Learning number theory with Gauß/Congruent numbers

Definition and elementary theorems edit

Definition 1.1:

Let   and  .   and   are said to be congruent to each other modulo   if and only if  . In this case, we write

 .

Theorem 1.2:

Let   and  . Then the numbers congruent to   modulo   are exactly the numbers

 .

Proof:

First we note that   for each  . Then let  , i. e.   for an  . Then  . 

Corollary 1.3:

The relation   is an equivalence relation.

Proof 1:

We prove the corollary from theorem 1.2.

Reflexivity: Setting  , we note that  .

Symmetry: Let  , i.e.  . Then  , where  .

Transitivity: Let   and  , i.e.   and  . Then  . 

Proof 2:

We prove the corollary directly.

Reflexivity:  .

Symmetry:  .

Transitivity:   and   implies  ,   for some  . Hence,  . 

Definition 1.4:

Let  ,  . Then the equivalence class of   with respect to the equivalence relation from the last theorem is called congruency class or residue class of   modulo  .

Theorem 1.5 (Euclid; division with remainder):

Let  ,  . Then there exist   and   such that   and

 .

Proof:

Consider the set

 .

Since this set by definition is contained in  , it has a least element. Call this element  . By definition, we obtain the existence of   such that

 .

We define  . Assume  . Then  , contradicting the minimality of  . 

We note at this point that rings   for which the above theorem holds with   instead of   and   (and a couple of more modifications, see the Wikipedia article) are called Euclidean rings.

Theorem 1.6:

There are exactly   different congruency classes modulo  , where  .

Proof 1:

We prove the theorem from division with remainder.

We claim that   define pairwise different congruency classes. Indeed, assume that two of those elements defined the same congruency class, call them   and  . Then  , and hence in particular   or  . The latter is a contradiction since we may make   as large as possible and   as small as possible.

Let now  . By division with remainder, we obtain  . Then  , where   is in the range  .

In summary, we obtain that each element   represents a distinct congruency class, and further that each other number is contained in one of the congruency classes represented by these elements. Since they are   many, the theorem follows. 

Proof 2:

We prove the theorem from Corollary 1.7 (this is not nonsensical because we give two proofs of this corollary which are not based upon this lemma).

First of all, we note that   define pairwise different congruency classes, since if  , by corollary 1.7 there exists exactly one number among   congruent to   (  itself).

Then we note that each congruence class is represented by at least one  , which in turn by corollary 1.7 is congruent to one of the elements  . Hence, each congruence class is represented by an element in  . Since these are exactly   different elements, the theorem follows.

Corollary 1.7:

Let  ,   and let   successive integers   be given. If   is any other integer, then exactly one of the integers   is congruent to   modulo  .

Proof 1:

We prove the corollary using theorem 1.2.

Indeed, by theorem 1.2 the integers congruent to   modulo   are given exactly by  . Assume none of those integers were contained within  . Define   to be the largest integer such that   and   to be the smallest integer such that  . By assumption  , since otherwise   would be larger than the largest integer   such that   and smaller than the smallest integer   such that   and hence   contrary to our assumption. But the difference between   and   is exactly  , and in particular there are   numbers enclosed between the two. This contradicts the assumption that the numbers   are enclosed within the two; those are   different numbers.

Assume two of the integers   were congruent to   modulo  . Call them   and  . Then  ,   for some   by theorem 1.2. Hence,  . In particular, the difference between   and   is either zero or greater or equal to   in contradiction to the fact that the difference between any to numbers within   is less or equal to   (for, we may maximise the difference by moving the lower element to   and the higher element to  ). 

Proof 2:

We prove the corollary using theorem 1.2 and fractions.

If   is an integer, then  . Otherwise,   is a fraction. Let   be the next larger integer. Then

 

and hence   is contained within  . Since furthermore for all  

 ,

only one of the fractions   can be an integer, which is equivalent to  . 

Proof 3:

We prove the corollary from the notion of congruence classes and lemma 1.6.

Indeed, in complete analogy to lemma 1.6, we prove that   represent pairwise distinct congruence classes. Since there are exactly   many of them, the elements   indeed represent all of the congruency classes. Since the equivalence classes of any equivalence relation form a partition, we conclude that each   is congruent to one of  . 

Residues edit

Definition 1.7:

Let   and  .   is called a residue of   modulo   if and only if  .

From theorem 1.2 now directly follows that the residues modulo   of   are exactly the numbers

 .

Lemma 1.8

Let   and  . Then   has exactly one residue within   and exactly one residue within  . They coincide if and only if  .

Proof:

This follows from corollary 1.6 with   or   respectively. 

Definition 1.9:

The two residues from lemma 1.8 are called negative least residue and positive least residue respectively.

Theorem 1.10:

Let  ,  ,   be the negative least residue and   be the positive least residue. Then one of four cases occurs:

  1.  
  2.  
  3.  
  4.  

Proof:

We have   and hence  , i.e.  . We separate four cases:

  1.  . Then since  , zero is the positive least residue and the negative least residue at once.

Theorem and definition 1.11:

Let   and  . There exists exactly one residue in the range

The rings of residue classes edit

Divisibility criteria edit

Theorem 1.?:

Let   be given by the decimal expansion  , i. e.

 .

Then   is divisible by three or nine if and only if   is.

Proof:

Since   (and hence  ), we have

 ,  . 

Exercises edit

  • Exercise 1.?.1: