MINIMAL SUBMANIFOLDS

arXiv:math/0504158v2 [math.DG] 18 Nov 2005

TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

Contents
1. Introduction 2

Part 1. Classical and almost classical results 2

1.1. The Gauss map 3
1.2. Minimal graphs 4
1.3. The maximum principle 5
2. Monotonicity and the mean value inequality 7
3. Rado’s theorem 8
4. The theorems of Bernstein and Bers 9
5. Simons inequality 10
6. Heinz’s curvature estimate for graphs 11
7. Embedded minimal disks with area bounds 11
8. Stable minimal surfaces 12
9. Regularity theory 13
9.1. ǫ–regularity and the singular set 14
9.2. Tangent cone analysis 14

Part 2. Embedded minimal surfaces 15

10.1. Multi–valued graphs 15
10.2. The sublinear growth of the separation 16
11. Embedded minimal disks 17
12. Fixed genus 20
12.1. Uniformly locally simply connected 20
12.2. Fixed genus 23

Part 3. Global theory of minimal surfaces in R3 25

13.1. Minimal surfaces with finite total curvature 25
13.2. The uniqueness of the catenoid 25
14. Global theory of embedded minimal surfaces 25
15. The Calabi-Yau conjectures 27

Part 4. Constructing minimal surfaces 29

16. The Plateau Problem 29
17. The Weierstrass representation 30
18. Area–minimizing surfaces 31
19. The min–max construction of minimal surfaces 32

The authors were partially supported by NSF Grants DMS 0104453 and DMS 0405695.
1
2 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

Part 5. Some applications of minimal surfaces 33

20.1. The positive mass theorem 34
20.2. Black holes 35
20.3. Constant mean curvature surfaces 36
20.4. Finite extinction for Ricci flow 36
References 37

1. Introduction
In this article, we survey some old and new results about minimal surfaces and
submanifolds. The field of minimal surfaces has its origin in the mid eighteenth
century with the work of Euler and Lagrange but it has very recently seen major
advances that have solved many long standing open conjectures in the field. In what
follows, we give a quick tour through the field, starting with the definition and the
classical results and ending up with current areas of research. Many references are
given for further reading.
Soap films, soap bubbles, and surface tension were extensively studied by the
Belgian physicist and inventor (the inventor of the stroboscope) Joseph Plateau in
the first half of the nineteenth century. At least since his studies, it has been known
that the right mathematical model for soap films are minimal surfaces – the soap
film is in a state of minimum energy when it is covering the least possible amount
of area. Minimal surfaces and equations like the minimal surface equation have
served as mathematical models for many physical problems.
The field of minimal surfaces dates back to the publication in 1762 of Lagrange’s
famous memoir “Essai d’une nouvelle méthode pour déterminer les maxima et les
minima des formules intégrales indéfinies”. Euler had already in a paper published
in 1744 discussed minimizing properties of the surface now known as the catenoid,
but he only considered variations within a certain class of surfaces. In the almost
one quarter of a millennium that has past since Lagrange’s memoir minimal surfaces
has remained a vibrant area of research and there are many reasons why. The study
of minimal surfaces was the birthplace of regularity theory. It lies on the intersection
of nonlinear elliptic PDE, geometry, topology and general relativity.

Part 1. Classical and almost classical results

Let Σ ⊂ Rn be a smooth k-dimensional submanifold (possibly with boundary)
and C0∞ (N Σ) the space of all infinitely differentiable, compactly supported, normal
vector fields on Σ. Given Φ in C0∞ (N Σ), consider the one–parameter variation
(1.1) Σt,Φ = {x + t Φ(x)|x ∈ Σ} .
The so called first variation formula of volume is the equation
d
Z
(1.2) Vol(Σt,Φ ) = hΦ , Hi ,
dt t=0 Σ

where H is the mean curvature (vector) of Σ. Here, and throughout this paper,
integration is with respect to dvol. (When Σ is noncompact, then Σt,Φ in (1.2) is
 MINIMAL SUBMANIFOLDS 3

replaced by Γt,Φ , where Γ is any compact set containing the support of Φ.) The
submanifold Σ is said to be a minimal submanifold (or just minimal) if

d
(1.3) Vol(Σt,Φ ) = 0 for all Φ ∈ C0∞ (N Σ)
dt t=0

or, equivalently by (1.2), if the mean curvature H is identically zero. Thus Σ is

minimal if and only if it is a critical point for the volume functional. (Since a
critical point is not necessarily a minimum the term “minimal” is misleading, but
it is time honored. The equation for a critical point is also sometimes called the
Euler–Lagrange equation.)
Suppose now, for simplicity, that Σ is an oriented hypersurface with unit normal
nΣ . We can then write a normal vector field Φ ∈ C0∞ (N Σ) as Φ = φnΣ , where the
function φ is in the space C0∞ (Σ) of infinitely differentiable, compactly supported
functions on Σ. Using this, a computation shows that if Σ is minimal, then

d2
Z
(1.4) Vol(Σt,φnΣ ) = − φ LΣ φ ,
dt2 t=0 Σ

where

(1.5) LΣ φ = ∆Σ φ + |A|2 φ

is the second variational (or Jacobi) operator. Here ∆Σ is the Laplacian on Σ

and A is the second fundamental form. So |A|2 = κ21 + κ22 + · · · + κ2n−1 , where
κ1 , . . . κn−1 are the principal curvatures of Σ and H = (κ1 + · · · + κn−1 ) nΣ . A
minimal submanifold Σ is said to be stable if
d2
(1.6) Vol(Σt,Φ ) ≥ 0 for all Φ ∈ C0∞ (N Σ) .
dt2 t=0

Integrating by parts in (1.4), we see that stability is equivalent to the so called

stability inequality
Z Z
(1.7) |A|2 φ2 ≤ |∇φ|2 .

More generally, the Morse index of a minimal submanifold is defined to be the

number of negative eigenvalues of the operator L. Thus, a stable submanifold has
Morse index zero.

1.1. The Gauss map. Let Σ2 ⊂ R3 be a surface (not necessarily minimal). The
Gauss map is a continuous choice of a unit normal n : Σ → S2 ⊂ R3 . Observe that
there are two choices n and −n corresponding to a choice of orientation of Σ. If Σ
is minimal, then the Gauss map is an (anti) conformal map since the eigenvalues
of the Weingarten map are κ1 and κ2 = −κ1 . Moreover, for a minimal surface

(1.8) |A|2 = κ21 + κ22 = −2 κ1 κ2 = −2 KΣ ,

where KΣ is the Gauss curvature. It follows that the area of the image of the Gauss
map is a multiple of the total curvature.
4 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

1.2. Minimal graphs. Suppose that u : Ω ⊂ R2 → R is a C 2 function. The

graph of u
(1.9) Graphu = {(x, y, u(x, y)) | (x, y) ∈ Ω} .
has area
Z
(1.10) Area(Graphu ) = |(1, 0, ux) × (0, 1, uy )|
ZΩ q Z p
= 2 2
1 + ux + uy = 1 + |∇u|2 ,
Ω Ω

and the (upward pointing) unit normal is

(1, 0, ux) × (0, 1, uy ) (−ux , −uy , 1)
(1.11) n= = p .
|(1, 0, ux) × (0, 1, uy )| 1 + |∇u|2
Therefore for the graphs Graphu+tη where η|∂Ω = 0 we get that
Z p
(1.12) Area(Graphu+tη ) = 1 + |∇u + t ∇η|2 ,
Ω
hence
d h∇u , ∇ηi
Z
(1.13) Area(Graphu+tη ) = p
dt t=0 Ω 1 + |∇u|2
!
∇u
Z
= − η div p .
Ω 1 + |∇u|2
It follows that the graph of u is a critical point for the area functional if and only
if u satisfies the divergence form equation
!
∇u
(1.14) div p = 0.
1 + |∇u|2
We will refer to (1.14) as the minimal surface equation.
Next we want to show that the graph of a function on Ω satisfying the minimal
surface equation is not just a critical point for the area functional but is actually
area-minimizing amongst surfaces in the cylinder Ω × R ⊂ R3 . To show this, first
extend first the unit normal n of the graph in (1.11) to a vector field, still denoted
by n, on the entire cylinder Ω × R by setting
(1.15) n(x, y, z) = n(x, y, u(x, y)) .
Let ω be the two-form on Ω × R given by the condition that for X, Y ∈ R3
(1.16) ω(X, Y ) = det(X, Y, n) .
An easy calculation shows that
! !
∂ −ux ∂ −uy
(1.17) dω = p + p = 0,
∂x 1 + |∇u|2 ∂y 1 + |∇u|2
since u satisfies the minimal surface equation. In sum, the form ω is closed and,
given any X and Y at a point (x, y, z),
(1.18) |ω(X, Y )| ≤ |X × Y | ,
 MINIMAL SUBMANIFOLDS 5

where equality holds if and only if

(1.19) X, Y ⊂ T(x,y,u(x,y)) Graphu .
Such a form ω is called a calibration. From this, we have that if Σ ⊂ Ω × R is any
other surface with ∂Σ = ∂ Graphu , then by Stokes’ theorem since ω is closed,
Z Z
(1.20) Area(Graphu ) = ω= ω ≤ Area(Σ) .
Graphu Σ

This shows that Graphu is area-minimizing among all surfaces in the cylinder and
with the same boundary.
If the domain Ω is convex, the minimal graph is absolutely area-minimizing.
To see this, observe first that if Ω is convex, then so is Ω × R, and hence the
nearest point projection P : R3 → Ω × R is a distance nonincreasing Lipschitz
map that is equal to the identity on Ω × R. If Σ ⊂ R3 is any other surface with
∂Σ = ∂ Graphu , then Σ′ = P (Σ) has Area(Σ′ ) ≤ Area(Σ). Applying (1.20) to Σ′ ,
we see that Area(Graphu ) ≤ Area(Σ′ ) and the claim follows.
If Ω ⊂ R2 contains a ball of radius r, then, since ∂Br ∩ Graphu divides ∂Br
into two components at least one of which has area at most (Area(S2 )/2) r2 , we get
from (1.20) the crude estimate
Area(S2 ) 2
(1.21) Area(Br ∩ Graphu ) ≤ r .
2
Very similar calculations to the ones above show that if Ω ⊂ Rn−1 and u : Ω → R
is a C 2 function, then the graph of u is a critical point for the area functional if
and only if u satisfies (1.14). Moreover, as in (1.20), the graph of u is actually
area-minimizing. Consequently, as in (1.21), if Ω contains a ball of radius r, then
Vol(Sn−1 ) n−1
(1.22) Vol(Br ∩ Graphu ) ≤ r .
2
1.3. The maximum principle. The first variation formula, (1.2), showed that a
smooth submanifold is a critical point for area if and only if the mean curvature
vanishes. We will next derive the weak form of the first variation formula which
is the basic tool for working with “weak solutions” (typically, stationary varifolds).
Let X be a vector field on Rn . We can write the divergence div Σ X of X on Σ as
(1.23) div Σ X = div Σ X T + div Σ X N = div Σ X T + hX, Hi ,
where X T and X N are the tangential and normal projections of X. In particular,
we get that, for a minimal submanifold,
(1.24) div Σ X = div Σ XT .
Moreover, from (1.23) and Stokes’ theorem, we see that Σ is minimal if and only if
for all vector fields X with compact support and vanishing on the boundary of Σ,
Z
(1.25) div Σ X = 0 .
Σ
The key point is that (1.25) makes sense as long as we can define the divergence of
a vector field on Σ.
The two most common notions of weakly minimal submanifolds are minimal
currents and stationary varifolds. A k-varifold Σ on Rn is a Radon measure on
Rn × G(k, n), where G(k, n) is the space of (unoriented) k-planes through the
6 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

origin in Rn (so that G(n − 1, n) is projective (n − 1) space). If X is a C 1 vector

field, then we define the divergence with respect to (x, ω) ∈ Rn × G(k, n) by
k
X
(1.26) div ω X = hEi , ∇Ei Xi ,
i=1
where Ei is an orthonormal basis for ω and ∇ is the Euclidean derivative. In
particular, the equation (1.25) makes sense when Σ is a varifold and we will say
that a varifold satisfying (1.25) is stationary.
As a consequence of (1.25), we have the following proposition:
Proposition 1.1. Σk ⊂ Rn is minimal if and only if the restrictions of the coor-
dinate functions of Rn to Σ are harmonic functions.
Proof. Let η be a smooth function on Σ with compact support and η|∂Σ = 0, then
Z Z Z
(1.27) h∇Σ η, ∇Σ xi i = h∇Σ η , ei i = div Σ (η ei ) .
Σ Σ Σ
From this, the claim follows easily. 
Recall that if Ξ ⊂ Rn is a compact subset, then the smallest convex set contain-
ing Ξ (the convex hull, Conv(Ξ)) is the intersection of all half–spaces containing Ξ.
The maximum principle forces a compact minimal submanifold to lie in the convex
hull of its boundary (this is the “convex hull property”):
Proposition 1.2. If Σk ⊂ Rn is a compact minimal submanifold, then Σ ⊂
Conv(∂Σ).
Proof. A half–space H ⊂ Rn can be written as
(1.28) H = {x ∈ Rn | hx, ei ≤ a} ,
for a vector e ∈ Sn−1 and constant a ∈ R. By Proposition 1.1, the function
u(x) = he, xi is harmonic on Σ and hence attains its maximum on ∂Σ by the
maximum principle. 
Another application of (1.24), with a different choice of vector field X, gives that
for a k-dimensional minimal submanifold Σ
(1.29) ∆Σ |x − x0 |2 = 2 div Σ (x − x0 ) = 2k .
Later we will see that this formula plays a crucial role in the monotonicity formula
for minimal submanifolds.
The argument in the proof of the convex hull property can be rephrased as
saying that as we translate a hyperplane towards a minimal surface, the first point
of contact must be on the boundary. When Σ is a hypersurface, this is a special
case of the strong maximum principle for minimal surfaces (see [14] for a proof):
Lemma 1.3. Let Ω ⊂ Rn−1 be an open connected neighborhood of the origin. If
u1 , u2 : Ω → R are solutions of the minimal surface equation with u1 ≤ u2 and
u1 (0) = u2 (0), then u1 ≡ u2 .
Since any smooth hypersurface is locally a graph over a hyperplane, Lemma
1.3 gives a maximum principle for smooth minimal hypersurfaces. There have
been several interesting extensions of the maximum principle to the singular case.
 MINIMAL SUBMANIFOLDS 7

For example, L. Simon, [97], proved that the strong maximum holds when both
hypersurfaces are area-minimizing even in the presence of singularities. B. Solomon
and B. White, [100], showed that it holds when at least one of the hypersurfaces
is smooth (the other may be just a stationary varifold). Finally, T. Ilmanen, [48],
showed that it holds for stationary hypersurfaces as long as each singular set has
zero codimension two measure.
Thus far, the examples of minimal submanifolds have all been smooth. The
simplest non-smooth example is given by a pair of planes intersecting transversely
along a line. To get an example that is not even immersed, one can take three
half–planes meeting along a line with an angle of 2π/3 between each adjacent pair.

