in B are declared occupied. Then NB (0) =
NB (1) = k a.s. When k < ∞, NB (0) = NB (1) = k only when B intersects at
most one infinite connected component, and we conclude that the number of
infinite connected components which intersect with B is at most 1 a.s. When
B ↑ V, this number increases to N , and we arrive at the claim that if k < ∞,
then k ≤ 1. This completes the proof that N = k a.s., for some k ∈ {0, 1, ∞}.
Instead of considering bond percolation, one can also study site percolation,
for which we independently and with fixed probability 1−p remove the vertices in
V, and we are only allowed to make use of edges for which both endpoints are kept.
In the literature (see e.g. (Grimmett, 1999)), the main focus has been on bond
percolation, despite the fact that, as we shall show now, site percolation is more
general. Indeed, we shall show that for each bond percolation model indicated by
G, there exists a site percolation model, denoted by Gs , which is equivalent to
bond percolation on G. Indeed, we take Vs to contain the edges in E, and say that
{as , bs } ∈ Es precisely when, in G, the edges to which as and bs are identified
share a common endpoint. In this setting, bond percolation on G becomes site
percolation on Gs , and the connected component of v ∈ V is infinite precisely
when there exists an edge b ∈ E such that the connected component of b ∈ Vs is
infinite. The reverse is not true, i.e., not every site percolation model is equivalent
to a bond percolation model. In this paper, we shall restrict to bond percolation,
having in mind that almost all arguments can be straightforwardly adapted to
the site percolation setting, possibly except for certain duality arguments which
play an important role in two dimensions, and are described in more detail in
Section 1.2.4. Interestingly, already in (Hammersley, 1961) it was shown that the
critical value for bond percolation is never larger than the one for site percolation.
It is useful to note that obviously percolation is monotone in the parameter
p. To make this notion precise, we say that an event E is increasing, when for
every ω ∈ E and η ≥ ω, where η ≥ ω when η(e) ≥ ω(e) for every edge e ∈ E, we
have that also η ∈ E. For example, the event {|C(v)| = ∞} is increasing. Then,
we can couple percolation for all probabilities p simultaneously as follows. We
let {Ue }e∈E be i.i.d. uniform random variables, and note that percolation with
percolation probability p is obtained by declaring an edge e occupied precisely
when Ue ≤ p (see (Hammersley, 1963)). This implies that when p1 ≤ p2 and E
an arbitrary increasing event, we have that

Pp1 (E) ≤ Pp2 (E). (1.4)

In particular, we obtain that θ(p1 ) ≤ θ(p2 ), i.e., p 7→ θ(p) is non-decreasing. We

also say that a random variable X is increasing when {X ≥ x} is increasing for
all x ∈ R.
 Percolation 5

We mention two inequalities that play a profound role in percolation the-

ory, namely the FKG and BK-inequalities. The FKG-inequality in the context
of percolation is called the Harris’ inequality and was first proved in (Harris,
1960). The more general FKG inequality, which, for example, also applies to the
Ising model, was derived in (Fortuin, Kasteleyn and Ginibre, 1971). The Harris’
inequality states that for two increasing events E and F ,

Pp (E ∩ F ) ≥ Pp (E)Pp (F ), (1.5)

the FKG-inequality gives the same conclusion under weaker assumptions on the
measure involved. In words, for increasing events E and F , the occurrence of
E makes the simultaneous occurrence of F more likely. For example, the FKG-
inequality yields that, for every x, y, u, v, ∈ V, we have Pp (x ←→ y, u ←→ v) ≥
Pp (x ←→ y)Pp (u ←→ v). The intuition for the FKG-inequality is that if the
increasing event E holds, then this makes it more likely for edges to be occupied,
and, therefore, it becomes more likely that the increasing event F also holds.
Thus, Pp (F |E) ≥ Pp (F ), which is equivalent to (1.5). See (Häggström, 2007) for
a Markov chain proof of the FKG-inequality.
The BK-inequality gives, in a certain sense, an opposite inequality. We shall
only state it in the case of increasing events, for which it was proved in (van den
Berg and Kesten, 1985). The most general version is proved in (Reimer, 2000).
For K ⊆ E and ω ∈ {0, 1}E , we write ωK (e) = ω(e) for e ∈ K, and ωK (e) = 0
otherwise. Let E and F again be increasing events, and write

E ◦ F = {ω : ∃K ⊆ E such that ωK ∈ E, ωK c ∈ F }. (1.6)

Then, the van den Berg-Kesten (BK) inequality states that

Pp (E ◦ F ) ≤ Pp (E)Pp (F ). (1.7)

For example, the event {x ←→ y} ◦ {u ←→ v} is the event that there are edge-
disjoint occupied paths from x to y and from u to v, and (1.7) implies that
Pp ({x ←→ y} ◦ {u ←→ v}) ≤ Pp (x ←→ y)Pp (u ←→ v). Intuitively, this can be
understood by noting that, if x ←→ y and u ←→ v must occur disjointly, then we
can first fix an occupied path connecting x and y in a certain arbitrary manner,
and remove the occupied edges used in this path. Then {x ←→ y} ◦ {u ←→ v}
occurs when in the configuration with the edges removed, we still have that
u ←→ v. Since we have removed the edges in the occupied path from x to y, this
event now has smaller probability than Pp (u ←→ v).
Many objects we shall study are increasing or decreasing. For example, |C(v)|
is obviously increasing. The number of infinite connected components N is an
example of a random variable that is neither increasing nor decreasing, and we
shall see the complications of this fact later on.
We next discuss an important tool to study probabilities which goes under
the name of Russo’s formula (Russo, 1981). Let E be an increasing event. Then
we say that the bond (u, v) is pivotal for the event E when E occurs when the
6 Percolation and random graphs

status of (u, v) in the (possibly modified) configuration where (u, v) is turned

occupied, while E does not occur in the (possibly modified) configuration where
(u, v) is turned vacant. Thus, the bond (u, v) is essential for the occurrence of the
event E. The set of pivotal bonds for an event is random, as it depends on which
other bonds are occupied and vacant in the configuration. Russo’s Formula states
that for every increasing event E which depends on a finite number of bonds,
d X

Pp (E) = Pp (u, v) is pivotal for E . (1.8)
dp
(u,v)∈E

Russo’s Formula allows us to study how the probability of an event changes as

p varies. The fact that (1.8) is only valid for events that only depend on a finite
number of bonds is a nuisance, and there are many settings in which Russo’s
Formula can be extended to events depending on infinitely many bonds by an
appropriate cutting procedure.
We shall be interested in several key functions that describe the connections
in bond percolation. The susceptibility p 7→ χ(p) is the expected cluster size

χ(p) = Ep [|C(v)|]. (1.9)

Clearly, we have that χ(p) = ∞ for p > pc , since then, with probability θ(p) > 0,
we have that |C(v)| = ∞. Further, p 7→ χ(p) is clearly increasing. Define the
critical value pT = pT (G) by

pT = sup{p : χ(p) < ∞}. (1.10)

The subscript T in pT (G) is in honor of Temperley.

A natural question is whether pT (G) = pc (G), i.e., is χ(p) < ∞ for every p <
pc ? The latter indeed turns out to be true by the celebrated results independently
proved by (Menshikov, 1986) and (Aizenman and Barsky, 1987), as we shall
discuss in more detail below.
For p ∈ [0, 1], let
χf (p) = Ep [|C(v)|1l{|C(v)|<∞} ] (1.11)
denote the mean finite cluster size. Clearly, for p < pT , we have that χf (p) =
χ(p), but for p > pT , this may not be true. We define the two-point function
τp : V × V → [0, 1] by

τpf (x, y) = Pp (x ←→ y, |C(x)| < ∞). (1.12)

On Zd , when the model is translation invariant, we have that τpf (x, y) = τpf (y −
x, 0) ≡ τpf (y − x). Also for transitive G, the two-point function is characterized
by τpf (x) = τpf (o, x), where o is an appropriately chosen origin in the model. In
terms of τpf (x), we can identify χf (p) as
X
χf (p) = τpf (x). (1.13)
x∈V
 Percolation 7

We shall also be interested in the mean number of clusters per vertex κ(p), which
is defined as
κ(p) = Ep [1/|C(v)|]. (1.14)

The significance of κ(p) is that it measures the average number of connected

components in large volumes. Indeed, let B(n) = {x ∈ V : d(o, x) ≤ n} de-
note a ball of radius n around o, where d(x, y) denotes an appropriately chosen
distance function on G. Then, with Cn the number of different connected compo-
nents obtained when making every edges not entirely in B(n) vacant, and when
∂B(n) = o(B(n)), Cn /|B(n)| → κ(p) a.s.
An important measure of the spatial extent of clusters is the correlation length
ξ(p) defined by
s
1 X 2 f
ξ(p) = f
|x| τp (x), (1.15)
χ (p)
x∈V

where we write |x| = d(o, x) for the distance of x to an appropriately chosen origin
o ∈ V. For many graphs G, there are several ways of defining the correlation
length, many of them being equivalent in the sense that they are bounded above
and below by finite and positive constants times ξ(p) defined in (1.15). The
correlation length measures the dependence between finite clusters at a given
distance. If d(x, y)  ξ(p) and, for p > pc , x and y are in different finite clusters,
then we can think of C(x) and C(y) as being close to independent, while if
d(x, y)  ξ(p), then C(x) and C(y) are quite dependent.

1.2.1 Critical behavior.

The behavior of percolation models is most interesting and richest for p values
which are close to the critical value. Clearly, the precise value of pc (G) depends
sensitively on the nature of the underlying graph G. By drawing an analogy to
physical systems, physicists predict that the behavior of percolative systems close
to criticality is rather insensitive to the precise details of the model, and it is
only characterized by the macroscopic behavior. Thus, percolation is expected to
behave in a universal manner. For example, it is predicted that the critical nature
of finite-range percolation systems on Zd , under suitable symmetry conditions, is
similar in the sense that the nature of the critical behavior are similar. While this
prediction is far from rigorous, it does offer us a way of summarizing percolation
models by only looking at their simplest examples. It is one of the key challenges
in percolation of making the paradigm of universality rigorous.
We shall now make the notion of universality more tangible, by discussing
critical exponents. The critical nature of many physical systems is believed to be
characterized by the validity of power laws, the exponent of which is a robust
or universal measure of the underlying critical behavior. We start by giving an
example of a critical exponent. It is predicted that

θ(p) ∼ (p − pc )β , as p ↓ pc , (1.16)
8 Percolation and random graphs

for some β > 0. The value of β is expected to be different for different G, but
(1.16) remains valid. The symbol ∼ in (1.16) can have several meanings, which
we now elaborate on. We say that the critical exponent β exists in the logarithmic
form if
log θ(p)
lim = β, (1.17)
p↓pc log (p − pc )

while β exists in the bounded-ratios form if there exist 0 < c1 < c2 < ∞ such
that, uniformly for p ≥ pc ,

c1 (p − pc )β ≤ θ(p) ≤ c2 (p − pc )β , (1.18)

Finally, we say that β exists in the asymptotic form if, as p ↓ pc , there exists a
c > 0 such that
θ(p) = c(p − pc )β (1 + o(1)). (1.19)
The existence of a critical exponent is a priori unclear, and needs a mathematical
proof. Unfortunately, in general such a proof is missing, and we can only give
proofs of the existence in special cases, which we shall discuss below in quite
some detail. Indeed, the existence of the critical exponent β > 0 is stronger than
continuity of p 7→ θ(p), which is unknown in general, and is arguably the holy
grail of percolation theory. Indeed, p 7→ θ(p) is clearly continuous on [0, pc ), and
it is also continuous (and even infinitely differentiable) on (pc , 1] by the results
of (van den Berg and Keane, 1984) (for infinite differentiability of p 7→ θ(p) for
p ∈ (pc , 1], see (Russo, 1978)). Thus, continuity of p 7→ θ(p) is equivalent to the
statement that θ(pc (G)) = 0.
We now introduce several more critical exponents. The critical exponent γ
for the expected cluster size is given by

χf (p) ∼ |p − pc |−γ , p → pc . (1.20)

More precisely, we can think of (1.20) as defining the critical exponents γ, γ 0 > 0
defined by
0
χ(p) ∼ (p − pc )−γ , p ↑ pc , χf (p) ∼ (p − pc )−γ , p ↓ pc , (1.21)

with the predicted equality γ = γ 0 . For (1.20) and (1.21), we are implicitly
assuming that pT (G) = pc (G), this equality shall be discussed in more detail
below.
Further, ν, ν 0 are defined by
0
ξ(p) ∼ (p − pc )−ν , p ↑ pc , ξ(p) ∼ (p − pc )−ν , p ↓ pc , (1.22)

again with the prediction that ν = ν 0 . The exponent −1 ≤ α < 0 is defined by

the blow up of the third derivative of p 7→ κ(p) at p = pc , i.e.,

κ000 (p) ∼ |p − pc |−1−α , (1.23)

 Percolation 9

while the gap exponent ∆ > 0 is defined by, for k ≥ 1,

 
Ep |C|k+1 1l{|C|<∞}
  ∼ |p − pc |−∆ . (1.24)
Ep |C|k+1 1l{|C|<∞}

Also α and ∆ can be defined, similarly to (1.21), as an exponent for p ↑ pc and

one for p ↓ pc , the values being equal, and we shall always use the convention
that α and ∆ denote the p ↑ pc versions, while α0 and ∆ denote the p ↓ pc
versions.
As mentioned before, it is highly unclear that these critical exponents are
well-defined, and that the value of ∆ > 0 does not depend on k. However, there
are good physical reasons why these exponents are defined as they are. The
exponents β, γ, ν, α, ∆ can be thought of as approach exponents which measure
the blow-up of various aspects of the cluster size as p approaches the critical
value p = pc (G).
We finally define two critical exponents at criticality. The exponent δ ≥ 1
measures the power-law exponent of the critical cluster tail, i.e.,

Ppc (|C| ≥ n) ∼ n−1/δ , n → ∞, (1.25)

the assumption that δ ≥ 1 following from the prediction that χ(pc ) = ∞. Further,
we define the exponent ρ > 0 by

Ppc (o ←→ ∂B(n)) ∼ n−1/ρ , n → ∞. (1.26)

Finally, η is defined by
X
τpc (x) ∼ n2−η ,
 
Epc |C ∩ B(n)| = n → ∞, (1.27)
x:|x|≤n

where we recall that |x| = d(o, x) is the distance of x to o in G.

The above definitions give rise to eight critical values that each describe a
certain aspect of the (near-)critical behavior of the percolation system. Several
relations between these critical exponents are predicted by (non-rigorous) physics
argument, and go under the name of scaling relations. These scaling relations
assert that

γ + 2β = β(δ + 1) = 2 − α = ∆ + β, γ = ν(2 − η). (1.28)

The validity of the scaling relations in (1.28) is widely accepted, but few proofs
exist, in particular since the existence of the critical exponents is in general un-
known. We can intuitively understand the scaling relations in (1.28) by assum-
ing the existence of certainscaling functions, which describe certain percolation
10 Percolation and random graphs

quantities close to criticality. An example is to assume that there exist scaling

functions f+ and f− , as well as some exponents σ, τ > 0, such that
(
n−σ f− (n/ξ(p)τ ) for p ≤ pc
Pp (|C| = n) ∼ (1.29)
n−σ f+ (n/ξ(p)τ ) for p ≥ pc ,
for some sufficiently smooth and integrable functions f+ and f− . When working
out the consequences of (1.29), we can see that it must imply the first three
scaling relations in (1.28), and when also assuming that a scaling function exists
for x 7→ τpf (x), the relation γ = ν(2−η) follows. The existence of scaling functions
is, as far as we know, unknown except for percolation on a tree. Nice discussions of
scaling functions can be found in (Grimmett, 1999, Sections 9.1-9.2) or (Hughes,
1996, Section 4.2). The most complete reference to results on critical exponents
until 1996 is (Hughes, 1996, Chapter 4).
Percolation is a paradigm model in statistical physics. A central notion in
this field is universality, a notion strongly rooted in theoretical physics which
states that the near-critical behavior of a system does is rather insensitive to
the microscopic details of the system. An example in the setting of percolation is
that any finite-range percolation model on a vertex space V has the same critical
behavior. In particular, it has the same critical exponents. While universality
is quite plausible when describing real physical systems from the viewpoint of
statistical physics, and while universality is a very useful notion since it allows
us to study only the simplest finite-range model available, there are very few
examples where universality can be rigorously proved. We shall discuss a few
universality results below.
So far, we have discussed percolation in full generality. We shall now treat
examples of percolation models. In Section 1.2.2, we shall describe percolation
on regular trees, in Section 1.2.3 we discuss percolation on Zd for general d, in
Section 1.2.4 we specialize to the two-dimensional setting, in Section 1.2.5, we
study the high-dimensional case for which d > 6 and in Section 1.2.6, we study
oriented or directed percolation. Finally, in Section 1.2.7, we study the case of
percolation on non-amenable graphs and we close this section in Section 1.2.8
by discussing continuum percolation and its applications.
1.2.2 Percolation on the regular tree.
In this section, we study percolation on the regular tree. Let Tr denote the r-
regular tree of degree r. The advantage of trees is that they do not contain cycles,
which make explicit computations possible. In order to compute the critical
exponents ν and η, we first identify
p
|x| = h(x), (1.30)
where h(x) is the height of the vertex x in the tree, i.e., the length of the shortest
path linking o and x, so that (1.30) is the Euclidean distance in the tree. We
shall first prove that the critical exponents for percolation on a regular tree exist
and identify their values in the following theorem:
 Percolation 11

Theorem 1.1. (Critical behavior on the r-regular tree) On the r-regular

tree Tr , pc = pT = 1/(r − 1), and β = γ = γ 0 = 1, δ = ∆ = ∆0 = ρ = 2,
ν = ν 0 = 1/2 and α = α0 = −1 in the asymptotic sense.
Proof We shall make substantial use of the fact that percolation on a tree can
be described in terms of branching processes. Recall that o is the root of the tree.
For x 6= o, we write CBP (x) for the forward cluster of x, i.e., the vertices y ∈ Tr
which are connected to x and for which the unique path from x to y only moves
away from the root o. Then, clearly,
X
|C(o)| = 1 + I(o,e) |CBP (e)|, (1.31)
e∼o

where the sum is over all neighbors e of o, I(o,e) is the indicator that the edge (o, e)
is occupied, and CBP (e) is the forward cluster of e. The random vector {I(o,e) }e∼o
forms a collection of r independent Bernoulli random variables with success
probability p, and {|CBP (e)|}e∼o is an i.i.d. sequence independent of {I(o,e) }e∼o .
Equation (1.31) allows us to deduce all information concerning |C(o)| from the
information of |CBP (e)|. Also, for each e, |CBP (x)| satisfies the formula
X
|CBP (x)| = 1 + I(x,e) |CBP (e)|, (1.32)
e∼x:h(e)>h(x)

where h(x) is the height of x in Tr , and {|CBP (e)|}e∼x:h(e)>h(x) is a set of r − 1

independent copies of |CBP (x)|. Thus, |CBP (x)| is the total progeny of a branching
process. As a result,
1
χBP (p) = E[|CBP (x)|] = . (1.33)
1 − (r − 1)p
From (1.31), we then obtain that, for p < 1/(r − 1),
rp 1+p
χ(p) = 1 + rpχBP (p) = 1 + = , (1.34)
1 − (r − 1)p 1 − (r − 1)p
while, for p > 1/(r − 1), χ(p) = ∞. In particular, pT = 1/(r − 1), and γ = 1 in
the asymptotic sense. The computation of χ(p) can also be performed without
the use of (1.32), by noting that, for p ∈ [0, 1],
τp (x) = ph(x) , (1.35)
and the fact that, for n ≥ 1, there are r(r − 1)n−1 vertices in Tr at height n, so
that, for p < 1/(r − 1),
∞
X r + 1 − (r − 1)p
χ(p) = 1 + r(r − 1)n−1 pn = . (1.36)
n=1
1 − (r − 1)p

However, for related results for percolation on a tree, the connection to branching
processes in (1.32) is vital. We defer the proof that γ 0 = 1 to later.
12 Percolation and random graphs

Let θBP (p) = Pp (|CBP (x)| = ∞). Then θBP (p) is the survival probability of a
branching process with a binomial offspring distribution with parameters r − 1
and p. Thus, θBP (p) satisfies the equation

r−1
r−1
θBP (p) = 1 − 1 − p + p(1 − θBP (p)) = 1 − 1 − pθBP (p) . (1.37)

To compute θBP (p), it is more convenient to work with the extinction probability
ζBP (p) = 1 − θBP (p), which is the probability that the branching process dies out.
The extinction probability ζBP (p) satisfies

r−1
ζBP (p) = 1 − p + pζBP (p) . (1.38)

This equation can be solved explicitly when r = 2, when the unique solution is
θBP (p) = 0 for p ∈ [0, 1) and θBP (1) = 1, so that pc = 1. When r = 3, we obtain
that
p2 ζBP (p)2 + (2p(1 − p) − 1)ζBP (p) + (1 − p)2 = 0, (1.39)
so that
1 − 2p(1 − p) ± |2p − 1|
ζBP (p) = . (1.40)
2p2
Since ζBP (0) = 1, ζBP (1) = 0, we must have that

1 − 2p(1 − p) − |2p − 1|
ζBP (p) = , (1.41)
2p2

so that ζBP (p) = 1 for p ∈ [0, 1/2], while, for p ∈ [1/2, 1],

1 − 2p(1 − p) + (1 − 2p) 2 − 4p + 2p2 1 − p
2

ζBP (p) = = = . (1.42)
2p2 2p2 p

As a result, we have the explicit form θBP (p) = 0 for p ∈ [0, 1/2] and

1 − p
2 2p − 1
θBP (p) = 1 − = , (1.43)
p p2

for p ∈ [1/2, 1], so that pc = 1/2. In particular, p 7→ θBP (p) is continuous, and,
for p ↓ pc ,
θBP (p) = 8(p − pc )(1 + o(1)). (1.44)
It is not hard to see that (1.44) together with (1.31) implies that

θ(p) = 12(p − pc )(1 + o(1)). (1.45)

Thus, for r = 3, the percolation function is continuous, and β = 1 in the asymp-

totic sense. It is not hard to extend the asymptotic analysis in (1.44)–(1.45) to
r ≥ 4, for which pc (Tr ) = p|sssT (Tr ) = 1/(r − 1), but we omit the details.
 Percolation 13

In order to study χfBP (p) = Ep [|CBP (x)|1l{|CBP (x)|<∞} ] for p > pc = 1/(r − 1),
we make use of the fact that

χfBP (p) = (1 − θBP (p))Ep |CBP (x)| |CBP (x)| < ∞ ,

 
(1.46)

and the conditional law of percolation on the tree given that |CBP (x)| < ∞ is
percolation on a tree with p replaced by the dual percolation probability pd given
by
pd = p(1 − θBP (p)). (1.47)

The crucial fact is that pd < pc (Tr ), which follows from the equality 1 − θBP (p) =
ζBP (p), (1.38) and the fact that

r−1
r−2
(r − 1)pζBP (p) = (r − 1)p 1 − p + pζBP (p) < (r − 1)p 1 − p + pζBP (p)
d
= (1 − p + ps)r−1 , (1.48)
ds s=ζBP (p)

which, since ζBP (p) is the smallest solution of (1 − p + ps)r−1 = s, implies that
the derivative of (1 − p + ps)r−1 at s = ζBP (p) is strictly bounded above by 1 for
p > pc (Tr ). Thus, by conditioning a supercritical cluster in percolation on a tree
to die out, we obtain a subcritical cluster at an appropriate subcritical pd which
is related to the original percolation parameter. This fact is sometimes called the
discrete duality principle.
We conclude that
1
χfBP (p) = (1 − θBP (p))Ep |C| |C| < ∞ = (1 − θBP (p))
 
