EMPG-11-12

DAMTP-2011-49
Geometric Models of Matter
M. F. Atiyah,
School of Mathematics, University of Edinburgh,
Kings Buildings, Edinburgh EH9 3JZ, UK.
M.Atiyah@ed.ac.uk
N. S. Manton,
DAMTP, Centre for Mathematical Sciences,
University of Cambridge,
Wilberforce Road, Cambridge CB3 0WA, UK.
N.S.Manton@damtp.cam.ac.uk
B. J. Schroers,
Department of Mathematics, Heriot-Watt University,
Riccarton, Edinburgh EH14 4AS, UK.
bernd@ma.hw.ac.uk
August 2011
Abstract
Inspired by soliton models, we propose a description of static particles in terms of
Riemannian 4-manifolds with self-dual Weyl tensor. For electrically charged particles,
the 4-manifolds are non-compact and asymptotically bred by circles over physical 3-
space. This is akin to the Kaluza-Klein description of electromagnetism, except that we
exchange the roles of magnetic and electric elds, and only assume the bundle structure
asymptotically, away from the core of the particle in question. We identify the Chern
class of the circle bundle at innity with minus the electric charge and the signature of
the 4-manifold with the baryon number. Electrically neutral particles are described by
compact 4-manifolds. We illustrate our approach by studying the Taub-NUT manifold
as a model for the electron, the Atiyah-Hitchin manifold as a model for the proton,
CP
2
with the Fubini-Study metric as a model for the neutron, and S
4
with its standard
metric as a model for the neutrino.
1 Introduction
Geometry and the quantum mechanics of particles have an uneasy relationship which is
why general relativity is hard to incorporate into quantum eld theory. String theory is an
1
a
r
X
i
v
:
1
1
0
8
.
5
1
5
1
v
1


[
h
e
p
-
t
h
]


2
5

A
u
g

2
0
1
1
ambitious and remarkable attempt at unication with many successes but the nal theory
has proved mysterious and elusive.
Einstein and Bohr fought a long battle on this front and Bohr was generally deemed to have
won, with the Copenhagen interpretation of quantum mechanics accepted. But Einsteins
belief in the role of geometry made a partial come-back with the adoption of gauge theories
as models of particle physics.
A more modest and limited role for geometry in nuclear physics was proposed by Skyrme
[1] with the solitonic model of baryons, i.e. proton, neutron and nuclei, now known as
Skyrmions. These have been shown to be approximate models of the physical baryons
occurring in gauge theories of quarks and gluons, and have been extensively studied [2] with
considerable success.
In this paper we explore a geometric model of particles which is inspired by Skyrmes idea
but with potential applications to both baryonic and leptonic particle physics. Our model
diers from the Skyrme model in that it uses Riemannian geometry rather than eld theory,
and so it is closer in spirit to Einsteins ideas. Another key dierence is that we absorb the
Kaluza-Klein idea of an extra circle dimension to incorporate electromagnetism. However,
we exchange the roles of electricity and magnetism relative to the standard Kaluza-Klein
approach, so that the extra circle dimension is magnetic rather than electric, and it is the
electric charge that is topologically quantised by the famous Dirac argument. We will also,
initially, ignore time and dynamics, focusing on purely static models.
Our geometric models will therefore be 4-dimensional Riemannian manifolds. We require
the manifolds to be oriented and complete but not usually compact: we use non-compact
manifolds to model electrically charged particles, and compact manifolds for neutral particles.
Both baryon number and electric charge will be encoded in the topology, with baryon number
(at least provisionally) identied with the signature of the 4-manifold
1
and electric charge
with minus the Chern class of an asymptotic bration by circles. In particular, the number
of protons and the number of neutrons will therefore be determined topologically.
The manifolds will also have an asymptotic structure which captures their relation to
physical 3-space. The non-compact manifolds we consider have an asymptotic region which
is bred over physical 3-space, so no additional structure is required. For the compact
(neutral) models, however, we x a distinguished embedded surface X where the 4-manifold
M intersects physical 3-space. We also call M \ X the inside of M.
For single particles, the symmetry group of rotations should x all of the above data. It
should act isometrically on the 4-manifold and preserve the asymptotic structure. In the non-
compact cases this means that it should be a bundle map in the asymptotic region, covering
the usual SO(3)-action on physical 3-space. In the compact cases it should preserve the
distinguished surface X. In order to capture the fermionic nature of the particles considered
1
The denition of signature for non-compact manifolds is reviewed in Sect. 8.1.
2
in this paper we also require spin structures on the non-compact manifolds and on the inside
(in the sense dened above) of compact manifolds. Moreover, the lift of the rotation group
action to the spin bundle (over the entire manifold in the non-compact case and over the
inside in the compact case) should necessarily be an SU(2)-action, and this is what we mean
by saying that our models are fermionic.
A key restriction on (the conformal classes of) our manifolds is that they are self-dual.
Recall that the Riemann curvature is made up of the Ricci tensor plus the Weyl tensor W,
which is conformally invariant. In dimension 4, W is the sum of self-dual and anti-self-dual
parts
W = W
+
W

. (1.1)
A 4-manifold M is said to be self-dual if W

= 0. These manifolds are precisely those

which have twistor spaces in the sense of Penrose [3]. The latter are 3-dimensional complex
manifolds Z with a real S
2
-bration over M. The complex structure of Z (together with a
real involution which is the antipodal map on each S
2
) encodes the entire conformal structure
of M. Even the Einstein metric can be captured by complex data on Z.
A simply-connected self-dual 4-manifold, for which the Ricci tensor is also zero, is a hy-
perk ahler manifold, whose structure group reduces to SU(2). It is a complex Kahler manifold
for an S
2
-family of complex structures, but for any of these complex structures, and for the
complex orientation, it would be anti-self-dual. Since we want self-dual manifolds we choose
the opposite orientation.
While some of our particles, including the proton, will be modelled by hyperk ahler mani-
folds, we do not want to be so restrictive. Instead we will only require our 4-manifold models
of particles to be self-dual and Einstein, so there can be a non-zero scalar curvature. Our
model for the neutron is of this type, distinguishing it from the proton. The neutron will of
course also have electric charge zero.
We should point out that reversing the orientation of a 4-manifold turns a self-dual manifold
into an anti-self-dual one. This should be interpreted as giving the geometric model of an
anti-particle. The existence of anti-particles follows from CPT invariance, and our models
are compatible with this.
Self-dual 4-manifolds are, in many ways, the 4-dimensional analogue of Riemann surfaces,
with H
2
replacing H
1
in homology. In particular there are theorems [4] which assert that
such manifolds admit connected sums although, unlike in the case of Riemann surfaces, there
are restrictions on when this is possible. Such connected sums model composite objects, like
nuclei. Although we focus at present on static particles we do envisage a deformation theory,
using the moduli space of self-dual manifolds, which could underlie particle interactions.
Fortunately a lot is now known about self-dual 4-manifolds with many metrics explicitly
calculated. This makes it possible to put forward some denite models for the proton and
neutron. Even though our ideas are inspired by Skyrmes theory of baryons, it turns out that
3
geometric models of leptons, i.e. the electron and (electron-)neutrino, are even simpler, and
we shall describe them too in the class of self-dual manifolds. Thus, somewhat surprisingly,
our framework of self-dual manifolds allows us to describe baryons and leptons in a unied
fashion.
The language and spirit of our model for particles is close to that of general relativity
and suggests the possibility of a unication with gravity, but we do not address this issue
here. In particular, we do not specify an action functional. Instead, we focus on how our
model describes general features of particles such as their various quantum numbers. We are
aware that a description of elementary particles as 4-dimensional Riemannian manifolds is
radically dierent from established treatments in terms of quantum eld theory. What we
aim to show in this paper is that such a geometric approach is possible, and that it has some
surprising and attractive features, such as the possibility of describing the electron and the
proton in one framework. While we do propose denite identications of certain 4-manifolds
with specic particles in Sects. 3-5 of this paper, these should be seen as illustrations of the
geometric approach, not necessarily as nal proposals.
The paper is organised as follows. In Sect. 2 we outline the genesis of our geometric
models of particles, starting with the Skyrme model of baryons. Electrically charged particles
are necessarily described by non-compact 4-manifolds in our approach, and in Sect. 3 we
explain how to model the electron and proton in terms of the Taub-NUT and Atiyah-Hitchin
manifolds, respectively. Neutral particles are described by compact 4-manifolds, and this is
discussed in Sects. 4-5. We propose CP
2
as a model for the neutron and S
4
as a model for
the neutrino. These are the simplest choices, but we also discuss some more sophisticated
versions. In Sect. 6 we describe how our particle models glue into empty space, and how
the particles may interact with each other. Sect. 7 contains an outline of how our geometric
models capture the spinorial nature of the particles they describe. In Sect. 8 we give the
dictionary which translates topological properties of 4-manifolds into the electric charge and
baryon number of particles, and discuss in some detail how these charges are related to elds
and densities used in conventional Lagrangian models of particle physics. Sect. 9 contains
our conclusion and some ideas for follow-up work. Conventions and calculations are collected
in appendices A,B and C.
2 From Skyrmions to 4-manifolds
We begin by spelling out in detail how the Skyrme model suggests our 4-manifold model.
The Skyrme model is based on a group-valued eld from R
3
,
U : R
3
G, (2.1)
where the Lie group G is usually taken to be SU(2), and U(x) 1 as |x| . The degree
of U as a map S
3
SU(2) is identied with baryon number. The minima of the Skyrme
energy, for each baryon number, are called Skyrmions.
4
Skyrmions are free to rotate, both in physical space and through conjugation by elements
of SU(2). Quantising this motion gives the Skyrmions spin and electric charge. The proton
and neutron, for example, are distinct quantum states of the essentially unique Skyrmion of
degree 1.
In [5] it was shown how to generate such Skyrme elds naturally by starting with an SU(2)
Yang-Mills gauge eld on R
4
and calculating the holonomy along the 4th direction. Suitable
asymptotic behaviour on R
4
guarantees a well-dened map U. Although this construction
does not preserve the respective energy functionals it does provide a good way of using
instantons on R
4
(i.e. self-dual gauge elds) to construct approximate minima of the Skyrme
energy. It also identies instanton number with the Skyrme degree. See also the recent
papers [6, 7] where the dierence between the Yang-Mills and Skyrme energy functionals is
interpreted as due to an innite tower of mesons.
Since the Yang-Mills energy functional in dimension 4 is conformally invariant we could
replace the decomposition
R
4
= R
3
R
1
(2.2)
by
S
4
\ S
2
= H
3
S
1
, (2.3)
where H
3
is hyperbolic 3-space. In fact we can vary the curvature of H
3
provided we rescale
the circle S
1
the opposite way, so that large circles correspond to almost at H
3
. We can
now x a gauge eld on S
4
\ S
2
and take the holonomy round the circles. There are some
technicalities (due to base-points) which we shall ignore but basically we expect to end up
with a Skyrmion on H
3
, an idea which has been explored in [8].
Now replace H
3
S
1
by any Riemannian 4-manifold M which is asymptotically bred
by circles over R
3
. This is the kind of Kaluza-Klein 4-manifold we are going to consider.
An SU(2) gauge eld on M would then give a Skyrme eld on the quotient of M by S
1
.
Since we do not want to assume there is a global circle bration, this Skyrme eld will only
be dened asymptotically outside some core. But an oriented 4-manifold has two natural
SU(2) bundles over it, the two spin bundles S
+
and S

