Page 331 - DMTH402_COMPLEX_ANALYSIS_AND_DIFFERENTIAL_GEOMETRY
P. 331

Complex Analysis and Differential Geometry




                    Notes          We also have that detB = detD. Thus taking into the account that (7) holds for D:

                                                                                  
                                                                                 
                                                                        
                                                              
                                                      logdetB  logdetD     tr D D   tr B B . 
                                                                               1
                                                                                        1
                                                                               
                                                                                       
                                   In order to prove the general case, it is more convenient to look at the equivalent identity:
                                                                    
                                                                    
                                                                             
                                                                                1
                                                                               
                                                               detB  tr    detB B B .                             ...(8)
                                   Note that  by Kramer’s  rule, the  matrix (detB)B   is the  matrix  of co-factors  of  B, hence  its
                                                                          –1
                                   components being determinants of minors of B, are multivariate polynomials in the components
                                   of B. Thus, both sides of the identity (8) are linear polynomials
                                                                n                 n
                                                         p B ;B      pij(B)b ,  ij   q B ;B     qij(B)b ,  ij
                                                               i,j 1             i,j 1
                                                                
                                                                                  
                                   in the components  b   of B, whose  coefficients p (B) and q (B)  are themselves  multivariate
                                                                                    ij
                                                    ij
                                                                            ij
                                   polynomials in the components b  of B. Since the set of matrices with distinct eigenvalues is an
                                                             ij
                                                                               
                                                                                      
                                               
                                   open set  U  S n n ,  we have already proved that   p B ;B    q B ;B  holds for all values of B, and
                                                                            
                                                                                    
                                              
                                                                           
                                                                                  
                                                                        
                                   all B  U. For each such B  U the equality   p B ;B    q B ;B  for all B implies that p (B) = q (B)
                                                                                
                                                                                                      ij
                                                                                                            ij
                                   for i, j = 1, …, n. Since this holds for all B in an open set, we conclude that p  = q , and hence
                                                                                                     ij
                                                                                                 ij
                                   p = q.
                                   We remark  that the  more general  identity (8)  in fact holds, as easily shown,  for all  square
                                   matrices B. An immediate consequence of the proposition is that:
                                                                      
                                                                             1
                                                                detB    1 tr B B  detB,                    ...(9)
                                                                             
                                                                        2
                                   for any continuously differentiable family of symmetric positive definite matrices B. We are
                                   now ready to prove the proposition.
                                   Proof of Proposition 6. Differentiating the area (4) under the integral sign, and using (9), we get:
                                                   dA (V)    1  g ij dg ij  det g  1  2  1  g ij dg ij dA.
                                                      F
                                                     dt    2  V  dt     du du   2  V  dt
                                                                         ij
                                   Since Y is smooth, we have at t = 0 that dF /dt = (dF/dt)  = Y , and thus
                                                                                   i
                                                                               i
                                                                    i
                                                            dg
                                                         g ij  ij   g ij  Y X   X X  2g X Y .
                                                                                  ij
                                                                           
                                                                                     
                                                                               j
                                                            dt      i  j  i         i  j
                                   This completes the proof of the proposition.
                                   Since the variation of the area dA (V)/dt is a linear functional in the generator dF/dt of the
                                                               F
                                   variation, it is possible to decompose any variation into tangential and normal components. We
                                   begin by showing that the area doesn’t change under a tangential variation. This is simply the
                                   infinitesimal version of Proposition (5).
                                   Proposition 7. Let  X : U    be a parametric surface, and let F(u; t) be a compactly supported
                                                          3
                                   tangential variation. If V  U is open with  V  compact in U, and the support of F contained in V,
                                   then dA (V)/dt = 0.
                                         F
          324                               LOVELY PROFESSIONAL UNIVERSITY
   326   327   328   329   330   331   332   333   334   335   336