Scottish Court of Session Decisions

OUTER HOUSE, COURT OF SESSION [2009] CSOH 125
CA98/07	OPINION OF LORD MENZIES in the cause ALBERT BARTLETT & SONS (AIRDRIE) LIMITED Pursuers; against GILCHRIST & LYNN LIMITED Defenders; and ATLAS WARD STRUCTURES LIMITED First Third Party; and BRIGGS AMASCO LIMITED Third Party to the First Third Party: ­­­­­­­­­­­­­­­­­________________

Pursuers: Ellis, Q.C., Sanders; MacRoberts

Defenders: Howie, Q.C., G. Walker; Wallace Construction Law

First Third Party: Simpson & Marwick

Third Party to the First Third Party: Semple Fraser

8 September 2009

Introduction
[1] The pursuers carry on business as processors, packagers and suppliers of potatoes and other vegetables to the retail, wholesale and food service trade. They operate from four main sites across the United Kingdom, including premises in Airdrie. In about October 2001, the pursuers invited tenders for the design and construction of a major new potato processing and packaging plant at New Monkland, Stirling Road, Airdrie. In December 2001, a contract was concluded between the pursuers and the defenders whereby the defenders were responsible for the design and construction of the new building. By reason of sub-contracts, or otherwise, each of the third parties had some involvement in the design or construction of the building; it is not necessary for present purposes to set out here either the nature or the contractual basis of their involvement.

[2] The building is approximately 100 metres long and 171 metres wide. The roof area is approximately 23,500 square metres. The design of the roof comprises several ridges and gulleys; the pitch of the roof between each ridge and gulley is approximately 4 degrees. The cladding of the roof is by means of insulated metal sheets, the metal of which was coated with plastic at the time of manufacture; interspersed at intervals between these insulated metal sheets were corrugated plastic roof light panels.

[3] During the course of the construction of the building, concerns were expressed on behalf of the pursuers to the defenders and others involved in the construction of the building about apparent water ingress through the roof. It was pointed out to the pursuers that the works were not yet complete and that any defects would be attended to once the works were complete. The pursuers took possession of the new building on 17 November 2003. Since then there has been considerable, widespread and persistent ingress of water at many points through the roof.

[4] In these proceedings the pursuers seek damages from the defenders for the loss and damage which they have suffered through the defenders' breach of contract. The defenders admit that the roof of the premises is not watertight, and they admit for the purpose of these proceedings that they are in breach of contract. Issues of liability, relief and apportionment between the defenders and the various third parties then arose. A proof at large was allowed on all the issues raised by the various parties. However, thereafter agreement was reached between the defenders and the various third parties as to issues of liability inter se, relief and apportionment, and an interlocutor was pronounced assoilzing the second third party. The only live issue between the pursuers and the defenders (and the remaining third parties) was the proper measure of damages to which the pursuers were entitled. I heard a proof lasting some 9 days on this issue.

[5] At the risk of oversimplification of parties' positions, the pursuers maintain that they are entitled to damages assessed on the basis of an overcladding solution, which would involve draining the skylights in the roof and thereafter installing over the existing roof cladding a properly fixed single ply profiled roof sheet. The defenders and the remaining third parties contend that this is unreasonably expensive, and would not, in any event, resolve the problems of water ingress to the building. They contend that a cheaper and more effective solution would be to apply the Kemperol 2K PUR system to those joints and areas of the roof which leak.

[6] In the course of the proof, parties lodged a joint minute (No. 63 of process) agreeing "that the scope of the present diet of proof should be restricted to the question of whether the pursuers' damages are to be assessed on the basis of:

(i) the 'overcladding system' as averred by the pursuer in Article 8 of the Condescendence; or

(ii) the "Kemperol system" as averred by Briggs Amasco in Answer 8 of the Condescendence; and

(iii) that all other matters should be reserved meantime for further procedure."

It is to these issues that this Opinion is addressed.

The Evidence

1. Evidence for the pursuers
[7] Seven witnesses gave evidence on behalf of the pursuers.

[8] Mr Alexander Colvan spoke to his written statement (No. 6/59 of process). He was employed as engineering manager of the pursuers, and spoke to the dimensions of their new building, the activities which were carried out in the various parts of the building, and the problems of water ingress from leaks in the roof which were noticeable as soon as the pursuers took possession of the premises in November 2003, and which were apparent from the various sets of photographs in No. 6/9 of process. These leaks have become progressively worse, they are more apparent during rainfall, and there are still continuing problems with leaks across most of the area of the roof. He found it difficult to be specific as to any areas of the premises where there were no leaks (except in Area G on the plan which forms tab 48 of No. 6/9 of process, which is the canopy covering the loading bay for articulated vehicles, in which there were no leaks). The leaks in the roof caused problems for employees who suffered slipping accidents because of water on the floor of the factory; No. 6/18 of process contained extracts from the pursuers' accident book between 2006 and 2008 which included reports of slippages as a result of water leaking from the roof. He agreed in cross-examination that the pursuers had not taken steps to repair or replace the roof as yet, except to the limited extent indicated in his written statement. The pursuers maintained a system of maintenance of the roof which involved inspection of the roof and the cleaning of roof gutters annually. The remedial works carried out by Brett Martin in about May 2005 to the roof lights involved the drilling of holes in the roof lights to allow drainage of water trapped therein, and the refilling of the holes with Mastic. This did not resolve the problem of trapped water in the roof lights.

[9] Mr Iain Stewart Fergusson was a chartered architect who had practiced as an architect since 1974. His professional qualifications and experience are set out fully in the curriculum vitae which forms Appendix 4 to his updated report dated 10 December 2008 (No. 6/9 of process). He spoke to that report, and to his commentary on other reports dated 8 January 2009 (No. 6/13 of process). His practice had centred principally on problems in buildings since about 1998, and he first became involved in the pursuers' building in December 2005, first inspecting it in January 2006. He was asked to give advice on why the roof was leaking and how the problem should be resolved. In order to prepare his reports, he had walked over the roof on many occasions and spent many hours on it.

[10] Mr Fergusson stated that the roof was formed of composite panels with an insulation core, which rested on Z Purlins on a steel structure. As recorded in the experts' joint statement with drawing attached (which forms tab No. 48 to No. 6/9 of process), the problem of water leakage occurs principally in roofs A-F on that drawing. Although there are a few additional isolated leaks, the principal problems with leakage are identified in paragraphs 3 and 5 of the experts' joint statement, as follows:

"3. Water seen on the floors and office ceilings within the building is the result of rain water falling on the roof and penetrating the end laps. The pattern of drips indicates that this is a wholesale problem affecting essentially all end laps on slopes A-F, including roof light/panel end laps. There are of the order of 1,000 linear metres of end laps in slopes A-F.

5. Water vapour is able to penetrate into a significant proportion of double skin roof lights and this has resulted in a significant accumulation of water in these roof lights. This condensate does not contribute to water dripping from the roofs."

[11] The specification of roof cladding which was provided by the defenders was as stated in tab 41 of No. 6/9 of process, in which the pitch of the roof was specified as 4 degrees and the metal sheets were described as manufactured by Ward Building Components, panel reference IP1000 with a panel thickness of 40mm (80mm over the production area). The facing material was specified as steel with a galvanised coating, external finish/colour being HPS 200/goosewing grey. Mr Fergusson considered that 4 degrees was a shallow pitch for such a roof, but overall the pitch of the roof conformed to the design pitch of 4 degrees, and the metal panels were designed for such a pitch.

[12] Under reference to paragraphs 6.02.5 to 6.02.11 of his report (No. 6/9 process) Mr Fergusson considered whether the roof leaks at the end laps were more likely to be caused by condensation or by rainfall penetration. He concluded that condensation was not the problem, and that the cause of rainwater penetration was that the manufacturers' recommendations were not complied with in that no end lap stitches in the valleys were installed at the overlap joints of the roof sheets (all as agreed in paragraphs 10-14 of the experts' joint statement, tab 48 of No. 6/9 of process). The reason that installing stitching screws at the end laps now would not resolve the roof leaks was that the butyl sealant which had been part of the original design for the end laps of the roof sheets had decomposed and deteriorated. It could not be replaced with new butyl sealant without causing major damage to the roof sheets. It appeared that in some places silicon sealant had been inserted in an attempt to plug the gaps at the end laps, but this had merely had the effect of forcing the panels apart.

[13] The method of resolving the problem of water ingress which Mr Fergusson favoured was by overcladding the existing roof with a further single skin roof. This new overcladding roof would be about 11/2 to 2 inches above the existing roof and would be supported on a metal grid. It would be a single skin of metal sheets, of the same material as the existing roof but with a different profile as the single skin required increased rigidity. Vapour control tape would be applied to all the joints, and as an extra precaution the space between the existing roof and the overcladding roof would be insulated to prevent condensation forming.

[14] Because the processing and packaging plant was in operation every week of the year, it was not a viable option to strip off the existing roof and replace it. The pursuers' operations required the temperature in the premises to be managed and controlled, it would not be satisfactory to have tarpaulins covering the food packing processes, and there was an unacceptable risk of items falling on people working below. Moreover, it would be more expensive to replace the existing panels than it would to adopt Mr Fergusson's preferred overcladding solution.

[15] Mr Fergusson favoured the overcladding solution because it was the only reasonable way to resolve the problems caused by the lack of end lap stitching in the existing roof, and it would not be possible to replace the butyl sealant without causing major damage to the existing roof sheets. The possibility of condensation forming between the existing roof and the overcladding roof would be eliminated by a vapour barrier tape which would be applied to the existing roof structure and would prevent vapour rising. This should have the same lifespan as the existing roof. As an additional precaution to prevent condensation, insulation would be inserted between the existing roof and the overcladding roof, which would cause the dew point to be pushed outwith the building. The different profile of the overcladding roof would provide adequate rigidity to compensate for its single skin. The specification of the overcladding option which was provided by Mr Fergusson's firm is to be found in the appendix to the form of tender which is No. 6/19 of process, H 31:10A-87. Mr Fergusson explained that he regarded this as a two-stage tender process, with the precise details being worked out with the contractors at the second stage of the process. This would probably have cost implications for the client. He would expect the contractor ultimately to make provision for items such as safety nets, cherry pickers, and an allowance for disruption to the contractors' work as a result of having to work around the pursuers' continuing operations.

[16] Mr Fergusson did not favour the option of applying Kemperol 2K PUR to the areas of water ingress. Although he conceded that this would certainly be cheaper than his preferred overcladding solution, he did not consider that it would be satisfactory for the pursuers. He described it as "essentially a sticking plaster approach". In examination in chief, his reservations about applying Kemperol to the end laps and other areas of water ingress may be summarised as follows:

(i) As appears from the technical information for the Kemperol 2K PUR system (see Appendix X1 to No. 35/3 of process), the system is built up using a number of layers; when complete, it will create a "dam effect" causing debris to remain on the roof. The design of the roof relies on free run off of water to keep it clean. The Kemperol system will inhibit this water flow, allowing dirt to settle on the roof sheeting. This will encourage organic dirt to accumulate. If this is not cleared off at frequent intervals, the guarantee of the existing roof sheeting provided by Corus may be affected. In order to avoid this, the pursuers will face the additional expense and responsibility of a regime of frequent cleaning and maintenance.

(ii) Not only could the Kemperol system impact on the guarantee of the metal sheeting provided by Corus, it could impact on the guarantee of the cladding provided by the first third party. The warranty proposed by Kemperol (at Appendix X4 of No. 35/3 of process) is limited to faulty materials on Kemperols' part, and extends only to Kemperol materials.

(iii) The Kemperol system can only bridge a gap of up to 2mm. There are numerous places on this roof in which the gaps and end laps are significantly greater than 2mm, as is apparent from several of the photographs in Appendix I to No. 35/1 of process. That the Kemperol system is only designed to bridge a gap of up to 2mm is clear from Kemperol's own technical information (Appendix X1 to No. 35/3 of process), where it is stated that "the product is designed to bridge cracks up to 2mm", and also from the British Board of Agrément ("BBA") certificate, (Appendix X5 to 35/3 of process at paragraph 16.5). This limitation in the Kemperol system is particularly important given the distortions in many of the roof panels on the existing roof. Mr Fergusson described the distorted metal as "having a memory" and even if an attempt is made to reduce these distortions, the metal sheets will tend to revert to their distorted shape and thereby cause the Kemperol to be strained and possibly become ruptured.

(iv) There were concerns about the durability of the Kemperol system, particularly having regard to paragraph 15 of the BBA certificate. It has only been used in Germany since 1998, and more recently in the United Kingdom. Although the BBA certificate states that "accelerated weathering tests confirm the satisfactory retention of physical properties and indicate that, with adequate maintenance and repair, the system should have a service life of up to 25 years", there was no definition of what constituted "adequate maintenance and repair", nor is it suggested that the system will have a service life of 25 years - the service life may be less than 25 years. Moreover, at the end of its service life it will be necessary to strip off the Kemperol system and replace it; this would give rise to a risk of damage to the roof sheets, further damaging the roof itself.

(v) Silicon is not specified as a suitable substrate to which the Kemperol system can be applied, either in Table 1 of the technical information or in paragraph 8 of the BBA certificate.

(vi) The substrate to which the Kemperol system is applied must be dry. It was apparent to Mr Fergusson on his several inspections of this roof that there was retained water at the end laps, and this would tend to come out of the roof joints when pressure was applied to them - e.g. when the Kemperol system was being applied.

(vii) Mr Roberts placed some reliance on the quotation from K. Pendlebury & Sons Ltd dated 10 December 2008 (Appendix Y1 to No. 35/3 of process). However, Mr Fergusson understood that nobody from Pendlebury visited the site. It was necessary to see the roof to understand what preparations were required. The quotation made no provision for stitching screws at the end laps, nor did it comment on the fact that there were numerous gaps wider than 2mm. Moreover, the contractors were unable to provide a fully insured warranty, the maximum period of insurance backing for their warranty being 10 years.