(assuming M is a spin-manifold, i.e.

w
2
(M) = 0). Picking one of these, say S
+
, we then get from the connection on S
+
a natural
construction of an asymptotic Skyrme eld on R
3
.
This is roughly the genesis of our idea to model particles by 4-manifolds, but the topology of
these asymptotic Skyrme elds does not quite t, and would not give integer baryon numbers
as dened in our model. The reason lies in a fundamental dierence between the topology
of gauge bundles which can have arbitrary instanton number (or second Chern class) and
the topology of tangent bundles, where there are divisibility theorems. For example the
rst Pontrjagin class of a 4-manifold (compact and oriented) is divisible by 3 and then gives
the signature. An example is CP
2
which has signature 1 and Pontrjagin class 3 (times the
generator of H
4
(CP
2
)).
5
In the Skyrme model the basic idea is that baryon number is identied with the degree
of the map U in (2.1), or equivalently with the instanton number (or second Chern class)
of the SU(2) bundle over R
4
. This diers from the 4-manifold model we want to explore,
where baryon number is identied with the signature of the 4-manifold. The signature is
additive under taking connected sums of 4-manifolds [9], and this captures the additivity of
baryon number for composites of particles, for example, fusion of nuclei. The integrality of
the signature is linked to it being an index of an elliptic operator. This means we are in the
realm of K-theory rather than cohomology. A consequence of this change of viewpoint is
that the geometry of the 4-manifold model is important for us, but we will not try to dene
a global 3-dimensional Skyrme eld U.
Recall that in the Skyrme model, baryon number is cohomological and electric charge arises
at the quantum level. For our 4-manifold model, electric charge is cohomological, arising,
as already explained, from the rst Chern class of the asymptotic S
1
-bration, while baryon
number as just indicated should be seen as an index.
To sum up our discussion, we see that our model goes beyond the Skyrme model in aiming
to understand topologically both the basic integer physical invariants, baryon number and
electric charge. The two models are dierent, but possibly dual in a suitable sense. We hope
to explore this in detail at a later stage.
3 Models for the electron and proton
Models for the basic particles should exhibit a high degree of symmetry and we expect the
rotation group SO(3) of R
3
, or its double cover SU(2), to act as isometries. For electrically
charged particles, we take our geometric models to be non-compact hyperk ahler manifolds.
We also assume that the volume grows with the third power of the radius, to allow for an
interpretation of the asymptotic region in terms of physical 3-space. As recently shown in
[10], this forces the hyperk ahler manifold to be ALF. We are therefore looking for rotationally
symmetric and complete ALF hyperk ahler manifolds. There are just two possibilities:
1. The Taub-NUT manifold [11, 12] depending on a positive parameter m (interpreted as
mass in the gravitational context). For brevity we denote it by TN.
2. The Atiyah-Hitchin manifold, the (simply-connected double cover of the) moduli space
of centred SU(2)-monopoles of charge two [13]. For brevity we denote it by AH.
Note that we could also single out TN and AH among non-compact, complete and rota-
tionally symmetric hyperk ahler manifolds by demanding that the SU(2)- (or SO(3)-) action
rotates the complex structures, see our discussion following (B.12). This turns out to play a
role in recovering the usual rotation action on physical 3-space in the asymptotic region of
our geometric models, as discussed in Sect. 7.
6
Both TN and AH can be parametrised in terms of a radial coordinate r and angular
coordinates on SU(2) (for TN) or SO(3)/Z
2
(for AH). Details are given in Appendix B. In
terms of the left-invariant 1-forms dened in (B.4), the metrics of both TN and AH can be
written as
ds
2
= f(r)
2
dr
2
+a(r)
2

2
1
+b(r)
2

2
2
+c(r)
2

2
3
, (3.1)
with the functions f, a, b, c satisfying the self-duality equations
2bc
f
da
dr
= (b c)
2
a
2
, + cyclic , (3.2)
where + cyclic means we add the two further equations obtained by cyclic permutation of
a, b, c. We adopt the convention
f(r) =
b(r)
r
, (3.3)
where (for reasons that will emerge later) the radial coordinate r has the range [0, ) for
TN and [, ) for AH. The self-duality equations (3.2) become
da
dr
=
1
2rc
(a
2
(b c)
2
) ,
db
dr
=
b
2rca
(b
2
(c a)
2
) ,
dc
dr
=
1
2ra
(c
2
(a b)
2
) . (3.4)
This system has solutions in terms of elementary functions
a(r) = b(r) = r
_
+
m
r
and c =
m
_
+
m
r
, (3.5)
with parameters , m > 0, associated to the TN manifold. The topology is that of R
4
, and
as 0 the metric tends to the at metric. For > 0 the manifold is asymptotic to an S
1
bre-bundle over R
3
with the length of the circle being 4m/

. There is a U(1) symmetry

acting along the bres, with just one xed point at the origin, r = 0. The whole isometry
group is U(2). As 0 the U(1)-action becomes the scalar action on C
2
. The complex
orientation of C
2
determines the orientation of TN as a self-dual manifold; this is opposite
to the orientation given by any of the complex structures in the hyperkahler family, see
Appendix A.6 for a discussion. At innity, the U(1)-action gives the standard Hopf line
bundle over CP
1
with Chern class +1; details are given in Appendix A.
The TN metric with coecient functions (3.5) has the following behaviour under scaling
by non-vanishing real numbers , :
r

r , m m,
2
, then ds
2

2
ds
2
. (3.6)
7
We use rescaling by to set = 1, and rescaling by to set m = 2 from now on. This
amounts to picking a unit of length for the radial coordinate r and to xing an overall scale
for the metric. Our choice is motivated by the asymptotic form of the AH metric, to be
discussed below. Note that, with this choice, the length of the asymptotic circle, in the
length units chosen, is 8.
The solution which gives rise to AH has the asymptotic form, for large r,
a(r) b(r) r
_
1
2
r
and c(r)
2
_
1
2
r
. (3.7)
These asymptotic expressions, a TN metric with = 1, m = 2, also satisfy (3.4). However
a(r) is not actually equal to b(r), and r only extends down to . For r near , (3.7) is a
poor approximation. Instead, one nds the leading terms
a(r) 2(r ) , b(r) +
1
2
(r ) , c(r) +
1
2
(r ) , (3.8)
which we will need later in this paper.
The manifold AH is the complement of RP
2
(the real projective plane) embedded in CP
2
,
and the complex orientation of CP
2
determines the orientation of AH as a self-dual manifold.
As for TN this is opposite to the orientation given by any of the complex structures in the
hyperk ahler family; see our discussion in Appendix A.6. AH has an SO(3) symmetry with
just one 2-dimensional orbit at r = , which is a minimal 2-sphere. We refer to this minimal
2-sphere, which is the totally imaginary conic in CP
2
and determined by z
2
1
+ z
2
2
+ z
2
3
= 0
in the homogeneous coordinates introduced in Sect. 4.1, as the core. Asymptotically, the
manifold is bred by circles. As further discussed below, neither the circles nor the base
space of this asymptotic bration are oriented because of a Z
2
-identication, given explicitly
in (B.7).
The manifold TN is usually interpreted as the geometry of a Dirac monopole at the origin
[14, 15, 16]. For us, with electric and magnetic charges reversed, it has to be interpreted as
an electrically charged particle. Since the signature of TN is zero (we discuss this further
in Sect. 8) the particle is leptonic. We therefore interpret TN as a model for the electron.
Down on R
3
, after factoring by U(1), any 2-sphere surrounding the origin has an electric
ux emerging from it due to the electron, which carries charge 1. This implies that there
is a sign change in going from the Chern class to the electric charge.
The manifold AH has the opposite asymptotic behaviour with a sign change for m and
an orientation change (see Appendix A) and so would lead us naturally to expect electric
charge +1. Also, the topology at the core is dierent, with a 2-sphere instead of a point. As
a result, AH has signature 1 (again discussed further in Sect. 8) and looks like the model we
want for the proton (rather than the positron).
8
However things are not quite that simple, as we shall now explain. The asymptotic
boundary of AH is not S
3
as for TN but the boundary of a tubular neighbourhood of RP
2
in CP
2
, which is S
3
divided by a cyclic group of order 4. Moreover the base of this unoriented
circle bration is RP
2
, not S
2
, and is non-orientable. This means that the 3-manifold which
is the base of the asymptotic bration is not R
3
, and has a fundamental group of order 2. It
is not orientable.
This might seem to be a disaster, but we shall argue that, while unexpected, it is not as
bad as it looks. The most convincing argument in its favour is to show that the electric
charge is well-dened and equal to +1 as hoped. This is done in detail in Appendix A.
The lack of orientability in physical 3-space should be thought of as follows. The lack of
orientation in R
3
locally is compensated by a corresponding ambiguity in the sign of electric
charge (non-orientability of the circle bres). Physically the geometric orientation is not felt.
4 The neutron
4.1 Complex projective plane
Having put forward a denite proposal for the proton we now have to face the neutron.
Since the neutron has no electric charge any non-compact model would need to have a
trivial asymptotic circle bration. The 4-manifold should have signature 1 and it should
resemble the AH model of the proton in its SO(3) orbit structure. However, the latter
requirement rules out asymptotically trivial circle bundles over physical 3-space since the
generic SO(3) orbits would be 2-dimensional in that case. We therefore consider compact
4-manifolds. In fact, there is an obvious choice which is just the complex projective plane
CP
2
with its Fubini-Study metric (and its natural complex orientation). This is a self-dual
manifold of positive scalar curvature.
CP
2
has even more symmetry than we need, since it is acted on by SU(3). The rotation
group SO(3) sits inside as the subgroup that preserves the real structure given by complex
conjugation. This preserves RP
2
as in the case of the proton, so we x this RP
2
as the
distinguished surface where the 4-manifold intersects physical 3-space, thus breaking the
symmetry to SO(3). The global SU(3) symmetry might give us a link to quarks, but this
remains to be explored.
The Fubini-Study metric is often written in coordinates which exhibit the invariance under
U(2) SU(3). This brings out the parallels with TN [17] but is not the symmetry we want
in our neutron model. For the interpretation of CP
2
as a neutron and for a comparison
with the AH model of the proton, we need to write the Fubini-Study metric in coordinates
adapted to the SO(3)-action, which is discussed in [18] and [19]. We write the results of [18]
in the conventions used in our discussion of the AH metric.
In terms of homogeneous coordinates z C
3
\{0} (with the identication z z, C

