The Aperiodical This blog is part of The Aperiodical by Peter Rowlett, Katie Steckles and Christian Perfect. Find Features, News, Videos and other Columns, all aimed at a mathematical audience. If you like this blog, you should check out the rest.

I also write and edit blog posts at Second-Rate Minds for ACMEScience.com.

Math/Maths PodcastMath/Maths Podcast: Peter Rowlett in the UK talks to Samuel Hansen in the US about news & current affairs.

railway display boardTravels in a Mathematical World Podcast: Mathematicians speaking about their work.

Institute of Mathematics and its Applications (IMA) Find out about the Institute of Mathematics and its Applications (IMA).
I guest blog over at IMA maths blogger.

British Society for the History of Mathematics (BSHM) I am a member of the British Society for the History of Mathematics (BSHM).

Jon's coloured balls solution

Read the puzzle and have a go before reading the solution.

First some notation. Let the six balls be r1, r2, w1, w2, b1, b2.

Denote a weighing by: LHS /\ RHS
where /\ is the pivot.

Denote the weight of ball r1 by m(r1) and the others similarly.

Weigh: r1, w1 /\ w2, b1

What you do next depends on what happens.

1. If the scales balance, then you know m(r1) = m(w2) and m(w1) = m(b2) basically because the two white balls must be different weights. Swap b1 for b2 and see which way the scales tip. This will tell you which of the b's is heavy and the rest follows.

2. If the scales don't balance:

Say they tip to the left; if they tip to the right that is a mirror of what I describe below.

Then you know w1 must be heavier than w2, that is m(w1) > m(w2) and r1 can't be light if b1 is heavy (or they would balance like case 1).
Set them up again as: r1, b2 /\ w1, b1

Then you have to think carefully through each case:

a. If the scales balance, you know: m(r2) > m(r1), m(b2) > m(b1) and m(w1) > m(w2).

b. If the scales tip left, you know: m(r1) > m(r2), m(b2) > m(b1) and m(w1) > m(w2).

c. If the scales tip right, you know: m(r1) > m(r2), m(b1) > m(b2) and m(w1) > m(w2).

0 comments: