$$\begin{array}{cc}X({\pi}_{1}\overline{M},{SL}_{2}(\mathbb{C}\left)\right)& \u27f6{\mathbb{C}}^{N+\tau}\end{array}$$ 
$$\begin{array}{cc}\chi & \mapsto \left({t}_{{\mu}_{1}}\right(\chi ),\dots ,{t}_{{\mu}_{N}}(\chi ),{t}_{{m}_{1}}(\chi ),\dots ,{t}_{{m}_{\tau}}(\chi \left)\right)\end{array}$$ 
$$\begin{array}{}\end{array}$$ 
If we fix
${m}_{i},{l}_{i}$
generators of
${\pi}_{1}{T}_{i}^{2}$
, where
${T}_{i}^{2}$
is the horospherical crosssection of the
$i$
th toral cusp, we can pass to Dehn surgery coefficients, cf. for example [CHK] , i.e. we obtain a local parametrization
For the
$i$
th toral cusp,
$[{x}_{i},{y}_{i}]=\infty $
corresponds to the cusped structure, whereas lines through the origin with rational slope correspond to (families of ) conemanifold structures on a fixed topological filling.
$$\begin{array}{cc}X({\pi}_{1}\overline{M},{SL}_{2}(\mathbb{C}\left)\right)& \u27f6{\mathbb{C}}^{N}\times ({\mathbb{R}}^{2}\cup \{\infty \}/\pm 1{)}^{\tau}\end{array}$$ 
$$\begin{array}{cc}\chi & \mapsto \left({t}_{{\mu}_{1}}\right(\chi ),\dots ,{t}_{{\mu}_{N}}(\chi ),[{x}_{1},{y}_{1}],\dots ,[{x}_{\tau},{y}_{\tau}\left]\right).\end{array}$$ 
$$\begin{array}{}\end{array}$$ 
More precisely, if
${x}_{i}/{y}_{i}\in \mathbb{Q}\cup \{\infty \}$
, write
${x}_{i}/{y}_{i}={p}_{i}/{q}_{i}$
with
${p}_{i},{q}_{i}$
coprime integers. The conemanifold structure corresponding to
$[{x}_{i},{y}_{i}]$
is obtained by glueing in a singular solid torus with coneangle
$2\pi {p}_{i}/{x}_{i}$
such that the curve
${p}_{i}{m}_{i}+{q}_{i}{l}_{i}$
bounds the singular disk. Note in particular that moving out to
$\infty $
along the line with slope
${p}_{i}/{q}_{i}$
corresponds to monotonously decreasing the coneangle towards
$0$
while preserving the topological type of the cone3manifold.
Dehn surgery coefficients different from those above do not give rise to conemanifold structures. We will generally say that a hyperbolic structure which is obtained by varying the
$i$
th Dehn surgery coefficient has a Dehn surgery type singularity at the
$i$
th toral cusp.
If we restrict to characters corresponding to holonomies of conemanifold structures, i.e. meridians are mapped to pure rotations with rotationangle given by the coneangle, then we can recover the trace of meridian from the coneangle, cf. for example [HK] , [Wei] . Since via mapping a hyperbolic structure to the character corresponding to its holonomy the deformation space of hyperbolic structures on
$M$
is locally homeomorphic to
$X({\pi}_{1}\overline{M},{SL}_{2}(\mathbb{C}\left)\right)$
, cf. [Gol] , we obtain the following local deformation theorem:
Theorem 2.16Let
$X$
be an orientable hyperbolic cone3manifold of finite volume with coneangles
$\le \pi $
, at most finitely many ends which are (smooth or singular) cusps with compact crosssections and possibly boundary consisting of totally geodesic hyperbolic turnovers. Suppose that
${\mathbf{E}}^{2}(\pi ,\pi ,\pi ,\pi )$
doesn't occur as cusp crosssection. Then assigning the vector of coneangles
$({\alpha}_{1},\dots ,{\alpha}_{N})$
and the family of Dehn surgery coefficients
$\left(\right[{x}_{1},{y}_{1}],\dots ,[{x}_{\tau},{y}_{\tau}\left]\right)$
to a hyperbolic conemanifold structure with Dehn surgery type singularities yields a local parametrization of the space of such structures near the given one.
