\documentclass[12pt]{article}
\setlength{\topmargin}{-11mm}
\setlength{\headheight}{6mm}
\setlength{\headsep}{8mm} 
\setlength{\oddsidemargin}{4mm}
\setlength{\evensidemargin}{4mm}
\setlength{\textheight}{220mm}
\setlength{\textwidth}{155mm}
\setlength{\footskip}{18mm}


\setcounter{secnumdepth}{2}
\makeatletter
\makeatother
\renewcommand{\thefootnote}{\fnsymbol{footnote}}

\pagestyle{plain}

\begin{document}


\baselineskip 6mm
\begin{flushright}
{\tt TMI-00-1 \\
     December 2000}
\end{flushright}
\thispagestyle{empty}
\vspace*{5mm}
\begin{center}
   {\Large {\bf Can Subquark Model explain CP violation ?}}
\vspace*{20mm} \\
  {\sc Takeo~~Matsushima}
 \footnote[1]
  {
   $\dagger$ e-mail : mtakeo@eken.phys.nagoya-u.ac.jp
 }
\vspace{5mm} \\
  \sl{1-17-1 Azura Ichinomiya,$^\dagger$}\\
  \it{Aichi-Prefecture, Japan}
\end{center}

\vspace{15mm}  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{center}
{\bf Abstract} \\
\vspace{3mm}
\begin{minipage}[t]{120mm}
\baselineskip 5mm
{\small
We study CP violation phenomena with the subquark model 
which is previously proposed by us. 
Our subquark model claims that ``direct'' and 
 ``indirect'' CP violation originate from different sources, which 
 are both belonging to subquark dynamics.
We discuss KTeV experiment and  CP-asymmetry of $(B_{d}\to{J/{\psi}K_{s}})$-
 and $(B_{d}\to{\pi}{\pi})$-decay.   
 This model predicts :
 the phases of indirect CP violation : ${\theta}_K={\theta}_D={\theta}_
  {B_s}={\theta}_{T_c}\simeq(1/2){\theta}_{B_d}\simeq
  (1/2){\theta}_{T_u}$.
   This model also predicts the CKM matrix elements :
  $|V_{ts}|=2.6\times{10^{-2}}$,$|V_{td}|=1.4\times{10^{-3}}$; 
 the neutral pseudoscalar meson mass differences : ${\Delta}
 M_D\approx10^{-14}$ GeV, 
 ${\Delta}M_{B_s}\approx10^{-11}$
  GeV, ${\Delta}M_{T_u}\approx10^{(-10\sim{-9})}$ GeV and 
  ${\Delta}M_{T_c}\approx10^{(-8\sim{-7})}$ GeV  
 %  and also predict the existence of 
 % the scalar partner of ``composite'' $Z^{0}$-boson 
 % and the mass value of the former is larger than that of 
 % the latter in contrast with $m_{\rho}>m_{\pi}$.
 % \hspace{1cm}
}
\end{minipage}


\end{center}

  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


\newpage
\baselineskip 18pt
\section{Introduction}
\hspace*{\parindent}
  The discovery of the top-quark[1] has finally confirmed the 
 existence of three quark-lepton symmetric generations.
 So far the standard $SU(2)_{L}\otimes{U(1)}$ model
 (denoted by SM)
 has successfully explained various experimental evidences.
 Nevertheless, as is well known, the SM is not regarded as
 the final theory because it has many arbitrary parameters, e.g.,
 quark and lepton masses, quark-mixing parameters and the Weinberg 
 angle, etc. . Therefore it is meaningful to investigate the
 origins of these parameters and the relationship among them.
  In order to overcome such problems some attempts have done, e.g.,
 Grand Unification Theory (GUT), Supersymmetry, Composite model, etc. .
 In the GUT scenario quarks and leptons are elementary fields
 in general. On the contrary in the composite scenario they are
 literally the composite objects constructed from the elementary
 fields (so called ``preon''). The lists of various related 
 works are in ref.[2]. If quarks and leptons are 
 elementary, in order to solve the above problems it is necessary 
 to introduce some external relationship or symmetries among them. 
  On the other hand the composite models have ability to explain
 the origin of these parameters in terms of the substructure 
 dynamics of quarks and leptons. Further, the composite scenario 
 naturally leads us to the thought that the intermediate vector bosons
 of weak interactions (${\bf W,Z}$) are not elementary gauge fields
 (which is so in the SM)
 but composite objects constructed from preons (same as ${\bf \rho}$
 -meson from quarks). Many studies based on such conception have 
 done after Bjorken's[3] and Hung and Sakurai's[4] suggestions of the 
 alternative way to unified weak-electromagnetic gauge theory[5-11].
 In this scheme the weak interactions are regarded as the effective
 residual interactions among preons. The fundamental fields for
 intermediate forces are massless gauge fields belonging to
 some gauge groups and they confine preons into singlet states
 to build quarks and leptons and ${\bf W,Z}$.
\par
  The conception of our model is that the fundamental interacting
 forces are all originated from massless gauge fields belonging
 to the adjoint representations of some gauge groups which have 
 nothing to do with the spontaneous breakdown and that the 
 elementary matter fields are only one kind of spin-1/2 preon
 and spin-0 preon carrying common ``$e/6$'' electric charge ($e>0$).
 Quarks, leptons and ${\bf W,Z}$ are all composites of them and
 usual weak interactions are regarded as effective residual
 interactions. Based on such idea we consider the underlying gauge 
 theory in section 2 and composite model in section 3. In section
  4 we discuss about ${\Delta}F=1$ phenomena such as quark-flavor-mixing
  and direct CP violation. In section 5 ${\Delta}F=2$ phenomena 
  such as $P^0$-$\overline{P^0}$ mixing and indirect CP violation
   are investigated.
   
 \section{  Gauge theory inspiring quark-lepton composite scenario}
 \hspace*{\parindent}
 
  In our model the existence of fundamental matter fields (preon)
 are inspired by the gauge theory with Cartan connections[14]. 
 Let us briefly summarize the basic features of that. Generally 
 gauge fields, including gravity, are considered as geometrical  
 objects, that is, connection coefficients of principal fiber 
 bundles. It is said that there exist some different points 
 between Yang-Mills gauge theories and gravity, though both 
 theories commonly possess fiber bundle structures. The latter has 
 the fiber bundle related essentially to 4-dimensional space-time
 freedoms but the former is given, in an ad hoc way, the one with 
 the internal space which has nothing to do with the space-time 
 coordinates. In case of gravity it is usually considered that 
 there exist ten gauge fields, that is, six spin connection fields 
 in $SO(1,3)$ gauge group and four vierbein fields in $GL(4,R)$ 
 gauge group from which the metric tensor ${\bf g}^{{\mu}{\nu}}$ 
 is constructed in a bilinear function of them. Both altogether 
 belong to  Poincar\'e group $ISO(1,3)=SO(1,3)\otimes{R}^4$ 
 which is semi-direct product. In this scheme spin connection 
 fields and vierbein fields are independent but only if there is 
 no torsion, both come to have some relationship. Seeing this, 
 $ISO(1,3)$ gauge group theory has the logical weak point not to  
 answer how two kinds of gravity fields are related to each other 
 intrinsically.
\par   
In the theory of Differential Geometry, S.Kobayashi has investigated 
 the theory of ``Cartan connection''[15]. This theory, in fact, 
 has ability to reinforce the above weak point. The brief 
 recapitulation is as follows. Let $E(B_n,F,G,P)$ be a fiber bundle
 (which we call Cartan-type bundle) associated with a principal 
 fiber bundle $P(B_n,G)$ where $B_n$ is a base manifold with  
 dimension ``$n$'', $G$ is a structure group, $F$ is a fiber space 
 which is homogeneous and diffeomorphic with $G/G'$ where $G'$ 
 is a subgroup of $G$. Let $P'=P'(B_n,G')$ be a principal 
 fiber bundle, then $P'$ is a subbundle of $P$.  
 Here let it be possible to decompose the Lie algebra ${\bf g}$ of 
 $G$ into the subalgebra ${\bf g}'$ of $G'$ and a vector space 
 ${\bf f}$ such as : 
 \begin{equation}
     {\bf g}={\bf g}'+{\bf f},\hspace{1cm}{\bf g}'\cap{\bf f}=0,      \label{1}
 \end{equation}
 
 \begin{equation}
     [{\bf g'},{\bf g'}]\subset {\bf g'},                             \label{2}
 \end{equation}
 
 \begin{equation}
     [{\bf g'},{\bf f}]\subset {\bf f},                               \label{3}
 \end{equation}
 
 \begin{equation}
     [{\bf f},{\bf f}]\subset {\bf g'},                               \label{4}  
 \end{equation}
 where $dim{\bf f}=dimF=dimG-dimG'=dimB_n=n$. 
 The homogeneous space $F=G/G'$ is said to be ``weakly reductive'' 
 if there exists a vector space ${\bf f}$ satisfying Eq.(1) and (3).
 Further $F$ satisfying Eq(4) is called ``symmetric space''. 
 Let ${\bf \omega}$ denote the connection form of $P$ and 
 $\overline{\bf \omega}$ be the restriction of ${\bf \omega}$ to
 $P'$. Then $\overline{\bf \omega}$ is a ${\bf g}$-valued linear
 differential 1-form and we have :
 \begin{equation}
     {\bf \omega}=g^{-1}\overline{\bf \omega}g+g^{-1}dg,              \label{5}
 \end{equation}
 where $g\in{G}$, $dg\in{T_g(G)}$. ${\bf \omega}$ is called the  
 form of ``Cartan connection'' in $P$. 
\par
 Let the homogeneous space $F=G/G'$ be weakly reductive. The 
 tangent space $T_O(F)$ at $o\in{F}$ is isomorphic with ${\bf f}$
 and then $T_O(F)$ can be identified with ${\bf f}$ and also 
 there exists a linear ${\bf f}$-valued differential 
 1-form(denoted by ${\bf \theta}$) which we call the 
 ``form of soldering''. Let ${\bf \omega}'$ 
 denote a ${\bf g}'$-valued 1-form in $P'$, we have :
 \begin{equation}
    \overline{\bf \omega}={\bf \omega}'+{\bf \theta}.                  \label{6}
 \end{equation}
 The dimension of vector space ${\bf f}$ and the dimension of  
 base manifold $B_n$ is the same ``$n$'', and then ${\bf f}$  
 can be identified with the tangent space of $B_n$ at the same 
 point in $B_n$ and ${\bf \theta}$s work as $n$-bein fields. In this 
 case  ${\bf \omega}'$ and ${\bf \theta}$ unifyingly belong to 
 group $G$. Here let us call such a mechanism ``Soldering Mechanism''.
\par
 Drechsler has found out the useful aspects of this theory and 
 investigated a gravitational gauge theory based on 
 the concept of the Cartan-type bundle equipped with the 
 Soldering Mechanism[16]. He considered $F=SO(1,4)/SO(1,3)$ 
 model. Homogeneous space $F$ with $dim=4$ solders 4-dimensional 
 real space-time. The Lie algebra of $SO(1,4)$ corresponds to 
 ${\bf g}$ in Eq.(1), that of $SO(1,3)$ corresponds to ${\bf g}'$
 and ${\bf f}$ is 4-dimensional vector space. The 6-dimensional 
 spin connection fields are ${\bf g}'$-valued objects and vierbein 
 fields are ${\bf f}$-valued, both of which are unified into 
 the members of $SO(1,4)$ gauge group. We can make the metric 
 tensor ${\bf g}^{{\mu}{\nu}}$ as a bilinear function of 
 ${\bf f}$-valued vierbein fields. Inheriting Drechsler's study 
 the author has investigated the quantum theory of gravity[14]. 
 The key point for this purpose is that $F$ is a symmetric 
 space because ${\bf f}$s are satisfied with Eq.(4). 
 Using this symmetric nature we can pursue making a 
 quantum gauge theory, that is, constructing ${\bf g}'$-valued 
 Faddeev-Popov ghost, anti-ghost, gauge fixing, gaugeon 
 and its pair field as composite fusion fields of 
 ${\bf f}$-valued gauge fields by use 
 of Eq.(4) and also naturally inducing BRS-invariance. 
\par
 Comparing such a scheme of gravity, let us consider 
 Yang-Mills gauge theories. Usually when we make the Lagrangian 
 density ${\cal L}=tr({\cal F}\wedge{\cal F}^{\ast})$ 
 (${\cal F}$ is a field strength), we must 
 borrow a metric tensor ${\bf g}^{{\mu}{\nu}}$ from gravity to 
 get ${\cal F}^{\ast}$ and also for Yang-Mills gauge fields to 
 propagate in the 4-dimensional real space-time. This seems to 
 mean that ``there is a hierarchy between gravity and other three 
 gauge fields (electromagnetic, strong, and weak)''. But is it   
 really the case ? As an alternative thought we can think that  
 all kinds of gauge fields are ``equal''. Then it would be natural 
 for the question ``What kind of equality is that ?'' to arise. 
 In other words, it is the question that ``What is the minimum 
 structure of the gauge mechanism which four kinds of forces 
 are commonly equipped with ?''. For answering this question, 
 let us make a assumption : ``\begin{em}Gauge fields 
 are Cartan connections equipped with Soldering 
 Mechanism\end{em}.'' In this meaning all 
 gauge fields are equal. If it is the case three gauge fields 
 except gravity are also able to have their own metric 
 tensors and to propagate in the real space-time without 
 the help of gravity. Such a model has already investigated 
 in ref.[14]. 
\par
 Let us discuss them briefly. It is found that there 
 are four types of sets of classical groups with small dimensions 
 which admit Eq.(1,2,3,4), that is, $F=SO(1,4)/SO(1,3)$, 
 $SU(3)/U(2)$, $SL(2,C)/GL(1,C)$ and $SO(5)/SO(4)$ with 
 $dimF=4$[17]. Note that the quality of ``$dim\hspace{1mm}4$''
  is very important because it guarantees $F$ to solder 
 to 4-dimensional real space-time and all gauge fields 
 to work in it. The model of $F=SO(1,4)/SO(1,3)$ for gravity 
 is already mentioned. 
 Concerning other gauge fields, it seems to be appropriate  
 to assign $F=SU(3)/U(2)$ to QCD gauge fields, 
 $F=SL(2,C)/GL(1,C)$ to QED gauge fields and 
 $F=SO(5)/SO(4)$ to weak interacting gauge fields. Some 
 discussions concerned are following. In general, matter 
 fields couple to ${\bf g}'$-valued gauge fields. As for   
 QCD, matter fields couple to the gauge fields of $U(2)$
 subgroup but $SU(3)$ contains, as is well known, three types 
 of $SU(2)$ subgroups and then after all they couple 
 to all members of $SU(3)$ gauge fields. In case of QED,  
 $GL(1,C)$ is locally isomorphic with $C^1\cong{U(1)}\otimes{R}$. 
 Then usual Abelian gauge fields are assigned to $U(1)$ 
 subgroup of $GL(1,C)$. Georgi and Glashow suggested that  
 the reason why the electric charge is quantized comes from 
 the fact that $U(1)$ electromagnetic gauge group is 
 a unfactorized subgroup of $SU(5)$[18]. Our model is 
 in the same situation because $GL(1,C)$ a unfactorized 
 subgroup of $SL(2,C)$. For usual electromagnetic $U(1)$ 
 gauge group, the electric charge unit ``$e$''$(e>0)$ is for   
 $one\hspace{2mm} generator$ of $U(1)$ but in case of $SL(2,C)$ 
 which has $six\hspace{2mm} generators$, the minimal 
 unit of electric charge 
 shared per one generator must be ``$e/6$''. This suggests 
 that quarks and leptons might have the substructure 
 simply because $e,\hspace{1mm}2e/3,\hspace{1mm}e/3>e/6$.
  Finally as for weak interactions we adopt 
 $F=SO(5)/SO(4)$. It is well known that $SO(4)$ is 
 locally isomorphic with $SU(2)\otimes{SU(2)}$. Therefore 
 it is reasonable to think it the left-right symmetric 
 gauge group : $SU(2)_L\otimes{SU(2)}_R$. As two $SU(2)$s are 
 direct product, it is able to have coupling constants 
 (${\bf g}_L,{\bf g}_R$) independently. 
 This is convenient to explain the fact of the disappearance 
 of right-handed weak interactions in the low-energy region. 
 Possibility of composite structure of quarks and leptons 
 suggested by above $SL(2,C)$-QED would introduce 
 the thought that the usual left-handed weak interactions 
 are intermediated by massive composite vector bosons as 
 ${\bf \rho}$-meson in QCD and that they are residual 
 interactions due to substructure dynamics of quarks 
 and leptons. The elementary massless gauge fields ,``\begin{em}
 as connection fields''\end{em}, relate intrinsically to the
  structure of the real space-time manifold but on the other hand 
 the composite vector bosons have nothing to do with it. 
 Considering these discussions, we set the assumption : 
 ``\begin{em}All kinds of gauge fields are elementary massless 
 fields, belonging to spontaneously unbroken 
 $SU(3)_C\otimes{SU(2)}_L\otimes{SU(2)}_R\otimes{U(1)}_{e.m}$ 
 gauge group and quarks and leptons and {\bf W, Z} are all 
 composite objects of the elementary matter fields\end{em}.''