2. Monotonicity and the mean value inequality

Monotonicity formulas and mean value inequalities play a fundamental role in
many areas of geometric analysis. Before we state and prove the monotonicity
formula of volume for minimal submanifolds, we will need to recall the coarea
formula. This formula asserts (see, for instance, [34] for a proof) that if Σ is a
manifold and h : Σ → R is a Lipschitz function on Σ, then for all locally integrable
functions f on Σ and t ∈ R
Z Z t Z
(2.1) f |∇h| = f dτ .
{h≤t} −∞ h=τ

When we apply this formula below, h will be Euclidean distance to a fixed point
x0 .
Proposition 2.1. Suppose that Σk ⊂ Rn is a minimal submanifold and x0 ∈ Rn ;
then for all 0 < s < t
|(x − x0 )N |2
Z
(2.2) t−k Vol(Bt (x0 )∩Σ)−s−k Vol(Bs (x0 )∩Σ) = k+2
.
(Bt (x0 )\Bs (x0 ))∩Σ |x − x0 |

Proof. Within this proof, we set Bt = Bt (x0 ). Since Σ is minimal,

(2.3) ∆Σ |x − x0 |2 = 2 div Σ (x − x0 ) = 2k .
By Stokes’ theorem integrating this gives
Z Z
(2.4) 2 k Vol(Bs ∩ Σ) = ∆Σ |x − x0 |2 = 2 |(x − x0 )T | .
Bs ∩Σ ∂Bs ∩Σ

Using this and the coarea formula (i.e., (2.1)), an easy calculation gives
d −k |x − x0 |
Z
−k−1 −k


s Vol(Bs ∩ Σ) = −k s Vol(Bs ∩ Σ) + s
ds ∂Bs ∩Σ |(x − x0 )T |
2
 
|x − x0 |
Z
−k−1 T
(2.5) =s − |(x − x0 ) |
∂Bs ∩Σ |(x − x0 )T |
|(x − x0 )N |2
Z
= s−k−1 T
.
∂Bs ∩Σ |(x − x0 ) |

Integrating and applying the coarea formula once more gives the claim. 
Notice that (x − x0 )N vanishes precisely when Σ is conical about x0 , i.e., when
Σ is invariant under dilations about x0 . As a corollary, we get the following:
8 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

Corollary 2.2. Suppose that Σk ⊂ Rn is a minimal submanifold and x0 ∈ Rn ;

then the function
Vol(Bs (x0 ) ∩ Σ)
(2.6) Θx0 (s) =
Vol(Bs ⊂ Rk )
is a nondecreasing function of s. Moreover, Θx0 (s) is constant in s if and only if
Σ is conical about x0 .
Of course, if x0 is a smooth point of Σ, then lims→0 Θx0 (s) = 1. We will later
see that the converse is also true; this will be a consequence of the Allard regularity
theorem.
The monotonicity of area is a very useful tool in the regularity theory for minimal
surfaces — at least when there is some a priori area bound. For instance, this
monotonicity and a compactness argument allow one to reduce many regularity
questions to questions about minimal cones (this was a key observation of W.
Fleming in his work on the Bernstein problem; see Section 4). Similar monotonicity
formulas have played key roles in other geometric problems, including harmonic
maps, Yang–Mills connections, J–holomorphic curves, and regularity of limit spaces
with a lower Ricci curvature bound.
Arguing as in Proposition 2.1, we get a weighted monotonicity:
Proposition 2.3. If Σk ⊂ Rn is a minimal submanifold, x0 ∈ Rn , and f is a
twice differentiable function on Σ, then
Z Z
(2.7) t−k f − s−k f
Bt (x0 )∩Σ Bs (x0 )∩Σ

t
|(x − x0 )N |2 1
Z Z Z
= f + τ −k−1 (τ 2 − |x − x0 |2 ) ∆Σ f dτ .
(Bt (x0 )\Bs (x0 ))∩Σ |x − x0 |k+2 2 s Bτ (x0 )∩Σ

We get immediately the following mean value inequality for the special case of
non–negative subharmonic functions:
Corollary 2.4. Suppose that Σk ⊂ Rn is a minimal submanifold, x0 ∈ Rn , and f
is a non–negative subharmonic function on Σ; then
Z
(2.8) s−k f
Bs (x0 )∩Σ

is a nondecreasing function of s. In particular, if x0 ∈ Σ, then for all s > 0

R
Bs (x0 )∩Σ f
(2.9) f (x0 ) ≤ .
Vol (Bs ⊂ Rk )

3. Rado’s theorem
One of the most basic questions is what does the boundary ∂Σ tell us about
a compact minimal submanifold Σ? We have already seen that Σ must lie in the
convex hull of ∂Σ, but there are many other theorems of this nature. One of the
first is a beautiful result of T. Rado which says that if ∂Σ is a graph over the
boundary of a convex set in R2 , then Σ is also graph (and hence embedded). The
proof of this uses basic properties of nodal lines for harmonic functions.
 MINIMAL SUBMANIFOLDS 9

Theorem 3.1. Suppose that Ω ⊂ R2 is a convex subset and σ ⊂ R3 is a simple

closed curve which is graphical over ∂Ω. Then any minimal disk Σ ⊂ R3 with
∂Σ = σ must be graphical over Ω and hence unique by the maximum principle.
Proof. (Sketch.) The proof is by contradiction, so suppose that Σ is such a minimal
disk and x ∈ Σ is a point where the tangent plane to Σ is vertical. Consequently,
there exists (a, b) 6= (0, 0) such that
(3.1) ∇Σ (a x1 + b x2 )(x) = 0 .
By Proposition 1.1, a x1 + b x2 is harmonic on Σ (since it is a linear combination of
coordinate functions). The local structure of nodal sets of harmonic functions (see,
e.g., [14]) then gives that the level set
(3.2) {y ∈ Σ | (a x1 + b x2 )(y) = (a x1 + b x2 )(x)}
has a singularity at x where at least four different curves meet. If two of these
nodal curves were to meet again, then there would be a closed nodal curve which
must bound a disk (since Σ is a disk). By the maximum principle, a x1 + b x2 would
have to be constant on this disk and hence constant on Σ by unique continuation.
This would imply that σ = ∂Σ is contained in the plane given by (3.2). Since
this is impossible, we conclude that all of these curves go to the boundary without
intersecting again.
In other words, the plane in R3 given by (3.2) intersects σ in at least four points.
However, since Ω ⊂ R2 is convex, ∂Ω intersects the line given by (3.2) in exactly
two points. Finally, since σ is graphical over ∂Ω, σ intersects the plane in R3 given
by (3.2) in exactly two points, which gives the desired contradiction. 

4. The theorems of Bernstein and Bers

A classical theorem of S. Bernstein from 1916 says that entire (i.e., defined over
all of R2 ) minimal graphs are planes. This remarkable theorem of Bernstein was
one of the first illustrations of the fact that the solutions to a nonlinear PDE, like
the minimal surface equation, can behave quite differently from solutions to a linear
equation.
Theorem 4.1. [5] If u : R2 → R is an entire solution to the minimal surface
equation, then u is an affine function.
Proof. (Sketch.) We will show that the curvature of the graph vanishes identically;
this implies that the unit normal is constant and, hence, the graph must be a plane.
The proof follows by combining two facts. First, the area estimate for graphs (1.21)
gives
(4.1) Area(Br ∩ Graphu ) ≤ 2 π r2 .
This quadratic area growth allows one to construct a sequence of non-negative
logarithmic cutoff functions φj defined on the graph with φj → 1 everywhere and
Z
(4.2) lim |∇φj |2 = 0 .
j→∞ Graphu
Moreover, since graphs are area-minimizing, they must be stable. We can therefore
use φj in the stability inequality (1.7) to get
Z Z
(4.3) φ2j |A|2 ≤ |∇φj |2 .
Graphu Graphu
10 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

Combining these gives that |A|2 is zero, as desired. 

Rather surprisingly, this result very much depended on the dimension. The
combined efforts of E. De Giorgi [31], F. J. Almgren, Jr. [2], and J. Simons [98]
finally gave:
Theorem 4.2. If u : Rn−1 → R is an entire solution to the minimal surface
equation and n ≤ 8, then u is an affine function.
However, in 1969 E. Bombieri, De Giorgi, and E. Giusti [8] constructed entire
non–affine solutions to the minimal surface equation on R8 and an area–minimizing
singular cone in R8 . In fact, they showed that for m ≥ 4 the cones
(4.4) Cm = {(x1 , . . . , x2m ) | x21 + · · · + x2m = x2m+1 + · · · + x22m } ⊂ R2m
are area–minimizing (and obviously singular at the origin).
In contrast to the entire case, exterior solutions of the minimal graph equation,
i.e., solutions on R2 \ B1 , are much more plentiful. In this case, L. Bers proved
that ∇u actually has an asymptotic limit:
Theorem 4.3. [6] If u is a C 2 solution to the minimal surface equation on R2 \B1 ,
then ∇u has a limit at infinity (i.e., there is an asymptotic tangent plane).
Proof. (Sketch.) To get a rough idea of why Bers’ theorem should hold, recall
that the Gauss map n gives a holomorphic map from the graph of u to the upper
hemisphere. Hence, composing n with stereographic projection gives a bounded
holomorphic function on the graph of u.
The rest of the argument is to show that the point at infinity is a removable
singularity for this bounded holomorphic function. Since the graph of u is a topo-
logical annulus, it must be conformal to an annulus of the form {a < |z| ≤ 1} in
the plane for some a ≥ 0. The point is to show that a = 0. Arguing as in the proof
of Bernstein’s theorem (cf. (4.1)) shows that the graph of u has quadratic area
growth and, hence, we can construct logarithmic cutoff functions “at infinity” with
arbitrarily small energy. It follows that we must have a = 0 and hence n extends
smoothly across the puncture. This gives a limiting value for n and, consequently,
also a limiting value for ∇u. 
Bers’ theorem was extended to higher dimensions by L. Simon:
Theorem 4.4. [94] If u is a C 2 solution to the minimal surface equation on Rn \B1 ,
then either
• |∇u| is bounded and ∇u has a limit at infinity.
• All tangent cones at infinity are of the form Σ × R where Σ is singular.
Bernstein’s theorem has had many other interesting generalizations, some of
which will be discussed later.

5. Simons inequality
In this section, we recall a very useful differential inequality for the Laplacian of
the norm squared of the second fundamental form A of a minimal hypersurface Σ
in Rn and illustrate its role in a priori estimates. This inequality, originally due to
J. Simons (see [14] for a proof and further discussion), is:
 MINIMAL SUBMANIFOLDS 11

Lemma 5.1. [98] If Σn−1 ⊂ Rn is a minimal hypersurface, then

(5.1) ∆Σ |A|2 = −2 |A|4 + 2|∇Σ A|2 ≥ −2 |A|4 .
An inequality of the type (5.1) on its own does not lead to pointwise bounds
on |A|2 because of the nonlinearity. However, it does lead to estimates if a “scale–
invariant energy” is small. For example, H. Choi and R. Schoen used (5.1) to
prove:
Theorem 5.2. [12] There exists ǫ > 0 so that if 0 ∈ Σ ⊂ Br (0) with ∂Σ ⊂ ∂Br (0)
is a minimal surface with
Z
(5.2) |A|2 ≤ ǫ ,

then
(5.3) |A|2 (0) ≤ r−2 .

6. Heinz’s curvature estimate for graphs

One of the key themes in minimal surface theory is the usefulness of a priori
estimates. A basic example is the curvature estimate of E. Heinz for minimal
graphs over a disk Dr0 of radius r0 in the plane.
Theorem 6.1. [41] If Dr0 ⊂ R2 and u : Dr0 → R satisfies the minimal surface
equation, then for Σ = Graphu and 0 < σ ≤ r0
(6.1) σ 2 sup |A|2 ≤ C .
Dr0 −σ

Proof. (Sketch.) Observe first that it suffices to prove the estimate for σ = r0 , i.e.,
to show that
(6.2) |A|2 (0, u(0)) ≤ C r0−2 .
Recall that minimal graphs are automatically stable. As in the proof of Theorem
4.1, the area estimate for graphs (1.21) allows us to use a logarithmic cutoff function
in the the stability inequality (1.7) to get that
C
Z
(6.3) |A|2 ≤ .
Br1 ∩Graphu log(r0 /r1 )
Taking r0 /r1 sufficiently large, we can then apply Theorem 5.2 to get (6.2). 
Heinz’s estimate gives an effective version of the Bernstein’s theorem; namely,
letting the radius r0 go to infinity in (6.1) implies that |A| vanishes, thus giving
Bernstein’s theorem.

