MATHEMATICAL COMBINATORICS (INTERNATIONAL BOOK SERIES), Vol. 2/2015

ISBN 978-1-59973-349-4
VOLUME 2, 2015
MATHEMATICAL COMBINATORICS
(INTERNATIONAL BOOK SERIES)
Edited By Linfan MAO
EDITED BY
THE MADIS OF CHINESE ACADEMY OF SCIENCES AND
ACADEMY OF MATHEMATICAL COMBINATORICS & APPLICATIONS
June, 2015
Vol.2, 2015 ISBN 978-1-59973-349-4
MATHEMATICAL COMBINATORICS
(INTERNATIONAL BOOK SERIES)
Edited By Linfan MAO
The Madis of Chinese Academy of Sciences and

Academy of Mathematical Combinatorics & Applications
June, 2015
Aims and Scope: The Mathematical Combinatorics (International Book Series) is a
fully refereed international book series, quarterly comprising 100-150 pages approx. per volume,
which publishes original research papers and survey articles in all aspects of Smarandache
multi-spaces, Smarandache geometries, mathematical combinatorics, non-euclidean geometry
and topology and their applications to other sciences. Topics in detail to be covered are:
Smarandache multi-spaces with applications to other sciences, such as those of algebraic
multi-systems, multi-metric spaces, , etc.. Smarandache geometries;
Topological graphs; Algebraic graphs; Random graphs; Combinatorial maps; Graph and
map enumeration; Combinatorial designs; Combinatorial enumeration;
Differential Geometry; Geometry on manifolds; Low Dimensional Topology; Differential
Topology; Topology of Manifolds; Geometrical aspects of Mathematical Physics and Relations
with Manifold Topology;
Applications of Smarandache multi-spaces to theoretical physics; Applications of Combi-
natorics to mathematics and theoretical physics; Mathematical theory on gravitational fields;
Mathematical theory on parallel universes; Other applications of Smarandache multi-space and
combinatorics.
Generally, papers on mathematics with its applications not including in above topics are
also welcome.
It is also available from the below international databases:
Serials Group/Editorial Department of EBSCO Publishing
10 Estes St. Ipswich, MA 01938-2106, USA
Tel.: (978) 356-6500, Ext. 2262 Fax: (978) 356-9371
http://www.ebsco.com/home/printsubs/priceproj.asp
and
Gale Directory of Publications and Broadcast Media, Gale, a part of Cengage Learning
27500 Drake Rd. Farmington Hills, MI 48331-3535, USA
Tel.: (248) 699-4253, ext. 1326; 1-800-347-GALE Fax: (248) 699-8075
http://www.gale.com
Indexing and Reviews: Mathematical Reviews (USA), Zentralblatt Math (Germany), Refer-
ativnyi Zhurnal (Russia), Mathematika (Russia), Directory of Open Access (DoAJ), Interna-
tional Statistical Institute (ISI), International Scientific Indexing (ISI, impact factor 1.416),
Institute for Scientific Information (PA, USA), Library of Congress Subject Headings (USA).
Subscription A subscription can be ordered by an email directly to
Linfan Mao
The Editor-in-Chief of International Journal of Mathematical Combinatorics
Chinese Academy of Mathematics and System Science
Beijing, 100190, P.R.China
Email: maolinfan@163.com
Price: US$48.00
Editorial Board (3nd)
Editor-in-Chief
Baizhou He
Linfan MAO Beijing University of Civil Engineering and
Chinese Academy of Mathematics and System Architecture, P.R.China
Science, P.R.China Email: hebaizhou@bucea.edu.cn
and
Xiaodong Hu
Academy of Mathematical Combinatorics &
Chinese Academy of Mathematics and System
Applications, USA
Science, P.R.China
Email: maolinfan@163.com
Email: xdhu@amss.ac.cn
Deputy Editor-in-Chief Yuanqiu Huang
Hunan Normal University, P.R.China
Guohua Song Email: hyqq@public.cs.hn.cn
Beijing University of Civil Engineering and
H.Iseri
Architecture, P.R.China
Mansfield University, USA
Email: songguohua@bucea.edu.cn
Email: hiseri@mnsfld.edu
Editors Xueliang Li
Nankai University, P.R.China
Said Broumi Email: lxl@nankai.edu.cn
Hassan II University Mohammedia Guodong Liu
Hay El Baraka Ben Msik Casablanca Huizhou University
B.P.7951 Morocco Email: lgd@hzu.edu.cn
Junliang Cai W.B.Vasantha Kandasamy
Beijing Normal University, P.R.China Indian Institute of Technology, India
Email: caijunliang@bnu.edu.cn Email: vasantha@iitm.ac.in
Yanxun Chang Ion Patrascu
Beijing Jiaotong University, P.R.China Fratii Buzesti National College
Email: yxchang@center.njtu.edu.cn Craiova Romania
Jingan Cui Han Ren
Beijing University of Civil Engineering and East China Normal University, P.R.China
Architecture, P.R.China Email: hren@math.ecnu.edu.cn
Email: cuijingan@bucea.edu.cn
Ovidiu-Ilie Sandru
Shaofei Du Politechnica University of Bucharest
Capital Normal University, P.R.China Romania
Email: dushf@mail.cnu.edu.cn
ii International Journal of Mathematical Combinatorics
Mingyao Xu
Peking University, P.R.China
Email: xumy@math.pku.edu.cn
Guiying Yan
Chinese Academy of Mathematics and System
Science, P.R.China
Email: yanguiying@yahoo.com
Y. Zhang
Department of Computer Science
Georgia State University, Atlanta, USA
Famous Words:
A man is not old as long as he is seeking something. A man is not old until
regrets take the place of dreams.
By J.Barrymore, an American actor.

Math.Combin.Book Ser. Vol.2(2015), 1-31
Mathematics After CC Conjecture
Combinatorial Notions and Achievements
Linfan MAO
Chinese Academy of Mathematics and System Science, Beijing 100190, P.R.China
Academy of Mathematical Combinatorics with Applications, Colorado, USA
E-mail: maolinfan@163.com
Abstract: As a powerful technique for holding relations in things, combinatorics has expe-
rienced rapidly development in the past century, particularly, enumeration of configurations,
combinatorial design and graph theory. However, the main objective for mathematics is
to bring about a quantitative analysis for other sciences, which implies a natural question
on combinatorics. Thus, how combinatorics can contributes to other mathematical sciences,
not just in discrete mathematics, but metric mathematics and physics? After a long time
speculation, I brought the CC conjecture for advancing mathematics by combinatorics, i.e.,
any mathematical science can be reconstructed from or made by combinatorialization in my
postdoctoral report for Chinese Academy of Sciences in 2005, and reported it at a few aca-
demic conferences in China. After then, my surveying paper Combinatorial Speculation and
Combinatorial Conjecture for Mathematics published in the first issue of International Jour-
nal of Mathematical Combinatorics, 2007. Clearly, CC conjecture is in fact a combinatorial
notion and holds by a philosophical law, i.e., all things are inherently related, not isolated
but it can greatly promote the developing of mathematical sciences. The main purpose
of this report is to survey the roles of CC conjecture in developing mathematical sciences
with notions, such as those of its contribution to algebra, topology, Euclidean geometry and
differential geometry, non-solvable differential equations or classical mathematical systems
with contradictions to mathematics, quantum fields after it appeared 10 years ago. All of
these show the importance of combinatorics to mathematical sciences in the past and future.
Key Words: CC conjecture, Smarandache system, GL -system, non-solvable system of

equations, combinatorial manifold, geometry, quantum field.
AMS(2010): 03C05,05C15,51D20,51H20,51P05,83C05,83E50.
1. Introduction
There are many techniques in combinatorics, particularly, the enumeration and counting with
graph, a visible, also an abstract model on relations of things in the world. Among them,
1 Reported at the International Conference on Combinatorics, Graph Theory, Topology and Geometry, Jan-
uary 29-31, 2015, Shanghai, P.R.China.

2 Received October 20, 2014, Accepted May 8, 2015.
2 Linfan MAO
the most interested is the graph. A graph G is a 3-tuple (V, E, I) with finite sets V, E and a
mapping I : E V V , and simple if it is without loops and multiple edges, denoted by
(V ; E) for convenience. All elements v in V , e in E are said respectively vertices and edges.
A graph with given properties are particularly interested. For example, a path Pn in a graph
G is an alternating sequence of vertices and edges u1 , e1 , u2 , e2 , , en , un1 , ei = (ui , ui+1 ) with
distinct vertices for an integer n 1, and if u1 = un+1 , it is called a circuit or cycle Cn . For
example, v1 v2 v3 v4 and v1 v2 v3 v4 v1 are respective path and circuit in Fig.1. A graph G is
connected if for u, v V (G), there are paths with end vertices u and v in G.
A complete graph Kn = (Vc , Ec ; Ic ) is a simple graph with Vc = {v1 , v2 , , vn }, Ec =
{eij , 1 i, j n, i 6= j} and Ic (eij ) = (vi , vj ), or simply by a pair (V, E) with V =
{v1 , v2 , , vn } and E = {vi vj , 1 i, j n, i 6= j}.
A simple graph G = (V, E) is r-partite for an integer r 1 if it is possible to partition V into
r subsets V1 , V2 , , Vr such that for e(u, v) E, there are integers i 6= j, 1 i, j r such that
u Vi and v Vj . If there is an edge eij E for vi Vi , vj Vj , where 1 i, j r, i 6= j,
then, G is called a complete r-partite graph, denoted by G = K(|V1 |, |V2 |, , |Vr |). Thus a
complete graph is nothing else but a complete 1-partite graph. For example, the bipartite graph
K(4, 4) and the complete graph K6 are shown in Fig.1.
K(4, 4) K6
Fig.1
Notice that a few edges in Fig.1 have intersections besides end vertices. Contrast to this
case, a planar graph can be realized on a Euclidean plane R2 by letting points p(v) R2 for
vertices v V with p(vi ) 6= p(vj ) if vi 6= vj , and letting curve C(vi , vj ) R2 connecting points
p(vi ) and p(vj ) for edges (vi , vj ) E(G), such as those shown in Fig.2.
e1 e2
v1 e5 v2
e6
e9 e10 e7
v4 e8 v3
e4 e3
Fig.2
Generally, let E be a topological space. A graph G is said to be embeddable into E ([32])

Mathematics After CC Conjecture 3
if there is a 1 1 continuous mapping f : G E with f (p) 6= f (q) if p 6= q for p, q G, i.e.,

edges only intersect at vertices in E . Such embedded graphs are called topological graphs.
There is a well-known result on embedding of graphs without loops and multiple edges in
n
R for n 3 ([32]), i.e., there always exists such an embedding of G that all edges are straight
segments in Rn , which enables us turn to characterize embeddings of graphs on R2 and its
generalization, 2-manifolds or surfaces ([3]).
However, all these embeddings of G are established on an assumption that each vertex
of G is mapped exactly into one point of E in combinatorics for simplicity. If we put off this
assumption, what will happens? Are these resultants important for understanding the world?
The answer is certainly YES because this will enables us to pullback more characters of things,
characterize more precisely and then hold the truly faces of things in the world.
All of us know an objective law in philosophy, namely, the integral always consists of its
parts and all of them are inherently related, not isolated. This idea implies that every thing in
the world is nothing else but a union of sub-things underlying a graph embedded in space of
the world.
{c}
1 2
{a, c} {c, e} {d, e}
{e}
3 4
Fig.3
Formally, we introduce some conceptions following.
Definition 1.1([30]-[31], [12]) Let (1 ; R1 ), (2 ; R2 ), , (m ; Rm ) be m mathematical

S
m
systems, different two by two. A Smarandache multisystem e is a union i with rules
i=1

e = S Ri on ,
m
R e denoted by ;e Re .
i=1

Definition 1.2([11]-[13]) For any integer m 1, let ; e R e be a Smarandache multisystem
consisting of m mathematical
h i (1 ; R1 ), (2 ; R2 ), , (m ; Rm ). An inherited topolog-
systems
L e e e e
ical structure G ; R of ; R is a topological vertex-edge labeled graph defined following:
h i
V GL ; e Re = {1 , 2 , , m },
h i
E GL ; e Re = {(i , j )|i T j 6= , 1 i 6= j m} with labeling
T
L : i L(i ) = i and L : (i , j ) L(i , j ) = i j
h i
for integers 1 i 6= j m, also denoted by GL ; e R
e for ; e R
e .
4 Linfan MAO
For example, let 1 = {a, b, c}, 2 = {c, d, e}, 3 = {a, c, e}, 4 = {d, e, f }and Ri =
for integers 1 i 4, i.e., all these system are sets. Then the multispace ; e Re with
h i
e = S i = {a, b, c, d, e, f } and Re = underlying a topological graph GL ;
4
e Re shown
i=1
in Fig.3. Combinatorially, the Smarandache multisystems can be classified by their inherited
topological structures, i.e., isomorphic labeled graphs following.
Definition 1.3 ([13]) Let
m m
! n n
!
[ (1)
[ (1)
[ (2)
[ (2)
G1 L1 = i ; Ri and G2 L2 = i ; Ri .
i=1 i=1 i=1 i=1
be two Smarandache multisystems underlying topological graphs G1 and G2 , respectively. They

Sm
(1) Sn
(2)
are isomorphic if there is a bijection : G1 L1 G2 L2 with : i i and
i=1 i=1
S
m
(1) S
n
(2)
: Ri Ri such that
i=1 i=1

(1) (1)
|i aRi b = |i (a)|i Ri |i (b)
(1)
for a, b i , 1 i m, where |i denotes the constraint of on (i , Ri ).
Consequently, the previous discussion implies that
Every thing in the world is nothing else but a topological graph GL in space of the world,
and two things are similar if they are isomorphic.
After speculation over a long time, I presented the CC conjecture on mathematical sciences
in the final chapter of my post-doctoral report for Chinese Academy of Sciences in 2005 ([9],[10]),
and then reported at The 2nd Conference on Combinatorics and Graph Theory of China in 2006,
which is in fact an inverse of the understand of things in the world.
CC Conjecture([9-10],[14]) Any mathematical science can be reconstructed from or made by

combinatorialization.
Certainly, this conjecture is true in philosophy. It is in fact a combinatorial notion for

developing mathematical sciences following.
Notion 1.1 Finds the combinatorial structure, particularly, selects finite combinatorial rulers
to reconstruct or make a generalization for a classical mathematical science.
This notion appeared even in classical mathematics. For examples, Hilbert axiom system
for Euclidean geometry, complexes in algebraic topology, particularly, 2-cell embeddings of
graphs on surface are essentially the combinatorialization for Euclidean geometry, topological
spaces and surfaces, respectively.
Notion 1.2 Combine different mathematical sciences and establish new enveloping theory on
topological graphs, with classical theory being a special one, and this combinatorial process will
never end until it has been done for all mathematical sciences.
A few fields can be also found in classical mathematics on this notion, for instance the
topological groups, which is in fact a multi-space of topological space with groups, and similarly,
the Lie groups, a multi-space of manifold with that of diffeomorphisms.
Even in the developing process of physics, the trace of Notions 1.1 and 1.2 can be also
found. For examples, the many-world interpretation [2] on quantum mechanics by Everett in
1957 is essentially a multispace formulation of quantum state (See [35] for details), and the
unifying the four known forces, i.e., gravity, electro-magnetism, the strong and weak nuclear
force into one super force by many researchers, i.e., establish the unified field theory is nothing
else but also a following of the combinatorial notions by letting Lagrangian L being that a
combination of its subfields, for instance the standard model on electroweak interactions, etc..
Even so, the CC conjecture includes more deeply thoughts for developing mathematics by
combinatorics i.e., mathematical combinatorics which extends the field of all existent mathemat-
ical sciences. After it was presented, more methods were suggested for developing mathematics
in last decade. The main purpose of this report is to survey its contribution to algebra, topol-
ogy and geometry, mathematical analysis, particularly, non-solvable algebraic and differential
equations, theoretical physics with its producing notions in developing mathematical sciences.
All terminologies and notations used in this paper are standard. For those not mentioned
here, we follow reference [5] and [32] for topology, [3] for topological graphs, [1] for algebraic
systems, [4], [34] for differential equations and [12], [30]-[31] for Smarandache systems.
2. Algebraic Combinatorics
Algebraic systems, such as those of groups, rings, fields and modules are combinatorial them-
selves. However, the CC conjecture also produces notions for their development following.
Notion 2.1 For an algebraic system (A ; O), determine its underlying topological structure
GL [A , O] on subsystems, and then classify by graph isomorphism.
Notion 2.2 For an integer m 1, let (1 ; R1 ), (2 ; R2 ), , (m ; Rm ) all be algebraic systems

h i S
m S
m
in Definition 1.2 and Ge; O underlying GL Ge; O with Ge = i and O = Ri , i.e., an
i=1 i=1
algebraic multisystem. Characterize Ge; O and establish algebraic theory, i.e., combinatorial

algebra on Ge; O .
For example, let

hG1 ; 1 i = a, b|a 1 b = b 1 a, a2 = bn = 1

hG2 ; 2 i = b, c|b 2 c = c 2 b, c5 = bn = 1

hG3 ; 3 i = c, d|c 3 d = d 3 c, d2 = c5 = 1
be groups with respective operations 1 , 2 and 3 . Then the set (Ge; {1 , 2 , 3 }) is an algebraic
S3
multisyatem with Ge = Gi .
i=1
6 Linfan MAO
2.1 K2L -Systems
A K2L -system is such a multi-system consisting of exactly 2 algebraic systems underlying a

topological graph K2L , including bigroups, birings, bifields and bimodules, etc.. For example, an
algebraic field (R; +, ) is a K2L -system. Clearly, (R; +, ) consists of groups (R; +) and (R\{0}; )

underlying K2L such as those shown in Fig.4, where L : V K2L {(R; +), (R \ {0}; )} and

L : E K2L {R \ {0}}.
R \ {0}
(R; +) (R \ {0}, )
Fig.4
A generalization of field is replace R \ {0} by any subset H R in Fig.4. Then a bigroup

comes into being, which was introduced by Maggu [8] for industrial systems in 1994, and then
Vasantha Kandasmy [33] further generalizes it to bialgebraic structures.
Definition 2.3 A bigroup (biring, bifield, bimodule, ) is a 2-system (G ; , ) such that

S
(1) G = G1 G2 ;
(2) (G1 ; ) and (G2 ; ) both are groups (rings, fields, modules, ).
f be a permutation multigroup action on

For example, let P e with
[ [
f = P1
P e = {1, 2, 3, 4, 5, 6, 7, 8}
P2 and {1, 2, 5, 6, 9, 10, 11, 12},
f is a per-
where P1 = h(1, 2, 3, 4), (5, 6, 7, 8)i and P2 = h(1, 5, 9, 10), (2, 6, 11, 12)i. Clearly, P
mutation bigroup.
Let (G1 ; 1 , 1 ) and ((G2 ; 2 , 2 )) be bigroups. A mapping pair (, ) with : G1 G2 and
: {1 , 1 } {2 , 2 } is a homomorphism if
(a b) = (a)()(b)
for a, b G1 and {1 , 1 } provided a b existing in (G1 ; 1 , 1 ). Define the image Im(, )

and kernel Ker(, ) respectively by
Im(, ) = { (g) | g G1 },
Ker(, ) = { g G1 | (g) = 1 , {2 , 2 }},
where 1 denotes the unit of (G ; ) with G a maximal closed subset of G on operation .

For subsets H e G,e O O, define (H; e O) to be a submultisystem of G; e O if (H; e O)

is multisystem itself, denoted by H; e O G; e O , and a subbigroup (H ; , ) of (G ; , ) is
normal, denoted by H G if for g G ,
g H = H g,
where g H = {g h|h H provided g h existing} and H g = {h g|h H provided h

g existing} for {, }. The next result is a generalization of isomorphism theorem of group
in [33].
Theorem 2.4([11]) Let (, ) : (G1 ; {1 , 1 }) (G2 ; {2 , 2 }) be a homomorphism. Then
G1 /Ker(, ) Im(, ).
Particularly, if (G2 ; {2 , 2 }) is a group (A ; ), we know the corollary following.
Corollary 2.5 Let (, ) : (G ; {, }) (A ; ) be an epimorphism. Then
G1 /Ker(, ) (A ; ).
Similarly, a bigroup (G ; , ) is distributive if
a (b c) = a b a c
hold for all a, b, c G . Then, we know the following result.
Theorem 2.6([11]) Let (G ; , ) be a distributive bigroup of order 2 with G = A1 A2 such

that (A1 ; ) and (A2 ; ) are groups. Then there must be A1 6= A2 . consequently, if (G ; ) it a
non-trivial group, there are no operations =
6 on G such that (G ; , ) is a distributive bigroup.
2.2 GL -Systems
Definition 2.2 is easily generalized also to multigroups, i.e., consisting of m groups underlying a
topological graph GL , and similarly, define conceptions of homomorphism, submultigroup and
normal submultigroup, of a multigroup without any difficult.
For example, a normal submultigroup of (Ge; O) e is such submutigroup (Hf; O) that holds
g Hf = Hf g
for g Ge, O, and generalize Theorem 2.3 to the following.
Theorem 2.7([16]) Let (, ) : (Ge1 ; O

e1 ) (Ge2 ; O
e2 ) be a homomorphism. Then
Ge1 /Ker(, ) Im(, ).
f be a permutation
e let P
Particularly, for the transitive of multigroup action on a set ,
e with Pf= S
m
e = S
m
multigroup action on Pi , i and for each integer i, 1 i m, the
i=1 i=1
8 Linfan MAO
permutation group Pi acts on i , which is globally k-transitive for an integer k 1 if for

any two k-tuples x1 , x2 , , xk i and y1 , y2 , , yk j , where 1 i, j m, there are
permutations 1 , 2 , , n such that
x1 1 2 n = y1 , x2 1 2 n = y2 , , xk1 2 n = yk
and abbreviate the globally 1-transitive to that globally transitive of a permutation multigroup.
The following result characterizes transitive multigroup.
f be a permutation multigroup action on

Theorem 2.8([17]) Let P e with
m
[ m
[
f=
P e=
Pi and i ,
i=1 i=1
i each integers 1 i m. Then

where, each permutation group Pi transitively acts on hi for
f e L e
P is globally transitive on if and only if the graph G is connected.
S
m
Similarly, let R e= Ri be a completed multisystem with a double operation set O R e =
S i=1
O1 O2 , where O1 = { i , 1 i m}, O2 = {+i , 1 i m}. If for any integers i, 1 i m,
e e
(Ri ; +i , i ) is a ring, then R is called a multiring, denoted by R; O1 O2 and (+i , i ) a
e and an integer i, 1 i m,
double operation for any integer i, which is integral if for a, b R
i b = b i a,1i =
a 6 0+i and a i b = 0+i implies that a = 0+i or b = 0+i . Such a multiring
e O1 O2 is called a skew multifield or a multifield if each (R; +i , i ) is a skew field or a
R;
field for integers 1 i m. The next result is a generalization of finitely integral ring.
Theorem 2.9([16]) A finitely integral multiring is a multifield.
For multimodule, let O = { +i | 1 i m}, O1 = {i |1 i m} and O2 = {+ i |1 i

m} be operation sets, (M ; O) a commutative multigroup with units 0+i and (R; O1 O2 )
a multiring with a unit 1 for O1 . A pair (M ; O) is said to be a multimodule over
(R; O1 O2 ) if for any integer i, 1 i m, a binary operation i : R M M is defined
by a i x for a R, x M such that the conditions following
(1) a i (x +i y) = a i x +i a i y;
i b) i x = a i x +i b i x;
(2) (a+
(3) (a i b) i x = a i (b i x);
(4) 1i i x = x.
hold for a, b R, x, y M , denoted by Mod(M (O) : R(O1 O2 )). Then we know the
following result for finitely multimodules.
D E
Theorem 2.10([16]) Let Mod(M (O) : R(O1 O2 )) = b
S|R be a finitely generated
multimodule with Sb = {u1 , u2 , , un }. Then
Mod(M (O) : R(O1 O2 ))

= Mod(R (n) : R(O1 O2 )),
where Mod(R (n) : R(O1 O2 )) is a multimodule on R (n) = {(x1 , x2 , , xn ) | xi R, 1

i n} with
i y1 , x2 +
(x1 , x2 , , xn ) +i (y1 , y2 , , yn ) = (x1 + i y2 , , xn +
i yn ),
a i (x1 , x2 , , xn ) = (a i x1 , a i x2 , , a i xn )
for a R, integers 1 i m. Particularly, a finitely module over a commutative ring

(R; +, ) generated by n elements is isomorphic to the module R n over (R; +, ).
3. Geometrical Combinatorics
Classical geometry, such as those of Euclidean or non-Euclidean geometry, or projective geome-

try are not combinatorial. Whence, the CC conjecture produces combinatorial notions for their
development further, for instance the topological space shown in Fig.5 following.
T
P1 P2
P1 P2
T T
P1 P0 P2 P0
T T
P1 P3 P0 P2 P4
T T
P3 P0 P4 P0
P3 T P4
P3 P4
Fig.5
Notion 3.1 For a geometrical space P, determine its underlying topological structure GL [A , O]
on subspaces, for instance, n-manifolds and classify them by graph isomorphisms.
Notion 3.2 For an integer m 1, let P1 , P2 , , Pm all be geometrical spaces in Definition

h i
f = S Pi , i.e., a geometrical multispace. Characterize
m
f underlying GL P
1.2 and P f with P
i=1
f and establish geometrical theory, i.e., combinatorial geometry on P.
P f
3.1 Euclidean Spaces
Let 1 = (1, 0, , 0), 2 = (0, 1, 0 , 0), , n = (0, , 0, 1) be the normal basis of a

Euclidean space Rn in a general position, i.e., for two Euclidean spaces Rn , Rn , Rn Rn 6=
Rmin{n ,n } . In this case, let Xv be the set of orthogonal orientations in Rnv , . Then
Rn Rn = Xv Xv , which enables us to construct topological spaces by the combination.
For an index set , a combinatorial Euclidean space EGL (n ; ) underlying a connected
graph GL is a topological spaces consisting of Euclidean spaces Rn , such that

V GL = { Rn | };
10 Linfan MAO

E GL = { (Rn , Rn ) | Rn Rn 6= , , } and labeling
T
L : Rn Rn and L : (R n , Rn ) Rn Rn

for (Rn , Rn ) E GL , , .
Clearly, for any graph G, we are easily construct a combinatorial Euclidean space under-
lying G, which induces a problem following.
Problem 3.3 Determine the dimension of a combinatorial Euclidean space consisting of m

Euclidean spaces Rn1 , Rn2 , , Rnm .
Generally, the combinatorial Euclidean spaces EGL (n1 , n2 , , nm ) are not unique and to
determine dimEGL (n1 , n2 , , nm ) converts to calculate the cardinality of |Xn1 Xn2 Xnm |,
where Xni is the set of orthogonal orientations in Rni for integers 1 i m, which can be
determined by the inclusion-exclusion principle, particularly, the maximum dimension following.
Pm
Theorem 3.4([21]) dimEGL (n1 , , nm ) 1 m + ni and with the equality holds if and
i=1
only if dim (Rni Rnj ) = 1 for (Rni , Rnj ) E GL , 1 i, j m.
To determine the minimum dimEGL (n1 , , nm ) is still open. However, we know this
number for G = Km and ni = r for integers 1 i m, i.e., EKm (r) by following results.
Theorem 3.5([21]) For any integer r 2, let EKm (r) be a combinatorial Euclidean space of
Rr , , Rr , and there exists an integer s, 0 s r 1 such that
| {z }
m

r+s1 r+s
<m .
r r
Then
dimmin EKm (r) = r + s.
Particularly,

3, if m = 1,

4, if 2 m 4,
dimmin EKm (3) =

5, if 5 m 10,

2 + m, if m 11.
3.2 Manifolds
An n-manifold is a second countable Hausdorff space of locally Euclidean n-space without

boundary, which is in fact a combinatorial Euclidean space EGL (n). Thus, we can further
replace these Euclidean spaces by manifolds and to get topological spaces underlying a graph,
such as those shown in Fig.6.
Fig.6
f
Definition 3.6([22]) Let
h Mi be a topological space consisting of finite manifolds M , .
in f f is such a graph with
An inherent graph G M of M
h i
f = {M , };
V Gin M
h i
f = {(M , M ) , 1 i + 1|M M 6= , , },
E Gin M i
where + 1 is the hnumber

i of arcwise connected components in M M for , .
in f
Notice that G M is a multiple graph. If replace all multiple edges (M , M )i , 1 i
f], also an underlying graph of M
+ 1 by (M , M ), such a graph is denoted by G[M f.
Clearly, if m = 1, then M f(ni , i ) is nothing else but exactly an n1 -manifold by

definition. Even so, Notion 3.1 enables us characterizing manifolds by graphs. The following
result shows that every manifold is in fact homeomorphic to combinatorial Euclidean space.
Theorem 3.7([22]) Any locally compact n-manifold M with an alta A = { (U ; )| } is

a combinatorial manifold Mf homeomorphic to a combinatorial Euclidean space EGL (n, )
with countable graphs G [M ]
in
= G.
Topologically, a Euclidean space Rn is homeomorphic to an opened ball Bn (R) = {(x1 , x2 ,

, xn )|x21 +x22 + +x2n < R}. Thus, we can view a combinatorial Euclidean space EG (n, )
as a graph with vertices and edges replaced by ball Bn (R) in space, such as those shown in
Fig.6, a 3-dimensional graph.
Definition 3.8 An n-dimensional graph M fn [G] is a combinatorial ball space B

e of B n ,
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 ;
12 Linfan MAO
(3) the boundary ei ei of ei consists of one or two B n and each pair (ei , ei ) is homeo-
n
morphic to the pair (B , B n );
(4) a subset A M fn [G] is open if and only if A ei is open for 1 i m.
Particularly, a topological graph T [G] of a graph G embedded in a topological space P is

1-dimensional graph.
According to Theorem 3.7, an n-manifold is homeomorphic to a combinatorial Euclidean

space, i.e., n-dimensional graph. This enables us knowing a result following on manifolds.
Theorem 3.9([22]) Let A [M ] = { (U ; ) | } be a atlas of a locally compact n-manifold

L1 L2
M . Then the labeled graph GL|| of M is a topological invariant on ||, i.e., if H|| and G|| are
two labeled n-dimensional graphs of M , then there exists a self-homeomorphism h : M M
L1
such that h : H|| GL|| naturally induces an isomorphism of graph.
2
Theorem 3.9 enables us listing manifolds by two parameters, the dimensions and inherited
graph. For example, let || = 2 and then Amin [M ] = {(U1 ; 1 ), (U2 ; 2 )}, i.e., M is double
L
covered underlying a graphs D0, 12 +1,0
shown in Fig.7,
e1
e2
e3
U1 U2
e12 +1

Fig.7
For example, let U1 = R2 , 1 = z, U2 = (R2 \ {(0, 0)} {}, 2 = 1/z and 12 = 0. Then
the 2-manifold is nothing else but the Riemannian sphere.
The GL -structure on combinatorial manifold M f can be also applied for characterizing a few
topological invariants, such as those fundamental groups, for instance the conclusion following.
h i
f , if M1 M2 is simply connected, then
Theorem 3.10([23]) For (M1 , M2 ) E GL M

O h i
f O f .
1 M = 1 (M ) 1 Gin M
f ])
MV (G[M
Particularly, for a compact n-manifold M with charts {(U , )| : U Rn , },

if U U is simply connected for , , then

1 (M )
= 1 Gin [M ] .
3.3 Algebraic Geometry
The topological group, particularly, Lie group is a typical example of K2L -systems that of algebra
with geometry. Generally, let
AX = (b1 , b2 , , bm )T (LEq)
be a linear equation system with

a11 a12 a1n x1

a a22 a2n x
21 2
A= and X =

am1 am2 amn xn
for integers m, n 1, and all equations in (LEq) are non-trivial, i.e., there are no numbers
such that (ai1 , ai2 , , ain , bi ) = (aj1 , aj2 , , ajn , bj ) for any integers 1 i, j m.
y
6 2x y = 2
2x y = 2
O
B
-x
C x + 2y = 2
D
x + 2y = 2
Fig.8
It should be noted that the geometry of a linear equation in n variables is a plane in Rn .

Whence, a linear system (LEq) is non-solvable or not dependent on their intersection is empty
or not. For example, the linear system shown in Fig.8 is non-solvable because their intersection
is empty.
Definition 3.11 For any integers 1 i, j m, i 6= j, the linear equations
ai1 x1 + ai2 x2 + ain xn = bi ,

aj1 x1 + aj2 x2 + ajn xn = bj
are called parallel if there no solution x1 , x2 , , xn hold both with the 2 equations.
Define a graph GL [LEq] on linear system (LEq) following:

V GL [LEq] = { the solution space Si of ith equation |1 i m},
14 Linfan MAO
T
E GL [Eq] = { (Si , Sj ) | Si Sj = 6 , 1 i, j m} and with labels
T
L : Si Si and L : (Si , Sj ) Si Sj

for Si V GL [LEq] , (Si , Sj ) E GL [LEq] . For example, the system of equations shown
in Fig.8 is

x + 2y = 2

x + 2y = 2

2x y = 2

2x y = 2
and C4L is its underlying graph GL [LEq] shown in Fig.9.
S1 B S4
A D
S3 C S2
Fig.9
Let Li be the ith linear equation. By definition we divide these equations Li , 1 i m

into parallel families
C1 , C2 , , Cs
by the property that all equations in a family Ci are parallel and there are no other equations
parallel to lines in Ci for integers 1 i s. Denoted by |Ci | = ni , 1 i s. Then, we can
characterize GL [LEq] following.
Theorem 3.12([24]) Let (LEq) be a linear equation system for integers m, n 1. Then
GL [LEq] KnL1 ,n2 , ,ns
with n1 + n + 2 + + ns = m, where Ci is the parallel family with ni = |Ci | for integers

1 i s in (LEq) and (LEq) is non-solvable if s 2.
Notice that this result is not sufficient, i.e., even if GL [LEq] Kn1 ,n2 , ,ns , we can not
claim that (LEq) is solvable or not. How ever, if n = 2, we can get a necessary and sufficient
condition on non-solvable linear equations.
Let H be a planar graph with each edge a straight segment on R2 . Its c-line graph LC (H)
is defined by
V (LC (H)) = {straight lines L = e1 e2 el , s 1 in H};

E(LC (H)) = {(L1 , L2 )| if e1i and e2j are adjacent in H f or L1 = e11 e12 e1l , L2 =
e21 e22 e2s , l, s 1}.
Theorem 3.13([24]) A linear equation system (LEq2) is non-solvable if and only if GL [LEq2]
LC (H)), where H is a planar graph of order |H| 2 on R2 with each edge a straight segment
Similarly, let
n+1
P1 (x), P2 (x), , Pm (x) (ESm )
be m homogeneous polynomials in n + 1 variables with coefficients in C and each equation

Pi (x) = 0 determine a hypersurface Mi , 1 i m in Rn+1 , particularly, a curve Ci if n = 2.
We introduce the parallel property following.
Definition 3.14 Let P (x), Q(x) be two complex homogeneous polynomials of degree d in n + 1
variables and I(P, Q) the set of intersection points of P (x) with Q(x). They are said to be
parallel, denoted by P k Q if d > 1 and there are constants a, b, , c (not all zero) such that
for x I(P, Q), ax1 + bx2 + + cxn+1 = 0, i.e., all intersections of P (x) with Q(x) appear
at a hyperplane on Pn C, or d = 1 with all intersections at the infinite xn+1 = 0. Otherwise,
P (x) are not parallel to Q(x), denoted by P 6k Q.
n+1
Define a topological graph GL ESm in Cn+1 by
n+1
V GL ESm = {P1 (x), P2 (x), , Pm (x)};
L
n+1
E G ESm = {(Pi (x), Pj (x))|Pi 6k Pj , 1 i, j m}
with a labeling
L : Pi (x) Pi (x), i (x), Pj (x)) I(Pi , Pj ),

(P
n+1
where 1 i 6= j m, and the topological graph of GL ESm without labels is denoted by
n+1
G ESm . The following result generalizes Theorem 3.12 to homogeneous polynomials.
n+1
Theorem 3.15([26]) Let n 2 be an integer. For a system (ESm ) of homogeneous polyno-
mials with a parallel maximal classification C1 , C2 , , Cl ,
n+1
G[ESm ] K(C1 , C2 , , Cl )
and with equality holds if and only if Pi k Pj and Ps 6k Pi implies that Ps 6k Pj , where
K(C1 , C2 , , Cl ) denotes a complete l-partite graphs
n+1
Conversely, for any subgraph G K(C1 , C2 , , Cl ), there are systems (ESm ) of homo-
geneous polynomials with a parallel maximal classification C1 , C2 , , Cl such that
n+1
G G[ESm ].
3
Particularly, if n = 2, i.e., an (ESm ) system, we get the condition following.

