Graphs with Contributions to Fundamental Groups

Linfan MAO

Chinese Academy of Mathematics and System Science, Beijing 100190, P.R.China

Beijing Institute of Architectural Engineering, Beijing 100044, P.R.China

E-mail: maolinfan@163.com

A manifold M inherits a labeled n-dimensional graph GL [M] structure consisting of

its charts. This structure enables one to generalize spaces to combinatorial spaces
by combination and find fundamental groups for more spaces by generalizing the
Seifert-Van Kampen theorem to such a case of maybe not arcwise-connected, i.e.,
there are C1 , C2 ,· · · , Cm , m ≥ 1 arcwise-connected components in U ∩ V .

Contents

§1. What is a Combinatorial Space?

1.1 Topological space with fundamental groups
1.2 Smarandache multi-space with geometry
1.3 Combinatorial manifolds with topology

§2 Classical Seifert-Van Kampen theorem with Applications

§3. Dimensional Graphs

3.1 What is a dimensional graph?
3.2 Fundamental groups of dimensional graphs
§4. Generalized Seifert-Van Kampen Theorem
4.1 Topological space attached graphs
4.2 Generalized Seifert-Van Kampen theorem

§5. Fundamental Groups of Space

5.1 Determine fundamental groups of combinatorial spaces
5.2 Determine fundamental groups of manifolds

§6. Furthermore Discussions

1
Reported at Beijing Normal University, June 18, 2010.
§1. What is a Combinatorial Space?

1.1 Topological spaces with fundamental groups

Examples of Topological Space:

(1) Real numbers R. Complex numbers C.
(2) Euclidean space Rn , Spheres Sn for n ≥ 1;
(3) Product of spaces, such as S2 × Sn−2 for n ≥ 4.

Terminology Reviewing: Topological space, Hausdorff space, Open or closed

sets, Open neighborhood, Cover, Basis, Compact space, ..., in [1]-[3] following.

[1] John M.Lee, Introduction to Topological Manifolds, Springer-Verlag New York,

Inc., 2000.
[2] W.S.Massey, Algebraic Topology: An Introduction, Springer-Verlag, New York,
etc.(1977).
[3] Munkres J.R., Topology (2nd edition), Prentice Hall, Inc, 2000.

Definition 1.2 Let S be a topological space and I = [0, 1] ⊂ R. An arc a in S is

a continuous mapping a : I → S with initial point a(0) and end point a(1), and S
is called arcwise connected if every two points in S can be joined by an arc in S.
An arc a : I → S is a loop based at p if a(0) = a(1) = p ∈ S. A degenerated loop
e : I → x ∈ S, i.e., mapping each element in I to a point x, usually called a point
loop.

Example Let G be a planar 2-connected graph on R2 and S is a topological space

consisting of points on each e ∈ E(G). Then S is a arcwise connected space by
definition. For a circuit C in G, we choose any point p on C. Then C is a loop ep
in S based at p, such as those shown in Fig.1.1.
eu
u

e1 e2

e3
v w
e4

Fig.1.1

2
Definition 1.3 Let a and b be two arcs in a topological space S with a(1) = b(0).
A product mapping a · b of a with b is defined by
(
a(2t), if 0 ≤ t ≤ 21 ,
a · b(t) =
b(2t − 1), if 12 ≤ t ≤ 1

and an inverse mapping a = a(1 − t) by a.

Definition 1.4 Let S be a topological space and a, b : I → S two arcs with a(0) =
b(0) and a(1) = b(1). If there exists a continuous mapping

H :I ×I →S

such that H(t, 0) = a(t), H(t, 1) = b(t) for ∀t ∈ I, then a and b are said homotopic,
denoted by a ≃ b and H a homotopic mapping from a to b.
a(1)=b(1)
s
1 a
H- b
a(0)=b(0)
O 1 t

Fig.1.2

Theorem 1.1 The homotopic ≃ is an equivalent relation, i.e, all arcs homotopic
to an arc a is an equivalent arc class, denoted by [a].

Definition 1.5 For a topological space S and x0 ∈ S, let π1 (S, x0 ) be a set consisting
of equivalent classes of loops based at x0 . Define an operation ◦ in π1 (S, x0 ) by

[a] ◦ [b] = [a · b] and [a]−1 = [a−1 ].

Theorem 1.2 π1 (S, x0 ) is a group.