.
1 − (r − 1)p(1 − θBP (p))
(1.49)
Using that β = 1 in the asymptotic sense then gives that

Cγ 0 + o(1)
χfBP (p) = . (1.50)
p − pc

By (1.31), this can easily be transferred to χf (p), so that γ 0 = 1 in the asymptotic

sense.
To compute ν, we note that, by (1.30) and (1.35), we have that, for p <
1/(r − 1),
∞
1 X 1 r
ξ(p)2 = nr(r − 1)n−1 pn = f , (1.51)
χfBP (p) n=1 χBP (p) [1 − (r − 1)p]2

so that ν = 1/2 in the asymptotic sense by (1.50). We refrain from proving that
ν 0 = ν, and only remark that this follows again by using the duality used in
(1.47).
14 Percolation and random graphs

To compute η, we note that

2
n
 X
(r − 1)m pm 2

Epc |CBP (x) ∩ B(n)| = c =n , (1.52)
m=1

so that η = 0. We can compute δ by using the random walk hitting time theorem,
see (Grimmett, 1999, Prop. 10.22) and (van der Hofstad and Keane, 2007), where
a very simple proof is given applying to general branching processes. This result
yields that
Pp (|CBP (x)| = k) = P(X1 + . . . + Xk = k − 1), (1.53)
where {Xi }∞
i=1 is an i.i.d. sequence of binomial random variables with parameter
r − 1 and success probability p. Thus,
 
k(r − 1) k−1
Pp (|CBP (x)| = k) = p (1 − p)k(r−1)−(k−1) . (1.54)
k−1
To prove that δ = 2, we note that for p = pc = 1/(r−1), by a local limit theorem,
we obtain
1
Ppc (|CBP (x)| = k) = (Cδ + o(1)) √ , (1.55)
k3
so that, by summing over k ≥ n, we obtain δ = 2 in an asymptotic sense. We
can compute ρ by noting that
θn = Ppc (∃y ∈ CBP (x) such that h(y) = n + h(x)) (1.56)
satisfies the recursion relation
1 − θn = (1 − pc θn−1 )r−1 . (1.57)
pc (Tr ) = 1/(r − 1) implies that
It is not hard to see that (1.57) together with p
θn = (Cρ + o(1))/n.√Thus ρ = 1/2, since |x| = h(x) and Ppc (o ←→ ∂B(n)) =
θ√n ∼ (Cρ + o(1))/ n.
2
The computation of the key objects for percolation on a tree is feasible due to the
close relationship to branching processes, a topic which has attracted substan-
tial interest in the probability community. See (Athreya and Ney, 1972), (Harris,
1963) and (Jagers, 1975) for detailed discussions about branching processes. As
it turns out, the computations on a tree also have direct consequences for per-
colation on general graphs, with and without loops. We shall now discuss some
of these consequences, the first being that pc > 0 on any graph with bounded
degree:
Theorem 1.2. (Percolation threshold is strictly positive) Let G = (V, E)
be a graph for which the degree of every vertex is bounded by r. Then, for every
x ∈ V, and p < 1/(r − 1),
Ep [|C(x)|] < ∞. (1.58)
In particular, for transitive graphs G with degree equal to r, pc (G) ≥ pT (G) ≥
1/(r − 1).
 Percolation 15

Proof Let ω = (ω1 , . . . , ωn ) be a nearest-neighbor path in G. We call ω self-

avoiding when ωi 6= ωj for all 0 ≤ i < j ≤ n. We let cn (x, y) denote the number of
n-step nearest-neighbor self-avoiding walks starting at x with endpoint y. Then,
we note that if x ←→ y, then there must be a self-avoiding walk path consisting
of occupied bonds. As a result,
∞
X
τp (x, y) ≤ pn cn (x, y). (1.59)
n=0

Therefore,
X ∞
XX ∞
X
Ep [|C(x)|] = τp (x, y) ≤ pn cn (x, y) = pn cn (x), (1.60)
y∈V y∈V n=0 n=0

where cn (x) denotes the number of n-step self-avoiding walk paths starting at
x. Since the degree of G is bounded by r, we have that, uniformly in x ∈ V,
cn (x) ≤ r(r − 1)n−1 . (1.61)
Thus, we arrive at
∞
X r 1
Ep [|C(x)|] ≤ pn r(r − 1)n−1 = , (1.62)
n=0
r − 1 1 − p(r − 1)

so that Ep [|C(x)|] < ∞ for p < 1/(r − 1). This completes the proof. 2
For percolation on a tree, it is not hard to see that the number of infinite clusters
N equals infinity a.s. To see this, fix a root o of the tree Tr . For any vertex v
unequal to o, let u be the unique vertex in Tr that is closer to the root. Then, the
probability that the bond (u, v) is vacant is strictly positive. If p > pc = 1/(r−1),
then with probability θBP (p) > 0, the vertex v will lie in an infinite component.
Thus, with strictly positive probability, there will be at least two bonds (o, e1 )
and (o, e2 ) that are vacant, and of which e1 and e2 lie in an infinite cluster. Thus,
Pp (N ≥ 2) > 0, so that, since N ∈ {0, 1, ∞} a.s., we must have that N = ∞ a.s.
We conclude that percolation on a tree is very well understood, that all of its
critical exponents exist in an asymptotic sense and can be explicitly identified.
Moreover, for p ≤ pc (Tr ) = 1/(r − 1), there is no infinite connected component,
while, for p ∈ (pc , 1), there are infinitely many of them. For p = 1, there is
a unique infinite cluster. We now proceed to study percolation on graphs with
loops, starting with the paradigm example of G = Zd .

1.2.3 Percolation on Zd .
In this section, we study percolation on Zd . We start by proving that the phase
transition is non-trivial:
Theorem 1.3. (Phase transition on Zd is non-trivial) For nearest-neighbor
1
percolation on Zd with d ≥ 2, pc (Zd ) ∈ [ 2d−1 , 1).
16 Percolation and random graphs

Proof The lower bound on pc (Zd ) is immediate from Theorem 1.2, so that we
are left to prove that pc (Zd ) < 1. For this, we must prove that, on Zd , θ(p) > 0
for p sufficiently close to 1. We first show that it suffices to prove this for d = 2.
For this, we fix p ∈ (0, 1) and note that |C| = |C(o)| for percolation on Zd
with parameter p is stochastically larger than |C| = |C(o)| for percolation on Z2
with parameter p, since by only using the nearest-neighbor edges in Z2 × {~0d−2 },
where ~0d−2 ∈ Zd−2 denotes the origin in Zd−2 , we obtain a cluster which is not
larger than the one using all nearest-neighbor edges in Zd . Thus, it suffices to
prove that θ(p) > 0 for p sufficiently close to 1 for d = 2.
We shall make use of duality, a notion which is crucial for d = 2 (see also
Section 1.2.4 below). For a set of vertices A, let the boundary edges of A be
∂e A = {{x, y} : x ∈ A, y 6∈ A}. Clearly, for C(o) the cluster of the origin, we
have that all edges in ∂e C(o) are vacant. Define the dual lattice L∗ consist of the
vertices Vd = {x + (1/2, 1/2) : x ∈ Z2 } and an edge exists between x + (1/2, 1/2)
and y + (1/2, 1/2) if and only if x and y are nearest-neighbors in Z2 . We note
that each edge in L∗ intersects precisely one nearest-neighbor edge in Zd . We
perform percolation on L∗ by identifying the status of the edge e ∈ L∗ by the
status of the unique edge it intersects in Zd . Then, the set of vacant edges ∂e C
is identified with a set of vacant edges in L. We next study the structure of the
set of edges.
We call a path ω = (ω0 , . . . , ωn ) a self-avoiding polygon of length n on L
when ωi ∈ Vd , ω0 = ωn and when ωi 6= ωj for every 0 ≤ i < j ≤ n for which
i, j 6= 0, n. A self-avoiding polygon separates L∗ into the outside and the inside of
the polygon, and we say that a self-avoiding polygon surrounds a point x ∈ Z2
when the point lies on the inside of the polygon. For Z2 , we observe that if
|C(o)| < ∞, then there must be a self-avoiding polygon ω = (ω1 , . . . , ωn ) of
length n ≥ 4 on L∗ which surrounds the origin o = (0, 0) of which each edge is
vacant in L∗ . This is worked out in detail in (Kesten, 1982, Page 386). Thus,
∞
X
1 − θ(p) ≤ mn (1 − p)n , (1.63)
n=4

where mn is the number of self-avoiding loops of length n on L surrounding the

origin o = (0, 0). Clearly, we have that mn ≤ n3n−1 , since there are at most n
possible positions where the self-avoiding loop crosses the half-line {(x, 0) : x ≥
0}, and the number of self-avoiding loops starting from any vertex is bounded
by 3n−1 , so that
X∞
1 − θ(p) ≤ n3n−1 (1 − p)n < 1, (1.64)
n=4

when p < 1 is sufficiently close to 1. Thus, for such p, we have that θ(p) > 0,
proving that p > pc (Z2 ). 2
Having established that the phase transition is non-trivial, the natural question
is what the critical value is. Below, we shall give some results on critical values,
 Percolation 17

particularly in 2 dimensions (see Section 1.2.4). An excellent reference to both

numerical values as well as rigorous bounds is (Hughes, 1996, Chapter 3). For ex-
ample, see (Hughes, 1996, Table 3.3) for some exact values of critical values, and
(Hughes, 1996, Table 3.6) for numerical values of pc (Zd ) on the nearest-neighbor
lattice, showing that the inequality pc (Zd ) ≥ 1/(2d − 1) is only a few percent off
in dimensions d ≥ 5. We now move on to the problem of the uniqueness of the
phase transition, i.e., whether pc (Zd ) = pT (Zd ):
Theorem 1.4. (Phase transition on Zd is unique) For nearest-neighbor per-
colation on Zd with d ≥ 2, pc (Zd ) = pT (Zd ). In particular, χ(p) < ∞ for all
p < pc (Zd ).
The proof of Theorem 1.4 was given independently in (Menshikov, 1986) and
(Aizenman and Barsky, 1987). Menshikov investigates Pp (An ), where An is the
event that there exists an open path from o to ∂S(n) and S(n) = {x ∈ Zd :
|x1 | + . . . + |xd | ≤ n}. Since p < pc , we have that Pp (An ) ↓ 0 as n → ∞. In order
to prove that χ(p) < ∞, it suffices to prove that Pp (An ) ↓ 0 converges to zero
sufficiently fast, since
∞
X ∞
X
χ(p) = Pp (|C(o)| > n) ≤ Pp (Adn1/d e ). (1.65)
n=0 n=0

Menshikov shows that Pp (An ) is close to exponentially small in n for rather

general graphs, (Grimmett, 1999, Theorem 5.4) improves this to
Pp (An ) ≤ e−nψ(p) (1.66)
for some ψ(p) which is strictly positive for p < pc . The proof of (1.66) makes
clever use of Russo’s formula, together with an appropriate renewal type argu-
ment to bound the expected number of pivotal bonds for An .
The proof in (Aizenman and Barsky, 1987) makes use of differential inequal-
ities, a powerful technique which can also be used to prove bounds on critical
exponents. In more detail, (Aizenman and Barsky, 1987) prove that for any p for
which χf (p) = ∞, we have that either θ(p) > 0 or θ(p) = 0 and θ(p0 ) ≥ p10 (p0 − p)
for all p0 > p. This shows that χ(p) < ∞ for p < pc . Indeed, suppose that
χ(p) = ∞. Then, since p < pc , we have that
χf (p) = Ep [|C(o)|1l{|C(o)|<∞} ] = Ep [|C(o)|] = ∞, (1.67)
where we made use of the fact that since p < pc , we have θ(p) = 0. Then, by
the above result and again since θ(p) = 0, θ(p0 ) ≥ p10 (p0 − p) > 0 for all p0 > p,
so that p ≥ pc = pc (Zd ). This is in contradiction to p < pc . We shall not present
further details of the proof of Theorem 1.4, as this is beyond the scope of this
paper. We do mention that both the proofs in (Aizenman and Barsky, 1987)
and in (Menshikov, 1986) are valid more generally. The proof in (Aizenman and
Barsky, 1987) holds for general partially-oriented models, including also finite-
range models as well as oriented percolation models. The proof in (Menshikov,
18 Percolation and random graphs

1986) does not rely on a lattice structure, but rather on the fact that the bound-
a
ary of a ball of size n grows not faster than ecn for some a < 1. These extensions
will be relevant below.
We have seen that, in general, the number of infinite clusters N is a.s. con-
stant, and that N ∈ {0, 1, ∞} a.s. When p < pc , then clearly N = 0 a.s. This
leaves the question what the value of N is in the supercritical regime. On the
tree, we have seen that N = ∞ a.s., and the main question is whether this can
also occur on finite-dimensional lattices. This question was beautifully answered
by (Burton and Keane, 1989) using a geometric argument. A nice survey of the
implications of the Burton-Keane proof can be found in (Meester, 1994).

Theorem 1.5. (Uniqueness of infinite cluster on Zd ) For nearest-neighbor

percolation on Zd with d ≥ 2, there is at most one infinite cluster. More precisely,
for p such that θ(p) > 0,

Pp (there is exactly one infinite cluster) = 1. (1.68)

A first proof of this result appeared in (Aizenman, Kesten and Newman, 1987).

Proof As discussed on page 3, we only need to rule out the possibility that
N , the number of infinite clusters, is equal to +∞ a.s. Indeed, when θ(p) = 0,
then N = 0 a.s., while if θ(p) > 0, then the proportion of vertices in infinite
clusters will be strictly positive, so that N ≥ 1 a.s. Since N is a.s. constant and
N ∈ {0, 1, ∞} a.s., we conclude that N = 1 a.s. when N < ∞.
We will argue by contradiction. We call x ∈ V a trifurcation when it satisfies

(1) |C(x)| = ∞;
(2) there are precisely three occupied edges incident to x;
(3) the removal of the three edges incident to x splits the infinite cluster C(x)
into precisely three disjoint infinite clusters, and no finite clusters.

We write Ix for the indicator that x is a trifurcation, and Pp (Ix ) = Pp (I0 )

for every x ∈ V, and any reference point 0. We shall first argue that when
N = ∞ a.s., then Pp (I0 ) > 0. Let MB denote the number of infinite clusters
touching B, and MB (0) the number of infinite clusters touching B when all
edges with both vertices in B are declared vacant. Since MB (0) ≥ MB , and
Pp (MB(n) ≥ 3) → Pp (N ≥ 3) = 1, by our assumption, we can take n so large
that Pp (MB(n) (0) ≥ 3) ≥ 1/2. Let ω be such that {MB(n) (0) ≥ 3} occurs, and
fix x, y, z ∈ ∂B(n) such that x, y, z lie in three distinct infinite clusters in the
configuration ω̃ for which all edges with both vertices in B are declared vacant.
We make the choice of x, y, z unique in an arbitrary way. Denote by Kx,y,z the
event that MB(n) (0) ≥ 3 and the three points are x, y, z.
Then, there exist 3 self-avoiding paths that only cross in the origin, and have
only x, y, z, respectively, as elements in ∂B(n). Let Jx,y,z be the event that all
the edges along the 3 paths are occupied, while all other edges are vacant. Since
 Percolation 19

B(n) is finite, and letting R(n) denote the number of edges in B(n), we obtain
that
Pp (Jx,y,z ) ≥ (min{p, 1 − p})R(n) = ε > 0. (1.69)
Finally, since, for all fixed x, y, z, the events Jx,y,z and Kx,y,z are independent,
we obtain that
X X
Pp (I0 ) ≥ Pp (Jx,y,z ∩ Kx,y,z ) = Pp (Jx,y,z )Pp (Kx,y,z )
x,y,z∈∂B(n) x,y,z∈∂B(n)
X
≥ε Pp (Kx,y,z ) = εPp (MB(n) (0) ≥ 3) ≥ ε/2 > 0, (1.70)
x,y,z∈∂B(n)

i.e., the probability that x is a trifurcation is strictly positive. By translation

invariance, we therefore obtain that
h X i
Ep Ix ≥ |B(n)|ε/2. (1.71)
x∈B(n)

We now derive a contradiction by noting that the number of trifurcations in B(n)

can be no more than ∂B(n + 1). Let τ denote the number of trifurcations and
u1 , u2 , . . . , uτ denote the trifurcations in B(n). Next, we informally argue that
we can identify to each trifurcation uk a point yk ∈ ∂B(n + 1) such that yi 6= yj
for each 1 ≤ i < j ≤ τ . This can be argued by noting that each trifurcation ui ∈
B(n) in a cluster K splits the intersection K ∩∂B(n+1) into three disjoint parts.
This can be formalized into a family of so-called compatible families of distinct 3-
partitions of K ∩∂B(n+1), and their number is restricted by |K ∩∂B(n+1)|−2,
see (Grimmett, 1999, Lemma 8.5). As a result, the number of trifurcations in
B(n) of a cluster K is bounded by |K ∩ ∂B(n + 1)| − 2 ≤ |K ∩ ∂B(n + 1)|.
Summing out over the distinct clusters yields
X
Ix ≤ |∂B(n + 1)|. (1.72)
x∈B(n)

By (1.71), we obtain
|B(n)|ε/2 ≤ |∂B(n + 1)|, (1.73)
for some ε > 0, which causes a contradiction when n → ∞ whenever |∂B(n +
1)|/|B(n)| → 0. As we see, this argument applies more generally than merely on
Zd . 2
Theorems 1.3, 1.4 and 1.5 answer many of the important questions for percola-
tion, namely, the phase transition is non-trivial, it is unique in the sense that
pc (Zd ) = pT (Zd ) and the infinite cluster, when it exists, is unique. However,
the critical behavior is less well understood. One of the fundamental questions
in percolation is whether there is an infinite component at criticality, i.e., is
θ(pc ) = 0, so that p 7→ θ(p) is continuous? Remarkably, this question is in gen-
eral wide open, even though in d = 2 and in high dimensions, it is affirmatively
20 Percolation and random graphs

answered. We shall discuss these special cases in more detail below. In general,
the behavior of percolation systems at criticality is still ill understood, and there
are no general arguments for the existence of critical exponents. Proofs so far
concentrate on proving the existence by computing their values.
Interestingly, several relations can be drawn between the critical exponents
on the tree and the ones on Zd . This is the content of the following theorem:
Theorem 1.6. (Inequalities for critical exponents on Zd ) For nearest-neighbor
bond percolation on Zd , β ≤ 1, γ ≥ 1 and β ≥ 2/δ, when, for each of the inequal-
ities, we assume that the critical exponent in question exists.
We note that on Tr , the bounds in Theorem 1.6 hold with equality, since
β = γ = 1, δ = 2. In general, the equalities are thus changed to inequalities. We
shall prove that γ ≥ 1 below, making use of Russo’s formula as well as the BK-
inequality. See (Hughes, 1996, Sections 4.3 and 4.4) for a more complete account
of inequalities for critical exponents.
Proof that γ ≥ 1. We use Russo’s formula (1.8) to obtain

d X
τp (x) = Pp ((u, v) is pivotal for o ←→ x). (1.74)
dp
(u,v)∈E

Summing over x yields

d X X
χ(p) = Pp ((u, v) is pivotal for o ←→ x). (1.75)
dp
x∈V (u,v)∈E

We next note that if (u, v) is pivotal for o ←→ x, then there exist two disjoint
paths of occupied bonds connecting 0 and u, and v and x, respectively. Thus,
{o ←→ u} ◦ {v ←→ x} occurs. By the BK-inequality (1.7), we obtain

d X X
χ(p) ≤ τp (u)τp (x − v) = rχ(p)2 , (1.76)
dp
x∈V (u,v)∈E

where r is the degree of the graph G. To make (1.76) precise, one has to overcome
a few technicalities, since Russo’s formula in (1.8) can only be applied to events
E depending on a finite number of bonds, which is not the case for {o ←→ x}.
For details, see e.g., (Heydenreich, van der Hofstad and Sakai, 2008).
To complete the argument, note that we can rewrite (1.76) as

d
χ(p)−1 ≥ −r. (1.77)
dp

Integrating (1.77) over [p, pc ], and noting that χ(pc )−1 = 0 by Theorem 1.4, we
obtain that
χ(pc )−1 − χ(p)−1 = −χ(p)−1 ≥ −r(pc − p), (1.78)
 Percolation 21

so that
1
χ(p) ≥ . (1.79)
r(pc − p)
This proves that γ ≥ 1 if γ exists. 2
As remarked before, no general arguments exist that critical exponents exist.
As we shall explain in more detail below, there are results that guarantee their
existence in special cases, and that the critical exponents satisfy the scaling
relations.
For percolation on Zd , apart from the scaling relations in (1.28), theoretical
physics arguments predict that for low enough dimensions, two further scaling re-
lations going under the name hyperscaling relations, which involve the dimension
d, hold:
dρ = δ + 1, dν = 2 − α. (1.80)
More precisely, it is predicted that the scaling relations in (1.28) hold for all d,
while the hyperscaling relations in (1.80) only hold for d below the so-called upper
critical dimension. Above the upper critical dimension, the critical exponents
are equal to the ones on the tree as appearing in Theorem 1.1. There has been
some work in the direction of establishing the scaling and hyperscaling relations.
In (Borgs, Chayes, Kesten and Spencer, 1999), it is shown that if the critical
exponent ρ exists in the logarithmic sense, and if the probability at criticality
p = pc that there is an easy-way crossing in the cube [0, n]×[0, 3n]d−1 is uniformly
bounded by 1 − ε for some ε > 0, then also δ and η exist in the logarithmic sense,
and satisfy dρ = δ + 1 and 2 − η = d(δ − 1)/(δ + 1). The latter equation is, when
we assume the scaling relations in (1.28), equivalent to dν = 2 − α.
In (Tasaki, 1987a), (Tasaki, 1987b) and (Chayes and Chayes, 1987) it is shown
that when the critical exponents take on the values on a tree, then d ≥ 6,
suggesting that the upper critical dimension is at least 6.
After this discussion on percolation on Zd in general dimension, we now move
to two special cases where much more is rigorously known, two dimensions and
high dimensions.

