There are some techniques that can be used to solve the nonlinear evolution equations, such as inverse scattering transformation, Hirota method, Darboux transformation and the tanh method [1] -[4] . Among this methods, the bilinear method and bilinear Bäcklund transformation have proved particularly powerful. Through the dependent variable transformations, some nonlinear evolution equations can be transformed into bilinear forms. Applying the bilinear method developed by Hirota, we can obtian the soliton solutions and quasiperiodic wave solutions [5] -[7] . The construction of the bilinear Bäcklund transformation [8] by using Hirota method relies on a par- ticular skill in using appropriate exchange formulas which are connected with the linear presentation of the system. Yet, the construction of bilinear Bäcklund transformation is complicated. Recently, Lambert, Gilson et al. [9] -[11] proposed an alternative procedure based on the use of the Bell polynomials which enabled one to obtain parameter families of bilinear Bäcklund transformation and Lax pairs for the soliton equations in a lucid and systematic way. In Ref [12] , Fan has constructed bilinear formalism, bilinear Bäcklund transformation, Lax pairs and infinite conservation laws for the nonisospectral and variable-coefficient KdV equation.

In this paper, we will extend the Binary Bell polynomials to deal with the modified Kadomtsev-Petviashvili (mKP) equation. First, we derive the bilinear form for the mKP equation by the binary Bell polynomials. Second, the bilinear Bäcklund transformation and Lax pairs are obtained in a quick and natural manner.

The mail tool used here is a class of generalized multi-dimensional binary Bell polynomials. First, we give some notations on the Bell polynomials to easily understand our presentation.

Lambert et al. proposed a generalization of the Bell polynomial [9] -[11] . Let, denote arbitrary integers, be a multi-variable function, the following polynomials

is called multi-dimensional Bell polynomial (generalized Bell polynomial or Y-polynomials). If all partial

derivatives are taken as different variable elements, then the

generalized Bell polynomial is the multivariable polynomial with respect to these variable

elements. The subscripts in the notation denote the highest order derivatives of

with respect to the variable respectively.

For the special case, the associated two-dimensional Bell polynomials defined by (1) read

Base on the use of above Bell polynomials (1), the multidimensional binary Bell polynomials (-poly- nomials) can be defined as follows

which is a multivariable polynomials with respect to all partial derivatives and

The binary Bell polynomials also inherits the easily recognizable partial structure of the Bell polynomials. The lowest order binary Bell polynomials are

The link between binary Bell polynomials and the standard Hirota bilinear equation

can be given by an identity

in which, and operators are classical Hirota bilinear operators defined by

In the particular case when, the formula (7) becomes

in which the P-polynomials can be characterized by an equally recognizable even part partitional structure

The formulaes (7),(9) and (10) will prove particular useful in connecting nonlinear equations with their corresponding bilinear equations. This means that once a nonlinear equation is expressible as a linear combi- nation of the P-polynomials, then it can be transformed into a linear equation.

The binary Bell polynomials can be separated into P-polynomials and Y-polynomials

The key property of the multi-dimensional Bell polynomials

implies that the binary Bell polynomials can still be linearized by means of the Hopf-Cole

transformation, that is ,.

The formulaes (11) and (13) will then provide the shortest way to the associated Lax system of nonlinear equations.

In this paper we consider the mKP equation

Let a potential field be

Substituting (15) into (14), we have

Introducing two new variables

using the binary Bell polynomials (5) and (6), equation (16) can be written into

A possible choice of such constraint maybe

then

Substituting (20) into (18) and using the relation (19), we have

Therefore, from (19) and (21), we deduce a couple system of -polynomials for the mKP equation

By application of the identity (7) and the transformation (17), equations (22) and (23) lead to the bilinear form for the mKP equation

Using Hirota's bilinear method, it is easy to solve the multisoliton solutions for the mKP equation. For example, the one-soliton solution reads

where and are variable constant.

In this section, we consider the bilinear Bäcklund transformation and Lax pair for the mKP equation.

Set

be two different solutions of (16), respectively. We associate the two-field condition

By the relation

Equation (28) can be transformed into

Let

so Equation (31) becomes

Similar to the (21), by the relation

Equation (36) can be transformed into

Then from (27) to (39), we get the system of -polynimials

Using the link between Bell Polynomials and Hirota bilinear bilinear Bäcklund transformation (7), the bilinear Bäcklund transformation can be written as

Through the bilinear Bäcklund transformation, we can get the soliton solutions for the mKP equation.

In the following, we will give the Lax pair for the mKP equation. By transformations

and the relation (13), the formulaes (48) and (50) become

Set

then by the relation (53) and (54) grow

which is the Lax pair of the mKP equation.

Similar to the (56) and (57). Let, (49) and (51) make

Let, then, by the relation (58) and (59) develop into

which is the Lax pair for the mKP equation.

Binary Bell Polynomials play an important role in the characterization of bilinear equation. By the Binary Bell Polynomials, we give the bilinear form, bilinear Bäcklund transformation and Lax pairs for the modified Kadomtsev-Petviashvili equation. This method is a lucid and systematic way. This method can be extended to the other soliton equations.

The work is supported by the National Natural Science Foundation of China (11301183).