## Isomorphic groupoid ${C}^{*}$-algebras associated with different Haar systems

### Mădălina Roxana BuneciThis work was partly supported by the MEC-CNCSIS grant At127/2004 and by the Postdoctoral Training Program HPRN-CT-2002-0277. University Constantin Brâncuşi, Târgu-Jiu

Abstract
We shall consider a locally compact groupoid endowed with a Haar system $\nu$  and having proper orbit space. We shall associated to each appropriated cross section $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  for ${d}_{F}:{G}^{F}\to {G}^{\left(0\right)}$  (where $F$  is a Borel subset of ${G}^{\left(0\right)}$  meeting each orbit exactly once) a ${C}^{*}$  -algebra ${M}_{\sigma }^{*}\left(G,\nu \right)$  . We shall prove that the ${C}^{*}$  -algebras ${M}_{\sigma }^{*}\left(G,{\nu }_{i}\right)$  $i=1,2$  associated with different Haar systems are $*$  -isomorphic AMS 2000 Subject Classification: 22A22, 43A22, 43A65, 46L99.
Key Words: locally compact groupoid, ${C}^{*}$  -algebra, $*$  -isomorphism.

1 Introduction

The ${C}^{*}$  -algebra of a locally compact groupoid was introduced by J. Renault in [8. The construction extends the case of a group: the space of continuous functions with compact support on groupoid is made into a $*$  -algebra and endowed with the smallest ${C}^{*}$  -norm making its representations continuous. In order to define the convolution on groupoid one needs to assume the existence of a Haar system which is an analogue of Haar measure on a group. Unlike the case for groups, Haar systems need not to be unique. A result of Paul Muhly, Jean Renault and Dana Williams establishes that the ${C}^{*}$  -algebras of $G$  associated with two Haar systems are strongly Morita equivalent (Theorem 2.8/p. 10 [3). If the groupoid is transitive they have proved that if $G$  is transitive then the ${C}^{*}$  -algebra of $G$  is isomorphic to ${C}^{*}\left(H\right)\otimes \mathcal{K}\left({L}^{2}\left(\mu \right)\right)$  , where $H$  is the isotropy group ${G}_{u}^{u}$  at any unit $u\in {G}^{\left(0\right)}$  , $\mu$  is an essentially unique measure on ${G}^{\left(0\right)}$  , ${C}^{*}\left(H\right)$  denotes the group ${C}^{*}$  -algebra of $H$  , and $\mathcal{K}\left({L}^{2}\left(\mu \right)\right)$  denotes the compact operators on ${L}^{2}\left(\mu \right)$  (Theorem 3.1/p. 16 [3). Therefore the ${C}^{*}$  -algebras of a transitive groupoid $G$  associated with two Haar systems are $*$  -isomorphic.
In [7Arlan Ramsay and Martin E. Walter have associated to a locally compact groupoid $G$  a ${C}^{*}$  algebra denoted ${M}^{*}\left(G,\nu \right)$  . They have considered the universal representation $\omega$  of ${C}^{*}\left(G,\nu \right)$  -the usual ${C}^{*}$  -algebra associated to a Haar system $\nu =\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  (constructed as in [8) . Since every cyclic representation of ${C}^{*}\left(G,\nu \right)$  is the integrated form of a representation of $G$  , it follows that $\omega$  can be also regarded as a representation of ${\mathcal{ℬ}}_{c}\left(G\right)$  , the space of compactly supported Borel bounded function on $G$  . Arlan Ramsay and Martin E. Walter have used the notation ${M}^{*}\left(G,\nu \right)$  for the operator norm closure of $\omega \left({\mathcal{ℬ}}_{c}\left(G\right)\right)$  . Since $\omega$  is an $*$  -isomorphism on ${C}^{*}\left(G,\nu \right)$  , we can regarded ${C}^{*}\left(G,\nu \right)$  as a subalgebra of ${M}^{*}\left(G,\nu \right)$  .
We shall assume that the orbit space of the groupoid $G$  is proper and we shall choose a Borel subset $F$  of ${G}^{\left(0\right)}$  meeting each orbit exactly once and such that $F\cap \left[K\right]$  has a compact closure for each compact subset $K$  of ${G}^{\left(0\right)}$  . For each appropriated cross section $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  for ${d}_{F}:{G}^{F}\to {G}^{\left(0\right)}$  , ${d}_{F}\left(x\right)=d\left(x\right)$  , we shall construct a ${C}^{*}$  -algebra ${M}_{\sigma }^{*}\left(G,\nu \right)$  which can be viewed as a subalgebra of ${M}^{*}\left(G,\nu \right)$  . If ${\nu }_{1}=\left\{{\nu }_{1}^{u},u\in {G}^{\left(0\right)}\right\}$  and ${\nu }_{2}=\left\{{\nu }_{2}^{u},u\in {G}^{\left(0\right)}\right\}$  are two Haar systems on $G$  , we shall prove that the ${C}^{*}$  -algebras ${M}_{\sigma }^{*}\left(G,{\nu }_{1}\right)$  and ${M}_{\sigma }^{*}\left(G,{\nu }_{2}\right)$  are $*$  -isomorphic.
For a transitive (or more generally, a locally transitive) groupoid $G$  we shall prove that the ${C}^{*}$  -algebras ${C}^{*}\left(G,\nu \right)$  , ${M}^{*}\left(G,\nu \right)$  and ${M}_{\sigma }^{*}\left(G,\nu \right)$  coincide.
For a principal proper groupoid $G$  , we shall prove that ${C}^{*}\left(G,\nu \right)\subset {M}_{\sigma }^{*}\left(G,\nu \right)\subset {M}^{*}\left(G,\nu \right)\text{.}$  Let $\pi :{G}^{\left(0\right)}\to {G}^{\left(0\right)}/G$  be the quotient map and let ${\nu }_{i}=\left\{{\varepsilon }_{u}×{\mu }_{i}^{\pi \left(u\right)},u\in {G}^{\left(0\right)}\right\}$  , $i=1,2$  be two Haar systems on the principal proper groupoid $G$  . We shall also prove that if the Hilbert bundles determined by the systems of measures ${\left\{{\mu }_{i}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  have continuous bases in the sense of Definition  24 , then $*$  -isomorphism between ${M}_{\sigma }^{*}\left(G,{\nu }_{1}\right)$  and ${M}_{\sigma }^{*}\left(G,{\nu }_{2}\right)$  can be restricted to a $*$  -isomorphism between ${C}^{*}\left(G,{\nu }_{1}\right)$  and ${C}^{*}\left(G,{\nu }_{2}\right)$  .
For establishing notation, we include some definitions that can be found in several places (e.g. [8, [4). A groupoid is a set $G$  endowed with a product map $\left(x,y\right)\to xy\left[:{G}^{\left(2\right)}\to G\right]$  where ${G}^{\left(2\right)}$  is a subset of $G×G$  called the set of composable pairs, and an inverse map $x\to {x}^{-1}\left[:G\to G\right]$  such that the following conditions hold:
$\left(1\right)$  If $\left(x,y\right)\in {G}^{\left(2\right)}$  and $\left(y,z\right)\in {G}^{\left(2\right)}$  , then $\left(xy,z\right)\in {G}^{\left(2\right)}$  , $\left(x,yz\right)\in {G}^{\left(2\right)}$  and $\left(xy\right)z=x\left(yz\right)$  .
$\left(2\right)$  ${\left({x}^{-1}\right)}^{-1}=x$  for all $x\in G$  .
$\left(3\right)$  For all $x\in G$  , $\left(x,{x}^{-1}\right)\in {G}^{\left(2\right)}$  , and if $\left(z,x\right)\in {G}^{\left(2\right)}$  , then $\left(zx\right){x}^{-1}=z$  .
$\left(4\right)$  For all $x\in G$  , $\left({x}^{-1},x\right)\in {G}^{\left(2\right)}$  , and if $\left(x,y\right)\in {G}^{\left(2\right)}$  , then ${x}^{-1}\left(xy\right)=y$  .
The maps $r$  and $d$  on $G$  , defined by the formulae $r\left(x\right)=x{x}^{-1}$  and $d\left(x\right)={x}^{-1}x$  , are called the range and the source maps. It follows easily from the definition that they have a common image called the unit space of $G$  , which is denoted ${G}^{\left(0\right)}$  . Its elements are units in the sense that $xd\left(x\right)=r\left(x\right)x=x$  .
Units will usually be denoted by letters as $u,v,w$  while arbitrary elements will be denoted by $x,y,z$  . It is useful to note that a pair $\left(x,y\right)$  lies in ${G}^{\left(2\right)}$  precisely when $d\left(x\right)=r\left(y\right)$  , and that the cancellation laws hold (e.g. $xy=xz$  iff $y=z$  ) $.$  The fibres of the range and the source maps are denoted ${G}^{u}={r}^{-1}\left(\left\{u\right\}\right)$  and ${G}_{v}={d}^{-1}\left(\left\{v\right\}\right)$  , respectively. More generally, given the subsets $A$  , $B$  $\subset {G}^{\left(0\right)}$  , we define ${G}^{A}={r}^{-1}\left(A\right)$  , ${G}_{B}={d}^{-1}\left(B\right)$  and ${G}_{B}^{A}={r}^{-1}\left(A\right)\cap {d}^{-1}\left(B\right)$  . The reduction of $G$  to $A\subset {G}^{\left(0\right)}$  is $G|A={G}_{A}^{A}$  . The relation $u~v$  iff ${G}_{v}^{u}\ne \phi$  is an equivalence relation on ${G}^{\left(0\right)}$  . Its equivalence classes are called orbits and the orbit of a unit $u$  is denoted $\left[u\right]$  . A groupoid is called transitive iff it has a single orbit. The quotient space for this equivalence relation is called the orbit space of $G$  and denoted ${G}^{\left(0\right)}/G$  . We denote by $\pi :{G}^{\left(0\right)}\to {G}^{\left(0\right)}/G,\pi \left(u\right)=\stackrel{˙}{u}$  the quotient map. A subset of ${G}^{\left(0\right)}$  is said saturated if it contains the orbits of its elements. For any subset $A$  of ${G}^{\left(0\right)}$  , we denote by $\left[A\right]$  the union of the orbits $\left[u\right]$  for all $u\in A$  .
A topological groupoid consists of a groupoid $G$  and a topology compatible with the groupoid structure. This means that:
$\left(1\right)x\to {x}^{-1}\left[:G\to G\right]$  is continuous.
$\left(2\right)$  $\left(x,y\right)\left[:{G}^{\left(2\right)}\to G\right]$  is continuous where ${G}^{\left(2\right)}$  has the induced topology from $G×G$  .
We are exclusively concerned with topological groupoids which are second countable, locally compact Hausdorff. It was shown in [6that measured groupoids may be assume to have locally compact topologies, with no loss in generality.
If $X$  is a locally compact space, ${C}_{c}\left(X\right)$  denotes the space of complex-valuated continuous functions with compact support. The Borel sets of a topological space are taken to be the $\sigma$  -algebra generated by the open sets. The space of compactly supported bounded Borel function on $X$  is denoted by ${\mathcal{ℬ}}_{c}\left(X\right)$  . For a locally compact groupoid $G$  , we denote by ${G}^{\prime }=\left(x\in G:r\left(x\right)=d\left(x\right)\right)$  the isotropy group bundle of $G$  . It is closed in $G$  .
