Table of Contents
Konigsberg Bridges Puzzle: Kids Meet Graph Theory
The Konigsberg bridges puzzle for kids, taped out on your floor: try to walk all seven bridges once, fail, then prove with counting why it cannot be done.
Most puzzles a child is handed have an answer. This one does not, and that is the point. The Konigsberg bridges puzzle for kids works because the failure is not a failure of cleverness: the walk is provably impossible, and a ten-year-old can follow the proof. You tape four paper islands to the floor, join them with seven masking-tape bridges, and ask your kid to cross every bridge exactly once. They will try for fifteen minutes. Then you count the bridges meeting each island, and the impossibility falls out of four small numbers.
Key Takeaways
- You’ll tape a walkable map of four landmasses and seven bridges on the floor, try to cross each bridge once, then redraw it as dots and lines.
- It teaches graph theory: that a network’s behaviour depends on how many connections meet at each point, not on distances or shapes.
- Best from 9. Younger children enjoy the walking but will not follow the parity argument, which is where the value is.
- The result is a rule they can use: a route crossing every edge once exists only if exactly zero or two points have an odd number of connections.
- Expect genuine frustration around attempt ten. Say in advance that the answer might be “impossible”, or it reads as a failure.
What You’ll Need
| Item | How much | Where to get it | If you don’t have it |
|---|---|---|---|
| Large sheets of paper | 4, A3 or newspaper pages | Printer, or a newspaper | Four cushions, or four chalk circles outdoors |
| Masking tape | 1 roll | Toolbox or craft drawer | Strips of paper; do not use duct tape on wood floors |
| Marker | 1 | Stationery | Any pen |
| Clear floor | About 2 × 2 m | Living room | Do it outside with chalk |
| Plain paper and pencil | 3 sheets | Anywhere | Back of an envelope |
| Small sticky notes | 8 | Desk | Torn paper squares to mark used bridges |
Time: setup 15 min, activity 35 min, cleanup 5 min Cost: a roll of masking tape Mess level: low, but you need the floor
Do the Experiment: The Konigsberg Bridges Puzzle Kids Can Walk
Step 1 – Lay out the four landmasses
Put the four sheets on the floor in a diamond: one in the middle as the island, one above, one below, one to the right. Label them with the marker: A for the island, B above, C below, D to the right. These are the two riverbanks, the island and the strip of land between the river’s two branches, which is the real geography of eighteenth-century Konigsberg.
Step 2 – Tape the seven bridges
Run masking-tape strips: two from A to B, two from A to C, one from A to D, one from B to D, and one from C to D. That is seven. Make each strip at least 40 cm long so a child can actually walk along it, and press the ends down firmly.
Step 3 – Set the rule and let them walk
One rule: cross every bridge exactly once. You can start anywhere, finish anywhere, and visit a landmass as many times as you like. Give them a sticky note to drop on each bridge as they use it, so the record is physical and arguments about “did I do that one” do not happen.
Step 4 – Count the failures out loud
They will get to six bridges and be stranded. Then again. After five or six attempts, ask the question that turns this into mathematics: “Is it always the same bridge left over?” It usually is not, which is a useful surprise. Let them reach attempt ten before you move on.
Step 5 – Redraw it as dots and lines
On paper, draw four dots labelled A, B, C, D and connect them with the same seven lines. Point out what you have thrown away: the size of the islands, the length of the bridges, the shape of the river. None of it mattered. The only thing you kept is which places are joined.
Step 6 – Write the number of bridges at each dot
Count the lines touching each dot. A has 5, B has 3, C has 3, D has 3. Write those numbers on the drawing. Then ask how many are odd. All four.
Step 7 – Explain the two-at-a-time argument, then test it
Every time you walk through a landmass without starting or finishing there, you use two bridges: one to arrive, one to leave. So any middle landmass must have an even number of bridges. Only the place you start and the place you finish can be odd, and you only get two of those. Konigsberg has four. Now test the rule: peel off the bridge from B to D, recount, and you get A=5, B=2, C=3, D=2. Two odd. Walk it starting on A and finishing on C. It works the first time.
Step 8 – Invent three layouts and predict each one
Draw three new dot-and-line maps on paper. Before tracing, have your kid count the odd dots and predict whether a one-pass route exists. Then try. Predicting correctly three times in a row is the moment the rule becomes theirs.
What you should see: The first six or seven walking attempts end with one bridge unreachable, and the leftover bridge changes from attempt to attempt, which makes the impossibility feel arbitrary. It is not. Once the four odd numbers are written down, the argument takes about ninety seconds and the frustration flips into something better. Removing one bridge works immediately: the route from A to C is easy to find, usually on the first try, which is strong confirmation that the rule is doing real work.
What’s Actually Going On
This is the origin of graph theory, and the way it discards information is the whole lesson. A graph is a set of points, called vertices, joined by lines, called edges. Lengths, areas, angles and positions are deleted. What remains is connectivity, and for this problem connectivity is enough.
