Math Circle Session 8: The Circle Game & Puzzles

Today was the 8th session of our math circle for primary school children. We only had four of the original six children today, for a variety of reasons, but it seemed to work well.

We first ran through a puzzle we set the kids to think about over the last week. The idea of the puzzle, taken from Martin Gardner’s book, is as follows. Take a piece of paper shown below. Imagine this is a map, that you have to fold up (I find folding up maps really hard!)

Screen Shot 2015-03-17 at 17.10.16You have to fold only along the lines shown, and must form a folded “map” with the 1 on top, facing up. The next number down after the 1 must be the 2, after the 2 must be the 3, and so on until the 8 lies on the bottom.

Some kids had tried this at home, some had not (as a policy, we never make “homework” compulsory) but none had found the solution, so we covered this today.

We then moved onto the main part of the session, “The Circle Game”, adapted from Rozhkovskaya’s book. The kids were each given a worksheet consisting of 16 points equally spaced around a circle, and asked to play a game in pairs. The rules of the game are as follows:

  • take it in turns
  • on each turn, connect one point to another point with a straight line; any pair of points can be chosen, but no line drawn can cross another line already drawn

The last player to be able to draw a line is the winner.

The kids enthusiastically played this game, trying to outsmart each other. However, at the end of play, we had three games in which the first player to move had always won (though the points selected were different). I also asked the kids to count the number of moves made, which was found to be 29 in each game. The question was posed: “why does the first player always win?”

One child quickly pointed out that if the number of moves is an odd number, it must always be the first player who wins. Another child suggested that we should try 17 points around the circle as then “it would be an even number of moves and Player 2 would win”. I suggested that instead of making the game more complicated, we simplify it to the smallest possible number of points on the circle, which the children identified as two points.

Children very quickly found that two points would always result in one move, three points in three moves, and four points in five moves. At this point the child who suggested that 17 points would result in a win for Player 2 withdrew his opinion, and all children were convinced that the number of moves would always be odd and therefore Player 1 would always win.

We then looked at generalising the number of moves required for n points. Children very quickly noticed that the number of moves increased by two for each new point added. Due to the significant difference in ages in the group, I wasn’t sure how comfortable children would be with an algebraic generalisation of the result, but it turns out that even the youngest was comfortable with the formula moves = 2 \times points - 3, which they got to via my suggestion that if the numbers were going up by two each time, try looking at the relationship between moves and 2 \times points.

I was surprised by how quickly we had reached this point in the session, so I rounded off the exploration by demonstrating an inductive proof of this formula in graphical form on the whiteboard (see, for example, Theorem 1.8 at

One child was keen to play this game with his dad, and it was suggested that the “youngest goes first” rule should be applied, possibly after agreeing a prize for winning!

With the remaining spare time for the session, we set the following puzzle – also from Gardner’s book – to begin during the session and to continue to think about over the week. Consider the grid below.

Screen Shot 2015-03-17 at 17.35.27The puzzle is to fill each empty cell with a single digit, so that the bottom row forms a number between 0 and 10^{10}-1. The challenge is that the digit under the heading zero must equal the total number of zero digits in this number, the digit under the 1 must equal the total number of 1 digits in this number, and so on. I thought it might be hard for the kids to understand the puzzle, but actually they understood it quite easily. They were amazed when I told them that the answer is unique: there is only one of the 10 billion possible numbers that works! It was interesting to observe the different reactions to this: some kids decided they’d never be able to find the solution, and therefore decided not to put much effort into the task. Others decided this was a challenge, and kept plugging away. One child was determined to take it home so that his dad would help.

Overall, I think this session went very well.

Update: I received an email tonight from the school headteacher with the solution to this number puzzle. I love that he’s so into it!

Math Circle Session 7: Combinatorics, Pascal’s Triangle, and the Sierpinski Triangle Again

Today saw Session 7 of our math circle. No blog on Session 6, as I was away for work – my wife explored The Towers of Hanoi in my absence.

Today’s session was based on a combination of two sessions in Rozhkovskaya’s book. The idea is as follows. Imagine a maze at a tourist attraction. The maze is such that there is one entrance, but seven exits labelled 0 to 6. Imagine that you are standing facing north at the entrance. At this entrance, and at each intersection, you can move ahead-left on a bearing of 315 degrees by one unit or ahead-right on a bearing of 45 degrees by one unit until you meet the next intersection or exit. Thus the maze decomposes into 7 levels (including the entrance and exits). Now further imagine that you must roll a die at each decision point, taking the left route if you roll even and the right route if you roll odd.

