<?xml version="1.0" encoding="utf-8"?><?xml-stylesheet type="text/xsl" href="math.xslt"?><html xmlns="http://www.w3.org/1999/xhtml"><head><link rel="stylesheet" type="text/css" href="style.css"/><title>Non-Noether symmetries and their influence on phase space geometry</title></head><body><h1>Non-Noether symmetries and their influence on phase space geometry</h1><div class="author">George Chavchanidze</div><div class="affiliation">Department of Theoretical Physics,A. Razmadze Institute of Mathematics,1 Aleksidze Street, Tbilisi 0193, Georgia</div><div class="abstract"><strong class="cap">abstract. </strong>We disscuss some geometric aspects of the concept of non-Noether symmetry.It is shown that in regular Hamiltonian systems such a symmetry canonically leadsto a Lax pair on the algebra of linear operators on cotangent bundle over the phase space.Correspondence between the non-Noether symmetries and other wide spread geometricmethods of generating conservation laws such as bi-Hamiltonian formalism,bidifferential calculi and Frölicher-Nijenhuis geometry is considered.It is proved that the integrals of motion associated with thecontinuous non-Noether symmetry are in involution whenever thegenerator of the symmetry satisfies a certain Yang-Baxter type equation.</div><div class="keywords"><strong class="cap">keywords. </strong>Non-Noether symmetry; Conservation law; bi-Hamiltonian system; Bidifferential calculus; Lax pair; Frölicher-Nijenhuisoperator;</div><div class="msc"><strong class="cap">msc. </strong> 70H33; 70H06; 53Z05</div><div class="reference">J. Geom. Phys. 48 (2003) 190-202</div><div class="paragraph">In the present paper we would like to shed more light on geometric aspects ofthe concept of non-Noether symmetry and to emphasize influence of such a symmetries on the phase space geometry.Partially the motivation for studying these issues comes from the theory of integrable modelsthat essentially relies on different geometric objects used for constructing conservationlaws. Among them are Frölicher-Nijenhuisoperators, bi-Hamiltonian systems, Lax pairs and bicomplexes. And it seems that the existance of these importantgeometric structures could be related to the hidden non-Noether symmetries of the dynamical systems.We would like to show how in Hamiltonian systems presence of certain non-Noether symmetries leads to theabove mentioned Lax pairs, Frölicher-Nijenhuis operators, bi-Hamiltonian structures,bicomplexes and a number of conservation laws.</div><div class="paragraph">Let us first recall some basic knowledge of the Hamiltonian dynamics. The phase space ofa regular Hamiltonian system is a Poisson manifold – a smooth finite-dimensionalmanifold equipped with the Poisson bivector field <math xmlns="http://xml-maiden.com">W</math>subjected to the following condition<formula xml:id="e1" xmlns="http://xml-maiden.com">[W , W] = 0</formula>where square bracket stands for Schouten bracket or supercommutator(for simplicity further it will be referred as commutator). In a standard manner Poissonbivector field defines a Lie bracket on the algebra of observables(smooth real-valued functions on phase space) called Poisson bracket:<formula xmlns="http://xml-maiden.com">{f , g} = W(df ∧ dg)</formula>Skew symmetry of the bivector field <math xmlns="http://xml-maiden.com">W</math> provides the skew symmetry ofthe corresponding Poisson bracket and the condition<a href="#e1">(1)</a> ensures that for every triple <math xmlns="http://xml-maiden.com">(f, g, h)</math> of smoothfunctions on the phase space the Jacobi identity<formula xmlns="http://xml-maiden.com">{f{g , h}} + {h{f , g}} + {g{h , f}} = 0.</formula>is satisfied. We also assume that the dynamical system under considerationis regular – the bivector field <math xmlns="http://xml-maiden.com">W</math> has maximalrank, i. e. its <math xmlns="http://xml-maiden.com">n</math>-th outer power, where  <math xmlns="http://xml-maiden.com">n</math> is a half-dimension ofthe phase space, does not vanish <math xmlns="http://xml-maiden.com">W<sup>n</sup> ≠ 0</math>.In this case <math xmlns="http://xml-maiden.com">W</math> gives rise to a well known isomorphism<math xmlns="http://xml-maiden.com">Φ</math> between the differential 1-forms andthe vector fields defined by<formula xml:id="e4" xmlns="http://xml-maiden.com">Φ(u) = W(u)</formula>for every 1-form <math xmlns="http://xml-maiden.com">u</math> and could be extended to higher degreedifferential forms and multivector fields by linearity and multiplicativity<math xmlns="http://xml-maiden.com">Φ(u ∧ v) = Φ(u) ∧ Φ(v)</math>.</div><div class="paragraph">Time evolution of observables (smooth functions on phase space) is governed by the Hamilton's equation<formula xml:id="e5" xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>Ô = {h , Ô}</formula>where <math xmlns="http://xml-maiden.com">h</math> is some fixed smooth function on the phase space called Hamiltonian.Let us recall that each vector field <math xmlns="http://xml-maiden.com">E</math> on the phase space generatesthe one-parameter continuous group of transformations<math xmlns="http://xml-maiden.com">g<sub>z</sub> = e<sup>zL<sub>E</sub></sup></math> (here <math xmlns="http://xml-maiden.com">L</math> denotes Lie derivative)that acts on the observables as follows<formula xmlns="http://xml-maiden.com">g<sub>z</sub>(Ô) = e<sup>zL<sub>E</sub></sup>(Ô) = f + zL<sub>E</sub>Ô + ½(zL<sub>E</sub>)<sup>2</sup>Ô + ⋯</formula>Such a group of transformation is called symmetry of Hamilton's equation <a href="#e5">(5)</a>if it commutes with time evolution operator<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction> g<sub>z</sub>(Ô) = g<sub>z</sub>(<fraction><num>d</num><den>dt</den></fraction>Ô)</formula>in terms of the vector fields this condition means that the generator<math xmlns="http://xml-maiden.com">E</math> of the group <math xmlns="http://xml-maiden.com">g<sub>z</sub></math> commutes with the vector field<math xmlns="http://xml-maiden.com">W(h) = {h , }</math>, i. e.