1.2.4 Percolation in two dimensions

In this section, we study percolation in the plane. We start with the fact that
pc (Z2 ) = 1/2 for bond percolation on the square lattice:
Theorem 1.7. (Harris-Kesten Theorem: pc (Z2 ) = 1/2) For nearest-neighbor
bond percolation on Z2 , θ(1/2) = 0 and θ(p) > 0 for p > 1/2.
The fact that θ(1/2) = 0, which implies that pc (Z2 ) ≥ 1/2, is sometimes
called Harris’ Theorem and was proved in (Harris, 1960). The proof that θ(p) >
0 for p > 1/2 is sometimes called Kesten’s Theorem (Kesten, 1980), and is
considerably more involved. We shall prove Harris’ Theorem below, making use
of an important tool in 2-dimensional percolation going under the name of Russo-
Seymour-Welsh or RSW-theory (Russo, 1981), (Seymour and Welsh, 1978). We
22 Percolation and random graphs

state the version of the RSW Theorem from (Russo, 1981). We first introduce
some notation. We let Rn,k (p) be the probability that there is an occupied path
from left to right crossing the rectangle [0, kn] × [0, n].
Theorem 1.8. (RSW Theorem) For any p ∈ (0, 1) and n, k ≥ 1 with n even,
and for percolation on the square lattice Z2 , the following bounds hold:
h q i6 h q i12
Rn,2 (p) ≥ Rn,1 (p) 1− 1 − Rn,1 (p) , Rn,3 (p) ≥ Rn,1 (p)3 1− 1 − Rn,1 (p) .
(1.81)
We start by proving the RSW Theorem:
Proof of the RSW Theorem 1.8. The main tool will be the Harris-inequality in
(1.5). We start by deducing the second inequality from the first. We note that if
we take the two rectangles [0, 2n]×[0, n] and [n, 3n]×[0, n], then their intersection
is [n, 2n] × [0, n]. Further, if there are left-to-right crossings in [0, 2n] × [0, n] and
in [n, 3n]×[0, n], and a top-to-bottom crossing in [n, 2n]×[0, n], then there is also
a left-to-right crossing in [0, 3n] × [0, n]. Let LRn,1 , LRn,2 , TBn,3 be the events
that the three respective crossings exist. Then,

Rn,3 (p) ≥ Pp (LRn,1 ∩ LRn,2 ∩ TBn,3 ). (1.82)

Now, LRn,1 , LRn,2 and TBn,3 are all increasing events with

Pp (LRn,1 ) = Pp (LRn,2 ) = Rn,2 (p), and Pp (TBn,3 ) = Rn,1 (p). (1.83)

Thus, we arrive at
Rn,3 (p) ≥ Rn,1 (p)Rn,2 (p)2 , (1.84)
showing that the first inequality in (1.81) implies the second.
To see the first inequality in (1.81), in exactly the same way as above, we can
show that
Rn,2 (p) ≥ Rn,1 (p)Rn,3/2 (p)2 . (1.85)
We shall frequently make use of a clever consequence of Harris’ inequality, some-
times called the square root trick. This trick states that, for any two increasing
events A1 and A2 with equal probability Pp (A1 ) = Pp (A2 ), we have
q
Pp (A1 ) ≥ 1 − 1 − Pp (A1 ∪ A2 ). (1.86)

To see (1.86), we note that

[1 − Pp (A1 )]2 = Pp (Ac1 )2 = Pp (Ac1 )Pp (Ac2 ) ≤ Pp (Ac1 ∩ Ac2 ) = 1 − Pp (A1 ∪ A2 ),

(1.87)
so that (1.86) follows by taking the square root and rearranging terms. We shall
now apply (1.86). For this, let Hu be the event that there exists a left-to-right
crossing in [0, n] × [0, n] starting in the line au = {0} × [n/2, n] and Hl the
event that there exists a left-to-right crossing in [0, n] × [0, n] starting in the line
 Percolation 23

al = {0} × [0, n/2]. Then, by (1.86), and the fact that there is a left-to-right
crossing in [0, n] × [0, n] precisely when Hu ∪ Hl holds, we have
q
Pp (Hu ) ≥ 1 − 1 − Rn,1 (p). (1.88)

Now we introduce some more notation. Let s = (s1 , . . . , sm ) be a path connecting

the left and right sides of [0, n] × [0, n], starting on the left side and taking values
in [0, n] × [0, n]. We let Sl denote the set of such paths for which sa ∈ al . Let Es
be the event that s is the lowest occupied path connecting the left and right sides
of [0, n]×[0, n]. For such an s, and with a = au ∪al = {0}×[0, n], we let sa be the
last intersection with a, and we write sr = (sa , . . . , sm ) for the part of the path s
0
after its last visit to a, and sr for the reflection of sr in the line {n} × [0, n]. Let
Fs be the event that there exists a path in [n/2, 3n/2] × [0, n] connecting the top
0
of [n/2, 3n/2] × [0, n] to the path sr , and which always remains above sr ∪ sr .
It is not hard to see that, for every s, we have that Es ∩ Fs ∩ Hu ⊂ LRn,3/2 ,
where we write LRn,3/2 for the event that there is a left-to-right crossing in
[0, 3n/2] × [0, n]. Thus, with
[
G= Es ∩ Fs , (1.89)
s∈Sl

we obtain that G ∩ Hu ⊂ LRn,3/2 , which implies that

Rn,3/2 (p) ≥ Pp (G ∩ Hu ). (1.90)

Applying the Harris’ inequality once more and noting that both G and Hu are
increasing, we obtain that

Rn,3/2 (p) ≥ Pp (G)Pp (Hu ). (1.91)

We have already derived a lower bound on Pp (Hu ) in (1.88), so we are left to

lower bound Pp (G). Since G is a disjoint union, we have that
X
Pp (G) = Pp (Es ∩ Fs ). (1.92)
s∈Sl

Now, for a fixed s, Fs only depends on edges in [0, 3n/2] × [0, n] that are above
0 0
sr ∪ sr , while Es only depends on edges in [0, n] × [0, n] that are below sr ∪ sr ,
so that Es and Fs are independent. Thus,
X
Pp (G) ≥ Pp (Fs )Pp (Es ). (1.93)
s∈Sl

For fixed s, denote by Fs0 the event that there exists a path in [n/2, 3n/2] × [0, n]
0
connecting the top of [n/2, 3n/2]×[0, n] to the path sr , and which always remains
24 Percolation and random graphs

0
above sr ∪sr . Then, clearly, Pp (Fs ) = Pp (Fs0 ) and both Fs and Fs0 are increasing,
so that by the square root trick (1.86), we have
q q
Pp (Fs ) ≥ 1 − 1 − Pp (Fs ∪ Fs0 ) ≥ 1 − 1 − Rn,1 (p), (1.94)

since Pp (Fs ∪ Fs0 ) ≤ Rn,1 (p). Thus, using that the union over s ∈ Sl of Es equals
Hu ,
h q iX h q i
Pp (G) ≥ 1 − 1 − Rn,1 (p) Pp (Es ) = 1 − 1 − Rn,1 (p) Pp (Hu ). (1.95)
s∈Sl

Combining (1.91) with (1.95) yields,

h q i h q i3
Rn,3/2 (p) ≥ 1 − 1 − Rn,1 (p) Pp (Hu )2 ≥ 1 − 1 − Rn,1 (p) , (1.96)

which, combined with (1.85), yields the first claim in (1.81). 2

We continue by discussing duality, a notion which has been extremely im-
portant in two-dimensional percolation. We shall assume that we are working on
a two-dimensional planar lattice L, i.e., a graph that can be embedded into R2
in such a way that different edges can only meet at the vertices of the lattice.
We shall assume that the graph is translation invariant, and that the embedding
of the lattice divides R2 into an infinite tiling of identical and bounded faces.
Then, the vertices of the dual lattice L∗ are the faces of the embedding of L, and
we connect two vertices in L∗ when their corresponding faces share a boundary,
which is a bond in L. Thus, to each bond in L∗ we can identify a unique bond
in L. Thus, to a bond percolation configuration on L, we have identified a bond
percolation configuration on L∗ by identifying the occupation status of a bond
on L∗ to the one of the bond on L to which it is identified. It is sometimes con-
venient to identify the vertices of the dual lattice with the centers of the faces,
and in this representation, we have that the nearest-neighbor square lattice on
Z2 is dual to the nearest-neighbor square lattice on Z2 + (1/2, 1/2), as discussed
in Section 1.2.3. As a result, percolation on the nearest-neighbor square lattice
Z2 is self-dual.
We now investigate the event LRn that there is a left-right crossing of oc-
cupied bonds in the rectangle [0, n + 1] × [0, n]. Also, denote by TB∗n the event
that there is a top to bottom crossing of vacant bonds in the dual lattice on
[1/2, n + 1/2] × [−1/2, n + 1/2]. Clearly, one of the two must happen, so that
Pp (LRn ) + Pp (TB∗n ) = 1. (1.97)
By construction, we have that Pp (TB∗n ) = P1−p (LRn ), so that we obtain, for all
p ∈ [0, 1] and n ≥ 0,
Pp (LRn ) + P1−p (LRn ) = 1. (1.98)
When we pick p = 1/2, we thus obtain that P1/2 (LRn ) = 1/2 for every n ≥ 1.
Thus, Rn,1 (1/2) ≥ P1/2 (LRn ) = 1/2 for every n ≥ 1, so that Rn,2 (1/2) ≥
 Percolation 25

p
[1 − 1/2]6 /2 > 0 uniformly in n ≥ 1. We shall see that this is sufficient to
show that θ(1/2) = 0. Followed by the proof of θ(1/2) = 0, we shall discuss some
further consequences of duality.

Proof that θ(1/2) = 0 for bond percolation on the two-dimensional square lattice.
Fix n ≥ 1, and let Gn be the event that there exists an occupied path from
∂[−n, n] × [−n, n] to ∂[−3n, 3n] × [−3n, 3n], where we write ∂[a, b] × [c, d] for the
boundary of the rectangle [a, b] × [c, d]. Then,
k
Y k
Y
P1/2 (o ←→ ∂B(3k )) ≤ P1/2 (∩kl=1 G3l ) = P1/2 (G3l ) = [1 − P1/2 (Gc3l )],
l=1 l=1
(1.99)
since the events {G3l }∞ l=1 depend on the occupation status of disjoint sets of
bonds, and are therefore independent. By duality, the event Gcn occurs pre-
cisely when there is a dual path of vacant edges in [−3n, 3n] × [−3n, 3n] \
[−n, n] × [−n, n] surrounding the square [−n, n] × [−n, n]. Denote this event
by O(n). Then, P1/2 (Gcn ) = P1/2 (O(n)). The event O(n) is a subset of the event
that there exists left-to-right crossings in the rectangles [−3n, 3n] × [n, 3n] and
[−3n, −3n] × [−n, −3n], and top-to-bottom crossings in the rectangles [n, 3n] ×
[−3n, 3n] and [−n, −3n] × [−3n, 3n]. The probability of each of these events is
equal to Rn,3 (1/2). By Harris’ inequality, the crossings are positively correlated:

P1/2 (O(n)) ≥ Rn,3 (1/2)4 . (1.100)

By Theorem 1.8 and since Rn,1 (1/2) ≥ 1/2, we obtain that Rn,3 (1/2) ≥ a for
some explicit a > 0, and uniformly in n ≥ 1, so that
k
Y
P1/2 (o ←→ ∂B(3k )) ≤ (1 − a) = (1 − a)k . (1.101)
l=1

As a result, we even obtain that P1/2 (o ←→ ∂B(n)) ≤ (1 − a)blog n/ log 3c =

O(n−a/ log 3 ), so that, if ρ > 0 exists, we obtain that ρ ≤ log 3/a. In particular,
we have that θ(1/2) ≤ P1/2 (o ←→ ∂B(n)) for each n ≥ 1, which tends to 0 when
n → ∞. 2

Duality can be used in several more convenient ways. For example, we can define
˜ by the limit
an alternative correlation length ξ(p)

log τpf (0, ne1 )

˜ −1 = − lim
ξ(p) . (1.102)
n→∞ n
˜ is of the same order of magnitude as ξ(p) in (1.15)
It can be expected that ξ(p)
when p is close to critical, as they both describe the maximal distance between
vertices for which there is dependence between their clusters. Then, (Chayes,
26 Percolation and random graphs

Chayes, Grimmett, Kesten and Schonmann, 1989) use duality to show that
˜
ξ(p) ˜ − p)/2 for p > 1/2. Thus, in particular, if ν exists, then so does
= ξ(1
0
ν and it takes the same value.
Using arguments such as duality and RSW-theory, the critical values of
several other lattices in two dimensions have been established. An example is
psite
c (T ) = 1/2, where T is the triangular lattice, which will play an important
role later on. Also site percolation on the triangular lattice turns out to be self-
dual, which explains why its critical value is 1/2.
Recently, Bollobás and Riordan adapted the RSW-ideas in such a way that
they are more generally applicable. For example, in (Bollobás and Riordan,
2006c), they used this new methodology to give a simpler proof of the Harris-
Kesten Theorem. The adapted RSW-methods have also been crucial in order
to prove that certain critical values for site percolation on certain tesselations
equal 1/2. For example, take a Poisson point process. For each vertex in R2 ,
attach it to the closest point(s) in the Poisson point process. This divides R2
in cells, and we draw an edge between two cells when they share a line of their
respective boundaries. Then, when we color the cells independently green with
probability p and yellow with probability 1 − p, (Bollobás and Riordan, 2006a)
shows that the critical value of the occurrence of an infinite green connected
component again is 1/2. In (Bollobás and Riordan, 2008), this result is extended
to other two-dimensional tesselations, such as the Johnson-Mehl tesselation and
two-dimensional slices of three-dimensional Voronoi tesselations.
In the remainder of this section, we shall work with site percolation on the tri-
angular lattice, for which in the past decade tremendous and remarkable progress
has been made. We start by giving the exact values of the critical exponents,
the values of which have been predicted early on in the physics community (see
(Nienhuis, 1984)) for general two-dimensional lattices:
Theorem 1.9. (Critical exponents on the triangular lattice) For site per-
colation on the two-dimensional triangular lattice, the critical exponents β, γ, ν, η, δ
and ρ exist in the logarithmic sense, and take on the values
5 43 4 5 91 48
β= , γ = γ0 = , ν = ν0 = , η= , δ= ρ= . (1.103)
36 18 3 24 5 5
Theorem 1.9 is one of the major breakthroughs in modern probability theory,
particularly since its proof has shed light not only on the existence and scaling
limit of two-dimensional percolation, but rather of the critical behavior of a wide
class of two-dimensional statistical physical models. So far, this technology has
not only been used for percolation, but also for loop-erased random walk and
uniform spanning trees (Schramm, 2000), and a proof for the Ising model has
been announced by Smirnov. The proof of Theorem 1.9 is a consequence of the
connection between critical percolation and so-called stochastic Loewner evolu-
tion (SLE), a topic which we will discuss in some detail below. The history is
that Schramm (Schramm, 2000) first identified a class of continuous models, so-
called SLE, which are conformally invariant models in the plane of which the
 Percolation 27

properties depend on its parameter κ > 0. Schramm continued by noting that if

the scaling limit of two-dimensional percolation would be conformally invariant,
then it must be equal to SLE with parameter κ = 6. Smirnov (Smirnov, 2001)
proved that indeed the scaling limit of critical percolation on the triangular lat-
tice is conformally invariant. This is the celebrated result by Smirnov (Smirnov,
2001), which we shall discuss in more detail below. Schramm already noted that
SLE with parameter κ = 6 has similar critical exponents as in (1.103), when de-
fined in an appropriate way. Smirnov and Werner (Smirnov and Werner, 2001)
realized that the values listed in (1.103) follow from the two statements

P1/2 (A1R ) = R−5/48+o(1) (1.104)

and
P1/2 (A2R ) = R−5/4+o(1) , (1.105)
where A1R is the event that the origin is connected to the boundary of a ball
of radius R, while A2R is the probability that there exist two neighbors of the
origin of which one has a green connection and another has a yellow connection
to the boundary of the ball of width R. The first of these identities simply states
that ρ exists and takes the value 48/5. The fact that these two statements imply
the existence and values of the critical exponents listed in (1.103) is non-trivial
and due to (Kesten, 1987). The asymptotics in (1.104) was shown in (Lawler,
Schramm and Werner, 2002), the one in (1.105) in (Smirnov and Werner, 2001).
The equalities γ = γ 0 and ν = ν 0 follow from the self-duality of site percolation
on the triangular lattice. Theorem 1.9 identifies almost all critical exponents,
an exception being α, which has always remained to be somewhat mysterious.
While ∆ does not appear in Theorem 1.9, we believe that its derivation should
be easier than that of α.
We now discuss the recent work on the scaling limit of critical percolation
on the triangular lattice in more detail. In order to do so, we must start with
the notion of conformal invariance, a notion which is crucial in the study of
two-dimensional critical systems. We work on C, and we let D ⊂ C be a simply
connected domain. We say that a map f : C → C is conformally invariant when
it preserves angles. In order to explain when a map f is angle-preserving we
introduce some notation. Let t 7→ γ1 (t) and t 7→ γ2 (t) be two crossing curves
with γ1 (t), γ2 (t) ∈ C for all t. Suppose that γ1 and γ2 are sufficiently smooth,
then for t close to s, the curves γ1 (t) and γ2 (t) cross each other at a certain angle.
f is called a conformal map or preserves angles when the curves f (γ1 (t)) and
f (γ2 (t)) cross at the same angle as γ1 and γ2 for every pair of crossing curves γ1
and γ2 in D. Important examples of conformal maps are Möbius transformations
given by
az + b
f (z) = , (1.106)
cz + d
where a, b, c, d ∈ C with ad − bc 6= 0. Let D ⊂ C be a (sufficiently smooth)
domain, with four points P1 , P2 , P3 and P4 on the boundary which are such
28 Percolation and random graphs

that Pi is in between Pi−1 and Pi+1 (where, by convention, P5 ≡ P1 ). We call

D4 = (D; P1 , P2 , P3 , P4 ) a 4-marked domain.
We shall investigate general lattices L in two dimensions, and we shall rescale
the lattice with a small factor δ that shall later tend to zero (note that this δ
has nothing to do with the critical exponent δ in (1.25)). For example, for the
triangular lattice, we can think of δ as being the width of the edges in the lattice.
Then, (Langlands, Pouliot and Saint-Aubin, 1994) studied crossing probabilities
of the form that the boundary of D between P1 and P2 has an occupied path to
the part of the boundary between P3 and P4 . Denote this event by Cδ (D4 ). When
we work on a rectangle with P1 = (0, 0), P2 = (0, n), P3 = (n, m), P4 = (m, 0),
then this is nothing but the statement that the rectangle has a left-to-right
crossing. In (Langlands, Pouliot and Saint-Aubin, 1994), the hypothesis was
made that
P (D4 ) = lim Ppc (Cδ (D4 )) (1.107)
δ↓0

exists and lies in (0, 1) when the points P1 , P2 , P3 , P4 are different. These as-
sumptions are already highly non-trivial, but the main assumption in (Lang-
lands, Pouliot and Saint-Aubin, 1994) is that the limit P (D4 ) is conformally
invariant. This is what is often meant with the assumption that the scaling limit
of percolation is conformally invariant. Let us now explain what this assumption
means in more detail. The limit P (D4 ) is conformally invariant when, for D40 be-
ing the image under a conformal map of D4 , we have that that P (D4 ) = P (D40 ).
This means that if we would consider the intersection of D40 with the discretized
lattice of width δ, and we compute the limit

P (D40 ) = lim Ppc (Cδ (D40 )), (1.108)

δ↓0

then in fact this limit exists and equals P (D40 ) = P (D4 ).

The above ‘hypothesis’ now goes under the name of conformal invariance
of percolation, and is in fact what the celebrated paper (Smirnov, 2001) has
proved on the triangular lattice. Since there are many conformal maps, the con-
formal invariance hypothesis is actually quite strong. In fact, Cardy (Cardy,
1992) used it to make a prediction of the exact limit of crossing probabilities in
various domains, using mathematically non-rigorous arguments from conformal
field theory. In order to explain Cardy’s conjecture, we note that, since the limit
of crossing probabilities is invariant under conformal maps, and for any 4-marked
domain D4 there exists a conformal map that maps D to the circle B(1) and
Pi to zi on the boundary of the circle, the limit of crossing probabilities is de-
termined by the crossing probabilities on the circle. For such special 4-marked
domains, we define the cross-ratio η (which again has nothing to do with the
critical exponent η defined in (1.27)) by

(z4 − z3 )(z2 − z1 )
η= . (1.109)
(z4 − z2 )(z3 − z1 )
 Percolation 29

It turns out that η ∈ (0, 1) and that two 4-marked domains on the circle can
be mapped to one another by a conformal map if and only if they have the
same cross-ratio η. Thus, 4-marked domains on the circle can be characterized
by their cross-ratios. In turn, there is a unique conformal map mapping any 4-
marked domain to a 4-marked domain on the circle, so that we can define the
cross-ratio of a general 4-marked domain to be the cross-ratio of the image under
the unique conformal map to the circle. Thus, we see that two 4-marked domains
are conformally equivalent precisely when their cross-ratios are equal, and we can
reformulate the hypothesis of (Langlands, Pouliot and Saint-Aubin, 1994) to say
that the limiting crossing probabilities are a function of their cross-ratio, i.e.,
there exists a function f : (0, 1) 7→ (0, 1) such that P (D4 ) = f (η(D4 )) when
η(D4 ) is the cross-ratio of the 4-marked domain D4 . Based on this assumption,
(Cardy, 1992) shows that in fact

3Γ(2/3) 1/3
P (D4 ) = η 2 F1 (1/3, 2/3; 4/3; η), (1.110)
Γ(1/3)2

where 2 F1 is a hypergeometric function. Carleson noted that Cardy’s conjecture

takes a particularly√ appealing form on an equilateral triangle, i.e., we take P1 =
(1, 0), P2 = (1/2, 3/2), P3 = (0, 0) and P4 = (x, 0) where x ∈ (0, 1), and D
is the interior of the equilateral triangle spanned by P1 , P2 , P3 . Then, (1.110) is
equivalent to stating that, for all x ∈ (0, 1),

P (D4 ) = x. (1.111)

Note that any 4-marked domain can be conformally mapped to the equilateral
triangle, the only degree of freedom being the value of x in Px = (x, 0). In
his seminal paper, Smirnov (Smirnov, 2001) showed (1.111). We shall not go
into the proof in (Smirnov, 2001), as this would take up many more pages than
were allotted to us. Instead, we give some more references to the literature.
The first idea of a possible scaling limit of critical percolation is in the seminal
paper of Schramm, which introduce the limiting stochastic process, which goes
under the name of Stochastic Loewner Evolution, or sometimes Schramm Loewner
Evolution (SLE). Now, SLE has developed into the main tool for studying two-
dimensional critical systems, and it is likely that much more progress shall be
made in this direction in the coming years. We refer to (Bollobás and Riordan,
2006b, Chapter 7) for an expository discussion of Smirnov’s proof, as well as its
consequences. Reviews on SLE and its consequences can be found in (Lawler,
2004; Lawler, 2005; Kager and Nienhuis, 2004; Werner, 2004; Werner, 2005),
and we refer there for more detailed discussions. We close this discussion by
noting that, while the physics community has always predicted that the same
critical behavior should be valid for a wide range of two-dimensional critical
percolation models, the proof on the triangular lattice is, to date, still basically
the only proof of conformal invariance for two-dimensional percolation models.
In particular, the corresponding result for two-dimensional bond percolation on
30 Percolation and random graphs

the square lattice is unknown. One reason for this is that the proof in (Smirnov,
2001) makes essential use of the three-fold rotational symmetry of the triangular
lattice, and it is, up to date, unclear how these symmetries can be replaced by
the different sets of symmetries on the square lattice.
SLE has also proved useful to understand the so-called near-critical phase of
percolation, where p = 1/2 + θδ 4/3 , a nice survey of these results, as well as of
the proof of convergence of the percolation exploration process which explored
the boundary of clusters can be found in (Camia, 2008).
The two-dimensional percolation problem is, after percolation on the tree,
the best understood percolation problem, and the results described above give a
rather complete overview of the depth and wealth of two-dimensional percolation
theory. Yet, several results are not yet known and are worth considering: (a) proof
of existence of scaling functions (see e.g., (Hughes, 1996, (4.295))); (b) a closer
investigation of near-critical percolation (see (Camia, 2008) for an overview);
(c) improvement of our understanding of universality in two-dimensional perco-
lation, for example, by proving conformal invariance of two-dimensional bond
percolation on the square lattice.

1.2.5 Percolation in high dimensions

In this section, we study percolation in high-dimensions. We consider G = (V, E)
with V = Zd and with edge set E either the nearest-neighbor bonds in sufficiently
high dimension, or the spread-out bonds

E = {{x, y} : kx − yk∞ ≤ L} (1.112)

for some L sufficiently large. The main result in high dimensions is the following:

Theorem 1.10. (Mean-field critical exponents for high-d percolation)

For percolation on Zd , for either d sufficiently large in the nearest-neighbor
model, or d > 6 and L sufficiently large in the spread-out model, β = γ = 1,
ν = 1/2 and δ = ∆ = 2 in the bounded-ratio sense, while η = 0 in the asymptotic
sense.