)
9
the Fubini-Study metric on CP
2
is
ds
2
=
|z
2
|dz

dz z

dzz
t
d z
|z|
4
. (4.1)
For calculations we can x |z| = 1 and parametrise
z = e
i
Rz
0
, (4.2)
where R SO(3) can, in turn, be parametrised in terms of Euler angles as shown in Appendix
B.1. The reference vector z
0
(which depends on one parameter) should be a unit vector and
we can assume, by adjusting the phase if necessary, that its real and imaginary parts are
orthogonal. For our purpose, it is convenient to single out the 1-axis and pick
z
0
=
_
_
0
a
2
ia
3
_
_
, a
2
2
+a
2
3
= 1 . (4.3)
Then we nd the following expression for the Fubini-Study metric in terms of the left-
invariant forms (B.4) on SO(3) (for details of an analogous calculation we refer the reader
to [18]):
ds
2
= da
2
2
+da
2
3
+ (a
2
3
a
2
2
)
2

2
1
+a
2
3

2
2
+a
2
2

2
3
. (4.4)
Parametrising
a
2
= cos
_

2
+

4
_
, a
3
= sin
_

2
+

4
_
,
_
0,

2
_
, (4.5)
we obtain, nally, the Fubini-Study metric in the form
ds
2
=
1
4
d
2
+ sin
2

2
1
+ sin
2
_

2
+

4
_

2
2
+ cos
2
_

2
+

4
_

2
3
. (4.6)
There is a simple interpretation of the geometry of CP
2
and its orbit structure in terms of
orientated ellipses up to scale [20], which is useful for comparison with the AH metric. As
already exploited above, we can adjust the phase in the homogeneous coordinate z = u+iv
(no longer xed to satisfy |z| = 1) so that the real vectors u and v are orthogonal: if
z
2
= 0 this is automatic, and if z
2
= 0 we multiply by a unit complex number to set
Im(z
2
) = 2u v = 0 and Re(z
2
) = u
2
v
2
< 0 (we pick the negative sign to agree with the
choice made in (4.5) above). We can interpret v and u as the major- and minor-axis of an
ellipse. This ellipse is only determined up to scale (we can still rescale z by any positive real
number) but it is orientated. The totally degenerate case z = 0 is excluded by the denition
of homogeneous coordinates, but circles (z
2
= 0 or |u| = |v|) and lines (|u| = 0) can occur.
In terms of our parametrisation (4.2), the reference vector (4.3) and the denition of in
(4.5), we see that for = 0 the ellipse is a circle and for =

2
it degenerates to a line. For
generic values of , the SO(3)-orbit is SO(3)/Z
2
, with the Z
2
generated by the 180

-rotation
about the 1-axis, but for = 0 the orbit is a 2-sphere and for =

2
the orbit is RP
2
. This
10
is the same as the orbit structure of AH compactied by an RP
2
at innity, although the
metric is of course dierent.
We also note that the K ahler form
= i
|z
2
|dz
t
d z z

dz z
t
d z
|z|
4
(4.7)
takes the simple form
= cos
2

3
sin d
1
, (4.8)
which should be compared to the expression given in [17] in coordinates adapted to the
U(2) symmetry of CP
2
. The form is invariant under the 180

-rotation about the 1-axis

and hence well-dened on the generic SO(3)-orbits. It is manifestly closed, but not exact:
for > 0 we can write = d(cos
1
) but this expression is not valid on the exceptional
SO(3)-orbit where = 0, since
1
is not well-dened there.
The Kahler form is self-dual with respect to the complex orientation (with volume element
dV =
2
= sin(2) d
1

2

3
). Since it is closed, it is also harmonic. The existence
of a non-exact harmonic, self-dual form on CP
2
follows from the fact that the signature of
CP
2
is 1. In view of our interpretation of signature as baryonic charge one might expect
there to be a baryonic interpretation of . We return to this question when discussing the
AH model of the proton further in Sect. 8.
4.2 Hitchins one-parameter family of Einstein metrics
If the CP
2
-model for the neutron turns out to be too nave, there is a more sophisticated
variant which could be explored. This arises from the sequence of self-dual Einstein manifolds
M(k), for k > 2, studied by Hitchin [21]. The manifolds M(k) for even k 4 are all dened
on the same space as our proton model, namely CP
2
but with RP
2
removed. M(4) is CP
2
with the Fubini-Study metric and all the metrics on M(k), k > 4 and even, are incomplete
on the open set in CP
2
\ RP
2
, but can be completed to metrics on CP
2
with a conical
singularity of angle 4/(k 2) along RP
2
. For odd k 3 the manifolds are dened on S
4
,
with M(3) being S
4
with its standard metric. This time the metrics of M(k), k > 3 and
odd, are incomplete on the open set in S
4
\RP
2
, but can be completed to metrics on S
4
with
a conical singularity of angle 2/(k 2) along RP
2
.
The sequence of conical manifolds M(k) for even k 4 and starting with CP
2
, has de-
creasing scalar curvature and converges to AH as k . It may turn out that some other
value of k gives a better model for the neutron than k = 4. Note that for k > 4 the conical
singularity breaks the symmetry down to SO(3). Even for k = 4 we shall see later that other
factors break the symmetry in this way.
Hitchin also pointed out [21, 22] that the family M(k) can be extended to real parameter
values. For any real k > 0, M(k) is related to the moduli space of centred SU(2) monopoles
11
over hyperbolic space of curvature 1/p
2
, where p = (k 4)/4. When k is not an integer,
the conical angle is not a rational multiple of and M(k) is not an orbifold. Consequently,
the explicit methods of [21] do not then apply. Nonetheless, having k as a real parameter
gives useful room for manoeuvre in modelling the neutron and may provide contact with
conventional nuclear models. In particular, 1/k may play a role as a small parameter that
controls the breaking of isospin symmetry.
The forthcoming paper [23] contains a signature formula for Riemannian manifolds with
conical singularities, like the Hitchin manifolds M(k). We summarise that result in Sect. 8.2.
5 The neutrino
Having put forward models of the electron, proton and neutron, it is then natural to look for
a similar model of the neutrino. Since it has no electric charge it should, like the neutron,
be modelled by a compact manifold. It should have symmetry similar to the U(2) symmetry
of the electron and should have positive curvature. It should have zero baryon number, that
is, vanishing signature.
Just as CP
2
is the most obvious model for the neutron, the standard 4-sphere, S
4
, is the
most obvious model for the neutrino. Again this has more symmetry than we need, SO(5)
instead of SO(3). Just as a distinguished RP
2
in CP
2
picks out the smaller SO(3) symmetry,
so a distinguished S
2
in S
4
is needed to cut down the symmetry of the neutrino.
To exhibit the symmetry, we parametrise S
4
in terms of vectors x R
3
and y R
2
satisfying the constraint
xx +yy = 1 . (5.1)
The metric is then
ds
2
= dxdx +dydy . (5.2)
The group SO(3) SO(2) acts in the obvious way on the pair of vectors (x, y) S
4
and
preserves the metric. In order to compare with other metrics discussed in this paper we
parametrise
x = sin (sin cos , sin sin , cos ) ,
y = cos (cos , sin ) , (5.3)
with [0,