[17] In cross-examination, Mr Fergusson stated that before the meeting between the experts for the various parties in this case he had not met Mr Roberts, although he had heard of him and it appeared from his curriculum vitae that he was a specialist in roofing and cladding systems. Mr Fergusson knew Mr Blois-Brooke, who had been involved in a number of building problems; he was a reasonably well known name and Mr Fergusson had worked both with him and against him in the past.

[18] Mr Fergusson confirmed that when he referred to the risk of Kemperol causing ponding which would negate any guarantee offered by Corus (see paragraph 7.02.5 of his report No. 6/9 of process), he was referring to the Corus confidex guarantee, a blank form of which is at tab 45 of his report. He had never seen such a guarantee in relation to this roof; this was because the pursuers never accepted the roof, so no guarantee was issued. He accepted that a manufacturers guarantee applied to the Kemperol system for a period of 20 years from installation, and so he could not support averments to the contrary to be found towards the foot of page 46 of the Closed Record. He agreed that the roof had been up since about October or November 2003; the expected life of 40 years which he gave for the roof cladding panels if properly maintained (6/9 of process at paragraph 6.03.3) was a rule of thumb, with 25 years being a reasonable expectation until the first maintenance.

[19] Mr Fergusson agreed that the Kemperol repair scheme was very considerably cheaper than the overcladding scheme which he preferred, the overcladding scheme being estimated to cost in excess of £900,000 and the Kemperol scheme being estimated at about £132,000 plus VAT (Appendix Z of No.  35/3 of process). He also agreed that the roof on this building was surrounded by a parapet, as shown in Appendix I of No. 35/1 of process.

[20] He accepted that when he designed and prepared the specification for his overcladding proposal, he had not considered what was now being proposed as the alternative, namely the Kemperol 2K PUR system with affected end laps being stitched down with six stitcher screws and three primary screws, application of the Kemperol system in association with a breaker, and painting over the whole with goosewing grey paint. This had not been suggested at the time of the design of his overcladding proposals. What was then being suggested as an alternative was the Aperture System of applied liquid sealant, which he agreed was quite different from the present proposal. However, he still retained the concerns which he had expressed in examination in chief. Could the gaps at the end laps be closed sufficiently? Could the Kemperol system be applied over silicone sealant? If not, could the silicone sealant be removed without damaging the roof? Could the "memory" of the metal sheets be changed? Were screws 50mm back from the edge able to exert sufficient pressure on the ends of the roofing sheets? And will the Kemperol system cause a slowing of water run off, allowing debris to accumulate and perhaps impacting on guarantees? He remained unhappy about each of these aspects.

[21] He accepted that his overcladding specification was a "two stage process, and embryonic", and that a number of matters required to be resolved before it could be built up. The criticisms made of the overcladding scheme contained in Section 11.2 of Mr Roberts' report (No. 35/3 of process) were put to Mr Fergusson and he responded as follows:

(a) Interstitial condensation was a potential problem. However, the application of vapour tape would counter this, and this would be backed up by insulation between the two roof layers.

(b) The concerns raised in paragraphs 11.2.2 to 11.2.4 of Mr Roberts' report were put to Mr Fergusson, namely that the new roof would be supported by the Ashgrid support rail system, which would involve about 49,000 new fixing holes through the existing roof, which would require to pass through the existing roof at an oblique angle before being attached to the purlin below, with a risk of clashing with existing screws and also of missing the purlin, and in any event giving rise to very many further points of water ingress through the existing roof. Mr Fergusson's response was that this was something which the contractors would have to do properly. It was put to him that these problems had been identified by one of the quoting contractors, and raised in their letter of 11 May 2007 (No. 6/21 of process); his answer was that this was the advantage of having a two stage tender process, which allowed the contractor's input to be taken into account.

(c) There would be difficulties with the roof lights because they would be effectively built in, as observed in paragraph 11.2.6 of Mr Roberts' report. However, these difficulties were not insuperable, and in any event the lifespan of the roof lights would be extended by the overcladding system, as they would not be exposed to ultra violet light and atmospheric contaminants.

(d) Although there are at present leaks at the roof lights, Mr Fergusson did not expect a repeat of this problem, and he would expect them to be sealed with butyl sealant. There should be no dew point problem causing condensation at the roof lights, because the underside of the roof lights are less cold because of the material of which they are constructed.

[22] Mr Fergusson reiterated his concerns about the Kemperol system. Despite the reference to the contrary in the BBA certificate he was surprised that the material was suitable for the repair of flat or shallow pitched roofs; the impediment which it would cause to the free flow of water would inevitably increase the requirement for cleaning and maintenance, and he considered that any form of applied solution on this scale halfway down the shallow slope of the roof and across the whole breadth of the roof was not appropriate. He accepted that Kemperol was seamless and elastic, and if used with a bond breaker would be well placed to deal with movement; however, a bond breaker was not specified in the Pendlebury quotation, and he would expect this to be included if it was part of the proposal, because it would comprise extra material and would also involve extra labour costs. Although the debonding layer would cover the areas of silicon sealant, if silicon was left at the end laps then this would mean that the pitch of the roof was less than 4 degrees, thereby increasing the problem of impediment to water flow. Placing stitcher screws 50mm from the end lap would not apply enough pressure to close this gap. Moreover, he was still concerned that it would be difficult to provide a suitably dry, clean and uncontaminated substrate so that Kemperol would accept that it was properly installed. To achieve this would add to the costs of the Kemperol system.

[23] Although a debonding layer would cover a crack of more than 2mm, Mr Fergusson questioned whether it would give structural support. The Kemperol system does not have capacity to provide itself with structural support over gaps of more than 2mm, so it might fail and a crack or hole might develop. This is what Mr Fergusson took from the wording of Appendix X1; he had no experience of the Kemperol system. He would need more specific information about how Kemperol was tested for the BBA certificate, what accelerated weathering tests were used and what experience of the system has been in the United Kingdom before he felt able to rely on the BBA certificate.

[24] In re-examination Mr Fergusson observed that no guarantee had been provided in this case because the architect had not accepted the roof, but he expected that a Corus guarantee such as the sample at tab 45 of No. 6/9 of process would be applicable in this case. He was not happy with the extent of the insured warranty referred to in the Pendlebury quotation (Appendix Y1 of No. 35/3 of process). Moreover, the Pendlebury quotation provided only for primary stitcher screws being applied, despite the indication in the costings in Appendix Z that additional primary and secondary stitcher screws would be applied. There was no provision in the costings for making good gaps at end laps or removing sealant, and he would expect these costs to be included. It would be necessary to dry the space between the end laps and clean out any debris and sealant from it before applying the Kemperol. This might cause distortion to the sheet end, and the action of applying Kemperol to the end lap would tend to exude further water. No costs were allowed for in this respect in Appendix Z. Moreover, no provision was made for the cost of protecting the pursuers' labour force while these works were being carried out - there was a potential risk of injury to the pursuers' employees working below, and also a potential risk of contamination of the food being processed and packaged. No allowance was made for these costs in Appendix Z, nor was there any allowance for interruption or disruption of the pursuers' operations, or interruptions to the contractors' workings.

[25] With regard to the possible costs of his overcladding solution, Mr Fergusson considered that the application of vapour tape would probably be sufficient to prevent the risk of interstitial condensation, but insulation may or may not be necessary, and he considered that the cost of such insulation should be retained in the contractors' costings, although it would probably not be spent.

[26] Mr Fergusson himself suggested the Ashgrid system for supporting the overcladding roof. He conceded that it was always a potential problem affixing screws through an existing roof, at an angle and connecting them to an unseen location, but that was a matter for the contractor to achieve. Although he suggested the Ashgrid system, it might be that there was another system which could be used more effectively. He was confident that the problem of brackets could be resolved at the second stage of the tender process.

[27] Mr William Hope spoke to his written statement (No. 6/57 of process). He was a FRICS, and had worked as a quantity surveyor since 1974. He worked for a local authority, but in addition he operated the firm of Hope Partnership part-time, and in this latter capacity was the quantity surveyor in relation to the construction of the pursuers' new building. He prepared the invitations to tender for the overcladding option, an example of which was No. 6/19 of process. The specification of the H31 metal profiled/flat sheet cladding/covering which formed part of that document was provided by Mr Fergusson; the specific limitations on method/sequence/timing at Part A35 on page 1/A19 were inserted after discussions with the pursuers and Mr Fergusson. Mr Hope prepared the report on tenders (No. 6/17 of process) which recommended a tender which would result in an overall project cost of £1,207,588.00 excluding VAT and inflation beyond the acceptance period. He recommended this tender because the other two tenderers had underestimated the cost of health and safety provision, craneage and the like. Under reference to No. 6/22 of process, Mr Hope accepted that no provision had been made in the tenders for a performance bond in respect of the contractors' default or liquidation; surety for this was normally about 10% of the value of the contract.

[28] Mr Hope observed that he found it difficult to envisage what day works might be in a design and build contract, and if there was no provision for day works then the employer would not pay anything. If a particular contingency did not arise, nothing would be payable by the employer in that regard. In relation to the works which were the subject of his report on tenders, Mr Hope envisaged that immediately beneath the area of working on the roof there would be temporary provision for internal protection provided by the contractors; anything beyond or below this would be the responsibility of the pursuers, who might wish to go further than the protection provided by the contractors. In re-examination he accepted that the invitation to tender provided (at page 1/A19) that "no contamination of the produce can be allowed", and that Price Roofing, who were one of the tendering contractors (whose tender was No. 6/16 of process) made no provision for the avoidance of contamination, except perhaps the sum of about £6,000 "protection", which would not meet the cost of providing for the avoidance of contamination over a 26 week contract. It did not look as if these tenderers had priced for this element of cost at all.

[29] Mr Andrew Oswald had practised as a chartered surveyor since 1979, and had been a partner with Knight Frank for about 15 years before the proof. He specialised in valuation of all types of commercial property. He spoke to his report dated 9 November 2008 and letter dated 18 December 2008 (Nos. 6/10 and 6/12 of process respectively). His purpose was to report on the valuation implications of the options to repair an acknowledged defect to the pursuers' roof. He described the options as "either to provide a new 'over-roof' across the entire structure, or to seal the original roof by a proprietary system known as the Kemperol V210 system". He explained his understanding that

"the difference between the cost of the two repairing methods is approximately £1,000,000 for the over-roofing and approximately £70,000 for the sealant".

He went on to express the opinion

"that any building surveyor would recommend the over-roofing system, although considerably more costly, does provide a permanent solution to the problem. I do not believe that on the open market a hypothetical person would accept the sealant system proposed".

He was of the opinion that the market value of the feuhold interest in the property, with vacant possession, assuming that there is no roof defect as at November 2008 was in the sum of £10,000,000 whereas on the special assumption that only the sealant system is applied to the roof defect the market value was in the sum of £8,800,000. He explained that he reached this figure because a potential purchaser would take account of the fees involved, the risk of a higher repair cost, increased management time and loss of business. If the overcladding option was carried out there would be no effect on the value of the premises; if the Kemperol solution was adopted, this would be seen by potential purchasers as a patch or temporary solution, and the value of the premises would be diminished by about £1,200,000.

[30] Mr Oswald explained that his discipline was valuation of commercial properties, and he relied for specialist advice on the proposed remedies from Mr John Gallagher of Knight Frank. Mr Gallagher was a building surveyor and was concerned in the "nuts and bolts" of properties. It was Mr Gallagher who gave him the background to the two potential remedies, and the likely costs of each of these: these were not within his area of expertise. It was Mr Gallagher's view that the sealant system was a short-term patch, and this was what Mr Oswald factored into his valuation. The comment made at page 15 of No. 6/10 of process that any building surveyor would recommend the over-roofing system (quoted above) was made in reliance on the views of Mr Fergusson and Mr Gallagher. He thought it unlikely that any bank or other lender would be prepared to lend to a potential purchaser if the roof had been the subject of a patch repair. However, he accepted that if the bank or lender had been advised by reputable building surveyors that a sealant was a permanent repair, no doubt the bank would take its own view on this - he was not a banker. The main point was that any potential purchaser would wish a permanent repair to the roof - if the roof needed to be repaired every 10 or 20 years this would not be acceptable.

[31] Mr John Gallagher was a chartered surveyor and a partner in Knight Frank. He had been employed by Knight Frank for 15 years and had been head of their building consultancy for about 9 years. Building consultancy involved the technical side of building, including maintaining, refurbishing and designing buildings. He had experience of looking at similar buildings and of assessing the condition of roofs, refurbishing them and overcladding them.

[32] Mr Gallagher knew what Kemperol consisted of, and if a roof had been repaired with this he would notice it and report it. There were several waterproofing products on the market; he might not recognise it as Kemperol, but he would recognise it as a repair and this would prompt further investigation.

[33] He prepared his report dated October 2008 (part of No. 6/12 of process) to enable Mr Oswald to have an understanding of the options being considered for repair. The fact that he referred in the report to "Kemporal V 210" although what is now proposed is Kemperol 2K PUR is largely irrelevant - the V210 system came from the documents which he had reviewed, and this had now changed to K2 PUR. In any event, Kemperol was not a suitable option for remedying the defects. This was a new build project, which failed from day one. The only situation which would allow business continuity in Mr Gallagher's opinion was the overcladding solution - if there had not been a requirement for the pursuers' business to continue without any break, he would have required a replacement of the roof. The use of Kemperol would void any guarantees relating to the supplier of the roof panels. It would be to introduce a second product to a roof that has failed; if there was an insurance claim, one would be met with the argument that the roof was not repaired using recognised and approved methods, and the claim would be rejected. It was important to recognise that this fault occurred during the erection of the new building, and was apparent from the outset - it was not something that occurred 20 years down the line. Mr Gallagher had reviewed Wards' literature and could find no reference to Kemperol in it, so this was not a material approved by Wards. There might be individual guarantees of individual elements of the roof, but these could not make up for the absence of a blanket guarantee covering the whole roof. He had seen Appendix Y1 of No. 35/3 of process, the Pendlebury quotation, and he observed that the Kemperol system was guaranteed for 20 years and that workmanship was guaranteed for 10 years, but this did not alter his view that this was not equivalent to a guarantee of the whole roof. This is something which he would expect any building surveyor to pick up and be concerned about. He did not regard Kemperol as a satisfactory solution to the problems in the pursuers' roof.