The first question: If I were an ice-cream vendor, where should I place my van to capture most people coming out? Children had an intuitive feeling that most people would come out in the middle. Only one child (the oldest) offered a reasoned explanation for this: “we expect as many odd throws as even throws, which means the most likely exit would be the middle one”.

We got the children to experiment by rolling routes through the maze, and collected a tally of how many trials ended up at each exit, confirming this suspicion. Children naturally noticed that exits zero and seven were unlikely, as they would require rolling “all odd” or “all even”.

This naturally led onto the next part of the exercise, to annotate each decision point with the number of ways in which it could be reached. Children found this hard, and made many mistakes, because they were trying to count all the ways to reach a given decision point from the maze entrance from scratch for each decision point. I explained that there are only at most two ways to reach any decision point, so it is sufficient to sum the possibilities found to these two preceding points.

Thus, Pascal’s triangle was revealed. We asked the kids to sum the rows of Pascal’s triangle, and they discovered they summed to successive powers of two.

Finally, we had the kids colour all the even numbers in a large version of Pascal’s triangle (pre-printed). One of the kids immediately recognised a fractal-like shape. Once complete, we pointed out the resemblance to Sierpinski’s triangle, created in Session 7. Oddly (to me) this wasn’t immediately apparent, even to the kids who were there.

Overall, I was a little disappointed with today’s session. I think the material was good, but it was a sunny day, and two of the kids showed no sign of wanting to be there, inside after school. This meant I had to spend a lot more time than I ever imagined trying to get these kids on task, meaning a lot less time to discuss the mathematics of what we were seeing, which was a shame. My one rule for Math Circle was that the kids should want to be there and, for whatever reason, I don’t think this was true for 1/3 of them this time. I hope it’s a temporary blip, but I need to ask for views from primary school teachers; I was again reminded today how different kids are to the adults I’m used to!

Math Circle Session 5: Powers, Limits, and Fractals

Yesterday saw the next installment of our math circle with primary school children. Again, following but extending Rozhkovskaya’s book, we used fractals as an interesting way in to consolidate (and in the case of the younger pupils, introduce) powers of numbers, to explore the children’s intuition about infinite series, and to make some pretty pictures!

We started by producing a fractal tree. The rule is simple: after Year 1, the trunk has grown. At the end of Year 2, two branches have grown from that, after Year 3, two further branches from each existing branch, and so on. Children were quickly able to tell us that the number of branches doubled each year, some took great pleasure in calculating or reciting powers of two. The template for this exercise was taken from Rozhkovskaya’s book, and is quite clever: lines are drawn across the page heights of 8 squares, 12 squares, 14 squares, 15 squares, and 15.5 squares, and each successive year’s branches should be drawn to reach the corresponding height. A bird is drawn flying around 18 squares up. The question is posed: will the tree ever reach the bird?

The children had mixed views on whether the tree will reach the bird. The most common view was that it must. One of the older children was able to provide a line of reasoning: “Each time, the height of the tree increases. Since it is always increasing, even by small amounts, it must eventually reach any height.” We then looked at how much gap there is between the height of the tree and a line of height 16 as the years progress: 16 at the beginning, 8 at the end of Year 1, 4 at the end of Year 2, and so on. Children could see that there was always a positive gap between the height of the tree and the height 16, for any finite number of years. However, this clashed with several children’s intuition, and it took quite a lot of discussion before it was generally accepted that an infinite series can sum to a finite limit. One of the youngest children in the group asked some very probing questions, such as “is infinity a number”, which led to some useful side discussions. Random banter between children later in the session about “hacking” their siblings’ passwords led onto a consolidation of this discussion by asking the question “If you had to press an infinite number of keys on a keyboard, could you do it in a finite amount of time? No? What if the first key took 8 seconds to press, the next 4, the next 2, and so on…”IMG_3794

We then moved onto a practical activity of constructing a sequence of approximations to the Sierpinski triangle by sticking little white triangles on coloured paper (pictured). One of the children noticed quickly, and was able to clearly explain, that the number of white triangles was increasing in powers of three. At this point we ran out of time, and our session ended. I intend to pick up and use the Sierpinski triangle again soon in math circle, through Pascal’s triangle modulo 2.

Math Circle Session 4: Ciphers

Today saw the fourth session of our math circle. Attendance was down, and there was a sombre mood, due to a tragic event to hit the the school community today.

