Pursuers: McBrearty; Ledingham Chalmers LLP

First Defender: Balfour; HBM Sayers

Second Defenders and Third Party: Ennis; Simpson & Marwick

13 December 2013

Introduction

[1] The pursuers are the former owners of the ground floor flat of a two-storey flatted dwellinghouse in Mount Vernon, Glasgow, having purchased the flat in about December 2004 and sold it in March 2010. The first defender is and, at the material time, was the co-proprietor, along with his wife, of the upper flat. The property is a large detached south-facing sandstone villa which, according to the first defender, was built in about 1880 and converted into two flats in the late 1950s. The first defender and his wife have resided there since 1965. On the evening of 4 March 2006, at a time when all of the four proprietors were absent from home, a fire broke out in the property resulting in severe fire, smoke and water damage to both flats and to their contents. It is common ground that the outbreak of fire was associated with a gas central heating and hot water boiler located in the pursuers' kitchen. In these proceedings the pursuers claim that the loss and damage which they sustained was caused by the fault of the first defender in allegedly obstructing the boiler flue outlet, and also by fault and breach of contract on the part of the second defenders, with whom the pursuers had a service contract in respect of the boiler, and who were initially brought into the action as a third party by the first defender. The action came before me for a proof on liability only, all issues relating to quantum having been agreed by joint minute.

The boiler and its operation

[2] The boiler with which this action is concerned was a Myson Midas SF1 model. This is a gas-fired combination boiler that is designed to operate as a balanced flue room-sealed appliance, with combustion air being drawn into it, and expelled from it, outside the property and not from and into the room in which the boiler is located. The boiler housing is mounted on a wall plate fixed to the inside of an external wall. Within the housing there is a combustion chamber containing the gas burner, heat exchanger, and a fan which assists in drawing air into the boiler and expelling exhaust gases. The boiler flue passes horizontally through holes in the rear of the boiler housing and wall plate and in the wall itself, and consists of two concentric tubes with spacers to support the inner tube. Fresh air is drawn into the boiler housing through the outer tube. The air is mixed with gas in the combustion chamber, where the gas is ignited and burned. The hot gases are drawn upwards by the fan (and by convection) through the heat exchanger and out to atmosphere via the inner tube of the flue. The electrical circuitry of the boiler is located behind a fascia panel towards the bottom of the front of the boiler housing.

[3] For the purposes of this case, it is necessary to describe the flue and its attachment to the boiler, when properly installed, in more detail. (I note here that during the proof there were available in court both the flue pipe from the pursuers' boiler and also the flue pipe and housing of the specimen boiler of the same type and age which had been used by the second defenders in the test which they carried out and which is described at paragraph 25 below.) Externally, the inner (exhaust) tube protrudes further than the outer (inlet) tube and has a grill whose slats should be vertical when the flue is fitted. The flue is telescopic, i.e. its length is adjustable to fit the distance in a particular building between the external wall face and the rear face of the boiler housing. At the boiler end of the flue, the inner tube again protrudes further than the outer tube. This end of the inner tube fits inside a sleeve protruding from the hood of the fan in the combustion chamber, thereby providing a sealed route for exhaust gases to pass from the chamber via the inner tube to atmosphere. At the boiler end of the outer tube, there is a swaged section or lip. This does not prevent the flue from passing through the holes in the wall mounting plate and the rear of the boiler. However, attached to the inside face of the rear of the boiler there is a Stratford clip or clamping ring against which the swaged section abuts, preventing the flue from being inserted any further into the combustion chamber within the boiler housing. A plastic flue gasket is fitted between the inside rear face of the boiler and the swaged section of the flue and Stratford clip, providing a seal and holding the flue in place but not entirely precluding it from being pulled out through the hole.

[4] The foregoing is a general description of the operation and installation of the type of boiler with which this action is concerned. I now turn to the specifics of the boiler in the pursuers' kitchen. It is agreed by the parties that it was installed in about 1993, i.e. some years before the pursuers acquired the flat. The identity of the installer is unknown. It was located inside a timber kitchen cupboard, secured at top and bottom by screws to wooden battens in a plasterboard wall. The external wall of the building was stone, with a cavity between stone and plasterboard. The garden area immediately outside the pursuers' kitchen belonged to the proprietors of the upper flat. The flue accordingly vented its exhaust gases to atmosphere in an area of garden ground, on the west side of the property, which did not belong to the pursuers and to which they did not have access. The flue outlet was not visible from any window in the lower flat. It was approximately 1.9 metres above the ground and was located under a porch on the first floor to which a stair led giving access to the upper flat. Measurements taken by the expert witnesses instructed on behalf of the pursuers and the first defender indicated that the flue was not installed precisely horizontally but sloped slightly downwards towards the exterior of the property at an angle of approximately 4 degrees. It was explained in evidence by Mr Malcolm Finlay, a regional technical compliance and health and safety manager employed by British Gas, that this was a common practice among installers and was intended to minimise the ingress of rain to the flue and to allow any rain which did penetrate to run out.

The pursuers' service contract

[5] When the pursuers purchased the flat in 2004, they transferred an existing service contract with the second defenders to the boiler. The second defenders' obligations under the service contract were set out in a document entitled "The British Gas HomeCare Guide" which included detailed terms and conditions. The contract provided, inter alia, for an "initial safety inspection", carried out to confirm whether or not the second defenders were willing to assume contractual responsibility for the maintenance of the boiler and, if so, for annual servicing thereafter. Two of the specified tasks of the engineer carrying out the initial inspection were "A visual examination of the appliance and system to detect any obvious hazards and existing design faults" and "Perform ventilation, appliance flue and basic electrical safety checks". The second defenders' terms and conditions further stated (at page 12 of the Guide) that at the initial inspection "...We will inspect your system or appliance (or both) to make sure they are safe and in good working order".

[6] On the occasion of an initial inspection, and on annual inspections thereafter, the second defenders' engineer was required to undertake work in accordance with British Gas Operating Procedures for appliance servicing. These procedures included the following checks:

"...3. Check the general condition of the appliance, and that the installation conforms to appropriate standards, to also include:

· Check Ventilation in accordance with Operational Procedure 9,

· Check Flue in accordance with Operational Procedure 10,

· Check for appropriate appliance and compartment warning notices,

· Check for signs of spillage on appliance and adjacent decoration.

...5. Check the system for signs of water leaks and repair as necessary. (This will necessitate the casing to be removed as far as is required to visually examine appropriate areas of the boiler and may also reveal signs of spillage, heat stress or mechanical deterioration not apparent with the case in position. The inside of the casing should be cleaned as necessary. Check all appliance seals and joints, replacing as necessary.)

