Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: A historical 28/97 Egyptian fraction series
Replies: 4   Last Post: May 3, 2000 8:26 AM

 Messages: [ Previous | Next ]
 eppstein@euclid.ics.uci.edu Posts: 128 Registered: 12/8/04
Re: A historical 28/97 Egyptian fraction series
Posted: May 2, 2000 2:23 AM

milogardner@juno.com (milo gardner) writes:
> With scanning the book can anyone come up with the unit fractions
> series that only used denominators < 1000?

You won't be happy -- it requires five terms:

28/97 = 1/4 + 1/40 + 1/97 + 1/485 + 1/776

There are other five-term solutions as well:

28/97 = 1/4 + 1/30 + 1/388 + 1/582 + 1/970
28/97 = 1/5 + 1/12 + 1/388 + 1/582 + 1/970

Of course you could get more than five terms, e.g. by expanding the
terms with smaller denominators... Here are the best three and four
term representations (there is no two term representation):

28/97 = 1/4 + 1/36 + 1/97 + 1/1746
28/97 = 1/4 + 1/26 + 1/5044

> Tomorrow I'll post Bunch's series, and ask: how did the Egyptian
> scribe come up with the series.

No idea, but my software found it by trying combinations of multiples of
1/97 until it found one for which the remaining fraction after
subtracting those multiples no longer had a factor of 97 in the
denominator. The detailed algorithm involves modular arithmetic and
dynamic programming, neither of with were likely known in that form to
the scribes...

> Since Bunch suggested that the Egyptian fraction method was awkward
> modern mathematics should have little intellectual difficulty with
> the problem.

Certainly none of these calculations took much computer time.
--
David Eppstein UC Irvine Dept. of Information & Computer Science
eppstein@ics.uci.edu http://www.ics.uci.edu/~eppstein/

Date Subject Author
5/1/00 Milo Gardner
5/2/00 eppstein@euclid.ics.uci.edu
5/2/00 Steve Moss
5/3/00 Milo Gardner
5/3/00 Milo Gardner