It is believed, by invoking the paradigm of universality, that the critical

exponents for any finite-range system which has sufficient symmetry are equal.
Thus, Theorem 1.10 suggests that also β = γ = 1, ν = 1/2 and δ = ∆ = 2 for the
nearest-neighbor model with d > 6. However, since the paradigm of universality
is not mathematically rigorous, we cannot conclude this. Note that Theorem 1.10
does support the prediction of universality, since, in particular, the values of the
critical exponents do not depend on the precise values of L, when L is sufficiently
large. Also, for d sufficiently large, the critical values agree for all values of L.
The reason for the fact that L or d needs to be big in Theorem 1.10 is that
the proof of Theorem 1.10 makes use of a perturbation expansion called the lace
expansion. We shall now first discuss the history of the proof, before discussing
the details.
 Percolation 31

In (Aizenman and Newman, 1984), it was proved that γ = 1 when the so-
called triangle condition, a condition on the percolation model, holds. The tri-
angle condition states that

X
∇(pc ) = τpc (0, x)τpc (x, y)τpc (y, 0) < ∞, (1.113)
x,y∈V

where we recall that τp (x, y) = Pp (x ←→ y) is the two-point function. In (Barsky

and Aizenman, 1991), it was shown that, under the same condition, β = 1 and
δ = 2. Needless to say, without the actual verification of the triangle condition,
this would not prove anything. The triangle condition was proved to hold in
the setting in Theorem 1.10 in (Hara and Slade, 1990) by the use of the lace
expansion, a method which has since proved to be extremely powerful in order
to characterise mean-field behavior of various models in high dimensions. Later,
the results for ν, ∆ and η were proved in (Hara, 1990), (Nguyen, 1987) and
(Hara, 2005; Hara, van der Hofstad and Slade, 2003), again using lace expansion
arguments. Several related results on high-dimensional percolation, in particu-
lar suggesting that the scaling limit of large critical clusters is a process called
Integrated Brownian excursion, can be found in (Hara and Slade, 2000a; Hara
and Slade, 2000b), where also the fact that δ = 2 in the asymptotic sense was
proved.
We now discuss the methodology in high dimensions. We start with the proof
that γ = 1 if the triangle condition holds. Recall the argument that shows that
γ ≥ 1 below Theorem 1.6, and (1.75) in particular. The BK-inequality gives an
upper bound on (1.75), and, in order to prove that γ = 1, a matching lower bound
needs to be obtained. For this, we can use the independence of the occupation
status of the bonds to explicitly write

(u,v)
Pp ((u, v) is pivotal for o ←→ x) = Ep [1l{o←→u} τ C̃ (o)
(v, x)], (1.114)

where, for a set of sites A, the restricted two-point function τ A (v, x) is the
probability that v is connected to x using only bonds with both endpoints outside
A, and C̃ (u,v) (o) consists of those sites which are connected to 0 without using
(u,v)
the bond (u, v). Clearly, τ C̃ (o)
(v, x) ≤ τ (v, x), and this reproves the upper
bound previously obtained using the BK-inequality. We note that

A
τ (v, x) − τ A (v, x) = Pp (v ←→ x), (1.115)

A
where we write that v ←→ x when every path of occupied bonds from v to x
contains a bond containing a vertex in A. Thus,
32 Percolation and random graphs

d X X
χ(p) = Ep [1l{o←→u} ]τp (v, x)
dp
x∈V (u,v)∈E
X X C̃ (u,v) (o)
− Ep [1l{o←→u} Pp (v ←→ x)]
x∈V (u,v)∈E
X X C̃ (u,v) (o)
= rχ(p)2 − Ep [1l{o←→u} Pp (v ←→ x)]. (1.116)
x∈V (u,v)∈E

Now, for any A ⊆ Zd ,

A
X
Pp (v ←→ x) ≤ Pp ({v ←→ a} ◦ {a ←→ x}), (1.117)
a∈A

which leads to
d X
χ(p) ≥ rχ(p)2 − χ(p) Ep [1l{o←→u,a} ]Pp (v ←→ a). (1.118)
dp
(u,v)∈E

Applying the BK-inequality yields that

X
Pp (o ←→ u, a) ≤ Pp ({o ←→ z} ◦ {z ←→ u} ◦ {z ←→ a}) (1.119)
z
X
≤ τp (z)τp (u − z)τp (a − z),
z

so that X
Pp (o ←→ u, a)Pp (v ←→ a) ≤ χ(p)[∇(p) − 1], (1.120)
(u,v)∈E

implying that
d
χ(p) ≥ χ(p)2 [2 − ∇(pc )]. (1.121)
dp
If we know that ∇(pc ) < 2, then we can integrate the above equation in a similar
fashion as around (1.77) to obtain that γ = 1. When we only have the finiteness
of the triangle, then some more work is necessary to make the same conclusion
(see (Aizenman and Newman, 1984) for details).
In the lace expansion, the above argument is adapted to deal with τp (x)
directly, using rewrites as in (1.114)–(1.115) repeatedly, instead of using the
upper bound in (1.117). We refer to (Hara and Slade, 1990) or the monograph
(Slade, 2006) for detailed derivations of the lace expansion.
The lace expansion can also be used to prove asymptotics of the critical
value in high dimensions, either when L → ∞ for d > 6 fixed or for the nearest-
neighbor model and d → ∞. In (van der Hofstad and Sakai, 2005), the asymp-
totics of the critical point for percolation, as well as for self-avoiding walk, the
 Percolation 33

contact process and oriented percolation, were investigated for d > 6 and L → ∞.
It was shown that
1 + cd Ld + o(L−d )
pc (L, Zd ) = , (1.122)
(2L + 1)d − 1
for some explicit constant cd > 0, and where pc (L, Zd ) is the critical value of
spread-out percolation with edge set E in (1.112). The best asymptotics of pc (Zd )
when d → ∞ are in (Hara and Slade, 1993; Hara and Slade, 1995; van der Hofstad
and Slade, 2005; van der Hofstad and Slade, 2006), where it is shown that, when
d → ∞, pc (Zd ) has an asymptotic expansion in terms of inverse powers of (2d)
with rational coefficients, i.e., for each n, we can write
n
X
pc (Zd ) = ai (2d)−i + O((2d)n+1 ), (1.123)
i=1

where the ai are rational coefficients with a1 = a2 = 1, a3 = 7/2. We refer to the

references in (van der Hofstad and Sakai, 2005) and (van der Hofstad and Slade,
2006) for the literature on asymptotics of percolation critical points.
We close this section by discussing finite-size scaling in high-dimensional per-
colation. In (Aizenman, 1997), it was assumed that a version of η = 0 holds (more
precisely, that τpc (x) is bounded above and below by positive and finite constants
times |x|−(d−2) ) in order to show that, at criticality, the largest intersection of a
cluster with a cube of width 2r + 1 grows like r4 times logarithmic corrections.
This corresponds to the bulk boundary condition. The condition used was ver-
ified in (Hara, 2005; Hara, van der Hofstad and Slade, 2003) in the setting of
Theorem 1.10.
Aizenman proceeds to conjecture that at criticality, with periodic boundary
conditions, the largest critical cluster grows like r2d/3 , i.e., like V 2/3 , where V =
(2r + 1)d is the volume of the cube. This was proved in (Heydenreich and van
der Hofstad, 2007) making crucial use of the combined results in (Borgs, Chayes,
van der Hofstad, Slade and Spencer, 2005a; Borgs, Chayes, van der Hofstad,
Slade and Spencer, 2005b). Such behavior is dubbed random graph asymptotics,
as V 2/3 growth at criticality is best known for the Erdős-Rényi random graph
discussed in Section 1.3 below. Related results in this direction can be found
in (van der Hofstad and Luczak, 2006) and (Nachmias, 2007), using alternative
methods. An interesting question is what proper definitions of the critical value
or window are in the general context of percolation on finite graphs, and what the
proper conditions on the graph are such that percolation on it has random graph
asymptotics close to criticality. While the combined examples in (Borgs, Chayes,
van der Hofstad, Slade and Spencer, 2005a; Borgs, Chayes, van der Hofstad,
Slade and Spencer, 2005b; Heydenreich and van der Hofstad, 2007; Nachmias,
2007) provide some initial ideas, the general picture is not yet clear.
Despite the fact that detailed results are available in high dimensions, sev-
eral results are not yet known and are worth considering: (a) proof of existence
and mean-field values of ν 0 , γ 0 , ρ (particularly the supercritical regime in high-
dimensions is still ill understood); (b) proof of existence of scaling functions (see
34 Percolation and random graphs

e.g., (Hughes, 1996, (4.295)); (c) improvement of our understanding of univer-

sality in high-dimensional percolation, by, for example, showing that bond per-
colation on the nearest-neighbor lattice has the critical exponents in Theorem
1.10 for any d > 6.

1.2.6 Oriented percolation

In this section, we study so-called oriented or directed percolation. In this case,
G = (V, E) is given by V = Zd × Z+ , and E = {((x, n), (y, n + 1) : |x − y| = 1},
and G is considered as a directed graph, i.e., we remove all bonds independently
with probability p and (x, m) −→ (y, n) is only possible when m ≥ n. Thus, we
can only traverse edges in the direction of increasing last coordinate, and this
last coordinate has the convenient interpretation of time. We define the forward
cluster C(x, n) of (x, n) ∈ Zd × Z+ to be

C(x, n) = {(z, l) : (x, n) −→ (z, l)}, (1.124)

so that, in particular, C(x, n) ⊂ Zd × {n, n + 1, . . .}.

In some cases, we shall, similarly to the setting in Section 1.2.5, also deal
with the spread-out model, in which E = {((x, n), (y, n + 1) : kx − yk∞ ≤ L}
for some L ≥ 1. While one might expect that percolation on oriented lattices is
quite similar to percolation on unoriented lattices, this turns out not to be the
case:
Theorem 1.11. (Continuity of oriented percolation) For oriented perco-
lation on Zd × Z+ , for d ≥ 1, there is no infinite cluster at p = pc (Zd × Z+ ),
i.e., θ(pc (Zd × Z+ )) = 0.
Theorem 1.11 was first proved in (Bezuidenhout and Grimmett, 1990) for di-
rected percolation, which is a slight variation of the model defined here. The
results were extended to the oriented percolation setting described above in
(Grimmett and Hiemer, 2002).
The proof of Theorem 1.11 makes use of a block renormalization which was
also used in (Barsky, Grimmett and Newman, 1991a; Barsky, Grimmett and
Newman, 1991b) to prove that percolation does not occur in half-spaces. The
proof in (Bezuidenhout and Grimmett, 1990) also applies to the contact process,
a continuous-time adaptation of oriented percolation. The deep relation between
the contact process and oriented percolation has proved to be quite useful, and
results in one model can typically also be proved for the other.
In (Durrett, 1980), the one-dimensional contact process and oriented perco-
lation models were studied, focussing on the growth of the vertices in the cluster
of the origin (0, 0) ∈ Zd × Z+ at time n. These results basically show that when
the cluster of the origin is infinite, then the part of it at time n grows linearly in
n with a specific growth constant. In (Sakai, 2002), the hyperscaling inequalities
for oriented percolation and the contact process were proved, indicating that
mean-field critical exponents can only occur for d > 4, thus suggesting that the
upper critical dimension of oriented percolation equals dc = 4. Indeed, as proved
 Percolation 35

thereafter, the orientation of the percolation problem implies that mean-field

behavior already occurs for d > 4:
Theorem 1.12. (Mean-field critical exponents for oriented percolation)
For oriented percolation on Zd × Z+ , for either d sufficiently large in the nearest-
neighbor model, or d > 4 and L sufficiently large in the spread-out model,
β = γ = 1 and δ = ∆ = 2 in the bounded-ratio sense, while η = 0 in the
Fourier-asymptotic sense.
The proof of Theorem 1.12 is given in (Nguyen and Yang, 1993; Nguyen and
Yang, 1995; van der Hofstad and Slade, 2003) and follows a similar strategy as
the proof of Theorem 1.10 by employing the results in (Aizenman and Barsky,
1987; Aizenman and Newman, 1984) assuming the triangle condition, and using
the lace expansion as in (Hara and Slade, 1990). In (van der Hofstad, den Hol-
lander and Slade, 2007; van der Hofstad, den Hollander and Slade, 2007) it is
proved that in the spread-out setting, for d > 4, the probability that there is an
occupied path at criticality connecting (0, 0) to {(x, n) : x ∈ Zd } is asymptotic to
1/(Bn)(1 + o(1)). This can be seen as a version of the statement that the critical
exponent ρ exists and takes the mean-field value ρ = 1/2. In (Sakai, 2002), hy-
perscaling inequalities are shown that imply that critical exponents cannot take
their mean-field values when d < 4. The main results in (van der Hofstad and
Slade, 2003) make a connection between clusters at criticality for the spread-out
oriented percolation model above 4 dimensions, and a measure-valued process
called super-Brownian motion, a model which can be seen as the scaling limit
of critical branching random walk. See (Dawson, 1993; Dynkin, 1994; Etheridge,
2000; Le Gall, 1999; Perkins, 2002) for expositions on super-processes.
It would be of interest to prove that scaling functions exist for high-dimensional
percolation, and to prove further results concerning the critical exponents. For
example, we do not know ν, or that γ 0 exists and γ 0 = γ = 1 for oriented
percolation above four dimensions.

1.2.7 Percolation on non-amenable graphs.

We start by defining what an amenable graph is. For a finite set of vertices V ,
we denote its edge boundary by

∂E V = {(u, v) : u ∈ V, v 6∈ V }. (1.125)

The notion of amenability is all about whether the size of ∂E V is of equal order as
that of V , or is much smaller. To formalize this, we denote the Cheeger constant
of a graph G by
|∂E V |
h(G) = inf . (1.126)
V ⊂V:|V |<∞ |V |

A graph is called amenable when h(G) = 0, and is it called non-amenable oth-

erwise. Key examples of amenable graphs are finite-range translation invariant
graphs G with vertex set Zd , the simplest example of a non-amenable graph is
36 Percolation and random graphs

the regular tree Tr with r ≥ 3. For the regular tree Tr with r ≥ 3, it is not
hard to see that h(Tr ) = r − 2. (Benjamini and Schramm, 1996) contains cer-
tain preliminary results of percolation on non-amenable graphs, and many open
questions, some of which have been settled in the mean time. For example, (Ben-
jamini and Schramm, 1996, Theorem 1) shows that pc (G) ≤ 1/(h(G) + 1), so
that pc (G) < 1 for every non-amenable graph.
A related definition of non-amenability can be given in terms of the spectral
radius of a graph. Let pn (u, v) be the probability that simple random walk on G
starting at u ∈ V is at time n at v ∈ V. The spectral radius of G is defined as

ρ(G) = lim p2n (u, u)1/(2n) . (1.127)

n→∞

By Kesten’s Theorem (Kesten, 1959a; Kesten, 1959b), see also (Dodziuk, 1984),
when G has bounded degree, ρ(G) < 1 precisely when h(G) > 0. This exemplifies
the fact that there is a close relationship between graph theoretic properties, and
the behavior of stochastic processes on the graph. A similar relation between the
existence of invariant site percolation and amenability of Cayley graphs is proved
in (Benjamini, Lyons, Peres and Schramm, 1999b, Theorem 1.1).
As we have seen for percolation on Zd in Theorem 1.5, in the super-critical
regime, the infinite cluster is unique. It turns out (see e.g., the discussion fol-
lowing (Kesten, 2002, Theorem 4)) that the uniqueness of the infinite cluster
is valid for all amenable graphs. As the proof of Theorem 1.5 shows, there is a
close relation between the ratio of the size of the boundary and its volume and
the uniqueness of the infinite cluster, which helps to explain the uniqueness for
all amenable graphs. On the other hand, for trees, the number of infinite com-
ponents equals N = ∞ a.s. in the supercritical phase, which can be attributed
to the fact that it we remove one edge, then a tree falls apart into two infinite
graphs which will each have at least one infinite component a.s., so that, in total
there will be infinitely many infinite clusters. Thus, this phenomenon is more
related to there not being any cycles rather than the boundary being large.
In order to investigate the number of infinite clusters, we define the unique-
ness critical value by

pu = pu (G) = inf{p : Pp − a.s. there is a unique infinite cluster}. (1.128)

For the regular tree with r ≥ 3, pu = 1, while for Zd , pu = pc . Below, we shall

give examples where pc < pu < 1.
While the existence of an infinite cluster is clearly an increasing event, the
uniqueness of the infinite cluster is not. Therefore, it is a priori not obvious that
for all p > pu , the infinite cluster will be unique. This is the main content of the
following theorem. In its statement, we will write N (p) for the number of infinite
clusters in the coupling of the percolation models for all p ∈ [0, 1] described
above (1.4).
 Percolation 37

Theorem 1.13. (Uniqueness transition) For percolation on a connected, quasi-

transitive, infinite graph of bounded degree, a.s.,

0
 for p ∈ [0, pc );
N (p) = ∞ for p ∈ (pc , pu ); (1.129)

1 for p ∈ (pu , 1].


The proof of this theorem can be found in (Schonmann, 1999), and related
results appeared in (Häggström and Peres, 1999; Häggström, Peres and Schon-
mann, 1999). Note that, in general, not much is known for the critical cases
p = pc and p = pu .
(Benjamini and Schramm, 1996, Theorem 4) gives a criterion in terms of the
spectral radius which implies that pc < pu . Indeed, it shows that if ρ(G)pr < 1,
where r is the maximal degree of G, then there are a.s. infinitely many infinite
clusters. Thus, if ρ(G)pc r < 1, then pc < pu . Interesting examples arise by
looking at Cartesian products of graphs. Let G1 = (V1 , E1 ) and G2 = (V1 , E1 )
be two graphs, and let G = G1 × G2 have vertex set V = V1 × V2 and edge set

E = {(u1 , u2 ), (v1 , v2 )} : (u1 , v1 ) ∈ E2 or (u2 , v2 ) ∈ E2 . (1.130)

Then, clearly, when the maximal degrees of G1 and G2 , respectively, are denoted
by r1 and r2 respectively, the maximal degree of G = G1 × G2 is r1 + r2 . Also,
pc (G1 × G2 ) ≤ pc (G2 ). (Benjamini and Schramm, 1996, Corollary 1) gives many
examples of graphs with pc < pu by looking at G = G1 × Tr , where G1 is quasi
transitive and Tr is the r-regular tree of degree r ≥ 3. Indeed, by (Benjamini
and Schramm, 1996, Theorem 4), and pc (G1 × Tr ) ≤ pc (Tr ) = (r − 1)−1 , we
obtain that ρ(G)pc r < 1 when r is sufficiently large since ρ(G) → 0 as r → ∞. A
simple example of a graph where pc < pu < 1 is Z × Tr as proved in (Grimmett
and Newman, 1990).
We continue by studying the nature of the phase transition on non-amenable
graphs. The first result concerns the continuity of the phase transition for Cayley
graphs of non-amenable groups. We start by defining what a Cayley graph is.
Let Γ be a group, and let S = {g1 , . . . , gn } ∪ {g1−1 , . . . , gn−1 } be a finite set of
generators. The Cayley graph G = G(Γ) has vertex set V = Γ, and edge set
E = {{g, h} : g −1 h ∈ S}.
Theorem 1.14. (Continuity on non-amenable Cayley graphs) For perco-
lation on a Cayley graph of a finitely generated non-amenable group, there is no
infinite cluster at p = pc (G), i.e., θ(pc (G)) = 0.
This result was proved in (Benjamini, Lyons, Peres and Schramm, 1999a;
Benjamini, Lyons, Peres and Schramm, 1999b), and generalized earlier work by
(Wu, 1993) on Z × Tr with r ≥ 7. The proof makes use of the mass-transport
technique to a clever choice of the mass-transport function.
We complete this section by describing some results on critical exponents. For
this, we need to introduce the notions of planar and unimodular graphs. A graph
38 Percolation and random graphs

is called planar when it can be embedded into R2 with vertices being represented
by points in R2 and edges by lines between the respective vertices such that the
edges only intersect at their end-points. For x ∈ V, let the stabilizer of x S(x) be
the set of automorphisms of G that keep x fixed, i.e., S(x) = {γ : γ(x) = x}. The
graph G is called unimodular if |{γ(y) : γ ∈ S(x)}| = |{γ(x) : γ ∈ S(y)}| for every
x, y ∈ V. Unimodularity turns out to be an extremely useful notion, particularly
since it turns out to be imply the so-called mass-transport principle (MTP).
Indeed, let G be a transitive unimodular graph. We say that f : V×V → [0, ∞) is
diagonally invariant under the automorphisms of G when f (x, y) = f (γ(x), γ(y))
for every automorphism γ : V → V. Then, the MTP states that
X X
f (x, y) = f (y, x). (1.131)
y∈V y∈V

For a proof, see e.g. (Benjamini, Lyons, Peres and Schramm, 1999a). In most
cases, the MTP in (1.131) is used in the following way. We take φ : V × V ×
2G → [0, ∞) such that φ(x, y, ω) = φ(γ(x), γ(y), γ ◦ ω), for any configuration
ω and where γ ◦ ω is the configuration (γ ◦ ω)(x) = ω(γ(x)). We interpreted
φ(x, y, ω) as the mass which x sends to y in the configuration κ. Then, we take
f (x, y) = Ep [φ(x, y, ω)], and the MTP implies that the mass sent out by x is
equal to the amount of mass x receives, which explains the name mass transport
principle. For non-unimodular graphs, an adaptation of (1.131) holds, where the
left hand side is multiplied by w(x) and the right hand side by w(y), where
w(x) = |Sx o|/|So x|, but this relation is not as powerful. All amenable graphs are
unimodular, an example of a non-unimodular graph is the so-called grandmother
graph, which is obtained by adding a connection between any vertex of the tree
to its grandmother (i.e., the unique vertex which is two steps close to the root
than the vertex itself). See (Timár, 2006) for a wealth of related examples of
non-unimodular graphs.
Finally, the number of ends of a graph G is
E(G) = sup {number of infinite connected components of G \ S}. (1.132)
S⊂V:|S|<∞

Then, (Schonmann, 2001; Schonmann, 2002) prove that percolation has mean-
field critical exponents in the following cases:
Theorem 1.15. (Mean-field critical exponent on non-amenable graphs)
For percolation on a locally finite, connected, transitive, non-amenable graph G,
β = γ = 1, δ = ∆ = 2 in the bounded-ratio sense, in the following cases:
√
1. Graphs G for which h(G) > ( 2r2 − 1 − 1)/2, where r is the degree of the
graph;
2. Graphs G which are planar and have one end;
3. Graphs G which are unimodular and have finitely many ends.
Since percolation in high-dimensions is known to have mean-field critical
exponents, which are the critical exponents on the tree, one would expect that,
 Percolation 39

in great generality, percolation on non-amenable graphs do so too. Theorem 1.15

is a step into the direction of proving this belief, but a general result to this extent
is still missing. It would be of interest to investigate this form of universality in
more detail, as well as the existence of scaling functions for general non-amenable
graphs.
An application of percolation on non-amenable graphs to image analysis can
be found in (Kendall and Wilson, 2003).

