Layering Theory & Practice

Layered glazes produce effects impossible with a single coat because the boundary between two different glaze compositions creates a reaction zone with its own chemistry. Understanding the mechanisms behind these interactions — eutectic melting, viscosity differences, surface tension, and colorant chemistry — is the key to controlling layered results rather than leaving them to chance.

Every glaze has a temperature range at which it softens and flows, but when two different glazes sit one atop the other, the contact zone between them can begin melting at a temperature lower than either glaze would melt on its own. The reason is eutectic behavior: certain oxide ratios melt at a single, sharply defined temperature that falls below the melting point of any component in the mixture. Pure silica melts at 3110 °F (1710 °C) and pure alumina at 3722 °F (2050 °C), yet a mixture of roughly 10% alumina and 90% silica melts at only 2813 °F (1545 °C) — hundreds of degrees below either endpoint.^[1][2]

In practical layering, the interface between two glazes is where eutectic chemistry matters most. As oxides from the upper and lower layers inter-diffuse during firing, the boundary zone creates new oxide ratios that neither glaze contains on its own. If those ratios happen to approach a eutectic composition, the interface melts earlier and flows more than the surrounding single-glaze areas. This is why overlap zones so often run farther than the individual glazes beside them, and why kiln cookies or catch plates are non-negotiable when testing layered combinations.^[3][4][2]

The eutectic concept also explains certain visual effects. When a eutectic mixture cools, all its components remain molten together until they solidify simultaneously, forming a clear, transparent glass. If the melt also contains non-eutectic ingredients, those can solidify separately while the surrounding glass is still liquid, leaving tiny crystals suspended in the matrix. The result is opacity, opalescence, or crystalline texture — effects potters prize in reactive overlap zones.^[1]

Digitalfire notes that eutectics have limited direct utility in everyday glaze formulation because commercial frits are already pre-melted glasses with broad softening ranges, but the concept remains critical for understanding why layered glazes interact the way they do. The effect is strongest when both glazes are already close to maturity at the target cone; in underfired systems where neither layer develops much melt, eutectic enhancement is less dramatic.^[2][4]

Melt fluidity — the degree to which a glaze flows when molten — is governed by the ratio of fluxes (boron, lithium, sodium, potassium) to refractories (alumina, silica). High-flux glazes run; high-alumina, high-silica glazes stay put. But chemistry alone cannot predict behavior: fritted materials produce more fluid melts than raw minerals sourcing identical oxides, and finer particle sizes dissolve faster, increasing flow even at the same oxide ratio.^[5]

When two glazes of different fluidities are layered, four broad scenarios arise:

• Two fluid layers — the combination must be kept very thin or the overlap will run off vertical surfaces.
• Fluid base under a non-fluid overlay — the weight of the stiffer top coat can be pulled downward by the mobile layer beneath.
• Non-fluid base under a fluid overlay— the safest combination, because the stable base anchors the system while the fluid top melts into it, creating streaks, color gradients, and the “icicle” drip effects potters seek.
• Two non-fluid layers — largely defeats the purpose of layering, since the glazes will not interact enough to produce new visual effects.

Scenario 3 — stiff base plus fluid top — is the workhorse strategy for decorative layering and underpins the design of several commercial systems. Coyote's Texas Two-Step line, for example, pairs non-fluid Step One Undercoats with fluid Step Two Overcoats specifically to produce oilspot, cascade, and mottled effects at cone 5–6 without excessive running.^[4][6]

Two standard bench tests reveal melt fluidity before a glaze goes on ware. The Melt Flow Runway Test (Digitalfire code GLFL) races two glazes down an inclined ceramic runway, showing comparative viscosity and surface tension at a glance. The Ball Test (GBMF) fires a 10-gram dried glaze sphere on porcelain, revealing flow distance, bubble behavior, surface tension characteristics, and transparency. Either test is invaluable for predicting how two glazes will behave when stacked.^[5]

Opacity in ceramic glazes arises from five distinct mechanisms:

• Dispersion of refractory micro-particles such as zircon or tin oxide that reflect and refract light.
• Crystallization of opaque phases during cooling, particularly from high-CaO melts.
• Matte surfaces whose non-flat topology scatters light.
• Phase separation — microscopic islands of differing glass composition within the melt.
• Suspended micro-bubbles that scatter light.