We went ahead with the three children who came, looking at ciphers. Following the guidance and resources in Rozhkovskaya’s book, we looked at the following activities:

  • We started with some plaintext words and one ciphertext word, and the kids very quickly spotted, without any help, which plaintext word corresponded to the ciphertext word, on the basis that both had a double letter in a certain place and they had the same number of symbols.
  • We showed how a coding wheel can be used to produce a Caesar shift cipher, and got them to actually build such a wheel.
  • We queried how many possible ciphers could be constructed with such a wheel. Interestingly, this was not obvious to the children at first, and a common answer was “infinite”. On questioning, this is because they were considering all integer shifts, i.e. mapping the letter n to the letter n+k for some key k, and forgetting the modulo arithmetic in play, despite having the wheel in front of them.
  • We got them to produce coded messages to each other using the wheel, which they seemed to enjoy. Our initial plan was to have them pass the message and the key to the next child to decode. However, they immediately attempted to decode without they key by looking for patterns, so we didn’t stop this. One key pattern they spotted included the limited number of two letter words in English, allowing them to enumerate possible keys on two-letter words in the ciphertext. Another was the limited number of letters in the alphabet that can appear twice consecutively in an English word, again allowing for enumeration.

There’s a lot more that can be done with ciphers, and I expect we’ll continue this for another session.

Math Circle Session 3: Knots

Today saw the third session of the primary school math circle we’ve been running for kids in Year 2 to Year 6. (I was absent for Session 2, where my wife covered mathematical card tricks.)

Today we looked at knots, drawing inspiration from Rozhkovskaya’s book as well as Adams’ introduction to knot theory. In particular, we covered the following points:

  • (Medial) graph representations of knot projections, moving backward and forward between them (see
  • The importance of assigning an (under/over) decision to crossings
  • Getting the kids to think of their own ideas for what knot equivalence might mean (shape, size, rotation, deformation – of what type) etc.

The key innovation we used here, which I think really brought the session to life, was to get the kids actually physically making the knots using Wikkistix, wax-coated string which allowed them to make, break, and remake their knots.

The kids really ran with this, and made their own discoveries, in particular:

  • One child discovered that some knots corresponding to the graph K_3 could be manipulated to produce the unknot, some could not.
  • A child discovered that it is possible to produce two interconnected knots, forming a link. Another child came to the same conclusion from the graph representation.
  • Consider the graph with vertex set \{v_1,\ldots,v_n\} and edge set \{\{v_i,v_{i+1}\} | 1 \leq i \leq n-1\}\} \cup \{\{v_n,v_1\}\} (is there a name for these graphs?). One child completely independently found that for n even, this corresponds to a link of two knots, whereas for n odd, it corresponds to a single knot.

I certainly had a lot of fun this afternoon – I hope they did too!

Starting a Math Circle

I’ve been intrigued for a couple of years by the idea of math circles. Over Christmas I finally plucked up the courage to start one at a local primary school with my wife, a maths teacher. The school was happy, and recruited six pupils from Year 2 to Year 6 to attend.

Armed with a number of publications from the American Mathematical Society’s Mathematical Circles Library, our first task was to find a suitable topic for Session 1. Our primary goal was to find a topic that was clearly not everyday school maths, preferably as far away from the National Curriculum as possible, and ideally including practical activities.

In the end, we decided to look at Möbius strips. In the traditional of journal keeping for Mathematical Circles, I thought I would report the experience here in case anyone else wants to give it a go. In particular, we took the following approach:

  1. Make zero half-turn bands (loops) and colour the outside in one colour.  Then colour the inside in a different colour.
  2. Repeat with a one half-turn band. This caused some surprise when it became apparent that one colour coloured the whole band.
  3. Predict what would happen if you cut the zero half-turn band down the centre (prediction was universally that you’d get two zero half-turn bands). Try it.
  4. Now predict for the one half-turn band. Children were less sure about this case, but the most popular prediction was two half-turn bands. More surprise when it turned out to create a single four half-turn band. One child then went off on his own exploring what happened when you cut one of these four half-turn bands (two four half-turn bands).

By this time some of children were already off under their own steam, trying out their own experiments. This was great, but even with only six children and with two adults, I found it hard to pull together the outcomes of these experiments in any systematic way in real time.

Eventually we discovered together that:

  • If the initial number of half-turns is odd, cutting gives you one larger band with more half turns. (I was hoping we’d be able to quantify this, but it turns out to be very difficult to count the number of half-turns in a strip – even for me, let alone for the younger children!)
  • If the initial number of half-turns is even, cutting gives you two interlinked bands with the same number of half-turns.

This took up pretty much the whole 50mins we had for Session 1, though I did briefly try to show them an explanation of why this might be the case, following the diagrammatic representation in this document from the University of Surrey. I probably didn’t leave enough time to do this properly, and they were anyway keen on cutting and exploring by that time, so with hindsight I probably should have just left them to it.

What delighted me was the child who wanted to take home his Möbius strip to show his dad. So, not a bad start to the Math Circle. Let’s see how we get on!