Let $G$  be a locally compact second countable groupoid equipped with a Haar system, i.e. a family of positive Radon measures on $G$  , $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  , such that 1) For all $u\in {G}^{\left(0\right)}$  , $supp\left({\nu }^{u}\right)={G}^{u}$  .
2) For all $f\in {C}_{c}\left(G\right)$  , $u\to \int f\left(x\right)d{\nu }^{u}\left(x\right)\left[:{G}^{\left(0\right)}\to \mathbf{C}\right]$  is continuous.
3) For all $f\in {C}_{c}\left(G\right)$  and all $x\in G$  , $\int f\left(y\right)d{\nu }^{r\left(x\right)}\left(y\right)=\int f\left(xy\right)d{\nu }^{d\left(x\right)}\left(y\right)$  As a consequence of the existence of continuous Haar systems, $r,d:G\to {G}^{\left(0\right)}$  are open maps ([10). Therefore, in this paper we shall always assume that $r:G\to {G}^{\left(0\right)}$  is an open map If $\mu$  is a measure on ${G}^{\left(0\right)}$  , then the measure $\nu =\int {\nu }^{u}d\mu \left(u\right)$  , defined by $\int f\left(y\right)d\nu \left(y\right)=\int \left(\int f\left(y\right)d{\nu }^{u}\left(y\right)\right)d\mu \left(u\right)\text{,}f\ge 0\text{Borel}$  is called the measure on $G$  induced by $\mu$  . The image of $\nu$  by the inverse map $x\to {x}^{-1}$  is denoted ${\nu }^{-1}$  . $\mu$  is said quasi-invariant if its induced measure $\nu$  is equivalent to its inverse ${\nu }^{-1}$  A measure belongings to the class of a quasi-invariant measure is also quasi-invariant. We say that the class is invariant.
If $\mu$  is a quasi-invariant measure on ${G}^{\left(0\right)}$  and $\nu$  is the measure induced on $G$  , then the Radon-Nikodym derivative $\Delta =\frac{d\nu }{d{\nu }^{-1}}$  is called the modular function of $\mu$  .
In order to define the ${C}^{*}$  -algebra of a groupoid the space of continuous functions with compact support on groupoid is made into a $*$  -algebra and endowed with the smallest ${C}^{*}$  -norm making its representations continuous. For $f$  , $g\in {C}_{c}\left(G\right)$  the convolution is defined by: $f*g\left(x\right)=\int f\left(xy\right)g\left({y}^{-1}\right)d{\nu }^{d\left(x\right)}\left(y\right)$  and the involution by ${f}^{*}\left(x\right)=\overline{f\left({x}^{-1}\right)}\text{.}$  Under these operations, ${C}_{c}\left(G\right)$  becomes a topological $*$  -algebra.
A representation of ${C}_{c}\left(G\right)$  is a $*$  -homomorphism from ${C}_{c}\left(G\right)$  into $\mathcal{ℬ}\left(H\right)$  , for some Hilbert space $H$  , that is continuous with respect to the inductive limit topology on ${C}_{c}\left(G\right)$  and the weak operator topology on $\mathcal{ℬ}\left(H\right)$  . The full ${C}^{*}$  -algebra ${C}^{*}\left(G\right)$  is defined as the completion of the involutive algebra ${C}_{c}\left(G\right)$  with respect to the full ${C}^{*}$  -norm $\parallel f\parallel =sup\parallel L\left(f\right)\parallel$  where $L$  runs over all non-degenerate representation of ${C}_{c}\left(G\right)$  which are continuous for the inductive limit topology.
Every representation $\left(\mu ,{G}^{\left(0\right)}*\mathcal{ℋ},L\right)$  (see Definition 3.20/p.68 [4) of $G$  can be integrated into a representation, still denoted by $L$  , of ${C}_{c}\left(G\right)$  . The relation between the two representation is: $〈L\left(f\right){\xi }_{1},{\xi }_{2}〉=\int f\left(x\right)〈L\left(x\right){\xi }_{1}\left(d\left(x\right)\right),{\xi }_{2}\left(r\left(x\right)\right)〉{\Delta }^{-\frac{1}{2}}\left(x\right)d{\nu }^{u}\left(x\right)d\mu \left(u\right)$  where $f\in$  ${C}_{c}\left(G\right)$  , ${\xi }_{1},{\xi }_{2}\in {\int }_{{G}^{\left(0\right)}}^{\oplus }\mathcal{ℋ}\left(u\right)d\mu \left(u\right)$  .
Conversely, every non-degenerate $*$  -representation of ${C}_{c}\left(G\right)$  is obtained in this fashion (see [8or [4).

2 The decomposition of a Haar system over the principal groupoid

First we present some results on the structure of the Haar systems, as developed by J. Renault in Section 1 of [9and also by A. Ramsay and M.E. Walter in Section 2 of [7.
In Section 1 of [9Jean Renault constructs a Borel Haar system for ${G}^{\prime }$  . One way to do this is to choose a function ${F}_{0}$  continuous with conditionally support which is nonnegative and equal to $1$  at each $u\in {G}^{\left(0\right)}.$  Then for each $u\in {G}^{\left(0\right)}$  choose a left Haar measure ${\beta }_{u}^{u}$  on ${G}_{u}^{u}$  so the integral of ${F}_{0}$  with respect to ${\beta }_{u}^{u}$  is $1.$  Renault defines ${\beta }_{v}^{u}=x{\beta }_{v}^{v}$  if $x\in {G}_{v}^{u}$  (where $x{\beta }_{v}^{v}\left(f\right)=\int f\left(xy\right)d{\beta }_{v}^{v}\left(y\right)$  as usual). If $z$  is another element in ${G}_{v}^{u}$  , then ${x}^{-1}z\in {G}_{v}^{v}$  , and since ${\beta }_{v}^{v}$  is a left Haar measure on ${G}_{v}^{v}$  , it follows that ${\beta }_{v}^{u}$  is independent of the choice of $x$  . If $K$  is a compact subset of $G$  , then ${sup}_{u,v}{\beta }_{v}^{u}\left(K\right)<\infty$  . Renault also defines a 1-cocycle $\delta$  on $G$  such that for every $u\in {G}^{\left(0\right)}$  , $\delta {|}_{{G}_{u}^{u}}$  is the modular function for ${\beta }_{u}^{u}$  . $\delta$  and ${\delta }^{-1}=1/\delta$  are bounded on compact sets in $G$  .
Let $R=\left(r,d\right)\left(G\right)=\left\{\left(r\left(x\right),d\left(x\right)\right),x\in G\right\}$  be the graph of the equivalence relation induced on ${G}^{\left(0\right)}$  . This $R$  is the image of $G$  under the homomorphism $\left(r,d\right)$  , so it is a $\sigma$  -compact groupoid. With this apparatus in place, Renault describes a decomposition of the Haar system $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  for $G$  over the equivalence relation $R$  (the principal groupoid associated to $G$  ). He proves that there is a unique Borel Haar system $\alpha$  for $R$  with the property that ${\nu }^{u}=\int {\beta }_{t}^{s}d{\alpha }^{u}\left(s,t\right)\text{for all}u\in {G}^{\left(0\right)}\text{.}$  In Section 2 [7A. Ramsay and M.E. Walter prove that ${sup}_{u}{\alpha }^{u}\left(\left(r,d\right)\left(K\right)\right)<\infty \text{, for all compact}K\subset G$  For each $u\in {G}^{\left(0\right)}$  the measure ${\alpha }^{u}$  is concentrated on $\left\{u\right\}×\left[u\right]$  . Therefore there is a measure ${\mu }^{u}$  concentrated on $\left[u\right]$  such that ${\alpha }^{u}={\varepsilon }_{u}×{\mu }^{u}$  , where ${\varepsilon }_{u}$  is the unit point mass at $u$  . Since $\left\{{\alpha }^{u},u\in {G}^{\left(0\right)}\right\}$  is a Haar system, we have ${\mu }^{u}={\mu }^{v}$  for all $\left(u,v\right)\in R$  , and the function $u\to \int f\left(s\right){\mu }^{u}\left(s\right)$  is Borel for all $f\ge 0$  Borel on ${G}^{\left(0\right)}$  . For each $u$  the measure ${\mu }^{u}$  is quasi-invariant (Section 2 [7). Therefore ${\mu }^{u}$  is equivalent to ${d}_{*}\left({v}^{u}\right)$  (Lemma 4.5/p. 277 [5).
If $\eta$  is a quasi-invariant measure for $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  , then $\eta$  is a quasi-invariant measure for $\left\{{\alpha }^{u},u\in {G}^{\left(0\right)}\right\}$  . Also if ${\Delta }_{R}$  is the modular function associated to $\left\{{\alpha }^{u},u\in {G}^{\left(0\right)}\right\}$  and $\eta$  , then $\Delta =\delta {\Delta }_{R}\circ \left(r,d\right)$  can serve as the modular function associated to $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  and $\eta$  .
Since ${\mu }^{u}={\mu }^{v}$  for all $\left(u,v\right)\in R$  , the system of measures ${\left\{{\mu }^{u}\right\}}_{u}$  may be indexed on the elements of the orbit space ${G}^{\left(0\right)}/G$  .
Definition 1 We shall call the pair of the system of measures $\left({\left\{{\beta }_{v}^{u}\right\}}_{\left(u,v\right)\in R},{\left\{{\mu }^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}\in {G}^{\left(0\right)}/G}\right)$  (described above) the decomposition of the Haar system $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  over the principal groupoid associated to $G$  . Also we shall call $\delta$  the $1$  -cocycle associated to the decomposition.
Remark 2 Let us note that the system of measures $\left\{{\beta }_{v}^{u}\right\}$  and the $1$  -cocycle in the preceding definition do not depend on the Haar system.
Lemma 3 Let $G$  be a locally compact second countable groupoid with the bundle map $r{|}_{{G}^{\prime }}$  of ${G}^{\prime }$  open. Let $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  be a Haar system on $G$  and $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }^{\stackrel{˙}{u}}\right\}\right)$  its decomposition over the principal groupoid associated to $G$  .
Then for each $f\in {C}_{c}\left(G\right)$  the function $x\to \int f\left(y\right)d{\beta }_{d\left(x\right)}^{r\left(x\right)}\left(y\right)$  is continuous on $G$  .
Proof. By Lemma 1.3/p. 6 [9, for each $f\in {C}_{c}\left(G\right)$  the function $u\to \int f\left(y\right)d{\beta }_{u}^{u}\left(y\right)$  is continuous $.$  Let $x\in G$  and ${\left({x}_{i}\right)}_{i}$  be a sequence in $G$  converging to $x$  . Let $f\in {C}_{c}\left(G\right)$  and let $g$  be a continuous extension on $G$  of $y\to f\left(xy\right)\left[:{G}^{d\left(x\right)}\to \mathbf{C}\right]$  . Let $K$  be the compact set $\left({\left\{x,{x}_{i},i=1,2,..\right\}}^{-1}supp\left(f\right)\cup supp\left(g\right)\right)\cap {r}^{-1}\left(\left\{d\left(x\right),d\left({x}_{i}\right),i=1,2,...\right\}\right).$  We have $|\int f\left(y\right)d{\beta }_{d\left(x\right)}^{r\left(x\right)}\left(y\right)-\int f\left(y\right)d{\beta }_{d\left({x}_{i}\right)}^{r\left({x}_{i}\right)}\left(y\right)|$  $=|\int f\left(xy\right)d{\beta }_{d\left(x\right)}^{d\left(x\right)}\left(y\right)-\int f\left({x}_{i}y\right)d{\beta }_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}\left(y\right)|$  $=|\int g\left(y\right)d{\beta }_{d\left(x\right)}^{d\left(x\right)}\left(y\right)-\int f\left({x}_{i}y\right)d{\beta }_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}\left(y\right)|$  $\le |\int g\left(y\right)d{\beta }_{d\left(x\right)}^{d\left(x\right)}\left(y\right)-\int g\left(y\right)d{\beta }_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}\left(y\right)|+$  $+|\int g\left(y\right)d{\beta }_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}\left(y\right)-\int f\left({x}_{i}y\right)d{\beta }_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}\left(y\right)|$  $\le |\int g\left(y\right)d{\beta }_{d\left(x\right)}^{d\left(x\right)}\left(y\right)-\int g\left(y\right)d{\beta }_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}\left(y\right)|+$  $+{sup}_{y\in {G}_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}}|g\left(y\right)-f\left({x}_{i}y\right)|{\beta }_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}\left(K\right)$  A compactness argument shows that ${sup}_{y\in {G}_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}}|g\left(y\right)-f\left({x}_{i}y\right)|$  converges to $0$  . Also $|\int g\left(y\right)d{\beta }_{d\left(x\right)}^{d\left(x\right)}\left(y\right)-\int g\left(y\right)d{\beta }_{d\left({x}_{i}\right)}^{d\left({x}_{i}\right)}\left(y\right)|$  converges to $0$  because the function $u\to \int f\left(y\right)d{\beta }_{u}^{u}\left(y\right)$  is continuous $.$  Hence $|\int f\left(y\right)d{\beta }_{d\left(x\right)}^{r\left(x\right)}\left(y\right)-\int f\left(y\right)d{\beta }_{d\left({x}_{i}\right)}^{r\left({x}_{i}\right)}\left(y\right)|$  converges to $0.$
Definition 4 A locally compact groupoid $G$  is proper if the map $\left(r,d\right):G\to {G}^{\left(0\right)}×{G}^{\left(0\right)}$  is proper (i.e. the inverse map of each compact subset of ${G}^{\left(0\right)}×{G}^{\left(0\right)}$  is compact).(Definition 2.1.9/p. 37 [1).
Throughout this paper we shall assume that $G$  is a second countable locally compact groupoid for which the orbit space is Hausdorff and the map $\left(r,d\right):G\to R,\left(r,d\right)\left(x\right)=\left(r\left(x\right),d\left(x\right)\right)$  is open, where $R$  is endowed with the product topology induced from ${G}^{\left(0\right)}×{G}^{\left(0\right)}$  . Therefore $R$  will be a locally compact groupoid. The fact that $R$  is a closed subset of ${G}^{\left(0\right)}×{G}^{\left(0\right)}$  and that it is endowed with the product topology is equivalent to the fact $R$  is a proper groupoid.
Throughout this paper by a groupoid with proper orbit space we shall mean a groupoid $G$  for which the orbit space is Hausdorff and the map $\left(r,d\right):G\to R,\left(r,d\right)\left(x\right)=\left(r\left(x\right),d\left(x\right)\right)$  is open, where $R$  is endowed with the product topology induced from ${G}^{\left(0\right)}×{G}^{\left(0\right)}$  .
Proposition 5 Let $G$  be a second countable locally compact groupoid with proper orbit space. Let $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  be a Haar system on $G$  and $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }^{\stackrel{˙}{u}}\right\}\right)$  its decomposition over the principal groupoid associated to $G$  . Then .for each $g\in {C}_{c}\left({G}^{\left(0\right)}\right)$  , the map $u\to \int g\left(v\right)d{\mu }^{\pi \left(u\right)}\left(v\right)$  is continuous.
Proof. Let $g\in {C}_{c}\left({G}^{\left(0\right)}\right)$  and ${u}_{0}\in {G}^{\left(0\right)}$  . Let ${K}_{1}$  be a compact neighborhood of ${u}_{0}$  and ${K}_{2}$  be the support of $g$  . Since $G$  is locally compact and $\left(r,d\right)$  is open from $G$  to $\left(r,d\right)\left(G\right)$  , there is a compact subset $K$  of $G$  such that $\left(r,d\right)\left(K\right)$  contains $\left({K}_{1}×{K}_{2}\right)\cap \left(r,d\right)\left(G\right)$  . Let ${F}_{1}\in {C}_{c}\left(G\right)$  be a nonnegative function equal to $1$  on a compact neighborhood $U$  of $K$  . Let ${F}_{2}$  $\in {C}_{c}\left(G\right)$  be a function which extends to $G$  the function $x\to {F}_{1}\left(x\right)/\int {F}_{1}\left(y\right)d{\beta }_{d\left(x\right)}^{r\left(x\right)}\left(y\right)$  , $x\in U$  . We have $\int {F}_{2}\left(y\right)d{\beta }_{v}^{u}\left(y\right)=1$  for all $\left(u,v\right)\in \left(r,d\right)\left(K\right)$  . Since for all $u\in {K}_{1}$  ,
 $\begin{array}{ccc}\int g\left(v\right)d{\mu }^{\pi \left(u\right)}\left(v\right)& =& \int g\left(v\right)\int {F}_{2}\left(y\right)d{\beta }_{v}^{u}\left(y\right)d{\mu }^{\pi \left(u\right)}\left(v\right)\end{array}$
 $\begin{array}{ccc}& =& \int g\left(d\left(y\right)\right){F}_{2}\left(y\right)d{\nu }^{u}\left(y\right)\text{,}\end{array}$
it follows that $u\to \int g\left(v\right)d{\mu }^{\pi \left(u\right)}\left(v\right)$  is continuous at ${u}_{0}.$
Remark 6 Let $G$  be a locally compact second countable groupoid with proper orbit space. Let $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  be a Haar system on $G$  and $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }^{\stackrel{˙}{u}}\right\}\right)$  be its decomposition over the associated principal groupoid. If $\mu$  is a quasi-invariant probability measure for the Haar system, then ${\mu }_{1}=\int {\mu }^{\pi \left(u\right)}d\mu \left(u\right)$  is a Radon measure which is equivalent to $\mu .$  Indeed, let $f\ge 0$  Borel on ${G}^{\left(0\right)}$  such that $\mu \left(f\right)=0$  . Since $\mu$  is quasi-invariant, it follows that for $\mu a.a.u$  ${\nu }^{u}\left(f\circ d\right)=0$  , and since ${\mu }^{\pi \left(u\right)}$  is equivalent to ${d}_{*}\left({v}^{u}\right)$  , it results ${\mu }^{\pi \left(u\right)}\left(f\right)=0$  for $\mu$  $a.a.u$  .
Conversely, if ${\mu }_{1}\left(f\right)=0$  , then ${\mu }^{\pi \left(u\right)}\left(f\right)=0$  for $\mu$  $a.a.u$  , and therefore ${\nu }^{u}\left(f\circ d\right)=0$  . Thus the quasi-invariance of $\mu$  implies $\mu \left(f\right)=0$  . Thus each Radon quasi-invariant measure is equivalent to a Radon measure of the form $\int {\mu }^{\stackrel{˙}{u}}d\stackrel{~}{\mu }\left(\stackrel{˙}{u}\right)$  , where $\stackrel{~}{\mu }$  is a probability measure on the orbit space $G/{G}^{\left(0\right)}.$

3 A ${C}^{*}$  -algebra associated to a locally compact groupoid with proper orbit space

Let $G$  be a locally compact second countable groupoid with proper orbit space. Let $\pi :{G}^{\left(0\right)}\to {G}^{\left(0\right)}/G$  be the quotient map. Since the quotient space is proper, ${G}^{\left(0\right)}/G$  is Hausdorff. Let us assume that the range map $r$  is open. As a consequence, the map $\pi$  is open. Applying Lemma 1.1 [2to the locally compact second countable spaces ${G}^{\left(0\right)}$  and ${G}^{\left(0\right)}/G$  and to the continuous open surjection $\pi :{G}^{\left(0\right)}\to {G}^{\left(0\right)}/G$  , it follows that there is a Borel set $F$  in ${G}^{\left(0\right)}$  such that:
• 1. $F$  contains exactly one element in each orbit $\left[u\right]={\pi }^{-1}\left(\pi \left(u\right)\right)$  .
• 2. For each compact subset $K$  of ${G}^{\left(0\right)}$  , $F\cap \left[K\right]=F\cap {\pi }^{-1}\left(\pi \left(K\right)\right)$  has a compact closure.
For each unit $u$  let us define $e\left(u\right)=F\cap \left[u\right]$  ( $e\left(u\right)$  is the unique element in the orbit of $u$  contained in $F$  ). For each Borel subset $B$  of ${G}^{\left(0\right)}$  , $\pi$  is continuous and one-to-one on $B\cap F$  and hence $\pi \left(B\cap F\right)$  is Borel in ${G}^{\left(0\right)}/G$  . Therefore the map $e:{G}^{\left(0\right)}\to {G}^{\left(0\right)}$  is Borel (for each Borel subset $B$  of ${G}^{\left(0\right)}$  , ${e}^{-1}\left(B\right)=\left[B\cap F\right]={\pi }^{-1}\left(\pi \left(B\cap F\right)\right)$  is Borel in ${G}^{\left(0\right)}$  ). Also for each compact subset $K$  of ${G}^{\left(0\right)}$  , $e\left(K\right)$  has a compact closure because $e\left(K\right)\subset F\cap \left[K\right]$  .
Since the orbit space ${G}^{\left(0\right)}/G$  is proper the map $\left(r,d\right):G\to R,\left(r,d\right)\left(x\right)=\left(r\left(x\right),d\left(x\right)\right)$  is open and $R$  is closed in ${G}^{\left(0\right)}×{G}^{\left(0\right)}$  . Applying Lemma 1.1 [2to the locally compact second countable spaces $G$  and $R$  and to the continuous open surjection $\left(r,d\right):G\to R$  , it follows that there is a regular cross section ${\sigma }_{0}:R\to G$  . This means that ${\sigma }_{0}$  is Borel, $\left(r,d\right)\left({\sigma }_{0}\left(u,v\right)\right)=\left(u,v\right)$  for all $\left(u,v\right)\in R$  , and ${\sigma }_{0}\left(K\right)$  is relatively compact in $G$  for each compact subset $K$  of $R$  .
Let us define $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  by $\sigma \left(u\right)={\sigma }_{0}\left(e\left(u\right),u\right)$  for all $u$  . It is easy to note that $\sigma$  is a cross section for $d:{G}^{F}\to {G}^{\left(0\right)}$  and $\sigma \left(K\right)$  is relatively compact in $G$  for all compact $K\subset {G}^{\left(0\right)}$  .
Replacing $\sigma$  by $v\to \sigma {\left(e\left(v\right)\right)}^{-1}\sigma \left(v\right)$  we may assume that $\sigma \left(e\left(v\right)\right)=e\left(v\right)$  for all $v$  . Let us define $q:G\to {G}_{F}^{F}$  by $q\left(x\right)=\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1},x\in G\text{.}$  Let $\nu =\left\{{\nu }^{u}:u\in {G}^{\left(0\right)}\right\}$  be a Haar system on $G$  and let $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }^{\stackrel{˙}{u}}\right\}\right)$  be its decompositions over the principal groupoid . Let $\delta$  be the $1$  -cocycle associated to the decomposition.
Let us denote by ${\mathcal{ℬ}}_{\sigma }\left(G\right)$  the linear span of the functions of the form $x\to {g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)$  where ${g}_{1},{g}_{2}$  are compactly supported Borel bounded function on ${G}^{\left(0\right)}$  and $g$  is a Borel bounded function on ${G}_{F}^{F}$  such that if $S$  is the support of $g$  , then the closure of $\sigma {\left({K}_{1}\right)}^{-1}S\sigma \left({K}_{2}\right)$  is compact in $G$  for all compact subsets ${K}_{1}$  , ${K}_{2}$  of ${G}^{\left(0\right)}$  . ${\mathcal{ℬ}}_{\sigma }\left(G\right)$  is a subspace of ${\mathcal{ℬ}}_{c}\left(G\right)$  , the space of compactly supported Borel bounded function on $G$  .
If ${f}_{1}$  , ${f}_{2}\in {\mathcal{ℬ}}_{\sigma }\left(G\right)$  are defined by
 $\begin{array}{ccc}{f}_{1}\left(x\right)& =& {g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)\end{array}$
 $\begin{array}{ccc}{f}_{2}\left(x\right)& =& {h}_{1}\left(r\left(x\right)\right)h\left(q\left(x\right)\right){h}_{2}\left(d\left(x\right)\right)\end{array}$
then ${f}_{1}*{f}_{2}\left(x\right)=g*h\left(q\left(x\right)\right){g}_{1}\left(r\left(x\right)\right){h}_{2}\left(d\left(x\right)\right){〈{g}_{2},\overline{{h}_{1}}〉}_{\pi \left(r\left(x\right)\right)}$  ${f}_{1}^{*}\left(x\right)=\overline{{g}_{2}\left(r\left(x\right)\right)}\overline{g\left(q{\left(x\right)}^{-1}\right)}\overline{{g}_{1}\left(d\left(x\right)\right)}$  Thus ${\mathcal{ℬ}}_{\sigma }\left(G\right)$  is closed under convolution and involution.
Let $\omega$  be the universal representation of ${C}^{*}\left(G,\nu \right)$  the usual ${C}^{*}$  -algebra associated to a Haar system $\nu =\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  (constructed as in [8) . Since every cyclic representation of ${C}^{*}\left(G,\nu \right)$  is the integrated form of a representation of $G$  , it follows that $\omega$  can be also regarded as a representation of ${\mathcal{ℬ}}_{c}\left(G\right)$  , the space of compactly supported Borel bounded function on $G$  . Arlan Ramsay and Martin E. Walter have used the notation ${M}^{*}\left(G,\nu \right)$  for the operator norm closure of $\omega \left({\mathcal{ℬ}}_{c}\left(G\right)\right)$  . Since $\omega$  is an $*$  -isomorphism on ${C}^{*}\left(G,\nu \right)$  , we can regarded ${C}^{*}\left(G,\nu \right)$  as a subalgebra of ${M}^{*}\left(G,\nu \right)$  .
Definition 7 We denote by ${M}_{\sigma }^{*}\left(G,\nu \right)$  the operator norm closure of $\omega \left({\mathcal{ℬ}}_{\sigma }\left(G\right)\right)$  .
Lemma 8 Let ${\left\{{\mu }_{1}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  and ${\left\{{\mu }_{2}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  be two systems of measures on ${G}^{\left(0\right)}$  satisfying:
• 1. $supp\left({\mu }_{i}^{\stackrel{˙}{u}}\right)=\left[u\right]$  for all $\stackrel{˙}{u}$  , $i=1,2$
• 2. For all compactly supported Borel bounded function $f$  on ${G}^{\left(0\right)}$  the function $u\to \int f\left(v\right){\mu }_{i}^{\pi \left(u\right)}\left(v\right)$  is bounded and Borel.
Then there is a family ${\left\{{U}_{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  of unitary operators with the following properties:
• 1. ${U}_{\stackrel{˙}{u}}:{L}^{2}\left({\mu }_{1}^{\stackrel{˙}{u}}\right)\to {L}^{2}\left({\mu }_{2}^{\stackrel{˙}{u}}\right)$  is a unitary operator for each $\stackrel{˙}{u}\in {G}^{\left(0\right)}/G$  .
• 2. For all Borel bounded function $f$  on ${G}^{\left(0\right)}$  , $u\to {U}_{\pi \left(u\right)}\left(f\right)$  is a bounded Borel function with compact support.
• 3. For all Borel bounded function $f$  on ${G}^{\left(0\right)}$  , ${U}_{\pi \left(u\right)}\left(\overline{f}\right)=\overline{{U}_{\pi \left(u\right)}\left(f\right)}$
Proof. Using the same argument as in [6(p. 323) we can construct a sequence ${f}_{1},{f}_{2},...$  of real valued Borel bounded function on ${G}^{\left(0\right)}$  such that $dim\left({L}^{2}\left({\mu }_{1}^{\stackrel{˙}{u}}\right)\right)=\infty$  if and only if ${\parallel {f}_{n}\parallel }_{2}$  for $n=1,2,...$  and then $\left\{{f}_{1},{f}_{2},...\right\}$  gives an ortonormate basis of ${L}^{2}\left({\mu }_{1}^{\stackrel{˙}{u}}\right)$  , while $dim\left({L}^{2}\left({\mu }_{1}^{\stackrel{˙}{u}}\right)\right)=k<\infty$  if and only if ${\parallel {f}_{n}\parallel }_{2}=1$  for $n\le k$  , and ${\parallel {f}_{n}\parallel }_{2}=0$  for $n>k$  and then $\left\{{f}_{1},{f}_{2},...{f}_{k}\right\}$  gives an ortonormate basis of ${L}^{2}\left({\mu }_{1}^{\stackrel{˙}{u}}\right)$  . Let ${g}_{1},{g}_{2},...$  be a sequence with the same properties as ${f}_{1}.{f}_{2},...$  corresponding to ${\left\{{\mu }_{2}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  . Let us define ${U}_{\stackrel{˙}{u}}:{L}^{2}\left({\mu }_{1}^{\stackrel{˙}{u}}\right)\to {L}^{2}\left({\mu }_{2}^{\stackrel{˙}{u}}\right)$  by ${U}_{\stackrel{˙}{u}}\left({f}_{n}\right)={g}_{n}\text{for all}n$  Then the family ${\left\{{U}_{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  has the required properties.
Theorem 9 Let $G$  be a locally compact second countable groupoid with proper orbit space. Let $\left\{{\nu }_{i}^{u},u\in {G}^{\left(0\right)}\right\}$  , $i=1,2$  be two Haar systems on $G$  . Let $F$  be a Borel subset of ${G}^{\left(0\right)}$  containing only one element $e\left(u\right)$  in each orbit $\left[u\right]$  . Let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a cross section for $d:{G}^{F}\to {G}^{\left(0\right)}$  with $\sigma \left(e\left(v\right)\right)=e\left(v\right)$  for all $v\in {G}^{\left(0\right)}$  and $\sigma \left(K\right)$  relatively compact in $G$  for all compact sets $K\subset {G}^{\left(0\right)}$  .
Then the ${C}^{*}$  -algebras ${M}_{\sigma }^{*}\left(G,{\nu }_{1}\right)$  and ${M}_{\sigma }^{*}\left(G,{\nu }_{2}\right)$  are $*$  -isomorphic.
Proof. Let $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }_{i}^{\stackrel{˙}{u}}\right\}\right)$  be the decompositions of the Haar systems over the principal groupoid . Let $\delta$  be the $1$  -cocycle associated to the decompositions, $i=1,2$  .
We shall denote by ${〈\cdot ,\cdot 〉}_{i,\stackrel{˙}{u}}$  the inner product of $\left({L}^{2}\left({G}^{\left(0\right)},\delta \left(\sigma \left(\cdot \right)\right){\mu }_{i}^{\stackrel{˙}{u}}\right)\right)$  , $i=1,2$  .
Let us define $q:G\to {G}_{F}^{F}$  by $q\left(x\right)=\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1},x\in G\text{.}$  We shall define a $*$  -homomorphism $\Phi$  from ${\mathcal{ℬ}}_{\sigma }\left(G\right)$  to ${\mathcal{ℬ}}_{\sigma }\left(G\right)$  . It suffices to define $\Phi$  on the set of function on $G$  of the form $x\to {g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)$  Let ${\left\{{U}_{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  be the family of unitary operators with the properties stated in Lemma  8 , associated to the systems of measures ${\left\{\delta \left(\sigma \left(\cdot \right)\right){\mu }_{i}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  , $i=1,2$  .