...9. Check the burner operating pressure and flame picture, adjust if necessary. (Where the boiler is range-rated, set the burner pressure to the maximum shown on the appliance data badge.)

10. Carry out a Performance Test. (In accordance with Section 4.4 of Part 21 of these Procedures. If the test ratio is greater than 0.004, or when your other checks or discussions with the customer have indicated there may be problems, carry out the additional strip-down servicing requirement procedure on the boiler as detailed in Steps 17 to 22."

The reference to the "test ratio" is a reference to the boiler's carbon monoxide/carbon dioxide ratio. The test affords an indication of the efficiency of combustion within the boiler and also, to some extent, of the effectiveness of ventilation.

[7] It is a matter of agreement between the parties that employees of the second defenders had attended on four occasions in 1997 and 1998 at the flat subsequently purchased by the pursuers. No records were produced in relation to work carried out on any of those occasions. It is also agreed that during the period of the pursuers' ownership, prior to the fire, the second defenders' engineers attended at the flat on three occasions, namely 20 March, 7 June and 14 December, all in 2005. The visit on 20 March 2005 was made by an engineer named Jason Wilson who gave evidence at the proof but had no recollection of the visit. Records of a report made by Mr Wilson at the time of the visit indicate that this was a first visit to the premises under the pursuers' service contract, although the boiler had previously been serviced by the second defenders under a contract with a previous owner. The report discloses that Mr Wilson tested the burner pressure and found it to be "OK". He also used a probe inserted into the external end of the flue to test the boiler's CO/CO₂ ratio. It produced a satisfactory reading of 0.0004. The engineer did note that the boiler flue did not conform to current standards in that the outlet was not protected by a wire guard to protect people from burning themselves by touching it. The visit on 7 June 2005 was in response to a complaint of low water pressure and resulted in the fitting of a new pressure relief valve in the boiler. The burner pressure was tested and found to be satisfactory; there is no record of a test of the CO/CO₂ ratio. The visit on 14 December 2005 was in response to a complaint of a leaking water pipe. No test of burner pressure or CO/CO₂ ratio was recorded.

The occurrence of the fire
[8] The second pursuer gave evidence that on the morning of 4 March 2006, which was a Saturday, she had a brief conversation with the first defender in the course of which he told her that he had been noticing a burning smell in the back of his house for about a month. He asked the second pursuer whether she had noticed it. She said she had not, but wondered if it might be related to work which had been carried out in connection with a damp course. She told the first defender that she would speak to the first pursuer about it. The first defender agreed that this conversation had taken place, but stated that he had been aware of a burning smell on two or three occasions over a period of months during the previous winter and then again a few weeks before the day of the fire. It had smelt like wood burning but he could not locate the source. He had spoken to the pursuers about it.

[9] The first defender and his wife spent the evening and night of 4 March 2006 away from home. The pursuers were also out for the evening, having left the house during the afternoon. At approximately 8.30 pm, Strathclyde Fire & Rescue received a call reporting that the house was on fire. Watch Manager Bruce Kinnear was the officer in charge of the first fire appliance to arrive at the scene. He walked up the west side of the building and looked through a west-facing downstairs window. He could see a fire burning in the pursuers' kitchen which appeared to be well developed. He ascended the external stair to the upper flat and forced entry to it, whereupon he discovered that the floor inside the door was missing and that he could see down through the hole in the floor to the flames in the kitchen below. The fire had reached the roof and the loft space above the upper flat was well alight. As safe access to the upper flat from the stair was not possible, a ladder was pitched to the front windows by which means Officer Kinnear was able to enter. His further investigations indicated that the fire was established throughout the roof space, including the area above that part of the upper flat which was not yet damaged by fire. The fire was fought from a turntable ladder. At around 10.00 pm, Officer Kinnear re-entered the first floor to see how water damage could be minimised. It looked as if the roof was going to collapse. Items in the lower flat were sheeted to protect them from damage but it thereafter became too dangerous for the fire crew to remain within the building. The fire continued to be tackled from the turntable ladder and was extinguished by about midnight.

[10] The first investigation of the fire was carried out the following morning by Watch Manager David Lynn, a fire investigation officer employed by Strathclyde Fire & Rescue. He noted that the property was severely affected by the fire: approximately 90% of the roof members had disappeared and there were three unsupported chimney heads. The worst affected area of the building was the west side. The fire appeared to have developed at a low level and spread upwards to the roof. In the course of his examination of the property, Officer Lynn noticed that a number of long pieces of timber, which he took to be wooden clothes poles, had been placed vertically against the external end of the ground floor boiler flue. The poles were neatly stacked in a tepee formation so that they butted against one another at the top, where the flue was, and fanned out on the ground. They covered the flue vent behind. Officer Lynn noticed a pattern of soot on the inside faces of the poles. He was unable to carry out a full inspection of the interior of the property because much of it was too dangerous to enter. He regarded the presence of the timber set against the flue outlet as an important piece of evidence but reached no firm conclusion as to the cause of the fire.

Investigations on behalf of insurers
[11] Mr Stuart Mortimore, FIET, MIFireE, a principal member of Dr J H Burgoyne & Partners LLP, consulting scientists and engineers, was instructed by loss adjusters on behalf of the pursuers' insurers to carry out an investigation of the cause of the fire. He visited the premises on 7 March 2006 to undertake a scene investigation. On 15 March 2006, he returned along with Mr Frank Duffy, who holds an MSc in fire investigation and was at that time employed as a senior fire investigator by International Fire Investigators and Consultants Ltd, and who was instructed by loss adjusters on behalf of the first defender's insurers. Mr Mortimore and Mr Duffy decided that the second defenders should be offered an opportunity to view the property and to participate in any examination of the pursuers' boiler. On 12 May 2006, Mr Mortimore and Mr Duffy visited the premises along with representatives of the second defenders and their insurers. On 16 May 2006 the second defenders advised Mr Mortimore that they did not wish to participate in any further scene investigation work and accordingly Mr Mortimore and Mr Duffy returned to the premises to complete their examination. In the course of these visits Mr Mortimore took a large number of photographs which were available in A4 size at the proof.