Example: (1) π1 (Rn , x0 ), x0 ∈ Rn and π1 (Sn , y0), y0 ∈ Sn is trivial for n ≥ 2;

= Z and π1 (T 2 , z0 ) ∼
(2) π1 (S, y0 ) ∼ = Z 2 , z0 ∈ T 2 .

1.2 Smarandache multi-space with geometry

Definition 1.6 A rule on a set Σ is a mapping

3
 | × Σ{z
Σ · · · × Σ} → Σ
n

for some integers n. A mathematical system is a pair (Σ; R), where Σ is a set
consisting mathematical objects, infinite or finite and R is a collection of rules on
Σ by logic providing all these resultants are still in Σ, i.e., elements in Σ is closed
under rules in R.

Definition 1.7 A rule in a mathematical system (Σ; R) is said to be Smarandachely

denied if it behaves in at least two different ways within the same set Σ, i.e., validated
and invalided, or only invalided but in multiple distinct ways.
A Smarandache system (Σ; R) is a mathematical system which has at least
one Smarandachely denied rule in R. Particularly, if all systems in (Σ; R) is a
geometrical space, such a Smarandache system is called Smarandache geometry.

[4] F.Smarandache, Paradoxist mathematics, Collected Papers, Vol.II, 5-28, Uni-

versity of Kishinev Press, 1997.
[5] H.Iseri, Smarandache Manifolds, American Research Press, Rehoboth, NM,
2002.
[6] L.F.Mao, Smarandache Multi-Space Theory, Hexis. Phoenix, USA 2006.

Example: Consider a geometry induced from a Euclidean planar geometry by

planar maps. Let a complete graph K4 be embedded in a Euclidean plane R2 ,
where points 1, 2 are elliptic, 3 is Euclidean but the point 4 is hyperbolic. Then all
lines in the field A do not intersect with L, but each line passing through the point
4 will intersect with the line L. Therefore, (M, µ) is a Smarandache geometry by
denial the axiom (E5) with these axioms (E5), (L5) and (R5).

3
L2 - 1
A
-
L1 4
z*
2 3
L

Fig.1.3

4
Definition 1.8 For an integer m ≥ 2, let (Σ1 ; R1 ), (Σ2 ; R2 ), · · ·, (Σm ; Rm ) be m
mathematical systems different two by two. A Smarandache multi-space is a pair
m
[ [m
e e e
(Σ; R) with Σ = e
Σi and R = Ri .
i=1 i=1

Definition 1.9 A combinatorial system CG is a union of mathematical systems

m
[ m
[
(Σ1 ; R1 ),(Σ2 ; R2 ), · · ·, (Σm ; Rm ) for an integer m, i.e., CG = ( Σi ; Ri ) with
i=1 i=1
an underlying connected graph structure G, where

V (G) = {Σ1 , Σ2 , · · · , Σm },
T
E(G) = { (Σi , Σj ) | Σi Σj 6= ∅, 1 ≤ i, j ≤ m}.

CC Conjecture(2005, MAO) A mathematical science can be reconstructed from

or made by combinatorialization.

(i) There is a combinatorial structure and finite rules for a classical math-
ematical system, which means one can make combinatorialization for all classical
mathematical subjects.
(ii) One can generalizes a classical mathematical system by this combinatorial
notion such that it is a particular case in this generalization.
(iii) One can make one combination of different branches in mathematics and
find new results after then.
(iv) One can understand our WORLD by this combinatorial notion, establish
combinatorial models for it and then find its behavior, and so on.

[7] L.F.Mao, Automorphism Groups of Maps, Surfaces and Smarandache Geome-

tries, American Research Press, 2005.
[8] L.F.Mao, Combinatorial speculation and combinatorial conjecture for mathe-
matics, International J.Math. Combin., Vol.1(2007), 1-19.

1.3 Combinatorial manifolds with topology

An n-dimensional manifold is a second countable Hausdorff space such that each

point has an open neighborhood homomorphic to a Euclidean space Rn of dimension
n, abbreviated to n-manifold.
Loosely speaking, a combinatorial manifold is a combination of finite manifolds,

5
such as those shown in Fig.1.4.

