Tile Work of the 17th and 18th Centuries

Migration of Techniques and Styles

Lesley Durbin



  Eight-ponted stars and feathery foliage decorate an area of Delft tile-work
  Despite the complexity of the repeated pattern on these Delft tiles, the hand-painting displays an easy, flowing style.

The history of what we have largely come to recognise as the 17th- and 18th-century tile work of northern Europe is a fascinating trail of migrating communities and craft techniques.

Glazed ceramic tile- and brick-making is an ancient art first found in Egypt, Mesopotamia and China. Tile-making among the Islamic cultures of North Africa, the Middle and Near East was, and still is, widespread, as are tiles originating from medieval and modern Spain, particularly Talavera de la Reina, Toledo, and the Triana district of Seville. This strand of tile-making can be traced from Moorish Spain and also from Italy up into Holland and the Low Countries at the beginning of the 16th century.

The Portuguese tradition of tile-making was inextricably linked with both Flemish tile-making and the Moorish tradition throughout the 17th century. Flemish ceramic artists of the highest calibre travelled to Portugal to work on royal interiors. The exchange of talent and technique between southern and northern Europe was commonplace for over three centuries.

The Spanish tradition migrated across to Mexico and later into California. The Dutch and Low Country tradition first appeared in London during the mid-16th century as people fled wars of religious persecution. However, it wasn’t until the 18th century that the technique for tile-making firmly established itself in Britain with centres in Liverpool, Bristol and London. Thousands of tiles were also imported from Holland in response to the soaring popularity of ‘Delft tiles’ (although, as we will see, the term Delft can be misleading).

By the mid-18th century the industry in England was at its height and, in addition to the home market, tiles were shipped in their thousands from the ports of Europe across the Atlantic to the Dutch and English east coast settlements of the New World.

Tile-making continued in Holland, where it still survives. The English scene saw the slow decline and end of the ‘Delft’ tradition as developments in the English ceramics industry began to produce the robust and distinctive product which eventually became the Victorian tile. There was, however, a brief revival of imported Dutch tiles between 1890 and 1910 when they were favoured by Morris & Co and other Aesthetic Movement designers.


The Spanish and Delft styles of manufacture are broadly speaking the same: a twice-fired, hand-decorated, tin-glazed tile. However, it is the later, northern hemisphere examples of manufacturing technique and conservation that will be the focus of this article.

The city of Delft, in Holland, was generally thought to produce the best tiles so ‘Delft’ has, over time, become the commonly used generic term. English tile-makers went to great lengths to make their product indistinguishable from the Dutch tile and it requires an expert eye to differentiate between them.

  Tile panel depicting drummer boy and other figures
  Tile panel, Triana, Seville, Spain
  Area of vertical tile work featuring various designs including horse-riders and bucolic scenes
  Typical use of Delft tiles surrounding a fireplace in
the UK
  Tiles featuring hand-painted sailing ships
  Hand painted images in cobalt and manganese
  Small area of tile work with two tiles missing exposing badly cracked plaster backing
  Lime plaster adhesive which has shrunk and fractured while drying

In Holland, Spain and Portugal elaborately hand-painted tile panels were used as interior and exterior decoration in their own right (above). However, in England their use seems to have remained firmly utilitarian in dairies and kitchens although they were used more decoratively in fireplace surrounds (below left).

The clay used for ‘Delft’ tile-making is specific, it tends to be rich in calcium carbonate and fine sand deposited as silt from river beds and has added marl (a naturally occurring sedimentary clay). This produces a tile which is light in weight and tends towards porosity. Clay was usually dug locally but, as with all large-scale ceramic industry production, imported clay often eventually replaced locally dug clay.

The clay was weathered in the open air over a period of time, allowing the natural actions of frost and rain to break down its solid mass. The production technique began with rolling and ‘pugging’, mixing and folding the wet clay to remove stones, impurities and small pockets of air. The wet clay was then rolled out onto a sanded board to thicknesses as little as 6-7mm, allowed to dry leather-hard and cut into squares.

After a further drying period the undecorated tiles were put into a kiln for the first biscuit firing to temperatures of up to 1000°c to produce a hard ceramic object.

The biscuit tile was then ready for glazing and decorating. Tin glaze is prepared from a clear lead glaze with added tin oxide, which gives the glaze an opaque white colour. Shades of white could vary widely and eventually sand and an alkali were added to glazes to consolidate and strengthen the final product.