The geometry of the link of a vertex is determined by the coneangles
$\alpha ,\beta $
and
$\gamma $
of the adjacent edges, cf. Lemma 2.7 :
If
$\alpha +\beta +\gamma >2\pi $
, the link is spherical, i.e. given by a spherical turnover
${\mathbf{S}}^{2}(\alpha ,\beta ,\gamma )$
. The vertex is just an ordinary vertex sitting inside the cone3manifold. If
$\alpha +\beta +\gamma =2\pi $
, the link is horospherical, i.e. given by a horospherical Euclidean turnover
${\mathbf{E}}^{2}(\alpha ,\beta ,\gamma )$
. The vertex itself should now be considered to sit at infinity. Finally, if
$\alpha +\beta +\gamma <2\pi $
, the link is hyperspherical, i.e. given by a totally geodesic hyperbolic turnover
${\mathbf{H}}^{2}(\alpha ,\beta ,\gamma )$
. The vertex should now be considered to sit beyond the totally geodesic boundary.
By leaving the Dehn surgery coefficients untouched, we obtain a local rigidity result for a class of hyperbolic cone3manifolds of finite volume analogous to the one contained in [Wei] , cf. also Theorem 1.1 :
Corollary 2.17 (local rigidity: finitevolume case)Let
$X$
be an orientable hyperbolic cone3manifold of finite volume with coneangles
$\le \pi $
, at most finitely many ends which are (smooth or singular) cusps with compact crosssections and possibly boundary consisting of totally geodesic hyperbolic turnovers.
Suppose that
${\mathbf{E}}^{2}(\pi ,\pi ,\pi ,\pi )$
doesn't occur as cusp crosssection.
Then assigning the vector of coneangles
$({\alpha}_{1},\dots ,{\alpha}_{N})$
to a hyperbolic conemanifold structure yields a local parametrization of the space of such structures near the given one.
Similarly, by leaving the coneangles untouched, we obtain the conclusion of Thurston's hyperbolic Dehn surgery theorem in the setting of hyperbolic cone3manifolds of finite volume:
Corollary 2.18 (hyperbolic Dehn surgery)Let
$X$
be an orientable hyperbolic cone3manifold of finite volume with coneangles
$\le \pi $
, at most finitely many ends which are (smooth or singular) cusps with compact crosssections and possibly boundary consisting of totally geodesic hyperbolic turnovers. Suppose that
${\mathbf{E}}^{2}(\pi ,\pi ,\pi ,\pi )$
doesn't occur as cusp crosssection. Then for each toral cusp the conclusion of Thurston's hyperbolic Dehn surgery theorem holds, in particular all but finitely many Dehn fillings (per cusp) are hyperbolic.
Note that by Theorem 2.16 we can moreover independently perturb the coneangles and perform the operation of hyperbolic Dehn surgery at a toral cusp. This will be of importance in the proof of the main theorem.
3 Global Rigidity
3.1 Geometric properties of cone3manifolds
In this section we mostly review material contained in [BLP2] . We will state results only as far as directly needed for our applications. For proofs and more details we ask the reader to consult [BLP2] .
In the following let
$X$
be a cone3manifold of curvature
$\kappa $
. For
$r>0$
(and
$r\le \pi /\sqrt{\kappa}$
if
$\kappa >0$
) and
$S$
a spherical conesurface, the standard ball of radius
$r$
with link
$S$
is simply the
$\kappa $
cone over
$S$
, which we as usual denote by
${cone}_{\kappa ,(0,r)}S$
.
For
$p\in X$
consider
${B}_{r}\left(p\right)$
, the embedded
$r$
ball around p in
$X$
.
Recall that for
$\varepsilon >0$
small enough we have by definition of conemanifold structure
$${B}_{\varepsilon}\left(p\right)\sim ={cone}_{\kappa ,(0,\varepsilon )}S\left(p\right),$$
where
$S\left(p\right)$
is the link of
$p$
in
$X$
.
