The lemma Bézout  (after Etienne Bézout  1730-1783) () in the number theory  says that the greatest common divisor  of two integers and as a linear combination  of and represent with integer coefficients can.
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
  
Statement and formal presentation Expressed formally, the following applies:
  
    
      
        ∀ 
        a 
        , 
        b 
        ∈ 
        
          Z 
         
          
        ∃ 
        s 
        , 
        t 
        ∈ 
        
          Z 
         
        : 
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
        = 
        s 
        ⋅ 
        a 
        + 
        t 
        ⋅ 
        b 
       
     
    {\ displaystyle \ forall a, b \ in \ mathbb {Z} \ \ exists s, t \ in \ mathbb {Z}: \ operatorname {ggT} (a, b) = s \ cdot a + t \ cdot b} 
   
 Are and coprime  , then exist  such  that
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        s 
        , 
        t 
        ∈ 
        
          Z 
         
       
     
    {\ displaystyle s, t \ in \ mathbb {Z}} 
   
  
  
    
      
        1 
        = 
        s 
        ⋅ 
        a 
        + 
        t 
        ⋅ 
        b 
       
     
    {\ displaystyle 1 = s \ cdot a + t \ cdot b} 
   
 applies.
comment 
A kind of reversal also applies here: if there is with , then there is .
  
    
      
        s 
        , 
        t 
        ∈ 
        
          Z 
         
       
     
    {\ displaystyle s, t \ in \ mathbb {Z}} 
   
 
  
    
      
        s 
        a 
        + 
        t 
        b 
        = 
        1 
       
     
    {\ displaystyle sa + tb = 1} 
   
 
  
    
      
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
        = 
        1 
       
     
    {\ displaystyle \ operatorname {ggT} (a, b) = 1} 
   
  
 
The coefficients and can be calculated efficiently using the extended Euclidean algorithm  .
  
    
      
        s 
       
     
    {\ displaystyle s} 
   
 
  
    
      
        t 
       
     
    {\ displaystyle t} 
   
  
The lemma can be generalized to more than two whole numbers: If there are whole numbers, then there are integral coefficients with
  
    
      
        
          a 
          
            1 
           
         
        , 
        ... 
        , 
        
          a 
          
            n 
           
         
       
     
    {\ displaystyle a_ {1}, \ dotsc, a_ {n}} 
   
 
  
    
      
        
          s 
          
            1 
           
         
        , 
        ... 
        , 
        
          s 
          
            n 
           
         
       
     
    {\ displaystyle s_ {1}, \ dotsc, s_ {n}} 
   
 
  
    
      
        
          s 
          
            1 
           
         
        
          a 
          
            1 
           
         
        + 
        ⋯ 
        + 
        
          s 
          
            n 
           
         
        
          a 
          
            n 
           
         
        = 
        gcd 
         
        ( 
        
          a 
          
            1 
           
         
        , 
        ... 
        , 
        
          a 
          
            n 
           
         
        ) 
       
     
    {\ displaystyle s_ {1} a_ {1} + \ dotsb + s_ {n} a_ {n} = \ operatorname {ggT} (a_ {1}, \ dotsc, a_ {n})} 
   
  More generally, Bézout's lemma holds true in every principal ideal ring  , even in a non-commutative one  ; for the exact statements see there.
The question of which numbers can even be represented as coefficients with natural numbers is the subject of the coin problem  .
proof The proof of the lemma is based on the possibility of division with remainder  . Thus it can easily be transferred to
 Euclidean rings  .
For and can be set, so we assume, without loss of generality  , that holds. Among all the numbers with there are certainly also those that are positive and are. Be the smallest number among these. Since both and shares, shares also .
  