The practical consequence for layering is that transparent and opaque glazes interact with light in fundamentally different ways. In a transparent glaze, light travels through the glass layer, reflects off the clay body or an underlying glaze, and returns to the eye colored by the colorants it passed through twice. Thickness controls intensity: thicker pools appear darker, thin edges appear lighter. When such a glaze is opacified, the light-scattering particles block the round-trip path, subduing color — “a deep rich blue can turn into a dull pastel blue when the glaze is opacified.” Ryan Coppage notes that an opaque glaze may need roughly five to ten times more metallic oxide colorant than a translucent glaze to reach the same apparent color intensity.^[7][8]

The choice of opacifier matters for layering chemistry. Tin oxide requires only about 7% for complete whiteness and produces bluer whites; zircon (Zircopax) requires 10–20% and tends toward yellower whites while also stiffening the melt and hardening the surface, potentially causing blistering or pinholing at higher concentrations. Critically, tin-opacified glazes are susceptible to chrome flashing (turning pink near chrome-bearing glazes in the kiln), while zircon-opacified glazes are immune to this reaction.^[7]

Transparent overcoats are not automatically safe in complex surfaces. Digitalfire notes that transparent glazes applied too thickly often cloud and can obscure underglaze decoration, because excess thickness changes optical depth and can encourage micro-bubble retention that reduces clarity. A transparent top coat should usually be thinner and more carefully controlled than an opaque one.^[9][10]

Partially opacified glazes — termed “milky” — occupy a middle ground and can produce striking depth effects in layered systems. The majolica tradition exploits this principle directly: an opaque white tin glaze provides a reflective canvas for transparent colored overglaze brushwork, where the translucent colors gain brilliance from the white backdrop beneath.^[7]

Phase separation occurs when a molten glaze separates at the microscopic scale into two or more glassy phases with different compositions. Certain oxide combinations lack full compatibility in silicate melts, and the resulting internal interfaces between phases scatter light, producing milkiness, haze, or opalescence even when no crystals are present. Colorants concentrate preferentially in one phase, creating soft mottling and color variation. Increased alumina helps suppress separation, while high boron promotes it.^[11]

This mechanism is directly relevant to layered glazes because when two glazes create a new composition at their interface, phase separation may occur where it would not in either glaze alone. The sky-blue opalescence of Chinese Jun (Chun) ware, for example, is produced by liquid-liquid phase separation forming droplets 1–100 nm in diameter. These droplets create a short-range ordered structure that acts as an amorphous photonic crystal, producing structural color through coherent light scattering — the visual effect combines chemical color from iron ions with this structural color.^[12][13]

A related unwanted effect is “boron blue.” When boron levels are excessively high — especially alongside ample SiO₂ and Al₂O₃ — boron forms crystal phases that turn transparent glazes milky or cloudy blue. This can appear unexpectedly when two boron-rich glazes are layered together, concentrating boron at the interface beyond the threshold either glaze would reach alone.^[7]

Surface tension is the thermodynamic tendency of a liquid to minimize its surface area. In glazes, low surface tension promotes wetting and spreading across the clay body; high surface tension resists wetting and causes the melt to contract, contributing to edge pull-back, beading, crawling, and incomplete coverage. Surface tension and viscosity are distinct properties but interact: a glaze can have high surface tension yet level well if its viscosity is low enough.^[14]

Specific oxides shift surface tension predictably:

This matters in layering: a high-surface-tension glaze brushed over a low-surface-tension glaze may bead up or crawl rather than smoothly coating the layer beneath.^[14]