[12] The following is an amalgamation of the findings of Mr Mortimore and Mr Duffy, taken from their respective reports and evidence to the court. Except where I have stated otherwise, I did not understand there to be any disagreement between them as regards these matters. There were no smoke deposits on the external faces of the walls of the house above windows, doors or vents on either the first or second floor, except for the boiler flue. Six pieces of timber stood next to the flue outlet. Two of the pieces had semi-circular sooty deposits which aligned with the (inner) exhaust tube of the flue. The outer tube, being shorter, would not have been blocked and fresh air would have been able to enter the flue even if, as appeared to be the case, the timber pieces had stood against the end of the inner tube. Within the lower flat, only the southwest corner of the kitchen had been attacked by fire. Damage began at floor level where fire debris from the collapsed ceiling and roof above had accumulated. There was no heat damage at high level in the kitchen and no significant smoke contamination within the kitchen or elsewhere in the flat. Furniture and other items elsewhere in the kitchen showed little or no sign of heat damage. The cupboard which had housed the boiler had been virtually destroyed by the fire and fire fighting activities. The upper flat and contents were destroyed by the collapse of the roof, fire and water damage.

[13] At the time of the inspections by Mr Mortimore and Mr Duffy, the boiler remained attached to the wall by screws at the bottom of its wall plate. However the wooden batten to which the upper screws had been attached had burned away and the plasterboard against which the wall plate would have been fixed had disintegrated, so the boiler was leaning outward at a slight angle from where the wall had been (Mr Duffy estimated that it was leaning about 4-5 cm out at the top). It was not connected to the flue pipe which was mortared in place at the exterior of the stone wall. The Stratford clip and the plastic flue gasket remained fixed to the inner face of the boiler housing. The wiring of the boiler fan was unburned. The condition of the combustion chamber and flue pipes, including the exhaust tube, was inconsistent with a fire originating within this equipment. Mr Mortimore and Mr Duffy carefully attempted to push the boiler back into the position it would have occupied before the fire, using a wooden batten to mark where they measured the inside face of the plasterboard to have been. They concluded that with the boiler in its pre-fire position, there were gaps between the boiler and the flue pipe capable of allowing hot exhaust gases to escape from the flue into the wall void. According to Mr Mortimore's measurements, the gap between the outlet from the fan and the end of the inner flue pipe would have been approximately two or three millimetres, and the gap between the back of the boiler and the end of the outer flue pipe would have been about 15 or 16 millimetres. He considered that an absence of any relevant witness marks or "protection" marks on either pipe of the flue indicated that prior to the fire these gaps had been present around the whole circumference of both pipes. Mr Duffy's conclusion was slightly different. When it was his turn to measure, with Mr Mortimore holding the boiler up, he measured the gap from back of the boiler to the outer pipe as being about 12 millimetres. He also concluded, when he came to write his report in 2013, that because (as already noted) the flue had been fitted at a 4° angle, and having regard to certain marks which he identified on the underside of the inner flue pipe, this tube might have just overlapped with the fan outlet at the bottom, though not at the top. Mr Duffy considered that when the flue was hot, the inner flue pipe would expand longitudinally at the boiler end, and that such thermal movement would reduce the size of the gap. This would explain why there was not a single witness mark to indicate the location, prior to the fire, of the inner tube in relation to the fan outlet sleeve. The gap in the inner flue pipe would have been about 3‑4 millimetres. Mr Mortimore regarded Mr Duffy's conclusions as inconsistent with what they had both observed when carrying out their joint investigation. However, neither Mr Mortimore nor Mr Duffy considered any of these differences of measurement or opinion to be material to their shared conclusion as to the cause of the fire, discussed below. Both experts considered that the gaps which they identified had been present since the installation of the appliance.

[14] Mr Mortimore and Mr Duffy observed electrical wiring in the wall space in the vicinity of the flue. However, it was found to be redundant and would have been de-energised during the fire. Neither found any evidence to suggest that the fire had been caused by an electrical defect.

The placing of wood against the flue outlet
[15] It was not disputed that prior to the fire pieces of wood were placed over the flue outlet in such a way as to obstruct the escape of hot exhaust gases from the inner pipe. Officer Lynn observed them there, stacked in a tight formation, on the morning after the fire. They were in the course of being removed to one side when the first pursuer arrived at the property that morning. The neatness and tightness of the stacking left no doubt that they had been placed over the flue outlet deliberately. The first defender accepted that the wood had been in that position prior to the fire and that he must have seen it there when walking between his house and his garage. He explained that the pieces of wood were not clothes poles but timber that had been salvaged by him from an old greenhouse which had stood at the side of the house before being demolished some years ago.

[16] The issue of who had placed the wood against the flue outlet, and why, was not a matter of agreement. The first pursuer described conversations he had had during 2005 with the first defender, who complained that emissions from the flue were causing the sensor of his outside light to activate when the boiler fired and asked the first pursuer to have his boiler checked. The first pursuer agreed that the light had been coming on as he had seen it too. According to his recollection, one of the visits by the second defenders' engineers was to investigate this matter (although this is not supported by the second defenders' records). No further conversation about the light took place between himself and the first defender thereafter. On 24 March 2006, Officer Lynn was interviewed by Mr Mortimore and stated that when the first defender arrived at the premises on the morning after the fire, he told Mr Lynn that he had left the poles there. Officer Lynn reiterated this statement in his evidence to the court. He was sure of it as he had recognised at the time that this was a critical piece of evidence. The first defender denied having placed the pieces of wood against the flue outlet. He at first denied having told Officer Lynn that he placed them there but subsequently accepted that although he had not left the wood there, it was possible that he had said that he did. He agreed that there had been a problem with his external light sensor and also agreed, in cross-examination, that he had mentioned it to the first pursuer because it might have been caused by emissions from the flue, although he disagreed that he had asked for the boiler to be checked. He hinted somewhat obliquely while giving evidence that the wood might have been placed over the flue by his wife because she was annoyed by the fumes. He agreed that he could not have avoided noticing the pieces of wood but asserted that he had no idea that they were blocking a flue outlet behind.

[17] I did not find the first defender's evidence on this matter to be credible. Especially incredible was his assertion that although he had not put the wood there himself, he saw it there but did not realise that it was covering the flue outlet from the pursuers' kitchen. The wood was deliberately placed against the flue outlet. The first defender was aware of the existence of the flue outlet, which for the previous 13 years had been located in a wall bounding his own garden. He had discussed it with the first pursuer. Other than an oblique reference by the first defender in the course of his evidence to having been told something by his wife, there has never been any suggestion that she was responsible for placing pieces of wood, which had been salvaged by the first defender and retained by him in his garden for future use, against the flue outlet. No evidence was led by the first defender that might have implicated his wife or anyone else. I accept Officer Lynn's evidence that on the morning after the fire the first defender told him that it was he who had placed the wood there. It is difficult to see any reason why the first defender might have made this statement contemporaneously with the fire incident unless it was true. An explanation for the first defender's actions is afforded by the problems that he had experienced with the triggering of the light sensor. In these circumstances I find that the first defender did place the wood against the flue outlet with the intention of obstructing the emission of fumes from it, and that his evidence to the contrary was an attempt to avoid blame for the fire and was not truthful.

