Anthony & Ors v The Coal Authority

This judgment is arranged in the following sections:

This is a claim in nuisance brought by seven residents of the villages of Grovesend and Waungron near Swansea, West Glamorgan, against the successor of the National Coal Board (NCB) and British Coal Corporation, The Coal Authority. The source of the alleged nuisance is a spoil tip designated No. 52 by NCB. Tip 52 served Brynlliw underground colliery principally during the period 1957-72 although coal was earlier recovered from Brynlliw between about 1908 and 1925. Tip 52 was closed in about 1971 and was declared a disused tip in 1983 under the Mines & Quarries (Tips) Regulations 1971. In 1987, with a view to returning the land to those with commons rights, NCB re-shaped and partially landscaped the tip. On 22 June 1995 the land on which the restored tip was situated was conveyed to a group of West Glamorgan Commoners for a consideration of £1.

During the summer of 1996 a fire broke out at the south western flank of the tip causing, in time, clouds of smoke and noxious fumes adversely affecting, say the claimants, the use and enjoyment of their neighbouring properties. The fire remained seated for a period of upwards of three years until remedial works commissioned by West Glamorgan County Council finally extinguished it in 2000. On nine occasions conditions were such that visibility along the M4 motorway, situated a few hundred yards to the east, became dangerously reduced causing the motorway to be temporarily closed to traffic.

Eileen Anthony, Eirwina Richards, Caroline and Peter Arthur, and Carey Knox are all home owners in Pentre Road, Grovesend. Eleanor and Richard Hill own their home in New Road, Grovesend. Mairwen Hughes owns her home at Maes yr Ellen, Waungron.

Brynlliw Colliery was situated some 9 kms to the north west of Swansea. Tip 52 occupied an area of land aligned about 500 metres long north to south and 300 metres wide west to east between the villages of Waungron in the north and Grovesend in the south. Pentre Road (also called by several witnesses Pontardulais Road) joins the two villages along a north-south route immediately to the west of the tip. Maes yr Ellen is a farmhouse to the east of Waungron and New Road is in the centre of Grovesend village.

The claimant's case is that during tipping operations on Tip 52 in the period 1957- 1972 NCB created an unreasonable fire hazard because, as formed, the tip was foreseeably liable to catch fire. The Coal Authority accepts responsibility for the acts of its predecessors. The fire which broke out in 1996 was the result of spontaneous combustion caused by the hazard created by NCB; alternatively, the tip was foreseeably liable to combustion ignited by an external source. The claimants' case is pleaded in public and private nuisance. It is the claimants' case that NCB created the nuisance. Accordingly, the test is not one of a measured duty of care but foreseeability only. Further, or alternatively, NCB knew or should have known when they re-formed the tip in 1987 that it presented a fire hazard and failed to take reasonable steps to render it safe.

The defendant's case is that the creation of the tip was a reasonable user of land. The tip as formed was safe. Damage from the effects of fire caused by spontaneous combustion was unforeseeable. The 1996 fire was ignited by a third party. Even if the tip was liable to spontaneous combustion the risk of fire was so small that the use of the land was nevertheless reasonable. Nothing was done in 1987 to enhance the risk. When the land was returned to the commoners in 1995 NCB acted reasonably in requiring a covenant for aftercare of the tip.

It is agreed that the principal issues I need to address are:

The first three issues identified all require analysis of the evidence how the tip was constructed and the combustibility of the material within it.

Before extensive mechanisation after the Second World War coal was won by hand tools and the separation of coal from waste largely took place underground. Waste material was used to pack the voids created by the advance. Following the use of mechanised face cutters a good deal of waste material was brought to the surface together with and/or attached to the coal. That which was commercially recoverable was treated in the coal preparation plant at Brynlliw. The residue from the plant and all other mining waste was tipped.

It is agreed that the first waste tipped between 1957 and 1960 was untreated material generated by the deepening of the shafts. It probably contained pure coal as the shafts were driven through the seams. Coal production began in 1960 and for 7 years Tip 52 was the only recipient for waste. It was then known as Tip No 1. In 1967 tipping commenced on the adjoining but separate Tip 750 (later called 850) situated on the east side of the railway line which bisected the site. At first, Tip 750 was called No 2. I shall henceforward refer to the two tips as Nos 52 and 750 respectively. Between 1960 and 1969 tipping at the southern end of Tip 52 included coal duff (small sized coal) and washery filter cake (very fine dirt/coal collected by the washery filters). During the years 1969 and 1970 the southern end was extended with coarse material to shield Grovesend village from the duff and filter cake. In 1972 tipping on Tip 52 ceased and had probably been declining for a year or so before. Peaty soil was imported from a nearby common and the flanks planted with trees.

There was mined at Brynlliw both anthracite and dry steam coal. These are both high rank coals. In the United Kingdom coals are classified by rank determined by its carbon and volatile matter content. Anthracite (rank 101/102) has a high carbon and a low volatile matter content. Lignite (rank 901) has a low carbon and high volatile matter content. It is agreed that the coal waste at Brynlliw would have comprised Ranks 102 and 201. High rank coals are the least susceptible to spontaneous combustion (self heating). The proportion of the lower rank dry steam coal in the tip is not agreed and I shall return to the issue below.

Spontaneous combustion is a phenomenon the product of natural oxidation of a mass of coal or other carbonaceous materials (e.g. wood, shale, hay). They react chemically with oxygen to produce heat (the exothermic process). If the heat is not dissipated by conduction or convection the temperature of the mass increases. The rate of chemical reaction increases until the process becomes self-sustaining and the ignition temperature of volatile matter in the material is reached. Heating may take place without ignition. Only if the ignition temperature is reached will fire break out. It was necessary on occasions during the evidence to seek clarification from witnesses whether they were using the term 'heating' in the pre- ignition or post- ignition sense.

Other factors are believed to influence the risk of self heating to ignition. Small particle size provides greater surface area per unit mass creating an increased exposure of carbon to the air. High moisture content, particularly caused by sudden wetting, can feed the process with oxygen. A high pyrite content is believed to make a minor contribution to self heating properties. Pyrite will itself oxidise causing a volume increase and micro fracturing of the coal of which it is part leading to an increase in surface area per unit mass. The susceptibility of a mass of coal or coal waste to spontaneous combustion, while known, is not perfectly understood. It is common ground that whether spontaneous combustion actually occurs will depend upon a number of factors including rank, geology, topography and local meteorological conditions.

Anthracite has the highest ignition temperature of 520ºC in coal dust layers while bituminous coal (including dry steam coal) has the lowest at 170ºC.

It is generally accepted that, in the absence of disturbance, the longer the mass has been exposed the less likely it is to reach ignition since the mass pre-oxidises or 'weathers' with time.

The documentary evidence in the hands of the defendant has been examined by Dr J S Edwards an expert mining engineer instructed on behalf of the defendant. Following graduation Dr Edwards was employed in 1968 by NCB as a mining engineer. In 1969 he became Undermanager at Cadley Hill Colliery in the South Midlands Area. It was a mine beset with risk from spontaneous combustion in some of its coal seams. Dr Edwards gained experience in the prevention and fighting of fire underground. In 1972 he joined the Mining Department at Nottingham University where he remained until 2004. He is now a consultant with Nottingham University Consultants Limited.

Dr Edwards has noted that Mr E H Browne prepared a paper dated 10 November 1953 (3/2) for the NCB Finance Committee upon the South Western Division's proposals to redevelop Brynlliw Colliery. The proposal was to mine Swansea steam coal. The proposal was referred back for improvement and was later re-presented. Mr W V Sheppard on 3 April 1957 prepared a further paper (3/5) for the Finance Committee and again identified an intention to mine Swansea steam coal. In describing the available reserves of 40 million tons he referred to the presence of coal rank 201 in each of the Swansea 3ft, 4ft and 5ft seams, and coal rank 102/201 in the Swansea 6ft seam.

The Finance Committee recorded a Minute that marketing section intended to sell the coal produced at Brynlliw as anthracite and dry steam coal.

The Guide to the Coalfields (1957-1983) and the Colliery Year Book (1959-1964), published annually relying upon information provided by the coal industry, are inconclusive.

Dr Edwards, with the assistance of Mr Ian Williamson, consultant geologist specialising in coalfield geology, has attempted a more sensitive analysis. He refers to HF Adams' monograph of The Seams of the South Wales Coalfield (1967) (7/7) Figure 16 to demonstrate that the shafts at Brynlliw fell within the anthracite 102 coalfield (volatile content 6.1-9%). Brynlliw was, however, within ¼ mile of the low volatile (9.1-13.5%) dry steam coal field. Figure 16 itself is qualified by the words 'where two of classes coal occur together the shading denotes the dominant coal class'. In Dr Edwards' opinion the greater part of the workings at Brynlliw was within the anthracite zone. He estimates that the waste produced by the colliery would have included coal in the proportion 70% anthracite and 30% dry steam coal. Even the dry steam coal should, in Dr Edwards' opinion, be sub-classified as 201a (volatile matter 9.1-11.5%) rather than 201b (11.6-13.5%) since this would be consistent with the analysis of material recovered from the tip by Dr Blandford in 1998.

Dr Michael Richards of SRK Consulting, instructed on behalf of the claimants, differs little from Dr Edwards in his estimation of the waste material in Tip 52. Dr Richards also has personal experience of spontaneous combustion problems in coal mining, in his case in waste tips in South Africa where he worked in production mining. Like Dr Edwards he became a lecturer and senior lecturer at Nottingham University until he left to join SRK in 1995. He points out that coal is classified as bituminous rather than anthracite if it has a volatile content above 9%. The adjusted dry ash free volatiles content of the samples taken from the tip in 1998 was 10.2- 11.2% consistent with a 201a dry steam coal (bituminous) classification. It is agreed between Dr Richards and Dr Edwards that the coal recovered from the Swansea 6ft seam would have been 'dirty' and friable. Friability is relevant since fine fractions are difficult to separate. The coal won may not have been recoverable in the washing process and, if not, the waste discarded would have contained a high proportion of coal. The differences in coal content found upon analysis of Dr Blandford's samples (para 24 below) tend to confirm the unpredictability of coal content in the tip.

Of the coal content within the waste, Dr Richards estimates that the proportion would have been about 60% anthracite and 40% dry steam coal. He reaches this estimation by an examination of Adams' monograph and an analysis of the faces worked at Brynlliw until 1971. The picture is further complicated by the possibility, accepted by both sides, that coal from Mountain (dry steam) and Cynheidre (anthracite) collieries was washed at Brynlliw and some of the waste deposited on Tip 52 before 1970. He has left this possible complication out of account in reaching his proportion of 60% anthracite. Both Dr Richards and Dr Edwards agree that at best their analysis of the available evidence has resulted in a 'guestimate'. Dr Richards observes that the boundary in classification is, in any case, arbitrary and the burning properties between the two coals should be regarded as marginal. He accepts that ordinarily a tip composed of the waste products to be found on Tip 52 would be regarded as presenting a low risk of spontaneous combustion. That part of the tip which represented a risk was comprised of dry steam coal waste. Anthracite alone is not considered a risk for spontaneous combustion.

Dr Malcolm Blandford is a chartered geologist and principal of Blandford Consulting. He has long standing experience as a geologist in the South Wales coalfield where he was employed by NCB for 12 years. Dr Blandford was instructed in 1998 to make a site investigation and to provide technical information with a view to commercial recovery of coal from Tip 52. If viable that recovery could have paid for extinction of the fire and the reclamation of the tip. He provided to the Trustees of Mynydd Lliw Lands Trust and CDN Developments Ltd his 'Results of Site Investigation' (8/2), and to The Vale of Neath Coal Company Limited a 'Technical Statement' (8/3) and 'Environmental Statement' (8/4). Twelve boreholes were sunk in the tip, BH 1-4 in the lower third across the widest part of the tip from east to west, BH 5 in the north east corner, BH 6 and 7 in the upper third across the narrowest part of the tip, BH 8-11 across the southern part of the tip from east to west, and BH 12 at the central northern boundary of the tip. The Grovesend Upper and Lower Seams were located. Five Trial Pits TP 1-5 were dug into the surface of the tip at locations along and near, but not on, its western edge. Samples were taken from TP 2 (central western extremity), TP 3 (south western edge) and TP 5 (west side of neck at narrowest point northern third). Coal content varied between 2% (TP 3) and 23% (TP 5), inadequate for commercial exploitation as part of a tip removal development. It was these samples which produced the volatile matter analyses to which I have already referred.

Dr Blandford has since been instructed as an expert by the claimants. He referred in his report prepared in 2000 to an additional analysis in his Summary of Analyses (2/1/13). One excavation appeared to be pure coal. A sample was recovered which appeared on analysis to be filter cake or coal duff. On analysis the sample contained dry ash free volatile matter of 13.3%. The sample was taken by The Vale of Neath Coal Company Ltd and not by Dr Blandford. He cannot identify the location from which the sample was taken but infers that it must have been the south eastern portion of the tip since that is where records indicate the filter cake and duff were deposited. Dr Blandford is not surprised by the increased percentage of volatile matter in the sample. In his view, a range is to be expected at Brynlliw where the quality of coal mined from the Swansea seams was variable.

The evidence establishes that the coal content of the tip was variable. In parts it was small but in others it was substantial. I accept that the volatile matter content of the coal waste tipped at Brynlliw lay generally within the range 9.1%-11.2% although there may well have been pockets of coal with a content up to 13%.

