
We relate certain abelian invariants of a knot, namely the Alexander
polynomial, the Blanchfield form, and the Arf invariant, to intersection data
of a Whitney tower in the 4ball bounded by the knot. We also give a new
3dimensional algorithm for computing these invariants.

We give a formula for the duality structure of the 3manifold obtained by
doing zeroframed surgery along a knot in the 3sphere, starting from a diagram
of the knot. We then use this to give a combinatorial algorithm for computing
the twisted Blanchfield pairing of such 3manifolds. With the twisting defined
by CassonGordon style representations, we use our computation of the twisted
Blanchfield pairing to show that some subtle satellites of genus two ribbon
knots yield nonslice knots. The construction is subtle in the sense that, once
based, the infection curve lies in the second derived subgroup of the knot
group, and that we identify these infection curves explicitly.

We determine for which complex numbers on the unit circle the LevineTristram
signature and the nullity give rise to link concordance invariants.

We show that closed, connected 4manifolds up to connected sum with copies of
the complex projective plane are classified in terms of the fundamental group,
the orientation character and an extension class involving the second homotopy
group. For fundamental groups that are torsion free or have one end, we reduce
this further to a classification in terms of the homotopy 2type.

We define an obstruction for a knot to be Z[Z]homology ribbon, and use this
to provide restrictions on the integers that can occur as the triple linking
numbers of derivative links of knots that are either homotopy ribbon or doubly
slice. Our main application finds new nondoubly slice knots. In particular
this gives new information on the doubly solvable filtration of Taehee Kim:
doubly algebraically slice ribbon knots need not be doubly (1)solvable, and
doubly algebraically slice knots need not be (0.5,1)solvable. We also discuss
potential connections to unsolved conjectures in knot concordance, such as
generalised versions of Kauffman's conjecture. Moreover it is possible that our
obstruction could fail to vanish on a slice knot.

We investigate the disparity between smooth and topological almost
concordance of knots in general 3manifolds Y. Almost concordance is defined by
considering knots in Y modulo concordance in Yx[0,1] and the action of the
concordance group of knots in the 3sphere that ties in local knots. We prove
that the trivial free homotopy class in every 3manifold other than the
3sphere contains an infinite family of knots, all topologically concordant,
but not smoothly almost concordant to one another. Then, in every lens space
and for every free homotopy class, we find a pair of topologically concordant
but not smoothly almost concordant knots. Finally, as a topological
counterpoint to these results, we show that in every lens space every free
homotopy class contains infinitely many topological almost concordance classes.

We present the Round Handle Problem, proposed by Freedman and Krushkal. It
asks whether a collection of links, which contains the Generalised Borromean
Rings, are slice in a 4manifold R constructed from adding round handles to the
four ball. A negative answer would contradict the union of the surgery
conjecture and the scobordism conjecture for 4manifolds with free fundamental
group.

We give a refined value group for the collection of triple linking numbers of
links in the 3sphere. Given two links with the same pairwise linking numbers
we show that they have the same refined triple linking number collection if and
only if the links admit homeomorphic surface systems. Moreover these two
conditions hold if and only if the link exteriors are bordant over $B
\mathbb{Z}^n$, and if and only if the third lower central series quotients
$\pi/\pi_3$ of the link groups are isomorphic preserving meridians and
longitudes. We also show that these conditions imply that the link groups have
isomorphic fourth lower central series quotients $\pi/\pi_4$, preserving
meridians.

We give a simple obstruction for a knot to be amphichiral, in terms of the
homology of the 2fold branched cover. We work with unoriented knots, and so
obstruct both positive and negative amphichirality.

We prove a decomposition formula for twisted Blanchfield pairings of
3manifolds. As an application we show that the twisted Blanchfield pairing of
a 3manifold obtained from a 3manifold Y with a representation $\phi:
Z[\pi_1(Y)] \to R$, infected by a knot J along a curve $\eta$ with $\phi(\eta)
\neq 1$, splits orthogonally as the sum of the twisted Blanchfield pairing of Y
and the ordinary Blanchfield pairing of the knot J, with the latter tensored up
from $Z[t,t^{1}]$ to R.

To a special type of grope embedded in 4space, that we call an admissible
grope, we associate a length function for each real number q at least 1. This
gives rise to a family of pseudometrics d^q, refining the slice genus metric,
on the set of concordance classes of knots, as the infimum of the length
function taken over all possible grope concordances between two knots. We
investigate the properties of these metrics. The main theorem is that the
topology induced by this metric on the knot concordance set is not discrete for
all q>1. The analogous statement for links also holds for q=1. In addition we
translate much previous work on knot concordance into distance statements. In
particular, we show that winding number zero satellite operators are
contractions in many cases, and we give lower bounds on our metrics arising
from knot signatures and higher order order signatures. This gives further
evidence in favor of the conjecture that the knot concordance group has a
fractal structure.

Given a 3manifold $Y$ and a free homotopy class in $[S^1,Y]$, we investigate
the set of topological concordance classes of knots in $Y \times [0,1]$
representing the given homotopy class. The concordance group of knots in the
3sphere acts on this set. We show in many cases that the action is not
transitive, using two techniques. Our first technique uses Reidemeister torsion
invariants, and the second uses linking numbers in covering spaces. In
particular, we show using covering links that for the trivial homotopy class,
and for any 3manifold that is not the 3sphere, the set of orbits is infinite.
On the other hand, for the case that $Y=S^1 \times S^2$, we apply topological
surgery theory to show that all knots with winding number one are concordant.

