Glazed vs Unglazed Terracotta: Why Glaze Ruins an Olla

01 · THE SHORT ANSWER

Glazed vs unglazed terracotta ollas: why glaze ruins it

An unglazed terracotta olla releases water through its wall. A glazed one does not. That single sentence is the whole debate. Everything else — shape, capacity, lid design, neck height, wall thickness — changes how fast or how evenly an olla delivers water. Glaze changes whether it delivers water at all^[1][3].

This matters commercially because “olla” has become a marketing word. Search for one and you will find beautifully glossy vessels in cobalt, celadon, and speckled cream, sold as garden watering pots. Many of them are lovely. A good number of them are, functionally, buried vases. The gloss you are looking at is the exact thing that stops the pot from working.

If you want the broader introduction to the method before going deep on materials, start with what is an olla — the complete guide to self-watering clay pots. This article is specifically about the wall.

02 · THE PHYSICS

How an unglazed olla actually moves water

Fired earthenware is a sponge made of rock. During firing, the clay platelets sinter together but do not fully melt, leaving a network of open pores — typically somewhere in the range of 10–25% of the body’s volume in a low-fired terracotta^[4]. Those pores connect. Water inside the pot wicks into them by capillary action, saturating the wall, and reaches the outer surface.

What happens next is the elegant part, and it is why an olla is not simply a slow leak. Once the outer surface is wet, further outflow is controlled by soil matric potential — the suction that dry soil exerts on water^[6]. Dry soil pulls hard, and the olla gives up water quickly. Wet soil pulls weakly, the gradient across the wall collapses, and outflow slows to almost nothing. The plant’s own water use, which dries the soil around the roots, is what opens the tap. This is demand-driven irrigation with no sensors, no timer, and no moving parts^[2][5].

Two consequences follow. First, an olla buried in soil and left alone will not empty itself into saturated ground — it self-regulates. Second, and critically for this article, the whole mechanism depends on an unbroken capillary path from the water inside to the soil outside. Interrupt that path anywhere and the device stops. A glaze layer is a total interruption.

Field and greenhouse work on buried clay pot irrigation consistently finds very high application efficiency compared with surface watering, precisely because water is delivered subsurface and on demand rather than sprayed across an evaporating surface^[2][11]. We covered the numbers in how clay ollas save 70% water vs surface irrigation. None of that efficiency is available from a sealed pot.

03 · WHAT GLAZE IS

Glaze is glass, not paint

People often picture glaze as a coloured coating sitting on top of the clay, like enamel on metal. It is closer to a chemical weld. A ceramic glaze is a suspension of silica, a flux such as sodium, potassium, calcium or lead oxide, and an alumina stabiliser. In the kiln the flux lowers silica’s melting point far enough that the mixture melts into a true glass and fuses with the clay surface beneath it^[4].

Glass is amorphous and non-porous. That is the entire historical point of glazing: it was developed so earthenware could hold liquid, wine, oil, water, without sweating it away or absorbing flavours, and so vessels could be wiped clean^[4]. Every desirable property of glaze on a dinner plate is an undesirable property on an olla. The technology was invented to defeat exactly the behaviour the olla depends on.

Nor is the effect partial in any useful way. A glaze does not slow seepage to a gentler rate; it stops the capillary path. A crazed glaze — one with a fine network of cracks from thermal mismatch — may pass a trickle of water through the cracks, but that gives you a few random point leaks rather than an evenly wetted wall, which defeats the uniform wetting front an olla is prized for. Buy a crazed glazed pot and you get unpredictable dribbling, not irrigation.

04 · PARTIAL GLAZING

When a glazed neck is fine — and even helpful

The rule is not “never glaze”. The rule is “never glaze the buried body”. Anything above the soil line contributes nothing to soil delivery, so glazing there costs you no performance and buys you a few real advantages.

• Less algae in the neck. The exposed neck of an olla gets light and holds water, which is all algae needs. A smooth glazed interior neck gives algae far less to grip and wipes clean with a cloth in seconds, where a porous neck stains green and needs scrubbing.
• Less evaporation from the exposed collar. A porous collar standing proud of the soil wicks water and evaporates it straight to the air — a small but genuinely wasted loss, worst in hot, windy weather. Glazing the collar keeps that water inside for the roots.
• A cleaner, more durable rim. The rim is the part you handle, knock with a watering can, and seat a lid onto. Glaze resists chipping and staining there, and a glazed rim seats a lid more snugly, which matters for keeping mosquitoes and debris out.
• No effect on the wetting front. Water leaves an olla from the buried body, and the wetting bulb forms around that surface. A glazed collar sitting in air changes none of that geometry. Performance is unchanged.

What to check when you see a partially glazed pot: how far down does the glaze go, and how deep do you intend to bury it? Glaze that stops one inch below the rim is fine. Glaze that runs a third of the way down the shoulder cuts a meaningful slice off your seeping surface if that shoulder ends up underground. Measure before you buy, and if in doubt bury the pot so that the glaze line sits at or just above soil level.

