Worldbuilding Stack Exchange is a question and answer site for writers/artists using science, geography and culture to construct imaginary worlds and settings. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

A character of my story is in the following situation: He is in the middle of a forest and has lost his way out. He wants to be sure to keep straight on and not to move in a circle.

  • He has no compass
  • The sun and stars do not exist, just diffuse light.
  • A second person is with him, they have no special equipment (I could give them some things that they have in their pockets, but no large gadgets).
  • He is a scientist and very clever.

Is there a way my characters can make sure that they walk straight on for several kilometers under this conditions?

EDIT: I asked this question yesterday and since then I got a lot of nice answers, thank you.

But I decided to describe the setting a little bit better since some answers do not quite fit and there are a few questions in the comments.

My story takes place in a fantasy world. In this world sun, moon and stars do not exist. The light comes uniformly from all directions (not from the underground), so one can not use it for orientation. Weather-phenomenons also are distributed isotropically. Furthermore there is no magnetic field so a compass would not work.

The forest itself is very homogeneous. There is only one sort of trees, they are about three meters high and not very close together. There is no underwood, only grassy forest floor. The underground is flat, there are no hills or troughs and no rivers. I described it in my story as 'like the sea, always the same but always different'.

I hope the setting and the reason why even my clever charakter got lost and has problems to find a way out is a bit clearer now.

share|improve this question
How straight do you need to walk? This is a forest after all. Straight is often a very expensive way to travel. – Cort Ammon Jan 18 at 1:39
How dense is the forest? – aroth Jan 18 at 4:47
Navigate by GPS, of course! – Emilio M Bumachar Jan 18 at 7:13
There is no underwood, only grassy forest floor. In my experience, forests generally don't have grass because there isn't enough light getting below the canopy, but maybe there is a reason in your world. – Steve Jan 18 at 21:27
@GarethRees Hey, I didn't mean you should delete your answer! I just thought that you might want to edit it a little to fit the updated question! It was a good answer and useful information, just slightly out of date. – AndyD273 Jan 19 at 16:07

17 Answers 17

This answer was in response to an earlier version of the question where it seemed possible that the story environment might resemble Earth on a cloudy or misty day.

Navigation in dense woodland without a compass is very hard. Here are some tactics:

  1. If the wood has a distinctive underlying topography, then you can follow that. For example, you can follow a ridge (easier uphill), or a valley (easier downstream). You can follow a contour line by refusing to climb or descend.

  2. If the region has strong prevailing winds, then the trees may show the direction: they may have stronger growth on the sheltered side, or bigger root systems on the windward side.

  3. If sun often penetrates into the wood, then vegetation will distinguish the sunny and shady directions (south and north, respectively, in the northern hemisphere). Flowers prefer to point in the sunny direction; moss prefers the damp conditions on the shady side of trees; phototropic climbers prefer the sunny sides of trees.

  4. If the trees are climbable, then you can climb them and sight distant landmarks.

I use tactic (1) all the time—but it relies on knowing the general topographic layout of the landscape, for example from looking at a map. The observations in (2) and (3) are from Tristan Gooley's The Walker's Guide to Outdoor Clues & Signs. But I find these techniques far from easy to apply in practice and I would hate to have to rely on them. Tactic (4) is pretty desperate—even if you're lucky enough to find a tree tall enough to give you a view over the canopy, it's very risky.