[34] In cross-examination Mr Gallagher stated that Kemperol was a material with which he was acquainted; it creates a waterproof membrane and is in his view a temporary solution, and was not something which he would specify as a material in a new build contract.

[35] Mr Colin Thomas Campbell began employment with the pursuers in November 1986 and since 2002 has been their production and HR Director. He is in charge of the operating area of the premises, and looks after all staff, the running of the building and the structure of the building. He spoke to his statement, No. 58 of process. He was involved with the design team, and during construction he attended every second day; leaks were apparent in the roof before the pursuers took possession, but they were under a strict deadline to take possession before Christmas 2003. There were some areas of the roof where the problem end laps had been taped; he had seen panels similar to No. 35/5I of process (which was an example of roof panelling with the Kemperol 2K PUR system applied), although the strips that he had seen on the roof were narrower than the Kemperol production. He had observed that water did not flow over the tape, and there was a build up of soot or silt on the leading edge, i.e. the upper edge of the tape. Under reference to the last four sentences of paragraph 32 of his statement, he stated that some of the gaps at the end laps were in excess of 10mm; he guessed that 35-40% of the roof area had end laps with such gaps. He has been on the roof quite often, and has shown roofing experts around it.

[36] Under reference to paragraph 37 of his statement, Mr Campbell reiterated that from the pursuers' perspective any form of patch up job was unacceptable. They did not purchase a roof with Kemperol, they purchased a proper roof and they did not receive that. If they had received roof panels when they took entry to the premises with Kemperol applied to the end laps, they would not have accepted this. Operationally, he confirmed that it would not be possible to take the roof off and put a new one on the premises; the pursuers produced between 3,500 and 4,000 tonnes of potatoes per week, and the basic hours of the factory are 6.00am to 1.30am between Monday and Thursday, 6.00am to 5.00pm on Friday, 9.00am to 2.00pm on Saturday and 9.00am to 6.00pm on Sunday. The pursuers supply seven major supermarkets and they dispatch produce from the premises every day of the week between 7.00am and 7.00pm. But for these operational issues, they would have insisted on a new roof; overcladding is the next best thing. This was a high profile site in the packing industry, and it was embarrassing for the pursuers to show influential customers around the site and to have the problem of water ingress through the roof still not sorted.

[37] Mr Ronnie Alexander Bartlett joined the pursuers in 1979; he became a director in 1984 and had been joint managing director since about 1999. He is in charge of the potato side of the business. He spoke to his statement, No. 56 of process.

[38] He confirmed that it was not practicable to remove the existing roof and replace it, because of the disruption to the pursuers' business that this would cause. Were it not for this, the pursuers would be looking for a completely new roof. Overcladding, although a compromise, is the next best thing to a replacement roof. He confirmed that if damages were awarded to the pursuers, they intended to proceed to overclad the roof.

[39] Repair by means of the Kemperol system was not an acceptable option for the pursuers - they had bought a new building, and they would not have accepted the roof if it had Kemperol strips on it when they occupied the premises. They wanted something new and to best standards, and that was not what they received. Mr Bartlett had been advised about the Kemperol system by Mr Fergusson. He considered that Kemperol was just a patch up job; it might be acceptable in a 30 year old building, but not for a new build. Mr Fergusson's advice was that Kemperol was unsuitable. The product did not have long life, and if they could not guarantee it, why should the pursuers accept it? It could also adversely impact the value of the building. If the court awarded damages on the basis of the Kemperol solution, the pursuers would be in a different position from the position to which they were contractually entitled. The reason that remedial works had not been carried out yet was that the pursuers had been in dispute with the builders ever since they took occupation, and they wanted to be sure of a proper solution to the problem. In cross-examination he emphasised that it was not the money which the pursuers wanted; they just wanted a new overclad roof. This position was taken on professional advice; they had bought a new roof and they should get what they paid for.

2. Evidence for the defenders and third parties
[40] Two witnesses gave evidence on behalf of the defenders and third parties.

[41] Mr Keith Roberts spoke to his two reports dated 28 March 2008 and 11 December 2008 (Nos. 35/1 and 35/3 of process respectively), subject to minor amendments made at the outset of his evidence. He was aged 50, a chartered civil and structural engineer and managing director and principal consultant of Roberts Consulting. His qualifications, experience, employment history and publications are set out at Appendix A to No. 35/1 of process. Over the last 8 or 10 years he has specialised in roofing and cladding, and in advising on roofing disputes. Since 1994 he has been the author of the technical page of Roofing and Cladding magazine.

[42] Mr Roberts had been involved in inspecting and advising on remedial works to 12 composite panel roofs in the year before the proof. He was well acquainted with the Kemperol 2K PUR product, and had seen the material first in about 1992. He had seen Kemperol 2K PUR laid at a site. He had visited the present site once, over a two-day visit in March 2008. He explained that the site was categorised as severely exposed; it was about 185 metres above sea level and exposed to strong wind speeds and many days of rainfall each year. The roof of the building was shallow-pitched at 4° (and as was apparent from the roof surveys at 35/2 of process and Appendix O of 35/3 of process, there were several places at which the slope was shallower than 4°.) The shallow pitch of the roof slopes was a critical factor in the design of the roof, and special precautions were required for such a shallow pitched roof (see BS5427 at Appendices L2 and L3 of No 35/1 of process). In particular, the roof slope for these roof lights was too shallow, and no special precautions had been provided for this. They would inevitably leak, and they would continue to leak if Mr Fergusson's overcladding solution was adopted. This would be to repeat the same fundamental error, as the slope of the roof is too shallow. The roof lights were made from glass reinforced plastic, which expanded and contracted much more than the surrounding steel sheeting, causing distress to the steel and to fixings. Although the shape in profile of the roof lights was similar to the steel sheeting, it was not exactly the same, resulting in an inaccurate fit over the metal roof sheeting. This occurred in about 330 places on the existing roof, which compared with a total of 830 leaks.

[43] Mr Roberts explained the various defects in the existing roof as these were demonstrated in the photographs which form Appendix I to his first report (No. 35/1 of process). Many of these photographs showed an absence of end lap stitcher screws, gaps at end laps and scope for thermal movement there. Photograph 22 showed a tear in the roof light which allowed condensation to form and become trapped within the roof light. There was a particular problem at four-way intersections between roof lights and metal sheeting, as shown in photograph 25 and discussed at Section 3.5 of No. 35/1 of process. This was an inevitable problem of the present system, and the problem would be repeated in the proposed overcladding solution. Moreover, the overcladding solution did not make any provision for leakage around structural posts which project through the existing roofing, as shown in photograph 30, of which there might be perhaps two dozen in the entire roof.

[44] Kemperol had recommended their 2K PUR system in preference to their V210 system, partly because the latter gives off a strong smell as it cures, which could taint the food being processed below, and partly because the former can be overpainted with goose wing grey paint. In answer to a question from the Court, he explained that it would be impractical to provide an overcladding solution to this roof with a pitch of 6° or more, because of the size of the roof. Roof ridges would have to be raised by about 0.8 metres and the perimeter would need to be raised by a similar amount. Extensive steelwork would be required and additional bracing would also be needed. With regard to Mr Fergusson's proposal for overcladding, Mr Roberts expanded on the problems that he perceived with such a solution as discussed at Section 11.2 of his report (No. 35/3 of process). He confirmed that the Ashgrid support rail system is widely used for over roofing, but because the screws required to be inserted at an oblique angle of about 10°, and because they required to pass through 80mm of insulation, there was a risk of conflict, or that the screw might miss the purlin. However, his greatest concern was that the screws were not sealed, and water would find its way through the new screwhole into the building. There was a real risk that water would find its way through the outer skin, (by leakage from end laps, or leakage around roof lights, or condensation or ridges or penetration) and would lie on top of the composite panels of the existing roof. There it would be hidden from view, and inspection would not be possible. It would then find its way by redundant fixing holes into the building. This was a problem which was highlighted by one of the tendering contractors for the overcladding solution - see the letter dated 11 May 2007 from W F Price (Roofing) Ltd (No 6/21 of process), which drew attention to "a high risk of missing purlin lines and creating redundant holes in the composite panels". Vapour seal tapes, on which Mr Fergusson placed reliance, do not have a long life. They tear readily, and are not robust or reliable and tend to fail within five years. They would have to accommodate thermal movement in the roof; although this would be reduced to some extent by the overcladding roof, there would still be movement caused by cold weather, and the vapour tape would have to accommodate this. Mineral wool insulation would not improve the risk of condensation, as it allows vapour to pass through. If it became wet, the insulation material would act like a sponge. Mr Roberts anticipated that the first element of the overcladding system which would fail would be the roof lights.

[45] Turning to the Kemperol alternative, Mr Roberts had experience of this system being applied to coated steel sheeting, and provided that the proper primer was used it worked well. If the works specified in Appendices W1 and W2 to No. 35/3 of process were carried out correctly, there would be no reason for an engineer or surveyor to advise a lender that the value of the building would be reduced. Mr Roberts would not describe the Kemperol solution as a temporary patch-up solution; rather, it was a long-term resolution of a water penetration issue. It is an appropriate, considered and documented proposal for remedying the existing problem, and was documented and supported by guarantees.

[46] As described in paragraph 11.3.5 of his supplementary report (No 35/3 of process) the specification for the Kemperol solution required the fleece to be feathered back at the edges to minimise steps. This meant that each layer was progressively wider than the one below it. There would always be a short step on the upper side of the waterproofing strip, but this would not reduce the life of the system. With regard to Mr Fergusson's concerns about ponding, he accepted that there would be a step which would be apparent on such a shallow roof, and droplets of water would lie at the upper end of the Kemperol system. However, there would be a steelseal topcoat which would go beyond the Kemperol system, and this would protect the metal sheeting from any ponding. There would be retention of some droplets of water, particularly when the Kemperol was fresh and oily, and after time he would expect some dirt to accumulate there, just as happens at the junctions with roof lights. This would not be deleterious to the roof: the important thing was to stop the leaks. Mr Roberts was quite satisfied that the quotation from Pendlebury's included the provision of stitcher screws.

[47] If applied properly, Mr Roberts had heard of no problems with the Kemperol system. It was critical that it was applied by experienced contractors in suitable weather conditions. He estimated the life of the Kemperol system to be at least 20 years, and that there was no difference between the life span of the Kemperol system and that of the Plastisol coated sheeting of the roof. Kemperol was an expensive coating system compared to other waterproofing systems, and he believed it to be the top of the range. With regard to the costs of the Kemperol option, the costings at Appendix Z to No. 35/3 of process were provided by Briggs Amasco; he considered them to be a fair and reasonable estimate, although it would be reasonable to have some provision for contingencies, which might be properly stated at 10%.

[48] With regard to the ability of the Kemperol system to bridge gaps of the size found on this roof, Mr Roberts was confident that the inclusion of a bond-breaker tape in the Kemperol system would bridge these gaps. The references in Appendices X1 and X5 to No. 35/3 of process to the system being designed to bridge cracks up to 2mm did not concern him - the bond-breaker would bridge the gap and reduce any strain of movement on the Kemperol. Bond-breaker tape could accommodate thermal movement of 13mm. He described Mr Fergusson's concern about the distance from the end lap edge to the stitcher screws as "nonsense" and immaterial - the important point was that the bond-breaker would bridge the gap. An experienced contractor would easily be able to apply Kemperol to those places where "gun grade" silicon sealant had been applied and had raised the end lap edge by 1 or 2mm. Morever, in his opinion soft carbon steel such as that used for the construction of the existing roof sheeting would not have a memory which would cause it to spring back, as Mr Fergusson had suggested might be the case.

[49] Mr Roberts was asked to comment on the averments for the pursuers at the top of page 47 of the Closed Record (No 62 of process) to the effect that the available property market would not accept a building such as that of the pursuers with the Kemperol solution, and that the pursuers would have a considerably less valuable and not reasonably marketable building unless they overclad the roof. He disagreed with these averments and observed that he had given advice both to purchasers and lenders in relation to buildings to which the Kemperol system has been applied, to the effect that the Kemperol system was fine. His advice had been accepted, and he considered that the property market would accept this building with the Kemperol system.

[50] He summarised the advantages and disadvantages of the two options as follows. The disadvantages of Kemperol were that it depended on suitable weather conditions when it was being laid, so it could only be applied in spring or summer; it was dependent on an experienced and well-supervised workforce applying it; and there have been a limited number of quoted projects where the K2 PUR system has been used on metal roofing applications. Against these disadvantages, the advantages of the Kemperol solution were that it provided a flexible, continuous coating system which was compatible with the substrate; it would give an even colour of goose wing grey across the whole roof surface; and he was confident that it would resolve the problem of leakage at end laps, which was the principal issue on this roof. With regard to the overcladding option, the advantage was that it would give the appearance of a new, continuous metal roof. However, there were several disadvantages to it. It does not address the leakage issue at the composite panel end laps, where the slope of the roof is 4° yet the manufacturer of the roof lights requires a slope of 6°; the Ashgrid system requires oblique screw fixings, which would give rise to further leaks; condensation would continue to form in the underside of the metal roofing and would fall into the building; it would not solve the roof leaks which presently exist; and the numerous penetrations through the metal roof would require the application of Kemperol or another liquid coating to make them weather tight. He described the overcladding option as disproportionate to the problems of the roof; it would extend over the whole area of the roof, yet leaks are generally confined to end laps. The total length of end laps on the building was 1070 metres, and the widest part of the Kemperol coating was 350mm, so there would be almost 500 square metres of the Kemperol coating on the roof. In his opinion the best possible system to solve the problems on this roof was the Kemperol system, and he was satisfied that it would solve the problem. He did not consider that the overcladding system would solve the problem; it did not comply with standards, and it would leak.

