How to Find North Without a Compass (and Why Moss Lies to You)
Old camp joke: "You'll never get lost if you remember one rule — moss always grows on the north side of your compass."
01Folk navigation: an audit
Moss points north. Anthills face south. Branches grow thicker on the south side, bark thicker on the north. This ruleset has been handed down for generations, and almost nobody has tested it.
We tested it. Every known method sorted into three bins: works, works conditionally, doesn't work — with an error figure for each, because "roughly north" without a number is not information, it's a mood.
The short version up front: the most accurate compass in the forest is a stick and two pebbles. Details below.
02Moss first, because it's the most popular
Moss doesn't respond to cardinal directions. It responds to moisture and shade.
On a lone tree in an open field, the north side really is shadier — that's where the moss grows, and that's where the myth comes from. But people don't get lost in fields. They get lost in forests, where shade comes from every direction and humidity is set by terrain, wind and canopy. In such a forest, moss grows wherever it likes. In a wet climate — everywhere at once: on rocks, on fences, on all sides of a trunk simultaneously.
Two facts finish the method off. First: in a good share of surveyed woods, moss is actually thicker on the south side — it's warmer there, and the plant approves. So the myth is worse than imprecise: it periodically points the opposite way. Second: the green film on bark that people read as "moss" is often not moss at all but a single-celled alga, Pleurococcus. Generations of mushroom pickers have navigated forests by an alga.
Sometimes moss does agree with north — with the same reliability with which a stopped clock shows the correct time twice a day.
03One criterion instead of memorizing
You don't need to memorize which method works. One criterion covers it:
Methods driven by the sun and stars are geometry. Methods driven by living things are statistics with a wide spread.
The sun and stars follow celestial mechanics — identical yesterday, today, and in a century. Moss, anthills, crowns and bark depend on moisture, wind, soil and the neighbors: too many variables to make an instrument out of.
04 · WORKSThe methods that survive contact with reality
The shadow-stick method (gnomon). A stick planted vertically, about a meter tall. Mark the shadow tip. Fifteen minutes later — mark it again. The line between the marks runs east–west; the first mark is always west, the second always east, in both hemispheres. The perpendicular is north–south.
The method includes a mandatory fifteen-minute pause — a rare case of a measuring procedure scheduling your rest for you. Use it as intended.
Accuracy ±2–5°. The non-obvious detail: accuracy is worst at noon, when the shadow is short and its azimuth moves fastest. Best hours: morning and mid-afternoon. The intuition "I'll wait until the sun is higher" works exactly backwards.
Polaris. ±1–2° — the most accurate no-instrument method there is. Found from the Big Dipper's bowl. Polaris is not the brightest star in the sky — it ranks around forty-eighth; whoever "takes the brightest one" takes Venus and follows a planet. We covered the finding procedure, and how to read your latitude off Polaris, separately.
Orion in winter. Orion's belt — three stars in a row — rises almost due east and sets almost due west. When clouds have taken Polaris and Orion hangs above the horizon, take your direction from it. The method has run for several thousand years without a design change.
Sun plus time of day. Morning east, evening west, noon south (northern hemisphere). Low precision, top speed: orienting a map takes ten seconds.
The watch method. Hour hand at the sun; the bisector between the hand and 12 o'clock points south. ±10–15°; mind daylight saving time. Southern hemisphere — invert it: point the 12 at the sun instead, and the same bisector points north. On a digital watch face you'll have to imagine the hands — progress left the method without its hardware.
FIELD SKILLS · SUNDIAL COMPASS Practise the sun-and-shadow bearing before you need it. →05 · WORKS CONDITIONALLYAdmissible as confirmation, never as a source
Anthills. The gentler slope statistically faces south more often: the colony collects heat. But slope also answers to terrain, wind and the nearest tree. Admissible as confirmation of the shadow method. Inadmissible as a sole source.
A lone tree's crown. In a field — yes, the south side is fuller. In a forest the crown grows toward the gap in the canopy, not toward the south: two neighboring trees can point opposite ways.
