Services & Resources / Wolfram Forums
MathGroup Archive
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 1993

[Date Index] [Thread Index] [Author Index]

Search the Archive

Graph Theory

  • To: mathgroup at
  • Subject: Graph Theory
  • From: puglisi at
  • Date: Thu, 7 Oct 93 18:00:17 MDT

Milan, 7 Oct. 93

 Hi, I am reading a book on "Graph Theory" by Harary.

I am interested in the "Turan" theorem which state that 
the maximum number of edges a graph with p points can have 
without containing a complete subgraph of n vertices (i.e. Kn)
is given by the formula:  ( page 18)

           (n-2)(p^2 - r^2)
ex(p,Kn) = ---------------- + Binomial[r,2]  with  (n<=p)
where p = r mod(n-1) and (0<= r <n-1)

Now "p" is at most equal to n-2 which seems to contradict the entire formula.

Can anybody help me understand if there is a typo error or something else?

	Thanks Anyway

Alberto Puglisi
Italtel R&D
Milan, Italy		Internet:   puglisi at

  • Prev by Date: Re: Double axes in LogPlot?
  • Next by Date: gl-library on SGI?
  • Previous by thread: Polar coordinates
  • Next by thread: gl-library on SGI?