[51] In cross-examination Mr Roberts agreed that the roof was not performing properly at present, but he was of the opinion that the overcladding option would repeat a lot of the problems in the present roof. What he was proposing was a hybrid between a waterproof and watershedding roof, which was not uncommon. He knew that the Kemperol system provided a permanent solution. It was put to him that a properly functioning overcladding system would be nearer to reinstatement to a new roof than patching with Kemperol; Mr Roberts replied that a properly formed liquid solution is today accepted as equivalent to overcladding. There were several causes of the problems in the existing roof - the shallow fall of about 4° (particularly for GRP inline roof lights, and the lack of fit where these join with the metal sheeting), the exposure of the site and the length of the roof. He considered that these were inadequacies of design of the original roof, and that this roof would always have required waterproof liquid protection at the junctions of roof sheets. It would not be impossible to design an overcladding system with a 6° fall (although, as explained earlier, this would give rise to complications, extra work and extra cost), but although this would resolve the problem about the pitch of the roof lights, it would not resolve the problem at four-way junctions. Detailed design of the overcladding option might reduce the amount of rainwater entering the building, but there would still be rainwater entering. Overcladding with continuous lengths of aluminium sheets (as proposed by Price Roofing in their quotation dated 13 October 2008, No. 6/16 of process) would solve the problem, and has been shown to be reliable at shallow slopes, but this is a quite different solution from that being proposed by Mr Fergusson and the pursuers.

[52] Condensation would continue to be a problem in the overcladding option. Mr Roberts is a civil and structural engineer and was qualified to calculate dew points, which he had done. He disagreed with the conclusions of the reports prepared by Morris Engineering Design Services and referred to by Mr Fergusson at paragraph 6.02.3 of No. 6/9 of process; Mr Morris' survey was carried out in May, but a humidity survey should be carried out during the winter months. Mr Morris concluded that the building had an internal environment classed as normal to low, but Mr Roberts had measured the humidity and assessed it as humidity class 3 (in a range 1-5). He disagreed with Mr Morris and took the view that there would be a dew point in the gap between the existing roof and an overcladding roof. The prospect of condensation would be reduced if there was a continuous vapour barrier, but vapour control tape has a short term life. Condensation would be sufficient to enter the building by end laps, side laps and screw fixings during the winter months. Mr Roberts accepted that the problem of structural members penetrating both roofs could be solved by careful details, and the problem of oblique screws could be resolved by ferrule fixings, but there was a need for a lot of development to work up Mr Fergusson's scheme.

[53] With regard to the ability of Kemperol to bridge gaps of more than 2mm, Mr Roberts maintained that the passage at paragraph 16.5 of the BBA certificate did not include provision of a bond-breaker. The Kemperol system with a bond-breaker would bridge a gap of more than 2mm. He had not asked Kemperol about gaps of more than 2mm, because his proposal was that a bond-breaker should be used, nor had he asked Kemperol if the system would bridge gaps of more than 2mm with a bond-breaker. He had however spoken to Kemperol about support, and he was satisfied that the use of a bond-breaker would provide adequate support. He was asked if he knew of any examples of the Kemperol system failing, and he replied that there was one area at the Chelsea Football Stadium site where an area of Kemperol had not fully bonded, because it was laid during a shower of rain. It was re-laid again and thereafter performed satisfactorily.

[54] A pitch of 4° was in Mr Roberts' opinion very shallow for a metal water shedding roof (such as the roof of the pursuers' premises) but not for a waterproof roof, which could be flat. He agreed that even small rises could result in ponding, but he did not consider that the step of 2-3mm caused by the waterproofing system would act as a dam, nor would it form a material obstruction - there might be a slowing of the water run-off, but there would be no issue as to waterproofing. Even if there was ponding above the Kemperol system, this would not be an issue for the integrity of the roof, as there would be two coats of steelseal coating above the Kemperol, and extending about 45mm beyond it. This steelseal paint had a life span of up to ten years, after which it would be necessary to repaint the steelseal. This coating is applied to give an even decorative finish, and would not be needed to protect against ponding. Mr Roberts accepted that it would have been desirable if both Kemperol and Pendleburys had visited the pursuers' building; he advised them to do so, but there were problems of obtaining access to the site in the summer of 2008. Neither had seen the site, although he had spoken at length to both, each had seen his photographs, and he agreed with them that the Kemperol system would work at the site.

[55] Mr Roberts agreed that it was undesirable that rainwater should stand on the composite panels, but the steelseal coating would protect against this. Ponding might affect the longevity of the plastisol coating of the panels, but his only concern about this would be if there was an industrial plant producing sulphuric acid within 20 kilometres upwind of the site. It was true that Corus recognised that long term ponding was an issue for their roofing material, and this might result in an exclusion from the Corus guarantee, but there was no guarantee here; in any event, the Corus guarantee is most unlikely to be useful, because it required an annual survey to be carried out and sent to Corus. However, he accepted that once the plastisol coating failed, after a period of many years, the underlying steel would rust. He confirmed that plastisol is a "plasticised PVC" for the purposes of Table 3 in the BBA certificate, and was therefore a suitable substrate and provided adequate adhesion for Kemperol. With regard to the roof lights, Mr Roberts accepted that they might have a life span of as much as 25 years. However, no professional could go along with inline roof lights such as these at a 4° slope in this degree of exposure. It would be possible to fit barrel-shaped or domed roof lights as part of an overcladding solution, but this was not what was being proposed on behalf of the pursuers, and in any event barrel-shaped or domed roof lights would rely on a reinforced liquid coating such as Kemperol to make the bottom end lap watertight.

[56] Mr Roberts accepted that it would be obvious to any surveyor looking at this roof if it was repaired by the Kemperol option that the roof had been repaired, and that this was not part of the original roofing system. A competent and experienced surveyor would know about the problems associated with composite roofing; he would note and record the treatment, and this would trigger further assessment of the roof.

[57] With regard to the issue of condensation arising in the overcladding option, Mr Roberts considered that Mr Morris should have referred to the British Standard on Controls of Condensation in Buildings, BS5250. Mr Morris was recording measurements during the summer, but the measurements should have been carried out during the winter when condensation problems would have been likely to be most acute. Mr Roberts was of the opinion that condensation, particularly in the area of the roof lights, was a real issue in the overcladding option, and he believed that condensation would continue to occur as it does at present. The lifespan of any vapour tape used to prevent such condensation would be about 5 years; after this time, one would be in the same position as if no vapour tape had been applied.

[58] In re-examination Mr Roberts explained that any question of ponding at the upper edge of the Kemperol system would be a small volume of water, amounting to droplets rather than gallons or litres. He did not believe that this would give rise to any realistic prospect of water ingress to the building, nor of any damage to the roof panels. Although the primary purpose of the steelseal coating was decorative, to enable the goose wing colour to match the rest of the roof, it had the happy by-product that it offered additional protection to the plastisol coating of the roofing panels.

[59] The fact that Kemperol and Pendlebury had not visited the site did not preclude them from properly assessing the nature of the job. They had seen his photographs and specification, and the job was in many ways similar to a job which they completed successfully on composite panel roofing in Lancashire. The costings in Appendix Z to his supplementary report were prepared by Briggs Amasco, who had visited the site. The surface coating of the overcladding option specified by Mr Fergusson was the same as that of the composite panels on the existing roof, so it had a 25 year life span before a decision to repaint was required. It was reasonable to proceed on the basis that either option would involve repainting in order to achieve a 40 year life span. If the cost of repainting the Kemperol solution with steelseal was added to the costs in Appendix Z to No. 35/3 of process, he estimated that this would add about £18,000-£20,000 to the total sum. Mr Roberts adhered to the opinion expressed in the conclusions to his supplementary report, namely that the Kemperol treatment was the correct way to resolve the problems of leakage on this roof, and that it was preferable to the overcladding option.

[60] Mr Thomas Robin Eardley Blois-Brooke was aged 57 and a consulting engineer. He became a chartered engineer in 1977 and joined the practice of William J Marshall and Partners, Consulting Engineers and Architects in 1979, becoming a partner in 1985. At the date of the proof he was senior partner of that firm. His curriculum vitae is set out at Appendix D to his report No. 15/4 of process. He spoke to the terms of that report and his supplementary report, No. 15/10 of process. He had experience of designing and advising on problems arising from several other large roofs similar to the one in question. He was acquainted with the other expert witnesses and had read their reports.

[61] In amplification of his reports, Mr Blois-Brooke explained that at paragraph 3.3 of No. 15/10 of process he was discussing a first recoat of the existing roof after a period of 25-30 years (the estimate of life not being a precise science) because of cosmetic appearance rather than because of any structural or protective requirement. Cosmetic appearance on this roof was a controversial subject itself, because the roof could not be seen from anywhere because of the surrounding parapets of the building. The existing roof would require to be repainted with a different material, such as steelseal or an equivalent, which would have an estimated life of about 10 years. The exercise of repainting was not particularly difficult, but it did involve the cleaning and preparing of the roof, and standing the size of the present roof, a large amount of work would be involved. By contrast to the metal roof panels, the expected life of the roof lights is 25 years, and when they reach the end of their life they become fragile and would probably require to be replaced.

[62] The principle of the pursuers' overcladding scheme caused him concerns. As currently formulated, he did not consider that it would solve the problems, and it would be difficult, if not impossible, to construct as envisaged. Several aspects caused him particular concern. There was a probability of the Ashgrid system screws clashing, or being so close as not to give purchase; it might work where the existing roof cladding is 40mm deep, but not where it was about 80mm deep, as it was over the production area. He observed that "somebody needs to think this through." There was a probability that the single sheet overcladding would deflect, particularly as it will be attached at 1.8m centres to match the existing purlins. Moreover, the overcladding roof would be at the same 4º pitch as the existing roof, and would still suffer from the same problems at endlaps, where water would be likely to be blown in. It was difficult to seal these with traditional butyl tape. The in-line roof lights would be installed at a pitch less than the recommended pitch of 6º. He agreed with the proposition that overcladding would simply repeat the problems experienced with the existing roof. Overcladding as proposed was not a solution to the problem - it does not work at present, and the overcladding proposal needed more work.

[63] The overcladding solution was a disproportionate response to the existing defects, which are predominantly confined to the end laps. He approached the matter of competing remedial options on the basis of what he would advise a client to do if there was no litigation and the client was paying for the works. He regarded overcladding as an excessive response to what is a local defect. The Kemperol option focused on the known area of defect and made it waterproof. It was an appropriate engineering solution. He knew Kemperol as a product, and had specified it on one project. He had used this kind of material for many years. He had spoken to several owners of buildings where Kemperol had been used, and all reported to him that they were satisfied with it and had experienced no problems with it. It has the reputation in the industry of being at the upper end of the market. He agreed with Mr Roberts' view that Kemperol was a better solution than overcladding - as he put it, it was a targeted drug, not a generic one.

[64] Under reference to Table 4 of the BBA certificate, Mr Blois-Brooke was of the opinion that Kemperol would cope with movements of up to 30%, and on this roof could cope with 5mm of thermal movement if a bond breaker was applied. He doubted if there would remain any gaps of as much as 5mm after the application of stitcher screws, which would have the effect of drawing the gaps together and which were included within the Kemperol option; however, if any such gaps remained, the bond breaker would bridge such a gap, whether it was vertical or horizontal. Provided that a proper bond breaker was incorporated, Kemperol could be used where gaps exceeded 2mm. He disagreed with the proposition that the internal structure of the Kemperol system is not enough to support its own weight; before curing, it has an inherent strength which can bridge a gap, and with the addition of a bond breaker before it is applied, additional substrate or support is provided. Once the bond breaker has lost adhesion, as it is designed to do, the Kemperol is sufficiently strong to bridge the gap. He had no concern about using Kemperol with a bond breaker over gaps of more than 2mm in width or height. The bond breaker together with the sealant were elastomeric; they comprised a bandage joint which he considered to be an appropriate solution. The references to gaps or cracks of up to 2mm in the Kemperol literature were references to the system without a bond breaker, whereas Mr Blois-Brooke was considering the system together with a bond breaker. The warning about using the neat material on cracks over 2mm wide does not apply if a bond breaker is used.

[65] The Kemperol system would not have any deleterious effect on the existing metal sheeting, nor should it affect any guarantee in that respect. Mr Blois-Brooke was aware that a competing manufacturer of similar panels had expressed no concerns about having Kemperol 2K PUR on the panels, and he was aware of another case in which Corus had stated that the application of Kemperol to the polyurethane covering of their metal sheeting would not affect their guarantee. With regard to the estimate of a service life of up to 25 years for the Kemperol system, as contained in paragraph 15 of the BBA certificate, Mr Blois-Brooke explained that BBA have a very good reputation in the industry, and it is quite an arduous task to obtain an agrément certificate for a product. Engineers and architects would take very seriously an assertion by BBA that a product has a service life of up to 25 years. Although estimation of life span is not a precise science, BBA tend to estimate in 5 year periods, so an estimated service life of up to 25 years might perhaps be between 20 and 25 years. He would expect the product to last longer than the manufacturer's warranty. In the present case, the original roofing panels were installed in 2003, so they will reach the end of their life in 25 to 30 years from that date, being 2028-2033. If Kemperol were to be applied in 2009, it would reach the end of its life in about 2034. Kemperol was therefore not a temporary repair, and would last for as long as the existing roofing, and indeed the 30 year commercial life of the building itself.