The claimants rely with the agreement of the defendant upon a supplementary report of Dr Blandford of June 2005 in which he entered the debate about the rank of coal mined at Brynlliw. He points out that although Dr Edwards has accurately transposed from the line drawn by Adams the limits of the anthracite and dry steam coal coalfields in the vicinity of Brynlliw, the scale of the original drawing does not permit precise identification of the boundary. Furthermore, the original drawing should not be regarded as an exercise in precision. Dr Blandford's view is that Adams cannot be regarded as a reliable source for an estimation in what proportion anthracite and dry steam coal were extracted from Brynlliw. Dr Blandford does not hazard an estimate of his own except to express the view that 'the greater proportion of coal mined at Brynlliw Colliery, when all seams are taken into consideration, would have been of dry steam rank'.

Dr Blandford expressed the opinion in his original report (2/1/26) that, 'The analyses of the coals show that they have a fairly high volatile matter content and are, therefore, a rank of coal that could ignite fairly easily'. In evidence Dr Blandford told me that this did not represent his opinion and he wished to change it as follows: 'The analyses of coals show that they have a volatile matter content of 9%-11.2% and are, therefore, a rank of coal that has been known to combust spontaneously'. Dr Blandford conceded that he had no specific expertise in fire in coal tips and would defer to the other experts who did. When asked in cross examination what propensity he believed 201 rank coal had to combust, low, medium or high, Dr Blandford replied, "Medium, I would say". Pressed to comment on the agreement reached between Dr Richards and Dr Edwards, Dr Blandford said he would accept 'low' but not 'very low'. In reaching that view he relied upon the reputation of the South Wales coalfield which he had learned over many years.

Both Dr Richards and Dr Edwards have long term relevant experience. They have reached agreement upon the relevant available evidence. The agreement upon which I shall act is that the spoil contained waste anthracite and dry steam coal which the industry regarded and continues to regard as presenting amongst the lowest propensity for spontaneous combustion in the scale of rank 100 (anthracite-lowest) to 900 (noncoking coals- highest). I find the proportion of anthracite to dry steam coal was about 60%-70%. The difference is marginal to the issue whether the tip created a fire hazard since the experts agree that, while the risk of spontaneous combustion could not be assessed upon a scientific consideration of its coal content at zero, it was 'very low'.

A good deal of expert time and research has been devoted to the method of construction of Tip 52. The debate has centred upon the quality of compaction of the tip. Tips which have been laid in layers, properly compacted, and have shallow flanks reduce the liability of air to ingress the mass of the tip, and thus to contribute to the exothermic process leading via heating to ignition.

The significance of the dispute is in large measure dissipated by the knowledge that the fire which caused the alleged nuisance commenced in the south western flank of the tip. Dr Richards and Mr Bacon agreed that "the surface of the western flank of Tip 52 would not have been compacted. On account of its slope [the evidence suggests 30%-34%] which was roughly the angle of repose of the material, up to a few metres of loose material would be likely to be left uncompacted close to the face of the flank". In other words, when the waste material was spread and levelled along the top surface or plateau of the tip it would, at the flanks, be pushed over the edge and left where it came to rest. They agreed, "that the uncompacted surface tip material in the western flank was more likely to ignite due to spontaneous combustion (if it contained coal of the type that is liable to self combustion), to be ignited by an external source of heat and to allow the spread of fire than if it was compacted...[T]hat Tip No 52 was not remarkable in its construction compared with other tips built by the NCB prior to Aberfan. Mobile earthmoving plant began to be introduced at this time, though many tips continued to be operated using fixed transport, systems which resulted in loosely placed spoil and steep perimeter slopes. Steep perimeter slopes remained the norm with the earthmoving plant operated tips but the layered spoil received more compaction from the passage of heavily laden vehicles".

I have considered the documentary evidence available to the experts. I accept that there is room for disagreement as to the degree to which the majority (what I have called the plateau) of the tip was compacted. The agreements reached as to the flanks, however, represent sound scientific inference or sound common sense and I accept them. Since the fire of 1996 commenced in the south western flank of the tip, considerations relevant to foreseeability of spontaneous combustion and the cause of the fire of 1996-2000 should be focused, not on the body of the tip, which may (or not) have been sufficiently compacted to prevent a fire, but upon the western flank which, Mr Bacon confirmed in evidence, was probably not compacted at all. As will become apparent all incidents of burning in Tip 52 relied upon by the claimants to prove susceptibility occurred on the flanks. It is not necessary, in my view, to reach a conclusion upon the degree of compaction within the body of the tip.

Dr Richards has concluded that the scientific evaluation of the coal content and coal rank at Brynlliw as a very low risk for self- heating is displaced by evidence of the history of coal seams and tips in the area. Experience has shown that a tip composed of low rank coal may not combust spontaneously while, on the other hand, a tip comprised of high rank coal will. The history of the tip and local conditions must be explored. The history of the coalfield in the vicinity of Brynlliw, in Dr Richards' view, exposes an identifiable risk of spontaneous combustion in Tip 52 despite the comparatively high rank of coal within it.

Dr Richards relied upon the following:

At a Symposium on The Prevention of Spontaneous Combustion held at Harrogate in November 1970, Paper No 15 was presented by RA Evans, District Rescue Stations Manager (South Wales) and R Hart, Assistant Area Ventilation Engineer, West Wales Area. They reported, among other incidents, spontaneous heating in the roadside pack of a longwall advancing face underground at Morlais. The incident occurred in September 1968. The volatile content of coal in the heated zone was 10%-11%. Morlais mined the 'high rank Swansea Five Feet seam' (3/21/90). Digging out the area of heating enabled fresh air to reach it resulting in an open fire. The Symposium discussed methods of packing the roadway to reduce the risk of heating. In his summing up to delegates, which was recorded and distributed to NCB areas, Dr Willett said that it was most unusual for spontaneous combustion to occur underground in any coal seam of higher rank than 500 but made particular mention of 'the five cases of spontaneous combustion in South Wales in high rank coals (coal rank code 200) which are reported in Paper No 15'. This, in my view, was ample evidence that dry steam coal could not be regarded as risk free. The volatile content of the coal which ignited in each case reported in the paper fell within the range 10%- 13%.

Tip 60 at Morlais was visited by an NCB surveyor, AN Lane, on 27 January 1970. The south western part of the tip was burning. In 1974 when a proposal for the removal of red and black shale from the tip was being considered the colliery manager, D Roberts, recommended to the Chief Mining Engineer for the Area that the remainder of the tip would need to be consolidated to protect against the risk of heating. On 1 March 1978 the District Civil Engineer recommended that in order to avoid the risk of spontaneous combustion the coal waste when tipped should be rolled in 300mm layers. Here is a case of coal waste with a similar volatile matter content to that at Brynlliw, a short distance away, being treated with suspicion and caution.

It follows that despite the high rank of coal NCB had knowledge of its susceptibility to spontaneous combustion.

Huw Jones, a witness for the defendant, was a civil engineer working for WS Atkins Consultants Limited. He was instructed by the former Lliw Valley Borough Council to advise upon the identification and treatment of burning colliery tips at Garncoch Nos 1 and 3, within 4 miles to the south east of Brynlliw. In April 1977 a local newspaper reported the fire to be burning fiercely in the edge of the tip. There is no direct evidence of the cause of the fire available to me. This is a fire recalled by Eric Davies who was NUM secretary at Brynlliw and a member of the Mines and Rescue Team. He referred to it as a heating which became a fire on spontaneous combustion. About 12-18 months after the colliery closed at Garncoch the fire caused problems for the residents of Penllergaer. The fire eventually burned through the whole tip. It is common ground that once established a tip fire can be intractable whatever the original cause of ignition.

Dr Richards identified fires in the tips at both Broad Oak and Caeduke collieries. I have received no specific evidence about them.

Mr Huw Jones was also instructed to advise Lliw Valley Borough Council in respect of a tip fire at Mountain Colliery, a mile or so to the south of Brynlliw. This must have been in the early 1980s. The tip was introduced to Mr Jones as a case of spontaneous combustion. He understood that the tip had been burning for many years.

Alan Webster was employed by NCB and British Coal between 1969 and 1995. From 1973 to 1989 he was an engineering geologist responsible for the Brynlliw tips. He too recalls the fire in the tip at Mountain Colliery. The Mountain tip was of a chemical and physical type similar to that at Brynlliw. In his witness statement Mr Webster thought the fire at Mountain was alight in the 1970s. In oral evidence he recalled an occasion in the 1980s when he visited the burning tip with Mr Snell, No 1 District Civil Engineer. The winder house at the top of the tip had been cut up by people looking for scrap. The winding rope had been cut up with oxyacetylene equipment. The fire had caught in timber from the winding house. Mr Webster agreed that what he observed was not the first fire at Mountain. The burning he observed was in a different part of the tip.

On 18 November 1976 West Glamorgan County Council issued a Notice under section 14(1) Mines and Quarries (Tips) Act 1969 requiring remedial operations to remove burning material from the tip. There was a dispute between the Council and NCB as to the lawfulness of the notice although the existence of the fire was not challenged.

Mr Leighfield, former chief surveyor and minerals manager for British Coal, has researched the history of Brynlliw and other tips in the area. Mountain tip comprised an amalgamation of Tip Nos 53-56. Tip 54 was a Maclane (cone shaped) tip, loosely compacted, with a large coal content. The remainder were constructed in the form of a plateau. The site was occupied by Lliw Valley Borough Council in September 1980 under a compulsory purchase order. By March 1982, when the land was conveyed to the Council, the fire had been extinguished.

Eric Davies went to live in Gorseinon next to the Mountain colliery site in August 1955. The tip was burning then and neighbours told him it had been burning since the end of the Second World War. According to Mr Davies it was widely accepted that the origin of the fire was spontaneous combustion. As will later appear, Mr Webster, in 1987, did not regard the tips in the area as at risk of spontaneous combustion because, whenever he had visited, fires could always be explained by external sources of ignition.

John Michael Lewis was employed by NCB from 1957 to 1974. He was one of a team of surveyors responsible for a group of collieries which included Brynlliw between 1957 and 1968. From 1968 to 1974 he was assistant colliery surveyor at Brynlliw. Mr Lewis remembers that Tip 52 contained a good deal of coal in the early years, certainly enough to prompt the erection of signs warning against coal picking from the tip. As a member of the surveying team Mr Lewis' job included the preparation of annual plans recording the progress and development of the tip. After the Aberfan disaster tips required more frequent inspection and the surveying team would decide what additional information should be included on the plans. There were incidents of heating at the tip between 1968 and 1974. Mr Lewis said he recorded some of them on plans submitted over the years. The plans would on completion be kept in a colliery office. They would be available to the area engineers and surveyors. In his supplementary witness statement he drew the symbol which he used. He would write the word 'BURNING' enclosed within a box drawn to represent the area subject to heating. He recalled that, partic ularly on damp days, the characteristic sulphurous smell would be noticeable. Grass and vegetation on the surface would be seen to die under the influence of a deep seated heat. Burning material would be removed and spread to cool. He would be required to inspect and survey the cooled material before it was reburied in the tip. He would prepare a report both on the proposal to deal with the fire and the success of the remedial work adopted. There would, therefore, be two reports for each excavation. Occasio nally scientists would be called to the tip to advise whether burning material had been properly extinguished. Mr Lewis recalled the existence of a team whose task it was to deal with heatings. They had access to an area gang to assist if necessary. Water would not extinguish a fire and rain would feed a tip fire by releasing oxygen. The outbreak of fire was unpredictable. They could occur in quick succession or with a considerable break between.

A number of plans have been recovered from the Coal Authority's Mining Records Department which held the abandonment records for Brynlliw. The plans were deposited by HM Inspectorate of Mines. The source of the plans which came into possession of the Inspectorate is not known. The plans recovered are dated annually from 1968-1983 (TB vol 8). It was demonstrated to Mr Lewis by Mr Darling QC that none of them contained any reference to burning areas. Mr Lewis responded that he had a specific recollection of a fire in the south western corner of the tip in 1970/71. He could recall marking the area on one of his plans. He saw the fire for himself. It was not just a heating. There was a form 320T which referred to the fire. He consulted it before completing the plan. Mines and Rescue advised about the danger of using water in an attempt to extinguish it and he thought it lasted 3 days. The fire was dug out and spread. Water was used to cool the burning material. The fire was so hot that a rubber wheeled vehicle had to be replaced by a tracked excavator. As to the other plans, he explained that a burning would not have been included if it had been extinguished by the time the plan came to be submitted. He would not record on each plan the position of an historical burning, only a burning which was causing a problem at or about the time the plan was being prepared. There would be a form 320T recording the fire which was also retained in the book kept at the colliery. Since the plan was prepared annually it may well be that although he had been aware of a burning he had not included it on the plan because it was no longer alight. In his experience the same applied to six monthly inspections by the engineers. If a burning had taken place but had been successfully dealt with before the engineer inspected, he would not expect to find reference to the burning in the engineer's report.

I have examined each of these plans. Several do indeed contain a key which explains the symbols used to locate burnings, boreholes, tree planting and other features. The symbol for burning is as Mr Lewis recalled it. In my view, these plans are an unreliable source of the history of heating and burning on Tip 52. First, it is common ground that they do not record at all in some years and do not record accurately in others the position of boreholes weekly inspected on Tip 750. There are written records for the water level readings at boreholes 3 and 4 which simply do not appear on any of the plans. Second, it is common ground, as Mr Leighfield found in his historical research, that in 1972/73 peaty soil was imported from Mynydd Lliw common and spread on the flanks of Tip 52. It was then planted with trees and seeded with grass. The trees planted on the western flank matured and were retained during the 1987 re-landscaping of the tip. Nowhere in the plans which post-dated this considerable development is there any indication that it had taken place. Third, It is common ground that significant fires took place in the tip in 1979 and 1981. Neither of them appears in the plans for those or succeeding years despite the existence of reports in which they are described. Fourth, none of the plans is signed, nor is the report of which the plan is part, identified, nor is any drawing numbered. My conclusion is that I cannot rely on these plans either as true copies of the originals or as contemporaneous documents which contradict Mr Lewis' recollection.