The quantity that decides everything is degree: how many edges meet at a vertex. A walk that uses every edge exactly once is called an Eulerian path. Here is the argument, complete, in two sentences. Each visit to a vertex that is neither the start nor the end consumes exactly two edges, one in and one out, so every such vertex must have even degree. The start and the end are the only vertices allowed an odd degree, which means a graph can have at most two odd vertices, and if it has zero then the route must begin and end in the same place.
Konigsberg’s graph has degrees 5, 3, 3, 3. Four odd vertices. No Eulerian path exists, and no amount of cleverness produces one. That is a different kind of answer from “I couldn’t find it”, and getting a child to feel the difference is worth an afternoon.
The history has a wrinkle worth knowing, because dates for this get quoted wrongly everywhere. Euler worked the problem out in 1735 and his paper, Solutio problematis ad geometriam situs pertinentis, was published in 1741 in volume 8 of the St Petersburg Academy’s Commentarii. That volume covers the year 1736, which is why the date 1736 is the one you usually see. The Euler Archive, which holds the catalogue of everything he wrote, records it as written 1735 and published 1741.
A second wrinkle matters more mathematically. Euler’s paper proves the hard half: if a graph has more than two odd vertices, no such walk exists. The converse, that having zero or two odd vertices is actually enough in a connected graph, he asserted rather than fully proved, and completing it took later mathematicians most of the following century. Brian Hopkins and Robin Wilson laid this out in a 2004 article in the College Mathematics Journal titled, pointedly, “The Truth about Konigsberg”. It is a good thing to tell a child: the person who founded a field did not finish it.
What grew out of this is not a curiosity. Route planning for bin lorries, street sweepers and snow ploughs is an Eulerian problem with repeats allowed. Circuit board layout, railway timetabling, protein interaction maps, the structure of the internet and the friend graph of a social network are all graph theory. The subject exists because somebody took a town’s bridges seriously.
How to Explain It to Your Kid
Ages 5–8: cross every bridge once
Do the floor version and nothing else. Do not draw the graph. Say: “Every bridge once. No bridge twice.” Let them walk, get stuck, and walk again. Then peel off one bridge without explaining why and let them succeed. The thing a six-year-old takes from this is that some puzzles change when you change the board, which is quietly a big idea.
Ages 9–12: count the odd ones
This is the right age for the whole thing. Get the four numbers written down, then do the in-and-out argument with your hands: “you come in on one bridge, you leave on another, that’s two.” Then give them the rule and the three invented layouts from step 8. The test of understanding is not whether they can recite the rule, it is whether they can look at a new drawing and call it before tracing.
Ages 13+: parity, proof, and the postman
Give the argument as a proof about parity and ask them to write it out in their own words, which is harder and more useful than it sounds. Then set the real extension: the Chinese postman problem. If a route covering every edge once is impossible, what is the shortest route that covers every edge at least once? Have them find it for Konigsberg by hand, counting repeated bridges. That is an optimisation problem used in actual logistics, and it is solvable on a diagram this small.
The question to ask: “Here is a map with six places. Two of them have an odd number of roads and four have an even number. Without tracing anything, tell me whether I can drive every road exactly once, and if so, where I have to start.”
Why Teaching This Early Pays Off
Most school mathematics is about quantity. Graph theory is about structure, and a child who has done this once knows that you can answer a question about a real object by deliberately throwing most of the object away. That move, modelling, is the skill that the Framework for K-12 Science Education names as developing and using models, and it is the one that transfers furthest.
The second payoff is a specific and unusual one: an encounter with a proof of impossibility. Children meet plenty of problems they cannot solve. They almost never meet a problem that nobody can solve, with a reason attached that they can check themselves. That distinction between “unsolved” and “impossible” is load-bearing later, in computer science especially, where the whole theory of what algorithms can and cannot do rests on it. Jeannette Wing’s framing of computational thinking puts this kind of reasoning about problem structure ahead of writing code, which is the right order.
The usual caveat applies. One afternoon with masking tape does not make anyone better at maths tests, and nobody should pretend it does. Kirschner, Sweller and Clark’s warning is relevant here specifically: left to explore alone, a child doing this learns that bridges are annoying. The parity argument has to be handed over. Fifteen minutes of walking plus two minutes of counting is the combination that works.