7. Embedded minimal disks with area bounds

In the early nineteen–eighties R. Schoen and L. Simon extended the theorem
of Bernstein to complete simply connected embedded minimal surfaces in R3 with
quadratic area growth; see [90]. A surface Σ is said to have quadratic area growth if
for all r > 0, the area of the intersection of the surface with the ball in R3 of radius
r and center at the origin is bounded by C r2 for a fixed constant C independent
of r.
12 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

Theorem 7.1. [90] Let 0 ∈ Σ2 ⊂ Br0 = Br0 (x) ⊂ R3 be an embedded simply

connected minimal surface with ∂Σ ⊂ ∂Br0 . If µ > 0 and either
(7.1) Area(Σ) ≤ µ r02 , or
Z
(7.2) |A|2 ≤ µ ,
Σ
then for the connected component Σ′ of Br0 /2 (x0 ) ∩ Σ with 0 ∈ Σ′ we have
(7.3) sup |A|2 ≤ C r0−2
Σ′
for some C = C(µ).
In corollary 1.18 in [17], this was generalized to quadratic area growth for in-
trinsic balls (this generalization played an important role in analyzing the local
structure of embedded minimal surfaces). We will use Br (p) to denote the intrinsic
ball of radius r centered at a point p in a surface Σ, i.e., Br (p) is the set of points
q in Σ that can be connected to p by a path γ in Σ of length less than r.
Theorem 7.2. [17] Given a constant CI , there exists CP so that if B2r0 ⊂ Σ ⊂ R3
is an embedded minimal disk satisfying either
(7.4) Area(B2r0 ) ≤ CI r02 or
Z
(7.5) |A|2 ≤ CI ,
B2r0

then
(7.6) sup |A|2 ≤ CP s−2 .
Bs

As an immediate consequence, letting r0 → ∞ gives Bernstein-type theorems for

embedded simply connected minimal surfaces with either bounded density or finite
total curvature.
The classical minimal surfaces known as Enneper’s surface and the catenoid
show that neither “embedded” nor “simply-connected” can be removed. Enneper’s
surface is a complete immersed minimal disk but is not flat and is not embed-
ded (see (17.3) for the definition of Enneper’s surface). This example shows that
embeddedness is essential for these estimates. Similarly, the catenoid shows that
simply-connected is essential. The catenoid is the minimal surface in R3 given by
(7.7) {(cosh s cos t, cosh s sin t, s) | s, t ∈ R} ;
see figure 5.

8. Stable minimal surfaces

It turns out that stable minimal surfaces have a priori estimates. Since minimal
graphs are stable, the estimates for stable surfaces can be thought of as gener-
alizations of the earlier estimates for graphs. These estimates have been widely
applied and are particularly useful when combined with existence results for stable
surfaces (such as the solution of the Plateau problem). The starting point for these
estimates is that, as we saw in (1.4), stable minimal surfaces satisfy the stability
inequality
Z Z
(8.1) |A| φ ≤ |∇φ|2 .
2 2
 MINIMAL SUBMANIFOLDS 13

We will mention two such estimates. The first is R. Schoen’s curvature estimate
for stable surfaces:
Theorem 8.1. [88] There exists a constant C so that if Σ ⊂ R3 is an immersed
stable minimal surface with trivial normal bundle and Br0 ⊂ Σ \ ∂Σ, then
(8.2) sup |A|2 ≤ C σ −2 .
Br0 −σ

The second is an estimate for the area and total curvature of a stable surface;
for simplicity, we will state only the area estimate:
Theorem 8.2. [15] If Σ ⊂ R3 is an immersed stable minimal surface with trivial
normal bundle and Br0 ⊂ Σ \ ∂Σ, then
(8.3) Area (Br0 ) ≤ 4π r02 /3 .
As mentioned, we can use (8.3) to bound the energy of a cutoff function in the
stability inequality and, thus, bound the total curvature of sub-balls. Combining
this with the curvature estimate of Theorem 5.2 gives Theorem 8.1; see [15]. Note
that the bound (8.3) is surprisingly sharp; even when Σ is a plane, the area is πr02 .
We will explain next why Theorem 8.2 holds. The stability inequality can be
used to get upper bounds for the total curvature in terms of the area of a minimal
surface. On the other hand, we can use either the Gauss–Bonnet theorem or the
Jacobi equation to get the opposite bound. Combining these two bounds will give
the a priori bound on the area of intrinsic balls in a stable surface. More precisely,
integrating the Jacobi equation and using the Gauss equation KΣ = −|A|2 /2 gives
Z R Z tZ Z
2 2
(8.4) 4 (Area (BR ) − π R ) = 2 |A| = |A|2 (R − r)2 .
0 0 Bs BR
RR
The second equality uses two integrations by parts (i.e., 0 f (t) g ′′ (t) dt with f (t) =
RtR 2 2
0 Bs |A| and g(t) = (R − t) ). See corollary 1.7 of [17] for further details.
Using φ = R − r (where r(x) = distΣ (0, x)) in the stability inequality gives
Z Z
2 2 2
(8.5) 4 (Area (BR ) − π R ) = |A| (R − r) ≤ |∇(R − r)|2 = Area (BR ) .
BR BR
2
Consequently, Area (BR ) ≤ 4π R /3.

9. Regularity theory
In this section, we survey some of the key ideas in classical regularity theory,
such as the role of monotonicity, scaling, ǫ-regularity theorems (such as W. Allard’s
theorem) and tangent cone analysis (such as F. Almgren’s refinement of H. Federer’s
dimension reducing).
The starting point for all of this is the monotonicity of volume for a minimal
k–dimensional submanifold Σ. Namely, Corollary 2.2 gives that the density
Vol(Bs (x0 ) ∩ Σ)
(9.1) Θx0 (s) =
Vol(Bs ⊂ Rk )
is a monotone non-decreasing function of s. Consequently, we can define the density
Θx0 at the point x0 to be the limit as s → 0 of Θx0 (s). It also follows easily from
monotonicity that the density is semi–continuous as a function of x0 .
14 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

9.1. ǫ–regularity and the singular set. An ǫ–regularity theorem is a theorem

giving that a weak (or generalized) solution is actually smooth at a point if a
scale–invariant energy is small enough there. The standard example is the Allard
regularity theorem:
Theorem 9.1. [1] There exists δ = δ(k, n) > 0 such that if Σ ⊂ Rn is a k–
rectifiable stationary varifold (with density at least one a.e.), x0 ∈ Σ, and
Vol (Br (x0 ) ∩ Σ)
(9.2) Θx0 = lim < 1+δ,
r→0 Vol (Br ⊂ Rk )

then Σ is smooth in a neighborhood of x0 .

Similarly, the small total curvature estimate of Theorem 5.2 may be R thought of
as an ǫ-regularity theorem; in this case, the scale–invariant energy is |A|2 .
As an application of the ǫ–regularity theorem, Theorem 9.1, we can define the
singular set S of Σ by
(9.3) S = {x ∈ Σ | Θx ≥ 1 + δ} .
It follows immediately from the semi–continuity of the density that S is closed.
In order to bound the size of the singular set (e.g., the Hausdorff measure), one
combines the ǫ–regularity with simple covering arguments.
This preliminary analysis of the singular set can be refined by doing a so–called
tangent cone analysis.

9.2. Tangent cone analysis. It is not hard to see that scaling preserves the space
of minimal submanifolds of Rn . Namely, if Σ is minimal, then so is
(9.4) Σy,λ = {y + λ−1 (x − y) | x ∈ Σ} .
(To see this, simply note that this scaling multiplies the principal curvatures by
λ.) Suppose now that we fix the point y and take a sequence λj → 0. The mono-
tonicity formula bounds the density of the rescaled solution, allowing us to extract
a convergent subsequence and limit. This limit, which is called a tangent cone at
y, achieves equality in the monotonicity formula and, hence, must be homogeneous
(i.e., invariant under dilations about y).
The usefulness of tangent cone analysis in regularity theory is based on two
key facts. For simplicity, we illustrate these when Σ ⊂ Rn is an area minimizing
hypersurface. First, if any tangent cone at y is a hyperplane Rn−1 , then Σ is
smooth in a neighborhood of y. This follows easily from the Allard regularity
theorem since the density at y of the tangent cone is the same as the density at y
of Σ. The second key fact, known as “dimension reducing,” is due to F. Almgren,
[3], and is a refinement of an argument of H. Federer. To state this, we first stratify
the singular set S of Σ into subsets
(9.5) S0 ⊂ S1 ⊂ · · · ⊂ Sn−2 ,
where we define Si to be the set of points y ∈ S so that any linear space contained
in any tangent cone at y has dimension at most i. (Note that Sn−1 = ∅ by Allard’s
Theorem.) The dimension reducing argument then gives that
(9.6) dim (Si ) ≤ i ,
 MINIMAL SUBMANIFOLDS 15

where dimension means the Hausdorff dimension. In particular, the solution of the
Bernstein problem then gives codimension 7 regularity of Σ, i.e., dim (S) ≤ n − 8.
See lecture 2 in [96] for a proof of (9.6).
Tangent cones produced by scalings as above may very well depend on the par-
ticular convergent subsequence. In some cases, one can prove uniqueness of the
tangent cone and this is often quite useful (see, for instance, section 3.4 in [96] for
one such application).

Part 2. Embedded minimal surfaces

Thus far, the results for embedded minimal surfaces have assumed some addi-
tional a priori bound, such as bounds on area or total curvature, and the proofs
break down without these a priori bounds. In this part, we will focus on recent
results for embedded minimal surfaces without a priori bounds.

10.1. Multi–valued graphs. We have earlier studied minimal graphs. It is useful

also to consider multi-valued minimal graphs. Intuitively, an (embedded) multi–
valued graph is a surface such that over each point of the annulus, the surface
consists of N graphs. To make this notion precise, let P be the universal cover
of the punctured plane C \ {0} with global polar coordinates (ρ, θ) so ρ > 0 and
θ ∈ R. An N –valued graph on C \ {0} is a single valued graph of a function u over
P. For working purposes, we generally think of the intuitive picture of a multi–
sheeted surface in R3 , and we identify the single–valued graph over the universal
cover with its multi–valued image in R3 .
We will also need the notion of an N -valued graph over the annulus {r < ρ ≤
s} ⊂ C. In this case, the function is single-valued on
(10.1) {(ρ, θ) | r < ρ ≤ s , |θ| ≤ N π} .
The multi–valued graphs that we will consider will all be embedded, which cor-
responds to a nonvanishing separation between the sheets (or the floors). Here the
separation is the function (see figure 2)
(10.2) w(ρ, θ) = u(ρ, θ + 2π) − u(ρ, θ) .
The helicoid is a minimal surface consisting of two multi-valued graphs glued
together along an axis. The helicoid looks like a “double spiral staircase” (see [23])
and is parametrized by:
Example 2: (Helicoid; see figure 1). The helicoid is the minimal surface in R3
given by the parametrization
(10.3) (s cos t, s sin t, t) , where s, t ∈ R .
If Σ is the helicoid, then Σ \ {x3 − axis} = Σ1 ∪ Σ2 , where Σ1 , Σ2 are ∞–valued
graphs on C \ {0}. Σ1 is the graph of the function u1 (ρ, θ) = θ and Σ2 is the graph
of the function u2 (ρ, θ) = θ + π. (Σ1 is the subset where s > 0 in (10.3) and Σ2
the subset where s < 0.) In either case the separation w = 2 π. A multi–valued
minimal graph is a multi–valued graph of a function u satisfying the minimal surface
equation.
Note that for an embedded multi–valued graph, the sign of w determines whether
the multi–valued graph spirals in a left–handed or right–handed manner, in other
16 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

x3 -axis
x3-axis

One half rotation

u(ρ, θ + 2π)

w
u(ρ, θ)

Figure 1. Multi–valued Figure 2. The separation w

graphs. The helicoid is ob- grows/decays in ρ at most sub-
tained by gluing together two linearly for a multi–valued min-
∞–valued graphs along a line. imal graph; see (10.5).
First published in the Notices
of the American Mathematical
Society in 2003, published by
the American Mathematical
Society.

words, whether upwards motion corresponds to turning in a clockwise direction or

in a counterclockwise direction.

10.2. The sublinear growth of the separation. As we have seen, the separation
is constant for the multi–valued graphs coming from each half of the helicoid. This
can be viewed as a type of Liouville Theorem reflecting the conformal properties of
an infinite–valued graph. In Proposition II.2.12 of [16], we proved a corresponding
gradient estimate (i.e., a pointwise estimate for |∇ log |w||) for a general multi-
valued graph. Integrating this gradient estimate gives that the separation grows
sublinearly (see figure 2):
Proposition 10.1. [16] Given α > 0, there exists N so that if u satisfies the
minimal surface equation on
(10.4) {e−N r1 ≤ ρ ≤ eN r2 , −N ≤ θ ≤ 2π + N } ,
|∇u| ≤ 1, and has separation w 6= 0, then
 α
r2
(10.5) |w|(r2 , 0) ≤ |w|(r1 , 0) .
r1
Proof. (Sketch.) As mentioned above, the inequality (10.5) follows from integrating
the gradient estimate
α
(10.6) |∇ log |w||(r, 0) ≤
r
in the same way that the Harnack inequality for positive harmonic functions follows
from integrating the gradient estimate.
 MINIMAL SUBMANIFOLDS 17

To see why a gradient estimate like (10.6) holds, observe that u(·, ·) and its
2π–rotation u(·, · + 2π) are both solutions of the minimal surface equation and,
thus, the difference w is (almost) a positive solution of the linearized equation. The
linearized equation is itself a perturbation of the Laplace equation and it is not
difficult to get an estimate
C
(10.7) |∇ log |w||(r, 0) ≤ ,
r
for some constant C.
The point now is to show that if N is large, then we can make the constant C in
(10.7) small. For simplicity, suppose that w is actually a positive harmonic func-
tion on the region (10.4). Since harmonic functions are invariant under conformal
transformations, w ◦ ez is a positive harmonic function on the “rectangle”
(10.8) {(x + iy) | − N + log r1 < x < N + log r2 and |y| < N } .
In the extreme case when N = ∞, the positive harmonic function w ◦ ez is defined
on the entire plane and, hence, is constant by the Liouville theorem. It is not hard
to see that applying the gradient estimate on this rectangle when N is large, and
then translating this back to the original function w, gives (10.6). 