The second source of contemporaneous recording of events on the tip is the tip report. Mr Bacon informed me that tipping commenced before and was largely completed before the first NCB mandatory requirements for the management and control of tips were introduced on 1 January 1968 (NCB Production Instruction PI/1967/16 - see 6/2). The Instruction anticipated the regulations which came into effect three years later. It required the tip foreman to record the condition of the tip weekly in form P221. The form was to record, amongst other things, any fire hazard, including any visible burning, steam or smoke, fumes or smells of burning, cracking of the surface and collapse inwards or cavitation. The colliery mechanical engineer was to inspect and report on the tip once a month, also on form P221. The area civil engineer was to ensure that the tip was inspected by a competent person every six months and that a report be provided to the area chief engineer on form P222.

Following the disaster at Aberfan in 1966 the industry carried out a wholesale review of its tips procedures. In consequence a new regime for the construction of tips and their inspection was confirmed in the Mines & Quarries (Tips) Regulations 1971. NCB issued a new Code and Rules in 1971 (6/5).

Regulation 11 required that an active tip be inspected at least once a week by 'a competent person' who was to make a report of the condition of the tip and of any defects within it, and the action taken to remedy any defect. By NCB's Codes and Rules 1971, the competent person was the tip foreman. The form he was required to use to record the condition of the tip was P224 (formerly P221) and, to record any unusual occurrence, form 320T. A report by the colliery mechanical engineer or surface superintendent was required every month on form P225. A report by a competent civil engineer was required by Regulation 9 every six months on form P226 (formerly P222). Regulation 10 required a report by the colliery manager every three months. A tip report by a chartered civil engineer was required by regulation 12 every 2 years, and after a dangerous occurrence, and after any relevant change in the tip's design or composition.

Following closure of a tip the inspection regime changed. Regulation 18 required a report within 10 years, or 2 years if no previous report had been made under the regulations, and after a dangerous occurrence, by a chartered civil engineer. Twelve monthly inspections and reports from a competent civil engineer were required by regulation 17. However, by regulation 18(2)(b) the area civil engineer was to decide whether more frequent inspections were needed and, if so, P226 was the form to be used. The statutory requirement for weekly inspections and reports by the tip foreman came to an end. Responsibility for the precise regime adopted was that of the colliery manager and area civil engineer.

Mr Lewis told me that form 320T (none of which have survived) would but forms P222 and P226 prepared by the engineers (some of which have survived) would not necessarily record an historical burning, provided there was no sign of burning when the inspection took place. It is noteworthy that form P222 asks 'any fire?' and invites specification whether it is old or new, increasing, slight, moderate or large. P226 asks 'any signs of burning?' and invites specification whether the fire is new or worse. These forms are designed, it seems to me, to record contemporaneous rather than historical events unless the historical event continues to exist. It is in my view likely that unless the fire was substantial or was still causing a problem the engineer would make no reference to it in his report. The other contents of form P226, in particular, make plain that its purpose was to identify current causes of concern and measures required to remedy them.

Tip 52 closed in about 1971. On 2 November 1976 Mr Snell prepared a regulation 18 report. It was countersigned by the area civil engineer and area chief engineer. In appendix 12(c) to the report are listed (pre-1971 Rules) completed forms P221 for 21 June 1968 and 14 June 1970, and forms P222 for 9 February 1968, 1 May 1968, 8 October 1970 and 5 April 1971. Seven six monthly completed forms P226 between November 1971 and April 1975 (post-1971 Rules) are listed. None of these reports has survived. Mr Snell did not state in his report either that there had been no incidents of heating or that there was no risk of heating in Tip 52. However, if there had been a known, continuing and significant risk of combustion in Tip 52 reference to it in Mr Snell's regulation 18 report could, in my view, have been expected, and there was none. Risks dealt with in the report concern poor drainage and the risk of slippage. I am driven to infer that although heatings had occurred in this tip during Mr Lewis' time as assistant surveyor they were not regarded by the area engineering team as a continuing problem five years after the tip had closed.

Between 26 October 1972 and 1 April 1981 the assistant civil engineer, Edgar John carried out 6 monthly inspections and reported the results in form P226. Almost all have survived. On 1 October 1979 Mr John reported a single instance of a tip fire. Mr Snell completed a form 320T which has not survived. On 20 November 1981 Alan Webster, Mr John's successor as assistant civil engineer, reported on form P226 a further fire. Both forms P226 featured in cross examination of the tip foremen (paras 60 and 65 below).

I have already stated that I accept John Michael Lewis' evidence that he was aware of heatings and burnings in the tip, one of which, in particular, he drew on his plan having consulted the relevant unusual occurrence entry in form 320T. I have to consider the question whether the absence of records of heatings and burnings in form P226 should cause me to doubt the reliability of the evidence of the tip foremen.

Derek Phillips worked underground at Brynlliw until 1971 when he became tip foreman above ground. He is a friend of Eric Davies and became a witness for the claimants when Mr Davies informed their solicitor of the nature of Mr Phillips' former employment. Mr Phillips made a handwritten statement which formed the basis for a typed summary of evidence attested for the first time when he attended to give evidence. When he began his wo rk as tip foreman there were two tips at Brynlliw, Tip 52 and 750. Tip 52, which Mr Phillips erroneously called Tip 50 in his handwritten statement, was the nearer to Pontardulais Road. It was almost full. Mr Phillips told me that he would carry out inspections of both tips at Brynlliw, mostly on Saturdays, throughout his time as foreman, notwithstanding there was no statutory requirement for weekly inspections to be made of Tip 52. He was paid overtime for the work. He worked to a routine, walking Tip 750 first, and then made his way to Tip 52. Since Tip 750 was active Mr Phillips was required to complete form P224 on which he was to report any adverse conditions including evidence of heatings. Mr Phillips could recall no evidence of heatings in Tip 750. While Mr Phillips did not complete the statutory form for Tip 52 he did report heatings to the mechanical engineer Mr Ceri Morgan. They took place at apparently random intervals. On one occasion he recalled there were two incidents of heating in quick succession followed by a period of 12 months or so with no activity. Mr Morgan would tell Mr Phillips to deal with it, Mr Phillips would take remedial action and he would report back the result to Mr Morgan. When he found an area of heating he would use a temperature probe to check its dimensions from the centre outwards. The probe would be inserted 3ft-4ft into the surface but an extension was available if required. He plotted his readings in his notebook.

Mr Phillips attached to his handwritten statement a sketch plan upon which he marked the position of seven areas of heating. No 1, which he thought was first in time, was at the southern end of the tip. No 2 was in the north east. No 3 was at the northern end. No 4 was in the coal stock pile in the Allt yr Graban Coal yard, physically separated from Tip 52 by Allt yr Graban Road. No 5 was in the central western edge of the tip. No 6 was an area in the central plateau of the tip where Mr Phillips thought attempts to seed with grass had failed. Mr Phillips' recollection is in this respect accurate. York University made a failed attempt to grass the plateau of the tip in 1979. Mr Phillips could not be sure that he had the areas in the correct chronological order but he was firm that he had observed these heatings and reported them. He thought all of them took place before 1979. A seventh area of heating, which Mr Phillips had not numbered, he had marked on his plan in an east facing flank close to the powder magazine. It is noteworthy that Mr Phillips located all the heatings on Tip 52, save No 6, at its edges. Mr Phillips could recall no evidence of an extraneous cause for the heatings he found. As far as he was concerned they were spontaneous.

Mr Phillips recollected that in about 1976 a single borehole was sunk at the central northern end of Tip 52 so he would inspect that borehole having checked those in Tip 750 first. No plan retained by the defendant shows the position of the borehole to which Mr Phillips referred, although it is conceded that while only one borehole was marked on a statutory plan of Tip 750, at least two were present and record cards survive on which the tip foremen have recorded measurements at four boreholes said to have been sunk on Tip 750. It was suggested to Mr Phillips in cross examination that if he was saying he went to Tip 52 to inspect a borehole he was mistaken because none existed. His response was, "I am adamant I went there". He spoke of the sulphurous smell which would be emitted from Tip 52 when the tip was subject to heatings. All of the heatings he observed were accompanied by burned or scorched vegetation.

In his Area Civil Engineer's Classified Closed Tip regulation 18 report on 2 November 1976 Mr Snell stated, "No boreholes have been drilled in tip 52 and no piezometers have been installed within the tip material to monitor water levels. The water level has been assumed from observation". The geological report which formed appendix 12(b), and was dated September 1975, referred to the existence of the 4 boreholes on Tip 850 (750). In its conclusions the geological report points out that its analysis of piezometer results relates to 'borehole positions which are all some distance from the tip under review'. A recommendation was made that Tip 52 should be regularly inspected in accordance with regulation 17. The report neither supports nor contradicts Mr Phillips' evidence as to the presence of a borehole on Tip 52. His recollection that there was no borehole until 1976 is not inconsistent with the report. While I have no independent information whether it came about or how it may have come about, it is possible that a decision was made to sink a borehole in Tip 52 in order to check the water table at the northern end. At paragraph 8 of the regulation 18 report it was stated, "Only on the short north face of the tip is there any real danger to life, a failure in this area could reach the Allt yr Graban Road". What the report does do is support Mr Phillips' evidence that he inspected Tip 52 since it refers, in relation to the period up to September 1975, to the fact the water level there had been visually observed rather then measured.

On 1 October 1979 Mr Edgar John, assistant civil engineer, prepared a form P226 statutory closed tip report in which he made reference to a fire on Tip 52: "On 1/8/79 Flames and smoke reported from the flank facing the Pontardulais Grovesend Road. A length of peat & vegetation affected-70m. Fire Brigade called. Dragline excavated burning material and spread to cool. All visible combustion extinguished by 7am 3/8/79. Entry in 320T by R Snell Esq., District Civil Engineer." Here is an occasion when, contrary to Mr Lewis' recollection, a reference was made to an historical fire, from form 320T, during a six monthly inspection by an engineer. Mr Phillips' attention was drawn to a memorandum (4/18/72) written on 6 November 1979 by Mr Snell to the Production Manager, Mr HN Roberts. In it, Mr Snell referred to a conversation with Mr Roberts on 24 August about the fire in Tip 52. In his concluding paragraph Mr Snell mentioned "the achievement of Mr D Phillips, the Surface General Foreman and his team, whose untiring efforts during those three consecutive shifts, prevented the fire spreading through the bund, and into the filter cake area, thus averting a far more serious incident." On 13 November 1979 Mr Snell added a handwritten postscript to the foot of his memorandum for the information of a Mr Davies in answer to his question- has the cause of the fire been established? : "Yes. 9- 12 year old children had been throwing Molotov cocktails at the tip slope. This action ignited the vegetation, then the peat top soiling, and finally the shale. The children were apprehended by the local police."

It was suggested to Mr Phillips that the fire referred to in the statutory report was the fire he had marked on the southern facing slope of the tip as No 1. Mr Phillips said he thought it was not. He went back underground during the autumn of 1979 and thought the fire to which he had been referring was older than that. Mr Phillips' last statutory report for Tip 750 was dated 27 October 1979 before his place as surface foreman was taken by Eiros Wyn Vaughan whose first report was 4 November 1979. In the summary of his evidence to which he attested at the hearing Mr Phillips said he was aware of the fire because he was heavily involved in its treatment. When asked about it in evidence, however, he had no specific recollection of it and although he was aware of the allegation that children had thrown Molotov cocktails he had no knowledge whether that was so.

Mr Phillips' fire No 1 was not in the western edge of the tip facing Pontardulais Road but in the southern edge facing Station Road. The fire he marked in the western edge was his No 5. In closing argument Mr Paul Darling QC, on behalf of the defendant, acknowledged that on reflection his suggestion to Mr Phillips that fire No 1 occurred in 1979 may have been erroneous. It was more likely, if there was any connection to be made between Mr Phillips' plan and the 1979 fire, that it was No 5 to which the statutory report was referring. Mr Phillips' recollection was that fire No 1 was the only fire which required the hiring in of machinery to extinguish it. Fire Nos 1, 2, 3 and 5 were dealt with by dragging up the sides of the tip up with a dragline to remove the heating and allow it to cool. A bulldozer was then used to alter the angle of slope. "This is how I was taught", he said. In the case of fire No 2 he recalled the top was taken off by the bulldozer and spread on the adjoining Tip 750 to cool. Then they brought the material back and spread it on Tip 52. They adopted the same procedure with fire No 3 but did not remove the materia l to Tip 750; they spread it alongside on Tip 52 and then put it back when it had cooled. Fire No 5 in the western edge was dug out and the material taken to Tip 750 where it was spread and left. Mr Phillips' team had to go back two or three times because they thought they had extinguished the fire but found that they had not.

Eiros Wyn Vaughan was a surface worker at Brynlliw from about 1976. From time to time he would accompany Mr Phillips in his work during the week but not at weekends when Mr Phillips was making his weekly inspections. Mr Vaughan was promoted to surface foreman in 1979, a post which he retained until he left Brynlliw to work at Cynheidre in 1982. He too carried out tip inspections at weekends and worked to a routine, first walking to Tip 52 and then to Tip 750. He would test with a probe for heatings on Tip 750 but not on Tip 52. His recollection was that there was in his time one water borehole on Tip 750 and another on Tip 52 at which he was required to take measurements (in the same general position as that described by Mr Phillips), although he conceded that with the passage of time he might be wrong; both boreholes might have been on Tip 750.