<formula xml:id="e8" xmlns="http://xml-maiden.com">[E , W(h)] = 0.</formula> However we would like to consider more generalcase where <math xmlns="http://xml-maiden.com">E</math> is time dependent vector field on phase space. In this case<a href="#e8">(8)</a> should be replaced with<formula xml:id="e9" xmlns="http://xml-maiden.com"><fraction><num>∂</num><den>∂t</den></fraction>E = [E , W(h)].</formula>If in addition to <a href="#e8">(8)</a> the vector field <math xmlns="http://xml-maiden.com">E</math> does not preserve Poissonbivector field <math xmlns="http://xml-maiden.com">[E , W] ≠ 0</math> then <math xmlns="http://xml-maiden.com">g<sub>z</sub></math> is called non-Noether symmetry.</div><div class="paragraph">Now let us focus on non-Noether symmetries. We would like to show that the presence ofsuch a symmetry could essentially enrich the geometry of the phase spaceand under the certain conditions could ensure integrability of the dynamical system.Before we proceed let us recall that the non-Noether symmetry leads to a number ofintegrals of motion<a href="#r4">[4]</a>. More precisely therelationship between non-Noether symmetries and the conservation laws is described bythe following theorem.</div><div class="theorem"><strong class="cap">theorem. </strong>Let <math xmlns="http://xml-maiden.com">(M , h)</math> be regular Hamiltonian system on the <math xmlns="http://xml-maiden.com">2n</math>-dimensionalPoisson manifold <math xmlns="http://xml-maiden.com">M</math>. Then, if the vector field <math xmlns="http://xml-maiden.com">E</math> generatesnon-Noether symmetry, the functions<formula xml:id="e10" xmlns="http://xml-maiden.com">Y<sup>(k)</sup> = <fraction><num>V[k]</num><den>V[0]</den></fraction>           k = 1,2, ... n</formula>where <math xmlns="http://xml-maiden.com">V[k] = Ŵ<sup>k</sup> ∧ W<sup>n − k</sup></math> are multivector fields of maximal degree constructed by means of Poisson bivector <math xmlns="http://xml-maiden.com">W</math> and its Lie derivative <math xmlns="http://xml-maiden.com">Ŵ = [E , W]</math>, are integrals of motion. </div><div class="proof"><strong class="cap">proof. </strong>By the definition<formula xmlns="http://xml-maiden.com">Ŵ<sup>k</sup> ∧ W<sup>n − k</sup> = Y<sup>(k)</sup>W<sup>n</sup>.</formula>(definition is correct since the space of <math xmlns="http://xml-maiden.com">2n</math> degree multivector fields on <math xmlns="http://xml-maiden.com">2n</math>degree manifold is one dimensional).Let us take time derivative of this expression along the vector field <math xmlns="http://xml-maiden.com">W(h)</math>,<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>Ŵ<sup>k</sup> ∧ W<sup>n − k</sup> = (<fraction><num>d</num><den>dt</den></fraction>Y<sup>(k)</sup>)W<sup>n</sup> + Y<sup>(k)</sup>[W(h) , W<sup>n</sup>]</formula>or<formula xml:id="e13" xmlns="http://xml-maiden.com">k(<fraction><num>d</num><den>dt</den></fraction>Ŵ) ∧ Ŵ<sup>k − 1</sup> ∧ W<sup>n − k</sup><line/>+ (n − k)[W(h) , W] ∧ Ŵ<sup>k</sup> ∧ W<sup>n − k − 1</sup> = <line/>(<fraction><num>d</num><den>dt</den></fraction>Y<sup>(k)</sup>)W<sup>n</sup> + nY<sup>(k)</sup>[W(h) , W] ∧ W<sup>n − 1</sup></formula>but according to the Liouville theorem the Hamiltonian vector field preserves <math xmlns="http://xml-maiden.com">W</math> i. e.<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>W = [W(h) , W] = 0</formula>hence, by taking into account that<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>E= <fraction><num>∂</num><den>∂t</den></fraction>E + [W(h) , E] = 0</formula> we get<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>Ŵ = <fraction><num>d</num><den>dt</den></fraction>[E , W] = [<fraction><num>d</num><den>dt</den></fraction>E, W] + [E[W(h) , W]] = 0.</formula>and as a result <a href="#e13">(13)</a> yields<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>Y<sup>(k)</sup> W<sup>n</sup> = 0</formula>but since the dynamical system is regular (<math xmlns="http://xml-maiden.com">W<sup>n</sup> ≠ 0</math>)we obtain that the functions <math xmlns="http://xml-maiden.com">Y<sup>(k)</sup></math> are integrals of motion.</div><div class="remark"><strong class="cap">remark. </strong> Instead of conserved quantities<math xmlns="http://xml-maiden.com">Y<sup>(1)</sup> ... Y<sup>(n)</sup></math>, thesolutions <math xmlns="http://xml-maiden.com">c<sub>1</sub> ... c<sub>n</sub></math> of the secular equation<formula xml:id="e18" xmlns="http://xml-maiden.com">(Ŵ − cW)<sup>n</sup> = 0</formula>could be associated with the generator of symmetry.By expanding expression <a href="#e18">(18)</a> it is easy to verify that the conservation laws<math xmlns="http://xml-maiden.com">Y<sup>(k)</sup></math> can be expressed in terms of the integrals of motion<math xmlns="http://xml-maiden.com">c<sub>1</sub> ... c<sub>n</sub></math> in the following way<formula xml:id="e19" xmlns="http://xml-maiden.com">Y<sup>(k)</sup> = <fraction><num>(n − k)! k!</num><den>n!