Theorem 3.16([26]) Let GL be a topological graph labeled with I(e) for e E GL . Then
3
3
there is a system ESm of homogeneous polynomials such that GL ESm GL if and only if
16 Linfan MAO

there are homogeneous polynomials Pvi (x, y, z), 1 i (v) for v V GL such that

(u) (v)
Y Y
I(e) = I Pui , Pvi
i=1 i=1

for e = (u, v) E GL , where (v) denotes the valency of vertex v in GL .
These GL -system of homogeneous polynomials enables us to get combinatorial manifolds,

for instance, the following result appeared in [26].
n+1
Theorem 3.17 Let (ESm ) be a GL -system consisting of homogeneous polynomials P1 (x), P2 (x),
, Pm (x) in n + 1 variables with respectively hypersurfaces S1 , S2 , , Sm . hThen
i there his ia
combinatorial manifold M f in Cn+1 such that : M f Se is 1 1 with GL M f GL Se ,
S
m
where, Se = Si .
i=1

Particularly, if n = 2, we can further determine the genus of surface g Se by closed
formula as follows.
Theorem 3.18([26]) Let C1 , C2 , , Cm be complex curves determined by homogeneous poly-

nomials P1 (x, y, z), P2 (x, y, z), , Pm (x, y, z) without common component, and let
deg(Pi )deg(Pj ) eij deg(Pi )deg(Pj )

Y k
X X
RPi ,Pj = cij
kz bij
ky , i,j = 1
k=1 k=1 eij
k 6=0
be the resultant of Pi (x, y, z), Pj (x, y, z) for 1 i 6= j m. Then there is an orientable surface
Se in R3 of genus

D E X m
(deg(P ) 1)(deg(P ) 2) X
i i
g Se = G Ce + (pi )
i=1
2 i
p Sing(Ci )
X X X h \ \ i
i
+ (i,j 1) + (1) c Ck1 Cki 1
T T
1i6=jm i3 Ck1 Cki 6=
S
m
with a homeomorphism : Se C
e= Ci . Furthermore, if C1 , C2 , , Cm are non-singular,
i=1
then
D E X m
(deg(Pi ) 1)(deg(Pi ) 2)
g Se = G Ce +
i=1
2
X X X h \ \ i
+ (i,j 1) + (1)i c Ck1 Cki 1 ,
T T
1i6=jm i3 Ck1 Cki 6=
where
i 1 Pi i 1 i
(p ) = Ipi Pi , (p ) + | (p )|
2 y
is a positive integer with a ramification index (pi ) for pi Sing(Ci ), 1 i m.
Theorem 3.17 enables us to find interesting results in projective geometry, for instance the
following result.
Corollary 3.19 Let C1 , C2 , , Cm be complex non-singular curves determined by homogeneous
T
polynomials P1 (x, y, z), P2 (x, y, z), , Pm (x, y, z) without common component and Ci Cj =
T
m Tm
Ci with Ci = > 0 for integers 1 i 6= j m. Then the genus of normalization Se of
i=1 i=1
curves C1 , C2 , , Cm is
m
X
e = g(S)
e = ( 1)(m 1) + (deg(Pi ) 1)(deg(Pi ) 2)
g(S) .
i=1
2
Particularly, if C1 , C2 , , Cm are distinct lines in P2 C with respective normalizations of

spheres
S1 , D , Sm . Then there D
S2 ,E is aE normalization of surface Se of C1 , C2 , , Cm with
genus G L e . Furthermore, if G L e ) is a tree, then Se is homeomorphic to a sphere.
3.4 Combinatorial Geometry
Furthermore, we can establish combinatorial geometry by Notion 3.2. For example, we have
3 classical geometries, i.e., Euclidean, hyperbolic geometry and Riemannian geometries for de-
scribing behaviors of objects in spaces with different axioms following:
Euclid Geometry:
(A1) There is a straight line between any two points.

(A2) A finite straight line can produce a infinite straight line continuously.
(A3) Any point and a distance can describe a circle.
(A4) All right angles are equal to one another.
(A5) If a straight line falling on two straight lines make the interior angles on the same
side less than two right angles, then the two straight lines, if produced indefinitely, meet on that
side on which are the angles less than the two right angles.
Hyperbolic Geometry:
Axioms (A1) (A4) and the axiom (L5) following:
(L5) there are infinitely many lines parallel to a given line passing through an exterior
point.
Riemannian Geometry:
Axioms (A1) (A4) and the axiom (R5) following:
there is no parallel to a given line passing through an exterior point.
Then whether there is a geometry established by combining the 3 geometries, i.e., partially
Euclidean and partially hyperbolic or Riemannian. Today, we have know such theory really
exists, called Smarandache geometry defined following.
18 Linfan MAO
Definition 3.20([12]) An axiom is said to be Smarandachely denied if the axiom behaves in at

least two different ways within the same space, i.e., validated and invalided, or only invalided
but in multiple distinct ways.
A Smarandache geometry is a geometry which has at least one Smarandachely denied axiom
(1969).
L2 L1 L
C E C
D F E
l1
A D
L3
A F G B B
(a) (b)
Fig.10
For example, let us consider a Euclidean plane R2 and three non-collinear points A, B and
C shown in Fig.10. Define s-points as all usual Euclidean points on R2 and s-lines any Euclidean
line that passes through one and only one of points A, B and C. Then such a geometry is a
Smarandache geometry by the following observations.
Observation 1. The axiom (E1) that through any two distinct points there exist one line
passing through them is now replaced by: one s-line and no s-line. Notice that through any
two distinct s-points D, E collinear with one of A, B and C, there is one s-line passing through
them and through any two distinct s-points F, G lying on AB or non-collinear with one of A, B
and C, there is no s-line passing through them such as those shown in Fig.10(a).
Observation 2. The axiom (E5) that through a point exterior to a given line there is
only one parallel passing through it is now replaced by two statements: one parallel and no
parallel. Let L be an s-line passes through C and D on L, and AE is parallel to CD in the
Euclidean sense. Then there is one and only one line passing through E which is parallel to L,
but passing a point not on AE, for instance, point F there are no lines parallel to L such as
those shown in Fig.10(b).
Generally, we can construct a Smarandache geometry on smoothly combinatorial manifolds

f
M , i.e., combinatorial geometry because it is homeomorphic to combinatorial Euclidean space
EGL (n1 , n2 , , nm ) by Definition 3.6 and Theorem 3.7. Such a theory is founded on the results
for basis of tangent and cotangent vectors following.
Theorem 3.21([15]) For any point p M f(n1 , n2 , , nm ) with a local chart (Up ; [p ]), the
f(n1 , n2 , , nm ) is
dimension of Tp M
s(p)
P
f(n1 , n2 , , nm ) = sb(p) +
dimTp M (ni sb(p))
i=1

with a basis matrix =
x s(p)ns(p)

1 1
s(p) x11 s(p) x1bs(p) x1(bs(p)+1)
x1n1 0

1
1

0
s(p) x21 s(p) x2bs(p) x2(bs(p)+1) x2n2

1 1
s(p) xs(p)1 s(p) xs(p)bs(p) xs(p)(bs(p)+1)
s(p)(ns(p) 1) s(p)ns(p)
x x
where xil = xjl for 1 i, j s(p), 1 l sb(p), namely there is a smoothly functional matrix
f(n1 , n2 , , nm ),
[vij ]s(p)ns(p) such that for any tangent vector v at a point p of M

v = [vij ]s(p)ns(p) , [ ]s(p)ns(p) ,
x
P
k P
l
where h[aij ]kl , [bts ]kl i = aij bij , the inner product on matrixes.
i=1 j=1
Theorem 3.22([15]) For p (M f(n1 , n2 , , nm ); A)

e with a local chart (Up ; [p ]), the dimen-
f
sion of Tp M (n1 , n2 , , nm ) is
s(p)
P
f(n1 , n2 , , nm) = sb(p) +
dimTp M (ni sb(p))
i=1
with a basis matrix [dx]s(p)ns(p) =

dx11 dx1bs(p)
s(p) s(p) dx1(bs(p)+1) dx1n1 0

dx21
dx2bs(p)
dx2(bs(p)+1) dx2n2 0
s(p) s(p)

dxs(p)1 dxs(p)bs(p)
s(p) s(p) dx s(p)+1)
s(p)(b
dxs(p)ns(p) 1 dxs(p)ns(p)
where xil = xjl for 1 i, j s(p), 1 l sb(p), namely for any co-tangent vector d at a point
f(n1 , n2 , , nm ), there is a smoothly functional matrix [uij ]s(p)s(p) such that,
p of M
D E
d = [uij ]s(p)ns(p) , [dx]s(p)ns(p) .
Then we can establish tensor theory with connections on smoothly combinatorial manifolds
([15]). For example, we can establish the curvatures on smoothly combinatorial manifolds, and
e formula following.
get the curvature R
f be a finite combinatorial manifold, R

Theorem 3.23([18]) Let M e : X (M
f) X (M f) X (M
f)
f f f f
X (M ) C (M ) a curvature on M . Then for p M with a local chart (Up ; [p ]),
e=R
R e()()()() dx dx dx dx ,
20 Linfan MAO
where
e()()()() 1 2 g()() 2 g()() 2 g()() 2 g()()

R = ( + )
2 x x x x x x x x
o o
+
()() ()() g(o)() ()() ()() g(o)() ,
and g()() = g( x , x ).
This enables us to characterize the combination of classical fields, such as the Einsteins
gravitational fields and other fields on combinatorial spacetimes and hold their behaviors ( See
[19]-[20] for details).
4. Differential Equations Combinatorics
Let

f1 (x1 , x2 , , xn+1 ) = 0

f (x , x , , x
2 1 2 n+1 ) = 0
(Eqm )

........................

fm (x1 , x2 , , xn+1 ) = 0
be a system of equations. It should be noted that the classical theory on equations is not
combinatorics. However, the solutions of an equation usually form a manifold in the view of
geometry. Thus, the CC conjecture bring us combinatorial notions for developing equation
theory similar to that of geometry further.
Notion 4.1 For a system (Eqm ) of equations, solvable or non-solvable, determine its un-
derlying topological structure GL [Eqm ] on each solution manifold and classify them by graph
isomorphisms and transformations.
Notion 4.2 For an integer m 1, let D1 , D2 , , Dm be the solution manifolds of an

h i
e with De = S Di , i.e.,
m
e underlying GL D
equation system (Eqm ) in Definition 1.2 and D
i=1
a combinatorial solution manifold. Characterize the system (Eqm ) and establish an equation
theory, i.e., equations combinatorics on (Eqm ).
Geometrically, let
Sfi = {(x1 , x2 , , xn+1 )|fi (x1 , x2 , , xn+1 ) = 0} Rn+1
the solution-manifold in Rn+1 for integers 1 i m, where fi is a function hold with conditions
of the implicit function theorem for 1 i m. Then we are easily finding criterions on the
solubility of system (ESm ), i.e., it is solvable or not dependent on
m
\
Sfi 6= or = .
i=1
Whence, if the intersection is empty, i.e., (ESm ) is non-solvable, there are no meanings in
classical theory on equations, but it is important for hold the global behaviors of a complex
thing. For such an objective, Notions 4.1 and 4.2 are helpful.
Let us begin at a linear differential equations system such as those of
X = A1 X, , X = Ak X, , X = Am X 1
(LDESm )
or
[0] [0]

x(n) + a11 x(n1) + + a1n x = 0

x(n) + a[0] x(n1) + + a[0] x = 0
21 2n n
(LDEm )

(n) [0] [0]
x + am1 x(n1) + + amn x = 0
with
[k] [k] [k]

a11 a12 a1n x1 (t)
[k]
a [k]
a22 a2n
[k] x (t)
2
Ak = 21 and X =

[k] [k] [k]
an1 an2 ann xn (t)
[k]
where each aij is a real number for integers 0 k m, 1 i, j n.
For example, let (LDE62 ) be the following linear homogeneous differential equation system

x
+ 3x + 2x = 0 (1)

x
+ 5x + 6x = 0 (2)

x
+ 7x + 12x = 0 (3)

x
+ 9x + 20x = 0 (4)

x
+ 11x + 30x = 0 (5)

x
+ 7x + 6x = 0 (6)
Certainly, it is non-solvable. However, we can easily solve equations (1)-(6) one by one and
get their solution spaces as follows:

S1 = et , e2t = {C1 et + C2 e2t |C1 , C2 R} = {x|x + 3x + 2x = 0}

2t 3t
S2 = e , e = {C1 e2t + C2 e3t |C1 , C2 R} = {x|
x + 5x + 6x = 0}

3t 4t 3t 4t
S3 = e , e = {C1 e + C2 e |C1 , C2 R} = {x| x + 7x + 12x = 0}

4t 5t 4t 5t
S4 = e , e = {C1 e + C2 e |C1 , C2 R} = {x| x + 9x + 20x = 0}

5t 6t
S5 = e , e = {C1 e5t + C2 e6t |C1 , C2 R} = {x|
x + 11x + 30x = 0}

6t t 6t t
S6 = e , e = {C1 e + C2 e |C1 , C2 R} = {x|x + 7x + 6x = 0}
Replacing each i by solution space Si in Definition 1.2, we get a topological graph

22 Linfan MAO
GL [LDE62 ] shown in Fig.11 on the linear homogeneous differential equation system (LDE62 ).
Thus we can solve a system of linear homogeneous differential equations on its underlying graph
GL , no matter it is solvable or not in the classical sense.

t 2t
2t
2t 3t
e ,e e e ,e

3t
het i e

6t t

3t 4t
e , e e ,e

6t
4t
e e

5t 6t
5t
4t 5t

e ,e e e ,e
Fig.11
Generally, we know a result on GL -solutions of homogeneous equations following.
1 n
Theorem 4.3([25]) A linear homogeneous differential equation system (LDESm ) (or (LDEm ))

has a unique GL -solution, and for every H L labeled with linear spaces i (t)ei t , 1 i n on
vertices such that

\

i (t)ei t , 1 i n j (t)ej t , 1 j n 6=

if and only if there is an edge whose end vertices labeled by i (t)ei t , 1 i n and j (t)ej t ,
1 j ni respectively, then there is a unique linear homogeneous differential equation system
1 n
(LDESm ) (or (LDEm )) with GL -solution H L , where i is a ki -fold zero of the characteristic
equation, k1 + k2 + + ks = n and i (t) is a polynomial in t with degree ki 1.
Applying GL -solution, we classify such systems by graph isomorphisms.
Definition 4.4 A vertex-edge labeling : G Z+ is said to be integral if (uv) min{(u), (v)}

for uv E(G), denoted by GI , and two integral labeled graphs GI1 and GI2 are called iden-
S
tical if G1 G2 and (x) = ((x)) for any graph isomorphism and x V (G1 ) E(G1 ),
denoted by GI1 = GI2 .
For example, GI1 = GI2 but GI1 6= GI3 for integral graphs shown in Fig.12.
3 2 4 4 2 3 3 1 3
1 1 2 2 1 1 2 2 2
4 2 3 3 2 4 4 1 4
GI1 GI2 GI3
Fig.12
1 n
The following result classifies the systems (LDESm ) and (LDEm ) by graphs.
1 1 n n
Theorem 4.5([25]) Let (LDESm ), (LDESm ) (or (LDEm ), (LDEm ) ) be two linear homo-

geneous differential equation systems with integral labeled graphs H, H . Then (LDESm 1
)
1 n n
(LDESm ) (or (LDEm ) (LDEm ) ) if and only if H = H .
For partial differential equations, let

F1 (x1 , x2 , , xn , u, ux1 , , uxn ) = 0

F (x , x , , x , u, u , , u ) = 0
2 1 2 n x1 xn
(P DESm )

......................................

Fm (x1 , x2 , , xn , u, ux1 , , uxn ) = 0
be such a system of first order on a function u(x1 , , xn , t) with continuous Fi : Rn Rn

such that Fi (0) = 0.
Definition 4.6 The symbol of (P DESm ) is determined by

F1 (x1 , x2 , , xn , u, p1 , , pn ) = 0

F (x , x , , x , u, p , , p ) = 0
2 1 2 n 1 n

...................................

Fm (x1 , x2 , , xn , u, p1 , , pn ) = 0,
i.e., substitutes ux1 , ux2 , , uxn by p1 , p2 , , pn in (P DESm ), and it is algebraically contra-

dictory if its symbol is non-solvable. Otherwise, differentially contradictory.
For example, the system of partial differential equations following

(z y)ux + (x z)uy + (y x)uz = 0

zux + xuy + yuz = x2 + y 2 + z 2 + 1

yux + zuy + xuz = x2 + y 2 + z 2 + 4
is algebraically contradictory because its symbol

(z y)p1 + (x z)p2 + (y x)p3 = 0

zp1 + xp2 + yp3 = x2 + y 2 + z 2 + 1

yp1 + zp2 + xp3 = x2 + y 2 + z 2 + 4
is contradictory. Generally, we know a result for Cauchy problem on non-solvable systems of

partial differential equations of first order following.
24 Linfan MAO
Theorem 4.7([28]) A Cauchy problem on systems

F1 (x1 , x2 , , xn , u, p1 , p2 , , pn ) = 0

F (x , x , , x , u, p , p , , p ) = 0
2 1 2 n 1 2 n
......................................

Fm (x1 , x2 , , xn , u, p1 , p2 , , pn ) = 0
of partial differential equations of first order is non-solvable with initial values

0
xi |xn =xn = xi (s1 , s2 , , sn1 )
0
u|xn =x0n = u0 (s1 , s2 , , sn1 )

pi |xn =x0n = p0i (s1 , s2 , , sn1 ), i = 1, 2, , n
if and only if the system
Fk (x1 , x2 , , xn , u, p1 , p2 , , pn ) = 0, 1 k m
is algebraically contradictory, in this case, there must be an integer k0 , 1 k0 m such that
Fk0 (x01 , x02 , , x0n1 , x0n , u0 , p01 , p02 , , p0n ) 6= 0
or it is differentially contradictory itself, i.e., there is an integer j0 , 1 j0 n 1 such that
n1
X x0
u0
p0i i
6= 0.
sj0 i=0
s j0
According to Theorem 4.7, we know conditions for uniquely GL -solution of Cauchy problem
on system of partial differential equations of first order following.
Theorem 4.8([28]) A Cauchy problem on system (P DESm ) of partial differential equations

[k0 ] [k] [k0 ]
of first order with initial values xi , u0 , pi , 1 i n for the kth equation in (P DESm ),
1 k m such that
[k] Xn [k0 ]
u0 [k0 ] xi
pi =0
sj i=0
sj
is uniquely GL -solvable, i.e., GL [P DES] is uniquely determined.
Applying the GL -solution of a GL -system (DESm ) of differential equations, the global

stability, i.e, sum-stable or prod-stable of (DESm ) can be introduced. For example, the sum-
stability of (DESm ) is defined following.
C
Definition 4.9 Let (DESm ) be a Cauchy problem on a system of differential equations in Rn ,

H G DESm a spanning subgraph, and u[v] the solution of the vth equation with initial
L L C
[v]
value u0 , v V H L . It is sum-stable on the subgraph H L if for any number > 0 there

exists, v > 0, v V H L such that each GL (t)-solution with

[v] [v]
u 0 u0 < v , v V H L
exists for all t 0 and the inequality

X X
[v]

u u [v]
<
vV (H L ) vV (H L )
H
holds, denoted by GL [t] GL [0] and GL [t] GL [0] if H L = GL DESm
C
. Furthermore, if
L
L
there exists a number v > 0, v V H such that every G [t]-solution with

[v] [v]
u 0 u0 < v , v V H L
satisfies

X [v] X

lim u [v]
u = 0,
t
vV (H) vV (H L )
H
then the GL [t]-solution is called asymptotically stable, denoted by GL [t] GL [0] and GL [t]

GL [0] if H L = GL DESm C
.
C
For example, let the system (SDESm ) be

u
= Hi (t, x1 , , xn1 , p1 , , pn1 ) C

t 1im SDESm
[i]
u|t=t0 = u0 (x1 , x2 , , xn1 )
[i] [i] [i] [i] [i]

and a point X0 = (t0 , x10 , , x(n1)0 ) with Hi (t0 , x10 , , x(n1)0 ) = 0 for an integer 1 i
C C
m is equilibrium of the ith equation in (SDESm ). A result on the sum-stability of (SDESm )
is obtained in [30] following.
[i] C
Theorem 4.10([28]) Let X0 be an equilibrium point of the ith equation in (SDESm ) for each
integer 1 i m. If
Xm Xm
Hi
Hi (X) > 0 and 0
i=1 i=1
t
P
m
[i] C
for X 6= X0 , then the system (SDESm ) is sum-stable, i.e., GL [t] GL [0].
i=1
Furthermore, if
m
X Hi
<0
i=1
t
P
m
[i]
for X 6= X0 , then GL [t] GL [0].
i=1
26 Linfan MAO
5. Fields Combinatorics
The modern physics characterizes particles by fields, such as those of scalar field, Maxwell field,
Weyl field, Dirac field, Yang-Mills field, Einstein gravitational field, , etc., which are in fact
spacetime in geometry, isolated but non-combinatorics. Whence, the CC conjecture can bring
us a combinatorial notion for developing field theory further, which enables us understanding
the world and discussed extensively in the first edition of [13] in 2009, and references [18]-[20].
Notion 5.1 Characterize the geometrical structure, particularly, the underlying topological
structure GL [D] of spacetime D on all fields appeared in theoretical physics.
Notice that the essence of Notion 5.1 is to characterize the geometrical spaces of particles.
Whence, it is in fact equivalent to Notion 3.1.
Notion 5.2 For an integer m 1, let D1 , D2 , , Dm be spacetimes in Definition 1.2

h i
e = S Di , i.e., a combinatorial spacetime. Select suitable
m
e underlying GL D
and D e with D
i=1
Lagrangian or Hamiltonian density L fto determine field equations of D,
e hold with the principle
of covariance and characterize its global behaviors.
There are indeed such fields, for instance the gravitational waves in Fig.13.
Fig.13
e is a combination of fields underlying a topological graph GL

A combinatorial field D
with actions between fields. For this objective, a natural way is to characterize each field
Ci , 1 i n of them by itself reference frame {x}. Whence, the principles following are
indispensable.

Action Principle of Fields. There are always exist an action A between two fields C1 and
C2 of a combinatorial field if dim(C1 C2 ) 1, which can be found at any point on a spatial
direction in their intersection.
h i
Thus, a combinatorial field depends on graph GL D e , such as those shown in Fig.14.
C1

C2
C1C2

C4C1 C2C3
C3

C4
C3C4
Fig.14
For understanding the world by combinatorial fields, the anthropic principle, i.e., the born
of human beings is not accidental but inevitable in the world will applicable, which implies the
generalized principle of covariance following.
Generalized Principle of Covariance([20]) A physics law in a combinatorial field is invari-

ant under all transformations on its coordinates, and all projections on its a subfield.
Then, we can construct the Lagrangian density L f and find the field equations of combi-
e which are divided into two cases ([13], first edition).
natorial field D,
Case 1. Linear
In this case, the expression of the Lagrange density LGL [De ] is
n
X X
LGL [De ] = ai L D i + bij Tij ,
i=1 e ])
(Di ,Dj )E (GL [D
where ai , bij are coupling constants determined only by experiments.
Case 2. Non-Linear
In this case, the Lagrange density LGL [De ] is a non-linear function on LDi and Tij for
h i
1 i, j n. Let the minimum and maximum indexes j for (Mi , Mj ) E GL D e are il and
iu , respectively. Denote by
x = (x1 , x2 , ) = (LD1 , LD2 , , LDn , T11l , , T11u , , T22l , , ).
If LGL [De ] is k + 1 differentiable, k 0, by Taylors formula we know that
" # " 2 #
n
X LGL [De ] 1 X LGL [De ]
n
LGL [De ] = LGL [De ] (0) + xi + xi xj
i=1
xi 2! i,j=1 xi xj
xi =0 xi ,xj =0
" #
1 X n k LGL [De ]
++ xi1 xi2 xik
k! i ,i , ,i =1 xi1 xi2 xik
1 2 k xij =0,1jk
+R(x1 , x2 , ),
28 Linfan MAO
where
R(x1 , x2 , )
lim = 0,
kxk0 kxk
and choose the first k terms
" # " 2 #
n
X LGL [De ] 1 X LGL [De ]
n
LGL [De ] (0) + xi + xi xj
i=1
xi 2! i,j=1 xi xj
xi =0 xi ,xj =0
" k #
1 X n L GL [D e]
++ xi1 xi2 xik
k! i ,i , ,i =1 xi1 xi2 xik
1 2 k xij =0,1jk
to be the asymptotic value of Lagrange density LGL [De ] , particularly, the linear parts
" # " #
n
X LGL [De ] X LGL [De ]
LGL [De ] (0) + LMi + Tij .
i=1
LDi Tij
LDi =0 (Mi ,Mj f ])
)E(GL [M Tij =0
Notice that such a Lagrange density maybe intersects. We need to consider those of
Lagrange densities without intersections. For example,
4
X 4
X
LGL [De ] = LC2i L

C i C i+1
i=1 i=1
for the combinatorial field shown in Fig.14.
Then, applying the Euler-Lagrange equations, i.e.,
LGL [De ] LGL [De ]

= 0,
De De
e
where De is the wave function of combinatorial field D(t), we are easily find the equations of
e
combinatorial field D.
For example, for a combinatorial scalar field De , without loss of generality let
n
X
De = ci Di
i=1
n
1X X
LGL [De ] = ( D i Di m2i 2Di ) + bij Di Dj ,
2 i=1 i i
e ])
(Di ,Dj )E (GL [D
i.e., linear case

n
X X
LGL [De ] = L Di + bij Tij
i=1 e ])
(Di ,Dj )E (GL [D
with LDi = 21 (i Di i Di m2i 2Di ), Tij = Di Dj , i = Di and constants bij , mi , ci for

integers 1 i, j n. Then the equation of combinatorial scalar field is
Xn X
1 Mj Mi
( i + m2i )Mi bij + = 0.
i=1
c i ci cj
f
(Mi ,Mj )E(GL [M])
Similarly, we can determine the equations on combinatorial Maxwell field, Weyl field, Dirac
field, Yang-Mills field and Einstein gravitational field in theory. For more such conclusions, the
reader is refers to references [13], [18]-[20] in details.
Notice that the string theory even if arguing endlessly by physicists, it is in fact a combina-
torial field R4 R7 under supersymmetries, and the same also happens to the unified field theory
such as those in the gauge field of Weinberg-Salam on Higgs mechanism. Even so, Notions 5.1
and 5.2 produce developing space for physics, merely with examining by experiment.
6. Conclusions
The role of CC conjecture to mathematical sciences has been shown in previous sections by
examples of results. Actually, it is a mathematical machinery of philosophical notion: there
always exist universal connection between things T with a disguise GL [T ] on connections,
which enables us converting a mathematical system with contradictions to a compatible one
([27]), and opens thoroughly new ways for developing mathematical sciences. However, is a
topological graph an element of a mathematical system with measures, not only viewed as a
geometrical figure? The answer is YES!

Recently, the author introduces G -flow in [29], i.e., an oriented graph G embedded in a
topological space S associated withan injective mappings L : (u, v) L(u, v) V such that

L(u, v) = L(v, u) for (u, v) X G holding with conservation laws
X

L (v, u) = 0 for v V G ,
uNG (v)

where V is a Banach space over a field F and showed all these G -flows G V form a Banach
space by defining X
L
G = kL(u, v)k

(u,v)X G

for G L G V , and furthermore, Hilbert space by introducing inner product similarly, where

kL(u, v)k denotes the norm of F (uv ) in V , which enables us to get G -flow solutions, i.e.,
combinatorial solutions on differential equations.
References
[1] G.Birkhoff and S.MacLane, A Survey of Modern Algebra (4th edition), Macmillan Publish-
ing Co., Inc, 1977.
30 Linfan MAO
[2] H.Everett, Relative state formulation of quantum mechanics, Rev.Mod.Phys., 29(1957),

454-462.
[3] J.L.Gross and T.W.Tucker, Topological Graph Theory, John Wiley & Sons, 1987.
[4] Fritz John. Partial Differential Equations(4th Edition). New York, USA: Springer-Verlag,
1982.
[5] John M.Lee, Introduction to Topological Manifolds, Springer-Verlag New York, Inc., 2000.
[6] Tian Ma and Shouhong Wang, Unified field equations coupling four forces and principle of
interaction dynamics, arXiv: 1210.0448.
[7] Tian Ma and Shouhong Wang, Unified field theory and principle of representation invari-
ance, arXiv: 1212.4893.
[8] P.L.Maggu, On introduction of bigroup concept with its applications industry, Pure Appl.
Math. Sic., 39(1994), 171-173.
[9] Linfan Mao, On the Automorphisms of Maps & Klein Surfaces, a post doctoral report for
Chinese Academy of Sciences, 2005.
[10] Linfan Mao, On Automorphisms groups of Maps, Surfaces and Smarandache geometries,
Sientia Magna, Vol.1, No.2(2005), 55-73.
[11] Linfan Mao, Automorphism Groups of Maps, Surfaces and Smarandache Geometries, First
edition published by American Research Press in 2005, Second edition is as a Graduate
Textbook in Mathematics, Published by The Education Publisher Inc., USA, 2011.
[12] Linfan Mao, Smarandache Multi-Space Theory(Second edition), First edition published
by Hexis, Phoenix in 2006, Second edition is as a Graduate Textbook in Mathematics,
Published by The Education Publisher Inc., USA, 2011.
[13] Linfan Mao, Combinatorial Geometry with Applications to Field Theory, First edition pub-
lished by InfoQuest in 2009, Second edition is as a Graduate Textbook in Mathematics,
Published by The Education Publisher Inc., USA, 2011.
[14] Linfan Mao, Combinatorial speculation and combinatorial conjecture for mathematics,
International J.Math. Combin. Vol.1(2007), No.1, 1-19.
[15] Linfan Mao, Geometrical theory on combinatorial manifolds, JP J.Geometry and Topology,
Vol.7, No.1(2007),65-114.
[16] Linfan Mao, Extending homomorphism theorem to multi-systems, International J.Math.
Combin., Vol.3(2008), 1-27.
[17] Linfan Mao, Action of multi-groups on finite multi-sets, International J.Math. Combin.,
Vol.3(2008), 111-121.
[18] Linfan Mao, Curvature equations on combinatorial manifolds with applications to theoret-
ical physics, International J.Math. Combin., Vol.1 (2008), 16-35
[19] Linfan Mao, Combinatorial fields-an introduction, International J. Math.Combin., Vol.1(2009),
Vol.3, 1-22.
[20] Linfan Mao, Relativity in combinatorial gravitational fields, Progress in Physics, Vol.3(2010),
39-50.
[21] Linfan Mao, A combinatorial decomposition of Euclidean space Rn with contribution to
visibility, International J.Math.Combin., Vol.1(2010), 47-64.
[22] Linfan Mao, Graph structure of manifolds with listing, International J.Contemp. Math.
Sciences, Vol.5, 2011, No.2,71-85.
[23] Linfan Mao, A generalization of Seifert-Van Kampen theorem for fundamental groups, Far
East Journal of Math.Sciences, Vol.61 No.2 (2012), 141-160.
[24] Linfan Mao, Non-solvable spaces of linear equation systems, International J. Math. Com-
bin., Vol.2 (2012), 9-23.
[25] Linfan Mao, Global stability of non-solvable ordinary differential equations with applica-
tions, International J.Math. Combin., Vol.1 (2013), 1-37.
[26] Linfan Mao, Geometry on GL -systems of homogenous polynomials, International J.Contemp.
Math. Sciences, Vol.9 (2014), No.6, 287-308.
[27] Linfan Mao, Mathematics on non-mathematics - A combinatorial contribution, Interna-
tional J.Math. Combin., Vol.3(2014), 1-34.
[28] Linfan Mao, Cauchy problem on non-solvable system of first order partial differential equa-
tions with applications, Methods and Applications of Analysis, Vol.22, 2(2015), 171-200.