Mr Vaughan said he carried out weekly inspections of both tips without fail. Occasionally he would be accompanied by the engineer, Mr Herdman, or his deputy, David Elwyn Davies. His reports would be completed at the engineers' offices and countersigned by the available engineer. Mr Vaughan recalled, when he made his witness statement, 'several incidents of heatings' at Tip 52. He had experienced vapour and smoking of the tip on some occasions and the burning of tip material on others. He would arrange for machinery to dig out the suspect material to cool. Once cool the red ash or shale remaining would be placed back into the tip. Sometimes it would be necessary to use a dragline with a front bucket to dig out the heated area and he might be engaged for a whole weekend dealing with it. On no occasion did Mr Vaughan come across a fire which had been ignited on the surface. He had no doubt that what he saw were instances of spontaneous combustion.

In cross examination Mr Vaughan recalled what he called the big fire in Tip 52 referred to by Alan Webster in the P226 report dated 20 November 1981. Mr Webster reported: "Southern slope caught fire during the summer (boys lighting a small bonfire). Turf and trees caught. The fire was confined to the surface (40m x 40m). Fresh grass growing back on the burnt area." Mr Vaughan remembered that the fire started on a Saturday evening. He was called out on the Sunday morning. The fire service was present. He deployed machinery to excavate to about 6ft and spread the burning area for cooling. Fire hoses were used to dowse the spread material. The operation took 3 days to complete. This fire was not, according to Mr Vaughan, on the southern slope but on the western edge within the southern half of the tip.

Mr Vaughan could give no particulars of other instances of fire or heatings on Tip 52. He said that they were all less serious than the 1981 fire. In re-examination Mr Vaughan departed from his previous evidence to say that the temperature raising took place on Tip 750. He could not remember any heatings which were not on Tip 750 other than that which led to the fire in 1981. It was clear that Mr Vaughan had become confused. There is no other evidence of heatings on Tip 750 and it is not the defendant's case that, in this respect, Mr Vaughan has confused one tip with the other. I pointed out the inconsistency in the evidence to Mr Vaughan and asked him to concentrate on the Tip 52 where the 1981 fire had occurred. Were there any other heatings on Tip 52 he could recall and if so roughly how many? Putting on one side the significant fire in 1981 he thought he had come across and had to deal with between one and five heatings on Tip 52 during his time as foreman. He could recall digging out on Tip 52 once or twice. It was a long time ago and Mr Vaughan conceded that his memory was suspect.

David Elwyn Davies was the deputy unit mechanical engineer at Brynlliw from January 1975 until the colliery closed in April 1983. He began his career as a fitter with NCB at Nantgarw in 1951 rising to deputy mechanical engineer at Lady Windsor in 1965. He was employed by the Central Electricity Generating Board at Aberthaw Power Station between 1967 and 1971, managing coal stocks, and returned to the coal industry in 1975. Mr Davies acquired knowledge of the behaviour of coal stocks at Aberthaw because part of his responsibility was to investigate heatings. He attended several seminars and discussion groups conducted within the electricity industry and took part in research at Aberthaw. Before he rejoined NCB in 1975 Mr Davies had no real appreciation of the risk of spontaneous combustion in spoil tips as opposed to coal stocks. He had, however, come across burned or brown vegetation and steaming of waste if wet. Mr Davies' experience in the coal industry before his appointment to Brynlliw was confined to the eastern sector of the South Wales coalfield.

When he started at Brynlliw Mr Davies was made aware that Tip 52 had been closed since about 1971. He noted the steep slopes of the tip, particularly on the southern flank facing Station Road, Grovesend.

Mr Davies said in his witness statement that he recalled both Derek Phillips and Eiros Wyn Vaughan performing their statutory responsibilities as tip foremen. Although Tip 52 was a closed tip their inspections included Tip 52. They would complete written reports and submit them for counter signature by the mechanical engineer or his deputy (Mr Davies himself) and then to the colliery manager or his deputed official. He could recall countersigning reports on Tip 52 which disclosed evidence of heating. The incidents were reported to the area civil engineer whose office was at Pontardulais. They were heatings within the tip accompanied by no apparent external explanation for ignition. The burning material would be removed and extinguished, a process which could take the team days if not weeks. Mr Davies expressed the opinion that there was no reason to suppose that heatings would cease merely because the tip was closed. So long as the tip continued to be regularly inspected and monitored any heatings would be detected early and remain small. The longer a heating was allowed to burn without being detected the more difficult it would be to extinguish.

It was pointed out to Mr Davies in cross examination that the statutory regime did not require weekly inspection and report of Tip 52. Having considered the regulations he agreed. He conceded he might be wrong about the completion of P224 forms for Tip 52 after its closure. He did recall countersigning reports and was not surprised if they had gone missing. He agreed that he did not distinguish between the two tips by number. He recalled the routine was that he and Ceri Morgan would work alternate weekends and whichever of them was in work at the weekend would tend to countersign the forms. The tip foreman would complete the P224s in the vestibule to the offices where the electrical, mechanical and planned maintenance engineers worked. The books when completed were put into storage somewhere on the colliery premises. Heatings on Tip 52 would be included in the reports, identified by description in the form. Mr Davies was saying that even if there was no P224 specifically for Tip 52, he thought a heating on Tip 52 would be reported by description in the form for Tip 750. He too recalled the existence of the borehole to the west side of the railway line which bisected the spoil heaps. I asked Mr Davies when was the first time he realised there was no statutory requirement for weekly P224 reports on the closed Tip 52. He replied, "Today". In view of the contrary recollections of both Mr Phillips and Mr Vaughan I cannot accept that heatings in Tip 52 were recorded on form P224.

Eric Davies it will be recalled was a union official at Brynlliw throughout the 1970s until the colliery closed in 1983. He made two witness statements, the first dealing with his knowledge of Tip 52 and the second with his knowledge of coalfield fires generally in the area. In his first witness statement Mr Davies said that he was aware of the propensity of Tip 52 to spontaneous combustion. In evidence he explained that he was a friend of Derek Phillips. They had a standing joke between them that Mr Phillips only inspected the tips on the weekend to attract the overtime pay. In his turn Mr Davies was teased by his colleagues that he was working full time above ground while he was nominally a face worker. As was the convention in the mining industry the lodge secretary was given generous 'facility time' and in Mr Davies' case that was virtually full time. He was aware contemporaneously of heatings on Tip 52 from Mr Phillips. In evidence Mr Davies said he had seen for himself the heatings marked in Mr Phillips' plan at Nos 1 and 5. He was aware of the heating in the coal stock marked as No 4. He thought fire No 1 took place in about 1976. At the time he put it down to spontaneous combustion. It was the only occasion he could recall seeing a fire engine in attendance. However, there may well have been others which he had forgotten because he could recall the fact of other fires but not their circumstances.

In cross examination Mr Davies was asked about the 1979 fire on the western side of the tip, possibly Mr Phillips' fire No 5. He said that he would not have been involved in the discussions recorded in Mr Snell's memorandum and could not make the link with fire No 5 as he was invited. He agreed that the fire he recalled at the No 5 position occurred after 1976. He did not recall the reference to Molotov cocktails but he would not be surprised if blame was placed on an external source. He had made known to management his opinion that the cause of fire No 1 was spontaneous combustion but they had expressed a contrary view.

It was suggested to Mr Davies that the 1981 fire referred to in Mr Webster's P226 may have been Mr Phillips' fire No 1. Mr Davies said he was not aware of a fire caused by small boys lighting a bonfire in 1981 specifically. It could well have been fire No 1. However, Mr Davies' concession pre-supposes that the 1981 fire occurred, as P226 records, at the southern edge of the tip. Mr Vaughan said it did not and he was the tip foreman called out to deal with it. While his recollection of other details was modest Mr Vaughan had a lively recollection of the 1981 fire. He had reason.

Mr Davies said that the colliery's priority while he was secretary was its profitability. Brynlliw was under the constant threat of closure. This may explain the absence of documentary evidence of heatings. He had thought he would find reference to them in minutes of colliery safety meetings but the records had not survived. He was, however, surprised that the closed tip statutory reports which had survived did not make reference to heatings which had taken place. Mr Davies was satisfied that the tip foremen did make inspections of Tip 52 and did report heatings.

When the 1996 fire broke out Mr Davies was the constituency agent for the local Member of Parliament, Mr Martin Caton. A meeting between Coal Authority representatives and Brynlliw employees was called on 23 November 1998 at Mr Caton's invitation to impress upon the Authority that the fire was its responsibility despite its return of the land to the commoners. Notes of the Meeting were kept (5/24) by the Licensing Manager, Mr Wilson, from which it can be seen that the following contributions were made, amongst others: (1) David Elwyn Davies, mechanical engineer, said that in 1975 when he went to the tip a heating had occurred. (2) Derek Phillips said that heatings had regularly taken place since 1970. He asked what had become of the weekly and monthly reports. He was told by David Elwyn Davies that the reports for 1971 to 1981 were missing. (3) Mr Schofield, for the Coal Authority, said that those records which had survived did not support the case for spontaneous combustion. They referred only to fires caused by vandals in 1979 and 1981. At the time of the sale to the commoners the Authorit y had the regulation 18 report for 1976 and it made no reference to burning. He agreed to research the records further. (4) Eric Davies said that there had never been a tip for the Swansea 5ft and 6ft seams which did not have a fire. It could be established that Tip 52 had spontaneously combusted. He could get half a dozen witnesses to establish the details.

I derive from the evidence of this meeting a consistency of account by Eric Davies and Derek Phillips, supported by David Elwyn Davies, the mechanical engineer.

Eric Davies gave evidence of other fires in tips in the area, to which I have already referred. In his second witness statement he said that he knew of a fire which had occurred underground at Brynlliw in the junction of the T6/T7 faces which were mining dry steam coal in the Swansea 3ft seam. Mr Davies amplified his account during cross examination. The fire took place in the waste (or 'gob') underground. He knew about it because he led the second mines and rescue team which went underground to switch the airway pumps back on. It occurred in 1976 or 1977. They had to re-start the turbines to power the pumps. He had particular reason to remember the incident because the following day he went to Switzerland for a fortnight. When he returned he went back underground to inspect the area. He found that the fire had started in the gob and spread to the arches in the heading. It looked like a fire in a grate and he concluded it was spontaneous combustion. When he returned above ground he spoke to the colliery manager and they agreed the cause. This was an important event. Suspicion of an external cause for ignition underground would have led to an investigation and disciplinary action if the culprit was identified. The roadway was reinforced and the area regularly monitored. There would undoubtedly have been a record made at the rescue station and by the engineers who would have reported further to the area and divisional people and the Mines and Quarries Inspectorate. No records have been located.

When he gave evidence Dr Edwards told me that he had examined the abandonment plans for the T6/T7 district having heard Mr Davies' evidence and believed he could explain why spontaneous combustion had occurred. This was a zone with severe roof problems. The team came to the finishing line of the coal face. They paused and resumed mining coal further along the seam leaving coal in the roof behind them. This, in a material respect, replicated the circumstances of the fire which broke out at Morlais colliery in 1968. The geological feature created by the exigencies of the mining effort helped create the conditions which led to the combustion. This was a classic example which illustrated Dr Willett's warning when summing up at the symposium in 1970, "Don't just think of rank, think of geology as well".

I accept the thrust of the evidence that Tip 52 had, during the period 1970 to 1981 a history of heatings within it. Some of them at least led to burning which Mr Phillips and Mr Vaughan were required to dig out. Mr Phillips' fires Nos 1, 2 and 3 certainly fall into this category. I am satisfied that, additionally, there were, in 1979, during Mr Phillips' time as tip foreman, and in 1981 during Mr Vaughan's time as tip foreman, fires on the western slope of the tip. It is not possible at this distance of time to be sure whether they were in fact started by trespassers. There is no satisfactory evidence that the deep seated burning was caused by surface fires, which, it is agreed, would have taken time to ignite the coal waste and go underground, despite the contemporaneous references in the documents to the arrest of youngsters and to the presence of a bonfire. The 1996 fire is an example of a fire which took hold, and attracted children who then lit a further fire (para 94 below). It is at least possible, as Dr Richards has supposed, that the Molotov cocktails were thrown at a tip which was already burning and that a bonfire was lit near a fire which had already taken hold in the tip On the other hand, it is not possible to prove correct the opinion of the tip foremen whose job it was to assess and deal with the fires when they were found, and there is no reason to think that the contemporaneous records of those events and the opinions expressed were not carefully considered. On balance, I must conclude that the contemporaneous opinion revealed by the records is to be preferred. The 1979 and 1981 fires were either caused by trespassers or were believed on good evidence to have been so caused. Nevertheless, quite apart from the 1979 and 1981 fires I am satisfied that there was a sufficient body of evidence, concerning this and other mines and tips in the area, known to management to demonstrate, when the colliery closed, that this was a tip liable to heat and to catch fire spontaneously. There were several incidents noted by the tip foremen and reported orally to the engineers and surveyors which were undeniably deep seated on discovery, not attributed to a surface ignition, and required digging out. As at 1983 high rank of coal did not give comfort in the case of Tip 52.