share|improve this answer
I was assuming there is no distinctive topography to follow myself. Still you are right that the question does not actually state that. Tempted to copy it to my answer, but I'll resist the temptation. – Ville Niemi Jan 17 at 18:52
Moss trick wouldn't work in environment with "no sun". Although, how a forest or indeed a habitable planet can exist without a sun is an even bigger mystery... – Darrel Hoffman Jan 17 at 20:20
@DarrelHoffman: I am interpreting "no sun and no stars" to mean "the characters can't see the sun or stars when they are lost in the woods, for example because of thick cloud or mist." – Gareth Rees Jan 17 at 20:24
Following a contour line may well have you walking in circles, depending on the topography. Altitude is irrotational, though, meaning you won't ascend or descend indefinitely without getting somewhere. – Kevin Jan 18 at 23:46
@GarethRees I'm sad that you deleted your answer. If a question changed in a way that invalidated an answer then the question should be adjusted, not the answers. Would you consider undeleting so the community can figure out how to address the situation? It looks like a good answer and it'd be a shame to lose it. – Monica Cellio Jan 19 at 16:23

If he doesn't want to go in a particular direction but just wants to leave the forest without getting turned around, walking downhill is guaranteed to eventually work. Downhill is a better choice than uphill because in nature there is almost never a bowl shape without a downhill exit (due to the need for drainage), but peaks are common. In addition, civilization often clusters near water, which is found by going downhill.

Once water is found, this strategy becomes "follow the water", and there is little chance of getting turned around. Water never flows in a circle.

Here is a famous example of a girl who, after a plane crash, navigated out of the Amazon rain forest, travelling ten days to civilization, using this technique:

share|improve this answer
The problem may be that you arrive in the fork of two rivers too deep or fast to ford. At that point you have to choose between swimming (risky), building a raft (need tools), or proceeding upstream in search of a ford. – nigel222 Jan 18 at 8:32
@nigel222 who cares - once you've found a river, you have a navigable feature to follow without getting lost (and a high probability of finding someone/civilization). – gbjbaanb Jan 18 at 16:34
"... in nature there is almost never a bowl shape without a downhill exit ..." What about ponds and lakes? – Peregrine Rook Apr 27 at 19:18
The water flows out of the pond or lake somewhere, or else the pond/lake would be deeper – QuadmasterXLII Apr 30 at 4:56

I felt this would be better suited for a separate answer from my original, though it might be contrary to the criteria: have the characters make their own compass.

If you granted the characters a magnet and a small metallic object (such as a paper clip or needle) in their pockets, the latter could be magnetized and placed on top of a (very) small leaf or something else that would float in water. The object will then turn with the leaf and point north. The body of water would have to be still and possibly portable somehow (an uncovered thermos, perhaps?) unless you combined my other answer with this one- find north (or whatever way you wanted to go) and leave a trail.

share|improve this answer
When using this technique, how to tell the difference between north or south? – Kyslik Jan 18 at 11:07
If you use the magnet and strike from the eye of the needle towards the pointy end, then the pointy end will be north. – Esben Boye-Jacobsen Jan 18 at 13:21
@EsbenBoye-Jacobsen: Not if you strike with the other pole of the magnet instead. – Henning Makholm Jan 18 at 18:54
You don't need to tell the difference between north and south, you just need to consistently follow the same end of the needle (i.e. don't perform any 180 degree turns). The questioner isn't bothered whether they go north or south, just that they don't go in circles. Unfortunately, the questioner has also now added that there is no magnetic field. – Steve Jessop Jan 19 at 11:09

First person places stick where he is. Second person chooses the direction to go to and advances, they must both remain visible to each other.

The second person stops when he thinks he has advanced enougth and signals the first person.

The first person advances half of the way and place another stick in the middle.

The first person then advances beyond the second person until he can barely see the newly placed stick.

I think you can see where I'm going with this....