[29] Linfan Mao, Extended Banach G -flow spaces on differential equations with applications,
Electronic J.Mathematical Analysis and Applications, Vol.3, No.2 (2015), 59-90.
[30] F.Smarandache, Paradoxist Geometry, State Archives from Valcea, Rm. Valcea, Romania,
1969, and in Paradoxist Mathematics, Collected Papers (Vol. II), Kishinev University
Press, Kishinev, 5-28, 1997.
[31] F.Smarandache, Multi-space and multi-structure, in Neutrosophy. Neutrosophic Logic, Set,
Probability and Statistics, American Research Press, 1998.
[32] J.Stillwell, Classical Topology and Combinatorial Group Theory, Springer-Verlag New York
Inc., 1980.
[33] W.B.Vasantha Kandasmy, Bialgebraic Structure and Smarandache Bialgebraic Structure,
American Research Press, 2003.
[34] Wolfgang Walter, Ordinary Differential Equations, Springer-Verlag New York, Inc., 1998.
[35] A.Yu.Kamenshchil and O.V.Teryaev, Many-worlds interpretation of quantum theory and
mesoscopic anthropic principle, Concepts of Physics, Vol.V, 4(2008), 575-592.
Timelike-Spacelike Mannheim Pair Curves

Spherical Indicators Geodesic Curvatures and Natural Lifts
S
uleyman SENYURT and Selma DEMET
(Ordu University, Faculty of Arts and Science, Department of Mathematics, Ordu/Turkey)
E-mail: senyurtsuleyman@hotmail.com, selma demet @hotmail.com
Abstract: In this paper, Mannheim curve is a timelike curve, by getting partner curve as a
spacelike curve which has spacelike binormal, with respect to IL3 Lorentz Space, S21 Lorentz
sphere, or H02 Hyperbolic sphere, we have calculated arc lengths of Mannheim partner curves
(T ) , (N ), (B ) spherical indicator curves, arc length of (C ) fixed pol curve, and we have
calculated geodesic curves of them, and also we have figured some relations among them. In
addition, if the natural lifts geodesic spray of spherical indicator curvatures of Mannheim
partner curve is an integral curve, we have expressed how Mannheim Curve is supposed to
be.
Key Words: Lorentz space, Mannheim curve, geodesic curvatures, geodesic spray, natural
lift.
AMS(2010): 53B30, 53C50
1. Introduction
There are a lot of researches to be done in 3-dimentional Euclidian Space on differential geometry
of the curves. Especially, many theories were obtained by making connections two curves
mutual points and between Frenet Frames. Well known researches are Bertrand curves and
Involute-Evolute curves, [6], [4], [7], [19]. Those curves were studied carefully in different
spaces, therefore, so many results were gained. In Euclidian Space and Minkowski Space,
Bertrand curves Frenet frames and Involute-Evolute curves Frenet frames create spherical
indicator curves on unit sphere surface. Those spherical indicator curves natural lifts and
geodesic sprays are defined in [5], [16], [3], [8].
Mannheim curve was firstly defined by A. Mannheim in 1878. Any curve can be a

Mannheim curve if and only if = 2 + 2 , is a nonzere constant, where curvature
of curve is and curvature of torsion is . After a time, Manheim curve was redefined by Liu ve
Wang. According to this new definition, when first curves principal normal vector and second
curves binormal vector are linearly dependent, first curve was named as Mannheim curve, and
second curve was named as Mannheim partner curve, [21], [10]. There are so many researches
Timelike-Spacelike Mannheim Pair Curves Spherical Indicators Geodesic Curvatures and Natural Lifts 33
to be done by Mathematicians after Liu ve Wangs definition [12], [15], [15], [2].
2. Preliminaries
Let : I E3 , (t) = (1 (s), 2 (s), 3 (s)) be unit speed differentiable curve. If we symbolize
: I E3 curves Frenet as {T, N, B}, curvature as , and torsion as , there are some
equations among them;

T (s) = (s)N (s)

N (s) = (s)T (s) + (s)B(s) (2.1)

B (s) = (s)N (s).
By using (2.1), we can get W Darboux vector as;
W = T + B (2.2)
If is the angle which is between W and B, the unit Darboux vector is that;
C = sin T + cos B (2.3)
Let X be differentiable vector space on M .( Note that M is any vector space)
d
( (s)) = X ( (s)) (2.4)
ds
d
curve is an integral curve of X if and only if ds ( (s)) = X ( (s)).
S
Suppose that T M = P M TM (P ), then,
: I T M, (s) = ( (s) , (s)) (2.5)
: I T M curve is natural lift of : I M and for v T M
X (v) = hv, S (v)i N |P
X vector space is called geodesic spray [5], [17]. Where
DX Y = DX Y + hS (X) , Y i N (2.6)
The equation of (2.6) is a Gauss equation on M . (T ) , (N ) and (B) spherical indicator

34 S
uleyman S
ENYURT and Selma DEMET
curves equations and (C) pol curves equations are given respectively;

T (s) = T (s)

(s) = N (s)
N
B (s) = B (s)

C (s) = C (s)
With respect to E3 , arc lengths and geodesic curvatures of those curves are given respec-
tively;
Z s Z s Z s Z s
sT = ds, sN = kW k ds, sB = ds, sC = ds (2.7)
0 0 0 0

1

kT = ,

cos

r

2

kN = 1 + ,

kW k
(2.8)

1

kB = ,

sin

r

2

kC = 1 + kW
k
.
With respect to S 2 , geodesic curvatures are given;
kW k
T = tan , N = , B = cot , C = (see[9]) (2.9)
kW k
Let : I (M ) be natural lift of : I M . X geodesic spray is an integral curve if

and only if there is a geometric curve on M , (see [5]).
g : R3 R3 R , g(X, Y ) = x1 y1 + x2 y2 x3 y3
This inner product space is defined as Lorentz Space and symbolized as IL3 .X IL3 vectors
p
norm is kXkIL = |g(X, X)|. For X = (x1 , x2 , x3 ) and Y = (y1 , y2 , y3 ) IL3
X Y = (x3 y2 x2 y3 , x1 y3 x3 y1 , x1 y2 x2 y1 )
X Y is called vector product of X and Y , [1]. Let T be tangent vector of : I IL3 .

: I IL3 is respectively defined as:
(1) If g(T, T ) > 0, curve is a spacelike curve;

(2) If g(T, T ) < 0, curve is a timelike curve;
(3) If g(T, T ) = 0, curve is a lightlike or null curve, (see [11]).
Let : I IL3 be differentiable timelike curve. In this case, T is timelike, N and B are
spacelike, and Frenet formulas are given;

T = N

N = T B (2.10)

B = N.
(see [22]). Where

W = T B (2.11)
(see [20]). In this situation, there are two cases for W Darboux vector; if W is spacelike, the
Lorentzian timelike angle which is between B and W, then;
= kW k cosh , = kW k sinh (2.12)
C = sinh T cosh B (2.13)
and unit Darboux vector is;

If W timelike, and are formulized;
= kW k sinh , = kW k cosh (2.14)
and unit Darboux vector is;
C = cosh T sinh B (2.15)
Let : I IL3 be spacelike curve which has spacelike binormal. In this case, curves
Frenet vectors vector product are respectively;
T N = B , N B = T , B T = N and Frenet formulas are found as;

T = N

N = T + B (2.16)

B = N
(see [22]).In this case, Darboux vector will be;
W = T + B, (see[20]). (2.17)
Let be the angle which is between B and W . Then,
= kW k cos , = kW k sin (2.18)
and unit Darboux vector is given as;
C = sin T + cos B (2.19)

36 S
uleyman S
Let M be a Lorentz manifold, and M be a hypersurface of M. Suppose that S is a shape

is the connection
operator which is obtained from N normal of M , D is the connection on M, D
on M ,
For X, Y (M ), Gauss Equation is;
X Y + g(S(X), Y )N
DX Y = D (2.20)
Where S(X) = DX N and = g(N, N ), [18].

S12 (r) = X IR13 g(X, X) = r2 , r IR, r = f ixed
is defined as Lorentz sphere,

H02 (r) = X IR13 g(X, X) = r2 , r IR, r = f ixed
is defined as hyperbolic sphere.
Let : I E3 and : I E3 be two differentiable curves. Suppose that Frenet

Frames on the points of (s) and (s) are respectively given as {T (s), N (s), B(s)} and
{T (s), N (s), B (s)}. If curves principal normal vector and curves binormal vector
are linearly dependent, curve is named Mannheim curve, curve is named Mannheim part-
ner curve, and it is shown as (, ), [21]. Mannheim curves equation is given as;
(s ) = (s) N (s) or (s) = (s ) + B (s ) [12].
There are some following equations among those curves;

T = cos T + sin N

N = B (2.21)

B = sin T + cos N .
ds ds
cos = , sin = (2.22)
ds ds

T = cos T sin B

N = sin T + cos B (2.23)

B = N.
Where (T,T )=, [3] .Let be curvature of , be torsion of , and let be curvature of
, be torsion of . Then, there are the following equations;
ds

=
sin

ds
(2.24)

= cos ds .
ds

= d

ds (2.25)
ds
= ( sin cos ) .
ds

= , [3] (2.26)

Let : I E3 be Mannheim curve, : I E3 be Mannheim partner curve. Suppose
that Frenet frames are respectively given as {T (s), N (s), B(s)} and {T (s), N (s), B (s)}. Let
be the angle which is between T and T , and let be the angle which is between B and W.
In this case, the following equations hold.
C = T (2.27)

sin = cos
(2.28)
cos = sin
(see [15]). If we consider (2.28), (2.8), (2.9), (2.22) and (2.23) will respectively turn the following
equations;

cos = , sin = (2.29)
kW k kW k

1

kT = ,

sin

s

2

k
N = 1 + ,

kW k

(2.30)

1

kB = , ,

cos

s

2

kW k

kC = 1 + .

kW k
T = cot , N = , B = tan , C =
kW k
ds ds
sin = , cos = (2.31)
ds ds
(see [15]).
3. Timelike-Spacelike Mannheim Curve Pairs
Definition 3.1 Let : I IL3 be timelike curve and let : I IL3 be spacelike curve
which has spacelike binormal. Suppose that curves Frenet frames on the point of (s) is
38 S
uleyman S
{T (s), N (s), B(s)} and curves Frenet frames on the point of (s) is {T (s), N (s), B (s)}
If curves principal normal vector and curves binormal vektorare linearly dependent,
curve is called Mannheim curve and curve is called Mannheim partner curve. This pair
curve is briefly symbolized as (, ) and it is named timelike-spacelike Mannheim curve pairs.
Theorem 3.1 The distance which is between (, ) timelike-spacelikeMannheim curve pairs

is constant.
Proof It can be written that;
(s) = (s ) + (s )B (s ).
If this equation is derived with respect to s parameter, we can write that;
ds
T = T + N + B
ds
If we get inner product of the last equation and B , then;
= 0.
From the definition of Euclidean distance, we can write that;
d( (s ), (s)) = k(s) (s )k
= || = cons tan t 2
Theorem 3.2 Let (, ) be timelike-spacelike Mannheim curve pairs. Suppose that curves
and curves Frenet frames are respectively {T, N, B} and {T , N , B }. In this case, there
are the following equations;

T = sinh T + cosh B

N = cosh T + sinh B (3.1)

B = N.
ds ds
sinh = , cosh = (3.2)
ds ds

T = sinh T cosh N

N = B (3.3)

B = cosh T sinh N .
Proof If we derive (s ) = (s) N (s) with respect to s parameter, we can write that;
ds
T = (1 )T (s) B (3.4)
ds
If we get inner product of (3.4) and T , then;
ds
sinh = 1 (3.5)
ds
If we get inner product of (3.4) and B, then;
ds
cosh = (3.6)
ds
If (3.5) and (3.6) are plugged into (3.4), we can write that;
T = sinh T + cosh B
From Frenet formulas, the following equations can be found.
N = cosh T + sinh B,
B = N
Obviously, we have shown that the equation of (3.1). If we arrange this equation with respect
to T and B, we can find the equation of (3.3). If (s) = (s ) + B (s ) is derived with
respect to sparameter, it can be found that;
ds ds
T = T + N
ds ds
If we consider the corresponding value of T from (3.3), the equation of (3.2) is proven.
Theorem 3.3 Let(, )be timelike-spacelike Mannheim curve pairs. Let be curvature of ,
and let be torsion of . In this case, there is the following equation
= 1
Proof If we divide (3.5) to (3.6), we can write that;
1
tanh =

If we get = tanh , the result is proven. 2
Theorem 3.4 Let (, ) be timelike-spacelike Mannheim curve pairs. In this case, there are
the following equations among curvatures.
ds ds
= cosh , = sinh (3.7)
ds ds
d ds ds ds
=
= , = cosh + sinh . (3.8)
ds ds ds ds
40 S
uleyman S
ds
Proof If hT, B i = 0 is derived, then; = cosh ,
ds
ds
If hB, B i = 0 is derived, then; = sinh ,
ds
d ds
If hT, T i = sinh is derived, then; = = ,
ds ds
ds ds
If hN, N i = 0 is derived, then; = cosh + sinh . Therefore, the result is
ds ds
proven. 2
Theorem 3.5 Let(, ) be timelike-spacelike Mannheim curve pairs. Let be curvature of
, let be torsion of , and let be torsion of . The following equation holds.

= . (3.9)

Proof If we get equations from (3.2) and if we multiply side by side (3.5) and (3.6), the
result is proven. 2
Theorem 3.6 Let (, ) be timelike-spacelike Mannheim curve pairs. There is the following
equation between curves W Darboux vector and curves T tangent vector.
ds
W = T (3.10)
ds
Proof We know that W = T B. If we get the corresponding values of T and B from
(3.3), and then if we plug into those values in (3.10),Then, if we get the corresponding values
of and from (3.7), and then if we plug into those values in (3.10), the result is proven. 2
Result 3.1 Let (, ) be timelike-spacelike Mannheim curve pairs. Let be the angle which
is between curves Darboux vector and curves binormal vector. There are the following
equations between and .
If W is a spacelike vector;
sinh = sinh
(3.11)
cosh = cosh
If W is timelike vector;
cosh = sinh
(3.12)
sinh = cosh
= (3.13)
Proof Suppose that W is a spacelike vector, and the following equations hold because of
the equations of (2.13) and (3.1),
C = sinh T cos hB,
T = sinh T + cosh B
If we consider that C = T , we can write that;

sinh = sinh
,
cosh = cosh
Similarly, suppose that W is a timelike vector, and from the equation of (2.15), we can write
that;
C = cosh T sin h B
If we consider that C = T , we can write that;

cos h = sinh
sin h = cosh
If we divide two equations to each other in the equation of (3.12), we can easily write that;
cot h = tanh
If we derive last equation, the following equation holds.
= .
Theorem 3.7 Let (, ) be timelike-spacelike Mannheim curve pairs. There is the following
equation which is between curves W Darboux vector and curves W Darboux vector;
1
W = W N (3.14)
sinh kW k
Proof It can be written T = W + B because Manheim partner curves W

vector is spacelike. If this equation is plugged into (3.10), and if we consider that B = N, the
following equation holds.
ds
W = (W + N )
ds
ds
If the corresponding value of from (3.2) is written in this equation, then;
ds
1
W = W + N (3.15)
sinh
ds
On the other hand, from the equation of (3.7), we can write that = . If the corre-
ds kW k
sponding value of in (3.9) is written in this equation, then;
ds
= .
ds kW k
42 S
uleyman S
ds
If we plug this equation into = , the following equation holds.
ds

= (3.16)
kW k
If we write this value of in (3.15), the result is proven.
Let sT be : I IL3 timelike Mannheim curves (T ) tangent indicators arc length, then;
Z s
sT = ds (3.17)
0
Similarly, (N ) principal normal, (B) binormal, and (C) fixed pol curves arc lengths are respec-
tively; Z s
sN = kW k ds, (3.18)
0
Z s
sB = | | ds, (3.19)
0
Z s
sC = | | ds (3.20)
0
Let kT be (T ) tangent indicators geodesic cuvature on IL3 . Suppose that TT is unit tangent
vector of (T ), then;
kT = kDTT TT k
If T (s) = T (s) tangent indicator is derived with respect to sT parameter, we can write that;
TT = N (3.21)
If we derive one more time and simplify the equation, the following equation holds.

DTT TT = T B (3.22)

From the definition of geodesic curvature;
s
2
kT = 1 + (3.23)
2
If we consider the equations of (2.12) and (3.11), we can write that;
1
kT = (3.24)
cosh
Similarly, if N (s) = N (s) principal normal indicatoris derived with respect to sN parameter,
and the equation of (2.12) is plugged into (3.25), and then the equation of (3.11) is considered,
we can write that;
TN = cosh T + sinh B (3.25)
If we derive one more time, the following equation holds.
1
DTN TN = ( sinh T + kW k N cosh B) (3.26)
kW k

v
u
u 2

kN = t 1 + (3.27)
kW k
If B (s) = B(s) binormal indicator is derived with respect to sB parameter, and if we chose
the positive routing.

DTB TB = T B (3.28)

From the equation of (2.12);
1
kB = (3.29)
sinh
If C (s) = C(s) fixed pol curve is derived with respect to sC parameter, we can write that;
TC = cosh T sinh B
If we derive one more time, and if we consider the equations of (3.11) and (3.13), we can write
that;
kW k
DTC TC = sinh T + cosh B N (3.30)

v
u 2
u kW k
kC = t 1 + (3.31)

Result 3.2 Let (, ) be timelike-spacelike Mannheim curve pairs. Mannheim curves

spherical indicators are (T ) , (N ), (B) and also Mannheim curves fixed pol curve which is
(C), with respect to IL3 , geodesic curvatures of (T ) , (N ),(B) and (C) are respectively;

1

kT =

cosh

v ,

u
u 2

t 1 +

kN =
kW k

1

kB =

sinh

v
u 2

u
t kW k

kC =
1 +

.

44 S
uleyman S
be : I IL3 Mannheim
Let D be : I IL3 Mannheim curves connection on IL3 , let D
3
curves connection on S1 , and let D be : I IL Mannheim curves connection on H02 .
2
Suppose that is unit normal vector space of S12 and H02 , then;
X Y + g(S(X), Y ),
DX Y = D = g(, )
Y + g(S(X), Y ),
DX Y = D = g(, ).
X
Where S is shape operator of S12 and H02 , and corresponding matrix is;

1 0
S=
0 1
(see[?]). Let T be (T ) tangent indicators geodesic curvature on H02 , then;

T = D TT TT
From Gauss Equation, we can write that;
T + g(S(T ), T )T
DTT TT = D TT T T T
where
= g(T, T ) = 1, S(TT ) = TT , and g(S(TT ), TT ) = 1
If we write those values in Gauss equation, and if we consider (3.22), we can write that;
T = B.
D (3.32)
TT T

And also from the equation of (2.12) and (3.11);
T = tanh . (3.33)
Similarly, Let N be geodesic curvature of (N ) principal normal indicator on S12 , then;

N = D TN TN
From Gauss Equation, we can write that;
TN TN + g(S(TN ), TN )N
DTN TN = D
where
= g(N, N ) = +1, S(TN ) = TN , and g(S(TN ), TN ) = +1.
If those values are written in Gauss equation, and if we consider (3.26), we can write that;
TN TN =
D (sinh T cosh B) (3.34)
kW k
If we get norm of the equation;

N = (3.35)
kW k
Let B be geodesic curvature of (B) binormal indicator on S12 , then;

B = D TB TB .
From Gauss equation, we can write that;
TB TB + g(S(TB ), TB )B
DTB TB = D
where
= g(B, B) = +1, S(TB ) = TB and g(S(TB ), TB ) = 1.
If we write those values in Gauss equation, and if we consider (3.28), we can write that;
TB TB = T
D (3.36)

From the equations of (2.12) and (3.11), it can be written that;
B = coth (3.37)
Let C be geodesic curvature of (C) fixed pol curve on S12 , then;

C = D TC TC .
From Gauss equation

TC TC + g(S(TC ), TC )C
DTC TC = D
where
= g(C, C) = +1, S(TC ) = TC and g(S(TC ), TC ) = 1.
If we write those values in Gauss equation, and if we consider (2.13), (3.11) and (3.30), we can
write that;
kW k
C = . (3.38)

Result 3.3 Let (, ) be timelike-spacelike Mannheim curve pairs. Mannheim curves

spherical indicators are (T ) , (N ), (B) and also Mannheim curves fixed pol curve is (C),
with respect to S12 Lorentz sphere or H02 hyperbolic sphere, geodesic curvatures of (T ) , (N ),
(B) and (C) are respectively;
kW k
T = tanh , N = , B = coth , C = .
kW k
Let : I IL3 be spacelike Mannheim curve which has spacelike binormal, and let sT
46 S
uleyman S
be : I IL3 curves (T ) tangent indicators arc length, then,

Z s
sT = ds (3.39)
0
Similarly, arc lengths of (N ), (B ) and (C ) are found;

Z s
r
2 2
sN = ( ) + kW k ds (3.40)
0
Z s
sB = kW k ds (3.41)
0
Z s
sC = ( ) ds. (3.42)
0
On the other hand, let be the angle which is between W and B . unit Darboux vector can
be written as;
C = sin T + cos B
where

sin =
and cos =
tan = .
kW k kW k
C is derived and then if we simplify the equation, we can write that;

( ) =
2
1+
If the values of and are written in the equation, we can write that;
( Note that the values of and are corresponding values of (3.8) and (3.12)

kW k

( ) = 2 (3.43)
kW k
1+
If (3.43) is written in the equation of (3.42), we can easily find that;

Z s
kW k

sC = 2 ds (3.44)
0 kW k
1+
If (3.23) is considered, we can obtain that;

p
2 1
kC
( ) = 2 (3.45)
kC
If (3.45) is written in the equation of (3.42), we can find that;

p
Z s 2 1
kC
sC = 2 ds. (3.46)
0 kC
Result 3.4 Let (, ) be timelike-spacelike Mannheim curve pairs. On the point of (s),
curves Frenet vectors spherical indicator curves and (C ) fixed pol curve which is drawn by
the unit Darboux vector. In terms of IL3 , (C ) fixed pol curves arc lengths are respectively;
Z s

s = ds,

T

0

Z sp

s = |( )2 + kW k2 ds,

N

0
Z s

s B = kW kds,

0

Z s

( kW k
)
s
C = kW k
ds.
0 1 + ( )2
Result 3.5 Let (, ) be timelike-spacelike Mannheim curve pairs, and let kC be timelike
Mannheim curves geodesic curvature. In this case, the arc length of (C ) fixed pol curve is;
p
Z s 2 1
kC
sC = 2 ds.
0 kC
In terms of IL3 , Let kT be T (s) = T (s) tangent indicators geodesic curvature, let sT be
arc parameter , and let TT be unit tangent vector . Then, we can say that;
TT = cosh T + sinh B (3.47)
If we derive one more time the equation of (3.47), it can be written that;
kW k
DTT TT = sinh T + cosh B + N (3.48)

or if we get norm of (3.48), we can write that;
v
u 2
u
t kW k
kT = 1 + (3.49)

Similarly, in terms of IL3 , Let kN be N (s) = N (s) principal normal indicators geodesic
curvature, let sN be arc parameter, and let TN be unit tangent vector. Then, we can say
48 S
uleyman S
that;
1 1
TN = s ( sinh T + cosh B) + s N (3.50)
2 2
1 + kW k 1 +
kW k
If we consider the equations of (3.27) and (3.31), it can be written that;
1 1
TN = ( sinh T + cosh B) + N (3.51)
kC kN
If we derive one more time the equation of (3.51), after simplifying, it can be written that;

( sinh
kC ) + ( kN ) ( k1N )
DTN TN = p T+ p N (3.52)
|( )2 + kW k2 | |( )2 + kW k2 |
cosh !
( kC ) kN
+ p B
|( )2 + kW k2 |
If we get norm of (3.52), we can write that;

v
u 2 2 2
u
u sinh
+ kN + k1N + cosh

u kC kC kN
kN = t . (3.53)
2
( )2 + kW k
In terms of IL3 , Let kB be B (s) = B (s) binormal indicators geodesic curvature, let
sB be arc parameter, and let TB be unit tangent vector. Then, we can say that;

TB = T B (3.54)
kW k kW k
If we consider the equations of (2.12) and (3.11), it can be written that;
TB = cosh T + sinh B
If we derive, TB after simplifying, it can be written that;

DTB TB = ( sinh T + cosh B) + N (3.55)
kW k
or if we get norm of TB , we can write that;
s 2

kB = 1+ (3.56)
kW k
In terms of IL3 , Let kC be C (s) = C (s) fixed pol curves geodesic curvature, let sC be arc
parameter, and let TC be unit tangent vector. Then, we can say that;
TC = cos T + sin B (3.57)
If we derive one more time the equation of (3.57), it can be written that;
kW k
DTC TC = sin T + cos B N (3.58)
( )
v
u 2
u kW k
kC =t 1 + (3.59)
( )
p
If the values of and are written in kW k = |( )2 + ( )2 |, we can find that ( Note
that the value of and are corresponding values of (3.8) and (3.9).)
v
u 2
u t kW k
kW k = 1 +
()
From the equation of (3.31) and (3.45), it can be written that;
kW k (kC )3
= p (3.60)
( ) 2 1
kC
If the value of (3.60) is written in, we can say that;
v 2
u
u
u (kC )3
kC =u
t 1 + p
2 1
kC
Result 3.6 Let (, ) be timelike-spacelike Mannheim curve pairs. In terms of IL3 , curves
(T ), (N ), (B ) spherical indicator curves and (C ) fixed pol curves geodesic curvatures are
50 S
uleyman S
respectively;
r

kW k 2

kT = 1 + ( ) ,

s

[( sinh
kC ) + ( kN )]2 + [( k1N ) ]2 + [( cosh

kC )
2
kN ]

k = ,

N
( )2 + kW k2

s

2

kB = 1 +

,

kW k

v

u 2

u (kC )3

t
k = 1 + p .

C 2 1)
( kC
Let T be : I IL3 spacelike binormal spacelike Mannheim partner curves T (s) =

T (s) tangent indicators geodesic curvature in S12 . Then;

T = D T TT
T
From Gauss equation, it can be written that;
T TT + g(S(TT ), TT )T
DTT TT = D T
where
= g(T , T ) = +1, S(TT ) = TT and g(S(TT ), TT ) = +1.
If those values are written in Gauss equation, and if the equation of (3.1) and (3.48) are
considered, we can say that;
T TT = kW k N
D (3.61)
T

kW k
T =

DTT TT = 0 if and only if T curve geodesic spray is an integral curve. From the equation
of (3.61), we can find that; = 0, = 0 which means is a straight line.
Result 3.7 Let (, ) be timelike-spacelike Mannheim curve pairs. Mannheim curve does
not have any partner curve because Mannheim curveis a straight line.
Let N be geodesic indicator in H02 for N (s) = N (s) principal normal indicator. We
can write that;

N = D T TN
N
From Gauss equation, it can be written that;
T TN + g(S(TN ), TN )N
DTN TN = D N
where
= g(N , N ) = +1, S(TN ) = TN and g(S(TN ), TN ) = +1.
If those values are written in Gauss equation, and if the equations of (3.1) and (3.52) are
considered, we can say that

( sinh ) + ( kN )
T TN
D = p kC
+ cosh T + (3.62)
N
|( )2 + kW k2 |
cosh
( k1N ) ( ) kN
p N + p kC sinh B
|( )2 + kW k2 | |( )2 + kW k2 |

2 2
( sinh
) + ( kN ) ( k1N )
N = p kC + cosh + p
|( )2 + kW k2 | |( )2 + kW k2 |
2 ! 2
1
cosh
( ) kN
+ p kC sinh
|( )2 + kW k2 |

DTN TN = 0 if and only if N curve geodesic spray is an integral curve. In this case, we can
write that;

sinh

kC + kN

r

+ cosh = 0,

2 2

( ) + kW k

1
kN
r = 0,

2 2

( ) + kW k

cosh

kN

r
kC

sinh = 0

2 2
( ) + kW k
This value cannot be 0.
Result 3.8 Let (, ) be timelike-spacelike Mannheim curve pairs. There is no Mannheim

partner curve on hyperbolic sphere to be made Mannheim partner curves (N ) principal
normal indicator geodezic spray is an integral curve.
Let B be geodesic curve in S12 for B (s) = B (s) principal normal indicator. In this
case, we can write that;
52 S
uleyman S

B = D T
TB B

From Gauss Equation, it can be written that;

DTB TB = D
TB TB + g(S(TB ), TB )B
where
= g(B , B ) = 1, S(TB ) = TB and g(S(TB ), TB ) = 1.
If those values are written in Gauss Equation, then, B = N . And if the equation of (3.55) is
considered, we can say that;

D TB TB = ( sinh T + cosh B) (3.63)
kW k

B =
kW k

D curve geodesic spray is an integral curve. In this case, from
TB TB = 0 if and only if B
the equation of (3.63), we can write that;

sinh

= 0,

kW k

cosh = 0

kW k

where = 0. In this case, from the equations of (3.13) and (2.18), = constant. This means

Mannheim curve is a helix.
Result 3.9 Let (, ) be timelike-spacelike Mannheim curve pairs. If Mannheim curve is a

helix, Mannheim partner of is a straight line.
Let C be geodesic curve in S12 for C (s) = C (s). In this case, we can write that;

C = D T TC
C
From Gauss Equation, it can be written that;
T TC + g(S(TC ), TC )C
DTC TC = D C
where
= g(C , C ) = +1, S(TC ) = TC and g(S(TC ), TC ) = 1.

If those values are written in Gauss equation, then,
C = sin T + cos B
And if the equation of (3.57) is considered, we can say that;

T TC = kW k N
D (3.64)
C
( )

If the equation of (3.60) is considered, geodesic curve is,
(kC )3
C = p
2 1
kC

DTC TC = 0 if and only if C curve geodesic spray is an integral curve. In this case, from
the equation of (3.64), we can write that; = = 0. And from the equation of (3.9), = 0.
Result 3.10 Let (, ) be timelike-spacelike Mannheim curve pairs. Mannheim curve does
not have any partner curve because Mannheim curveis a straight line.
Result 3.11 Let (, ) be timelike-spacelike Mannheim curve pairs. In terms of S12 or H02 ,
curves (T ), (N ) and (B ) spherical indicator curves and (C ) fixed pol curves geodesic
curvatures are respectively;
kW k
T = C =

2 2
( sinh
) + ( kN ) ( k1N )
N = p kC + cosh + p
|( )2 + kW k2 | |( )2 + kW k2 |
2 ! 2
1

( cosh
kC ) kN

+ p sinh
|( )2 + kW k2 |
(kC )3
B = , C = .
kW k kC
References
[1] Akutagawa K. and Nishikawa S., The Gauss map and spacelike surfaces with prescribed
mean curvature in Minkowski 3-space, Tohoko Math. J., 42, 67-82, 1990.
[2] Azak A. Zeynep, Timelike Mannheim Partner curves in L3 , Sakarya Univ., Journal of
Sci.II, 35-45, 2009.
[3] Bilici M., C The natural lift curves and the geodesic sprays for
alskan M. and Aydemir I.,
the spherical indicatrices Of the pair of evolute-involute curves, International Journal of
App. Math., (11), 415-420, 2002.
54 S
uleyman S
[4] Burke J.F., Bertrand curves associated with a pair of curves, Mathematics Magazine, 34(1),
60-62, 1960.
[5] C , Hacsalihoglu H. H., Some characterizations for the natural
alskan M., Sivridag A.I.

lift curves and the geodesic sprays, Ankara Univ., Fac. of Sci.,commun., 235-242, 1984.
[6] Ferrandez A., Lucas P., On surfaces in the 3-dimensional Lorentz - Minkowski space, Pasific
J. of Math., 152(1), 1992.
[7] Gorg ul
u A., Ozdamar E., A generalizations of the Bertrand curves as general inclined
n
curves in E , Communications de la Fac. Sci. Uni. Ankara, Series A1, 53-60, 1986.
[8] Ergun E. and C - integral curves and M
alskan M., On the M - geodesic sprays in Minkowski
3-space, Int.J.Contemp. Math. Sci., 6(39), 1935-1939, 2011.
[9] Hacsalioglu H.H., Differential Geometry, In onu Univ., Fac.of Arts and Sci. pub., (7), 1983.
[10] Liu H. and Wang F., Mannheim partner curves in 3-space, Journal of Geometry, 88(1-2),
120-126(7), 2008.
[11] Oneill B., Semi Riemann Geometry, Academic Press, New York, London, 1983.
[12] Orbay K. and Kasap E., On Mannheim partner curves in E3 , International Journal of
Physical Sciences, Vol.4, 261-264, 2009.
[13] Ratcliffe J. G., Foundations of Hyperbolic Manifolds, Springer-Verlag New York, Inc., New
York, 736, 1994.
[14] Senyurt S. and Bektas O., Timelike-Spacelike Mannheim partner curves in E3 , Interna-
tional Journal of The Physical Sciences, 7(1), 100-106, 2012.
[15] Senyurt S., Natural lifts and the geodesic sprays for the spherical indicatrices of the
Mannheim partner curves in E3 , International Journal of The Physical Sciences, 7(16),
2414-2421, 2012.
C
[16] Sivridag A.I., alskan M., On the M-integral curves and M-geodesic sprays, Erciyes Univ.,
J. of Sci., 7(1-2), 1283-1287, 1991.
[17] Thorpe J. A., Elementary Topics in Differential Geometry, New York, Heidelberg, Berlin:
Springer-Verl., 1979.
[18] Turgut A., 3- Boyutlu Minkowski Uzaynda Spacelike ve Timelike Regle Y uzeyler, Gazi
Univ. Institute of Sci., Ph D thesis, 1995.
[19] Turgut A., Esin E., Involute- evolute curve couples of higher order in Rn and their hori-
zontal lift in T IRn , Commun. Fac.Sci. Univ. Ankara, Series A1, 41(3), 125-130, 1992.
[20] Ugurlu H.H., On the geometry of timelike surfaces, Commun. Fac. Sci. Ank., Series A1
(46), 211-223, 1997.
[21] Wang F. and Liu H., Mannheim partner curves in 3-space, Proceedings of the Eleventh
International Workshop on Diff. Geom., 25-31, 2007.
[22] Woestijne V.D.I., Minimal surfaces of the 3-diemensional Minkowski space, Proc. Congres
Geometrie differentielle et aplications Avignon (30 May 1988), World Scientific Publishing.
Singapore, 344-369.
Smarandache-R-Module and Mcrita Context
N.Kannappa
(PG & Research Department of Mathematics, TBML College, Porayar-609307, TamilNadu, India)
P.Hirudayaraj
(Department of Mathematics, RVS College of Arts and Science, Karaikal-609609, Puducherry, India)
E-mail: sivaguru91@yahoo.com, hiruthayaraj99@gmail.com
Abstract: In this paper we introduced Smarandache-2-algebraic structure of R-module

namely Smarandache-R-module. A Smarandache-2-algebraic structure on a set N means
a weak algebraic structure A0 on N such that there exist a proper subset M of N , which
is embedded with a stronger algebraic structure A1, stronger algebraic structure means
satisfying more axioms, by proper subset one understands a subset from the empty set, from
the unit element if any, from the whole set. We define Smarandache-R-module and obtain
some of its characterization through S-algebra and Morita context. For basic concept we
refer to Raul Padilla.
Key Words: R-module, Smarandache-R-module, S-algebras, Morita context and Cauchy

modules.
AMS(2010): 53C78.
1. Preliminaries
Definition 1.1 Let S be any field. An S-algebra A is an (R, R)-bimodule together with module
morphisms : AR A A and : R A called multiplication and unit linear maps respectively
such that
1A
A R A R A = A R A A with ( 1A ) = (1A ) and
1A
1A
R = A R A
A with ( 1A ) = (1A ).
1A
Definition 1.2 Let A and B be S-algebras. Then f : A B is an S-algebra homomorphism

if B (f f ) = f A and f A = B .
Definition 1.3 Let S be a commutative field with 1R and A an S-algebra M is said to be a

left A-module if for a natural map : A R M M , we have (1A ) = ( 1M ).
Definition 1.4 Let S be a commutative field. An S-coalgebra is an (R, R)-bimodule C with R-

linear maps : C C R C and : C R, called comultiplication and counit respectively such
1 Received September 6, 2014, Accepted May 10, 2015.
56 N.Kannappa and P.Hirudayaraj
1C 1C
that C C R C C R C R C with (1C ) = (1C ) and C C R C R
1C 1C
with (1C ) = 1C = ( 1C ) .
Definition 1.5 Let C and D be S-coalgebras. A coalgebra morphism f : C D is a module

morphism if it satisfies D f = (f f ) C and D f = C .
Definition 1.6 Let A be an S-algebra and C an S-coalgebra. Then the convolution product is
defined by f g = (f g) with 1HomR (C, A) = (1R ) for all f, g HomR (C, A).
Definition 1.6 For a commutative field S, an S-bialgebra B is an R-module which is an algebra

(B, , ) and a coalgebra (B, , ) such that and are algebra morphisms or equivalently
and are coalgebra morphisms.
Definition 1.7 Let R, S be fields and M an (R, S)-bimodule. Then, M = HomR (M, R)
is an (S, R)-bimodule and for every left R-module L, there is a canonical module morphism