I found Mr Phillips to be an impressive witness. He did not exaggerate; he was thoughtful; he was careful; he had been consis tent; and the circumstantial detail of some passages of his evidence was compelling. Notwithstanding that 34 years has passed since the relevant events began, Mr Phillips had reason to recall significant events in his working life and, within the natural limits of human recollection, it is my view that he succeeded. Mr Vaughan was a less able witness, simply because of the passage of time and the importance of the occasion of giving evidence. I accept the thrust of Mr Vaughan's evidence that there was at least one heating, on Tip 52 during his time as foreman caused by spontaneous combustion. I accept that these heatings were known to the engineering staff and surveyors at Brynlliw and to the NUM secretary, Mr Eric Davies. I accept the evidence of Mr Davies that there was a small fire underground in about 1976/77 which, he subsequently agreed with the colliery manager, was the result of spontaneous combustion. I accept Mr Lewis' evidence that some of the forms 320T which he had inspected did indeed make reference to heatings. I found the points at which these witnesses' recollections met rendered their evidence mutually supportive. Where there were inconsistencies between the witnesses, for example and principally, in David Elwyn Davies' recollection of weekly statutory reports for Tip 52 following closure, the explanation for them is satisfactory and they do not undermine their combined recollection of reports of heating to the engineers. The full picture provided by the evidence of these witnesses is compelling. I detected no sign that any of them had a wish to represent the historical picture in anything other than honest and straightforward terms. So much was conceded by the defendants.

Mr Darling concentrated in his submissions to me not upon the credibility of the evidence but upon the inferences legitimately to be drawn from it. Dr Edwards agreed in cross examination that he had not been surprised to hear witnesses describe the presence of heatings and fires on tips. However, he cautioned against treating them as a consequence of spontaneous combustion. A heating, in the case of a working colliery, might be caused by various things, such as local colliery fires, boiler ash, or trespassers. As I have observed, trespassers are a probable exp lanation for the fires of 1979 and 1981, but I do not see how colliery fires or boiler ash could explain burning at the edges, distant from the colliery buildings, of a tip which had been closed and on which tipping no longer took place. Mr John Hand QC, for the claimants, reminded Dr Edwards of the presence of heating in the coal stocks at Allt yr Graban. Dr Edwards responded that all coal stocks are liable to variations in temperature without constituting a heating. I am satisfied that the heating in the coal stocks observed by Mr Phillips was not an inconsequential event in the life of the colliery. Mr Phillips was responsible for its management and had every reason to remember its significance. This was, in my view, a further demonstration, if one was needed, that it was not possible to dismiss burnings of the high rank coal within Tip 52 as the work of trespassers.

In my view, the existence of a risk of tip combustion was well known to NCB management in South Wales, notwithstanding the high ranks of coal being mined. While the first regulation 12 report for Tip 750, dated 28 June 1972, (3/31/275) stated 'No danger of fire is anticipated on this tip', there was added to the third report dated 14 July 1978 (4/11/17) an additional appendix which contained the following warning:

Mr Bacon notes in his report that this addition "suggests that there had been a problem of fires at a colliery or on a tip, not necessarily at Brynlliw, which had been caused by the uncontrolled burning of rubbish". It is true that the appendix continues, at (2) and (3), to require the burning of rubbish in a safe place away from the tip and that braziers and other burning should be properly controlled and supervised. However, instruction (1) is quite specific in its reference to the avoidance of spontaneous combustion by the inclusion of combustible material in the tip. Combustible material other than coal had indeed been disposed of in Tip 52, as Dr Blandford's excavations demonstrated (para 98 below). In my view, the only sensible explanation for the inclusion of paragraph 1 in this appendix to the reports for Tip 750 can have been its relevance in warning of a known risk of spontaneous combustion in tips in this area of the coalfield.

Those 6 monthly and annual reports of Tip 52 prepared by surveying and engineering staff which have survived make no reference to any problems of combustion save those in 1979 and 1981 which I have extracted earlier in this judgment. Otherwise, any comments concerned problems of drainage and spring issues at the flanks from time to time. The following extracts should be noted:

In each of the foregoing reports the maker has placed a cross in the 'No' box against the question, 'Any signs of burning?'. However, several references were made to the drainage problem at the foot of the southern and western flanks and erosion of the western flank was specifically referred to in 1985. No work was done to the western flank in 1987 and any process of erosion could only have been arrested by natural ground cover. It is apparent that the principal concern of the engineers was stability. As will appear later, continuing erosion of the western flank was relevant not only to the stability of the tip but also to the risk that combustible material would become exposed.

On 5 September 1983 the Brynlliw colliery and washery ceased production. The land, including Tip 52 and Tip 850, would eventually be returned to the Commoners and, under a deed of release of 1965, NCB was required to "restore the land, reseed it and remove from it all buildings, structures, engines, fences or other works erected or placed thereon". On 24 November 1983 NCB met with West Glamorgan County Council and Lliw Valley Borough Council to discuss arrangements for restoration. NCB made it clear that it did not propose to alter the contours of the western flank unless a grant was available since it was already in an advanced state of vegetation. The District Civil Engineer, Mr Snell, was present at the meeting. There was no recorded discussion of any fire risk. On 5 June 1984 a further meeting took place with West Glamorgan County Council. NCB proposed making no changes to the outward facing slopes. The remainder would be graded to form a shallow valley through the centre of the colliery surface. The Commoners Association had requested that there should no tree planting on the top surface of the tips. They wished it to be restored to grazing land. Pessimism was expressed for the prospect of a grant from Welsh Development Agency for an improved scheme. West Glamorgan requested NCB by letter of 30 October 1984 to consider a more extensive tree planting programme and proposals for protection and aftercare of the restored area. Mr Snell advised Mr Leighfield, the Area Surveyor, that the proposal was expensive and against the wishes of the commoners. At a meeting held on 29 November West Glamorgan accepted Mr Snell's advice together with his proposal to screen the tip from Station Road by planting trees. NCB would dispose of the site to the commoners subject to their acceptance of an aftercare agreement. The regime contemplated by such an agreement and the reasons for it are the not the subject of any contemporaneous record. Meanwhile, NCB was disposing of the coal stocks at Brynlliw. Regrading of the tips could not commence until that process was completed.

On 7 January 1987 Mr K Lisk, Area Civil Engineer, proposed to Mr Snell and members of the Area Tip Committee that partial restoration could be achieved by removing the top 3 metres of spoil from the crest of Tip 52. That material would be used to cover the colliery yard and stores area by spreading it eastwards. The cost would be about £180,000. Restoration of Tip 850 (750) would follow.

The tender conditions for the contract to perform the work imposed upon the contractor an obligation "to ensure that if heating is detected of any material forming part of site the fact is reported immediately to the Colliery manager, of which the site forms part and to the Engineer".

The work on Tip 52 was completed by February 1988. Restoration of Tip 850 (750) took place after the coal stocks were lifted commencing April 1988. It is not suggested that any report of heating or burning was made by the contractor to NCB or by any engineer whose job it was to inspect the tip. Since the work did not involve disturbance of the bulk of the west flank it is not likely that any heating which may have been present was not observed. However, I shall proceed on the assumption that there was at that time none to be seen.

Attached to Dr Blandford's report is an aerial photograph taken by a company called Geonex on 9 June 1992. Dr Blandford commented in his report, "[the photograph] shows the site to be covered with grass and the trees along its western edge [appear] to be maturing. There is no evidence in the photograph of any combustion within the spoil heap". The height from which the photograph was taken does not permit close analysis of the condition of the flanks but the general impression is one of a grassed plateau with wooded flanks.

Royston James Davies has lived all his life and still lives with his family at Myrtle Grove Cottage, adjacent to the south western flank of Tip 52. Mr Davies routinely walked his dog over the flanks and plateau of the tip. During the summer of 1996 he noticed that his dog was retrieving an increased number of animals. That may be a recollection made with the assistance of hindsight. However, for a period of 1-2 months before the fire was reported to the fire service he noticed that there was a smell of sulphur on the tip. It was eventually sufficiently disturbing for his wife to report the smell to their general practitioner, Dr Brown. Trees began to fall apparently as a result of root burning rather than a surface or grass fire. Mr Davies found dead birds. He became convinced that the tip was on fire when at night he saw a blue flame characteristic of a gas cooker. The fire service was called. The identity of the caller is not known. Mr Davies had recently purchased a camcorder and used it to record the flank of the tip as it was when the fire service was first in attendance. He thought that was at the beginning of August.

On one occasion, which must have been after the tip fire had taken a strong hold, Mr Davies was walking on the top of the tip close to its flank when his dog appeared to be pawing at the ground and burned its paws. Mr Davies poked his walking stick into the surface and dislodged a clod of earth which appeared to be red hot.

Mark Lowis James was, at the time, Station Officer at Gorseinon Fire Station. He is now Assistant Divisional Officer and Community Fire Safety Officer. Mr James recalled that the service was first called to a grass fire. The first officers to arrive could see they were not dealing with just a grass fire and Mr James was asked to attend within ½ hr-1hr of the service's arrival. Although Mr Davies thought that he filmed the fire service on the day of its arrival, having viewed the film, Mr James thought that it was the second day. Mr James pointed out to me signs that the fire being investigated was a tip fire and not a surface or grass fire. What was to be seen were birch trees which had toppled as a result of the fire burning their roots from beneath the surface. The trunks of the trees at their base were hardly affected, and the upper trunks and branches, not at all.

Mr James was asked to consider a file note made by an environmental officer employed by West Glamorgan County Council on 26 September 1996. Mr Emyr Evans wanted to know what the fire service was doing with regard to the fire. An operations officer at Morriston told him that it was a deep seated fire which the fire service was treating as low priority. They had been dowsing the fire daily with water and were monitoring it. There was no present intention to dig the fire out. Mr Evans walked the area and found that the fire had spread 25 yards or so in 10 days. On 27 September he spoke to a Station Officer Morgan who said that the situation was aggravated by children who had lit a surface fire earlier in September. (This is the occasion to which I referred at para 79 when, but for the interval of time since the fire was first reported, it might have been thought that children started it). The fire service was currently considering digging out the fire because it did not seem to be deep seated. Mr James knew nothing about these conversations. He said he did not know of a Station Officer Morgan at that time although there was a Divisional Officer Morgan. He said that following his inspection of the fire, he did not understand how anyone could have expressed the view that the fire was not deep seated. He had seen a deep fissure in the ground in which he could see burning underground. Water was used to dowse the fire but that was largely a cosmetic exercise to reassure the public that something was being done. He knew that water was unlikely to be a solution.

Michael Peacey, a fire officer also stationed at Gorseinon, confirmed that he was called to what was reported as a grass fire. It was plain on arrival that it was not. After several days an attempt was made to dig out the fire with a mechanical digger and to extinguish the burning area with water. It did not work. Having been engaged in the digging operation Mr Peacey formed the view that the core of the fire was 15ft-20ft beneath the surface. In common with Mr James, Mr Peacey did not think this fire originated on the surface. He had on earlier occasions been called to grass fires on the common land in the area, and adjacent to Station Road, but this was not a similar event.

Mr Royston Davies had further important evidence to give. His home was 50 yards away from the location on the south western flank where he had first observed signs of fire and he walked regularly in the area. He said he not seen any surface fire in the area that summer. No other witness was called to speak of any surface fire taking place in this area during 1996.

Huw Jones, said that when travelling Pontardulais Road in 1995 or 1996 he saw tree felling operations on the west side of the tip and fires being lit in the vicinity of the tip. In his witness statement he made a temporal connection between his observation and the subsequent tip fire. It became apparent during cross examination, however, that the activity Mr Jones had seen was nowhere near the outbreak of fire in 1996. The fires he saw comprised felled willow and silver birch trees in piles on flat land on the road side of the fence which ran along the foot of the western slope further to the north. He saw no fire on the flank itself. He acknowledged he could not have seen the area where the fire started near Myrtle Cottage from his car travelling along Pontardulais Road. It is clear that Mr Jones' observation does not assist a judgment what was the cause of the 1996 fire. His fires were not a possible source of ignition of the tip fire.

Expert opinion on the cause of the fire was divided. Dr Blandford described the result of his trial pits dug to a depth of between 5.3m and 6m below the surface of the tip. The trial pits were dug 11 years after the height of the tip had been lowered during the 1987 reclamation. All of them were within a short distance of the western flank but none was in the western flank. In each he found loose, black and grey colliery spoil and frequently yellow stained gravel of weathered mudstone with coal. He found in addition cobbles of mudstone and ironstone, pit props, rubber belting and sections of steel framework. All the pits were dry but the spoil material was sufficiently loose that rainwater would pass freely through it.

Thermographic surveys were carried out on 26 November 1998 and 17 May 1999. Ambient temperature was 8ºC. Temperatures recorded, during probes to a depth of 200mm along the western edge of the tip, exceeded 250ºC in places. The fire was found to be present in the western face over a surface area 400m long south to north, and it extended into the tip by up to 40m. When the exercise was repeated in 1999 Dr Blandford discovered that the fire had progressed a further 20m into the tip at its southern end and 35m into the tip at its northern end. The fire had progressed northwards along the length of the tip a further 200m.

In reaching his opinion that the cause of the fire was, on a balance of probability, spontaneous combustion Dr Blandford relied on the following:

While Dr Blandford's view is formed upon a consideration of several relevant features of the tip, its history and the circumstances, I am bound to treat his opinion with circumspection in consequence of factor (iv). I consider that Dr Blandford's preparedness to express himself so forthrightly (and wrongly) about the known combustible qualities of rank 102/201 coal generally taints his opinion as to cause. Dr Blandford told me that he did not intend to express himself as he did. He intended only to say that high rank coal was known to have been subject to spontaneous combustion in the past. If that is so I am surprised that Dr Blandford expressed himself as he did, since the two positions are wide apart. I will rely upon Dr Blandford's evidence of his trial pit results and thermographic survey but not upon his conclusion as to cause.

