March 12th, 2014, 02:18 AM  #1 
Senior Member Joined: Oct 2013 From: Far far away Posts: 431 Thanks: 18  Pythagorean triples
Prove or disprove: If x^2 + y^2 = z^2 for whole numbers x, y and z then, either x or y is a multiple of 3 My attempt: Assume that x and y are not divisible by 3 So x and y must be of the form 3m + 1 or 3m + 2 (leaving a remainder 1 or 2 when divided by 3) We have 3 possibilities: 1) x = 3a + 1 and y = 3b + 1 SO has remainder = 2 2) x = 3a + 1 and y = 3b + 2 SO has a remainder 2 3) x = 3a + 2 and y = 3b + 2 [latex](3a+2)^{2}+(3b+2)^{2}\,=\,9a^2+12a+4+9b^2+12b+4\,= \,3(3a^2+3b^2+4a+4b+2)+2[latex] SO has a remainder of 2 Now consider z. It can be only one of the following forms 3n, 3n + 1 and 3n + 2 1) z = 3n SO has a remainder 0 2) z = 3n + 1 SO has a remainder 1 3) z = 3n + 2 SO has a remainder 1 If x and y are not multiples of 3 then the remainder of (x^2 + y^2)/3 is 2 BUT the remainder of (x^2+y^2)/3 cannot be 2 (it has to be 0 or 1). Therefore at least one of x or y has to be a multiple of 3. Is this proof acceptable? Is there a better proof? 
March 12th, 2014, 08:57 AM  #2 
Math Team Joined: Dec 2013 From: Colombia Posts: 7,690 Thanks: 2669 Math Focus: Mainly analysis and algebra  Re: Pythagorean triples
A = There exist integers x, y, z such that x^2 + y^2 = z^2 B = Either x or y is a multiple of three You have proved not(B) => not(A), which is equivalent to A => B. So yes, that would be a valid proof. Of course, the proof has nothing to say about the existence of such triplets, but neither do the conditions of the hypothesis. 
March 12th, 2014, 09:31 AM  #3 
Global Moderator Joined: Dec 2006 Posts: 21,035 Thanks: 2271  Re: Pythagorean triples
Is the notation used below acceptable? It would allow a more succinct presentation. As 2² ? 1 (mod 3), n ? 2 (mod 3) implies n² ? 1 (mod 3). 
March 12th, 2014, 11:13 AM  #4 
Math Team Joined: Jul 2011 From: North America, 42nd parallel Posts: 3,372 Thanks: 234  Re: Pythagorean triples
Just as a note , you can reduce your work on the proof by 2/3 if you consider Primitive Pythagorean Triples. For Primitive Pythagorean Triples:By elementary arguments , 1) gcd(x,y) = 1 2) x and y cannot both be odd It follows that one of the two legs must be even , the other leg must be odd and the hypotenuse must be odd. So you only need to consider (WLOG) the case x = 3a + 1 , y = 3b + 2 Here are some fun facts about PT's and PPT's. http://en.m.wikipedia.org/wiki/Pythagorean_triple 
March 12th, 2014, 02:51 PM  #5 
Global Moderator Joined: May 2007 Posts: 6,835 Thanks: 733  Re: Pythagorean triples
You could use the following which applies to all P triples. m and n integers, m > n x = m²  n², y = 2mn (z = m² + n²) If m or n is a multiple of 3, then y is. If m?n (mod 3), then mn is a multiple of 3, therefore x is. Otherwise m+n is a multiple of 3, and x is. 

Tags 
pythagorean, triples 
Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Infinitely many pythagorean triples  shunya  Geometry  3  March 20th, 2014 07:12 AM 
A property of Pythagorean triples  elim  Number Theory  8  August 21st, 2011 11:29 AM 
How many Pythagorean triples?  Pell's fish  Number Theory  3  August 5th, 2011 10:38 AM 
Pythagorean Triples  julian21  Number Theory  2  November 13th, 2010 12:01 PM 
Pythagorean Triples II  julian21  Number Theory  3  November 12th, 2010 11:58 AM 