\section{Composite model}
 \hspace*{\parindent}
 Our direct motivation towards compositeness of quarks and leptons 
 is one of the results of the arguments in Sect.2, that is, 
 $e,\hspace{1mm}2e/3,\hspace{1mm}e/3>e/6$. 
 However, other several phenomenological 
 facts tempt us to consider a composite model, e.g.,  
 repetition of generations, quark-lepton parallelism of weak 
 isospin doublet structure, quark-flavor-mixings, etc..
 Especially Bjorken[3]'s and Hung and Sakurai[4]'s suggestion 
 of an alternative to unified weak-electromagnetic gauge theories
 have invoked many studies of composite models including 
 composite weak bosons[5-11]. Our model is in the line of   
 those studies. There are two ways to make composite 
 models, that is, ``Preons are all fermions.'' or ``Preons are   
  both fermions and bosons (denoted by FB-model).'' 
 The merit of the former is that it can avoid the problem 
 of a quadratically divergent self-mass of elementary scalar 
 fields. However, even in the latter case such a disease 
 is overcome if both fermions and bosons are the 
 supersymmetric pairs, both of which carry the same quantum 
 numbers except the nature of Lorentz transformation (
 spin-1/2 or spin-0)[19]. Pati and Salam have suggested    
 that the construction of a neutral composite object 
 (neutrino in practice) needs both kinds of preons, fermionic 
 as well as bosonic, if they carry the same charge for the 
 Abelian gauge or belong to the same 
 (fundamental) representation for the non-Abelian gauge[20].  
 This is a very attractive idea for constructing the minimal 
 model. Further, according to the representation theory of  
 Poincar\'e group both integer and half-integer 
 spin angular momentum occur equally for massless particles[21], 
 and then if nature chooses ``fermionic monism'', 
 there must exist the additional special reason to select it. 
 Therefore in this point also, the thought of the FB-model 
 is minimal. Based on such considerations we propose 
 a FB-model of \begin{em}``only one kind of spin-1/2 
 elementary field 
 (denoted by $\Lambda$) and of spin-0 elementary field
 (denoted by $\Theta$)''\end{em} (preliminary 
 version of this model has appeared in Ref.[14]). 
 Both have the same electric charge of ``$e/6$''  
 (Maki has first proposed the FB-model with the minimal 
 electric charge $e/6$. \cite{22})
\renewcommand{\thefootnote}{\arabic{footnote}}
\footnote{The notations of $\Lambda$ and $\Theta$ are 
 inherited from those in Ref.[22]. After this we 
 call $\Lambda$ and $\Theta$ ``Primon'' named by Maki 
 which means ``primordial particle''[22].}
 and the same transformation properties of the 
 fundamental representation ( 3, 2, 2) under the spontaneously unbroken   
 gauge symmetry of $SU(3)_C\otimes{SU(2)_L}\otimes{SU(2)_R}$
 (let us call $SU(2)_L\otimes{SU(2)_R}$ ``hypercolor gauge symmetry''). 
 Then $\Lambda$ and $\Theta$ come into the supersymmetric pair 
 which guarantees 'tHooft's naturalness condition[23]. 
 The $SU(3)_C$, $SU(2)_L$ and $SU(2)_R$ gauge fields 
 cause the confining forces with confining energy scales of 
 $\Lambda_c<< \Lambda_L<(or \cong) \Lambda_R$ (Schrempp and Schrempp 
 discussed this issue elaborately in Ref.[11]). 
 Here we call positive-charged 
 primons ($\Lambda$, $\Theta$) ``$matter$'' and negative-charged 
 primons ($\overline\Lambda$, $\overline\Theta$) ``$anti
 matter$''. Our final goal is to build quarks, leptons 
 and ${\bf W, Z}$ from $\Lambda$ ($\overline\Lambda$)
  and $\Theta$ ($\overline\Theta$).   
 Let us discuss that scenario next.
 \par
 At the very early stage of the development of the universe, 
 the matter fields ($\Lambda$, $\Theta$) and their antimatter fields 
 ($\overline{\Lambda}$, $\overline{\Theta}$) must have broken out  
 from the vacuum. After that they would have combined with each  
 other as the universe was expanding. That would be the 
 first step of the existence of composite matters. 
 There are ten types of them : 
\newline
 \hspace*{2mm}$spin\displaystyle{\frac{1}{2}}$\hspace{2.4cm}$spin0$
 \hspace{2.7cm}$e.m.charge$
 \hspace{1.2cm}$Y.M.representation$
{
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1}
 \setcounter{equation}{0} 
 \renewcommand{\theequation}{\theenumi\alph{equation}}
 \begin{eqnarray}
     \Lambda\Theta\hspace{2.5cm}\Lambda\Lambda,
     \Theta\Theta\hspace{3.1cm}
     \frac{1}{3}e\hspace{1.8cm}(\overline{3},1,1)\hspace{2mm}
     (\overline{3},3,1)\hspace{2mm}(\overline{3},1,3),\\
     \Lambda\overline\Theta,\overline\Lambda\Theta\hspace{2cm}
     \Lambda\overline\Lambda,\Theta\overline\Theta\hspace{3.2cm}
     0\hspace{2.0cm}(1,1,1)\hspace{2mm}(1,3,1)\hspace{2mm}(1,1,3),\\
     \overline\Lambda\overline\Theta\hspace{2.5cm}\overline\Lambda
     \overline\Lambda,\overline\Theta\overline\Theta\hspace{2.7cm}
     -\frac{1}{3}e\hspace{1.7cm}(3,1,1)\hspace{2mm}(3,3,1)
     \hspace{2mm}(3,1,3)\hspace{1mm}.                                 \label{7}
 \end{eqnarray}
 \setcounter{equation}{\value{enumi}}}
 In this step the confining forces are, in kind, in $SU(3)\otimes
 {SU(2)}_L\otimes{SU(2)}_R$ gauge symmetry but the 
 $SU(2)_L\otimes{SU(2)}_R$ confining forces must be main 
 because of the energy scale of $\Lambda_L,\Lambda_R>>\Lambda_c$
 and then the color gauge coupling $\alpha_s$ and e.m. coupling 
 constant $\alpha$ are negligible. As is well known,
 the coupling constant of $SU(2)$ confining force are 
 characterized 
 by $\varepsilon_i=\sum_a\sigma_p^a\sigma_q^a$,where 
 ${\sigma}s$ are $2\times2$ matrices of $SU(2)$, $a=1,2,3$, 
 $p,q=\Lambda,\overline\Lambda,\Theta,\overline
 \Theta$, $i=0$ for singlet and $i=3$ for triplet. 
 They are calculated as 
 $\varepsilon_0=-3/4$ which causes the attractive force and 
 and $\varepsilon_3=1/4$ causing the repulsive force. Next, 
 $SU(3)_C$ octet and sextet states are repulsive but singlet,
 triplet and antitriplet states are attractive and then the formers 
 are disregarded. Like this, two primons are confined into composite 
 objects in more than  one singlet state of any 
 $SU(3)_C,SU(2)_L,SU(2)_R$. Note that three primon systems 
 cannot make the singlet states of $SU(2)$. Then we omit them. 
 In Eq.(7b), the $(1,1,1)$-state is the ``most attractive channel''. 
 Therefore $(\Lambda\overline\Theta)$, $(\overline\Lambda
 \Theta)$, $(\Lambda\overline\Lambda)$ and $(\Theta\overline
 \Theta)$ of $(1,1,1)$-states with neutral e.m. charge must 
 have been most abundant in the universe. 
 Further $(\overline{3},1,1)$- and 
 $(3,1,1)$-states in Eq.(7a,c) are next attractive. 
 They presumably go into $\{(\Lambda\Theta)(\overline\Lambda
 \overline\Theta)\}, \{(\Lambda\Lambda)(\overline\Lambda 
 \overline\Lambda)\}$, etc. of $(1,1,1)$-states 
 with neutral e.m. charge.
 These objects may be the candidates for the ``cold dark matters'' 
 if they have even tiny masses. 
 It is presumable that the ratio of the quantities between 
 the ordinary matters and the dark matters 
 firstly depends on the color and hypercolor charges 
 and the quantity of the latter much excesses that of the former 
 (maybe the ratio is more than $1/(3\times3)$). 
 Finally the $(*,3,1)$-and $(*,1,3)$-states are remained 
 ($*$ is $1,3,\overline{3}$). 
 They are also stable
 because $|\varepsilon_0|>|\varepsilon_3|$. They are, so to say,  
 the ``intermediate clusters'' towards constructing 
 ordinary matters(quarks,leptons and ${\bf W,Z}$).
\footnote{Such thoughts have been proposed by Maki in Ref.[22]}
  Here we call such intermediate clusters ``subquarks'' 
  and denote them as follows :
\newline
\hspace*{6.5cm}$Y.M.representation$\hspace{1.5cm}$spin$
 \hspace{0.5cm}$e.m.charge$
{
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1}
 \setcounter{equation}{0} 
 \renewcommand{\theequation}{\theenumi\alph{equation}}
 \begin{eqnarray}
    {\bf \alpha}&=&(\Lambda\Theta)\hspace{2.2cm}{\bf \alpha}_L:
    (\overline{3},3,1)\hspace{3mm}{\bf \alpha}_R:(\overline{3},1,3)
    \hspace{1.2cm}\frac{1}{2}\hspace{1.2cm}\frac{1}{3}e\\
    {\bf \beta}&=&(\Lambda\overline\Theta)\hspace{2.2cm}{\bf \beta}_L: 
    (1,3,1)\hspace{3mm}{\bf \beta}_R:(1,1,3)\hspace{1.3cm}\frac{1}{2}
    \hspace{1.3cm}0\\
    {\bf x}&=&(\Lambda\Lambda,\hspace{2mm}
    \Theta\Theta)\hspace{1.2cm}{\bf x}_L:
    (\overline{3},3,1)\hspace{3mm}{\bf x}_R:(\overline{3},1,3)
    \hspace{1.3cm}0\hspace{1.3cm}\frac{1}{3}e\\
    {\bf y}&=&(\Lambda\overline\Lambda\hspace{2mm}\Theta\overline\Theta)
    \hspace{1.4cm}{\bf y}_L:(1,3,1)\hspace{3mm}{\bf y}_R:(1,1,3)
    \hspace{1.3cm}0\hspace{1.3cm}0,                                 \label{8}
 \end{eqnarray}
 \setcounter{equation}{\value{enumi}}}
and there are also their antisubquarks[9].
\footnote{The notations of ${\bf \alpha}$,${\bf \beta}$,
 ${\bf x}$ and ${\bf y}$ are inherited from those in Ref.[9] 
 written by Fritzsch and Mandelbaum, because ours is, 
 in the subquark level, similar to theirs
 with two fermions and two bosons.
 R. Barbieri, R. Mohapatra and A. Masiero proposed 
 the similar model[9].}
\par
 Now we come to the step to build quarks and leptons. The gauge 
 symmetry of the confining forces in this step is also 
 $SU(2)_L\otimes{SU(2)}_R$ because the subquarks are in the
 triplet states of $SU(2)_{L,R}$ and then they are combined 
 into singlet states by the decomposition of $3\times3=1+3+5$
 in $SU(2)$. We make the first generation of quarks and leptons 
 as follows :
\newline
 \hspace*{8cm}$e.m.charge$\hspace{1.2cm}$Y.M.representation$
 {
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1}
 \setcounter{equation}{0}
 \renewcommand{\theequation}{\theenumi\alph{equation}} 
 \begin{eqnarray}
       <{\bf u}_h|&=&<{\bf \alpha}_h{\bf x}_h|\hspace{3.4cm}
                     \frac{2}{3}e\hspace{2,95cm}(3,1,1)\\
        <{\bf d}_h|&=&<\overline{\bf \alpha}_h\overline{\bf x}_h
                     {\bf x}_h|\hspace{2.5cm}-\frac{1}{3}e
                     \hspace{2.9cm}(3,1,1)\\
        <{\bf \nu}_h|&=&<{\bf \alpha}_h\overline{\bf x}_h|
                     \hspace{3.5cm}0\hspace{3.2cm}(1,1,1)\\
        <{\bf e}_h|&=&<\overline{\bf \alpha}_h\overline{\bf x}_h
                     \overline{\bf x}_h|\hspace{2.6cm}-e
                     \hspace{3.2cm}(1,1,1),                      \label{9}
 \end{eqnarray}
 \setcounter{equation}{\value{enumi}}}
 where $h$ stands for $L$(left handed) or $R$(right handed)[5].
\footnote{Subquark configurations in Eq.(9) are essentially the 
 same as those in Ref.[5] written by Kr\' olikowski, 
 who proposed the model of one fermion and one boson 
 with the same e.m. charge $e/3$}.
 Here we note that ${\bf \beta}$ and ${\bf y}$ do not appear.  
 In practice ($({\bf \beta}{\bf y}):(1,1,1)$)-particle 
 is a candidate for neutrino. But as Bjorken has pointed
 out[3], non-vanishing charge radius of neutrino is necessary
 for obtaining the correct low-energy effective weak interaction
 Lagrangian[11]. Therefore ${\bf \beta}$ is assumed not to 
 contribute to forming ordinary quarks and leptons.
 However $({\bf \beta}{\bf y})$-particle may be a candidate
 for ``sterile neutrino''. 
 Presumably composite (${\bf \beta}$${\bf \beta}$)-;
 (${\bf \beta}\overline{\bf \beta}$)-;($\overline{\bf \beta}
 \overline{\bf \beta}$)-states may go into the dark matters. 
 It is also noticeable that in this model the leptons have 
 finite color charge radius and then $SU(3)$ gluons interact 
 directly with the leptons at energies of the order of, or 
 larger than $\Lambda_{L}$ or $\Lambda_{R}$[19]. 
 \par
 Concerning the confinements of primons and subquarks, the  
 confining forces of two steps are in the same spontaneously 
 unbroken $SU(2)_L\otimes{SU(2)}_R$ gauge symmetry.
 It is known that the running coupling constant 
 of the $SU(2)$ gauge theory satisfies the following equation : 
 {
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1}
 \setcounter{equation}{0}
 \renewcommand{\theequation}{\theenumi\alph{equation}}
 \begin{eqnarray}
       \frac{1}{\alpha^{a}_{W}(Q_{1}^{2})}&=&
                \frac{1}{\alpha^{a}_{W}(Q_{2}^{2})}
                +b_{2}(a)\ln\left(\frac{Q_{1}^{2}}{Q_{2}^{2}}\right),\\
       b_{2}(a)&=&\frac{1}{4\pi}\left(\frac{22}{3}-\frac{2}{3}
       \cdot{N}_{f}-\frac{1}{12}\cdot{N}_{s}\right),                 \label{10}
 \end{eqnarray}
 \setcounter{equation}{\value{enumi}}}
 where $N_f$ and $N_s$ are the numbers of fermions and scalars   
 contributing to the vacuum polarizations,
 ($a=q$) for the confinement subquarks in quark and ($a=sq$)
 for confinement primons in subquark.
 We calculate $b_2(q)=0.35$ which comes from that the number of 
 confined fermionic subquarks are $4$ (${\bf \alpha}_{i}, i=1,2,3$
 for color freedom, ${\bf \beta}$) and $4$ for bosons (${\bf x}_i,
 {\bf y}$) contributing to the vacuum polarization, and 
 $b_2(sq)=0.41$ which  is calculated with three kinds of 
 $\Lambda$ and $\Theta$ owing to three color freedoms. 
 Experimentary it is reported that $\Lambda_q>1.8$ TeV(CDF exp.)
 or $\Lambda_q>2.4$ TeV(D0 exp.)[12].
 Extrapolations of $\alpha^{q}_{W}$ and $\alpha^{sq}_{W}$ to near 
 Plank scale are expected to converge to the same point and then
  tentatively, setting 
 $\Lambda_q=5$ TeV, $\alpha^{q}_{W}(\Lambda_q)=\alpha^{sq}_{W}
 (\Lambda_{sq})=\infty$, we get $\Lambda_{sq}=10^3\Lambda_q$,   
\par
 Next let us see the higher generations. Harari and Seiberg 
 have stated that the orbital and radial excitations 
 seem to have the wrong energy scale ( order of 
 $\Lambda_{L,R}$) and then the most likely type of 
 excitations is the addition of preon-antipreon pairs[6,25].
 In our model the essence of generation is like ``$isotope$"
 in case of nucleus.
 Then using ${\bf y}_{L,R}$ in Eq.(8,d) 
 we construct them as follows :
 {  
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1} 
 \setcounter{equation}{0}
 \renewcommand{\theequation}{\theenumi\alph{equation}}
 \begin{eqnarray}
 &&\left\{
 \begin{array}{lcl}
          <{\bf c}\hspace{0.5mm}|
                  &=&<{\bf \alpha}{\bf x}{\bf y}|\\
          <{\bf s}\hspace{0.5mm}|
                  &=&<\overline{\bf \alpha}\overline{\bf x}
                      {\bf x}{\bf y}|,
 \end{array} 
 \right.
 \hspace{6mm}
 \left\{
 \begin{array}{lcl}
          <{\bf \nu_\mu}\hspace{0.3mm}|
                       &=&<{\bf \alpha}\overline{\bf x}{\bf y}|\\
          <{\bf \mu}\hspace{2mm}|&=&<\overline{\bf \alpha}\overline{\bf x}
                                    \overline{\bf x}{\bf y}|,
 \end{array}
 \right.
 \hspace{0.6cm}\mbox{2nd generation}\\
 &&\left\{
 \begin{array}{lcl}
          <{\bf t}\hspace{0.5mm}|
                   &=&<{\bf \alpha}{\bf x}{\bf y}{\bf y}|\\
          <{\bf b}|&=&<\overline{\bf \alpha}\overline{\bf x}
                      {\bf x}{\bf y}{\bf y}|,
 \end{array}
 \right.
 \hspace{0.3cm}
 \left\{
 \begin{array}{lcl}
          <{\bf \nu_\tau}\hspace{0.5mm}|
                &=&<{\bf \alpha}\overline{\bf x}
                              {\bf y}{\bf y}|\\
          <{\bf \tau}\hspace{2mm}|&=&<\overline{\bf \alpha}
          \overline{\bf x}\overline{\bf x}{\bf y}{\bf y}|,
 \end{array}
 \right.
 \hspace{4mm}\mbox{3rd generation}                               \label{11}
 \end{eqnarray} 
 \setcounter{equation}{\value{enumi}}}
 where the suffix $L,R$s are omitted for brevity. 
 We can also make vector and scalar particles with (1,1,1) : 
 {
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1} 
 \setcounter{equation}{0}
 \renewcommand{\theequation}{\theenumi\alph{equation}}
 \begin{eqnarray}&&\left\{
 \begin{array}{lcl}
             <{\bf W}^+|&=&<{\bf \alpha}^\uparrow{\bf \alpha}^
                             \uparrow{\bf x}|\\
             <{\bf W}^-|&=&<\overline{\bf \alpha}^\uparrow
                           \overline{\bf \alpha}^\uparrow\overline{\bf x}|,
 \end{array}
 \right.\hspace{6mm}
 \left\{
 \begin{array}{lcl}
             <{\bf Z}_1^0|&=&<{\bf \alpha}^\uparrow\overline
                             {\bf \alpha}^\uparrow|\\
             <{\bf Z}_2^0|&=&<{\bf \alpha}^\uparrow\overline
                             {\bf \alpha}^\uparrow{\bf x}\overline{\bf x}|, 
 \end{array} 
 \right.\hspace{0.75cm}\mbox{Vector}\\
 &&\left\{
 \begin{array}{lcl}
             <\hspace{2mm}{\bf S}^+\hspace{0.5mm}|
                  &=&<{\bf \alpha}^\uparrow{\bf \alpha}^
                           \downarrow{\bf x}|\\
             <\hspace{2mm}{\bf S}^-\hspace{0.5mm}|
                  &=&<\overline{\bf \alpha}^
                          \uparrow\overline{\bf \alpha}^\downarrow{\bf x}|,
 \end{array}
 \right.
 \hspace{6mm}
 \left\{
 \begin{array}{lcl}
            <{\bf S}_1^0|&=&<{\bf \alpha}^\uparrow\overline
                             {\bf \alpha}^\downarrow|\\
            <{\bf S}_2^0|&=&<{\bf \alpha}^\uparrow\overline
                            {\bf \alpha}^\downarrow{\bf x}\overline{\bf x}|,
 \end{array}
 \right.
 \hspace{8mm}\mbox{Scalar}                                       \label{12}
 \end{eqnarray}
 \setcounter{equation}{\value{enumi}}}
 where the suffix $L,R$s are omitted for brevity and $\uparrow, 
 \downarrow$ indicate $spin\hspace{1mm}up, spin\hspace{1mm}down$ states.
 They play the role of intermediate bosons same as ${\bf \pi}$, 
 ${\bf \rho}$ in the strong interactions. As Eq.(9) and Eq.(12) 
 contain only ${\bf \alpha}$ and ${\bf x}$ subquarks, we can 
 draw the ``$line\hspace{1mm}diagram$'' of weak interactions as seen 
 in Fig (1). Eq.(9d) shows that the electron is constructed 
 from antimatters only. We know, phenomenologically, that this 
 universe is mainly made of protons, electrons, neutrinos,
 antineutrinos and unknown dark matters. It is said that  
 the universe contains  almost the same number of protons 
 and electrons.
  Our model show that one proton has the configuration
 of $({\bf u}{\bf u}{\bf d}): (2{\bf \alpha}, \overline{\bf \alpha}, 
 3{\bf x}, \overline{\bf x})$; electron: $(\overline{\bf \alpha}, 
 2\overline{\bf x})$; neutrino: $({\bf \alpha}, \overline{\bf x})$; 
 antineutrino: $(\overline{\bf \alpha}, {\bf x})$ and the dark 
 matters are presumably constructed from the same amount of matters 
 and antimatters because of their neutral charges. Note that 
 proton is a mixture of matters and anti-matters and electrons
 is composed of anti-matters only. This may lead the thought 
 that ``the universe is the matter-antimatter-even object.'' 
 And then there exists a conception-leap between 
 ``proton-electron abundance'' and ``matter abundance'' 
 if our composite scenario is admitted 
 (as for the possible way to realize the proton-electron 
 excess universe, see Ref.[14]).
 This idea is different from the current thought that
 the universe is made of matters only. Then the question 
 about CP violation in the early universe does not occur.
\par
 Our composite model contains two steps, namely the first is 
 ``subquarks made of primons'' and the second is ``quarks and 
 leptons made of subquarks''. Here let us discuss about 
 the mass generation mechanism  of quarks and leptons as 
 composite objects. Our model has only one  kind of fermion 
 : $\Lambda$  and boson : $\Theta$. The first  step of 
 ``subquarks made of primons'' seems to have nothing 
 to do with 'tHooft's anomaly matching condition[23] 
 because there is no global symmetry with 
 $\Lambda$ and $\Theta$. Therefore from this line of thought 
 it is  impossible to say anything about that ${\bf \alpha}$, 
 ${\bf \beta}$,  ${\bf x}$ and ${\bf y}$ are massless or massive. 
 However, if it is the case that the neutral  
 (1,1,1)-states of primon-antiprimon composites (as is stated above) 
 construct the dark matters, the masses of them are presumably less 
 than the order of MeV from the phenomenological aspects of 
 astrophysics. In this connection it is interesting that Kr\'olikowski
 has showed one possibility of constructing ``massless'' composite
 particles(fermion-fermion or fermion-boson pair) controlled
 by relativistic two-body equations[34].
 Then we may assume that these subquarks are 
 massless or almost massless compared with $\Lambda_{L,R}$ 
 in practice, that is, utmost a few MeV. 
 In the second step, the arguments of 'tHooft's anomaly 
 matching condition are meaningful. The confining of subquarks 
 must occur at the energy scale of $\Lambda_{L,R}>>\Lambda_c$ 
 and then it is natural that $\alpha_s, \alpha \rightarrow0$ 
 and that the gauge symmetry group is the spontaneously 
 unbroken $SU(2)_L\otimes{SU(2)}_R$ gauge group. 
 Seeing Eq.(9), we find quarks and leptons are composed of 
 the mixtures of subquarks and antisubquarks. 
 Therefore it is proper to 
 regard subquarks and antisubquarks as different kinds of 
 particles. From Eq.(8,a,b) we find eight kinds of fermionic 
 subquarks ( 3 for ${\bf \alpha}$, $\overline{\bf \alpha}$ and 
 1 for ${\bf \beta}$, $\overline{\bf \beta}$). So the global 
 symmetry concerned is $SU(8)_L\otimes{SU(8)}_R$. Then we  
 arrange :
 \begin{equation}
        ({\bf \beta},\overline{\bf \beta},{\bf \alpha}_i,\overline
        {\bf \alpha}_i\hspace{3mm}i=1,2,3\hspace{1mm})_{L,R}
        \hspace{2cm}in\hspace{1cm}(SU(8)_L\otimes{SU(8)}_R)_{global},                                                                           \label{13}
 \end{equation}
 where $i$ is color freedom.
 Next, the fermions in Eq.(13) are confined into the singlet 
 states of the local $SU(2)_L\otimes{SU(2)}_R$ gauge symmetry 
 and make up quarks and leptons as seen in Eq.(9) (eight fermions). 
 Then we arrange :
 \begin{equation}
        ({\bf \nu_e},{\bf e},{\bf u}_i,{\bf d}_i\hspace{3mm}i=1,2,3
        \hspace{1mm})_{L,R}\hspace{2cm}in\hspace{1cm}(SU(8)_L\otimes
        {SU(8)}_R)_{global},                                    \label{14}
 \end{equation}
 where $i$s are color freedoms. From Eq.(13) and Eq.(14) 
 the anomalies of the subquark level and the quark-lepton level 
 are matched and then all composite quarks and leptons (in the 1st 
 generation) are remained massless or almost massless. 
 Note again that presumably, 
 ${\bf \beta}$ and $\overline{\bf \beta}$ in Eq.(13) are composed 
 into ``bosonic'' (${\bf \beta}$${\bf \beta}$), 
 (${\bf \beta}$$\overline{\bf \beta}$) and 
 ($\overline{\bf \beta}$$\overline{\bf \beta}$), 
 which vapour out to the dark matters. Schrempp and Schrempp have 
 discussed about a confining $SU(2)_L\otimes{SU(2)}_R$ gauge 
 model with three fermionic preons and stated that it is 
 possible that not only the left-handed quarks and leptons   
 are composite but also the right-handed are so on the condition  
 that $\Lambda_R/\Lambda_L$ is at least of the order of $3$[11].
  As seen in Eq.(12a) the existence of composite 
 ${\bf W}_R$, ${\bf Z}_R$ is predicted. As concerning, 
 the fact that they are not observed yet means  that the masses of 
 ${\bf W}_R$, ${\bf Z}_R$ are larger than those of 
 ${\bf W}_L$, ${\bf Z}_L$ because of $\Lambda_R>\Lambda_L$. 
 Owing to 'tHooft's 
 anomaly matching condition the small mass nature of the 1st 
 generation comparing to $\Lambda_L$ is guaranteed but 
 the evidence that the quark masses of the 2nd and the 3rd 
 generations become larger as the generation numbers increase 
 seems to have nothing to do 
 with the anomaly matching mechanism in our model, because, as 
 seen in Eq.(11a,b), these generations are obtained by just 
 adding neutral scalar ${\bf y}$-particles. 
  This is different from Abott and Farhi's model 
 in which all fermions of three generations are equally
  embedded in $SU(12)$ global symmetry group and all 
  members take part in the anomaly matching mechanism[8,26]. 
  Concerning this, let us discuss a little about subquark 
  dynamics inside quarks. 
 According to ``Uncertainty Principle'' the radius of 
 the composite particle is, in general, roughly inverse 
 proportional to the kinetic energy of the constituent 
 particles moving inside it. 
 The radii of quarks may be around $1/\Lambda_{L,R}$ .
 So the kinetic energies of subquarks may be more than 
 hundreds GeV and then it is considered that the masses of 
 quarks essentially depend on the kinetic energies of 
 subquarks and such a large binding energy as 
 counterbalances them. As seen in Eq.(11a,b) our model 
 shows that the more the generation number increases 
 the more the number of the constituent particles 
 increases. So assuming that the radii 
 of all quarks do not vary so much (because we have no 
 experimental evidences yet), the interaction length among 
 subquarks inside quarks becomes shorter as generation 
 numbers increase and accordingly the average kinetic 
 energy per one subquark may increase. 
 Therefore integrating out the details of subquark 
 dynamics it could be said that 
 the feature of increasing masses of the 2nd and 
 the 3rd generations is essentially 
 described as a increasing function of the sum of 
 the kinetic energies of constituent subquarks. 
 From Review of Particle Physics[29] 
 we can phenomenologically parameterized the mass 
 spectrum of quarks and leptons as follows : 
 {
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1}
 \setcounter{equation}{0}
 \renewcommand{\theequation}{\theenumi\alph{equation}}
 \begin{eqnarray}
         M_{UQ}&=&1.2\times10^{-4}\times(10^{2.05})^n\hspace{1cm}
                  \mbox{GeV}\hspace{1.5cm}\mbox{for}\hspace{2mm}
                  \mbox{{\bf u},{\bf c},{\bf t}},\\
         M_{DQ}&=&3.0\times10^{-4}\times(10^{1.39})^n\hspace{1cm}
                 \mbox{GeV}\hspace{1.5cm}\mbox{for}\hspace{2mm}
                 \mbox{{\bf d},{\bf s},{\bf b}},\\
         M_{DL}&=&3.6\times10^{-4}\times(10^{1.23})^n\hspace{1cm}
                 \mbox{GeV}\hspace{1.5cm}\mbox{for}\hspace{2mm}
                 \mbox{{\bf e},${\bf \mu}$,${\bf \tau}$},      \label{15}
 \end{eqnarray}
 \setcounter{equation}{\value{enumi}}}
 where $n=1,2,3$ are the generation numbers and input data 
 are quark masses of 2nd and 3rd generation. 
 They seem to be  geometricratio-like. 
 The slope parameters of the up-quark 
 sector and down-quark sector are different, 
 so it is likely  that each has different aspects 
 in subquark dynamics. 
 It is interesting that the slope parameters of both down 
 sectors of quark and lepton are almost equal, which suggests 
 that there exist similar properties in substructure dynamics 
 and if it is the case, the slope parameter of up-leptonic(
 neutrino) sector may be the same as that of up-quark sector,
 that is, $M_{UL}\sim10^{2n}$.
 From Eq.(15) we obtain $M_{\bf u}=13.6$ MeV, 
 $M_{\bf d}=7.36$ MeV and $M_{\bf e}=6.15$ MeV. 
 These are a little unrealistic compared with the 
 experiments[29]. But considering the above discussions 
 about the anomaly matching conditions (
 Eq.(13,14)), it is natural that the masses of the 
 members of the 1st generation are roughly equal to 
 those of the subquarks, that is, a few MeV. 
 The details of their real mass-values may also depend on 
 the subquark dynamics owing to the effects of 
 electromagnetic and color gauge interactions. 
 These mechanism has studied by Weinberg[32] 
 and Fritzsch[33].
\par
 One of the experimental evidences inspiring the SM is 
 the ``universality'' of the coupling strength among the 
 weak interactions. Of course if the intermediate bosons are 
 gauge fields, they couple to the matter fields universally. 
 But the inverse of this statement is not always true, namely 
 the quantitative equality of the coupling strength of the 
 interactions does not necessarily imply that the intermediate 
 bosons are elementary gauge bosons. In practice the interactions 
 of ${\bf \rho}$ and ${\bf \omega}$ are regarded as indirect 
 manifestations of QCD. In case of chiral 
 $SU(2)\otimes{SU(2)}$ the pole dominance works very well
 and the predictions of current algebra and PCAC seem  
 to be fulfilled within about $5$\%[19]. 
 Fritzsch and Mandelbaum[9,19] and Gounaris, K\"ogerler 
 and Schildknecht[10,27] have elaborately discussed about 
 universality of weak interactions appearing as a 
 consequence of current algebra and ${\bf W}$-pole 
 dominance of the weak spectral functions from the 
 stand point of the composite model. 
 Extracting the essential points from their  
 arguments we mention our case as follows. 
 In the first generation let the weak charged currents be 
 written in terms of the subquark fields as :
 \begin{equation}
           {\bf J}_{\mu}^{+}=\overline{U}h_{\mu}D,\hspace{2cm}
            {\bf J}_{\mu}^{-}=\overline{D}h_{\mu}U,             \label{16}
 \end{equation}
 where $U=({\bf \alpha}{\bf x})$, $D=(\overline{\bf \alpha}
 \overline{\bf x}{\bf x})$ and $h_{\mu}=\gamma_{\mu}
 (1-\gamma_5)$.
 Reasonableness of Eq.(16) may given by the fact that
 $M_W<<\Lambda_{L,R}$ (where $M_W$ is ${\bf W}$-boson mass).
 Further, let $U$ and $D$ belong to the doublet of 
 the global weak isospin $SU(2)$ group and ${\bf W}^+$, 
 ${\bf W}^-$, $(1/\sqrt{2})({\bf Z}_1^0-{\bf Z}_2^0)$ 
 be in the triplet and $(1/\sqrt{2})({\bf Z}_1^0+
 {\bf Z}_2^0)$ be in the singlet of $SU(2)$. 
 These descriptions seem to be natural 
 if we refer the diagrams in Fig.(1). 
 The universality of the weak interactions are inherited 
 from the universal coupling strength of the algebra 
 of the global weak isospin $SU(2)$ group with the 
 assumption of ${\bf W}$-, ${\bf Z}$-pole dominance. 
 The universality including the 2nd and the 3rd 
 generations are investigated in the next section 
 based on the above assumptions  
 and in terms of the flavor-mixings.