T2
3 2
B1 M T B1 B1

(a) (b)

Fig.1.4

Definition 1.10 A combinatorial Euclidean space is a combinatorial system CG of

Euclidean spaces Rn1 , Rn2 , · · ·, Rnm underlying a connected graph G defined by

V (G) = {Rn1 , Rn2 , · · · , Rnm },

T
E(G) = { (Rni , Rnj ) | Rni Rnj 6= ∅, 1 ≤ i, j ≤ m},

denoted by EG (n1 , · · · , nm ) and abbreviated to EG (r) if n1 = · · · = nm = r, which

enables us to view an Euclidean space Rn for n ≥ 4.

e 1 , · · · , nm ) is the combinatorial Eu-

Definition 1.11 A combinatorial fan-space R(n
clidean space EKm (n1 , · · · , nm ) of Rn1 , Rn2 , · · ·, Rnm such that

\ m
\
R ni R nj = R nk .
k=1

for any integers i, j, 1 ≤ i 6= j ≤ m.

Definition 1.12 For a given integer sequence 0 < n1 < n2 < · · · < nm , m ≥ 1,
a combinatorial manifold M f is a Hausdorff space such that for any point p ∈ M f,
f and a
there is a local chart (Up , ϕp ) of p, i.e., an open neighborhood Up of p in M
e 1 (p), n2 (p), · · · , ns(p) (p)), a combinatorial fan-space
homoeomorphism ϕp : Up → R(n
with

{n1 (p), n2 (p), · · · , ns(p) (p)} ⊆ {n1 , n2 , · · · , nm },

S
{n1 (p), n2 (p), · · · , ns(p) (p)} = {n1 , n2 , · · · , nm },
f
p∈M

f 1 , n2 , · · · , nm ) or M
denoted by M(n f on the context and

Ae = {(Up , ϕp )|p ∈ M
f(n1 , n2 , · · · , nm ))}

6
 f(n1 , n2 , · · · , nm ).
an atlas on M
f is finite if it is just combined by finite manifolds
A combinatorial manifold M
with an underlying combinatorial structure G without one manifold contained in the
union of others.

Question: Can we find the fundamental groups of finitely combinatorial manifolds?

§2 Classical Seifert-Van Kampen Theorem with Applications

Theorem 2.1(Seifert and Van-Kampen) Let X = U ∪ V with U, V open subsets

and let X, U, V , U ∩ V be non-empty arcwise-connected with x0 ∈ U ∩ V and H a
group. If there are homomorphisms

φ1 : π1 (U, x0 ) → H and φ2 : π1 (V, x0 ) → H

and
- π1(U, x0)
i1 φ1

j
?1 ?
π1 (U ∩ V, x0 ) - π1(X, x0) Φ -H
6j2 6
- π1(V, x0)
i2 φ2

with φ1 · i1 = φ2 · i2 , where i1 : π1 (U ∩ V, x0 ) → π1 (U, x0 ), i2 : π1 (U ∩ V, x0 ) →

π1 (V, x0 ), j1 : π1 (U, x0 ) → π1 (X, x0 ) and j2 : π1 (V, x0 ) → π1 (X, x0 ) are homomor-
phisms induced by inclusion mappings, then there exists a unique homomorphism
Φ : π1 (X, x0 ) → H such that Φ · j1 = φ1 and Φ · j2 = φ2 .

Theorem 2.2(Seifert and Van-Kampen theorem, classical version) Let spaces X, U, V

and x0 be in Theorem 2.1. If

j : π1 (U, x0 ) ∗ π1 (V, x0 ) → π1 (X, x0 )

is an extension homomorphism of j1 and j2 , then j is an epimorphism with kernel

7
Kerj generated by i−1
1 (g)i2 (g), g ∈ π1 (U ∩ V, x0 ), i.e.,

π1 (U, x0 ) ∗ π1 (V, x0 )
π1 (X, x0 ) ∼
=
−1 N .
i1 (g) · i2 (g)| g ∈ π1 (U ∩ V, x0 )

Corollary 2.1 Let spaces X, U, V and x0 be in Theorem 2.1. If U ∩ V is simply

connected, then
π1 (X) = π1 (U, x0 ) ∗ π1 (V, x0 ).
S
n
Application: Let Bn = Si1 be a bouquet shown in Fig.2.1 with vi ∈ Si1 ,
i=1
Wi = Si1 − {vi } for 1 ≤ i ≤ n and
[ [ [ [ [ [
U = S11 W2 · · · Wn and V = W1 S21 · · · Sn1 .

O v1 v2 vn

Fig.2.1
T
Then U V = S1.n , an arcwise-connected star. Whence,


π1 (Bn , O) = π1 (U, O) ∗ π1 (V, O) ∼
= π1 (Bn−1 , O) ∗ S11 .

By induction, we easily get that


π1 (Bn , O) = Si1 , 1 ≤ i ≤ n .