1.2.8 Continuum percolation.

Continuum percolation is a close brother of percolation, where instead of working
on a lattice, we work in the continuum. While there are many possible models,
we shall restrict to the simplest version, the so-called Boolean model. For details
on the model, see the monograph (Meester and Roy, 1996), or (Penrose, 2003,
Section 9.6), and the references therein. In the Boolean model, we start with a
Poisson point process PPP of a given intensity λ > 0, and each point x ∈ PPP
is assigned a radius. The radii of the different vertices are independent random
variables, an important special case is when all radii are fixed. For x ∈ PPP, let
Rx be its corresponding radius. We create an occupied region by looking at all
vertices contained in the union over x of the balls of radius Rx , i.e., the occupied
region is given by [
O= B(x, Rx ), (1.133)
x∈PPP

where B(x, r) is the ball of radius r centered at x ∈ Rd . We denote by C(x) the

connected part of O that contains x, we let θ(λ) be the probability that C(x) is
unbounded, and we define
χ(λ) = Eλ [|C(o)|], (1.134)
where, for a region C ⊂ Rd , we write |C| for the Lebesgue measure of C. The
function θ(λ) plays a similar role in continuum percolation as the percolation
function in (discrete) percolation, while χ(λ) plays a similar role as the expected
cluster size. With the above definitions at hand, continuum percolation as a
model is quite similar to discrete percolation described above, and most of the
results discussed above for percolation on Zd carry over to continuum perco-
lation on Rd . In fact, many proofs make crucial use of the discrete result, by
an appropriate discretization procedure. However, by varying the random radii,
certain phenomena arise that are not present in bond percolation, such as the
possibility that cluster have, a.s., finitely many Poisson points, yet the expected
number of Poisson points is infinite.
Continuum percolation is an important model from the point of applications,
as it can be seen as a simple model for a communication network where trans-
mitters have a finite transmitting distance. When the points in the Poisson point
process PPP mark the locations of sensors in an ad hoc network, and the radii
correspond to their transmission distance, then the fact that |C(x)| = ∞ cor-
responds to the fact that the sensors can jointly transmit over an unbounded
40 Percolation and random graphs

domain. In this light, continuum percolation is becoming an important tool in

the investigation of various telecommunication networks. For examples of the
application of continuum percolation ideas to communications, see e.g. (Dousse,
Franceschetti, Macris, Meester and Thiran, 2006), (Baccelli and Blaszczyszyn,
2001) or the recent book (Franceschetti and Meester, 2008). In (Grossglauser
and Thiran, 2006), you can find a discussion on the relation between percolation
problems and the engineering of wireless telecommunication models, touching
upon navigability of networks, and the relation between connectivity and capac-
ity.
Random geometric graphs are obtained by taking only a bounded domain and
performing a similar strategy as described above on it. More precisely, random
geometric graphs are characterized by two parameters and can be constructed
as follows. We consider the square [0, 1]2 and we put n points in it, uniformly
at random. We can think of these n points as being a population of adhoc or
peer-to-peer network users, who wish to communicate data to each other. Then
we connect pairs of points within distance r for some appropriate r (possibly
depending on n). Questions of interest are how large the radius should be as
a function of the intensity in order for all points to be connected, and what
the minimal degree of the vertices in this graph is. The connectivity of the
graph is essential for each of the network users to be able to transmit data to
each other. Clearly, this model is only a mere caricature of reality, and a better
understanding of the real-world properties of networks is necessary. However,
this caricature model does already shed light on some of the basic problems in
geometric wireless networks. We refer to (Penrose, 2003) for detailed results on
this model, as well as on the literature.

1.3 Random graphs

1.3.1 Motivation
Real-world networks. In the past decade, many examples have been found of
real-world networks that are small worlds, scale-free and highly clustered. We
shall start by discussing these notions one by one.
The small-world phenomenon states that distances in a network are relatively
small. This is related to the well-known ‘Six degrees of separation’ paradigm
in social networks, which conjectures that any pair of individuals in the world
can be connected by a chain of persons knowing each other on a first name
basis, the chain consisting of at most 6 intermediary individuals. This paradigm
has attracted considerable attention, see e.g. (Newman, Watts and Barabási,
2006) for a historical account, including the original papers. The notion of real
networks being small worlds is inherently a bit imprecise, as we do not define
what ‘relatively’ small is. For the time being one can have in mind that distances
in many real networks are at most 10-20, later, we shall give a more precise
definition what it means for a process of random graphs to be a small world
process.
 Random graphs 41

The scale-free phenomenon states that the degree sequences in a network

satisfies a power law. In more detail, the degree sequence of a network of size n,
∞
which we denote by {p(n)k }k=0 , is given by

n
1X
pk(n) = 1l{di =k} , (1.135)
n i=1

where di is the degree of vertex i in the network, i.e., the number of neighbors
of vertex i. Then, a real network is scale free when pk(n) is approximately propor-
tional to a power law with a certain exponent τ ∈ [1, ∞), i.e., for k sufficiently
large, we have that there is a constant C > 0 such that

pk(n) ≈ Ck −τ . (1.136)

Naturally, also the notion of a real network being scale free is inherently some-
what vague. For example, how precise must the approximation in (1.136) be?
Equation (1.136) can only be valid up to a certain point, since, for a simple
graph of size n (i.e., a graph without self-loops and multiple edges), the maximal
degree is equal to n − 1, so that the left hand side of (1.136) equals 0 for k ≥ n,
while the right hand side remains positive for all k. In practice, (1.136) is verified
by taking the logarithm on both sides and noticing that log p(n)k is close to linear
in log k:
log pk(n) ≈ log C − τ log k, (1.137)

i.e., a loglog plot of the degree sequence should be close to linear, and the slope
is given by −τ , where τ denotes the power-law exponent.
A network is highly clustered when two typical neighbors of an arbitrary
vertex are more likely to be connected to each other as well than an arbitrary
pair of vertices, i.e., many wedges are in fact closed to become triangles. When
we draw two different vertices uniformly at random from a graph of size n, the
probability that there is an edge between the drawn vertices is equal to

2E
, (1.138)
n(n − 1)

¯ where
where E is the number of edges. Note that 2E/n = d,
n
1X
d¯ = di (1.139)
n i=1

2E
is the average degree of all vertices in the network. Thus, we see that n(n−1) =
¯
d/(n − 1). In most real-world networks, the average degree is much smaller than
the size of the network, so that the probability that two uniformly drawn vertices
42 Percolation and random graphs

share an edge is rather small. The clustering coefficient CG of a graph G = (V, E)

is defined by P
i,j,k∈V 1l{ij,ik,jk∈E}
CG = P , (1.140)
i,j,k∈V 1l{ij,ik∈E}

i.e., the proportion of wedges that forms a triangle. A network is highly clustered
2E
when CG is much larger than n(n−1) . Again, this notion is inherently imprecise,
as we do not define how much larger CG needs to be. The reason why many
networks are highly clustered is that there often is a certain group structure.
For example, in a collaboration network, if a mathematician has published with
another mathematician, then they are likely to be in the same community. Thus,
when a mathematician has published with two other mathematicians, then they
are likely to all be in the same community, which increases the likelihood that
the two mathematicians have also published together. Similar effects play a role
in the World-Wide Web and in social networks.
As explained above, all these notions are merely empirical, and we shall give
a proper mathematical definition before we introduce the mathematical random
graph models that we shall consider in this section, and which are aimed at
describing real networks. The aim of this section is not to define these notions
precisely for empirical networks, but instead to define these notions precisely for
random graph models for them. See (Albert and Barabási, 2002; Dorogovtsev and
Mendes, 2002; Newman, 2003) for reviews on complex networks, and (Barabási,
2002) for a more expository account. We do notice that these real-world complex
networks are not at all like classical random graphs (see (Alon and Spencer, 2000;
Bollobás, 2001; Janson, Luczak and Rucinski, 2000) and the references therein),
particularly since the classical models do not have power-law degrees. As a result,
the empirical findings of real-world networks have ignited enormous research
on adaptations of the classical random graph that do have power-law degree
sequences. We shall survey some of these results, and we shall start by defining
the notions of small-world, scale-free and highly-clustered random graphs in a
precise mathematical way.
Small-world, scale-free and highly-clustered random graph processes. As de-
scribed in the above motivation, many real-world complex networks are large.
Many of them, such as the World-Wide Web and collaboration networks, grow
in size as time proceeds. Therefore, it is reasonable to consider graphs of grow-
ing size, and to define the notions of scale-free, small-world and highly-clustered
random graphs as a limiting statement when the size of the random graphs tends
to infinity. This naturally leads us to study graph sequences. We shall denote a
sequence of random graphs by {Gn }∞ n=1 , where n denotes the size of the graph
Gn , i.e., the number of vertices in Gn .
Denote the proportion of vertices with degree k by Pk(n) , i.e.,
n
1X
Pk(n) = 1l (n) , (1.141)
n i=1 {Di =k}
 Random graphs 43

where Di(n) denotes the degree of vertex i ∈ {1, . . . , n} in the graph Gn , and recall
that the degree sequence of Gn is given by {Pk(n) }∞ k=0 . We use capital letters in
our notation to indicate that we are dealing with random variables, due to the
fact that Gn is a random graph. This explains why there are capitals in (1.141),
but not in (1.135). Now we are ready to define what it means for a random graph
process {Gn }∞n=1 to be scale free.
We call a random graph process {Gn }∞ n=1 sparse when

lim Pk(n) = pk , (1.142)

n→∞

for some deterministic limiting probability distribution {pk }∞ k=0 . Since the limit
pk in (1.142) is deterministic, the convergence in (1.142) can equivalently be
taken as convergence in probability or in distribution. Also, since {pk }∞ k=0 sums
up to one, for large n, most of the vertices have a bounded degree, which explains
the phrase sparse random graphs.
We further call a random graph process {Gn }∞ n=1 scale free with exponent τ
when it is sparse and when
log pk
lim =τ (1.143)
k→∞ log (1/k)
exists. Thus, for a scale-free random graph process its degree sequence converges
to a limiting probability distribution as in (1.142), and the limiting distribution
has asymptotic power-law tails described in (1.143). This gives a precise mathe-
matical meaning to a random graph process being scale free. In some cases, the
definition in (1.143) is a bit too restrictive, particularly when the probability
mass function k 7→ pk is not very smooth. Instead, we can also replace it by
log [1 − F (k)]
lim = τ − 1, (1.144)
k→∞ log (1/k)
P
where F (x) = y≤x py denotes the distribution function corresponding to the
probability mass function {pk }∞
k=0 . In particular models, we shall use the version
that is most appropriate in the setting under consideration.
We say that a graph process {Gn }∞ n=1 is highly clustered when

lim CGn = CG∞ > 0. (1.145)

n→∞

We finally define what it means for a graph process {Gn }∞ n=1 to be a small
world. Intuitively, a small world should have distances that are much smaller than
those in a lattice or torus. When we consider the nearest-neighbor torus Tr,d ,
then, and we draw two vertices uniformly at random, their distance will be of the
order r. Denote the size of the torus by n = (2r + 1)d , then the typical distance
between two uniformly chosen vertices is of the order n1/d , so that it grows as a
positive power of n. We shall be dealing with random graph processes {Gn }∞ n=1
for which Gn is not necessarily connected. Let Hn denote the distance between
44 Percolation and random graphs

two uniformly chosen connected vertices, i.e., we pick a pair of vertices uniformly
at random from all pairs of connected vertices, and we let Hn denote the graph
distance between these two vertices. We shall call Hn the typical distance of Gn .
Then, we say that a random graph process {Gn }∞ n=1 is a small world when there
exists a constant K such that

lim P(Hn ≤ K log n) = 1. (1.146)

n→∞

Note that, for a graph with a bounded degree dmax , the typical distance is at least
(1−ε) log n/ log dmax , with high probability, so that a random graph process with
bounded degree is a small world precisely when the order of the typical distance
is optimal.
For a graph Gn , let diam(Gn ) denote the diameter of Gn , i.e., the maximal
graph distance between any pair of connected vertices. Then, we could also have
chosen to replace Hn in (1.146) by diam(Gn ). However, the diameter of a graph
is a rather sensitive object which can easily be changed by making small changes
to a graph in such a way that the scale-free nature and the typical distance Hn
do not change. For example, by adding a sequence of m vertices in a line, which
are not connected to any other vertex, the diameter of the graph becomes at
least m, whereas, if m is much smaller than n, Hn is not changed very much.
This explain why we have a preference to work with the typical distance Hn
rather than with the diameter diam(Gn ).
In some models, we shall see that typical distances can be even much smaller
than log n, and this is sometimes called an ultra-small world. More precisely, we
say that a random graph process {Gn }∞ n=1 is an ultra-small world when there
exists a constant K such that

lim P(Hn ≤ K log log n) = 1. (1.147)

n→∞

There are many models for which (1.147) is satisfied, but for which at the same
time diam(Gn )/ log n converges in probability to a positive limit. This once more
explains our preference to work with the typical graph distance Hn .
We have given precise mathematical definitions for the notions of random
graphs being scale free, highly clustered and small worlds. This has not been
done in the literature so far, and our definitions are based upon a summary of
the relevant results proved for random graph models. We believe it to be a sound
step forward to make the connection between the theory of random graphs and
the empirical findings on real-life networks.
The remainder of this section is organised as follows. In Section 1.3.2, we
study three models for random graphs without geometry that can have rather
general degree sequences, namely, inhomogeneous random graphs, the configu-
ration model and preferential attachment models. We discuss results concerning
the phase transitions and distances in such models. In Section 1.3.5, we shall
discuss random graphs with geometry. The results in Section 1.3.5 are closely
 Random graphs 45

related to percolation questions discussed in Section 1.2. The main distinction

between the random networks discussed in this section and the percolation net-
works discussed in Section 1.2 is that the random graphs discussed here shall be
finite, while the networks in Section 1.2 are all infinite. This raises interesting
new questions, such as how the phase transition can be defined (a cluster can
never be infinite), to what extent the phase transition is unique, and what the
distance between two uniformly chosen vertices is.

1.3.2 Models without geometry

Extensive discussions of scale-free random graphs are given in (Chung and Lu,
2006a; Durrett, 2007), monographs on classical random graphs are (Bollobás,
2001; Janson, Luczak and Rucinski, 2000). We now discuss three particular ex-
amples of random graphs with power-law degree sequences, namely, the inho-
mogeneous random graph, the configuration model, and preferential attachment
models.
(a) Inhomogeneous random graphs. The simplest imaginable random graph is
the so-called Erdős-Rényi random graph, which consists of n vertices and each of
the n(n − 1)/2 edges is present or occupied with probability p, independently of
the occupation status of the other edges. Denote the resulting graph by G(n, p).
This model was introduced by (Gilbert, 1959), while (Erdős and Rényi, 1959)
introduced a model where a fixed number of edges is chosen uniformly at ran-
dom and without replacement. The two models are quite comparable, and most
asymptotic results in one of the two models can easily be transferred to asymp-
totic results in the other. A model with a fixed number of edges being chosen
with replacement, so that possibly multiple edges between vertices arise, can be
found in (Austin, Fagen, Penney and Riordan, 1959). The name Erdős-Rényi
random graph is given to this class of models due to the fact that the first rigor-
ous results were derived in the seminal paper (Erdős and Rényi, 1960), which can
be seen as having founded the field of random graphs, and which has inspired
research questions for decades to follow (see also the books (Bollobás, 2001; Jan-
son, Luczak and Rucinski, 2000)). The above model with independent edges can
be viewed as percolation on the complete graph, the main difference to the theory
in Section 1.2 being that the graph is finite.
One of the charming features of the Erdős-Rényi random graph is the fact
that its vertices are completely exchangeable. For example, every vertex v ∈
[n], where we write [n] = {1, . . . , n} for the vertex set, has a degree that is
distributed as a binomial random variable with parameters n − 1 and p. Thus,
when np → ∞, the average degree tends to infinity, while if np = λ, for some
λ, the average degree remains uniformly bounded. When p = λ/n, the average
degree of each vertex is roughly equal to λ, and the degree is a vertex converges
in distribution to a Poisson random variable with parameter λ. It can be seen
that also the proportion of vertices with degree k, as defined in (1.141), converges
in probability to the Poisson probability mass function pk = e−λ λk /k!. Thus, the
Erdős-Rényi random graph process is sparse.
46 Percolation and random graphs

Note that the tails of a Poisson distribution are quite thin, even subexpo-
nentially thin. As a result, the Erdős-Rényi random graph process is not scale
free. This problem can be overcome by stepping away from the assumption that
the edge probabilities are equal, instead taking them unequal. This is the cel-
ebrated inhomogeneous random graph (IRG), about which the seminal paper
(Bollobás, Janson and Riordan, 2007) proves substantial results in full general-
ity. See also the references in (Bollobás, Janson and Riordan, 2007) for several
examples which have been studied in the literature, and which they generalize.
We shall not go into the precise definition of the model in (Bollobás, Janson and
Riordan, 2007), but rather look at some simpler examples which already allow
for general degree sequences.
To give the general setting, we let G(n, p) denote a general inhomogeneous
random graph, where p = {pij }1≤i<j≤n is such that pij is the probability that
the edge ij = (i, j) is occupied, and where different edges are independent. The
Erdős-Rényi random graph is retrieved when taking pij = p for all ij. We now
generalize the definition in such a way that power-law degree sequences can be
obtained. WePassign weights {wi }ni=1 to the vertices, wi being the weight of vertex
n
i. Let ln = i=1 wi be the total weight. Then, we can take
pij = 1 − e−wi wj /ln . (1.148)
In this way, we retrieve the Poisson random graph (Norros and Reittu, 2006).
Alternatively, for pij = max{wi wj /ln , 1} we retrieve the random graph with given
expected degree as studied in detail by Chung and Lu (Chung and Lu, 2002a;
Chung and Lu, 2002b; Chung and Lu, 2003; Chung and Lu, 2006b; Chung and
Lu, 2006a; Chung, Lu and Vu, 2004). Note that, in this model, if we assume that
maxni=1 wi2 < n, and if we allow for a single self-loop at vertex i with probability
wi2 /ln , the expected degree of vertex i is precisely equal to wi , which explains the
name of this model. A final example occurs when we take pij = wi wj /(ln +wi wj ),
which is called the generalized random graph (Britton, Deijfen and Martin-Löf,
2006). In (Janson, 2008), conditions are given as to when two inhomogeneous
random graph processes {G(n, p)}∞ 0 ∞
n=1 and {G(n, p )}n=1 are equivalent, i.e.,
when events have asymptotically the same probability for G(n, p) and G(n, p0 )
as n → ∞.
Since the expected degree of vertex i is close to wi , in order to obtain a
specified degree sequence in the graph, we need to pick the weights {wi }ni=1
appropriately. In the sequel, we shall take
wi = wi(n) = [1 − F ]−1 (i/n), i ∈ [n], (1.149)
−1
where [1 − F ] is the generalized inverse function of the survival function x 7→
1 − F (x) given by
(1 − F )−1 (u) = inf{s : [1 − F ](s) ≤ u}, u ∈ (0, 1). (1.150)
We call G(n, p), with p as in (1.148) and {wi }ni=1 as in (1.149) the rank-1 in-
homogeneous random graph with deterministic weights according to F . In this
 Random graphs 47

case, the degrees of G(n, p) can be seen to have a mixed Poisson distribution
with mixing distribution F (Britton, Deijfen and Martin-Löf, 2006), i.e., the
asymptotic probability that a uniform vertex has degree k is

W k −W
pk = E[ e ], (1.151)
k!
where W has distribution function F (see, for example, (Britton, Deijfen and
Martin-Löf, 2006, Theorem 3.1) or (van der Hofstad, 2008, Chapter 6)). Thus,
the rank-1 inhomogeneous random graph is sparse. It is not hard to see, using
concentration techniques on a Poisson random variable, that the mixed-Poisson
distribution in (1.151) satisfies (1.143) when W is a continuous random variable
with density fW satisfying
log fW (w)
lim = τ. (1.152)
w→∞ log (1/w)

The condition in (1.152) is sufficient, but not necessary. However, it does illus-
trate that there is a close relation between the tails of a mixed Poisson random
variable and that of its mixing distribution.
Instead of taking the weights deterministic as in (1.149), we can also take wi =
Wi , where {Wi }ni=1 is an i.i.d. sequence of random variables with distribution
function F . We call G(n, p) with p as in (1.148) and {Wi }ni=1 an i.i.d. sequence of
random variables with distribution F , the rank-1 inhomogeneous random graph
with random weights according to F . The advantage of random degrees is that
the vertices in the resulting graph are exchangeable, the disadvantage is that
the edges are no longer independent, their dependence being moderated by their
random weights.
IRGs give a flexible class of models for random graphs with flexible degree
sequences of mixed Poisson form.
(b) The configuration model. The class of IRGs is a flexible class of random
graphs, allowing for rather general degree sequences. One slight disadvantage of
the model is the fact that one cannot specify the degree sequence more precisely.
For example, one cannot have fixed degrees (so-called random regular graphs),
or force the degrees to be at least 1 (which might be appropriate when using the
model for certain real networks). An alternative model, allowing for more general
degree sequences, is the configuration model (CM). We shall first introduce the
model.
Fix an integer n. Consider a sequence of degrees d1 , d2 , . . . , dn . We will con-
struct an undirected graph with P n vertices where vertex j has degree dj . Obvi-
n
ously, we will assume that ln = j=1 dj is even, since in any graph, the total
degree is twice the number of edges.
To construct the graph, we have n separate vertices and incident to vertex
j, we take dj stubs. All stubs need to be connected to another stub to build the
graph. The stubs are numbered in an arbitrary order from 1 to ln . We start by
48 Percolation and random graphs

connecting at random the first stub with one of the ln − 1 remaining stubs. Once
paired, two stubs form a single edge of the graph. We continue the procedure
of randomly choosing and pairing the stubs without replacement, until all stubs
are connected. Unfortunately, vertices having self-loops may occur. However,
self-loops are scarce when n → ∞. See (Janson, 2006) for the best results on
the simplicity of the CM. In general, it turns out that the vector of the number
of cycles of all lengths (i.e., the vector consisting of the number of self-loops,
the number of multiple edges, the number of triangles, the number of squares,
etc) converges in distribution
Pn to an independent sequence of Poisson random
variables, when νn = l1n j=1 dj (dj − 1) converges to a certain ν < ∞, and
√
maxi di = o( n). See also (Bollobás, 2001, Section 2.4). The CM with fixed
degrees has a long history. It was introduced in (Bender and Canfield, 1978) to
study uniform random graphs with a given degree sequence (see also (Bollobás,
2001, Section 2.4)). One specific example is to take the degrees all equal, in which
case we speak of a random regular graph.
Similarly to the rank-1 case of IRG described above, there are two cases
which have been studied for the CM, depending on whether the degrees are de-
terministic or random. In the case of deterministic degrees, a possible choice to
obtain a degree sequence with distribution function F is to take F the distribu-
tion function of an integer random variable, and to take the number of vertices
with degree k to be equal to dnF (k)e − dnF (k − 1)e. We shall call this the con-
figuration model with deterministic degrees according to F . For random degrees,
we can take the degrees {Di }ni=1 to be i.i.d. random variables with distribution
function F , which we shall call the configuration model with random degrees ac-
cordingPto F . In the latter case, unless D1 is even with probability 1, the sum
n
Ln = i=1 Di will be odd with probability exponentially close to 1/2. To avoid
this problem, we can increase the degree of vertex n by 1, so that the degree
of vertex n equals Dn + 1l{Dn is odd} . This hardly makes any difference to the
properties of the graph under consideration, and we will ignore this effect in the
sequel.
In (Britton, Deijfen and Martin-Löf, 2006, Theorem 2.1), it is shown that
for the CM with random degrees according to F , where F is a distribution
function of a random variable with finite (1 + ε)th moment, if we erase all self-
loops and contract all multiple edges to a single edge, the resulting graph is
simple and sparse (recall (1.142)), where pk = F (k) − F (k − 1). Thus, multiple
edges and self-loops only form a small proportion of the total number of edges.
Thus, in order to obtain a scale-free random graph process, we need to take
pk = F (k) − F (k − 1) to satisfy (1.143). In (Britton, Deijfen and Martin-Löf,
2006), also other constructions to create a CM with a prescribed degree sequence
are analyzed.