[66] With regard to the costings in Appendix Z to Mr Roberts' report No. 35/3 of process, Mr Blois-Brooke revised the terms of paragraph 3.15 of his own supplemental report (No. 15/10 of process) because he now understood that stitcher screws were included within the sums quoted for by Pendlebury's. He was familiar with the specification at Appendices W1 and W2 of No. 35/3 of process, and he considered these to be adequate to deal with the problems of the leaking roof, including the few (about 14 in number) other leaks which were not at endlaps. Although he was not a quantity surveyor, the figures were of the order that he would expect. In answer to a question from the Court, he accepted that it was appropriate to include provision for contingencies. This he considered should be approximately 10-15%, although he would not expect that this was all used up.

[67] Mr Blois-Brooke was asked to consider Mr Fergusson's concerns about ponding at the upper edge of the Kemperol system. He observed that as the Kemperol system was about 3mm thick, there would be a small amount of water lying to the upper edge of the Kemperol from time to time; however, the proposal was to feather off the Kemperol, so this would reduce the thickness to about 1.5mm. The water would evaporate, and Mr Blois-Brooke regarded this as de minimis. This was not really ponding as envisaged by Corus, who were concerned about "freeze/thaw action"; this would be a very thin sliver of water which would not have the robustness of mass to affect the body of plastisol. The prime cause of failure of plastisol was at a cut edge, where it can peel back and expose the underlying steel. There was no evidence of peel back or any other defect in these sheets. He would not expect any water lying at the top edge of the Kemperol system to have any effect. He disagreed with Mr Fergusson on this point, and also disagreed with him that such a small area of surface water would negate any Corus guarantee (if indeed these panels were covered by a guarantee). Moreover, in answer to questions from the Court, he agreed that all the water would be resting on the Kemperol feathering or the steelseal coating extending beyond it. He did not consider that there was any real risk of this water giving rise to water ingress or damaging the panels. He also disagreed with Mr Fergusson's concerns about the memory of the steel sheeting and whether the application of further stitcher screws to the end laps before the Kemperol was applied would be effective. He considered that the stitcher screws would pull the sheets together; the sheeting was quite flimsy and could be pulled down by stitcher screws effectively, as had been done in the past. There would be some tension, but, as he put it, the screws were "man enough" to overcome this. In his opinion Kemperol was the most appropriate solution to these problems; it was not a temporary solution, and it is the option which he recommended.

[68] In cross-examination Mr Blois-Brooke agreed that in appearance a functioning over clad roof was closer to the original functioning roof than Kemperol was, but he observed that these roofs are not visible to anybody unless one is actually standing on them. The way in which an overcladding roof would work, if properly designed, would be to prevent water ingress at the end laps. The advantage of Kemperol is that it makes these end laps watertight, so it would be an enhancement. If one could actually achieve a functioning over clad roof (about which he harboured doubts) he agreed that it would be closer to the original functioning roof - but the difficulty was that it would replicate the shallow pitch of that roof, with all the problems (particularly at end laps and roof lights) which exist in that roof. He accepted that the Kemperol system would signal to anyone that there had been a problem on the roof and that this was not an original element of the roof scheme, but it showed that the problem had been repaired. He agreed that a building owner would strive for a functioning overcladding roof, as the next best thing to the roof he contracted for, but he would need to bear in mind the costs and the alternatives. The pursuers' present proposals for overcladding were not a solution to the problem, because as currently developed they would not prevent water ingress. The pursuers' overcladding proposals required significant further work - they were not thought out fully at all. Mr Blois-Brooke described it as "a very poor scheme as it stands", and if implemented as it stands it would not solve the problems. However, it might be made to work, but would need significant further development. There were several aspects of the overcladding proposal which caused him concern:

He did not believe that Wards manufacture IP1000 single ply sheeting. In any event, the IP1000 panel has a particular profile, which would be inappropriate for a single sheet overcladding system. It has only 3 major corrugations per metre; single ply sheeting will tend to sag, and in order to provide a more robust panel it is necessary to increase the number of corrugations. The risk of sagging in this case is increased because there is a span of 1.8m between the purlins. It would not be advisable to introduce intermediate spans, as they would be supported off the panel and would involve trying to drill fixing screws through very thin sheets of metal. More purlins could be inserted, but this would result in a much more major and expensive job than that currently proposed. The current proposals are not sufficiently robust and will result in an over clad roof which sags.

The pursuers' proposals provide for the installation of stitcher screws at the end laps at each pan. A single stitcher screw was not enough to stop movement. The present proposals contain inadequate provision of stitcher screws at the end laps.

The problems of clashing of screws (and of potentially missing the purlins) inherent in the use of the Ashgrid system had already been discussed, and remained a problem.

The overcladding roof would retain the same 4º pitch as the existing roof. This would cause problems at the in-line roof lights (where the manufacturers required a minimum pitch of 6º) and also the potential for gaps at fourway interchanges. On a slope as shallow as 4º, water will tend to seek out these gaps. It will pond in the end laps, and butyl sealant does not function well in ponded water. In-line roof lights are to be avoided if at all possible, but if they were properly sealed (for example with the Kemperol system) it would be possible to use them. However, a proper system of sealing the roof lights would require to be specified. It would also be possible to use barrel roof lights with proper detailing, but this was not what was specified. At all points of penetration it would be necessary to use a waterproofing system such as Kemperol.

Mr Blois-Brooke considered that a design for an overcladding roof could be made to work, but the pursuers' present proposals would have to be developed considerably.

[69] With regard to other concerns that had been raised about the overcladding option, he considered that Mr Morris in his two reports about condensation focused on the wrong point. The concern was not the risk of condensation on the inside face of the insulated panels, but rather to prevent penetration of warm moist air from the building into the interstitial area between the two roofs, because if it did so it would meet the cold single sheet of overcladding metal which would cause condensation and drips. Mr Blois-Brooke considered that the use of vapour tape should work, and also that insulation would serve to reduce the risk of condensation. Mineral wool insulation, when it gets wet, gets heavy and one would have to be careful about such extra weight, but there are other forms of insulation available. However, he observed that all these items are moving further and further from the scheme as presently proposed. He agreed that the overcladding proposal would tend to prolong the life of the existing roof lights, but some of the existing condensation in the roof lights may arise from inside the building itself, which would not be addressed by the overcladding scheme.

[70] Turning to the criticisms made of the Kemperol option, Mr Blois-Brooke accepted that the Kemperol system should be applied to a dry substrate. The references to its capacity to bridge a gap of up to 2mm were all references to the "neat" system, without any bond breaker. A bond breaker was clearly part of the proposals for the application of the system in the present case. He had spoken to the manufacturers of Kemperol and had discussed the existing gaps with them; part of the proposals in the Kemperol option included drawing these gaps together with adequate numbers of stitcher screws, but in any event the manufacturers expressed no concern about the capacity of their product to bridge these gaps. He accepted that it would have been helpful if Kemperol and Pendlebury had visited the site, but each had seen the photographs and indicated that there would be no problems. He did not agree with the suggestion that if the Kemperol system was applied to a gap of over 2mm, the Kemperol guarantee would not be available - Kemperol were aware of the circumstances, and if the system was applied with a bond breaker they would stand by their guarantee. What was being proposed was not an unusual scheme of works.

[71] Mr Blois-Brooke had observed gaps at end laps larger than 2mm on this roof, but stitcher screws would be effective in pulling together the 2 pieces of metal, as they had been effective in the past. He would place the stitcher screws closer than 50mm to the edge. Any water sitting on the upper edge of the application would be on the Kemperol or on the steelseal coating, and not on the plastisol of the roofing sheets. It would require a much larger gap than any he had seen on this roof to extend any water beyond the limits of the Kemperol system. The durability of the system might possibly vary with its situation, but the situation of these premises was not unusual. He had no concerns about the adhesion of the Kemperol system to the plastisol covered sheeting.

[72] With regard to lifespans, the original plastisol covered sheets had a lifespan of 25-30 years before requiring repainting. The roof lights had a lifespan of up to 25 years, when they would require to be replaced. Kemperol would have a lifespan which would expire at about the same time, or later. When the Kemperol began to fail, the pursuers would have two options - either they could cover it with another layer of Kemperol, or they could remove it and start again. Either would result in some expense, but they would have had problems with replacement of the existing butyl sealant in any event. He favoured the option of leaving the Kemperol on and applying a further coating of it, which would be feasible in the circumstances. Kemperol came with a 10 year workmanship guarantee and a 20 year materials guarantee; these guarantees would apply in this case, because the Kemperol would be applied with the approval and knowledge of the manufacturers, who were fully aware of what was being proposed.

[73] In response to the suggestion that other experienced construction professionals might take a different view regarding Kemperol, Mr Blois-Brooke observed that this would depend on their experience of remedial work. If they did not have much experience in advising on this, they might not be aware of the advantages of Kemperol; however, experienced people in this field all consider that this is a good product, and he did not think that any person experienced in this field would express a different view. Mr Fergusson has had experience of remedial works, and although he raised a few issues about Kemperol Mr Blois-Brooke did not regard him as giving a damning condemnation of system. In Mr Blois-Brooke's view, Kemperol would be the appropriate solution to this problem.

[74] In re-examination, Mr Blois-Brooke expressed the view that if the overcladding option was fully developed as it required to be in order to render it viable, this would add significantly to its cost. He was unable to estimate this increase in cost, but he had in mind something in the range of an increase of 20-50% over the figures given by the pursuers.

[75] On the question of the likely lifespan of vapour tape, he observed that this required to be equivalent to the lifespan of the overcladding system, ie. 25 years. He was sure that such a vapour tape could be procured; it would be situated in a relatively protected environment under the overcladding roofing, and it should be possible to obtain tape with such a lifespan.

[76] With regard to the question of proportionality, Mr Blois-Brooke was asked if he regarded it as reasonable for the pursuers to adopt the overcladding solution and all the extra works that would be required to develop this. He stated that he considered that the original overcladding proposal, as developed to date, was an excessive response, and that if it cost between 20-50% more to develop it fully, then it was all the more excessive, when a practical and available scheme which would solve the problem existed in the form of the Kemperol option.

Submissions by counsel
[77] In addition to full submissions presented in Court, both senior counsel very helpfully provided me with a written outline of their submissions, for which I am most grateful. These written outline submissions are in the Court process. I took account of both oral and written submissions for each party; what I seek to do here is simply to provide as brief an outline as possible of the positions adopted on behalf of the respective parties.

Submissions for the pursuers
Senior counsel for the pursuers submitted that they were entitled to be put in the same position, so far as money can achieve this, as if the contract had been performed. The normal measure of damages in respect of a defective building is the cost of reinstatement, i.e. putting the building into its contracted for state. It is only when the cost of reinstatement is unreasonable and out of all proportion to the benefit to be obtained by reinstatement that reinstatement no longer becomes an appropriate measure of damages. The pursuers' choices should not be weighed too finely, and they were entitled to reinstatement unless this was seriously disproportionate. Senior counsel accepted that the pursuers were under a duty to mitigate their loss, so if an alternative to overcladding was as close to the contractual standard, there may be a failure to mitigate. Where a cheaper solution was found to be as good as, or better than, the solution proposed by the pursuers, damages may be assessed on that basis. However, the onus is on the defenders to show that the remedy is disproportionate unless this is obviously apparent. Incidental betterment falls to be disregarded. The question for the court is whether, in all the circumstances, it is reasonable for the pursuers to overclad the roof, and the decision should not be weighed in too fine a scale; the onus rests on the defenders to show significant disproportion. In accordance with general legal principle, a man may not benefit from his own wrong. In support of these propositions, senior counsel referred me to Ruxley Electronics and Construction v Forsyth [1966] 1 AC 344; East Ham Corporation v Bernard Sunley & Sons Ltd [1966] AC 406; McLaren Murdoch & Hamilton Ltd v Abercromby Motor Company 2003 SCLR 323; McGlinn v Waltham Contractors Ltd (No. 3), [2007] EWHC 149 (TCC) [2008] 111 Con LR 1; George Fischer v Multi Design Consultants Ltd [1998] 61 Con LR 85; Harbutts Plasticine Ltd v Wayne Tank and Pump Co Ltd [1970] 1 QB 447; Alghussein Establishment v Eton College [1998] 1 WLR 587; Hudson's Building & Engineering Contracts (11^th edition) paragraphs 8.119-8.122; and McGregor on Damages (17^th edition) at paragraphs 7-019 and 1-027.

[78] Senior counsel submitted that the overcladding solution was reasonable in the circumstances of this case, because it would put the pursuers back as nearly as may be to the position if the contractual standard had been achieved. The pursuers had paid approximately £13 million for a new building, and the roof has failed since they took possession of it in 2003 to fulfil its basic function of keeping water out of the building. The contractual standard to be achieved was a new roof which functioned to keep water out with an expected leak-free lifespan without significant repair of about 25 years, and longer with some repairs. Even if the Kemperol solution was technically adequate, it was clearly a repair solution - it was a "bandage joint", and not a material which is specified for such joints on new roofs. It would be obvious that the roof had been repaired. On the evidence of Mr Oswald and Mr Gallagher (which was essentially unchallenged) this would have an adverse effect on the value of the property. A potential purchaser of the property would not find a 10 or 20 year repair acceptable. The Kemperol bandage joints would draw attention to the fact that that there had been wide scale repairs to the roof, which would not be regarded as being as good as a new roof covering. This provided powerful justification for the pursuers' wish to overclad the roof.