Surface tension also governs bubble behavior. Low surface tension enhances bubble merging and rupture, while high surface tension strengthens bubble films, allowing them to persist or remain trapped — a key factor in blistering, pinholing, and the orange-peel texture that occurs when trapped bubbles shrink during cooling. In layered systems, bubble trapping is exacerbated when an early-melting top layer seals over a still-degassing bottom layer, a common source of pinholing in two-glaze combinations.^[14]

Crawling — where molten glaze withdraws from the ceramic body, leaving bare patches — is driven by high surface tension combined with poor adhesion from thick application, dusty bisque surfaces, or excessive shrinkage during drying. In layered work, applying a high-shrinkage glaze over a non-gummed base coat dramatically increases crawling risk. Adding CMC gum (roughly 1%) to the base coat dramatically improves adhesion and crawling resistance.^[14][4]

Chrome-Tin Reactions: Chrome-bearing glazes can volatilize enough chromium oxide at peak temperature to turn nearby tin-opacified whites pink, even when the chrome and tin are not mixed in the same bucket. The mechanism is kiln atmosphere transport: chrome vapor leaves one glaze and deposits on another receptive, tin-bearing surface. This is as much a kiln-loading caveat as a layering one — simply placing a chrome green piece next to a tin white piece in the same kiln can produce unwanted pink flashing on the white. Adequate kiln ventilation helps but does not eliminate the risk; the safest approach is to keep chrome greens and tin whites apart. Zircon-opacified glazes are immune.^[15][16]

For intentional chrome-tin pinks, only 0.1–0.5% chrome oxide with 5% tin oxide is needed, but the system is chemically demanding. Calcium (CaO) must be present in the receiving glaze or the pink color is completely absent. Zinc must be excluded or the color turns brown. The pink is stable in oxidation up to about 1250 °C but fades severely in reduction or with overfiring. Chrome-tin reds do not work well above cone 8. Substituting calcium with a magnesium/strontium oxide blend makes the glaze chemistry hostile to chrome-tin pink formation — the documented Pfaltzgraff factory solution when they needed chrome-safe whites. Commercial chrome-tin stains are generally more reliable than raw chrome-plus-tin layering experiments.^[17][15][18]

Chrome oxide behavior varies dramatically by glaze chemistry beyond the tin interaction. In high boron/alkaline bases with under 1% chrome, it produces bright glossy transparent greens. In zinc-containing bases, it generates browns. In high calcium/strontium bases without zinc, it creates pinks through burgundy. Chrome is amphoteric — it plays an intermediate role between fluxes and silica, sometimes acting as a flux itself. This means the same chrome-bearing glaze can produce completely different colors depending on what glaze it is layered over or under.^[18]

Copper Behavior: In oxidation firing, copper oxide normally produces green. In alkaline glazes (high Na₂O, K₂O), copper shifts toward turquoise blue, reminiscent of Egyptian faience and southwestern pottery. In reduction, copper in alkaline glazes containing tin oxide can produce copper reds, with the specific hue influenced by alumina, magnesium, and boron content. When layering a copper-bearing glaze over an alkaline base glaze, the copper at the interface may shift color due to the alkaline flux environment it encounters — an intentional layering strategy for achieving turquoise effects.^[19][20]

Iron Oxide:Iron oxide produces amber, yellow, or brown in oxidation and celadon green or tenmoku black in reduction. Beyond its role as a colorant, iron also acts as a flux, increasing melt fluidity at higher concentrations. In alkaline glazes, iron produces warmer, more amber tones; in high-calcium glazes, more classic amber-to-brown progressions. Iron's dual role as colorant and flux makes it particularly interactive in layered systems: a high-iron glaze layered over a low-iron glaze will flux the interface, increasing flow and color blending at the overlap.^[20][19]

During firing, the liquid glaze melt attacks the clay body surface, penetrating into it and forming a buffer zone of intermediate compositions. This interfacial layer develops differently by clay type: porcelain develops highly bonded interfaces where glaze and body become nearly inseparable; earthenware produces minimal glass development and weaker adhesion; stoneware falls between. Temperature, soaking time, cooling rate, and material chemistry all affect interface development.^[21][22]

