### November 27, 2006

2000 Mathematics Subject Classification. Primary 46L65, Secondary 46L08.Partially supported by Dinacyt (Proyecto Clemente Estable 8013), Uruguay.
<ph f="cmbx">Morita equivalence for Quantum Heisenberg Manifolds</ph>

Centro de Matematicas. Facultad de Ciencias. Igua 4225, CP 11 400, Montevideo, Uruguay. E-mail address : abadie@cmat.edu.uy
• Abstract. We discuss Morita equivalence within the family $\left\{Dc\mu \nu :c\in \mathbb{Z},c>0,\mu ,\nu \in \mathbb{R}\right\}$  of quantum Heisenberg manifolds. Morita equivalence classes are described in terms of the parameters $\mu$  , $\nu$  and the rank of the free abelian group ${G}_{\mu \nu }=2\mu \mathbb{Z}+2\nu \mathbb{Z}+\mathbb{Z}$  associated to the $C*$  -algebra $Dc\mu \nu$  .
Introduction. Quantum Heisenberg manifolds $\left\{Dc\mu \nu :c\in \mathbb{Z},c>0,\mu ,\nu \in \mathbb{R}\right\}$  were constructed by Rieffel in [Rf4as a quantization deformation of certain homogeneous spaces $H/{N}_{c}$  , $H$  being the Heisenberg group.
It was shown in [Ab1,3.4that ${K}_{0}\left(Dc\mu \nu \right)\sim ={\mathbb{Z}}^{3}\oplus {\mathbb{Z}}_{c}$  , which implies that $Dc\mu \nu$  and ${D}_{{\mu }^{\prime }{\nu }^{\prime }}^{{c}^{\prime }}$  are not isomorphic unless $c={c}^{\prime }$  . Besides, $Dc\mu \nu$  and ${D}_{{\mu }^{\prime }{\nu }^{\prime }}^{c}$  are isomorphic when $\left(2\mu ,2\nu \right)$  and $\left(2{\mu }^{\prime },2{\nu }^{\prime }\right)$  belong to the same orbit under the usual action of $\text{GL}2\left(\mathbb{Z}\right)\text{}$  on ${\mathbb{T}}^{2}$  ( [AE,Theorem2.2; see also [Ab2,3.3).
The range of traces on $Dc\mu \nu$  was discussed in [Ab2, where it was shown that the range of the homomorphism induced on ${K}_{0}\left(Dc\mu \nu \right)$  by any tracial state on $Dc\mu \nu$  has range ${G}_{\mu \nu }=\mathbb{Z}+2\mu \mathbb{Z}+2\nu \mathbb{Z}$  . As a consequence ([Ab2,3.17), the isomorphism condition stated above turns out to be necessary when the rank of ${G}_{\mu \nu }$  is either 1 or 3. Rieffel showed in [Rf4that $Dc\mu \nu$  is simple if and only if $\left\{1,\mu ,\nu \right\}$  is linearly independent over the field of rational numbers (i.e rank ${G}_{\mu \nu }=3$  ); it might be interesting to know whether in this case the classification can be made by means of the results of Elliott and Gong ([EG).
The quantum Heisenberg manifold $Dc\mu \nu$  was described in [AEEas a crossed product by a Hilbert $C*$  -bimodule. In order to discuss Morita equivalence within this family, we adapt to this setting some of the techniques employed in the analogous discussion for non-commutative tori ([Rf3) and Heisenberg $C*$  -algebras ([Pa2). Thus we generalize in Section 1 Green's result (discussed by Rieffel in [Rf2,Situation10) on the Morita equivalence of the crossed products ${C}_{0}\left(M/K\right)\mathbb{\setminus }H$  and ${C}_{0}\left(M/H\right)\mathbb{\setminus }K$  , for free and proper commuting actions on a locally compact space $M$  . This result provides the main tool to discuss, in Section 2, Morita equivalence for quantum Heisenberg manifolds.

1 Morita equivalence of crossed products by certain Hilbert $C*$  -bimodules over commutative $C*$  -algebras.

For a Hilbert $C*$  -bimodule $X$  over a $C*$  -algebra $A$  , the crossed product $A\mathbb{\setminus }X$  was introduced in [AEE(see also [Pi) as the universal $C*$  -algebra for which there exist a *-homomorphism ${i}_{A}:A⟶A\mathbb{\setminus }X$  and a continuous linear map ${i}_{X}:X⟶A\mathbb{\setminus }X$  such that
 $\begin{array}{c}{i}_{X}\left(ax\right)={i}_{A}\left(a\right){i}_{X}\left(x\right),{i}_{A}\left(〈x,y{〉}_{L}\right)={i}_{X}\left(x\right){i}_{X}\left(y{\right)}^{*}.\end{array}$
 $\begin{array}{c}{i}_{X}\left(xa\right)={i}_{X}\left(x\right){i}_{A}\left(a\right),{i}_{A}\left(〈x,y{〉}_{R}\right)={i}_{X}\left(x{\right)}^{*}{i}_{X}\left(y\right).\end{array}$
The crossed product $A\mathbb{\setminus }X$  carries a dual action $\delta$  of ${S}^{1}$  , defined by ${\delta }_{z}\left({i}_{A}\left(a\right)\right)={i}_{A}\left(a\right)$  , ${\delta }_{z}\left({i}_{X}\left(x\right)\right)=z{i}_{X}\left(x\right)$  , for $a\in A$  , $x\in X$  and $z\in {S}^{1}$  .
Moreover, if a $C*$  -algebra $B$  carries an action $\delta$  of ${S}^{1}$  such that $B$  is generated as a $C*$  -algebra by the fixed point subalgebra ${B}_{0}=\left\{b\in B:{\delta }_{z}\left(b\right)=b\forall z\in {S}^{1}\right\}$  and the first spectral subspace ${B}_{1}=\left\{b\in B:{\delta }_{z}\left(b\right)=zb\forall z\in {S}^{1}\right\}$  , then $B$  is isomorphic to ${B}_{0}\mathbb{\setminus }{B}_{1}$  (where ${B}_{1}$  has the obvious Hilbert $C*$  -bimodule structure over ${B}_{0}$  ), via an isomorphism that takes the action $\delta$  into the dual action.
If $X$  is an $A$  -Hilbert $C*$  -bimodule and $\alpha \in \text{Aut}\left(A\right)$  , we denote by ${X}_{\alpha }$  the Hilbert $C*$  -bimodule over $A$  obtained by leaving unchanged the left structure, and by setting $x{\cdot }_{{X}_{\alpha }}a:=x\alpha \left(a\right),〈x,y{〉}_{R}^{{X}_{\alpha }}:={\alpha }^{-1}\left(〈x,y{〉}_{R}\right),$  where the undecorated notation refers to the original right structure of $X$  .