[79] The test as to whether the overcladding solution is disproportionate as advanced by Mr Blois-Brooke was not correctly stated. Instead of approaching the question on the basis of what he would have advised a client who was having to pay for the repair themselves, as Mr Blois-Brooke did, the appropriate test was to ask what would put the pursuers in as nearly the same position as if the contract had been properly performed. What a client may do if he is not able to recover damages may be quite different from his entitlement in damages for a breach of contract. In any event, the cost of overcladding was not disproportionate to the benefit to be obtained. Although there was a significant cost differential, when the contract price of the building is taken into account, and also the effect on the value of the building by the use of the Kemperol solution as opposed to the overcladding solution, the cost of overcladding is not disproportionate. With regard to the costs of the overcladding scheme, Mr Ellis suggested as a starting figure the sum brought out at page 6 of the revised quotation from Price Roofing dated 13 October 2008 (No. 6/16 of process), being £757,579.00. In addition to this he accepted that there would require to be provision for protective works, and that the sum would fall to be increased by between 20% and 50% to reflect extra design works to provide a fully worked up proposal. He submitted that such a developed proposal was properly within the pursuers' averments, which were wide enough to enable detailed aspects of design to be changed. For example, it may be the case that IP1000 was not manufactured as single ply sheeting, but a different number or configuration of corrugations was simply a modification or development of what was already on record, and did not amount to a major departure or change. (In the course of his submissions, Mr Ellis moved to amend the pursuers' pleadings in the penultimate line of page 45 of the Closed Record, No. 62 of process, to give effect to this submission. His motion to amend was not opposed, and I granted it.) It was accepted that the defenders might have a stronger case that the pursuers were being unreasonable in maintaining their preference for an overcladding solution if that solution could not be made to work; however, it was submitted that this was not the case, and the pursuers' specification to tenderers for the overcladding solution was in very general terms.

[80] By comparison to the figures indicated above for the overcladding solution, the costs of the Kemperol solution on the basis of Appendix Z to No. 35/3 of process appeared to be in the region of about £131,755.00, or thereby. Senior counsel emphasised that both these figures were "ballpark" figures and had not been subjected to scrutiny by quantity surveyors, but he accepted that they gave an indication of the approximate scale of the costs of the alternative schemes. Senior counsel submitted that the defenders had not discharged the onus on them of establishing that the overcladding scheme could not work. Moreover, even if the defenders had succeeded in establishing that Kemperol was a technically adequate solution, it does not follow that it would be unreasonable for the pursuers to proceed with overcladding, because this remains the solution which would put the pursuers back in the position, as far as money can do this, that they would have been in but for the breach of contract. Any design problems with the overcladding solution can be resolved; by contrast, there were significant drawbacks with the Kemperol proposal. The steelseal paint would have a 10 year lifespan. The Kemperol itself would require to be renewed or replaced in about 2028 to 2033 with additional expense. There are concerns about the durability of the Kemperol system, and workmanship is only guaranteed for 10 years and material for 20 years. There is nothing in the literature relating to Kemperol to suggest that it is capable of bridging gaps of more than 2mm with a bond breaker - the literature suggests that a 2mm gap is the limit of the Kemperol system's capacity, and no one has asked the manufacturers about this. Neither the manufacturers nor the proposed contractors have visited the site. Ponding of water remains an issue for the effectiveness of the Corus guarantee. The pursuers and their advisers were entitled to take the view that there was insufficient certainty about the performance of Kemperol that it should not be regarded as a satisfactory remedial solution.

Submissions for the defenders and third parties
[81] Senior counsel for the defenders and third parties accepted that the measure of loss in the case of a building contract is prima facie the cost of cure. However, this general rule was subject to the important qualification that the cost must be reasonable - if the cost of repair was out of all proportion to the benefit, damages would fall to be measured otherwise. As Lord Bridge of Harwich observed in Ruxley Electronics,

"to hold in a case such as this that the measure of the building owner's loss is the cost of reinstatement, however unreasonable it would be to incur that cost, seems to me to fly in the face of common sense."

Senior counsel accepted that he did not challenge Mr Bartlett's evidence that he intended to adopt the overcladding solution if awarded damages, although he suggested that there was a question mark over the pursuers' intention in this regard. However, the pursuers' contention that they were entitled to overcladding as the next best thing to a new roof was a misleading gloss on the rule - they were only entitled to reinstatement if this is reasonable in all the circumstances. If a cheaper option was technically suitable for the task, the law will not allow the pursuers to adopt a more expensive option, as it would not be reasonable to do so. He submitted that where there is a competition between a cheaper option and a more expensive option, if the pursuers wished to adopt the more expensive option, the onus rested with them to show that it was reasonable to do so. In the present case, the onus rested with the pursuers to satisfy the court that the Kemperol option would not work. In support of these submissions, senior counsel relied on the cases referred to above, and in addition, referred me to Iggleden v Fairview New Homes Ltd [2007] EWHC 1573 (TCC); Southampton Container Terminals Ltd v Schiffahrtsgesellshcaft "Hansa Australia" MbH & Co (the "Maersk Colombo") [2001] 2 Lloyd's LR 275 [2001] EWCA Civ 717; and Pegler Ltd v Wang (UK) Ltd [2000] 70 Con LR 68.

[82] Turning to an assessment of the evidence, senor counsel made no challenge to the credibility of Mr Fergusson; however, he was by experience and qualifications an architect, and he was giving expert evidence on an area principally of engineering. His expertise was not equivalent to that of Mr Roberts on this type of roof. Senior counsel suggested that he was at (or beyond) the limits of his field, and he had no good answer to the criticisms made by Mr Roberts and Mr Blois-Brooke to the effect that the overcladding solution which he proposed did not work. Where there was any conflict between Mr Fergusson on the one hand and Mr Blois-Brooke and Mr Roberts on the other, senior counsel invited me to prefer Mr Blois-Brooke and Mr Roberts. Mr Fergusson was forced to accept that his scheme was embryonic, and that it would repeat the problems of the existing roof. By contrast, Mr Roberts was very experienced in roofing problems, and his evidence could be relied on with confidence. He set out clearly his criticisms of the overcladding solution, and his concerns were reiterated by Mr Blois-Brooke. Senior counsel accepted that Mr Roberts' attitude towards cross-examination might with some justification be criticised, but this may have been because he was tired after two days of giving evidence. In any event, both Mr Blois-Brooke and Mr Roberts were firmly of the view that the overcladding scheme did not work. Mr Blois-Brooke gave his evidence in a very measured, credible and reliable way, and was an impressive witness.

[83] Considering the various factors raised in the authorities, the aesthetic merits of the two competing solutions are irrelevant, as this roof is hidden by the parapets of the building and cannot be seen at all. There is a substantial differential between the costs of the two options: unless the Kemperol system does not work or there is another good reason which outweighs this price differential, it would be reasonable to prefer the Kemperol option. The evidence of Mr Oswald and Mr Gallagher should not be accepted as to the diminution in value arising from the Kemperol solution. Mr Oswald's evidence as to valuation was based on Mr Gallagher's advice; that advice, as attached to No. 6/12 of process and as spoken to by Mr Gallagher, was badly flawed. The overcladding solution would alert any building surveyor to the fact that there had been problems with this roof, just as much as the Kemperol solution would do. The Kemperol solution was not temporary, and its lifespan would match that of the existing roof. Both Mr Blois-Brooke and Mr Roberts had experience of advising owners of buildings with defective roofs which had been repaired with Kemperol, and their advice had been made available to lenders and potential purchasers. Their understanding of the Kemperol system was vastly greater than that of Mr Gallagher, and their evidence should be preferred to his. Any competent adviser would say that there was no diminution in the value of the building as a result of the Kemperol solution being used.

[84] The pursuers' position was that there were a number of areas of uncertainty relating to the Kemperol solution which justified their preferring the overcladding solution despite the large discrepancy in cost. However, the evidence about these alleged areas of uncertainty was unsatisfactory. With regard to lifespan of the Kemperol system, the evidence was that it had a probable lifespan of 20-25 years. This was supported by the BBA certificate and by both Mr Roberts and Mr Blois-Brooke. If repairs were necessary to the Kemperol at all, they would not be required before the roof lights needed to be replaced. There would be no extra maintenance work. It was not the case that when Kemperol came to the end of its lifespan there was a risk that the panels would be damaged, because the preferred solution would be to overlay with a further layer of Kemperol. There was no extra burden of maintenance, because steelseal would require to be applied to the Kemperol just as it will require to be applied to the whole roof.

[85] With regard to durability, it mattered not that neither Kemperol or Pendlebury had visited the site - they had the specifications and photographs and had discussed the problem. There was no substance to any concern as to the durability of the Kemperol product, which was the subject of the BBA certificate and also the manufacturer's warranty. There was nothing about this site which should cast doubt on the anticipated durability of the product. With regard to guarantees, Mr Blois-Brooke's evidence was important - he regarded guarantees as an extra or bonus, but the important object was to obtain a system that actually works, as Kemperol does.

[86] With regard to Kemperol's ability to bridge gaps, it is accepted that at present there are gaps in excess of 2mm, some being 5mm and possibly some even to 10mm. How many would continue to exist after the installation of adequate numbers of stitcher screws and fixer screws is open to question. Mr Blois-Brooke observed that the stitcher screws would be "man enough" to hold down the steel panels, and he discounted the suggestion that the thin steel sheeting would retain a memory which would exert sufficient tension on the screws. Mr Roberts also expressed the view that the stitcher screws would be adequate to reduce these gaps. In any event, even if there are some gaps remaining, the bond breaker would be an adequate solution to the problem. It would provide initial structural strength until the Kemperol was cured, and it would cope with thermal movement and easily bridge the gaps, whether these were horizontal or vertical. The standard Kemperol 2K PUR system does not include a bond breaker and so such a bond breaker is not part of the system certified by the BBA. The court should have no concerns about Kemperol's ability to bridge a gap of more than 2mm.

[87] Ponding was not a serious issue - the amount of any water at the upper edge of the Kemperol strip would be minimal, amounting to no more than slivers or droplets. Most would run off or evaporate. Any water would not be resting on the plastisol but would be on the steelseal covering. Any such water would have no prospect of damaging the plastisol nor of causing any water ingress, and if there were any Corus guarantees in relation to this roof, they would not be affected by this.

[88] In light of a proper analysis of all of the above features, it was apparent that Kemperol was a technically effective solution and adequate to resolve the problem, so there was nothing to make it reasonable for the pursuers to adopt a more expensive system, irrespective of the problems of the overcladding system. When one takes into account the problems with the overcladding system, which was at present embryonic, incomplete and in its present form (to use Mr Blois-Brooke's phrase) "it isn't a solution", there is even less reason to justify the additional expenditure associated with the overcladding option. If it can be made to work, it would be between 20% and 50% more expensive than the costings presently before the court, there would be problems associated with it, and there would be no real benefits to it. It hardly bears comparison with the Kemperol solution, which is well thought through and complete.

Response for the pursuers
[89] In responding on behalf of the pursuers, senior counsel observed that the "Maersk Colombo" case related to moveables, to which different rules applied, and for which the prima facie measure of damages was market value, not reinstatement. The onus of establishing that reinstatement was reasonable might rest on pursuers in such a case; in a case such as the present, reinstatement was the norm, and the onus rested on the defenders to show that the pursuers' preferred solution was unreasonable. The pursuers' attitude towards reasonableness should not be weighed in too fine a balance. The mere stopping of leaks in the roof did not necessarily amount to restitutio in integrum, and in the present case it does not amount to this, for the reasons already explained. Mr Blois-Brooke's approach to proportionality, based as it was on advice that he would give to a client who was paying for repairs to the roof himself, was flawed. The proper test was to put himself in the position of advising the reasonable owner of a building who is the victim of a breach of contract; someone faced with a leak in a building might simply choose to plug the leak, but if he has a contractual right against someone, he might quite properly choose to do more than this. What one is measuring is whether their response to a breach of contract is proportionate or not.

Discussion
[90] I deal first with my assessment of the evidence. In doing so, I make it clear that no questions of credibility arose in relation to any of the witnesses - I was satisfied that each of the witnesses was doing his best to be truthful and to be of assistance to the Court. Indeed, neither counsel made any real suggestion to the contrary: the issue before me turned largely on an assessment of competing expert evidence. I accepted the evidence of Mr Alexander Colvan, Mr Colin Campbell and Mr Ronnie Bartlett in its entirety; in particular, I accepted that the present roof has never functioned properly to keep water out, and that there have been serious problems of water ingress to the building since the pursuers took possession of it. I also accept that it is the pursuers' strong preference, on the basis of the professional advice which they have received, to proceed with the overcladding solution and that they intend to do so in the event that damages are awarded on this basis. I also accepted the evidence of Mr William Hope as to the tender process for the overcladding works and his report on the revised tenders.