11. Embedded minimal disks

There are two classical local models for embedded minimal disks (by an embedded
disk we mean a smooth injective map from the closed unit ball in R2 into R3 ). One
model is the plane (or, more generally, a minimal graph) and the other is a piece
of a helicoid.
The helicoid was discovered by Meusnier in 1776. Meusnier had been a student
of Monge. He also discovered that the surface now known as the catenoid is minimal
in the sense of Lagrange, and he was the first to characterize a minimal surface as
a surface with vanishing mean curvature. Unlike the helicoid, the catenoid is not
topologically a plane but rather a cylinder.
It turns out that these two classical examples (graphs and helicoids) completely
capture the local structure of an embedded minimal disk. This is made concrete in
the compactness theorem for embedded minimal disks, Theorem 11.1 below.
To avoid tedious dependence of various quantities we state Theorem 11.1 not
for a single embedded minimal disk with sufficiently large curvature at a given
point but instead for a sequence of such disks where the curvatures are blowing up.
Theorem 11.1 says that a sequence of embedded minimal disks mimics the following
behavior of a sequence of rescaled helicoids:
Consider the sequence Σi = ai Σ of rescaled helicoids where ai → 0.
(That is, rescale R3 by ai , so points that used to be distance d apart
will in the rescaled R3 be distance ai d apart.) The curvatures of
this sequence of rescaled helicoids are blowing up along the vertical
axis. The sequence converges (away from the vertical axis) to a
foliation by flat parallel planes. The singular set S (the axis) then
consists of removable singularities.
Let now Σi ⊂ B2R be a sequence of embedded minimal disks with ∂Σi ⊂ ∂B2R .
Clearly (after possibly going to a subsequence) either (A) or (B) occurs:
(A) supBR ∩Σi |A|2 ≤ C < ∞ for some constant C.
18 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

(B) supBR ∩Σi |A|2 → ∞.

In (A) (by a standard argument; see, e.g., lemma 2.2 in [14]) the intrinsic ball Bs (yi )
is a graph for all yi ∈ BR ∩ Σi , where s depends only on C. Thus the main case is
(B) which is the subject of the next theorem.

Using the notion of multi–valued graphs, the lamination theorem (the main the-
orem of [19]), can now be stated:

One half of Σ. The other half.

S

Figure 3. Theorem 11.1 – the singular set, S, and the two multi–valued
graphs. First published in the Notices of the American Mathematical
Society in 2003, published by the American Mathematical Society.

Theorem 11.1. (Theorem 0.1 in [19]). See figure 3. Let Σi ⊂ BRi = BRi (0) ⊂ R3
be a sequence of embedded minimal disks with ∂Σi ⊂ ∂BRi where Ri → ∞. If
(11.1) sup |A|2 → ∞ ,
B1 ∩Σi

then there exists a subsequence, Σj , and a Lipschitz curve S : R → R3 such that

after a rotation of R3 :
(1) x3 (S(t)) = t. (That is, S is a graph over the x3 –axis.)
(2) Each Σj consists of exactly two multi–valued graphs away from S (which
spiral together).
(3) For each 1 > α > 0, Σj \ S converges in the C α –topology to the foliation,
F = {x3 = t}t , of R3 .
(4) supBr (S(t))∩Σj |A|2 → ∞ for all r > 0, t ∈ R. (The curvatures blow up
along S.)
In (2) and (3), the statements that Σj \ S are multi–valued graphs and converges
to F means that for each compact subset K ⊂ R3 \ S and j sufficiently large K ∩Σj
consists of multi–valued graphs over (part of) {x3 = 0} and K ∩ Σj → K ∩ F in
the sense of graphs.
A key point in Theorem 11.1 is that there is no useful monotonicity formula
or natural a priori bound. The main tools for overcoming these difficulties are a
“classification of singularities” which describes a neighborhood of points of large
curvature and our one-sided curvature estimate (Theorem 11.2 below).
The one-sided curvature estimate says roughly that if an embedded minimal disk
lies in a half–space above a plane and comes close to the plane, then it is a graph
over the plane. Precisely, this is the following theorem:
Theorem 11.2. (Theorem 0.2 in [19]). See figure 4. There exists ǫ > 0, so that if
Σ ⊂ B2r0 ∩ {x3 > 0} ⊂ R3
 MINIMAL SUBMANIFOLDS 19

is an embedded minimal disk with ∂Σ ⊂ ∂B2 r0 , then for all components Σ′ of

Br0 ∩ Σ which intersect Bǫr0 we have
(11.2) sup |AΣ |2 ≤ r0−2 .
Σ′

Bǫr0 x3 = 0

Br0
B2r0

Figure 4. Theorem 11.2 – the one–sided curvature estimate for an embedded

minimal disk Σ in a half–space with ∂Σ ⊂ ∂B2r0 : The components of Br0 ∩ Σ
intersecting Bǫr0 are graphs. First published in the Notices of the American
Mathematical Society in 2003, published by the American Mathematical Society.
Using the minimal surface equation and that Σ′ has points close to a plane,
it is not hard to see that, for ǫ > 0 sufficiently small, (11.2) is equivalent to the
statement that Σ′ is a graph over the plane {x3 = 0}.
We will often refer to Theorem 11.2 as the one–sided curvature estimate.

x3

Rescaled catenoid.
x2

x1 x3 = 0
Figure 5. The catenoid given Figure 6. Rescaling the
by revolving x1 = cosh x3 catenoid shows that simply
around the x3 –axis. First pub- connected (and embedded) is
lished in the Notices of the needed in the one–sided curva-
American Mathematical Society ture estimate. First published
in 2003, published by the Amer- in the Notices of the American
ican Mathematical Society. Mathematical Society in 2003,
published by the American
Mathematical Society.

Note that the assumption in Theorem 11.2 that Σ is simply connected is crucial
as can be seen from the example of a rescaled catenoid. Recall that the catenoid is
the minimal surface in R3 given by
(11.3) (cosh s cos t, cosh s sin t, s)
20 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

where s, t ∈ R; see figure 5. Under rescalings this converges (with multiplicity two)
to the flat plane; see figure 6. Likewise, by considering the universal cover of the
catenoid, one sees that embedded, and not just immersed, is needed in Theorem
11.2.
As an almost immediate consequence of Theorem 11.2 and a simple barrier ar-
gument we get that if in a ball two embedded minimal disks come close to each
other near the center of the ball then each of the disks are graphs. Precisely, this
is the following:

graph

graph

Figure 7. Corollary 11.3: Two sufficiently close components of an em-

bedded minimal disk must each be a graph.

Corollary 11.3. (Corollary 0.4 in [19]). See figure 7. There exist c > 1, ǫ > 0 so
that the following holds:
Let Σ1 and Σ2 ⊂ Bcr0 ⊂ R3 be disjoint embedded minimal surfaces with ∂Σi ⊂
∂Bcr0 and Bǫ r0 ∩ Σi 6= ∅. If Σ1 is a disk, then for all components Σ′1 of Br0 ∩ Σ1
which intersect Bǫ r0
(11.4) sup |A|2 ≤ r0−2 .
Σ′1

12. Fixed genus

Following our results on embedded minimal disks described in the previous sec-
tion, we proved two main structure theorems for non-simply connected embedded
minimal surfaces of any given fixed genus in [20].
The first of these asserts that any such surface without small necks can be ob-
tained by gluing together two oppositely–oriented double spiral staircases; see figure
8.
The second gives a pair of pants decomposition of any such surface when there
are small necks, cutting the surface along a collection of short curves; see figure
9. After the cutting, we are left with graphical pieces that are defined over a disk
with either one or two sub–disks removed (a topological disk with two sub–disks
removed is called a pair of pants).
Both of these structures occur as different extremes in the two-parameter family
of minimal surfaces known as the Riemann examples.
12.1. Uniformly locally simply connected. Sequences of surfaces which are
not simply connected are, after passing to a subsequence, naturally divided into
two separate cases depending on whether or not the topology is concentrating at
points. To distinguish between these cases, we will say that a sequence of surfaces
 MINIMAL SUBMANIFOLDS 21

The two main structure theorems for non-simply connected surfaces:

One of the “pair of pants” (in bold).

The curves that we cut along.

Figure 8. Absence of necks: Figure 9. Presence of necks:
The surface can be obtained by The surface can be decomposed
gluing together two oppositely– into a collection of pair of pants
oriented double spiral staircases. by cutting along short curves.

Σ2i ⊂ R3 is uniformly locally simply connected (or ULSC) if for each compact subset
K of R3 , there exists a constant r0 > 0 (depending on K) so that for every x ∈ K,
all r ≤ r0 , and every surface Σi
(12.1) each connected component of Br (x) ∩ Σi is a disk.
For instance, a sequence of rescaled catenoids where the necks shrink to zero is not
ULSC, whereas a sequence of rescaled helicoids is.
Remark 12.1. If each component of the intersection of a minimal surface with a
ball of radius r0 is a disk, then so are the intersections with all sub–balls by the
convex hull property (see, e.g., lemma C.1 in [19]). Therefore, it would be enough
that (12.1) holds for r = r0 .
We will next describe briefly the case of ULSC sequences. See [20] for more on
this and for the general case of fixed genus.
We will assume here that the surfaces are not disks (the case of disks was dealt
with in the previous subsection). In particular, we will assume that for each i, there
exists some yi ∈ R3 and si > 0 so that
(12.2) some component of Bsi (yi ) ∩ Σi is not a disk.
Loosely speaking, the next result shows that when the sequence is ULSC (but
not simply connected), a subsequence converges to a foliation by parallel planes
away from two lines S1 and S2 ; see figure 10. The lines S1 and S2 are disjoint and
orthogonal to the leaves of the foliation and the two lines are precisely the points
where the curvature is blowing up. This is similar to the case of disks, except that
we get two singular curves for non-disks as opposed to just one singular curve for
disks.
Theorem 12.2. [20] Let Σi ⊂ BRi = BRi (0) ⊂ R3 be a sequence of compact
embedded minimal surfaces with fixed genus and with ∂Σi ⊂ ∂BRi where Ri → ∞.
Suppose that each Σi is ULSC and satisfies (12.2) with si = R > 1 and yi = 0. If
(12.3) sup |A|2 → ∞ ,
B1 ∩Σi

then there exists a subsequence Σj , two disjoint parallel lines S1 and S2 , and a
rotation of R3 so that:
22 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

Limit foliation by planes.

Singular lines S1 and S2 .

Figure 10. Theorem 12.2: Limits of sequences of non-simply connected,
yet ULSC, surfaces with curvature blowing up. The singular set consists
of two lines S1 and S2 and the limit is a foliation by flat parallel planes.

(A) For each 1 > α > 0, Σj \ (S1 ∪ S2 ) converges in the C α -topology to the
foliation {x3 = t} by parallel planes.
(B) supBr (x)∩Σj |A|2 → ∞ as j → ∞ for all r > 0 and x ∈ S1 ∪ S2 . (The
curvatures blow up along S1 and S2 .)
(Culsc ) Away from S1 ∪ S2 , each Σj consists of exactly two multi-valued graphs
spiraling together. Near S1 and S2 , the pair of multi-valued graphs form
double spiral staircases with opposite orientations at S1 and S2 . Thus,
circling only S1 or only S2 results in going either up or down, while a path
circling both S1 and S2 closes up (see figure 11).
(Dulsc ) S1 and S2 are orthogonal to the leaves of the foliation.

We should point out that the multi-valued graphs in (Culsc ) are defined over
the doubly-punctured plane; see figure 11. The multi-valued graphs considered
previously were defined over a plane punctured at just one point.

Locally graphical except over two points;

those points correspond to the two axes.

The spiral staircases around each of the axes connect to

each other between the axes.
Figure 11. A multi-valued graph over the doubly-punctured plane. The
spiral staircases near each puncture are oppositely–oriented.
 MINIMAL SUBMANIFOLDS 23

Despite the similarity of Theorem 12.2 to the case of disks, it is worth noting
that the results for disks do not alone give this theorem. Namely, even though the
ULSC sequence consists locally of disks, the compactness result for disks was in the
global case where the radii go to infinity. One might wrongly think that Theorem
12.2 could be proven using the results for disks and a blow up argument. However,
local examples constructed in [24] show the difficulty with such an argument.

12.2. Fixed genus. In the previous section, we analyzed sequences of embedded

minimal surfaces with fixed genus where the topology does not concentrate at any
points (i.e., ULSC sequences). We will now consider general sequences where the
topology is allowed to concentrate. For simplicity, we will restrict to the case of
genus zero, i.e., to planar domains (see [20] for the general case).
One way of locally distinguishing sequences where the topology does not con-
centrate from sequences where it does comes from analyzing the singular set. The
singular set S is defined to be the set of points where the curvature is blowing up.
That is, a point y in R3 is in S for a sequence Σi if
(12.4) sup |A|2 → ∞ as i → ∞ for all r > 0.
Br (y)∩Σi

For embedded minimal surfaces, S consists of two types of points. The first type
is roughly modelled on rescaled helicoids and the second on rescaled catenoids:
• A point y in R3 is in Sulsc if the curvature for the sequence Σi blows up at
y and the sequence is ULSC in a neighborhood of y.
• A point y in R3 is in Sneck if the sequence is not ULSC in any neighborhood
of y. In this case, a sequence of closed non-contractible curves γi ⊂ Σi
converges to y.
The sets Sneck and Sulsc are obviously disjoint and the curvature blows up at both,
so Sneck ∪ Sulsc ⊂ S. An easy argument shows that, after passing to a subsequence,
we can assume that
(12.5) S = Sneck ∪ Sulsc .
Note that Sneck = ∅ is equivalent to that the sequence is ULSC as is the case
for sequences of rescaled helicoids. On the other hand, Sulsc = ∅ for sequences of
rescaled catenoids.
We will show that every sequence Σi has a subsequence that is either ULSC or
for which Sulsc is empty. This is the next “no mixing” theorem. These two different
cases give two very different structures.
Theorem 12.3. [20] If Σi ⊂ BRi = BRi (0) ⊂ R3 is a sequence of compact em-
bedded minimal planar domains with ∂Σi ⊂ ∂BRi where Ri → ∞, then there is a
subsequence with either Sulsc = ∅ or Sneck = ∅.
The case where Sneck = ∅ was dealt with in the previous section.
Common for both the ULSC case and the case where Sulsc is empty is that the
limits are always laminations by flat parallel planes and the singular sets are always
closed subsets contained in the union of the planes. This is the content of the next
theorem:
Theorem 12.4. [20] Let Σi ⊂ BRi = BRi (0) ⊂ R3 be a sequence of compact
embedded minimal surfaces with fixed genus and with ∂Σi ⊂ ∂BRi where Ri → ∞.
24 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