05 · FIRING TEMPERATURE

The hidden failure: over-firing and vitrification

Unglazed is necessary but not sufficient. A pot can be entirely free of glaze and still barely seep, because porosity also depends on how hot and how long the clay was fired.

As firing temperature climbs, fluxes within the clay body begin to melt and flow into the pore spaces — a process called vitrification. Earthenware fired in the roughly 950–1100 °C band keeps a high open porosity and stays absorbent. Push the same body toward stoneware temperatures, around 1200 °C and above, and the glassy phase fills the pore network from the inside^[4]. The result looks matte and unglazed, rings with a hard high note when tapped, and passes almost no water. It is a self-glazed pot.

This is why an olla is a genuinely engineered object rather than any old flowerpot. Clay body, grog content, wall thickness and firing schedule all have to land in a window that gives durable, frost-resistant, handleable ceramic while preserving the open pore structure^[5]. Too soft and it crumbles or fails in freeze–thaw; too hard and it stops working. We go through the material specification in detail in what makes a quality olla — clay, build, and why it matters.

Two quick field cues. A porous earthenware body sounds dull and low when flicked with a fingernail; a vitrified body rings. And a porous body, dipped briefly in water, darkens visibly within seconds as it drinks. A vitrified one shrugs the water off in beads.

06 · THE HOME TEST

How to test olla porosity at home in 30 minutes

You do not need a lab. Porosity announces itself. Do this before you bury anything.

Step 1 — Start dry. The pot must be bone dry, inside and out. A pot that is already damp from storage will not show a contrast. Leave it somewhere warm for a day if you are unsure.

Step 2 — Fill it and stand it in a dry sink or on newspaper. Fill to the neck with plain water. Do not stand it in a saucer of water, which will wet the outside from below and confuse the result.

Step 3 — Watch the outer surface for 15–30 minutes. A properly porous olla darkens from the bottom upward as the wall saturates. Within roughly half an hour the whole body below the water line should look wet and noticeably deeper in colour, and by an hour or two you will usually see fine beads of moisture standing on the surface, and often a damp ring on the newspaper. That is a working wall.

Step 4 — Read the result honestly. If the outside is still dry and the same colour after an hour, the pot is glazed, sealed, or vitrified. It will not irrigate. If only patches darken, you have partial sealing or a very uneven body — expect uneven delivery. If the pot visibly drips or empties in an hour or two, it is too coarse or cracked; check for a hairline fracture.

Step 5 — Weigh it, if you want a number. Weigh the pot dry, soak it in water for 24 hours, wipe the surface and weigh again. The percentage weight gain is roughly its open porosity by mass. Earthenware suitable for olla use commonly lands in the mid-teens or higher; a vitrified body will barely register^[4].

Bear in mind that a brand-new olla often seeps faster in its first days than it will later, as the wall wets through and fine particles settle into the pores. That settling is normal. What is not normal is a wall that never wets at all.

07 · DECORATIVE FAKES

Why decorative glazed pots sold as “ollas” fail

The failure mode is quietly frustrating, which is why it is worth naming. A glazed pot sold as an olla does not announce that it is broken. It sits in the bed looking handsome. The water level goes down slowly — from evaporation at the neck, not seepage — which reads as “working, just slowly”. Meanwhile the plants around it are living entirely off rainfall, and the gardener concludes that ollas are overrated.

Several things push glazed pots into the market. Glaze photographs beautifully, and product listings are won on photos. Glazing hides variation in the clay body, so a factory can use cheaper mixed clay. A glazed pot does not develop the mineral bloom and algae staining that an honest working olla acquires, so it looks new for longer in a showroom. And “olla” is not a protected term — anyone can print it on a label.

Some related near-misses to watch for. A sealed or waterproofed planter, sold with a food-safe or waterproof liner coating, has exactly the same problem as glaze with none of the shine to warn you. A painted terracotta pot is often sealed too, since paint plus sealer is standard for outdoor decorative terracotta. And a plain drainage-hole flowerpot inverted or plugged is a common DIY substitute; it will seep, but garden centre pots vary wildly in firing and wall thickness, so results are a lottery.

One more thing to expect on a genuine olla: a whitish crust may develop on the outer wall over a season or two. That is carbonate scale from dissolved minerals in hard tap water, left behind as water evaporates and seeps. It is proof of function, not a defect, and it can be reduced by using rainwater and cleared by scrubbing or a soak in dilute vinegar^[13].

08 · CHOOSING WELL

What to look for when buying an olla

Buy on the wall, not the look. Ask or check for four things: an unglazed body below the soil line; a clay body fired to earthenware, not stoneware, temperatures; a consistent wall thickness, since thin patches seep fast and thick patches barely at all; and a lid, because an open olla neck breeds mosquitoes, collects leaf litter, and loses water to evaporation.

Beyond the material, sizing and refilling are the practical variables that decide whether the system fits your garden. Capacity relative to bed size sets your refill interval — see how often should you refill an olla for the working numbers.

Finally, treat the surface finish as a diagnostic. An olla that has been in the ground a season should look used: darker where it has been wet, perhaps faintly mineral-crusted, with soil clinging to the wall when you lift it. A pot that comes out of the ground looking exactly as it went in has not been doing anything.