2
], [0, 2) and the usual range for the polar coordinates , on S
2
, and nd
the expression
ds
2
= d
2
+ sin
2
(
2
1
+
2
2
) + cos
2
d
2
. (5.4)
Here we used that d
2
+sin
2
d
2
=
2
1
+
2
2
in terms of the left-invariant 1-forms dened in
(B.4). The generic SO(3) SO(2) orbit is S
2
S
1
, but this collapses to S
1
when = 0 and
to S
2
when = /2.
12
Note that S
4
is conformally at, so the Weyl tensor vanishes and is trivially self-dual. S
4
has no middle-dimensional homology, so the signature is zero, and hence our model neutrino
has zero baryon number, as required. Since S
4
also has an orientation-reversing isometry,
our model seems to suggest that the neutrino coincides with the anti-neutrino. For this and
other reasons (see later) our choice of S
4
is very tentative and provisional. As with the CP
2
model for the neutron it should be regarded at present as a prototype.
To address the symmetry breaking issue, and several others, we will in the next section
discuss how our various models are supposed to t into conventional 3-space.
6 Attaching the models to space
So far our models are abstract objects, 4-manifolds on their own, which are supposed to
model four basic particles of nature. How are we to view them in the real world?
Let us begin with the easiest case, that of the electron. Thought of originally as the
Dirac monopole, the idea is well-known. We consider Kaluza-Klein space as a Riemannian
4-manifold with a circle action. Away from matter this space is assumed to be a circle bundle
over R
3
. Outside a given region in R
3
which is electrically neutral the bundle is assumed to
be (topologically) just the product space. If a region is electrically charged the circle bundle
over the boundary is supposed to have a Chern class equal to minus the charge. If we start
from the vacuum, then inserting one electron amounts to attaching a truncated version of
TN to the boundary. This truncation turns the idealized model into a more realistic model of
a particle. If other particles are present, TN will be an approximation to the precise metric.
This approximation is some measure of the force exerted on the electron. In a dynamic
theory, forces should emerge from the equations, a task for the future.
Next let us move on to the proton. This is similar, using a truncated version of AH.
However, as pointed out earlier, the circle bration is now not oriented so that the asymptotic
3-space is not R
3
(unless our region contains equal numbers of protons and anti-protons).
The model for the neutron is compact, so there is no way to attach it to a boundary. Instead
we propose that our model neutron (a copy of CP
2
) intersects our 4-space in a surface. This
surface should project to a surface in 3-space which is the boundary of the neutron as seen
by an observer. Since we want to keep the neutron similar to the proton (except for the
charge) it seems reasonable to take this surface to be a copy of RP
2
. But since the charge
is now zero the circle bundle over the surface should be trivial. This could arise as follows.
Pick a point in R
3
and blow it up to give an RP
2
, so that we are modifying 3-space in this
neighbourhood. Keep the circle bundle trivial, so making the charge zero. Lift this RP
2
into
the total space of the circle bundle and let the CP
2
neutron model intersect 4-space in this
RP
2
, which is the distinguished RP
2
in CP
2
(hence breaking the larger symmetry of CP
2
). In
this construction the metric on CP
2
need not be changed. The only change that is needed is
the change in metric on the background 4-space got from the blowing up process in 3-space.
13
Finally we come to the neutrino. From the other examples it is clear what is required.
This time we just take a 3-ball in R
3
with boundary a 2-sphere. The circle bundle over it is
trivial and we lift the sphere to the total 4-space. We now require the S
4
neutrino model to
intersect our 4-space in the chosen 2-sphere. The choice of the 2-sphere in S
4
again breaks
the symmetry, down to SO(3) SO(2). The 2-sphere need not be a great (geodesic) sphere
and this provides a parameter to play with.
Note that in this case we could reinterpret the picture as the surgery that kills o the circle
and leaves a 2-sphere. This means that for the electron, the proton and the neutrino we can
still think of our space as a 4-manifold. But this does not seem to work for the neutron
where we have to settle for a 4-space with intersecting components like a complex algebraic
surface with double curves.
The four models that we have proposed for the four basic particles should be geometrically
related in some way to account for the process of beta decay in which a neutron breaks up
into a proton, an electron and an anti-neutrino. The opposite asymptotic behaviour of AH
and TN is a good start but the dierence in the asymptotic fundamental groups presents a
problem. This suggests that the model of the neutrino should somehow bridge the gap and
it argues against the simplicity of the 4-sphere. We hope to pursue this question.
7 Spin 1/2
In all our models we have a natural action of the symmetry group of rotations (SO(3) or
SU(2)) preserving the metric and the asymptotics, the details of which dier according to
the cases. For the neutral, compact models we interpret asymptotic to mean the behaviour
near the distinguished surface where the 4-manifold intersects 3-space, which is either an
RP
2
or an S
2
. For the electrically charged, non-compact models we have an asymptotic
bration by circles over physical 3-space; rotations preserve this bration and induce an
SO(3)-action on the base. The hyperk ahler structures on TN [24] and AH [25] can be
used to construct Cartesian coordinates on physical 3-space with the physically required
transformation properties under spatial rotations. Here we make essential use of the fact
that the complex structures on TN and AH transform as vectors under rotations, as explained
at the beginning of Sect. 3.
For a model to represent a particle of spin 1/2 we must include the data necessary to lift the
rotation group action to an SU(2) action, and to construct its 2-dimensional representation.
To achieve this for non-compact (electrically charged) models we require a spin structure
on the 4-manifold while for compact (neutral) models we only require a spin structure on
the inside, obtained by removing a distinguished surface X from the 4-manifold. In this
short section we explain why, in each of the models considered in this paper, the lift of the
rotation group action to the spin bundle is an SU(2)-action. We do not attempt to construct
naturally associated spin 1/2 representations, but comment on how this may be done.
14
For the TN model of the electron, there is nothing to do since the rotation group action
on TN is an SU(2)-action. For the neutron model we view the required data as the com-
pactication of the proton model AH, not just topologically, but also with the action of the
rotation group. In particular the RP
2
at innity is part of the data. We now simply require
the extra data of a spin structure on the inside CP
2
\ RP
2
, i.e., on AH.
It might be thought that, since the spin structure on AH is unique, there is nothing gained
by the additional data, but this is to ignore the interaction with the symmetries. We must
now require the rotation group to lift to the spin bundle, and this may require us to pass to
SU(2), in which case we label the model as fermionic. Otherwise we call it bosonic. To see
that, in principle, either case could occur, consider the two manifolds Y
1
= SO(3) R and
Y
2
= R
3
R. We take the left-translation action of SO(3) on Y
1
and the standard action on
Y
2
. For Y
1
, the tangent bundle is trivial and we can choose the trivial spin bundle (though
since Y
1
is not simply-connected there is another choice). The SO(3)-action extends without
going to SU(2), so Y
1
is bosonic in our terminology. For Y
2
the xed point at the origin of
R
3
means that we can only lift to the spin bundle after passing to SU(2), so Y
2
is fermionic.
For AH we have to show that, with its action of SO(3), it is fermionic. There are several
ways to do this. Perhaps the simplest (in line with the example Y
1
above) is to note that the
action is not free and that the isotropy group of a point on the core is SO(2). To lift even
this subgroup to the spin bundle over the xed point requires us to go to the double cover.
A comparison between our models for the electron and proton is illuminating. As we
pointed out, SU(2) acts on the electron but only SO(3) acts on the proton. Thus the
fermionic natures of our two models dier. In one case it is inherent in the symmetry while
in the other it is geometric or topological.
Our tentative model for the neutrino is just the round 4-sphere, with a distinguished 2-
sphere at innity given by the decomposition R
5
= R
3
R
2
and the corresponding action of
SO(3). The inside, got by removing S
2
, has an innite cyclic fundamental group so there
are innitely many spin structures, but only one extends to the whole of S
4
. If we pick this,
then it is easy to see that the lift of SO(3) to the spin structure requires us to pass to SU(2)
(for example we can use an SO(2) xing a point at innity and argue as with the proton).
Thus our model of the neutrino is fermionic.
Our discussion so far shows that our geometric approach furnishes fermionic models, but
it does not establish that they necessarily give spin 1/2. This requires constructing the 2-
dimensional representation of SU(2) and relating it to the asymptotic region. We expect that
the required 2-dimensional representations can be constructed in terms of eigenspaces of the
Dirac operator on our model manifolds, possibly twisted by a U(1)-bundle with curvature
proportional to one of the harmonic 2-forms discussed in Sects. 4.1 and 8.
15
8 Charges, energies and uxes
8.1 General remarks
So far we have focussed on topological and geometrical features of our models and explained
how they describe general properties of particles like baryon number, electric charge and
location in space. We want to keep an open mind about how our geometric models make
quantitative contact with the physics of elementary particles. In particular, we do not assume
that this should necessarily happen in the standard framework of Lagrangian eld theory,
where dynamics, conservation laws and even the quantum theory are all derived from an
action functional.
The purpose of this section is to illustrate that our geometric models for particles nev-
ertheless contain natural candidates for the kind of quantities which arise in Lagrangian
models, like energy density and electric elds. We show that electric charge as dened in
our model can be represented by a harmonic 2-form, thus making contact with the usual
description of electric ux. One important feature of the densities and elds considered in
this section is that they are dened on the 4-manifold so that they can only be interpreted
as conventional spatial densities and elds in the asymptotic region of the 4-manifold which
canonically projects down to physical 3-space.
We begin with a summary of the topological quantities and their physical interpretation for
each of the 4-manifolds considered thus far. For compact manifolds the electric charge is zero
and for non-compact manifolds it is minus the self-intersection number X
2
of the manifold
X representing innity in their compactication. The compactication is CP
2
for both TN
and AH, but X = CP
1
for TN with self-intersection number X
2
= 1, while X = RP
2
for
AH, with self-intersection number X
2
= 1; see Appendix A.6 for details.
The baryon number is identied with the signature of the 4-manifold. For a non-compact
oriented manifold, the signature is dened as the signature of the image of the compactly
supported cohomology in the full cohomology [26]. The topology of TN is that of R
4
so the
signature vanishes. The signature of AH is 1. This follows from the fact that its 2-dimensional
homology is generated by the core 2-sphere, and that the self-intersection number of this
2-sphere is positive (in fact equal to +4). The same argument applies to the sequence of
Hitchin manifolds M(k) for k 4 and even, reviewed in Sect. 4.2, which are all compact
and topologically equivalent to CP
2
.
In Table 1 we list the electric charge and baryon number as well as the Euler characteristic
and the squared L
2
-norm ||R||
2
of the Riemann curvature for the four 4-manifolds mainly
discussed in this paper. The Euler characteristic is homotopy invariant, so can be computed
for TN and AH by noting that the former retracts to a point and the latter to a 2-sphere.
16
X
2
(electric charge) (baryon number) ||R||
2
/(8
2
)
TN (electron) 1 0 1 1
AH (proton) 1 1 2 2
CP
2
(neutron) 0 1 3 3
S
4
(neutrino) 0 0 2 2
Table 1: Geometric properties of 4-manifolds and their physical interpretation
8.2 Signature, Euler characteristic and L
2
-norm of the Riemann curvature
For compact Riemannian 4-manifolds there are formulae for the Euler characteristic and
signature in terms of integrals over the 4-manifold involving the Riemann curvature, which
we shall review below. In the non-compact cases, these bulk contributions need to be sup-
plemented by boundary integrals and (for the signature) a subtle spectral contribution (-
invariant), see e.g. [12] and [23] for a summary. For manifolds with conical singularities like
the sequence of Hitchin manifolds reviewed in Sect. 4.2, a signature formula was recently
found [23], which we also review below. We now discuss the bulk contributions, relegating
most detailed calculations to Appendix B.2.
Writing R for the Riemann tensor as in Appendix B, we dene the squared L
2
-norm of R
(for compact and non-compact manifolds) as
||R||
2
=
_
M

a<b
R
ab
R
ab
, (8.1)
where R
ab
is the Hodge dual of R
ab
. The form that integrates to the rst Pontrjagin class
on a compact manifold is
p
1
=
1
4
2

a<b
R
ab
R
ab
, (8.2)
and the form which integrates to the signature in the compact case is
S =
1
3
p
1
, (8.3)
so that
(M) =
_
M
S =
1
3
_
M
p
1
=
1
12
2
_
M

a<b
R
ab
R
ab
. (8.4)
The form that integrates to the Euler characteristic in the compact case is
e =
1
16
2