Let us define $\Phi$  by $\Phi \left(f\right)=\left(x\to {U}_{\pi \left(r\left(x\right)\right)}\left({g}_{1}\right)\left(r\left(x\right)\right)g\left(q\left(x\right)\right){U}_{\pi \left(d\left(x\right)\right)}\left({g}_{2}\right)\left(d\left(x\right)\right)\right)$  where $f$  is defined by $f\left(x\right)={g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)$  If ${f}_{1}$  and ${f}_{2}$  are defined by
 $\begin{array}{ccc}{f}_{1}\left(x\right)& =& {g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)\end{array}$
 $\begin{array}{ccc}{f}_{2}\left(x\right)& =& {h}_{1}\left(r\left(x\right)\right)h\left(q\left(x\right)\right){h}_{2}\left(d\left(x\right)\right)\end{array}$
then ${f}_{1}*{f}_{2}\left(x\right)=g*h\left(q\left(x\right)\right){g}_{1}\left(r\left(x\right)\right){h}_{2}\left(d\left(x\right)\right){〈{g}_{2},\overline{{h}_{1}}〉}_{1,\pi \left(r\left(x\right)\right)}$  and consequently
 $\begin{array}{ccc}& & \Phi \left({f}_{1}*{f}_{2}\right)\end{array}$
 $\begin{array}{ccc}& =& g*h\left(q\left(x\right)\right){U}_{\pi \left(r\left(x\right)\right)}\left({g}_{1}\right)\left(r\left(x\right)\right){U}_{\pi \left(r\left(x\right)\right)}\left({h}_{2}\right)\left(d\left(x\right)\right){〈{g}_{2},\overline{{h}_{1}}〉}_{1,\pi \left(r\left(x\right)\right)}\end{array}$
 $\begin{array}{ccc}& =& \Phi \left({f}_{1}\right)*\Phi \left({f}_{2}\right)\text{.}\end{array}$
Let $\stackrel{~}{\eta }$  be a probability measure on ${G}^{\left(0\right)}/G$  and ${\eta }_{i}=\int {\mu }_{i}^{\stackrel{˙}{u}}d\stackrel{~}{\eta }\left(\stackrel{˙}{u}\right),i=1,2$  . Let ${L}_{1}$  be the integrated form of a representation $\left(L,\mathcal{ℋ}*{G}^{\left(0\right)},{\eta }_{1}\right)$  and ${L}_{2}$  be the integrated form of $\left(L,\mathcal{ℋ}*{G}^{\left(0\right)},{\eta }_{2}\right)$  . Let $B$  be the Borel function defined by: $B\left(u\right)=L\left(\sigma \left(u\right)\right)$  and $W:{\int }_{{G}^{\left(0\right)}}^{\oplus }\mathcal{ℋ}\left(u\right)d{\eta }_{1}\left(u\right)\to {\int }_{{G}^{\left(0\right)}}^{\oplus }$  $\mathcal{ℋ}\left(e\left(u\right)\right)d{\eta }_{1}\left(u\right)$  be defined by $W\left(\zeta \right)=\left(u\to B\left(u\right)\left(\zeta \left(u\right)\right)\right)$  Since every element of ${L}^{2}\left({G}^{\left(0\right)},\delta \left(\sigma \left(\cdot \right)\right){\mu }_{1}^{\stackrel{˙}{w}},\mathcal{ℋ}\left(e\left(w\right)\right)\right)$  is a limit of linear combinations of elements $u\to a\left(u\right)\xi$  with $a\in {L}^{2}\left({G}^{\left(0\right)},\delta \left(\sigma \left(\cdot \right)\right){\mu }_{1}^{\stackrel{˙}{w}}\right)$  and $\xi \in \mathcal{ℋ}\left(e\left(w\right)\right)$  , we can define a unitary operator ${V}_{\stackrel{˙}{w}}:{L}^{2}\left({G}^{\left(0\right)},\delta \left(\sigma \left(\cdot \right)\right){\mu }_{1}^{\stackrel{˙}{w}},\mathcal{ℋ}\left(e\left(w\right)\right)\right)\to {L}^{2}\left({G}^{\left(0\right)},\delta \left(\sigma \left(\cdot \right)\right){\mu }_{2}^{\stackrel{˙}{w}},\mathcal{ℋ}\left(e\left(w\right)\right)\right)$  by ${V}_{\stackrel{˙}{w}}\left(u\to a\left(u\right)\xi \right)={U}_{\stackrel{˙}{w}}\left(a\right)\xi$  Let $V:{\int }_{{G}^{\left(0\right)}}^{\oplus }\mathcal{ℋ}\left(e\left(u\right)\right)d{\eta }_{1}\left(u\right)\to {\int }_{{G}^{\left(0\right)}}^{\oplus }$  $\mathcal{ℋ}\left(e\left(u\right)\right)d{\eta }_{2}\left(u\right)$  be defined by $V\left(\zeta \right)=\left(u\to {V}_{\stackrel{˙}{u}}\left(\zeta \left(u\right)\right)\right)$  If ${\zeta }_{1},{\zeta }_{2}\in {\int }_{{G}^{\left(0\right)}}^{\oplus }$  $\mathcal{ℋ}\left(e\left(u\right)\right)d{\eta }_{1}\left(u\right)$  and $f$  is of the form $f\left(x\right)={g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)\text{,}$  we have
 $\begin{array}{ccc}& & 〈W{L}_{1}\left(f\right){W}^{*}{\zeta }_{1},{\zeta }_{2}〉\end{array}$
 $\begin{array}{ccc}& =& \int \int g\left(x\right)\delta {\left(x\right)}^{\frac{-1}{2}}〈L\left(x\right){A}_{1}\left(\stackrel{˙}{w}\right),{B}_{1}\left(\stackrel{˙}{w}\right)〉d{\beta }_{e\left(w\right)}^{e\left(w\right)}\left(x\right)d\stackrel{~}{\eta }\left(\stackrel{˙}{w}\right)\end{array}$
where
 $\begin{array}{ccc}{A}_{1}\left(\stackrel{˙}{w}\right)& =& \int {g}_{2}\left(v\right){\zeta }_{1}\left(v\right)\delta {\left(\sigma \left(v\right)\right)}^{\frac{1}{2}}d{\mu }_{1}^{\stackrel{˙}{w}}\left(v\right)\end{array}$
 $\begin{array}{ccc}{B}_{1}\left(\stackrel{˙}{w}\right)& =& \int {g}_{1}\left(u\right){\zeta }_{2}\left(u\right)\delta {\left(\sigma \left(u\right)\right)}^{\frac{1}{2}}d{\mu }_{1}^{\stackrel{˙}{w}}\left(u\right)\end{array}$
Moreover, if $f$  is of the form $f\left(x\right)={g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)$  and ${\zeta }_{1},{\zeta }_{2}\in {\int }_{{G}^{\left(0\right)}}^{\oplus }$  $\mathcal{ℋ}\left(e\left(u\right)\right)d{\eta }_{2}\left(u\right)$  , then
 $\begin{array}{ccc}& & 〈VW{L}_{1}\left(f\right){W}^{*}{V}^{*}{\zeta }_{1},{\zeta }_{2}〉\end{array}$
 $\begin{array}{ccc}& =& \int \int g\left(x\right)\delta {\left(x\right)}^{\frac{-1}{2}}〈L\left(x\right){A}_{2}\left(\stackrel{˙}{w}\right),{B}_{2}\left(\stackrel{˙}{w}\right)〉d{\beta }_{e\left(u\right)}^{e\left(u\right)}\left(x\right)d\stackrel{~}{\eta }\left(\stackrel{˙}{w}\right)\end{array}$
 $\begin{array}{ccc}& =& 〈W{L}_{2}\left(\Phi \left(f\right)\right){W}^{*}{\zeta }_{1},{\zeta }_{2}〉\end{array}$
where
 $\begin{array}{ccc}{A}_{2}\left(\stackrel{˙}{w}\right)& =& \int {g}_{2}\left(v\right){V}^{*}{\zeta }_{1}\left(v\right)\delta {\left(\sigma \left(v\right)\right)}^{\frac{1}{2}}d{\mu }_{1}^{\stackrel{˙}{w}}\left(v\right)\end{array}$
 $\begin{array}{ccc}& =& \int {U}_{\stackrel{˙}{v}}\left({g}_{2}\right)\left(v\right){\zeta }_{1}\left(v\right)\delta {\left(\sigma \left(v\right)\right)}^{\frac{1}{2}}d{\mu }_{2}^{\stackrel{˙}{w}}\left(v\right)\end{array}$
 $\begin{array}{ccc}{B}_{2}\left(\stackrel{˙}{w}\right)& =& \int {g}_{1}\left(v\right){V}^{*}{\zeta }_{2}\left(v\right)\delta {\left(\sigma \left(v\right)\right)}^{\frac{1}{2}}d{\mu }_{1}^{\stackrel{˙}{w}}\left(v\right)\end{array}$
 $\begin{array}{ccc}& =& \int {U}_{\stackrel{˙}{u}}\left({g}_{1}\right)\left(u\right){\zeta }_{2}\left(u\right)\delta {\left(\sigma \left(u\right)\right)}^{\frac{1}{2}}d{\mu }_{2}^{\stackrel{˙}{w}}\left(u\right)\end{array}$
Therefore $\parallel {L}_{1}\left(f\right)\parallel =\parallel {L}_{2}\left(\Phi \left(f\right)\right)\parallel$  . Consequently we can extend $\Phi$  to a $*$  -homomorphism between the ${M}_{\sigma }^{*}\left(G,{\nu }_{1}\right)$  and ${M}_{\sigma }^{*}\left(G,{\nu }_{2}\right)$  . It is not hard to see that $\Phi$  is in fact a $*$  -isomorphism: ${\Phi }^{-1}\left(f\right)=\left(x\to {U}_{\pi \left(r\left(x\right)\right)}^{*}\left({g}_{1}\right)\left(r\left(x\right)\right)g\left(q\left(x\right)\right){U}_{\pi \left(d\left(x\right)\right)}^{*}\left({g}_{2}\right)\left(d\left(x\right)\right)\right)$  for each $f$  of the form $f\left(x\right)={g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right).$

4 The case of locally transitive groupoids

A locally compact transitive groupoid $G$  is a groupoid for which all orbits $\left[u\right]$  are open in ${G}^{\left(0\right)}$  . We shall prove that if $G$  is a locally compact second countable locally transitive groupoid endowed with a Haar system $\nu$  , then ${C}^{*}\left(G,\nu \right)={M}^{*}\left(G,\nu \right)={M}_{\sigma }^{*}\left(G,\nu \right)$  for any regular cross section $\sigma .$
Notation 10 Let $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  be a fixed Haar system on $G$  . Let $\mu$  be a quasi-invariant measure, $\Delta$  its modular function, ${\nu }_{1}$  be the measure induced by $\mu$  on $G$  and ${\nu }_{0}={\Delta }^{-\frac{1}{2}}{\nu }_{1}$  . Let us denote by $I{I}_{\mu }\left(G\right)$  the set $\left\{f\in {L}^{1}\left(G,{\nu }_{0}\right):{\parallel f\parallel }_{II,\mu }<\infty \right\}\text{,}$  where ${\parallel f\parallel }_{II,\mu }$  is defined by ${\parallel f\parallel }_{II,\mu }=sup\left\{\int |f\left(x\right)j\left(d\left(x\right)\right)k\left(r\left(x\right)\right)|d{\nu }_{0}\left(x\right),\int {|j|}^{2}d\mu =\int {|k|}^{2}d\mu =1\right\}\text{.}$  If ${\mu }_{1}$  and ${\mu }_{2}$  are two equivalent quasi-invariant measures, then ${\parallel f\parallel }_{II,{\mu }_{1}}={\parallel f\parallel }_{II,{\mu }_{2}}$  , because ${\parallel f\parallel }_{II,\mu }=\parallel I{I}_{\mu }\left(|f|\right)\parallel$  for each quasi-invariant measure $\mu$  , where $I{I}_{\mu }$  is the one dimensional trivial representation on $\mu$  Define ${\parallel f\parallel }_{II}$  to be $sup\left\{{\parallel f\parallel }_{II,\mu }\text{:}\mu \text{quasi-invariant Radon measure on}{G}^{\left(0\right)}\right\}$  The supremum can be taken over the classes of quasi-invariant measure.
If $\parallel \parallel$  is the full ${C}^{*}$  -norm on ${C}_{c}\left(G\right)$  , then $\parallel f\parallel \le {\parallel f\parallel }_{II}\text{for all}f\text{.}$  (see [7)
Lemma 11 Let $G$  be a locally compact second countable groupoid with proper orbit space. Let $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  be a Haar system on $G$  , $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }^{\stackrel{˙}{u}}\right\}\right)$  its decomposition over the principal groupoid associated to $G$  and $\delta$  the associated $1$  -cocycle. If $f$  is a universally measurable function on $G$  , then ${\parallel f\parallel }_{II}\le {sup}_{\stackrel{˙}{w}}{\left(\int \int {\left(\int |f\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)\right)}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}.$
Proof. Each Radon quasi-invariant measure is equivalent with a Radon measure of the form $\int {\mu }^{\stackrel{˙}{u}}d\stackrel{~}{\mu }\left(\stackrel{˙}{u}\right)$  , where $\stackrel{~}{\mu }$  is a probability measure on the orbit space $G/{G}^{\left(0\right)}.$  Therefore for the computation of ${\parallel .\parallel }_{II}$  it is enough to consider only the quasi-invariant measures of the form $\mu =\int {\mu }^{\stackrel{˙}{u}}d\stackrel{~}{\mu }\left(\stackrel{˙}{u}\right)$  , where $\stackrel{~}{\mu }$  is a probability measure on ${G}^{\left(0\right)}/G$  . It is easy to see that the modular function of $\int {\mu }^{\stackrel{˙}{u}}d\stackrel{~}{\mu }\left(\stackrel{˙}{u}\right)$  is $\Delta =\delta$  .
