
A pseudoedge graph of a convex polyhedron K is a 3connected embedded graph
in K whose vertices coincide with those of K, whose edges are distance
minimizing geodesics, and whose faces are convex. We construct a convex
polyhedron K in Euclidean 3space with a pseudoedge graph with respect to
which K is not unfoldable. The proof is based on a result of Pogorelov on
convex caps with prescribed curvature, and an unfoldability obstruction for
almost flat convex caps due to Tarasov. Our example, which has 340 vertices,
significantly simplifies an earlier construction by Tarasov, and confirms that
Durer's conjecture does not hold for pseudoedge unfoldings.

We show that the torsion of any simple closed curve $\Gamma$ in Euclidean
3space changes sign at least $4$ times provided that it is starshaped and
locally convex with respect to a point $o$ in the interior of its convex hull.
The latter condition means that through each point $p$ of $\Gamma$ there passes
a plane $H$, not containing $o$, such that a neighborhood of $p$ in $\Gamma$
lies on the same side of $H$ as does $o$. This generalizes the four vertex
theorem of Sedykh for convex space curves. Following Thorbergsson and Umehara,
we reduce the proof to the result of Segre on inflections of spherical curves,
which is also known as Arnold's tennis ball theorem.

We prove that any properly oriented $C^{2,1}$ isometric immersion of a
positively curved Riemannian surface M into Euclidean 3space is uniquely
determined, up to a rigid motion, by its values on any curve segment in M. A
generalization of this result to nonnegatively curved surfaces is presented as
well under suitable conditions on their parabolic points. Thus we obtain a
local version of CohnVossen's rigidity theorem for convex surfaces subject to
a Dirichlet condition. The proof employs in part Hormander's unique
continuation principle for elliptic PDEs.

We prove the existence of a center, or continuous selection of a point, in
the relative interior of $C^1$ embedded $k$disks in Riemannian $n$manifolds.
If $k\le 3$ the center can be made equivariant with respect to the isometries
of the manifold, and under mild assumptions the same holds for $k=4=n$. By
contrast, for every $n\ge k\ge 6$ there are examples where an equivariant
center does not exist. The center can be chosen to agree with any of the
classical centers defined on the set of convex compacta in the Euclidean space.

The width $w$ of a curve $\gamma$ in Euclidean space $R^n$ is the infimum of
the distances between all pairs of parallel hyperplanes which bound $\gamma$,
while its inradius $r$ is the supremum of the radii of all spheres which are
contained in the convex hull of $\gamma$ and are disjoint from $\gamma$. We use
a mixture of topological and integral geometric techniques, including an
application of Borsuk Ulam theorem due to Wienholtz and Crofton's formulas, to
obtain lower bounds on the length of $\gamma$ subject to constraints on $r$ and
$w$. The special case of closed curves is also considered in each category. Our
estimates confirm some conjectures of Zalgaller up to $99\%$ of their stated
value, while we also disprove one of them.

We prove that in Euclidean space $R^{n+1}$ any compact immersed nonnegatively
curved hypersurface $M$ with free boundary on the sphere $S^n$ is an embedded
convex topological disk. In particular, when the $m^{th}$ mean curvature of $M$
is constant, for any $1\leq m\leq n$, $M$ is a spherical cap or an equatorial
disk.

We prove that the torsion of any closed space curve which bounds a simply
connected locally convex surface vanishes at least 4 times. This answers a
question of Rosenberg related to a problem of Yau on characterizing the
boundary of positively curved disks in Euclidean space. Furthermore, our result
generalizes the 4 vertex theorem of Sedykh for convex space curves, and thus
constitutes a far reaching extension of the classical 4 vertex theorem. The
proof involves studying the arrangement of convex caps in a locally convex
surface, and yields a Bose type formula for these objects.

The total diameter of a closed planar curve $C\subset R^2$ is the integral of
its antipodal chord lengths. We show that this quantity is bounded below by
twice the area of $C$. Furthermore, when $C$ is convex or centrally symmetric,
the lower bound is twice as large. Both inequalities are sharp and the equality
holds in the convex case only when $C$ is a circle. We also generalize these
results to $m$ dimensional submanifolds of $R^n$, where the "area" will be
defined in terms of the mod $2$ winding numbers of the submanifold about the
$nm1$ dimensional affine subspaces of $R^n$.