\section{${\Delta}F=1$ flavor-mixing by subquark dynamics}
\hspace*{\parindent}
\par 
\vspace{1cm}
{\bf a. Flavor-mixing matrix element}
\newline
\par
 The quark-flavor-mixings in the weak interactions are 
 usually expressed by Cabbibo-Kobayashi-Maskawa (CKM) 
 matrix based on the SM.  
 Its nine matrix elements (in case of three generations) 
 are "free'' parameters (in practice four parameters 
 with the unitarity) 
 and this point is said to be one of the drawback 
 of the SM along with non-understanding of 
 the origins of the quark-lepton mass spectrum 
 and generations. 
 In the SM, the quark fields (lepton fields also) 
 are elementary and then we are able to investigate, 
 at the utmost, the external relationship among them. 
 On the other hand if quarks are the composites of 
 substructure constituents, 
 the quark-flavor-mixing phenomena must be understood 
 by the substructure dynamics and the values of CKM
 matrix elements become materials for studying these. 
 Terazawa and Akama have investigated quark-flavor-mixings 
 in a three spinor subquark model with higher generations 
 of radially excited state of the up (down) quark 
 and stated that a quark-flavor-mixing matrix element 
 is given by an overlapping integral of two radial 
 wave functions of the subquarks which depends on 
 the momentum transfer between quarks[28,31].
 \par 
 In our model ``\begin{em}the quark-flavor-mixings occur 
 by creations or annihilations 
 of ${\bf y}$-particles inside quarks\end{em}''. 
 The ${\bf y}$-particle is a neutral scalar   
 subquark in the {\bf3}-state of $SU(2)_L$ group 
 and then couples to two hypercolor gluons 
 (denoted by ${\bf g}_h$) (see Fig.(2)). 
 Here we propose the important assumption 
 : ``The \begin{em}
 (${\bf y}\rightarrow{2}{\bf g}_h$)-process is 
 factorized from the net 
 ${\bf W}^{\pm}$ exchange interactions\end{em}.'' 
 This assumption is plausible because the effective 
 energy of this process may be in a few TeV energy 
 region comparing to a hundred GeV energy region of 
 {\bf W}-exchange processes.
 Let us write the contribution of 
 (${\bf y}\rightarrow{2}{\bf g}_h$)-process to charged 
 weak interactions as :
 \begin{equation}
          A_{i}=\alpha^{q}_{W}(Q_{i}^{2})^{2}\cdot{B}\hspace{2cm}
                \mbox{$i$\hspace{0.5mm}=\hspace{2mm}
                {\bf s},{\bf c},{\bf b},{\bf t}},               \label{17}
 \end{equation}
 where $\alpha_{W}$ is a running coupling constant 
 of the hypercolor gauge theory appearing in Eq.(10)
 , $Q_{i}$ is the effective four momentum of 
 ${\bf g}_h$-exchange among subquarks inside 
 the $i$-quark and $B$ is a dimensionless 
 ``$complex$'' free parameter originated from 
 the unknown primon dynamics and may depend on 
 $<0|f(\overline\Lambda\cal{O}$${\Lambda}$,{\rm and/or},
 $\overline\Theta\cal{O}$${\Theta})|{\bf y}>$
 ($\cal{O}$ is some operator).
 \par
  The weak charged currents of quarks are 
 taken as the matrix elements of subquark currents 
 between quarks which are not the eigenstates of 
 the weak isospin[28]. 
 Using Eq.(11), (16) and (17) with the above 
 assumption we have :
{
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
    V_{ud}\overline{\bf u}h_{\mu}{\bf d}&=&<{\bf u}|
                        \overline{U}h_{\mu}D|{\bf d}>,\\
    V_{us}\overline{\bf u}h_{\mu}{\bf s}&=&<{\bf u}|
                        \overline{U}h_{\mu}(D{\bf y})|{\bf s}>\cong
                        <{\bf u}|\overline{U}h_{\mu}D|{\bf s}>
                        \cdot{A}_s,\\
    V_{ub}\overline{\bf u}h_{\mu}{\bf b}&=&<{\bf u}|
                        \overline{U}h_{\mu}(D{\bf y}{\bf y})|
                        {\bf b}>\cong<{\bf u}|\overline{U}h_{\mu}D|
                        {\bf b}>{\bf \cdot}{2}A_{b}^{2},\\
    V_{cd}\overline{\bf c}h_{\mu}{\bf d}&=&<{\bf c}|
                        (\overline{U}{\bf y})h_{\mu}D|{\bf d}>\cong
                        <{\bf c}|\overline{U}h_{\mu}D|{\bf d}>
                        \cdot{A}_c,\\
    V_{cs}\overline{\bf c}h_{\mu}{\bf s}&=&<{\bf c}|
                        (\overline{U}{\bf y})h_{\mu}(D{\bf y})|
                        {\bf s}>,\\
    V_{cb}\overline{\bf c}h_{\mu}{\bf b}&=&<{\bf c}|
                        (\overline{U}{\bf y})h_{\mu}(D{\bf y}{\bf y})
                        |{\bf b}>\cong<{\bf c}|(\overline{U}{\bf y})
                         h_{\mu}(D{\bf y})|{\bf b}> \cdot{A}_b,\\
    V_{td}\overline{\bf t}h_{\mu}{\bf d}&=&<{\bf t}|
                        (\overline{U}{\bf y}{\bf y})h_{\mu}D|{\bf d}>
                        \cong<{\bf t}|\overline{U}h_{\mu}D|{\bf d}>
                        \cdot{2}A_{t}^{2},\\
    V_{ts}\overline{\bf t}h_{\mu}{\bf s}&=&<{\bf t}|
                        (\overline{U}{\bf y}{\bf y})h_{\mu}(D{\bf y})
                        |{\bf s}>\cong
                        <{\bf t}|(\overline{U}{\bf y})h_{\mu}
                        (D{\bf y})|{\bf s}>\cdot{A}_t,\\
    V_{tb}\overline{\bf t}h_{\mu}{\bf b}&=&<{\bf t}|
                        (\overline{U}{\bf y}{\bf y})h_{\mu}
                        (D{\bf y}{\bf y})|{\bf b}>,             \label{18}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 where $V_{ij}$s are CKM-matrices and \{${\bf u}$,
 ${\bf d}$, ${\bf s}$, etc.\} in the left sides of 
 the equations are quark-mass eigenstates. 
 Here we need some explanations. 
 In transitions from the 3rd to the 1st 
 generation in Eq.(18c,g) there are two types :  
 One is that two (${\bf y}\rightarrow{2}{\bf g}_h$)-processes 
 occur at the same time and the other is that 
 ${\bf y}$ annihilates into $2{\bf g}_{h}$ in a cascade 
 way . Then we can describe the case of Eq.(18c) as :
\begin{eqnarray}
   <{\bf u }|\overline{U}h_{\mu}(D{\bf y}{\bf y})|{\bf b}>
       &\cong&<{\bf u}|\overline{U}h_{\mu}D|{\bf b}>
       \cdot{A}_{b}^{2}+<{\bf u}|\overline{U}h_{\mu}(D{\bf y})
       |{\bf b}>\cdot{A}_{b}\nonumber\\
       &\cong&<{\bf u}|\overline{U}h_{\mu}D|{\bf b}>
              \cdot{A}_{b}^{2}+<{\bf u}|\overline{U}h_{\mu}D
              |{\bf b}>\cdot{A}_{b}^{2}\nonumber\\
       &=&<{\bf u}|\overline{U}h_{\mu}D|{\bf b}>
          \cdot{2}A_{b}^{2}.                                    \label{19}
\end{eqnarray}
 The case of Eq.(18g) is also 
 same as this (here the phase-difference between 
 the 1st and the 2nd term is disregarded for simplicity). 
 If we admit the assumption of factorizability of (${\bf y}
 \rightarrow{2}{\bf g}_{h}$)-process, it is natural that the 
 universality of the net weak interactions among three generations 
 are realized. The net weak interactions are essentially same 
 as $({\bf u}\rightarrow{\bf d})$-transitions(Fig.(1)). 
 Then we may think that : 
{
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
      |<{\bf u}|\overline{U}h_{\mu}D|{\bf d}>| &\cong&
      |<{\bf u}|\overline{U}h_{\mu}D|{\bf s}>| \cong
      |<{\bf u}|\overline{U}h_{\mu}D|{\bf b}>|\hspace{1cm}\nonumber\\                            
                                               &\cong&
      |<{\bf c}|\overline{U}h_{\mu}D|{\bf d}>|\cong|
      <{\bf t}|\overline{U}h_{\mu}D|{\bf d}>|,\\
      |<{\bf c}|(\overline{U}{\bf y})h_{\mu}(D{\bf y})|{\bf s}>|
                                              &\cong&
      |<{\bf c}|(\overline{U}{\bf y})h_{\mu}(D{\bf y})|{\bf b}>
                                             |\cong
      |<{\bf t}|(\overline{U}{\bf y})h_{\mu}
                 (D{\bf y})|{\bf s}>|,\hspace{1cm}             \label{20}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 and additionally we  may assume :
\begin{equation}
      |<{\bf u}|\overline{U}h_{\mu}D|{\bf d}>|\cong
      |<{\bf c}|(\overline{U}{\bf y})h_{\mu}(D{\bf y})|{\bf s}>|
      \cong|<{\bf t}|(\overline{U}{\bf y}{\bf y})h_{\mu}
      (D{\bf y}{\bf y})|{\bf b}>|.                              \label{21}
\end{equation}
 In Eq.(20b) and (21) ${\bf y}$-particles are the 
 ``$spectators$'' for the weak interactions.
 Concerning the left sides of Eq.(18a-i), The 
 \{$\overline{\bf u}h_{\mu}{\bf d}$, $\overline
 {\bf u}h_{\mu}{\bf s}$, etc.\} operate coordinately as 
 the function of the current operator (that is, 
 just as the function of coupling to the ``$common$'' 
 $W$-boson current)when only weak interactions switch on. 
 In practice weak interactions occur as the residual 
 ones commonly among subquarks inside any kinds of quarks. 
 Therefore in this scenario (quark-subquark correspondence)
 it seems natural to assume 
 that such equations work in the weak interactions as :
\begin{equation}
       \overline{\bf u}h_{\mu}{\bf d}=\overline{\bf u}h_{\mu}
       {\bf s}=\overline{\bf u}h_{\mu}{\bf b}=\overline{\bf c}
        h_{\mu}{\bf d}=\cdots.\hspace{5cm}                      \label{22}
\end{equation}
 Using Eq.(17),(18),(20),(21) and (22) we find :
{
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
      \frac{|V_{us}|}{|V_{ud}|}&=&
             |A_{s}|=\alpha^{q}_{W}(Q_{s}^{2})^{2}\cdot|B|,\\
      \frac{|V_{cd}|}{|V_{ud}|}&=&
             |A_{c}|=\alpha^{q}_{W}(Q_{c}^{2})^{2}\cdot|B|,\\
      \frac{|V_{cb}|}{|V_{cs}|}&=&
             |A_{b}|=\alpha^{q}_{W}(Q_{b}^{2})^{2}\cdot|B|,\\
      \frac{|V_{ts}|}{|V_{cs}|}&=&
             |A_{t}|=\alpha^{q}_{W}(Q_{t}^{2})^{2}\cdot|B|,\\
      \frac{|V_{ub}|}{|V_{ud}|}&=&
             2|A_{b}|^{2}=2\{\alpha^{q}_{W}
                               (Q_{b}^{2})^{2}\cdot|B|\}^2,\\
      \frac{|V_{td}|}{|V_{ud}|}&=&
             2|A_{t}|^{2}=2\{\alpha^{q}_{W}
                               (Q_{t}^{2})^{2}\cdot|B|\}^2.
                                                             \label{23}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}.
\par 
 Here let us investigate the substructure dynamics 
 inside quarks referring to the above equations. 
 In our composite model quarks are composed of 
 ${\bf \alpha}$, ${\bf x}$, ${\bf y}$. Concretely 
 from Eq.(11) ${\bf c}$-quark is composed of 
 three subquarks; ${\bf t}$-quark : four subquarks; 
 ${\bf s}$-quarks : four subquarks; ${\bf b}$-quark : 
 five subquarks. From the discussions in Sect.3, 
 let the quark mass be proportional to the sum of 
 the average kinetic energies of the subquarks 
 (denoted by $<T_{i}>$,\hspace{1mm}$i={\bf s}, 
 {\bf c},{\bf b},{\bf t}$). 
 The proportional constants are assumed common in the 
 up (down)-quark sector and different between the up-
  and the down-quark sector according to the 
  discussions in Sect.3. 
 Then we denote them by $K_{s}\hspace{1mm}(s=up,down)$. 
 The $<T_{i}>$ may considered inverse proportional 
 to the average interaction length among subquarks
 (denoted by $<r_{i}>$). Further, it is presumable 
 that $\sqrt{Q^2_{i}}$($Q_{i}$ is the effective 
 four momentum of 
 ${\bf g}_h$-exchange among subquarks inside the 
 $i$-quark in Eq.(17)) 
 is inverse proportional to $<r_{i}>$. 
\par
 Then we have :
{
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
      \frac{M_{b}}{M_{s}}&=&\frac{5K_{down}<T_{b}>}{4K_{down}<T_{s}>}
      =\frac{5}{4}\cdot\frac{<r_{s}>}{<r_{b}>}\nonumber\\
      &=&\frac{5}{4}\cdot\sqrt{\frac{Q_{b}^2}{Q_{s}^2}},\\
      \frac{M_{t}}{M_{c}}&=&\frac{4K_{up}<T_{t}>}{3K_{up}<T_{c}>}
      =\frac{4}{3}\cdot\frac{<r_{c}>}{<r_{t}>}\nonumber\\
      &=&\frac{4}{3}\cdot\sqrt\frac{Q_{t}^2}{Q_{c}^2},         \label{24}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 where $M_{i}$ is the mass of $i$-quark.
 In the Review of Particle Physics[29] we find 
 : $M_{b}/M_{s}=30\pm15$  
 and $M_{t}/M_{c}=135\pm35$, using which we get by Eq.(24) : 
{
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
          \frac{Q_{b}^2}{Q_{s}^2}&\cong&(24)^2,\\
          \frac{Q_{t}^2}{Q_{c}^2}&\cong&(100)^2.                \label{25}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 Note again  that it seems to be meaningless to estimate 
 $Q_{s}^2/Q_{t}^2$ or $Q_{c}^2/Q_{b}^2$ 
 because the up-quark sector and the down-quark sector 
 possibly have the different aspects of 
 substructure dynamics(that is $K_{up} \neq K_{down}$). 
\par
 The absolute values of CKM-matrix  elements: $|V_{ij}|$s 
 are reported as the``experimental'' 
 results(without unitarity assumption)[29] that : 
\begin{eqnarray}
         |V_{ud}|&=&0.9735\pm0.0008,\hspace{1cm}|V_{us}|=0.2196
                    \pm0.0023,\nonumber\\
         |V_{cd}|&=&0.224\pm0.016,\hspace{1.5cm}|V_{cb}|=0.0402
                    \pm0.0019,\\
         |V_{cs}|&=&1.04\pm0.16,\hspace{1.8cm}|V_{ub}|/|V_{cb}|
                    =0.090\pm0.025.\nonumber
                                                                \label{26}
\end{eqnarray}
 Relating these data to the scheme of our composite model, 
 we investigate the quark-flavor-mixing phenomena 
 in terms of the substructure dynamics. 
 Using Eq.(23a,c) and $|V_{us}|$, 
 $|V_{cb}|$ in Eq.(26) we get :