    
      
        a 
        = 
        0 
       
     
    {\ displaystyle a = 0} 
   
 
  
    
      
        s 
        = 
        0 
       
     
    {\ displaystyle s = 0} 
   
 
  
    
      
        t 
        = 
        ± 
        1 
       
     
    {\ displaystyle t = \ pm 1} 
   
 
  
    
      
        a 
        ≠ 
        0 
       
     
    {\ displaystyle a \ neq 0} 
   
 
  
    
      
        x 
        = 
        s 
        ⋅ 
        a 
        + 
        t 
        ⋅ 
        b 
       
     
    {\ displaystyle x = s \ cdot a + t \ cdot b} 
   
 
  
    
      
        s 
        , 
        t 
        ∈ 
        
          Z 
         
       
     
    {\ displaystyle s, t \ in \ mathbb {Z}} 
   
 
  
    
      
        ≠ 
        0 
       
     
    {\ displaystyle \ neq 0} 
   
 
  
    
      
        d 
        = 
        s 
        ⋅ 
        a 
        + 
        t 
        ⋅ 
        b 
       
     
    {\ displaystyle d = s \ cdot a + t \ cdot b} 
   
 
  
    
      
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
       
     
    {\ displaystyle \ operatorname {ggT} (a, b)} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
       
     
    {\ displaystyle \ operatorname {ggT} (a, b)} 
   
 
  
    
      
        d 
       
     
    {\ displaystyle d} 
   
  
We now show that and is also a factor . Division by remainder gives us a representation of the form , where . If you insert for the representation and solve the equation for , you get . Because of the minimality of must be, so is a divisor of . Correspondingly, it also holds that is a divisor of , and thus holds . We had already seen before that is a divisor of . So it applies .
  
    
      
        d 
       
     
    {\ displaystyle d} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        q 
        ⋅ 
        d 
        + 
        r 
       
     
    {\ displaystyle q \ cdot d + r} 
   
 
  
    
      
        0 
        ≤ 
        r 
        < 
        d 
       
     
    {\ displaystyle 0 \ leq r <d} 
   
 
  
    
      
        d 
       
     
    {\ displaystyle d} 
   
 
  
    
      
        s 
        ⋅ 
        a 
        + 
        t 
        ⋅ 
        b 
       
     
    {\ displaystyle s \ cdot a + t \ cdot b} 
   
 
  
    
      
        r 
       
     
    {\ displaystyle r} 
   
 
  
    
      
        r 
        = 
        ( 
        1 
        - 
        q 
        ⋅ 
        s 
        ) 
        ⋅ 
        a 
        + 
        ( 
        - 
        q 
        ⋅ 
        t 
        ) 
        ⋅ 
        b 
       
     
    {\ displaystyle r = (1-q \ cdot s) \ cdot a + (- q \ cdot t) \ cdot b} 
   
 
  
    
      
        d 
       
     
    {\ displaystyle d} 
   
 
  
    
      
        r 
        = 
        0 
       
     
    {\ displaystyle r = 0} 
   
 
  
    
      
        d 
       
     
    {\ displaystyle d} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        d 
       
     
    {\ displaystyle d} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        d 
        ≤ 
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
       
     
    {\ displaystyle d \ leq \ operatorname {ggT} (a, b)} 
   
 
  
    
      
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
       
     
    {\ displaystyle \ operatorname {ggT} (a, b)} 
   
 
  
    
      
        d 
       
     
    {\ displaystyle d} 
   
 
  
    
      
        d 
        = 
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
       
     
    {\ displaystyle d = \ operatorname {ggT} (a, b)} 
   
 
Main ideals If one uses the concept of the ideal  from ring theory  , it is fundamentally true that the main ideals   and are contained in the main ideal . So also the ideal is to contain. One can also formulate Bézout's lemma in such a way that applies to the ring (or generally to Euclidean rings)
  
    
      
        a 
        R. 
       
     
    {\ displaystyle aR} 
   
 
  
    
      
        b 
        R. 
       
     
    {\ displaystyle bR} 
   
 
  
    
      
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
        R. 
       