a<b

abcd
R
ab
R
cd
. (8.5)
17
Note that, with our conventions and for compact Einstein manifolds [27],
(M) =
_
M
e =
1
8
2
||R||
2
, (8.6)
which determines ||R||
2
for CP
2
and S
4
in terms of their topology.
The Riemann curvature on a 4-manifold may also be thought of as a mapping of 2-forms.
Exploiting the fact that the space
2
of 2-forms on an oriented 4-manifold decomposes into
-eigenspaces

of the Hodge star operator , we get a corresponding decomposition of the

Riemann curvature into irreducible pieces [12]
R =
_
W
+
+
s
12
B
B W

+
s
12
_
. (8.7)
Here W

are the self-dual and anti-self-dual parts of the Weyl tensor, s is the scalar curvature
and B amounts to the tracefree part of the Ricci curvature. Then the signature of a compact
manifold can also be expressed as [27]
(M) =
1
12
2
_
||W
+
||
2
||W

||
2
_
. (8.8)
For a self-dual manifold W

= 0 so, up to the factor 12

2
, is given by the L
2
-norm of the
Weyl tensor W
+
, and is non-negative.
Since the metrics on both TN and AH are hyperkahler, B and s vanish, so the full Riemann
curvature is self-dual. Therefore the bulk contribution to both the signature and the Euler
characteristic can be expressed in terms of the L
2
-norm of the Riemann curvature, which
equals the L
2
-norm of W
+
. In Appendix B.2, we show that
||R||
2
TN
= 8
2
, ||R||
2
AH
= 16
2
. (8.9)
Using the self-duality of the Riemann curvature, we deduce that the bulk contributions to
the Euler characteristic are both in agreement with the topological results listed above:
_
TN
e =
8
2
8
2
= 1 ,
_
AH
e =
16
2
8
2
= 2 . (8.10)
The bulk contributions to the signatures, on the other hand, turn out to be fractional:
_
TN
S =
8
2
12
2
=
2
3
,
_
AH
S =
16
2
12
2
=
4
3
. (8.11)
As shown in [23], the fall-o of the spin connection and curvature imply that the boundary
integrals do not contribute in the limit to either the Euler characteristic or the signature.
However, the fractional values of the bulk integrals for the signature show that there must
be a non-zero contribution from the -invariant.
18
In [23], a signature formula is derived for Riemannian 4-manifolds with conical singular
metrics. As reviewed in Sect. 4.2, the Hitchin manifolds M(k) are of this type, and M() can
be identied with AH. The general formula derived in [23] for the signature of a Riemannian
4-manifold M whose metric has a conical singularity of angle 2/ along a surface X is
(M) =
1
3
_
M
p
1

1
3
_
1
1

2
_
X
2
, (8.12)
where p
1
is the form (8.2) which integrates to the Pontrjagin class, and X
2
is the self-
intersection number of X in M.
Since the signature is interpreted as the baryon number in our model, it is tempting to
interpret the integrand of the bulk contribution to the signature as a baryon number density.
Our calculations above show that this cannot be the whole story, since the signature also
receives a contribution from the -invariant of the boundary. We nevertheless compute and
plot the integrands of the bulk contributions to the signature below. Since overall factors
are not important here, we look at the squared L
2
-norm (8.1) of the Riemann curvature.
In the notation of Appendix B, the integrand of (8.1) in both the TN and AH case can be
written as
dF
1

2

3
=
r
ab
2
c
dF
dr
dV , (8.13)
where dV is the volume element, so that the combination
r
ab
2
c
dF
dr
(8.14)
may be interpreted as a density. We plot this density for TN and AH in Fig. 1. It is nite
at, respectively, the origin and the core, and for large r falls o like 1/r
6
. In the AH case we
compute the functions a, b, c and hence F numerically; in the TN case we have the explicit
formula
r
ab
2
c
dF
dr
=
24
(r + 2)
6
. (8.15)
8.3 Electric and baryonic uxes
We now construct elds on TN and AH which carry the electric ux measured by the
self-intersection numbers X
2
tabulated in Table 1. We do this by systematically studying
rotationally symmetric harmonic forms on both TN and AH, with most of the details given
in Appendix C.
One nds [28, 29, 30, 31] that the only square-integrable harmonic and rotationally sym-
metric 2-form on TN is, up to an overall arbitrary constant,

+
3
=
_
r
r + 2

1

2
+
2
(r + 2)
2
dr
3
_
= d
_
r
r + 2

3
_
. (8.16)
19
Figure 1: The density (8.14) for TN (left) and AH (right)
We claim that this form is a harmonic representative of the Poincare dual of the surface
X = CP
1
at innity in TN, the surface parametrised by and . Our calculation also gives
an alternative computation of the electric charge as minus the self-intersection number of
CP
1
in CP
2
. Let p be this self-intersection number, then

CP
1 =
p
4

+
3
(8.17)
is Poincare dual to CP
1
since _
CP
1

CP
1 = p . (8.18)
However, by the denition of the self-intersection numbers in terms of cohomology, we also
have _
TN

CP
1
CP
1 = p . (8.19)
Evaluating the integral in Appendix C we nd
p
2
= p , (8.20)
so that p = 1, conrming that, with the self-dual orientation, the self-intersection of CP
1
in
CP
2
is 1.
Since the 2-form
CP
1 is harmonic and since its total ux through innity equals the
electric charge we interpret it as the electric eld of the electron. Although we have adopted a
viewpoint dual to standard electromagnetism, with a purely spatial 2-form being interpreted
as electric rather than magnetic ux, it is interesting that the self-dual 2-form (8.16) also
contains a term which allows for a conventional electric interpretation: when we contract

CP
1 with the vector eld

along the bres (using

3
(

) = 1), we obtain a purely radial

eld which, asymptotically, falls o like 1/r
2
.
20
The integral (8.19) is the squared L
2
-norm of the electric eld. Ignoring overall factors
and working with
+
3
we write
||
+
3
||
2
=
_

+
3
dV , (8.21)
with dV dened as in (B.17), and interpret the integrand

+
3
(r) =
2
(r + 2)
4
(8.22)
as an electric energy density. For comparison with the AH case below, we plot the prole
function g
+
3
(r) = r/(r + 2) and the energy density
+
3
in Fig. 2.
Figure 2: The electric prole function g
+
3
and the energy density
+
3
Turning now to AH, we review in Appendix C why there are only two SO(3)-invariant
harmonic forms

1
on AH which respect the identication (B.7). They have the structure

1
= G

1

2

3
+dG

1

1
, (8.23)
for functions G

1
of r which satisfy the ordinary dierential equations for g

1
in (C.3). Using
the asymptotic formulae (C.13) and (C.14) one shows [24, 32, 33] that only G
+
1
is nite at
the core and decays at innity. In fact, the solution decays exponentially fast at innity,
with the leading term proportional to e
r/2
. We normalise G
+
1
() = 1, so that near the core
G
+
1
(r) 1
1

2
(r )
2
, for (r ) small . (8.24)
The 2-form
+
1
is not dual to the surface X = RP
2
at innity in AH and has no interpretation
in terms of electric ux. However, it is dual to the core CP
1
in AH. With

core
=
p
4

+
1
(8.25)
21
one nds _
core

core
= p . (8.26)
As a check, we use this form to compute the self-intersection number of the core. Recalling
that the volume of SO(3)/Z
2
is 4
2
, and using (C.15), we have
_
AH

core

core
=
p
2
16
2
(4
2
)(0
2
1
2
) =
p
2
4
, (8.27)
so that
p
2
4
= p , (8.28)
conrming p = 4 for the self-dual orientation (this agrees with [13], where the anti-self-dual
orientation gives the opposite sign).
The harmonic form (4.8) on CP
2
is related to the signature of CP
2
through the fact that
the signature of a compact 4-manifold equals the dierence of the dimensions of the spaces
of self-dual and anti-self-dual harmonic representatives of the second de Rham cohomology.
It seems likely that the (up to scale) unique bounded and rotationally symmetric self-dual
harmonic form
core
on AH is related to the self-dual harmonic form (4.8) via the sequence
of Hitchin manifolds reviewed in Sect. 4.2.
Since the existence of
core
on AH is linked to the signature of AH being 1, and since
signature represents baryon number in our approach, it may be possible to interpret the
detailed structure of
core
in baryonic terms. The exponential fall-o exhibited by
core
is
reminiscent of the protons pion eld in the Yukawa description of the nuclear force. In
Fig. 3 we plot the numerically computed prole function G
+
1
and the associated energy
density (C.17), which turns out to be

+
1
= 2
(G
+
1
)
2
b
2
c
2
. (8.29)
This is nite at the core and decays exponentially according to e
r
for large r.
To nd a harmonic 2-form on AH which can play the role of the protons electric eld
we need to go to a branched cover of AH, denoted AH. The metric on the branched cover
is not smooth at the core, but this will not aect the following calculations near innity.
Dropping the requirement that forms are invariant under the identication (B.7), we nd
that the closure condition on the 2-forms (C.11) has only one further solution which is nite
at the core and which remains bounded for large r. This is the 2-form

3
= G

3

1

2
+dG

3

3
, (8.30)
with a radial function G

3
satisfying (C.3). The solution vanishes at the core as
G

3
(r) C

r , for (r ) small , (8.31)

22
Figure 3: The baryonic prole function G
+
1
and the energy density
+
1
for some constant C. The large r behaviour is
G

3
(r)

C
r 2
r
, (8.32)
where

C is another constant. This leads to an anti-self-dual form on AH, which is square-
integrable (as we shall show below) and which has not been considered previously. It is our
candidate for the electric eld of the proton.
As a manifold, AH compacties to S
2
S
2
, as discussed in Appendices A.4 and A.5. The
surface X = RP
2
which compacties AH to CP
2
becomes the anti-diagonal S
2
in S
2
S
2
.
Choosing

C = 1 in (8.32), a harmonic representative of the Poincare dual class to this S
2
is

S
2 =
p
4

3
. (8.33)
Then _
S
2

S
2 = p , (8.34)
and since the volume of SO(3) is 8
2
by (B.8), we also compute
_
AH