\begin{equation}
 \frac{\alpha^{q}_{W}(Q_{s}^{2})}{\alpha^{q}_{W}(Q_{b}^{2})}=2.32,
                                                                \label{27}
\end{equation}
 where we assume $|V_{ud}|=|V_{cs}|$.
 Applying $N_{f}=N_{s}=4$ (as is stated in Sect.3) 
 to Eq.(10b) 
 we have :
\begin{equation}
       b_{2}(q)=0.35.                                           \label{28}
\end{equation}
 Here we rewrite Eq.(10a) as :
\begin{equation}
 \alpha^{q}_{W}(Q_{1}^{2})=\frac{1-\displaystyle{\frac{\alpha^{q}_{W}
 (Q_{1}^{2})}{\alpha^{q}_{W}(Q_{2}^{2})}}}
 {b_{2}(q)\ln\left(\displaystyle{\frac{Q_{1}^{2}}{Q_{2}^{2}}}\right)}.
                                                                \label{29}
\end{equation}
 Inserting the values of Eq.(25,a), (27) and (28) into Eq.(29)
 we have :
 \begin{equation}
          \alpha^{q}_{W}(Q_{s}^{2})=0.602,                      \label{30}
 \end{equation}
 where $Q_{s}$,$(Q_{b})$ corresponds to $Q_{1}$,$(Q_{2})$ in Eq.(29). 
 Combining $|V_{ud}|$, $|V_{us}|$ in Eq.(26) and Eq.(30) 
 with Eq.(23a) we obtain :
\begin{equation}
          |B|=0.629,                                            \label{31}
\end{equation}
 and using Eq.(30) to Eq.(27) we get :
\begin{equation}
          \alpha^{q}_{W}(Q_{b}^{2})=0.259.                      \label{32}
\end{equation}
 By use of $|V_{ud}|$, $|V_{cd}|$ in Eq.(26) and Eq.(31) 
 to Eq.(23b) we have :
\begin{equation}
          \alpha^{q}_{W}(Q_{c}^{2})=0.605.                      \label{33}
\end{equation}
 Using Eq.(10a) with Eq.(25b), (28) and (33) 
 (setting $t$ ($c$) to $1$ ($2$)) we obtain :
\begin{equation}
          \alpha^{q}_{W}(Q_{t}^{2})=0.207.                      \label{34}
\end{equation} 
\par
 Inserting Eq.(31), (32) to the right side of Eq.(23e) we have :
\begin{equation}
          |V_{ub}|=3.45\times10^{-3}.                          \label{35}    
\end{equation}
 Comparing this with the experimental value 
 of $|V_{ub}|=0.0037\pm0.0012$ (obtained from the values of
 $|V_{cb}|$ and $|V_{ub}|/|V_{cb}|$ in Eq.(26)), the consistency  
 between the prediction and the experiment seems good. 
 This result is also consistent with the first
 exclusive determinations of $|V_{ub}|$ from the decay 
 $B\rightarrow{\pi}l{\nu}_l$ and $B\rightarrow{\rho}l{\nu}_l$ by the 
 CLEO experiment to obtain $|V_{ub}|=(3.3\pm0.4\pm0.7)
 \times10^{-3}$[59].
\par
 Finally using Eq.(31), (34) to Eq.(23d,f) we predict :
\begin{equation} 
         |V_{ts}|=2.62\times10^{-2},\hspace{1.5cm}
         |V_{td}|=1.40\times10^{-3},                          \label{36} 
\end{equation}
 where we use $|V_{ud}|=0.974$, $|V_{cs}|=0.974$[29].
 Comparing the values of Eq.(36) with 
 $|V_{ts}|=0.039\pm0.004$ and  
 $|V_{td}|=0.0085\pm0.0045$[29] obtained by assuming 
 the three generations with unitarity, 
 we find that our results are smaller by a factor than them. 
 The origin of these results is presumably in that 
 the top-quark mass is heavy.
 We wish the direct measurements of ($t\rightarrow{d},s$) 
 transitions in leptonic 
 and/or semileptonic decays of top-quark mesons .
 \par
 So far we have discussed absolute values of $V_{qq^{'}}$ 
 but in practice they are generally ``$complex$'' because
 $B$ (in Eq.(17)) is originated from ${\bf y}$-subquark 
 annihilation(creation) to(from) vacuum and then it may be that 
 $<0|f({\bf \Lambda}{\cal O}\overline{\bf \Lambda},
 \rm{and/or,} 
 {\bf \Theta}{\cal O}{\overline{\bf \Theta}})|{\bf y}>
 \sim|B|e^{i{\theta}}$(${\cal O}$ is some operator).
 Then preparing for discussions in next subsection {\bf b}, 
 let the off-diagonal matrix elements of $V_{qq^{'}}$ 
 be parametrized as :
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}

\begin{eqnarray}
      V_{us}&=&{\lambda}e^{i\delta}\hspace{1.3cm}
      V_{cb}={\lambda}^{2}e^{i\delta}
      \hspace{1.6cm}V_{ub}={\lambda}^{3}e^{i{\delta}^{'}},\\
      V_{cd}&=&-{\lambda}e^{-i{\delta}}\hspace{1cm}
      V_{ts}=-{\lambda}^{2}e^{-i{\delta}}
      \hspace{1cm}V_{td}={\lambda}^{3}e^{-i{\delta}^{'}},      \label{37}
\end{eqnarray}

\setcounter{equation}{\value{enumi}}}
  here $\delta(\delta^{'})$corresponds to one(two) ${\bf y}$-
 subquark(s) creation from vacuum and $-\delta(-\delta^{'})$ 
 one(two) ${\bf y}$-subquark(s) annihilation to vacuum and we use 
 $\lambda=0.22$ from Wolfenstein's parametrization[70].
 If we see Eq.(23e,f) and Eq.(37), it may well be expected that
 ${\delta}^{'}=2{\delta}$ but here ${\delta}^{'}$ is set 
 as a free parameter, which are interestingly discussed 
 in next subsection {\bf b}.
 In case of diagonal matrix elements ${\bf y}$-subquark is a spectator 
 and then $V_{qq^{'}}$ are real, which we set for simplicity :               
 \begin{equation}
        \hspace{1.5cm}V_{ud}=V_{cs}=V_{tb}=1.                \label{38}
 \end{equation}
 
\par
\vspace{1cm}
\clearpage
{\bf b. Direct CP violation}
\newline
\par
The KTeV experiment(E832) at Fermilab in 1999[64] has crucial 
impact to theoretical analyses of direct CP violations 
in nonleptonic neutral K meson decays. This experiment 
shows that a naive superweak model cannot be the sole 
source of CP violation in the K meson system.
Our model, like CKM-model, is equipped with complex elements 
in ${\Delta}F=1$ vertex parts as seen in Eq.(37a,b).
Therefore it has ability to account for direct CP violation 
phenomena, the origin of which is creation(annihilation)
of ``${\bf y}$''-subquark from(to) vacuum as stated before.
\par
First we investigate $K^{0}(\overline{K^{0}})\to{\pi}{\pi}$ decays. 
Decay amplitudes can be described by two parts, which are in 
isospin $I=1$ and $I=2$ states due to Bose statistics of 
S wave pions. On the other hand in the description of quark 
levels they contain the ``tree''- and the ``penguin''-type graphs.
The former has both of $I=0$ and $I=2$ contributions 
but the latter has only $I=0$ contribution.  
Then let us write decay amplitudes(denoted by $A_{I}, I=0,2$
for isospin) as :
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
  A_{0}&=&V^{*}_{us}V_{ud}|A_{0T}|e^{i{\theta}^{T}_{0}}
  +\left(V^{*}_{us}V_{ud}|A^{u}_{0P}|+V^{*}_{cs}V_{cd}|A^{c}_{0P}|
  +V^{*}_{ts}V_{td}|A^{t}_{0P}|\right)e^{i{\theta}^{P}_{0}},\\
  A_{2}&=&V^{*}_{us}V_{ud}|A_{2T}|e^{i{\theta}_{2}},              \label{39}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
where it means that $T$ : tree-type graph; $P$ : penguin-type graph; 
$u,c,t,$ : quarks contributing to penguin-type graphs virtually;
${\theta}^{T,{\rm{or}} P}_{0,{\rm{or}} 2}$ are strong final-state 
interaction phase shifts.
\par
Using Eq.(37a,b) and Eq.(38) to Eq.(39a,b) and assuming 
${\theta}^{T}_{0}\approx{\theta}^{P}_{0}(\equiv{\theta}_{0})$ 
(because we have no experimental information ), we have :  
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}

\begin{eqnarray}
  A_{0}&=&\left({\lambda}e^{-i{\delta}}|A_{0T}|+{\lambda}e^{-i{\delta}}
         |A_{0P}^{u}|-{\lambda}e^{-i{\delta}}|A_{0P}^{c}|
         -{\lambda}^{5}e^{i({\delta}-{\delta}^{'})}|A_{0P}^{t}|
         \right)e^{i{\theta}_{0}}\nonumber\\
       &=&\left\{{\lambda}e^{-i{\delta}}|A_{0T}|
          \left(1+\frac{|A_{0P}^{u}|}{|A_{0T}|}-\frac{|A_{0P}^{c}|}
          {|A_{0T}|}\right)
          -{\lambda}^{5}e^{i({\delta}-{\delta}^{'})}|A_{0P}^{t}|\right\}
          e^{i{\theta}_{0}},\\
  A_{2}&=&{\lambda}e^{-i{\delta}}|A_{2T}|e^{i{\theta}_{2}}.      \label{40}         
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
\par 
Roughly assuming $|A_{0P}^{u}|\simeq|{A_{0P}^{c}}|
\simeq{|A_{0P}^{t}|}(\equiv{|A_{0P}|})$ 
and introducing ${\gamma}_{K}\equiv{|A_{0P}/A_{0T}|}$;
${\Delta}\equiv{\delta}-{\delta}^{'}$, we rewrite Eq.(40a) as 

 \begin{equation}
   A_{0}={\lambda}|A_{0T}|\left(e^{-i{\delta}}-{\lambda}^{4}{\gamma}_{K}
         e^{i{\Delta}}\right)e^{i{\theta}_{0}}.                                                                                                                            \label{41} 
 \end{equation}
 