The same interfacial reaction occurs between two glaze layers. Metal oxides from each glaze migrate into the other at the boundary, modifying color development, viscosity, and surface texture in the contact zone. This exchange creates a gradient composition that produces visual effects impossible to achieve with either glaze alone — the “third glaze” effect that potters observe in overlap zones, and the primary reason most potters layer glazes in the first place.^[21][23]

A credit-card thickness (roughly 0.75–1 mm) is the general standard for a single glaze application, with stoneware glazes typically targeting about 0.5 mm fired thickness. In layered systems, the combined thickness of both layers matters more than either individual layer's thickness, and the total should generally not exceed what would be appropriate for a single glaze unless the glazes are specifically formulated for heavy application.^[9]

Transparent colored glazes demonstrate intentional color variation through thickness — darker where pooled thickly, lighter at thin edges. Reactive glazes that variegate or crystallize usually require a specific thickness range to develop their characteristic effects. Too-thin application results in washed-out color and incomplete coverage; too-thick application causes blistering, crawling, running, and drying cracks. In layered work, a thin under-layer allows the top glaze to dominate, while a thicker under-layer creates more dramatic interaction effects.^[9][23]

Excessive combined thickness is the most common cause of glazes running off pieces in layered work. Thickly applied glazes assert their shrinkage during drying, compromising the bond with the body below; the cracks that appear become bare patches (crawling) after firing. This is especially problematic when the second coat rewets and adds shrinkage stress to the first. Adding CMC gum to the base coat and allowing complete drying before the second application are the primary solutions.^[9][4]

With the science of glaze interaction established, this section turns to practical strategies for applying multiple glazes — from the foundational base-plus-top-coat method to advanced three-layer combinations, gravity management, and commercial systems designed for layering.

The most reliable approach to layered glazing pairs a stable base coat with a more reactive or fluid top coat. The base anchors the system to the clay body while the top melts into and interacts with it, creating the color blending, mottling, and flow effects that make layering worthwhile.^[4][24]

For brushing — the primary application method for most commercial glaze users — the base coat should be built up in two to three even coats, alternating brush stroke direction with each pass (first horizontal, second diagonal, third vertical) to eliminate streak marks and even out thickness. Apply one coat, let the wet look disappear, then apply the next coat perpendicular to the previous.^[25][26]

Traditional dipping glazes (powdered minerals with 15–25% clay for suspension) develop fragile bonds with bisque ware that fail when a second coat is applied — the rewetting and shrinkage of the second coat pulls the first layer away from the body. Adding CMC gum (approximately 1% as a starting point) creates functional base coats that tolerate overlapping applications, though it trades off slower drying and reduced thixotropic properties. Commercial brushing glazes already contain higher gum percentages, which is why they tolerate layering better than studio dipping glazes.^[4][27]

Glazes intended for layering should also pass recipe sanity checks. Excessive clay content causes extreme shrinkage; high bentonite percentages (7%+) create excessive drying stress; high zinc oxide combined with heavy clay causes peeling. Many studio layering failures trace to poor raw glaze properties rather than fired chemistry incompatibility.^[4]

For subsequent layers beyond the first, use lower specific gravity slurry, submerge for shorter durations when dipping, or switch to brushing or spraying for thinner coats. Each subsequent coat takes longer to dry because the previous coat has saturated the bisqueware's porous surface, reducing its absorption rate.^[3][25]

When testing a new layered combination for the first time, the Ceramic Arts Network advises limiting the second layer to the top one-third of the pot initially, to observe flow behavior before committing to full coverage. A hydrometer for checking specific gravity is especially useful when building three-layer surfaces, where slight differences in consistency compound across layers.^[3][28]

The standard recommendation for brushing is to allow each glaze layer to dry to a chalky, matte appearance before applying the next coat. Digitalfire specifically advises allowing “a whole day” for complete drying. Wet-on-wet brushing results in uneven thickness, poor adhesion, and the second coat disturbing or lifting the first. Gum in the base coat expands options by giving the raw glaze more green strength and adhesion, but a wet, unbound layer is easily disturbed by the next brush pass.^[4][27][25]