For $\alpha \in \text{Aut}\left(A\right)$  and the usual $A$  -Hilbert $C*$  -bimodule structure on $A$  , the crossed product $A\mathbb{\setminus }{A}_{\alpha }$  is easily checked to be the usual crossed product $A{\mathbb{\setminus }}_{\alpha }\mathbb{Z}$  .
Definition 1.1. Given a proper action $\alpha$  of $\mathbb{Z}$  on a locally compact Hausdorff space $M$  and a unitary $u\in {C}_{b}\left(M\right)$  , let ${X}^{\alpha ,u}$  denote the set of functions $f\in {C}_{b}\left(M\right)$  satisfying $f=u\alpha \left(f\right)$  , and such that the map $x↦|f\left(x\right)|$  , which is constant on $\alpha$  -orbits, belongs to ${C}_{0}\left(M/\alpha \right)$  .
Then $X\alpha ,u$  is a Hilbert $C*$  -bimodule over ${C}_{0}\left(M/\alpha \right)$  for pointwise multiplication on the left and the right, and inner products given by $〈f,g{〉}_{L}=f\overline{g}$  , $〈f,g{〉}_{R}=\overline{f}g$  .
Proposition 1.2. Let $\alpha$  and $\beta$  be free and proper commuting actions of $\mathbb{Z}$  on a locally compact Hausdorff space $M$  , and let $u$  be a unitary in ${C}_{b}\left(M\right)$  . Then the $C*$  -algebras ${C}_{0}\left(M/\alpha \right)\mathbb{\setminus }X\alpha ,{u}_{\beta }$  and ${C}_{0}\left(M/\beta \right)\mathbb{\setminus }X\beta ,u{*}_{\alpha }$  are Morita equivalent.
• Proof. Let $U:\mathbb{Z}×\mathbb{Z}\mathbb{⟶}\mathcal{U}\left({C}_{b}\left(M\right)\right)$  be given by $U\left(n,k\right)=\left\{\begin{array}{cc}1& \text{if either}n=0\text{or}k=0\text{.}\\ {\prod }_{i\in {S}_{k},j\in {S}_{n}}{\alpha }^{i}{\beta }^{j}\left(u\right),& \text{for}n,k>0\text{}\\ {\prod }_{i\in {S}_{k},j\in {S}_{n}}{\alpha }^{i}{\beta }^{j}\left({u}^{*}\right),& \text{for either}n\text{or}k<0\text{, and}nk\ne 0\text{},\end{array}$  where ${S}_{l}=\left\{0,1,...l-1\right\}$  if $l>0$  and ${S}_{l}=\left\{-1,-2,...l\right\}$  if $l<0$  .
Straightforward computations show that $U\left(m+n,k\right)=U\left(m,k\right){\beta }^{m}\left(U\left(n,k\right)\right)$  , and $U\left(n,k+l\right)=U\left(n,k\right){\alpha }^{k}\left(U\left(n,l\right)\right)$  . Consider the proper actions ${\gamma }^{\alpha }$  and ${\gamma }^{\beta }$  of $\mathbb{Z}$  on ${C}_{0}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}$  and ${C}_{0}\left(M\right){\mathbb{\setminus }}_{\alpha }\mathbb{Z}$  , respectively, given by: $\left[{\gamma }_{k}^{\alpha }\left(\phi \right)\right]\left(n\right)=U\left(n,k\right){\alpha }^{k}\left[\phi \left(n\right)\right]\text{and}\left[{\gamma }_{n}^{\beta }\left(\psi \right)\right]\left(k\right)={U}^{*}\left(n,k\right){\beta }^{n}\left[\psi \left(k\right)\right],$  for $\phi \in {C}_{c}\left(\mathbb{Z},{C}_{0}\left(M\right)\right)\subset {C}_{0}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}$  and $\psi \in {C}_{c}\left(\mathbb{Z},{C}_{0}\left(M\right)\right)\subset {C}_{0}\left(M\right){\mathbb{\setminus }}_{\alpha }\mathbb{Z}$  .
These two actions correspond, respectively, to ${\gamma }^{\alpha ,U}$  and ${\gamma }^{\beta ,{U}^{*}}$  in [Ab1,Propositions1.2and2.1. By virtue of [Ab1,Theorem2.12, the generalized fixed-point algebras, in the sense of [Rf5,Definition1.4, ${D}^{\alpha }$  and ${D}^{\beta }$  of ${C}_{0}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}$  and ${C}_{0}\left(M\right){\mathbb{\setminus }}_{\alpha }\mathbb{Z}$  under the actions ${\gamma }^{\alpha }$  and ${\gamma }^{\beta }$  , respectively, are Morita equivalent. The result will then be proved once we show that ${D}^{\alpha }\sim ={C}_{0}\left(M/\alpha \right)\mathbb{\setminus }X\alpha ,{u}_{\beta }$  and ${D}^{\beta }\sim ={C}_{0}\left(M/\beta \right)\mathbb{\setminus }X\beta ,u{*}_{\alpha }$  .
Recall from [Ab1,Proposition2.1that ${D}^{\alpha }$  is defined to be the closed span in $\mathcal{ℳ}\left({C}_{0}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}\right)$  of the set $\left\{{P}_{\alpha }\left({\phi }^{*}*\psi \right):\phi ,\psi \in {C}_{c}\left(\mathbb{Z}×M\right)\right\}$  , where ${P}_{\alpha }\left(\phi \right)\left(x,n\right)={\sum }_{k\in \mathbb{Z}}\left[{\gamma }_{k}^{\alpha }\left(\phi \right)\right]\left(x,n\right),$  for $\phi \in {C}_{c}\left(\mathbb{Z}×M\right)\subset {C}_{0}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z},x\in M$  , and $n\in \mathbb{Z}$  .
The $C*$  -algebra ${D}^{\alpha }$  can also be described ([Ab1,Proposition2.8) as the closure in $\mathcal{ℳ}\left({C}_{0}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}\right)$  of the $*$  -subalgebra ${C}^{\alpha }=\left\{F\in {C}_{c}\left(\mathbb{Z},{C}_{b}\left(M\right)\right):{\gamma }^{\alpha }\left(F\right)=F$  and ${\pi }_{\alpha }\left(\text{supp}F\left(n\right)\right)$  is precompact for all $n\in \mathbb{Z}\right\}$  , where ${\pi }_{\alpha }$  denotes the canonical projection ${\pi }_{\alpha }:M⟶M/\alpha$  .