Dr Richards reached the same view, having adopted a more sceptical approach to the combustible qualities of coal in the tip. In Dr Richards' opinion, "the empirical evidence displaces the scientific view" for reasons which I have already identified. He would have taken into account the history of the seams and the area before reaching a view as to risk. Apart from Graig Merthyr there was an arc of collieries around Brynlliw which had suspected cases of spontaneous combustion in their tips. Dr Richards' impression was that post event rationalisation had produced theories of third party causes for ignition when experience demonstrated that the coalfield's tips presented an inherent problem. The known propensity of coal stocks at Aberthaw and Allt yr Graban to overheat was enough to displace the scientific orthodoxy.

Dr Richards accepted in cross examination that he knew of no decision to review existing tips in South Wales in consequence of a known or suspected risk of spontaneous combustion. It was suggested to him that the fact a tip was not compacted would not help assess the risk of spontaneous combustion, only the risk of spreading if a fire was ignited. Dr Richards responded that there had never been a fire in a properly compacted tip save where it had been laid against a non-compacted tip in a vulnerable area. If someone had considered in 1987 the features referred to by Dr Richards in his report, including rank and condition of the coal waste, the presence of other combustible material, history and experience, loose uncompacted waste tipped over the flanks, that person must have concluded there was a risk of spontaneous combustion. In common with Mr Bacon and Dr Edwards, Dr Richards did not accept that tree roots would make a significant contribution and I prefer their opinion. Asked why this could not have been a fire started by a bonfire on the surface, Dr Richards said that the bonfire would have to be a substantial long-standing fire. In reexamination he expressed the view that an ordinary grass fire would not be enough. It would be of insufficient intensity and duration. He could not see in the video any evidence of a surface bonfire. In his view, this was a fire which started beneath the surface and fed itself to air at the surface. At that point the fire became manifest.

Dr Richards was asked in cross examination why there was no report of red shales on the tip if it was right that there was a history of fire. Tip shale turns red when oxidised in heat. Dr Richards replied he could not answer for Dr Blandford's observations but he had been to Tip 52 and had seen red shale dotted all over it. I have noted that Dr Blandford made no mention of red shale in his report upon the contents of his trial pits, but the trial pits were set back from the edge of the western face and there were five in number over a distance of about 500 metres. By the time Dr Blandford had dug out his trial pits the top surface of the west side of the tip had been lowered by 3 metres during the restoration in 1987. Furthermore, Dr Blandford was never asked whether he saw or recovered any red shale in 1998/1999. In any event, I consider the point of limited assistance to the defendant since it is the defendant's own case that there were two substantial fires on the southern and western flanks of the tips. If that is right there must have been red shale present in the flanks. Dr Richards was asked whether the interval of time between 1983 and 1996 would not have caused weathering of the shale and, accordingly, have reduced even further the likelihood of spontaneous combustion. He replied that new areas for oxidation can be produced by drainage gulleys, and he saw in the video film taken by Mr Royston Davies what looked to him to be just such a new gulley. It looked as though fresh movement had occurred with the effect of exposing fresh coal. Dr Richards had in his supplementary report analysed the contents of the video film taken by Mr Davies on his camcorder. Dr Richards described a gulley "to the left hand side of the burning area...The drainage channel indicates that the gulley has been present for some time." In the third segment of the film Dr Richards identified a close-up of the same gulley. I have viewed the gulley for myself in the video film. It is plainly a natural channel formed by water draining off the tip from top to bottom of the flank. In its centre is a narrower and deeper split in the surface of the flank, apparently caused by a concentrated flow of water such as one would see caused by a fast flowing stream. It seemed to Dr Richards that the gulley had deepened over time as a consequence of scouring by surface water. Dr Richards' observation is consistent with the report of weather erosion on the western flank made by DJ May in his P226 on 15 April 1985. Water run off would explain the frequent reference to ponding at the foot of the western flank. Dr Richards saw that the fire had started to burn the vegetation and tree roots and trunks in a discrete area immediately adjacent to and to the right of the gulley. The vegetation on either side of the sub-surface fire was intact. In other words the extent of the surface burn was limited by the sub-surface fire. The effect of water directed into the burning shale was to produce clouds of steam. The ground beneath the surface is plainly extremely hot. All this is to be observed in Mr Davies' video film.

Asked whether the statutory records appeared to document overall a satisfactory history of the life of the tip, Dr Richards replied that they did not. This was an inactive tip in respect of which appropriate records of heatings observed by the tip foremen were not kept. Every time a heating occurred there should, in Dr Richards' view, have been at least a form 320T completed, as a record of an unusua l occurrence, but the only 320T book for Tip 52 available, for 1978, was simply marked up that the tip was closed. In re-examination Dr Richards explained that the reason why the forms 320T should have been completed was that they were required as a record to ensure the correct aftercare of the tip.

Mr Bacon's opinion is that the more likely explanation for ignition of the fire in 1996 is an external source such as an inadvertent or deliberate surface fire. In his view, the source of heat could have been comparatively small. In his report, this view was reached without an analysis of the evidence of Mr Royston Davies or Mr James, and upon the impression that employees at the colliery may have been mistaken about the presence of heatings on Tip 52 in the period 1970-1982. He was not, therefore, satisfied of the presence of empirical evidence sufficient to displace the scientific presumption that the risk of spontaneous combustion was very low. In these circumstances, it is hardly surprising that Mr Bacon reached the conclusion he did. In cross examination Mr Bacon conceded that a surface fire sufficient to ignite the tip would have to be big enough to be seen in the vicinity in which the fire took hold. While the incubation period could be very variable he would estimate that it could have been about a week before the scene was filmed by Mr Davies. That, of course, would not explain Mr Davies' experience of the tip for a month or two before the fire became patent. Mr Bacon was to driven to assert that a surface fire may have taken place about 50 yards from Mr Davies' home in an area in which he regularly walked but that if he was not observing the flank 24 hrs a day Mr Davies could have missed it. Before trial Mr Bacon had made the following agreement with Dr Richards: "[A] fire ignited by an external source might go undetected for an extended period during an incubation period. However, in locations which were vegetated, fairly well- trafficked and adjacent to housing, it might be expected that a fire which was of sustainable size would be noticed relatively quickly..."

Dr Edwards reached the same view. He too discounts the employee evidence but not because they may have mistaken the location of heatings, but because the heatings do not indicate a propensity to ignition. He accepted that the tip might heat but expressed the view that does not alone demonstrate a likelihood of combustion. In his report Dr Edwards wrote, "on the balance of probabilities, having regard to the type of coal discard in the tip, the long period of dormancy of the tip and the photographic evidence of mature, healthy trees growing on the tip in the area in which the fire started, combined with the absence of documentary evidence, spontaneous combustion is not a likely explanation for the fire". Dr Edwards' opinion thus expressed needs to be qualified by the terms of the agreement subsequently reached with Dr Richards. The loosely compacted slope in a steep west facing flank increased the risk of fire whether spontaneous or ignited externally.

In cross examination Dr Edwards acknowledged the evidence of heating in the tip over time, the occurrence of two fires underground, the steep slope of the flank, prevailing winds from the west partly protected by tree cover, and the absence of evidence of a surface fire. He said a conclusion depended upon how the various factors were 'put together'. In his view, the fact that the tip had been inactive for 20 years or so before 1996 was influential. It was too long a period for spontaneous combustion to appear.

In his closing argument Mr Darling submitted that it was not possible to conclude on a balance of probability that the fire of 1996 was caused by spontaneous combustion. At best there was a stalemate in the contemporaneous and scientific evidence. The scientific implausibility of spontaneous combustion arising in a tip such as this after such a prolonged period of inactivity should, in the end, prevail. Mr Hand relied upon the history of the tip as described by NCB employees and the absence of evidence of an external source of ignition, evidence which was to be expected if that was the cause. The video evidence as explained by Mr James was important, as was the contemporaneous observation of the behaviour of the tip by Mr Royston Davies. Since both sides accepted that spontaneous combustion would take place only upon a collision of the precise circumstances which allowed it, it was likely that those circumstances came along infrequently. The interva l of time since the last known incident of heating did not, therefore, undermine the force of the other evidence in favour of spontaneous combustion.

Mr Bacon when asked in cross examination, "In the years after the closure of Brynlliw there is no evidence of heating or of a fire but we do not know do we whether there was or was not?" he replied, "If heating was present you would expect them to develop as they did in 1996". Since it was Mr Bacon's evidence that spontaneous combustion was not a probable cause of the fire I found this answer to be curiously inconsistent with the thrust of his evidence. Furthermore, it betrayed a recognition that if heatings were indeed taking place spontaneous combustion was at least a foreseeable consequence. Mr Bacon disavowed any intention by his answer to imply the cause of ignition in 1996, but when in re-examination Mr Darling asked Mr Bacon, "If you have heatings does it mean you are going to have spontaneous combustion?" he replied, "Not necessarily, obviously. The temperature will not have reached a critical level". There was a further inconsistency here which, in my view, tends to undermine the weight of Mr Bacon's expert opinion as to cause. Mr Bacon offered no explanation why it was perfectly possible scientifically for the tip to have been heating in the period 1983 to 1996 without reaching ignition temperature, while denying that it was probable the tip was heating if did not as a result combust.

Dr Edwards in his report was explicit as to the factors which would have to be present to generate sufficient heat to cause combustion: "All coal seams have some propensity to spontaneous combustion. The significance of the problem depends upon a complex relationship among a range of factors, the more important of which are:

In cross examination Dr Edwards agreed that although one incident of combus tion would not condemn a coal seam, those responsible for safety would have to factor the event into a reassessment of risk. He accepted that heatings may occur which did not reach ignition temperature. He went to say that he would expect coal stocks to show elevated temperatures under some conditions without reaching ignition. His research had led him to conclude that the prevailing west south west wind blew 40% of the time. It tended to be a summer prevailing wind. Dr Edwards said that he did not regard 40% as a sufficiently prevailing wind to create a risk of combustion but then went on to say that between November and April the prevailing wind was northerly. It would seem to follow, on Dr Edwards' analysis, that the summer wind did indeed prevail in the sense which may assist combustion. Dr Edwards explained that the ability of air to ingress the combustible material in a waste tip was governed by the differential air pressure at the bottom and top of the slope. What is required is a balance between too little and too much air. Hence the importance of precise conditions prevailing at any given time. He accepted (only as a generality) my naïve comparison with a chimney. The loosely compacted material on the slope lay against the harder compacted interior. I have seen an album of photographs taken by Mr Peter Arthur along the top of the tip on 23 November 1998 when it was plainly burning fiercely. There were fissures in the surface parallel with the western flank where the loosely compacted material of the face rested against body of the tip. Smoke is to be seen emerging from the fissures. As to the opinion of Mr Bacon and Dr Edwards that tree cover would have provided a beneficial reduction in differential air pressure, it seems to me, upon examining the video film, that there was little or no tree or other cover at the foot of the south western flank, where the underground fire was first observed by Mr Royston Davies and Mr James. I have no expert or other evidence with which to assess the likelihood that meteorological conditions favoured heating and combustion in 1996 but I am satisfied on a balance of probability that the fire of 1996 combusted spontaneously. I hesitated to reach this conclusion in the face of the undeniable fact that there was no positive evidence of heatings or combustion between 1983 and 1996 but I am persuaded that Mr Hand is right. Scientifically, it is perfectly possible that the right conditions combine in unknown and unpredictable meteorological and other circumstances. Once there is a propensity created by some of Dr Edwards' identified factors, it will require the combination of others to bring about the risk of combustion, and they may be present rarely, if at all. Furthermore, Dr Richards identified precisely the kind of event (exposure of fresh coal by the scouring action of water), which would create conditions of risk in a tip which had shown itself to be susceptible in the past. Neither Mr Bacon nor Dr Edwards dealt with Dr Richards' opinion in any detail in evidence. Mr Bacon expressed the view that the video film did not assist him to determine the cause of the fire. He personally did not see anything which pointed to spontaneous combustion. Dr Edwards thought that a tip which had been inactive for 20 years was not at risk of spontaneous combustion. They did not, however, challenge the scientific basis for Dr Richards' view that inactivity between 1983 and 1996 did not determine risk if the circumstances for a combustion were recently created, and they did not appear to wish to engage with the details and significance of Dr Richards' observations of the video film.

It seems to me that the inference that this was not an ignition caused by inadvertent or deliberate fire setting on the surface is to be preferred. The fire started in an area where burning vegetation was likely to be observed and it was not. The tip had been behaving, Mr Davies observed, in a manner consistent with an underground rather than surface ignition. Mr James found no evidence when he inspected the area of any trees burned other than through the roots. The ground cover on the slope where the fire was found was comparatively sparse. There was no evidence of an extensive surface fire. The trees and vegetation on either side of the underground fire were intact. Exposure of new sources of combustion to air would explain an outbreak of spontaneous combustion in a flank which had suffered heatings in the past. I prefer the evidence of Dr Richards upon the cause of the fire.