§3. Dimensional Graphs

3.1 What is a dimensional graph?

A topological graph T [G] of a graph G is a 1-dimensional graph in a topological

space.

Definition 3.1 A topological graph T [G] is a pair (X, X 0 ) of a Hausdorff space X

with its a subset X 0 such that

8
 (1) X 0 is discrete, closed subspaces of X;
(2) X − X 0 is a disjoint union of open subsets e1 , e2 , · · · , em , each of which is
homeomorphic to an open interval (0, 1);
(3) the boundary ei − ei of ei consists of one or two points. If ei − ei consists of
two points, then (ei , ei ) is homeomorphic to the pair ([0, 1], (0, 1)); if ei − ei consists
of one point, then (ei , ei ) is homeomorphic to the pair (S 1 , S 1 − {1});
(4) a subset A ⊂ Te[G] is open if and only if A ∩ ei is open for 1 ≤ i ≤ m.

Theorem 3.1([2]) Any tree is contractible.

Theorem 3.2([2]) Let Tspan be a spanning tree in the topological graph T [G], {eλ :
λ ∈ Λ} the set of edges of T [G] not in Tspan and αλ = Aλ eλ Bλ ∈ π(T [G], v0 ) a
loop associated with eλ = aλ bλ for ∀λ ∈ Λ, where v0 ∈ T [G] and Aλ , Bλ are unique
paths from v0 to aλ or from bλ to v0 in Tspan . Then

π(T [G], v0 ) = hαλ |λ ∈ Λi .

fn [G] is a combinatorial Euclidean space

Definition 3.2 An n-dimensional graph M
EG (n) of Rnµ , µ ∈ Λ underlying a combinatorial structure G such that

(1) V (G) is discrete consisting of B n , i.e., ∀v ∈ V (G) is an open ball Bvn ;

fn [G] \ V (M
(2) M fn [G]) is a disjoint union of open subsets e1 , e2 , · · · , em , each
of which is homeomorphic to an open ball B n ;
(3) the boundary ei − ei of ei consists of one or two B n and each pair (ei , ei ) is
n
homeomorphic to the pair (B , B n );
fn [G] is open if and only if A ∩ ei is open for 1 ≤ i ≤ m.
(4) a subset A ⊂ M

u u

- Bn -
v

v
v

Fig.2.2

9
3.2 Fundamental groups of dimensional graphs

fn [G].
Theorem 3.3 For any integer n ≥ 1, T0 [G] is a deformation retract of M

fn [G] = T0 [G] is itself a topological graph. So

Sketch of Proof If n = 1, then M
we assume n ≥ 2.
For n ≥ 2, letf (x, t) = (1 − t)x + tx0 be a mapping f : Mfn [G] × I → M fn [G]
for ∀x ∈ Mfn [G]1 , t ∈ I, where x0 = Ov if x ∈ Bvn , and x0 = p(x) if x ∈ ei , where
p : uv → euv a projection for 1 ≤ i ≤ m, such as those shown in Fig.2.3.

p
s ? ? ? ? ? ?
p
Ov
* 6I
Ou Ov
p
6 6 6p 6 6

Bvn uv

Fig.2.3

Then f is such a deformation retract. 

§4. Generalized Seifert-Van Kampen Theorem

4.1 Topological space attached graphs

Definition 4.1 A topological space X attached with a graph G is a space X ⊙ G

such that

\
X G 6= ∅, G 6⊂ X
T
and there are semi-edges e+ ∈ (X G) \ G, e+ ∈ G \ X.

An example for X ⊙ G can be found in Fig.4.1.

10
 u

v
X G

X ⊙G

Fig.4.1

Theorem 4.1 Let X be arc-connected space, G a graph and H the subgraph X ∩ G

in X ⊙ G. Then for x0 ∈ X ∩ G,

π1 (X, x0 ) ∗ π1 (G, x0 )
π1 (X ⊙ G, x0 ) ∼
=
−1 N ,
i1 (αeλ )i2 (αeλ )| eλ ∈ E(H) \ Tspan )
where i1 : π1 (H, x0 ) → X, i2 : π1 (H, x0 ) → G are homomorphisms induced by
inclusion mappings, Tspan is a spanning tree in H, αλ = Aλ eλ Bλ is a loop associated
with an edge eλ = aλ bλ ∈ H \ Tspan , x0 ∈ G and Aλ , Bλ are unique paths from x0 to
aλ or from bλ to x0 in Tspan .