Now, since ${C}^{\alpha }$  is contained in ${C}_{b}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}$  , which is closed in $\mathcal{ℳ}\left({C}_{0}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}\right)$  , so is ${D}^{\alpha }$  . Moreover, the $C*$  -algebra ${D}^{\alpha }$  is invariant under the dual action $\stackrel{^}{\beta }$  of $\mathbb{T}$  on ${C}_{b}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}$  :  $\begin{array}{cc}\left[{\gamma }^{\alpha }\left({\stackrel{^}{\beta }}_{z}F\right)\right]\left(n,x\right)& =U\left(n,1\right)\left(x\right)\left({\stackrel{^}{\beta }}_{z}\left(F\right)\right)\left(n,{\alpha }^{-1}x\right)\end{array}$
 $\begin{array}{cc}& =U\left(n,1\right)\left(x\right){z}^{n}F\left(n,{\alpha }^{-1}x\right)\end{array}$
 $\begin{array}{cc}& ={z}^{n}F\left(n,x\right)\end{array}$
 $\begin{array}{cc}& =\left({\stackrel{^}{\beta }}_{z}F\right)\left(n,x\right),\end{array}$
for $F\in {C}^{\alpha }$  , $x\in M$  , $n\in \mathbb{Z}$  , and $z\in \mathbb{T}.$  Besides, $\text{supp}\left({\stackrel{^}{\beta }}_{z}\left(F\right)\left(n\right)\right)=\text{supp}F\left(n\right)$  for all $n\in \mathbb{Z}$  , so ${\stackrel{^}{\beta }}_{z}\left(F\right)\in {C}^{\alpha }$  for all $z\in \mathbb{T}$  .
We next show that the action $\stackrel{^}{\beta }$  on ${D}^{\alpha }$  is semi-saturated. That is, that, as a $C*$  -algebra, ${D}^{\alpha }$  is generated by the fixed-point subalgebra ${D}_{0}$  and the first spectral subspace ${D}_{1}=\left\{d\in {D}^{\alpha }:{\stackrel{^}{\beta }}_{z}\left(d\right)=zd\forall z\in \mathbb{T}\right\}$  for the restriction of the dual action $\stackrel{^}{\beta }$  .
Since the maps ${P}_{i}:{D}^{\alpha }⟶{D}_{i}$  given by ${P}_{i}\left(a\right)={\int }_{\mathbb{T}}{z}^{-i}{\stackrel{^}{\beta }}_{z}\left(a\right)dz$  are surjective contractions, ${D}_{i}$  is the closure of ${P}_{i}\left({C}^{\alpha }\right)$  . Now, for $i=0,1$  , ${C}_{i}={C}^{\alpha }\cap {F}_{i}$  , and ${D}_{i}={D}^{\alpha }\cap {F}_{i}$  , where ${F}_{0}$  and ${F}_{1}$  are, respectively, the fixed-point subalgebra and the first spectral subspace of ${C}_{b}\left(M\right){\mathbb{\setminus }}_{\beta }\mathbb{Z}$  , which are known to be the ${\delta }_{i}$  -maps; that is, ${F}_{i}=\left\{F\in {C}_{c}\left(\mathbb{Z},{C}_{b}\left(M\right)\right):\text{supp}F=\left\{i\right\}\right\}$  .
Notice that ${C}^{\alpha }\cap {F}_{0}=\left\{f{\delta }_{0}:f\in {C}_{b}\left(M\right):{\pi }_{\alpha }\left(\text{supp}f\right)\text{is precompact and}f=\alpha \left(f\right)\right\}$  can be identified with ${C}_{c}\left(M/\alpha \right)$  via $f{\delta }_{0}↦\stackrel{~}{f}$  , where $\stackrel{~}{f}\circ {\pi }_{\alpha }=f$  , and that this map extends to a $*$  -isomorphism between ${D}_{0}$  and ${C}_{0}\left(M/\alpha \right)$  . Now, ${D}_{1}$  is a Hilbert $C*$  -bimodule over ${D}_{0}$  for
 $\begin{array}{c}\begin{array}{}\text{(1b)}& \begin{array}{}\text{(1a)}& \left(f{\delta }_{0}\right)*\left(g{\delta }_{1}\right)=\left(fg\right){\delta }_{1}\text{and}\left(g{\delta }_{1}\right)*\left(f{\delta }_{0}\right)=\left(g\beta \left(f\right)\right){\delta }_{1},〈f{\delta }_{1},g{\delta }_{1}{〉}_{L}=\left(f{\delta }_{1}\right)*\left(g{\delta }_{1}{\right)}^{*}=\left(f\overline{g}\right){\delta }_{0},〈f{\delta }_{1},g{\delta }_{1}{〉}_{R}=\left(f{\delta }_{1}{\right)}^{*}*\left(g{\delta }_{1}\right)=\left({\beta }^{-1}\left(\overline{f}g\right)\right){\delta }_{0}.\end{array}\end{array}\end{array}$ (1c)
Notice that ${D}_{1}$  is full on the left (and on the right, by a similar argument) as a Hilbert $C*$  -bimodule over ${C}_{0}\left(M/\alpha \right)$  . For $<{D}_{1},{D}_{1}{>}_{L}$  , the closed linear span in ${C}_{0}\left(M/\alpha \right)\sim ={D}_{0}$  of the set $\left\{〈f{\delta }_{1},g{\delta }_{1}{〉}_{L}:f{\delta }_{1},g{\delta }_{1}\in {D}_{1}\right\}$  , is a closed ideal of ${C}_{0}\left(M/\alpha \right)$  . Therefore, unless $<{D}_{1},{D}_{1}{>}_{L}={C}_{0}\left(M/\alpha \right)$  , there exists ${x}_{0}\in M$  such that $f\left({x}_{0}\right)=0$  for all $f{\delta }_{1}\in {D}_{1}$  .
Now, given ${x}_{0}\in M$  , we can choose ([Rf2,Situation10) a neighborhood $U$  of ${x}_{0}$  such that $U\cap {\alpha }^{k}\left(U\right)=\varnothing$  for $k\ne 0$  . Let $g\in {C}_{c}\left(M{\right)}^{+}$  be such that $\text{supp}g\subset U$  and $g\left({x}_{0}\right)=1$  .
Then
$\left[{P}_{\alpha }\left(\left({g}^{1/2}{\delta }_{0}{\right)}^{*}*\left({g}^{1/2}{\delta }_{1}\right)\right)\right]\left(x,n\right)==\left({P}_{\alpha }\left(g{\delta }_{1}\right)\right)\left(x,n\right)=\left({\sum }_{k}U\left(1,k\right)\left(x\right)g\left({\alpha }^{-k}\left(x\right)\right)\right){\delta }_{1}\left(n\right),$
so ${P}_{\alpha }\left(\left({g}^{1/2}{\delta }_{0}{\right)}^{*}*\left({g}^{1/2}{\delta }_{1}\right)\right)\in {D}_{1}$  , and equals 1 at $\left({x}_{0},1\right)$  .
In order to prove that ${C}^{\alpha }\subset {C}^{*}\left({D}_{0},{D}_{1}\right)$  , it suffices to show that $f{\delta }_{k}\in {C}^{*}\left({D}_{0},{D}_{1}\right)$  for $f{\delta }_{k}\in {C}^{\alpha }$  , $k\in \mathbb{Z}$  ; since ${C}^{\alpha }$  is closed under involution, we may assume that $k\ge 0$  . We show this fact, which clearly holds for $k=0$  and $k=1$  , by induction on $k$  .