The fire began on or about 1 August 1996. Conditions gradually worsened during 1997, 1998 and 1999 until about February 2000 when remedial work undertaken by West Glamorgan County Council gradually restored normality. I have already referred to the progression of the fire northwards along the western face of the tip and inwards into the face as Dr Blandford noted it in 1998 and 1999. In the edited video film provided for my use I have seen television news footage of the effect locally of an easterly wind blowing towards Pentre Road and a westerly wind taking smoke in the direction of the M4 motorway, stopping traffic. I have viewed photographs of the fire taken on 28 November 1998. I have seen a photograph taken by Eleanor Hill in 1996 or 1997 from the football ground in Grovesend, near the south western flank of the tip, in a north easterly direction on a fairly clear day, depicting smoke billowing into the sky from west to east. I have seen other photographs taken in the vicinity of the home of a Mr Burder who lives a few metres from the western flank of the tip about midway between Grovesend and Waungron. Keith Hughes, now sadly deceased, kept a diary which I have read of relevant events between 23 November 1998 and 5 February 2000 (5/70/271-323).

I have considered the witness statements of each of the claimants (1/11/107-1/18/131) in which were described the effects upon each household. These accounts were not challenged by the defendant. They are consistent. It is plain that once it had taken hold the fire generated large quantities of smoke and dust and a characteristic and pungent smell of sulphur. When weather conditions were unfavourable (particularly low cloud and light winds) the smog hung like a pall over the whole area, including, on some of these occasions, the M4. Significant interference with the use and enjoyment of property was a regular and occasionally prolonged event over a period of almost 4 years. I have no difficulty in concluding that each claimant suffered loss of amenity and enjoyment of his or her property. I accept that the effect upon each property would vary according to the wind direction, although I noted the evidence that the sulphurous smell pervaded the area in conditions of light wind, whatever its direction. Some of the subject properties were at the north western end of the tip, one was to the north and the other to the south of the tip. The prevailing summer wind was westerly. Conditions were worse in the autumn and winter but when the smoke came down in the summer, householders were unable to enjoy their gardens.

Conditions would require householders intermittently to close their windows, to bring in washing, to stay indoors and to wash and clean their belongings. On bad days breathing was affected. Some of the witnesses referred to 'whiteouts' when it would be difficult to see to park a car or to move about a property outside.

Mr Darling and Mr Daiches did not, on behalf of the defendant, endeavour to persuade me that if the other elements of the tort of nuisance were present the effects noted were other than "undue interference with [its] neighbour in the comfortable and convenient enjoyment of his land": Clerk & Lindsell on Torts, 18th Edition 19-06, cited with approval in Thompson-Schwab v. Costaki [1956] 1 WLR 335 at 338 per Lord Evershed MR). Mr Daiches, who addressed me on behalf of the defendant on quantum, drew my attention to two first instance decisions of awards for loss of use and enjoyment intermittently over a prolo nged period. There were similarities and dissimilarities between those cases and this. He reminded me that an award for this category of nuisance was not the equivalent of an award for personal injuries but compensation for a species of proprietory loss. When invited by me, Mr Daiches suggested that an appropriate award for each household would be based upon £1,000 per annum for each property. Mr Hand on behalf of the claimants, while inviting me to make my own assessment, did not submit that Mr Daiches' figure was significantly wide of the appropriate mark, nor that I should distinguish between one neighbouring property and another.

In Sedleigh-Denfield v. O'Callaghan [1940] AC 880 (HL) an occupier of land was held liable for the continuation of a nuisance caused by the installation by a trespasser, Middlesex County Council, of a drainage pipe in an agricultural ditch on the occupier's land which enabled the ditch and pipe to be covered over. The trespasser placed a grating on top of the pipe instead of placing it in the ditch a short distance upstream of the pipe. As a consequence the pipe periodically became blocked with leaves which a servant or agent of the occupier cleaned out twice a year. However, during a period of heavy rainfall the pipe was blocked with refuse and the ditch overflowed causing damage to a neighbour's land. Since the occupier had through his servant or agent knowledge of the hazard created by the pipe and its incorrectly situated grating he had continued and was responsible for the nuisance which caused damage to his neighbour. Lord Atkin, at page 986, described the tort of nuisance as follows:

Lord Wright in a much quoted passage from his speech at page 903 said: "A balance has to be maintained between the right of the occupier to do what he likes with his own, and the right of his ne ighbour not to be interfered with. It is impossible to give any precise or universal formula, but it may broadly be said that a useful test is perhaps what is reasonable according to the ordinary usages of mankind living in society, or more correctly in a particular society. The forms which nuisance may take are protean."

At page 904 Lord Wright continued:

Lord Porter, at page 919, considered the argument that although the occupier knew that the grating should have been situated in front and not on top of the pipe, he had no reason for suspecting that any trouble would ensue. The occupier contended that he had no knowledge of the existence of a nuisance, only of the potentiality of a nuisance, i.e., the possibility that the pipe might at some future time become blocked and cause a flood on his neighbour's land:

The claimants' case here is that the defendant's predecessor, NCB, created a potential nuisance. Its potentiality is not a defence and the defendant does not argue that it is. The defendant's first line of defence is that the use of its land was, in all the circumstances, reasonable and, its second, that damage to its neighbours was not foreseeable.

The use of land for the storage of mining waste was considered by the House of Lords in Attorney General v. Cory Bros. & Co [1921] 1 AC 521. Cory Brothers were found liable both in negligence and under the rule in Rylands v. Fletcher. In describing the foundation for liability Viscount Haldane said at page 536:

I draw attention to this passage with which the majority, in other words, agreed in the knowledge that non-natural and extraordinary use are not necessarily the same thing as unreasonable user of land. It seems to me, however, that the creation of an artificial structure which is potentially dangerous (which the House of Lords regarded as nonnatural or extraordinary use of land for the purpose of Rylands v. Fletcher liability) may also be an unreasonable use of land in proof of private nuisance. To lay a spoil heap would be an ordinary use of land for the purpose of considering liability for escape unless the manner in which the spoil is heaped creates a danger to others: see Rickards v. Lothian [1913] AC 263 at 280. In my view, apart from a case of escape, such a structure, giving rise to the potentiality of damage to one's neighbour may be an unreasonable use of land, for which the creator would be liable.

In his review of the law of nuisance in Cambridge Water Co v. Eastern Counties Leather plc [1994] 2 AC 264, Lord Goff of Chieveley said, at page 299D:

Critical to the issue of liability in private nuisance is the question whether the damage done was foreseeable. Having quoted from the opinion of the Board delivered by Lord Reid in Overseas Tankship (UK) Ltd v. Miller Steamship Co. Pty. (The Wagon Mound (No 2)) [1967] 1 AC 617:

There is in the present case a further consideration arising from the use by Lord Wright in Sedleigh-Denfield of the words "latent defect". As a matter of fact the creator of a hazardous state of affairs may be unaware that the state he is creating is hazardous. In Cambridge Water Co the leather company spilled over several years a substance which experience led it to believe would evaporate upon contact with the ground. Unknown and, with the expertise then reasonably available to the company, unknowable, some of the spillage seeped into the concrete floor of its premises and entered the chalk aquifer beneath, and thence contaminated by means of micro drainage channels a source of drinking water a neighbour wished to tap. Liability was not established in nuisance or under Rylands v. Fletcher because the company "could not at the relevant time reasonably have foreseen that the damage in question might occur...", per Lord Goff at page 307E. By the time the scientific possibility for the route of contamination became known the substance had become irretrievably lost in the ground below.

Towards the end of his speech Lord Go ff considered the trial judge's finding, made for the purpose of rejecting the Rylands v. Fletcher claim, that the storage of the chemical which caused contamination was a natural use of land. Lord Goff, at page 309, disagreed:

It follows I think that the creation of a state of affairs on land which, at the time, carries an unforeseen and unforeseeable risk of damage to one's neighbour is not actionable in nuisance. If, however, by the improvement of knowledge, scientific or otherwise, the risk becomes foreseeable, one is under a duty to abate that state of affairs and, if one fails to fulfil the duty to abate it, then, on the occurrence of damage, the nuisance is actionable. The test cannot be more favourable to the creator of the nuisance than it was to the occupier who came across a state of affairs which already existed. But what is the extent of the duty? The claimants' case is that the hazard must be removed. "A private nuisance arises out of a state of things on one man's property whereby his neighbour's property is exposed to danger", per Atkinson J in Spicer v. Smee [1946] All ER at page 393. The defendant, on the other hand, argues for a measured duty of care similar to that which was held by the Court of Appeal to exist in Leakey v. National Trust [1980] 1 QB 485 and Holbeck Hall Hotel Ltd. v. Scarborough Borough Council [2000] QB 836.

In neither case did the defendant create the hazardous state of affairs. Nature created a state of affairs which rendered the defendant's land a danger to the claimant. The principle was derived from the decision of the Privy Council in Goldman v. Hargrave [1967] 1 AC 645. In Leakey, Megaw LJ at page 524E described the scope of the duty as follows:

Leakey was considered in Holbeck Hall. At page 858B, Stuart-Smith LJ further analysed the circumstances in which the measured duty of care arises. It arises when:

The duty did not extend to the expenditure of large sums of money upon geological investigation to ascertain whether the danger was substantially greater than appeared upon reasonable investigation to be the case.

I am aware that I am not dealing with an accident of nature but with the creation of a structure which subsequently acquired experience demonstrated carried a foreseeable risk of damage. In my view, however, the duty to abate the risk of damage is not an absolute one. The original duty of the occupier was to make reasonable use of his land and no more. If at the time of the creation what he did was reasonable because damage to his neighbour's use and enjoyment of land was unforeseeable, then it seems to me he is in no morally different position (and should be in no different legal position) from the occupier who suffers the acts of trespassers (see Smith v. Littlewoods Organisation Ltd [1987] 1 AC 241), or the occupier who is a victim of the forces of nature which render his land a danger to the land of his neighbour. I accept the submissions made to me by Mr Darling that the occupier fixed with knowledge of the relevant risk is expected to do what is reasonable in the circumstances to abate that risk. What steps should have been taken, if any, will depend upon factors to which both Megaw and Stuart-Smith LJJ referred in Leakey and Holbeck Hall. In a case such as this an assessment of what is reasonable will depend, it seems to me, largely upon the magnitude of the risk of damage and the extent of damage should the risk transpire, the cost and convenience of abating the risk, and the resources of the occupier who created it.

The issue of foreseeability was fully argued in the Court of Appeal in another Coal Authority case, Arscott & Ors v. The Coal Authority & Anr [2005] Env LR6, [2004] EWCA Civ 892. NCB used coal spoil to raise the height of the banks of the River Taf in Aberfan. The spoil created an enhanced risk of flooding of homes and, in a time of heavy rainfall, flooding occurred. The Coal Authority successfully argued that it was unforeseeable that the works increased rather than diminished the flood risk. On appeal the claimants submitted, relying on Hughes v. Lord Advocate [1963] AC 837, that it was not necessary to prove that the defendant should have foreseen the precise damage occasioned by the precise mechanics which in the event occurred. Laws LJ said at paragraph 58:

Hindsight is of no va lue. I am required to look at the state of knowledge of and the knowledge available to the defendant's predecessor between 1957, when in modern times the tip was first used, and 1995, when they disposed of their interest in the land to another.

I need to consider three periods, 1957-1971 when the tip was active, 1972-1988, when the tip was closed, disused and re-landscaped, and 1989-1995, when the tip was in its restored condition.

In March 1945 the Minister for Fuel and Power presented to Parliament the Coal Mining Report of the technical Advisory Committee (Cmd 6610). Paragraph 604 read as follows:

The conclusion reached and recommendation made in Section D, 'Refuse Disposal' included "(iii) Effective measures should be taken to prevent the serious nuisance caused by burning tip heaps".

It follows that it was well known in the coal industry that spontaneous combustion was liable to occur in coal spoil tips and that, if it did, serious nuisance might be created.

Following the disaster at Aberfan, NCB required a much improved system of tip inspection, not just to deal with stability but also to deal with fire (Production Department Instruction PI/1967/16-see 6/2/167). Any signs of fire hazard, "visible burning, steam or smoke, fumes or smell of burning, heating of surface, cracking of surface, collapse inwards because of cavitation" were to "be looked for and reported on. If they are of any size, and particularly if there appears to be change taking place, then immediate action is necessary and must be taken".

The publication 'Spoil Tip Management' produced by Production Department was revised in February 1968. At section 5 'Methods' the instruction explained the danger of loosely compacted material:

In dealing with inspections at section 7.1, "Method of Carrying out Inspections" the advice proceeds:

In its Technical Handbook, 'Spoil Heaps and Lagoons', September 1970 NCB said in the Preface:

Paragraph 1.2.1 draws attention to the fact that many existing tips which have been constructed with loose slopes comprising dry material such as run-of-mine or development dirt, tipped over high faces, and a low water table or where the tip has been built over springs, will be partially or completely burnt. Tip 52 was such a tip.

At section 6 the Handbook deals with Improvement, Precautionary and Re medial Work. Remedial work took place in 1987, primarily for aesthetic reasons. Paragraph 3 recommended that when remedial work takes place care should be taken to ensure that adverse secondary effects do not outweigh the benefits. Aspects to be considered included erosion caused by mud runs, gullying caused by uncontrolled drainage, the starting of fire or encouragement of burning.

In section 8.2 the Handbook draws attention to the risk of accidental ignition of spoil heaps by hot ashes and the lighting of fires but "The most common cause of burning, however, is spontaneous combustion of carbonaceous material, often aggravated by the presence of pyrite. This is an oxidation process in which the material combines with oxygen from the air with the evolution of heat. Oxidation usually proceeds very slowly at ambient temperatures but increases rapidly and progressively as the temperature rises". Section 8.3 explained that materials such as wood could in the presence of moisture ignite spontaneously during the process of heating under the influence of ambient temperatures despite the fact that their ignition temperatures on exposure to oxygen were normally within the range 296º-300º. In general lower rank coals were more reactive and hence more susceptible to self heating than higher rank coals. It is the oxidation of pyrite and organic sulphur in the coal which forms sulphur dioxide and the characteristic smell from which a heating can provide a means of detection of heating. At relatively low temperatures an increase in free moisture increases the risk of spontaneous heating. With fine material having small air voids the air remains stagnant and the heat generated is retained in the mass; but when the available oxygen is consumed the heating stops. With intermediate gradings and voids the conditions for spontaneous heating are ideal, and the heated parts may form hotspots and eventually break into flame.

