There is a difference between a cutting that roots in water and a plant that lives in water, and most articles blur it. A pothos cutting will throw roots in a glass in ten days — that is propagation, and everyone knows it works. The question that actually matters is what happens in month eight, when the initial reserves are gone and the jar is either a thriving plant or a stick in green sludge.

I have run about twenty species through permanent water culture over four years. Eleven of them are still alive. Here is what separates the two groups.

Why Some Plants Live in Water and Others Drown

Roots need oxygen. In soil they get it from air pockets between particles; that is the entire reason drainage matters and the entire reason overwatering kills. Submerge a normal soil-grown root system and it suffocates within days.

But plants can grow a different kind of root. Water roots are structurally distinct from soil roots — thinner walls, more aerenchyma (internal air channels), less root hair development. They are adapted to extract dissolved oxygen directly from the water column. Species that readily produce water roots live in jars indefinitely. Species that cannot only ever sit in water and slowly rot.

This is why you cannot simply wash the soil off a mature plant and move it to a vase. The existing roots are the wrong type. You either start from a cutting that grows water roots from scratch, or you accept a die-back-and-regrow transition that many plants do not survive.

It is worth being straight about something here, because it looks like a contradiction. University extension services generally discourage rooting houseplants in water — Illinois Extension says plainly that it does not recommend the practice, and notes that roots which develop in water are often weak and do not adapt as easily to potting mix as those formed in a rooting medium. That is correct, and it does not contradict this article. Their concern is the handoff: water roots are built for a medium that never dries, so when you pot them into soil they frequently collapse and the plant stalls or dies. That warning is about water as a waypoint. This article is about water as the destination. If the plant never leaves the jar, the weakness that worries them never gets tested — which is precisely why permanent water culture works for species that would struggle with the usual water-then-soil route.

Dissolved oxygen is also why the shape of the vessel matters more than its size. A wide, shallow vessel has more surface area for gas exchange than a tall narrow one of the same volume. If you have watched a jar go stagnant and cloudy, geometry was probably half the cause.

The Eleven That Work

Plant Botanical name Speed Notes
Pothos Epipremnum aureum Fast The benchmark. Thrives for years
Spider plant Chlorophytum comosum Fast Plantlets root even faster than the parent
Lucky bamboo Dracaena sanderiana Slow Sold this way; genuinely permanent
Arrowhead vine Syngonium podophyllum Fast Colour holds well in water
Wandering dude Tradescantia zebrina Very fast Roots in under a week, needs frequent trims
Chinese evergreen Aglaonema spp. Moderate Surprisingly durable long-term
Heartleaf philodendron Philodendron hederaceum Fast Equal to pothos in reliability
Peace lily Spathiphyllum Moderate Will flower in water with feeding
Coleus Coleus scutellarioides Very fast Short-lived but spectacular colour
English ivy Hedera helix Moderate Prone to algae; use opaque vessel
Dracaena Dracaena spp. Slow Corn plant and marginata both fine

Pothos remains the one I recommend first. If you want the species-specific version of that, growing pothos in water goes deeper into long-term jar maintenance than I can here.

The Three That Always Fail

Worth naming explicitly, because they are constantly recommended by people who stopped observing at week six.

Monstera deliciosa. Roots readily, looks great for three months, then declines. The mature plant's oxygen demand outruns what a still jar delivers, and the large leaves transpire more than passive uptake supports. It works in semi-hydro with an air gap; it does not work in a vase.

ZZ plant. The rhizome rots. It roots from a leaf cutting in water — this is a real and well-documented propagation method — but the resulting rhizome needs a solid medium to bulk up in. Left in water it stalls, then softens.

Succulents and cacti, all of them. Some will produce roots in water out of sheer stubbornness. None will live there. Their entire physiology is built around drought cycling. Do not let a rooted echeveria cutting convince you otherwise.

Nutrients: The Step Everyone Skips

Plain tap water contains trace minerals and essentially no nitrogen, phosphorus or potassium. A cutting coasts on stored reserves for roughly two to four months. After that, without feeding, you get the classic water-culture look: alive, pale, and producing progressively smaller leaves.

The dose I use: a standard hydroponic nutrient at one-quarter of the label's strength, once a month. Quarter strength is the important part. Full-strength solution in a still jar burns water roots, because there is no substrate to buffer concentration and no drainage to flush excess.