However, extended drying can also cause problems. If the first coat dries too long and becomes dusty, the second coat may not adhere well. Timing is a balance.^[4]

For dipping (less common for the commercial-glaze audience but sometimes used for base coats), the timing rules differ. When triple-glazing by dipping, the second and third glazes should be dipped as soon as the pot is dry enough to accept more glaze — not after extended drying. Long gaps between dips can cause adhesion problems. The technique involves progressively shorter immersion times: approximately 2 seconds for the first dip, 1 second for the second. The key distinction is that dipping deposits glaze quickly without brush friction to disturb the layer below.^[28]

Triple glazing involves applying three distinct glaze layers sequentially, typically with graduated coverage: the first layer covers 100% of the exterior, the second covers approximately 80%, and the third covers approximately 65%. This creates zones of single, double, and triple glaze interaction on the same piece, producing “fluid, complex colors that would generously accentuate significantly altered forms.” Thinner-than-usual glaze consistency is essential, monitored with a hydrometer for specific gravity.^[28]

A documented successful three-layer combination: (1) entire pot in Odyssey White Gloss, (2) top third in Chun Celadon, (3) rim only in Strontium Crystal Magic. This demonstrates strategic placement — confining reactive glazes to the upper portion where gravity aids rather than hinders their interaction. The article advises confining third and fourth glazes to the top quarter or rim when first testing, to limit potential running.^[3]

For three or more layers, keep each layer thin to control flow. Overlap zones create new colors via mixing, and gravity causes fluid upper layers to pull downward on verticals. Always test on vertical test tiles before committing to ware.^[4][29]

Application patterns should generally be designed so that no part of the piece has more than two glaze thicknesses, unless specifically testing three-layer interactions. Overlap zones can be straight lines (from dipping), irregular shapes (from pouring, spraying, or brushing), or diagonal patterns (from dipping at an angle). Each creates different visual movement across the form.^[23][3][30]

One sophisticated gradient technique layers a dark (near-black) glaze as undercoat, then a colored glaze on top, then in selected areas a transparent high-fusibility glaze whose melting increases the fluidity of all lower layers, allowing the dark undercoat to bleed through and darken the color above. The transparent glaze acts as a localized flux, creating gradient effects that cannot be achieved any other way.^[23]

Wax resist is a practical tool for controlling overlap zones. Apply wax to any area where a second glaze should not go — rims, foot rings, or specific design boundaries. Wax burns away in the kiln but prevents glaze adhesion during application.^[4]

Vertical surfaces emphasize dripping “icicles” and downward flow, while horizontal surfaces form bubbly, frosting-like textures. Glazes formulated for vertical surfaces must be more viscous than tile glazes designed for horizontal application. Gravity creates much of the primary visual interest in layered glazes — the fluid top coat flows over and through the stiffer base, creating streaks, drips, and color gradients that follow the form's contours. Kari Radasch specifically watches “how glazes move, melt and flow depending upon their mass, temperature and location on the pot.” ^[31][5]

The safest strategy for vertical work is placing the most fluid glaze at the top of the form, where it has the least distance to travel before reaching the foot. If multiple layers all have high fluidity, each must be thin enough that the final combined thickness will not run excessively.^[4][3]

When glazing vessels, the interior should be glazed first and allowed to dry completely before glazing the exterior. This maintains adequate bisque absorbency for the exterior coat — if the exterior is glazed first, moisture saturates the wall, reducing absorption capacity for the interior application.^[9]

On interior surfaces, running glazes pool at the bottom of the vessel rather than dripping onto kiln shelves, making interiors more forgiving for fluid layered glazes. However, excessive pooling can create thick glass accumulations that crack during cooling from thermal expansion mismatch. On exterior surfaces, running fuses the piece to the kiln shelf, requiring kiln cookies and careful thickness control. A common strategy is to use more adventurous layering on interiors and more conservative combinations on exteriors.^[4][9]