</den></fraction> <under><ope>∑</ope><sat>m[i] &gt; m[j]</sat></under> c<sub>m[1]</sub>c<sub>m[2]</sub> ⋯ c<sub>m[k]</sub></formula></div><div class="example"><strong class="cap">example. </strong> Let <math xmlns="http://xml-maiden.com">M</math> be <math xmlns="http://xml-maiden.com">R<sup>4</sup></math> with coordinates<math xmlns="http://xml-maiden.com">z<sub>1</sub>, z<sub>2</sub>, z<sub>3</sub>, z<sub>4</sub></math> and Poisson bivector field<formula xml:id="e20" xmlns="http://xml-maiden.com">W = D<sub>1</sub> ∧ D<sub>3</sub> + D<sub>2</sub> ∧ D<sub>4</sub></formula>(<math xmlns="http://xml-maiden.com">D<sub>m</sub></math> just denotes derivative with respect to <math xmlns="http://xml-maiden.com">z<sub>m</sub></math> coordinate)and let's take<formula xmlns="http://xml-maiden.com">h = ½z<sub>1</sub><sup>2</sup> + ½z<sub>2</sub><sup>2</sup> + e<sup>z<sub>3</sub> − z<sub>4</sub></sup></formula>Then the vector field<formula xmlns="http://xml-maiden.com">E = <under><over><sat>4</sat><ope>∑</ope></over><sat>m = 1</sat></under>E<sub>m</sub>D<sub>m</sub></formula>with components<formula xml:id="e23" xmlns="http://xml-maiden.com">E<sub>1</sub> = ½z<sub>1</sub><sup>2</sup> − e<sup>z<sub>3</sub> − z<sub>4</sub></sup> −<fraction><num>t</num><den>2</den></fraction>(z<sub>1</sub> + z<sub>2</sub>)e<sup>z<sub>3</sub> − z<sub>4</sub></sup><line/>E<sub>2</sub> = ½z<sub>2</sub><sup>2</sup> + 2e<sup>z<sub>3</sub> − z<sub>4</sub></sup> +<fraction><num>t</num><den>2</den></fraction>(z<sub>1</sub> + z<sub>2</sub>)e<sup>z<sub>3</sub> − z<sub>4</sub></sup><line/>E<sub>3</sub> = 2z<sub>1</sub> + ½z<sub>2</sub> + <fraction><num>t</num><den>2</den></fraction>(z<sub>1</sub><sup>2</sup> + e<sup>z<sub>3</sub> − z<sub>4</sub></sup>)<line/>E<sub>4</sub> = z<sub>2</sub> − ½z<sub>1</sub> + <fraction><num>t</num><den>2</den></fraction>(z<sub>2</sub><sup>2</sup> + e<sup>z<sub>3</sub> − z<sub>4</sub></sup>)</formula>satisfies <a href="#e9">(9)</a> condition and as a result generates symmetry of the dynamical system.The symmetry appears to be non-Noether with Schouten bracket <math xmlns="http://xml-maiden.com">[E , W]</math> equal to<formula xml:id="e24" xmlns="http://xml-maiden.com">Ŵ = [E , W] = z<sub>1</sub>D<sub>1</sub> ∧ D<sub>3</sub> +z<sub>2</sub>D<sub>2</sub> ∧ D<sub>4</sub> +e<sup>z<sub>3</sub> − z<sub>4</sub></sup>D<sub>1</sub> ∧ D<sub>2</sub> +D<sub>3</sub> ∧ D<sub>4</sub></formula>calculating volume vector fields<math xmlns="http://xml-maiden.com">Ŵ<sup>k</sup> ∧ W<sup>n − k</sup></math> gives rise to<formula xmlns="http://xml-maiden.com">W ∧ W = − 2D<sub>1</sub> ∧ D<sub>2</sub> ∧ D<sub>3</sub> ∧ D<sub>4</sub><line/>Ŵ ∧ W = − (z<sub>1</sub> + z<sub>2</sub>)D<sub>1</sub> ∧ D<sub>2</sub> ∧ D<sub>3</sub> ∧ D<sub>4</sub><line/>Ŵ ∧ Ŵ = − 2(z<sub>1</sub>z<sub>2</sub> − e<sup>z<sub>3</sub> − z<sub>4</sub></sup>) D<sub>1</sub> ∧ D<sub>2</sub> ∧ D<sub>3</sub> ∧ D<sub>4</sub></formula>and the conservation laws associated with this symmetry are just<formula xmlns="http://xml-maiden.com">Y<sup>(1)</sup> = <fraction><num>Ŵ ∧ W</num><den>W ∧ W</den></fraction> = ½(z<sub>1</sub> + z<sub>2</sub>)<line/>Y<sup>(2)</sup> = <fraction><num>Ŵ ∧ Ŵ</num><den>W ∧ W</den></fraction> = z<sub>1</sub>z<sub>2</sub> − e<sup>z<sub>3</sub> − z<sub>4</sub></sup></formula></div><div class="paragraph">Presence of the non-Noether symmetry not only leads to a sequence of conservation laws, but alsoendows the phase space with a number of interesting geometric structures and it appears that such asymmetry is related to many important concepts used in theory of dynamical systems.One of the such concepts is Lax pair.Let us recall that Lax pair of Hamiltonian system on Poisson manifold <math xmlns="http://xml-maiden.com">M</math> isa pair <math xmlns="http://xml-maiden.com">(L , P)</math> of smooth functions on <math xmlns="http://xml-maiden.com">M</math> with values in someLie algebra <math xmlns="http://xml-maiden.com">g</math> such that the time evolution of <math xmlns="http://xml-maiden.com">L</math> is governedby the following equation<formula xml:id="e27" xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>L = [L , P]</formula>where <math xmlns="http://xml-maiden.com">[ , ]</math> is a Lie bracket on <math xmlns="http://xml-maiden.com">g</math>. It is well known that each Laxpair leads to a number of conservation laws. When <math xmlns="http://xml-maiden.com">g</math> is some matrix Lie algebrathe conservation laws are just traces of powers of <math xmlns="http://xml-maiden.com">L</math><formula xml:id="e28" xmlns="http://xml-maiden.com">I<sup>(k)</sup> = Tr(L<sup>k</sup>)</formula>It is remarkable that each generator of the non-Noethersymmetry canonically leads to the Lax pair of a certain type.In the local coordinates <math xmlns="http://xml-maiden.com">z<sub>m</sub></math>, where the bivector field<math xmlns="http://xml-maiden.com">W</math> and the generator of the symmetry <math xmlns="http://xml-maiden.com">E</math> have thefollowing form<formula xmlns="http://xml-maiden.com">W = <under><ope>∑</ope><sat>km</sat></under>W<sub>ab</sub>D<sub>k</sub> ∧ D<sub>m</sub>              E = <under><ope>∑</ope><sat>m</sat></under>E<sub>m</sub>D<sub>m</sub></formula>corresponding Lax pair could be calculated explicitly.Namely we have the following theorem:</div><div class="theorem"><strong class="cap">theorem. </strong>Let <math xmlns="http://xml-maiden.com">(M , h)</math> be regular Hamiltonian system on the <math xmlns="http://xml-maiden.com">2n</math>-dimensionalPoisson manifold <math xmlns="http://xml-maiden.com">M</math>.Then, if the vector field <math xmlns="http://xml-maiden.com">E</math> on <math xmlns="http://xml-maiden.com">M</math> generates the non-Noether symmetry,the following <math xmlns="http://xml-maiden.com">2n×2n</math> matrix valued functions on <math xmlns="http://xml-maiden.com">M</math><formula xml:id="e30" xmlns="http://xml-maiden.com">L<sub>ab</sub> = <under><ope>∑</ope><sat>dc</sat></under> (W<sup>−1</sup>)<sub>ad</sub> (E<sub>c</sub>D<sub>c</sub>W<sub>db</sub>− W<sub>cb</sub>D<sub>c</sub>E<sub>d</sub> + W<sub>dc</sub>D<sub>c</sub>E<sub>b</sub>)<line/>P<sub>ab</sub> = <under><ope>∑</ope><sat>c</sat></under> D<sub>a</sub> (W<sub>bc</sub>D<sub>c</sub>h)</formula>form the Lax pair <a href="#e27">(27)</a> of the dynamical system <math xmlns="http://xml-maiden.com">(M , h)</math>.