Sketch of Proof Let U = X and V = G. Applying the Seifert-Van Kampen

theorem, we get that

π1 (X, x0 ) ∗ π1 (G, x0 )
π1 (X ⊙ G, x0 ) ∼
=
−1 ,
i1 (g)i2(g)| g ∈ π1 (X ∩ G, x0 )
Applying Theorem 3.2, We finally get the following conclusion,

π1 (X, x0 ) ∗ π1 (G, x0 )
π1 (X ⊙ G, x0 ) ∼
=
−1 N
i1 (αeλ )i2 (αeλ )| eλ ∈ E(H) \ Tspan )


Corollary 4.1 Let X be arc-connected space, G a graph. If X ∩ G in X ⊙ G is a

tree, then

π1 (X ⊙ G, x0 ) ∼
= π1 (X, x0 ) ∗ π1 (G, x0 ).

11
Particularly, if G is graphs shown in Fig.2.2 following

x0 x1 xm
x0 x1 x2 xm

T T
Bm Sm

Fig.4.2

and X ∩ G = K1,m , Then

T
π1 (X ⊙ Bm , x0 ) ∼
= π1 (X, x0 ) ∗ hLi |1 ≤ i ≤ mi ,
T
where Li is the loop of parallel edges (x0 , xi ) in Bm for 1 ≤ i ≤ m − 1 and

T
π1 (X ⊙ Sm , x0 ) ∼
= π1 (X, x0 ).

Theorem 4.2 Let Xm ⊙ G be a topological space consisting of m arcwise-connected

spaces X1 , X2 , · · · , Xm , Xi ∩ Xj = ∅ for 1 ≤ i, j ≤ m attached with a graph G,
V (G) = {x0 , x1 , · · · , xl−1 }, m ≤ l such that Xi ∩ G = {xi } for 0 ≤ i ≤ l − 1. Then

m
!
Y
π1 (Xm ⊙ G, x0 ) ∼
= π1 (Xi∗ , x0 ) ∗ π1 (G, x0 )
i=1
m
!
Y
∼
= π1 (Xi , xi ) ∗ π1 (G, x0 ),
i=1
S
where Xi∗ = Xi (x0 , xi ) with Xi ∩ (x0 , xi ) = {xi } for (x0 , xi ) ∈ E(G), integers
1 ≤ i ≤ m.

Sketch of Proof The proof is by induction on m with Theorem 4.1 and the
Seifert-Van Kampen theorem. 

T T
Corollary 4.2 Let G be the graph Bm or a star Sm . Then

m
!
Y
T
π1 (Xm ⊙ Bm , x0 ) ∼
= π1 (Xi∗ , x0 ) T
∗ π1 (Bm , x0 )
i=1

12
 m
!
Y
∼
= π1 (Xi , xi−1 ) ∗ hLi |1 ≤ i ≤ mi ,
i=1

T
where Li is the loop of parallel edges (x0 , xi ) in Bm for integers 1 ≤ i ≤ m and

m
Y
π1 (Xm ⊙ T
Sm , x0 ) ∼
= π1 (Xi∗ , x0 ).
i=1

Corollary 4.3 Let X = Xm ⊙ G be a topological space with a simply-connected

space Xi for any integer i, 1 ≤ i ≤ m and x0 ∈ X ∩ G. Then we know that

π1 (X, x0 ) ∼
= π1 (G, x0 ).

4.2 Generalized Seifert-Van Kampen theorem

Theorem 4.3 Let X = U ∪ V , U, V ⊂ X be open subsets and X, U, V arcwise-

connected and let C1 , C2 , · · · , Cm be arcwise-connected components in U ∩ V for an
integer m ≥ 1, xi−1 ∈ Ci , b(x0 , xi−1 ) ⊂ V an arc : I → X with b(0) = x0 , b(1) = xi−1
S
and b(x0 , xi−1 ) ∩ U = {x0 , xi−1 }, CiE = Ci b(x0 , xi−1 ) for any integer i, 1 ≤ i ≤ m,
H a group and there are homomorphisms

[
φi1 : π1 (U b(x0 , xi−1 ), x0 ) → H, φi2 : π1 (V, x0 ) → H

such that
ii1 - π1(U ∪ b(x0, xi−1, x0)) φi1

?ji1 ?
π1 (CiE , x0 ) - π1 (X, x0 )
Φ - H
6ji2 6
- π1 (V ), x0 )
ii2 φi2

with φi1 · ii1 = φi2 · ii2 , where ii1 : π1 (CiE , x0 ) → π1 (U ∪ b(x0 , xi−1 ), x0 ), ii2 :
π1 (CiE , x0 ) → π1 (V, x0 ) and ji1 : π1 (U∪b(x0 , xi−1 , x0 )) → π1 (X, x0 ), ji2 : π1 (V, x0 )) →
π1 (X, x0 ) are homomorphisms induced by inclusion mappings, then there exists a