In short by using the sticks they can maintain a straight line.

share|improve this answer
That technic would do. But because they are in a forest it would be difficult to see the sticks or the other person in a distance more than a few meters, so they would walk on very slowly (the more so because they have to cut or collect the sticks). Maybe there is a qicker way? – Kiki Jan 17 at 18:10
Have you ever been in a forest? there are lots of things (I think they are called trees) that prevent you from going straight up, or from seeing signals left a few metres behind. – SJuan76 Jan 17 at 18:10
@SJuan76 I disagree. Forests normally don't block visibility that fast--there's not enough light to support that much vegetation. Dense vegetation only occurs where there's a break in the forest--something I don't think the scenario includes. – Loren Pechtel Jan 17 at 19:59
It really depends a lot on the type of forest... some have lots of meter-high undergrowth, others are indeed almost “empty halls” under the canopy. But even those (except human-grown pine plantages) tend to have enough obstacles in them to make it impossible to extend lines of sight to an actual straight course. Some zig-zagging is not such a big problem when you have an absolute reference and can always take up the old course, but with your method you'd need to measure the exact corner angle at each step. Deviations will add up. So, I agree with SJuan that the method will probably not work. – leftaroundabout Jan 18 at 0:18
They don't have to collect the sticks unless they care about people not following them, there are enough sticks laying on the ground in a forest. If they can't see the sticks you can try and use the trees themselves by placing a mark on the bark high enough for you to see it. I agree, deviations will add up but I think it would still allow them to advance in a fairly straight line for several kilometers which is what the OP asked for. – Jonathan Jan 18 at 8:45

I don't think there is a certain always works method. Trees obstruct your line of sight and make it impossible to actually walk straight over extended distances. That said people walk in forests without compass or other navigation aids all the time and mostly come back alive.

First advice would be not getting lost. This is not the same as knowing where you are, you just need to pay attention when walking in the forest so that you can walk back the same way you came if necessary. This basically amounts to recognizing things you passed when walking the other way. In practice people also recognize things they passed the other times they were in the same forest so that they can navigate familiar forests with ease.

So basically saying a very clever person got lost in the forest kind of assumes he thought the other guy was navigating for them. And the other guy obviously thought the opposite. In such case it is possible neither of them paid attention to their surroundings and cannot recognize anything.

Even so the goal should be going back the way you came in hopes you will recognize something. So it is generally a good idea to have some idea which direction you are going when you go to a forest otherwise trying to determine directions after you get lost is an exercise in futility since it won't really make a difference which way you go.

Assuming they know the original direction relative to either sun or wind they can try looking up. Even when the light is diffuse and the wind non-existent in the ground, the tops of the trees might have clear light and shadowed sides and so some wind. This is because the light and wind on the forest floor are being obstructed by the trees. This is probably the best bet for "something a clever guy would realize" as it hinges on realizing the reason the light is so diffuse and the wind non-existent.

If the wind is really almost non-existent it might still be possible to detect direction by using very light object. Similarly the direction of sun can be detected by using a sunstone. Although it is extremely unlikely they are carrying one. Still recognizing one you see and being able to use it would be pretty impressive. And many crystals have similar properties, remembering you carry one and that it can be used for detecting the direction of sun would also be pretty impressive.

There are also moss and lichen that grow better on specific sides of the tree trunk. If common in your forest this can be spotted and used to get a direction. Specifics vary, it can depend on either sunlight (north-south) or prevailing winds (east-west). EDIT: This is actually same but bit different than what Gareth Rees mentions in his answer. What I am talking about is only available in some forests, but is actually more practical to use if available since it is more consistent.

I said the tress make it impossible to walk straight. Still you can and should avoid walking in circles. Pick a tree. Walk toward it. Pick a tree behind the first tree. Walk toward it. And so on. This doesn't allow you to walk straight, but it helps avoid the natural tendency to walk in circles. If you are not sure which way to go anyway, this is good enough.

share|improve this answer
“Errors in direction should average out” – that doesn't quite work. If, at each step, you make a course error of δ, then after n steps you'll have an error of δ√n. Aften sufficiently many steps, the direction will be completely random even if δ is very small. And that's assuming you don't make systematic errors to either side; those will of course add up even more quickly. – leftaroundabout Jan 18 at 0:33
@leftaroundabout True enough, there is no reason it would actually average out. Funny thing is that I explicitly said myself it won't allow you to walk straight, it just avoids the "systematic error". And then wrote a next sentence that claimed it makes you walk approximately straight if the distance is large enough. Offending sentence edited out. Thanks! – Ville Niemi Jan 18 at 0:42

I believe the question is aimed at a scientific 'Macgyver' solution, in this regard, consider this:

The scientist has a key-chain flashlight, by filing down (hunk of glass/costume jewl/quartz from the earth) he is able to focus the light to a point.