</div><div class="proof"><strong class="cap">proof. </strong>Let us consider the following operator on a space of 1-forms<formula xml:id="e31" xmlns="http://xml-maiden.com">Ŕ<sub>E</sub>(u) = Φ<sup>− 1</sup>([E , Φ(u)]) − L<sub>E</sub>u</formula>(here <math xmlns="http://xml-maiden.com">Φ</math> is the isomorphism <a href="#e4">(4)</a>).It is obvious that <math xmlns="http://xml-maiden.com">Ŕ<sub>E</sub></math>is a linear operator and it is invariantsince time evolution commutes with both<math xmlns="http://xml-maiden.com">Φ</math>(as far as <math xmlns="http://xml-maiden.com">[W(h) , W] = 0</math>) and <math xmlns="http://xml-maiden.com">E</math> (because <math xmlns="http://xml-maiden.com">E</math> generatessymmetry). In the terms of the local coordinates <math xmlns="http://xml-maiden.com">Ŕ<sub>E</sub></math> has the following form<formula xmlns="http://xml-maiden.com">Ŕ<sub>E</sub> = <under><ope>∑</ope><sat>ab</sat></under>L<sub>ab</sub> dz<sub>a</sub> ⊗ D<sub>b</sub></formula>and the invariance condition<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>Ŕ<sub>E</sub> = L<sub>W(h)</sub>Ŕ<sub>E</sub> = 0</formula>yields<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>Ŕ<sub>E</sub> =<fraction><num>d</num><den>dt</den></fraction><under><ope>∑</ope><sat>ab</sat></under>L<sub>ab</sub> dz<sub>a</sub> ⊗ D<sub>b</sub><line/>= <under><ope>∑</ope><sat>ab</sat></under>(<fraction><num>d</num><den>dt</den></fraction>L<sub>ab</sub>) dz<sub>a</sub> ⊗ D<sub>b</sub> +<under><ope>∑</ope><sat>ab</sat></under>L<sub>ab</sub> (L<sub>W(h)</sub>dz<sub>a</sub>) ⊗ D<sub>b</sub><line/>+ <under><ope>∑</ope><sat>ab</sat></under>L<sub>ab</sub> dz<sub>a</sub> ⊗ (L<sub>W(h)</sub>D<sub>b</sub>) =<under><ope>∑</ope><sat>ab</sat></under>(<fraction><num>d</num><den>dt</den></fraction>L<sub>ab</sub>) dz<sub>a</sub> ⊗ D<sub>b</sub><line/>+ <under><ope>∑</ope><sat>abcd</sat></under>L<sub>ab</sub>D<sub>c</sub>(W<sub>ad</sub>D<sub>d</sub>h)dz<sub>c</sub> ⊗ D<sub>b</sub> +<under><ope>∑</ope><sat>abcd</sat></under>L<sub>ab</sub>D<sub>b</sub>(W<sub>cd</sub>D<sub>d</sub>h)dz<sub>a</sub> ⊗ D<sub>c</sub><line/>= <under><ope>∑</ope><sat>ab</sat></under>(<fraction><num>d</num><den>dt</den></fraction>L<sub>ab</sub> + <under><ope>∑</ope><sat>c</sat></under>(P<sub>ac</sub>L<sub>cb</sub> − L<sub>ac</sub>P<sub>cb</sub>))dz<sub>a</sub> ⊗ D<sub>b</sub> = 0</formula>or in matrix notations<formula xmlns="http://xml-maiden.com"><fraction><num>d</num><den>dt</den></fraction>L = [L , P].</formula>So, we have proved that the non-Noether symmetry canonically yields a Lax pairon the algebra of linear operators on cotangent bundle over the phase space.</div><div class="remark"><strong class="cap">remark. </strong> The conservation laws <a href="#e28">(28)</a>associated with the Lax pair <a href="#e27">(27)</a> can be expressed in terms of theintegrals of motion <math xmlns="http://xml-maiden.com">c<sub>i</sub></math> in quite simple way:<formula xml:id="e36" xmlns="http://xml-maiden.com">I<sup>(k)</sup> = Tr(L<sup>k</sup>) = <under><ope>∑</ope><sat>m</sat></under> c<sub>m</sub><sup>k</sup></formula>This correspondence follows from the equation <a href="#e18">(18)</a>and the definition of the operator <math xmlns="http://xml-maiden.com">Ŕ<sub>E</sub></math> <a href="#e31">(31)</a>.</div><div class="example"><strong class="cap">example. </strong>Let us calculate Lax matrix associated with non-Noether symmetry <a href="#e23">(23)</a>.Using <a href="#e30">(30)</a> it is easy to check that Lax matrix has eight nonzero elements<formula xmlns="http://xml-maiden.com">L<sub>11</sub> = L<sub>33</sub> = z<sub>1</sub>;           L<sub>22</sub> = L<sub>44</sub> = z<sub>2</sub><line/>L<sub>14</sub> = − L<sub>23</sub> = e<sup>z<sub>3</sub> − z<sub>4</sub></sup>;           L<sub>32</sub> = − L<sub>41</sub> = 1</formula>The conservation laws associated with this Lax matrix are<formula xml:id="e38" xmlns="http://xml-maiden.com">I<sup>(1)</sup> = Tr(L) = 2(z<sub>1</sub> + z<sub>2</sub>)<line/>I<sup>(2)</sup> = Tr(L<sup>2</sup>) = 2z<sub>1</sub><sup>2</sup> + 2z<sub>2</sub><sup>2</sup> + 4e<sup>z<sub>3</sub> − z<sub>4</sub></sup></formula></div><div class="paragraph">Now let us focus on the integrability issues. We know that<math xmlns="http://xml-maiden.com">n</math> integrals of motion are associated with each generator of non-Noethersymmetry and according to the Liouville-Arnold theorem Hamiltonian system iscompletely integrable if it possesses <math xmlns="http://xml-maiden.com">n</math> functionally independent integrals ofmotion in involution (two functions <math xmlns="http://xml-maiden.com">f</math> and <math xmlns="http://xml-maiden.com">g</math> are said to bein involution if their Poisson bracket vanishes <math xmlns="http://xml-maiden.com">{f , g} = 0</math>).Generally speaking the conservation laws associated with symmetry might appear to be neitherindependent nor involutive.However it is reasonable to ask the question – what condition should be satisfiedby the generator of the symmetry to ensure the involutivity(<math xmlns="http://xml-maiden.com">{Y<sup>(k)</sup> , Y<sup>(m)</sup>} = 0</math>) of conserved quantities?In Lax theory such a condition is known asClassical Yang-Baxter Equation (CYBE). Since involutivity of the conservation lawsis closely related to the integrability it is essential to have some analog of CYBE for the generatorof non-Noether symmetry. To address this issue we would like to propose the following theorem.