M M
L : M R L HomR (M, L) defined by L (m l)(m) = m (m)l for all m M, m M
and l L. If M
L is an isomorphism for each left R-module L, then R MS is called a Cauchy
module.
Definition 1.8 Let R, S be fields with multiplicative identities M , an (S, R)-bimodule and N ,
an (R, S)-bimodule. Then the six-tuple datum K = [R, S, M, N, h, iR , h, iS ] is said to be a Morita
context if the maps h, iR : N S M R and h, iS : M R N S are binmodule morphisms
satisfying the following associativity conditions:
m hn, miR = hm , nis m and hn, miRn = nhm, n is
h, iR and h, iS are called the Morita maps.
2. Smarandache-R-Modules
Definition 2.1 A Smarandache-R-module is defined to be such an R-module that there exists

a proper subset A of R which is an S-Algebra with respect to the same induced operations of R.
3. Results
Theorem 3.1 Let R be a R-module. There exists a proper subset A of R which is an S-coalgebra
iff A is an S-algebra.
Proof Let us assume A is an S-algebra. For proving that A is an S-coalgebra we check

the counit conditions as follows:
1A A
: A A R S A R A R.
Smarandache-R-Module and Mcrita Context 57
Next, we check the counit condition as follows:
1A EndS (A)
:A A R S R S A R (A R A) R A
1A A 1 A
A
(A R A) R (A R A) A
R
(A R A) R (A R A)
1A A
(A R A) R A (A R A) R A A R A A.
Thus, A is an S-coalgebra.
Conversely, Let us assume A is an S-coalgebra. Now to prove that A is an S-algebra, we
check the unit conditions as follows
EndS (A)
:R A R A 1A A A.
We check the multiplication conditions as follows A is a Cauchy module. Notice that
A R A R,
1A EndS (A)
A A R A R R A R A R A R R A A ,
1
: A R A A A R A R R A A .
Thus, A is an S-algebra. By definition, R is a smarandache R-module. 2

Theorem 3.2 Let R be an R-module. Then there exists a proper subset EndS (M ) of R which
is an S-algebra.
Proof Let us assume that R be an R-module. For proving that EndS (M ) is an S-coalgebra
which satisfies multiplication and unit conditions : EndS (M ) R EndS (M ) EndS (M ) and
: R EndS (M ), we check the comultiplication condition as follows:
1End(M ) n
: EndS (M ) EndS (M )R EndS (M ) R EndS (M ).
Next, we check the counit conditions as follows:
1End(M )
: EndS (M ) EndS (M ) R R EndS (M ) R EndS (M )

HomR (M, M ) R HomR (M, M )
M M
(M R M ) R (M R M ) R R R R.
Thus EndS (M ) is an S-coalgebra. By Theorem 3.1, EndS (M ) is an S-algebra. Hence, R is a

Smarandache R-module. 2
Theorem 3.3 Let R be an R-module. Then there exists a proper subset M R M of R which
is an S-algebra.
Proof For proving that M R M is an S-algebra, we check the multiplication and unit
conditions as follows:

: (M R M ) (M R M ) M R (M R M ) R M
1M M 1M
M R R R M
1M M
M R M .
As M is a Cauchy module, we have

: R EndS (M ) M R M ,
which implies that M R M is an S-algebra. Hence, R is a Smarandache R-module. 2
Theorem 3.4 Let R be an R-module. Then there exists a proper subset the datum [R, M, N, h, iR ]
a morita context (M R N ) of R which is an S-algebra.
Proof Let us assume that R be an R-module. For proving that M R N is an S-algebra,

we have
: (M R N ) R (M R N ) M R (N R M ) R N
1M h,i1N
M R R R N M R N,
which shows that the multiplication condition is satisfied.
Also, since M and N are Cauchy R-modules, there exist maps
EndR (M ) : R M R M and EndS (N ) : R N R N
that can be used to prove the unit condition as follows:
EndS (M)EndS (N )
: R R R R (M R M ) R (N R N )
1M N
(M R N ) R (M R N )
1M N
(M R N ) R (M R N )
1M N
R R (M R N )

R R (M R N ) (M R N ),
which implies that M R N is an S-algebra. By definition, R is a Smarandache R-module.
Theorem 3.5 Let R be an R-module. Then there exists a proper subset the datum [R, M, N, h, iR]
a morita context M R N of R which is an S-coalgebra.
Proof Let us assume that R be an R-module. For proving that (M R N ) is an S-coalgebra,

Smarandache-R-Module and Mcrita Context 59
we have
: M R N = (M R N ) R (RR )
1M N EndS (M)EndS (N )
(M R M ) R (N R N )
1M N
(M R N ) R (M R N ) R (M R N )
1M N
(M R N ) R (M R N ) R (M R N )
1M N h,iR
(M R N ) R (M R N ) R R

(M R N ) R (M R N ).
Also, we have the counit condition as follows:
1M N EndS (M)
: M R N (M R N ) R R (M R N ) R (M R M )
h.iR 1M M M
R R M R M M R M R,
which implies that = M R N is an S-coalgebra. Hence, R is a Smarandache R-module. 2
Theorem 3.6 Let R be an R-module. Then there exists a proper subset the datum [R, M, N, h, iR]
a Morita context iff M R N is an S-bialgebra.
Proof First, if M R N is an S-bialgebra by Theorem 3.5, we know that M R N is an

S-algebra and M R N is an S-coalgebra. Hence by definition, R is a Smarandache R-module.
If M R N is an S-bialgebra, we have the map
= h.iR : M R N R.
Associativity of the map = h, iR holds because the diagram

(M R N ) R M M R (N R M )
1M 1M
M
is commutative. Hence the datum [R, M, N, h, iR] is a Morita context. 2
References
[1] Atiyah M.F. and Macdonald I.G., Introduction to Commutative Algebra, Addison-Wesley
Publishing Co., Reading, Mass- London- Don Mills, Ont., 1969.
[2] Avramov L.L. and Golod E.S, The homology of algebra of the Koszul complex of a local
Gorenstein ring, Mat Zametki, 9 (1971), 53-58.
[3] Benson D.J. and Greenlees J.P.C., Commutative algebra for cohomology rings of classifying
spaces of compact Lie groups, J.Pure Appl.Algebra, 122(1997).
[4] Bousfield A.K., On the homology spectral sequence of a cosimplicial space, Amer. J. Math.,
109 (1987), No. 2, 361-394.
[5] Dwyer W.G., Greenless J.P.C. and Iyengar S., Duality in Algebra and Topology on 12 Oct
2005.
[6] Florentin Smarandache, Special Algebaric Structures ,University of New Mexico, USA(1991).
[7] Klein J.R., The dualizing spectrum of a topological group, Math. Ann., 319 (2001), No.
3, 421-456.
[8] Pliz Gunter, Near-Rings, North Holland Press, Amsterdam(1977).
[9] Raul Padilla, Smarandache Algebraic Structures, Universidade do Minho,Portugal (1999).
[10] Shipley B., HZ-algebra spectra are differential graded algebras, 2004.
Generalized Vertex Induced Connected Subsets of a Graph
B.Chaluvaraju, Medha Itagi Huilgol, N.Manjunath and S.Syed Asif Ulla

(Department of Mathematics, Central College Campus, Bangalore University, BANGALORE - 560 001, India)
E-mail: bchaluvaraju@gmail.com, medha@bub.ernet.in
Abstract: Let k be a positive integer. A graph G = (V, E) is said to be a k - connected

graph if for any given subset S of V (G) with |S| = k, the subgraph induced by S is connected.
In this paper, we explore some properties of k - connectedness and its minimality conditions
with respect to other graph theoretic parameters.
Key Words: Graph, subgraph, k - connected graph, minimal k -connected graph.

AMS(2010): 05C15, 05C69.
1. Introduction
In this article, we consider finite, undirected, simple and connected graphs G = (V, E) with
vertex set V and edge set E. As such p =| V | and q =| E | denote the number of vertices and
edges of a graph G, respectively. In general, we use hXi to denote the sub graph induced by the
set of vertices X V . N (v) and N [v] denote the open and closed neighborhoods of a vertex
v, respectively. A non-trivial graph G is called connected if any two of its vertices are linked
by a path in G. A graph G is called n-connected (for n N ) if |V (G)| > n and G X(the
graph that results from removing all vertices in X and all edges incident with these vertices) is
connected for any vertex set X V (G) with |X| < n. The greatest integer n such that G is n
- connected is called the connectivity (G) of G. A cut-edge or cut-vertex of G is an edge or
a vertex whose deletion increases the number of components. Unless mentioned otherwise for
terminology and notation the reader may refer [1] and [5].
The general problem consists of selecting a set of land parcels for conservation to ensure
species availability. This problem is also related to site selection, reserve network design and
corridor design. Biologists have highlighted the importance of addressing negative ecological
impacts of habitats fragmentation when selecting parcels for conservation. Ways to increase
the spatial coherence among the set of parcels selected for conservation have been investigated.
Conservation planning via k -connected graph model is an important conservation method, in
this model they increase the genetic diversity and allow greater mobility for better response to
predation and stochastic events such as fire, as well as long term climate change. This motivated
us to study k - connectedness in the following manner:
1 Received October 26,2014, Accepted May 12, 2015.

62 B.Chaluvaraju, Medha Itagi Huilgol, N.Manjunath and S.Syed Asif Ulla
For any positive integer k. A graph G is said to be a k - connected graph if for any given
subset S of V (G) with |S| = k, the subgraph induced by S is connected.
For more detail, we refer [2]-[4], [6] [10] and [13].
2. k -Connected Graphs
Proposition 2.1 For any graph G with p 1 vertices is a 1 - connected graph.
Proposition 2.2 For a given p 3 vertices, there exist a 3 - connected graph of a graph G.
Proof Removal of t independent edges from a complete graph on p vertices results into a
3 - connected graph, where 0 t p2 if p is even and 0 t p1
2 if p is odd. 2
Proposition 2.3 Let be the number of edges required to make a totally disconnected graph
which is a 3 - connected graph and hence

p2 2p if p is even
2
(K p ) =
p2 2p+1 if p is odd
2
For a complete bipartite graph Km,n , the number of edges to be added to make it a 3 - connected
graph is given by (K m ) + (K n ).
Proposition 2.4 In general the number of edges required to make complete n- partite graph
Kx1 ,x2 ,x3 ,...,xn as a 3 - connected graph is given by
(K x1 ) + (K x2 ) + + (K xn ).
Proposition 2.5 The complete bipartite graph Km,n is a 3 - connected graph if m = 1, 2 and
n = 1, 2.
Theorem 2.1 For any graph G with p 3 vertices is a 3 - connected graph if and only if
deg(vi ) p 2 for all vi V (G).
Proof Let G be a 3 - connected graph. Suppose on contrary, there exists vj such that
deg(vj ) p 3. Let v1 and v2 be any two vertices which are not adjacent to vj , thus the graph
induced by v1 , v2 and vj is not connected, which is a contradiction. Hence deg(vi ) p 2 for
all vi G.
Conversely, suppose deg(vi ) p 2 for all vi G, let v1 , v2 , v3 be any three vertices. Then
v1 is adjacent to at least one of v2 and v3 . v2 is adjacent to at least one of v1 and v3 . Also v3
is adjacent to at least one of v1 and v2 . Therefore G is a 3 - connected graph. 2
Generalized Vertex Induced Connected Subsets of a Graph 63
Theorem 2.2 If a graph G is a k - connected graph, then

t if p = t(k 1) + 1
(G)
t + 1 if p = t(k 1) + r + 1, 1 r k 2
Proof Let G be a k connected graph with p = t(k 1) + 1 vertices. Suppose on contrary

that (G) = s < t. Let v be a vertex with deg(v) = s. Now we partition the remaining p 1
vertices into t vertex disjoint sets such that each set contains a vertex adjacent to v. Since
s < t, there exists at least one set N which has no vertex adjacent to v. Then hN {v}i is a
disconnected subgraph on k vertices, which is a contradiction.
Now, let p 1 = t(k 1) + r. Suppose on contrary that (G) = s < t + 1, let v be a
vertex such that deg(v) = s. Now partition the remaining p 1 vertices excluding the vertex v
in t + 1 number of vertex disjoint subsets having t subsets with cardinality k 1 and one with
cardinality r in such a way that each subset contains exactly one vertex adjacent to v. Since
deg(v) < t + 1, there exist at least one subset having no vertex adjacent to v. If the cardinality
of such a subset, say T is k 1 then hT {v}i is a disconnected subgraph on k - vertices, again
a contradiction. If the cardinality of such a subset, say D is r, then take a vertex which is
adjacent to v from a subset A with cardinality k 1 to D and any one vertex from D to A, then
hA {v}i is a disconnected subgraph which is induced by k - vertices, a contradiction. Hence

t if p 1 = t(k 1)
(G)
t + 1
2
if p 1 = t(k 1) + r, 1 r k 2
Theorem 2.3 Let G be a k - connectedHence the result follows. Then G is a k+1 - connected
graph with 2 k p 1.
Proof On contrary, suppose G is a k - connected graph but not a k+1 - connected

one. Let S be set of k + 1 vertices on which the graph induced has more than one component.
Clearly a subset T consisting of k vertices on which the graph induced is disconnected which
is a subgraph of G, which is a contradiction. Hence G is also a k+1 - connected graph with
2 k p 1. 2
Theorem 2.4 A graph G is a k -connected graph for all k, 1 k p if and only if G is
isomorphic to Kp .
Proof Let G be a k - connected graph for all k, 1 k p. Clearly G is isomorphic to a

complete graph Kp as G is a 2 - connected graph.
On the other hand, let G be a complete graph on p vertices. In Kp , every pair of vertices
are adjacent. Hence G is a 2 - connected graph. As we have proved, every 2 - connected
graph is t - connected graph for all t, 3 t p. Therefore G is a k - connected graph with
1 k p. 2
Theorem 2.5 For any Square tree T 2 of a tree T with diameter d(T ) 5 is a p1 - connected
graph.
Proof Let T be any tree with diameter d(T ) 5. Consider any four non pendent vertices
vi+1 , vi+2 , vi+3 , vi+4 in T such that vi+1 is adjacent to vi+2 , vi+2 is adjacent to vi+3 , vi+3 is
adjacent to vi+4 , then in square of T , vi+2 is adjacent to vi+1 , vi+3 , vi+4 and vi+3 is adjacent
to vi+1 , vi+2 , vi+4 . Removal of vi+2 and vi+3 disconnects the square and since no vertex is a
cut vertex, removal of any one vertex does not disconnect T 2 . Hence square of any tree T with
diameter d(T ) 5 is a p1 - connected graph. 2
2
Proposition 2.6 For any integer m 1, K1,m , K1,m are 2 - connected graphs.
Proof Let K1,m be any star. Then the square of a star is a complete graph on m + 1
vertices and hence a 2 - connected graph. 2
3. Minimality Conditions on k -Connected Graphs
A k - connected graph G is said to be a vertex minimal k - connected graph if G is not a

k1 - connected graph. A vertex minimal k - connected graph G is said to be a partially
vertex - edge minimal k - connected graph if G e is not a k - connected graph for some
e E(G). A vertex minimal k - connected graph G is said to be a totally vertex - edge
minimal k - connected graph if G e is not a k - connected graph for every e E(G). For
more details, refer [12].
Proposition 3.1 For any cycle Cp ; p 4 vertices, the number of edges to be added to make it
a 3 - connected graph is given by

p2 4p if p is even
2
q 2
p 4p+1
if p is odd
2
where, the equality holds when the resulting graph is a partially vertex - edge minimal 3 -
connected graph.
Proof Let Cp ; p 4 vertices be a cycle. Then we have the following cases.
Case 1. Suppose p is even. In any 3 - connected graph, the degree each vertex is at least
p 2. Hence the number of edges in any 3 - connected graph is always greater than or
equal to p(p2)
2 . Therefore the number of edges to be added to Cp is greater than or equal to
p2 2p p2 4p
2 p = 2 .
Case 2. Suppose p is odd. In any 3 - connected graph on odd number of vertices, the degree
of each of the p 1 vertices is at least p 2 and the degree of one vertex is p 1. Hence the
number of edges in any 3 - connected graph on odd number of vertices is always greater than
2
or equal to (p1)
2 . Hence the number of edges to be added to Cp is greater than or equal to
p2 2p+1
2 p = p2 4p+1
2 . 2
Proposition 3.2 For any path Pp ; p 4 vertices, the number of edges to be added to make it
a 3 - connected graph is given by

p2 4p + 1 if p is even
2
q 2
p 4p+1
+ 1 if p is odd
2
where the equality holds when the resulting graph is a partially vertex - edge minimal 3 -
connected graph.
Proof The proof follows on the same lines as in the above proposition. 2
Theorem 3.1 A connected graph G is a vertex minimal p - connected graph if and only if it
has at least one cut vertex.
Proof Let a connected graph G be a p - connected graph, that is, G is not a p1 -

connected graph. There exist a vertex v such that the graph induced by V (G)v is disconnected.
Hence v is cut vertex. Conversely, let G be a connected graph with a cut vertex, say v, therefore
the subgraph induced by the vertices V (G) v is disconnected. Hence the graph G is a vertex
minimal p - connected graph. 2
Theorem 3.2 For a given k = 2l + 1, l 3, there exists k - connected graph.
Proof Let k = 2l + 1, l 3 and G be a 3 - connected graph on k 3 vertices with

V (G) = {1, 2, 3, , k3} and let G be a graph with V (G ) = {1, 2, 3, , k3, k2} obtained

by adding a vertex k 2 and making it adjacent to all the vertices of G. Now take prism of G ,

label the vertices of second copy of G in the prism as {f (1), f (2), f (3), , f (k 3), f (k 2)}
such that f (i) is the mirror image of i and remove the edge (k 2, f (k 2)) from the prism. In
the resulting graph H (say), the subgraph induced by any subset S V (H)/{k 2, f (k 2)}
containing k 1 vertices is connected. The subgraph induced by V (G) {k 2, f (k 2)}
disconnected on k 3 + 1 + 1 = k 1 vertices and hence every subgraph induced by k vertices
is connected. Hence H is k - connected. 2
Observation 3.1 The graph obtained in the above theorem is regular when G is a partially
vertex - edge minimal 3 - connected graph, which is having even order.
1 3 f (3) f (1)
4 f (4)
2 f (2)
5 f (5)
Figure 1. Prism of a 3 - connected graph having odd order

For illustration, we construct the above prism of a 3 - connected graph having odd order,
where prism of a graph G is defined as the cartesian product G K2 ; that is, take two disjoint
copies of G and add a matching joining the corresponding vertices in the two copies, [8].
Theorem 3.3 For any vertex minimal k - connected graph can be embedded in a vertex
minimal k+i - connected graph, where i 0.
Proof Let G1 and G2 be two vertex minimal k and k+i - connected graphs. Now we
construct a vertex minimal k+i - connected graph in which G1 is an induced vertex minimal
k - connected subgraph. Make each vertex of G1 adjacent to each vertex in G2 and let the
resulting graph be G. Let S be any set of k + i vertices from G. In the following cases we prove
the graph induced by S is connected.
Case 1. Suppose S V (G1 ) V (G2 ) = 6 , then the graph induced by S is connected since
each vertex in S V (G1 ) is adjacent to every vertex of S V (G2 ).
Case 2. Suppose S V (G1 ) 6= and S V (G2 ) = , then |S| k and S V (G1 ), the graph
induced by S is connected as G1 is vertex minimal k - connected.
Case 3. Suppose S V (G1 ) = and S V (G2 ) 6= , then the graph induced by S is connected
since S is completely contained in V (G2 ) and G2 is a vertex minimal k+i - connected graph.
In all the three cases the graph induced by S is connected and G is not k+i1 - connected
graph since G2 is a vertex minimal k+i - connected graph. Hence the graph G is a vertex
minimal k+i - connected graph having vertex minimal k - connected graph G1 as its induced
subgraph.
Thus the result follows. 2
Theorem 3.4 A connected graph G is a partially vertex - edge minimal p - connected graph
if and only if it has at least one cut edge.
Proof Let G be a connected graph with a cut edge say e = uv. Here u is a cut vertex and
also G e is disconnected, hence G is a partially vertex - edge minimal p - connected graph.
Conversely, let G be a partially vertex - edge minimal p - connected graph then G e is not
a p - connected graph. Hence e is a cut edge. 2
To prove our next result we make use of the following observations.
Observation 3.2 Removal of p2 independent edges from a complete graph Kp on even number
of vertices results into a partially vertex - edge minimal 3 - connected graph having p(p2)
2
edges.
p1
Observation 3.3 Removal of 2 independent edges from a complete graph on odd number
(p1)2
of vertices results into a partially vertex - edge minimal 3 - connected graph having 2
edges.
Observation 3.4 Partially vertex - edge minimal 3 - connected graph having even order is
a regular graph with regularity p 2.
Theorem 3.5 Let G1 and G2 be two partially vertex - edge minimal 3 - connected graph. If
V (G1 ) = V (G2 ), then G1 is isomorphic to G2 .
Proof Let G1 and G2 be partially vertex - edge minimal 3 - connected graph having same
order. Then, there are following cases:
Case 1. Suppose p is even. As the graphs G1 and G2 are partially vertex - edge minimal 3
connected graphs, deg(v) = p 2, for all v V (G1 ) and deg(v) = p 2 for all v V (G2 ).
Clearly G1 is isomorphic to G2 .
Case 2. Suppose p is odd. As the graphs G1 and G2 are partially vertex - edge minimal 3 -
connected graphs, in the graphs G1 and G2 degree of each of p 1 vertices is p 2 and degree
of one vertex is p 1. Hence in this case also G1 is isomorphic to G2 . 2
Theorem 3.6 Any graph G with order p is a totally vertex - edge minimal p - connected
graph if and only if G is isomorphic to a tree on p vertices.
Proof Let G be a graph of order p which is a strongly critical p - connected graph, that
is, G e is not p - connected for all e V (G) and G is not a p1 - connected graph. The
first condition in a totally vertex - edge minimal p - connected graph which implies that every
edge in G is a bridge and the second condition in a totally vertex - edge minimal p - connected
graph which again implies that every vertex in G is a cut vertex, clearly G is isomorphic to a
tree on p vertices.
Conversely, let G is isomorphic to a tree on p vertices. Since every internal vertex is a cut
vertex, we have a disconnected induced subgraph on p 1 number of vertices and since every
edge is a bridge, G e is not a p - connected graph for all e V (G). Hence G is a totally
vertex - edge minimal p - connected graph. 2
Theorem 3.7 Any graph G having even number of vertices is a totally vertex - edge minimal
3 - connected graph if and only if deg(vi ) = p 2 for all vi V (G).
Proof Let G be a totally vertex - edge minimal 3 - connected graph on even number of
vertices, implies deg(v) p 2 for all v V (G) from the Theorem 2.1. Suppose deg(v) > p 2
for some v V (G), i.e., deg(v) = p 1. There exist a vertex w adjacent to v such that
deg(w) = p 1. The graph G vw is still a 3 - connected graph. Hence G is not a totally
vertex - edge minimal 3 - connected graph, which is a contradiction. Hence deg(v) = p 2
for all v V (G).
Conversely, let G be a graph such that deg(v) = p 2 for all v V (G). For every vertex
in G there exist an unique non adjacent vertex in G. Hence in G e there exist two vertices
say v and w non adjacent to some vertex say u. The graph induced by these three vertices is
disconnected and hence the graph is a totally vertex - edge minimal 3 - connected graph. 2
Observation 3.5 Any complete graph on p 3 vertices is a 3 - connected graph but not a
partially vertex - edge minimal and totally vertex - edge minimal 3 - connected graph.
Acknowledgement
We thank Prof.E.Sampathkumar for introducing this new concept and giving us his valuable
time for discussing about this concept.
References
[1] J. A. Bondy and U. S. R.Murty, Graph Theory with Applications, Macmillan, London,
1976.
[2] E.A.Bender, P.J.Cameron, A.M.Odlyzko and L.B.Richmond, Connectedness classes and
cycle index, Probab. Comput., 8(1-2)(1999) 31-43.
[3] Carla P. Gomes, Willem-Jan van Hoeve and Ashish Sabharwal, Connections in networks:
A hybrid approach, In CPAIOR, 5015(2008)303-307.
[4] S.A.Choudum and S.Sivagurunathan, Optimal fault - tolerant networks with a server,
Networks, 35(2) (2000) 157 - 160.
[5] F.Harary, Graph theory, Addison Wesley,Reading Mass., 1969.
[6] F.Harary and J.P.Hayes, Node fault tolerance in graphs, Networks, 27 (1996) 19 - 23.
[7] J.P.Hayes, A graph model for fault - tolerant computing systems, IEEE Trans. on Comp.,
C - 25 (1976) 875 - 884.
[8] G.Sabidussi, Graph derivatives, Math.Z., 76(1961), 385-401.
[9] E.Sampathkumar, Report on UGC Emeritus Fellowship, No.F.6-120/2000(SA), 2004.
[10] E.Sampathkumar, Connectivity of a graph-A generalization, J.Combinatorics, Information
and System Sciences, Vol.9, No. 2, 71-78(1984).
[11] S.Spacapan, Connectivity of Cartesian products of graphs, Appl. Math. Lett., 21(2008),
682 - 685.
[12] H.B.Walikar, B.D.Acharya and E.Sampathkumar, Recent developments in the theory of
domination in graphs, MRI Lecture Notes in Math., 1 (1979).
[13] J.C.Williams and Stephanie A. Snyder, Restoring habitat corridors in fragmented land-
scapes using optimization and percolation models, Environmental Modeling and Assess-
ment, 10(3)(2005), 239-250.
b-Chromatic Number of Splitting Graph of Wheel
Samir.K.Vaidya
(Saurashtra University, Rajkot - 360005, Gujarat, India)
Minal.S.Shukla
(Atmiya Institute of Technology and Science, Rajkot - 360005, Gujarat, India)
E-mail: samirkvaidya@yahoo.co.in, shuklaminal19@gmail.com
Abstract: A proper k-coloring is called a b-coloring if there exits a vertex (b-vertex) that
has neighbour(s) in all other k 1 color classes. The largest integer k for which G admits
a b-coloring is called the b-chromatic number denoted as (G). If b-coloring exists for every
integer k satisfying (G) 6 k 6 (G) then G is called b-continuous. The b-spectrum Sb (G)
of a graph G is the set of k integers(colors) for which G has a b-coloring. We investigate
b-chromatic number of the splitting graph of wheel and also discuss its b-continuity and
b-spectrum.
Key Words: b-Coloring, b-continuity, b-spectrum.

AMS(2010): 05C15, 05C76.
1. Introduction
A proper k-coloring of a graph G = (V (G), E(G)) is a mapping f : V (G) {1, 2, , k} such

that every two adjacent vertices receives different colors. The chromatic number of a graph
G is denoted by (G), is the minimum number for which G has a proper k-coloring. The set
of vertices with a specific color is called a color class. A b-coloring of a graph G is a variant
of proper k-coloring such that every color class has a vertex which is adjacent to at least one
vertex in every other color classes and such a vertex is called a color dominating vertex. If v
is a color dominating vertex of color class c then we denote it as cdv(c) = v. The b-chromatic
number (G) is the largest integer k such that G admits a b-coloring with k colors. The
concept of b-coloring was originated by Irving and Manlove [1] and they also observed that
every coloring of a graph G with (G) colors is obviously a b-coloring. In the same paper they
have introduced the concepts of b-continuity and b-spectrum. If the b-coloring exists for every
integer k satisfying (G) 6 k 6 (G) then G is called b-continuous and the b-spectrum Sb (G)
of a graph G is the set of k integers(colors) for which G has a b-coloring. Kouider and Maheo [2]
have obtained lower and upper bounds for the b-chromatic number of the cartesian products of
two graphs while Vaidya and Shukla [3,4,5,6] have investigated b-chromatic numbers for various
70 Samir.K.Vaidya and Minal.S.Shukla
graph families. The concept of b-coloring has been extensively studied by Faik [7], Kratochvil
et al.[8], Alkhateeb [9] and Balakrishnan et al. [10].
Definition 1.1 The splitting graph S (G) of a graph G is obtained by adding new vertex v
corresponding to each vertex v of G such that N (v) = N (v ), where N (v) and N (v ) are the
neighborhood sets of v and v respectively in S (G).
Here we investigate b-chromatic number for splitting graph of wheel.
Definition 1.2([1]) The m-degree of a graph G, denoted by m(G), is the largest integer m such
that G has m vertices of degree at least m 1.
Proposition 1.3([1]) If graph G admits a b-coloring with m-colors, then G must have at least
m vertices with degree at least m 1.

3, n is even
Proposition 1.4 Let Wn = Cn + K1 . Then (Wn ) =
4, n is odd.
Proposition 1.5([11]) (G) 6 (G) 6 m(G).
Proposition 1.6([12]) For any graph G, (G) > 3 if and only if G has an odd cycle.
2. Main Results
Lemma 2.1 For a wheel Wn ,

4, n is odd
[S (Wn )] =
3, n is even
Proof Let v1 , v2 , . . . , vn be the rim vertices of wheel Wn which are duplicated by the
vertices v1 , v2 , . . . , vn respectively and let v denotes the apex vertex of Wn which is duplicated
by the vertex v . Let e1 , e2 , . . . , en be the rim edges of Wn . Then the resultant graph S [Wn ]
will have order 2(n + 1) and size 6n.
Case 1. n is odd
In this case S [Wn ] contains odd Wn as an induced subgraph. Since (Wn ) = 4

[S (Wn )] = 4.
Case 2. n is even
In this case S [Wn ] contains even Wn as an induced subgraph. Since (Wn ) = 3

[S (Wn )] = 3. 2
b-Chromatic Number of Splitting Graph of Wheel 71
Theorem 2.2 For a wheel Wn ,

4, n = 3

3, n = 4

[S (Wn )] =

5, n = 5, 6, 8

6, n = 7

6, n > 9
Proof To prove the result we continue with the terminology and notations used in Lemma
2.1 and consider the following cases.
Case 1. n=3
In this case the graph S (W3 ) contains an odd cycle. Then by Proposition 1.6, [S (W3 )] >
3. As m[S (W3 )] = 4 and by Lemma 2.1, [S (W3 )] = 4. We have 4 6 [S (W3 )] 6 4 by
Proposition 1.5. Thus, [S (W3 )] = 4.
Case 2. n=4
3. As m[S (W4 )] = 5 and by Lemma 2.1 [S (W4 )] = 3. Then by Proposition 1.5 we have
3 6 [S (W4 )] 6 5.
If [S (W4 )] = 5 then by Proposition 1.3, the graph S (W4 ) must have five vertices of
degree at least 4 which is possible. But due to the adjacency of vertices of the graph S (W4 )
any proper coloring with five colors have at least one color class which does not have color
dominating vertices hence it will not be b-coloring for the graph S (W4 ). Thus, [S (W4 )] 6= 5.
Suppose [S (W4 )] = 4. Now consider the color class c = {1, 2, 3, 4} and define the color
function as f : V {1, 2, 3, 4} as f (v) = 4 = f (v ), f (v1 ) = 1, f (v2 ) = 2, f (v1 ) = 1, f (v2 ) =
2, f (v3 ) = 3, f (v4 ) = 3 which in turn forces to assign f (v3 ) = 1, f (v4 ) = 2. This proper
coloring gives the color dominating vertices for color classes 1, 2 and 4 but not for 3 which
is contradiction to our assumption. Thus, [S (W4 )] 6= 4. Hence, we can color the graph by
three colors. For b-coloring, consider the color class c = {1, 2, 3} and define the color function
as f : V {1, 2, 3} as f (v1 ) = 1 = f (v1 ), f (v2 ) = 2 = f (v2 ), f (v3 ) = 1 = f (v3 ), f (v4 ) =
2 = f (v4 ), f (v) = 3 = f (v ). This proper coloring gives the color dominating vertices as
cdv(1) = v1 , cdv(2) = v2 , cdv(3) = v. Thus [S (W4 )] = 3.
Case 3. n = 5, 6, 8
Subcase 3.1 n = 5
3. As m[S (W5 )] = 6 and by Lemma 2.1, [S (W5 )] = 4. Then by Proposition 1.5 we have
4 6 (S (W5 ) 6 6.
If (S (W5 ) = 6 then by Proposition 1.3, the graph S (W5 ) must have six vertices of degree
at least five which is possible. But due to the adjacency of vertices of the graph S (W5 ) any
proper coloring with six colors have at least one color class which does not have color dominating
vertices. Hence it will not be b-coloring for the graph S (W5 ). Thus, (S (W5 ) 6= 6.
Suppose (S (W5 ) = 5. Now consider the color class = {1, 2, 3, 4, 5} and define the color
function as f : V {1, 2, 3, 4, 5} as f (v) = 5 = f (v ), f (v1 ) = 3, f (v2 ) = 1, f (v3 ) = 2, f (v4 ) =
3, f (v5 ) = 4, f (v1 ) = 2, f (v2 ) = 4, f (v3 ) = 4, f (v4 ) = 1, f (v5 ) = 1. This proper coloring gives
the color dominating vertices as cdv(1) = v2 , cdv(2) = v3 , cdv(3) = v4 , cdv(4) = v5 , cdv(5) = v.
Thus, (S (W5 ) = 5.
Subcase 3.2 n = 6, 8
In this case the graph S (Wn ) contains an odd cycle. Then by Proposition 1.6, [S (Wn )] >
3. As m[S (Wn )] = 7 and by Lemma 2.1, [S (Wn )] = 3. Then by Proposition 1.5 we have
3 6 [S (Wn )] 6 7.
If [S (Wn )] = 7 then by Proposition 1.3, the graph S (Wn ) must have seven vertices of
degree at least six which is possible. But due to the adjacency of the vertices of graph S (Wn )
any proper coloring with seven colors have at least one color class which does not have color
dominating vertices. Hence it will not be b-coloring for the graph S (Wn ). Thus, [S (Wn )] 6= 7.
Suppose [S (Wn )] = 6. Now consider the color class = {1, 2, 3, 4, 5, 6} and define the color

function as f : V {1, 2, 3, 4, 5, 6} as f (v) = 6 = f (v ), f (v1 ) = 3, f (v2 ) = 1, f (v3 ) = 2, f (v4 ) =
3, f (v5 ) = 4, f (v1 ) = 4, f (v2 ) = 4, f (v3 ) = 5, f (v4 ) = 5, f (v5 ) = 1 which in turn forces to assign
f (v6 ) = 2, f (v6 ) = 1. This proper coloring gives the color dominating vertices for color classes
1, 2, 3, 4 and 6 but not for 5 which is contradiction to our assumption. Thus, [S (Wn )] 6= 6.
Suppose that S (Wn ) has b-coloring with 5 colors. Now consider the color class = {1, 2, 3, 4, 5}

and define the color function as f : V (G) {1, 2, 3, 4, 5} as f (v) = 5 = f (v ), f (v1 ) = 3, f (v2 ) =
1, f (v3 ) = 2 = f (v3 ), f (v4 ) = 3 = f (v4 ), f (v5 ) = 4, f (v6 ) = 2, f (v1 ) = 4, f (v2 ) = 4, f (v5 ) =
1, f (v6 ) = 1. This proper coloring gives the color dominating vertices as cdv(1) = v2 , cdv(2) =
v3 , cdv(3) = v4 , cdv(4) = v5 , cdv(5) = v. Thus, [S (Wn )] = 5.
Case 4. n=7
4 6 [S (W7 )] 6 7.
Suppose [S (W7 )] = 7. Now consider the color class = {1, 2, 3, 4, 5, 6, 7} and define the
color function as f : V {1, 2, 3, 4, 5, 6, 7} as f (v) = 7, f (v ) = 6, f (v1 ) = 5, f (v2 ) = 1, f (v3 ) =
2, f (v4 ) = 3, f (v5 ) = 1, f (v6 ) = 4, f (v1 ) = 1, f (v2 ) = 4, f (v3 ) = 4, f (v4 ) = 5, f (v5 ) = 5, f (v6 ) =
4 which in turn forces to assign f (v7 ) = 2, f (v7 ) = 3. This proper coloring gives the color
dominating vertices for color classes 1, 2, 3, 4 and 5 but not for 6 and 7 which is contradiction
to our assumption. Thus, [S (W7 )] 6= 7.
Suppose that S (W7 ) has b-coloring with 6 colors. Now consider the color class = {1, 2, 3, 4,
5, 6} and define the color function f : V {1, 2, 3, 4, 5, 6} as f (v) = 6 = f (v ), f (v1 ) =
5, f (v4 ) = 5, f (v5 ) = 1, f (v6 ) = 1, f (v7 ) = 5. This proper coloring gives the color dominating
vertices as cdv(1) = v2 , cdv(2) = v3 , cdv(3) = v4 , cdv(4) = v5 , cdv(5) = v7 , cdv(6) = v. Thus,
[S (W7 )] = 6.
Case 5. n>9
For n = 9, the graph S (W9 ) contains an odd cycle. Then by Proposition 1.6, [S (W9 )] >
4 6 [S (W7 )] 6 7.
Suppose [S (W9 )] = 7. Consider the color class = {1, 2, 3, 4, 5, 6, 7} and define the color

function f : V {1, 2, 3, 4, 5, 6, 7} as f (v) = 6, f (v ) = 7, f (v1 ) = 3, f (v2 ) = 1, f (v3 ) =
2, f (v7 ) = 5, f (v7 ) = 5, f (v8 ) = 3, f (v8 ) = 4 which in turn forces to assign f (v9 ) = 2 = f (v9 ).
This proper coloring gives the color dominating vertices for color classes 1, 2, 3, 4 and 5 but not
for 6 and 7 which is contradiction to our assumption. Thus, [S (W9 )] 6= 7.
Suppose that S (W9 ) has b-coloring with 6 colors. Consider the color class = {1, 2, 3, 4, 5, 6}
and define the color function f : V {1, 2, 3, 4, 5, 6} as f (v) = 6 = f (v ), f (v1 ) = 3, f (v2 ) =
1, f (v3 ) = 2, f (v4 ) = 3, f (v5 ) = 4, f (v6 ) = 2, f (v7 ) = 5, f (v8 ) = 3, f (v9 ) = 1 = f (v9 ), f (v1 ) =
4, f (v2 ) = 4, f (v3 ) = 5, f (v4 ) = 5, f (v5 ) = 1, f (v6 ) = 1, f (v7 ) = 5, f (v8 ) = 4. This proper
coloring gives the color dominating vertices as cdv(1) = v2 , cdv(2) = v3 , cdv(3) = v4 , cdv(4) =
v5 , cdv(5) = v7 , cdv(6) = v. Thus, [S (W9 )] = 6.
For n > 9, we repeat the colors as in the above graph S (W9 )for the vertices {v1 , v2 , . . . , v9 ,
v1 , v2 , . . . , v9 , v, v } and for the remaining vertices assign the colors as f (v) = 6 = f (v ), f (v3k+7 )
= 1 = f (v3k+7
), f (v3k+8 ) = 2 = f (v3k+8 ) where k N . Hence, [S (W9 )] = 6, for all n > 9. 2
Theorem 2.3 Let Wn be a wheel. Then, S (Wn ) is b-continuous.
Proof To prove this result we continue with the terminology and notations used in Lemma
2.1 and consider the following cases.
Case 1. n=3
In this case the graph S (W3 ) is b-continuous as [S (W3 )] = [S (W3 )] = 4.
Case 2. n=4
In this case the graph S (W4 ) is b-continuous as [S (W4 )] = [S (W4 )] = 3.
Case 3. n=5
In this case by Lemma 2.1, [S (W5 )] = 4 and by Theorem-2.2, [S (W5 )] = 5. Hence,

b-coloring exists for every integer satisfying [S (W5 )] 6 k 6 [S (W5 )](Here k = 4, 5). Thus,
S (W5 ) is b-continuous.
Case 4. n=6
In this case by Lemma 2.1, [S (W6 )] = 3 and by Theorem-2.2, [S (W6 )] = 5. It is

obvious that b-coloring for the graph S (W6 ) is possible using the number of colors k = 3, 5.
Now for k = 4 the b-coloring for the graph S (W6 ) is as follows.
Consider the color class = {1, 2, 3, 4} and define the color function f : V {1, 2, 3, 4} as
f (v) = f (v ) = 4, f (v1 ) = f (v1 ) = 3, f (v2 ) = f (v2 ) = 1, f (v3 ) = f (v3 ) = 2, f (v4 ) = f (v4 ) =

3, f (v5 ) = f (v5 ) = 1, f (v6 ) = f (v6 ) = 2. This proper coloring gives the color dominating
vertices as cdv(1) = v2 , cdv(2) = v3 , cdv(3) = v4 , cdv(4) = v. Thus, S (W6 ) is four colorable.