S
2
S
2 =
p
2
16
2
(8
2
)(1
2
0
2
) =
p
2
2
. (8.35)
Hence

p
2
2
= p , (8.36)
which is solved by p = 2. Dividing by 2, to take us back to RP
2
, we conrm the self-
intersection number of RP
2
in CP
2
as 1, again for the self-dual orientation.
Since the 2-form
S
2 on AH is harmonic and since its total ux through innity, suitably
interpreted, equals the electric charge, we think of
S
2 as representing the electric eld
23
of the proton. The comments made after (8.16) about the possibility of recovering a con-
ventional electric eld by contracting the electric 2-form on TN with the vector eld along
asymptotic bres apply, suitably modied, to AH.
In Fig. 4 we plot the numerically computed prole function G

3
for C = 1 (as dened in
(8.31)) and the associated energy density (C.17), which turns out to be

3
= 2
(G

3
)
2
a
2
b
2
. (8.37)
This function has a (r )
1
singularity at r = , and falls o like r
4
for large r as in the
TN case.
Figure 4: The electric prole function G

3
and the energy density

3
9 Conclusion and outlook
In introducing and illustrating geometric models of matter in this paper we have concentrated
on general, global properties of particles like baryon number and electric charge. The most
striking phenomenological success of our geometric approach is its prediction of precisely two
stable electrically charged particles, one leptonic and one baryonic, with opposite electric
charges.
An important theme throughout this paper is the implementation of rotational symmetry
in our models. The rotation action preserves the metric and, in the compact cases, the
distinguished surface where the 4-manifold intersects physical 3-space. We have shown that
our models are fermionic in the sense that the lift of the rotation action to spin bundles (over
the inside of the 4-manifold in the compact cases) is necessarily an SU(2)-action, but we
have left the explicit construction of spin 1/2 states for future work.
24
Non-vanishing electric charge and baryon number give rise to harmonic 2-forms on our
model manifolds. These have allowed us to make contact with the conventional description
of charged particles in terms of associated elds and uxes.
There are various geometrically natural candidates for measuring energy or mass in our
models, but we have not committed ourselves to any particular energy functional in this pa-
per. In the absence of an energy measure, we are not able to make contact with experimental
data about particle masses or forces between particles.
We end our summary with some observations about relative scales predicted by our models
for charged particles. The AH model relates three scale parameters: the size of the core
(radius in our geometric units), the size of the asymptotic circles (radius 2 in geometric
units) and the scale 1 implicit in the exponential corrections to the asymptotic form of the
coecient functions (3.7) for the AH metric, which are proportional to e
r
(see [34, 35] for a
discussion of these corrections in the context of magnetic monopoles). Interpreting the core
radius as the proton radius and the scale in the exponential decay as the Compton wavelength

of the pion, we nd that our model correctly assigns the same order of magnitude to those
two quantities. The details are not quite right (experimentally, the proton radius is just over
half of

) but this is not surprising given our very rudimentary understanding of how our
model relates to experiment. In any case, these considerations suggest that we should pick
a length unit of about 1 fermi or 10
15
m in the AH model.
Since the asymptotic bration by circles arises for all electrically charged particles, we
expect the size of the asymptotic circle to have a purely electric interpretation, and we also
assume that it is the same for both AH and TN (this was implicit in the way we xed
scale and units for TN). One natural guess, at least for TN, is to relate the size of the
asymptotic circle to the classical electron radius. Our models would then equate the orders
of magnitude of the classical electron radius with the Compton wavelength of the pion, which
is phenomenologically right. It is an attractive feature of TN as a model for the electron
that it has a length scale, but no core structure.
Many avenues remain to be explored. In our geometric approach, fusion and ssion of
nuclei as well as decay processes involving both baryons and leptons (like the beta decay
of the neutron) should have a description in terms of gluing and deformation of self-dual
4-manifolds. In order to study masses and binding energies of particles we need to pick an
energy measure. This could involve the norms of curvatures and harmonic forms discussed
here, but may take a less conventional form. Our ideas about spin 1/2 need to be eshed
out. The Dirac operator on our model manifolds is likely to play a role in combining energy
and spin considerations, and Seiberg-Witten theory on the model manifolds seems relevant,
too. In order to move beyond static consideration, time needs to be introduced.
The list of open issues may seem daunting, but in each case the geometric framework
introduced here suggests natural lines of attack. We hope to pursue them in future work.
25
Acknowledgment We thank Nigel Hitchin and Claude LeBrun for assistance and Jose
Figueroa-OFarrill for working with us during the early stages of the project.
Appendices
A Geometry and sign conventions
A.1 Self-intersection numbers
Let L be a complex line bundle over a compact oriented surface X. Then
c
1
(L)[X] = X
2
(A.1)
where X
2
denotes the self-intersection number of the zero section in the total space of L.
This is essentially one of the denitions of the rst Chern class. Note that both X and L are
oriented, with the bres having the natural orientation of the complex numbers. However the
self-intersection X
2
depends only on the orientation of the total space, since (X)
2
= X
2
.
A.2 The Hopf bundle
Let us review the denition of the Hopf bundle H, the standard line bundle over the complex
projective line, paying particular attention to orientations. We start with C
2
, and CP
1
as
the space of complex lines through the origin. This complex line bundle is dened to be
the dual of the line bundle H. The reason for this apparently perverse choice is that H has
holomorphic sections (given by C
2
) and its dual H

does not. This is clear from the exact

sequence of holomorphic vector bundles over CP
1
:
0 H

C
2
H 0 . (A.2)
Since holomorphic intersections are always positive, the rst Chern class of H gives +1 when
evaluated on the fundamental class of CP
1
.
If we compactify C
2
by adding a CP
1
at innity then the self-intersection number of this
line is clearly +1, so that the normal bundle is H.
A.3 Non-orientable surfaces
Now assume that X is a non-orientable surface in an oriented 4-manifold M. This denes a
mod 2 homology class in M and so it would appear that its self-intersection number is only
dened modulo 2. However a more careful look shows that we can dene an integer self-
intersection. There are several (equivalent) ways to see this. First we note that the tangent
bundle and the normal bundle of X in M have isomorphic orientation real line bundles L.
26
Then both the Euler class of the normal bundle and the fundamental class of X are dened
as classes twisted by L. The evaluation (A.1) therefore makes sense and denes X
2
. An
alternative way is to pass to the double covering of a neighbourhood of X in M, where X
acquires an orientation, compute the self-intersection there and divide by 2. A third way
is to deform X into a transverse surface

X and, near each intersection point of X and

X,
choose an orientation of X and the corresponding orientation of

X given by the deformation.
Now compute the local intersection numbers of X and

X and sum them up.
A.4 An example
The example we want to study is that of the real projective plane X = RP
2
inside the
complex projective plane M = CP
2
. We give M its standard complex orientation. We plan
to show that X
2
= 1.
To do this we will use the double branched covering of CP
2
given by the product of two
copies of the complex projective line CP
1
. We can identify CP
2
with the symmetric product
of the two copies, branched along the diagonal D. This map is holomorphic and compatible
with the action of SL(2, C). Now introduce a metric on CP
1
identifying it with a 2-sphere,
and consider in CP
1
CP
1
the graph D

of the antipodal map. The image of D

in CP
2
is
the standard embedding of RP
2
into CP
2
. Outside the branch locus D the map from the
product of the two 2-spheres to CP
2
is a double covering so we can compute X
2
as half of
(D

)
2
. This is most easily done by observing that, in the cohomology of the product, with x
and y the two generators of H
2
, the classes of D and D

are
D = x +y , D

= x y . (A.3)
Since x
2
= y
2
= 0 and xy = 1 it follows that
D
2
= 2 and (D

)
2
= 2 (A.4)
(while D.D

= 0 agreeing with the fact that D and D

are disjoint). Dividing by 2 to get

back down to CP
2
we end up with X
2
= 1 as stated.
As a check we can also use a deformation of D

in M by choosing a dierent metric on

CP
1
giving a dierent antipodal map. In particular, consider the 2-sphere as the boundary
of the unit ball in R
3
and move the origin so that antipodal points are now the end points of
lines through the new origin. The line joining the two origins denes a unique intersection
point of the two copies of RP
2
in CP
2
. Since this line acquires dierent orientations from
the two origins the intersection number is 1.
A third sign check is to observe that the normal bundle to the diagonal in the product
of the two 2-spheres is the tangent bundle with Euler number 2, while for the antidiagonal
(graph of the antipodal map) it is the cotangent bundle with Euler number 2.
27
A.5 More on the example
It is enlightening to analyze the SO(3) action on the spaces in the example above. In addition
we can introduce the branched double covering CP
2
S
4
given by complex conjugation
[36, 37]. This is branched over RP
2
.
Thus we have maps
S
2
S
2
CP
2
S
4
, (A.5)
which are compatible with the SO(3) actions. Here S
4
is best thought of as the space of real
3 3 symmetric matrices of zero trace and xed norm.
The generic orbits are, in each case, 3-dimensional but there are two exceptional orbits of
dimension 2. Inside the product S
2
S
2
they are the diagonal and anti-diagonal (both S
2
),
in S
4
they are both RP
2
, while in CP
2
one is S
2
and one is RP
2
. Each of the two maps is
branched along a surface: S
2
for the rst map and RP
2
for the second.
To examine the self-intersection numbers of the exceptional orbits it is useful to note two
general rules:
A. If X in M is double covered by X