     
    {\ displaystyle \ operatorname {ggT} (a, b) \; R} 
   
 
  
    
      
        a 
        R. 
        + 
        b 
        R. 
       
     
    {\ displaystyle aR + bR} 
   
 
  
    
      
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
        R. 
       
     
    {\ displaystyle \ operatorname {ggT} (a, b) \; R} 
   
 
  
    
      
        R. 
        = 
        
          Z 
         
       
     
    {\ displaystyle R = \ mathbb {Z}} 
   
 
  
    
      
        a 
        R. 
        + 
        b 
        R. 
        = 
        c 
        R. 
       
     
    {\ displaystyle aR + bR = cR} 
   
 
  
    
      
        c 
        = 
        gcd 
         
        ( 
        a 
        , 
        b 
        ) 
       
     
    {\ displaystyle c = \ operatorname {ggT} (a, b)} 
   
  Main ideal rings  are rings in which every ideal is a main ideal. There is always an element for elements and the ring , so that the ideal is the main ideal . is then on the one hand a common divisor of and , and on the other hand a linear combination of and . In main ideal rings, Bézout's lemma applies to a certain extent by definition if the element is regarded as that of and .
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        c 
       
     
    {\ displaystyle c} 
   
 
  
    
      
        a 
        R. 
        + 
        b 
        R. 
       
     
    {\ displaystyle aR + bR} 
   
 
  
    
      
        c 
        R. 
       
     
    {\ displaystyle cR} 
   
 
  
    
      
        c 
       
     
    {\ displaystyle c} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        c 
       
     
    {\ displaystyle c} 
   
 
  
    
      
        gcd 
       
     
    {\ displaystyle \ operatorname {ggT}} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
Inferences Bézout's lemma is of elementary importance for mathematics and especially for number theory. So you can z. B. derive the lemma of Euclid  , which results in the uniqueness of the prime number decomposition  . The Chinese remainder theorem  is a further consequence of the Bézout lemma. For linear Diophantine equations  , Bézout's lemma yields a criterion for their solvability.
literature 
Kurt Meyberg: Algebra - Part 1  . Hanser 1980, ISBN 3-446-13079-9  , p. 43 
Stephen Fletcher Hewson: A Mathematical Bridge: An Intuitive Journey in Higher Mathematics  . World Scientific, 2003, ISBN 978-981-238-555-0  , pp. 111 ff. 
 
Web links References and comments 
↑  Because if and is a common factor , so and , then, is , therefore, a factor of 1. ■
  
    
      
        d 
        ∈ 
        
          Z 
         
       
     
    {\ displaystyle d \ in \ mathbb {Z}} 
   
 
  
    
      
        a 
       
     
    {\ displaystyle a} 
   
 
  
    
      
        b 
       
     
    {\ displaystyle b} 
   
 
  
    
      
        a 
        = 
        
          a 
          
            ′ 
           
         
        d 
       
     
    {\ displaystyle a = a ^ {\ prime} d} 
   
 
  
    
      
        b 
        = 
        
          b 
          
            ′ 
           
         
        d 
       
     
    {\ displaystyle b = b ^ {\ prime} d} 
   
 
  
    
      
        1 
        = 
        s 
        
          a 
          
            ′ 
           
         
        d 
        + 
        t 
        
          b 
          
            ′ 
           
         
        d 
        = 
        ( 
        s 
        
          a 
          
            ′ 
           
         
        + 
        t 
        
          b 
          
            ′ 
           
         
        ) 
        d 
        = 
        1 
       
     
    {\ displaystyle 1 = sa ^ {\ prime} d + tb ^ {\ prime} d = (sa ^ {\ prime} + tb ^ {\ prime}) \, d = 1} 
   
 
  
    
      
        d 
       
     
    {\ displaystyle d} 
   
  
 
 
 
<img src="https://de.wikipedia.org/wiki/Special:CentralAutoLogin/start?type=1x1" alt="" title="" width="1" height="1" style="border: none; position: absolute;">