Hence b-coloring exists for every integer k satisfy [S (W6 )] 6 k 6 [S (W6 )] (Here k = 3, 4, 5).
Consequently S (W6 ) is b-continuous.
Case 5. n=7
By Lemma 2.1, [S (W7 )] = 4 and by Theorem 2.2, [S (W7 )] = 6. It is obvious that

b-coloring for the graph S (W7 ) is possible using the number of colors k = 4, 6. Now for k = 5
the b-coloring for the graph S (W7 ) is as follows.
Consider the color class = {1, 2, 3, 4, 5} and define the color function f : V {1, 2, 3, 4, 5}
as f (v) = f (v ) = 5, f (v1 ) = 3, f (v1 ) = 4, f (v2 ) = 1, f (v2 ) = 4, f (v3 ) = 2, f (v3 ) = 2, f (v4 ) =
3, f (v4 ) = 1, f (v5 ) = 4, f (v5 ) = 1, f (v6 ) = 2, f (v6 ) = 2, f (v7 ) = 1 = f (v7 ). This proper
coloring gives the color dominating vertices as cdv(1) = v2 , cdv(2) = v3 , cdv(3) = v4 , cdv(4) =
v5 , cdv(5) = v. Thus, S (W7 ) is five colorable. Hence, b-coloring exists for every integer k
satisfy [S (W7 )] 6 k 6 [S (W7 )](Here k = 4, 5, 6). Hence S (W7 ) is b-continuous.
Case 6. n=8
By Lemma 2.1, [S (W8 )] = 3 and by Theorem 2.2,[S (W8 )] = 5. It is obvious that

b-coloring for the graph S (W8 ) is possible using the number of colors k = 3, 5. Now for k = 4
the b-coloring for the graph S (W8 ) is as follows.
Consider the color class = {1, 2, 3, 4} and define the color function as f : V {1, 2, 3, 4}
as f (v) = f (v ) = 4, f (v1 ) = 3 = f (v1 ), f (v2 ) = 1 = f (v2 ), f (v3 ) = 2 = f (v3 ), f (v4 ) = 3 =

f (v4 ), f (v5 ) = 1 = f (v5 ), f (v6 ) = 2 = f (v6 ), f (v7 ) = 1 = f (v7 ), f (v8 ) = 2 = f (v8 ). This proper
coloring gives the color dominating vertices as cdv(1) = v2 , cdv(2) = v3 , cdv(3) = v4 , cdv(4) = v.
Thus, S (W8 ) is four colorable. Hence, b-coloring exists for every integer k satisfy [S (W8 )] 6
k 6 [S (W8 )](Here k = 3, 4, 5). Thus, S (W8 ) is b-continuous.
Case 7. n>9
For n = 9, by Lemma 2.1, [S (W9 )] = 4 and by Theorem 2.2,[S (W9 )] = 6. It is obvious

that b-coloring for the graph S (W9 ) is possible using the number of colors k = 4, 6. Now for
k = 5 the b-coloring for the graph S (W9 ) is as follows.
Consider the color class = {1, 2, 3, 4, 5} and define the color function as f : V {1, 2, 3, 4, 5}
as f (v) = f (v ) = 5, f (v1 ) = 3, f (v1 ) = 4, f (v2 ) = 1, f (v2 ) = 4, f (v3 ) = 2, f (v3 ) = 2, f (v4 ) =

3, f (v4 ) = 1, f (v5 ) = 4, f (v5 ) = 1, f (v6 ) = 2, f (v6 ) = 2, f (v7 ) = 1 = f (v7 ), f (v8 ) = 2, f (v8 ) =

2, f (v9 ) = f (v9 ) = 1. This proper coloring gives the color dominating vertices as cdv(1) =
v2 , cdv(2) = v3 , cdv(3) = v4 , cdv(4) = v5 , cdv(5) = v. Thus, S (W9 ) is five colorable. Hence,
b-coloring exists for every integer k satisfy [S (W9 )] 6 k 6 [S (W9 )](Here k = 4, 5, 6). Hence,
S (W9 ) is b-continuous.
For odd n > 9, we repeat the colors as in S (W9 ) for the vertices {v1 , v2 , v9 , . . . , v1 , v2 ,
. . . , v 9 , v, v } and for the remaining vertices gives the colors as follows:
When k = 5, f (v ) = f (v) = 5, f (v3k+7 ) = f (v 3k+7 ) = 1, f (v3k+8 ) = f (v 3k+8 ) = 2, k

N.
For even n > 9, we repeat the color assignment as in case n = 8 discussed above for the
vertices {v, v , v1 , . . . , v8 , v1 , v2 , . . . , v8 } and for remaining vertices gives the colors as follows:
When k = 4, f (v ) = f (v) = 4, f (v2k+7 ) = 1 = f (v 2k+7 ), f (v2k+8 ) = 2 = f (v 2k+8 ), k

N and when k = 5, f (v ) = f (v) = 5, f (v2k+8 ) = 1 = f (v 2k+8 ), f (v2k+9 ) = 2 = f (v 2k+9 ), k
N. 2
Any coloring with (G) is a b-coloring, we state the following obvious result.
Corollary 2.4 Let Wn be a wheel. Then

{4}, n=3

{3}, n=4

{4, 5}
n=5

Sb [S (Wn )] = {3, 4, 5}, n = 6, 8

{4, 5, 6}, n=7

{4, 5, 6} f or odd n > 9

{3, 4, 5} f or even n > 9
3. Concluding Remarks
A discussion about b-coloring of wheel is carried out by Alkhateeb [9] while we investigate
b-chromatic number of splitting graph of wheel. We also obtain b-spectrum and show that
splitting graph of wheel is b-continuous.
Acknowledgement
The authors are grateful to the anonymous referee for careful reading of first draft of this paper.
References
[1] R.W.Irving and D.F.Manlove, The b-chromatic number of a graph, Discrete Applied Math-
ematics, 91, (1999), 127-141.
[2] M.Kouider, M.Maheo, Some Bounds for the b-chromatic number of a graph, Discrete Math-
ematics, 256, (2002), 267-277.
[3] S.K.Vaidya and M.S.Shukla, b-chromatic number of some cycle related graphs, Interna-
tional Journal of Mathematics and Soft Computing, 4, (2014), 113-127.
[4] S.K.Vaidya and M.S.Shukla, b-chromatic number of some wheel related graphs, Malaya
Journal of Matematik, 2(4), (2014), 482-488.
[5] S.K.Vaidya and M.S.Shukla, Some new results on b-coloring of graphs, Advances and Ap-
plications in Discrete Mathematics, (In Press).
[6] S.K.Vaidya and M.S.Shukla, Switching of vertex in path and b-coloring, International Jour-
nal of Mathematics And its Applications, (In Press)
[7] T.Faik, About the b-continuity of graphs, Electronic Notes in Discrete Mathematics, 17,
(2004), 151-156.
[8] J.Kratochvil, Z.Tuza and M.Voight, On b-Chromatic Number of Graphs, Lecture Notes in
Computer Science, Springer, Berlin, 2573, (2002), 310-320.
[9] M.Alkhateeb, On b-coloring and b-continuity of graphs, Ph.D Thesis, Technische Univer-
sit
at Bergakademie, Freiberg, Germany, (2012).
[10] R.Balakrishnan, S. Francis Raj and T.Kavaskar, b-Chromatic Number of Cartesian Product
of Some Families of Graphs ,Discrete Applied Mathematics, 160, (2012), 2709-2715.
[11] R.Balakrishnan and K.Ranganathan, A Textbook of Graph Theory, 2/e, Springer, New
York, 2012.
[12] J.Clark and D.A.Holton,A First Look at Graph Theory, World Scientific, pp.193, (1969).
Eccentric Connectivity and Connective Eccentric Indices

Of Generalized Complementary Prisms and Some Duplicating Graphs
S.Arockiaraj
Department of Mathematics
Mepco Schlenk Engineering College, Sivakasi- 626 005, Tamilnadu, India)
Vijaya Kumari
Research Scholar of Karpagam University, Coimbatore
Department of Mathematics, Vijaya College, Mulky-574154, Karnataka, India
E-mail: psarockiaraj@gmail.com, vijayasullal@gmail.com
Abstract: The connected eccentricity index and the eccentric connectivity index of a graph
G is respectively defined by
X deg(v) X
ce (G) = and c (G) = deg(v) ecc(v),
ecc(v)
vV (G) vV (G)
where ecc(v) is the eccentricity of a vertex v in G. In this paper, we have obtained the bounds
for connective eccentricity index of those generalized complementary prisms and eccentric
connective index of duplication of some graphs.
Key Words: Eccentricity, radius, diameter, complementary prism.
AMS(2010): 05C15, 05C38.
1. Introduction
Throughout this paper all graphs we considered are simple and connected. For a vertex v
V (G), deg(v) denotes the degree of v, (G) and (G) represent the minimum and maximum
degree of G respectively. For vertices u, v V (G), the distance d(u, v) is defined as the length of
the shortest path between u and v in G. The eccentricity ecc(v) of a vertex v is the maximum
among the distance from v to the remaining vertices of G. The diameter d(G) of the graph
G is the maximum eccentricity of the vertices of G, while the radius r(G) of the graph G is
the minimum eccentricity of the vertices of G. The total eccentricity of the graph G, denoted
by (G) is defined as the sum of eccentricities of all the vertices of the graph G. That is

78 S.Arockiaraj and Vijaya Kumari
P
(G) = ecc(v). The eccentric connectivity index of G, denoted by c (G) is defined as
vV (G)
X
c (G) = deg(v) ecc(v).
vV (G)
In [4], the connective eccentricity index (CEI) of a graph G was defined as
X deg(v)
ce (G) = .
ecc(v)
vV (G)
Kathiresan and Arockiaraj introduced some generalization of complementary prisms and

studied the Wiener index of those generalized complementary prisms ([8]).
Let G and H be any two graphs on p1 and p2 vertices, respectively and let R and S be
subsets of V (G) = {u1 , u2 , . . . , up1 } and V (H) = {v1 , v2 , . . . , vp2 } respectively. The comple-
mentary product G(R)2H(S) has the vertex set {(ui , vj ) : 1 i p1 , 1 j p2 } and (ui , vj )
and (uh , vk ) are adjacent in G(R)2H(S)
(1) if i = h, ui R and vj vk E(H), or if i = h, ui

/ R and vj vk
/ E(H) or
(2) if j = k, vj S and ui uh E(G), or if j = k, vj / S and ui uh
/ E(G).
In other words, G(R)2H(S) is the graph formed by replacing each vertex ui R of G
by a copy of H, each vertex ui / R of G by a copy of H, each vertex vj S of H by a
copy of G and each vertex vj / S of H by a copy of G. If R = V (G) (respectively, S =
V (H)), the complementary product can be written as G2H(S) (respectively, G(R)2H). The
complementary prism GG obtained from G is G2K2 (S) with |S| = 1. That is, GG has a copy
of G and a copy of G with a matching between the corresponding vertices. In GG, we have
an edge vv for each vertex v in G. The authors of [?] consider this edge as K2 or K1,1 or P2 .
By taking m copies of G and n copies of G, they generalize the complementary prism as a
graph G2H(S), where H = Km+n (or Km,n ) and S is a subset of V (H) having m vertices and
H = C2m (or P2m ) whose vertex set is {v1 , v2 , . . . , v2m } and S = {v1 , v3 , . . . , v2m1 } ([8]).
Let G be a graph with vertex set V (G) and edge set E(G). The graph EV (G) obtained by
duplicating each edge by a vertex of a graph G is defined as follows. The vertex set of EV (G)
is V (G) E(G). Two vertices x, y in the vertex set of EV (G) are adjacent in EV (G) in case
one of the following holds:
(1) x and y are adjacent vertices in G;

(2) x is in V (G), y is in E(G) and x, y are incident in G.
Motivated by these works, we have obtained the bounds for connective eccentric index of
those generalized complimentary prisms and eccentric connective index of duplication of some
graphs.
Theorem 1.1([8]) For the complementary prism GG, r(GG) = 2 and

2 if d(G) = d(G) = 2
d(GG) =
3 otherwise.
Eccentric Connectivity and Connective Eccentric Indices of Graphs 79
Theorem 1.2([8]) For any connected graph G with p 2,

2 if d(G) = d(G) = 2 and m = n = 1
d(Gm+n ) =
3 otherwise.
Theorem 1.3([8]) For any connected graph G with p 2,

2 if d(G) = d(G) = 2 and m = n = 1
d(Gm,n ) =
3 otherwise.
Theorem 1.4([8] ) For any connected graph G with p 2,

2m
if m > 1
d(Gpm,m ) = 2 if m = 1 and d(G) = d(G) = 2

3 otherwise.
Theorem 1.5([8]) For any connected graph G with p 2 d(Gcm,m ) = 2r + 1 if m = 2r 2 and

r is a positive integer.
2. Main Results
Theorem 2.1 For any connected graph G on p vertices,
p(p + 1) p(p + 1)
ce (GG) .
3 2
Proof For any connected graph G on p vertices, by Theorem 1.1, GG F22 while G F22
and GG F23 while G
/ F22 . When GG F22 , ecc(v) = 2 for all v V (GG). So
X deg(v) 1 X p(p + 1)
ce (GG) = = deg(v) = .
ecc(v) 2 2
vV (GG) vV (GG)
When GG F23 , 2 ecc(v) 3 for all v V (GG). This implies that
1 1 1

3 ecc(v) 2
for all v V (GG) and hence

deg(v) deg(v) deg(v)

3 ecc(v) 2
for all v V (GG). Therefore
X deg(v) X deg(v) X deg(v)

.
3 ecc(v) 2
vV (GG) vV (GG) vV (GG)
Thus
p(p + 1)
3
ce (GG)
p(p + 1)
2
. 2
Theorem 2.2 For any connected graph G on p vertices and q edges with m, n 1

2 p m+n ce p m+n
(m n)q + n + (Gm+n ) (m n)q + n + .
3 2 2 2 2

Proof The number of edges in Gm+n is (m n)q + n p2 + m+n 2 . By Theorem 1.2,
Gm+n F22 when m = 1, n = 1 and Gm+n F23 otherwise. Hence by Theorem 1.2, the result
follows. 2
Theorem 2.3 For any connected graph G on p vertices and q edges,

2 p ce p
(m n)q + n + mn (Gm,n ) (m n)q + n + mn.
3 2 2
Proof The number of edges in Gm,n is

p
(m n)q + n + mn.
2
By Theorem 1.3, Gm,n F22 when m = n = 1 and Gm,n F23 otherwise. Hence by
Theorem 2.1, the result follows. 2
Theorem 2.4 For any connected graph G on p vertices and q edges,

p 1 ce p p 1
+p 2 (Gm,m ) 2 +p 2 .
2 m 2 m

Proof In Gpm,m , the number of edges is m p2 + p(2m 1). By Theorem 1.4, r(Gpm,m ) = m
and d(Gpm,m ) = 2m. So
deg(v) deg(v) deg(v)

2m ecc(v) m
and hence

1
m
m
p
2
ce p
+ p(2m 1) (Gm,m )
2
m
m
p
2
+ p(2m 1) . 2
Theorem 2.5 For any connected graph G on p vertices and for any even integer m 2,
2mp(p + 1) 4mp(p + 1)
ce (Gcm,m ) .
m+1 m+2
Proof For even integer m 2, by Theorem 1.5,

m
+ 1 ecc(v) m + 1
2
p

for all v V (Gcm,m ). Also the number of edges in Gcm,m is m 2 + 2mp. Therefore

2 p 2 p
m + 2mp ce (Gcm,m ) m m + 2mp
m+1 2 2 +1 2
and hence
2mp
m+1
[p + 1] ce (Gcm,m )
4mp
m+2
[p + 1]. 2
Now we determine the exact value of c (G) for some graph families.
Proposition 2.6 For any n 3,

1 2
16n + 7),
2 (9n n is odd
c (EV (Pn )) =
1 2
16n + 8),
2 (9n n is even.
Proof Let v1 , v2 , , vn be the vertices on the path and x1 , x2 , , xn1 be the vertices
corresponding to the edges of the path Pn . Then,
n
n i, 1i 2 n i, 1 i n2
ecc(vi ) = n and ecc(xi ) = n
i 1, +1in i, + 1 i n 1.
2 2
Also,

2, i = 1, n
deg(vi ) = and deg(xi ) = 2, 1 i n 1.
4, 2in1
Therefore
X
c (EV (Pn )) = deg(v) ecc(v)
vV (EV (Pn ))
n1
X n1
X
=4 ecc(vi ) + 4(n 1) + 2 ecc(xi )
i=2 i=1
X
2
n
n1
X
=4 (n i) + 4 (i 1) + 4(n 1)
i=2 i= n
2 +1
X
2
n
n1
X
+ 2(n 1) + 2 (n i) + 2 i
i=2 i= n
+1
2
X
n
2 n1
X j n k
=6 (n i) + 6 i4 n1 + 6(n 1)
i=2
2
i= n
2 +1
j n k X
2
n
n1
X jnk
= 6n 1 6 i+6 i 4n + 4 + 4 + 6n 6
2 i=2
2
i= n
2 +1
j n k X
n
2 n1
X jnk
= 6n 1 12 i+6 i + 2n 2 + 4 .
2 i=2 i=2
2
jnk X
2
n
n1
X jnk
= 6n 6n 12 i + 12 + 6 i 6 + 2n 2 + 4
2 i=1 i=1
2
jnk n n !
2 2 +1 n(n 1)
= (6n + 4) 4n + 4 12 +6
2 2 2
jnk j n k j n k
= (6n + 4) 4n + 4 6 + 1 + 3n2 3n
2 2 2
jnk j n k2
2
= (6n 2) 6 + 3n 7n + 4
2 2
1 (9n2 16n + 7), n is odd
= 2
1 (9n2 16n + 8),
2
2 n is even.

3n2 n, n is odd
c (EV (Cn )) =
3n2 + 2n, n is even.
Proof Let v1 , v2 , , vn be the vertices on the cycle of length n and x1 , x2 , , xn be the

vertices corresponding to the edges of Cn so that vi xi , xi vi+1 E(EV (Cn )), 1 i n, where
vn+1 = v1 . Then,

n1 n+1
2 , n is odd 2 , n is odd
ecc(vi ) = and ecc(xi ) =
n n+2
n is even, for 1 i n.
2, n is even 2 ,
Also, deg(vi ) = 4, 1 i n and deg(xi ) = 2, 1 i n. Therefore,

X
c (EV (Cn )) = deg(v) ecc(v)
vV (EV (Cn ))
n
X n
X
=4 ecc(vi ) + 2 ecc(xi )
i=1 i=1

4n n1 + 2n n+1 , n is odd
2 2
=
4n n + 2n n+2 , n is even
2 2

3n2 n, n is odd
=
3n2 + 2n,
2
n is even.

6,
n=2
c
(EV (Kn )) = 36, n=3

7n(n 1), n 4.

Proof Let v1 , v2 , . . . , vn be the vertices of Kn and x1 , x2 , . . . , xn , m = n2 be the vertices
corresponding to the edges of Kn . Then ecc(vi ) = 2 for 1 i n and ecc(xi ) = 3 for 1 i m.
Also deg(vi ) = 2n 2, 1 i n and deg(xi ) = 2, 1 i m. Therefore,
X
c (EV (Kn )) = deg(v) ecc(v)
vV (EV (Kn ))
X X
= deg(vi ) ecc(vi ) + deg(xi ) ecc(xi )
vi V (EV (Kn )) xi V (EV (Kn ))
= 2n(2n 2) + 6m
n(n 1)
= 4n(n 1) + 6 = 7n(n 1).
2
When n = 3, ecc(vi ) = ecc(xi ) = 2. Therefore c (EV (K3 )) = 36. When n = 2, ecc(vi ) =
ecc(xi ) = 1 and deg(vi ) = 2 for i = 1, 2 and deg(x1 ) = 2. So c (EV (K2 )) = 6. 2
Proposition 2.9 The eccentric connectivity index of K1,n is 6n.
Proof Let v0 be the central vertex and v1 , v2 , . . . , vn be the pendent vertex of K1,n . Let
x1 , x2 , . . . , xn be the vertices corresponding to the edges of K1,n in EV (K1,n ). Then ecc(v0 ) = 1,
ecc(vi ) = 2, 1 i n and ecc(xi ) = 2, 1 i n. Also deg(v0 ) = 2n and deg(vi ) = deg(xi ) = 2
for 1 i n. Therefore,
X
c (EV (K1,n )) = deg(v) ecc(v)
vV (EV (K1,n ))
n
X n
X
= 2n + 4+ 4 = 6n. 2
i=1 i=1

28n,
n = 3, 4
c
(EV (Wn )) = 34n, n=5

36n, n 6.
Proof Let v0 be the central vertex and v1 , v2 , . . . , vn be the vertices on the cycle of Wn .
Let xi , 1 i n be the vertices corresponding to the edges on the cycle and xn+i , 1 i n
be the vertices corresponding to the edges v0 vi , 1 i n.
Assume that n 6. In EV (Wn ), ecc(v0 ) = 2, ecc(vi ) = ecc(xi ) = 3, 1 i n and ecc(xi ) =

4, n + 1 i 2n. Also deg(v0 ) = 2n, deg(vi ) = 6, 1 i n and deg(xi ) = 2, 1 i 2n.
Therefore
X
c (EV (Wn )) = deg(v) ecc(v)
vV (EV (Wn ))
n
X
= deg(v0 ) ecc(v0 ) + deg(vi ) ecc(vi )
i=1
n
X 2n
X
+ deg(xi )ecc(xi ) + deg(xi ) ecc(xi )
i=1 i=n+1
= 4n + 18n + 6n + 8n = 36n.
When n = 5, ecc(v0 ) = 2, ecc(vi ) = 3, 1 i n, ecc(xi ) = 3, 1 i 2n and hence

c (EV (Wn )) = 34n. When n = 3 and 4, ecc(v0 ) = 2, ecc(vi ) = 2, 1 i n, ecc(xi ) =
3, 1 i 2n and hence c (EV (Wn )) = 28n. 2

27 2
+n 13
2 n 2 , when n is odd
c (EV (Ln )) =
27 2
+ n 4,
2 n when n is even.
Proof Let u1 , u2 , . . . , un and v1 , v2 , . . . , vn be the vertices on the path of length n 1. Let

xi and yi be the duplicating vertices of the edges ui ui+1 and vi vi+1 respectively, 1 i n 1
and zi be the duplicating vertex of the edge ui vi , 1 i n. In EV (Ln ),

n + 1 i, 1 i n2
ecc(ui ) = n
ecc(un+1i ),
2 + 1 i n,
ecc(vi ) = ecc(ui ), 1 i n,

n + 1 i, 1 i n2
ecc(xi ) = n
ecc(xni ), + 1 i n 1,
2
ecc(yi ) = ecc(xi ), 1 i n 1 and

n + 1 i, 1 i n2
ecc(zi ) = n
ecc(zn+1i ), + 1 i n.
2
Also,

4, i = 1, n
deg(ui ) = deg(vi ) = and deg(xi ) = deg(yi ) = deg(zi ) = 2.
6, 2 i n 1
Therefore,
X
c (EV (Ln )) = deg(v) ecc(v)
vV (EV (Ln ))
n
X n1
X
=2 deg(ui )ecc(ui ) + 2 deg(xi )ecc(xi )
i=1 i=1
Xn
+ deg(zi )ecc(zi )
i=1
n1
X n1
X n
X
= 16n + 12 ecc(ui ) + 4 ecc(xi ) + 2 ecc(zi ).
i=2 i=1 i=1
When n is odd,

c n+1 n+3
(EV (Ln )) = 16n + 12 (n 2) + 2 1+ 2 + +
2 2

n+3 n3
+ 4 (n 1) + 2 1 + 2 + +
2 2

n+3 n1
+2 n + 2 1 + 2 + +
2 2
= 16n + 6(n 2)(n + 1) + 4(n 3)(n 1)

n1 n+1 27 2 13
+ 2(n 1)(n + 3) + n(n + 3) + 2 = n +n .
2 2 2 2
When n is even,

n2 n4
c (EV (Ln )) = 16n + 12 (n 2) + 2 1+ 2 + +
2 2

n+2 n2
+ 4 (n 1) + 2 1 + 2 + +
2 2

n+4 n2
+2 n + 2 1 + 2 + +
2 2
= 16n + 6(n2 4) + 2(n 1)(n + 2) + n(n + 4)
3
+ 3(n 4)(n 2) + (n(n 2)) =
2
27 2
2
n + n 4. 2

6n2 + 12n, n is odd
c (EV (Cn K1 )) =
6n2 + 16n, n is even.
Proof Let v1 , v2 , . . . , vn be the vertices in the cycle Cn and ui be the pendent vertex
attached at vi , 1 i n, in Cn K1 . Let xi , 1 i n be the vertices corresponding to the
edges of the cycle Cn and yi , 1 i n be the vertices corresponding to the pendent edges of
Cn K1 in EV (Cn K1 ). In EV (Cn Kn ), for 1 i n,

n+1
2 , if n is odd
ecc(vi ) =
n+2
2 , if n is even,

n+3
2 , if n is odd
ecc(ui ) =
n+4
2 , if n is even,

n+3
2 , if n is odd
ecc(xi ) = and
n+2
2 , if n is even
ecc(yi ) = ecc(ui ).
Also deg(vi ) = 6 and deg(ui ) = deg(xi ) = deg(yi ) = 2, 1 i n. When n is odd,

X
c (EV (Cn K1 )) = deg(v) ecc(v)
vV (EV (Cn K1 ))
n
X n
X n
X
=6 ecc(vi ) + 4 ecc(ui ) + 2 ecc(xi )
i=1 i=1 i=1

n+1 n+3 n+3
= 6n + 4n + 2n
2 2 2
= 6n2 + 12n.
When n is even,
X
c (EV (Cn K1 )) = deg(v) ecc(v)
vV (EV (Cn K1 ))
n
X n
X n
X
=6 ecc(vi ) + 4 ecc(ui ) + 2 ecc(xi )
i=1 i=1 i=1

n+2 n+4 n+2
= 6n + 4n + 2n
2 2 2
= 6n2 + 16n. 2
References
[1] A.R.Ashrafi, M.Saheli, M.Ghorbani, The eccentric connectivity index of nanotubes and
nanotori, J. Comput. Appl. Math., 235 (2011), 45614566.
[2] B.Eskender and E.Vumar, Eccentric connectivity index and eccentric distance sum of some
graph operations, Transactions on Combinatorics, 2(1) (2013), 103111.
[3] Guihaiyu and Lihua Feng, On eccentric connectivity index of graphs, MATCH COMMUN.
Math. Comput. Chem., 69 (2013), 611628.
[4] Guihaiyu and Lihua Feng, On eccentric connectivity index of graphs, MATCH COMMUN.
Math. Comput. Chem., 69 (2013), 611628.
[5] I.Gutman, R.Geez and J.Rada, Wiener index of eulerian graphs, Discrete Appl. Math.,
162 (2014), 247250.
[6] H.Hua, S.Zhang and K.Xu, Further results on the eccentric distance sum, Discrete Appl.
Math., 160 (2012), 170180.
[7] A.Ilic and I.Gutman, Eccentric connectivity index of chemical trees, MATCH. COMMUN.
MATH. COMPUT. CHEM., 65 (2011), 731744.
[8] KM.Kathiresan and S.Arockiaraj, On the wiener index of generalized complimentary prism,
Bull. Inst. Combin. Appl., 59 (2010), 3145.
[9] D.Michalak, On middle and total graphs with coarseness number equal I, Springer-verlag
graph theory, Lagow proceedings, Berlin, New York, (1981), 139150.
[10] B.Zhou and Dee, On eccentric connectivity index, MATCH COMMUN. Math. Comput.
Chem., 63 (2010), 181198.
The Moving Coordinate System and Euler-Savarys Formula for the