Traditional glaze colours for decoration were produced by adding cobalt to produce blue and manganese for purple. After dipping the biscuit tile into the tin glaze and allowing it to dry, the decorator could then apply the design by hand painting in blue or purple glaze directly onto the dipped, unfired surface.

The tile was then ready for a second firing of up to 800°c. Repeated patterns were achieved using ‘pouncing’, a method of tracing designs onto tiles by drawing the outline of the design onto paper and then making pinpricks at regular intervals along the lines of the pattern. The perforated design is then placed over a tile and dusted with burnt bone or charcoal, transferring a dotted outline of the pattern onto the tile.

Because the pattern is only transferred in outline no two tiles are exactly alike. The charm of Delft tiles lies in this elegant variation, in the ease and swiftness of the hand-drawn design and in the way that the blues and purples migrate into the white background during firing (see title illustration).

In Liverpool, in 1756, John Sadler and Guy Green introduced the technique of transfer printing onto tiles which produced an entirely different style of decoration formed by a printed outline transferred from a copper plate engraving fired onto the surface of the tin glaze. This allowed repeated intricate designs to be printed onto tiles at great speed.

While transfer printing later became a mainstay of decoration for the pottery industry as a whole, no other tile-makers followed suit at the time.


Delft tiles in dairies and kitchens were for the most part swept away in favour of the new, more robust Victorian tile. Those that remain are usually conserved with great care or taken away to be displayed as ‘art’ pieces. In the UK, Delft tiles are rarely found undisturbed in their original location in significant numbers.

The chief failure of Delft tiles is the inherent problem of the glaze sufficiently melding with the clay body to form a single robust entity. The glaze and body too often shrank at different rates when cooling from the kiln resulting in the glaze ‘crazing’ (forming a network of fine cracks) and flaking away from the body.

The resulting clay body, in ceramic terms, would be described as ‘soft’, meaning that it tends towards high porosity, friability – particularly around the edges – and easy fragmentation. Much of the conservation work on 17th- and 18th-century tiles is associated with consolidation of damage directly resulting from the manufacturing technique.

The soft body of the tile is particularly prone to moisture absorption, which in turn can result in efflorescence caused by salt crystallisation. Salt crystal growth can force the glaze to become detached from the tile body. This is a well-known and understood problem with soft-body ceramic objects. Moisture absorption also leads to the ready ingress of stains into the tile body, particularly if tiles have been used in fireplace surrounds.

The glaze itself is also soft and will readily become worn on the surface, losing the typical soft shine which tin glaze produces and again leaving the glaze open to absorption of dirt.


The most damaging and frequently encountered inappropriate repair is the use of Portland cement either to reinstate loose tiles or to install them in a new location following the failure of the original lime plaster fixative.

The relative hardness, lack of plasticity and ability to hold soluble salts all combine to make the use of Portland cement as an adhesive for Delft tiles highly damaging. Little can be done to reverse the use of cement without the loss of original tile material.


Delft tiles can be cleaned using a solution of de-ionised water and a neutral pH detergent such as Synperonic A. Usually, a moistened soft cloth is sufficient to clean the glaze but particular attention can be paid to dirty grout lines using a tooth brush. Buffing to a shine with a dry soft cotton cloth brings out the soft lustre of the glaze. If tiles are overwetted there is a danger that crystallisation of soluble salts may take place, and if loose tiles are soaked over a period of time in distilled water to remove salts or staining, care must be taken to dry the tiles quickly and thoroughly.

  Horizontal hearth tiles with cracks and losses near fire grate
  Fractured hearth tiles with worn glaze surface
  Filled narrow crack in tile with simple repeat flower pattern
  Fractured Sadler and Green printed tile, glued and filled with plaster of Paris

Organic staining, such as soot, if left alone, will continue to migrate into the tile body. Removing organic and iron oxide staining is difficult and requires the expertise of a ceramic conservator.

Site-related problems

Lime plaster was used to fix tiles during the 18th century and it is usually found to be fractured behind the tiles (above left). Shrinkage at the time of drying causes a complex polygonal matrix of fractures to form which can cause the tiles to sound hollow or to rattle when tapped, but it may not be problematic if the tiles are adhered and the grout between the tiles is sound.


It is quite common to find 18th-century tiles in fireplace surrounds. The severe loss of glaze surface on hearth tiles due to wear cannot be repaired. Impact damage such as fractures can be repaired using normal ceramic restoration techniques especially if the tile is also loose from its setting.