</div><div class="theorem"><strong class="cap">theorem. </strong>If the vector field <math xmlns="http://xml-maiden.com">E</math> on <math xmlns="http://xml-maiden.com">2n</math>-dimensionalPoisson manifold <math xmlns="http://xml-maiden.com">M</math>satisfies the condition<formula xml:id="e39" xmlns="http://xml-maiden.com">[[E[E , W]]W] = 0</formula>and <math xmlns="http://xml-maiden.com">W</math> bivector field has maximal rank (<math xmlns="http://xml-maiden.com">W<sup>n</sup> ≠ 0</math>)then the functions <a href="#e10">(10)</a> are in involution<formula xmlns="http://xml-maiden.com">{Y<sup>(k)</sup> , Y<sup>(m)</sup>} = 0</formula></div><div class="proof"><strong class="cap">proof. </strong> First of all let us note thatthe identity <a href="#e1">(1)</a> satisfied by the Poissonbivector field <math xmlns="http://xml-maiden.com">W</math> is responsible for the Liouville theorem<formula xml:id="e41" xmlns="http://xml-maiden.com">[W , W] = 0               L<sub>W(f)</sub>W = [W(f) , W] = 0</formula>By taking the Lie derivative of the expression <a href="#e1">(1)</a>we obtain another useful identity<formula xmlns="http://xml-maiden.com">L<sub>E</sub>[W , W] = [E[W , W]] = [[E , W] W] + [W[E , W]] = 2[Ŵ , W] = 0.</formula>This identity gives rise to the following relation<formula xml:id="e43" xmlns="http://xml-maiden.com">[Ŵ , W] = 0          ⇔          [Ŵ(f) , W] = − [Ŵ , W(f)]</formula>and finally condition <a href="#e39">(39)</a> ensures third identity<formula xmlns="http://xml-maiden.com">[Ŵ , Ŵ] = 0</formula>yielding Liouville theorem for <math xmlns="http://xml-maiden.com">Ŵ</math><formula xml:id="e45" xmlns="http://xml-maiden.com">[Ŵ , Ŵ] = 0          ⇔          [Ŵ(f) , Ŵ] = 0</formula>Indeed<formula xmlns="http://xml-maiden.com">[Ŵ , Ŵ] = [[E , W]Ŵ] = [[Ŵ , E]W]<line/>= − [[E , Ŵ]W] = − [[E[E , W]]W] = 0</formula>Now let us consider two different solutions <math xmlns="http://xml-maiden.com">c<sub>i</sub> ≠ c<sub>j</sub></math>of the equation <a href="#e18">(18)</a>. By taking the Lie derivative of the equation<formula xmlns="http://xml-maiden.com">(Ŵ − c<sub>i</sub>W)<sup>n</sup> = 0</formula>along the vector fields <math xmlns="http://xml-maiden.com">W(c<sub>j</sub>)</math> and<math xmlns="http://xml-maiden.com">Ŵ(c<sub>j</sub>)</math>  and using Liouville theorem for<math xmlns="http://xml-maiden.com">W</math> and <math xmlns="http://xml-maiden.com">Ŵ</math> bivectors we obtain the following relations<formula xml:id="e48" xmlns="http://xml-maiden.com">(Ŵ − c<sub>i</sub>W)<sup>n − 1</sup>(L<sub>W(c<sub>j</sub>)</sub>Ŵ − {c<sub>j</sub> , c<sub>i</sub>}W) = 0,</formula>and<formula xml:id="e49" xmlns="http://xml-maiden.com">(Ŵ − c<sub>i</sub>W)<sup>n − 1</sup>(c<sub>i</sub>L<sub>Ŵ(c<sub>j</sub>)</sub>W + {c<sub>j</sub> , c<sub>i</sub>}<sub>•</sub>W) = 0,</formula>where<formula xmlns="http://xml-maiden.com">{c<sub>i</sub> , c<sub>j</sub>}<sub>•</sub> = Ŵ(dc<sub>i</sub> ∧ dc<sub>j</sub>)</formula>is the Poisson bracket calculated by means of the bivector field <math xmlns="http://xml-maiden.com">Ŵ</math>.Now multiplying <a href="#e48">(48)</a> by <math xmlns="http://xml-maiden.com">c<sub>i</sub></math> subtracting <a href="#e49">(49)</a> and usingidentity <a href="#e43">(43)</a> gives rise to<formula xml:id="e51" xmlns="http://xml-maiden.com">({c<sub>i</sub> , c<sub>j</sub>}<sub>•</sub> − c<sub>i</sub>{c<sub>i</sub> , c<sub>j</sub>})(Ŵ − c<sub>i</sub>W)<sup>n − 1</sup>W = 0</formula>Thus, either<formula xml:id="e52" xmlns="http://xml-maiden.com">{c<sub>i</sub> , c<sub>j</sub>}<sub>•</sub> − c<sub>i</sub>{c<sub>i</sub> , c<sub>j</sub>} = 0</formula>or the volume field<math xmlns="http://xml-maiden.com">(Ŵ − c<sub>i</sub>W)<sup>n − 1</sup>W</math>vanishes. In the second case we can repeat<a href="#e48">(48)</a>-<a href="#e51">(51)</a> procedure forthe volume field<math xmlns="http://xml-maiden.com">(Ŵ − c<sub>i</sub>W)<sup>n − 1</sup>W</math>yielding after <math xmlns="http://xml-maiden.com">n</math>iterations <math xmlns="http://xml-maiden.com">W<sup>n</sup> = 0</math> that according to ourassumption (that the dynamical system is regular) is not true.As a result we arrived at <a href="#e52">(52)</a> and by the simpleinterchange of indices <math xmlns="http://xml-maiden.com">i ↔ j</math> we get<formula xml:id="e53" xmlns="http://xml-maiden.com">{c<sub>i</sub> , c<sub>j</sub>}<sub>•</sub> − c<sub>j</sub>{c<sub>i</sub> , c<sub>j</sub>} = 0</formula>Finally by comparing <a href="#e52">(52)</a> and <a href="#e53">(53)</a> we obtain thatthe functions <math xmlns="http://xml-maiden.com">c<sub>i</sub></math>  are in involution with respect to the bothPoisson structures (since <math xmlns="http://xml-maiden.com">c<sub>i</sub> ≠ c<sub>j</sub></math>)<formula xmlns="http://xml-maiden.com">{c<sub>i</sub> , c<sub>j</sub>}<sub>•</sub> = {c<sub>i</sub> , c<sub>j</sub>} = 0</formula>and according to <a href="#e19">(19)</a> the same is true for the integrals of motion<math xmlns="http://xml-maiden.com">Y<sup>(k)</sup></math>.</div><div class="corollary"><strong class="cap">corollary. </strong>Each generator of non-Noether symmetry satisfying equation <a href="#e39">(39)</a> endowsdynamical system with the bi-Hamiltonian structure – couple (<math xmlns="http://xml-maiden.com">W , Ŵ</math>)of compatible (<math xmlns="http://xml-maiden.com">[W , Ŵ] = 0</math>)Poisson (<math xmlns="http://xml-maiden.com">[W , W] = [Ŵ , Ŵ] = 0</math>)bivector fields.</div><div class="remark"><strong class="cap">remark. </strong> Theorem 3 is useful in multidimentional dynamical systems where involutivity ofconservation laws can not be checked directly.</div><div class="example"><strong class="cap">example. </strong> One can check that the non-Noether symmetry <a href="#e23">(23)</a> satisfiescondition <a href="#e39">(39)</a> and the bivector fields <math xmlns="http://xml-maiden.com">W</math> and <math xmlns="http://xml-maiden.com">Ŵ</math> defined by<a href="#e20">(20)</a> and <a href="#e24">(24)</a> form bi-Hamiltonian system<math xmlns="http://xml-maiden.com">[W , W] = [W , Ŵ] = [Ŵ , Ŵ] = 0</math>.