He removes some bark from a tree (point A), focus' his light-source at another tree (Point B), and walks to it, removing some bark from the next tree, and pointing the source away from the first tree (A), he decides on the next point to travel to (Point C).

Or, given a laser-pointer.... But that's too easy.

None the less, it uses the play that light always travels in strait lines.

This type of mechanism is often used for maintaining alignment of tunnels during mining - though on a more accurate scale.

share|improve this answer
Nice answer John, welcome to the site. – James Jan 18 at 15:44
@James , Thank you :) – John Bargman Jan 18 at 15:51

Apocryphally, moss always grows on the northern side of the tree (well, in the northern hemisphere at least).

If you actually go out and try this, you'll find it works as poorly as you would expect. Surveying a large number of trees you might be able to establish a side on which moss grows best. But is that because that side is to leeward, northern-exposure, pointing-uphillest, or something else?

[Edit: thinking about it, that possibly doesn't matter, so long as it's not the last option. If the moss is consistently directional, then you can use it for making sure your path does not diverge, regardless of whether you know what direction you are traveling in. So I suppose this does answer the question as asked, even though there are far better responses in other answers. Sadly, things like leaf direction won't tell you the direction you expect the sun to be, since the light is diffuse. This fact may also affect the moss thing: direct sunlight would make the effect far stronger.]

share|improve this answer

If it isn't important which straight line you walk in, just that it's straight, then I can get you there with a fishing line and a stick.

Take your stick and draw a straight line. Have your companion hold one end of the fishing line while you hold the other. Walk to the end of the fishing line while your companion keeps the line parallel to the line you've drawn on the ground.

When you reach the limit, draw another line on the ground at your feet parallel to the fishing line. Have your companion come meet you.

Repeat as necessary.

share|improve this answer
Welcome to the site Ian. – James Jan 18 at 19:10

A method that doesn't take any equipment:
Choose two trees where one is in front of the other by some amount, and walk so that they always line up. When you get to the closest of the two trees pick two more that line up past the second. Every time you get to one of your trees mark it by peeling off some bark, breaking a branch so it hangs down, or something so that by looking back you'll be able to see where you came from.
This will help in case you somehow get disoriented and lose your chosen trees.

You'll have to stop at night, so putting several stakes into the ground to mark the direction you want to go before dark will help you find the next two trees when it gets light again. You'd also be able to look back and get a pretty good idea of where you want to go.

This will not get you moving in any particular direction, but it will take you in a straight line so eventually you'll get to the edge of the forest.

share|improve this answer
My parents once lived about a year in the jungles of Brazil; this was the recommended method for anyone not familiar with the area. If you need to walk straight, line up two trees in front of you, as far ahead as you can, and walk towards them. There will be accumulating error, but not as much as guessing. One fellow got lost, but managed to make it back to the village he was staying in (several miles away) using this method. It took a while, but he made it. – ArmanX Jan 18 at 22:28

Grass and trees, when not wet, burn... pull up grass, find those sticks, build a fire. always keep the rising column of smoke at your back, or better, have your friend point a long stick at the smoke, go to the end nearest the fire, look for the most distant object in line with the stick's far end. keep it in your gaze and walk to it, repeat. All the they need is a lighter and long stick and dry fuel..clear a burn barrier around your fire or you may burn the whole forest down...yes this method has been used before..but not always a safe option to use!

share|improve this answer
Given the fantasy restrictions I think this is the only sensible option. – BlindKungFuMaster Jan 19 at 10:40

Your character could use a simple method that I myself have used to make safe landings on a rocky, cliffy coast at night. We used the steep and rocky foreshore at the landing site to place two light-sticks, one above the other, so placed that they and the safe entrance to the cove formed a straight line. No closing with the shore until the two are lined up!