If $f{\delta }_{k}\in {C}^{\alpha }$  and $\epsilon >0$  , since ${\pi }_{\alpha }\left(\text{supp}f\right)$  is precompact in $M/\alpha$  , and ${D}_{1}$  is full over ${C}_{0}\left(M/\alpha \right)$  , we can find ${\phi }_{i},{\psi }_{i}\in {D}_{1},i=1,...,p,$  such that $\parallel {\sum }_{i}\left({\phi }_{i}*{\psi }_{i}^{*}\right)*f{\delta }_{1}-f{\delta }_{1}{\parallel }_{{D}^{\alpha }}=\parallel {\sum }_{i}〈{\phi }_{i},{\psi }_{i}{〉}_{L}f-f{\parallel }_{{C}_{b}\left(M\right)}<\epsilon$  .
Now, since ${\phi }_{i}$  and ${\psi }_{i}^{*}*f$  belong to ${C}^{*}\left({D}_{0},{D}_{1}\right)$  for $i=1,...,p$  , so does $f$  . This shows that ${D}^{\alpha }={C}^{*}\left({D}_{0},{D}_{1}\right)$  and, consequently, by [AEE,Theorem3.1, that ${D}^{\alpha }\sim ={D}_{0}\mathbb{\setminus }{D}_{1}$  .
It only remains to notice now that ${D}_{0}\mathbb{\setminus }{D}_{1}\sim ={C}_{0}\left(M/\alpha \right)\mathbb{\setminus }{X}_{\beta }^{\alpha ,u}$  . As noticed above, ${D}_{0}$  is isomorphic to ${C}_{0}\left(M/\alpha \right)$  . On the other hand, the map $f{\delta }_{1}↦f$  takes ${C}^{\alpha }\cap {F}_{1}$  to ${X}_{\beta }^{\alpha ,u}$  . By keeping track of the formulae in  1 , one easily checks that that map is an isometry, so it extends to an isometry from ${D}_{1}$  to ${X}_{\beta }^{\alpha ,u}$  , which is onto because its image contains the dense set: ${X}_{0}^{\alpha ,u}=\left\{f\in {X}_{\beta }^{\alpha ,u}\text{: the map}x↦|f\left(x\right)|\text{is compactly supported on}M/\alpha \text{}\right\}.$  (Notice that ${X}_{0}^{\alpha ,u}$  is dense in ${X}_{\beta }^{\alpha ,u}$  , because, if $\left\{{e}_{\lambda }\right\}$  is an approximate identity for ${C}_{c}\left(M/\alpha \right)$  , then ${e}_{\lambda }f$  converges to $f$  for all $f\in {X}_{\beta }^{\alpha ,u}$  ). This shows that ${D}^{\alpha }$  is isomorphic to ${C}_{0}\left(M/\alpha \right)\mathbb{\setminus }{X}_{\beta }^{\alpha ,u}$  . Analogously, ${D}^{\beta }$  is isomorphic to ${C}_{0}\left(M/\beta \right)\mathbb{\setminus }{X}_{\alpha }^{\beta ,{u}^{*}}$  .

2 Morita equivalence for quantum Heisenberg manifolds

In [AEE(see also [AE,2) the quantum Heisenberg manifold $Dc\mu \nu$  was shown to be the crossed product of $C\left({\mathbb{T}}^{2}\right)$  , the $C*$  -algebra of continuous functions on the torus, by the Hilbert $C*$  -bimodule ${M}_{{\alpha }_{\mu \nu }}^{c}$  , where ${\alpha }_{\mu \nu }\left(x,y\right)=\left(x+2\mu ,y+2\nu \right)$  , and ${M}^{c}=\left\{f\in {C}_{b}\left(\mathbb{R}×\mathbb{T}\right):f\left(x+1,y\right)={e}^{-2\pi icy}f\left(x,y\right)\right\}$  is the Hilbert $C*$  -bimodule obtained by letting $C\left({\mathbb{T}}^{2}\right)$  act by pointwise product, and by defining the inner products $〈f,g{〉}_{L}=f\overline{g}$  , $〈f,g{〉}_{R}=\overline{f}g$  .
Remark 2.1. The $C*$  -algebras $Dc\mu \nu$  and $Dc{\mu }^{\prime }{\nu }^{\prime }$  are isomorphic when the projections of $\left(2\mu ,2\nu \right)$  and $\left(2{\mu }^{\prime },2{\nu }^{\prime }\right)$  on the torus are in the same orbit under the usual action of $\text{GL}2\left(\mathbb{Z}\right)\text{}$  ([AE,Theorem2.2, see also [Ab2,Remark3.3).
Proposition 2.2. Let $\mu \ne 0$  . Then $Dc\mu \nu$  and ${D}_{\frac{1}{4\mu },\frac{\nu }{2\mu }}^{c}$  are Morita equivalent.
• Proof. We follow the lines of [Rf3,1.1and apply Proposition  1.2 to the following setting: $\alpha$  and $\beta$  consist of translation on $\mathbb{R}×\mathbb{T}$  by $\left(\frac{1}{2\mu },0\right)$  and $\left(1,2\nu \right)$  , respectively, and $u\in {C}_{b}\left(\mathbb{R}×\mathbb{T}\right)$  is given by $u\left(x,y\right)=e\left(-cy\right)$  , where $\mathbb{T}$  is viewed as $\mathbb{R}/\mathbb{Z}$  and, for a real number $h$  , $e\left(h\right)={e}^{2\pi ih}$  .