As is well known, direct CP violation in K meson system is analysed 
in CKM phase convention by :

 
\begin{equation} 
   {\varepsilon}^{'}=\frac{i}{\sqrt{2}}{\omega}(t_{2}-t_{0})
   e^{i({\theta}_{2}-{\theta}_{0})},                           \label{42}
\end{equation}   
 where ${\omega}=$Re$A_{2}/$Re$A_{0}$, $t_{I}
 =$Im$A_{I}/$Re$A_{I}, (I=0, 2)$.
 \par
 On the other hand ``${\varepsilon}$'' which describes indirect CP 
 violation in K meson system has been confirmed experimentally as :
 \begin{equation}
   {\varepsilon}=(1.569+i1.646){\times}10^{-3}.                \label{43}
 \end{equation}
 In CPT symmetry limit phases of ${\varepsilon}$ and 
 ${\varepsilon}^{'}$ are accidentally almost equal and then 
 usually we discuss by using the equation :
 \begin{equation}
   \rm{Re}\left(\frac{{\varepsilon}^{'}}{\varepsilon}\right)
   \approx\left|\frac{{\varepsilon}^{'}}{\varepsilon}\right|.  \label{44}
 \end{equation}
\par
From Eq.(42) with ${\varepsilon}$ we have :
 {  
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1}
 \setcounter{equation}{0}
 \renewcommand{\theequation}{\theenumi\alph{equation}}
 \begin{eqnarray}
   \left|\frac{{\varepsilon}^{'}}{\varepsilon}\right|&=&
        \frac{\omega}{\sqrt{2}|{\varepsilon}|}{\kappa},\\
        {\kappa}&=&\left|\frac{{\rm{Im}}A_{2}}{{\rm{Re}}A_{2}}
                 -\frac{{\rm{Im}}A_{0}}{{\rm{Re}}A_{0}}\right|. \label{45}         
 \end{eqnarray}
 \setcounter{equation}{\value{enumi}}}
By using Eq.(40b), Eq.(41) and Eq.(45b) we obtain :
 \begin{equation}
   {\kappa}=
      \left|-\tan{\delta}\left(1-\frac{1+{\lambda}^{4}{\gamma}_{K}
      \displaystyle{\frac{\sin{\Delta}}{\sin{\delta}}}}
       {1-{\lambda}^{4}{\gamma}_{K}\displaystyle
       {\frac{\cos{\Delta}}{\cos{\delta}}}}\right)\right|.                                                                                                       \label{46}
 \end{equation}
 \par
Next let us discuss the issues about CP asymmetries of B meson decays 
such as $B_{d}^{0}\to{J}/{\psi}K_{s}$ and 
         $B_{d}^{0}\to{\pi}^{+}{\pi}^{-}$.
We write $(B_{d}^{0}\to{J}/{\psi}K_{s})$-decay amplitude as :
 \begin{eqnarray} 
    A(J/{\psi}K_{s})&=&
      V_{cb}^{*}V_{cs}|A_{T}|e^{i{\theta}^{T}}
     +\left(V_{ub}^{*}V_{us}|A_{P}^{u}|
     +V_{cb}^{*}V_{cs}|A_{P}^{c}|
     + V_{tb}^{*}V_{ts}|A_{P}^{t}|\right)e^{i{\theta}^{P}}\nonumber\\
                    &=&
      {\lambda}^{2}e^{-i{\delta}}|A_{T}|e^{i{\theta}^{T}}
     +{\lambda}^{2}\left(e^{-i{\delta}}|A_{P}^{c}|
     -e^{-i{\delta}}|A_{P}^{t}|
     +{\lambda}^{2}e^{-i{\Delta}}|A_{P}^{u}|\right)
      e^{i{\theta}^{P}}                                      \label{47}
 \end{eqnarray}
where we use Eq.(37a,b) and Eq.(38). 
It is obtained from Eq.(47) that :
 \begin{equation}
    A(J/{\psi}K_{s})=
             {\lambda}^{2}|A_{T}|
             e^{i{\theta}^{T}}e^{-i{\delta}}
             \left(1+{\lambda}^{2}{\gamma}_{B}e^{i{\Theta}}
             e^{i({\delta}+{\Delta})}\right),                \label{48}
 \end{equation}            
where we roughly assume $|A_{P}^{u}|\simeq|A_{P}^{c}|\simeq
         |A_{P}^{t}|(\equiv|A_{P}|)$
and define ${\Theta}\equiv{\theta}^{P}-{\theta}^{T}$
and ${\gamma}_{B}\equiv{|A_{P}/A_{T}|}$.
For $(\overline{B_{d}^{0}}\to{J}/{\psi}K_{s})$-decay we have :
 \begin{equation}
     \overline{A}(J/{\psi}K_{s})=
             {\lambda}^{2}|A_{T}|
             e^{i{\theta}^{T}}e^{i{\delta}}
             \left(1+{\lambda}^{2}{\gamma}_{B}e^{i{\Theta}}
             e^{-i({\delta}+{\Delta})}\right),              \label{49}
 \end{equation} 
 where we consider CP invariance of strong 
 interaction($|\overline{{A}_{T}}|
 =|A_{T}|$, ${\Theta}^{T}$ and ${\Theta}$ does not change its phase).
Here defining $\overline{Z}\equiv\overline{A}/A$ and using 
Eq.(48) and Eq.(49) we obtain : 
 \begin{equation}
 \overline{Z}(J/{\psi}K_{s})=e^{2i{\delta}}
             \frac{1+{\lambda}^{2}{\gamma}_{B}e^{i{\Theta}}
             e^{-i({\delta}+{\Delta})}}
             {1+{\lambda}^{2}{\gamma}_{B}e^{i{\Theta}}
             e^{i({\delta}+{\Delta})}}.                    \label{50} 
 \end{equation}
Estimating ${\lambda}^{2}{\gamma}_{B}\leq{O}(10^{-3})\ll{1}$ because 
${\lambda}^{2}\sim(10^{-2})$ and ${\gamma}_{B}\leq{O}(1)$, 
$\overline{Z}$ is described as [65] :
 \begin{equation}
     \overline{Z}(J/{\psi}K_{s})\simeq
         e^{2i{\delta}}
         \left\{1-2i{\lambda}^{2}{\gamma}_{B}e^{i{\Theta}}
         \sin({\delta}+{\Delta})\right\}.                 \label{51}    
 \end{equation}
Indirect CP violation is usually defined as :
 \begin{equation}
    {\LARGE{m}}\simeq\sqrt{\frac{M_{12}^{*}}{M_{12}}}
    \equiv{e}^{i2{\theta}_{B_{d}}},                       \label{52}
 \end{equation}
 where $M_{12}$ is the off-diagonal part of the mass matrix and 
 $|m|\simeq{1}$ is considered from 
 exp. : ${\Delta}{\Gamma}_{B}/{\Delta}M_{B}\ll{1}$.
By use of Eq.(51) and Eq.(52) the CP-asymmetry (denoted by ${\cal A}$) is 
given as[65] :
 \begin{eqnarray}
        {\cal A}(J/{\psi}K_{s})&=&{\rm{Im}}\{m\overline{Z}
                           (J/{\psi}K_{s})\}\nonumber\\
                 &\simeq&{\sin(2{\delta}+2{\theta}_{B_{d}})-
                 2{\lambda}^{2}{\gamma}_{B}
                 \sin({\delta}+{\Delta})
                 \cos(2{\delta}+{\Theta}
                  +2{\theta}_{B_{d}})}.                   \label{53}
 \end{eqnarray}                
In case of ($B_{d}^{0}\to{\pi}^{+}{\pi}^{-}$)-decay we have :
 \begin{eqnarray}  
      A({\pi}^{+}{\pi}^{-})&=&
           V^{*}_{ub}V_{ud}|A_{T}^{'}|e^{i{\theta}^{T'}}
  +\left(V^{*}_{ub}V_{ud}|A^{u'}_{P}|+V^{*}_{cb}V_{cd}|A^{c'}_{P}|
  +V^{*}_{tb}V_{td}|A^{t'}_{P}|\right)e^{i{\theta}^{P'}}\nonumber\\
                   &=&
      {\lambda}^{3}\left\{e^{-i{\delta}^{'}}|A_{T}^{'}
      |e^{i{\theta}^{T'}}
      +\left(e^{-i{\delta}^{'}}|A_{P}^{u'}|
      +e^{-i{\delta}^{'}}|A_{P}^{t'}|
      -e^{-2i{\delta}}|A_{P}^{c'}|\right)
       e^{i{\theta}^{P'}}\right\},                       \label{54}
 \end{eqnarray}    
 where we use Eq.(37a,b) and Eq.(38) and we get :
 \begin{equation} 
   A({\pi}^{+}{\pi}^{-})=
             {\lambda}^{3}\left(|A_{T}^{'}|+2e^{i{\Theta}^{'}}
             |A_{P}^{'}|\right)
             e^{i{\theta}^{T'}}e^{-i{\delta}^{'}}
             \left(1-{\gamma}^{'}_{B}e^{i{\Theta}{'}}
             e^{-i({\delta}+{\Delta})}\right),           \label{55}
 \end{equation} 
 where we roughly assume 
 $|A_{P}^{u'}|\simeq|{A}_{P}^{c'}|\simeq|{A}_{P}^{t'}|
 (\equiv|{A}_{P}^{'}|)$ and define 
 ${\gamma}_{B}^{'}\equiv{|}A_{P}^{'}|/
 (|A_{T}^{'}|+2e^{i{\Theta}^{'}}|A_{P}^{'}|)$ and 
 ${\Theta}^{'}\equiv
 {\theta}^{P'}-{\theta}^{T'}$.
 We obtain for ($\overline{B^{0}_{d}}\to{\pi}^{+}
 {\pi}^{-}$)-decay process : 
 \begin{equation} 
   \overline{A}({\pi}^{+}{\pi}^{-})=
             {\lambda}^{3}\left(|A_{T}^{'}|+2e^{i{\Theta}^{'}}
             |A_{P}^{'}|\right)
             e^{i{\theta}^{T'}}e^{i{\delta}^{'}}
             \left(1-{\gamma}^{'}_{B}e^{i{\Theta}{'}}
             e^{i({\delta}+{\Delta})}\right),              \label{56}
 \end{equation} 
where it is assumed that $|A_{T}^{'}|=|\overline{A_{T}^{'}}|$,
$|A_{P}^{'}|=|\overline{A_{P}^{'}}|$ because of CP invariance of 
strong interaction. 
By definition of $|\tilde{{\gamma}}^{'}_{B}|
\equiv|A_{P}^{'}/A_{T}^{'}|$  
we have $|A_{P}^{'}|/(|A_{T}^{'}|+2e^{i{\Theta}^{'}}|A_{P}^{'}|)
\approx\tilde{{\gamma}}^{'}_{B}/(1+4\tilde{{\gamma}}^{'}_{B}
\cos{\Theta}^{'})$ in order of 
$O(\tilde{{\gamma}}^{'}_{B})$. 

  Then we get :
  \begin{eqnarray}
     {\cal A}({\pi}^{+}{\pi}^{-})&=&
                 {\rm{Im}}\{m\overline{Z}({\pi}^{+}{\pi}^{-})\}\nonumber\\
                 &\simeq&{\sin(2{\delta}^{'}+2{\theta}_{B_{d}})
                 -2\frac{\tilde{{\gamma}}^{'}_{B}}{1+
                 4\tilde{{\gamma}}^{'}_{B}
                 \cos{\Theta}^{'}}\sin({\delta}+{\Delta})
                  \cos(2{\delta}^{'}+{\Theta}^{'}
                 +2{\theta}_{B_{d}})},\hspace{7mm}                     \label{57}
 \end{eqnarray} 
where $|\tilde{{\gamma}}^{'}_{B}|\ll{1}$ is assumed.
\par
From here let us extract numerical informations of above quantities 
from experimental results. In $K$ meson system it is reported that :
 \begin{equation} 
   {\omega}=2.27\times{10}^{-3},                           \label{58}
 \end{equation}
and also that :
{  
 \setcounter{enumi}{\value{equation}}
 \addtocounter{enumi}{1}
 \setcounter{equation}{0}
 \renewcommand{\theequation}{\theenumi\alph{equation}}
 \begin{eqnarray}
    \rm{Re}\left(\frac{{\varepsilon}^{'}}{\varepsilon}\right)&=&
 [23\pm{6.5}]\times{10}^{-4}\hspace{3.3cm}\mbox{NA31[66]}\\
 &=&[28\pm3.0(\rm{stat})\pm2.8(\rm{sys})]
 \times{10}^{-4}\hspace{4mm}
 \mbox{E832\hspace{1mm}[64]}.                               \label{59}
 \end{eqnarray}
 \setcounter{equation}{\value{enumi}}}
                                                                
Concerning CP-asymmetry in the ($B\to{J}/{\psi}K_{s}$)-decay 
CDF Collaboration reported that[67] :
 \begin{equation}
    {\cal A}(J/{\psi}K_{s})=0.79_{-0.44}^{+0.41}\hspace{1mm}.       \label{60}
 \end{equation}     
Here using Eq.(53) and Eq.(60) let us estimate the numerical value of 
${\delta}$. Eq.(53) is rewritten as :
 \begin{equation}
        \sin(2{\delta}+2{\theta}_{B_{d}})=
                 {\cal A}(J/{\psi}K_{s})+2{\lambda}^{2}{\gamma}_{B}
                 \sin({\delta}+{\Delta})
                 \cos(2{\delta}+{\Theta}+2{\theta}_{B_{d}}).          \label{61}
 \end{equation}

The facts : ${\lambda}^{2}\sim{O}(10^{-2})$,
$|{\gamma}_{B}|\le{1}$, $|\sin({\delta}+{\Delta})|\le{1}$ and
$|\cos(2{\delta}+{\Theta}+2{\theta})|\le{1}$  lead :
    $|{\lambda}^{2}{\gamma}_{B}\sin({\delta}+{\Delta})
      \cos(2{\delta}+{\Theta}+2{\theta}_{B_{d}})|
     \le{O(10^{-3})}$ at most.
On the other hand Eq.(60) shows : ${\cal A}
(J/{\psi}K_{s})\sim{O}(10^{-1})$,
then it  may well be said :
 \begin{equation}
    \sin(2{\delta}
    +2{\theta}_{B_{d}})\simeq{{\cal A}(J/{\psi}K_{s})}.                      \label{62}
 \end{equation}
Therefore Eq.(60) and Eq.(62) leads : 
 \begin{equation}
     20^{\circ}<2{\delta}+2{\theta_{B_{d}}}\le90^{\circ}.
 \end{equation}                                                       \label{63}
From discussions about ${\theta_{B_{d}}}$ in next section 
our model predicts : ${\theta_{B_{d}}}=2{\theta_{K}}\simeq{1.3}\times
{10^{-2}}\rm{rad}=0.75^{\circ}$, then neglecting ${\theta_{B_{d}}}$ in Eq.(63)
we have :
 \begin{equation}  
    10^{\circ}<{\delta}\le45^{\circ}.
 \end{equation}                                                        \label{64}
 
Next let us see $(K\to{\pi}{\pi})$-decay. 
By use of Eq.(44).(45a).(58) and (59b) we obtain :
 \begin{equation}
      {\kappa}=2.05\times{10}^{-4}.
 \end{equation}                                                        \label{65} 

Let Eq.(46) be rewritten as :
 \begin{equation}
    \frac{\sin{\Delta}}{\sin{\delta}}+\left(1-\frac{\kappa}{|\tan{\delta}|}
    \right)\frac{\cos{\Delta}}{\cos{\delta}}=\frac{{\pm}\kappa}
   {{\lambda}^{4}{\gamma}_{K}|\tan{\delta}|}. 
 \end{equation}                                                        \label{66}
From Eq.(64) and (65) we have $2.05\times10^{-4}<{\kappa}/
|\tan{\delta}|<1.16\times10^{-3}$ and then considering 
$(1-{\kappa}/|\tan{\delta}|)\approx1$ we obtain : 

 \begin{equation}
   \sqrt{\left(\frac{1}{\sin{\delta}}\right)^{2}+
         \left(\frac{1}{\cos{\delta}}\right)^{2}}\sin({\Delta}+{\alpha})=
   \frac{{\pm}\kappa}{{\lambda}^{4}{\gamma}_{K}|\tan{\delta}|},
   \hspace{1.0cm}\tan{\alpha}=\frac{\sin{\delta}}{\cos{\delta}}.
 \end{equation}                                                        \label{67}
 
 
Concerning estimation of the hadronic matrix elements various calculations
have been made[69] but there are still many uncertainties. Then here 
we treat ${\gamma}_{K}$ as free parameter. Using Eq.(64), (65) and (67) 
with ${\Delta}={\delta}-{\delta}^{'}$ and setting 
${\gamma}_{K}={\lambda}(=0.22)\sim{1}$
we obtain :
 \begin{eqnarray}
 &&\left\{
 \begin{array}{lcl}
  {\delta}&=&10^{\circ}\hspace{1cm}{\delta}^{'}=
  15^{\circ}({\gamma}_{K}=1)\sim{-3}^{\circ}
  ({\gamma}_{K}={\lambda})\\
  {\delta}&=&{45}^{\circ}\hspace{1cm}{\delta}^{'}=
  86^{\circ}({\gamma}_{K}=1)\sim\hspace{1mm}{78}^{\circ}
  ({\gamma}_{K}={\lambda}),\\
  \end{array}
  \right.
  \end{eqnarray}                                                    \label{68}
  for $+{\kappa}$ in Eq.(67).
  \begin{eqnarray}
 %  \mbox{(B)}
 &&\left\{
 \begin{array}{lcl}
  {\delta}&=&10^{\circ}\hspace{1cm}{\delta}^{'}=
  25^{\circ}({\gamma}_{K}=1)\sim\hspace{2mm}{43}^{\circ}
  \hspace{0mm}({\gamma}_{K}={\lambda})\\
  {\delta}&=&{45}^{\circ}\hspace{1cm}{\delta}^{'}=
  93^{\circ}({\gamma}_{K}=1)\sim{102}^{\circ}
  ({\gamma}_{K}={\lambda}),\\
  \end{array}
  \right.
 \end{eqnarray}                                                    \label{69}
 for $-{\kappa}$ in Eq.(67).
 \par