Snow in spring. South-facing slopes clear first. This one is reliable — the same solar geometry, written onto the terrain. A spring forest is the easiest to read.
Resin on conifers. Flows more on the warmer side. Warmer — not necessarily southern: a tree beside a sun-heated rock takes its heat from the rock.
The crescent moon's horns. A line through both horns, extended to the horizon, points roughly south (northern hemisphere). Coarse — but at night without Polaris, coarse geometry beats confident intuition; on intuition's reliability, see the experiment below. What the moon phase actually predicts →
06 · DOESN'T WORKRetired with honors
Moss = north. Covered above.
Tree rings on a stump. Ring width tracks competition for light: a neighbor fell, rings widened toward the gap — regardless of compass direction. A stump documents the forest's history, not a bearing.
Bark thickness. Depends on species, age, wind, and mechanical damage. No correlation with cardinal directions usable for navigation.
Branches denser on the south side. In a forest — see crowns. Separately: try comparing, by eye, the branch density of a mature tree from four sides. The result of that comparison will describe you, not the tree.
Sunflowers follow the sun. Only young plants track the sun. A mature field locks facing east — permanently. So sunflowers are a compass after all; they just point not "at the sun" but fixed east. A rare case where the myth is wrong and a method falls out of it anyway.
"A compass points where you need to go." We've heard this one too. No. The mechanism for that is not provided.
07North in a village or a town
Settled landscape is marked up better than wilderness — and hardly anyone uses it.
Geostationary satellites sit over the equator, so in the northern hemisphere every dish looks into the southern sky. Unlike anthills, dishes don't disagree with each other.
A centuries-old tradition orients the altar east. Not absolute — terrain and street grids sometimes beat the canon — but as a hint in an unfamiliar old town, it works.
Turned south for the same reason as the dishes: the owner wants maximum sun and has already done the calculation for you.
The architects of past centuries didn't know they were building navigation infrastructure; it came out well regardless.
08The price of error, in meters
Degrees are an abstraction. Converted to drift:
| ERROR | METHOD | DRIFT / 1 KM | OVER 5 KM |
|---|---|---|---|
| ±1° | Polaris | 17 m | 87 m |
| ±2° | shadow stick, unhurried | 35 m | 175 m |
| ±5° | shadow stick, rushed | 87 m | 436 m |
| ±10° | sun + time of day | 176 m | 882 m |
| ±15° | analog watch method | 268 m | 1,340 m |
| ±30° | moss | 577 m | 2,887 m |
The Polaris method runs ±1–2° depending on how carefully you sight the Dipper's pointers — the table takes the tighter half. At the worse end of that range (2°), drift matches the shadow-stick row above: about 35 m per kilometre, not 17 m.
Moss appears in the table out of respect for tradition.
Now the targets. A forest road is a strip five meters wide and kilometers long: you'll cross it even with a 30° error. A tent, a car, a bridge are point targets: 2.6 km of moss-grade drift means you pass half a kilometer from your own camp and never know it.
Hence the rule older than the compass: aim at a line, not at a point. A river, a road, a power line, a forest edge — impossible to miss. Reach the line, then walk it to the point.
09The experiment that closed the "I can feel the direction" argument
People with GPS trackers were sent to walk "as straight as possible" — for hours, through a German forest and the Sahara. Souman et al., Current Biology, 2009 (Max Planck Institute for Biological Cybernetics).
Results. Those who could see the sun or the moon walked nearly straight. Forest participants under overcast skies walked in circles, crossing their own tracks without noticing. The desert participant held his course while the moon was up — and began to loop once it set. First conclusion: the brain holds a heading only against an external reference, and does so without informing the owner.
Then participants were blindfolded. The circles tightened to under twenty meters in diameter — a person cannot cross a football pitch the short way while convinced they're walking straight. Second conclusion, which buried a popular theory: "one leg is stronger and pulls you sideways" — false. The same people looped left on one trial and right on the next. The cause is accumulated small errors in the sense of "straight ahead," which nothing resets without an external fix.