Then, by Proposition  1.2 , $C\left(\left(\mathbb{R}×\mathbb{T}\right)/\alpha \right)\mathbb{\setminus }{X}_{\beta }^{\alpha ,u}$  and $C\left(\left(\mathbb{R}×\mathbb{T}\right)/\beta \right)\mathbb{\setminus }{X}_{\alpha }^{\beta ,{u}^{*}}$  are Morita equivalent, where
${X}^{\alpha ,u}=\left\{F\in {C}_{b}\left(\mathbb{R}×\mathbb{T}\right):F\left(x-\frac{1}{2\mu },y\right)=e\left(cy\right)F\left(x,y\right)\right\}\text{and}{X}^{\beta ,{u}^{*}}=\left\{F\in {C}_{b}\left(\mathbb{R}×\mathbb{T}\right):F\left(x-1,y-2\nu \right)=e\left(-cy\right)F\left(x,y\right)\right\}=\left\{F\in {C}_{b}\left(\mathbb{R}×\mathbb{T}\right):F\left(x+1,y+2\nu \right)=e\left(c\left(y+2\nu \right)\right)F\left(x,y\right)\right\}.$
Let ${H}_{\alpha }:C\left({\mathbb{T}}^{2}\right)⟶C\left(\left(\mathbb{R}×\mathbb{T}\right)/\alpha \right)$  and ${H}_{\beta }:C\left({\mathbb{T}}^{2}\right)⟶C\left(\left(\mathbb{R}×\mathbb{T}\right)/\beta \right)$  be the isomorphisms given by: $\left({H}_{\alpha }\phi \right)\left(x,y\right)=\phi \left(2\mu x,y\right),\left({H}_{\beta }\phi \right)\left(x,y\right)=\phi \left(x,2\nu x-y\right),$  and, for $\left({\mu }^{\prime },{\nu }^{\prime }\right)=\left(\frac{1}{4\mu },\frac{\nu }{2\mu }\right)$  , set ${J}_{\alpha }:{M}_{{\alpha }_{\mu \nu }}^{c}⟶{X}_{\beta }^{\alpha ,u}\text{and}{J}_{\beta }:{M}_{{\alpha }_{{\mu }^{\prime }{\nu }^{\prime }}}^{c}⟶{X}_{\alpha }^{\beta ,{u}^{*}},$  $\left({J}_{\alpha }f\right)\left(x,y\right)=f\left(2\mu x,y\right),\left({J}_{\beta }f\right)\left(x,y\right)=e\left(cx\left(x+1\right)\nu \right)f\left(x,2\nu x-y\right).$  Notice that $\left({J}_{\alpha }f\right)\left(x-\frac{1}{2\mu },y\right)=f\left(2\mu x-1,y\right)=e\left(cy\right)\left({J}_{\alpha }f\right)\left(x,y\right),$  and  $\begin{array}{cc}\left({J}_{\beta }f\right)\left(x+1,y+2\nu \right)& =e\left(c\left(x+1\right)\left(x+2\right)\nu \right)f\left(x+1,2\nu x-y\right)\end{array}$
 $\begin{array}{cc}& =e\left(c\left(x+1\right)\left(x+2\right)\nu \right)e\left(-c\left(2\nu x-y\right)\right)f\left(x,2\nu x-y\right)\end{array}$
 $\begin{array}{cc}& =e\left(c\left(y+2\nu \right)\right)\left({J}_{\beta }f\right)\left(x,y\right),\end{array}$
so the definitions make sense.
For $i=\alpha ,\beta$  , it is easily checked that ${J}_{i}$  is a bijection and that, for $\phi \in C\left({\mathbb{T}}^{2}\right)$  , $f,g\in {M}^{c}$  : ${J}_{i}\left(\phi \cdot f\right)={H}_{i}\left(\phi \right)\cdot {J}_{i}\left(f\right),{J}_{i}\left(f\cdot \phi \right)={J}_{i}\left(f\right)\cdot {H}_{i}\left(\phi \right),$  $〈{J}_{i}f,{J}_{i}g{〉}_{L}={H}_{i}\left(〈f,g{〉}_{L}\right),〈{J}_{i}f,{J}_{i}g{〉}_{R}={H}_{i}\left(〈f,g{〉}_{R}\right).$  This shows that $Dc\mu \nu =C\left({\mathbb{T}}^{2}\right)\mathbb{\setminus }{M}_{{\alpha }_{\mu \nu }}^{c}$  and $Dc{\mu }^{\prime }{\nu }^{\prime }=C\left({\mathbb{T}}^{2}\right)\mathbb{\setminus }{M}_{{\alpha }_{{\mu }^{\prime }{\nu }^{\prime }}}^{c}$  are isomorphic, respectively, to $C\left(\left(\mathbb{R}×\mathbb{T}\right)/\alpha \right)\mathbb{\setminus }{X}_{\beta }^{\alpha ,u}$  and $C\left(\left(\mathbb{R}×\mathbb{T}\right)/\beta \right)\mathbb{\setminus }{X}_{\alpha }^{\beta ,{u}^{*}}$  , and they are, consequently, Morita equivalent to each other.
Corollary 2.3. Let $\mu \notin \mathbb{Q}$  , and let $A=\left(\begin{array}{cc}a& b\\ c& d\end{array}\right)\in \text{GL}2\left(\mathbb{Z}\right)\text{}$  . If $2{\mu }^{\prime }=\frac{2a\mu +b}{2c\mu +d}\text{and}2{\nu }^{\prime }=\frac{2\nu }{2c\mu +d},$  then the quantum Heisenberg manifolds $Dc\mu \nu$  and $Dc{\mu }^{\prime }{\nu }^{\prime }$  are Morita equivalent.
• Proof. It suffices to check the statement for ${A}_{1}=\left(\begin{array}{cc}1& 1\\ 0& 1\end{array}\right)$  and ${A}_{2}=\left(\begin{array}{cc}0& 1\\ 1& 0\end{array}\right)$  , since ${A}_{1}$  and ${A}_{2}$  generate $\text{GL}2\left(\mathbb{Z}\right)\text{}$  ([Ku,AppendixB), and $\left(\mu ,\nu \right)↦\left({\mu }^{\prime },{\nu }^{\prime }\right)$  defines an action of $\text{GL}2\left(\mathbb{Z}\right)\text{}$  on $\left(\mathbb{R}\\mathbb{Q}\right)×\mathbb{R}$  . For $A={A}_{1}$  we get isomorphic $C*$  -algebras by Remark  2.1 . For $A={A}_{2}$  , we get $\left({\mu }^{\prime },{\nu }^{\prime }\right)=\left(\frac{1}{4\mu },\frac{\nu }{2\mu }\right)$  , and the result follows from Proposition  2.2 .
Proposition 2.4. Let $\left\{1,\mu ,\nu \right\}$  be linearly independent over $\mathbb{Q}$  , and let $A=\left(\begin{array}{ccc}a& b& c\\ d& e& f\\ g& h& i\end{array}\right)\in \text{GL}3\left(\mathbb{Z}\right)\text{}$  . If $2{\mu }^{\prime }=\frac{2a\mu +2b\nu +c}{2g\mu +2h\nu +i}\text{and}2{\nu }^{\prime }=\frac{2d\mu +2e\nu +f}{2g\mu +2h\nu +i},$  then the quantum Heisenberg manifolds $Dc\mu \nu$  and $Dc{\mu }^{\prime }{\nu }^{\prime }$  are Morita equivalent.
• Proof. As in the proof of Theorem 1.7 in [Pa2, $A={A}_{1}{A}_{2}{A}_{3}$  , where ${A}_{1}=\left(\begin{array}{ccc}A& B& C\\ D& E& F\\ 0& 0& 1\end{array}\right),{A}_{2}\left(\begin{array}{ccc}G& 0& H\\ 0& 1& 0\\ I& 0& J\end{array}\right),{A}_{3}=\left(\begin{array}{ccc}K& L& 0\\ M& N& 0\\ 0& 0& 1\end{array}\right),$  and ${A}_{i}\in \text{GL}3\left(\mathbb{Z}\right)\text{}$  , for $i=1,2,3$  .