The most common problem is soot blackening at the edges of tiles. The worst effects of this can usually be removed as described above but often the best option is to remove and replace the old grout altogether.

Salt crystallisation

Salt crystallisation is a well understood problem. In damp conditions, chemical changes resulting from salt crystallisation degrade lime plaster causing it to soften and crumble. The problem can be exacerbated in the case of 18th-century tile work because the lime plaster used as a fixative often contained gypsum, which is particularly prone to salt action where there are prevalent damp conditions. The only remedy in this situation is to create an ambient environment with a relative humidity no greater than 60 per cent, which will prevent crystallisation.

Mortar degradation

Thermal shock and water penetration are common causes of degradation. The mortar will first expand and then shrink on cooling or drying at a different rate to the tile, causing cracked mortar or complete detachment. Even if the mortar has failed, however, the tiles may remain in place, held together by the grout. Failed mortar is detected by a hollow or rattling sound or bulging surface plane and loose tiles. Lime plaster adhesives are extremely corruptible and easily subject to degradation. Corroded ironwork may also cause fracturing due to expansion of the ironwork.

While problematic in itself, structural movement rarely has an impact on 18thcentury tile work unless, of course, movement causes tiles to become unseated and fall. Lime plaster adhesive will fracture and shift without damaging tiles because it is inherently weaker.

Mortar consolidation

Consolidation can be achieved by introducing an adhesive liquid by capillary action which will creep into voids and give additional adhesion between tiles and mortar. The advantage of this method is that both the tiles and the original mortar are left largely undisturbed except for the intrusion of the adhesive liquid. The disadvantage is the near impossibility of complete re-adhesion as there will always be some areas which the liquid will not reach.

Compatibility of materials and reversibility are important considerations when choosing a consolidant. It must not be stronger or harder than the ceramic tile and because Delft tiles have relatively weak bodies, only a few consolidating materials are suitable.

  Picture of two vertically adjacent tiles with flower vase decoration before and after restoration
  Composite image showing tiles before and after painting restoration

Re-fixing tiles

There are various methods for re-fixing 18th-century Delft tiles. The substrate behind the adhesive layer must be relatively weak or sacrificial. This will protect the tiles into the future. In a new location this can be a layer of plaster board on a stud frame. In a traditional location the substrate must be lime mortar.

Lime used as a tile adhesive must be naturally hydraulic. The inclusion of plaster of Paris with a hydraulic lime produces a ‘selenitic’ lime, which gives very good grab but is not suitable for damp conditions.

Repairing 18th-century tiles

Paraloid b72 is a standard consolidant adhesive for Delft tiles and other soft-bodied ceramics. It can be applied, diluted in acetone, with a fine paintbrush around the fragile edges of distressed glaze and over vulnerable crazed areas.

Restoration also works well as a form of conservation – filling missing areas of glaze or body clay is preferable to leaving damaged edges and surfaces uncovered and therefore unprotected. The original material of the tile will be better preserved if there is a sound surface area. The introduction of a white plaster of Paris infill where fragments are missing in a scheme will not greatly detract from the overall visual aesthetic (see printed tile repair, above right).

If the intention is to restore the colour and decoration to fragmented and missing areas of tiles then it is advisable to use an acrylic- and calcium carbonate-based fine surface filler as it will provide a smooth dense surface finish for paint application. The decoration can then be applied using an acrylic clear glaze mixed with artists’ powder pigments to match (above left).


Good maintenance relies on positive action to maintain an ambient environment suitable for soft-bodied tin-glazed tiles. Damp conditions with excessive moisture ingress into the mortar or substrate are the most common cause of damage.


Recommended Reading

L Durbin, Architectural Tiles: Conservation and Restoration, 2nd edition, Routledge, London, 2014

English Heritage, Practical Building Conservation: Earth, Brick and Terracotta, Ashgate, Farnham, 2014

Expert guidance on the care and conservation of historic tiles can be found on the Jackfield Conservation Studio website


The Building Conservation Directory, 2015


LESLEY DURBIN ACR began working in conservation in 1983 and is senior conservator in The Jackfield Conservation Studio, based at Jackfield Tile Museum, Ironbridge. She has worked in the conservation of architectural tiles for over 30 years and is the author of Architectural Tiles: Conservation and Restoration (see Recommended Reading list at the bottom of this page).

Further information





Architectural terracotta

Ceramics conservators

Floor and wall tile suppliers

Interior conservators and decorators

Interior designers and consultants


Site Map