A practical schedule:

  1. Every two weeks: dump and refill with fresh room-temperature water.
  2. Every four weeks: make that refill a quarter-strength nutrient solution.
  3. Every three months: wash the vessel with hot soapy water and rinse thoroughly. Biofilm accumulates on glass and hosts the bacteria that cause root rot.

If you are unsure which product to buy, the A-B-C nutrient breakdown explains what the numbering on the bottles means. For houseplants you want a general-purpose formula, not a bloom formula.

Tap water is usually fine. Chlorine dissipates and rarely causes problems at municipal levels. Chloramine does not dissipate, and if your utility uses it, leaving water out overnight achieves nothing — a carbon filter does. Softened water is the real problem: ion-exchange softeners replace calcium with sodium, and sodium accumulates in a closed vessel to levels that damage roots.

Vessel and Light

Opaque beats clear. Clear glass looks better and grows algae. Algae competes for nutrients and dissolved oxygen, and once established in a jar it is difficult to remove without disturbing roots. If you want to see the roots — which is most of the appeal — accept that you will be cleaning glass, or compromise with coloured glass that admits less light to the water column.

Fill to cover roots only. Submerging the stem base is the single most common failure. Water should cover the root mass and stop there. Leaves and stem nodes above the waterline stay above it.

Light needs are unchanged. Water culture does not alter what the foliage requires. A pothos still wants bright indirect light. What does change is temperature: a jar of water on a cold windowsill in winter runs several degrees below room air, and cold roots stop taking up nutrients even when the leaves look fine.

For a multi-jar display, a water propagation jar garden covers the build side, and a tabletop water garden scales the same idea into a single larger vessel. If you would rather have the tidiness of a pot with the low-maintenance of water, converting to leca semi-hydro is the middle path — clay pebbles give the air gap that pure water lacks.

Edibles Are a Different Question

Regrowing kitchen scraps in water is adjacent but distinct: those are short-term regrowth projects, not permanent culture. Regrowing green onions in water works well and takes ten days, but the plant exhausts itself after two or three cuts because there is no nutrient input to sustain it. That is a feature for a windowsill, not a failure.

Frequently Asked Questions

Q: Can a houseplant really live in water forever, or does it eventually need soil?

A: It can live in water indefinitely, provided you feed it. Pothos, spider plant and lucky bamboo routinely survive for many years in water culture. What kills long-term water plants is not the absence of soil but the absence of nutrients — plain water runs out of nitrogen, phosphorus and potassium after two to four months.

Q: How often should I change the water?

A: Every two weeks for a full change, topping up in between as it evaporates. Every fourth change, use a quarter-strength hydroponic nutrient solution instead of plain water. Wash the vessel itself with hot soapy water every three months to remove the biofilm that harbours rot bacteria.

Q: Why are my water-grown plant's leaves turning yellow?

A: Almost always nutrient depletion, if the plant has been in the same water for more than two months. Yellowing that starts on the oldest, lowest leaves is a classic nitrogen shortage. If the yellowing is accompanied by brown mushy roots and a bad smell, it is rot instead, usually from stem submersion or a stagnant vessel.

Q: Can I move a soil-grown plant straight into water?

A: Not reliably. Soil roots and water roots are structurally different, and existing soil roots usually die when submerged. The success rate is much higher if you take a fresh cutting and let it grow water roots from the start. If you must convert a whole plant, wash all soil off, expect significant root die-back, and keep the water scrupulously clean during the transition.

Q: Do I need an air pump or airstone for houseplants in water?

A: No, not for a jar. Passive gas exchange at the water surface is sufficient for foliage houseplants at this scale, which is why wide vessels outperform tall narrow ones. Air pumps become worthwhile for larger deep-water culture systems and for high-demand plants like monstera, which is precisely why monstera fails in a still vase.

Permanent water culture is not a compromise version of growing in soil — it is a different system with its own failure mode, and that failure mode is nutrient starvation rather than root rot. Pick a species that grows real water roots, use a wide opaque vessel, cover the roots and not the stem, and feed at quarter strength once a month. Do that and a pothos in a jar will outlive most of the potted plants on the same shelf.