The injectivity radius of
$X$
at
$p$
,
${r}_{inj}\left(p\right)$
, is the supremum over all
$r>0$
such that
$${B}_{r}\left(p\right)\sim ={cone}_{\kappa ,(0,r)}S\left(p\right).$$
The coneinjectivity radius of
$X$
at
$p$
,
${r}_{coneinj}\left(p\right)$
, is the supremum over all
$r>0$
such that
${B}_{r}\left(p\right)$
is contained in an embedded standard ball, i.e. such that there exists
$q\in X$
and
$R>0$
with
${B}_{r}\left(p\right)\subset {B}_{R}\left(q\right)$
and
$${B}_{R}\left(q\right)\sim ={cone}_{\kappa ,(0,R)}S\left(q\right).$$
For
$\varrho >0$
,
$X$
is said to be
$\varrho $
thick (at a point
$p$
) if it contains an embedded smooth standard ball of radius
$\varrho $
(centered at
$p$
).
Let
$X$
and
$Y$
be metric spaces and
$\varepsilon >0$
. Following [BLP2] we call a map
$\phi :X\to Y$
a
$(1+\varepsilon )$
bilipschitz embedding if
$$(1+\varepsilon {)}^{1}\cdot d({x}_{1},{x}_{2})<d(\phi \left({x}_{1}\right),\phi \left({x}_{2}\right))<(1+\varepsilon )\cdot d({x}_{1},{x}_{2})$$
holds for all
${x}_{1},{x}_{2}\in X$
.
Definition 3.1 (geometric convergence)Let
$({X}_{n},{x}_{n}{)}_{n\in \mathbb{N}}$
be a sequence of pointed cone3manifolds. We say that the sequence
$({X}_{n},{x}_{n})$
converges geometrically to a pointed cone3manifold
$({X}_{\infty},{x}_{\infty})$
if for every
$R>0$
and
$\varepsilon >0$
there exists
$N=N(R,\varepsilon )\in \mathbb{N}$
such that for all
$n\ge N$
there is a
$(1+\varepsilon )$
bilipschitz embedding
${\phi}_{n}:{B}_{R}\left({x}_{\infty}\right)\to {X}_{n}$
satisfying:
Note that if the
${X}_{n}$
have curvature
${\kappa}_{n}\in \mathbb{R}$
, then
${X}_{\infty}$
will have curvature
${\kappa}_{\infty}={lim}_{n\to \infty}{\kappa}_{n}$
. The coneangle along an edge of
${\Sigma}_{\infty}$
will be the limit of the coneangles along the corresponding edge in
${\Sigma}_{n}$
. Note however that part of the singular locus of the approximating cone3manifolds may disappear in the limit.  1. $d\left({\phi}_{n}\right({x}_{\infty}),{x}_{n})<\varepsilon $ ,
 2. ${B}_{(1\varepsilon )R}\left({x}_{n}\right)\subset {\phi}_{n}\left({B}_{R}\right({x}_{\infty}\left)\right)$ , and

3.
${\phi}_{n}\left({B}_{R}\right({x}_{\infty})\cap {\Sigma}_{\infty})={\phi}_{n}\left({B}_{R}\right({x}_{\infty}\left)\right)\cap {\Sigma}_{n}$
.
Given a geometrically convergent sequence
$({X}_{n},{x}_{n})$
we may without loss of generality assume that the
$(1+\varepsilon )$
bilipschitz embeddings
${\phi}_{n}$
restrict to smooth
$(1+\varepsilon )$
bilipschitz diffeomorphisms
$${F}_{n}:{B}_{R}\left({x}_{\infty}\right)\cap {M}_{\infty}\to {\phi}_{n}\left({B}_{R}\right({x}_{\infty}\left)\right)\cap {M}_{n},$$
where
${M}_{\infty}$
, resp.
${M}_{n}$
denotes the smooth part of
${X}_{\infty}$
, resp.
${X}_{n}$
. Let us assume in the following that
${\kappa}_{n}=\kappa $
independent of
$n$
and therefore also
${\kappa}_{\infty}=\kappa $
. We may further assume that (possibly after composing with isometries)
$${dev}_{n}\circ {\stackrel{~}{F}}_{n}:\stackrel{~}{{B}_{R}\left({x}_{\infty}\right)\cap {M}_{\infty}}\to \mathbf{M}3\kappa $$
converges to
${dev}_{\infty}$
restricted to
$\stackrel{~}{{B}_{R}\left({x}_{\infty}\right)\cap {M}_{\infty}}$
uniformly on compact sets (together with all derivatives), and that
$${hol}_{n}\circ ({F}_{n}{)}_{*}:{\pi}_{1}({B}_{R}\left({x}_{\infty}\right)\cap {M}_{\infty})\to Isom+\mathbf{M}3\kappa $$
converges to
${hol}_{\infty}$
restricted to
${\pi}_{1}\left({B}_{R}\right({x}_{\infty})\cap {M}_{\infty})$
in the compactopen topology.