Since the map $\left(\mu ,\nu \right)↦\left({\mu }^{\prime },{\nu }^{\prime }\right)$  defines an action of $\text{GL}3\left(\mathbb{Z}\right)\text{}$  on the set $\left\{\left(\mu ,\nu \right)\in {\mathbb{R}}^{2}:\left\{1,\mu ,\nu \right\}\text{is linearly independent over}\mathbb{Q}\text{}\right\}$  , it suffices to check the statement for ${A}_{i}$  , $i=1,2,3.$  For $A={A}_{1}$  and $A={A}_{3}$  the $C*$  -algebras $Dc\mu \nu$  and $Dc{\mu }^{\prime }{\nu }^{\prime }$  are isomorphic by Remark  2.1 . Thus it suffices to show the result for $A={A}_{2}$  . The map $\left(\begin{array}{cc}G& H\\ I& J\end{array}\right)↦\left(\begin{array}{ccc}G& 0& H\\ 0& 1& 0\\ I& 0& J\end{array}\right)$  is a group homomorphism from $\text{GL}2\left(\mathbb{Z}\right)\text{}$  into $\text{GL}3\left(\mathbb{Z}\right)\text{}$  , and $\text{GL}2\left(\mathbb{Z}\right)\text{}$  is generated by $\left(\begin{array}{cc}1& 1\\ 0& 1\end{array}\right)$  and $\left(\begin{array}{cc}0& 1\\ 1& 0\end{array}\right)$  , which implies that we only need to prove the statement for ${A}_{1}=\left(\begin{array}{ccc}1& 0& 1\\ 0& 1& 0\\ 0& 0& 1\end{array}\right)$  and ${A}_{2}=\left(\begin{array}{ccc}0& 0& 1\\ 0& 1& 0\\ 1& 0& 0\end{array}\right)$  . For ${A}_{1}$  we get $2{\mu }^{\prime }=2\mu +1$  , $2{\nu }^{\prime }=2\nu$  , so $Dc\mu \nu$  and $Dc{\mu }^{\prime }{\nu }^{\prime }$  are isomorphic by Remark  2.1 . Proposition  2.2 takes care of the case $A={A}_{2}$  , since then we have $\left({\mu }^{\prime },{\nu }^{\prime }\right)=\left(\frac{1}{4\mu },\frac{\nu }{2\mu }\right)$  .
Notation 2.5. We denote by $\text{}G\mu \nu \text{}$  the subgroup of $\mathbb{R}$  generated by $\left\{1,2\mu ,2\nu \right\}$  . It was shown in [Ab2,Theorem3.16that the homomorphism induced on ${K}_{0}\left(Dc\mu \nu \right)$  by any tracial state on $Dc\mu \nu$  has range $\text{}G\mu \nu \text{}$  .
Remark 2.6. If rank $\text{}G\mu \nu \text{}=2$  , then there exist an irrational number ${\nu }^{\prime }$  and integers $p,q\in \mathbb{Z},p\ne 0$  , $\left(p,q\right)=1$  , such that $Dc\mu \nu$  and ${D}_{\frac{p}{2q},{\nu }^{\prime }}^{c}$  are isomorphic.
• Proof. We proceed as in [Pa1,Proposition1.5. Let ${\mu }_{0}=2\mu$  , ${\nu }_{0}=2\nu$  . Since the group generated by $\left\{1,{\mu }_{0},{\nu }_{0}\right\}$  has rank 2, either ${\mu }_{0}$  or ${\nu }_{0}$  is irrational.
We may assume that ${\nu }_{0}$  is irrational, because, by Remark  2.1 , $Dc\mu \nu$  and ${D}_{\nu \mu }^{c}$  are isomorphic. Besides, there exist $M,N,P\in \mathbb{Z}$  , with $N\ne 0$  such that $M+N{\mu }_{0}+P{\nu }_{0}=0$  , so we have ${\mu }_{0}=\frac{k}{l}{\nu }_{0}+\frac{m}{n}$  , with $\left(k,l\right)=1$  . If $k=0$  , then ${\mu }_{0}\in \mathbb{Q}$  , and we are done. Otherwise take $a,b\in \mathbb{Z}$  such that $ak+bl=1$  , so that $\left(\begin{array}{cc}-l& k\\ a& b\end{array}\right)\in \text{GL}2\left(\mathbb{Z}\right)\text{}$  , and set $\left({\mu }_{0}^{\prime },{\nu }_{0}^{\prime }\right)=\left(\begin{array}{cc}-l& k\\ a& b\end{array}\right)\left({\mu }_{0},{\nu }_{0}\right).$  Then ${\mu }_{0}^{\prime }=-l\left(\frac{k}{l}{\nu }_{0}+\frac{m}{n}\right)+k{\nu }_{0}=\frac{-lm}{n}\in \mathbb{Q}.$  and ${\nu }_{0}^{\prime }=a\left(\frac{k}{l}{\nu }_{0}+\frac{m}{n}\right)+b{\nu }_{0}=\frac{1}{l}{\nu }_{0}+\frac{am}{n}\notin \mathbb{Q},$  We now take ${\nu }^{\prime }={\nu }_{0}^{\prime }/2$  and $p/q={\mu }_{0}^{\prime }$  , in lowest terms. By Remark  2.1  $Dc\mu \nu$  and ${D}_{\frac{p}{2q},{\nu }^{\prime }}^{c}$  are isomorphic.
Proposition 2.7. Let $p$  and $q$  be non-zero integers such that $\left(p,q\right)=1$  , and let $\nu \in \mathbb{R}$  . Then ${D}_{\frac{p}{2q},\nu }^{c}$  is Morita equivalent to ${D}_{0,q\nu }^{c}$  .
• Proof. By Remark  2.1 we may assume that $p$  and $q$  are positive. By applying Proposition  2.2 to $\left(\mu ,\nu \right)=\left(q/2,\nu \right)$  , we get that ${D}_{0,\nu }^{c}\sim ={D}_{q/2,\nu }^{c}$  is Morita equivalent to ${D}_{\frac{1}{2q},\frac{\nu }{q}}^{c}$  , thus proving the proposition for $p=1$  . For $p>1$  , let ${r}_{0}=q$  , ${r}_{1}=p$  , and, if ${r}_{i+1}\ne 1$  , define ${r}_{i+2}$  by ${r}_{i}={m}_{i+1}{r}_{i+1}+{r}_{i+2}$  , where $0\le {r}_{i+2}<{r}_{i+1}$  , and ${m}_{i+1}\in \mathbb{Z}$  .