Food safety requires special attention with layered glazes. Commercial glazes labeled food-safe are tested in isolation. When layered with other glazes, a glaze that resists acid attack on its own may be “destabilized enough to leach metals.” Bright-colored glazes containing copper, cobalt, manganese, or chrome are particularly risky on food-contact surfaces when combined with other glazes. The recommendation is to use non-colored glazes on functional ware food surfaces and understand the complete recipe chemistry. Hesselberth and Roy's Mastering Cone 6 Glazes (2002) remains the definitive reference on cone 6 durability and leaching.^[4]

At low fire (cone 06 to cone 1), the glaze layer remains distinctly separate from the clay body with minimal interface development. Low-fire glazes offer the brightest and most varied color palette but can appear harsh. Layering at low fire produces more predictable results because lower temperatures allow less oxide migration between layers. Mayco Stroke & Coat and AMACO LUG (Low-fire Underglaze) are designed specifically for low-fire layering, and majolica is the classic low-fire layering tradition.^[32][2]

Mayco Stroke & Coat demonstrates how coat count changes results at low fire: one coat creates a translucent finish, two to three coats produce full opacity, and when applied over a matte glaze, one coat fires matte while two to three coats fire glossy. Note that chrome-tin pigmented Stroke & Coat colors (SC013 Grapel, SC033 Fruit of the Vine, SC085 Orkid) can develop a milky or cloudy haze when clear glazes are applied over them — a specific chemical incompatibility to watch for.^[33]

Mid-fire (cone 4 to cone 6) is the most popular range for layering experimentation because commercial glaze systems are specifically designed with layering in mind. AMACO Potter's Choice and Potter's Choice Flux (PCF) glazes are formulated for layering at cone 5/6, with an online layering tool that lets potters select top and bottom glazes and view combination results. Coyote's Texas Two-Step system and Shino series are designed for cone 5–6 layering, pairing stable undercoats with fluid overcoats. Mayco's stoneware line also includes recommended layering combinations.^[32][34][6]

At high fire (cone 8 to cone 12), a substantial body-glaze interfacial layer forms, and the body becomes dense and vitrified. The color palette narrows because extreme heat causes many colorants to burn out or shift dramatically — chrome-tin reds, for example, do not survive above cone 8. However, high-fire layering produces the most integrated, complex surfaces because greater heat allows more thorough mixing of layered glazes. Iron, cobalt, and rutile remain reliable colorants at high fire. John Britt's The Complete Guide to High-Fire Glazes (2007) covers cone 10 layering extensively.^[32][18]

A grid of glaze combinations should treat A-over-B as different from B-over-A. The lower coat controls absorption and early melt contact with the clay body, while the upper coat controls the exposed surface chemistry and runoff path. In practice, a combination matrix should include both orders rather than assuming the result is reversible. This is especially important when the two glazes differ strongly in opacity, fluidity, or gum content.^[35][36]

Commercial systems recognize this directionality. AMACO's online layering tool distinguishes between “over” and “under” for every Potter's Choice combination, because the results are visually different in each direction. Coyote's Texas Two-Step documentation specifies which glazes serve as Step One (under) and which as Step Two (over), explicitly acknowledging that the reverse order produces different — and sometimes undesirable — results.^[34][6]

For brushing-focused studios, several commercial glaze lines are designed explicitly for layered application:

AMACO Potter's Choice (Cone 5–6):A line of reactive glazes formulated for layering. AMACO provides an online layering tool showing photographic results for hundreds of two-glaze combinations in both application orders. Potter's Choice Flux (PCF) glazes are designed as fluid top coats over standard Potter's Choice bases.^[34]

Coyote Texas Two-Step (Cone 5–6): A system of non-fluid Step One Undercoats paired with fluid Step Two Overcoats, producing oilspot, cascade, and variegated effects. The system is designed so the base remains stable while the top coat creates the visual action. Coyote also offers a Shino series with layering recommendations.^[6]

Mayco Stroke & Coat (Cone 06 to Cone 6): Versatile glazes that fire translucent in one coat and opaque in two to three coats. Designed for layering over underglazes and with each other. Mayco provides a Ceramics 101 guide with layering examples including antiquing and majolica-style overlaps.^[33][29]

Mayco Stoneware Glazes (Cone 5–6): Include recommended layering pairings in manufacturer documentation.^[33]

These systems take much of the guesswork out of layering for studios that do not mix their own glazes, because the manufacturer has pre-tested combinations and provides guidance on application order and thickness.