We study closed, oriented 4manifolds whose fundamental group is that of a
closed, oriented, aspherical 3manifold. We show that two such 4manifolds are
stably diffeomorphic if and only if they have the same w_2type and their
equivariant intersection forms are stably isometric. We also find explicit
algebraic invariants that determine the stable classification for spin
manifolds in this class.

We define the twisted Blanchfield pairing of a symmetric triad of chain
complexes over a group ring Z[G], together with a unitary representation of G
over an Ore domain with involution.
We prove that the pairing is sesquilinear, and we prove that it is hermitian
and nonsingular under certain extra conditions. A twisted Blanchfield pairing
is then associated to a 3manifold together with a decomposition of its
boundary into two pieces and a unitary representation of its fundamental group.

We give a new proof that the LevineTristram signatures of a link give lower
bounds for the minimal sum of the genera of a collection of oriented, locally
flat, disjointly embedded surfaces that the link can bound in the 4ball. We
call this minimal sum the 4genus of the link. We also extend a theorem of
Cochran, Friedl and Teichner to show that the 4genus of a link does not
increase under infection by a string link, which is a generalised satellite
construction, provided that certain homotopy triviality conditions hold on the
axis curves, and that enough Milnor's invariants of the infection string link
vanish. We construct knots for which the combination of the two results
determines the 4genus.

We calculate Blanchfield pairings of 3manifolds. In particular, we give a
formula for the Blanchfield pairing of a fibred 3manifold and we give a new
proof that the Blanchfield pairing of a knot can be expressed in terms of a
Seifert matrix.

We prove that a Casson tower of height 4 contains a flat embedded disc
bounded by the attaching circle, and we prove disc embedding results for height
2 and 3 Casson towers which are embedded into a 4manifold, with some
additional fundamental group assumptions. In the proofs we create a capped
grope from a Casson tower and use a refined height raising argument to
establish the existence of a symmetric grope which has two layers of caps, data
which is sufficient for a topological disc to exist, with the desired boundary.
As applications, we present new slice knots and links by giving direct
geometric constructions of slicing discs. In particular we construct a family
of slice knots which are potential counterexamples to the homotopy ribbon slice
conjecture.

We give an example of a 3component smoothly slice boundary link, each of
whose components has a genus one Seifert surface, such that any metaboliser of
the boundary link Seifert form is represented by 3 curves on the Seifert
surfaces that form a link with nonvanishing Milnor triple linking number. We
also give a generalisation to mcomponent links and higher Milnor invariants.
We prove that our examples are ribbon and that all ribbon links are boundary
slice.

We use topological surgery in dimension four to give sufficient conditions
for the zero framed surgery manifold of a 3component link to be homology
cobordant to the 3torus, which arises from zero framed surgery on the
Borromean rings, via a topological homology cobordism preserving the homotopy
classes of the meridians. This enables us to give new examples of 3component
links with unknotted components and vanishing pairwise linking numbers, such
that any two of these links have homology cobordant zero surgeries in the above
sense, but the zero surgery manifolds are not homeomorphic. Moreover the links
are not concordant to one another, and in fact they can be chosen to be height
n but not height n+1 symmetric grope concordant, for each n which is at least
three.

We show that the following unlinking strategy does not always yield an
optimal sequence of crossing changes: first split the link with the minimal
number of crossing changes, and then unknot the resulting components.

Given a link in $S^3$ we will use invariants derived from the Alexander
module and the Blanchfield pairing to obtain lower bounds on the Gordian
distance between links, the unlinking number and various splitting numbers.
These lower bounds generalise results recently obtained by Kawauchi.
We give an application restricting the knot types which can arise from a
sequence of splitting operations on a link. This allows us to answer a question
asked by Colin Adams in 1996.

Given a sequence of oriented links L^1,L^2,L^3,... each of which has a
distinguished, unknotted component, there is a decomposition of the 3sphere
naturally associated to it, which is constructed as the components of the
intersection of an infinite sequence of nested solid tori. The Bing and
Whitehead continua are simple, well known examples. We give a necessary and
sufficient criterion to determine whether such a decomposition is shrinkable,
generalising previous work of F. Ancel and M. Starbird and others. This
criterion can effectively determine, in many cases, whether the quotient map
which identifies the elements of the decomposition to points can be
approximated by homeomorphisms.

The bipolar filtration introduced by T. Cochran, S. Harvey, and P. Horn is a
framework for the study of smooth concordance of topologically slice knots and
links. It is known that there are topologically slice 1bipolar knots which are
not 2bipolar. For knots, this is the highest known level at which the
filtration does not stabilize. For the case of links with two or more
components, we prove that the filtration does not stabilize at any level: for
any n, there are topologically slice links which are nbipolar but not
(n+1)bipolar. In the proof we describe an explicit geometric construction
which raises the bipolar height of certain links exactly by one. We show this
using the covering link calculus. Furthermore we discover that the bipolar
filtration of the group of topologically slice string links modulo smooth
concordance has a rich algebraic structure.

We prove that an embedded cobordism between manifolds with boundary can be
split into a sequence of right product and left product cobordisms, if the
codimension of the embedding is at least two. This is a topological counterpart
of the algebraic splitting theorem for embedded cobordisms of the first author,
A. Nemethi and A. Ranicki. In the codimension one case, we provide a slightly
weaker statement. We also give proofs of rearrangement and cancellation
theorems for handles of embedded submanifolds with boundary.

The splitting number of a link is the minimal number of crossing changes
between different components required, on any diagram, to convert it to a split
link. We introduce new techniques to compute the splitting number, involving
covering links and Alexander invariants. As an application, we completely
determine the splitting numbers of links with 9 or fewer crossings. Also, with
these techniques, we either reprove or improve upon the lower bounds for
splitting numbers of links computed by J. Batson and C. Seed using Khovanov
homology.