If
(12.6) sup |A|2 → ∞ ,
B1 ∩Σi

then there exists a subsequence Σj , a lamination L = {x3 = t}{t∈I} of R3 by

parallel planes (where I ⊂ R is a closed set), and a closed nonempty set S in the
union of the leaves of L such that after a rotation of R3 :
(A) For each 1 > α > 0, Σj \ S converges in the C α -topology to the lamination
L \ S.
(B) supBr (x)∩Σj |A|2 → ∞ as j → ∞ for all r > 0 and x ∈ S. (The curvatures
blow up along S.)
One can get both types of curvature blow-up by considering the family of embed-
ded minimal planar domains known as the Riemann examples. Modulo translations
and rotations, this is a two-parameter family of periodic minimal surfaces, where
the parameters can be thought of as the size of the necks and the angle from one
fundamental domain to the next. By choosing the two parameters appropriately,
one can produce sequences of Riemann examples that illustrate both of the two
structure theorems (cf. figures 8 and 9):
(1) If we take a sequence of Riemann examples where the neck size is fixed and
the angles go to π2 , then the surfaces with angle near π2 can be obtained
by gluing together two oppositely–oriented double spiral staircases. Each
double spiral staircase looks like a helicoid. This sequence of Riemann
examples converges to a foliation by parallel planes. The convergence is
smooth away from the axes of the two helicoids (these two axes are the
singular set S where the curvature blows up). The sequence is ULSC since
the size of the necks is fixed and thus illustrates the first structure theorem,
Theorem 12.2.
(2) If we take a sequence of examples where the neck sizes go to zero, then we
get a sequence that is not ULSC. However, the surfaces can be cut along
short curves into collections of graphical pairs of pants. The short curves
converge to points and the graphical pieces converge to flat planes except
at these points, illustrating the second structure theorem, Theorem 12.5
below.
With these examples in mind, we are now ready to state our second main struc-
ture theorem describing the case where Sulsc is empty.
Theorem 12.5. Let a sequence Σi , limit lamination L, and singular set S be as
in Theorem 12.4. If Sulsc = ∅ and
(12.7) sup |A|2 → ∞ ,
B1 ∩Σi

then S = Sneck by (12.5) and

(Cneck ) Each point y in S comes with a sequence of graphs in Σj that converge
to the plane {x3 = x3 (y)}. The convergence is in the C ∞ topology away
from the point y and possibly also one other point in {x3 = x3 (y)} ∩ S.
If the convergence is away from one point, then these graphs are defined
over annuli; if the convergence is away from two points, then the graphs are
defined over disks with two subdisks removed.
 MINIMAL SUBMANIFOLDS 25

Part 3. Global theory of minimal surfaces in R3

13.1. Minimal surfaces with finite total curvature. A complete, non-compact,
minimal immersion of a surface Σ in R3 is said to have finite total curvature (or
ftc) if
Z
(13.1) |A|2 < ∞ .
Σ

The simplest examples are the plane (where the total curvature is zero) and the
catenoid. In the case of the catenoid, the Gauss map gives a conformal diffeo-
morphism to the sphere punctured at the north and south poles. Since A is the
differential of the Gauss map, it follows that the catenoid has total curvature 8π.
The fundamental result for minimal surfaces with ftc is that they are all conformally
diffeomorphic to compact Riemann surfaces with a finite number of points removed.
Namely, we have the following result of R. Osserman (see [79]):
Theorem 13.1. Let Σ ⊂ R3 be a complete minimal immersion with finite total
curvature. Then:
(1) Σ is conformally diffeomorphic to a compact Riemann surface with a finite
set of points removed. Each point corresponds to an end of the surface.
(2) Σ is proper.
(3) The Weierstrass data extends across the punctures meromorphically. (See
Section 17 for the definition of the Weierstrass representation.)
(4) The total curvature is an integral multiple of 8π.
One application of this theorem is a classification of the ends of an embedded
minimal surface with finite total curvature. Namely, one can show that any such
end is asymptotic to either a plane or to half of a catenoid.

13.2. The uniqueness of the catenoid. Given the structure result, Theorem
13.1, it is natural to try to understand the space of minimal surfaces with finite
total curvature in terms of the genus and the number of ends. Two such results were
proven by R. Schoen and F. Lopez–A. Ros, respectively. The theorem of Schoen
says that the catenoid is the unique embedded minimal surface with finite total
curvature and exactly two ends.
Theorem 13.2. [89] If Σ ⊂ R3 is a connected embedded minimal surface with
finite total curvature and exactly two ends, then Σ is a catenoid.
The theorem of Lopez and Ros says that the catenoid and the plane are the
only embedded minimal surfaces with finite total curvature and genus zero. Here,
“genus zero” means conformal to the sphere with a finite set of points removed.
Theorem 13.3. [59] If Σ ⊂ R3 is an embedded minimal surface with finite total
curvature and genus zero, then Σ is a catenoid or a plane.

14. Global theory of embedded minimal surfaces

Recent years have seen breakthroughs on many long–standing problems in the
global theory of minimal surfaces in R3 . This is an enormous subject and, rather
than give a comprehensive treatment, we will mention a few important results which
fit well with the rest of this survey. Throughout this section, Σ will be a complete
26 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

properly embedded minimal surface in R3 (recall that properness here means that
the intersection of Σ with any compact subset of R3 is compact).
We say that Σ has finite topology if it is homeomorphic to a closed Riemann
surface with a finite number of punctures; the genus of Σ is then the genus of this
Riemann surface and the number of punctures is the number of ends. It follows
that a neighborhood of each puncture corresponds to a properly embedded annular
end of Σ. Perhaps surprisingly at first, the more restrictive case is when Σ has
more than one end. The reason for this is that a barrier argument gives a stable
minimal surface between any pair of ends. Such a stable surface is then asymptotic
to a plane (or catenoid), essentially forcing each end to live in a half–space. Using
this restriction, P. Collin proved:
Theorem 14.1. [30] Each end of a complete properly embedded minimal surface
with finite topology and at least two ends is asymptotic to a plane or catenoid.
In particular, outside some compact set, Σ is given by a finite collection of disjoint
graphs over a common plane (and has finite total curvature).
As mentioned above, Collin essentially proved Theorem 14.1 by showing that an
embedded annular end that lives in a half–space must have finite total curvature.
[27] used the one–sided curvature estimate to strengthen this from a half–space
to a strictly larger cone, and in the process give a very different proof of Collin’s
theorem.
Theorem 14.2. [27] There exists ǫ > 0 so that any complete properly embedded
minimal annular end contained in the cone
(14.1) {x3 ≥ −ǫ (x21 + x22 + x23 )1/2 }
is asymptotic to a plane or catenoid.

When Σ has only one end (e.g., for the helicoid), it need not have finite total
curvature so the situation is more delicate. However, the regularity results of the
previous section can be applied. For example, if Σ is a (non–planar) embedded mini-
mal disk, then we get a multi–valued graph structure away from a “one–dimensional
singular set.” Using Theorems 11.1 and 11.2, W. Meeks and H. Rosenberg proved
the uniqueness of the helicoid:
Theorem 14.3. [70] The plane and helicoid are the only complete properly embed-
ded simply–connected minimal surfaces in R3 .
This uniqueness has many applications. Recall that if we take a sequence of
rescalings of the helicoid, then the singular set S for the convergence is the vertical
axis perpendicular to the leaves of the foliation. In [64], W. Meeks used this fact
together with the uniqueness of the helicoid to prove that the singular set S in
Theorem 11.1 is always a straight line perpendicular to the foliation. Recently, W.
Meeks and M. Weber have constructed a local example (i.e., a sequence of embedded
minimal disks in a unit ball) where S is a circle.
We have not even touched on the case where Σ has infinite topology (e.g., when
Σ is one of the Riemann examples). This is an area of much current research, see
[18], the work of Meeks, J. Perez and A. Ros, [67], [68], and [69], the survey [66]
and references therein.
 MINIMAL SUBMANIFOLDS 27

We close this section with a local analog of the two–ended case. Namely, in [22],
we proved that any embedded minimal annulus in a ball (with boundary in the
boundary of the ball and) with a small neck can be decomposed by a simple closed
geodesic into two graphical sub–annuli. Moreover, we gave a sharp bound for the
length of this closed geodesic in terms of the separation (or height) between the
graphical sub–annuli. This serves to illustrate our “pair of pants” decomposition
from [18] in the special case where the embedded minimal planar domain is an
annulus (we will not touch on this further here). The catenoid
(14.2) {x21 + x22 = cosh2 x3 }
is the prime example of an embedded minimal annulus.
The precise statement of this decomposition for annuli is:
Theorem 14.4. [22] There exist ǫ > 0, C1 , C2 , C3 > 1 so: If Σ ⊂ BR ⊂ R3 is an
embedded minimal annulus with ∂Σ ⊂ ∂BR and π1 (BǫR ∩ Σ) 6= 0, then there exists
a simple closed geodesic γ ⊂ Σ of length ℓ so that:
R component of BR/C1 ∩ Σ containing it into
• The curve γ splits the connected
two annuli Σ+ , Σ− each with |A|2 ≤ 5 π.
• Furthermore, Σ± \ TC2 ℓ (γ) are graphs with gradient ≤ 1.
• Finally, ℓ log(R/ℓ) ≤ C3 h where the separation h is given by
(14.3) h= min |x+ − x− | .
x± ∈∂BR/C1 ∩Σ±

Here Ts (S) ⊂ Σ denotes the intrinsic s–tubular neighborhood of a subset S ⊂ Σ.

15. The Calabi-Yau conjectures

Recall that an immersed submanifold in Rn is proper if the pre-image of any
compact subset of Rn is compact in the surface. This property has played an
important role in the theory of minimal submanifolds and many of the classical
theorems in the subject assume that the submanifold is proper.
It is easy to see that any compact submanifold is automatically proper. On the
other hand, there is no reason to expect a general immersion (or even embedding) to
be proper. For example, the non-compact curve parametrized in polar coordinates
by
(15.1) ρ(t) = π + arctan(t) , θ(t) = t
spirals infinitely between the circles of radius π/2 and 3π/2. However, it was long
thought that a minimal immersion (or embedding) should be better behaved. This
principle was captured by the Calabi-Yau conjectures, dating back to the 1960s.
Much work has been done on them over the past four decades. Their original form
was given in 1965 in [10] where E. Calabi made the following two conjectures about
minimal surfaces (see also S.S. Chern, page 212 of [11] and S.T. Yau’s 1982 problem
list):
Conjecture 1. “Prove that a complete minimal hypersurface in Rn must be un-
bounded.”
Calabi continued: “It is known that there are no compact minimal submanifolds
of Rn (or of any simply connected complete Riemannian manifold with sectional
curvature ≤ 0). A more ambitious conjecture is”:
28 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

Conjecture 2. “A complete [nonflat] minimal hypersurface in Rn has an un-

bounded projection in every (n − 2)–dimensional flat subspace.”
The immersed versions of these conjectures turned out to be false. As mentioned
above, Jorge and Xavier, [50], constructed nonflat minimal immersions contained
between two parallel planes in 1980, giving a counter-example to the immersed
version of the more ambitious Conjecture 2. Another significant development came
in 1996, when N. Nadirashvili, [77], constructed a complete immersion of a minimal
disk into the unit ball in R3 , showing that Conjecture 1 also failed for immersed
surfaces; see [60], [57], [58], for other topological types than disks.
It is clear from the definition of proper that a proper minimal surface in R3
must be unbounded, so the examples of Nadirashvili are not proper. Much less
obvious is that the plane is the only complete proper immersed minimal surface in
a halfspace. This is however a consequence of the strong halfspace theorem of D.
Hoffman and W. Meeks:
Theorem 15.1. [44] A complete connected properly immersed minimal surface
Σ ⊂ {x3 > 0} ⊂ R3 must be a horizontal plane {x3 = Constant}.
The main result of [29] is an effective version of properness for disks, giving a
chord arc bound. Obviously, intrinsic distances are larger than extrinsic distances,
so the significance of a chord arc bound is the reverse inequality, i.e., a bound on
intrinsic distances from above by extrinsic distances. This is accomplished in the
next theorem:
Theorem 15.2. [29] There exists a constant C > 0 so that if Σ ⊂ R3 is an
embedded minimal disk, B2R = B2R (0) is an intrinsic ball in Σ \ ∂Σ of radius 2R,
and supBr0 |A|2 > r0−2 where R > r0 , then for x ∈ BR
(15.2) C distΣ (x, 0) < |x| + r0 .
The assumption of a lower curvature bound, supBr0 |A|2 > r0−2 , in the theorem
is a necessary normalization for a chord arc bound. This can easily be seen by
rescaling and translating the helicoid.
Properness of a complete embedded minimal disk is an immediate consequence
of Theorem 15.2. Namely, by (15.2), as intrinsic distances go to infinity, so do
extrinsic distances. Precisely, if Σ is flat, and hence a plane, then obviously Σ is
proper and if it is non-flat, then supBr0 |A|2 > r0−2 for some r0 > 0 and hence Σ is
proper by (15.2).
A consequence of Theorem 15.2 together with the one-sided curvature estimate
of [19] (i.e., theorem 0.2 in [19]) is the following version of that estimate for intrinsic
balls:
Corollary 15.3. [29] There exists ǫ > 0, so that if
(15.3) Σ ⊂ {x3 > 0} ⊂ R3
is an embedded minimal disk with B2R (x) ⊂ Σ \ ∂Σ and |x| < ǫ R, then
(15.4) sup |AΣ |2 ≤ R−2 .
BR (x)

As a corollary of this intrinsic one-sided curvature estimate we get that the sec-
ond, and more ambitious, of Calabi’s conjectures is also true for embedded minimal
disks.
 MINIMAL SUBMANIFOLDS 29

In fact, [29] proved both of Calabi’s conjectures and properness also for embedded
surfaces with finite topology. Recall that a surface Σ is said to have finite topology
if it is homeomorphic to a closed Riemann surface with a finite set of points removed
or “punctures”. Each puncture corresponds to an end of Σ.
The following generalization of the halfspace theorem gives Calabi’s second, more
ambitious, conjecture for embedded surfaces with finite topology:
Theorem 15.4. The plane is the only complete embedded minimal surface with
finite topology in R3 in a halfspace.
Likewise, we get the properness of embedded surfaces with finite topology:
Theorem 15.5. A complete embedded minimal surface with finite topology in R3
must be proper.
Compare also W. Meeks and H. Rosenberg, [71].
There has been extensive work on both properness and the halfspace property
assuming various curvature bounds. Jorge and Xavier, [49] and [50], showed that
there cannot exist a complete immersed minimal surface with bounded curvature
in ∩i {xi > 0}; later Xavier proved that the plane is the only such surface in a
halfspace, [107]. Recently, G.P. Bessa, Jorge and G. Oliveira-Filho, [7], and H.
Rosenberg, [86], have shown that if such a surface is embedded, then it must be
proper. This properness was extended to embedded minimal surfaces with locally
bounded curvature and finite topology by Meeks and Rosenberg in [70]; finite topol-
ogy was subsequently replaced by finite genus in [67] by Meeks, J. Perez and A.
Ros.
Inspired by Nadirashvili’s examples, F. Martin and S. Morales constructed in
[61] a complete bounded minimal immersion which is proper in the (open) unit
ball. That is, the preimages of compact subsets of the (open) unit ball are compact
in the surface and the image of the surface accumulates on the boundary of the unit
ball. They extended this in [62] to show that any convex, possibly noncompact or
nonsmooth, region of R3 admits a proper complete minimal immersion of the unit
disk; cf. [78].

