$${H}_{t}^{f\circ \phi \left(p\right)}:=f\circ \phi \left({B}_{t\wedge {\tau}_{u}}^{p}\right)f\circ \phi \left({B}_{t\wedge {\tau}_{u}}^{0}\right){\int}_{0}^{t\wedge {\tau}_{u}}{1}_{F}\left(s\right)L(f\circ \phi )\left({B}_{s}^{p}\right)ds,$$
is a local martingale, for any
$p\in U\backslash \left(\right(n2)skeleton),$
where
$F={\cup}_{i=1}^{\infty}\left]\right]{\sigma}_{i},{\tau}_{i}\left]\right]$
, with
$$\begin{array}{c}{\sigma}_{i}=inf\{t>{\tau}_{i1};\phi ({B}_{t}^{p})\in {U}_{N}\}\end{array}$$  
$$\begin{array}{c}{\tau}_{i}=inf\{t>{\sigma}_{i};\phi ({B}_{t}^{p})\notin {\overline{V}}_{N}\}\end{array}$$  
$$\begin{array}{c}{\sigma}_{0}=0\end{array}$$  
$$\begin{array}{c}{\tau}_{0}=0\end{array}$$  
$$\begin{array}{}\end{array}$$  
So, we have for any
$p\in U\backslash \left(\right(n2)skeleton),$
$$\frac{\partial}{\partial t}E\left[\right(f\circ \phi \left)\right({B}_{t\wedge {\tau}_{u}}^{p}\left)\right]=\frac{1}{2}\Delta (f\circ \phi )\left(p\right).$$
Since
$(f\circ \phi )\left({B}_{t\wedge {\tau}_{u}}^{p}\right)$
is a local submartingale, then
$$\frac{\partial}{\partial t}E\left[\right(f\circ \phi \left)\right({B}_{t\wedge {\tau}_{u}}^{p}\left)\right]\ge 0,$$
so
$\Delta (f\circ \phi )\left(p\right)\ge 0$
, for any
$p\in U\backslash \left(\right(n2)skeleton).$
Thus, by EellsFuglede (cf. [
11]
Theorem 12.1), we obtain that
$\phi $
is a harmonic map on
$U\backslash \left(\right(n2)skeleton).$
Now, using the same arguments as in the proof of Proposition 4.4 , for harmonic functions, we conclude that
$\phi $
is harmonic on each
$U$
.
$\square $
Theorem 5.2
Let
$(K,g)$
be an admissible Riemannian polyhedra endowed with a picewise smooth Riemannian metrics and
$(N,h)$
a smooth Riemannian manifold without boundary. Let
$\phi :K\to N$
a continuous map such that
$\phi \in {W}_{loc}^{1,2}(K,N)$
, i.e.
$K$
is covered by relatively compact subdomains
$U$
such that every
$\phi $
is of finite energy on each
$U$
. Moreover,
$\phi $
is considered at least of class
${\mathcal{C}}^{2}$
on the interior of each
$n$
simplex and the interior of each
$(n1)$
simplex. Then
$\phi $
is a harmonic morphism if and only if
$\phi $
maps
$K$
valued Brownian motions
$({B}_{t}^{p}{)}_{t\ge 0}$
, for any
$p\in K\backslash \left(\right(n2)skeleton)$
, to a Brownian motion on
$N$
, i.e. If
${(}^{N}{B}_{t}^{p}{)}_{t\ge 0}$
denote the Brownian motion the manifold
$N$
then there exist a continuous increasing process
$({A}_{t}{)}_{t\ge 0}$
such that:
${}^{N}{B}_{t}^{p}\circ {A}_{t}=\phi \circ {B}_{t}^{p}$
.
Remark 5.3
We will suppose the
$dimK\ge dimN$
. Otherwise,
$\phi $
is constant. Indeed, if
$dimK<dimN$
, by smooth theory (see [
1]
, p.46),
$\phi $
is constant on each interior of maximal simplex (of a choosen fine triangulation of
$K$
). On the other hand,
$\phi $
is continuous and
$K$
is
$(n1)$
chainable, so
$\phi $
is constant on
$K$
.
Proof of Theorem ( 5.2 ):
For the proof of the Theorem (
5.2 ), we will addapt and complete the proof given by Darling (see [
8]
) in the smooth case.
$\u201d\Rightarrow \u201d$
Suppose
$\phi :K\to N$
is a harmonic morphism. Then by EellsFuglede's result (cf [
11]
Theorem 13.2), there exists a function
$\lambda \in {L}_{loc}^{1}\left(K\right)$
called the dilation, such that:
$$\begin{array}{c}{\int}_{K}\langle \nabla \psi ,\nabla (v\circ \phi )\rangle ={\int}_{K}\psi \lambda \left[\right({\Delta}_{N}v)\circ \phi ],\end{array}$$ 
(10)

for every
$v\in {\mathcal{C}}^{2}\left(N\right)$
and
$\psi \in Li{p}_{c}\left(K\right)$
.
Let
$({B}_{t}^{p}{)}_{t\ge 0}$
a
$K$
valued Brownian motion, for any
$p\in K\backslash \left(\right(n2)skeleton)$
.
Take
$U$
as in the hypothesis of the theorem and suppose
$p\in U$
.
Define a continuous increasing process
$({A}_{t}{)}_{t\ge 0}$
by:
$$\begin{array}{c}{A}_{t}:={\int}_{0}^{t}\lambda \left({B}_{s}\right)ds,\end{array}$$ 
(11)

and it's inverse as:
${C}_{t}=inf\{u;{A}_{u}>u\}.$
Denote
$\phi \circ B$
by
$X=({X}_{t}{)}_{t\ge 0}$
on
$N$
. For any function
$f:N\to \mathbf{R}$
of class
${\mathcal{C}}^{2}$
and any
$p\in K\backslash \left(\right(n2)skeleton)$
we have:
$${\int}_{0}^{t}{\Delta}_{N}f({X}^{\phi \left(p\right)}\circ {C}_{s})ds={\int}_{0}^{{C}_{t}}{\Delta}_{N}f\left({X}_{u}^{\phi \left(p\right)}\right)d{A}_{u}.$$
By ( 11 ) we obtain:
$$\begin{array}{c}{\int}_{0}^{t}{\Delta}_{N}f({X}^{\phi \left(p\right)}\circ {C}_{s})ds={\int}_{0}^{{C}_{t}}\lambda \left({B}_{u}^{p}\right){\Delta}_{N}f\left({X}_{u}^{\phi \left(p\right)}\right)du.\end{array}$$ 
(12)

Using ( 10 ) , the right hand side of the equality ( 12 ) is equal to:
$2{\int}_{0}^{{C}_{t}}L(f\circ \phi )\left({B}_{u}^{p}\right)du,{\mu}_{g}\text{a.e.}$
, for every
$p\in K\backslash \left(\right(n2)skeleton)$
.
On the other hand, by Theorem
4.3 , the process:
$${\stackrel{~}{H}}_{t}^{f\circ \phi \left(p\right)}:={\chi}_{\{{B}_{t}^{p}/\in ((n2)skeleton\left)\right\}}(f\circ \phi )\left({B}_{t}^{p}\right)(f\circ \phi )\left(p\right){\int}_{0}^{t}{\chi}_{\{{B}_{s}^{p}/\in ((n2)skeleton\left)\right\}}L(f\circ \phi )\left({B}_{s}^{p}\right)ds$$
is a continuous local martingale for any
$p\in K\backslash \left(\right(n2)skeleton)$
.
Consider now the process:
$${\stackrel{~}{H}}^{f\circ \phi \left(p\right)}\circ {C}_{t}\stackrel{denote}{=}{\stackrel{~}{R}}_{t}^{f\circ \phi \left(p\right)}.$$
${\stackrel{~}{R}}_{t}^{f\circ \phi \left(p\right)}$
is obviously a continuous local martingale but it is also almost surely equal (using 10 and 12 ) to the process:
$${R}_{t}^{f\circ \phi \left(p\right)}:=f({X}^{\phi \left(p\right)}\circ {C}_{t})f\left(\phi \right(p\left)\right)\frac{1}{2}{\int}_{0}^{t}{\Delta}_{N}f({X}^{\phi \left(p\right)}\circ {C}_{s})ds.$$
So
${R}_{t}^{f\circ \phi \left(p\right)}$
is also a continuous local martingale for any
$p\in K\backslash \left(\right(n2)skeleton)$
, which means, by definition, that
${X}^{\phi \left(p\right)}\circ {C}_{s}$
is a Brownian motion on
$N$
.
$\u201d\Leftarrow \u201d$
Conversely, suppose that for any
$p\in K\backslash \left(\right(n2)skeleton)$
,
$\left(\phi \right({B}_{t}^{p}){)}_{t\ge 0}$
is a Brownian motion on
$N$
up to a change of time.
Let
$V$
be an open set of
$N$
such that
${\phi}^{1}\left(V\right)\subset U$
, where
$U$
is taken as in the hypothesis of the theorem. Let
$f:V\to \mathbf{R}$
be a local harmonic function on
$N$
.
Fix
${p}_{0}\in U\backslash \left(\right(n2)skeleton)$
with
$\phi \left({p}_{0}\right)\in V$
and
$\tau $
denote the first exit time of
$({B}_{t}^{{p}_{0}}{)}_{t\ge}$
from
${\phi}^{1}\left(V\right)$
.
By hypothesis, the process
$\left(f\right(\phi \left({B}_{t\wedge \tau}^{{p}_{0}}\right)){)}_{t\ge 0}=(f\circ \phi \left({B}_{t\wedge \tau}^{{p}_{0}}\right){)}_{t\ge 0}$
is equal to
$f{(}^{N}{B}^{\phi \left(p\right)}\circ {A}_{t\wedge \tau})$
.
But this last process is continuous local martingale (because, by definition the process
$\left(f{(}^{N}{B}_{t\wedge \tau}^{\phi \left(p\right)}\right){)}_{t\ge 0}$
is a continuous local martingale and martingale property is stable under change of time).
So we have shown that for every
${p}_{0}\in U\backslash \left(\right(n2)skeleton)$
and for every harmonic function on
$N$
,
$(f\circ \phi ({B}_{t\wedge \tau}^{{p}_{0}}){)}_{t\ge 0}$
is a continuous local martingale but by Corollary 4.4 this means that
$f\circ \phi $
is harmonic.
In other words,we have shown that
$\phi $
pullsback harmonic function on
$N$
to harmonic function on
$K\backslash \left(\right(n2)skeleton)$
. But we have already proved in the proof of the Theorem ( 4.3 ) that harmonic function on
$K\backslash \left(\right(n2)skeleton)$
are harmonic on
$K$
.
We conclude that
$\phi $
is a harmonic morphism.
$\square $
References

