BCD 2019

76 T H E B U I L D I N G CO N S E R VAT I O N D I R E C TO R Y 2 0 1 9 C AT H E D R A L COMMU N I C AT I ON S COLLYWESTON ROOFS JANINE DYKES C OLLYWESTON STONE slate takes its name from the village of Collyweston in Northamptonshire. It is a traditional roof covering which has been used in the East Midlands since the Roman era. Indeed, during excavations on Roman sites at Great Casterton and Godmanchester hexagonal shaped slates with a single peg hole at the top have been found (Clifton-Taylor, 1987) and similarly, excavations at the Roman villa south of Apethorpe Palace and the settlements at Barnwell and Collyweston have uncovered further examples of Collyweston slate (Burgess, 1991). GEOLOGY Collyweston slate is a sedimentary fissile oolitic limestone from the Jurassic period (140–190 million years ago). Sedimentary rocks are formed by the gradual accumulation of layers of sediment, such as on the floor of the sea, and may include seashells (calcium carbonate), sand (from the decay and disintegration of pre-existing rocks) mud and/or other mineral sediments. The layered formation is retained in the new rock as bedding planes, affecting its physical characteristics. As Collyweston slate was formed by precipitation of calcium carbonate into the sand bed, it is not true slate, which is a metamorphic rock formed from a fine-grained sedimentary or igneous rock. However, like slate it is fissile: in other words, it splits easily along planes of weakness into sheets, albeit less thin than those of a true slate. Like all limestones, Collyweston slate is high in calcium carbonate, but about 30 per cent of the slate is quartz grains (essentially silicon oxide), and other particles include fragments of fossil up to 4mm in size, iron oxide and white mica (a hydrated silicate of aluminium or potassium). Calcite is the cementing agent and at the bedding planes it is free of quartz particles, which causes the thin horizontal divisions known in Collyweston slate as the cliving planes (Sharp, 1873). THE IMPORTANCE OF LOCAL BUILDING MATERIALS Prior to the railways it was very difficult to transport heavy materials. As a result, most buildings were constructed with materials available near to the site, and only the most wealthy owners and the patrons of ecclesiastical buildings could afford to transport the materials they wished to use from further afield. This is why historically places were built from a small palette of materials, which created a strong local character. This has been diluted somewhat since the 19th century but is still strong in many areas across the country, including the use of flint stones in Norfolk, hard granite stones in the Lake District and limestone in the Cotswolds. Physical context strongly affects the image of a place in people’s minds and local materials are a fundamental element of what gives different places their character and identity. There is a close relationship between traditional materials and geology, which architectural historian and writer Alec Clifton-Taylor described as ‘the pattern of English building’ in his book of the same title. Local materials like Collyweston slate have distinctive physical and aesthetic qualities which carry on hundreds of years of traditional skills and detailing, and create strong connections between people and place. To allow their sense of pride and connection with the locality to continue for future generations, it is therefore important to replace local materials and their traditional detailing on a like-for-like basis. METHODS OF COLLYWESTON SLATING Collyweston slates are often nailed to battens, but traditionally they were fixed by a peg hooked over riven lath and with a dab of lime mortar at the head to secure it (head bedding). The newly laid Collyweston slates were then pointed with lime mortar (tail bedding) including the tail and the perpendicular joints, but no higher than the tail of the slate above. Internally, the underside of the slates was draft-proofed with a coat of lime plaster. Torching, as this practice is known, has largely been replaced by the use of roofing under felt, or underlay, but this can trap moisture where slates are fully bedded in mortar (see common failures, below). Collyweston slates are laid in diminishing courses. This means that the slates get smaller with each course, with the smallest slates at the ridge line and the largest at eaves level. The stone slates are laid using the double lap slating method. This means that the slates should overlap the slate in the course next but one below by two or three inches (head lap) and each slate should also overlap the slate in the course below laterally by about half of its width (side lap). To be sure that the roof is waterproof both laps are required, but unfortunately they are sometimes reduced so that less slates are needed and more money can be saved. This is known as stretching the slates. Unusually large slates may be an indication that the side lap has been reduced which can compromise the The newly renewed Collyweston slate roof of the 18th century watermill at Sacrewell, Cambridgeshire (Photo by Rob Harvey for Messenger BCR)