Part 4. Constructing minimal surfaces

Thus far, we have mainly dealt with regularity and a priori estimates but have
ignored questions of existence. In this part we survey some of the most useful
existence results for minimal surfaces. Section 16 gives an overview of the clas-
sical Plateau problem. Section 17 recalls the classical Weierstrass representation,
including a few modern applications, and the Kapouleas desingularization method.
Section 18 deals with producing area minimizing surfaces and questions of embed-
dedness. Finally, Section 19 recalls the min–max construction for producing unsta-
ble minimal surfaces and, in particular, doing so while controlling the topology and
guaranteeing embeddedness.

16. The Plateau Problem

The following fundamental existence problem for minimal surfaces is known as
the Plateau problem: Given a closed curve Γ, find a minimal surface with boundary
Γ. There are various solutions to this problem depending on the exact definition of a
30 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

surface (parameterized disk, integral current, Z2 current, or rectifiable varifold). We

shall consider the version of the Plateau problem for parameterized disks; this was
solved independently by J. Douglas and T. Rado. The generalization to Riemannian
manifolds is due to C. B. Morrey.
Theorem 16.1. Let Γ ⊂ R3 be a piecewise C 1 closed Jordan curve. Then there
exists a piecewise C 1 map u from D ⊂ R2 to R3 with u(∂D) ⊂ Γ such that the
image minimizes area among all disks with boundary Γ.
The solution u to the Plateau problem above can easily be seen to be a branched
conformal immersion. R. Osserman proved that u does not have true interior branch
points; subsequently, R. Gulliver and W. Alt showed that u cannot have false branch
points either.
Furthermore, the solution u is as smooth as the boundary curve, even up to the
boundary. A very general version of this boundary regularity was proven by S.
Hildebrandt; for the case of surfaces in R3 , recall the following result of J. C. C.
Nitsche:
Theorem 16.2. If Γ is a regular Jordan curve of class C k,α where k ≥ 1 and
0 < α < 1, then a solution u of the Plateau problem is C k,α on all of D̄.
The optimal boundary regularity theorem in higher dimensions was proven by
R. Hardt and L. Simon in [40].

17. The Weierstrass representation

The classical Weierstrass representation (see [43] or [79]) takes holomorphic data
(a Riemann surface, a meromorphic function, and a holomorphic one–form) and
associates to these data a minimal surface in R3 . To be precise, given a Riemann
surface Ω, a meromorphic function g on Ω, and a holomorphic one–form φ on Ω,
then we get a (branched) conformal minimal immersion F : Ω → R3 by
 
1 −1 i −1
Z
(17.1) F (z) = Re (g (ζ) − g(ζ)), (g (ζ) + g(ζ)), 1 φ(ζ) .
ζ∈γz0 ,z 2 2
Here z0 ∈ Ω is a fixed base point and the integration is along a path γz0 ,z from
z0 to z. The choice of z0 changes F by adding a constant. In general, the map
F may depend on the choice of path (and hence may not be well–defined); this is
known as “the period problem” (see M. Weber and M. Wolf, [104], for the latest
developments). However, when g has no zeros or poles and Ω is simply connected,
then F (z) does not depend on the choice of path γz0 ,z .
Three standard constructions of minimal surfaces from Weierstrass data are
(17.2) g(z) = z, φ(z) = dz/z, Ω = C \ {0} giving a catenoid ,
(17.3) g(z) = z, φ(z) = z dz, Ω = C giving Enneper’s surface ,
(17.4) g(z) = eiz , φ(z) = dz, Ω = C giving a helicoid .
The Weierstrass representation is particularly useful for constructing immersed
minimal surfaces. For example, in [77], N. Nadirashvili used it to construct a
complete immersed minimal surface in the unit ball in R3 (see also [50] for the case
of a slab). In particular, Nadirashvili’s surface is not proper, i.e., the intersections
with compact sets are not necessarily compact.
 MINIMAL SUBMANIFOLDS 31

Typically, it is rather difficult to prove that the resulting immersion is an embed-

ding (i.e., is 1–1), although there are some interesting cases where this can be done.
The first modern example was [44] where D. Hoffman and W. Meeks proved that
the surface constructed by Costa was embedded; this was the first new complete
finite topology properly embedded minimal surface discovered since the classical
catenoid, helicoid, and plane. This led to the discovery of many more such surfaces
(see [43] and [84] for more discussion).
Very recently in [45] (see also [46]), D. Hoffman, M. Weber, and M. Wolf have
used the Weierstrass representation to construct a genus one properly embedded
minimal surface asymptotic to the helicoid. They construct this “genus one heli-
coid” as the limit of a continuous one-parameter family of screw-motion invariant
minimal surfaces–also asymptotic to the helicoid–that have genus equal to one in
the quotient.
In [24], we used the Weierstrass representation to construct a sequence of em-
bedded minimal disks
(17.5) Σi ⊂ B1 = B1 (0) ⊂ R3
with ∂Σi ⊂ ∂B1 where the curvatures blow up only at 0 and Σi \ {x3 –axis} consists
of two multi–valued graphs for each i. Furthermore, Σi \ {x3 = 0} converges to two
embedded minimal disks Σ− ⊂ {x3 < 0} and Σ+ ⊂ {x3 > 0} each of which spirals
into {x3 = 0} and thus is not proper. (This should be contrasted with Theorem
11.1 where the complete limits are planes and hence proper.)
N. Kapouleas has developed another method to construct complete embedded
minimal surfaces with finite total curvature. For instance, in [54], he shows that
(most) collections of coaxial catenoids and planes can be desingularized to get
complete embedded minimal surfaces with finite total curvature. The Costa surface
above had genus one and three ends (that is to say, it is homeomorphic to a torus
with three punctures). In the Kapouleas construction, one could start with a plane
and catenoid intersecting in a circle and then desingularize this circle using suitably
scaled and bent Scherk surfaces to get a finite genus embedded surface with three
ends. (This desingularization process adds handles, i.e., increases the genus.) In
this manner, Kapouleas gets an enormous number of new examples; see also the
gluing construction of S.D. Yang, [108], which uses catenoid necks to glue together
nearby minimal surfaces.

18. Area–minimizing surfaces

Perhaps the most natural way to construct minimal surfaces is to look for ones
which minimize area, e.g., with fixed boundary, or in a homotopy class, etc. This
has the advantage that often it is possible to show that the resulting surface is
embedded. We mention a few results along these lines.
The first embeddedness result, due to W. Meeks and S.T. Yau, shows that if the
boundary curve is embedded and lies on the boundary of a smooth mean convex
set (and it is null–homotopic in this set), then it bounds an embedded least area
disk.
Theorem 18.1. [74] Let M 3 be a compact Riemannian three–manifold whose
boundary is mean convex and let γ be a simple closed curve in ∂M which is null–
homotopic in M ; then γ is bounded by a least area disk and any such least area disk
is properly embedded.
32 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

Note that some restriction on the boundary curve γ is certainly necessary. For
instance, if the boundary curve was knotted (e.g., the trefoil), then it could not
be spanned by any embedded disk (minimal or otherwise). Prior to the work of
Meeks and Yau, embeddedness was known for extremal boundary curves in R3 with
small total curvature by the work of R. Gulliver and J. Spruck [38]; see chapter 4
in [14] for other results and further discussion. Recently, in [33], T. Ekholm, B.
White, and D. Wienholtz proved that minimal surfaces whose boundary has total
curvature less than 4π also must be embedded.
If we instead fix a homotopy class of maps, then the two fundamental exis-
tence results are due to J. Sacks–K. Uhlenbeck and R. Schoen– S.T. Yau (with
embeddedness proven by W. Meeks–S.T. Yau and M. Freedman–J. Hass–P. Scott,
respectively):
Theorem 18.2. [87], [75] Given M 3 , there exist conformal (stable) minimal im-
mersions u1 , . . . , um : S2 → M which generate π2 (M ) as a Z[π1 (M )] module.
Furthermore,
• If u : S2 → M and [u]π2 6= 0, then Area(u) ≥ mini Area(ui ).
• Each ui is either an embedding or a 2–1 map onto an embedded 2–sided
RP 2 .
Theorem 18.3. [93], [36] If Σ2 is a closed surface with genus g > 0 and i0 : Σ →
M 3 is an embedding which induces an injective map on π1 , then there is a least
area embedding with the same action on π1 .
In [72], W. Meeks, L. Simon, and S.T. Yau find an embedded sphere minimizing
area in an isotopy class in a closed 3–manifold.
We end this section by mentioning two applications of Theorem 18.3. First, in
[26], we showed that any topological 3–manifold M had an open set of metrics so
that, for each such metric, there was a sequence of embedded minimal tori whose
area went to infinity. In [32], B. Dean showed that this was true for every genus
g ≥ 1.

19. The min–max construction of minimal surfaces

Variational arguments can also be used to construct higher index (i.e., non–
minimizing) minimal surfaces using the topology of the space of surfaces. There
are two basic approaches:
• Applying Morse theory to the energy functional on the space of maps from
a fixed surface Σ to M .
• Doing a min–max argument over families of (topologically non–trivial)
sweep–outs of M .
The first approach has the advantage that the topological type of the minimal
surface is easily fixed; however, the second approach has been more successful at
producing embedded minimal surfaces. We will highlight a few key results below
but refer to [13] for a thorough treatment.
Unfortunately, one cannot directly apply Morse theory to the energy functional
on the space of maps from a fixed surface because of a lack of compactness (the
Palais–Smale Condition C does not hold). To get around this difficulty, J. Sacks
and K Uhlenbeck, [87], introduce a family of perturbed energy functionals which
 MINIMAL SUBMANIFOLDS 33

do satisfy Condition C and then obtain minimal surfaces as limits of critical points
for the perturbed problems:

Theorem 19.1. [87] If πk (M ) 6= 0 for some k > 1, then there exists a branched
immersed minimal 2–sphere in M (for any metric).

This was sharpened somewhat by M. Micallef and D. Moore, [76], (showing

that the index of the minimal sphere was at most k − 2), who used it to prove a
generalization of the sphere theorem. See A. Fraser, [37], for a generalization to a
free boundary problem.
The basic idea of constructing minimal surfaces via min–max arguments and
sweep–outs goes back to G. Birkhoff, who developed it to construct simple closed
geodesics on spheres. In particular, when M is a topological 2–sphere, we can
find a 1–parameter family of curves starting and ending at point curves so that
the induced map F : S2 → S2 (see figure 12) has nonzero degree. The min–max
argument produces a nontrivial closed geodesic of length less than or equal to the
longest curve in the initial one–parameter family. A curve shortening argument
gives that the geodesic obtained in this way is simple.

Figure 12. A 1–parameter family of curves on a 2–sphere which induces a map

F : S2 → S2 of degree 1. First published in Surveys in Differential Geometry,
volume IX, in 2004, published by International Press.

In [82], J. Pitts applied a similar argument and geometric measure theory to

get that every closed Riemannian three manifold has an embedded minimal surface
(his argument was for dimensions up to seven), but he did not estimate the genus
of the resulting surface. Finally, F. Smith (under the direction of L. Simon) proved
(see [13]):

Theorem 19.2. [99] Every metric on a topological 3–sphere M admits an embedded

minimal 2–sphere.

The main new contribution of Smith was to control the topological type of the
resulting minimal surface while keeping it embedded; see also J. Pitts and J.H.
Rubinstein, [83], for some generalizations.

Part 5. Some applications of minimal surfaces

In this last part, we discuss very briefly a few applications of minimal surfaces.
As mentioned in the introduction, there are many to choose from and we have
selected just a few. See [4], [35], [39], [42], [101], and [102] for other applications.
34 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

20.1. The positive mass theorem. The (Riemannian version of the) positive
mass theorem of R. Schoen and S.T. Yau states that an asymptotically flat 3-
manifold M with non-negative scalar curvature must have positive mass. The Rie-
mannian manifold M here arises as a maximal space-like slice in a 3+1-dimensional
space-time solution of Einstein’s equations.
The asymptotic flatness of M comes from that the space-time models an isolated
gravitational system and hence is a perturbation of the vacuum solution outside a
large compact set. To make this precise, suppose for simplicity that M has only
one end; M is then said to be asymptotically flat if there is a compact set Ω ⊂ M so
that M \ Ω is diffeomorphic to R3 \ BR (0) and the metric on M \ Ω can be written
as
 4
M
(20.1) gij = 1 + δij + pij ,
2 |x|
where
(20.2) |x|2 |pij | + |x|3 |D pij | + |x|4 |D2 pij | ≤ C .
The constant M is the so called mass of M . Observe that the metric gij is a
perturbation of the metric on a constant-time slice in the Schwarzschild space-time
of mass M; that is to say, the Schwarzschild metric has pij ≡ 0.
A tensor h is said to be O(|x|−p ) if |x|p |h| + |x|p+1 |D h| ≤ C. For example, an
easy calculation shows that
(20.3) gij = (1 + 2 M/|x|) δij + O(|x|−2 ) ,
√
g ≡ det gij = 1 + 3 M |x|−1 + O(|x|−2 ) .
p

The positive mass theorem states that the mass M of such an M must be non-
negative:
Theorem 20.1. [92] With M as above, M ≥ 0.
There is a rigidity theorem as well which states that the mass vanishes only when
M is isometric to R3 :
Theorem 20.2. [92] If |∇3 pij | = O(|x|−5 ) and M = 0 in Theorem 20.1, then M
is isometric to R3 .
We will give a very brief overview of the proof of Theorem 20.1, showing in the
process where minimal surfaces appear.