Theorem 3.2 (compactness)[BLP2] Let
$({X}_{n},{x}_{n}{)}_{n\in \mathbb{N}}$
be a sequence of pointed cone3manifolds with curvatures
${\kappa}_{n}\in [1,1]$
, coneangles
$\le \pi $
and possibly with totally geodesic boundary. Suppose that for some
$\rho >0$
each
${X}_{n}$
is
$\rho $
thick at
${x}_{n}$
.
Then (possibly after passing to a subsequence) the sequence
$({X}_{n},{x}_{n})$
converges geometrically to a pointed cone3manifold
$({X}_{\infty},{x}_{\infty})$
with curvature
${\kappa}_{\infty}={lim}_{n\to \infty}{\kappa}_{n}$
.
In the hyperbolic case, M. Boileau, B. Leeb and J. Porti derive a thickthin decomposition in the spirit of the classical Margulis lemma applied to complete hyperbolic 3manifolds. To state the results we need some further notation, which is introduced in [BLP2] :
Recall that an embedded connected surface
$S$
in
${\mathbf{H}}^{3}$
is called umbilic, if all principal curvatures have the same value
$pc\left(S\right)$
. Its intrinsic curvature is then given by
${\kappa}_{S}=1+pc(S{)}^{2}$
. A umbilic surface
$S$
is called spherical if
${\kappa}_{S}>0$
, horospherical if
${\kappa}_{S}=0$
and hyperspherical if
${\kappa}_{S}<0$
.
Recall further that
${\mathbf{H}}^{3}$
carries the following natural foliations by umbilic surfaces:
Let
$S$
be a conesurface of curvature
${\kappa}_{S}\in \{1,0,1\}$
. Depending on the curvature
${\kappa}_{S}$
we define the complete tube over
$S$
,
${tube}_{1}S$
, as follows:
In any case,
${tube}_{1}S$
carries a natural foliation by umbilic surfaces equidistant to
$S$
. Its leaves are spherical if
${\kappa}_{S}=1$
, horospherical if
${\kappa}_{S}=0$
and hyperspherical if
${\kappa}_{S}=1$
.
 1. The foliation by distance spheres of a point $p\in {\mathbf{H}}^{3}$ : the leaves are spherical.
 2. The foliation by horospheres centered at $p\in {\partial}_{\infty}{\mathbf{H}}^{3}$ : the leaves are horospherical.

3.
The foliation by surfaces equidistant to a totally geodesic
${\mathbf{H}}^{2}\subset {\mathbf{H}}^{3}$
: the leaves are hyperspherical.

1.
If
${\kappa}_{S}=1$
, then
${tube}_{1}S$
is just the complete hyperbolic cone over
$S$
: $${tube}_{1}S={cone}_{1}S.$$
 2. If ${\kappa}_{S}=0$ , then ${tube}_{1}S$ is the complete hyperbolic cusp with horospherical crosssection $S$ : $${tube}_{1}S=\mathbb{R}\times S$$ with metric $${g}^{hyp}=d{t}^{2}+{e}^{2t}{g}^{S}.$$

3.
If
${\kappa}_{S}=1$
, then
${tube}_{1}S$
is the complete hyperbolic neck with totally geodesic central leaf
$S$
: $${tube}_{1}S=\mathbb{R}\times S$$ with metric $${g}^{hyp}=d{t}^{2}+cosh(t{)}^{2}{g}^{S}.$$
Following [BLP2] , a umbilic tube will be a closed connected subset of
${tube}_{1}S$
, which is a union of leaves of the natural umbilic foliation.