[91] I did not feel able to place as much reliance on the expert opinion of Mr Iain Fergusson as I did on the opinions of Mr Roberts and Mr Blois-Brooke. I did not consider that Mr Fergusson had such an intimate knowledge and understanding of the Kemperol 2K PUR system as the latter two experts. He accepted in cross-examination that he had not considered the Kemperol 2K PUR solution when he proposed his overcladding solution, because at that time, the suggested alternative was the Aperture System, which he agreed was quite different from the present proposal. However, he retained concerns and doubts about the Kemperol solution as presently proposed. I discuss these concerns in greater detail below, but they related to several factors - ponding of water and slowing of water run off, requiring extra maintenance and possibly affecting the Corus guarantee; the ability of Kemperol to bridge gaps of more than 2mm (particularly given the "memory" of distorted metal sheeting); whether a debonding layer would provide adequate structural support; the durability of the Kemperol system, together with the need to strip it off and replace it at the end of its life, with the risk of causing damage to the roofing sheets; whether silicon was a suitable substrate for the Kemperol system; and whether the substrate could be rendered sufficiently dry for Kemperol to be applied, standing the amount of water secreted under the end laps. I formed the view that these concerns arose from Mr Fergusson's less than complete understanding of the Kemperol solution being proposed. By contrast, his position in relation to the overcladding solution was to accept that it was a two-stage process and that the second stage had not yet been reached, and that the proposals were at present embryonic. He did not appear to consider it incumbent on him to present a fully developed, worked out proposal for overcladding, and his response to any criticisms of the present proposals was that these were matters for the contractors to sort out in due course. Any problems associated with creating 49,000 new fixing holes, together with the problems of screws passing at an oblique angle and risking clashing with existing screws and missing the purlin, were something which the contractors would have to do properly. He conceded that there would be difficulties with the roof lights, but these would not be insuperable. He appeared to have no concerns about Mr Morris' dew point calculations. In re-examination he conceded that although he suggested the Ashgrid system, it was always a potential problem affixing screws through an existing roof at an angle and connecting them to an unseen location, and it might be that there was another system which could be used more effectively. Ultimately, I reached the view that Mr Fergusson's expert evidence about the Kemperol solution was less than satisfactory because he did not have an adequate knowledge or understanding of it, and his evidence about the overcladding solution was unsatisfactory because his proposals were not sufficiently developed to be properly tested as to their effectiveness, practicality and cost.

[92] Mr Andrew Oswald's evidence as to the effect on the value of the premises as a result of the two alternative solutions depended entirely on the information and advice which he had received from Mr John Gallagher as to the nature of the two options. It was Mr Gallagher's view that the sealant system (which he described as Kemperol V210) was a "short term patch", and this was what Mr Oswald factored into his valuation. I did not feel able to place reliance on Mr Gallagher's evidence. He did not appear to be aware of the lifespan of the Kemperol system, nor about the technical capabilities of the system. He did not appear to have carried out any evaluation of the alternative overcladding solution, and whether it would be capable of solving the problems of water ingress to the building. He appears to have based his opinion on the view that the Kemperol proposal was a temporary repair, and that it would prompt any surveyor to carry out further investigations. He did not address the question of why the overcladding option would not have the same effect.

[93] Mr Keith Roberts was extremely well qualified to give evidence as to the technical merits and demerits of the two solutions under consideration, and had specialised for some years in disputes relating to roofing and cladding. He had detailed knowledge of the Kemperol 2K PUR system, he had experience of its use, and he had seen it applied on site. He was also very familiar with the products used in the existing roofing, and had carried out a detailed inspection of the site and a detailed assessment of the overcladding solution proposed by Mr Fergusson. I did have some concerns at times during Mr Roberts' evidence as to the way in which he answered questions in cross-examination; he was very reluctant to concede anything at all, at times he appeared evasive and reluctant to answer counsel's questions, and on occasions he appeared to fence with counsel in a way which is not usually associated with an expert witness giving evidence in these courts. However, his evidence lasted for some 21/2 days, and I formed the impression that this was much longer than he had anticipated. Towards the end of the second day and into the third day of his evidence he may have suffered from some tiredness. Despite these concerns, I found Mr Roberts to be an impressive witness as to the technical aspects of both the overcladding solution and the Kemperol solution, and I felt able to rely on his evidence.

[94] Mr Thomas Blois-Brooke was also highly qualified and greatly experienced in this field. He gave his evidence in a very measured and impressive way, without exaggeration and with clear and persuasive reasoning. I found him to be a most impressive witness, and felt able to rely on his evidence in almost its entirety. His evidence coincided generally (although not in every detail) with that of Mr Roberts. In so far as there was a conflict of evidence as between Mr Fergusson, Mr Oswald and Mr Gallagher on the one hand and Mr Roberts and Mr Blois-Brooke on the other, I preferred the evidence of the latter two witnesses.

[95] Against the foregoing assessment of the evidence before the Court, I now turn to the technical merits and demerits of each of these proposals. With regard to the Kemperol option, the experts were agreed that this would require to be laid in suitable weather conditions, which would mean that it would have to be laid in spring or summer. It requires to be laid by an experienced workforce; I am satisfied from the evidence of Mr Roberts and Mr Blois-Brooke that Pendleburys are suitable contractors with experience in applying Kemperol. Although it would have been preferable for Pendleburys to have visited the site before giving their quotation, I am satisfied that they had enough material by way of photographs and discussions with Mr Roberts and Mr Blois-Brooke to enable them to assess the task adequately. Kemperol also requires to be applied to a dry and suitable substrate. Plastisol is a suitable substrate for this purpose. I do not consider that there is a real risk to the proper application of Kemperol from the extrusion of water presently underlying the end laps, as suggested by Mr Fergusson. His concern was that the effect of the contractors' employees working at the end laps would tend to cause the water which is presently under the end laps to be extruded, causing the substrate to be sufficiently wet to be unsuitable for the application of Kemperol. However, the proposal includes provision of additional stitcher screws before the application of Kemperol. This will, I am satisfied, increase the rigidity of the roof at the end laps, reduce or eliminate existing gaps, and should result in minimal risk of material extrusion of water.

[96] I do not consider that there is substance to Mr Fergusson's concerns about ponding of water at the upper edge of the Kemperol strip. Although there will be some water at the upper edge of the Kemperol strip, I am satisfied that the volume of this water will be minimal - as it was put in evidence, it will amount to "droplets" or "a sliver of water". None of this water will be lying on the plastisol coated sheeting; it will all be on the feathered Kemperol or the steelseal covering which would extend beyond it. I do not consider that ponding of water will have any impact on the effectiveness or durability of the plastisol coated sheeting, nor do I consider that it would have any effect on the Corus guarantee (if there were such a guarantee effective in respect of this roof) nor any other such guarantee.

[97] I am satisfied that the concerns expressed about the ability of the Kemperol solution as proposed (i.e. including the provision of a bond breaker) to bridge gaps at end laps such as those seen in the various photographs of this roof are unfounded. First, I am persuaded that the provision of additional stitcher screws as proposed would have the result of pulling these gaps together, with the consequence that they will be closed or substantially reduced. Mr Fergusson was concerned that the existing roof sheeting would have a "memory" and would tend to spring back to its distorted position, thereby re-opening the gaps and putting strain on the Kemperol. If such thin metal sheeting does indeed have a "memory", I am persuaded by the evidence of Mr Roberts and Mr Blois-Brooke that the additional stitcher screws would be "man enough" to hold the end laps together, so that the gaps would be closed or reduced. I am also persuaded that the inclusion of a bond breaker in the Kemperol proposal means that it will be able to bridge gaps significantly in excess of 2mm. I am satisfied that the reference in the Kemperol literature and the BBA certificate to the ability of the product to bridge gaps up to 2mm relates to the "neat" product without the inclusion of a bond breaker, and that if a bond breaker is included then larger gaps can be adequately and satisfactorily bridged. I do not accept the suggestion that the Kemperol solution as proposed, including a bond breaker, would have inadequate structure to support itself; for the reasons explained by Mr Roberts, I am satisfied that the bond breaker would provide sufficient support until the Kemperol system has cured, and thereafter the Kemperol is capable of supporting itself over any remaining gaps at end laps. It is elastomeric, and capable of coping with any thermal movement that it is likely to encounter.

[98] The durability of the Kemperol proposal was another issue. I am satisfied that the Kemperol proposal will last for between 20 and 25 years from the date of its application. Once it has reached the end of its lifespan, the decision will have to be made whether to remove it and renew it, or to apply a further coating of Kemperol on top of it. If Kemperol were to be applied to this roof in 2010, it would therefore have a lifespan until about 2030-2035. On the evidence, this is longer than the anticipated lifespan of the roof lights, and decisions about replacement of the Kemperol or application of further Kemperol will require to be made at about the same time as decisions regarding maintenance of the plastisol coated roof sheeting. There is nothing in the geographical situation of the premises which would affect the estimate of the durability of the Kemperol solution.

[99] Kemperol is a waterproofing (as opposed to a water shedding) roofing system, and will accordingly work effectively to prevent water ingress on flat roofs, or roofs with a very shallow pitch. It will operate effectively on a roof with a pitch of 4°.

[100] Turning to the overcladding option, the proposal is to provide an overcladding roof at the same pitch as the present roof. I am satisfied that this will tend to replicate many of the problems experienced with the existing roof. The roof lights will remain at a 4° pitch, although designed for a minimum 6° pitch. The problems at the four-way junctions between roof lights and the metal cladding sheets will tend to be replicated. Moreover, the problem of water being blown into end laps at a 4° pitch may also be replicated (although this may not be so much of a problem if adequate stitcher screws are provided). Although it would be possible to construct an overcladding solution with a pitch greater than 4°, nobody appears to be suggesting that this should be done; it would involve increasing the height of the ridges and roof parapets substantially, it would be very much more expensive than the present overcladding proposal, and it would be a significant departure from it.

[101] The proposal is to overclad with a single sheet of metal sheeting. I am satisfied on the basis of Mr Blois-Brooke's evidence that there will be an increased risk of sagging and/or deflection of such a thin single skinned roof, particularly as there is a span of 1.8 metres between the purlins. Although additional purlins could be inserted, this would result in a much more major and expensive job than that currently proposed. Moreover, the present proposals contain inadequate provision of stitcher screws at end laps, which will not be enough to prevent movement.

[102] The single sheet of metal overcladding will be supported on the Ashgrid system. This will require about 49,000 new holes to be drilled into the existing roofing, and fixing screws applied at an oblique angle through the existing roofing to the purlins below. I am persuaded that there is a real risk that these screws will clash or will miss the purlins completely. Mr Roberts accepted that it might be possible to affix an overcladding roof and avoid these problems by careful detailing, the provision of ferule fixings and very considerable development work; Mr Blois-Brooke was of the view that there was a probability of the Ashgrid system screws clashing, or being so close as not to give purchase, and that while it might work where the existing roof cladding is 40mm deep, it would not work where it was about 80mm deep, as it was over the production area. I am satisfied on the basis of this evidence that while it may be possible to provide an overcladding solution, the problems of affixing a single sheet of metal overcladding to the existing roof by means of the Ashgrid system are such that the overcladding roof will not function to prevent ingress of water to the building.

[103] The existing inline roof lights will be retained in the overcladding proposal, despite being at a pitch less than the manufacturer's specified minimum pitch. This is likely to result in continuing problems of water ingress. As observed by Mr Roberts, it would be possible to specify domed or barrel shaped roof lights, which would tend to give rise to fewer problems than the existing roof lights; however, this would be a significant alteration to the overcladding proposals, and would add further to the costs. Moreover, any such roof lights would still require the application of Kemperol or another waterproofing system at end laps.

[104] Some time was spent on discussions regarding the possibility of condensation continuing as a problem under the overcladding solution. I do not attach any weight to the reports regarding condensation and dew points prepared by Morris Engineering Design Services and referred to at paragraph [52] above; I accept Mr Roberts' criticisms that these were based on tests carried out at the wrong time of the year, and focused unduly on the risk of condensation on the inside face of the insulated panels, instead of in the interstitial space. I am of the opinion that there is likely to be a continuing problem of condensation in the space between the two layers of roofing, which requires to be addressed. Indeed, Mr Ferguson accepted this in evidence, but was of the view that it would be resolved by the application of vapour tape, and that the provision of suitable insulation (although perhaps "belt and braces") would be an additional safeguard. Mr Roberts expressed the view that no vapour tape would have a lifespan of more than about 5 years. However, Mr Blois-Brooke did not share this concern and was of the opinion that it would be possible to obtain vapour tape with a lifespan equivalent to the lifespan of the overcladding solution, and that suitable insulation would indeed provide an additional safeguard. I prefer the evidence of Mr Blois-Brooke to that of Mr Roberts in this regard, and I accept that the pursuers' overcladding proposals make adequate provision to address the risk of interstitial condensation.

[105] Drawing together these observations on the various criticisms of the technical abilities of the two options, I am persuaded that the criticisms of the Kemperol solution are without foundation. I am satisfied that the application of Kemperol, together with stitcher screws and a bond breaker to the end laps on this roof as proposed by the defenders and third parties would amount to a satisfactory technical solution to the problem of water ingress on this roof. By contrast, the pursuers' proposals for an overcladding solution, as presently formulated (i.e. at the same pitch as the existing roof, with a single skin metal overcladding roof supported by the Ashgrid system attached to existing purlins) will tend to replicate many of the problems of the existing roof. These proposals will not result in a satisfactory technical solution to the problems of water ingress to the pursuers' premises. There may be other considerations which fall to be considered, to which I turn below, but on the question of which of these options provides the best technical solution to the problem of water leaking through this roof, I consider that the answer is the Kemperol solution. On the basis of the evidence that I have heard, if the Kemperol system as proposed by the defenders and third parties is applied correctly, this will put a stop to leaks in this roof until such time as the roof lights require to be replaced and the plastisol coated metal sheets require maintenance. At that time, a further layer of Kemperol can be applied. On the other hand, if the overcladding solution presently proposed is adopted, it is likely that problems of water ingress into the premises will continue.

[106] As discussed above, Mr Fergusson accepted that the pursuers' proposals were "embryonic" and that they had chosen to proceed by way of a two stage process of tendering. His response to many of the criticisms of the pursuers' proposals put to him in cross-examination was that these were matters which he would expect to be ironed out by the contractors at a later stage. In re-examination he accepted (at paragraph [26] above) that it was always a potential problem affixing screws through an existing roof at an angle and connecting them to an unseen location, and that although he suggested the Ashgrid system it might be that there was another system which could be used more effectively. Senior counsel for the pursuers in his submissions (at paragraph [79] above) suggested that a developed proposal was properly within the pursuers' averments, which were wide enough to enable detailed aspects of design to be changed. In the course of his submissions I allowed him to amend his pleadings in the penultimate line of page 45 of the Closed Record by deleting the words "IP1000 profile" and substituting therefore "profiled". The question arises as to how much latitude should be allowed to the pursuers in developing finalised proposals for an overcladding option which would meet some or all of the criticisms discussed above. As narrated at paragraph [6] above, it was agreed that the scope of the present diet of proof should be restricted to the question of whether the pursuers' damages are to be assessed on the basis of (i) the "overcladding system" as averred by the pursuer in Article 8 of the Condescendence; or (ii) the "Kemperol system" as averred by Briggs Amasco in Answer 8 of the Condescendence. What latitude should be allowed to parties to develop their proposals in response to criticism?