Let $j,k\in {L}^{2}\left({G}^{\left(0\right)},\mu \right)$  with $\int |j{|}^{2}d\mu =\int |k{|}^{2}d\mu =1$  . We have
 $\begin{array}{ccc}& & \int \int \int \int |f\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)|j\left(v\right)||k\left(u\right)|d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)d\stackrel{~}{\mu }\left(\stackrel{˙}{w}\right)\end{array}$
 $\begin{array}{ccc}& \le & \int {\left(\int \int {\left(\int |f\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)\right)}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}\cdot \end{array}$
 $\begin{array}{ccc}& & \cdot {\left(\int \int {|j\left(v\right)|}^{2}{|k\left(u\right)|}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}d\stackrel{~}{\mu }\left(\stackrel{˙}{w}\right)\end{array}$
 $\begin{array}{ccc}& =& {sup}_{\stackrel{˙}{w}}{\left(\int \int {\left(\int |f\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)\right)}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}\cdot \end{array}$
 $\begin{array}{ccc}& & \cdot \int {\left(\int {|j\left(v\right)|}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)\right)}^{\frac{1}{2}}{\left({|k\left(u\right)|}^{2}d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}d\stackrel{~}{\mu }\left(\stackrel{˙}{w}\right)\end{array}$
 $\begin{array}{ccc}& \le & {sup}_{\stackrel{˙}{w}}{\left(\int \int {\left(\int |f\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)\right)}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}\end{array}$
Consequently, ${\parallel f\parallel }_{II}\le {sup}_{\stackrel{˙}{w}}{\left(\int \int {\left(\int |f\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)\right)}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}$
If $G$  is locally transitive, each orbit $\left[u\right]$  is open in ${G}^{\left(0\right)}$  . Each measure ${\mu }^{\stackrel{˙}{u}}$  is supported on $\left[u\right]$  . Since $\left(\left[u\right]\right)$  is a partition of ${G}^{\left(0\right)}$  into open sets, it follows that there is a unique Radon measure $m$  on ${G}^{\left(0\right)}$  such that the restriction of $m$  at ${C}_{c}\left(\left[u\right]\right)$  is ${\mu }^{\stackrel{˙}{u}}$  for each $\left[u\right]$  .
Corollary 12 Let $G$  be a locally compact second countable locally transitive groupoid endowed with a Haar system $\nu =\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  . Let $f$  be a universally measurable function such that ${\parallel f\parallel }_{II}<\infty$  .
• 1. If ${\left({f}_{n}\right)}_{n}$  is a uniformly bounded sequence of universally measurable functions supported on a compact set, and if ${\left({f}_{n}\right)}_{n}$  converges pointwise to $f$  , then ${\left({f}_{n}\right)}_{n}$  converges to $f$  in the norm of ${C}^{*}\left(G,\nu \right).$
• 2. If ${\left({f}_{n}\right)}_{n}$  is an increasing sequence of universally measurable nonnegative functions on $G$  that converges pointwise to $f$  , then ${\left({f}_{n}\right)}_{n}$  converges to $f$  in the norm of ${C}^{*}\left(G,\nu \right).$
Proof. Let $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }^{\stackrel{˙}{u}}\right\}\right)$  be the decomposition of the Haar system over the principal groupoid associated to $G$  and $\delta$  the associated $1$  -cocycle. Let $m$  be the unique measure such that restriction of $m$  at ${C}_{c}\left(\left[u\right]\right)$  is ${\mu }^{\stackrel{˙}{u}}$  for each $\left[u\right]$  . Let ${\left({f}_{n}\right)}_{n}$  be a sequence of universally measurable functions supported on a compact set $K$  . Let us write $M={sup}_{u,v}{\beta }_{u}^{v}\left({K}^{-1}\right)\text{.}$  and let us assume that ${\left({f}_{n}\right)}_{n}$  converges pointwise to $f$  . According to Lemma  11 , we have ${\parallel f-{f}_{n}\parallel }_{II}\le {sup}_{\stackrel{˙}{w}}{\left(\int \int {\left(\int |f\left(x\right)-{f}_{n}\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)\right)}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}\text{.}$  Hence
 $\begin{array}{ccc}{\parallel f-{f}_{n}\parallel }_{II}& \le & {sup}_{\stackrel{˙}{w}}M{\left(\int \int \left(\int {|f\left(x\right)-{f}_{n}\left(x\right)|}^{2}d{\beta }_{v}^{u}\left(x\right)\right)d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}\end{array}$
 $\begin{array}{ccc}& \le & M{\left(\int \int \left(\int {|f\left(x\right)-{f}_{n}\left(x\right)|}^{2}d{\beta }_{u}^{v}\left(x\right)\right)dm\left(v\right)dm\left(u\right)\right)}^{\frac{1}{2}}\text{.}\end{array}$
If $\parallel \parallel$  denotes the ${C}^{*}$  -norm, then ${lim}_{n}\parallel f-{f}_{n}\parallel \le {lim}_{n}{\parallel f-{f}_{n}\parallel }_{II}=0\text{,}$  because $\int \int \left(\int {|f\left(x\right)-{f}_{n}\left(x\right)|}^{2}d{\beta }_{u}^{v}\left(x\right)\right)dm\left(v\right)dm\left(u\right)$  converges to zero, by the Dominated Convergence Theorem. Let ${\left({f}_{n}\right)}_{n}$  be an increasing sequence of universally measurable nonnegative functions that converges pointwise to $f.$  Since
 $\begin{array}{ccc}{\parallel f-{f}_{n}\parallel }_{II}& \le & {sup}_{\stackrel{˙}{w}}{\left(\int \int {\left(\int |f\left(x\right)-{f}_{n}\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)\right)}^{2}d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}\end{array}$
 $\begin{array}{ccc}& \le & {\left(\int \int {\left(\int |f\left(x\right)-{f}_{n}\left(x\right)|\delta {\left(x\right)}^{-\frac{1}{2}}d{\beta }_{v}^{u}\left(x\right)\right)}^{2}dm\left(v\right)dm\left(u\right)\right)}^{\frac{1}{2}}\end{array}$
it follows that ${lim}_{n}{\parallel f-{f}_{n}\parallel }_{II}=0\text{.}$
Proposition 13 Let $G$  be a locally compact second countable locally transitive groupoid endowed with a Haar system $\nu =\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  .Then any function in ${\mathcal{ℬ}}_{c}\left(G\right)$  , the space of compactly supported Borel bounded function on $G$  , can be viewed as an element of ${C}^{*}\left(G,\nu \right)$  .
Proof. Let $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }^{\stackrel{˙}{u}}\right\}\right)$  be the decomposition of the Haar system over the principal groupoid associated to $G$  , and $\delta$  the associated $1$  -cocycle. Let $m$  be the unique measure such that restriction of $m$  at ${C}_{c}\left(\left[u\right]\right)$  is ${\mu }^{\stackrel{˙}{u}}$  for each $\left[u\right]$  .
Let $m$  be a dominant for the family $\left\{{\mu }^{\stackrel{˙}{u}}\right\}$  . Let ${\nu }_{1}$  be the measure on $G$  define by $\int f\left(x\right)d{\nu }_{1}\left(x\right)=\left(\int \int \left(\int f\left(x\right)d{\beta }_{u}^{v}\left(x\right)\right)dm\left(v\right)dm\left(u\right)\right)$  for all Borel nonnegative function $f$  . If $f\in$  ${\mathcal{ℬ}}_{c}\left(G\right)$  , then $f$  is the limit in ${L}^{2}\left(G,{\nu }_{1}\right)$  of a sequence, ${\left({f}_{n}\right)}_{n}$  , in ${C}_{c}\left(G\right)$  that is supported on some compact set $K$  supporting $f$  . If we write $M={sup}_{u,v}{\beta }_{u}^{v}\left({K}^{-1}\right)\text{,}$  we have
 $\begin{array}{ccc}{\parallel f-{f}_{n}\parallel }_{II}& \le & {sup}_{\stackrel{˙}{w}}M{\left(\int \int \left(\int {|f\left(x\right)-{f}_{n}\left(x\right)|}^{2}d{\beta }_{v}^{u}\left(x\right)\right)d{\mu }^{\stackrel{˙}{w}}\left(v\right)d{\mu }^{\stackrel{˙}{w}}\left(u\right)\right)}^{\frac{1}{2}}\end{array}$
 $\begin{array}{ccc}& \le & M{\left(\int \int \left(\int {|f\left(x\right)-{f}_{n}\left(x\right)|}^{2}d{\beta }_{u}^{v}\left(x\right)\right)dm\left(v\right)dm\left(u\right)\right)}^{\frac{1}{2}}\text{.}\end{array}$
If $\parallel \parallel$  denotes the ${C}^{*}$  -norm, then ${lim}_{n}\parallel f-{f}_{n}\parallel \le {lim}_{n}{\parallel f-{f}_{n}\parallel }_{II}=0\text{.}$  Thus $f$  can be viewed as an element in ${C}^{*}\left(G,\nu \right)$  .
Proposition 14 If $G$  is a locally compact second countable locally transitive groupoid endowed with a Haar system $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  with bounded decomposition, then ${C}^{*}\left(G,\nu \right)={M}^{*}\left(G,\nu \right)\text{.}$
Proof. It follows from the Proposition  13 .
Remark 15 Let $G$  be locally compact locally transitive groupoid. Let $F$  be a subset of ${G}^{\left(0\right)}$  containing only one element $e\left(u\right)$  in each orbit $\left[u\right]$  . It is easy to see that $F$  is a closed subset of $G$  and that $F$  is a discrete space. Let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a regular cross section of ${d}_{F}$  . Let us endow ${\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  with the topology induced from ${G}^{\left(0\right)}×{G}_{F}^{F}×{G}^{\left(0\right)}$  . The topology of ${\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  is locally compact because ${\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  is a closed subset of the locally compact space ${G}^{\left(0\right)}×{G}_{F}^{F}×{G}^{\left(0\right)}$  . With the operations:
 $\begin{array}{ccc}\left(u,x,v\right)\left(v,y,w\right)& =& \left(u,xy,w\right)\end{array}$
 $\begin{array}{ccc}{\left(u,x,v\right)}^{-1}& =& \left(v,{x}^{-1},u\right)\end{array}$
${\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  becomes a groupoid. Let us define $\phi :G\to {\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  by $\phi \left(x\right)=\left(r\left(x\right),\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1},d\left(x\right)\right)$  and note that $\phi$  is a Borel isomorphism which carries compact sets to relatively compact sets .
Lemma 16 Let $G$  be locally compact second countable locally transitive groupoid.