 From Eq.(68) and Eq.(69) ``${\delta}^{'}\approx2{\delta}$'' is realized 
with rather large ${\delta}$ and ${\gamma}_{K}$. 
As is stated before concerning Eq.(37a,b),
 ``${\delta}^{'}\approx2{\delta}$'' means
that creations(annihilations) of two ${\bf y}$-subquarks 
from(to) vacuum in $(u\to{b})$- or $(t\to{d})$-flavor changing 
interactions may occur almost non-correlatively and then 
the phase is ``$additive$'' but if ${\delta}^{'}$ is 
``$exactly$'' equal to $2{\delta}$, 
it is found that no CP-violation in 
$(K\to{\pi}{\pi})$-decay processes occurs as seen in Eq.(41).
Lastly we estimate the asymmetry ${\cal A}({\pi}^{+}{\pi}^{-})$ of 
$(B\to{\pi}{\pi})$-decay. In Eq.(57) carrying out rather 
rough approximations 
\clearpage
 we obtain :

 \begin{equation}
    {\cal A}({\pi}^{+}{\pi}^{-})\approx\sin(2{\delta}^{'}).
 \end{equation}                                                      \label{70}
 
Then ${\cal A}({\pi}^{+}{\pi}^{-})$ is calculated as  :
 \begin{eqnarray}
 %&&\left\{
 \begin{array}{lcl}
  \hspace{2mm}0.5\hspace{2mm}({\delta}^{'}=15^{\circ}
  ,{\gamma}_{K}=1)&\sim&
  -0.1({\delta}^{'}={-3}^{\circ},{\gamma}_{K}={\lambda})
  \hspace{0.5cm}\mbox{for}\hspace{1mm}{\delta}=10^{\circ}\\
  \hspace{0mm}0.14\hspace{2mm}({\delta}^{'}=86^{\circ}
  ,{\gamma}_{K}=1)&\sim&
  \hspace{2mm}0.41({\delta}^{'}={78}^{\circ},{\gamma}_{K}={\lambda})
  \hspace{0.6cm}\mbox{for}\hspace{1mm}{\delta}=45^{\circ},\\       \label{71}
  \end{array}
  %\right.
  \end{eqnarray}                                                                      
  which are from Eq.(68) and also :
  \begin{eqnarray}
 %&&\left\{
 \begin{array}{lcl}
  \hspace{2mm}0.766\hspace{2mm}({\delta}^{'}=25^{\circ}
  ,{\gamma}_{K}=1)&\sim&
  \hspace{2mm}0.997({\delta}^{'}=\hspace{2mm}{43}^{\circ},
  {\gamma}_{K}={\lambda})
  \hspace{0.5cm}\mbox{for}\hspace{1mm}{\delta}=10^{\circ}\\
  \hspace{0mm}-0.10\hspace{3mm}({\delta}^{'}=93^{\circ}
  ,{\gamma}_{K}=1)&\sim&
  \hspace{1mm}-0.40({\delta}^{'}={102}^{\circ},{\gamma}_{K}
  ={\lambda})
  \hspace{0.5cm}\mbox{for}\hspace{1mm}{\delta}=45^{\circ},\\       \label{72}
  \end{array}
  %\right.
  \end{eqnarray}                                                                      
  which are from Eq.(69). From above results it is seen that 
  ${\cal A}({\pi}^{+}{\pi}^{-})$ has rather small value in large ${\delta}$
  and ${\gamma}_{K}$(that is, when ${\delta}^{'}\approx{2}{\delta}$ 
  realizes). 
  
  
  
  
    
  

\section{${\Delta}F=2$ flavor-mixing by subquark dynamics}
\hspace{\parindent}

 {\bf a. Mass difference ${\Delta}M_P$
 by $P^{0}-\overline{P^{0}}$ mixing} 
\newline
\par
 The typical ${\Delta}F=2$ phenomenon is the mixing 
 between a neutral pseudo scalar meson ($P^0$) and its 
 antimeson ($\overline{P^{0}}$). There are six types  
 , e.g., $K^{0}-\overline{K^{0}}$, 
 $D^{0}-\overline{D^{0}}$, $B^{0}_{d}-\overline{B^{0}_{d}}$, 
 $B^{0}_{s}-\overline{B^{0}_{s}}$, $T^{0}_{u}-
 \overline{T^{0}_{u}}$ and $T^{0}_{c}-\overline{T^{0}_{c}}$ 
 mixings. 
 Usually they have been considered to be the most sensitive 
 probes of higher-order effects of the weak interactions 
 in the SM. 
 The basic tool to investigate them is the ``box diagram''. 
 By using this diagram to the $K_{L}$-$K_{S}$ mass 
 difference, Gaillard and Lee predicted the mass of the 
 charm quark[38]. Later, Wolfenstein suggested that the 
 contribution of the box diagram which is called the 
 short-distance (SD) contribution cannot supply the whole 
 of the mass difference ${\Delta}M_K$ and there are 
 significant contributions arising from the long-distance 
 (LD) contributions associated with low-energy 
 intermediate hadronic states[38]. 
 As concerns, the LD-phenomena occur in the energy range 
 of few hundred MeV and the SD-phenomena around $100$ GeV 
 region. 
 Historically there are various investigations for 
 $P^{0}$-$\overline{P^{0}}$ mixing problems[36][39-48] and 
 many authors have examined them by use of LD- and 
 SD-contributions. 
 In summary, the comparison between the theoretical results 
 and the experiments about ${\Delta}M_P$ 
 ($P=K,D$ and $B_d$) are as follows :
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 {\Delta}M^{LD}_{K}&\approx&{\Delta}M^{SD}_{K}\approx
 {\Delta}M^{exp}_{K},\\
 {\Delta}M^{SD}_{D}&\ll&{\Delta}M^{LD}_{D}
 (\ll{\Delta}M^{exp}_{D},\rm{upper}\hspace{2mm}\rm{bound}),\\
 {\Delta}M^{LD}_{B_d}&\ll&{\Delta}M^{SD}_{B_d}\simeq
 {\Delta}M^{exp}_{B_d}.                                        \label{73}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 Concerning Eq.(732a) it is explain that ${\Delta}M_{K}=
 {\Delta}M^{SD}_{K}+D{\Delta}M^{LD}_{K}$ where ``$D$'' is 
 a numerical value of order $O(1)$. As for Eq(73c), 
 they found that ${\Delta}M^{LD}_{B_d}\approx10^{-16}$ GeV 
 and ${\Delta}M^{SD}_{B_d}\approx10^{-13}$ GeV, 
 then the box diagram is the most important 
 for $B^{0}_{d}$-$\overline{B^{0}_{d}}$ mixing. 
 Computations of ${\Delta}M^{SD}_{B_d}$ and 
 ${\Delta}M^{SD}_{B_s}$ from the box diagrams 
 in the SM give
\begin{equation}
 \frac{{\Delta}M^{SD}_{B_s}}{{\Delta}M^{SD}_{B_d}}\simeq
 \left|\frac{V_{ts}}{V_{td}}\right|^2
 \frac{B_{B_s}{f}^2_{B_s}}
 {B_{B_d}{f}^2_{B_d}}
 \frac{M_{B_s}}{M_{B_d}}\zeta,
                                                             \label{74}
\end{equation}

 where $V_{ij}$s stand for CKM matrix elements; $M_P$ : 
 P-meson mass; $\zeta$ : a QCD correction of order $O(1)$; 
 $B_B$ : Bag factor of B-meson and $f_B$ : decay constant 
 of B-meson.
 Measurements of ${\Delta}M^{exp}_{B_d}$ and 
 ${\Delta}M^{exp}_{B_s}$ are, therefore, said to be useful 
 to determine $|V_{ts}/V_{td}|$[49][50]. 
 Concerning Eq.(73b), they found that ${\Delta}M^{LD}_{D}
 \approx10^{-15}$ GeV and ${\Delta}M^{SD}_{D}
 \approx10^{-17}$ GeV[36][44] but the experimental 
 measurement is ${\Delta}M^{exp}_{D}<4.6\times10^{-14}$ 
 GeV with CL=$90\verb+%+$ [29]. 
 Further there is also a study that ${\Delta}M^{LD}_D$ is 
 smaller than $10^{-15}$ GeV by using the heavy quark 
 effective theory[45]. 
 Then many people state that it would be a signal of new 
 physics beyond the SM if the future experiments confirm 
 that ${\Delta}M^{exp}_D\simeq10^{-14}\sim10^{-13}$ 
 GeV[39-45][60].
 \par
 On the other hand some authors have studied these 
 phenomena in the context of the theory explained by 
 the single dynamical origin. 
 Cheng and Sher[68], Liu and Wolfenstein[47], and G\'erard 
 and Nakada[48] have thought that all 
 $P^{0}$-$\overline{P^0}$ mixings occur only by the 
 dynamics of the TeV energy region which is essentially 
 the same as the SW-idea originated by 
 Wolfenstein[35]. 
 They extended the original SW-theory (which explains indirect 
 CP violation in the $K$-meson system) to other flavors by 
 setting the assumption that ${\Delta}F=2$ changing 
 neutral $spin\hspace{2mm}0$ particle with a few TeV mass 
 (denoted by $H$) contributes to the ``real part'' of 
 $M_{ij}$ which determines ${\Delta}M_P$ and also the 
 ``imaginary part'' of $M_{ij}$ which causes the indirect 
 CP violation. 
 The ways of extensions are that $H$-particles couple to 
 quarks by the coupling proportional to $\sqrt{{m}_i
 {m}_j}$[47][68], $({m}_i/{m}_j)^n\hspace
 {2mm}n=0,1,2$[47] and $({m}_i+{m}_j)$[48] 
 where $i,j$ are flavors of quarks coupling to $H$. 
 It is suggestive that the SW-couplings depend on quark 
 masses (this idea is adopted in our model discussed 
 below). Cheng and Sher[68] and Liu and Wolfenstein[47] 
 obtained that ${\Delta}M_D=({m}_c/{m}_s)
 {\Delta}M^{exp}_{K}\approx10^{-14}$ GeV with the 
 assumption that $H$-exchange mechanism saturates the 
 ${\Delta}M^{exp}_K$ bound, which is comparable to 
 ${\Delta}M^{exp}_{D}<4.6\times10^{-14}$ GeV[29].
 Concerning $B$-meson systems they found that 
 ${\Delta}M_{B_s}/{\Delta}M_{B_d}={m}_s/{m}_d
 \simeq20$ which seems agreeable to $({\Delta}M_{B_s}/
 {\Delta}M_{B_d})_{exp}>22$[29]. 
 However using their scheme it is calculated that :
 
\begin{equation}
 \frac{{\Delta}M_{B_d}}{{\Delta}M_K}=
 \frac{B_{B_d}{f}^2_{B_d}}{B_K{f}^{2}_{K}}
 \frac{M_{B_d}}{M_{K}}
 \frac{m_b}{m_s}
 \simeq300,                                                \label{75}
\end{equation}

 where we use $m_b=4.3$ GeV, $m_s=0.2$ GeV, 
 $M_{B_d}=5.279$ GeV, $M_K=0.498$ GeV, 
 $B_{B_d}{f}^2_{B_d}=(0.22{\rm GeV})^2$, 
 $B_K{f}^2_K=(0.17{\rm GeV})^2$. 
 This is larger than 
 $({\Delta}M_{B_d}/{\Delta}M_K)_{exp}=89$[29] and 
 is caused by large b-quark mass value. 
\par
 Now let us discuss $P^0$-$\overline{P^0}$ mixings by 
 using our FB-model. The discussions start from the 
 assumption that the mass mixing matrix $M_{ij}(P)$ $(i(j)=
 1(2)$ denotes $P^0(\overline{P^0}))$ is described only by the 
 ``${\bf y}-exchange$'' interactions causing a direct 
 ${\Delta}F=2$ transitions. 
  we calculate ${\Delta}M_P$ as :
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 M_{12}(P)&=
 &<\overline{P^0}|{\cal H}_{\Delta F=2}^{\bf y}|P^0>,\\
 {\Delta}M_P\hspace{2mm}&=&M_H-M_L\simeq2|M_{12}(P)|,        \label{76}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
where ${\cal H}_{{\Delta}F=2}^{\bf y}$ is Hamiltonian for 
${\Delta}F=2$ transition interaction by ${\bf y}$-exchange ;
 we assume $ImM_{12}\ll{ReM_{12}}$ which is 
 experimentally acceptable[36][52], and $M_{H(L)}$ stands 
 for heavier (lighter) $P^0(\overline{P^{0}})$-meson mass. 
 Applying the vacuum-insertion calculation to the hadronic 
 matrix element as $<\overline{P^0}|[\overline{{q}_{i}}
 \gamma_{\mu}(1-\gamma_5){q}_j]^2|P^0>\sim
 B_P{f}^2_{P}M^2_P$[36] we get :
 
 \begin{equation}
        M_{12}(P)=\frac{1}{12\pi^2}B_P{f}^2_{P}M_{P}{\cal M}_P.    \label{77}
 \end{equation}
 
 The diagrams contributing to 
 ${\cal M}_P$ are seen in Fig.(3).  
 $P^0$-$\overline{P^0}$ mixings occur due to 
 ``{\bf y}-exchange'' between two quarks inside the present 
 $P^0(\overline{P^{0}})$-meson. 
 This is a kind of the realizations of Wolfenstein's 
 SW-idea[35]. The schematic illustration is as follows : 
 two particles (quarks) with radius order of 
 $1/{\Lambda}_q$ (maybe a few ${\rm TeV}^{-1}$ are moving  
 inside a sphere (meson) with radius order of 
 ${\rm GeV}^{-1}$. 
 The ${\bf y}$-exchange interactions would occur when 
 two quarks inside $P^0(\overline{P^{0}})$-meson 
 interact in contact with 
 each other because ${\bf y}$-particles are confined 
 inside quarks. 
 As seen in Fig.(3), the contributions of 
 ${\bf y}$-exchanges are common among various 
 $P^0(\overline{P^{0}})$-mesons. 
 Then we set the assumption :
 ``\begin{em}universality of the ${\bf y}$-exchange
 interactions\end{em}'', Let us describe ${\cal M}_P$ as :

\begin{equation}
        {\cal M}_P={n}_P{\eta}(P)
        {\tilde {\cal M}}_{l}(P),                              \label{78}
\end{equation}

 where $n_P=1$ for $P=K, D, B_d, T_u$; $n_P=2$ for $P=B_s, 
 T_c$, $l=1$ for $K, D, B_s, T_u$; $l=2$ for $B_d, T_c$. 
 Universality means explicitly that :
\begin{eqnarray}
 {\tilde {\cal M}}_{1}(K)&=&{\tilde {\cal M}}_{1}(D)=
 {\tilde {\cal M}}_{1}(B_s)={\tilde {\cal M}}_{1}(T_c),\nonumber\\
 &\simeq&{\tilde {\cal M}}_{2}(B_d)
 ={\tilde {\cal M}}_{2}(T_u).
                                                              \label{79}
\end{eqnarray}
 The explanation of ${n}_P$ is such that $K$ and $D$ 
 have one ${\bf y}$-particle and one ${\bf y}$-particle 
 exchanges; $B_d$ and $T_u$ have two ${\bf y}$-particles 
 and both of them exchange simultaneously, so 
 we set ${n}_P=1$ for them. On the other hand 
$B_s$ and $T_c$ have two 
 ${\bf y}$-particles but one of them exchanges, 
 so they have ${n}_P=2$ because the probability becomes 
 double. 
 The `` ${l}$ '' means the number of exchanging 
 ${\bf y}$-particles in the present diagram. 
 Concerning ${\eta}(P)$, we explain as follows : 
 In our FB-model $P^0$-$\overline{P^0}$ mixing occurs by 
 the ``contact interaction'' of two quarks colliding 
 inside $P^0(\overline{P^0})$-meson. 
 Therefore the probability of this interaction may be 
 considered inverse proportional to the volume of the 
 present $P^0(\overline{P^{0}})$-meson, e.g., 
 the larger radius $K$-meson 
 gains the less-valued probability of the colliding than 
 the smaller radius $D$- (or $B_s$-) meson. 
 The various aspects of hadron dynamics seem to be 
 successfully illustrated by the semi-relativistic 
 picture with ``Breit-Fermi Hamiltonian''[53]. 
 Assuming the power-law potential 
 $V(r)\sim{r}^{\nu}$($\nu$ is a real number), the radius 
 of $P^0(\overline{P^{0}})$-meson (denoted by ${\bf r}_P)$ is  
 proportional to ${\mu}_P^{-1/(2+\nu)}$, where 
 ${\mu}_P$ is the reduced mass of two quark-masses inside 
 $P^0(\overline{P^{0}})$-meson[53]. 
 Then the volume of $P^0(\overline{P^{0}})$-meson is 
 proportional to ${\bf r}_P^{3}\sim{\mu}_P^{-3/(2+\nu)}$. 
 After all we could assume for ${\eta}(P)$ in Eq.(78) as :
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 {\eta}(P)&=&{\xi}(\frac{{\mu}_P}
 {{\mu}_K})^{1.0}\hspace{2cm}{\rm for\hspace{3mm}
 linear-potential},\\
 &=&{\xi}(\frac{{\mu}_P}{{\mu}_K})^{1.5}
 \hspace{1.7cm}{\rm for\hspace{8mm}
 log-potential},                                      \label{80}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 where $\xi$ is a dimensionless numerical factor 
 depending on the details of the dynamics 
 of the quark-level. The  
 ${\eta}(P)$ is normalized by ${\mu}_K$ (reduced mass 
 of $s$- and $d$-quark in $K$ meson) for convenience. 