</div><div class="paragraph">Another concept that is often used in theory of dynamical systems and couldbe related to the non-Noether symmetry is the bidifferential calculus (bicomplex approach).Recently A. Dimakis and F. Müller-Hoissenapplied bidifferential calculi to the wide range of integrable modelsincluding KdV hierarchy, KP equation, self-dual Yang-Mills equation,Sine-Gordon equation, Toda models, non-linear Schrödingerand Liouville equations. It turns out that these models can be effectivelydescribed and analyzed using the bidifferential calculi <a href="#r1">[1]</a>, <a href="#r2">[2]</a>.</div><div class="paragraph">Under the bidifferential calculus we mean the graded algebra of differential forms<formula xmlns="http://xml-maiden.com">Ω = <under><over><sat>∞</sat><ope>∪</ope></over><sat>k = 0</sat></under> Ω<sup>(k)</sup></formula>(<math xmlns="http://xml-maiden.com">Ω<sup>(k)</sup></math> denotes the space of <math xmlns="http://xml-maiden.com">k</math>-degree differential forms)equipped with a couple of differential operators<formula xmlns="http://xml-maiden.com">d, đ : Ω<sup>(k)</sup> → Ω<sup>(k + 1)</sup></formula>satisfying<math xmlns="http://xml-maiden.com">d<sup>2</sup> = đ<sup>2</sup> = dđ + đd = 0</math>conditions (see <a href="#r2">[2]</a>).It is interesting that if generator of the non-Noether symmetry satisfiesequation <a href="#e39">(39)</a> then we are able to construct an invariant bidifferential calculusof a certain type. This construction is summarized in  the following theorem:</div><div class="theorem"><strong class="cap">theorem. </strong>Let <math xmlns="http://xml-maiden.com">(M , h)</math> be regular Hamiltonian system on the Poisson manifold <math xmlns="http://xml-maiden.com">M</math>.Then, if the vector field <math xmlns="http://xml-maiden.com">E</math> on <math xmlns="http://xml-maiden.com">M</math> generates the non-Noether symmetryand satisfies the equation <a href="#e39">(39)</a>, the differential operators<formula xml:id="e57" xmlns="http://xml-maiden.com">du = Φ<sup>− 1</sup>([W , Φ(u)])</formula><formula xml:id="e58" xmlns="http://xml-maiden.com">đu = Φ<sup>− 1</sup>([[E , W]Φ(u)])</formula>form invariant bidifferential calculus(<math xmlns="http://xml-maiden.com">d<sup>2</sup> = đ<sup>2</sup> = dđ + đd = 0</math>)over the graded algebra of differential forms on <math xmlns="http://xml-maiden.com">M</math>.</div><div class="proof"><strong class="cap">proof. </strong> First of all we have to show that <math xmlns="http://xml-maiden.com">d</math> and <math xmlns="http://xml-maiden.com">đ</math>are really differential operators , i.e., they are linear maps from<math xmlns="http://xml-maiden.com">Ω<sup>(k)</sup></math> into<math xmlns="http://xml-maiden.com">Ω<sup>(k + 1)</sup></math>, satisfy derivation property andare nilpotent (<math xmlns="http://xml-maiden.com">d<sup>2</sup> = đ<sup>2</sup> = 0</math>).Linearity is obvious and follows from the linearity of the Schouten bracket  <math xmlns="http://xml-maiden.com">[ , ]</math>and <math xmlns="http://xml-maiden.com">Φ, Φ<sup>− 1</sup></math>maps. Then, if <math xmlns="http://xml-maiden.com">u</math> is a <math xmlns="http://xml-maiden.com">k</math>-degree form<math xmlns="http://xml-maiden.com">Φ</math> maps it on <math xmlns="http://xml-maiden.com">k</math>-degree multivector field andthe Schouten brackets <math xmlns="http://xml-maiden.com">[W , Φ(u)]</math> and<math xmlns="http://xml-maiden.com">[[E , W]Φ(u)]</math> result the<math xmlns="http://xml-maiden.com">k + 1</math>-degree multivector fields that are mapped on <math xmlns="http://xml-maiden.com">k + 1</math>-degreedifferential forms by <math xmlns="http://xml-maiden.com">Φ<sup>− 1</sup></math>.So, <math xmlns="http://xml-maiden.com">d</math> and <math xmlns="http://xml-maiden.com">đ</math>are linear maps from <math xmlns="http://xml-maiden.com">Ω<sup>(k)</sup></math> into<math xmlns="http://xml-maiden.com">Ω<sup>(k + 1)</sup></math>.Derivation property follows from the same feature of the Schouten bracket<math xmlns="http://xml-maiden.com">[ , ]</math> and linearity of<math xmlns="http://xml-maiden.com">Φ</math> and<math xmlns="http://xml-maiden.com">Φ<sup>− 1</sup></math> maps.Now we have to prove the nilpotency of <math xmlns="http://xml-maiden.com">d</math> and <math xmlns="http://xml-maiden.com">đ</math>.Let us consider <math xmlns="http://xml-maiden.com">d<sup>2</sup>u</math><formula xmlns="http://xml-maiden.com">d<sup>2</sup>u = Φ<sup>− 1</sup>([W , Φ(Φ<sup>− 1</sup>([W , Φ(u)]))])<line/>= Φ<sup>− 1</sup>([W[W , Φ(u)]]) = 0</formula>as a result of the property <a href="#e41">(41)</a> and the Jacobi identity for <math xmlns="http://xml-maiden.com">[ , ]</math> bracket.In the same manner<formula xmlns="http://xml-maiden.com">đ<sup>2</sup>u = Φ<sup>− 1</sup>([[W , E][[W , E]Φ(u)]]) = 0</formula>according to the property <a href="#e45">(45)</a> of<math xmlns="http://xml-maiden.com">[W , E] = Ŵ</math> and the Jacobi identity.Thus, we have proved that <math xmlns="http://xml-maiden.com">d</math> and <math xmlns="http://xml-maiden.com">đ</math> are differential operators(in fact <math xmlns="http://xml-maiden.com">d</math> is ordinary exterior differential and the expression<a href="#e57">(57)</a> is its well known representation in terms of Poisson bivector field).It remains to show that the compatibility condition <math xmlns="http://xml-maiden.com">dđ + đd = 0</math>is fulfilled. Using definitions of <math xmlns="http://xml-maiden.com">d, đ</math> and the Jacobi identity we get<formula xmlns="http://xml-maiden.com">(dđ + đd)(u) = Φ<sup>− 1</sup>([[[W , E]W]Φ(u)]) = 0</formula>as far as <a href="#e43">(43)</a> is satisfied.So, <math xmlns="http://xml-maiden.com">d</math> and <math xmlns="http://xml-maiden.com">đ</math> form the bidifferential calculus over the gradedalgebra of differential forms.It is also clear that the bidifferential calculus <math xmlns="http://xml-maiden.com">d, đ</math>is invariant, since both <math xmlns="http://xml-maiden.com">d</math> and <math xmlns="http://xml-maiden.com">đ</math> commute with time evolutionoperator <math xmlns="http://xml-maiden.com">W(h) = {h, }</math>.