13
unique homomorphism Φ : π1 (X, x0 ) → H such that Φ · ji1 = φi1 and Φ · ji2 = φi2
for integers 1 ≤ i ≤ m.
S
Sketch of Proof Define U E = U { b(x0 , xi ) | 1 ≤ i ≤ m−1}. Then we get that
E E
X = U ∪ V , U , V ⊂ X are still opened with an arcwise-connected intersection
U E ∩ V = Xm ⊙ Sm T T
, where Sm is a graph formed by arcs b(x0 , xi−1 ), 1 ≤ i ≤ m.
S
m T
Notice that Xm ⊙ SmT = CiE and CiE CjE = {x0 }. Therefore, we get that
i=1
m
O
T
π1 (Xm ⊙ Sm , x0 ) = π1 (CiE , x0 ).
i=1

This fact enables us knowing that there is a unique m-tuple (h1 , h2 , · · · , hm ), hi ∈

π1 (CiE , xi−1 ), 1 ≤ i ≤ m such that
m
Y
I = hi
i=1
T
for ∀I ∈ π1 (Xm ⊙ Sm , x0 ) and inclusion maps

iE T E
1 : π1 (Xm ⊙ Sm , x0 ) → π1 (U , x0 ),

iE T
2 : π1 (Xm ⊙ Sm , x0 ) → π1 (V, x0 ),

j1E : π1 (U E , x0 ) → π1 (X, x0 ), j2E : π1 (V, x0 ) → π1 (X, x0 )

with iE E E E
1 |π1 (CiE ,x0 ) = ii1 , i2 |π1 (CiE ,x0 ) = ii2 , j1 |π1 (U ∪b(x0 ,xi−1 ,x0 )) = ji1 and j2 |π1 (V,x0 ) = ji2
for integers 1 ≤ i ≤ m.
Define φE E
1 and φ2 by
m
Y m
Y
φE
1 (I ) = φi1 (ii1 (hi )), φE
2 (I ) = φi2 (ii2 (hi )).
i=1 i=1

Then the following diagram

iE
1 - π1(U E , x0) φE
1

j1E
? ?
π1 (U E ∩ V, x0 ) - π1(X, x0) Φ - H
6j E 6
2

E
- π1(V, x0)
i2 φE
2

14
is commutative. Applying Theorem 2.1, we get the conclusion. 

Theorem 4.2 Let X, U, V , CiE , b(x0 , xi−1 ) be arcwise-connected spaces for any
S
integer i, 1 ≤ i ≤ m as in Theorem 3.1, U E = U { b(x0 , xi ) | 1 ≤ i ≤ m − 1} and
T
Bm a graph formed by arcs a(x0 , xi−1 ), b(x0 , xi−1 ), 1 ≤ i ≤ m, where a(x0 , xi−1 ) ⊂ U
is an arc : I → X with a(0) = x0 , a(1) = xi−1 and a(x0 , xi−1 ) ∩ V = {x0 , xi−1 }.
Then

T
π1 (U, x0 ) ∗ π1 (V, x0 ) ∗ π1 (Bm , x0 )
π1 (X, x0 ) ∼
= N ,
E −1
Qm
E
(i1 ) (g) · i2 (g)| g ∈ π1 (Ci , x0 )
i=1

where iE E E E E
1 : π1 (U ∩ V, x0 ) → π1 (U , x0 ) and i2 : π1 (U ∩ V, x0 ) → π1 (V, x0 ) are
homomorphisms induced by inclusion mappings.

Sketch of Proof By the proof of Theorem 4.1 we have known that there are
homomorphisms φE E E E E E
1 and φ2 such that φ1 · i1 = φ2 · i2 . Applying Theorem 2.2, we
get that

π1 (U E , x0 ) ∗ π1 (V, x0 )
π1 (X, x0 ) ∼
= N
.
h(iE E
1 ) (I ) · i2 (I )|I ∈ π1 (U ∩ V, x0 )i
−1 E

Notice that U E ∩ V E = Xm ⊙ Sm
T
and

π1 (U E , x0 ) ∼ T
= π1 (U, x0 ) ∗ π1 (Bm , x0 ).

