How to play Slitherlink
- Draw lines between neighbouring dots, across or down.
- Make each numbered square have exactly that many sides drawn.
- Mark an X on any edge you know is not part of the loop.
- Keep the line a single loop: no crossings, no branches, no dead ends.
- Close the loop when every number is satisfied.
Slitherlink rules
The grid
- The board is a rectangle of dots. Some of the squares between the dots hold a number: 0, 1, 2 or 3.
Drawing the loop
- Lines join two neighbouring dots horizontally or vertically, never diagonally.
- A numbered square has exactly that many of its four sides on the loop. A blank square has no limit.
- Every dot has either no lines or exactly two. This is what keeps the line one continuous loop with no branches or loose ends.
- There must be one loop only, not several separate ones.
Winning
- The loop is closed, it is a single loop, and every number is satisfied.
Slitherlink techniques, from first game to expert
Beginner
1. Clear around every 0
None of a 0's four sides is used. Mark all four with X. This often settles the sides of the numbers next to it.
Beginner
2. Learn the corner numbers
A 3 in a corner of the board uses both outer edges at that corner. A 1 in a corner uses neither: the corner dot would otherwise need two lines from the 1's sides, giving it 2.
Beginner
3. Read a 3 beside a 0
The side they share is empty because of the 0. So the 3 uses its other three sides, and the lines at both ends continue away from the 0.
Improving
4. Read two 3s side by side
When two 3s are next to each other, the edge between them is on the loop, and so are the two outer edges parallel to it. The short edges that continue the middle line beyond the pair are empty.
Improving
5. Read two 3s on a diagonal
For two 3s that touch only at a corner, the two outer edges at the far corner of each 3 are on the loop.
Improving
6. Count lines at every dot
After each move, check the dots around it. A dot with one line and only one free edge must continue along that edge. A dot with two lines gets X on its other edges. A dot with three X marks has no lines at all.
Expert
7. Avoid closing a small loop
When two loose ends of the line are next to each other, joining them closes a loop. If numbers are still unsatisfied elsewhere, that edge must be an X.
Expert
8. Use inside and outside
Every square is either inside the loop or outside it. Two neighbouring squares are on different sides exactly when the edge between them is a line. On hard boards, shading inside and outside settles edges that the numbers alone do not.
A worked example
A 4×4 board of squares (5×5 dots). Blank squares are shown as dots:
Row 1: . . 3 3
Row 2: 2 1 . .
Row 3: 2 0 0 2
Row 4: . . . .
- The two 0s in row 3. Mark X on every side of both. That includes the right side of the 2 at R3C1, the left side of the 2 at R3C4, the bottom of the 1 at R2C2 and the top of the square below each 0.
- The pair of 3s in row 1. They sit side by side, so the edge between them is a line, and so are the left side of R1C3 and the right side of R1C4. The short edge that continues the middle line downward, between R2C3 and R2C4, is X.
- The corner 3 at R1C4. It is in the board’s top-right corner, so its top and right sides are lines. With the middle edge, that is 3, so its bottom side is X.
- The dot between the two 3s, on the top edge. It already has two lines: the shared edge going down and R1C4’s top going right. A dot can only have two, so the top of R1C3 is X.
- R1C3 again. It has its left and right sides and its top is X, so its third line is its bottom.
- The dot at the top-left of R1C3. It has one line, R1C3’s left side, and its edge to the right is X. A dot cannot stop with one line, so the line continues left along the top of R1C2.
- Down the right edge. The dot under the corner 3 has one line coming down and an X to its left, so the line carries on down the board’s right edge.
The loop is starting to take shape: from the top of R1C2 it dips round R1C3, rises between the two 3s, crosses the top of R1C4 and runs down the right edge. The 2 at R3C4, whose left side is X, then decides how it turns back. Continue with the same two tools, the numbers and the two-lines-per-dot rule, and the finished loop goes down the right edge to row 3, steps in under R3C4, drops to the bottom edge, runs left along the bottom, climbs the left edge to the top of row 3, cuts in along the top of R3C1, then climbs between columns 1 and 2 to close at the top of R1C2. Check it: every number has exactly its count of sides.
History and variants
Slitherlink was popularised by the Japanese publisher Nikoli and is also published as Fences and Loop the Loop. Variants use triangular or hexagonal cells, but the square grid with numbers 0 to 3 is the standard form.
Where to play Slitherlink
Gridwise
Logic puzzles in one calm app
- Slitherlink is in the Gridwise collection, also named Loop puzzle and Fences.
- It offers 300 levels on a level path, with a trophy level closing each block of ten.
- Typical solve time is about 8 minutes, among the longer puzzles in the app.
- Pip can point at the next deduction, explain it, or draw the edge.
- Plays fully offline, with no lives to lose.
Questions about Slitherlink
What are the rules of Slitherlink?
Draw one closed loop along the grid lines so that each numbered square has exactly that many of its sides on the loop. The loop never crosses or branches.
Can the Slitherlink loop cross itself?
No. Each dot has either zero or two lines, so the loop cannot cross, branch or stop.
What does a 0 mean in Slitherlink?
None of that square's four sides is part of the loop.
How do you start a Slitherlink puzzle?
Mark X around every 0, then look for 3s next to 0s, pairs of adjacent 3s and numbers in the board's corners. Those patterns place the first lines.
Does a blank square have to be inside the loop?
No. A blank square can have any number of its sides on the loop, and it can end up inside or outside.
Facts checked against the game's own design documents on .