P. Baird, J.C. Wood: Harmonic Morphisms Between Riemannian Manifolds, Oxford University Press (2003).

W. Ballmann, M. Brin: Orbihedra of Nonpositive Curvature, Publications IHES, 82, 169209 (1995).

J.R. Baxter, R.V. Chacon: The equivalence of diffusions on networks to Brownian motion, Contemporary Mathematics, Vol.26, 3348 (1984).

T. Bouziane: Brownian motion in riemannian admissible complex to appear in Illinois J.Math. (2005) or arXiv:math.PR/0403080

M.R. Bridson: Geodesics and Curvature in Metric Simplicial Complexes, World Scientific, Eds. E. Ghys, A.Haefliger, A. Verjovsky (1990).

M.R. Bridson, A. Haefliger: Metric spaces of Nonpositive curvature, SpringerVerlag (1999).

M. Brin, Y. Kifer: Brownian motion, harmonic functions and hyperbolicity for Euclidean complexes, Math.Z. 237, 421468 (2001).

R.W.R. Darling: Martingales in Manifoldsdefinitions, examples, and behaviour under maps, Séminaire de Probabilités XVI, 1980/81 Supplément: Géométrie Différentielle Stochastique, Lecture Notes in Mathematics, no. 921.

M. Davis, T. Januszkiewicz: Hyperbolization of polyhedra, Journal of Differential Geometry, 34, no. 2, 347388 (1991).

E. B. Dynkin: Markov Process, vol.1, SpringerVerlag (1965).

J. Eells, B. Fuglede: Harmonic maps between Riemannian Polyhedra, Cambridge University Press (2001).

M. Emery: Stochastic Calculus in Manifolds, SpringerVerlag (1989).

B. Fuglede: Harmonic morphisms between Riemannian manifolds, Ann. Inst. Fourier 28, 107144 (1978).

E.P. Hsu: Stochastic Analysis on Manifold, Graduate Studies in Mathematics, vol.38, AMS.

A. Gray: Tubes, Progress in Mathematics Vol.221, Birkhäuser Verlag (2004).

R. E. Greene, H. Wu: Embedding of open Riemannian manifolds by harmonic functions, Ann. Inst. Fourier 25, 1, 215235 (1975).

T. Ishihara: A mapping of Riemannian manifolds which preserves harmonic functions, J. Math. Kyoto Univ.(JMKYAZ) 19, 215229 (1979).

N. J. Korevaar, R. M. Schoen: Sobolev spaces and harmonic maps for metric space targets, Comm. Anal. Geom. 1, no. 34, 561659 (1993).

P. A. Meyer: A differential geometric formalism for the Ito calculus, Springer Lecture Notes in Mathematics, Vol.851, 256270 (1981)

B. O'Neill :SemiRiemannian Geometry with applications to relativity, Academic Press (1983).

L. Schwartz :Semimartingales sur des variétés, et martingales conformes, Lect. Notes in Math., vol. 780, Springer (1980).

E. H. Spanier: Algebraic Topology, McGrawHill, New York (1966).

J. Tits: Buildings of spherical type and finite BNpairs, Lect. Notes in Math., vol. 386, SpringerVerlag (1974).

M.E. VázquezAbal: Harmonicity and Brownian motion, Proceedings Workshop: Recent Topics in Differential Geometry, Puerto de la Cruz (Tenerife, Spain), Secret.Public.Univ.de la Laguna, Serie Informes no.32, 123132 (1991).