(c) Preferential attachment models. While inhomogeneous random graphs and

the configuration model can have power-law degree sequences when the edge
probabilities or degrees are chosen appropriately, they do not explain why many
 Random graphs 49

real-world networks are scale free. A possible explanation was given by Barabási
and Albert (Barabási and Albert, 1999) by a phenomenon called preferential
attachment. Preferential attachment models the growth of the network in such
a way that new vertices are more likely to add their edges to already present
vertices with a high degree. For example, in a social network, a newcomer is more
likely to get acquainted to a person who is socially active, and, therefore, is likely
to already possess a high degree. Thus, large degrees are likely to become even
larger, which explains why this model is sometimes called the Rich-get-Richer
model.
Interestingly, in certain cases, preferential attachment models (PAMs) have
power-law degree sequences, and, therefore, preferential attachment offers a con-
vincing explanation why many real-world networks have power-law degree se-
quences. As a result, many papers appeared that study such models. See e.g.
(Aiello, Chung and Lu, 2002; Bollobás, Borgs, Chayes and Riordan, 2003; Bol-
lobás and Riordan, 2003a; Bollobás and Riordan, 2003b; Bollobás and Riordan,
2004a; Bollobás and Riordan, 2004b; Bollobás, Riordan, Spencer and Tusnády,
2001; Cooper and Frieze, 2003) and the references therein. The literature pri-
marily focusses on three main questions. The first is to prove that such random
graphs are indeed scale free (Aiello, Chung and Lu, 2002; Bollobás, Borgs, Chayes
and Riordan, 2003; Bollobás and Riordan, 2003a; Bollobás and Riordan, 2003b;
Bollobás, Riordan, Spencer and Tusnády, 2001; Cooper and Frieze, 2003). The
second is to show that the resulting models are small worlds by investigating the
distances in them. See for example (Bollobás and Riordan, 2004b) for a result
on the diameter. In non-rigorous work, it is often suggested that many of the
scale-free models, such as the CM or IRGs, the models in (Bollobás, Janson and
Riordan, 2007) and the PAMs, have similar properties for their distances. Dis-
tances in the CM have been shown to depend on the number of finite moments
of the degree distribution. The natural question is therefore whether the same
applies to PAMs. A third key question for PAMs is their vulnerability, for ex-
ample to deliberate attack (Bollobás and Riordan, 2003b) or to the spread of a
disease (Berger, Bollobás, Borgs, Chayes and Riordan, 2003).
We now formulate the preferential attachment model, which is an extension of
the Barabási-Albert model, formulated rigorously in (Bollobás, Riordan, Spencer
and Tusnády, 2001), by constructing a graph process {Gm (n)}∞ n=1 . We start by
formulating the model for m = 1, for which we start with G1 (1) consisting of a
single vertex with a single self-loop. We denote the vertices of the graph Gm (n)
by v1(m) , v2(m) , . . .. We denote the degree of vertex vi(m) by Di(n) , where a self-loop
increases the degree by 2.
Then, for m = 1, and conditionally on G1 (n), the growth rule to obtain G1 (n+1)
is as follows. We add a single vertex n + 1 having a single edge. The other end
of this edge is equal to i = n + 1 with probability proportional to 1 + δ, and to
i ∈ [n] with probability proportional to Di(n) + δ, where δ ≥ −1 is a parameter
of the model. Thus,
50 Percolation and random graphs

( 1+δ
(1) (1)

n(2+δ)+(1+δ) , for i = n + 1,
P vn+1 → vi G1 (n) = (n)
Di +δ (1.153)
n(2+δ)+(1+δ) , for i = 1, 2, . . . , n.

The model with integer m > 1, is defined in terms of the model for m = 1 as
follows. We take δ ≥ −m, and then start with G1 (mn), with δ 0 = δ/m ≥ −1 by
identifying the vertices v1(1) , v2(1) , . . . , vm
(1)
in G1 (mn) to be vertex v1(m) in Gm (n),
(1) (1)
and for 1 < j ≤ n, the vertices v(j−1)m+1 , . . . , vjm in G1 (mn) to be vertex vj(m)
in Gm (n); in particular the degree Dj (n) of vertex vj(m) in Gm (n) is equal to the
(1) (1)
sum of the degrees of the vertices v(j−1)m+1 , . . . , vjm in G1 (mn). This defines
the model for integer m ≥ 1. Observe that the range of δ is [−m, ∞).
The resulting graph Gm (n) has precisely mn edges and n vertices at time n, but
is not necessarily connected. For δ = 0, we obtain the original model studied in
(Bollobás, Riordan, Spencer and Tusnády, 2001), and further studied in (Bollobás
and Riordan, 2003b; Bollobás and Riordan, 2004a; Bollobás and Riordan, 2004b).
The extension to δ 6= 0 is crucial in our setting, as we shall explain in more detail
below.
There are several related ways in which we can define the model. For example,
(1)
we disallow self-loops when m = 1 by setting the probability that vn+1 connects
(1)
to vn+1 to be 0 in (1.153), and construct the model for m > 1 as in the paragraph
above. Alternatively, we can let the m edges incident to vertex n to be attached
(m)
independently of each other (in particular, in this case, vertex vn+1 cannot to
connect to itself, so that the graph is connected). For many of the results, this
precise choice is irrelevant, and we shall stick to the model in (1.153). The last
two versions have the nice feature that they lead to connected random graphs.
We continue to discuss the degree sequence of the above preferential attach-
ment model. Recall (1.141) for the definition of the degree sequence. Much of
the available literature on PAMs centers around the proof that the asymptotic
degree sequence obeys a power law, where the exponent τ depends in a sensitive
way on the parameters of the model. Thus, the PAM is scale free. For the PAM
considered here, the power-law exponent equals τ = 3 + δ/m, so that it can take
any value τ ∈ (2, ∞) by adjusting the parameter δ > −m. It is here that we rely
on the choice of the model in (1.153). A form of bias in growing networks towards
vertices with higher degree is, from a practical point of view, quite likely to be
present in various real networks, but it is unclear why the PA scheme should be
affine as in (1.153). However, only affine PA schemes give rise to power-law degree
sequences. See (Oliveira and Spencer, 2005; Rudas, Tóth and Valkó, 2007) for
examples of PAMs with (possibly) non-linear PA-mechanisms and their degree
sequences. We now explain how the affine PA-mechanism in (1.153) gives rise
to power-law degree sequences and highlight the proof. We start by introducing
some notation.
For m ≥ 1 and δ > −m, we define {pk }∞ k=0 to be the probability distribution
given by pk = 0 for k = 0, . . . , m − 1 and, for k ≥ m,
 Random graphs 51

Γ(k + δ)Γ(m + 2 + δ + δ/m)

pk = (2 + δ/m) (1.154)
Γ(m + δ)Γ(k + 3 + δ + δ/m)
Then the main result on the scale-free nature of preferential attachment models
is the following:
Theorem 1.16. (Degree sequence in the PAM) Fix δ > −m and m ≥ 1.
Then, there exists a constant C > 0 such that, as n → ∞,
r

(n) log n 
P max |Pk − pk | ≥ C = o(1). (1.155)
k n
Furthermore, there exists a constant C = C(m, δ) > 0 such that, as k → ∞,
1
pk = Ck −τ (1 + O( )), (1.156)
k
where
δ
τ =3+ > 2. (1.157)
m
In particular, Theorem 1.16 implies that the PA-random graph process is scale
free. Theorem 1.16 appears in many forms in various settings. The statement
which is closest to Theorem 1.16 is (Deijfen, van den Esker, van der Hofstad and
Hooghiemstra, 2007, Theorem 1.3), where also the setting where each vertex
enters the graph process with a random number of edges is considered. The first
proof of a result as in Theorem 1.16 appeared in (Bollobás, Riordan, Spencer
and Tusnády, 2001), they show a slightly weaker version of Theorem 1.16 when
δ = 0.
Virtually all proofs of asymptotic power laws in preferential attachment mod-
els consist of two steps: one step where it is proved that the degree sequence is
concentrated around its mean, and one where the mean degree sequence is iden-
tified. We shall now give an intuitive explanation of Theorem 1.16.
Let Nk(n) = nPk(n) be the number of vertices with degree k in Gm (n) (recall
(1.141)). We are interested in the limiting distribution of Pk(n) as n → ∞. This
distribution arises as the solution of a certain recurrence relation, of which we
will now give a short heuristic derivation. First note that, obviously,

E[Nk(n+1) |Gm (n)] = Nk(n) + E[Nk(n+1) − Nk(n) |Gm (n)]. (1.158)

Asymptotically, for n large, it is quite unlikely that a vertex will be hit by more
than one of the m edges added upon the addition of vertex n. Let us hence
ignore this possibility for the moment. The difference Nk(n+1) − Nk(n) between the
number of vertices with degree k at time n + 1 and time n respectively, is then
obtained as follows:
(a) Vertices with degree k in Gm (n) that are hit by one of the m edges em-
anating from vertex n are subtracted from Nk(n) . The conditional prob-
ability that a fixed edge is attached to a vertex with degree k is (k +
52 Percolation and random graphs

δ)Nk(n) /(n(2m + δ)), so that (ignoring multiple attachments to a single ver-

tex) the mean number of vertices to which this happens is approximately
m(k + δ)Nk(n) /(n(2m + δ)). We note that we have replaced the numerator,
which is n(2m + δ) + (e − 1)(2 + δ/m) + 1 + δ in the attachment of the
eth edge emanating from vertex n, by its approximate value n(2m + δ) for
large n.
(b) Vertices with degree k − 1 in Gm (n) that are hit by one of the m edges
emanating from vertex n are added to Nk(n) . By reasoning as above, it
follows that the mean number of such vertices is approximately m(k − 1 +
δ)Nk(n) /(n(2m + δ)).
(c) The new vertex n should be added if it has degree k. When we ignore the
case that vertex n attaches edges to itself, this happens precisely when
k = m.
Combining this gives
  m(k − 1 + δ) (n) m(k + δ) (n)
E Nk(n+1) − Nk(n) |Gm (n) ≈ Nk−1 − N + 1l{k=m} .
n(2m + δ) n(2m + δ) k
(1.159)

Substituting (1.159) into (1.158) and taking expectations, we arrive at

  m(k − 1 + δ) m(k + δ)
E Nk(n+1) ≈ E[Nk(n) ] + (n)
E[Nk−1 ]− E[Nk(n) ] + 1l{k=m} .
n(2m + δ) n(2m + δ)
(1.160)
Now assume that Pk(n) converges to some limit pk as n → ∞, so that hence
Nk(n) ∼ npk . Then, for n → ∞, and observing that E[Nk(n+1) ] − E[Nk(n) ] → pk
implies that n1 E[Nk(n) ] → pk , for all k, yields the recursion

m(k − 1 + δ) m(k + δ)
pk = pk−1 − pk + 1l{k=m} , (1.161)
2m + δ 2m + δ
or, equivalently,

k−1+δ 2 + δ/m
pk = pk−1 + 1l{k=m} , (1.162)
k + 2 + δ + δ/m k + 2 + δ + δ/m

By iteration, it can be seen that this recursion is solved by pk = 0 when k < m

and
k
2 + δ/m Y k−j+δ
pk = , k ≥ m. (1.163)
k + 2 + δ + δ/m j=m k − j + 2 + δ + δ/m

By rewriting the products in terms of Gamma-functions, we see that (1.163)

is equal to (1.154). It is not hard to see that, when k → ∞, pk ∼ Ck −τ with
τ = 3 + δ/m. This explains the occurrence of power-law degree sequences in
 Random graphs 53

affine PAMs. The above argument can be made rigorous in order to show that
maxk |E[Nk(n) ] − npk | remains uniformly bounded (see e.g., (Deijfen, van den
Esker, van der Hofstad and Hooghiemstra, 2007)).
In order to prove concentration of Nk(n) , all proofs in the literature make use
of a clever martingale argument from (Bollobás, Riordan, Spencer and Tusnády,
2001). Define the Doob martingale Mt by

Mt = E[Nk(n) |Gm (t)]. (1.164)

Then, M0 = E[Nk(n) ] while Mn = Nk(n) , so that Nk(n) − E[Nk(n) ] = Mn − M0 . The

key ingredient is the observation that, for all t ∈ [n], |Mt − Mt−1 | ≤ 2m a.s.,
since the only vertices that are affected by the information of Gm (t) instead of
Gm (t − 1) are the vertices affected by the attachment of the edges incident to
vertex t. Together, the concentration and the asymptotic mean give that Gm (n)
has an asymptotic degree sequence {pk }k≥m , where pk is close to a power-law
for k large.
A prediction of universality. In non-rigorous work, it is often suggested the var-
ious scale-free random graph models, such as the CM or various models with con-
ditional independence of the edges as in (Bollobás, Janson and Riordan, 2007),
behave similarly. For scale-free random graph processes, this informal statement
can be made precise by conjecturing that the phase transition or distances have
the same behavior in graphs with the same power-law degree exponent. We shall
discuss some of the results in this direction below.

1.3.3 Phase transition in models without geometry

In this section, we study the phase transition in random graphs. We first intro-
duce some notation. For the CM with deterministic or random degrees according
to F , we define
E[D(D − 1)]
ν= , (1.165)
E[D]
where D has distribution function F . For the rank-1 inhomogeneous random
graph with deterministic or random weights according to F , we define

E[W 2 ]
ν= , (1.166)
E[W ]
where W has distribution function F . For the PAM, we let ν = m. Below, we
say that a sequence of events {En }∞
n=1 occurs with high probability (whp) when
limn→∞ P(En ) = 1. It turns out that ν = 1 plays the role of a critical value for
all these random graphs:
Theorem 1.17. (Phase transition in random graphs) (a-b) For the config-
uration model with deterministic or random degrees according to F , and
for the rank-1 inhomogeneous random graph with deterministic or random
weights according to F , the largest connected component has, whp, size
54 Percolation and random graphs

o(n) when ν ≤ 1, and size ζn(1 + o(1)) for some ζ > 0 when ν > 1, where
n is the size of the graph.
(c) For the PAM of size n, whp, the largest connected component has size o(n)
when ν = m = 1, while the probability that the PAM is connected converges
to 1 for n → ∞ when ν = m > 1.
We write Theorem 1.17(a-b) to indicate that the result holds both for the IRG
(which is model (a)) and for the CM (which is model (b)). This notation shall
be used frequently below.
The result for IRG is a special case of (Bollobás, Janson and Riordan, 2007,
Theorem 3.1). Earlier versions for the random graph with given expected de-
grees appeared in (Chung and Lu, 2002b; Chung and Lu, 2006b) (see also the
monograph (Chung and Lu, 2006a)). For the CM, the first result in the gener-
ality of Theorem 1.17 appeared in (Molloy and Reed, 1995; Molloy and Reed,
1998) under stronger conditions than mentioned here. For the sharpest result,
see (Janson and Luczak, 2007). The connectivity of PAMs was investigated for
δ = 0 in (Bollobás and Riordan, 2004b), it was extended to all δ > −m in (van
der Hofstad, 2008, Chapter 11).
In (Deijfen, van den Esker, van der Hofstad and Hooghiemstra, 2007), also
the setting where the number of edges with which a vertex enters the random
graph is random. Indeed, denote the number of edges of vertex t by Wt , then in
(Deijfen, van den Esker, van der Hofstad and Hooghiemstra, 2007) it is assumed
that {Wt }∞t=1 is an i.i.d. sequence. In general, such models are also scale-free with
power-law exponent τ = min{τP , τW }, where τP = 3 + δ/µ and µ = E[Wt ] is the
“PA exponent”, while τW is the power-law exponent of the weight distribution,
i.e.,
log P(W1 ≥ k)
lim = τW − 1. (1.167)
k→∞ log (1/k)
Thus, one can summarize this by the fact that the effect with the least corre-
sponding power-law exponent determines the power-law exponent of the graph.
It would be of interest to study the phase transition for such more general models,
and to verify r under what condition a giant component exists.

1.3.4 Distances in models without geometry

In this section, we summarize the results on distances in power-law random
graphs. We combine the results in the three models discussed in Section 1.3.2
by the value of their respective power-law exponent. We define Hn to be the
typical distance in the graph of size n, i.e., the number of edges in the shortest
path between two uniformly chosen connected vertices. Note that even in a fixed
graph, Hn is a random variable, as it depends on the uniform vertices which are
chosen. We shall also discuss results on the diameter of the graph, which is the
maximum of the shortest path distances between any pair of connected vertices.
Both give information about distances in graphs, the typical distance being a
more robust and informative feature of the graph than the diameter.
 Random graphs 55

Distances in random graphs with finite variance degrees. The main results on
distances in power-law random graphs with power-law exponent τ > 3 are sum-
marized in the following theorem:
Theorem 1.18. (Distances in graphs with finite variance degrees) (a-b)
For the configuration model and the rank-1 inhomogeneous random graph
of size n, Hn / log n converges in probability to 1/ log ν, where ν is given
by (1.165) for the CM, and by (1.166) for the rank-1 IRG, when F in the
definition of the models satisfies that there exist c > 0 and τ > 3 such that

1 − F (x) ≤ cx−(τ −1) . (1.168)

(c) For the affine PAM of size n with δ > 0, so that τ = 3 + δ/m > 3, whp,
Hn / log n is bounded above and below by positive and finite constants.
The result for the rank-1 IRG can be found in (van den Esker, van der
Hofstad and Hooghiemstra, 2006a), where it is also shown that the fluctuations
of Hn around logν n remain bounded in probability, both in the case of i.i.d.
degrees as well as for deterministic weights under a mild further condition on
the distribution function. The first result in this direction was proved in (Chung
and Lu, 2002a; Chung and Lu, 2003) for the expected degree random graph, in
the case of rather general deterministic weights. A special case of the IRG with
finite variance degrees is the Erdős-Rényi random graph with edge probability
p = λ/n, for which ν = λ.
The result for the CM can be found in (van der Hofstad, Hooghiemstra and
Van Mieghem, 2005) in the case of i.i.d. degrees, where again also the fluctuations
are determined. The results for deterministic degrees in the CM is conjectured
in (van den Esker, van der Hofstad and Hooghiemstra, 2006a), but is not proved
anywhere. We expect that the methodology in (van der Hofstad, Hooghiemstra
and Van Mieghem, 2005) can be simply adapted to this case. The result for the
affine PAM was proved in (van der Hofstad and Hooghiemstra, 2007a). Unfortu-
nately the proof of convergence in probability is missing in this case. It would be
of interest to identify the constant to which Hn / log n converges in this setting.
Distances in random graphs with finite mean and infinite variance degrees. When
τ ∈ (2, 3), the variance of the degrees becomes infinite, which is equivalent to
thePstatement that, with Di(n) denoting the degree of vertex i in the graph of size
n
n, i=1 (Di(n) )2 grows much faster than n. The following theorem shows that, in
such cases, the distances are much smaller than log n:
Theorem 1.19. (Distances in graphs with τ ∈ (2, 3)) (a-b) For the config-
uration model and the rank-1 inhomogeneous random graph, Hn / log log n
converges in probability to 2/| log (τ − 2)|, when F in the definition of the
models satisfies that there exist c > 0 and τ ∈ (2, 3) such that

1 − F (x) = cx−(τ −1) (1 + o(1)). (1.169)

56 Percolation and random graphs

(c) For the affine PAM with δ < 0 and m ≥ 2, so that τ = 3 + δ/m ∈ (2, 3),
whp Hn / log log n is bounded above by a finite constant.
Theorem 1.19 shows that all three models are ultra-small worlds when the power-
law exponent τ satisfies τ ∈ (2, 3).
The result for the rank-1 IRG is proved in (Chung and Lu, 2002a; Chung
and Lu, 2003) for the expected degree random graph, in the case of certain
deterministic weights. The result for the CM can be found in (van der Hofstad,
Hooghiemstra and Znamenski, 2007) in the case of i.i.d. degrees, where again also
the fluctuations are determined and are proved to be bounded. The restrictions
on F are somewhat weaker than (1.169), as they also allow x 7→ xτ −1 [1 − F (x)]
to be slowly varying under certain conditions on the regularly varying function.
The results in (Fernholz and Ramachandran, 2007) mentioned earlier apply in
this case as well, and show that, when the proportion of vertices with degrees
1 and 2 is positive, the diameter divided by log n converges in probability to a
positive constant. In (van der Hofstad and Hooghiemstra, 2007b), it is shown
that the diameter in the CM is bounded above by a constant times log log n
when there are no vertices of degree 1 and 2.
The result for the affine PAM was proved in (van der Hofstad and Hooghiem-
stra, 2007a). Again, it would be of interest to identify the constant to which
Hn / log log n converges in this setting.
For the affine PA-model, we refer to (Bollobás and Riordan, 2004b) for a
proof of the fact that the diameter of the PA-model diam(Gm (n)), for δ = 0,
satisfies that diam(Gm (n)) × log log n/ log n converges in probability to 1. This
result is much sharper than the ones in Theorem 1.18(c) and Theorem 1.19(c),
and it would be of interest to investigate whether the methodology used there
can be adapted to the case where δ 6= 0.
Distances in random graphs with infinite mean degrees. Only in the CM and the
IRG, it is possible that the power-law exponent of the degrees or the weights of
the vertices τ satisfies τ ∈ (1, 2). In general, this is not very realistic, as it means
that either there are extremely many multiple edges (in the CM) or the power-
law exponent in the graph does not match the value of τ (in the IRG). Distances
in infinite mean random graphs have been studied in (van den Esker, van der
Hofstad, Hooghiemstra and Znamenski, 2006b; Norros and Reittu, 2006) and
show that distances remain uniformly bounded by three. The intuition behind
this is clear: all vertices are connected to vertices with extremely high degree,
and these vertices form a complete graph, so that typical distances are at most
3. We refrain from a further discussion of random graphs with infinite mean
degrees.
Conclusion on phase transition and distances. The main tool in order to study
the phase transition and distances in the CM and IRGs is a comparison of the
neighborhood of a vertex to a two-stage (multitype) branching process. In order
to prove distance results, one then has to further investigate the growth of the
number of vertices at a given distance using limit laws for branching processes.
 Random graphs 57

When (1.168) holds, the number of vertices at a given distance k grows propor-
tionally to ν k , which suggests that distances are of the order logν n, as stated in
Theorem 1.18.
Specifically, for the CM, with deterministic or i.i.d. degrees, the two-stage
branching process, which we denote by {Zk }k≥0 , starts from Z0 = 1, has offspring
distribution {fn }∞ n=1 , where fn = F (n) − F (n − 1) are the jump sizes of the
distribution F , in the first generation, and offspring distribution

(n + 1)fn+1
gn = , n ≥ 0, (1.170)
E[D]

in the second and further generations. It is not hard to verify that the parameter ν
in (1.165) is the expectation of the size-biased distribution {gn }∞
n=0 . For the rank-
1 IRG with deterministic or random weights, the branching process {Zk } has a
mixed-Poisson distribution with random parameter W in the first generation
and a mixed Poisson distribution with random parameter We , which has the
size-biased distribution of W , in the second and further generations. Thus, when
W has a continuous density w 7→ f (w), the density of We is equal to fe (w) =
wf (w)/E[W ]. It can be seen that these two mixed Poisson distribution are again
related through (1.170), and that again the parameter ν in (1.166) equals the
expectation of the size-biased distribution used as offspring distribution in the
second and further generations.
The condition ν > 1 assures that the branching process {Zk } is supercritical,
so that it can grow to a large size with positive probability (recall Theorem 1.17).
Intuitively, all vertices for which the connected component are large (say larger
than nε for some ε > 0) are connected and thus form a single giant component.
The constant ζ in Theorem 1.17 is the survival probability of the two stage
(multi-type) branching process {Zk }k≥0 . Now, for a branching process {Zk }k≥0
for which the offspring distribution has finite (1 + ε)th moment, we have that
Zk ν −k converges a.s. to a limiting random variable W that is not identically 0
(by the Kesten-Stigum Theorem). The core of the proof is to use that Hn is
the graph distance between two uniformly chosen connected vertices V1 and V2 .
Then, the neighborhood shells Zk(1,n) and Zk(2,n) consist of those vertices that are
at graph distance precisely equal to k from V1 and V2 , respectively. Hn is equal
(1,n)
to kn +1, where kn is the first time that any of the vertices in Zbk n /2c
connects to
(2,n)
a vertex in Zdkn /2e . The proof is then completed by coupling these neighborhood
shells to two independent two-stage branching processes as described above.
When (1.169) holds, then, by results of (Davies, 1978), the growth is super-
exponential, i.e., (τ −2)k log (Zk + 1) converges a.s. to a limiting random variable
Y , where Y > 0 precisely when the branching process survives. Thus, condi-
tionally on Y = y > 0, the number of individuals in generation k grows like
−k
e(τ −2) y(1+o(1)) suggesting that distances are of order log log n/| log(τ − 2)|.
The additional factor 2 in Theorem 1.19 is due to the fact that in order for two
vertices to meet, each of their neighborhoods needs to have size at least nε for
58 Percolation and random graphs

some ε > 0, each of which can be expected to occur in a generation k with

k ≈ log log n/| log(τ − 2)|.