One Parameter Motions On Galilean (Isotropic) Plane
M
ucahit AKBIYIK and Salim YUCE
Department of Mathematics, Facuty of Arts and Sciences
Yildiz Technical University, Esenler, 34210, Istanbul, Turkey
E-mail: makbiyik@yildiz.edu.tr, sayuce@yildiz.edu.tr
Abstract: In this article, one Galilean (or called Isotropic) plane moving relative to two
other Galilean planes (or Isotropic Planes), one moving and the other fixed, was taken into
consideration and the relation between the absolute, relative and sliding velocities of this
movement and pole points were obtained. Also a canonical relative system for one-parameter
Galilean planar motion was defined. In addition, Euler-Savary formula, which gives the
relationship between the curvature of trajectory curves, was obtained with the help of this
relative system.
Key Words: Kinematics, moving coordinate system, Euler Savary formula, Galilean plane
(isotropic plane).
AMS(2010): 53A17, 53A35, 53A40
1. Introduction
Galilean Geometry, is described by Yaglom, [1]. So far, many researcher has done a lot of studies
as [2-4], etc in the Galilean Plane (or Isotropic Plane) and Galilean Space. Also, Euler-Savarys
formula is very famous theorem. It gives relation between curvature of roulette and curvatures
of these base curve and rolling curve, [14]. It takes place in a lot of studies of engineering and
mathematics. A few of them are studies worked by Alexander and Maddocks,[5], Buckley and
Whitfield, [6], Dooner and Griffis, [7], Ito and Takahaski, [8], Pennock and Raje, [9].
uller, [10]; defined one-parameter planar motion in the Euclidean plane E 2 . He
In 1959, M
studied the moving coordinate system and Euler-Savarys formula during one parameter planar
motions. Then, Ergin in 1991 and 1992, [11-[12]; considering the Lorentzian Plane L2 , instead
of the Euclidean plane E 2 , introduced the one parameter planar motion in the Lorentzian plane
L2 and gave the relations between both the velocities and accelerations and also defined the
moving coordinate system. Furthermore, in 2002 Aytun [13] studied the Euler Savary formula
for the one parameter Lorentzian motions as using M ullers Method [10]. And in 2003, Ikawa
[14] gave the Euler-Savary formula on Minkowski without using M ullers Method [10].
1 Received December 26, 2014, Accepted May 28, 2015.
The Moving Coordinate System and Euler-Savarys Formula 89
In 1983, Otto Roschel, [15]; studied kinematics in the isotropic plane. He investigated
fundamental properties of the point-paths, developed a formula analog to the wellknown formula
of Euler-Savary and studied special motions: An isotropic elliptic motion and an isotropic four-
P P
bar-motion. And in 1985, he [16]; studied motions / 0 in the isotropic plane. Given
P P
C 2 -curve k in the moving frame he found the enveloped curve k0 in the fixed frame 0
and considered the correspondance between the isotropic curvatures A and A0 of k and k0 .
Then he investigated third - order properties of the point-paths. And then in 2013, Y uce,
2 2
[17], considering the Galilean Plane G , instead of the Euclidean Plane E or instead of the
Lorentzian Plane L2 , defined one parameter planar Galilean motion in Galilean Plane G2 analog
[10] or [11]. Moreover, they analyzed the relationships between the absolute, relative and sliding
velocities of one-parameter Galilean Planar motion as well as the related pole lines.
Now we investigate the moving coordinate system and Euler Savarys Formula during the
one parameter planar Galilean motion in Galilean Plane G2 analog [10] or [11] by using M
ullers
method.
2. Preliminaries
In this section, the basic information about Galilean geometry which is described by Ya-
glom, [1], will be given.
Let {x} and {x } be two relative frames and origin point O with velocity v on a line o
move according to relative frame {x } , that is, b (t) = b + vt where t is time and b is coordinate
of point O with respect to coordinate system {x } at the moment t = 0 (see, Figure 1).
b
O -v A
o O
x
Figure 1 The rectilinear motion
Then, relation between coordinates of x and x is
x = x + b (t) (1)
= x + b + vt. (2)
Also, since time would be t = t + a (example, t is Gregorian calendar, t is Hijra calendar), we

can write
x = x + vt + b
(3)
t = t + a.
This transformations (1.1) are called Galilean Transformations for rectilinear motions. If point
A (x, t) with coordinate x and t of a (two-dimensional) plane xOt (see, Figure 2) represents
position of point A (x) on a line o at time t, then two-dimensional Geometry which is invariant
under the Galilean Transformations for rectilinear motions is obtained.
90 M
6
t
A(x, t)
-x
O
Figure 2 xOt plane
So, this geometry is called the geometry of Galileos principle of relativity for rectilinear motions
or two-dimensional Galilean geometry and is represented by G2 . Since we shall only talk about
the two-dimensional Galilean geometry in this work, we shall shortly call Galilean plane. If
transformation (3) is arranged as x instead of t and y instead of x, we get
x = x + a
(4)
y = y + vx + b.
This transformation (4) composed of the shear transformation
x1 = x
(5)
y1 = y + vx
and the translation transformation

x = x1 + a
(6)
y = y1 + b.
Theorem 2.1([1]) Transformation (4) maps
(1) lines onto lines;

(2) parallel lines onto parallel lines;
C D
(3) collinear segments AB, CD onto collinear segments A B ,C D with A B = CD
AB ;
(4) a figure F onto a figure F of the same area.
In the Galilean plane, the vectors {g1 = (1, 0), g2 = (0, 1)} are called orthogonal basis vec-
tors of G2 , and also a vector which is parallel to vector g2 is called special vector. If {g1, g2 } are
orthogonal basis vectors and a, b G2 whose coordinates are (x1 , x2 ) and (y1 , y2 ) according to
this basis vectors {g1, g2 } , respectively, then the Galilean inner product of vectors a, b G2
with respect to bases {g1, g2 } is defined by
ha, biG = x1 y1 (7)
(also you can see in [4]). If a, b are special vectors, then the Galilean special inner product of
special vectors is defined by

ha, bi = x2 y2 . (8)
Hence, the norm of every vector a = (x1 , x2 ) G2 on the Galilean plane is denoted by kakG
and is defined by q
kakG = ha, aiG = |x1 | (9)
and the norm of every special vector a = (0, x2 ) G2 on the Galilean Plane is denoted by kak
and is defined by q
kak = ha, ai = |x2 | . (10)
The distance between points A (x1 , x2 ) and B (y1 , y2 ) on the Galilean Plane is denoted by dAB
and is defined by q
dAB = hAB, ABiG = y1 x1 , (11)
where y1 > x1 (see, Figure 3).
Figure 3 The distance between two points in G2
Figure 4 The special distance between two points
That is, dAB is equal to kPP1 k in the sense of Euclidean Geometry. If the distance between
92 M
A (x1 , x2 ) and B (y1 , y2 ) is equal to zero (x1 = y1 ), then special distance of the points A (x1 , x2 )
and B (y1 , y2 ) is denoted by AB and is defined by
q
AB = hAB, ABi = y2 x2 (12)
here y2 > x2 (see, Figure 4). The set of points M (x, y) whose distances from a fixed point
Q (a, b) have constant absolute value r is called a Galilean circle, and is denoted by S. Thus,
the circle S in the Galilean Plane is defined by
2
(x a) = r2 (13)
or
x2 + 2px + q = 0 (14)
where p = a, q = a2 r2 . Also in the Galilean Plane, lines are parallel to y-axis are separable
from class of lines and these lines are called special lines and others are called ordinary lines.
Therefore, the circle S in the Galilean Plane consists of two special lines whose distance from
Q is r (see, Figure 5).
Figure 5 The circle in G2
Figure 6 The angle between two intersecting lines

However, the angle between two ordinary lines y = kx + s and y = k1 x + s1 intersecting at a

point Q = (x0 , y0 ) (see, Figure 6) is defined by
ll1 = k1 k. (15)
But the right angle is defined by angle between ordinary line and special line in the Galilean
Plane. So, the special lines are perpendicular to ordinary lines and also special vectors are
perpendicular to ordinary vectors. Consequently, let S be a unit circle with centered at O and
M (x, y) be a point on S. Assume that l denotes line OM and denotes Ol (see, Figure 7).
Figure 7 The trigonometry in G2
Then, we have
cosg = 1 (16)
and
sing = . (17)
Also, suppose that l1 be another ordinary line and ll1 = . Then we get
cos g ( + ) = 1 (18)
and
sin g ( + ) = sin g cos g + cos g sin g. (19)
We can define a circle by another definition in Euclidean Geometry that the set of points M
from which a given ordinary segment AB (i.e., a segment on an ordinary line) is seen at a
constant directed angle . If we use this definition in the Galilean Plane, we have equation
ax2 + 2b1 x + 2b2 y + c = 0 (20)
which are (Euclidean) parabolas and this set is called a Galilean cycle and denoted by Z. Here
each of lines which are parallel to y-axis, is a diameter of cycle Z and it is denoted by d (see,
Figure 8).
94 M
Figure 8 The cycle in G2 and circle in E 2
Also, the length of an arc AB of a curve is equal to the length s = dAB of the cord AB (see,
Figure 9).
Figure 9 The length of an arc in G2
Thus, the radius of cycle Z is defined by
1
r= . (21)
2a
Furthermore, the curvature of at A is defined as the rate change of the tangent at A, that
is, the curvature of at A is

= lim (22)
s0 s
where = T AT0 is angle between the two neighboring tangents, s = arcAM is the scalar
arc element of such that M is a point of curve (see, Figure 10).
Figure 10 The curvature in G2
Therefore, the radius of curvature of at A, denoted r, is
1
r= . (23)

Now, let consider all cycles passing through the points of and having the same tangent AT = l
at A as . From these cycles, we select that for any other cycle Z0 which is closest distance
M M between points on and Z sufficiently close to A which project to the same point N on
l is larger than the distance between the corresponding points M and M0 of and Z0 . This
cycle Z0 is called the osculating cycle of curve at A (see, Figure 11), [1].
Figure 11 The cycle in G2
3. One Parameter Planar Galilean Motion

Let G and G G = G = G2 be moving and fixed Galilean planes and {O; g1 , g2 } and {O ; g1 , g2 }
be their coordinate systems, respectively. The motion defined by the transformation
x = x u (24)
96 M
is called as one parameter planar Galilean motion and denoted by B = G/G where
OO = u =u1 g1 + u2 g2 (25)
for u1 , u2 R, x, x are the coordinate vectors with respect to the moving and fixed rectangular
coordinate systems of a point X = (x1 , x2 ) G respectively. Also, these vectors x, x and u
and shear rotation angle between g1 and g1 are continuously differentiable functions of a
time parameter t (see, Figure 12).
Figure 12 The motion B = G/G
We can write
g1 = g1 + g2
(26)
g2 = g2
for the shear rotation angle = (t) between g1 and g1 . In this study, we suppose that
d
(t)
= 6= 0 (27)
dt
where . denotes the derivation with respect to t. By differentiating the equations (25) and
(26), the derivative formulae of the motion B = G/G are
g1 = g
2
g2 = 0 (28)
u = u1 g1 + (u2 + u1 )
g2 .
The velocity of the point X with respect to G is defined as the relative velocity Vr and is
founded by
Vr x1 g1 + x2 g2 . (29)
Furthermore, velocity of the point X G according to G is known as the absolute velocity,

and is found as
Va = u1 g1 + (u2 u1 + x1 )
g2 + Vr . (30)
Thus, we get the sliding velocity
Vf = u1 g1 + (u2 u1 + x1 )
g2 . (31)
In the general one parameter motions, the points whose sliding velocity is zero, i.e.,Vf = 0 are
called the pole point or instantaneous shear rotation pole point and in the Galilean Plane G,
the pole point P = (p1 , p2 ) G of the motion B = G/G is defined by

p = u + u2 (t)
1 1
(t)
p... (32)
p2 = p2 (t ())
for R (see, Figure 13) [17].
Figure 13 The pole line in Galilean plane
Corollary 3.1([17]) During one parameter planar motion B = G/G invariants points in both
planes at any instant t have been on a special line in the plane G. That is, there only exists
pole line in the Galilean Plane G at any instant t. For all t I, this pole lines are parallel
to y-axis and these pole lines form bundles of parallel lines. Using equations (31)and (32), for
sliding velocity, we can write
Vf = {0g1 + (x1 p1 ) g2 } .
(33)
Corollary 3.2([17]) During one parameter planar motion B = G/G , the pole ray PX =
(x1 p1 ) g1 + (x2 p2 ) g2 and Vf = {0g1 + (x1 p1 ) g2 } are perpendicular vectors, i.e.,

PX, Vf G = 0. Thus, under the motion B = G/G , the focus of the points X G is an orbit
curve that is normal pass through the shear rotation pole P .
Corollary 3.3([17]) Under the motion B = G/G , the norm of the sliding velocity Vf is
kVf k = kPXkG ||
. (34)
That is, during the motion B = G/G , all of the orbits of the points X G are such curves
98 M
whose normal lines pass thoroughly the pole point P . At any instant t, the motion B = G/G
is a Galilean instantaneous shear rotation with the angular velocity about the pole point P .
Since there exist pole points in every moment t, during the one-parameter plane motion
B = G/G any pole point P is situated various positions on the plane G and G . The position
of the pole point P on the moving plane G is usually a curve and this curve is called moving
pole curve and is denoted by (P ) . Also the position of this pole point P on the fixed plane G
is usually a curve and this curve is called fixed pole curve and is denoted by (P ) .
4. The Moving Coordinate System on the Galilean Planes
In this section, we study on three Galilean planes, and investigate relative, sliding and absolute
velocity, a point of X on a plane according to the other two plane and relations between the
pole points. Let A and G be moving and G be fixed Galilean plane and {B, a1 , a2 } , {O; g1 , g2 }
and {O ; g1 , g2 } their coordinate systems, respectively (see, Figure 14).
Figure 14 The two moving and one fixed coordinate system
Assume that and are rotation angles of one parameter planar motions A/G and A/G ,
respectively. Let us consider a point X with the coordinates of (x1 , x2 ) in moving plane A.
Since
BX = x1 a1 + x2 a2 (35)
OB = b = b1 a1 + b2 a2 (36)
O B = b = b1 a1 + b2 a2 (37)
are vectors on the moving system of A, we have
x = OX = OB + BX = b + x1 a1 + x2 a2 (38)
x = O X = O B + BX = b +x1 a1 + x2 a2 (39)
where vector x and x denote the point X with respect to the coordinate systems of G and G ,
respectively. Lets find the velocities of one parameter motion with the help of the differentiation
the equations (38) and (39). Assume that d... denotes the differential with respect to G and
d... denotes the differential with respect to G .
The derivative equations in motion B = A/G, are
da1 = da2 (40)
da2 = 0 (41)
db = db1 a1 + (b1 d + db2 ) a2 (42)
and the derivative equations in motion B = A/G taking d b = d b , are
d a1 = d a2 (43)
d a2 = 0 (44)
d b = db1 a1 + (b1 d + db2 ) a2 . (45)
So differential of X with respect to G is
dx = (1 + dx1 ) a1 + (2 + x1 + dx2 ) a2 (46)
where 1 = db1 , 2 = db2 + b1 d, = d. Therefore the relative velocity vector of X with

respect to G is
dx
Vr = (47)
dt
and also differential of X with respect to G is
d x = (1 + dx1 ) a1 + (2 + x1 + dx2 ) a2 (48)
where 1 = db1 , 2 = db2 + b1 d, = d. Thus, the absolute velocity vector of X with respect
to G is
d x
Va = . (49)
dt
Here 1 , 2 , 1 , 2 , and are the Pfaffian forms of one parameter motion with respect to t.
If Vr = 0 and Va = 0 then the point X is fixed in the planes G and G , respectively. Thus, the
conditions that the point is fixed in planes G and G become
dx1 = 1 , dx2 = 1 x1 (50)
and
dx1 = 1 , dx2 = 2 x1 , (51)
respectively. Substituting equation (50) into equation (48) and considering that the sliding
df x
velocity of the point X is Vf = dt , we have
df x = {1 1 } a1 + {(2 2 ) + ( )} a2 . (52)
100 M
Therefore from (46), (48) and (52) we may give the following theorem.
Theorem 4.1 If X is a fixed point on G, then we have
d x = df x + dx, (53)
that is, Va = Vf +Vr . Thus, velocities law is preserved.
e corresponds the differential of

Remark 4.2 In the motion A/G, the absolute velocity Va
dx = 1 a1 + {2 + x1 } a2 + dx1 a1 + dx2 a2 (54)
e r which is the velocity of X

according to plane G of the point X, and the relative velocity V
according to the plane A, is equal to the differential of
dx1 a1 + dx2 a2 (55)
e f with respect to motion A/G is

with respect to A of the point X. Thus the sliding velocity V
the differential of
1 a1 + {2 + x1 } a2 (56)
e is equal
according to G of the point X. Similarly, in the motion A/G , the absolute velocity V a
to the differential of
d x = 1 a1 + {2 + x1 } a2 + dx1 a1 + dx2 a2 (57)
e is the differential of
with respect to G of the point X, and the relative velocity V r
dx1 a1 + dx2 a2 (58)
e corresponds the differential of

with respect to A of the point X. So the sliding velocity V f
1 a1 + {2 + x1 } a2 (59)
with respect to G of the point X.Since the motion G/G is characterized by the inverse motion
of A/G and the motion A/G , we have the sliding motion df x when we subtract from the
sliding velocity of the motion A/G to the sliding velocity of the motion A/G. So we can write
e V
Vf = V ef. (60)
f
To avoid the cases of pure translation we assume that 6= 0, 6= 0.

In one parameter planar Galilean motions the pole point is characterised by vanishing
sliding velocity, i.e., df x = 0. So, the pole point P of the one parameter planar motion G/G is
obtained as
2
p1 = 2 (61)

p2 = p2 () , R (62)
where BP = p1 a1 + p2 a2 . Note that here we find
1 1 = 0. (63)
5. The Shear Rotation Poles for Moving Galilean Planes with

Respect to the Other
Let us have three planes such as A, G, G moving with respect to together and also occur in
two one parameter planar Galilean motion with respect to each other. In the determined time
t, pairs of plane (A, G), (A, G ) and (G, G ) have a determined shear rotation pole line, and in-
stantaneous shear rotation motions arise with angular velocity about the pole line. Accordingly,
three planes moving with respect to together constitute a three-member kinematic chain.
Motion A/G of plane A with respect to plane G is formulated by equation system (41).

Here d = is infinitesimal shear rotation angle, that is, dt is an angular velocity. Differential
of point X with respect to plane A
dBX =dx1 a1 + dx2 a2 . (64)
The differential corresponds to relative velocity with respect to plane A. If point X is fixed,
then we can write dBX = 0. In the equation (46), the differential of point X is given with
respect to plane G. From here, sliding velocity of point X with respect to motion A/G is
1 a1 + (2 + x1 ) a2 . (65)
However the shear rotation pole of motion is characterized by vanishing the sliding velocity.
So, for the shear rotation pole line q of motion A/G, we have
n 2
q... q1 = , q2 = q2 () R. (66)

Similarly for the shear rotation pole line q of motion A/G , we get

q ... q1 = 2 , q2 = q2 () , R.

(67)

And also, the angular velocity of motion G/G is
d ( )
= (68)
dt dt
and for the shear rotation pole line p, from equations (61) and (62) we can rewrite

p = 2 2
1
p... . (69)
p2 = p2 () , R
102 M
So, we give following theorem.
Theorem 5.1 If three Galilean planes form one parameter planar Galilean motions pairwisely,
there exist three shear rotation pole lines at every moment t, and each of these three lines is
parallel to the others.
Corollary 5.2 Generally, if there are nGalilean planes which form one parameter planar
Galilean motions pairwisely, then we tell of nmember
kinematic chain. If the each motions is
n
connected time (real) parameter t , there exit relative shear rotation pole lines at every
2
moment t and every each line is parallel to each others.
Theorem 5.3 The rate of the distance of three shear rotation poles is as the rate of their
angular velocities.
2 2
Proof Since q1 = 2 , q1 = 2 and p1 = , it is hold
(q1 q1 ) : (p1 q1 ) : (q1 p1 ) = ( ) : : . (70)
Thus, we can write

QQ : Q P : P Q = ( ) : : . (71)
G G G
6. Euler-Savarys Formula for One Parameter Motions in the Galilean Plane
We studied one parameter Galilean motion adding {B, a1 , a2 } moving system to the motion
of G with respect to G . Now, in this section, we choose a special relative system {B, a1 , a2 }
satisfying the following conditions:
(i) The initial point B of the system is a pole point P on the pole line that coordinates
are p1 and p2 .
(ii) The axes {B; a1 } coincides with the common tangent of the pole curves (P ) and (P ) .
If we consider the condition i), then from equations (61) and (62) we have
p1 = p2 = 0. (72)
Thus, we can write

1 = 1 , 2 = 2 . (73)
Therefore we get
db = dp = 1 a1 + 2 a2 = d p = db . (74)
Considering the condition ii), then we have 2 = 2 = 0. So if we choose 1 = 1 = , then the

derivative equations for the canonical relative system {P, a1 , a2 } are
da1 = a2 , d a1 = a2
da2 = 0, d a2 = 0 (75)

dp = a1 , d p = a1 .
Moreover, is the cotangent angle, that is, two neighboring tangets angle of curve (P ) , and
is also the cotangent angle of curve (P ) where, = ds is the scalar arc element of the pole
curves (P ) and (P ). And so : is the curvature of the moving pole curve (P ). Similarly,
: is the curvature of the fixed pole curve (P ). Hence from (23) the radius of curvature of
the pole curves (P ) and (P ) are

r= (76)

and

r = (77)

respectively. Moving plane G rotates the infinitesimal instantaneous angle of the d =
around the shear rotation pole P within the time scale dt with respect to fixed plane G .
Therefore the angular velocity of shear rotational motion of G with respect to G is
d
= = . (78)
ds ds
Hence we get
d 1 1
= = (79)
ds ds r r
from equations (76), (77) and (78). We accept that for the direction of unit tangent vector a1 ,
pole curves (P ) and (P ) are drawn to the positive x-axis direction that is, ds
dt > 0, and so we
have r > 0. Similarly we can write r > 0.
Now we will investigate case of the point X which is on the diameter d of osculating cycle
of trajectory curve which is drawn in the fixed plane G by a point X of moving plane G in
the movement G/G . In the canonical relative system, let coordinates of points X at plane G
and X at plane G be the (x1 , x2 ) and (x1 , x2 ), respectively. In the movement D/D , there
is a point X which is on center of curvature of osculating cycle of trajectory curve of X are
situated together with the instantaneous rotation pole P in every moment t such that
PX = x1 a1 + x2 a2
and
PX = x1 a1 + x2 a2
have same direction which passes the pole point P. So we can write
x1 : x2 = x1 : x2
104 M
or
x1 x2 x1 x2 = 0. (80)
Considering the condition ii) we obtain the condition that the point X to be the fixed in the
moving plane G is
dx1 = , dx2 = x1 (81)
and the point X to be the fixed in the moving plane G is
dx1 = , dx2 = x1 . (82)
Differentiating the equation(80) and from the conditions (81) and (82) we have
(x2 x2 ) + x1 x1 ( ) = 0. (83)
If the polar coordinates are passed, then we get
x1 = a cos g = a, x2 = a sin g = a (84)
x1 = a cos g = a , x2 = a sin g = a . (85)
Thus, we can write

(a sin g a sin g) + aa ( ) = 0. (86)
So from last equation and equation (79) we have

1 1 1 1 d
sin g = = . (87)
a a r r ds
Here, r and r are the radii of curvature of the pole curves P and P , respectively. ds represents
the scalar arc element and d represents the infinitesimal Galilean angle of the motion of the
pole curves. The equation (87) is called the Euler-Savary formula for one-parameter motion in
Galilean plane G.
Consequently, the following theorem can be given.
Theorem 6.1 Let G and G be the moving and fixed Galilean planes, respectively. A point
X G, draws a trajectory whose a point at the normal axis of curvature is X on the plane G
in one-parameter planar motion G/G . In the inverse motion of G/G , a point X assumed on
G draws a trajectory whose a point at the normal axis of curvature is X on the plane G. The
relation between the points X and X which is given by the Euler-Savary formula given in the
equation (87).
References
[1] Yaglom I.M., A simple non-Euclidean Geometry and its Physical Basis, Springer-Verlag,
New York, 1979.
[2] O. Roschel, Die Geometrie des Galileischen Raumes, Habilitationsschrift, Institut f
ur
Math. und Angew. Geometrie, Leoben, 1984.

[3] Pavkovic B.J. and Kamenarovi c, I., The equiform differential geometry of curves in the
Galilean space, Glasnik Matemati cki, 22(42), 449-457, 1987.
[4] Helzer G., Special relativity with acceleration, Mathematical Association of America, The
American Mathematical Monthly, Vol.107, No. 3, 219-237, 2000.
[5] Alexander J.C., Maddocks J.H., On the maneuvering of vehicles, SIAM J. Appl. Math.,
48(1): 38-52,1988.
[6] Buckley R., Whitfield E.V., The Euler-Savary formula, The Mathematical Gazette, 33(306):
297-299, 1949.
[7] Dooner D.B., Griffis M.W., On the spatial Euler-Savary equations for envelopes, J. Mech.
Design, 129(8): 865-875, 2007.
[8] Ito N., Takahashi K.,Extension of the Euler-Savary equation to hypoid gears, JSME
Int.Journal. Ser C. Mech Systems, 42(1): 218-224, 1999.
[9] Pennock G.R., Raje N.N., Curvature theory for the double flier eight-bar linkage, Mech.
Theory, 39: 665-679, 2004.
[10] Blaschke W., and M uller H.R., Ebene Kinematik, Verlag Oldenbourg, M unchen, 1959.
[11] A.A.Ergin, On the one-parameter Lorentzian motion, Comm. Fac. Sci. Univ. Ankara,
Series A 40, 5966, 1991.
[12] A.A.Ergin, Three Lorentzian planes moving with respect to one another and pole points,
Comm. Fac. Sci.Univ. Ankara, Series A 41,79-84,1992.
[13] I.Aytun, Euler-Savary Formula for One-Parameter Lorentzian Plane Motion and its Lorent-
zian Geometrical Interpretation, M.Sc. Thesis, Celal Bayar University, 2002.
[14] T.Ikawa, Euler-Savarys formula on Minkowski geometry, Balkan Journal of Geometry and
Its Applications, 8 (2), 31-36, 2003.
[15] Otto Roschel, Zur Kinematik der isotropen Ebene., Journal of Geometry,21,146156,1983.
[16] Otto Roschel, Zur Kinematik der isotropen Ebene. II., Journal of Geometry, 24, 112
122,1985.
[17] Akar M., Yuce S. and Kuruoglu N, One parameter planar motion in the Galilean plane,
International Electronic Journal of Geometry, 6(1), 79-88, 2013.
Laplacian Energy of Binary Labeled Graph
Pradeep G.Bhat, Sabitha DSouza and Swati S.Nayak

(Department of Mathematics, Manipal University, PIN 576104, Manipal, India)
E-mail: pg.bhat@manipal.edu, sabitha.dsouza@manipal.edu, swati.nayak@manipal.edu
Abstract: Let G be a binary labeled graph and Al (G) = (lij ) be its label adjacency
P
n
matrix. For a vertex vi , we define label degree as Li = lij . In this paper, we define label
j=1
Laplacian energy LEl (G). It depends on the underlying graph G and labels of the vertices.
We compute label Laplacian spectrum of families of graph. We also obtain some bounds for
label Laplacian energy.
Key Words: Label Laplacian matrix, label Laplacian eigenvalues, label Laplacian energy.
AMS(2010): 05C50, 15A15, 15A18.
1. Introduction
For an n-vertex graph G with adjacency matrix A whose eigenvalues are 1 > 2 > . . . > n ,
Pn
the energy of the graph G is defined as E(G) = |i |. The concept of Energy of graph
i=1
was introduced by Ivan Gutman, in connection with the -molecular energy. The matrix
L(G) = D(G) A(G) is the Laplacian matrix of (n, m) graph G. If 1 > 2 > . . . > n = 0
are the eigenvalues of L(G), then the Laplacian energy of G is defined as
P
n
2m
LE(G) = |i n |
i=1
However, in the last few years, research on graph energy has much intensified, resulting
in a very large number of publications which can be found in the literature [4, 5, 6, 7, 9, 16].
In spectral graph theory , the eigenvalues of several matrices like adjacency matrix, Laplacian
matrix [8], distance matrix [10] etc. are studied extensively for more than 50 years. Recently
minimum covering matrix, color matrix, maximum degree etc are introduced and studied in
[1,2,3].
Motivated by this, P.G. Bhat and S. DSouza have introduced a new matrix Al (G) called
label matrix [14] of a binary labeled graph G = (V, X), whose elements are defined as follows:
1 Received December 15, 2014, Accepted May 20, 2015.

Laplacian Energy of Binary Labeled Graph 107

a, if vi vj X(G) and l(vi ) = l(vj ) = 0,

b, if vi vj X(G) and l(vi ) = l(vj ) = 1,
lij =

c, if vi vj X(G) and l(vi ) = 0, l(vj ) = 1 or vice-versa,

0, otherwise.
where a, b, and c are distinct non zero real numbers. The eigenvalues 1 , 2 , . . ., n of Al (G) are
said to be label eigenvalues of the graph G and form its label spectrum. The label eigenvalues
satisfy the following simple relations:
n
X n
X
i = 0 and 2i = 2Q (1.1)
i=1 i=1
where
Q = n 1 a 2 + n 2 b 2 + n 3 c2 (1.2)
Where n1 , n2 and n3 denotes number of edges with (0, 0), (1, 1) and (0, 1) as end vertex labels
respectively.
P
n
The label degree of the vertex vi , denoted by Li , is given by Li = lij . A Graph G is said
i=1
to be k-label regular if Li = k for all i. The label Laplacian matrix of a binary labeled graph G
is defined as
Ll (G) = Diag(Li ) Al (G)
where Diag(Li ) denotes the diagonal matrix of the label degrees. Since Ll (G) is real symmetric,
all its eigenvalues i , i = 1, 2, . . . , n, are real and can be labeled as 1 > 2 > . . . > n . These
form the label Laplacian spectrum of G. Several results on Laplacian of Graph G are reported
in the Literature ([6, 11, 12, 13, 16]).
This paper is organized as follows. In section 2, we establish relationship between i and
i and some general results on Laplacian label eigenvalues i . In section 3, lower bound and
upper bounds for LEl (G) are obtained. In the last section label Laplacian spectrum is derived
for family of graphs.
2. Label Laplacian Energy
The following Lemma 2.1 shows the similarities between the spectra of label matrix and label
Laplacian matrix. For a labeled graph, let PA (x) and PL (x) denote the label and label Laplacian
characteristic polynomials respectively.
Lemma 2.1 If {1 , 2 , . . . , n } is the label spectrum of k-label regular graph G, then {k

n , k n1 , . . . , k 1 } is the label Laplacian spectrum of G.
Proof The label Laplacian characteristic polynomial for k-label regular graph G is given
108 Pradeep G.Bhat, Sabitha DSouza and Swati S.Nayak
by
PL (x) = det(Ll (G) xI) = (1)n det(Al (G) (k x)I) = (1)n PA (k x) (2.1)
Thus, if 1 > 2 > > n is the label spectrum of k-label regular graph G, then from equation
(2.1), it follows that k n > k n1 > . . . > k 1 is the label Laplacian spectrum of G.2
We first introduce the auxiliary eigenvalues i , defined as
1 P
n
i = i n Lj
j=1
Lemma 2.2 If 1 , 2 , , n are the label Laplacian eigenvalues of Ll (G), then

n
X n
X
2i = 2Q + L2i .
i=1 i=1
Proof We have
n
X n X
X n X n
X
2i = trace (Ll (G))2 = lij lji = 2 2
lij + 2
lii
i=1 i=1 j=1 i<j i=1
n
X Xn
= 2[n1 (a)2 + n2 (b)2 + n3 (c)2 ] + L2i = 2Q + L2i 2
i=1 i=1
P
n P
n
Lemma 2.3 Let G be a binary labeled graph of order n. Then i = 0 and i2 = 2R,
i=1 i=1
where
!2
1
P
n
1
P
n
R=Q+ 2 Li n Lj
i=1 j=1
and Q is given by equation (1.2).
Proof Note that
P
n P
n P
n P
n
i = tr(Ll (G)) = Li and 2i = L2i + 2Q
i=1 i=1 i=1 i=1
From which we have,
!
P
n P
n
1 P
n P
n P
n
i = i n Lj = i Lj = 0
i=1 i=1 j=1 i=1 j=1
and
2
n
X n
X Xn
i 1
i2 = Lj
i=1 i=1
n j=1
n n n n
!2
X 2X X 1X
= 2i Lj i + Lj
i=1
n i=1 i=1
n i=1
n n
!2 n
!2
X 2 X 1X
2
= Li + 2Q Lj + Lj
i=1
n i=1 n i=1
n n
!2
X 1X
= 2Q + Li
n i=1
Lj = 2R 2
i=1
Definition 2.1 Let G be a binary labeled graph of order n. Then the label Laplacian energy of
P
n
G, denoted by LEl (G), is defined as |i |, i.e.
i=1
P
n
1 P
n
LEl (G) = |i n Lj |
i=1 j=1
In 2006, I.Gutman and B.Zhou defined Laplacian energy LE(G) of a graph G. More on
Laplacian energy reader can refer ([8], [15], [17], [18]). In Chemistry, there are situations where
chemists use labeled graphs, such as vertices represent two distinct chemical species and the
edges represent a particular reaction between two corresponding species. We mention that this
paper deals only the mathematical aspects of label Laplacian energy of a graph and it is a new
concept in the literature.
Lemma 2.4 If G is k- label regular, then LEl (G) = El (G).
1 P
n
If G is k- label regular, then k = Li = n Lj for i = 1, 2, , n. Using Lemma 2.1,
j=1
1
P
n
i = i n Lj = (k n+1i ) k = n+1i
j=1
for i = 1, 2, , n. Hence, the lemma follows from the definitions of the label energy and label
Laplacian energy.
2. Bounds for the Label Laplacian Energy
Lemma 3.1([17]) Let a1 , a2 , . . . , an be non-negative numbers. Then

! n1 !2 ! n1
n n n n n n
1X Y X X 1X Y
n ai ai 6n ai ai 6 n(n 1) ai ai .
n i=1 i=1 i=1 i=1
n i=1 i=1
Theorem 3.1 Let G be a binary labeled graph with n vertices and m edges. Then
q q
2 2
2R + n(n 1) n 6 LEl (G) 6 2(n 1)R + n n ,
!
P
n
1
where = det Ll (G) n L j I .
j=1
Proof Note that

n
X n
X
|i | = LEl (G) and i2 = 2R,
i=1 i=1
!2
1 P n
1 P
n
where R = n1 (a)2 + n2 (b)2 + n3 (c)2 + Li n Lj .
2 i=1 j=1
Using Lemma 3.1, it can be easily checked that Theorem 3.1 is true if = 0. Now we
assume that 6= 0. By setting ai = i2 , i = 1, 2, . . . , n, and
! n1
n n
1X 2 Y
K = n i2 > 0.
n i=1 i i=1
From Lemma 3.1, we have
n n
!2
X X
K 6n i2 |i | 6 (n 1)K,
i=1 i=1
which can be further expressed as
K 6 2nR (LEl (G))2 6 (n 1)K
2nR (n 1)K 6 (LEl (G))2 6 2nR K, (3.1)
where
! n1
n
Y
1 1 2 2
K = n i2 2
i =n 2R n = 2R n n
n i=1
n
Substituting in equation (3.1), we obtain

q q
2 2
2R + n(n 1) n 6 LEl (G) 6 2R(n 1) + n n . 2
Theorem 3.2 Let G be a binary labeled graph of order n > 2. Then

2 R 6 LEl (G) 6 2nR.
Proof Consider the term
n X
n n n
! n
X 2
X X X
S = (|i | |j |) = 2n |i |2 2 |i | |j |
i=1 j=1 i=1 i=1 j=1
2 2
= 2n.2R 2(LEl (G)) = 4nR 2(LEl (G))

Note that S > 0, i.e., 4nR 2(LEl (G))2 > 0, which implies LEl (G) 6 2nR. We have
n 2
P
i = 0 and the fact that R > 0,
i=1
!2
n
X n
X X X X

2R = i2 = i 2
i j 6 2 i j 6 2 |i ||j | (3.2)
i=1 i=1 16i<j6n 16i<j6n 16i<j6n
Thus,
n
!2 n
X X X
2
LEl (G) = |i | = |i |2 + 2 |i ||j |
i=1 i=1 16i<j6n
> 2R + 2R = 4R

from Lemma 2.3 and equation (3.1). Hence, LEl (G) > 2 R. 2
Theorem 3.3 Let G be a labelled graph of order n. Then

v
n
u
u n
!2
1 X u 1X
LEl (G) 6 Li + t(n 1) 2R Li .
n i=1 n i=1
Proof We have
n n
1X 1X
n = 0 Li = Li .
n i=1 n i=1
Consider the non-negative term
n1
X n1
X
S= (|i | |j |)2
i=1 j=1
n n
! n

X X X
= 2(n 1) i2 2 |i | |j |
i=1 i=1 j=1
!2 !2
n n
1X 2 LEl (G) 1
X
= 2(n 1) 2R Li Li > 0.
n i=1 n i=1
Hence, v
u !2
Xn u n
1X
LEl (G) 6
1
n i=1
u
Li + t(n 1) 2R
n i=1
Li . 2
4. Label Laplacian Spectrum of Some Graphs
Theorem 4.1 For n > 2, the label Laplacian spectrum of complete graph Kn is

0 ma + (n m)c (n m)b + mc nc
1 m1 nm1 1
where m vertices are labeled zero, n m vertices are labeled one and 0 6 m 6 n.
Proof Let v1 , v2 , , vm vertices of Kn be labeled zero and vm+1 , vm+2 , , vn be labeled

1. Then the label degree of vertex vi is L(vi ) = (m 1)a + (n m)c for i = 1, 2, . . . , m,
L(vi ) = (n m 1)b + mc for i = m + 1, m + 2, . . . , n,

[ma + (n m)c]Im aJmm cJm(nm)
Ll (Kn ) =
cJ(nm)m [(n m)b + mc]Inm bJ(nm)(nm)
Consider
det (I Ll (Kn ))

[ {ma + (n m)c}]Im + aJmm cJm(nm)
= .

cJ(nm)m [ {(n m)b + mc}]Inm + bJ(nm)(nm)
Step 1 Replacing column C1 by C1 = C1 + C2 + . . .+ Cn , we obtain determinant det(B).
Step 2 In determinant B, replace the row Ri by Ri = Ri Ri1 for i = 2, 3, . . . , m, m +

2, m + 3, . . . , n, we obtain
m1 nm1
det(B) = ( {(m 1)a + (n m)c}) ( {(n m 1)b + mc}) det(C).
Step 3 By changing Ci by Ci = Ci + Ci+1 + . . . + Cn for i = m + 1 to n in determinant
C, we get a new determinant D of order m + 1, i.e.

