The Kratky method looks like a trick: fill a container with nutrient solution, set in a seedling, put the lid on, come back six weeks later for the harvest. No pump, no air stone, no topping up. It works exactly until you do the one thing that feels most sensible - add solution when the level drops. Topping up is what kills the plant in this system, and the reason lies in how its root is organized.
Quick glossary: Kratky method - non-circulating hydroponics where all the nutrient solution is added before planting and nothing is added afterward; developed by Bernard Kratky at the University of Hawaii. Oxygen roots - the part of the root system left in moist air above the solution, working on respiration. Water roots - the submerged part, taking up water and nutrients. Air gap - the layer of moist air between the solution surface and the lid, which grows as the plant drinks the solution down.
How It Works Without a Pump
The layout is the same in every variant: the container is filled with nutrient solution before planting, and a seedling in a net pot is set so the lower 3 cm of the pot is immersed, enough to wet the whole growing medium by capillary action so the seedling waters itself.
After that you do nothing: the solution level falls, the root follows it down, and a zone of moist air is left above the water.
The classic description (Kratky, Acta Horticulturae 843, 2009) gives three variants. A 4-liter bottle for a single plant. A tank with a lid supported by the frame. A long raceway where an extruded polystyrene board floats at first and then comes to rest on 10 cm support pipes. The mechanics are identical; only the scale differs.
Having no pump is the main practical advantage. The system does not depend on electricity at all, and under Ukrainian conditions with scheduled outages that removes a whole class of risk that NFT cannot defend against in principle: there, a few hours without the pump means the film dries and the batch is lost.
Two Groups of Roots: Why Topping Up Drowns the Plant
The root in this system splits into two functionally different parts. The part left in moist air above the solution handles respiration; the literature calls these oxygen roots, and they branch vigorously. The part in the solution takes up water and nutrients.
Kratky's rule fits in one sentence: the solution level may stay where it is or fall, but it must never be raised. A raised level submerges the oxygen roots and the plant drowns. That is the only reason the method forbids topping up, and it is the rule most often broken: the level has dropped 20 cm, it looks alarming, and the hand reaches for the watering can.
The air gap is a physical requirement rather than a preference. In trials where the cover floated on the solution and no air layer formed at all, yields fell by up to 23% compared with a cover supported by the tank frame. In a bottle the level drops 15-25 cm over the cycle; in a tank, 6-12 cm. That is the working gap.
Volume: The One Calculation You Have to Get Right
The fill volume cannot be corrected later, which makes it the one place where a mistake costs the whole cycle.
Kratky's benchmark for lettuce: 4-8 liters of solution per plant at filling. Actual consumption over a cycle is 3-6 liters, and in Hilo, Hawaii a 200 g plant drank 3-4 liters. The difference between what is poured in and what is drunk is the reserve that carries the system through to harvest.
The remaining parameters come from the same paper: solution EC of 1.5 mS, 5 cm net pots, 20x30 cm spacing, 14 cm tank depth, a marketable head of 150-250 g, and a cycle of 6-7 weeks from seeding or about 5 weeks from transplanting. Water use works out below 20 liters per kilogram of lettuce, with a best recorded figure of 11 liters.
The 4-liter bottle suits crops that need less than 4 liters for their entire lifespan. Anything that drinks more will not grow in a bottle, however carefully the nutrient recipe is calculated.
Where the Method Works and Where It Fails
It works on short crops that finish in 5-7 weeks: lettuce, leafy greens, basil. Plus anything where autonomy itself is the value - teaching trials, a plot with no water or power, a windowsill batch running alongside the main system.
It fails on fruiting crops and long cycles. A tomato or cucumber drinks for months, the required volume becomes unrealistic, and the solution composition drifts unevenly meanwhile: the plant removes elements at different rates and nothing can correct that, because there is no longer any access to the solution. That applies to a single fill. Kratky did grow tomatoes without a pump, but the solution was kept at a constant level below 2 inches (5 cm) for the whole cycle, never raised: 'Big Beef' produced 5.9 lb (2.68 kg) of fruit per plant over a 72-day harvest (2005). That's a separate setup with topping off to a fixed mark, not "fill it and forget it."
The second limit is heat. Against oxygen shortage the method is better insured than DWC: oxygen roots breathe air and do not depend on dissolved oxygen in the solution. But the water roots sit all cycle in still water that nobody chills, and in summer it warms up along with the room. Warm standing water carrying organic matter from dead root tissue is the environment Pythium multiplies in best. In hot weather a Kratky setup either goes in the coolest spot available or does not go up at all.
The third limit goes unmentioned in popular instructions: still, unaerated water under a lid is a breeding site for mosquitoes. The lid has to fit tightly and the pot openings have to match the pots.
Three Mistakes That Cost the Most
Topping up when the level drops. Topping up cancels the method: raised water covers the oxygen roots and the plant loses access to air. A fall of 15-25 cm per cycle in a bottle and 6-12 cm in a tank is the system working normally, not a warning sign. The other failure modes, from salts in the water to rain, with numbers from Kratky's trials, are covered in why Kratky setups fail.
Using a clear container and letting light reach the solution. Within a week the walls are green. Algae consume oxygen, add organic matter, and attract fungus gnats, whose larvae carry root pathogens. The container or the lid has to be opaque; the commercial version lines the tank with black polyethylene for exactly this reason.
Filling "roughly enough" without counting the crop. If the plant drinks everything three weeks before harvest, the root is left in an empty tank and the batch stalls. The arithmetic is simple: consumption over the cycle plus a reserve, which for lettuce means 4-8 liters per plant at filling.
Signs That a Kratky System Is Working
- The level falls steadily, the air gap grows, and there is still solution left at harvest
- Upper roots are white and densely branched, lower roots are white in the solution, with no slime and no smell
- The medium in the net pot stays moist for the whole cycle although you never watered it
- No algal film on the solution surface, no gaps around the lid
If the level barely moves under normal light and temperature, the plant is not drinking, and the cause is in the root rather than in the recipe. What a root that has stopped working looks like is covered in the article on root rot.
If you want to go deeper: DWC (Deep Water Culture): Oxygen, Temperature, and Common Failures - there the root also sits in water for the whole cycle, but an aerator supplies the oxygen, and comparing the two approaches shows exactly what lets Kratky work without electricity.