 %We may think that the ${\bf y}$-exchange  
 %is described by the overlapping of the wave functions of 
 %two quarks inside $P^0(\overline{P^{0}})$-meson. 
 %Then we write as :
\par 
 The present experimental results of $\Delta M_P$ are as 
 follows[29][51] : 
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 {\Delta}M_{K}^{exp}&=&(3.489\pm0.008) \times 10^{-15}
                 \hspace{2cm}{\rm GeV},\\
 {\Delta}M_{D}^{exp}&<&4.6 \times 10^{-14}
                 \hspace{4.2cm}{\rm GeV},\\
 {\Delta}M_{B_d}^{exp}&=&(3.12 \pm 0.11) \times 10^{-13}
                 \hspace{2.4cm}{\rm GeV},\\
{\Delta}M_{B_s}^{exp}&>&7.0 \times 10^{-12}
                 \hspace{4.2cm}{\rm GeV}.                 \label{81}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 Using Eq.(76), (77) and (81), we have :
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 |{\cal M}_D|     &<& \hspace{2mm}1.4  |{\cal M}_K|,\\
 |{\cal M}_{B_d}| &=&            4.92 |{\cal M}_K|,\\
 |{\cal M}_{B_s}| &>&            86.0 |{\cal M}_K|.        \label{82}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 At the level of ${\cal M}_P$, it seems that :

\begin{equation} 
  \frac{|{\cal M}_P|}{|{\cal M}_K|} \simeq O(1) 
  \sim O(100),                                              \label{83}
\end{equation}

 where $P=D, B_d, B_s$. 
\par 
Here adopting not ${\cal M}_P$ but ``${\tilde {\cal M}}_{l}(P)$``
let us make following discussions. 
By use of Eq.(78), (79), (80a,b) and (82b) we obtain :
{  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 {\mu}_{B_d}  &=&  4.91{\mu}_K
                \hspace{3cm}{\rm for\hspace{3mm}
 linear-potential},\\
              &=&  2.88{\mu}_K
                \hspace{3cm}{\rm for\hspace{8mm}
 log-potential},                                          \label{84}           
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 where $B_{B_d}{f}^2_{B_d}=(0.22{\rm GeV})^2$, 
 $B_K{f}^2_K=(0.17{\rm GeV})^2$ are used. 
 This result does not seem ``$unnatural$''.
 Comparing with the case of Eq.(75), 
 we can evade the large enhancement by $b$-quark mass 
 effect. 
 This is because the quark mass dependence is 
 introduced through the ``$reduced\hspace{2mm} mass$''
  (in which the 
 effect of heavier mass decreases). 
 Some discussions are as follows : 
 If we adopt the pure non-relativistic picture 
 it may be that 
 ${\mu}_K \simeq {\mu}_{B_d} \simeq {m}_d
 \simeq ({\mu}_D \simeq {\mu}_{T_u})$ but from the 
 semi-relativistic standpoint it seems preferable that 
 ${\mu}_K (< {\mu}_{D}) < 
  {\mu}_{B_d}(<{\mu}_{T_u})$ because the effective 
 mass value of ``$d$-quark'' in $B_d$-meson is considered 
 larger than that in $K$-meson. 
 It may be caused by that the kinetic energy of 
 ``$d$-quark'' in $B_d$-meson is larger than that in 
 $K$-meson owing to the presumption : 
 ${\bf r}_{B_d}<{\bf r}_K$ where ${\bf r}_p$ means 
 the radius of $p$-meson(Refer to discussions in 
 Sect.3).
 Then we can expect the plausibility of Eq.(84). 
 Of course it may be also a question whether 
 $|{\tilde {\cal M}}_{1}(K)|\simeq
 |{\tilde {\cal M}}_{2}(B_d)|$
 is good or not (this point influences Eq.(84)),
 which will become clear when the experimental 
 result about ${\Delta}M_{T_u}$ is confirmed in future and 
 compared with ${\Delta}M_{B_d}$. 
 \par
 Next, let us discuss ${\Delta}M_D$. 
 Here we write ${\Delta}M^{\bf y}_{P}$ as the mass difference 
 of $P^{0}$and $\overline{P^{0}}$ by ${\bf y}$-exchange interaction.
 If we set ${\mu}_D={\mu}_K$ tentatively in Eq,(80)(though 
 ${\mu}_D>{\mu}_K$ in practice) we obtain :
 
\begin{equation}
 ({\Delta}M_D^{\bf y})_{LL}=4.67 \times {\Delta}M_{K}^{exp}
           = 1.6  \times 10^{-14}\hspace{1cm} {\rm GeV},        \label{85}
\end{equation}

 where $LL$ means ``lowest limit'' and we use 
 $B_D{f}_D^2=(0.19 {\rm GeV})^2$ and Eq.(76), 
 (77), (78), (79) and (81a). In the same way, assuming 
 ${\mu}_D=1.5 \times {\mu}_K$ for example and 
 using Eq.(79) we have :
 
\begin{equation}
 {\Delta}M_D^{\bf y}
  =(2.9 \sim 5.4) \times 10^{-14}\hspace{3.1cm}{\rm GeV},       \label{86}
\end{equation}

where the parenthesis means that (linear-potential $\sim$ 
log-potential).
 This result is consistent and comparable with Eq.(81b).
 These values are similar to the results 
 by Cheng and Sher[68] and Liu and Wolfenstein[47]. 
\par 
 The study of ${\Delta}M_{B_s}$ is as follows. 
 Both $s$- and $b$-quark in $B_s$-meson are rather massive 
 and then supposing availability of the non-relativistic 
 scheme we have :
\begin{equation} 
 {\mu}_{B_s}
    =\frac{{m}_s {m}_b}{{m}_s+{m}_b}
    =0.19 \hspace{2.7cm}{\rm GeV},                            \label{87}
\end{equation}
 where ${m}_s=0.2$ GeV and ${m}_b=4.3$ GeV
 are used.
 If we adopt ${\mu}_K=0.01$ GeV($\simeq{m}_d$) 
 for example we obtain :
 {  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 {\eta}(B_s) &=& 19.0 {\xi}
\hspace{2cm}{\rm for\hspace{3mm}linear-potential},\\
                    &=& 82.8 {\xi}
\hspace{2cm}{\rm for\hspace{8mm}log-potential},                \label{88}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 By using Eq.(76b), (77), (78) and (79) we have :

\begin{equation} 
     {\Delta}M_{B_s}^{\bf y}= 
 \frac{2B_{B_s}{f}^2_{B_s}}{B_K{f}^2_K}
 \frac{M_{B_s}}{M_K}
 \frac{{\eta}(B_s)}{{\eta}(K)}
 {\Delta}M_K^{\bf y},                                        \label{89}
\end{equation}

 where factor 2 comes from $n_{B_s}=2$ in Eq.(78).
 Assuming that ${\Delta}M_K^{\bf y}={\Delta}M_K^{exp}$ 
 (that is, the ${\bf y}$-exchange saturates the 
 ${\Delta}M_K^{exp}$ bound) and using Eq.(88a,b) we obtain :
 
\begin{equation}
 {\Delta}M_{B_s}^{\bf y}=(0.31\sim1.4)\times10^{-11}
 \hspace{3cm}{\rm GeV},                                       \label{90}
\end{equation}

 where we use $B_{B_s}{f}^2_{B_s}=(0.25{\rm GeV})^2$[49]
 (the parenthesis means the same as Eq.(86)).
 This estimation is consistent with Eq.(81d).
 From Eq.(81c) and (90) we get :
 {  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 \frac{{\Delta}M_{B_s}^{\bf y}}{{\Delta}M_{B_d}^{\bf y}} &=& 
  (10 \sim 50),\\
            x_s={\Delta}M_{B_s}^{\bf y}{\tau}_{B_s}&=&
 (8 \sim 30),                                                \label{91}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 where we set ${\Delta}M_{B_d}^{\bf y}={\Delta}M_{B_d}^{exp}$ 
 and use ${\tau}_{B_s}=2.4
         \times10^{12}$ ${\rm GeV}^{-1}$[29],
 and also the parenthesis means the same as Eq.(86).        
 Note that the present experimental result is  
 ${\Delta}M_{B_s}^{exp}/{\Delta}M_{B_d}^{exp}>22$[29].
 If we adopt the box diagram calculation in the SM and 
 use Eq.(74) with the unitary assumption of CKM-matrix 
 elements, it is found that[50][51] :
\begin{equation} 
 \frac{{\Delta}M_{B_s}^{SD}}{{\Delta}M_{B_d}^{SD}} 
        =17\sim52.                                         \label{92}
\end{equation}
 Therefore, from the above studies of ${\Delta}M_{B_d}$ 
 and ${\Delta}M_{B_s}$ it is difficult to clarify 
 which scheme (${\bf y}$-exchange or SD in the SM) is favorable, 
 at least until the future experiments confirm 
 the values of $|V_{ts}/V_{td}|$ and ${\Delta}M_{B_s}$. 
\par  
 Finally let us estimate ${\Delta}M_{T_u}^{\bf y}$ and 
 ${\Delta}M_{T_c}^{\bf y}$. 
 Setting ${\mu}_{T_u}={\mu}_{B_d}$
 (though ${\mu}_{T_u}>{\mu}_{B_d}$ in practice)  
 and using Eq.(76b), (77), (78), (79) and (80) we estimate 
 the lowest limit of ${\Delta}M_{T_u}^{SW}$ (denoted by 
 $({\Delta}M_{T_u}^{\bf y})_{LL}$) as :
\begin{equation} 
      ({\Delta}M_{T_u}^{\bf y})_{LL}= 
 \frac{B_{T_u}{f}^2_{T_u}}{B_{B_d}{f}^2_{B_d}}
 \frac{M_{T_u}}{M_{B_d}}
 {\Delta}M_{B_d}^{\bf y}
      =7.3 \times 10^{-10}\hspace{1cm}{\rm GeV},         \label{93} 
\end{equation}
 where we use $B_{T_u}{f}^2_{T_u}=(1.9{\rm GeV})^2$[36], 
 $M_{B_d}=5.279$ GeV, $M_{T_u}=171$ GeV and set 
 ${\Delta}M_{B_d}^{\bf y}={\Delta}M_{B_d}^{exp}$ in Eq.(81c). 
 Note that $|{\tilde {\cal M}}_2(T_u)|=
 |{\tilde {\cal M}}_2(B_d)|$ is used in Eq.(93).
 Cheng and Sher's scheme[68] predicts 
 ${\Delta}M_{T_u}\simeq10^{-7}$ GeV which is order of $10^3$
 larger than Eq.(93). (In Ref.[68] they estimated
 ${\Delta}M_T\simeq 10^{-10}$ GeV using smaller 
 $t$-quark mass value than $170$ GeV). 
 For evaluating ${\Delta}M_{T_c}$, we calculate :
\begin{equation}
 {\mu}_{T_c}
    =\frac{{m}_c {m}_t}{{m}_c+{m}_t}
    =1.34 \hspace{2.7cm}{\rm GeV},                     \label{94}
\end{equation}
 where ${m}_c=1.35$ GeV and ${m}_t=170$ GeV
 are used. Then we get from Eq.(80a,b) :
 {  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 {\eta}(T_c) &=& 134 {\xi}
\hspace{2cm}{\rm for\hspace{3mm}linear-potential},\\
                    &=& 1551 {\xi}
\hspace{1.8cm}{\rm for\hspace{7.5mm}log-potential},     \label{95}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 where we set ${\mu}_K=0.01$ GeV for example.
 After all with Eq.(95a,b) we obtain :
\begin{equation} 
      {\Delta}M_{T_c}^{\bf y}= 
 \frac{2B_{T_c}{f}^2_{T_c}}{B_K{f}^2_K}
 \frac{M_{T_c}}{M_{K}}
 \frac{{\eta}(T_c)}{{\eta}(K)}
      {\Delta}M_K^{\bf y}
      =(4\sim47) \times 10^{-8}\hspace{5mm}{\rm GeV},    \label{96} 
\end{equation}
 where we adopt $n_{T_c}=2$, $B_{T_c}{f}^2_{T_c}=
 (1.9{\rm GeV})^2$[36], $M_{T_u}=171$ GeV and 
 ${\Delta}M_K^{\bf y}={\Delta}M_K^{exp}$
 and the parenthesis means the same as Eq.(86). 
 Note that $|{\tilde {\cal M}}_1(T_c)|=
 |{\tilde {\cal M}}_1(K)|$ is used in Eq.(96).
 
\par
\vspace{1cm}
  {\bf b. 
 Indirect CP violation in $P^0$-$\overline{P^0}$ mixing}
\newline
 Here we discuss 
 CP violation by mass-mixings which is assumed 
 to be saturated by the ``${\bf y}$-exchange interactions''. 
 In the CP-conserving limit in the $P^0
 (\overline{P^{0}})$-meson systems, 
 $M_{12}(P)$s are supposed to be real positive. 
 Note that $CP|P_H>=-|P_H>$ and 
 $CP|P_L>=|P_L>$ where $H$ $(L)$ means heavy (light). 
 If the CP-violating ${\bf y}$-exchange interactions are switched on, 
 $M_{12}(P)$ becomes complex. Following G\'erard and
 Nakada's notation[48][52], we write as :
 
\begin{equation}
 M_{12}=|M_{12}|\exp(i{\theta}_P),                              \label{97} 
\end{equation}
 
 with :
\begin{equation} 
 \tan{\theta}_P=\frac{{\rm{Im}}M_{12}(P)}{{\rm{Re}}M_{12}(P)}.       \label{98}
\end{equation}
 As we assume that the ${\bf y}$-exchange interaction saturates 
 CP violation, we can write :
 
\begin{equation} 
 {\rm{Im}}<\overline{P^0}|H^{\bf y}_{{\Delta}F=2}|P>=
       {\rm{Im}}M_{12}(P).                                          \label{99}
\end{equation} 
 From Eq.(76a), (77) and (78) we obtain :
  
\begin{equation} 
 {\rm{Im}}M_{12}(P)={\cal C} \cdot {\rm{Im}}
 {\tilde {\cal M}}_{l}(P),                                           \label{100}
\end{equation}

 where ${\cal C}=
   (1/12\pi^2)B_P{f}^2_PM_P{\eta}(P)$. 
 Therefore the origin of CP violation of $P^0(\overline
 {P^{0}})$-meson 
 system is only in ${\tilde {\cal M}_{l}(P)}$. 
 The Factor ``${\cal C}$'' in Eq.(100) is common also in 
 ${\rm{Re}}M_{12}(P)$ and then we have :
 
\begin{equation} 
  \frac{{\rm{Im}}M_{12}(P)}{{\rm{Re}}M_{12}(P)}=
     \frac{{\rm{Im}}{\tilde {\cal M}}_{l}(P)}
     {{\rm{Re}}{\tilde {\cal M}}_{l}(P)}.                          \label{101}
\end{equation}          

 If the universality of Eq.(79) is admitted, we obtain :
 {  
\setcounter{enumi}{\value{equation}}
\addtocounter{enumi}{1}
\setcounter{equation}{0}
\renewcommand{\theequation}{\theenumi\alph{equation}}
\begin{eqnarray}
 {\theta}_K &=& {\theta}_D=
 {\theta}_{B_s} = {\theta}_{T_c},\\
 {\theta}_{B_d} &=& {\theta}_{T_u}
 \simeq 2{\theta}_K,                                              \label{102}
\end{eqnarray}
\setcounter{equation}{\value{enumi}}}
 These are the predictions about indirect CP violation from 
 the stand point of our FB-model. Concerning Eq.(102b) 
 if two {\bf y}-particles in ${B_d}$ and ${T_u}$ (See Fig.(3).)
 exchange without any correlation, it is possible that
 CP phases become double of ${\theta}_K$ and if there exists
 some correlation they become more or less than $2{\theta}_K$. 
 As the experimental result it is reported that[47] :

\begin{equation} 
 {\theta}_K=(6.5\pm0.2) \times 10^{-3}.                            \label{103}
\end{equation}  


 \par
{\bf Acknowledgements}
\par
 We would like to thank Z.Maki for valuable suggestions 
 and discussions. 
 We would also like to thank N.Maru and Y. Matsui 
 and A.Takamura for useful discussions and 
 the hospitality at the laboratory of the 
 elementary particle physics in Nagoya University.






 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%      %%%%%%                       Reference               %%%%%%%%%%%%%        %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{thebibliography}{99}
 \bibitem{1}F.Abe, et.al. Phys. Rev. {\bf D50}(1994)2966; 
           F.Abe, et.al. 
            Phys. Rev. Lett. {\bf 73}(1994)
 \newline
            225.
 \bibitem{2}L. Lyons, Prog. in Partic. \& Nucl. Phys. 
            {\bf 10}(1984)227.
 \bibitem{3}J.D.Bjorken, Phys. Rev. {\bf D19}(1979)335.
 \bibitem{4}P.Q.Hung and J.J.Sakurai, Nucl. Phys. {\bf B143}
            (1978)81; J.J.Sakurai, Max-Plank-
\newline
            Inst. preprint MPI-PAE/PTh44/82. July 1982.
 \bibitem{5}W.Kr\' olikowski, Act. Phys. Pol. {\bf B11}(1980)431.  
 \bibitem{6}E.G.H.Harari and N.Seiberg, Phys. Lett. 
           {\bf 98B}(1981)269.
 \bibitem{7}D.W.Greenberg and J.Sucher, Phys. Lett. 
           {\bf 99B}(1981)239.
 \bibitem{8}L. Abbot and E.Farhi, Phys. Lett. {\bf 101B}(1981)69.
 \bibitem{9}H.Fritzsch and G.Mandelbaum, Phys. Lett. 
 {\bf 102B}(1981)319; ibid,{\bf 109B}(1982)224.
 \newline
         R.Barbieri, R.N.Mohapatra and A.Masiero, Phys. Lett.
          {\bf 105B}(1981)369.    
 \bibitem{10}G.Gounaris, R.K\"ogerler and D.Schildknecht, 
            Phys. Lett. {\bf 133B}(1983)118.
 \bibitem{11}B.Schrempp and F.Schrempp, Nucl. Phys. 
            {\bf B231}(1984)109. 
 \bibitem{12}F.Abe, et al., Phys. Rev. Lett.{\bf 77}(1996)5336
 ; ibid, {\bf 79}(1997)2198.
 \bibitem{13}J.Huston et al., Phys. Rev. Lett. 
             {\bf 77}(1996)444;E.W.N.Glover et al., Durham Univ.
 \newline
             preprint. DPT/96/22,RAL-TR-96-019(1996). 
 \bibitem{14}T.Matsushima, Nagoya Univ. preprint DPNU-89-31,
             unpublished; T.Matsushi-
\newline
             ma, Nuovo Cim.{\bf 106A}(1993)139.
  \bibitem{15}S.Kobayashi, Can. J. Math. {\bf 8}(1956)145; 
             Ann.di Math, {\bf43}(1957)119.
 \bibitem{16} W.Drechsler, J, Math. Phys. {\bf 18}(1977)1358; 
             Lecture Notes in Physics, 67,edited by 
             J.Ehles ,et.al. (Springer Verlarg) p.1.
 \bibitem{17}R.Gilmore, Lie Groups, Lie Algebras, and Some 
             of Their
             Applications(Jhon Wiley \& Sons 1974) p50.
 \bibitem{18}H.Geogi and S.L.Glashow, Phys. Rev. Lett. 
            {\bf 32}(1974)438. 
 \bibitem{19}H.Fritzsch, Lectures given at the International 
             School on Subnuclear Physics, 
\newline
             Erice, Sicily August(1984), Max-Plank-Inst. 
             preprint MPI-PAE/PTh85/84 Nove-
\newline
             mber(1984). 
 \bibitem{20}J.C.Pati and A.Salam, Nucl. Phys. 
             {\bf B214}(1983)109.
 \bibitem{21}Y.Ohnuki, Unitary Representation of 
             the Poincar\'e Group and Relativistic 
             Wave Equations(World Scientific, 
             Singapore, 1988)p.1.
 \bibitem{22}Z.Maki, Proc. of 1981 INS Symposium 
             on Quark and Lepton 
             Physics, edited. by K.Fujikawa et al. p325;
             Concerning ``Primon'', private communication.
 \bibitem{23}G.'tHooft, Recent Developments in Gauge Theories, 
             edited by G.'tHooft, et al.
             (Plenum Press, N.Y.,1980) p135.
 \bibitem{24}M.Peskin, Proc. of the 1981 Int. Symp. 
             on Lepton and Photon 
             Interactions at High Energy, W.Pfeil et al. 
             (Bonn, 1981) p880.
 \bibitem{25}H.Harari and N.Seiberg Nucl.Phys. {\bf B204}(1982)141.
 \bibitem{26}L.F.Abbot and E.Farhi, Nucl. Phys. {\bf B189}(1981)547.
 \bibitem{27}D.Schildknecht, Electro weak Effects at High 
             Energies edited by H.B.Newman
             (Plenum Pub. Comp. 1985)p.1.
 \bibitem{28}H.Terazawa and K.Akama, Phys. Lett. {\bf 101B}(1981)190; 
             H.Terazawa, Mod. Phys. Lett. {\bf A36}(1992)3373.
             H.Terazawa, Prog. Theor. Phys. {\bf 58}(1977)1276; ibid.
             {\bf 64}(1980)1763.
 \bibitem{29}Review of Particle Physics, The European 
             Physical Journal, {\bf C15}(2000)1. 
 \bibitem{30}I.S.Towner, Phys. Lett. {\bf B333}(1994)13; A.Sirlin, 
             New York Univ. New York. Rep. No. 
             , unpublished.
 \bibitem{31}H.Terazawa, Univ. of Tokyo; Inst. for Nuclear Study
             INS-Rep. 1131 Jan. 1996.
 \bibitem{32}S.Weinberg, Phys. Lett. {\bf 102B}(1981)401.
 \bibitem{33}H.Fritzsch, Lectures in Proc. Arctic School of 
          Physics, $\breve{A}$k$\ddot{a}$slompolo. Finland (1982).
 \bibitem{34}W.Kr\'olikowski, Act. Phys. Pol. {\bf B11}(1979)
 739; Phys. Lett. {\bf 85B}(1979)335.
 \bibitem{35}L.Wolfenstein, Phys. Rev. Lett. 
             {\bf 13}(1964)562.
 \bibitem{36}L.L.Chau. Phys. Rep. {\bf 95}(1983)1; A.Pascos and 
             U.T$\ddot{\rm u}$rke, Phys. Rep. {\bf 178}
             (1989)145; 
             P.J.Franzini, Phys. Rep. {\bf 173}(1989)1.
 