Theorem 3.3 (thin parts)[BLP2] For
$D>0$
and
$0<\zeta \le \eta <\pi $
there exist constants
$i=i(D,\zeta ,\eta )>0$
,
$P=P(\zeta ,D)>0$
,
$\rho =\rho (D,\zeta ,\eta )>0$
such that the following holds: If
$X$
is an orientable hyperbolic cone3manifold without boundary, coneangles
$\in [\zeta ,\eta ]$
and
$diam\left(X\right)\ge D$
, then
$X$
contains a (possibly empty) disjoint union
${X}^{thin}$
of submanifolds which belong to the following list:
Furthermore, the boundary of each component of
${X}^{thin}$
is nonempty, strictly convex with principal curvatures
$\le \pi $
and each of its (at most two) components contains a smooth point
$p$
with
${r}_{inj}\left(p\right)\ge \rho $
(measured in
$X$
); each component of
${X}^{thin}$
contains an embedded smooth standard ball of radius
$\rho $
; all singular vertices are contained in
${X}^{thin}$
, and on
$X\backslash {X}^{thin}$
holds
${r}_{coneinj}\ge i$
.
Not that the thin part is canonically foliated: smooth cusps by the horospherical crosssections, tubular neighbourhoods of (smooth or singular) geodesics by the distance tori and umbilic tubes by the natural umbilic foliations.
 1. smooth Margulis tubes, i.e. tubular neighbourhoods of closed geodesics and smooth cusps of rank one or two,
 2. tubular neighbourhoods of closed singular geodesics,

3.
umbilic tubes with turnover crosssections which have strictly convex boundary, i.e. standard (singular) balls, cusps and necks.
Corollary 3.4 (thickness)[BLP2] There exists
$r=r(D,\zeta ,\eta )>0$
such that if
$X$
is as in Theorem 3.3 , then
$X$
is
$r$
thick, i.e.
$X$
contains an embedded smooth standard ball of radius
$r$
.
Corollary 3.5 (finiteness)[BLP2] Let
$X$
be as in Theorem 3.3 and suppose in addition that
$vol\left(X\right)<\infty $
. Then
$X$
has finitely many ends and all of them are (smooth or singular) cusps with compact crosssections.
By doubling along the boundary we obtain corresponding statements for hyperbolic cone3manifolds with totally geodesic boundary.
3.2 The hyperbolic case
Let
$X$
be a hyperbolic cone3manifold of finite volume with coneangles
$\le \pi $
, at most finitely many ends which are (smooth or singular) cusps with compact crosssections and possibly boundary consisting of totally geodesic hyperbolic turnovers. Suppose that
${\mathbf{E}}^{2}(\pi ,\pi ,\pi ,\pi )$
does not occur as cusp crosssection. We will describe how to obtain a continuous family of hyperbolic cone3manifolds of the same topological type connecting the given one with a complete (nonsingular) hyperbolic 3manifold of finite volume, possibly with boundary consisting of thricepunctured spheres.
Let
$N$
denote the number of edges contained in
$\Sigma $
and
$({\alpha}_{1},\dots ,{\alpha}_{N})$
the vector of coneangles. Using local rigidity we may assume that the coneangles
${\alpha}_{i}$
are already strictly smaller than
$\pi $
, i.e.
${\alpha}_{i}\le \eta <\pi $
.
The argument will proceed in two steps:
 1. Given a vector $({\beta}_{1},\dots ,{\beta}_{N})$ of small targetangles, construct a continuous angledecreasing family of hyperbolic cone3manifolds $({X}_{\tau}{)}_{\tau \in [0,1]}$ of constant topological type connecting $X={X}_{1}$ with a hyperbolic cone3manifold ${X}_{0}$ with coneangles $({\beta}_{1},\dots ,{\beta}_{N})$ .

2.
Given a hyperbolic cone3manifold
$X$
with small coneangles, construct a nonsingular hyperbolic structure on the smooth part of
$X$
and connect it with
$X$
by a continuous angledecreasing family of hyperbolic cone3manifolds of constant topological type.
Remark 3.6A hyperbolic cone3manifold will be considered to have small coneangles if the coneangles are small enough to force all links to be hyperspherical, e.g. coneangles
$<\frac{2\pi}{3}$
would be sufficient to guarantee this.
3.2.1 Deforming into small coneangles
Let
$N$
denote the number of edges contained in
$\Sigma $
and let
$({\alpha}_{1},\dots ,{\alpha}_{N})$
be the vector of coneangles of
$X$
. Let
$({\beta}_{1},\dots ,{\beta}_{N})$
with
$0<{\beta}_{i}\le {\alpha}_{i}$
be a specified vector of targetangles. We consider the interval