We show that every convex polyhedron admits a simple edge unfolding after an
affine transformation. In particular there exists no combinatorial obstruction
to a positive resolution of Durer's unfoldability problem, which answers a
question of Croft, Falconer, and Guy. Among other techniques, the proof employs
a topological characterization for embeddings among the planar immersions of
the disk.

We characterize embedded $\C^1$ hypersurfaces of $\R^n$ as the only locally
closed sets with continuously varying flat tangent cones whose
measuretheoreticmultiplicity is at most $m<3/2$. It follows then that any
(topological) hypersurface which has flat tangent cones and is supported
everywhere by balls of uniform radius is $\C^1$. In the real analytic case the
same conclusion holds under the weakened hypothesis that each tangent cone be a
hypersurface. In particular, any convex real analytic hypersurface
$X\subset\R^n$ is $\C^1$. Furthermore, if $X$ is real algebraic, strictly
convex, and unbounded then its projective closure is a $\C^1$ hypersurface as
well, which shows that $X$ is the graph of a function defined over an entire
hyperplane.

We show that every smooth closed curve C immersed in Euclidean 3space
satisfies the sharp inequality 2(P+I)+V >5 which relates the numbers P of pairs
of parallel tangent lines, I of inflections (or points of vanishing curvature),
and V of vertices (or points of vanishing torsion) of C. We also show that
2(P'+I)+V >3, where P' is the number of pairs of concordant parallel tangent
lines. The proofs, which employ curve shortening flow with surgery, are based
on corresponding inequalities for the numbers of double points, singularities,
and inflections of closed curves in the real projective plane and the sphere
which intersect every closed geodesic. These findings extend some classical
results in curve theory including works of Moebius, Fenchel, and Segre, which
is also known as Arnold's "tennis ball theorem".

We show that Caratheodory's conjecture, on umbilical points of closed convex
surfaces, may be reformulated in terms of the existence of at least one umbilic
in the graphs of functions f: R^2>R whose gradient decays uniformly faster
than 1/r. The divergence theorem then yields a pair of integral equations for
the normal curvatures of these graphs, which establish some weaker forms of the
conjecture. In particular, we show that there are uncountably many principal
lines in the graph of f whose projection into R^2 are parallel to any given
direction.

Given any finite subset X of the sphere S^n, n>1, which includes no pairs of
antipodal points, we explicitly construct smoothly immersed closed orientable
hypersurfaces in Euclidean space R^{n+1} whose Gauss map misses X. In
particular, this answers a question of M. Gromov.

We uncover some connections between the topology of a complete Riemannian
surface M and the minimum number of vertices, i.e., critical points of geodesic
curvature, of closed curves in M. In particular we show that the space forms
with finite fundamental group are the only surfaces in which every simple
closed curve has more than two vertices. Further we characterize the simply
connected space forms as the only surfaces in which every closed curve bounding
a compact immersed surface has more than two vertices.

We study the topology of the space $\d\K^n$ of complete convex hypersurfaces
of $\R^n$ which are homeomorphic to $\R^{n1}$. In particular, using Minkowski
sums, we construct a deformation retraction of $\d\K^n$ onto the Grassmannian
space of hyperplanes. So every hypersurface in $\d \K^n$ may be flattened in a
canonical way. Further, the total curvature of each hypersurface evolves
continuously and monotonically under this deformation. We also show that,
modulo proper rotations, the subspaces of $\d\K^n$ consisting of smooth,
strictly convex, or positively curved hypersurfaces are each contractible,
which settles a question of H. Rosenberg.

We study the geometry and topology of immersed surfaces in Euclidean 3space
whose Gauss map satisfies a certain twopieceproperty, and solve the ``shadow
problem" formulated by H. Wente.

A skew loop is a closed curve without parallel tangent lines. We prove:
The only complete surfaces in euclidean 3space with a point of positive
curvature and no skew loops are the quadrics.
In particular, ellipsoids are the only closed surfaces without skew loops.
We also prove results about skew loops on cylinders and positively curved
surfaces.

We define a new class of knot energies (known as renormalization energies)
and prove that a broad class of these energies are uniquely minimized by the
round circle. Most of O'Hara's knot energies belong to this class. This proves
two conjectures of O'Hara and of Freedman, He, and Wang. We also find energies
not minimized by a round circle. The proof is based on a theorem of G. Luko on
average chord lengths of closed curves.