One of the study's authors said after the experiments that he would never again enter a large forest without a compass. A person who studied direction-keeping professionally drew exactly one conclusion from his own data. It is rational to draw the same one.
The practical summary: "I can feel the way" is not data. Data is the sun, a shadow, a star, or a bearing on a scale.
10Where our cards come in — and where they don't
The shadow method needs a stick and two pebbles. There is nothing we can sell you at this stage, hard as we tried.
The card matters at the next step. North is established — now the bearing has to be held along the route, and the eye is a poor holder: every detour around a bush eats a few degrees, and — see the experiment above — without an external anchor the errors only accumulate. A protractor on the card supplies the procedure: take an angle to a visible landmark, reach it, take the next. Scale reading resolution — down to ±0.7°, against whole degrees and tens of degrees for "somewhere that way." The Universal 4.0 also carries a sundial — a factory-made gnomon, with engraved markings instead of pebbles. For a longer trip, the click-in Wrench KIT adds field torque to the same slot.
And the limit of application, stated plainly: a planned trip takes a map and a compass. The card is for the day the compass stayed home and the phone died. A reserve — not a substitute for preparation.
Curious exactly what that protractor and the drill gauge resolve to? We put the Professional through a full teardown in What Size Is This? The Go/No-Go Physics of a Wallet Gauge Card. New to the card itself — start with the complete field guide.
See every cutout in 3D → Tool Explorer.
CARD FINDER Not sure which card fits your kit? Four questions, about 30 seconds. →Overcast, no shadow. Now what? +
Check first: in light overcast a shadow exists, just faint — hold up something pale, a sheet of paper or the card itself. None at all — switch to linear features: terrain leads to water, water leads to people, a road is heard before it's seen. Slopes and rivers work without the sun.
Isn't the phone compass enough? +
It is — while the phone is charged, calibrated, and not sitting next to metal. Three conditions, each questionable in a forest. The shadow method has no conditions. Keep both.
Is Polaris the brightest star? +
No — roughly forty-eighth. It's found from the Dipper, not by brightness. The brightest thing in the evening sky is usually Venus, which is useless for navigation: it keeps its own schedule.
Is "fire three shots in the air" a real distress signal? +
The principle is real; the phrasing is from the old joke — the one where the hunters fire and fire, and one finally says: good thing we still have the arrows. The universal distress pattern is anything in series of three with pauses: whistle, knock, flash. A whistle outlasts a voice. A polished card in the sun is a heliograph with kilometers of range.
Can anthills be trusted? +
Like statistics: as confirmation, yes; as proof, no. If the anthill agrees with your shadow, good. If it argues — the side with celestial mechanics wins.
"Follow the river downstream to people"? +
In settled Europe — mostly yes. In marshland the river leads into the marsh; in mountains, into an impassable canyon. The correct form: water is a linear feature if you know where it flows. If you don't — hold to a road or a power line: people built them, and they end at people.
I'm already lost. Procedure? +
Stop — kilometers walked in panic are the most expensive kind. Sit down, eat something: glucose restores decision-making. Re-establish the last point where your position was certain. Find north by shadow. Choose a linear target. Walk to it holding the bearing by the scale, not by feel — for why not by feel, see the experiment above.
Drift figures: arc length d·θ in radians (≈ d·tan θ at small angles; at ±30° the straight-line miss runs 500–577 m per km depending on the geometry you count). Circle-walking: Souman et al., Current Biology, 2009 (Max Planck Institute for Biological Cybernetics). Moss and Pleurococcus: field surveys and published myth-busting reviews.
Andrii Gurskyi, founder & engineer, MRF.Tools
In 2015 he went looking for a real tool for his own wallet, found only stamped junk or overpriced steel — and started machining his own. Ten funded Kickstarter campaigns, 4,000+ backers, every reward shipped — several of them through a war. Designed and developed in Ukraine. About MRF →