 
 \bibitem{37}M.K.Gaillard and B.W.Lee, Phys. Rev. 
             {\bf D10}(1974)897.
 \bibitem{38}L.Wolfenstein, Nucl. Phys. {\bf B160}(1979)501.
 \bibitem{39}A.Datta and D.Kumbhakar, Z. Phys. 
             {\bf C27}(1985)515.
 \bibitem{40}A.Datta, Phys. Lett. {\bf B154}(1985)287.
 \bibitem{41}A.D.Hadeed and B.Holdom, Phys. Lett. 
             {\bf B159}(1985)379.
 \bibitem{42}I.Picek, Phys. Lett. {\bf B159}(1985)385.
 \bibitem{43}L.Wolfenstein, Phys. Lett. 
             {\bf B164}(1985)170.
 \bibitem{44}J.F.Donoghue, E.Golowich, B.R.Holstein and 
             J.Trampeti\'c, 
\newline
             Phys. Rev. {\bf D33}(1986)179.
 \bibitem{45}H.Geogi, Phys. Lett. {\bf B279}(1992)353; T.Ohl,
             G.Ricciardi and E.H.Simmons, Nucl. Phys. {\bf 
             B403}(1993)605, E.Golowich and A.A.Peterov, 
             Phys. Lett. {\bf B427}(1998)172.
             
 \bibitem{46}A.Datta and J.C.Pati, Phys. Lett. 
             {\bf B164}(1985)173.
 \bibitem{47}J.Liu and L.Wolfenstein, Phys. Lett. 
             {\bf B197}(1987)536. 
 \bibitem{48}J.M.G\'erard and T.Nakada, Phys. Lett. 
             {\bf B261}(1991)474.
 \bibitem{49}A.Ali and D.London, Z. Phys. 
             {\bf C65}(1995)431.
 \bibitem{50}J.L.Rosner, Nucl. Phys. {\bf B}(Proc. Suppl){\bf 59}
             (1997)1.
 \bibitem{51}A.Stocchi, LAL 99-01 January(1999); G.Blaylock at
             the XIX International Symposium on Lepton Photon 
             Interactions at High Energies, Stanford University,
             1999(http://LP99. slac. stanford. edu/db/program.asp).
 \bibitem{52}T.Nakada, PSI-PR-91-02 January(1991).
 \bibitem{53}A.DeR\'ujula, H.Georgi and S.L.Glashow, Phys. 
             Rev. {\bf D12}(1975)147; C.Quigg 
             and J.D.Rosner, Phys. Rep. {\bf 56}(1979)167.
 \bibitem{54}Y.Yamanaka, H.Sugawara and S.Pakvasa, Phys.
             Rev. {\bf D25}(1982)1895.
 \bibitem{55}N.Cabibbo, Phys. Rev. Lett. {\bf 10}(1963)531; 
             M.Kobayashi and K.Maskawa, Prog. Theor. Phys. 
             {\bf 49}(1973)652.
 \bibitem{56}L.Lavoura, Int. Jour. of Mod. Phys. 
             {\bf A9}(1994)1873.
 \bibitem{57}G.C.Branco and A.I.Sanda, Phys. Rev. 
             {\bf D26}(1982)3176.
 \bibitem{58}J.M.Soares and L.Wolfenstein, Phys. Rev. 
             {\bf D47}(1993)1021.
 \bibitem{59}J.P.Alexander et al., Phys. Rev. Lett. 
             {\bf 77}(1996)5000.
 \bibitem{60}J.L.Hewett,
             SLAC-PUB-95-6821,April(1995),; 
 \newline            
             Y.Nir, WIS-95/28/Jul-PH,.
 \bibitem{61}Preliminary version is in T.Matsushima, TMI-97-1,
             March(1997)/.
 \bibitem{62}Z.Maki, M.Nakagawa and S.Sakata, Prog. Theor. Phys.
             {\bf 28}(1962)870.
 \bibitem{63}Y.Fukuda et al., Phys. Rev. Lett. {\bf 81}
             (1998)1562.
 \bibitem{64}A.Alavi-Harati et al., Phys. Rev. Lett. {\bf 83}(1999)22.            
 \bibitem{65}M.Grounau, Phys. Lett. {\bf B300}(1993)163.            
 \bibitem{66}G.D.Barr et al., Phys. Lett. {\bf B317}(1993)233.            
 \bibitem{67}T.Affolder et. al., Phys. Rev. {\bf D61}(2000)072005            
 \bibitem{68}T.P.Chen and M.Sher, Phys. Rev.{\bf 35}(1987)3484            
 \bibitem{69}R.Gupta             
 \bibitem{70}L.Wolfenstein, Phys. Rev. Lett. {\bf 51}(1983)1945            
 %\bibitem{63}             
             
             
             
             
             
             
             


\end{thebibliography}
\clearpage
  
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\setlength{\unitlength}{0.7mm}
\begin{figure}
\begin{center}
\begin{picture}(180,270)(15,-20)
     %Subquark-line diagrams of the weak interactions}
\put(47,216){\makebox(7,7){${\bf \overline{\alpha}}$}}
\put(43,207){\makebox(7,7){${\bf \overline{x}}$}}
\put(35,213){\makebox(7,7){${\bf \overline u}$}}
\put(49,170){\makebox(7,7){${\bf \overline{\alpha}}$}}
\put(45,176){\makebox(7,7){${\bf \overline x}$}}
\put(42,182){\makebox(7,7){${\bf x}$}}
\put(35,173){\makebox(7,7){${\bf d}$}}
\put(132,221){\makebox(7,7)[bl]{${\bf \overline{\alpha}}$}}
\put(135,215){\makebox(7,7)[bl]{${\bf \overline x}$}}
\put(138,209){\makebox(7,7)[bl]{${\bf \overline x}$}}
\put(145,215){\makebox(7,7){${\bf e^-}$}}
\put(140,180){\makebox(7,7){${\bf x}$}}
\put(135,171){\makebox(7,7){${\bf \overline{\alpha}}$}}
\put(145,173){\makebox(7,7){${\bf \overline{\nu}}$}}
\put(90,207){\makebox(10,10)[b]{${\bf W^-}$}}
\put(83,205){\line(-2,1){28}}
\put(75,197){\line(-2,1){24}}
\put(75,197){\line(-2,-1){23}}
\put(86,197){\line(-2,-1){32}}
\put(83,190){\line(-2,-1){27}}
\put(86,185){\vector(1,0){15}}
\put(83,205){\line(1,0){22}}
\put(86,197){\line(1,0){14}}
\put(83,190){\line(1,0){22}}
\put(105,205){\line(2,1){25}}
\put(100,197){\line(2,1){32}}
\put(112,197){\line(2,1){22}}
\put(112,197){\line(2,-1){26}}
\put(105,190){\line(2,-1){29}}
\put(46,147){\makebox(7,7)[br]{${\bf {\alpha}}$}}
\put(42,141){\makebox(7,7)[br]{${\bf x}$}}
\put(38,135){\makebox(7,7)[br]{${\bf x}$}}
\put(39,110){\makebox(7,7){${\bf \overline{x}}$}}
\put(42,105){\makebox(7,7){${\bf \overline x}$}}
\put(46,100){\makebox(7,7){${\bf \overline{\alpha}}$}}
\put(135,147){\makebox(7,7)[bl]{${\bf \alpha}$}}
\put(139,142){\makebox(7,7)[bl]{${\bf \alpha}$}}
\put(143,135){\makebox(7,7)[bl]{${\bf x}$}}
\put(143,110){\makebox(7,7)[l]{${\bf \overline{x}}$}}
\put(139,105){\makebox(7,7)[l]{${\bf \overline{\alpha}}$}}
\put(135,100){\makebox(7,7)[l]{${\bf \overline{\alpha}}$}}
\put(35,140){\makebox(10,10){${\bf e^+}$}}
\put(35,100){\makebox(10,10){${\bf e^-}$}}
\put(145,140){\makebox(10,10){${\bf W^+}$}}
\put(145,100){\makebox(10,10){${\bf W^-}$}}
\put(88,129){\makebox(10,10){${\bf Z_1^0}$}}
\put(83,130){\line(-2,1){30}}
\put(80,125){\line(-2,1){30}}
\put(70,125){\line(-2,1){22}}
\put(70,125){\line(-2,-1){22}}
\put(80,125){\line(-2,-1){30}}
\put(86,115){\vector(1,0){15}}
\put(83,120){\line(-2,-1){30}}
\put(83,130){\line(1,0){20}}
\put(83,120){\line(1,0){20}}
\put(103,130){\line(2,1){30}}
\put(106,125){\line(2,1){30}}
\put(116,125){\line(2,1){22}}
\put(116,125){\line(2,-1){22}}
\put(106,125){\line(2,-1){30}}
\put(103,120){\line(2,-1){30}}
\put(46,87){\makebox(7,7)[br]{${\bf {\alpha}}$}}
\put(42,81){\makebox(7,7)[br]{${\bf x}$}}
\put(38,75){\makebox(7,7)[br]{${\bf x}$}}
\put(39,50){\makebox(7,7){${\bf \overline{x}}$}}
\put(42,45){\makebox(7,7){${\bf \overline x}$}}
\put(46,40){\makebox(7,7){${\bf \overline{\alpha}}$}}
\put(135,87){\makebox(7,7)[bl]{${\bf \alpha}$}}
\put(139,82){\makebox(7,7)[bl]{${\bf x}$}}
\put(143,75){\makebox(7,7)[bl]{${\bf \alpha}$}}
\put(143,50){\makebox(7,7)[l]{${\bf \overline{\alpha}}$}}
\put(139,45){\makebox(7,7)[l]{${\bf \overline{x}}$}}
\put(135,40){\makebox(7,7)[l]{${\bf \overline{\alpha}}$}}
\put(35,80){\makebox(10,10){${\bf e^+}$}}
\put(35,40){\makebox(10,10){${\bf e^-}$}}
\put(145,80){\makebox(10,10){${\bf W^+}$}}
\put(145,40){\makebox(10,10){${\bf W^-}$}}
\put(88,69){\makebox(10,10){${\bf Z_2^0}$}}
\put(83,70){\line(-2,1){30}}
\put(76,67){\line(-2,1){25}}
\put(70,65){\line(-2,1){22}}
\put(70,65){\line(-2,-1){22}}
\put(76,63){\line(-2,-1){25}}
\put(86,55){\vector(1,0){15}}
\put(83,60){\line(-2,-1){30}}
\put(83,70){\line(1,0){20}}
\put(76,67){\line(1,0){34.5}}
\put(76,63){\line(1,0){34.5}}
\put(83,60){\line(1,0){20}}
\put(103,70){\line(2,1){30}}
\put(110.5,67){\line(2,1){25}}
\put(116,65){\line(2,1){22}}
\put(116,65){\line(2,-1){22}}
\put(110.5,63){\line(2,-1){25}}
\put(103,60){\line(2,-1){30}}
\end{picture}
\end{center}
\caption{Subquark-line diagrams of the weak interactions}
\end{figure}

\begin{figure}
\setlength{\unitlength}{0.7mm}
\begin{center}
\begin{picture}(180,250)(15,-20)
\put(0,115){\makebox(80,80)[l]{\Huge ${\bf y}$}}
\put(5,165){\makebox(30,30)[r]{${\Lambda}
             \hspace{1mm}({\Theta})$}}
\put(5,115){\makebox(30,30)[r]{${\overline{\Lambda}
             \hspace{1mm}(\overline{\Theta})}$}}
\put(40,180){\line(1,0){60}}
\put(40,130){\line(1,0){60}}
\put(100,130){\line(0,1){50}}
\multiput(100,180)(4,0){10}{\line(1,0){3}}
\multiput(100,130)(4,0){10}{\line(1,0){3}}
\put(120,165){\makebox(30,30)[r]{${\bf g}_h$}}
\put(120,115){\makebox(30,30)[r]{${\bf g}_h$}}
\end{picture}
\end{center}
\caption{The (${\bf y} \longrightarrow 2{\bf g}_h)$-process by 
 primon-level diagram}
\end{figure}


\setlength{\unitlength}{0.8mm}
\begin{figure}
\begin{center}
\begin{picture}(180,250)(15,-20)
\put(140,155){\makebox(30,30)[l]{${\overline{K^0}}
             \hspace{1mm}({\overline{D^0}})$}}
\put(0,155){\makebox(30,30)[r]{${K^0}
             \hspace{1mm}({D^0})$}}
\put(30,182){\makebox(15,10)[r]{${\bf {\overline{s}}
             \hspace{1mm}(c)}$}}
\put(60,190){\line(1,0){60}}
\put(67,191){\makebox(50,10)[b]{${\bf {\alpha x \overline{x}
                                 \hspace{1mm}(\alpha x)}}$}}
\put(130,182){\makebox(15,10)[l]{${\bf {\overline {d}}
             \hspace{1mm}(u)}$}}
\put(50,180){\makebox(7,7)[r]{${\bf y}$}}
\put(60,183){\line(1,0){30}}
\put(90,160){\line(0,1){23}}
\put(90,160){\line(1,0){30}}
\put(123,157){\makebox(7,7)[l]{${\bf y}$}}
\put(30,150){\makebox(15,10)[r]{${\bf {d\hspace{1mm}
             (\overline{u})}}$}}
\put(60,153){\line(1,0){60}}
\put(130,151){\makebox(15,10)[l]{${\bf {s\hspace{1mm}
              (\overline{c})}}$}}
\put(67,154){\makebox(50,10)[b]{${\bf {\overline{\alpha} 
    \overline{x} x \hspace{1mm}(\overline{\alpha} 
    \overline{x})}}$}}
\put(140,95){\makebox(30,30)[l]{${\overline{B_s^0}}
             \hspace{1mm}({\overline{T_c^0}})$}}
\put(0,95){\makebox(30,30)[r]{${B_s^0}
             \hspace{1mm}({T_c^0})$}}
\put(30,120){\makebox(15,10)[r]{${\bf {\overline{b}}
             \hspace{1mm}(t)}$}}
\put(60,130){\line(1,0){60}}
\put(67,131){\makebox(50,10)[b]{
            ${\bf {\alpha x \overline{x}
                                \hspace{1mm}(\alpha x)}}$}}
\put(130,123){\makebox(15,10)[l]{${\bf {\overline {s}}
             \hspace{1mm}(c)}$}}
\put(50,122){\makebox(7,7)[r]{${\bf y}$}}
\put(60,125){\line(1,0){60}}
\put(123,122){\makebox(7,7)[l]{${\bf y}$}}
\put(50,116){\makebox(7,7)[r]{${\bf y}$}}
\put(60,120){\line(1,0){30}}
\put(90,100){\line(0,1){20}}
\put(90,100){\line(1,0){30}}
\put(123,97){\makebox(7,7)[l]{${\bf y}$}}
\put(50,91){\makebox(7,7)[r]{${\bf y}$}}
\put(60,95){\line(1,0){60}}
\put(123,91){\makebox(7,7)[l]{${\bf y}$}}
\put(30,87){\makebox(15,10)[r]{${\bf {s
           \hspace{1mm}(\overline{c})}}$}}
\put(60,88){\line(1,0){60}}
\put(67,89){\makebox(50,10)[b]{${\bf {\overline{\alpha} 
    \overline{x} x \hspace{1mm}(\overline{\alpha} 
    \overline{x})}}$}}
\put(130,90){\makebox(15,10)[l]{${\bf {b\hspace{1mm}
             (\overline{t})}}$}}
\put(140,35){\makebox(30,30)[l]{${\overline{B_d^0}}
             \hspace{1mm}({\overline{T_u^0}})$}}
\put(0,35){\makebox(30,30)[r]{${B_d^0}
             \hspace{1mm}({T_u^0})$}}
\put(30,60){\makebox(15,10)[r]{${\bf {\overline{b}}
             \hspace{1mm}(t)}$}}
\put(60,70){\line(1,0){60}}
\put(67,71){\makebox(50,10)[b]{${\bf {\alpha x \overline{x}
                                \hspace{1mm}(\alpha x)}}$}}
\put(130,63){\makebox(15,10)[l]{${\bf {\overline {d}}
             \hspace{1mm}(u)}$}}
\put(50,62){\makebox(7,7)[r]{${\bf y}$}}
\put(60,65){\line(1,0){32}}
\put(92,40){\line(0,1){25}}
\put(50,56){\makebox(7,7)[r]{${\bf y}$}}
\put(60,60){\line(1,0){28}}
\put(88,35){\line(0,1){25}}
\put(92,40){\line(1,0){28}}
\put(123,37){\makebox(7,7)[l]{${\bf y}$}}
\put(88,35){\line(1,0){32}}
\put(123,31){\makebox(7,7)[l]{${\bf y}$}}
\put(30,27){\makebox(15,10)[r]{${\bf {d
           \hspace{1mm}(\overline{u})}}$}}
\put(60,28){\line(1,0){60}}
\put(67,29){\makebox(50,10)[b]{${\bf {\overline{\alpha} 
           \overline{x} x \hspace{1mm}
           (\overline{\alpha} \overline{x})}}$}}
\put(130,30){\makebox(15,10)[l]{${\bf {b\hspace{1mm}
             (\overline{t})}}$}}
\end{picture}
\end{center}
\caption{  Schematic illustrations of $P^0$-$\overline{P^0}$ mixings
  by ${\bf y}$-exchange interactions}
\end{figure}


\end{document}