1.3.5 Models with geometry

In this section, we study some models of finite or bounded random graphs with
geometry. We note that two of such models have been considered already in
Section 1.2 as they are close relatives of the (infinite) percolation models. Indeed,
in Section 1.2.5, the critical nature of percolation on high-dimensional tori was
discussed, while in Section 1.2.8, random geometric graphs, which can be viewed
as the restriction of continuum percolation to a bounded domain, have been
described. In Section 2.5, models for random directed and on-line networks were
discussed.
We now start by studying a spatial model for small worlds:
Small-world networks. The models described in Section 1.3.2 have flexible de-
gree sequences and small distances, but they tend not to be very highly clustered.
Also, these models do not incorporate geometry at all. An alternative approach
of explaining the small-world phenomenon is to start with a finite torus, and
to add random long range connections to them, independently for each pair of
vertices. This gives rise to a graph which is a small perturbation of the original
lattice, but has occasional long range connections that are crucial in order to
shrink distances. From a practical point of view, we can think of the original
graph as being the local description of acquaintances in a social network, while
the shortcuts describe the occasional acquaintances in the population living far
apart. The main idea is that, even though the shortcuts only form a tiny part of
the connections in the graph, they are crucial in order to make it a small world.
There are various ways of adding long-range connections (for example by
rewiring the existing edges), and we shall focus on the models in (Barbour and
Reinert, 2001; Barbour and Reinert, 2004; Barbour and Reinert, 2006), for which
the strongest mathematical results have been obtained. Small-world models were
first introduced and analyzed in (Moore and Newman, 2000; Newman, Moore and
Watts, 2000; Newman and Watts, 1999), and a non-rigorous mean-field analysis
of distances in small-world models was performed in (Newman, Moore and Watts,
2000). See (Barbour and Reinert, 2001) for a discussion of the differences between
the exact and mean-field analyses.
The simplest version of the model studied in (Barbour and Reinert, 2001) is
obtained by taking the circle of circumference n, and adding a Poisson number of
shortcuts with parameter nρ/2, where the starting and endpoints of the shortcuts
are chosen uniformly at random independently of each other. This model is
called the continuous circle model in (Barbour and Reinert, 2001). Distance is
measured as usual along the circle, and the shortcuts have, by convention, length
zero. Thus, one can think of this model as the circle where the points along the
random shortcut are identified, thus creating a puncture in the circle. Multiple
shortcuts then lead to multiple puncturing of the circle, and the distance is
then the usual distance along the punctured graph. Denote by Dn the distance
 Random graphs 59

between two uniformly chosen points along the punctured circle. Then, (Barbour
and Reinert, 2001, Theorem 3.9) states that as n → ∞, Dn (2ρ)/ log ρn converges
in probability to 1 when nρ → ∞, and that ρ(Dn − log ρn/2) converges in
distribution to a random variable T satisfying
Z ∞ −y
e dy
P(T > t) = . (1.171)
0 1 + e2t
The random variable T can also be described by
2t
W (1) W (2)
P(T > t) = E[e−e ], (1.172)

where W (1) , W (2) are two independent exponential random variables with param-
eter 1. Alternatively, it can be see that T = (G1 + G2 − G3 )/2, where G1 , G2 , G3
are three independent Gumbel distributions (see (Barbour and Reinert, 2006,
Page 1242)).
Interestingly, the method of proof of (Barbour and Reinert, 2001, Theorem
3.9) is quite close to the method of proof for Theorem 1.18. Indeed, again the
parts of the graph that can be reached in distance at most t are analyzed. Let
P1 and P2 be two uniform points along the circle, so that Dn has the same
distribution as the distance between P1 and P2 . Denote by R(1) (t) and R(2) (t)
the parts of the graph that can be reached within distance t. Then, Dn = 2Tn ,
where Tn is the first time that R(1) (t) and R(2) (t) have a non-zero intersection.
The proof then consists of showing that, up to time Tn , the processes R(1) (t) and
R(2) (t) are close to certain continuous-time branching processes, primarily due to
the fact that the probability that there are two intervals that are overlapping in
quite small. Then, W (1) and W (2) can be viewed as appropriate martingale limits
of these branching processes. In (Barbour and Reinert, 2001, Theorem 4.2), also
an extension to higher dimensions is given.
The proof was extended in (Barbour and Reinert, 2006) to deal with discrete
tori where the shortcuts also contribute one to the graphs distance, so that
distances are the usual distances on discrete graphs. For this, it was necessary
that the average number of shortcuts per vertex ρ ↓ 0, a restriction that does
not appear in (Barbour and Reinert, 2001). It would be of interest to extend the
results to the case of fixed ρ as well in the discrete setting.
A related model was considered in (Turova and Vallier, 2006). Indeed, (Turova
and Vallier, 2006) studies a mixture between subcritical percolation on a finite
cube and the Erdős-Rényi random graph. Using the methodology in (Bollobás,
Janson and Riordan, 2007), it is shown that the phase transition is similar to
the one described in Theorem 1.17. It would be of interest to verify whether the
distance results in (Bollobás, Janson and Riordan, 2007) can also be used to
prove that the distances grow like logν n, where n is the size of the graph, and
ν > 1 an appropriate constant.
A scale-free percolation network. In this section, we discuss the results in (Yu-
kich, 2006) on an infinite scale-free percolation model. Note that, for a transitive
60 Percolation and random graphs

graph with fixed degree r and percolation with a fixed percolation parameter p,
the degree of each vertex has a binomial distribution with parameters r and p.
Since r is fixed, this does not allow for a power-law degree sequence. As a result,
it is impossible to have a scale-free random graph when dealing with indepen-
dent percolation, so that we shall abandon the assumption of independence of
the different edges, while keeping the assumption of translation invariance.
The model considered in (Yukich, 2006) is on Zd , and, thus, the definition
of a scale-free graph process does not apply so literally. We adapt the definition
slightly by saying that an infinite random graph is scale-free when
pk = P(Do = k), (1.173)
where Dx is the degree of vertex x ∈ Zd and o ∈ Zd is the origin, satisfies (1.143).
This is a reasonable definition, since if let Br = [−r, r]d ∩ Zd be a cube of width
r around the origin, and denote n = (2r + 1)d , then, for each k ≥ 0,
1 X
Pk(n) = 1l{Dx =k} , (1.174)
n
x∈Br

which, assuming translation invariance and ergodicity, converges to pk .

We next describe the model in (Yukich, 2006). We start by taking an i.i.d.
sequence {Ux }x∈Zd of uniform random variables on [0, 1]. Fix δ ∈ (0, 1] and
q ∈ (1/d, ∞). The edge {x, y} ∈ Zd × Zd appears in the random graph precisely
when
|x − y| ≤ δ min{Ux−q , Uy−q }. (1.175)
We can think of the ball of radius δUx−q as being the region of influence of x, and
two vertices are connected precisely when each of them lies into the region of
influence of the other. This motivates the choice in (1.175). The parameter δ can
be interpreted as the probability that nearest-neighbors are connected, and in
the sequel we shall restrict ourselves to δ = 1, in which case the infinite connected
component equals Zd . We denote the resulting (infinite) random graph by Gq .
We next discuss the properties of this model, starting with its scale-free
nature. In (Yukich, 2006, Theorem 1.1), it is shown that, with τ = qd/(qd − 1) ∈
(1, ∞), the limit
lim k τ −1 P(Do ≥ k) (1.176)
k→∞

exists, so that the model is scale free with degree power-law exponent τ (recall
(1.144)). The intuitive explanation of (1.176) is as follows. Suppose we condition
on the value of Uo = u. Then, the conditional distribution of Do given that
Uo = u is equal to
X X
Do = 1l{|x|≤min{Uo−q ,Ux−q }} = 1l{|x|≤Ux−q } . (1.177)
x∈Zd x:|x|≤u−q

Note that the random variables {1l{|x|≤Ux−q } }x∈Zd are independent Bernoulli ran-
dom variables with probability of success equal to
 Random graphs 61

P(1l{|x|≤Ux−q } = 1) = P(U ≤ |x|−1/q ) = |x|−1/q . (1.178)

In order for Do ≥ k to occur, for k large, we must have that Uo = u is quite

small, and, in this case, a central limit theorem should hold for Do , with mean
equal to
X
E[Do |Uo = u] = |x|−1/q = cu−(qd−1) (1 + o(1)), (1.179)
x:|x|≤u−q

for some explicit constant c = c(q, d). Furthermore, the conditional variance of
Do given that Uo = u is bounded above by its conditional expectation, so that
the conditional distribution of Do given that Uo = u is highly concentrated.
We omit the detail, and merely note that this can be made precise by using
standard large deviations result. Assuming sufficient concentration, we obtain
that the probability that Do ≥ k is asymptotically equal to the probability that
U ≤ uk , where uk is determined by the equation that
−(qd−1)
E[Do |Uo = uk ] = cuk (1 + o(1)) = k, (1.180)

so that uk = (k/c)−1/(qd−1) . This suggests that

P(Do ≥ k) = P(U ≤ uk )(1 + o(1)) = (ck)−1/(qd−1) (1 + o(1)), (1.181)

which explains (1.176).

We next turn to distances in this scale-free percolation model. For x, y ∈ Zd ,
we denote by dGq (x, y) the graph distance (or chemical distance) between the
vertices x and y, i.e., the minimal number of edges in Gq connecting x and y.
The main result in (Yukich, 2006) is the following theorem:
Theorem 1.20. (Ultra-small distances for scale-free percolation) For all
d ≥ 1 and all q ∈ (1/d, ∞), whp as |x| → ∞,

dGq (o, x) ≤ 8 + 4 log log |x|. (1.182)

The result in Theorem 1.20 shows that distances in the scale-free percolation
model are much smaller than those in normal percolation models. It would be
of interest to investigate whether the limit dGq (o, x)/ log log |x| exists, and, if so,
what this limit is.
While Theorem 1.20 resembles the results in Theorem 1.19, there are a few
essential differences. First of all, Gq is an infinite graph, whereas the models con-
sidered in Theorem 1.19 are all finite. It would be of interest to extend Theorem
1.20 to the setting on finite tori, where the Euclidean norm |x − y| in (1.175)
is replaced by the Euclidean norm on the torus, and the typical distance Hn is
considered. This result is not immediate from the proof of Theorem 1.20. Sec-
ondly, in Theorems 1.18 and 1.19, it is apparent that the behavior for τ > 3 is
rather different compared to the behavior for τ ∈ (2, 3). This feature is missing
62 Percolation and random graphs

in Theorem 1.20. It would be of interest to find a geometric random graph model

where the difference in behavior between τ > 3 and τ ∈ (2, 3) also appears.
The result in Theorem 1.20 can be compared to similar results for long-range
percolation, where edges are present independently, and the probability that the
edge {x, y} is present equals |x − y|−s+o(1) for some s > 0. In this case, detailed
results exist for the limiting behavior of d(o, x) depending on the value of s. For
example, in (Benjamini, Kesten, Peres and Schramm, 2004), it is shown that the
diameter of this infinite percolation model is equal to dd/(d − s)e a.s. See also
(Biskup, 2004) and the references therein.
Spatial preferential attachment models. In the past years, several spatial pref-
erential attachment models have been considered. We shall now discuss three of
such models.
In (Flaxman, Frieze and Vera, 2006; Flaxman, Frieze and Vera, 2007), a class
of geometric preferential attachment models that combines aspects of random
geometric graphs and preferential attachment graphs is introduced and studied.
Let Gt = (Vt , Et ) denote the graph at time t. Let S be the sphere S in R3 with
area equal to 1. Then, we let Vt be a subset of S of size t.
The process {Gt }∞t=0 evolves as follows. At time t = 0, G0 is the empty graph.
At time t + 1, given Gt , we obtain Gt+1 as follows. Let xt+1 be chosen uniformly
at random from S, and denote Vt+1 = Vt ∪ {xt+1 }. We assign m edges to the
vertex xt+1 , which we shall connect independently of each other to vertices in
Vt (xt+1 ) ≡ Vt ∩ Br (xt+1 ), where Br (u) = {x ∈ S : kx − uk ≤ r} denotes the
spherical cap of radius r around u. Let
X
Dt (xt+1 ) = Dv(t) , (1.183)
v∈Vt (xt )

where Dv(t) denotes the degree of vertex v ∈ Vt in Gt . The m edges are connected
to vertices (y1 , . . . , ym ) conditionally independently given (Gt , xt+1 ), so that, for
all v ∈ Vt (xt+1 ),
Dv(t)
P(yi = v) = , (1.184)
max(Dt (xt+1 ), αmAr t)
while
Dt (xt+1 )
P(yi = xt+1 ) = 1 − , (1.185)
max(Dt (xt+1 ), αmAr t)
where Ar is the area of Br (u), α ≥ 0 is a parameter, and r is a radius which shall
be chosen appropriately. Similarly to the situation of geometric random graphs,
the parameter r shall depend on the size of the graph, i.e., we shall be interested
in the properties of Gn when r = rn is chosen appropriately. The main result
in (Flaxman, Frieze and Vera, 2006) is the study of the degree sequence of the
arising model. Take rn = nβ−1/2 log n, where β ∈ (0, 1/2) is a constant. Finally,
let α > 2. Then, there exists a probability distribution {pk }∞
k=m such that, whp,

Pk(n) = pk (1 + o(1)), (1.186)

 Random graphs 63

where {pk }∞k=m satisfies (1.143) with τ = 1 + α ∈ (3, ∞). The precise result is in
(Flaxman, Frieze and Vera, 2006, Theorem 1(a)) and is quite a bit sharper, as
detailed concentration results are proved as well. Further results involve the proof
of connectivity of Gn and an upper bound on the diameter when r ≥ n−1/2 log n,
m ≥ K log n for some large enough K and α ≥ 0 of order O(log (n/r)). In
(Flaxman, Frieze and Vera, 2007), these results were generalized to the setting
where, instead of a unit ball, a smoother version is used.
In (Aiello, Bonato, Cooper, Janssen and Pralat, 2007), a spatial preferential
attachment model with local influence regions is studied, as a model for the
Web graph. The model is directed, but it can be easily adapted to an undirected
setting. The idea behind the model in (Aiello, Bonato, Cooper, Janssen and
Pralat, 2007) is that for normal preferential attachment models, new vertices
should be aware of the degrees of the already present vertices. In reality, it is
quite hard to observe the degrees of vertices, and, therefore, in (Aiello, Bonato,
Cooper, Janssen and Pralat, 2007), vertices instead have a region of influence
in some metric space, for example the torus [0, 1]m for some dimension m, for
which the metric equals
d(x, y) = min{kx − y + uk∞ : u ∈ {0, 1, −1}m }. (1.187)
When the new vertex arrives, it is uniformly located somewhere in the unit cube,
and it connects to each of the older vertices in which region of influence they land
independently and with fixed probability p. These regions of influence evolve as
time proceeds, in such a way that the volume of the influence region of the vertex
i at time t is equal to
A1 Di(t) + A2
R(i, t) = , (1.188)
t + A3
where now Di(t) is the in-degree of vertex i at time t, and A1 , A2 , A3 are param-
eters which are chosen such that pA1 ≤ 1. One of the main results of the paper
is that this model is a scale-free graph process. Indeed, denote
k−1
pk Y jA1 + A2
pk = , (1.189)
1 + kpA1 + pA2 j=0 1 + A2 + pA2

then (Aiello, Bonato, Cooper, Janssen and Pralat, 2007, Theorem 1.1) shows
that whp, for k ≤ (n1/8 / log n)4pA1 /(2pA1 +1) , the degree sequence of the graph
of size n satisfies (recall (1.141))
Pk(n) = pk (1 + o(1)), (1.190)
and {pk }∞k=0 satisfies (1.143) with τ = 1 + 1/(pA1 ) ∈ [2, ∞). Further results
involve the study of maximal in-degrees and the total number of edges.
For a relation between preferential attachment graphs with so-called fertility
and aging, and a geometric competition-induced growth model for networks, we
refer to (Berger, Borgs, Chayes, D’Souza and Kleinberg, 2004; Berger, Borgs,
Chayes, D’Souza and Kleinberg, 2005) and the references therein.
64 Percolation and random graphs

1.4 Acknowledgements
The work of RvdH was supported in part by Netherlands Organisation for Sci-
entific Research (NWO). I am grateful to Gordon Slade for introducing me to
the field of percolation, which has proved to be an extremely beautiful and rich
world indeed! It has always been a delight to attack percolation problems with
Gordon, and I have enjoyed this dearly. I warmly thank Gerard Hooghiemstra
for jointly taking the first steps (and then many more) in the field of random
graphs. Let’s take plenty more steps together!
 REFERENCES

Aiello, W., Bonato, A, Cooper, C., Janssen, J., and Pralat, P. (Preprint (2007)).
A spatial web graph model with local influence regions. To appear in Internet
Mathematics.
Aiello, W., Chung, F., and Lu, L. (2002). Random evolution in massive graphs.
In Handbook of massive data sets, Volume 4 of Massive Comput., pp. 97–122.
Kluwer Acad. Publ., Dordrecht.
Aizenman, M. (1997). On the number of incipient spanning clusters. Nucl.
Phys. B , 485, 551–582.
Aizenman, M. and Barsky, D.J. (1987). Sharpness of the phase transition in
percolation models. Commun. Math. Phys., 108, 489–526.
Aizenman, M., Kesten, H., and Newman, C.M. (1987). Uniqueness of the infi-
nite cluster and continuity of connectivity functions for short and long range
percolation. Commun. Math. Phys., 111, 505–531.
Aizenman, M. and Newman, C.M. (1984). Tree graph inequalities and critical
behavior in percolation models. J. Stat. Phys., 36, 107–143.
Albert, R. and Barabási, A.-L. (2002). Statistical mechanics of complex net-
works. Rev. Modern Phys., 74(1), 47–97.
Alon, N. and Spencer, J. (2000). The probabilistic method (Second edn). Wiley-
Interscience Series in Discrete Mathematics and Optimization. John Wiley &
Sons, New York.
Athreya, K. and Ney, P. (1972). Branching processes. Springer-Verlag, New
York. Die Grundlehren der mathematischen Wissenschaften, Band 196.
Austin, T. L., Fagen, R. E., Penney, W. F., and Riordan, J. (1959). The number
of components in random linear graphs. Ann. Math. Statist, 30, 747–754.
Baccelli, F. and Blaszczyszyn, B. (2001). On a coverage process ranging from
the Boolean model to the Poisson-Voronoi tessellation with applications to
wireless communications. Adv. in Appl. Probab., 33(2), 293–323.
Barabási, A.-L. (2002). Linked: The New Science of Networks. Perseus Pub-
lishing, Cambridge, Massachusetts.
Barabási, A.-L. and Albert, R. (1999). Emergence of scaling in random net-
works. Science, 286(5439), 509–512.
Barbour, A. D. and Reinert, G. (2001). Small worlds. Random Structures
Algorithms, 19(1), 54–74.
Barbour, A. D. and Reinert, Gesine (2004). Correction: “Small worlds” [Ran-
dom Structures Algorithms 19 (2001), no. 1, 54–74; mr1848027]. Random
Structures Algorithms, 25(1), 115.
Barbour, A. D. and Reinert, G. (2006). Discrete small world networks. Electron.
J. Probab., 11, no. 47, 1234–1283 (electronic).
Barsky, D.J. and Aizenman, M. (1991). Percolation critical exponents under
66 References

the triangle condition. Ann. Probab., 19, 1520–1536.