Let $F$  be a subset of ${G}^{\left(0\right)}$  containing only one element $e\left(u\right)$  in each orbit $\left[u\right]$  . Let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a regular cross section of ${d}_{F}$  . Then any compactly supported Borel bounded function on $G$  is pointwise limit of a uniformly bounded sequence ${\left({f}_{n}\right)}_{n}$  of Borel functions supported on a compact set supporting $f$  , having the property that each ${f}_{n}$  is a linear combination of functions of the form $x\to {g}_{1}\left(r\left(x\right)\right)g\left(\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1}\right){g}_{2}\left(d\left(x\right)\right)$  where ${g}_{1},{g}_{2}$  are compactly supported Borel bounded function on ${G}^{\left(0\right)}$  and $g$  is a compactly supported Borel bounded function on ${G}_{F}^{F}$  .
Proof. Let us endow ${\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  with the topology induced from ${G}^{\left(0\right)}×{G}_{F}^{F}×{G}^{\left(0\right)}$  as in Remark. 15 . The topology of ${\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  is locally compact. Any compactly supported Borel bounded function on ${G}^{\left(0\right)}×{G}_{F}^{F}×{G}^{\left(0\right)}$  is pointwise limit of uniformly bounded sequences ${\left({f}_{n}\right)}_{n}$  of Borel functions supported on a compact set, such that each function ${f}_{n}$  is a linear combination of functions of the form $\left(u,x,v\right)\to {g}_{1}\left(u\right)g\left(x\right){g}_{2}\left(v\right)$  where ${g}_{1},{g}_{2}$  are compactly supported Borel bounded function on ${G}^{\left(0\right)}$  and $g$  is a compactly supported Borel bounded function on ${G}_{F}^{F}$  . Consequently, any compactly supported Borel bounded function on ${\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  has the same property. Since $\phi :G\to {\cup }_{\left[u\right]}\left[u\right]×{G}_{e\left(u\right)}^{e\left(u\right)}×\left[u\right]$  defined by $\phi \left(x\right)=\left(r\left(x\right),\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1},d\left(x\right)\right)$  is a Borel isomorphism which carries compact sets to relatively compact sets, it follows that any compactly supported Borel bounded function on $G$  can be represented as a pointwise limit of a uniformly bounded sequence ${\left({f}_{n}\right)}_{n}$  of Borel functions supported on a compact set supporting $f$  , having the property that each ${f}_{n}$  is a linear combination of functions of the form $x\to {g}_{1}\left(r\left(x\right)\right)g\left(\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1}\right){g}_{2}\left(d\left(x\right)\right)$
Corollary 17 Let $G$  be locally compact second countable locally transitive groupoid. Let $F$  be a subset of ${G}^{\left(0\right)}$  containing only one element $e\left(u\right)$  in each orbit $\left[u\right]$  . Let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a regular cross section of ${d}_{F}$  . Then the linear span of the functions of the form $x\to {g}_{1}\left(r\left(x\right)\right)g\left(\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1}\right){g}_{2}\left(d\left(x\right)\right)$  where ${g}_{1},{g}_{2}\in {\mathcal{ℬ}}_{c}\left({G}^{\left(0\right)}\right)$  and $g\in {\mathcal{ℬ}}_{c}\left({G}_{F}^{F}\right),$  is dense in the full ${C}^{*}$  -algebra of $G$  .
Proof. Let $f$  be a function on $G$  , defined by $f\left(x\right)={g}_{1}\left(r\left(x\right)\right)g\left(\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1}\right){g}_{2}\left(d\left(x\right)\right)$  where ${g}_{1},{g}_{2}\in {\mathcal{ℬ}}_{c}\left({G}^{\left(0\right)}\right)$  and $g\in {\mathcal{ℬ}}_{c}\left({G}_{F}^{F}\right)$  . Then $f\in {\mathcal{ℬ}}_{c}\left(G\right)$  , therefore it can be viewed as an element of the ${C}^{*}\left(G,\nu \right)$  as we note in Proposition  13 . Each $f\in {\mathcal{ℬ}}_{c}\left(G\right)$  (in particular in ${C}_{c}\left(G\right)$  ) is the limit (pointwise and consequently in the ${C}^{*}$  -norm according to Corollary  12 ) of a uniformly bounded sequence ${\left({f}_{n}\right)}_{n}$  of Borel functions supported on a compact set supporting $f$  , having the property that each ${f}_{n}$  is a linear combination of functions of the required form.
Proposition 18 Let $G$  is a locally compact second countable locally transitive groupoid endowed with a Haar system $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  . Let $F$  be a subset of ${G}^{\left(0\right)}$  containing only one element $e\left(u\right)$  in each orbit $\left[u\right]$  . Let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a regular cross section of ${d}_{F}$  . Then ${C}^{*}\left(G,\nu \right)={M}^{*}\left(G,\nu \right)={M}_{\sigma }^{*}\left(G,\nu \right)\text{.}$
Proof. We have proved that ${C}^{*}\left(G,\nu \right)={M}^{*}\left(G,\nu \right)$  . From the preceding corollary, it follows that the linear span of the functions of the form $x\to {g}_{1}\left(r\left(x\right)\right)g\left(\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1}\right){g}_{2}\left(d\left(x\right)\right)$  where ${g}_{1},{g}_{2}\in {\mathcal{ℬ}}_{c}\left({G}^{\left(0\right)}\right)$  and $g\in {\mathcal{ℬ}}_{c}\left({G}_{F}^{F}\right)$  is dense in ${C}^{*}\left(G,\nu \right)$  . But this space is contained in ${\mathcal{ℬ}}_{\sigma }\left(G\right)$  . Therefore ${C}^{*}\left(G,\nu \right)={M}^{*}\left(G,\nu \right)={M}_{\sigma }^{*}\left(G,\nu \right)$  .

5 The case of principal proper groupoids case

Notation 19 Let $G$  be a locally compact second countable groupoid with proper orbit space. Let $F$  be a Borel subset of ${G}^{\left(0\right)}$  containing only one element $e\left(u\right)$  in each orbit $\left[u\right]$  . Let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a regular cross section for ${d}_{F}:{G}^{F}\to {G}^{\left(0\right)},{d}_{F}\left(x\right)=d\left(x\right)$  with $\sigma \left(e\left(v\right)\right)=e\left(v\right)$  for all $v\in {G}^{\left(0\right)}$  . Let $q:G\to {G}_{F}^{F}$  be defined by $q\left(x\right)=\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1}$  We shall endow ${G}_{F}^{F}$  with the quotient topology induced by $q.$  We shall denote by ${\mathbf{C}}_{\sigma }\left(G\right)$  the linear span of the functions of the form $x\to {g}_{1}\left(r\left(x\right)\right)g\left(\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1}\right){g}_{2}\left(d\left(x\right)\right)$  where ${g}_{1},{g}_{2}\in {C}_{c}\left({G}^{\left(0\right)}\right)$  and $g\in {C}_{c}\left({G}_{F}^{F}\right)$  .
Proposition 20 With the Notation  19 , if the space of continuous functions with compact support on ${G}_{F}^{F}$  (with the respect to the quotient topology induced by $q$  ) separates the points of ${G}_{F}^{F}$  , then ${\mathbf{C}}_{\sigma }\left(G\right)$  is dense in ${C}_{c}\left(G\right)$  (for the inductive limit topology). In particular, if the quotient topology induced by $q$  on ${G}_{F}^{F}$  is a locally compact (Hausdorff ) topology, then ${\mathbf{C}}_{\sigma }\left(G\right)$  is dense in ${C}_{c}\left(G\right)$  .
Proof. If the space of continuous functions with compact support on ${G}_{F}^{F}$  (with the respect to the quotient topology induced by $q$  ) separates the points of ${G}_{F}^{F}$  , then ${\mathbf{C}}_{\sigma }\left(G\right)$  separates the points of $G$  . By Stone-Weierstrass Theorem, it follows that ${\mathbf{C}}_{\sigma }\left(G\right)$  is dense in ${C}_{c}\left(G\right)$  (for the inductive limit topology)
Proposition 21 Let $G$  be a locally compact principal groupoid. If $G$  is proper, then the quotient topology induced by $q$  on ${G}_{F}^{F}$  is a locally compact (Hausdorff ) topology. Consequently, ${\mathbf{C}}_{\sigma }\left(G\right)$  is dense in ${C}_{c}\left(G\right)$  for the inductive limit topology (we use the Notation  19 ).
Proof. Let $\pi :G\to {G}^{\left(0\right)}/G$  be the canonical projection. Let us note that for a principal groupoid the condition $q\left(x\right)=q\left(y\right)$  is equivalent with $\pi \left(r\left(x\right)\right)=\pi \left(r\left(y\right)\right)\text{.}$  First we shall prove that the topology on ${G}_{F}^{F}$  is Hausdorff. Let ${\left({x}_{i}\right)}_{i}$  and ${\left({y}_{i}\right)}_{i}$  be two nets with $q\left({x}_{i}\right)=q\left({y}_{i}\right)$  for every $i$  . Let us suppose that ${\left({x}_{i}\right)}_{i}$  converges to $x$  and ${\left({y}_{i}\right)}_{i}$  converges to $y$  . Then $lim\pi \left(r\left({x}_{i}\right)\right)=lim\pi \left(r\left({y}_{i}\right)\right)=\pi \left(r\left(x\right)\right)=\pi \left(r\left(y\right)\right)$  Hence $q\left(x\right)=q\left(y\right)$  , and therefore the topology on ${G}_{F}^{F}$  is Hausdorff. We shall prove that $q$  is open. If ${\left({z}_{i}\right)}_{i}$  is a net converging to $q\left(x\right)$  in ${G}_{F}^{F}$  , then $\pi \circ r\left({z}_{i}\right)$  converges to $\pi \circ r\left(x\right)$  . Since $\pi \circ r:G\to {G}^{\left(0\right)}/G$  is an open map, there is a net ${\left({x}_{i}\right)}_{i}$  converging to $x$  , such that $\pi \circ r\left({x}_{i}\right)=\pi \circ r\left({z}_{i}\right)$  , and consequently $q\left({x}_{i}\right)=q\left({z}_{i}\right)={z}_{i}$  . Hence $q$  is an open map and the quotient topology induced by $q$  on ${G}_{F}^{F}$  is locally compact.
Theorem 22 Let $G$  be a locally compact second countable groupoid with proper orbit space. Let $F$  be a Borel subset of ${G}^{\left(0\right)}$  meeting each orbit exactly once. Let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a regular cross section for $d:{G}^{F}\to G$  . Let us assume that the quotient topology induced by $q$  on ${G}_{F}^{F}$  is a locally compact (Hausdorff ) topology. Let $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  be a Haar system on $G$  . Then ${C}^{*}\left(G,\nu \right)\subset {M}_{\sigma }^{*}\left(G,\nu \right)\subset {M}^{*}\left(G,\nu \right)\text{.}$
Proof. From Proposition  20  ${\mathbf{C}}_{\sigma }\left(G\right)$  is dense in ${C}_{c}\left(G\right)$  for the inductive limit topology and hence is dense in ${C}^{*}\left(G,\nu \right)$  . Since ${\mathbf{C}}_{\sigma }\left(G\right)\subset {\mathcal{ℬ}}_{\sigma }\left(G\right)$  , it follows that ${C}^{*}\left(G,\nu \right)\subset {M}_{\sigma }^{*}\left(G,\nu \right)$  .