Before You Start: What Parents Should Know
| Best age | 9 and up for the mathematics. 5 and up for the walking, with no expectation that the rule lands. The jump at about 9 is the ability to hold “even” and “odd” as properties of a count rather than of a number. |
| Supervision | Low. You are the referee for whether a bridge has been used, which is why the sticky notes are not optional. |
| Real hazards | Tripping. Clear the floor properly, no socks on hard floors, no running. Masking tape only, because stronger tape lifts varnish and paint from wood and skirting boards. |
| Setup | About 2 × 2 m of clear floor, plus a table for the paper stage. The walking version does not work on carpet with a rug over it. |
| Frustration point | Attempt eight to ten. Children often decide they are not smart enough. Say at the start that one possible answer is “it cannot be done” and that finding out which is the job. |
| Don’t | Don’t tell them it is impossible before they have tried, and don’t skip the floor version to save time. The walked failure is what makes the counting feel like an explanation rather than a rule. |
When It Doesn’t Work
- They cross a bridge twice without noticing. The sticky notes solve this. Without a physical record, every attempt dissolves into argument and the pattern never emerges.
- The tape bridges peel up. Press both ends down hard and keep strips under 60 cm. On carpet, tape does not hold at all, so switch to strips of paper weighted with books, or go outside with chalk.
- They find a route. Then a bridge is missing or duplicated. Recount: you need exactly seven, two from A to B, two from A to C, one from A to D, one from B to D, one from C to D.
- The parity argument does not land. Walk it physically with them instead of describing it. Stand on landmass B, step onto one bridge and off the other, and say “two”. Repeat on C. The abstraction usually arrives through the feet.
- A younger child gets upset at an impossible task. Remove the B to D bridge and let them succeed, then put it back. Alternating between a possible and an impossible version is better teaching anyway.
Take It Further
Solve the postman version. Konigsberg has no one-pass route, so find the shortest route that crosses every bridge at least once, allowing repeats. Count how many bridges have to be doubled. Then ask which bridges, and whether the choice matters. This is the Chinese postman problem and it is how real street-cleaning routes are planned.
Make the Konigsberg graph solvable in two different ways. Adding a bridge or removing one can both work. Removing B to D leaves two odd vertices. Adding a bridge between B and C also leaves two. Have them find and justify both, then find a change that makes a closed loop possible, where you finish where you started.
Map your own neighbourhood. Draw the street layout of a few blocks as dots and lines, with junctions as vertices. Count the odd junctions. Then ask whether a dog walk that goes down every street exactly once is possible. The answer is almost always no, and the reason is now obvious to them.
What to Watch For Over the Next 3 Months
- Week 4: They test the rule on drawings in books and on puzzle pages, usually the “draw this envelope in one stroke” type, and get it right by counting.
- Month 2: The word “impossible” starts being used more carefully, as something you can show rather than something you feel.
- Month 3: The question worth encouraging is about networks they care about: subway maps, friend lists, game maps. Graph theory is the answer to most of those, and they already have the first idea of it.
Frequently Asked Questions
Is the Konigsberg puzzle really impossible?
Yes, and provably so. Four of the four landmasses have an odd number of bridges, and a walk using every bridge once can tolerate at most two odd landmasses, which must be the start and the finish. This is not a hard puzzle. It is a demonstration that the answer is no.
What age can actually understand the explanation?
About 9. The argument needs a child to treat odd and even as properties of a count and to accept a general statement about every possible route rather than the routes they tried. Below that the walking is still worth doing on its own terms.
Did Euler solve it in 1736 or 1735?
He worked it out in 1735. The paper appeared in 1741, in a volume of the St Petersburg Academy’s journal covering the year 1736, which is where the commonly quoted 1736 comes from. The Euler Archive lists the paper as written 1735 and published 1741.
What is graph theory actually used for?
Routing bin lorries and snow ploughs, laying out circuit boards, scheduling trains, modelling protein interactions, analysing social networks and running the internet. All of it turns on which things are connected to which, which is exactly what this puzzle isolates.
Can we do it without taping up the floor?
You can do the whole thing on paper, and it takes fifteen minutes instead of fifty. The floor version is better for children under 11 because the two-bridges-at-a-time argument is something they can feel with their feet before they can follow it as words.
About the author
Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.
Sources
- Euler, L. (1741). “Solutio problematis ad geometriam situs pertinentis.” Commentarii academiae scientiarum Petropolitanae, 8, 128–140. Euler Archive, record E053 (written 1735). https://scholarlycommons.pacific.edu/euler-works/53/
- Hopkins, B., & Wilson, R. J. (2004). “The Truth about Konigsberg.” The College Mathematics Journal, 35(3), 198–207. https://doi.org/10.1080/07468342.2004.11922073
- Wing, J. M. (2006). “Computational thinking.” Communications of the ACM, 49(3), 33–35. https://doi.org/10.1145/1118178.1118215
- Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). “Why Minimal Guidance During Instruction Does Not Work.” Educational Psychologist, 41(2), 75–86. https://doi.org/10.1207/s15326985ep4102_1
- National Research Council. (2012). A Framework for K-12 Science Education. National Academies Press. https://doi.org/10.17226/13165
- CS Unplugged, University of Canterbury. Unplugged computing activities, including network and routing units. https://www.csunplugged.org/en/topics/searching-algorithms/
Related reading: cutting a Mobius strip in half, colouring a map with four crayons, and computational thinking versus coding.