We finally get that

T
π1 (U, x0 ) ∗ π1 (V, x0 ) ∗ π1 (Bm , x0 )
π1 (X, x0 ) ∼
= N . 
Qm
(iE E
1 ) (g) · i2 (g)| g ∈
−1 π1 (CiE , x0 )
i=1

Theorem 4.3 Let X, U, V , C1 , C2 , · · · , Cm be arcwise-connected spaces, b(x0 , xi−1 )

S
arcs for any integer i, 1 ≤ i ≤ m as in Theorem 3.1, U E = U { b(x0 , xi−1 ) | 1 ≤
T
i ≤ m} and Bm a graph formed by arcs a(x0 , xi−1 ), b(x0 , xi−1 ), 1 ≤ i ≤ m. Then

T
π1 (U, x0 ) ∗ π1 (V, x0 ) ∗ π1 (Bm , x0 )
π1 (X, x0 ) ∼
= N ,
E −1 E
Qm
E
(i1 ) (g) · i2 (g)| g ∈ π1 (Ci , x0 )
i=1

15
where iE E E E E
1 : π1 (U ∩ V, x0 ) → π1 (U , x0 ) and i2 : π1 (U ∩ V, x0 ) → π1 (V, x0 ) are
homomorphisms induced by inclusion mappings.

Corollary 3.4 Let X = U ∪ V , U, V ⊂ X be open subsets and X, U, V and U ∩ V

arcwise-connected. Then

π1 (U, x0 ) ∗ π1 (V, x0 )
π1 (X, x0 ) ∼
=
−1 N ,
i1 (g) · i2 (g)| g ∈ π1 (U ∩ V, x0 )
where i1 : π1 (U ∩ V, x0 ) → π1 (U, x0 ) and i2 : π1 (U ∩ V, x0 ) → π1 (V, x0 ) are homo-
morphisms induced by inclusion mappings.

Corollary 3.5 Let X, U, V , Ci , a(x0 , xi ), b(x0 , xi ) for integers i, 1 ≤ i ≤ m be as

in Theorem 3.1. If each Ci is simply-connected, then

π1 (X, x0 ) ∼ T
= π1 (U, x0 ) ∗ π1 (V, x0 ) ∗ π1 (Bm , x0 ).

Corollary 3.6 Let X, U, V , Ci , a(x0 , xi ), b(x0 , xi ) for integers i, 1 ≤ i ≤ m be as

in Theorem 3.1. If V is simply-connected, then

T
π1 (U, x0 ) ∗ π1 (Bm , x0 )
π1 (X, x0 ) ∼
= N ,
E −1 E
Q
m
E
(i1 ) (g) · i2 (g)| g ∈ π1 (Ci , x0 )
i=1

where iE E E E E
1 : π1 (U ∩ V, x0 ) → π1 (U , x0 ) and i2 : π1 (U ∩ V, x0 ) → π1 (V, x0 ) are
homomorphisms induced by inclusion mappings.

§5. Fundamental Groups of Spaces

5.1 Determine fundamental groups of combinatorial spaces

Definition 5.1 Let Mf be a combinatorial manifold underlying a graph G[M

f]. An
f] is defined by
edge-induced graph Gθ [M

f]) = {xM , xM ′ , x1 , x2 , · · · , xµ(M,M ′ ) | f or ∀(M, M ′ ) ∈ E(G[M

V (Gθ [M f])},

f = {(xM , xM ′ ), (xM , xi ), (xM ′ , xi )| 1 ≤ i ≤ µ(M, M ′ )},

E(Gθ [M])

where µ(M, M ′ ) is called the edge-index of (M, M ′ ) with µ(M, M ′ ) + 1 equal to the
number of arcwise-connected components in M ∩ M ′ .