The cause of the fire

Opinion of Mr Mortimore

[18] On the basis of his observations and the eye witness account of Officer Kinnear that there had been fire in the ground floor kitchen before the roof was penetrated, Mr Mortimore considered that the fire almost certainly started in, or in the immediate vicinity of, the ground floor kitchen. His view was that the confinement of significant heat or smoke damage to one corner of the kitchen was not consistent with the fire having originated within the kitchen itself. The fire damage that did occur in the south‑west corner was caused at a later stage of the fire when the ceiling above that corner collapsed and burning debris fell down from above. Mr Mortimore concluded that the fire almost certainly started in, or immediately adjacent to, the kitchen wall void at or near the boiler flue. Having regard to the fact that the wiring of the boiler fan was unburned, the lack of combustible material within the part of the boiler that abutted the wall void, and the absence of significant deposits of soot within the combustion chamber or flue pipe, he felt able to narrow this down to the fire having started within the kitchen wall void and spread upwards into the ceiling void between the flats and the roof space above the top flat. The only likely source of ignition was heat energy emanating from the flue. The flue pipe had not been secured to the back of the boiler. While the flue outlet pipe remained unobstructed, it seemed likely that the exhaust gases passed along it with ease and without significant spillage from the unsecured connection. Even with gaps present between flue and boiler, the boiler housing was of a volume sufficient to contain more than enough air to supply the burners without producing an abnormal CO/CO₂ ratio. Once the exhaust flue was blocked, however, the pressure of the fan could have led to an appreciable discharge of gases into the wall void through the gap that existed between the rear of the boiler and the outer tube of the flue. Fresh air sufficient to enable the boiler to operate could have continued to be supplied not only from the outer flue (whose external end was not blocked by the wood) but also from the wall void. The discharge of hot gases through the gap in the outer flue pipe could, over a period of time, have ignited timber wall supports within the void. When exposed to temperatures in excess of about 135°C, timber can degrade by a chemical process called pyrolysis into more complex compounds that can be ignited at a lower temperature than that (Mr Mortimore gave a figure in his report of about 235°C) normally required to ignite timber. The process of pyrolysis was not fully understood but could take many days or even weeks. This time scale was consistent with reports of burning smells and of the discovery of the fire. Mr Mortimore's opinion was that, given the absence of any other plausible explanation, this mechanism constituted the only likely cause of the fire. The event that led directly to the fire was the blocking of the flue, but the fire would probably not have occurred if the installers of the boiler had connected it to the flue or if the second defenders' engineers had observed this installation defect when they serviced the appliance. If there had been no gap between the boiler and the flue, blockage of the exhaust pipe would have caused the boiler to cease operating.

[19] One element of Mr Mortimore's reasoning was an assertion that the boiler could not have been assembled in accordance with the manufacturer's instructions because the diameter of the swaged section of the flue was too great to allow the flue to pass through the hole in the back of the boiler without physical damage to one or both. This was said to support his conclusion that there must have been a gap between the end of the flue and the back of the boiler. In the course of cross-examination by counsel for the second defenders, Mr Mortimore's assertion was demonstrated to be incorrect: the swaged section did pass through the hole. Contrary to Mr Mortimore's understanding, the function of the Stratford clip was to prevent the flue from being passed any further into the boiler housing rather than to secure it against being pulled out. The gas seal was provided by the flue gasket secured in place by the Stratford clip. This did not, however, alter his view that prior to the fire there had been a gap.

[20] When it was put to Mr Mortimore in cross-examination that the most likely cause of the fire was a fault in the electrical circuitry of the boiler, he disagreed. In order to reach the flue the fire would have had to pass upward through the boiler housing and would have burned the wiring in the fan. A fire would have been unlikely to reach the wall void directly from the electrical components because the plasterboard behind the boiler would have operated as a fire barrier. The fire damage to the boiler electrics was caused by burning debris from the ceiling and roof above. It was also put to Mr Mortimore that his photographs and measurements indicated that there had in fact been little or no gap between the external stone wall and the wooden battens within the wall void, and that it could be demonstrated as a matter of arithmetic that if there had been a gap in the outer pipe the hot gases would have vented into the space between the back of the boiler and its wall plate, i.e. into a space within the kitchen rather than within the wall void. Mr Mortimore considered that his measurements, properly interpreted, did not support this conclusion.

Opinion of Mr Duffy

[21] Mr Duffy shared Mr Mortimore's opinion that the fire had originated in the wall void in the vicinity of the boiler, for the same reasons. He too concluded from examination of the flue pipe and from the joint exercise to push the boiler back into the position that it had occupied before the fire that there had been an opening in the flue pipe which permitted flue gases to escape at a temperature where they represented a competent ignition source for timber within the wall void, following upon a period of pyrolysis. During most of the life of the appliance, the amount of gas escaping from the flue pipe was very small. However, when the end of the pipe was obstructed by the pieces of wood arranged round it, this caused back pressure and an increase in the quantity of gas escaping. Because of the small size of the gap, it would escape as a jet across the width of the outer (inlet) pipe with a directional element. If the gap was very small, due to thermal expansion, when the boiler was hot, this would increase the strength of the jet. The time from blockage to ignition of the timbers would depend on a number of factors including the temperatures experienced, properties of the timber, distance to the timber, heat loss to the structure and ventilation conditions. The pressure sensor of the appliance might not detect an escape of flue gases since they were still being evacuated from the combustion chamber. The soot deposits on the pieces of wood placed round the flue terminal indicated that the blockage had affected combustion efficiency. This mechanism was consistent with the occupants of the upper flat noticing a hot smell during the period prior to the occurrence of the fire.

[22] In response to cross-examination on behalf of the second defenders, Mr Duffy insisted that the attempt by Mr Mortimore and himself to re-position the boiler to where it had been before the fire was sufficiently accurate to justify his conclusion that there were gaps between the boiler and the flue. He too disagreed that Mr Mortimore's photographs supported the proposition that there was no gap between the stone wall and the battens within the wall void. He regarded the absence of substantial high-level damage within the kitchen as a clear indication that the source of the fire had not been the boiler electrics.

Evidence on behalf of the second defenders

[23] The case for the second defenders in relation to the cause of the fire had three broad strands:

• firstly, it was unlikely that a fire could have originated through the mechanism suggested by Mr Mortimore and Mr Duffy;
• secondly, it was unlikely that a gap had existed at all between the boiler and the flue pipe; and,
• thirdly, there was an alternative scenario, namely that the fire had been caused by a fault in the boiler's electrical circuitry.