Actually, ${r}_{i+2}>0$  : otherwise ${r}_{i+1}$  divides ${r}_{i}$  , and it follows that ${r}_{i+1}$  divides ${r}_{j}$  for all $j\le i$  . In particular, ${r}_{i+1}$  divides both $p$  and $q$  , which contradicts the fact that ${r}_{i+1}\ne 1$  . Now, since ${r}_{i+1}<{r}_{i}$  , there is an index ${i}_{0}$  for which ${r}_{{i}_{0}}=1$  .
On the other hand, it follows from Proposition  2.2 that, for any real number $\kappa$  , ${D}_{\frac{{r}_{i}}{2{r}_{i-1}},\kappa }^{c}$  is Morita equivalent to ${D}_{\frac{{r}_{i-1}}{2{r}_{i}},\kappa \frac{{r}_{i-1}}{{r}_{i}}}^{c}$  , which in turn is isomorphic to ${D}_{\frac{{r}_{i+1}}{2{r}_{i}},\kappa \frac{{r}_{i-1}}{{r}_{i}}}^{c}$  .
Thus ${D}_{\frac{p}{2q},\nu }^{c}={D}_{\frac{{r}_{1}}{2{r}_{0}},\nu }^{c}$  is Morita equivalent to ${D}_{\frac{{r}_{j}}{2{r}_{j-1}},\frac{q\nu }{{r}_{j-1}}}^{c}$  for any $j\le {i}_{0}$  . In particular, for $j={i}_{0}$  , we have that ${D}_{\frac{p}{2q},\nu }^{c}$  is Morita equivalent to ${D}_{\frac{1}{2{r}_{{i}_{0}-1}},\frac{\nu q}{{r}_{{i}_{0}-1}}}^{c}$  , which, as shown above, is Morita equivalent to ${D}_{0,\nu q}^{c}$  .
Theorem 2.8. Two quantum Heisenberg manifolds $Dc\mu \nu$  , ${D}_{{\mu }^{\prime }{\nu }^{\prime }}^{{c}^{\prime }}$  are Morita equivalent if and only if $c={c}^{\prime }$  and there exists a positive real number $r$  such that $\mathbb{Z}+2\mu \mathbb{Z}+2\nu \mathbb{Z}=r\left(\mathbb{Z}+2{\mu }^{\prime }\mathbb{Z}+2{\nu }^{\prime }\mathbb{Z}\right).$  In particular, the rank of the free abelian group $\text{}G\mu \nu \text{}=\mathbb{Z}+2\mu \mathbb{Z}+2\nu \mathbb{Z}$  is the same for Morita equivalent quantum Heisenberg manifolds, and:
• (1) If rank $\text{}G\mu \nu \text{}=1=$  rank $\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}$  , then $Dc\mu \nu$  is Morita equivalent to ${D}_{{\mu }^{\prime },{\nu }^{\prime }}^{c}$  . In particular, $Dc\mu \nu$  is Morita equivalent to the commutative Heisenberg manifold ${D}_{0,0}^{c}$  .
• (2) If rank $\text{}G\mu \nu \text{}=2=$  rank $\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}$  , let $\left\{\alpha ,\frac{p}{q}\right\}$  and $\left\{{\alpha }^{\prime },\frac{{p}^{\prime }}{{q}^{\prime }}\right\}$  be bases of $\text{}G\mu \nu \text{}$  and $\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}$  , respectively, where $\alpha$  and ${\alpha }^{\prime }$  are irrational numbers and $p,{p}^{\prime },q,{q}^{\prime }\in \mathbb{Z}$  , $\left(p,q\right)=\left({p}^{\prime }{q}^{\prime }\right)=1$  . Then ${D}_{\mu ,\nu }^{c}$  and ${D}_{{\mu }^{\prime },{\nu }^{\prime }}^{c}$  are Morita equivalent if and only if there exists $\left(\begin{array}{cc}a& b\\ c& d\end{array}\right)\in \text{GL}2\left(\mathbb{Z}\right)\text{}$  such that ${q}^{\prime }{\alpha }^{\prime }=\frac{aq{\alpha }^{\prime }+b}{cq{\alpha }^{\prime }+d}.$  In particular, ${D}_{\frac{p}{2q},\nu }^{c}$  is Morita equivalent to ${D}_{0,q\nu }^{c}$  .
• (3) If rank $\text{}G\mu \nu \text{}=3=$  rank $\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}$  , then $Dc\mu \nu$  and $Dc{\mu }^{\prime }{\nu }^{\prime }$  are Morita equivalent if and only if there exists $\left(\begin{array}{ccc}a& b& c\\ d& e& f\\ g& h& i\end{array}\right)\in \text{GL}3\left(\mathbb{Z}\right)\text{}$  such that $2{\mu }^{\prime }=\frac{2a\mu +2b\nu +c}{2g\mu +2h\nu +i}\text{and}2{\nu }^{\prime }=\frac{2d\mu +2e\nu +f}{2g\mu +2h\nu +i}.$
• Proof. It was shown in [Ab1,3.4that ${K}_{0}\left(Dc\mu \nu \right)={\mathbb{Z}}^{3}\oplus {\mathbb{Z}}_{c}$  , which implies that $Dc\mu \nu$  and ${D}_{{\mu }^{\prime }{\nu }^{\prime }}^{{c}^{\prime }}$  are not Morita equivalent for $c\ne {c}^{\prime }$  .
Besides ([Ab2,Theorem3.16), all tracial states on $Dc\mu \nu$  induce the same homomorphism on ${K}_{0}\left(Dc\mu \nu \right)$  , whose range is the group ${G}_{\mu \nu }=2\mu \mathbb{Z}+2\nu \mathbb{Z}+\mathbb{Z}$  .
Since ([Rf1,2.2) there is a bijection between finite traces of Morita equivalent unital $C*$  -algebras we must have ${G}_{\mu \nu }=r{G}_{{\mu }^{\prime }{\nu }^{\prime }}$  for some positive real number $r$  when $Dc\mu \nu$  and $Dc{\mu }^{\prime }{\nu }^{\prime }$  are Morita equivalent. An immediate consequence of this fact is that the rank of ${G}_{\mu \nu }$  is invariant under Morita equivalence.
If rank $\text{}G\mu \nu \text{}=1$  , then, by [Ab2,Remark3.5, $Dc\mu \nu$  is isomorphic to ${D}_{0,\frac{1}{2p}}^{c}$  for some non-zero integer $p$  , so $Dc\mu \nu$  is isomorphic to ${D}_{\frac{1}{2p},0}^{c}$  by Remark  2.1 .
Now, by Proposition  2.7 , ${D}_{\frac{1}{2p},0}^{c}$  is Morita equivalent ${D}_{0,0}^{c}$  .
If rank $\text{}G\mu \nu \text{}=2=$  rank $\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}$  and ${G}_{\mu \nu }=r{G}_{{\mu }^{\prime }{\nu }^{\prime }}$  for some positive $r$  , let $\left\{\alpha ,\frac{p}{q}\right\}$  and $\left\{{\alpha }^{\prime },\frac{{p}^{\prime }}{{q}^{\prime }}\right\}$  be bases of $\text{}G\mu \nu \text{}$  and $\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}$  , respectively, where $\alpha ,{\alpha }^{\prime }$  are irrational numbers, and $p,{p}^{\prime },q,{q}^{\prime }$  are integers, with $\left(p,q\right)=\left({p}^{\prime },{q}^{\prime }\right)=1$  .
Since $\mathbb{Z}\subset \text{}G\mu \nu \text{}\left(\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}\right)$  we have that $p\left({p}^{\prime }\right)=±1$  and, by Remark  2.1 , we may assume $p={p}^{\prime }=1$  . Then we have that $\alpha \mathbb{Z}+1/q\mathbb{Z}=r\left({\alpha }^{\prime }\mathbb{Z}+1/{q}^{\prime }\mathbb{Z}\right)$  , which implies that $\alpha q\mathbb{Z}+\mathbb{Z}=\left(rq/{q}^{\prime }\right)\left({\alpha }^{\prime }{q}^{\prime }\mathbb{Z}+\mathbb{Z}\right)$  . A standard argument shows that $q\alpha =\frac{a{q}^{\prime }{\alpha }^{\prime }+b}{c{q}^{\prime }{\alpha }^{\prime }+d}\text{for some}\left(\begin{array}{cc}a& b\\ c& d\end{array}\right)\in \text{GL}2\left(\mathbb{Z}\right)\text{}.$  Therefore ${D}_{\frac{q\alpha }{2},0}^{c}$  and ${D}_{\frac{{q}^{\prime }{\alpha }^{\prime }}{2},0}^{c}$  are Morita equivalent by Corollary  2.3 . On the other hand, by Remark  2.6 , $Dc\mu \nu$  and ${D}_{{\mu }^{\prime },{\nu }^{\prime }}^{c}$  are isomorphic, respectively, to ${D}_{\frac{m}{2n},\beta }^{c}$  and ${D}_{\frac{{m}^{\prime }}{2{n}^{\prime }},{\beta }^{\prime }}^{c}$  , for some irrational numbers $\beta$  and ${\beta }^{\prime }$  and integers $m,{m}^{\prime },n,{n}^{\prime }$  such that $\left(m,n\right)=\left({m}^{\prime },{n}^{\prime }\right)=1$  . Therefore $\left\{2\beta ,\frac{1}{n}\right\}$  and $\left\{2{\beta }^{\prime },\frac{1}{{n}^{\prime }}\right\}$  are bases of $\text{}G\mu \nu \text{}$  and ${G}_{{\mu }^{\prime }{\nu }^{\prime }}$  , respectively, and it follows from the argument above that ${D}_{n\beta ,0}^{c}$  and ${D}_{{n}^{\prime }{\beta }^{\prime },0}^{c}$  are Morita equivalent. It only remains to notice now that, by Proposition  2.7 and Remark  2.1 , ${D}_{n\beta ,0}^{c}$  and ${D}_{{n}^{\prime }{\beta }^{\prime },0}^{c}$  are Morita equivalent to $Dc\mu \nu$  and ${D}_{{\mu }^{\prime }{\nu }^{\prime }}^{c}$  , respectively.
Finally, if rank $\text{}G\mu \nu \text{}=3=$  rank $\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}$  and $\text{}G\mu \nu \text{}=r\text{}G{\mu }^{\prime }{\nu }^{\prime }\text{}$  for some positive $r$  , then let $A=\left(\begin{array}{ccc}a& b& c\\ d& e& f\\ g& h& i\end{array}\right)\in \text{GL}3\left(\mathbb{Z}\right)\text{}$  be the transpose of the matrix that changes coordinates between the bases $\left\{2r{\mu }^{\prime },2r{\nu }^{\prime },r\right\}$  and $\left\{2\mu ,2\nu ,1\right\}$  of $\text{}G\mu \nu \text{}$  . Then $2{\mu }^{\prime }=\frac{2a\mu +2b\nu +c}{2g\mu +2h\nu +i}\text{and}2{\nu }^{\prime }=\frac{2d\mu +2e\nu +f}{2g\mu +2h\nu +i},$  which implies, by Proposition  2.4 , that $Dc\mu \nu$  and $Dc{\mu }^{\prime }{\nu }^{\prime }$  are Morita equivalent.
References

1. Abadie, B. [1995], Generalized fixed-point algebras of certain actions on crossed-products, Pacific J. Math. 171, 1-21.
2. Abadie, B. [2000], The range of traces on quantum Heisenberg manifolds, Trans. Amer. Math. Soc. 352, 5767-5780.
3. Abadie, B.; Exel, R. [1997], Hilbert $C*$  -bimodules over commutative $C*$  -algebras and an isomorphism condition for quantum Heisenberg manifolds, Rev. Math. Phys. 9, 411-423.
4. Abadie, B.; Eilers, S.; Exel, R. [1998], Morita equivalence for crossed products by Hilbert $C*$  -bimodules, Trans. Amer. Math. Soc. 350, 3043-3054.
5. Elliott, G.; Gong, G. [1996], On the classification of $C*$  -algebras of real rank zero, II, Ann. of Math. 144, 497-610.
6. Kurosh, A.G. [1960], The theory of groups, Vol. 2, Chelsea Publishing Company, Second edition.
7. Packer, J. [1987], C*-algebras generated by projective representations of the discrete Heisenberg group, J. Operator Theory 18, 41-66.
8. Packer, J. [1988], Strong Morita equivalence for Heisenberg C*-algebras and the positive cones of their ${K}_{0}$  -groups, Canad. J. Math. XL, 833-864.
9. Pimsner, M.V. [1997], A class of $C*$  -algebras generalizing both Cuntz-Krieger algebras and crossed products by $\mathbb{Z}$  , Fields Inst. Comm. 12, AMS, 189-212.
10. Rieffel, M. [1981], C*-algebras associated with irrational rotations, Pacific J. Math. 93, 415-429.
11. Rieffel, M. [1982], Applications of strong Morita equivalence to transformation C*-algebras, Proc. Symp. Pure Math. 38 (Part 1), 299-310.
12. Rieffel, M. [1983], The cancellation theorem for projective modules over irrational rotation C*-algebras, Proc. London Math. Soc. 3 (No 47), 285-302.
13. Rieffel, M. [1989], Deformation quantization of Heisenberg manifolds, Comm. Math. Phys. 122, 531-562.
14. Rieffel, M. [1990], Proper actions of groups on C*-algebras, Mappings of operator algebras, Proc. Japan-US joint seminar, Birkhäuser, 141-182.

Centro de Matematicas. Facultad de Ciencias. Igua 4225, CP 11 400, Montevideo, Uruguay. E-mail address : abadie@cmat.edu.uy