Your heroes just take out that handy little folding saw you left in their pockets and cut enough wood to make two tall, skinny masts which they tie to separate trees using some paracord they just happened to have, or perhaps improvising rope from sliced and twisted bark. Now, they just need to keep glancing backward as they walk - if the two masts are in alignment, then they have not deviated.

share|improve this answer

Try the breadcrumbs approach.

Have them take something from the forest (rocks, bark, flora, etc.) and leave it behind them in a trail. If they happen to walk in a circle back to where they started, the trail should be recognizable to them as their own creation.

share|improve this answer
With this method they would know when they walk in a circle, but they still do not know the right direction. – Kiki Jan 17 at 17:21
Your question was asking how your characters would know if they were walking straight on, not how they would know the correct direction to walk in to escape the forest. Technically, if they are in the middle of a forest like you said, any straight direction would eventually lead them out. – Wylia Jan 17 at 17:25
Yes, any direction would do, but my question was not how they would know if they were walking straight on, but how they can be sure that they walk straight on. When they found out that they moved in a circle they still do not know how to move not in a circle. – Kiki Jan 17 at 17:59
I'm not sure I understand what you're saying. – Wylia Jan 17 at 20:00
@Wylia - He's saying breadcrumbs are a form of error detection, but not prevention or correction. They won't stop you from going in the wrong direction in the first place, and intersecting your trail won't tell you which direction you need to go in order to get back on your intended heading. – aroth Jan 18 at 4:56

Battery powered gyroscope. Here's one that's been MacGyvered. Method of use:

  1. Suspend from string with axis approximately parallel to the ground.
  2. Hold axis in direction you want to go.
  3. Turn on, and wait for it to spin up, then let go.
  4. Only walk (on average) in the direction of the axis. (You may have to go around trees and other obstacles.)
  5. Except for torque similar to that experienced by a Foucault pendulum, the pointing should be stable to a few degrees until the battery runs down.

The torque in the last item will be very small because you aren't tall enough to have a long string and aren't intentionally rocking the gyroscope back and forth. That torque might not even exist in your setting if the "planet" doesn't rotate about some axis.

For substantially more information, read up on inertial navigation systems. Various implementations run quite a range of cost, size, and weight. However, single chip systems do exist, so the protagonist could already have one integrated into something in his pocket.

share|improve this answer

If he only needs to walk straight, all he needs is a straight rope that drags behind him. He should aim for the rope to always be as straight as possible and the longer it is the better.

share|improve this answer
This seems like an interesting approach, though it will only work in a flat environment, where you do not have to avoid trees, cavings or cliffs on the way – T3 H40 Jan 19 at 11:48
I think it fits the environment described perfectly. Trees can be walked around. Even through hills not requiring walk around it would be able to guide you, just with slightly higher degree error from the initial bearing. I think it is just an improved breadcrumb method. – mega_creamery Jan 19 at 16:02
Possible also to do the same by arranging and rearranging 3 objects in a straight line. Looking through them in a way that would reduce parallax error. – mega_creamery Jan 19 at 16:05

The Romans had a tool for ensuring roads went in a straight line. It was essentially a cross that could be mounted on a stick, in the middle, so that on each of the 4 points of the cross, a weight hung from a string.

Place a 1st stick in the ground at a point, then advance to a point where that stick is still visible, and place a 2nd stick in the ground, with your tool on top of the stick. Stand on the side of your tool, opposite to the first stick and rotate the cross, until both of the hanging strings line up, with the first stick and the centre stick of your tool (the 2nd stick), in a line.