[24] In relation to the first of these strands, evidence was given at the proof by Dr Christopher Wareham, a principal associate with Hawkins and Associates Limited, forensic investigators, with extensive experience in the investigation of fires. Dr Wareham was first instructed in about 2012 and did not therefore carry out any first hand investigations into the fire. No written report by Dr Wareham was lodged. He had examined the specimen boiler and flue pipe in preparation for giving evidence at the proof. He regarded it as unlikely that an escape of hot gas into the wall void for a period of about 6-8 weeks would lower the ignition temperature of the wood sufficiently to enable it to ignite. If the gap in the pipe was only at the top, this would render pyrolysis and subsequent ignition less likely because the volume of gas escaping from a smaller hole would be less. Even if the condition subsisted for six months, ignition would still not be probable. Pyrolysis would only be a competent method of ignition if the temperature was very hot and there was combustible material very close to the flue. The placing of timbers against the flue outlet could be associated with an electrical cause of the fire in that the obstruction might cause the boiler persistently to extinguish and re‑light (a process known as "cycling") and thus repeatedly overload the electrical circuit.

[25] Support for Dr Wareham's view was provided by the results of a test carried out by Mr Finlay, who located a Myson Midas SFI combination boiler ("the test boiler") of similar age and installation date to the pursuer's boiler in a house in Edinburgh. This boiler had also been on a service contract with British Gas since 1998 and had been visited regularly. There were, however, material dissimilarities between the property in which the test boiler was installed and the property in which the fire occurred. The "test" property had poured concrete walls with roughcasting on the exterior and a concrete skim on the inside; there was no wall cavity. With the flue remaining connected to the appliance, Mr Finlay placed timbers against the flue terminal to replicate the restriction on the pursuers' flue, but - for safety reasons - without a gap in the flue. The combustion of the boiler in this condition was tested. In hot water mode the boiler would only operate for very short periods before beginning to cycle. Operating the boiler in central heating mode did produce results including a very high carbon monoxide concentration and a CO/CO₂ ratio of 0.6486. The test boiler was then set up with gaps of about 3 millimetres between the tubes of the flue and the boiler housing and fan sleeve, but - again for safety reasons - without any timber against the terminal. Again the CO concentration and the CO/CO₂ ratio were unsatisfactory. A repeat of the second test with the exhaust flue butted against the fan sleeve produced a CO/CO₂ ratio of 0.0057, which, if discovered in the course of a service by the second defenders' engineers, would have required a full strip and clean service. Mr Finlay's conclusions from these tests were, firstly, that full obstruction of the flue terminal by timbers would have compromised the operation of the boiler to an extent that would have been observed by the house occupier and, secondly, that the presence of a gap in the flue would have affected the efficient operation of the boiler to an extent which might have been observed by the house occupier but which, in any event, would have been disclosed by the tests carried out by the engineer who serviced the boiler on 20 March 2005. These findings accorded with his own experience: in his many years working as an engineer he had never been aware of a fire caused by a gap in a boiler flue. The absence of any reported problem with the operation of the boiler and the satisfactory readings obtained by the engineer during his visit suggested that this was not the mechanism that had caused the fire. He disagreed with the proposition that the interference with air supply caused by gaps in the flue would be compensated by the drawing of air from the wall void: any escape of exhaust gases from the inner to the outer tube would cause contamination of the air supply and reduce the efficiency of combustion.

[26] The validity of the conclusions drawn by the second defenders' witnesses from the test results was doubted by Mr Richard Siddons, an engineer and principal member of Burgoynes who gave evidence on behalf of the pursuers. According to Mr Siddons, a flue with a gap was akin to an open flue rather than a room-sealed and balanced flue, and the consequences of the gap would be affected by a number of variables including the type of building, atmospheric pressure and wind speed. The fact that the test was carried out in a building without a wall void nullified any comparison with the premises where the fire occurred: if the wall was solid there was no opportunity for exchange of air.

[27] As regards the second of the second defenders' strands of argument, Dr Wareham doubted whether Mr Mortimore and Mr Duffy could reasonably be confident that they had pushed the boiler back into the correct position. Mr Mortimore's photographs of the fan flue, taken during his investigative visits, showed marks on the fan sleeve which were consistent with it having had the exhaust flue pipe inside it when fitted. Even if the flue pipe was installed at a 4° angle, it was possible to connect the fan flue to it with a sufficient overlap without causing the fan to sit high at the front.

[28] Support for this view was again provided by Mr Finlay and also by Mr David Bendle, who had held the position of operational incident manager with British Gas since 2006. Mr Bendle described the audit carried out by British Gas, consisting of site visits after jobs were completed, to ensure that its engineers were adhering to the operating procedures that I set out above. Mr Finlay described in detail the tasks which would be carried out by an engineer during a service visit in 1997 or 2005 in order to comply with those operating procedures. Among those tasks (in addition to taking the readings already mentioned) would be an inspection of the flue system, so far as reasonably practicable. The engineer would look inside the combustion chamber to check the seals and identify any water leaks. In the Myson Midas boiler, it would have been possible to see the location of the junction of the flue and the boiler. If there was a gap it would have been readily apparent. The engineer would have regarded this as a dangerous situation and (in 2005) would have sought the customer's permission to disconnect the gas supply. The fact that none of this occurred indicated that there was no such gap when Mr Wilson carried out his inspection on 20 March 2005. For his part, Mr Wilson asserted that although he could not recall his visit to the pursuers' premises he would have inspected the combustion chamber and would definitely have noticed had there been a gap between the flue and the back of the boiler. He would have labelled it as "immediately dangerous" and shut off the boiler. In his evidence on behalf of the pursuers, Mr Siddons agreed that an engineer performing a service would require, in accordance with the relevant regulations and British Gas Servicing Policy, to check that the combustion system, including the flue, was in good order, and to take action if he discovered a gap.