Proof. (Sketch.) The argument will by contradiction, so suppose that the mass is
negative. Note first that a “rounding off” argument shows that the metric on M
can be perturbed to have positive scalar curvature outside of a compact set and
still have negative mass.
It is not hard to prove that the slab between two parallel planes is mean-convex.
That is, we have the following:
Lemma 20.3. If M < 0 and M is asymptotically flat, then there exist R0 , h > 0
so that for r > R0 the sets
(20.4) Cr = {|x|2 ≤ r2 , −h ≤ x3 ≤ h}
have strictly mean convex boundary.
 MINIMAL SUBMANIFOLDS 35

Since the compact set Cr is mean convex, we can solve the Plateau problem (as
in Section 16) to get an area minimizing (and hence stable) surface Γr ⊂ Cr with
boundary
(20.5) ∂Γr = {|x|2 = r2 , x3 = h} .
Using the disk {|x|2 ≤ r2 , x3 = h} as a comparison surface, we get uniform local
area bounds for any such Γr . Combining these local area bounds with the a priori
curvature estimates for minimizing surfaces, we can take a sequence of r’s going to
infinity and find a subsequence of Γr ’s that converge to a complete area-minimizing
surface
(20.6) Γ ⊂ {−h ≤ x3 ≤ h} .
Since Γ is pinched between the planes {x3 = ±h}, the estimates for minimizing
surfaces implies that (outside a large compact set) Γ is a graph over the plane
{x3 = 0} and hence has quadratic area growth and finite total curvature. Moreover,
using the form of the metric gij , we see that |∇u| decays like |x|−1 and
Z
(20.7) kg = (2 π s + O(1)) (s−1 + O(s−2 )) = 2 π + O(s−1 ) ,
σs

where σs is the curve {x21 + x22 = s2 } ∩ Γ and kg is the geodesic curvature of σs

(considered as a curve in the surface Γ).
To get the contradiction, one combines stability of Γ with the positive scalar
curvature of M to see that no such Γ could have existed. Namely, substituting the
Gauss equation into the stability inequality in a general 3-manifold (see, e.g., [14])
gives
Z Z
(20.8) (|A|2 /2 + ScalM − KΣ )φ2 ≤ |∇φ|2 .
Γ Γ
Since Γ has quadratic area growth, we can choose a sequence of (logarithmic) cutoff
functions in (20.8) to get
Z Z
2
(20.9) 0 < (|A| /2 + ScalM ) ≤ KΣ < ∞ ;
Σ Σ
since KΣ may not be positive, we also used that Γ has finite total curvature. More-
over, we used that ScalM is positive outside a compact set to see that the first
integral in (20.9) was
R positive. Finally, substituting (20.9) into the Gauss-Bonnet
formula gives that σs kg is strictly less than 2π for s large, contradicting (20.7). 

20.2. Black holes. Another way that minimal surfaces enter into relativity is
through black holes. Suppose that we have a three-dimensional time-slice M in
a 3 + 1-dimensional space–time. For simplicity, assume that M is totally geodesic
and hence has non-negative scalar curvature. A closed surface Σ in M is said to be
trapped if its mean curvature is everywhere negative with respect to its outward
normal. Physically, this means that the surface emits an outward shell of light
whose surface area is decreasing everywhere on the surface. The existence of a
closed trapped surface implies the existence of a black hole in the space-time.
Given a trapped surface, we can look for the outermost trapped surface con-
taining it; this outermost surface is called an apparent horizon. It is not hard to
see that an apparent horizon must be a minimal surface and, moreover, a barrier
argument shows that it must be stable. Since M has non-negative scalar curvature,
36 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

stability in turn implies that it must be diffeomorphic to a sphere. See, for instance,
[9] and [47] for some results on black holes, horizons, etc.

20.3. Constant mean curvature surfaces. At least since the time of Plateau,
minimal surfaces have been used to model soap films. This is because the mean
curvature of the surface models the surface tension and this is essentially the only
force acting on a soap film. Soap bubbles, on other hand, enclose a volume and
thus the pressure gives a second counterbalancing force. It follows easily that these
two forces are in equilibrium when the surface has constant mean curvature.
For the same reason, constant mean curvature surfaces arise in the isoperimetric
problem. Namely, a surface that minimizes surface area while enclosing a fixed
volume must have constant mean curvature (or “cmc”). It is not hard to see that
such an isoperimetric surface in Rn must be a round sphere. There are two inter-
esting partial converses to this. First, by a theorem of H. Hopf, any cmc 2-sphere in
R3 must be round. Second, using the maximum principle (“the method of moving
planes”) A.D. Alexandrov showed that any closed embedded cmc hypersurface in
Rn must be a round sphere. It turned out, however, that not every closed immersed
cmc surface is round. The first examples were immersed cmc tori constructed by
H. Wente, [105]. N. Kapouleas constructed many new examples, including closed
higher genus cmc surfaces, [52], [53]. See also [63] and [55] for other results on the
space of such cmc surfaces.
Many of the techniques developed for studying minimal surfaces generalize to
general constant mean curvature surfaces.

20.4. Finite extinction for Ricci flow. We close this survey with indicating how
minimal surfaces can be used to show that on a homotopy 3-sphere the Ricci flow
become extinct in finite time (see [25], [80] for details).
Let M 3 be a smooth closed orientable 3–manifold and let g(t) be a one–parameter
family of metrics on M evolving by the Ricci flow, so

(20.10) ∂t g = −2 RicMt .
In an earlier section, we saw that there is a natural way of constructing minimal
surfaces on many 3-manifolds and that comes from the min–max argument where
the minimal of all maximal slices of sweep–outs is a minimal surface. The idea
is then to look at how the area of this min–max surface changes under the flow.
Geometrically the area measures a kind of width of the 3–manifold and as we will
see for certain 3–manifolds (those, like the 3–sphere, whose prime decomposition
contains no aspherical factors) the area becomes zero in finite time corresponding
to that the solution becomes extinct in finite time.
Fix a continuous map β : [0, 1] → C 0 ∩ L21 (S2 , M ) where β(0) and β(1) are
constant maps so that β is in the nontrivial homotopy class [β] (such β exists when
M is a homotopy 3-sphere). We define the width W = W (g, [β]) by
(20.11) W (g) = min max Energy(γ(s)) .
γ∈[β] s∈[0,1]

The next theorem gives an upper bound for the derivative of W (g(t)) under the
Ricci flow which forces the solution g(t) to become extinct in finite.
 MINIMAL SUBMANIFOLDS 37

The min-max surface.

Figure 13. The sweep–out, the min–max surface, and the width
W. First published in the Journal of the American Mathematical
Society in 2005, published by the American Mathematical Society.

Theorem 20.4. Let M 3 be a homotopy 3-sphere equipped with a Riemannian met-

ric g = g(0). Under the Ricci flow, the width W (g(t)) satisfies
d 3
(20.12) W (g(t)) ≤ −4π + W (g(t)) ,
dt 4(t + C)
in the sense of the limsup of forward difference quotients. Hence, g(t) must become
extinct in finite time.
The 4π in (20.12) comes from the Gauss–Bonnet theorem and the 3/4 comes
from the bound on the minimum of the scalar curvature that the evolution equation
implies. Both of these constants matter whereas the constant C depends on the
initial metric and the actual value is not important.
To see that (20.12) implies finite extinction time rewrite (20.12) as
d  
(20.13) W (g(t)) (t + C)−3/4 ≤ −4π (t + C)−3/4
dt
and integrate to get
h i
(20.14) (T + C)−3/4 W (g(T )) ≤ C −3/4 W (g(0)) − 16 π (T + C)1/4 − C 1/4 .
Since W ≥ 0 by definition and the right hand side of (20.14) would become negative
for T sufficiently large we get the claim.
As a corollary of this theorem we get finite extinction time for the Ricci flow.
Corollary 20.5. Let M 3 be a homotopy 3-sphere equipped with a Riemannian
metric g = g(0). Under the Ricci flow g(t) must become extinct in finite time.

References
[1] W. Allard, On the first variation of a varifold, Ann. of Math. (2) 95 (1972) 417–491.
[2] F. J. Almgren, Jr., Some interior regularity theorems for minimal surfaces and an extension
of Bernstein’s theorem, Ann. of Math. (2) 84 (1966) 277–292.
[3] F. J. Almgren, Jr., Almgren’s big regularity paper. Q-valued functions minimizing Dirichlet’s
integral and the regularity of area-minimizing rectifiable currents up to codimension 2, With
a preface by Jean E. Taylor and Vladimir Scheffer. World Scientific Monograph Series in
Mathematics, 1. World Scientific Publishing Co., Inc., River Edge, NJ, 2000.
[4] F. J. Almgren, Jr., Minimal surface forms, Math. Intell. 4 (1982) 164–172.
38 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

[5] S. Bernstein, Über ein geometrisches Theorem und seine Anwendung auf die partiellen Dif-
ferentialgleichungen vom ellipschen Typos. Math. Zeit. 26 (1927) 551–558 (translation of the
original version in Comm. Soc. Math. Kharkov 2-ème sér. 15 (1915–17) 38–45).
[6] L. Bers, Isolated singularities of minimal surfaces, Ann. of Math. (2) 53 (1951) 364–386.
[7] G.P. Bessa, L. Jorge and G. Oliveira-Filho, Half-space theorems for minimal surfaces with
bounded curvature, J. Diff. Geom. 57 (2001) 493–508.
[8] E. Bombieri, E. De Giorgi, and E. Giusti, Minimal cones and the Bernstein problem, Invent.
Math. 7 (1969) 243–268.
[9] H. Bray, Proof of the Riemannian Penrose inequality using the positive mass theorem, J.
Differential Geom. 59 (2001), no. 2, 177–267.
[10] E. Calabi, Problems in differential geometry, Ed. S. Kobayashi and J. Eells, Jr., Proceedings
of the United States-Japan Seminar in Differential Geometry, Kyoto, Japan, 1965. Nippon
Hyoronsha Co., Ltd., Tokyo (1966) 170.
[11] S.S. Chern, The geometry of G-structures, Bull. Amer. Math. Soc. 72 (1966) 167–219.
[12] H.I. Choi and R. Schoen, The space of minimal embeddings of a surface into a three-
dimensional manifold of positive Ricci curvature, Invent. Math. 81 (1985) 387–394.
[13] T.H. Colding and C. De Lellis, The min–max construction of minimal surfaces, Surveys in
differential geometry, Vol. 8, Lectures on Geometry and Topology held in honor of Calabi,
Lawson, Siu, and Uhlenbeck at Harvard University, May 3–5, 2002, Sponsored by the Journal
of Differential Geometry, (2003) 75–107, math.AP/0303305.
[14] T.H. Colding and W.P. Minicozzi II, Minimal surfaces, Courant Lecture Notes in Math., v.
4, 1999.
[15] T.H. Colding and W.P. Minicozzi II, Estimates for parametric elliptic integrands, Interna-
tional Mathematics Research Notices, no. 6 (2002) 291-297.
[16] T.H. Colding and W.P. Minicozzi II, The space of embedded minimal surfaces of fixed genus
in a 3-manifold I; Estimates off the axis for disks, Annals of Math., 160 (2004) 27–68,
math.AP/0210106.
[17] T.H. Colding and W.P. Minicozzi II, The space of embedded minimal surfaces of fixed
genus in a 3-manifold II; Multi-valued graphs in disks, Annals of Math., 160 (2004) 69–92,
math.AP/0210086.
[18] T.H. Colding and W.P. Minicozzi II, The space of embedded minimal surfaces of fixed genus in
a 3-manifold III; Planar domains, Annals of Math., 160 (2004) 523–572, math.AP/0210141.
[19] T.H. Colding and W.P. Minicozzi II, The space of embedded minimal surfaces of fixed
genus in a 3-manifold IV; Locally simply connected, Annals of Math., 160 (2004) 573–615,
math.AP/0210119.
[20] T.H. Colding and W.P. Minicozzi II, The space of embedded minimal surfaces of fixed genus
in a 3-manifold V; Fixed genus, math.DG/0509647.
[21] T.H. Colding and W.P. Minicozzi II, Multi-valued minimal graphs and properness of disks,
International Mathematics Research Notices, no. 21 (2002) 1111-1127.
[22] T.H. Colding and W.P. Minicozzi II, On the structure of embedded minimal annuli, Interna-
tional Mathematics Research Notices, no. 29 (2002) 1539–1552.
[23] T.H. Colding and W.P. Minicozzi II, Disks that are double spiral staircases, Notices of the
AMS, Vol. 50, no. 3, March (2003) 327–339.
[24] T.H. Colding and W.P. Minicozzi II, Embedded minimal disks: Proper versus nonproper -
global versus local, Transactions of the AMS, 356 (2004) 283-289, math.DG/0210328.
[25] T.H. Colding and W.P. Minicozzi II, Estimates for the extinction time for the Ricci flow
on certain 3–manifolds and a question of Perelman, JAMS, 18 (2005), no. 3, 561–569,
math.AP/0308090.
[26] T.H. Colding and W.P. Minicozzi II, Examples of embedded minimal tori without area
bounds, International Mathematics Research Notices, 99 no. 20 (1999) 1097–1100.
[27] T.H. Colding and W.P. Minicozzi II, Complete properly embedded minimal surfaces in R3 ,
Duke Math. J. 107 (2001) 421–426.
[28] T.H. Colding and W.P. Minicozzi II, Embedded minimal disks, Global theory of min-
imal surfaces, 405–438, Clay Math. Proc., 2, Amer. Math. Soc., Providence, RI, 2005,
math.DG/0206146.
[29] T.H. Colding and W.P. Minicozzi II, The Calabi–Yau conjectures for embedded surfaces,
preprint, math.DG/0404197.
 MINIMAL SUBMANIFOLDS 39