Now have your colleague advance in what they believe to be a straight line away from the 1st stick and your tool. At some point, when they can still see you, they stop and place a 3rd stick. You look down the line of your tool, and the 2 strings, with the first stick behind you, and indicate to your colleague whether they need move the 3rd stick to the left or right to get into line with the string.

When they're lined up, you remove your tool from the 2nd stick, and go to place it on the 3rd stick. Your colleague can now go forth and line up a 4th stick. Eventually you'll have a line of sticks, in a perfectly straight line.

Clearly in the forest you may have issues, and want to vary the distance between sticks and vary your line a little to circumvent trees, but as long as you keep it as straight as possible, and if you have to deviate, alternate between left and right, you'll end up with a roughly straight path through the woods.

share|improve this answer

The following method may not be super-precise because directional hearing is not as good as seeing, but could circumvent the line-of-sight problems of other solutions:

  1. Person A walks away for certain distance (say, 100 m)
  2. A calls repeatedly for B and B walks ahead following the source of the calls.
  3. Once B believes to be somewhat closer but not too close (50 m, say) he calls A to move ahead away from his (B's) repeating calls
  4. When A is at a distance of about 100 m again, return to step 2
share|improve this answer

Based upon the updates to the question:


If your characters need to orient themselves in a particular direction before they start walking, it will be very difficult for them. You've removed all the things that might typically be used. None of the following are available:

  • Sunrises/sunsets
  • Moonrise/moonset
  • Stars at night
  • Magnetic fields
  • Natural landmarks
  • Artificial landmarks
  • Water features
  • Predictable/persistent weather patterns

Thus I'm not sure of any practical orientation technique that would allow them to set out on a specific heading. Though perhaps that's not required?

If it is, maybe at night they can see a faint glow on the horizon in the general direction of the nearest settlement. Or maybe there are auroras or some other aspect of the fantasy world that would be usable to get a ballpark directional fix. Otherwise...I don't see how your characters can orient on a specific heading with no visual references whatsoever, unless they have a sixth sense or a very keen sense of smell or similar contrivance.

Maintaining a Heading

This is easier based upon the scenario you've described. The trees in your forest aren't that large or that dense, they're just homogeneous and cover a large, flat area with no built-in visual references.

In that situation, all your characters need to keep a straight heading is a bit of wire, a knife or other sharp tool for making marks (optional), and some math. Your character should make a loop out of the wire, and attach it to their headgear so that it sits in front of one eye. They can then line their makeshift targeting reticle up with a distant tree and walk towards it. They'll travel in a straight line as long as they keep the tree centered in the reticle. And they should keep a count of the total number of steps taken to reach the tree.

When your character reaches the tree, they should mark their inbound trajectory with the knife (for instance, in the ground leading up to the tree, or the tree itself). Continuing on that trajectory they should proceed to the far side of the tree and look for a new tree in the distance to use as the new reference point. Before they go anywhere, they should mark the angle to the new reference point, and take note of the angle by which it deviates from their inbound course (I'm assuming that the trees won't always allow for a perfectly straight path to be followed).

Now all they have to do is proceed to the next tree, and repeat the process. By tracking the angle of each course deviation and the distance traveled between each tree, your scientist character should have no problems holding to a straight course by making minor corrections when choosing the next reference point (i.e. if you deviated to the right by 5 degrees for 500 steps, you'll want to compensate left 5 degrees relative to the original heading for 500 steps).

That should allow them to maintain a reasonably straight heading over a relatively long distance. Or if you give your characters very accurate tools for measuring distance and angles (a pedometer and a protractor, maybe?), it should keep a straight heading essentially indefinitely as long as there are trees to use as reference points. They could even execute turns and other maneuvers to detour around problematic bits of terrain if need be, just by keeping track of the angle and distance of travel. It's essentially a poor-man's inertial navigation system.

But won't work if, for instance, they get chased by something and lose track of either bit of information.

share|improve this answer

protected by James Jan 19 at 16:21

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.