Systematic testing separates potters who get lucky once from potters who get reliable results every time. This section covers tile design, blending methods, combination grids, and the record-keeping that makes test results actually useful.

Test tiles must be made from the same clay body as finished pieces, because glaze behavior — color, texture, fit, and interaction — changes significantly between bodies due to variations in iron content, thermal expansion, porosity, and flux content. If you work with multiple clay bodies, you need separate sets of test tiles for each.^[37]

Effective test tiles include both flat areas for color and texture observation and vertical sections to assess glaze flow. A glaze that looks fine on a flat tile may run excessively on a vertical surface, and this gap is amplified in layered glazes where eutectic effects increase fluidity. L-shaped test tiles provide vertical and horizontal surfaces in one piece; small cylinders or extruded forms work equally well. John Britt recommends making test tiles that mimic actual ware — thrown, extruded, or handbuilt shapes with both vertical and horizontal surfaces — because flat tiles alone can give dangerously misleading results.^[37][38][39]

Surface texture matters too. Incorporate ridges, incised lines, or carved textures to observe how the glaze breaks over edges, pools in recesses, and changes color at different thicknesses. This is especially important for layering tests, where interactions may be most visible where glaze collects in carved details.^[37]

Produce multiple test tiles in advance — up to 100 or more for serious testing campaigns — in consistent size (typically around 2 x 2 inches for flat tiles, 2–3 inches tall for vertical forms) and thickness. Making a batch of identical tiles in advance removes a barrier to spontaneous testing. Group similar tests in the same firing for easier comparison, and place tiles in different kiln areas to study temperature variation.^[36][37]

Each test tile requires clear labeling for post-firing identification. Use oxide pencil, underglaze pencil, or iron-oxide wash to write directly on the back of tiles before bisque firing. For complex test series, create a numbering system with corresponding records in a notebook or spreadsheet. For layering tests specifically, label both the individual glazes and the layering order (which is over, which is under).^[37][36]

Line blends test two different recipes mixed in varying proportions — for example, 0% to 10% additional red iron oxide in 1% increments, yielding 11 separate tests. The increment size should match the strength of the variable being tested: cobalt is so powerful that 0.1% increments are more informative than 1% increments. If the step size is too large for a potent colorant, the test skips the useful middle ground entirely. Glazy recommends keeping the 100% endpoints in every test because each firing can shift a glaze enough that fresh controls remain useful.^[35]

The volumetric syringe method keeps small line blends consistent and reproducible. The process: (1) mix equal weights of each glaze, (2) add water and sieve thoroughly, (3) add water until both glazes have equal volume, (4) use a syringe to blend in measured increments. For a 20 ml syringe with 10% increments, approximately 110 ml of each base recipe (220 ml total) is needed. This approach is more practical than weighing dry ingredients for each intermediate cup, and because it equalizes volumes first, it ensures you are testing chemistry rather than accidentally testing water content.^[35]

Triaxial blends test three ingredients simultaneously on a triangular grid. Each corner represents 100% of one material; each side is a line blend of the two corner materials; interior points contain all three. The number of tests follows the triangular number sequence:

Two range options serve different purposes. A 0%–100% range puts simple two-variable line blends on the outer edges, useful for comprehensive exploration. A 20%–100% range ensures every test includes all three variables, better for targeted development when you already know you want all three components present. A five-row triaxial gives far more useful three-way information than a four-row one.^[35][40]

For layering theory, triaxial thinking is useful when testing base glaze, top glaze, and a third variable (opacifier, colorant, or flux) together rather than in isolated pairs. Robin Hopper's The Ceramic Spectrum (1984/2001) includes extensive use of flux variation triaxial tests and remains a foundational text for systematic color development.^[35]