Importantly, section 8.4 informs that there is no inexpensive method for preventing spontaneous combustion of existing heaps which are inclined to catch fire. However:

Section 8.5 identifies the hazards of burning spoil heaps, particularly and principally noxious gases.

Section 8.6 deals with methods of dealing with a tip fire. Its contents emphasise the scale of the problem and the expertise required. Methods include digging out, trenching, blanketing, injection with inert material and the judicious and informed use of water.

In July 1981 the Commission on Energy and the Environment issued the report 'Coal and the Environment' (7/8) in which, at 9.10, it was announced that burning on tips constructed by modern methods described in the Technical Handbook (1970) had been eliminated.

NCB published in 1984 the 'Composition and Engineering Properties of British Colliery Discards' review written by Dr Roy Taylor in 1983. In it, at 3.4 Dr Taylor wrote:

Dr Richards expressed the opinion in his supplementary report that experience of Tip 52, and knowledge within the industry generally, exemplified by the foregoing publications, demonstrated that the fire in Tip 52 was a foreseeable event.

Dr Richards agreed that in the period 1970-1975 people would associate the risk of spontaneous combustion with rank of coal. High rank was less likely to combust than low rank. But it was the history of the seam and the area which had to be taken into account. If he had been asked to assess the risk which the tip presented he would have included the history above and below ground at the colliery and in the surrounding area. Dr Richards pointed out that there were thermometer probes kept at Brynlliw which were used by the tip foreman. He posed the question why they should have been in use at all unless there was a knowledge of the risk of burning, whether by spontaneous combustion or from external ignition.

It is probable, in my view, that in the period 1957-1970 management at the colliery believed the risk of spontaneous combustion to be negligible. While the risk of combustion in waste tips was generally known, it was not known that the risk applied to the West Wales coalfield until about 1970 when the Symposium took place because it was thought that the rank of coal was too high to permit combustion in waste. I have insufficient information about causation in other tips, particularly at Mountain Colliery, to reach the conclusion that spontaneous combustion in the flanks of the Brynlliw tip before 1970 was foreseeable. However, during the period 1971-1983, after tipping has ceased, this tip was known to be a risk for spontaneous combustion. The danger presented by the tip was kept within bounds by the system of inspection which was instituted by management and performed by the tip foremen and engineers. When heating and combustion took place it was prevented by prompt and expert action from becoming a nuisance to neighbouring properties. Had NCB been considering the return of the land on which the tip was situated by 1983, I have no doubt the recent history of the tip would have required remedial measures to have been taken to render its flanks safe from spontaneous combustion. Without them the tip would have been subject to a risk of burning over which the commoners would have no control. Those measures would have included the need to streamline and compact the flanks as well as the body of the tip. In my judgment, the absence of reported or recorded instances of heating and combustion between 1983 and 1995 did not, in the case of Tip 52, render the risk of combustion or the prospect of damage to neighbours by the effects of fire unforeseeable. While it was the responsibility of the contractor to report any evidence of heating, the work the contractor was required to do did not include disturbance of the western flank, save to pull back and spread the top 3 metres. That work tells us nothing about the susceptibility of the undisturbed flank to heating. British Coal knew during the 1980s that the western flank was subject to erosion, precisely the mechanism by which I have found new coal waste was exposed to oxidation, regenerating the known risk of heating and, therefore, combustion.

NCB informed West Glamorgan County Council that once the land was handed back it would ensure aftercare of the tip by requiring an aftercare agreement from the commoners. The transfer (4/73) of 198.18 acres for a consideration of £1 is dated 22 June 1995, between British Coal Corporation and four named commoners, and contained a number of covenants, the burden of which was upon the purchasers. They included an obligation that the purchaser should "employ all practical means for preventing combustion of all refuse deposited on the transferred property from any mine or quarry and for preventing or minimising the emission of smoke or fumes therefrom". Other covenants dealt with avoidance of the risk of instability. In my view, an awareness of the danger of combustion and the effects of combustion of the tip is revealed by the inclusion of the requirement to employ all practical means. It is not limited to taking steps to prevent the lighting of fires on or in the vicinity of the tip but is general in its terms.

I have already considered the risk posed by Tip 52 in order to arrive at my conclusion as to the cause of the 1996 fire. Dr Richards gave evidence that what would have been required to reduce the risk of fire in the western flank was the removal of trees which had been maturing since 1972, the relaying and compacting of the flank in a more streamlined form, and re-planting of suitable trees and vegetation. The original cost of the work on Tip 52 (removal of the top 3 metres of spoil from the crest-120,000 cubic metres) was estimated by Mr Snell to be in the region of £80,000, later according to Dr Richards (2/4/158), £134,000 plus £17,000 for moving 35,000 cubic metres. An accurate final cost has not been supplied. The cost of remedial work to the burning tip was estimated in 1999 by Robert West Consulting at £1 million. I do not have a costed scheme to consider but it is not suggested by the defendant that the standard NCB should have reached ought to be adjusted for lack of means. The defendant contends that what NCB did was reasonable in the circumstances.

In the Technical Handbook (1970) section 7.1, the authors acknowledged that although landscaping was usually carried out for aesthetic reasons, it may provide additional benefits such as "(c) reducing the risk of combustion of the tip material".

Mr Bacon expressed the opinion in cross examination that unless there was a pressing need for additional expenditure mere landscaping was sufficient. It was suggested to him that negotiations with the local authorities and the commoners revealed that NCB did not intend to expend more than was necessary to satisfy the demands being made on them. He replied that the commoners and the local authorities were satisfied. You would not spend a lot of money if you could satisfy the planners. Mr Bacon did not agree that the tip was a fire risk, although he acknowledged the recommendation of the Technical Handbook (1970) (above) and agreed that Taylor (1984) (7/9/1254) identified as good practice in new tips streamlining the flanks, particularly in the face of a prevailing wind to reduce the risk of spontaneous combustion. Mr Bacon was asked in re-examination whether it was common, appropriate or necessary to consider modern tip construction practice in connection with the management of an old, closed tip. He replied that it was not the approach of NCB to apply tip management rules in retrospect. In the light of NCB's own advice in the Landscaping section of its Technical Handbook September 1970 (landscaping is an opportunity to prevent combustion), I find that information surprising. Mr Bacon went on to say that, on the other hand, great attention was given to all tips. The characteristics of all tips was considered. In my view, that was not the case with Tip 52. Close attention was plainly not given to the condition of Tip 52 after the tip foremen ceased their inspections in 1983 when the colliery closed. No-one carried out an assessment which took into account the history of the tip and no examination took place to ascertain whether the western flank was susceptible to erosion capable of affecting the tip's propensity to heat. Mr Bacon agreed in cross examination that a history of heatings would have to be taken into account and that re-landscaping, compacting and replanting could have been achieved. His view, referring to the evidence of Mr Webster, was that if there had been a history of heatings it would have been considered. Mr Bacon thought it was 'perfectly understandable and reasonable' that the tip should be left as it was. Trees had been standing for 20 years and a good covering of trees was a recognised feature of landscaping of old tips. The Handbook of 1970 recommended that tip surfaces which are sus taining a good covering of vegetation should be left alone as they are nearing equilibrium (6/4/321/7.1(g)). It seemed to me that Mr Bacon's selection from the terms of the Handbook depended for its justification upon his view that there was no risk of spontaneous combustion revealed by the tip's recent history. The difficulty with his conclusion is that it proceeds on the assumption which is not justifiable by experience.

Alan Webster was a member of the team which organised the restoration in 1987. He said he knew that there had been fires on the tip and that there had been problems with vandalism but he did not know of any evidence of deep internal heating. He had not heard of the heatings described by Mr Phillips and Mr Vaughan. He suggested that if those incidents took place there should have been records which came to his attention. Had he known of them he would not have touched the proposed landscaping. He acknowledged that the heatings the tip foremen described took place while the colliery was still open so that men with expertise could deal with them. Once the colliery closed that local control was no longer present; if a heating became a fire the chances of it taking hold were that much greater. He said that when he visited burning tip s there always seemed to be an account of external ignition and although he and colleagues in the civil engineering department discussed spontaneous combustion they concluded there was no clear evidence of it. Despite Taylor's recommendation that flanks should be streamlined in the face of the prevailing wind, he said that he did not believe that a steep flank facing a prevailing wind did increase the risk of spontaneous combustion. He said that the flank was soiled. In fact the soil placed on the tip was a peaty soil itself liable to catch fire, as it had in the past. Furthermore, Mr Webster's reply revealed a willingness to prefer his own assessment of the effects of the prevailing wind at odds with the guidance provided to the industry. He gave no explanation for his personal view. Mr Webster said that the object of leaving the vegetation was to maintain good vegetation and drainage on the flank. He said that there had in the past been bad flooding to the western foot of the tip. In fact, as I have observed, erosion caused by uncontrolled water run off had been noted in 1985 and it was still occurring 11 years later. He agreed that the flank could have been compacted by using rollers. Asked whether any consideration had been given to compacting the flank, he said, 'No, it was never considered'. Had he been aware of evidence of spontaneous combustion, he conceded it is very possible that there would have been a different sort of restoration of the site. During the restoration he had been on the site almost every day and he had seen no sign of spontaneous combustion.

I have no doubt that Mr Webster was accurately reporting area management's state of mind at the time restoration was being planned, but I am deeply unimpressed by the route taken to that state of mind. There was a history of heatings on this tip a few of which had led to fires which could not be put down to outside agency. They were, as I have found, well known to the tip foremen and the engineering staff to whom they were reported and the surveying team which prepared the annual reports. I agree with Dr Richards' impression of a culture of blame attached to outside agencies when the evidence told that there was, in addition, if not instead, an inherent problem. Either these heatings were not properly reported to area civil engineering staff or, if they were, they were not given the prominence they deserved. What is clear is that when the colliery closed, nobody consulted the colliery staff with a view to reaching an accurate appreciatio n of the remedial works necessary to restore the tip to the commoners in a safe condition. No-one gave a thought to the historical fire risk, only to satisfying the commoners and the planners, none of whom were told about any risk of spontaneous combustion.

I again accept the evidence of Dr Richards that, in the light of events at this and other collieries, what was required before restoration in 1987 was an assessment of the risks presented by them. NCB had knowledge by its responsible employees and managers, or had the means of knowledge, that there was a foreseeable risk of spontaneous combustion in their tip causing nuisance to its neighbours. If the risk became reality the consequences could be serious and far-reaching. Had that assessment been performed when the natural opportunity presented itself NCB must have concluded that relevant remedial action was required. Thus far, nuisance had been averted by prompt action by employees. Upon closure of the colliery that level of control was no longer available. On the face of it, either special arrangements to continue the regular weekly inspections indefinitely, or civil engineering steps to remove or to reduce significantly the risk of spontaneous combustion, were necessary. Since it was NCB's intention to restore the tip to grazing land for the commoners in any event, this was an opportunity, at what I infer would have been a reasonable cost well within the defendant's capacity, to take action to remove that risk or to reduce it to negligible proportions.

I consider that NCB and British Coal failed in the circumstances to take reasonable steps to abate the nuisance whose potential survived the 1995 transfer and became reality on or about 1 August 1996.

I give judgment for the claimants and award the sum of £3,500 damages for nuisance in respect of each of the neighbouring properties the subject of these claims.

I have concluded the issue of liability in favour of the claimants upon their principal case. Their alternative case was that the defendant was liable in nuisance even if the fire of 1996 was caused by external ignition. Had I decided that the fire was caused by trespassers, I would have found that the defendants were not liable in nuisance or negligence.

As to nuisance, it is my judgment that the defendant's use of its land was unreasonable only by reason of the exposure of its neighbours to the risk of spontaneous combustion. That the tip was combustible by means of a significant fire lit on its wooded flank by a trespasser was not, in my opinion, an unreasonable use of land, any more than would have been the storage of a large quantity of any other combustible material such as wood. It was argued that the risk of prolonged fire and noxious emissions from such a fire created the same, or at least a comparable, unreasonable hazard to the use and enjoyment of neighbouring property as did the risk of spontaneous combustion. That such a risk was present should a fire be deliberately started is clear from the findings I have made. However, risk may be present from different causes, all of which are foreseeable, including the work of trespassers. That does not mean that the occupier is liable for each risk which may be realised. In my judgment, it is only the risk associated with the unreasonable quality of the defendant's use which entitles a neighbour to a finding of nuisance.

As to negligence, I do not consider that a duty of care to its neighbours, however expressed, required NCB to protect the flank from fires deliberately started by trespassers, beyond the action it took to fence the foot of the flank following restoration and before transfer.

Finally, and in the further alternative the claimants argued that this was a case of public nuisance. Interesting arguments were addressed to me as to whether, apart from the necessity to prove a nuisance to the public, the elements of public nuisance were the same as those for private nuisance. Having seen a good deal of documentary evidence of the effect of this fire on the local population generally, I have no difficulty in concluding that there did exist a nuisance to the public. However, in the light of my finding that private nuisance is proved, it is unnecessary for me to embark upon an analysis whether compliance with a measured duty of care is a defence to an action for public nuisance.