Corollary 23 Let $G$  be a locally compact second countable principal proper groupoid. Let $F$  be a Borel subset of ${G}^{\left(0\right)}$  meeting each orbit exactly once. Let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a regular cross section for $d:{G}^{F}\to G$  . Let $\left\{{\nu }^{u},u\in {G}^{\left(0\right)}\right\}$  be a Haar system on $G$  . Then ${C}^{*}\left(G,\nu \right)\subset {M}_{\sigma }^{*}\left(G,\nu \right)\subset {M}^{*}\left(G,\nu \right)\text{.}$
Proof. Applying Proposition  21 , we obtain that the quotient topology induced by $q$  on ${G}_{F}^{F}$  is a locally compact (Hausdorff ) topology. Therefore $G$  satisfies the hypothesis of Theorem  22 .
Definition 24 Let ${\left\{{\mu }^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  be a system of measures on ${G}^{\left(0\right)}$  satisfying:
• 1. $supp\left({\mu }^{\stackrel{˙}{u}}\right)=\left[u\right]$  for all $\stackrel{˙}{u}$  .
• 2. For all compactly supported continuous functions $f$  on ${G}^{\left(0\right)}$  the function $u\to \int f\left(v\right){\mu }^{\pi \left(u\right)}\left(v\right)$  is continuous
We shall say that the Hilbert bundle determined by the system of measures ${\left\{{\mu }^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  has a continuous basis if there is sequence ${f}_{1},{f}_{2},...$  of real valued continuous functions on ${G}^{\left(0\right)}$  such that $dim\left({L}^{2}\left({\mu }^{\stackrel{˙}{u}}\right)\right)=\infty$  if and only if ${\parallel {f}_{n}\parallel }_{2}$  for $n=1,2,...$  and then $\left\{{f}_{1},{f}_{2},...\right\}$  gives an ortonormate basis of ${L}^{2}\left({\mu }^{\stackrel{˙}{u}}\right)$  , while $dim\left({L}^{2}\left({\mu }^{\stackrel{˙}{u}}\right)\right)=k<\infty$  if and only if ${\parallel {f}_{n}\parallel }_{2}=1$  for $n\le k$  , and ${\parallel {f}_{n}\parallel }_{2}=0$  for $n>k$  and then $\left\{{f}_{1},{f}_{2},...{f}_{k}\right\}$  gives an ortonormate basis of ${L}^{2}\left({\mu }^{\stackrel{˙}{u}}\right)$  .
Remark 25 Let ${\left\{{\mu }_{1}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  and ${\left\{{\mu }_{2}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  be two systems of measures on ${G}^{\left(0\right)}$  satisfying:
• 1. $supp\left({\mu }_{i}^{\stackrel{˙}{u}}\right)=\left[u\right]$  for all $\stackrel{˙}{u}$  , $i=1,2$  .
• 2. For all compactly supported continuous functions $f$  on ${G}^{\left(0\right)}$  the function $u\to \int f\left(v\right){\mu }_{i}^{\pi \left(u\right)}\left(v\right)$  is continuous
Let us assume that the Hilbert bundles determined by the systems of measures ${\left\{{\mu }_{i}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  have continuous bases. Let ${f}_{1}.{f}_{2},...$  be a continuous basis for Hilbert bundle determined by ${\left\{{\mu }_{1}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  and let ${g}_{1},{g}_{2},...$  be a continuous basis for Hilbert bundle determined by ${\left\{{\mu }_{2}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  . Let us define a unitary operator ${U}_{\stackrel{˙}{u}}:{L}^{2}\left({\mu }_{1}^{\stackrel{˙}{u}}\right)\to {L}^{2}\left({\mu }_{2}^{\stackrel{˙}{u}}\right)$  by ${U}_{\stackrel{˙}{u}}\left({f}_{n}\right)={g}_{n}\text{for all}n$  Then the family ${\left\{{U}_{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  has the following properties:
• 1. For all Borel bounded function $f$  on ${G}^{\left(0\right)}$  , $u\to {U}_{\pi \left(u\right)}\left(f\right)$  is a bounded Borel function with compact support.
• 2. For all Borel bounded function $f$  on ${G}^{\left(0\right)}$  , ${U}_{\pi \left(u\right)}\left(\overline{f}\right)=\overline{{U}_{\pi \left(u\right)}\left(f\right)}$
• 3. For all compactly supported continuous functions $f$  on ${G}^{\left(0\right)}$  there is a sequence ${\left({h}_{n}\right)}_{n}$  of compactly supported continuous functions on ${G}^{\left(0\right)}$  such that $su{p}_{\stackrel{˙}{u}}\int {|{U}_{\stackrel{˙}{u}}\left(f\right)-{h}_{n}|}^{2}d{\mu }_{2}^{\stackrel{˙}{u}}\to 0\left(n\to \infty \right)$
Indeed, we can define ${h}_{n}\left(v\right)={\sum }_{k=1}^{n}{g}_{k}\left(v\right)\int f\left(u\right){f}_{k}\left(u\right)d{\mu }_{1}^{\pi \left(v\right)}\left(u\right)\text{.}$
Remark 26 Let $G$  be a locally compact second countable groupoid with proper orbit space. Let $F$  be a Borel subset of ${G}^{\left(0\right)}$  containing only one element $e\left(u\right)$  in each orbit $\left[u\right]$  . Let us assume that $F\cap \left[K\right]$  has a compact closure for each compact subset $K$  of ${G}^{\left(0\right)}$  , and let $\sigma :{G}^{\left(0\right)}\to {G}^{F}$  be a regular cross section for ${d}_{F}:{G}^{F}\to {G}^{\left(0\right)}$  . Let us endow ${G}_{F}^{F}$  with the quotient topology induced by $q:G\to {G}_{F}^{F}$  $q\left(x\right)=\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1},x\in G$  If $g\in {C}_{c}\left({G}_{F}^{F}\right)$  and ${g}_{1},{g}_{2}$  are two functions on ${G}^{\left(0\right)}$  with the property that there is two sequences ${\left({h}_{n}^{1}\right)}_{n}$  and ${\left({h}_{n}^{2}\right)}_{n}$  of compactly supported continuous functions on ${G}^{\left(0\right)}$  such that $su{p}_{\stackrel{˙}{u}}\int {|{g}_{i}-{h}_{n}^{i}|}^{2}d{\mu }_{2}^{\stackrel{˙}{u}}\to 0\left(n\to \infty \right)$  for $i=1,2$  , then $x\to f{g}_{1}\left(r\left(x\right)\right)g\left(\sigma \left(r\left(x\right)\right)x\sigma {\left(d\left(x\right)\right)}^{-1}\right){g}_{2}\left(d\left(x\right)\right)$  can be viewed as an element of ${C}^{*}\left(G,\nu \right)$  . Indeed, it is easy to see that ${‖f-\left({h}_{n}^{1}\circ r\right)\left(g\circ q\right)\left({h}_{n}^{2}\circ d\right)‖}_{II}\to 0\left(n\to \infty \right).$
Proposition 27 Let $G$  be a locally compact second countable principal proper groupoid. Let ${\nu }_{i}=\left\{{\nu }_{i}^{u},u\in {G}^{\left(0\right)}\right\},$  $i=1,2$  be two Haar system on $G$  and $\left(\left\{{\beta }_{v}^{u}\right\},\left\{{\mu }_{i}^{\stackrel{˙}{u}}\right\}\right)$  the corresponding decompositions over the principal groupoid. If the Hilbert bundles determined by the systems of measures ${\left\{{\mu }_{i}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  have continuous bases, then the ${C}^{*}$  -algebras ${C}^{*}\left(G,{\nu }_{1}\right)$  and ${C}^{*}\left(G,{\nu }_{2}\right)$  are $*$  -isomorphic.
Proof. We use the Notation  19 . From Proposition  20  ${\mathbf{C}}_{\sigma }\left(G\right)$  is dense in ${C}_{c}\left(G\right)$  for the inductive limit topology and hence is dense in ${C}^{*}\left(G,{\nu }_{1}\right)$  . We shall define a $*$  -homomorphism $\Phi$  from ${\mathbf{C}}_{\sigma }\left(G\right)$  to ${C}^{*}\left(G,{\nu }_{2}\right)$  . It suffices to define $\Phi$  on the set of function on $G$  of the form $x\to {g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)$  where ${g}_{1},{g}_{2}\in {C}_{c}\left({G}^{\left(0\right)}\right)$  and $g\in {C}_{c}\left({G}_{F}^{F}\right)$  . Let ${\left\{{U}_{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  be the family of unitary operators with the properties stated in Remark  25 associated to the systems of measures ${\left\{{\mu }_{i}^{\stackrel{˙}{u}}\right\}}_{\stackrel{˙}{u}}$  , $i=1,2$  .
Let us define $\Phi$  by $\Phi \left(f\right)=\left(x\to {U}_{\pi \left(r\left(x\right)\right)}\left({g}_{1}\right)\left(r\left(x\right)\right)g\left(q\left(x\right)\right){U}_{\pi \left(d\left(x\right)\right)}\left({g}_{2}\right)\left(d\left(x\right)\right)\right)$  where $f$  is defined by $f\left(x\right)={g}_{1}\left(r\left(x\right)\right)g\left(q\left(x\right)\right){g}_{2}\left(d\left(x\right)\right)$  with ${g}_{1},{g}_{2}\in {C}_{c}\left({G}^{\left(0\right)}\right)$  and $g\in {C}_{c}\left({G}_{F}^{F}\right)$  .
We noted in Remark  26 that the functions of the form $\Phi \left(f\right)$  can be viewed as elements of ${C}^{*}\left(G,{\nu }_{2}\right)$  . With the same argument as in the proof of Theorem  9 , it follows that $\Phi$  can be extended to $*$  -isomorphism between ${C}^{*}\left(G,{\nu }_{1}\right)$  and ${C}^{*}\left(G,{\nu }_{2}\right)$  .
References

1. C. Anantharaman-Delaroche, J. Renault, Amenable groupoids, Monographie de L'Enseignement Mathematique No 36, Geneve, 2000.
2. G. Mackey, Induced representations of locally compact groups. $I$  , Ann. of Math., 55(1952), 101-139.
3. P. Muhly, J. Renault and D. Williams, Equivalence and isomorphism for groupoid C*-algebras, J. Operator Theory 17(1987), 3-22.
4. P. Muhly, Coordinates in operator algebra, (Book in preparation).
5. A. Ramsay, Virtual groups and groups actions, Adv. in Math. 6(1971), 253-322.
6. A. Ramsay, Topologies on measured groupoids, J. Funct. Anal. 47(1982), 314-343.
7. A. Ramsay and M. E. Walter, Fourier-Stieltjes Algebras of locally compact groupoids, J. Funct. Anal. 148(1997), 314-367.
8. J. Renault, A groupoid approach to ${C}^{*}$  algebras, Lecture Notes in Math., Springer-Verlag, 793, 1980.
9. J. Renault,The ideal structure of groupoid crossed product algebras, J. Operator Theory, 25(1991), 3-36.
10. J. Westman, Nontransitive groupoid algebras, Univ. of California at Irvine, 1967.

University Constantin Brâncuşi of Târgu-Jiu Bulevardul Republicii, Nr. 1, 210152 Târgu-Jiu , Gorj Romania e-mail: ada@utgjiu.ro