For testing glaze layering combinations systematically, create a grid matrix where rows represent base glazes (underneath) and columns represent top glazes (over). Each cell represents one specific combination. For N glazes, the full grid requires N x N tests, including each glaze over itself as a thickness control. This is often called a “combination tile board.”^[41][34]

AMACO's online layering tool follows this grid principle, allowing potters to select top and bottom glazes and view photographic results. When building your own grid, start small (4–6 glazes = 16–36 combinations) before attempting larger matrices, and keep the same clay body and firing schedule across the entire grid for valid comparisons. Use systematic grids for multi-variable testing (thickness x layers x overlap), and record bisque porosity as well as firing data.^[34][39][42]

For each glaze test, document: materials used with clear labeling, application method and order of application, number of coats and thickness, kiln temperature and cone numbers, firing schedule (including ramp rates and holds), kiln position, specific gravity of the glaze, and final results including photographs. The main goal is “to have enough information to be able to repeat, or avoid, the results you discovered.” Write procedures down first to ensure follow-through, and record application sequence immediately — it is easy to forget which order glazes were applied.^[36]

Take photos of both the glazing process (unfired glaze appearance contains useful information) and fired results. Two inexpensive LED lamps on either side with tracing paper or thin fabric as simple softboxes create diffused light that removes harsh reflections on glossy glazes. A tripod improves photos more than a new camera. Capture at least three views: front, angled (to show form), and close-up of glaze and texture detail. For test tile documentation, consistent lighting and camera angle across all tiles enables valid visual comparison. Include a gray card or color reference card in frame for color accuracy.^[43][44]

Glazy (glazy.org) provides a free online platform for recording glaze recipes, test results, and photographs with community sharing capabilities. It has become the de facto standard for digital glaze record-keeping. Spreadsheet programs also work well, enabling organized tracking with multiple worksheets for different test series.^[36]

Store physical test tiles in labeled boxes or on display boards, sorted by firing temperature, clay body, or glaze type. A mounted test-tile board arranged in grid matrix format (base glaze = rows, top glaze = columns) allows at-a-glance comparison and serves as a permanent studio reference.^[41][37]

Examine each test tile for six qualities:

• Color — intensity, richness, evenness, and how thickness or overlaps affect hue.
• Surface quality — smoothness, texture, glossiness, matteness.
• Flow and movement — pooling, breaking over edges, running on vertical surfaces.
• Defects — crazing (fine cracks indicating thermal expansion mismatch), shivering (glaze flaking off), pinholing (small holes from trapped gas), blistering (larger bubbles), crawling (bare patches from poor adhesion).
• Transparency and opacity — how well underlying clay or underglaze shows through.
• Functional quality — smoothness for lip contact on mugs, durability of surface.

For layered glazes specifically, pay close attention to the overlap zone compared to single-glaze zones. The overlap reveals the interaction effects — the “third glaze” — and is the whole point of the test.^[37][45]

A glaze fired flat can look very different from the same glaze on a vertical surface, with gravity creating pooling at the bottom and thinning at the top. Vertical test tiles reveal whether a combination will run, how far it flows, and where color breaks occur. This is essential for layering tests because layered glazes tend to be more fluid than single glazes due to eutectic effects. Always include a vertical element in layering test tiles.^[37][38]

Arrange fired test tiles side by side for direct comparison under identical lighting, and photograph them together for the most useful archival records. After visual evaluation, test durability by scratching the glaze surface with a fingernail, coin, or steel knife. A functional food-safe glaze should resist scratching; soft, easily scratched surfaces indicate underfiring, excessive flux, or insufficient silica — all of which may be exacerbated in layered glazes where eutectic effects can over-flux the interface. For definitive leaching safety, the only reliable test is an actual acid leaching test (citric acid or acetic acid), as described in Hesselberth & Roy's Mastering Cone 6 Glazes and in ASTM C738 standards.^[46][37][36]