[29] The evidence of Dr Wareham also formed the basis of the third strand of the second defenders' case. Dr Wareham disagreed with the conclusion of Mr Mortimore and Mr Duffy that there was no location other than the wall void at which the fire could have started. The most common cause of boiler fires was failure of electrical components. The pattern of damage in the present case was consistent with the fire having started within the kitchen cabinet containing the boiler and having remained in the cabinet for a period of time. If the cabinet door was closed, the fire could have developed and then burned through the ceiling above, or, if there were holes in the plasterboard in the vicinity of the boiler, the fire could have spread from the cabinet into the wall void. The hot gases would have taken whatever route was available to them. The lack of significant damage to the electrical components on the top of the fan was not inconsistent with an electrical origin: the fan was in a sealed unit within the boiler which the fire could not have penetrated. Dr Wareham agreed that the fire damage to the floor of the kitchen was caused by debris falling from above. He disagreed, however, with the view that a fire which originated in the cupboard would have caused more extensive damage to the kitchen: if it found less resistance in the ceiling above the cabinet it would have broken through there before spreading throughout the kitchen. The parties' reports of burning smells during the weeks prior to the fire could be explained by occlusion of the air supply resulting in the casing of the boiler overheating and pyrolysing the wall fixings.

Assessment: the seat of the fire
[30] In assessing which evidence to prefer I consider it appropriate to begin by making findings as to the location where the fire originated. The candidates are (i) the wall void behind the boiler, as proposed by Messrs Mortimore and Duffy and (ii) the electrical circuitry in the lower part of the front of the boiler casing, as proposed by Dr Wareham. Based upon interpretation of the eye witness evidence of Officer Kinnear and the pattern of fire and smoke damage to the property, I find the conclusions reached by Messrs Mortimore and Duffy to be more persuasive as to the seat of the fire. It seems to me to be clear from the evidence of Officer Kinnear that when he arrived at the scene the fire was well established throughout the roof void but only within a limited area of both the upper and lower flats. I accept the opinion of Mr Mortimore and Mr Duffy that that pattern of damage is more consistent with the fire having originated and spread within the voids of the building. This of itself is not inconsistent with Dr Wareham's scenario in which the fire could have spread through a hole in the ceiling above the cabinet containing the boiler, thereby reaching the ceiling void, the upper flat and the roof void but not the remainder of the downstairs kitchen. As Mr Duffy points out in his report, there was extensive smoke and hot gas damage to the upper flat which must have occurred prior to the collapse of the roof. However, I accept Mr Duffy's opinion that it is unlikely that the fire spread from the ceiling void downwards to the wall void behind the boiler, where extensive damage occurred. It is also unlikely, as Mr Mortimore observed, that the fire would have spread directly from the lower part of the cabinet, where the electrical circuitry was located, to the wall void as the plasterboard would have formed an effective fire barrier. I accept Mr Mortimore's view that spread of fire upwards from the electrics would have been inconsistent with the survival of the fan wiring which, even though located within the combustion chamber, would have melted as a result of the very high temperature of the gases in the surrounding boiler casing.

[31] It seems to me that in this as in other respects, Mr Mortimore and Mr Duffy had a considerable advantage over Dr Wareham in having been able to conduct an extensive investigation of the locus of the fire. Dr Wareham's view was based to a significant extent on his experience that a fault in electrical circuitry was the most common cause of fire. I do not doubt that this is so but I do not consider that it carries much weight in the circumstances of a particular fire which has been thoroughly investigated. I did not understand Dr Wareham to state in terms that he considered that on balance of probabilities an electrical fault was the most likely cause of the fire; his position was rather that the evidence was consistent with such a cause. My impression was that as that evidence was put to him he experienced some difficulty in maintaining this position. I have in mind, for example, his explanation of why the placing of timbers against the exhaust flue would not simply be a coincidence if the cause of the fire was electrical: the suggestion that this could cause the boiler to cycle and thereby overload the electrical circuit was not supported by any evidence of difficulties having been experienced with the boiler in the period prior to the date of the fire. Certain marks visible in photographs of the flue pipe which Dr Wareham regarded as evidence of overlap were, on closer inspection, clearly unrelated to the connection of the components. In the end I was left with the impression that Dr Wareham's views were coloured by a desire to find reasons to challenge the scenario advocated by Mr Mortimore and Mr Duffy, and that his evidence lost balance as a consequence. Dr Wareham's own scenario, i.e. that the fire could have originated in the electrical circuitry and spread from the cabinet containing the boiler by burning a hole in the ceiling or by utilising existing holes leading into the wall void, did not appear to me to be rendered more likely than not by any of the evidence to which he referred. I attach no weight to the fact that Mr Mortimore imprinted a V shape on one of his photographs to demonstrate that fire spreads upwards; at the time when fire broke out the wooden battens near the base of Mr Mortimore's V would have been behind the plasterboard and not accessible by a fire originating in the circuitry of the boiler itself.

The mechanism of ignition
[32] Having decided that the fire originated in the wall void, I now turn to consider the means by which the timbers in that void are likely to have caught fire. The mechanism proposed by Mr Mortimore and Mr Duffy, namely the escape of hot gases through a gap in the flue resulting in pyrolysis and eventually ignition of the timbers does provide such a means, provided that I am satisfied (i) that such a gap existed, and (ii) that pyrolysis to the requisite degree was capable of occurring at this location.

(i) Existence of a gap
[33] The strongest evidence of the existence of a gap, were I to accept it, would be the result of the experiment carried out by Mr Mortimore and Mr Duffy when they attempted to replace the boiler in its pre-fire position vis-à-vis the flue. I found both Mr Mortimore and Mr Duffy to be impressive witnesses. I was entirely satisfied that the aim of each of these two experts since first instruction has been to establish the true cause of the fire rather than to promote a possible cause favourable to the interests of a particular party. I accept that the re-positioning of the boiler was carried out after careful measurement of the likely location of the face of the wall, and that any error would be limited to one or two millimetres, i.e. not enough to call into question the validity of their conclusion that a gap had existed, as depicted in the photographs referred to in the course of the proof. I also accept that the gap in the outer pipe was of sufficient width to enable the gases to escape into the wall void rather than, as suggested on behalf of the second defenders, only into the gap between the boiler casing and the wall plate upon which it was mounted, which gap would have been within the kitchen itself. It cannot now, of course, be known what size of hole existed in the plasterboard immediately behind the wall plate, but the width of the gap measured by Messrs Mortimore and Duffy would in any event have been sufficient to allow gases to escape into the void, especially if there was a pressure difference or a directional element to the escaping jet.