1 a a . . . a (n m)c

0 1 0 ... 0 0

0 1 1 ... 0 0

det(D) = . .. .. .. ..
.. . . ... . .

0 0 0 ... 1 0

1 c c . . . c mc
Step 4 By expanding determinant D over the first column , we obtain det(D) = mc +

(1)m+2 (1)m+1 (n m)c = nc.
Step 5 By back substitution,
m1 nm1
det (I Ll (G)) = ( [ma + (n m)c]) ( [(n m)b + mc]) ( nc).
Hence, label Laplacian spectrum of Kn is,

1
2
m1 nm1 1
Corollary 4.1 For n > 2, the label Laplacian spectrum of Kn {(0, 0)} is

0 (m 2)a + (n m)c ma + (n m)c (n m)b + mc nc
1 1 m2 nm1 1
Corollary 4.2 For n > 2, the label Laplacian spectrum of Kn {(1, 1)} is

0 ma + (n m)c (n m 2)b + mc (n m)b + mc nc
1 m1 1 nm2 1
Theorem 4.2 The label Laplacian spectrum of star graph Sn is

0 a c +
2 2
1 m 2 n m 1 1 1
where m denotes the number of vertices including the central vertex labeled zero, remaining
p
vertices labeled one, m n, = ma + (n m + 1)c and = [ma + (n m + 1)c]2 4acn.
Proof Let v1 , v2 , , vm be labeled as zero and remaining vertices be labeled as one,where

v1 is the central vertex. Then, L(v1 ) = a(m 1) + c(n m) and

a, for i = 2, 3, , m
L(vi ) =
c, for i = m + 1, m + 2, , n

a(m 1) + c(n m) a a a c c c

a a 0 0 0 0 0

a 0 a 0 0 0 0

.. .. .. .. .. .. .. .. ..
. . . . . . . . .

a 0 0 0 0 0 ... 0

Ll (Sn ) =
a 0 0 a 0 0 0

c 0 0 0 c 0 0

c 0 0 0 0 c 0

.. .. .. .. .. .. .. .. ..
. . . . . . . . .

c 0 0 0 0 0 c
where rows and columns are denoted by v1 , v2 , , vm , vm+1 , vm+2 , , vn for the matrix Ll (Sn ).
Consider det(I Ll (G)).
Step 1 Replace the column C1 by C1 = C1 +C2 + +Cn . Then we see that det(ILl (G))
is of the form det(B).
Step 2 In det(B), replace Ri by Ri = Ri Ri1 , i = 3, 4, , m, m+2, , n. Simplifying

we get det(B) = ( a)m2 ( c)nm1 det(C).
Step 3 In det(C), replace Ci by Ci = Ci + Ci+1 + + Cn for i = m, m + 1, , n. Then

it reduces to the order m + 1.
Step 4 In det(C), Replacing Ci by Ci = Ci + Ci+1 + + Cm for i = 2, 3, , (m 1),

we get

1 (m 1)a (m 2)a . . . 2a a (n m)c

1 ( a) 0 ... 0 0 0

0 0 1 ... 0 0 0

det(C) = . .. .. . .. ..
.. . . . . . .. . .

0 0 0 ... 0 1 0

1 0 0 . . . 0 0 ( c)
Expanding over the last column, we get det(C) = 2 [am + (n m + 1)c] + acn. By
back substitution, we obtain

det(I Ll (G)) = ( a)m2 ( c)nm1 2 [am + (n m + 1)c] + acn .
Hence, label Laplacian spectrum of Sn is given by

0 a c +
2 2
1 m 2 n m 1 1 1
p
where = ma + (n m + 1)c and = [ma + (n m + 1)c]2 4acn. 2
Corollary 4.3 The label Laplacian spectrum of star graph Sn is

0 b c +
2 2
1 m 2 n m 1 1 1
where m denotes the number of vertices including the central vertex labeled zero, remaining
p
vertices labeled one, m n, = mb + (n m + 1)c and = [mb + (n m + 1)c] 4bcn.
Proof Let v1 be the central vertex . Let v1 , v2 , , vm be labeled as one and remaining
vertices be labeled as zero. Then , L(v1 ) = b(m 1) + c(n m) and

b, for i = 2, 3, , m
L(vi ) =
c, for i = m + 1, m + 2, , n
The remaining proof of this corollary is similar to Theorem 4.2. 2
Corollary 4.4 If the vertices of cycle C2n are labeled 0 and 1 alternately, then LEl (C2n ) =
cLE(C2n ) = cE(C2n ).
Corollary 4.5 If the vertices of path Pn are labeled 0 and 1 alternately, then LEl (Pn ) =
cLE(Pn ).
Lemma 4.1([10]) Let M, N, P, Q be matrices, M invertible and

M N
S=
P Q
Then det(S) = det(M ) det(Q P M 1 N ). Furthermore, if M , P are commute, then det(S) =

det(M Q P N ).
Theorem 4.3 The label Laplacian spectrum of complete bipartite graph K(r, s) with m1 6
r, m2 6 s, the number of zeros in the vertex set of order r, s respectively, is given by

0 (am + (s m )c) (cm + (s m )b) (am + (r m )c) (cm + (r m )b)
2 2 2 2 1 1 1 1
1 m1 1 r m1 1 m2 1 s m2 1
and the roots of
[3 2 {(a + c)(m1 + m2 ) + (b + c)((r + s) (m1 + m2 ))}

+{ac(m21 + m22 ) + (c2 + ab)(m2 (s m2 ) + m1 (r m1 ))
+(ab + bc + ca)(m1 (s m2 ) + m1 (r m1 )) + bc((r m1 )
+(s m2 ))2 + (b2 + c2 )(r m1 )(s m2 )} {(a + c)bc(r m1 )(s m2 )(m1 + m2 )
+abc (m1 (s m2 )(s + m1 m2 ) + m2 (r m1 )(r + m2 m1 ))
+bc2 (r m1 )(s m2 )(r + s m1 m2 )
+ac(a + c)m21 m2 + c(ac + c2 + ab)m1 m2 (s m2 ) + ac(b + c)m1 m2 (r m1 )}] = 0.
0} 111
Proof Let the labels of r + s vertices of K(r, s) be |000{z | {z 1} and |000{z
0} 111
| {z 1}.
m1 rm1 m2 sm2

Arr Brs aJm1 m2 cJm1 sm2
Ll (K(r, s)) = with B =
T
Bsr Css cJrm1 m2 bJrm1 sm2
(r+s)x(r+s) rxs
Characteristic polynomial of Ll (K(r, s)) is

(I A)rr Brs
(Ll (K(r, s)), ) = = |I A||(I C) B T A1 B|
(4.1)
T
Bsr (I C)ss
by Lemma 4.1. Let us denote the label degree of m1 , r m1 , m2 and s m2 vertices as

W = cm2 + (s m2 )b, X = cm2 + (s m2 )b, Y = am1 + (r m1 )c and Z = cm1 + (r m1 )b
respectively. Then A = Diag[ W, W, . . . , W, W, X, X, . . . , X],
C = Diag[ Y, Y, . . . , Y, Y, Z, Z, . . . , Z]. Note that
n o n o
m1 a2 c2 (rm1 ) acm1 bc(rm1 )
W + X Jm2 m2 W + X Jm2 sm2
B T A1 B = n o n
b2 (rm1 )
o
acm1 bc(rm1 ) c 2 m1
W + X Jsm2 m2 W + X Jsm2 sm2
By applying elementary transformations, det(C B T A1 B) reduces to order m2 + 1. Hence,
|C B T A1 B| = ( Y )m2 1 ( Z)sm2 1 det(E), (4.2)
where

( Y ) m2 G (m2 1)G (m2 2)G ... G (s m2 )H

0 1 0 ... 0 0

0 0 1 ... 0 0
det(E) =

.. ..
. ... .

m2 H (m2 1)H (m2 2)H . . . H ( Z) (s m2 )K
m1 a2 c2 (rm1 ) acm1 bc(rm1 ) c 2 m1 b2 (rm1 )

and G = W + X , H= W + X and K = W + X .
Now expression (4.1) becomes
(Ll (K(r, s), ) = ( W )m1 1 ( X)rm1 1 ( Y )m2 1 ( Z)sm2 1

[( W )( X)( Y )( Z) c2 m1 (s m2 )( X)( Y )
b2 (r m1 )(s m2 )( W )( Y ) a2 m1 m2 ( Z)( X)
c2 (r m1 )m2 ( Z)( W ) + (r m1 )m1 m2 (s m2 )(c4 + a2 b2 2abc2 )].
On further simplification, we obtain
(Ll (K(r, s), ) = ( W )m1 1 ( X)rm1 1 ( Y )m2 1 ( Z)sm2 1

[3 2 (X + Y + Z + W ) + (W X + XY + Y Z + ZW + W Y + XZ)
(XY Z + XY W + XW Z + W Y Z) c2 m1 (s m2 )( (X + Y ))
b2 (r m1 )(s m2 )( (Y + W )) a2 m1 m2 ( (X + Z))
c2 (r m1 )m2 ( (W + Z))].
Substituting W ,X,Y and Z and reducing the terms we get

0 (am + (s m )c) (cm + (s m )b) (am + (r m )c) (cm + (r m )b)
2 2 2 2 1 1 1 1
1 m1 1 r m1 1 m2 1 s m2 1
and the roots of
[3 2 {(a + c)(m1 + m2 ) + (b + c)((r + s) (m1 + m2 ))}

+{ac(m21 + m22 ) + (c2 + ab)(m2 (s m2 ) + m1 (r m1 ))
+(ab + bc + ca)(m1 (s m2 ) + m1 (r m1 )) + bc((r m1 )
+(s m2 ))2 + (b2 + c2 )(r m1 )(s m2 )} {(a + c)bc(r m1 )(s m2 )(m1 + m2 )
+abc (m1 (s m2 )(s + m1 m2 ) + m2 (r m1 )(r + m2 m1 ))
+bc2 (r m1 )(s m2 )(r + s m1 m2 )
+ac(a + c)m21 m2 + c(ac + c2 + ab)m1 m2 (s m2 ) + ac(b + c)m1 m2 (r m1 )}] = 0. 2
Theorem 4.4 Let S(m, n) be a double star graph with central vertices labeled zero and the
pendent vertices labeled one. Then the characteristic polynomial of label Laplacian matrix of
S(m, n) is

( c)m+n4 3 [(m + n)c + 2a]2 + [ac(m + n) + c2 mn + 2ac] [ac2 (m + n)] .
Proof Let vm and vm+1 be the central vertices of S(m, n) with zero labels. Remaining
m + n 2 vertices be given label one. Characteristic polynomial of Ll (S(m, n)) is

( c)Im1 cJm11 Om11 Om1n1

cJ1m1 ( (m 1)c a)I1 aI1 O1n1

|I Ll (S(m, n))| =
O1m1 aI1 ( (n 1)c a)I1 cJ1n1

On1m1 Om11 cJn11 ( c)In1
Using elementary transformations, we get

( mc a) a 0

m+n4

|I Ll (S(m, n))| = ( c) ac ( (n 1)c a) (n 1)c

c c ( c)
Hence, the characteristic polynomial of S(m, n) is
(Ll (S(m, n)), ) = ( c)m+n4 3 [(m + n)c + 2a]2 + [ac(m + n)

+c2 mn + 2ac] [ac2 (m + n)] . 2
Corollary 4.6 Let S(m, n) be a double star graph with central vertices labeled one and the
pendent vertices labeled zero. Then the characteristic polynomial of label Laplacian matrix of
S(m, n) is

( c)m+n4 3 [(m + n)c + 2b]2 + [bc(m + n) + c2 mn + 2bc] [bc2 (m + n)] .
Definition 4.1 The crown graph Sn0 for an integer n > 3 is the graph with vertex set
{u1 , u2 , . . . , un , v1 , v2 , . . . , vn } and edge set {ui vi : 1 6 i, j 6 n, i 6= j}.
Lemma 4.2([10]) Let

A0 A1
A=
A1 A0
be a 22 block symmetric matrix. Then the eigenvalues of A are the eigenvalues of the matrices
A0 + A1 and A0 A1 .
Theorem 4.5 The label Laplacian spectrum of crown graph Sn0 of order 2n is

0 ma + (n m)c (n m)b + mc nc (m 2)a + c(n m) +
2 2
1 m1 nm1 1 m1 1 1
where X = 2a(m 1) + 2b(n m 1) + cn, Y = 4ab(m 1)(n m 1) + 2bc(n m 1)(n

m) + 2acm(m 1), = X, = X 2 4Y and m denotes the number of vertices labelled zero
in each vertex set of the crown graph.
Proof Let the labels of n vertices of Sn0 be |000{z

. . . 0} 111
| {z 1} in each partite set. Then
m nm

A B
Ll (Sn0 ) =
B A
where
v1 v2 vm vm+1 vm+2 vn
A = Diag[ W W W Z Z Z ],
W and Z are the label degrees of the m zero label vertices and n m one label vertices
respectively given by W = a(m 1) + c(n m) and Z = b(n m 1) + cm. Note that

a(J I)mm cJmnm
B=
cJnmm a(J I)nmnm
From Lemma 4.2, the label Laplacian spectrum of Ll (Sn0 ) is the union of spectrum of A + B
and A B. Observe that A + B = Ll (Kn ). Hence, by Theorem 4.2, we obtain

Specl (A + B) = (4.3)
1 m1 nm1 1
Also, A B = A + Al (Kn ). Consider det(I (A + Al (Kn ))).
Step 1 Replacing Ri by Ri = Ri Ri1 , for i = 2, 3, , m, m + 2, m + 3, , n, we

obtain
det(I (A+Al (Kn ))) = ((a(m2)+c(nm)))m1 ((b(nm2)+cm)))nm1 det(E).
Step 2 Replacing Ci by Ci = Ci + Ci+1 + . . . + Cm , i = 1, 2, . . . , m 1 and replacing Cj

by Cj = Cj + Cj+1 + . . . + Cn , j = 1, 2, . . . , n 1, the det(E) reduces to a determinant of order
m + 1.
a(m 1) a(m 2) . . . a c(n m)

0 1 0 ... 0 0

0 0 1 ... 0 0

det(E) = . .. .. .. .. ,
.. . . ... . .

0 0 0 ... 1 0

cm c(m 1) c(m 2) . . . c
where = [2a(m 1) + c(n m)] and = [2b(n m 1) + cm].

Step 3 Expanding over the first column
det(E) = 2 [2a(m 1) + 2b(n m 1) + cm] + 4ab(m 1)(n m 1)

+2bc(n m 1)(n m) + 2cam(m 1).
Step 4 Substituting det(E) in Step 1,
det(I (A + Al (Kn ))) = ( (a(m 2) + c(n m)))m1

( (b(n m 2) + cm)))nm1 {2 [2a(m 1) + 2b(n m 1) + cm]
+4ab(m 1)(n m 1) + 2bc(n m 1)(n m) + 2cam(m 1)}.
Hence, label Laplacian spectrum of A B is

a(m 2) + c(n m) m1

b(n m 2) + cm nm1

Specl (A B) = (4.4)
X+ X 2 4Y
1

2
X X 2 4Y
2 1
where
X = 2a(m 1) + 2b(n m 1) + cn,

Y = 4ab(m 1)(n m 1) + 2bc(n m 1)(n m) + 2acm(m 1).
The union of expressions (4.3) and (4.4) is the label Laplacian spectrum of Sn0 . 2
References
[1] C.Adiga, A.Bayad, I.Gutman and Shrikanth A.S., The minimum covering energy of a
graph, Kragujevac J.Sci., 34 (2012), 39-56.
[2] C.Adiga, E.Sampathkumar, M.A.Sriraj and Shrikanth A. S., Color energy of a graph, Proc.
Jangjeon Math. Soc., 16 (2013) 335-351.
[3] C.Adiga and Smitha, On maximum degree energy of a graph, Int. J. Contemp. Math.
Sciences, 8 (2009), 385-396.
[4] R.Balakrishnan, The energy of a graph, Linear Algebra Appl., 387(2004), 287-295.
[5] R.B.Bapat, Graphs and Matrices, Springer - Hindustan Book Agency, London, 2011.
[6] R. Li, Some lower bounds for Laplacian energy of graphs, International Journal of Con-
temporary Mathematical Sciences, 4(2009), 219-223.
[7] D.M.Cvetkovic, M.Doob and H.Sachs, Spectra of Graphs, Theory and Applications, Aca-
demic Press, New York, 1980.
[8] I.Gutman and B.Zhou, Laplacian energy of a graph, Lin. Algebra Appl., 414(2006) 29-37.
[9] I.Gutman, The energy of a graph, Ber. Math. Stat. Sekt. Forschungsz. Graz., 103(1978)
1-22.
[10] G.Indulal, I.Gutman and A.Vijayakumar, On distance energy of graphs, MATCH Com-
mun.Math. Comput.Chem., 60 (2008), 461-472.
[11] Jia-Yu Shao, Li Zhang and Xi-Ying Yuan, On the second Laplacian eigenvalues of trees of
odd order, Lin. Algebra Appl., 419(2006), 475-485.
[12] Maria Robbiano and Raul Jimenez , Applications of a theorem by Ky Fan in the theory of
Laplacian, MATCH Commun.Math. Comput.Chem., 62(2009), 537-552.

[13] B.Mohar, The Laplacian spectrum of graphs, in Graph Theory, Combinatorics and Appli-
cations, Y. Alavi, G. Chartrand, O. E. Ollerman and A. J. Schwenk, Eds., John Wiley and
Sons, New York, USA, (1991), 871-898.
[14] Pradeep G. Bhat and Sabitha DSouza, Energy of binary labeled graph, Trans. Comb.,
Vol.2.No. 3(2013), 53-67.
[15] Tatjana Aleksic, Upper bounds for Laplacian energy of graphs, MATCH Commun.Math.
Comput.Chem.,60(2008), 435-439.
[16] E.R.Van Dam and W.H.Haemers, Which graphs are determined by their spectrum?, Linear
Algebra Appl., 373 (2003), 241-272.
[17] Bo Zhou, Ivan Gutman and Tatzana Aleksic, A note on Laplacian energy of graphs,
MATCH Commun.Math.Comput. Chem., 60 (2008), 441-446.
[18] Bo Zhou, More on energy and Laplacian energy, MATCH Commun.Math.Comput. Chem.,
64 (2010), 75-84.
Some Results on Total Mean Cordial Labeling of Graphs
R.Ponraj and S.Sathish Narayanan

(Department of Mathematics, Sri Paramakalyani College, Alwarkurichi-627412, India)
E-mail: ponrajmaths@gmail.com, sathishrvss@gmail.com
Abstract: A graph G = (V, E) with p vertices and q edges is said to be a Total Mean
Cordial graph if lthere exists m a function f : V (G) {0, 1, 2} such that for each edge xy
f (x)+f (y)
assign the label 2
where x, y V (G), and the total number of 0, 1 and 2 are
balanced. That is |evf (i) evf (j)| 1, i, j {0, 1, 2} where evf (x) denotes the total
number of vertices and edges labeled with x (x = 0, 1, 2). In this paper, we investigate the
total mean cordial labeling behavior of Ln K1 , S(Pn 2K1 ), S(Wn ) and some more graphs.
Key Words: Smarandachely total mean cordial labeling, cycle, path, wheel, union, corona,
ladder.
AMS(2010): 05C78.
1. Introduction
Throughout this paper we considered finite, undirected and simple graphs. The symbols V (G)
and E(G) will denote the vertex set and edge set of a graph G. A graph labeling is an assignment
of integers to the vertices or edges, or both, subject to certain conditions. Labeled graphs serves
as a useful mathematical model for a broad range of application such as coding theory, X-ray
crystallography analysis, communication network addressing systems, astronomy, radar, circuit
design and database management [1]. Ponraj, Ramasamy and Sathish Narayanan [3] introduced
the concept of total mean cordial labeling of graphs and studied about the total mean cordial
labeling behavior of Path, Cycle, Wheel and some more standard graphs. In [4,6], Ponraj and
Sathish Narayanan proved that Knc + 2K2 is total mean cordial if and only if n = 1, 2, 4, 6, 8
and they investigate the total mean cordial labeling behavior of prism, gear, helms. In [5],
Ponraj, Ramasamy and Sathish Narayanan investigate the Total Mean Cordiality of Lotus
inside a circle, bistar, flower graph, K2,n , Olive tree, Pn2 , S(Pn K1 ), S(K1,n ). In this paper
we investigate Ln K1 , S(Pn 2K1 ), S(Wn ) and some more graphs. If x is any real number.
Then the symbol x stands for the largest integer less than or equal to x and x stands for
the smallest integer greater than or equal to x. For basic definitions that are not defined here
are used in the sense of Harary [2].
1 Received October 31, 2014, Accepted June 2, 2015.

Some Results on Total Mean Cordial Labeling of Graphs 123
2. Preliminaries
Definition 2.1 A total mean cordial labeling of a graph

l G = (V, m E) is a function f : V (G)
f (x)+f (y)
{0, 1, 2} such that for each edge xy assign the label 2 where x, y V (G), and the
total number of 0, 1 and 2 are balanced. That is |evf (i) evf (j)| 1, i, j {0, 1, 2} where
evf (x) denotes the total number of vertices and edges labeled with x (x = 0, 1, 2). If there exists
a total mean cordial labeling on a graph G, we will call G is total mean cordial.
Furthermore, let H G be a subgraph of G. If there is a function
f from V (G) {0, 1, 2}
f (u) + f (v)
such that f |H is a total mean cordial labeling but is a constant for all edges
2
in G \ H, such a labeling and G are then respectively called Smarandachely total mean cordial
labeling and Smarandachely total mean cordial labeling graph respect to H.
The following results are frequently used in the subsequent section.
Definition 2.2 The product graph G1 G2 is defined as follows: Consider any two vertices
u = (u1 , u2 ) and v = (v1 , v2 ) in V = V1 V2 . Then u and v are adjacent in G1 G2 whenever
[u1 = v1 and u2 adj v2 ] or [u2 = v2 and u1 adj v1 ]. Note that the graph Ln = Pn P2 is
called the ladder on n steps.
Definition 2.3 Let G1 and G2 be two graphs with vertex sets V1 and V2 and edge sets E1 and
E2 respectively. Then their join G1 + G2 is the graph whose vertex set is V1 V2 and edge set
is E1 E2 {uv : u V1 and v V2 }. Also the graph Wn = Cn + K1 is called the wheel.
Definition 2.4 Let G1 , G2 respectively be (p1 , q1 ), (p2 , q2 ) graphs. The corona of G1 with G2 ,
G1 G2 is the graph obtained by taking one copy of G1 and p1 copies of G2 and joining the ith
vertex of G1 with an edge to every vertex in the ith copy of G2 .
Definition 2.5 The union of two graphs G1 and G2 is the graph G1 G2 with V (G1 G2 ) =
V (G1 ) V (G2 ) and E (G1 G2 ) = E (G1 ) E (G2 ).
Definition 2.6 The subdivision graph S (G) of a graph G is obtained by replacing each edge
uv of G by a path uwv.
Theorem 2.7([7]) Let G be a (p, q) Total Mean Cordial graph and n 6= 3 then G Pn is also
total mean cordial.
Main Results
Theorem 3.1 S(Wn ) is total mean cordial.
Proof Let V (S(Wn )) = {u, ui , xi , yi : 1 i n}, E(S(Wn )) = {ui yi , yi ui+1 : 1 i

n 1} {un yn , yn u1 } {uxi , xi ui : 1 i n}. Clearly |V (S(Wn ))| + |V (S(Wn ))| = 7n + 1.
Case 1. n 0 (mod 12).

124 R.Ponraj and S.Sathish Narayanan
Let n = 12t and t > 0. Define f : V (S(Wn )) {0, 1, 2} by f (u) = 0,
f (xi ) = 0, 1 i 12t
f (ui ) = 0, 1 i 2t
f (u2t+i ) = 2, 1 i 7t
f (u9t+i ) = 1, 1 i 3t
f (yi ) = 1, 1 i 2t 1
f (y2t1+i ) = 2, 1 i 7t
f (y9t1+i ) = 1, 1 i 3t + 1.
Here evf (0) = 28t + 1, evf (1) = evf (2) = 28t.
Case 2. n 1 (mod 12).
Let n = 12t + 1 and t > 0. Define a map f : V (S(Wn )) {0, 1, 2} by f (u) = 0,
f (xi ) = 0, 1 i 12t + 1
f (ui ) = 0, 1 i 2t
f (u2t+i ) = 2, 1 i 7t
f (u9t+i ) = 1, 1 i 3t + 1
f (yi ) = 1, 1 i 2t 1
f (y2t1+i ) = 2, 1 i 7t + 1
f (y9t+i ) = 1, 1 i 3t + 1.
Here evf (0) = evf (1) = 28t + 3, evf (2) = 28t + 2.
Case 3. n 2 (mod 12).
Let n = 12t + 2 and t > 0. Define a function f : V (S(Wn )) {0, 1, 2} by f (u) = 0,
f (xi ) = 0, 1 i 12t + 2
f (ui ) = 0, 1 i 2t
f (u2t+i ) = 2, 1 i 7t + 1
f (u9t+1+i ) = 1, 1 i 3t + 1
f (yi ) = 1, 1 i 2t
f (y2t+i ) = 2, 1 i 7t + 1
f (y9t+1+i ) = 1, 1 i 3t + 1.
In this case evf (0) = evf (1) = evf (2) = 28t + 5.
Case 4. n 3 (mod 12).
Let n = 12t 9 and t > 0. For n = 3, the Figure 1 shows that S(W3 ) is total mean cordial.
1b b0
2b
b0
b b
1 b 2
2 2 b
b
b 0
0
Figure 1
Now assume t 2. Define a map f : V (S(Wn )) {0, 1, 2} by f (u) = 0,
f (xi ) = 0, 1 i 12t 9
f (ui ) = 0, 1 i 2t 2
f (u2t2+i ) = 2, 1 i 7t 5
f (u9t7+i ) = 1, 1 i 3t 2
f (yi ) = 1, 1 i 2t 3
f (y2t3+i ) = 2, 1 i 7t 5
f (y9t8+i ) = 1, 1 i 3t 1.
In this case evf (0) = evf (1) = 28t 21, evf (2) = 28t 20.
Case 5. n 4 (mod 12).
Let n = 12t 8 and t > 0. The following Figure 2 shows that S(W4 ) is total mean cordial.
0b
1b b0
2b
2b
0 b b b b0
2 2
b b
1 2 b
b 0
0
Figure 2
Now assume t 2. Define f : V (S(Wn )) {0, 1, 2} by f (u) = 0,
f (xi ) = 0, 1 i 12t 8
f (ui ) = 0, 1 i 2t 2
f (u2t2+i ) = 2, 1 i 7t 5
f (u9t7+i ) = 1, 1 i 3t 1
f (yi ) = 1, 1 i 2t 3
f (y2t3+i ) = 2, 1 i 7t 4
f (y9t7+i ) = 1, 1 i 3t 1.
In this case evf (0) = 28t 19, evf (1) = evf (2) = 28t 18.
Case 6. n 5 (mod 12).
Let n = 12t 7 and t > 0. Define a function f : V (S(Wn )) {0, 1, 2} by f (u) = 0,
f (xi ) = 0, 1 i 12t 7
f (ui ) = 1, 1 i 4t 3
f (u4t3+i ) = 2, 1 i 7t 4
f (u11t7+i ) = 1, 1 i t
f (yi ) = 0, 1 i 4t 3
f (y4t3+i ) = 2, 1 i 7t 4
f (y11t7+i ) = 1, 1 i t.
In this case evf (0) = evf (1) = evf (2) = 28t 16.
Case 7. n 6 (mod 12).
f (xi ) = 0, 1 i 12t 6
f (ui ) = 0, 1 i 2t 1
f (u2t1+i ) = 2, 1 i 7t 4
f (u9t5+i ) = 1, 1 i 3t 1
f (yi ) = 1, 1 i 2t 2
f (y2t2+i ) = 2, 1 i 7t 3
f (y9t5+i ) = 1, 1 i 3t 1.
Case 8. n 7 (mod 12).
f (xi ) = 0, 1 i 12t 5
f (ui ) = 0, 1 i 2t 1
f (u2t1+i ) = 2, 1 i 7t 3
f (u9t4+i ) = 1, 1 i 3t 1
f (yi ) = 1, 1 i 2t 2
f (y2t2+i ) = 2, 1 i 7t 3
f (y9t5+i ) = 1, 1 i 3t.
Here evf (0) = evf (1) = 28t 11, evf (2) = 28t 12.
Case 9. n 8 (mod 12).

f (xi ) = 0, 1 i 12t 4
f (ui ) = 0, 1 i 2t 1
f (u2t1+i ) = 2, 1 i 7t 2
f (u9t3+i ) = 1, 1 i 3t 1
f (yi ) = 1, 1 i 2t 1
f (y2t1+i ) = 2, 1 i 7t 3
f (y9t4+i ) = 1, 1 i 3t.
Case 10. n 9 (mod 12).
f (xi ) = 0, 1 i 12t 3
f (ui ) = 0, 1 i 2t 1
f (u2t1+i ) = 2, 1 i 7t 2
f (u9t3+i ) = 1, 1 i 3t
f (yi ) = 1, 1 i 2t 2
f (y2t2+i ) = 2, 1 i 7t 1
f (y9t3+i ) = 1, 1 i 3t.
Case 11. n 10 (mod 12).
f (xi ) = 0, 1 i 12t 2
f (ui ) = 0, 1 i 2t 1
f (u2t1+i ) = 2, 1 i 7t 1
f (u9t2+i ) = 1, 1 i 3t
f (yi ) = 1, 1 i 2t 2
f (y2t2+i ) = 2, 1 i 7t 1
f (y9t3+i ) = 1, 1 i 3t + 1.
In this case evf (0) = 28t 5, evf (1) = evf (2) = 28t 4.
Case 12. n 11 (mod 12).

f (xi ) = 0, 1 i 12t 1
f (ui ) = 1, 1 i 4t 1
f (u4t1+i ) = 2, 1 i 7t
f (u11t1+i ) = 1, 1 i t
f (yi ) = 0, 1 i 4t 1
f (y4t1+i ) = 2, 1 i 7t 1
f (y11t2+i ) = 1, 1 i t + 1.
Hence S(Wn ) is total mean cordial. 2
Theorem 3.2 S(Pn 2K1 ) is total mean cordial.

Proof Let V (S(Pn 2K1 )) = {ui , vi , wi , xi , yi : 1 i n} {ui : 1 i n 1} and

E(S(Pn 2K1 )) = {ui ui , ui ui+1 : 1 i n 1} {ui vi , ui wi , vi xi , wi yi : 1 i n}. Clearly
|V (S(Pn 2K1 ))| + |V (S(Wn 2K1 ))| = 12n 3. Now we define a map f : V (S(Pn 2K1 ))
{0, 1, 2} by f (v1 ) = 0, f (w1 ) = 1, f (un ) = 0,

f (ui ) = f (ui ) = 0, 1 i n 1
f (vi ) = f (wi ) = 1, 2 i n
f (xi ) = f (yi ) = 2, 1 i n.
In this case evf (0) = evf (1) = evf (2) = 4n 1.
Hence S(Pn 2K1 ) is total mean cordial. 2
Theorem 3.3 Ln K1 is total mean cordial.
Proof Let V (Ln K1 ) = {ui , vi , xi , yi : 1 i n} and E(Ln K1 ) = {xi ui , ui vi ,

vi yi : 1 i n} {ui ui+1 , vi vi+1 : 1 i n 1}. Here |V (Ln K1 )| + |E(Ln K1 )| = 9n 2.
Define a map f : V (Ln K1 ) {0, 1, 2} by
f (ui ) = 0, 1 i n

f (xi ) = 0, 1 i n2
f (yi ) = 1, 1 i n

f (x n +i ) = 1, 1 i n2
2
f (vi ) = 2, 1 i n.
The following Table 1 shows that f is a total mean cordial labeling of Ln K1 .

Nature of n evf (0) evf (1) evf (2)

9n 2 9n 2 9n 2
n 0 (mod 2)
3 3 3
9n 2 9n 2 9n 2
n 1 (mod 2)
3 3 3
Hence Ln K1 is Total Mean Cordial. 2
Theorem 3.4 The graph P1 P2 . . . Pn is total mean cordial.
Proof We prove this theorem by induction on n. For n = 1, 2, 3 the result is true, see
Figure 3.
0b 0b
b
0 b b 1b 2b
1 2
b b b
0 2 0
Figure 3
Assume the result is true for P1 P2 . . . Pn1 . Then by Theorem 2.7, (P1 P2 . . .
Pn1 ) Pn is total mean cordial. 2
Theorem 3.5 Let Cn be the cycle u1 u2 . . . un u1 . Let GCn be a graph with V (GCn ) = V (Cn )
{vi : 1 i n} and E(GCn ) = E(Cn ) {ui vi , ui+1 vi : 1 i n 1} {un vn , u1 vn }. Then
GCn is total mean cordial.
Proof Clearly, |V (GCn )| + |E(GCn )| = 5n.
Case 1. n 0 (mod 3).
Let n = 3t and t > 0. Define f : V (GCn ) {0, 1, 2} by
f (ui ) = f (vi ) = 0, 1 i t
f (ut+i ) = f (vt+i ) = 2, 1 i t
f (u2t+i ) = f (v2t+i ) = 1, 1 i t 1
f (u3t ) = 1 and f (v3t ) = 0. In this case evf (0) = evf (1) = evf (2) = 5t.
Let n = 3t + 1 and t > 0. Define f : V (GCn ) {0, 1, 2} by
f (ui ) = f (vi ) = 0, 1 i t
f (ut+1+i ) = f (vt+i ) = 2, 1 i t
f (u2t+1+i ) = f (v2t+1+i ) = 1, 1 i t
f (ut+1 ) = 0, f (v2t+1 ) = 2. In this case evf (0) = 5t + 1, evf (1) = evf (2) = 5t + 2.
Let n = 3t + 2 and t > 0. Construct a vertex labeling f : V (GCn ) {0, 1, 2} by
f (ui ) = f (vi ) = 0, 1 i t + 1
f (ut+2+i ) = f (vt+1+i ) = 2, 1 i t
f (u2t+2+i ) = f (v2t+2+i ) = 1, 1 i t
f (ut+1 ) = 1, f (v2t+2 ) = 2. In this case evf (0) = evf (1) = 5t + 3, evf (2) = 5t + 4.
Hence GCn is total mean cordial. 2
Example 3.6 A total mean cordial labeling of GC8 is given in Figure 4.
1b b0
b
0 b
b
1b 1 0 b0
b1 0b
2b b2 1b
b0
2b
b b
2 2
Figure 4
Theorem 3.6 Let St(Ln ) be a graph obtained from a ladder Ln by subdividing each step exactly
once. Then St(Ln ) is total mean cordial.
Proof Let V (St(Ln )) = {ui , vi , wi : 1 i n} and E(St(Ln )) = {ui wi , wi vi : 1 i
n} {ui ui+1 , vi vi+1 : 1 i n 1}. It is clear that |V (St(Ln ))| + |E(St(Ln ))| = 7n 2.
Let n = 6t. Define a map f : V (St(Ln )) {0, 1, 2} as follows.
f (ui ) = 0, 1 i 6t
f (wi ) = 0, 1 i t
f (wt+i ) = 1, 1 i 5t
f (vi ) = 2, 1 i 5t
f (v5t+i ) = 1, 1 i t.
In this case evf (0) = evf (1) = 14t 1, evf (2) = 14t.