[30] P. Collin, Topologie et courbure des surfaces minimales proprement plonges de R3 , Ann. of
Math. (2) 145 (1997) 1–31.
[31] E. De Giorgi, Frontiere orientate di misura minima, Sem. Mat. Scuola Norm. Sup. Pisa
(1961) 1–56.
[32] B. Dean, Compact Embedded Minimal Surfaces of Positive Genus Without Area Bounds,
Geom. Ded. 102 (2003), 45–52.
[33] T. Ekholm, B. White, and D. Wienholtz, Embeddedness of minimal surfaces with total bound-
ary curvature at most 4π, Ann. of Math. (2) 155 (2002), no. 1, 209–234.
[34] H. Federer, Geometric measure theory, Springer-Verlag, Berlin–Heidelberg–New York, 1969.
[35] R. Finn, Capillary surface interfaces, Notices Amer. Math. Soc. 46 (1999), no. 7, 770–781.
[36] M.H. Freedman, J. Hass, and P. Scott, Least area incompressible surfaces in 3-manifolds,
Invent. Math. 71 (1983), no. 3, 609–642.
[37] A. Fraser, On the free boundary variational problem for minimal disks, Comm. Pure Appl.
Math. 53 (2000) 931–971.
[38] R. Gulliver and J. Spruck, On embedded minimal surfaces, Ann. of Math. (2) 103 (1976)
331–347; Ann. of Math. (2) 109 (1979) 407–412.
[39] R. Hardt, D. Kinderlehrer, and F. H. Lin, The variety of configurations of static liquid
crystals, Variational methods (Paris, 1988), 115–131, Progr. Nonlinear Differential Equations
Appl., 4, Birkhuser Boston, Boston, MA, 1990.
[40] R. Hardt and L. Simon, Boundary regularity and embedded solutions for the oriented Plateau
problem, Ann. of Math. (2) 110 (1979), no. 3, 439–486.
[41] E. Heinz, Über die Lösungen der Minimalflächengleichung, Nachr. Akad. Wiss. Göttingen
Math.–Phys. Kl, II (1952) 51–56.
[42] D. Hoffman, Mixing materials and mathematics, Nature 384 (1996) 28 - 29.
[43] D. Hoffman and H. Karcher, Complete embedded minimal surfaces with finite total curvature,
Geometry V (R. Osserman, ed.) Encyclopaedia Math. Sci. 90, Springer-Verlag, New York
(1997) 5–93.
[44] D. Hoffman and W. Meeks III, A complete embedded minimal surface in R3 with genus one
and three ends, J. Diff. Geom. 21 (1985) 109–127.
[45] D. Hoffman, M. Weber, and M. Wolf, An embedded genus-one helicoid, math.DG/0401080.
[46] D. Hoffman, M. Weber, and M. Wolf, An embedded genus-one helicoid, PNAS, November 15,
2005, vol. 102, no. 46.
[47] G. Huisken and T. Ilmanen, The inverse mean curvature flow and the Riemannian Penrose
inequality, J. Differential Geom. 59 (2001), no. 3, 353–437.
[48] T. Ilmanen, A strong maximum principle for singular minimal hypersurfaces, Calc. Var.
Partial Differential Equations 4 (1996), no. 5, 443–467.
[49] L. Jorge and F. Xavier, On the existence of complete bounded minimal surfaces in Rn , Bol.
Soc. Brasil. Mat. 10 (1979), no. 2, 171–173.
[50] L. Jorge and F. Xavier, A complete minimal surface in R3 between two parallel planes,
Annals of Math. (2) 112 (1980) 203–206.
[51] L. Jorge and F. Xavier, An inequality between the exterior diameter and the mean curvature
of bounded immersions, Math. Zeit. 178 (1981), no. 1, 77–82.
[52] N. Kapouleas, Compact constant mean curvature surfaces in Euclidean three-space, J. Dif-
ferential Geom. 33 (1991), no. 3, 683–715.
[53] N. Kapouleas, Complete constant mean curvature surfaces in Euclidean three-space, Ann. of
Math. (2) 131 (1990), no. 2, 239–330.
[54] N. Kapouleas, Complete embedded minimal surfaces of finite total curvature, J. Diff. Geom.,
47 (1997) 95–169.
[55] N. Korevaar and R. Kusner, The global structure of constant mean curvature surfaces, Invent.
Math. 114 (1993), no. 2, 311–332.
[56] H. B. Lawson, Lectures on minimal submanifolds, vol. I, Publish or Perish, Inc, Berkeley,
1980.
[57] F. Lopez, F. Martin, and S. Morales, Adding handles to Nadirashvili’s surfaces, J. Diff. Geom.
60 (2002), no. 1, 155–175.
[58] F. Lopez, F. Martin, and S. Morales, Complete nonorientable minimal surfaces in a ball of
R3 , preprint.
[59] F. Lopez and A. Ros, On embedded complete minimal surfaces of genus zero, J. Differential
Geom. 33 (1991), no. 1, 293–300.
40 TOBIAS H. COLDING AND WILLIAM P. MINICOZZI II

[60] F. Martin and S. Morales, A complete bounded minimal cylinder in R3 , Michigan Math. J.
47 (2000), no. 3, 499–514.
[61] F. Martin and S. Morales, On the asymptotic behavior of a complete bounded minimal surface
in R3 , Trans. Amer. Math. Soc. 356 (2004), 3985–3994.
[62] F. Martin and S. Morales, Complete proper minimal surfaces in convex bodies of R3 , Duke
Math. J. 128 (2005), no. 3, 559–593.
[63] R. Mazzeo, F. Pacard, and D. Pollack,Connected sums of constant mean curvature surfaces
in Euclidean 3 space, J. Reine Angew. Math. 536 (2001), 115–165.
[64] W. Meeks III, The regularity of the singular set in the Colding and Minicozzi lamination
theorem, Duke Math. Jour., 123 (2004), no. 2, 329–334.
[65] W. Meeks III, The lamination metric for the Colding-Minicozzi minimal lamination, Illinois
J. Math. 49 (2005), no. 2, 645–658.
[66] W. Meeks III and J. Perez, Conformal properties in classical minimal surface theory, Sur-
veys in Diff. Geom. IX: Eigenvalues of Laplacians and other geometric operators, Ed. by A.
Grigor’yan and S.T. Yau, International Press (2004), 275–335.
[67] W. Meeks III, J. Perez, and A. Ros, The geometry of minimal surfaces of finite genus I;
Curvature estimates and quasiperiodicity, J. Differential Geom. 66 (2004), 1–45.
[68] W. Meeks III, J. Perez, and A. Ros, The geometry of minimal surfaces of finite genus II;
Nonexistence of one limit end examples, Invent. Math. 158 (2004), no. 2, 323–341.
[69] W. Meeks III, J. Perez, and A. Ros, The geometry of minimal surfaces of finite genus III;
bounds on the topology and index of classical minimal surfaces, preprint.
[70] W. Meeks III and H. Rosenberg, The uniqueness of the helicoid and the asymptotic geometry
of properly embedded minimal surfaces with finite topology, Ann. of Math., 161 (2005), no. 2,
727–758.
[71] W. Meeks III and H. Rosenberg, The minimal lamination closure theorem, preprint.
[72] W. Meeks III, L. Simon, and S.T. Yau, Embedded minimal surfaces, exotic spheres and
manifolds with positive Ricci curvature, Ann. of Math. (2) 116 (1982) 621–659.
[73] W. Meeks III and M. Weber, Existence of bent helicoids and the regularity of the singular
set in the Colding-Minicozzi lamination theorem, in preparation.
[74] W. Meeks III and S.T. Yau, The classical Plateau problem and the topology of three dimen-
sional manifolds, Topology 21 (1982) 409–442.
[75] W.H. Meeks and S.T. Yau, Topology of three–dimensional manifolds and the embedding
problems in minimal surface theory, Ann. of Math. (2) 112 (1980), no. 3, 441–484.
[76] M.J. Micallef and J.D. Moore, Minimal two–spheres and the topology of manifolds with
positive curvature on totally isotropic two–planes, Ann. of Math. (2) 127 (1988) no. 1 199–
227.
[77] N. Nadirashvili, Hadamard’s and Calabi-Yau’s conjectures on negatively curved and minimal
surfaces, Invent. Math. 126 (1996) 457–465.
[78] N. Nadirashvili, An application of potential analysis to minimal surfaces, Mosc. Math. J. 1
(2001), no. 4, 601–604, 645.
[79] R. Osserman, A survey of minimal surfaces, Dover, 2nd. edition (1986).
[80] G. Perelman, Finite extinction time for the solutions to the Ricci flow on certain three–
manifolds, math.DG/0307245.
[81] J. Perez, Limits by rescalings of minimal surfaces: Minimal laminations, curvature decay
and local pictures, notes for the workshop ”Moduli Spaces of Properly Embedded Minimal
Surfaces”, American Institute of Mathematics, Palo Alto, California (2005).
[82] J.T. Pitts, Existence and regularity of minimal surfaces on Riemannian manifolds, Princeton
University Press, Princeton, N.J.; University of Tokyo Press, Tokyo (1981).
[83] J.T. Pitts and J.H. Rubinstein, Applications of minmax to minimal surfaces and the topology
of 3–manifolds, Miniconference on geometry and partial differential equations, Proceedings of
the CMA, Australia National University (1986).
[84] H. Rosenberg, Some recent developments in the theory of properly embedded minimal surfaces
in R3 , Seminare Bourbaki 1991/92, Asterisque No. 206 (1992) 463–535.
[85] H. Rosenberg, Some recent developments in the theory of minimal surfaces in 3-manifolds,
IMPA Mathematical Publications. 24th Brazilian Mathematics Colloquium, Instituto de
Matemtica Pura e Aplicada (IMPA), Rio de Janeiro, 2003.
[86] H. Rosenberg, A complete embedded minimal surface in R3 of bounded curvature is proper,
preprint.
 MINIMAL SUBMANIFOLDS 41

[87] J. Sacks and K. Uhlenbeck, The existence of minimal immersions of 2–spheres, Ann. of Math.
(2) 113 (1981) no. 1, 1–24.
[88] R. Schoen, Estimates for stable minimal surfaces in three–dimensional manifolds, In Semi-
nar on Minimal Submanifolds, Ann. of Math. Studies, vol. 103, Princeton University Press,
Princeton, N.J., (1983) 111–126.
[89] R. Schoen, Uniqueness, symmetry, and embeddedness of minimal surfaces, J. Differential
Geom. 18 (1983), no. 4, 791–809 (1984).
[90] R. Schoen and L. Simon, Regularity of simply connected surfaces with quasi-conformal Gauss
map, In Seminar on Minimal Submanifolds, Annals of Math. Studies, vol. 103, Princeton
University Press, Princeton, N.J., (1983) 127–145.
[91] R. Schoen, L. Simon, and S.T. Yau, Curvature estimates for minimal hypersurfaces, Acta
Math. 134 (1975) 275–288.
[92] Schoen, R. and Yau, S. T., On the proof of the positive mass conjecture in general relativity,
Comm. Math. Phys. 65 (1979), no. 1, 45–76.
[93] R. Schoen and S.T. Yau, Existence of incompressible minimal surfaces and the topology of
three dimensional manifolds with nonnegative scalar curvature, Ann. of Math. (2) 110 (1979)
127–142.
[94] L. Simon, Asymptotic behaviour of minimal graphs over exterior domains, Ann. Inst. H.
Poincar Anal. Non Linaire 4 (1987) 231–242.
[95] L. Simon, Remarks on curvature estimates for minimal hypersurfaces, Duke Math. J. 43
(1976) 545–553.
[96] L. Simon, Singularities of Geometric Variational Problems, In Nonlinear Partial Differential
Equations in Differential Geometry (R. Hardt and M. Wolf, Ed.), American Mathematical
Society, Providence (1996) 185–223.
[97] L. Simon, A strict maximum principle for area minimizing hypersurfaces, J. Differential
Geom. 26 (1987), no. 2, 327–335.
[98] J. Simons, Minimal varieties in Riemannian manifolds, Ann. of Math. (2) 88 (1968) 62–105.
[99] F. Smith, On the existence of embedded minimal 2–spheres in the 3–sphere, endowed with
an arbitrary Riemannian metric, supervisor L. Simon, University of Melbourne (1982).
[100] B. Solomon and B. White, A strong maximum principle for varifolds that are stationary
with respect to even parametric elliptic functionals, Indiana Univ. Math. J. 38 (1989), no. 3,
683–691.
[101] J. Taylor, Some mathematical challenges in materials science, Mathematical challenges of
the 21st century (Los Angeles, CA, 2000). Bull. Amer. Math. Soc. (N.S.) 40 (2003), no. 1,
69–87.
[102] D.W. Thompson, On growth and form, New ed., Cambridge Univ. Press, Cambridge (1942).
[103] M. Traizet, Adding handles to Riemann’s minimal surfaces, J. Inst. Math. Jussieu 1 (2002)
145–174.
[104] M. Weber and M. Wolf, Teichmüller theory and handle addition for minimal surfaces, Ann.
of Math. (2), 156 (2002) 713–795.
[105] H. Wente, Counterexample to a conjecture of H. Hopf, Pacific J. Math. 121 (1986), no. 1,
193–243.
[106] B. White, Evolution of curves and surfaces by mean curvature, Proceedings of the ICM
(2002).
[107] F. Xavier, Convex hulls of complete minimal surfaces, Math. Ann. 269 (1984) 179–182.
[108] S.D. Yang, A connected sum construction for complete minimal surfaces of finite total
curvature, Comm. Anal. Geom. 9 (2001), no. 1, 115–167.
[109] S.T. Yau, Nonlinear analysis in geometry, L’Eseignement Mathematique (2) 33 (1987) 109–
158.

Department of Mathematics, MIT, 77 Mass. Ave., Cambridge, MA 02139

Department of Mathematics, Johns Hopkins University, 3400 N. Charles St., Balti-

more, MD 21218
E-mail address: [email protected] and [email protected]