[34] I do not altogether dismiss the discrepancies between the evidence of Mr Mortimore and Mr Duffy as to the precise relationship of the inner end of the flue to the back of the boiler and the fan exhaust sleeve, although I do attach weight to the fact that neither of them regarded these discrepancies as significant. Were I to choose between them, I would prefer the analysis of Mr Duffy. His opinion that the gap was wider at the top of the pipe than at the bottom is consistent with the fact that the flue had been installed, presumably intentionally, at an angle to the horizontal. It is also possible to detect marks within a millimetre or two of the end of the inner pipe which are not quite parallel to the end of the pipe and which therefore support Mr Duffy's view that the lower though not the upper part of the inner pipe may have overlapped slightly with the fan outlet pipe, at least when the flue assembly was hot. This very small degree of possible overlap is not inconsistent with the findings of Mr Mortimore and Mr Duffy when they pushed the boiler back in place, given the margin of error accepted by Mr Mortimore. I do not regard Mr Mortimore's mistake regarding the feasibility of assembly of the apparatus in accordance with the manufacturer's instructions as significantly damaging to the experts' shared opinion that a gap existed between the outer flue pipe and the back of the boiler. The fact that there was nothing other than the flue gasket to prevent the pipe being pulled out was described by Mr Finlay as an unusual design, no longer in use, and it does not appear to me to make any material difference to the question that must be determined in this case, namely whether a gap existed between this particular flue and boiler.

[35] Turning to the arguments presented by the second defenders against the existence of a gap, I am satisfied that these fall to be rejected. In my opinion the test carried out by Mr Finlay is uninstructive with regard to the circumstances of the present case. I agree with Mr Siddons's opinion that the fact that the property in which the test boiler was installed had no wall void constitutes a clear and decisive distinction from the fire property. The existence of such a void is critical to the explanation by Mr Mortimore and Mr Duffy, which I accept, of how the boiler could operate satisfactorily for a period of many years despite the existence of a gap between boiler and flue. It is unsurprising that in the absence of a wall void a gap would cause the boiler to produce test results giving cause for concern and that occlusion of the exhaust pipe would interfere with its operation. It would not, in my opinion, be safe to draw any inference from these results in relation to the operation of a boiler where air could be drawn from, and hot gases could discharge into, a void in much the same manner as they would be drawn from and discharge into atmosphere outside the building. Nor do I accept that I should infer from the fact that the boiler was serviced on one or more occasions by the second defenders' employees without any gap being observed that no gap existed. Absence of observation of a gap is equally consistent with a breach of duty of care on the part of the second defenders' employees.

[36] Counsel for the second defenders placed emphasis on measurements taken by Mr Mortimore, and recorded in a diagram annexed to Mr Duffy's report, of horizontal distances associated with the flue pipe and the components of the wall and the rear of the boiler. She acknowledged that Mr Mortimore had measured a space of 29mm between the inner face of the external stone wall and a wooden batten to which the boiler was attached. However, it was put to Mr Mortimore that at least one of his photographs appeared to indicate that the batten was hard against the wall. If that were so, then the flue was capable of being inside the boiler and, in any event, any gap would have been located between the rear of the boiler casing and the wall mounting plate, i.e. effectively within the kitchen and not within the wall void. I am not persuaded that the space recorded by Mr Mortimore did not exist, although it may have been less than the 29mm measured after the fire. A photograph taken at an angle showing the aftermath of the fire does not, in my opinion, afford a reliable basis for a finding in fact that there was no space at all between the plasterboard and the batten. I am not therefore persuaded that if any gap in the flue existed it must have been wholly on the kitchen side of the wall mounting plate. There is, in any event, no evidence that prior to the fire the plasterboard behind the wall mounting plate was breached only by a neat hole round the flue (as in the photographs of the property at which the test boiler was installed) so as to create a barrier between the space behind the wall mounting plate and the wall void.
(ii) Occurrence of pyrolysis
[37] Mr Mortimore, Mr Duffy and Dr Wareham were in substantial agreement regarding the mechanism by which pyrolysis occurs and the factors relevant to determining whether it might or might not occur in particular circumstances, although they provided slightly differing figures for the temperatures at which timber would normally ignite and at which it could ignite if pyrolysed. Based on his experience of ignition of other substances, Dr Wareham considered that the likely temperature within the wall void was not conducive to pyrolysis, although he did not wholly exclude the possibility that this was what had happened. It seemed to me, once again, that Dr Wareham's opinion was based generally on his past experience and did not take due account of the very unusual circumstances of the present case, namely a boiler flue with gaps in its inner and outer pipes and obstructed at its outlet during the period prior to the fire. I accept the evidence of Mr Mortimore and Mr Duffy that the circumstances of the present case, including the existence of a gap, the proximity of wall timbers to the flue pipe, and the temperature of the escaping gases, were such that pyrolysis was a competent method of ignition of timber within the void. Its occurrence is consistent with the time scale during which the exhaust flue was obstructed (after many years of apparently trouble-free operation) and is supported by the reports of burning smells during the weeks and months prior to the fire. Having regard to the whole circumstances of the case, I find that this was what occurred. The pyrolysed timbers having ignited at some time during 4 March 2006, the fire spread upwards through the wall void into the ceiling void and the roof void, breaking through into the upper flat. Once the ceiling of the lower flat kitchen had been penetrated by the fire in the void and flaming debris from above, burning matter fell down into the kitchen, resulting in damage to the fittings in the south-west corner including the boiler and the cabinet within which it had been installed.

Conclusion: combination of contributory factors

[38] The opinion of both Mr Mortimore and Mr Duffy, based upon the mechanism that I have described, was that the fire occurred as a consequence of the combination of the gap in the flue pipe and the obstruction of the exhaust flue outlet. Without the gap in the pipe, obstruction of the flue outlet would have been likely to cause the boiler to cease to operate and the fire would not have occurred. Without the obstruction of the flue outlet, exhaust gases would have continued to escape to atmosphere, despite the existence of the gap, and the fire would not have occurred. As I have accepted the evidence of these expert witnesses as to the mechanism by which the fire started, I also accept their evidence that these two factors in combination were the cause.

Grounds of fault

(i) Case against the first defender

[39] The pursuers submit that the fire was caused by the negligence of the first defender in blocking the exhaust flue by placing the pieces of timber against it. This action was obviously dangerous and constituted a breach of the first defender's duty not to act in such a way as was likely to cause damage to his neighbours' property. Separately, the first defender committed a nuisance in respect that the blocking of the flue constituted an invasion of the pursuers' interest in their property which exceeded what was reasonably tolerable. Reference was made to Kennedy v Glenbelle Ltd 1996 SC 95.

[40] The submission in response on behalf of the first defender was that the pursuers had failed to prove that he was responsible for placing the pieces of timber over the flue outlet. For the reasons given earlier in this opinion, I reject that submission. It was not contended on behalf of the first defender that the blocking of the flue outlet by the pieces of timber was not a material contributory factor. Nor was it submitted that if the first defender was found to have blocked the flue outlet, this did not constitute negligence or nuisance.

[41] In Kennedy v Glenbelle Ltd, Lord President Hope made clear (page 99) that liability for negligence and nuisance are both species of delictual liability. His Lordship observed:

"According to the law of Scotland liability in reparation for damages arises either ex contractu or ex delicto. There is no other basis on which a liability in reparation for damages can arise, according to our law. A claim of damages for nuisance is a delictual claim, as it does not depend for its existence on any contract. It arises where there is an invasion of the pursuer's interest in land to an extent which exceeds what is reasonably tolerable. The plus quam tolerabilile test is peculiar to the liability in damages for nuisance. Where that test is satisfied and culpa is established, the requirements for delictual liability are fulfilled. Liability in damages for negligence, on the other hand, depends on a failure to take reasonable care where there is a foreseeable risk of injury. That is another species of delictual liability, the basis for which also depends upon culpa."

I am satisfied that in the circumstances of the present case the pursuers have established liability on the part of the first defender for damage caused by both negligence and nuisance. The blocking of the flue outlet created a foreseeable risk of injury to the pursuers' property, if only in that it could have caused damage to the boiler by preventing its normal and safe operation but potentially with much more serious consequences, as unfortunately occurred, and accordingly constituted a breach of the duty to take reasonable care owed by the first defender to the pursuers. It was also an invasion of the pursuers' interest to an extent which exceeded what was reasonably tolerable.

[42] For completeness I record that before commencement of the proof the pursuers amended their pleadings to insert an esto case on the hypothesis that I was not satisfied that the first defender was personally responsible for placing the wood in front of the flue outlet. On this hypothesis it was submitted that the first defender had owed a duty to the pursuers, on becoming aware that the flue was occluded, to unblock it by removing the pieces of wood. This submission was resisted by the first defender on the grounds, firstly, that no such duty was incumbent on the first defender as Scots law did not impose liability for pure omissions and, secondly, that the amendment came too late as it was made after expiry of the 5-year prescriptive period. In the light of my finding that the first defender did place the wood in front of the flue outlet it is unnecessary for me to express an opinion in relation to this alternative case.

(ii) Case against the second defenders

[43] The pursuers averred that the fire was caused by negligence and breach of contract on the part of the second defenders. Ultimately the pursuers' submission was focused only on material breach of the service contract between the pursuers and second defenders, consisting of a failure by the engineer (Mr Wilson) who carried out the visit in March 2005 to identify the gap that existed between the flue pipe and the boiler housing and to take appropriate action to make the necessary repair or, alternatively, to ensure that the boiler could not continue to be used in its unsafe condition. It was submitted that there was no doubt that the connection between the flue and the boiler housing ought to have been inspected in the course of that visit. Reference was made in this regard to unchallenged evidence given by Mr Siddons under reference to regulation 26(9) of the Gas Safety (Installation and Use) Regulations 1998, the industry practice documents produced at the time by CORGI (Council for Registered Gas Installers), and the second defenders' own instructions to their employees. I need not examine these sources in detail because I did not understand the second defenders to dispute that the service undertaken by their engineer in March 2005 ought to have included inspection of the flue system; as I have already noted, the evidence of Mr Finlay and Mr Bendle was that it would be the engineer's duty to inspect the flue system so far as reasonably practicable and that the flue connection in the boiler with which this case is concerned was capable of inspection. Proceeding on the basis that there was a gap, the pursuers submitted that I should find that the engineer who visited either failed to inspect the flue connection or inspected it but failed to notice the gap. On either scenario the second defenders were in breach of the express terms of the contract that I have set out at paragraph 5 above and also an implied term of the contract which required their employee to carry out the inspection to the standard required of a reasonably competent gas engineer. These submissions were adopted on behalf of the first defender.

[44] On behalf of the second defenders it was submitted that I should find that if there had been such an obvious gap between the boiler and the flue, Mr Wilson would have seen it and would have fixed it or shut off the boiler. No gas engineer who had serviced the boiler, whether in 2005 or 1997 or perhaps earlier, had seen the gap. It was stretching credulity to come to a view that such a dangerous defect had for 12 years gone undetected by engineers visiting to service and repair it, including taking readings of the CO/CO₂ ratio of its flue gases. The reason why no-one had detected the gap was that there was no gap.

[45] I have already found in fact that at the time of the fire there was a gap between the flue and the rear of the boiler casing. No-one contended after proof that this gap developed at a time after the boiler and flue had been installed (although this was the first defender's position on record); by the time of the parties' submissions it was common ground that if a gap existed then it must have been there since 1993. I recognise the experience and expertise of Mr Finlay and Mr Bendle and I do not reject their evidence in any respect merely because they have been employed by the second defenders throughout their careers and continue to be so employed. Nevertheless on the basis of my findings as to the cause of the fire I must reject their opinion that the boiler could not have given satisfactory gas readings if there was a gap in the flue. I must also conclude that the reason why Mr Wilson failed to identify the existence of the gap was that he failed to fulfil the duty incumbent upon the second defenders when carrying out his examination of the boiler, and in particular that he either failed to inspect the fan connection or inspected it but failed to observe a gap which a reasonably competent gas engineer ought to have observed. I therefore hold that the second defenders were in material breach of their contract with the pursuers and that such breach was a material contribution to the loss and damage sustained by the pursuers as a result of the fire.

Apportionment of liability
[46] Having found both the first and second defenders to blame for the occurrence of the fire, I require to make an apportionment of liability in accordance with section 3 of the Law Reform (Miscellaneous Provisions) (Scotland) Act 1940. In my assessment, a significantly greater share of the liability falls to be apportioned to the first defender. I find no basis for treating either defender's contribution as having greater causative potency. So far as blameworthiness is concerned, however, I regard the contribution of the first defender as significantly more blameworthy. I bear in mind that the second defenders are liable not as persons who carried out a negligent installation of the boiler and flue but rather as persons who failed, in breach of contractual duty, to detect that a negligent installation by someone else had occurred. The second defenders' failing was an omission by an employee to exercise the requisite degree of care. The first defender's action in obstructing the flue outlet was a deliberate act with obvious potential to cause danger. In these circumstances, I consider it just to apportion two-thirds of the blame to the first defender and one-third to the second defenders.

Disposal
[47] In order to give effect to my opinion I shall sustain the first plea-in-law for the pursuers, the fifth plea-in-law for the first defender and the eighth plea-in-law for the second defenders (the latter two being the defenders' respective pleas regarding apportionment of liability), and I shall repel the defenders' other pleas-in-law. I shall grant decree for payment to the pursuers by the defenders jointly and severally of the sums, and with interest thereon, all as set out in the parties' joint minute. Questions of expenses are reserved.