Let n = 6t + 1 and t 1. Define a function f : V (St(Ln )) {0, 1, 2} by
f (ui ) = 0, 1 i 6t + 1
f (wi ) = 0, 1 i t
f (wt+i ) = 2, 1 i 5t + 1
f (vi ) = 1, 1 i 4t + 1
f (v4t+1+i ) = 2, 1 i 2t.
Here evf (0) = 14t + 1, evf (1) = evf (2) = 14t + 2.
Let n = 6t + 2 and t 0. The Figure 5 shows that St(L2 ) is total mean cordial.
2b b0
2b b2
0b b0
Figure 5
Consider the case for t 1. Define f : V (St(Ln )) {0, 1, 2} as follows.
f (ui ) = 0, 1 i 6t + 2
f (wi ) = 0, 1 i t
f (wt+i ) = 1, 1 i 5t + 1
f (vi ) = 2, 1 i 5t + 1
f (v5t+1+i ) = 1, 1 i t.
and f (w6t+2 ) = 2, f (v6t+2 ) = 0. Here evf (0) = evf (1) = evf (2) = 14t + 4.
Let n = 6t 3 and t 1. Define a function f : V (St(Ln )) {0, 1, 2} by
f (ui ) = f (wi ) = f (vi ) = 0, 1 i 2t 2

f (u2t1+i ) = f (w2t1+i ) = f (v2t+i ) = 1, 1 i 2t 2
f (u4t2+i ) = f (w4t1+i ) = f (v4t2+i ) = 2, 1 i 2t 2
f (u2t1 ) = f (w2t1 ) = 0, f (u4t2 ) = f (w4t2 ) = f (w4t1 ) = 1 and f (u6t3 ) = f (v6t3 ) =

2. In this case evf (0) = 14t 7, evf (1) = evf (2) = 14t 8.

Let n = 6t 2 and t > 0. Define f : V (St(Ln )) {0, 1, 2} by
f (ui ) = 0, 1 i 6t 2
f (wi ) = 0, 1 i t
f (wt+i ) = 1, 1 i 5t 2
f (vi ) = 2, 1 i 5t 2
f (v5t2+i ) = 1, 1 i t.
Let n = 6t 1 and t > 0. Define a function f : V (St(Ln )) {0, 1, 2} by
f (ui ) = 0, 1 i 6t 1
f (wi ) = 0, 1 i t
f (wt+i ) = 1, 1 i 5t 1
f (vi ) = 2, 1 i 5t 1
f (v5t1+i ) = 1, 1 i t.
Here evf (0) = evf (1) = evf (2) = 14t 3.

Hence St(Ln ) is total mean cordial. 2
References
[1] J.A.Gallian, A Dynamic survey of graph labeling, The Electronic Journal of Combinatorics,
16 (2013) # Ds6.
[2] F.Harary, Graph theory, Narosa Publishing house, New Delhi (2001).
[3] R.Ponraj, A.M.S.Ramasamy and S.Sathish Narayanan, Total mean cordial labeling of
graphs, International Journal of Mathematical combinatorics, 4(2014), 56-68.
[4] R.Ponraj and S.Sathish Narayanan, Total mean cordiality of Knc + 2K2 , Palestine Journal
of Mathematics, 4(2)(2015), 431-438.
[5] R.Ponraj, A.M.S.Ramasamy and S.Sathish Narayanan, Total mean cordial labeling of some
graphs, Utilitas Mathematica, (Accepted for Publication).
[6] R.Ponraj and S.Sathish Narayanan, Total mean cordial labeling of some cycle related
graphs, Journal of Applied Mathematics and Informatics, 33(2015), No. 1- 2, pp. 101 -
110.
[7] R.Ponraj and S.Sathish Narayanan, Further results on total mean cordial labeling of graphs
(communicated).
Number of Regions in Any Simple Connected Graph
Mushtaq Ahmad Shah

(Department of Mathematics, Vivekananda Global University, Jaipur)
M.H.Gulzar
(Department of Mathematics, University of Kashmir, Srinagar)
Mridula Purohit
(Department of Mathematics, VIT East Jaipur)
E-mail: shahmushtaq81@gmail.com
Abstract: A graph G(v, e) is simple if it is without self loops and parallel edges and a
graph G(v, e) is connected if every vertex of graph is connected with each other. This paper
is dealing with the problem of finding the number of regions in any simple connected graph.
In other words this paper generalize the Eulers result on number of regions in planer graphs
to all simple non planar graphs according to Euler number of regions in planar graphs is
given by f = e v + 2. Now we extend Eulers result to all simple graphs. I will prove that
the number of regions in any simple connected graph is equal to
r1 X
X r
f = ev+2+ j Ci , rN
j=1 i=2
The minimum number of regions in any complete graph is

1 hni n 1 n2 n3 n2 3n + 4
f= +
4 2 2 2 2 2
Where [ ] represents greatest integer function, and n is the number of vertices of graph.
Key Words: Planar graph, simple graph, non-planar graph, complete graph, regions of a
graph, crossing number.
AMS(2010): 05C25.
1. Introduction
A planar graph is one that can be drawn on a two-dimensional plane such that no two edges
cross. A cubic graph is one in which all vertices have degree three. A three connected graph
is one that cannot be disconnected by removal of two vertices. A graph is said to be bipartite
1 Received January 4, 2015, Accepted June 10, 2015.

134 Mushtaq Ahmad Shah, M.H.Gulzar and Mridula Purohit
if the vertices can be colored using exactly two colors such that no two adjacent vertices have
the same color. [1] pp 16-20 A plane representation of a graph divides the plane into regions
also called windows, faces or meshes. A window is characterized by the set of edges or the set
of vertices forming its boundary. Note that windows is not defined in a non planer graph or
even in a planer graph not embedded in a plane. Thus a window is a property of the specific
plane representation of a graph. The window of a graph may be finite or infinite. The portion
of a plane lying outside a graph embedded in a plane is infinite in its extend. Since a planar
graph may have different plane representation Euler gives formula for number of windows in a
planar graph.[2] pp 88-100.
Lemma 1.1([1]) A graph can be embedded in a surface of a sphere if and only if it can be
embedded in a plane.
Lemma 1.2([1]) A planer graph may be embedded in a plane such that any specified region can
be made the infinite region.
Lemma 1.3(Euler theorem, [1], [2]) A connected planer graph with n vertices and e edges has
e n + 2 regions.
Lemma 1.4([2]) A plane graph is bipartite if and only if each of its faces has an even number
of sides.
Corollary 1.5([2]) In a simple connected planar graph with f regions n vertices and e edges
(e > 2) the following inequalities must hold.
3
e f and e 3n 6.
2
Theorem 1.6([2]) The spherical embedding of every planar 3-connected graph is unique.
The crossing number (sometimes denoted as c(G) of a graph G is the smallest number of
pair wise crossings of edges among all drawings of G in the plane. In the last decade, there has
been significant progress on a true theory of crossing numbers. There are now many theorems
on the crossing number of a general graph and the structure of crossing critical graphs, whereas
in the past, most results were about the crossing numbers of either individual graphs or the
members of special families of graphs. The study of crossing numbers began during the Second
World War with Paul Turan. In [4], he tells the story of working in a brickyard and wondering
about how to design an efficient rail system from the kilns to the storage yards. For each kiln
and each storage yard, there was a track directly connecting them. The problem he Consider
was how to lay the rails to reduce the number of crossings, where the cars tended to fall off the
tracks, requiring the workers to reload the bricks onto the cars. This is the problem of finding
the crossing number of the complete bipartite graph. It is also natural to try to compute the
crossing number of the complete graph. To date, there are only conjectures for the crossing
numbers of these graphs called Guys conjecture which suggest that crossing number of complete
Number of Regions in Any Simple Connected Graph 135
graph Kn is given by V (Kn ) = Z(n) [5] [6].

1 hni n 1 n 2 n 3
Z(n) = ,
4 2 2 2 2
where [ ] represents greatest integer function which can also be written as

1 2 2
64 n(n 2) (n 4) ... if n is even
Z(n) =
1 2 2
64 (n 1) (n 3) ... if n is odd
Guy prove it for n 10 in 1972 in 2007 Richter prove it for n 12 For any graph G, we
say that the crossing number c(G) is the minimum number of crossings with which it is possible
to draw G in the plane. We note that the edges of G need not be straight line segments, and
also that the result is the same whether G is drawn in the plane or on the surface of a sphere.
Another invariant of G is the rectilinear crossing number, c(G), which is the minimum number
of crossings when G is drawn in the plane in such a way that every edge is a straight line
segment. We will find by an example that this is not the same number obtained by drawing G
on a sphere with the edges as arcs of great circles. In drawing G in the plane, we may locate
its vertices wherever it is most convenient. A plane graph is one which is already drawn in the
plane in such a way that no two of its edges intersect. A planar graph is one which can be
drawn as a plane graph [9]. In terms of the notation introduced above, a graph G is planar
if and only if c(G) = 0. The earliest result concerning the drawing of graphs in the plane is
due to Fary [7] [10], who showed that any planar graph (without loops or multiple edges) can
be drawn in the plane in such a way that every Edge is straight. Thus Farys result may be
rephrased: if c(G) = 0, then c(G) = 0. In a drawing, the vertices of the graph are mapped into
points of a plane, and the arcs into continuous curves of the plane, no three having a point in
common. A minimal drawing does not contain an arc which crosses itself, nor two arcs with
more than one point in common. [8][11]In general for a set of n line segments, there can be up
to O(n2 ) intersection points, since if every segment intersects every other segment, there would
be
n(n 1)
= O(n2 )
2
intersection points. To compute them all we require is O(n2 ) algorithm.
2. Main Result
Before proving the main result we would like to give the detailed purpose of this paper. Euler
gives number of regions in planer graphs which is equal to f = e v + 2. But for non planar
graphs the number of regions is still unknown. It is obvious that every graph has different
representations; there is no particular representation of non planer simple graphs a graph G(v, e)
can be represented in different ways. My aim is to find the number of regions in any simple non
planer graph in whatever way we draw it. I will prove that number of regions of any simple
non planar graph is equal to
r1 X
X r
f =ev+2+ j Ci , rN
j=1 i=2
P
where C2 are the total number of intersection points where two edges have a common point,
P
C3 are the total number of intersection points where three edges have a common point, and
P
so on, Cr are the total number of intersecting points where r edges have a common point.
In whatever way we draw the graph. And the minimum number of regions in a complete
graph is equal to

1 hni n 1 n 2 n 3 n2 3n + 4
f= + , n = number of vertices
4 2 2 2 2 2
This result is depending upon Guys conjecture which is true for all complete graphs n 12
therefore my result is true for all complete graphs n 12 if conjecture is true for all n, then
my result is also true for all complete graphs.
Theorem 2.1 The number of regions in any simple graph is given by

r
X r
X
f = ev+2+ j Ci , rN
j=1 i=2
In particular number of regions in any complete graph is given by

1 hni n 1 n 2 n 3 n2 3n + 4
f= +
4 2 2 2 2 2
This result of complete graphs is true for all graphs n 12 it is true for all n if Guys conjecture
is true for all n.
Proof Let G(v, e) be a graph contains the finite set of vertices v and finite set of edges e.
It is obvious that every graph has a planar representation in a certain stage and in that stage
according to Euler number of regions are f = e v + 2. Let n edges remaining in the graph by
adding a single edge graph becomes non planar that in this stage it has maximum planarity so
if we start to add remaining n edges one by one intersecting points occur and number of regions
start to increase. Out of remaining n edges let us suppose that there are certain intersecting
points where two edges have a common points it is denoted by C2 and total such points can be
P
represented by C2 similarly let us suppose that there are certain intersecting points where
three edges have a common points it is denoted by C3 and total such points can be represented
P
by C3 and this process goes on and finally let us suppose that there are certain intersecting
points where r edges have a common point it is denoted by Cr and total such points can be
P
represented by Cr . It must be kept in mind that graphs cannot be defined uniquely and
finitely every graph has different representations. Since my result is true for all representations
in whatever way you can represent graph. We first show that if we have finite set of n edges in
a plane such that each pair of edges have one common point and no three edges have a common
point, number of regions is increased by one by each pair of edges. Let fn be the number of
regions created by finite set of n edges. It is not obvious that every finite set of n edges creates
the same number of regions, this follows inductively when we establish a recurrence f0 .
Fig.1
We begin with no edges and one region, so f0 = 1. We prove that
fn = fn1 + n
if n 1. Consider finite set of n edges, with n 1 and let L be one of these edges. The other
edges form a finite set of n 1 edges. We argue that adding L increases the number of regions
by n. The intersection of L with the other edges partition L into n portions. Each of these
portions cuts a region into two. Thus adding L increases the number of regions by n .since this
holds for all finite set of edges we have
fn = fn1 + n
if n 1. This determines a unique sequence starting with f0 = 1, and hence every finite set
of edges creates same number of regions. Thus it is clear that if two edges have a common
point number of region is increased by one we represent it by C2 and total number of such
P
intersecting point is denoted by C2 , similarly if three edges have a common point number
of regions is increased by two and we denote it by C3 and total number of such intersection
P P
points is denoted by C3 and number of regions are 2 C3 this process goes on and finally if
r lines have a common point number of regions is increased by r and it is denoted by Cr and
P
total number of such intersection points are denoted by Cr and number of regions increased
P
by (r 1) Cr it must be noted that every graph has different representations any number of
intersection points can occur. Thus we conclude that number of regions in any simple graph is
given by
f = e v + 2 + sum of all intersecting points where two edges have common point
+2(sum of all intersecting points where three edges have common point)
+3(sum of all intersecting points where four edges have common point)
+
+(r 1)(sum of all intersecting points where r edges have common point),
written to be
X X X X
f =ev+2+ C2 + 2 C3 + 3 C4 + ... + (r 1) Cr ,
which can be expressed as
r1 X
X r
f =ev+2+ j Ci , rN
j=1 i=2
It should be noted that Figures 2-4 below illustrate above result.
Figure 2
Figure 2 has 20 vertices and 30 edges, there are 9 intersection points where two edges have
common point, 2 intersection points where three edges have common point,1 intersection points
where four edges have common point, 1 intersection points where five edges have common point,
and number of regions is 32 we now verify it by above formula.
r
X r
X
f = ev+2+ j Ci
j=1 i=2
X X X X
= ev+2+ C2 + 2 C3 + 3 C4 + 4 C5 .
Substitute above values we get that
f = 30 20 + 2 + 9 + 4 + 3 + 4 = 32,
which verifies that above result.
Figure 3 below has 14 vertices 24 edges 15 intersecting points where two edges have common
point. 2 intersection points where three edges have common point,1 intersection points where
four edges have common point, and number of regions is 34 we now verify it by above formula.
r
X r
X
f = ev+2+ j Ci
j=1 i=2
X X X
= ev+2+ C2 + 2 C3 + 3 C4 = 24 14 + 2 + 15 + 4 + 3 = 34,
which again verifies that above result.
Figure 3
Figure 4 below has 6 vertices 11 edges 2 intersecting points where two edges have common
point. 1 intersection points where three edges have common point, and number of regions is 11
we now verify it by above formula.
r
X r
X
f = ev+2+ j Ci
j=1 i=2
X X
= ev+2+ C2 + 2 C3 = 11 6 + 2 + 2 + 2 = 11,
verifies that above result again.

Figure 4
n(n1)
Now if the graph is complete with n vertices then the number of edges in it is 2 and
minimum number of crossing points are given by Guys conjecture that is

1 hni n 1 n 2 n 3
Z(n) =
4 2 2 2 2
which is true for all n 12 thus above result is true for all n 12, if Guys conjecture is true,
then my result is true for all n. We know that every complete graph has a planar representation
in a certain stage. When we start to draw any complete graph we add edge one by one and a
stage comes when graph has maximum planarity in that stage number of regions according to
Euler is f = e v + 2, when we start to add more edges one by one number of crossing numbers
occur but according to definition of crossing numbers two edges have a common point and no
three edges have a common point it has been shown that if two edges have a common point
P
number of regions is increased by C2 , thus the number of regions is given by
X
f =ev+2+ C2 ,
where
X 1 hni n 1 n 2 n 3
C2 =
4 2 2 2 2
is the minimum number of crossing points ( Guys conjecture), e the number of edges and v
number of vertices.
Let us suppose that graph has n vertices and number of edges is n(n1)
2 substitute these
values above we get minimum number of regions in a complete graph is given by
X
f = ev+2+ C2

n(n 1) 1 hni n 1 n 2 n 3
= n+2+
2 4 2 2 2 2

1 hni n 1 n 2 n 3 n2 3n + 4
= + .
4 2 2 2 2 2
That proves the result. 2
Figures 5-6 below illustrates this result. The Figure 5 below is the complete graph of six
vertices and number of regions are as

1 hni n 1 n 2 n 3 n2 3n + 4
f = +
4 2 2 2 2 2
2
1 6 61 62 63 6 3(6) + 4
= +
4 2 2 2 2 2
1 36 18 + 4
= 3221+ = 14
4 2
This shows that the above result is true.
Figure 5
Figure 6 below is the complete graph of 5 vertices and number of regions are as

1 hni n 1 n 2 n 3 n2 3n + 4
f = +
4 2 2 2 2 2
2
1 5 51 52 53 5 3(5) + 4
= +
4 2 2 2 2 2
1 25 15 + 4
= 2211+ =8
4 2
Figure 6
Example 1 Find the number of regions of a complete graph of 8 vertices with minimum
crossings. Find number of regions?
Solution Apply the above result we get

1 hni n 1 n 2 n 3 n2 3n + 4
f = +
4 2 2 2 2 2
2
1 8 81 82 83 8 38+4
= +
4 2 2 2 2 2
1 64 24 + 4
= 4332+ = 40
4 2
Example 2 A graph has 10 vertices and 24 edges, there are three points where two edges
have a common point, and there is one point where three edges have a common point find the
number of regions of a graph?
Solution By applying above formula we get
r
X r
X
f = ev+2+ j Ci
j=1 i=2
X X
= ev+2+ C2 + 2 C3 = 24 10 + 2 + 3 + 2 = 21
Thus number of regions is 21.
Acknowledgement
We are highly thankful to Vivekananda Global University, Jaipur for facilitating us to complete
our work smoothly.
References
[1] Taro Herju, Lecture Notes on Graph Theory, Cambridge University Press, 2011, pages
16-20 and 61-65.
[2] NARSINGH DEO , Graph Theory with Applications, Prentice - Hall of India Private limited
New Delhi, 11000/2005, pp 88-100 and 96-100.
[3] A.Hertel, Hamiltonian Cycle in Sparse Graphs, Masters thesis, University of Toronto, 2004.
[4] P.Turan, A note of welcome, J. Graph Theory, 1(1977), 15.
[5] Erdos and R.Guy, crossing number problem, American Math Month, 80(1973), 52-57.
[6] R.Guy, A combinatorial problem, Bull. Malayan Math. Soc., 7 (1960),68-72.
[7] I.Fary, On straight line representation of planar graphs, Acta Univ. Szeged, 11 (1948),
229-233.
[8] F.Harry and A.Hill, on the number of crossings in a complete graphs, Proc. Edinburgh
Math. Soc., 13 (1962) 333-338.
[9] J.W.Moon, On the distribution of crossing in random complete graphs, SAIM. J. Appl
.Math., 13(1965) 506-510.
[10] L.Saaty, The minimum number of intersection in complete graph, Proceedings of the Na-
tional Academy of Science, USA, 1964 September, Vol 52(3), 688-690.
[11] R.B.Richter and G.Salazak, Crossing number, A complete chapter, www.math.uwaterloo.ca/
brichter/pubs/June2008.pdf.
A Characterization of Directed Paths
Ramya.S
(LG Soft India, Bangalore-560 103, India)
Nagesh.H.M
(Department of Science and Humanities, PES Institute of Technology, Bangalore-560 100, India)
E-mail: ramya.dewdrop@gmail.com, nageshhm@pes.edu
Abstract: In this note, the non-trivial connected digraphs D with vertex set V (D) =
n
X
{v1 , v2 , . . . , vn } satisfying d (vi ) d+ (vi ) = n 2 are characterized, where d (vi ) and
i=1
d+ (vi ) be the in-degree and out-degree of vertices of D, respectively.
Key Words: Directed path, directed cycle, directed tree, tournament.
AMS(2010): 05C20.
1. Introduction
Notations and definitions not introduced here can be found in [1]. For a simple graph G
with vertex set V (G) = {v1 , v2 , , vn }, V.R.Kulli[2] gave the following characterization. A
graph G is a non-empty path if and only if it is connected graph with n 2 vertices and
Xn
d2i 4n + 6 = 0, where di is the degree of vertices of G. In this note, we extend the
i=1
characterization of paths to directed paths, which is needed to characterize the maximal outer
planarity property of some digraph operator(digraph valued function).
We need some concepts and notations on directed graphs. A directed graph(or just digraph)
D consists of a finite non-empty set V (D) of elements called vertices and a finite set A(D) of or-
dered pair of distinct vertices called arcs. Here, V (D) is the vertex set and A(D) is the arc set of
D. A directed path from v1 to vn is a collection of distinct vertices v1 , v2 , v3 , . . . , vn together with
the arcs v1 v2 , v2 v3 , . . . , vn1 vn considered in the following order: v1 , v1 v2 , v2 , v2 v3 , . . . , vn1 vn , vn .
A directed path is said to be non-empty if it has at least one arc. An arborescence is a directed
graph in which, for a vertex u called the root(i.e., a vertex of in-degree zero) and any other
vertex v, there is exactly one directed path from u to v. A directed cycle is obtained from a
nontrivial directed path on adding an arc from the terminal vertex to the initial vertex of the
directed path. A directed tree is a directed graph which would be a tree if the directions on the
arcs are ignored. The out-degree of a vertex v, written d+ (v), is the number of arcs going out
1 Received January 8, 2015, Accepted June 8, 2015.

A Characterization of Directed Paths 145
from v and the in-degree of a vertex v, written d (v), is the number of arcs coming into v.
The total degree of a vertex v, written td(v), is the number of arcs incident with v. We
immediately have td(v) = d (v) + d+ (v). A tournament is a nontrivial complete asymmetric
digraph.
2. Characterization
Theorem 2.1 A connected digraph D with vertex set V (D) = {v1 , v2 , , vn }, n 2 is a

non-empty directed path if and only if
n
X
d (vi ) d+ (vi ) = n 2. (1)
i=1
Proof Let D be a directed path with n vertices v1 , v2 , , vn . Then it is easy to verify

that the sum of product of in-degree and out-degree of its vertices is (n 2).
To prove the sufficiency part, we are given that D is connected with n vertices v1 , v2 , . . . , vn
and equation (1) is satisfied. If n = 2, then the only connected digraph is a tournament with
two vertices(or a directed path with two vertices) and (1) is trivially verified.
Now, suppose that D is connected with n 3 vertices. We consider the following two
cases:
(i) The total degree of every vertex of D is at most two;

(ii) There exists at least one vertex of D whose total degree is at least three.
In the former case, since D is connected, it is either a directed path or a directed tree or
a directed cycle.
Suppose that D is a directed tree with n 3 vertices. Then there exists exactly two vertices
n
X
of total degree one, and (n2) vertices of total degree two. Thus, d (vi )d+ (vi ) = < n2
i=1
violating the condition (1), where is the number of vertices of D whose in-degree and out-
degree are both one. Hence D cannot be a directed tree. On the other hand, if D is a directed
Xn
cycle with n 3 vertices, then d (vi ) d+ (vi ) = n > n 2, again violating the condition
i=1
(1). Hence D cannot be a directed cycle also. In the latter case, we prove as follows.
Case 1. Suppose that a connected digraph D with n 3 vertices has exactly one vertex of
total degree three, and remaining vertices of total degree at most two. We consider the following
two subcases of Case 1.
Subcase 1. If D is a directed tree, then clearly it has three vertices of total degree one,
n
X
and (n 4) vertices of total degree two. Thus, d (vi ) d+ (vi ) < n 2, where is the
i=1
number of vertices of D whose in-degree and out-degree are both at least one.
Subcase 2. If D is cyclic, then it has a vertex of total degree one, and (n 2) vertices of
146 Ramya.S and Nagesh.H.M
n
X
total degree two. Thus, d (vi ) d+ (vi ) = n > n 2.
i=1
Case 2. Finally, consider any connected digraph with n vertices having more than one vertex
of total degree at least three. Clearly, such a digraph can be obtained by adding new arcs
joining pairs of non-adjacent vertices of a digraph described in Case 1. The addition of new
arcs increases the total degree of some vertices and there by the above inequality is preserved
in this case also. Therefore in all cases, we arrive at a contradiction if we assume that D has
some vertices of total degree at least three. Hence we conclude that D is a non-empty directed
path. This completes the proof. 2
Remark 2.1 It is known that a directed path is a special case of an arborescence. Hence
equation (1) is satisfied for an arborescence whose root vertex has out-degree exactly one. For
an example, see Fig.1, Fig.2. It is easy to verify that equation (1) is satisfied for an arborescence
showed in Fig.1, but not in Fig.2.
Root vertex
Root vertex
? R
s
w
? ^ ?
Fig.1 Fig.2
Acknowledgement
We thank Prof.R.Chandrasekhar and Prof.Mayamma Joseph, members of Monthly Informal

Group Discussion(MIGD), ADMA, Bangalore for very useful discussions in motivating and
developing this work.
References
[1] Jorgen Bang-Jensen, Gregory Gutin, Digraphs Theory, Algorithms and applications, Springer-
Verlag London Limited(2009).
[2] V.R.Kulli, A Characterization of Paths, The Mathematical Education, 1975, pp 1-2.
I want to bring out the secrets of nature and apply them for the happiness of
man. I dont know of any better service to offer for the short time we are in the
world.
By Thomas Edison, an American inventor.

Author Information
Submission: Papers only in electronic form are considered for possible publication. Papers
prepared in formats, viz., .tex, .dvi, .pdf, or.ps may be submitted electronically to one member
of the Editorial Board for consideration the Mathematical Combinatorics (International
Book Series. An effort is made to publish a paper duly recommended by a referee within a
period of 3 months. Articles received are immediately put the referees/members of the Editorial
Board for their opinion who generally pass on the same in six weeks time or less. In case of
clear recommendation for publication, the paper is accommodated in an issue to appear next.
Each submitted paper is not returned, hence we advise the authors to keep a copy of their
submitted papers for further processing.
Abstract: Authors are requested to provide an abstract of not more than 250 words, lat-
est Mathematics Subject Classification of the American Mathematical Society, Keywords and
phrases. Statements of Lemmas, Propositions and Theorems should be set in italics and ref-
erences should be arranged in alphabetical order by the surname of the first author in the
following style:
Books
[4]Linfan Mao, Combinatorial Geometry with Applications to Field Theory, InfoQuest Press,
2009.
[12]W.S.Massey, Algebraic topology: an introduction, Springer-Verlag, New York 1977.
Research papers
[6]Linfan Mao, Mathematics on non-mathematics - A combinatorial contribution, International
J.Math. Combin., Vol.3(2014), 1-34.
[9]Kavita Srivastava, On singular H-closed extensions, Proc. Amer. Math. Soc. (to appear).
Figures: Figures should be drawn by TEXCAD in text directly, or as EPS file. In addition,
all figures and tables should be numbered and the appropriate space reserved in the text, with
the insertion point clearly indicated.
Copyright: It is assumed that the submitted manuscript has not been published and will not
be simultaneously submitted or published elsewhere. By submitting a manuscript, the authors
agree that the copyright for their articles is transferred to the publisher, if and when, the
paper is accepted for publication. The publisher cannot take the responsibility of any loss of
manuscript. Therefore, authors are requested to maintain a copy at their end.
Proofs: One set of galley proofs of a paper will be sent to the author submitting the paper,
unless requested otherwise, without the original manuscript, for corrections after the paper is
accepted for publication on the basis of the recommendation of referees. Corrections should be
restricted to typesetting errors. Authors are advised to check their proofs very carefully before
return.
June 2015
Contents
Mathematics After CC Conjecture Combinatorial Notions and Achievements
By Linfan MAO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 01
Timelike-Spacelike Mannheim Pair Curves Spherical Indicators Geodesic
Curvatures and Natural Lifts
By S
uleyman SENYURT and Selma DEMET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Smarandache-R-Module and Mcrita Context
By N.Kannappa and P.Hirudayaraj . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Generalized Vertex Induced Connected Subsets of a Graph
By B.Chaluvaraju, Medha Itagi Huilgol, N.Manjunath and S.Syed Asif Ulla . . . . . . . . . . . . . 61
b-Chromatic Number of Splitting Graph of Wheel
By Samir.K.Vaidya and Minal.S.Shukla . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Eccentric Connectivity and Connective Eccentric Indices of Generalized
Complementary Prisms and Some Duplicating Graphs
By S.Arockiaraj and Vijaya Kumari . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
The Moving Coordinate System and Euler-Savarys Formula for the One
Parameter Motions On Galilean (Isotropic) Plane
By M
ucahit AKBIYIK and Salim YUCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Laplacian Energy of Binary Labeled Graph
By Pradeep G.Bhat, Sabitha DSouza and Swati S.Nayak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Some Results on Total Mean Cordial Labeling of Graphs
By R.Ponraj and S.Sathish Narayanan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Number of Regions in Any Simple Connected Graph
By Mushtaq Ahmad Shah, M.H.Gulzar and Mridula Purohit . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
A Characterization on Directed Paths
By Ramya.S and Nagesh.H.M . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
An International Book Series on Mathematical Combinatorics

MATHEMATICAL COMBINATORICS (INTERNATIONAL BOOK SERIES), Vol. 2/2015

Uploaded by

Copyright:

Available Formats

MATHEMATICAL COMBINATORICS (INTERNATIONAL BOOK SERIES), Vol. 2/2015

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MATHEMATICAL COMBINATORICS (INTERNATIONAL BOOK SERIES), Vol. 2/2015

Uploaded by

Copyright:

Available Formats

ISBN 978-1-59973-349-4

Edited By Linfan MAO

THE MADIS OF CHINESE ACADEMY OF SCIENCES AND

ACADEMY OF MATHEMATICAL COMBINATORICS & APPLICATIONS

Edited By Linfan MAO

The Madis of Chinese Academy of Sciences and

By J.Barrymore, an American actor.

Mathematics After CC Conjecture

Combinatorial Notions and Achievements

Academy of Mathematical Combinatorics with Applications, Colorado, USA

Key Words: CC conjecture, Smarandache system, GL -system, non-solvable system of

uary 29-31, 2015, Shanghai, P.R.China.

Generally, let E be a topological space. A graph G is said to be embeddable into E ([32])

if there is a 1 1 continuous mapping f : G E with f (p) 6= f (q) if p 6= q for p, q G, i.e.,

{a, c} {c, e} {d, e}

Formally, we introduce some conceptions following.

Definition 1.1([30]-[31], [12]) Let (1 ; R1 ), (2 ; R2 ), , (m ; Rm ) be m mathematical

Definition 1.3 ([13]) Let

be two Smarandache multisystems underlying topological graphs G1 and G2 , respectively. They

Consequently, the previous discussion implies that

CC Conjecture([9-10],[14]) Any mathematical science can be reconstructed from or made by

Certainly, this conjecture is true in philosophy. It is in fact a combinatorial notion for

Notion 2.2 For an integer m 1, let (1 ; R1 ), (2 ; R2 ), , (m ; Rm ) all be algebraic systems

For example, let

2.1 K2L -Systems

A K2L -system is such a multi-system consisting of exactly 2 algebraic systems underlying a

A generalization of field is replace R \ {0} by any subset H R in Fig.4. Then a bigroup

Definition 2.3 A bigroup (biring, bifield, bimodule, ) is a 2-system (G ; , ) such that

f be a permutation multigroup action on

for a, b G1 and {1 , 1 } provided a b existing in (G1 ; 1 , 1 ). Define the image Im(, )

where 1 denotes the unit of (G ; ) with G a maximal closed subset of G on operation .

normal, denoted by H G if for g G ,

where g H = {g h|h H provided g h existing} and H g = {h g|h H provided h

Theorem 2.4([11]) Let (, ) : (G1 ; {1 , 1 }) (G2 ; {2 , 2 }) be a homomorphism. Then

Particularly, if (G2 ; {2 , 2 }) is a group (A ; ), we know the corollary following.

Corollary 2.5 Let (, ) : (G ; {, }) (A ; ) be an epimorphism. Then

Similarly, a bigroup (G ; , ) is distributive if

hold for all a, b, c G . Then, we know the following result.

Theorem 2.6([11]) Let (G ; , ) be a distributive bigroup of order 2 with G = A1 A2 such

for g Ge, O, and generalize Theorem 2.3 to the following.

Theorem 2.7([16]) Let (, ) : (Ge1 ; O

Ge1 /Ker(, ) Im(, ).

permutation group Pi acts on i , which is globally k-transitive for an integer k 1 if for

f be a permutation multigroup action on

i each integers 1 i m. Then

Theorem 2.9([16]) A finitely integral multiring is a multifield.

For multimodule, let O = { +i | 1 i m}, O1 = {i |1 i m} and O2 = {+ i |1 i

Mod(M (O) : R(O1 O2 ))

where Mod(R (n) : R(O1 O2 )) is a multimodule on R (n) = {(x1 , x2 , , xn ) | xi R, 1

for a R, integers 1 i m. Particularly, a finitely module over a commutative ring

Classical geometry, such as those of Euclidean or non-Euclidean geometry, or projective geome-

Notion 3.2 For an integer m 1, let P1 , P2 , , Pm all be geometrical spaces in Definition

3.1 Euclidean Spaces

Let 1 = (1, 0, , 0), 2 = (0, 1, 0 , 0), , n = (0, , 0, 1) be the normal basis of a

Problem 3.3 Determine the dimension of a combinatorial Euclidean space consisting of m

An n-manifold is a second countable Hausdorff space of locally Euclidean n-space without

where + 1 is the hnumber

Clearly, if m = 1, then M f(ni , i ) is nothing else but exactly an n1 -manifold by

Theorem 3.7([22]) Any locally compact n-manifold M with an alta A = { (U ; )| } is

deg(Pi )deg(Pj ) eij deg(Pi )deg(Pj )