in M

then (X

)
2
= 2X
2
.
B. If X is a branch locus in the double covering then (X

)
2
=
1
2
X
2
.
Rule A is obvious and rule B is familiar in algebraic geometry, but it applies more generally
even in the non-orientable case.
Using rules A and B we can now check what happens in our two maps. In S
4
the two
RP
2
s have self-intersection numbers +2 and 2. In S
2
S
2
the two 2-spheres also have self-
intersection numbers +2 and 2. In CP
2
the 2-sphere, which is a conic, has self-intersection
number +4, while RP
2
has self-intersection number 1.
A.6 Self-dual manifolds
We want to analyze various sign conventions in relation to self-dual manifolds, and begin by
checking our sign conventions for two examples of complex K ahler surfaces. Referring to our
discussion of the signature and Euler characteristic in Sect. 8.2, we note the examples:
1. The K3 surface with the Yau metric, which makes it hyperkahler. With its complex
orientation its signature is 16 and its Euler characteristic is 24. With the opposite
orientation, the metric is self-dual and the signature is positive.
2. CP
2
with the Fubini-Study metric is self-dual for its complex orientation [12], agreeing
with the positive signature +1.
28
We will also be interested in non-compact examples similar to the above. The argument
that a hyperk ahler manifold is self-dual for the orientation opposite to the complex orienta-
tion (given by one of the family of complex Kahler structures) is purely local. It just depends
on the fact that the bundle of self-dual 2-forms for these complex orientations is at. More
generally, on any K ahler manifold, the bundle of self-dual 2-forms is the direct sum of the
canonical line bundle K, its dual and a trivial bundle generated by the K ahler form.
The rst non-compact example that interests us is the Taub-NUT manifold TN. This
has the isometry group U(2), its topology (and symmetry) is that of C
2
with the central
U(1) giving the scalar action. The other U(1) actions, inside SU(2), dene the complex
hyperk ahler structures which have the opposite orientation. Hence TN is self-dual for the
orientation that becomes that of C
2
in the limit when the parameter goes to 0, as discussed
after (3.5).
The second example is that of the (simply-connected version of the) Atiyah-Hitchin man-
ifold AH, which is an open subspace of CP
2
got by removing RP
2
. As shown in [13], it has
a complete hyperk ahler metric. This is self-dual for the orientation opposite to the complex
orientations given by the hyperk ahler metric. But this is just the orientation given by its
complex structure as an open subset of CP
2
. This can be checked directly, but it is best
seen by using the results of Hitchin [21], reviewed in Sect. 4.2, which give a whole sequence
of self-dual manifolds on the same space, starting from the Fubini-Study metric of CP
2
and
converging to AH.
In each case, for TN and AH, we have a hyperkahler manifold acted on by SU(2) (the
action descends to SO(3) for AH). The manifolds have a core region and an asymptotic
region with an asymptotic action of O(2). In addition to the 2-parameter family of complex
structures given by the hyperkahler metric (and rotated by SU(2)) there is another complex
structure compatible with the SU(2) action, but giving the opposite orientation to that of
the 2-parameter family. For TN this complex manifold is just C
2
and for AH it is CP
2
\RP
2
.
In both cases the asymptotic O(2) action gives us topologically a disc bundle at innity and
this can be identied with the normal bundle of the surface X that naturally compacties
the manifold. For TN this compactication is CP
2
with X = CP
1
, and for AH it is CP
2
with
X = RP
2
. We have shown that these give rise to opposite signs:
For TN X
2
= +1 while for AH X
2
= 1 .
As explained in the main text, TN and AH are our geometric models for the electron and
the proton, and we identify the self-intersection number of X with minus the electric charge.
29
B Metric properties of TN and AH
B.1 Coordinates and conventions
In this paper we need to parametrise SU(2) explicitly at various points, and we use generators
t
1
=
1
2
i
1
=
1
2
_
0 i
i 0
_
, t
2
=
1
2
i
2
=
1
2
_
0 1
1 0
_
, t
3
=
1
2
i
3
=
1
2
_
i 0
0 i
_
, (B.1)
where
a
, a = 1, 2, 3, are the Pauli matrices; the commutators are [t
a
, t
b
] =
abc
t
c
. We use
Euler angles [0, ), [0, 2) and [0, 4) for parametrising SU(2) as follows:
g(, , ) = e
t
3
e
t
2
e
t
3
=
_
e
i
2
(+)
cos
1
2
e
i
2
()
sin
1
2

e
i
2
()
sin
1
2
e

i
2
(+)
cos
1
2

_
. (B.2)
Dening left-invariant 1-forms
a
on SU(2) via
g
1
dg =
1
t
1
+
2
t
2
+
3
t
3
, (B.3)
we compute to nd

1
= sin d + cos sin d,

2
= cos d + sin sin d,

3
= d + cos d, (B.4)
satisfying d
i
=
1
2

ijk

j

k
. They descend to SO(3) by simply restricting to [0, 2).
Explicitly, generators of the Lie algebra of SO(3) are (T
a
)
bc
=
abc
. They also satisfy
[T
a
, T
b
] =
abc
T
c
. We can then parametrise SO(3) matrices via
R(, , ) = e
T
3
e
T
2
e
T
3
, (B.5)
with [0, ), [0, 2) and [0, 2). Then
R
1
dR =
1
T
1
+
2
T
2
+
3
T
3
(B.6)
provides an alternative denition of the left-invariant 1-forms on SO(3): one obtains the
same expressions as in (B.4), but with the range of angles automatically appropriate to
SO(3).
Both TN and AH can be parametrised in terms of Euler angles and a radial coordinate.
For TN, the angular ranges are [0, ), [0, 2) and [0, 4). For AH they are
[0, ), [0, 2) and [0, 2) with the additional Z
2
identication
(, , ) ( , +, ), (B.7)
30
which, in the asymptotic region, is the simultaneous reversal of spatial and bre direction.
The following angular integrals, which enter into various calculations in this paper, are
therefore
_
SU(2)

1

2

3
= 16
2
,
_
SO(3)

1

2

3
= 8
2
,
_
(SO(3)/Z
2
)

1

2

3
= 4
2
. (B.8)
Note that
1
is invariant under (B.7), but
2
and
3
change sign.
Recalling that the metrics on TN and AH are of the Bianchi IX form
ds
2
= f(r)
2
dr
2
+a(r)
2

2
1
+b(r)
2

2
2
+c(r)
2

2
3
, (B.9)
we introduce the tetrad

1
= a
1
,
2
= b
2
,
3
= c
3
,
4
= fdr . (B.10)
The self-duality of the Riemann tensor computed from (B.9) with respect to the volume
element
dV = a
1
b
2
c
3
fdr = fabc dr
1

2

3
(B.11)
is then equivalent to the set of ordinary dierential equations
2bc
f
da
dr
= (b c)
2
a
2
+ 2bc , + cyclic , (B.12)
where + cyclic means we add the two further equations obtained by cyclic permutation of
a, b, c, and is a parameter which has to be either 0 or 1. In all cases the resulting metrics
are hyperk ahler, but in order to obtain metrics whose hyperk ahler structures are rotated by
the SU(2) action we need to set = 0. This is the case for both TN and AH, so = 0 in
(B.12).
One checks the equivalence of the self-duality of the Riemann tensor and (B.12) as follows.
Denoting the Riemann tensor by R, its component 2-forms with respect to (B.10) are
R
12
= d
3

3
+ (
3

2
)
1

2
,
R
34
= d
3

3
+ (
3

1
)
1

2
, (B.13)
and similar expressions for the components obtained by cyclic permutation of 1, 2, 3, where

1
=
b
2
+c
2
a
2
2bc
,
2
=
c
2
+a
2
b
2
2ca
,
3
=
a
2
+b
2
c
2
2ab
(B.14)
31
and

1
=
1
f
da
dr
,
2
=
1
f
db
dr
,
3
=
1
f
dc
dr
. (B.15)
When (B.12) hold, then
i
=
i
+1, i = 1, 2, 3, with = 0 or = 1, and the self-duality
relations
R
12
= R
34
, R
23
= R
14
, R
31
= R
24
(B.16)
are easily veried. One also checks that each of the three independent 2-forms R
12
, R
23
, R
31
is self-dual with respect to (B.11), as expected.
The only solutions of (B.12) with = 0 which give rise to complete manifolds whose generic
SU(2)- or SO(3)-orbit is three-dimensional are the TN and AH metrics, whose coecient
functions we discussed in Sect. 3. It is important that the coecient function c in (3.1) is
negative for all r in the AH manifold. It implies in particular that the canonical volume
element/orientation dV in (B.11), with f = b/r,
dV = fabc dr
1

2

3
=
ab
2
c
r
sin dr d d d , (B.17)
has opposite signs for TN and AH, in the following sense. Assuming the same orientation

3
of the SU(2) and SO(3) orbits in TN and, respectively, AH, the radial direction is
oppositely oriented in the two cases: the natural radial line element fabc dr has the same
orientation as dr for TN, but the opposite orientation for AH. This will be important when
evaluating various integrals. It means that, when using the coordinates r, , , we can
use the conventional orientations for these coordinates when computing for AH, but should
use the opposite orientation when computing for TN. Thus we integrate in the negative
r-direction when calculating on TN.
B.2 L
2
-norms of the Riemann curvature
The squared L
2
-norm of the Riemann tensor is
||R||
2
M
=
_
M
2(R
12
R
12
+R
23
R
23
+R
31
R
31
) , (B.18)
where M is TN or AH, and we have used the self-duality of the curvature. In terms of
the functions (B.14) the integrand can be simplied (see also [38], where this calculation is
carried out for the AH metric) and becomes
2(R
12
R
12
+R
23
R
23
+R
31
R
31
) = dF
1

2

3
, (B.19)
with
F = 2(
1
+
2
+
3
1)
2
8
1

2
. (B.20)
We denote these functions in the TN and AH cases by F
TN
and F
AH
.
32
For TN, we can compute explicitly and nd
F
TN
(r) = 8
2r + 1
(r + 2)
4
. (B.21)
Integrating (B.19) and using (B.8) as well as our sign convention for the volume form we
nd
||R||
2
TN
= (16
2
)(F
TN
(0) F
TN
()) = 8
2
. (B.22)
For AH, similarly,
||R||
2
AH
= (4
2
)(F
AH
() F
AH
()) . (B.23)
To complete the calculation we need the behaviour of F
AH
at innity and at . One nds
for AH

1
() =
2
() =
c
2a
() = 0 ,
3
() = 1
c
2
2a
2
() = 1 , (B.24)
as well as

1
() = 1 ,
2
() =
1
2
,
3
() =
1
2
, (B.25)
where we took careful limits, using (3.7) and (3.8). Therefore
F
AH
() = 0 , F
AH
() = 4 , (B.26)
so
||R||
2
AH
= (4
2
) (0 4) = 16
2
. (B.27)
C Harmonic forms on TN and AH
C.1 Rotationally symmetric harmonic forms
The question of computing harmonic 2-forms on the TN and AH manifolds arose in physics
in the context of testing S-duality, see [33] for AH and [30, 31] for TN. Harmonic forms also
play a role as curvatures of line bundles over TN and AH, see [32] for a discussion of this for
AH where the form later used for testing S-duality [33] is the curvature of an index bundle.
This appendix contains a systematic discussion of these forms and also a harmonic form on
the branched cover of AH which has not previously been considered in the literature. The
physical interpretation of the forms as electric and baryonic uxes is given in Sect. 8.
On both TN and AH, rotationally symmetric 2-forms can be written in terms of the metric
coecient functions f, a, b, c appearing in the respective metrics and functions f
1
, f
2
and f
3
of the radial coordinate r as