16
 f] of a combinatorial manifold M
By definition, Gθ [M f is gotten by replacing each
f] by a subgraph T B T
edge (M, M ′ ) in G[M ′ shown in Fig.5.1 with xM = M and
µ(M,M )
′
xM′ =M.
xM ′

x1 x2 xµ(M,M ′ )

xM

Fig.5.1

f be a finitely combinatorial manifold. Then

Theorem 5.1 Let M

!
Q
π1 (M) f])
∗ π1 (Gθ [M
f])
M ∈V (G[M
f ∼
π1 (M) =* +N ,
Q T
(iE −1 E
1 ) (g) · i2 (g)| g ∈ π1 (M1 M2 )
f
(M1 ,M2 )∈E(G[M])

where iE E
1 and i2 are homomorphisms induced by inclusion mappings iM : π1 (M ∩
M ′ ) → π1 (M), iM ′ : π1 (M ∩ M ′ ) → π1 (M ′ ) such as those shown in the following
diagram:
iM - π1(M) jM

?
π1 (M ∩ M ′ )
ΦM M ′ -
....................................................... f
π1 (M)
6
- π1(M ′)
iM ′ jM ′

f]).
for ∀(M, M ′ ) ∈ E(G[M

f = 1. Notice that Gθ [M
Sketch of Proof This result is obvious for |G[M]| f] =
T
Bµ(M,M f
′ )+1 if V (G[M ]) = {M, M }. Whence, it is an immediately conclusion of
′

Theorem 4.2 for |G[M f]| = 2.

17
 f \ (M \ M
Let U = M f) and V = M. By definition, they are both opened.
Applying Theorem 3.2, we get that

f − M) ∗ π1 (M) ∗ π1 (B T )
π1 (M
f ∼
π1 (M) = m
N ,
E −1 E
Q
m
(i1 ) (g) · i2 (g)| g ∈ π1 (Ci )
i=1

f − M) and
where Ci is an arcwise-connected component in M ∩ (M

X
m= µ(M, M ′ ).
f])
(M,M ′ )∈E(G[M

Applying the induction assumption, we get that

!
Q f])
π1 (M) ∗ π1 (Gθ [M
f])
M ∈V (G[M
f) ∼
π1 (M =* +N . 
Q T
(iE −1 E
1 ) (g) · i2 (g)| g ∈ π1 (M1 M2 )
f])
(M1 ,M2 )∈E(G[M

f be a finitely combinatorial manifold. If for ∀(M1 , M2 ) ∈

Corollary 5.1([9],[10]) Let M
f]), M1 ∩ M2 is simply connected, then
E(GL [M
 
M M
f) ∼
π1 (M = π1 (M) f]).
π1 (G[M
f])
M ∈V (G[M

[9] L.F.Mao, Geometrical theory on combinatorial manifolds, JP J.Geometry and

Topology, Vol.7, No.1(2007),65-114.
[10] L.F.Mao, Combinatorial Geometry with Applications to Field Theory, Info-
Quest, USA, 2009.

5.2 Determine fundamental groups of manifolds

f to be a chart (Uλ , ϕλ ) with ϕλ : Uλ → Rn for λ ∈ Λ

If we choose M ∈ V (G[M])
in Theorem 5.1, i.e., an n-manifold, we get the fundamental group of n-manifold
following.

18
Theorem 5.2 Let M be a compact n-manifold with charts {(Uλ , ϕλ )| ϕλ : Uλ →
Rn , λ ∈ Λ)}. Then

π1 (Gθ [M])
π1 (M) ∼
=* +N ,
Q
(iE E
1 ) (g) · i2 (g)| g ∈
−1 π1 (Uµ ∩ Uν )
(Uµ ,Uν )∈E(G[M ])

where iE E
1 and i2 are homomorphisms induced by inclusion mappings iUµ : π1 (Uµ ∩
Uν ) → π1 (Uµ ), iUν : π1 (Uµ ∩ Uν ) → π1 (Uν ), µ, ν ∈ Λ.

Corollary 5.2 Let M be a simply connected manifold with charts {(Uλ , ϕλ )| ϕλ :

Uλ → Rn , λ ∈ Λ)}. Then Gθ [M] = G[M] is a tree.

Corollary 5.3 Let M be a compact n-manifold with charts {(Uλ , ϕλ )| ϕλ : Uλ →

Rn , λ ∈ Λ)}. If Uµ ∩ Uν is simply connected for ∀µ, ν ∈ Λ, then

π1 (M) ∼
= π1 (G[M]).

§6. Furthermore Discussions

If objects considered are smooth, then we can establish a differential theory on com-
binatorial manifold, i.e., combinatorially differential geometry (see [10] for detail),
which can be used to characterizing the behavior of multi-spaces in Universe. More
such applications can be found in references [11]-[12].

[11] L.F.Mao, Curvature equations on combinatorial manifolds with applications to

theoretical physics, International J.Math. Combin., Vol.1(2008), Vol.1, 1-25.
[12] L.F.Mao, Relativity in combinatorial gravitational fields, Progress in Physics,
Vol.3 (2010), 33-44.

19