[107] I accept that it may be appropriate to allow parties some latitude as to detail. For example, the reason for the amendment proposed by senior counsel for the pursuers in the course of his submissions (and thereafter granted in the absence of opposition) was to reflect the evidence of Mr Blois-Brooke that single ply metal sheeting may not be available in the IP1000 specification, and that in any event rather more frequent corrugations would be appropriate to provide additional strength and rigidity to a single ply metal sheet for overcladding. Senior counsel for the pursuers submitted, under reference to the well known passage in Lord Justice Clerk Thomson's opinion in Burns v Dixon's Ironworks 1961 SC 102, that this was a minor amendment and merely a variation, modification or development of what is already averred rather than something which is new, separate and distinct. It is because I accepted this distinction that I allowed the amendment. However, the overcladding system as averred by the pursuers in Article 8 of the Condescendence goes on to refer to the pursuers' invitations for tenders, and to the pricings of various tenders. The appendix to the invitation to tender defined the employer's requirements as being detailed in specification H31:10A-87. This specification was detailed; for example, it required the support structure for the metal overcladding to be "Ashgrid bars on 90mm Ashgrid brackets fixed through valleys of existing roof panels. Bars and brackets at centres to manufacturer's recommendation". The external sheets were specified to be: "Steel 2 BSEN 508-1. hot-dip zinc/aluminium coated steel 2 BSEN 10214: 1992". It specified that the manufacturer was to be "Ward or equal and approved", and that the thickness (nominal) was "gauge to suit span between Ashgrid bars".

[108] Against the background of this specification, I do not consider that it would be open to the pursuers to argue that they would be entitled to an overcladding system which was not supported by the Ashgrid support system; nor, for example, could they argue that a mill finish aluminium sheet overlay roof, such as referred to by W F Price (Roofing) Ltd in their quotation dated 11 May 2007 (No. 6/21 of process) fell within the concept of a variation, modification or development of their proposal. (I make it clear, in fairness to the pursuers, that no such suggestion was made on their behalf). The pursuers have come to court to argue in favour of the "overcladding system as averred by them in Article 8 of the Condescendence" and they refer in that Article to their invitation to tender for a design and build contract for the construction of a single skin overcladding of the roof. That invitation contains the detailed specification to which I have referred. No doubt minor details such as the number of corrugations in each metal sheet may be regarded as a variation, modification or development of what is averred, but I do not consider that it is open to them to depart from the essentials of the proposal, such as the Ashgrid support system, the pitch of the roof, the qualities of the metal sheeting to be used or the design and fastening of the plastic roof lights. Their proposals must be those spoken to by Mr Ferguson in his reports and his evidence, and available for criticism (both as to technical capacity to resolve the problem of water ingress, and as to price) by Mr Roberts and Mr Blois-Brooke.

[109] As indicated above, the issue which option best achieves a technical solution to the problem of water ingress is not the only issue to be considered. In deciding whether it is proportionate and reasonable to assess the pursuers' damages on the basis of the overcladding system rather than the Kemperol system, it is necessary to consider other factors, including the aesthetic appearance of the two options, any effect on the value of the pursuers' premises which may be caused by adopting one option rather than the other, and the relative costs of the two options.

[110] In some cases the aesthetic appearance of one remedial option may carry considerable weight, either in favour or against it. Nobody suggests that this is such a case. This roof cannot be seen from anywhere except by someone standing on it. Neither party suggested that the aesthetic appearance of either option was a factor to which any weight required to be attached.

[111] There may be cases in which the adoption of one solution, although providing a technically satisfactory result, may cause diminution in the value of the premises. If this could be established, this is surely a factor to which weight should be attached. The pursuers sought to rely on the evidence of Mr Oswald and Mr Gallagher to support the argument that this was such a case, and that if the overcladding option was carried out there would be no effect on the value of the premises, whereas if the Kemperol option was carried out the value of the premises would be diminished by about £1.2 million. However, Mr Oswald reached this position relying entirely on the evidence of Mr Gallagher as to the temporary nature of the Kemperol solution, that Kemperol was not a satisfactory solution to the problems in the pursuers' roof, and that it would cause any professional surveyor to report adversely to potential lenders or purchasers. For the reasons which I have already given, I did not accept the evidence of Mr Gallagher in this regard. I preferred the evidence of both Mr Roberts and Mr Blois-Brooke to the effect that whichever option was adopted, a professional surveyor would be aware that there had been a problem with this roof which had required remedial work. If such a surveyor was aware of the qualities of Kemperol and had experience of its use, he would know that it was the most effective solution to these problems. If he was not aware of it and had no such experience, he would research the product and would quickly discover its qualities and that it was the best solution to the problem. Standing my views as to the technical merits of the Kemperol solution, in light of Mr Oswald's methodology in assessing valuation, I do not consider that the use of the Kemperol option would have any adverse impact on the value of the premises by comparison to the overcladding option.

[112] The issue of the costs of the two options is clearly one to which weight must be attached when considering whether it is reasonable and proportionate to prefer one over the other. The costs of the Kemperol option appear to be relatively easily ascertainable. None of the witnesses took serious issue with the costings at Appendix Z No. 35/3 of process, which brought out a total of £131,755 exclusive of VAT. Both Mr Roberts and Mr Blois-Brooke accepted that it would be appropriate to add some figure for contingencies, although Mr Blois-Brooke would not expect this to be completely used up. Mr Blois-Brooke suggested 10%-15% for contingencies, and Mr Roberts suggested 10%. Allowing about 10% for contingencies gives a figure of £145,000 plus VAT. Mr Fergusson did suggest some additional costs in the course of his re-examination (see paragraph [24} above), but he did not touch on these in examination in chief or in cross-examination, and the defenders had no opportunity to test him on these, so I am not inclined to attach much, if any, weight to them. In any event, it is not clear to me why provision would be required for these items in the Kemperol solution, as there would be no penetration of the full thickness of the existing roofing panels.

[113] By contrast, the cost of the overcladding solution is to some extent a matter of conjecture. Mr Hope gave evidence to the effect that he reported on the revised tenders and recommended a tender which would result in an overall project cost of £1,207,588 excluding VAT and inflation beyond the acceptance period. He also accepted that no provision had been made in the tenders for a performance bond in respect of the contractors' default or liquidation, surety for which was normally about 10% of the value of the contract. He recommended the tender because the other two tenderers had underestimated the cost of health and safety provision, craneage and the like. This evidence from Mr Hope was led by Mr Ellis on behalf of the pursuers (see paragraphs [27]and [28] above). That being so, I found it surprising that Mr Ellis suggested in his submissions (at paragraph [79] above) that the starting figure should be that in the revised quotation from Price Roofing in the sum of £757,579. On the evidence I consider that the figure of £1,207,588 given by Mr Hope, together with an uplift of about 10% for a performance bond, is a more accurate reflection of the costs of the pursuers' proposals as they presently stand. Thereafter, Mr Ellis recognised that the proposals as they presently stand are unworkable and will not resolve the problem, and that a further 20-50% would be required to reflect extra design works to provide a fully worked up proposal. If it is legitimate for the pursuers to argue for such extra design works, (and for the reasons given above I do not consider that it is), this would result in costs for the fully worked up overcladding solution of somewhere between £1.5 million and £2 million. Even in their "embryonic" state at present, the cost of overcladding together with a performance bond would be in the region of £1.3 million (all figures being exclusive of VAT). In its present form the overcladding proposal would therefore cost approximately nine times as much as the Kemperol proposal. If the overcladding proposal were to be fully developed at an extra cost of between 20% and 50%, the cost of the fully developed overcladding proposal would be in the region of eleven to fourteen times the cost of the Kemperol proposal.

[114] I now turn to consider what light the authorities to which I was referred may shed on the principles to be applied in this case. In doing so, it is worth noting that this case is in a rather different category from many of the authorities to which I was referred, because neither party in the present action is arguing that damages fall to be measured on the basis of diminution in value. That was the essential issue between the parties in Ruxley Electronics, in which ultimately the House of Lords agreed with the trial judge that the measure of damages was not the cost of reinstatement but the diminution in the value of the work occasioned by the breach of contract. Here, both parties are agreed that the proper measure of damages is reinstatement; the issue is which of two alternative reinstatement options should be preferred. There are several observations in the authorities to which I was referred which bear on this question, some expressed in terms of the requirement for reasonableness in adopting a measure of damages, and others in terms of the duty on an innocent party who has been the subject of a breach of contract to mitigate his loss. Although the decision of the Court of Appeal in Ruxley Electronics was overturned by the House of Lords, one observation by Staughton LJ in the Court of Appeal (quoted by Lord Lloyd of Berwick at page 364G) is relevant to the present circumstances, and stands despite the House of Lords' subsequent decision, namely: "It is unreasonable of a plaintiff to claim an expensive remedy if there is some cheaper alternative which would make good his loss."

[115] The circumstances being considered George Fischer Holding Ltd at paragraphs 193 and 194 were strikingly similar to the circumstances in the present case. That case concerned a leaking roof. By the commencement of the trial it was agreed that replacement of the roof was to be excluded as uneconomic and unnecessarily disruptive of the operation of the plaintiff's business. There were two alternative schemes to remedy the situation, one of which was an "over roof" solution, and the other of which was the application of a waterproof membrane applied in liquid form under the trade name Soladex. In that case the waterproof Soladex membrane would cost less than half the over roofing alternative, a difference of over £800,000. His Honour Judge John Hicks QC observed that:

"Since Soladex would be so much the cheaper and cannot be said to be the more detrimental to the appearance of the buildings - I should have thought, if anything, the reverse - it must clearly be preferred unless the criticisms of its expected effectiveness are, taking the above considerations into account, made good on the balance of probabilities. In my judgement they are not. ... I therefore conclude that damages should be assessed on the basis that the appropriate remedial scheme is the Soladex scheme."

[116] This passage was referred to with approval in McGlinn v Waltham Contractors Ltd at paragraph [793] where the court observed:

"If there are two equally efficacious alternative remedial schemes, and one is cheaper than the other, then prima facie the claimant is obliged to put in hand the cheaper of the two schemes. In the Great Ormond Street case (1987) 19 Con LR 25 at 96, Judge Newey said:

'The plaintiff has, whether as part of the requirement that he act reasonably or otherwise, a duty to mitigate his loss. This may require him if presented with two or more choices to choose the one which will keep his losses to the minimum. ... The duty to mitigate may require the plaintiff to have regard to advice from third parties, or even from the defendant, or from the defendant's advisers.'"

[117] The principles which are brought out by these authorities are (1) that a pursuer who is the victim of a breach of contract cannot recover from the party in breach the cost of reinstatement where such cost is unreasonable and out of all proportion to the benefit that it will achieve, even in a case where the court is satisfied that the pursuer intends to adopt such an expensive scheme if awarded damages on that basis, and (2) that the duty to mitigate his loss may require the pursuer to put in hand the cheaper of two alternative schemes. The circumstances of the present case may perhaps be regarded as stronger for the defenders than either George Fischer Holding Ltd or McGlinn. The difference between the cost of the two schemes in George Fischer was £800,000, one scheme costing rather more than double the other. In the present case, the Kemperol scheme is between £1.1 million and about £1.8 million cheaper than the overcladding scheme, which is between nine and almost fourteen times more expensive. And unlike the passage quoted from McGlinn, this case is not concerned with two equally efficacious alternative remedial schemes, one of which is cheaper than the other - as I have held above, not only is the Kemperol scheme very much cheaper than the overcladding option, it is also technically far more efficacious.

[118] Senior counsel for the pursuers disagreed with Mr Blois-Brooke's statement of the test as to whether the overcladding solution is disproportionate (see paragraphs [79] an [89] above). Mr Blois-Brooke approached the question on the basis of what he would have advised a client who was having to pay for the repair himself; Mr Ellis argued that he should put himself in the position of advising the reasonable owner of a building who is the victim of a breach of contract. I am inclined to agree with Mr Ellis on this point - a successful litigant suing for damages for breach of contract may properly approach the question of what remedial scheme to choose slightly differently from the owner of the same property who was having to pay for the repairs himself. However, the principles stated above still apply - the successful litigant cannot insist on damages measured on the basis of an unreasonably expensive or grossly disproportionate remedial scheme, and he remains under a duty to mitigate his loss. The distinction which Mr Ellis sought to make may well be relevant in a finely balanced case, but I do not consider that it is of more than academic interest in a case such as this.

[119] In summary, I consider that all of the relevant factors favour the proposition that the pursuers' damages are to be assessed on the basis of the Kemperol system as averred by Briggs Amasco in Answer 8 of the Condescendence. I am satisfied that the Kemperol system is technically better suited to resolving the problem of water ingress in the pursuers' roof than the alternative overcladding system. Aesthetic considerations do not arise in this case. I do not consider that the adoption of the Kemperol system will result in a diminution of the value of the premises by comparison to the overcladding system; and the difference in the cost of the two systems is so great that this factor strongly favours the Kemperol system.

[120] For these reasons I answer the question contained in the Joint Minute for the parties (No. 63 of process) by determining that the pursuers' damages are to be assessed on the basis of the Kemperol system as averred by Briggs Amasco in Answer 8 of the Condescendence. Parties were agreed that all other matters should be reserved meantime for further procedure. Accordingly this case will be put out By Order to discuss further procedure in early course.