1
= f

1
(b
2
c
3
a
1
fdr) ,

2
= f

2
(c
3
a
1
b
2
fdr) ,

3
= f

3
(a
1
b
2
c
3
fdr) , (C.1)
33
with + standing for self-dual and standing for anti-self-dual. Introducing the functions
g

1
= f

1
bc , g

2
= f

2
ca , g

3
= f

3
ab , (C.2)
closure of the forms (C.1) implies the ordinary dierential equations
dg

1
dr
=
af
bc
g

1
,
dg

2
dr
=
bf
ca
g

2
,
dg

3
dr
=
cf
ab
g

3
. (C.3)
For solutions of these dierential equations (we shall discuss below in which cases regular
solutions exist), the forms (C.1) can be written

1
= g

1

2

3
+dg

1

1
,

2
= g

2

3

1
+dg

2

2
,

3
= g

3

1

2
+dg

3

3
. (C.4)
All these forms are locally exact, that is,

i
= d(g

i

i
) , i = 1, 2, 3 , (C.5)
but the 1-forms in brackets may not be globally dened and, when dened, they may not be
L
2
, so that the corresponding forms

i
are not necessarily L
2
-exact.
We are interested in regular and bounded solutions of the equations (C.3). Using our
convention f = b/r, the functions appearing in the dierential equations are then
af
bc
=
a
rc
,
bf
ca
=
b
2
rca
,
cf
ab
=
c
ra
. (C.6)
C.2 Taub-NUT
Substituting the TN coecient functions (3.5) (with = 1 and m = 2), the functions (C.6)
simplify further to

a
rc
=
b
2
rca
=
r + 2
2r
,
c
ra
=
2
r(r + 2)
. (C.7)
It is easy to check, and was shown in [30] and [31], and earlier in [28] and [29], that only the
equation for g
+
3
has a solution which is both regular at the origin and bounded as r .
Integrating, one nds explicitly
g
+
3
(r) = C
r
r + 2
, (C.8)
34
with an arbitrary constant C. Setting C = 1, (C.4) gives the associated normalisable,
harmonic form

+
3
=
_
r
r + 2

1

2
+
2
(r + 2)
2
dr
3
_
= d
_
r
r + 2

3
_
. (C.9)
Then, using (B.8),
_
TN

+
3

+
3
=
_
0

4r
(r + 2)
3
dr (16
2
) = 16
2
. (C.10)
C.3 Atiyah-Hitchin
On AH, the analysis of harmonic forms is analogous. We write G

i
for the solutions of (C.3)
with the radial functions a, b, c, f of the AH metric, and write the forms obtained by solving
these equations as

1
= G

1

2

3
+dG

1

1
,

2
= G

2

3

1
+dG

2

2
,

3
= G

3

1

2
+dG

3

3
. (C.11)
As for TN, all these forms are formally exact,

i
= d(G

i

i
), i = 1, 2, 3 , (C.12)
but the 1-forms in brackets may not be globally dened, as we shall see.
Note that only the forms

1
respect the identication (B.7). However, as explained in the
main text we also consider the branched double cover AH of AH and therefore keep all forms
in the discussion. On AH, we use (3.8) and (3.7) to nd that the functions appearing in the
dierential equations (C.3) have the following behaviour near the core (i.e. (r ) small),
a
rc

2

2
(r ) ,
b
2
rca

c
ra

1
2(r )
, (C.13)
and the following behaviour for large r,
a
rc

b
2
rca

1
2
_
1
2
r
_
,
c
ra

2
r(r 2)
. (C.14)
For the dierential equations (C.3) on AH, this means that only the equations for G
+
1
and
G

3
have solutions which are nite at the core and remain bounded for large r. Both of these
solutions are discussed and interpreted in the main text.
For any solution of (C.3) on AH, we note
_
AH

=
_
AH
d((G

)
2
)
1

2

3
= (4
2
)
_
(G

)
2
() (G

)
2
()
_
, (C.15)
35
where

and G

stand for any of the forms and coecient functions in (C.11), and we
again used (B.8). We also note that the L
2
-norm of each of these forms can be written as
||

||
2
=
_
AH

=
_
AH
dV , (C.16)
where dV is the volume element dened in (B.11) and is a density which we interpret as
an energy density in the main text. Its general form is
=
2
fabc
G

dG

dr
, (C.17)
which can be simplied in each of the cases, using the dierential equations (C.3) on AH.
References
[1] T. H. R. Skyrme, A non-linear eld theory, Proc. Roy. Soc. A260 (1961) 127138.
[2] G. E. Brown and M. Rho, The Multifaceted Skyrmion, Singapore, World Scientic, 2010.
[3] R. Penrose, The twistor programme, Rep. Math. Phys. 12 (1977) 6576.
[4] S. Donaldson and R. Friedman, Connected sums of self-dual manifolds and deformations
of singular spaces, Nonlinearity 2 (1989) 197239.
[5] M. F. Atiyah and N. S. Manton, Skyrmions from instantons, Phys. Lett. B222 (1989)
438442.
[6] T. Sakai and S. Sugimoto, Low energy hadron physics in holographic QCD, Prog. Theor.
Phys. 113 (2005) 843882.
[7] P. Sutclie, Skyrmions, instantons and holography, JHEP 1008:019 (2010).
[8] M. Atiyah and P. Sutclie, Skyrmions, instantons, mass and curvature, Phys. Lett.
B605 (2005) 106114.
[9] M. F. Atiyah and I. M. Singer, The index of elliptic operators: III, Ann. of Math. 87
(1968) 546604.
[10] V. Minerbe, On the asymptotic geometry of gravitational instantons, Ann. Scient.

Ec.
Norm. Sup. (4) 43 (2010) 883924.
[11] S. W. Hawking, Gravitational instantons, Phys. Lett. A60 (1977) 8183.
[12] T. Eguchi, P. B. Gilkey and A. J. Hanson, Gravitation, gauge theories and dierential
geometry, Physics Reports 66 (1980) 213393.
36
[13] M. Atiyah and N. Hitchin, The Geometry and Dynamics of Magnetic Monopoles,
M. B. Porter Lectures, Rice University, Princeton NJ, Princeton University Press, 1988.
[14] D. Pollard, Antigravity and classical solutions of ve-dimensional Kaluza-Klein theory,
J. Phys. A16 (1983) 565574.
[15] D. J. Gross and M. J. Perry, Magnetic monopoles in Kaluza-Klein theories, Nucl. Phys.
B226 (1983) 2948.
[16] R. D. Sorkin, Kaluza-Klein monopole, Phys. Rev. Lett. 51 (1983) 8790.
[17] G. W. Gibbons and C. N. Pope, CP
2
as a gravitational instanton, Commun. Math.
Phys. 61 (1978) 239248.
[18] C. Bouchiat and G. W. Gibbons, Non-integrable quantum phase in the evolution of
a spin-1 system: a physical consequence of the non-trivial topology of the quantum
state-space, J. Phys. France 49 (1988) 187199.
[19] A. S. Dancer and I. A. B. Strachan, Kahler-Einstein metrics with SU(2) action, Math.
Proc. Camb. Phil. Soc. 115 (1994) 513525.
[20] M. F. Atiyah and N. S. Manton, Geometry and kinematics of two Skyrmions, Commun.
Math. Phys. 153 (1993) 391422.
[21] N. J. Hitchin, A new family of Einstein metrics, in Manifolds and Geometry (Pisa,
1993), p. 190222, Sympos. Math. XXXVI, Cambridge, Cambridge University Press,
1996.
[22] N. J. Hitchin, Twistor spaces, Einstein metrics and isomonodromic deformations, J.
Di. Geom. 42 (1995) 30112 .
[23] M. F. Atiyah and C. LeBrun, The signature of 4-manifolds with conical singular metric,
in preparation.
[24] G. W. Gibbons and P. J. Ruback, The hidden symmetries of multi-center metrics,
Commun. Math. Phys. 115 (1988) 267300.
[25] D. Olivier, Complex coordinates and Kahler potential for the Atiyah-Hitchin metric,
Gen. Rel. and Grav. 23 (1991) 13491362.
[26] M. F. Atiyah, V. K. Patodi and I. M. Singer, Spectral asymmetry and Riemannian
geometry: I, Math. Proc. Camb. Phil. Soc. 77 (1975) 4369.
[27] C. LeBrun, Curvature functionals, optimal metrics, and the dierential topology of
4-manifolds, in Dierent Faces of Geometry, S. Donaldson, Ya. Eliashberg and M. Gro-
mov, eds., New York, Kluwer Academic/Plenum, 2004.
37
[28] D. R. Brill, Electromagnetic elds in a homogeneous, nonisotropic universe, Phys. Rev.
133 (1964) B845B848.
[29] C. N. Pope, Axial-vector anomalies and the index theorem in charged Schwarzschild
and Taub-NUT spaces, Nucl. Phys. B141 (1978) 432444.
[30] K. Lee, E. J. Weinberg and P. Yi, Electromagnetic duality and SU(3) monopoles, Phys.
Lett. B376 (1996) 97102.
[31] J. P. Gauntlett and D. A. Lowe, Dyons and S-duality in N = 4 supersymmetric gauge
theory, Nucl. Phys. B472 (1996) 194206.
[32] B. J. Schroers and N. S. Manton, Bundles over moduli spaces and the quantisation of
BPS monopoles, Annals of Physics 225 (1993) 290338.
[33] A. Sen, Dyon-monopole bound states, self-dual harmonic forms on the multi-monopole
moduli space, and SL(2, Z) invariance in string theory, Phys. Lett. B329 (1994) 217221.
[34] G. W. Gibbons and N. S. Manton, Classical and quantum dynamics of BPS monopoles,
Nucl. Phys. B274 (1986) 183224.
[35] B. J. Schroers, Quantum scattering of BPS monopoles at low energy, Nucl. Phys. B367
(1991) 177214.
[36] W. S. Massey, The quotient space of the complex projective plane under conjugation is
a 4-sphere, Geom. Dedicata 2 (1973) 371374.
[37] N. H. Kuiper, The quotient space of CP(2) by complex conjugation is the 4-sphere,
Math. Ann. 208 (1974) 175177.
[38] S. Sethi, M. Stern and E. Zaslow, Monopole and dyon bound states in N = 2 super-
symmetric Yang-Mills theories, Nucl. Phys. B457 (1995) 484510.
38