</div><div class="remark"><strong class="cap">remark. </strong>Conservation laws that are associated with the bidifferential calculus<a href="#e57">(57)</a> <a href="#e58">(58)</a> and form Lenard scheme  (see <a href="#r2">[2]</a>): <formula xmlns="http://xml-maiden.com">(k + 1)đI<sup>(k)</sup> = kdI<sup>(k + 1)</sup></formula>coincide with the sequence of integrals of motion <a href="#e36">(36)</a>.Proof of this correspondence lay outside the scope of present article,but could be done in the manner similar to <a href="#r1">[1]</a>.</div><div class="example"><strong class="cap">example. </strong> The symmetry <a href="#e23">(23)</a> endows <math xmlns="http://xml-maiden.com">R<sup>4</sup></math> with bicomplex structure <math xmlns="http://xml-maiden.com">d, đ</math> where <math xmlns="http://xml-maiden.com">d</math> is ordinary exterier derivative while <math xmlns="http://xml-maiden.com">đ</math> is defined by<formula xmlns="http://xml-maiden.com">đz<sub>1</sub> = z<sub>1</sub>dz<sub>1</sub> − e<sup>z<sub>3</sub> − z<sub>4</sub></sup>dz<sub>4</sub><line/>đz<sub>2</sub> = z<sub>2</sub>dz<sub>2</sub> + e<sup>z<sub>3</sub> − z<sub>4</sub></sup>dz<sub>3</sub><line/>đz<sub>3</sub> = z<sub>1</sub>dz<sub>3</sub> + dz<sub>2</sub><line/>đz<sub>4</sub> = z<sub>2</sub>dz<sub>4</sub> − dz<sub>1</sub></formula>and is extended to whole De Rham complex by linearity, derivation property andcompatibility property <math xmlns="http://xml-maiden.com">dđ + đd = 0</math>. The conservation laws<math xmlns="http://xml-maiden.com">I<sup>(1)</sup></math> and <math xmlns="http://xml-maiden.com">I<sup>(2)</sup></math> defined by <a href="#e38">(38)</a>form the simpliest Lenard scheme<formula xmlns="http://xml-maiden.com">2đI<sup>(1)</sup> = dI<sup>(2)</sup></formula></div><div class="paragraph">Finally we would like to reveal some features of the operator<math xmlns="http://xml-maiden.com">Ŕ<sub>E</sub></math><a href="#e31">(31)</a> and to show how Frölicher-Nijenhuis geometry could arise inHamiltonian system that possesses certain non-Noether symmetry.From the geometric properties of the tangent valued forms we knowthat the traces of powers of a linear operator <math xmlns="http://xml-maiden.com">F</math>on tangent bundle are in involution whenever its Frölicher-Nijenhuis torsion<math xmlns="http://xml-maiden.com">T(F)</math> vanishes, i. e. whenever for arbitrary vector fields <math xmlns="http://xml-maiden.com">X,Y</math> the condition<formula xmlns="http://xml-maiden.com">T(F)(X , Y) = [FX , FY] − F([FX , Y] + [X , FY] − F[X , Y]) = 0</formula>is satisfied.Torsionless forms are also called Frölicher-Nijenhuis operators and are widely used intheory of integrable models. We would like to showthat each generator of non-Noether symmetry satisfying equation <a href="#e39">(39)</a>canonnically leads to invariant Frölicher-Nijenhuis operator on tangentbundle over the phase space. Strictly speaking we have the following theorem.</div><div class="theorem"><strong class="cap">theorem. </strong>Let <math xmlns="http://xml-maiden.com">(M , h)</math> be regular Hamiltonian system on the Poisson manifold <math xmlns="http://xml-maiden.com">M</math>.If the vector field <math xmlns="http://xml-maiden.com">E</math> on <math xmlns="http://xml-maiden.com">M</math> generates the non-Noether symmetryand satisfies the equation  <a href="#e39">(39)</a> then the linear operator, defined forevery vector field <math xmlns="http://xml-maiden.com">X</math> by equation<formula xmlns="http://xml-maiden.com">R<sub>E</sub>(X) = Φ(L<sub>E</sub>Φ<sup>− 1</sup>(X)) − [E , X]</formula>is invariant Frölicher-Nijenhuis operator on <math xmlns="http://xml-maiden.com">M</math>.</div><div class="proof"><strong class="cap">proof. </strong>Invariance of <math xmlns="http://xml-maiden.com">R<sub>E</sub></math> follows from the invariance of the<math xmlns="http://xml-maiden.com">Ŕ<sub>E</sub></math> defined by <a href="#e31">(31)</a>(note that for arbitrary 1-form vector field <math xmlns="http://xml-maiden.com">u</math> and vector field <math xmlns="http://xml-maiden.com">X</math>contraction <math xmlns="http://xml-maiden.com">i<sub>X</sub>u</math> has the property<math xmlns="http://xml-maiden.com">i<sub>R<sub>E</sub>X</sub>u = i<sub>X</sub>Ŕ<sub>E</sub>u</math>,so <math xmlns="http://xml-maiden.com">R<sub>E</sub></math> is actually transposed to <math xmlns="http://xml-maiden.com">Ŕ<sub>E</sub></math>).It remains to show that the condition <a href="#e39">(39)</a> ensures vanishing of theFrölicher-Nijenhuis torsion <math xmlns="http://xml-maiden.com">T(R<sub>E</sub>)</math> of<math xmlns="http://xml-maiden.com">R<sub>E</sub></math>, i.e. for arbitrary vector fields <math xmlns="http://xml-maiden.com">X, Y</math> we must get<formula xml:id="e67" xmlns="http://xml-maiden.com">T(R<sub>E</sub>)(X , Y) = [R<sub>E</sub>(X) , R<sub>E</sub>(Y)]<line/> − R<sub>E</sub>([R<sub>E</sub>(X) , Y] + [X , R<sub>E</sub>(Y)] − R<sub>E</sub>([X , Y])) = 0</formula>First let us introduce the following auxiliary 2-forms<formula xml:id="e68" xmlns="http://xml-maiden.com">ω = Φ<sup>− 1</sup>(W),           ω<sup>•</sup> = Ŕ<sub>E</sub>ω           ω<sup>••</sup> = Ŕ<sub>E</sub>ω<sup>•</sup></formula>Using the realization <a href="#e57">(57)</a> of the differential <math xmlns="http://xml-maiden.com">d</math>and the property <a href="#e1">(1)</a> yields<formula xmlns="http://xml-maiden.com">dω = Φ<sup>− 1</sup>([W , W]) = 0</formula>Similarly, using the property  <a href="#e43">(43)</a> we obtain<formula xmlns="http://xml-maiden.com">dω<sup>•</sup> = dΦ<sup>− 1</sup>([E , W]) − dL<sub>E</sub>ω = Φ<sup>− 1</sup>([[E , W]W]) − L<sub>E</sub>dω = 0</formula>And finally, taking into account that<math xmlns="http://xml-maiden.com">ω<sup>•</sup> = 2Φ<sup>− 1</sup>([E , W])</math>and using the condition <a href="#e39">(39)</a>, we get<formula xmlns="http://xml-maiden.com">dω<sup>••</sup> = 2Φ<sup>− 1</sup>([[E[E , W]]W]) − 2dL<sub>E</sub>ω<sup>•</sup> = − 2L<sub>E</sub>dω<sup>•</sup> = 0</formula>So the differential forms<math xmlns="http://xml-maiden.com">ω, ω<sup>•</sup>, ω<sup>••</sup></math>are closed<formula xml:id="e72" xmlns="http://xml-maiden.com">dω = dω<sup>•</sup> = dω<sup>••</sup> = 0</formula>Now let us consider the contraction of <math xmlns="http://xml-maiden.com">T(R<sub>E</sub>)</math> and <math xmlns="http://xml-maiden.com">ω</math>.<formula xml:id="e73" xmlns="http://xml-maiden.com">i<sub>T(R<sub>E</sub>)(X , Y)</sub>ω = i<sub>[R<sub>E</sub>X , R<sub>E</sub>Y]</sub>ω −i<sub>[R<sub>E</sub>X , Y]</sub>ω<sup>•</sup> −i<sub>[X , R<sub>E</sub>Y]</sub>ω<sup>•</sup> +i<sub>[X , Y]</sub>ω<sup>••</sup><line/>= L<sub>R<sub>E</sub>X</sub>i<sub>Y</sub>ω<sup>•</sup> −i<sub>R<sub>E</sub>Y</sub>L<sub>X</sub>ω<sup>•</sup> −L<sub>R<sub>E</sub>X</sub>i<sub>Y</sub>ω<sup>•</sup> +i<sub>Y</sub>L<sub>R<sub>E</sub>X</sub>ω<sup>•</sup> − L<sub>X</sub>i<sub>R<sub>E</sub>Y</sub>ω<sup>•</sup><line/>+ i<sub>R<sub>E</sub>Y</sub>L<sub>X</sub>ω<sup>•</sup> +i<sub>[X , Y]</sub>ω<sup>••</sup> = i<sub>Y</sub>L<sub>X</sub>ω<sup>••</sup> −L<sub>X</sub>i<sub>Y</sub>ω<sup>••</sup> +i<sub>[X , Y]</sub>ω<sup>••</sup> = 0</formula>where we used <a href="#e68">(68)</a> <a href="#e72">(72)</a>,the property of the Lie derivative<formula xmlns="http://xml-maiden.com">L<sub>X</sub>i<sub>Y</sub>ω =i<sub>Y</sub>L<sub>X</sub>ω + i<sub>[X , Y]</sub>ω</formula>and the relations of the following type<formula xmlns="http://xml-maiden.com">L<sub>R<sub>E</sub>X</sub>ω = di<sub>R<sub>E</sub>X</sub>ω + i<sub>R<sub>E</sub>X</sub>dω = di<sub>X</sub>ω<sup>•</sup> = L<sub>X</sub>ω<sup>•</sup> − i<sub>X</sub>dω<sup>•</sup> = L<sub>X</sub>ω<sup>•</sup></formula>So we proved that for arbitrary vector fields <math xmlns="http://xml-maiden.com">X, Y</math>the contraction of <math xmlns="http://xml-maiden.com">T(R<sub>E</sub>)(X , Y)</math> and <math xmlns="http://xml-maiden.com">ω</math> vanishes.But since <math xmlns="http://xml-maiden.com">W</math> bivector is non-degenerate(<math xmlns="http://xml-maiden.com">W<sup>n</sup> ≠ 0</math>), its counter image<formula xmlns="http://xml-maiden.com">ω = Φ<sup>− 1</sup>(W)</formula>is also non-degenerate and vanishing of the contraction <a href="#e73">(73)</a>implies that the torsion <math xmlns="http://xml-maiden.com">T(R<sub>E</sub>)</math> itself is zero.So we get<formula xmlns="http://xml-maiden.com">T(R<sub>E</sub>)(X , Y) = [R<sub>E</sub>(X) , R<sub>E</sub>(Y)]<line/>− R<sub>E</sub>([R<sub>E</sub>(X) , Y] + [X , R<sub>E</sub>(Y)] − R<sub>E</sub>([X , Y])) = 0</formula></div><div class="example"><strong class="cap">example. </strong> Note that operator <math xmlns="http://xml-maiden.com">R<sub>E</sub></math> associated with non-Noethersymmetry <a href="#e23">(23)</a> reproduces well known Frölicher-Nijenhuis operator<formula xmlns="http://xml-maiden.com">R<sub>E</sub> =z<sub>1</sub>dz<sub>1</sub> ⊗ D<sub>1</sub> −dz<sub>1</sub> ⊗ D<sub>4</sub> +z<sub>2</sub>dz<sub>2</sub> ⊗ D<sub>2</sub> +dz<sub>2</sub> ⊗ D<sub>3</sub> + z<sub>1</sub>dz<sub>3</sub> ⊗ D<sub>3</sub><line/>+ e<sup>z<sub>3</sub> − z<sub>4</sub></sup>dz<sub>3</sub> ⊗ D<sub>2</sub> +z<sub>2</sub>dz<sub>4</sub> ⊗ D<sub>4</sub> − e<sup>z<sub>3</sub> − z<sub>4</sub></sup>dz<sub>4</sub> ⊗ D<sub>1</sub></formula>(compare with <a href="#r3">[3]</a>)</div><div class="summary"><strong class="cap">summary. </strong>In summary let us note that the non-Noether symmetries form quite interestingclass of symmetries of Hamiltonian dynamical system and lead not only toa number of conservation laws (that under certain conditions ensure integrability),but also enrich the geometry of the phase space by endowing it with several importantstructures, such as Lax pair, bicomplex,bi-Hamiltonian structure, Frölicher-Nijenhuis operators etc.The present paper attempts to emphasize deep relationship between differentconcepts used in construction of conservation laws and non-Noether symmetry.</div><div class="acknowledgements"><strong class="cap">acknowledgements. </strong> Author is grateful to Zakaria Giunashvili, George Jorjadze andMichael Maziashvili for constructive discussions and help.This work was supported by INTAS (00-00561).</div> <h2 class="references">References</h2><ol class="references"><li>	<span class="who">M. Crampin, W. Sarlet, G. Thompson</span>	<span class="what">Bi-differential calculi and bi-Hamiltonian systems</span>	<span class="where">J. of Phys. A: Math. Gen. 33 No. 22, L177-180</span>	<span class="when">2000</span></li><li>	<span class="who">A. Dimakis, F. Müller-Hoissen</span>	<span class="what">Bicomplexes and integrable models</span>	<span class="where">nlin.SI/0006029</span>	<span class="when">2000</span></li><li>	<span class="who">R. Fernandes</span>	<span class="what">On the master symmetries and bi-Hamiltonian structure of the Toda lattice</span>	<span class="where">J. of Phys. A: Math. Gen. 26, 3793-3803</span>	<span class="when">1993</span></li><li>	<span class="who">M. Lutzky</span>	<span class="what">New derivation of a conserved quantity for Lagrangian systems</span>	<span class="where">J. of Phys. A: Math. Gen. 15, L721-722</span>	<span class="when">1998</span></li></ol></body></html>
