Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
NCTM or The Math Forum.



Re: chessboard rectangles
Posted:
Oct 11, 1998 2:13 PM


Sarah,
I agree with your total (1296). With a little rearranging your method will produce the general formula for the number of rectangles on an n by n chessboard.
Cont the rectangles by rows.
In row 1 there are n 1 by 1s , n1 1 by 2s, ... 2 1 by n1s and 1 1 by n:
Row Total = n + (n1) + (n2) + ... + 3 + 2 + 1 = n(n+1)/2 rectangles.
But of course there are n rows giving n ( row sum )
Now count all rectangels of height 2 . Start in the bottom row.
There are n 2 by 1s and n1 2 by 2s and ... and 1 2 by n for a
Row Total = n + (n1) + (n2) + ... + 3 + 2 + 1 = n (n+1)/2
The row total is the same, as it will be for all the rectangles of height 3, 4, ... n because they all share the same bases at the bottom of the board. However, there are only n1 rows of height 2 and n2 rows of height 3 etc. Thus,
# 1 by any totals: n [n(n+1)/2] # 2 by any totals: (n1)[n(n+1)/2] # 3 by any totals: (n2) [n(n+1)/2]
...
#n by any totals: 1 [n(n+1)/2]
So the total number of rectangles is [n(n+1)/2](n + n1) + ... + 3+ 2+1)
or Total = [n(n+1)/2]^2
which, as you mentioned, is the sum of the cubes from 1 to n.
Alan Lipp alipp@crocker.com  On Sun, 11 Oct 1998, Sarah Seastone wrote: > > Someone asked the number of rectangles on a chessboard. The answer is > "Many, in fact 1296." > ...
> Have I missed any? Or added wrong? What if we reduce or augment the > size of the chessboard? Can we find a general formula? What is it? How > do we use induction to show that it is indeed general?