Barsky, D.J., Grimmett, G.R., and Newman, C.M. (1991a). Dynamic renor-
malization and continuity of the percolation transition in orthants. In Spa-
tial stochastic processes, Volume 19 of Progr. Probab., pp. 37–55. Birkhäuser
Boston, Boston, MA.
Barsky, D.J., Grimmett, G.R., and Newman, C.M. (1991b). Percolation in half-
spaces: equality of critical densities and continuity of the percolation proba-
bility. Probab. Theory Related Fields, 90(1), 111–148.
Bender, E.A. and Canfield, E.R. (1978). The asymptotic number of labelled
graphs with a given degree sequences. Journal of Combinatorial Theory
(A), 24, 296–307.
Benjamini, I., Kesten, H., Peres, Y., and Schramm, O. (2004). Geometry of the
uniform spanning forest: transitions in dimensions 4, 8, 12, . . . . Ann. of Math.
(2), 160(2), 465–491.
Benjamini, I., Lyons, R., Peres, Y., and Schramm, O. (1999a). Critical percola-
tion on any nonamenable group has no infinite clusters. Ann. Probab., 27(3),
1347–1356.
Benjamini, I., Lyons, R., Peres, Y., and Schramm, O. (1999b). Group-invariant
percolation on graphs. Geom. Funct. Anal., 9(1), 29–66.
Benjamini, I. and Schramm, O. (1996). Percolation beyond Zd , many questions
and a few answers. Electron. Comm. Probab., 1, no. 8, 71–82 (electronic).
van den Berg, J. and Keane, M. (1984). On the continuity of the percolation
probability function. In Conference in modern analysis and probability (New
Haven, Conn., 1982), Volume 26 of Contemp. Math., pp. 61–65. Amer. Math.
Soc., Providence, RI.
Berg, J. van den and Kesten, H. (1985). Inequalities with applications to per-
colation and reliability. J. Appl. Prob., 22, 556–569.
Berger, N., Bollobás, B., Borgs, C., Chayes, J., and Riordan, O. (2003). Degree
distribution of the FKP network model. In Automata, languages and program-
ming, Volume 2719 of Lecture Notes in Comput. Sci., pp. 725–738. Springer,
Berlin.
Berger, N., Borgs, C., Chayes, J. T., D’Souza, R. M., and Kleinberg, R. D.
((2004)). Competition-induced preferential attachment. In Automata, lan-
guages and programming, Volume 3142 of Lecture Notes in Comput. Sci., pp.
208–221. Springer, Berlin.
Berger, N., Borgs, C., Chayes, J. T., D’Souza, R. M., and Kleinberg, R. D.
(2005). Degree distribution of competition-induced preferential attachment
graphs. Combin. Probab. Comput., 14(5-6), 697–721.
Bezuidenhout, C. and Grimmett, G. (1990). The critical contact process dies
out. Ann. Probab., 18, 1462–1482.
Biskup, M. (2004). On the scaling of the chemical distance in long-range per-
colation models. Ann. Probab., 32(4), 2938–2977.
Bollobás, B. (2001). Random graphs (Second edn), Volume 73 of Cambridge
Studies in Advanced Mathematics. Cambridge University Press, Cambridge.
 References 67

Bollobás, B., Borgs, C., Chayes, J., and Riordan, O. (2003). Directed scale-free
graphs. In Proceedings of the Fourteenth Annual ACM-SIAM Symposium on
Discrete Algorithms (Baltimore, MD, 2003), New York, pp. 132–139. ACM.
Bollobás, B., Janson, S., and Riordan, O. (2007). The phase transition in
inhomogeneous random graphs. Random Structures Algorithms, 31(1), 3–122.
Bollobás, B. and Riordan, O. (2003a). Mathematical results on scale-free ran-
dom graphs. In Handbook of graphs and networks, pp. 1–34. Wiley-VCH,
Weinheim.
Bollobás, B. and Riordan, O. (2003b). Robustness and vulnerability of scale-free
random graphs. Internet Math., 1(1), 1–35.
Bollobás, B. and Riordan, O. (2004a). Coupling scale-free and classical random
graphs. Internet Math., 1(2), 215–225.
Bollobás, B. and Riordan, O. (2004b). The diameter of a scale-free random
graph. Combinatorica, 24(1), 5–34.
Bollobás, B. and Riordan, O. (2006a). The critical probability for random
Voronoi percolation in the plane is 1/2. Probab. Theory Related Fields, 136(3),
417–468.
Bollobás, B. and Riordan, O. (2006b). Percolation. Cambridge University Press,
New York.
Bollobás, B. and Riordan, O. (2006c). A short proof of the Harris-Kesten
theorem. Bull. London Math. Soc., 38(3), 470–484.
Bollobás, B. and Riordan, O. (2008). Percolation on random Johnson-Mehl
tessellations and related models. Probab. Theory Related Fields, 140(3-4),
319–343.
Bollobás, B., Riordan, O., Spencer, J., and Tusnády, G. (2001). The degree
sequence of a scale-free random graph process. Random Structures Algo-
rithms, 18(3), 279–290.
Borgs, C., Chayes, J., Hofstad, R. van der, Slade, G., and Spencer, J. (2005a).
Random subgraphs of finite graphs. I. The scaling window under the triangle
condition. Random Structures Algorithms, 27(2), 137–184.
Borgs, C., Chayes, J., Hofstad, R. van der, Slade, G., and Spencer, J. (2005b).
Random subgraphs of finite graphs. II. The lace expansion and the triangle
condition. Ann. Probab., 33(5), 1886–1944.
Borgs, C., Chayes, J. T., Kesten, H., and Spencer, J. (1999). Uniform bounded-
ness of critical crossing probabilities implies hyperscaling. Random Structures
Algorithms, 15(3-4), 368–413. Statistical physics methods in discrete proba-
bility, combinatorics, and theoretical computer science (Princeton, NJ, 1997).
Britton, T., Deijfen, M., and Martin-Löf, A. (2006). Generating simple random
graphs with prescribed degree distribution. J. Stat. Phys., 124(6), 1377–1397.
Broadbent, S. R. and Hammersley, J. M. (1957). Percolation processes. I. Crys-
tals and mazes. Proc. Cambridge Philos. Soc., 53, 629–641.
Burton, R. M. and Keane, M. (1989). Density and uniqueness in percolation.
Comm. Math. Phys., 121(3), 501–505.
Camia, F. (Preprint (2008)). Scaling limits of two-dimensional percolation: an
68 References

overview.
Cardy, John L. (1992). Critical percolation in finite geometries. J. Phys.
A, 25(4), L201–L206.
Chayes, J. T. and Chayes, L. (1987). On the upper critical dimension of
Bernoulli percolation. Comm. Math. Phys., 113(1), 27–48.
Chayes, J. T., Chayes, L., Grimmett, G. R., Kesten, H., and Schonmann, R. H.
(1989). The correlation length for the high-density phase of Bernoulli perco-
lation. Ann. Probab., 17(4), 1277–1302.
Chung, F. and Lu, L. (2002a). The average distances in random graphs with
given expected degrees. Proc. Natl. Acad. Sci. USA, 99(25), 15879–15882
(electronic).
Chung, F. and Lu, L. (2002b). Connected components in random graphs with
given expected degree sequences. Ann. Comb., 6(2), 125–145.
Chung, F. and Lu, L. (2003). The average distance in a random graph with
given expected degrees. Internet Math., 1(1), 91–113.
Chung, F. and Lu, L. (2006a). Complex graphs and networks, Volume 107 of
CBMS Regional Conference Series in Mathematics. Published for the Confer-
ence Board of the Mathematical Sciences, Washington, DC.
Chung, F. and Lu, L. (2006b). The volume of the giant component of a random
graph with given expected degrees. SIAM J. Discrete Math., 20, 395–411.
Chung, F., Lu, L., and Vu, V. (2004). The spectra of random graphs with given
expected degrees. Internet Math., 1(3), 257–275.
Cooper, C. and Frieze, A. (2003). A general model of web graphs. Random
Structures Algorithms, 22(3), 311–335.
Davies, P. L. (1978). The simple branching process: a note on convergence when
the mean is infinite. J. Appl. Probab., 15(3), 466–480.
Dawson, D.A. (1993). Measure-valued Markov processes. In Ecole d’Eté de
Probabilités de Saint-Flour 1991. Lecture Notes in Mathematics #1541, Berlin.
Springer.
Deijfen, M., Esker, H. van den, Hofstad, R. van der, and Hooghiemstra, G.
(Preprint (2007)). A preferential attachment model with random initial de-
grees.
Dodziuk, J. (1984). Difference equations, isoperimetric inequality and tran-
sience of certain random walks. Trans. Amer. Math. Soc., 284(2), 787–794.
Dorogovtsev, S.N. and Mendes, J.F.F. (2002). Evolution of networks. Advances
in Physics, 51, 1079–1187.
Dousse, O., Franceschetti, M., Macris, N., Meester, R., and Thiran, P. (2006).
Percolation in the signal to interference ratio graph. J. Appl. Probab., 43(2),
552–562.
Durrett, R. (1980). On the growth of one-dimensional contact processes. Ann.
Probab., 8(5), 890–907.
Durrett, R. (2007). Random graph dynamics. Cambridge Series in Statistical
and Probabilistic Mathematics. Cambridge University Press, Cambridge.
Dynkin, E. B. (1994). An introduction to branching measure-valued processes,
 References 69

Volume 6 of CRM Monograph Series. American Mathematical Society, Prov-

idence, RI.
Erdős, P. and Rényi, A. (1959). On random graphs. I. Publ. Math. Debrecen, 6,
290–297.
Erdős, P. and Rényi, A. (1960). On the evolution of random graphs. Magyar
Tud. Akad. Mat. Kutató Int. Közl., 5, 17–61.
Esker, H. van den, Hofstad, R. van der, and Hooghiemstra, G. (Preprint (2006a).
Universality for the distance in finite variance random graphs.
Esker, H. van den, Hofstad, R. van der, Hooghiemstra, G., and Znamenski, D.
(2006b). Distances in random graphs with infinite mean degrees. Extremes, 8,
111–140.
Etheridge, A. M. (2000). An Introduction to Superprocesses, Volume 20 of
University Lecture Series. American Mathematical Society, Providence, RI.
Fernholz, D. and Ramachandran, V. (2007). The diameter of sparse random
graphs. Random Structures Algorithms, 31(4), 482–516.
Flaxman, A., Frieze, A., and Vera, J. (2006). A geometric preferential attach-
ment model of networks. Internet Math., 3(2), 187–205.
Flaxman, A., Frieze, A., and Vera, J. (2007). A geometric preferential attach-
ment model of networks II. In Proceedings of WAW 2007.
Fortuin, C. M., Kasteleyn, P. W., and Ginibre, J. (1971). Correlation inequal-
ities on some partially ordered sets. Comm. Math. Phys., 22, 89–103.
Franceschetti, M. and Meester, R. (2008). Random networks for communication.
Cambridge University Press. from statistical physics to information systems.
Gilbert, E. N. (1959). Random graphs. Ann. Math. Statist., 30, 1141–1144.
Grimmett, G. (1999). Percolation (2nd edn). Springer, Berlin.
Grimmett, G. and Hiemer, P. (2002). Directed percolation and random walk.
In In and out of equilibrium (Mambucaba, 2000), Volume 51 of Progr. Probab.,
pp. 273–297. Birkhäuser Boston, Boston, MA.
Grimmett, G. R. and Newman, C. M. (1990). Percolation in ∞ + 1 dimensions.
In Disorder in physical systems, Oxford Sci. Publ., pp. 167–190. Oxford Univ.
Press, New York.
Grossglauser, M. and Thiran, P. (2006). Networks out of control: Models and
methods for random networks.
Häggström, O. (2007) Problem solving is often a matter of cooking up an
appropriate markov chain. Scandinavian Journal of Statistics, 34, 768–780.
Häggström, O. and Peres, Y. (1999). Monotonicity of uniqueness for percolation
on Cayley graphs: all infinite clusters are born simultaneously. Probab. Theory
Related Fields, 113(2), 273–285.
Häggström, O., Peres, Y., and Schonmann, R. H. (1999). Percolation on tran-
sitive graphs as a coalescent process: relentless merging followed by simulta-
neous uniqueness. In Perplexing problems in probability, Volume 44 of Progr.
Probab., pp. 69–90. Birkhäuser Boston, Boston, MA.
Hammersley, J. M. (1961). Comparison of atom and bond percolation processes.
J. Mathematical Phys., 2, 728–733.
70 References

Hammersley, J. M. (1963). A Monte Carlo solution for percolation in a cubic

lattice. In Methods in Computational Physics, Vol. I, pp. 281–298. Academic
Press.
Hara, T. (1990). Mean field critical behaviour for correlation length for perco-
lation in high dimensions. Probab. Th. Rel. Fields, 86, 337–385.
Hara, T. (2005). Decay of correlations in nearest-neighbour self-avoiding
walk, percolation, lattice trees and animals. Preprint. Available on
http://arxiv.org/abs/math-ph/0504021.
Hara, T., Hofstad, R. van der, and Slade, G. (2003). Critical two-point functions
and the lace expansion for spread-out high-dimensional percolation and related
models. Ann. Probab., 31(1), 349–408.
Hara, T. and Slade, G. (1990). Mean-field critical behaviour for percolation in
high dimensions. Commun. Math. Phys., 128, 333–391.
Hara, T. and Slade, G. (1993). Unpublished appendix to (Hara and Slade,
1995). Available on http://www.ma.utexas.edu/mp− arc/index-93.html.
Hara, T. and Slade, G. (1995). The self-avoiding-walk and percolation critical
points in high dimensions. Combinatorics, Probability and Computing, 4, 197–
215.
Hara, T. and Slade, G. (2000a). The scaling limit of the incipient infinite cluster
in high-dimensional percolation. I. Critical exponents. J. Statist. Phys., 99(5-
6), 1075–1168.
Hara, T. and Slade, G. (2000b). The scaling limit of the incipient infinite cluster
in high-dimensional percolation. II. Integrated super-Brownian excursion. J.
Math. Phys., 41(3), 1244–1293.
Harris, T. (1963). The theory of branching processes. Die Grundlehren der
Mathematischen Wissenschaften, Bd. 119. Springer-Verlag, Berlin.
Harris, T. E. (1960). A lower bound for the critical probability in a certain
percolation process. Proc. Cambridge Philos. Soc., 56, 13–20.
Heydenreich, M. and Hofstad, R. van der (2007). Random graph asymptotics
on high-dimensional tori. Comm. Math. Phys., 270(2), 335–358.
Heydenreich, M., Hofstad, R. van der, and Sakai, A. (Preprint (2008)). Mean-
field behavior for long- and finite range Ising model, percolation and self-
avoiding walk.
Hofstad, R. van der (2008). Random graphs and complex networks. In prepa-
ration, see http://www.win.tue.nl/ rhofstad/NotesRGCN.pdf.
Hofstad, R. van der, Hollander, F. den, and Slade, G. (2007). The survival
probability for critical spread-out oriented percolation above 4+1 dimensions.
I. Induction. Probab. Theory Related Fields, 138(3-4), 363–389.
Hofstad, R. van der and Hooghiemstra, G. (Preprint (2007a)). Diameters in
preferential attachment graphs.
Hofstad, R. van der and Hooghiemstra, G. (Preprint (2007b)). A phase transi-
tion for the diameter of the configuration model.
Hofstad, R. van der, Hooghiemstra, G., and Van Mieghem, P. (2005). Dis-
tances in random graphs with finite variance degrees. Random Structures
 References 71

Algorithms, 27(1), 76–123.

Hofstad, R. van der, Hooghiemstra, G., and Znamenski, D. (2007). Distances
in random graphs with finite mean and infinite variance degrees. Electron. J.
Probab., 12(25), 703–766 (electronic).
Hofstad, R. van der and Keane, M. (Preprint (2007)). An elementary proof of
the hitting-time theorem.
Hofstad, R. van der and Luczak, M. (Preprint (2006)). Random subgraphs of
the 2D Hamming graph: The supercritical phase.
Hofstad, R. van der and Sakai, A. (2005). Critical points for spread-out self-
avoiding walk, percolation and the contact process above the upper critical
dimensions. Probab. Theory Related Fields, 132(3), 438–470.
Hofstad, R. van der and Slade, G. (2003). Convergence of critical oriented
percolation to super-Brownian motion above 4 + 1 dimensions. Ann. Inst. H.
Poincaré Probab. Statist., 39(3), 413–485.
Hofstad, R. van der and Slade, G. (2005). Asymptotic expansions in n−1 for
percolation critical values on the n-cube and Zn . Random Structures Algo-
rithms, 27(3), 331–357.
Hofstad, R. van der and Slade, G. (2006). Expansion in n−1 for percolation
critical values on the n-cube and Zn : the first three terms. Combin. Probab.
Comput., 15(5), 695–713.
Hughes, B. D. (1996). Random walks and random environments. Vol. 2. Oxford
Science Publications. The Clarendon Press Oxford University Press, New York.
Random environments.
Jagers, P. (1975). Branching processes with biological applications. Wiley-
Interscience [John Wiley & Sons], London. Wiley Series in Probability and
Mathematical Statistics—Applied Probability and Statistics.
Janson, S. (Preprint (2006)). The probability that a random multigraph is
simple.
Janson, S. (Preprint (2008). Asymptotic equivalence and contiguity of some
random graphs.
Janson, S. and Luczak, M. (Preprint (2007)). A new approach to the giant
component problem.
Janson, S., Luczak, T., and Rucinski, A. (2000). Random graphs. Wiley-
Interscience Series in Discrete Mathematics and Optimization. Wiley-
Interscience, New York.
Kager, Wouter and Nienhuis, Bernard (2004). A guide to stochastic Löwner
evolution and its applications. J. Statist. Phys., 115(5-6), 1149–1229.
Kendall, W. S. and Wilson, R. G. (2003). Ising models and multiresolution
quad-trees. Adv. in Appl. Probab., 35(1), 96–122. In honor of Joseph Mecke.
Kesten, H. (1959a). Full Banach mean values on countable groups. Math.
Scand., 7, 146–156.
Kesten, H. (1959b). Symmetric random walks on groups. Trans. Amer. Math.
Soc., 92, 336–354.
Kesten, H. (1980). The critical probability of bond percolation on the square
72 References

lattice equals 21 . Comm. Math. Phys., 74(1), 41–59.

Kesten, H. (1982). Percolation theory for mathematicians, Volume 2 of Progress
in Probability and Statistics. Birkhäuser Boston, Mass.
Kesten, H. (1987). Scaling relations for 2D-percolation. Comm. Math.
Phys., 109(1), 109–156.
Kesten, H. (2002). Some highlights of percolation. In Proceedings of the In-
ternational Congress of Mathematicians, Vol. I (Beijing, 2002), Beijing, pp.
345–362. Higher Ed. Press.
Langlands, R., Pouliot, P., and Saint-Aubin, Y. (1994). Conformal invariance
in two-dimensional percolation. Bull. Amer. Math. Soc. (N.S.), 30(1), 1–61.
Lawler, G. F. (2004). An introduction to the stochastic Loewner evolution. In
Random walks and geometry, pp. 261–293. Walter de Gruyter GmbH & Co.
KG, Berlin.
Lawler, G. F. (2005). Conformally invariant processes in the plane, Volume 114
of Mathematical Surveys and Monographs. American Mathematical Society,
Providence, RI.
Lawler, G. F., Schramm, O., and Werner, W. (2002). One-arm exponent for
critical 2D percolation. Volume 7, pp. no. 2, 13 pp. (electronic).
Le Gall, J.-F. (1999). Spatial branching processes, random snakes, and partial
differential equations. Birkhäuser, Basel.
Meester, R. (1994). Uniqueness in percolation theory. Statist. Neer-
landica, 48(3), 237–252.
Meester, R. and Roy, R. (1996). Continuum percolation, Volume 119 of Cam-
bridge Tracts in Mathematics. Cambridge University Press, Cambridge.
Menshikov, M.V. (1986). Coincidence of critical points in percolation problems.
Soviet Mathematics, Doklady, 33, 856–859.
Molloy, M. and Reed, B. (1995). A critical point for random graphs with a
given degree sequence. In Proceedings of the Sixth International Seminar on
Random Graphs and Probabilistic Methods in Combinatorics and Computer
Science, “Random Graphs ’93” (Poznań, 1993), Volume 6, pp. 161–179.
Molloy, M. and Reed, B. (1998). The size of the giant component of a random
graph with a given degree sequence. Combin. Probab. Comput., 7(3), 295–305.
Moore, C. and Newman, M.E.J. (2000). Epidemics and percolation in small-
world networks. Phys. Rev. E , 61, 5678–5682.
Nachmias, A. (Preprint (2007)). Mean-field conditions for percolation in finite
graphs.
Newman, M. E. J. (2003). The structure and function of complex networks.
SIAM Rev., 45(2), 167–256 (electronic).
Newman, M.E.J., Moore, C., and Watts, D.J. (2000). Mean-field solution of
the small-world network model. Phys. Rev. Lett., 84, 3201–3204.
Newman, M.E.J. and Watts, D.J. (1999). Scaling and percolation in the small-
world network model. Phys. Rev. E , 60, 7332–7344.
Newman, M. E. J., Watts, D. J., and Barabási, A.-L. ((2006)). The Struc-
ture and Dynamics of Networks. Princeton Studies in Complexity. Princeton
 References 73

University Press.
Nguyen, B. G. (1987). Gap exponents for percolation processes with triangle
condition. J. Statist. Phys., 49(1-2), 235–243.
Nguyen, B.G. and Yang, W-S. (1993). Triangle condition for oriented percola-
tion in high dimensions. Ann. Probab., 21, 1809–1844.
Nguyen, B.G. and Yang, W-S. (1995). Gaussian limit for critical oriented per-
colation in high dimensions. J. Stat. Phys., 78, 841–876.
Nienhuis, B. (1984). Critical behavior of two-dimensional spin models and
charge asymmetry in the Coulomb gas. J. Statist. Phys., 34(5-6), 731–761.
Norros, I. and Reittu, H. (2006). On a conditionally Poissonian graph process.
Adv. in Appl. Probab., 38(1), 59–75.
Oliveira, R. and Spencer, J. (2005). Connectivity transitions in networks with
super-linear preferential attachment. Internet Math., 2(2), 121–163.
Penrose, M.D. (2003). Random geometric graphs, Volume 5 of Oxford Studies
in Probability. Oxford University Press, Oxford.
Perkins, E. (2002). Dawson-Watanabe superprocesses and measure-valued dif-
fusions. In Lectures on probability theory and statistics (Saint-Flour, 1999),
Volume 1781 of Lecture Notes in Math., pp. 125–324. Springer, Berlin.
Reimer, D. (2000). Proof of the van den Berg-Kesten conjecture. Combin.
Probab. Comput., 9(1), 27–32.
Rudas, A., Tóth, B., and Valkó, B. (2007). Random trees and general branching
processes. Random Structures Algorithms, 31(2), 186–202.
Russo, L. (1978). A note on percolation. Z. Wahrscheinlichkeitstheorie und
Verw. Gebiete, 43(1), 39–48.
Russo, L. (1981). On the critical percolation probabilities. Z. Wahrsch. Verw.
Gebiete, 56(2), 229–237.
Sakai, A. (2002). Hyperscaling inequalities for the contact process and oriented
percolation. J. Statist. Phys., 106(1-2), 201–211.
Schonmann, R. H. (2001). Multiplicity of phase transitions and mean-field
criticality on highly non-amenable graphs. Comm. Math. Phys., 219, 271–
322.
Schonmann, R. H. (1999). Stability of infinite clusters in supercritical percola-
tion. Probab. Theory Related Fields, 113(2), 287–300.
Schonmann, R. H. (2002). Mean-field criticality for percolation on planar non-
amenable graphs. Comm. Math. Phys., 225(3), 453–463.
Schramm, O. (2000). Scaling limits of loop-erased random walks and uniform
spanning trees. Israel J. Math., 118, 221–288.
Seymour, P. D. and Welsh, D. J. A. (1978). Percolation probabilities on the
square lattice. Ann. Discrete Math., 3, 227–245. Advances in graph theory
(Cambridge Combinatorial Conf., Trinity College, Cambridge, 1977).
Slade, G. (2006). The lace expansion and its applications, Volume 1879 of
Lecture Notes in Mathematics. Springer-Verlag, Berlin. Lectures from the
34th Summer School on Probability Theory held in Saint-Flour, July 6–24,
2004, Edited and with a foreword by Jean Picard.
74 References

Smirnov, S. (2001). Critical percolation in the plane: conformal invariance,

Cardy’s formula, scaling limits. C. R. Acad. Sci. Paris Sér. I Math., 333(3),
239–244.
Smirnov, S. and Werner, W. (2001). Critical exponents for two-dimensional
percolation. Math. Res. Lett., 8(5-6), 729–744.
Tasaki, H. (1987a). Geometric critical exponent inequalities for general random
cluster models. J. Statist. Phys., 49(3-4), 841–847.
Tasaki, H. (1987b). Hyperscaling inequalities for percolation. Comm. Math.
Phys., 113(1), 49–65.
Timár, Á. (2006). Percolation on nonunimodular transitive graphs. Ann.
Probab., 34(6), 2344–2364.
Turova, T.S. and Vallier, T. (Preprint (2006)). Merging percolation on Zd and
classical random graphs: Phase transition.
Werner, W. (2004). Random planar curves and Schramm-Loewner evolutions.
In Lectures on probability theory and statistics, Volume 1840 of Lecture Notes
in Math., pp. 107–195. Springer, Berlin.
Werner, W. (2005). SLE, conformal restriction, loops. In European Congress
of Mathematics, pp. 515–528. Eur. Math. Soc., Zürich.
Wu, C. C. (1993). Critical behavior or percolation and Markov fields on branch-
ing planes. J. Appl. Probab., 30(3), 538–547.
Yukich, J. E. (2006). Ultra-small scale-free geometric networks. J. Appl.
Probab., 43(3), 665–677.