Barry Edward McDonald (t/as B.E. McDonald Transport) v Girkaid Pty Ltd & 4 Ors; Robert Bryce & Co Ltd v Girkaid Pty Ltd & 4 Ors; Hudson Resources Pty Ltd & 5 Ors v Robert Bryce & Co Ltd

Structure of the proceedings 11 Girkaid and Inghams sued Mr McDonald in negligence and for breach of statutory duty said to have been created by regulation 19 of the Dangerous Goods Regulation 1978 (NSW). They were successful. Adams J held that the fire had been caused by the spontaneous combustion process for which Girkaid and Inghams contended, using the term "spontaneous combustion" as meaning self-heating to the point of spontaneous ignition. 12 The primary judge entered judgment in Girkaid's favour against McDonald for $543,014.84. He entered judgment in Inghams' favour against McDonald for $709,913.91. 13 The relative simplicity with which the proceedings were commenced belied the pleading complexity which emerged. The statement of claim spawned nine cross-claims which I describe below in the order in which they were filed. I have also sought to identify the outcome of the cross claims, although in some respects the orders entered did not appear to reflect the judgment. 14 McDonald denied it was liable to Girkaid and Inghams, but cross-claimed against Bryce claiming the fire and any resulting loss were caused by its breach of a term implied into the contract of bailment to give McDonald such information as would enable him to store and deliver Brytall safely. McDonald also claimed the fire and any loss Girkaid and Inghams suffered were caused by Bryce's breach of statutory duty under regulations 18, 248 and 249(3) of the Dangerous Goods Regulation and its breach of its common law duty of care. McDonald sought to recover damages as well as contribution and/or indemnity for any judgment he might suffer pursuant to the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) (the "first cross-claim"). The primary judge found that Bryce had breached the contract of bailment in failing to inform McDonald of the propensity of Brytall to undergo spontaneous combustion when mixed with a substance such as attapulgite. As the damages for breach of contract were not susceptible to reduction for contributory negligence (Astley v Austrust Limited [1999] HCA 6; (1999) 197 CLR 1) he ordered Bryce to indemnify McDonald in respect of his liability to Girkaid and Inghams. He also entered judgment on the first cross-claim in McDonald's favour for $179,940.00 in respect of losses McDonald suffered in the fire. 15 McDonald also cross-claimed against SLE claiming the fire and any resulting loss were caused by its breach of contract, breach of statutory duty and breach of its duty of care. He also sought contribution and/or indemnity for any judgment he might suffer pursuant to the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) (the "second cross-claim"). Adams J dismissed the second cross-claim. 16 Bryce cross-claimed against McDonald for breach of the contract of bailment of the goods McDonald held on its behalf at the warehouse on the night of the fire. It also claimed the fire and any loss Girkaid and Inghams suffered were caused by McDonald's breach of his duty of care and statutory duty (the "third cross-claim"). The primary judge entered judgment in Bryce's favour against McDonald in respect of the lost goods for $384,701.00 being their agreed value of $272,173.00 (see [52]) plus interest (see transcript 8 March 2002 at Black 247F). Bryce also sued EKA Chemicals in a fourth cross-claim but that claim was discontinued (see judgment at [57]). 17 SLE cross-claimed against Hudson seeking damages for breach of contract, breach of duty of care and breach of s 52 and s 53(a) of the Trade Practices Act 1974 (Cth). It also sought contribution and/or indemnity pursuant to s 87 of the Trade Practices Act and s 5 of the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) (the "fifth cross-claim"). The primary judge dismissed the fifth cross-claim. 18 SLE also cross-claimed against Bryce seeking indemnity and/or contribution pursuant to the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) (the "sixth cross-claim"). The primary judge dismissed the sixth cross-claim. 19 Hudson cross-claimed against McDonald, Bryce and SLE seeking contribution and/or indemnity pursuant to the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) (the "seventh cross-claim"). The primary judge gave judgment for SLE on the seventh cross-claim and dismissed the seventh cross-claim against Bryce. 20 Bryce cross-claimed against Hudson and SLE seeking damages in respect of the loss of its containers of tall oil as well as contribution and/or indemnity pursuant to the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) (the "eighth cross-claim"). The primary judge gave judgment for SLE on the eighth cross-claim. He entered judgment for Bryce against Hudson for $153,880 (40% of $384,701). 21 McDonald cross-claimed against Hudson seeking damages as well as contribution and/or indemnity pursuant to the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) (the "ninth cross-claim"). The primary judge entered judgment in favour of McDonald against Hudson in the sum of $179,940 as well as ordering Hudson to indemnify McDonald for 40% of the judgments in favour of Girkaid and Inghams and for 40% of the judgment in favour of Bryce. 22 In summary, Girkaid and Inghams succeeded against McDonald. Bryce succeeded against McDonald in its claim for the loss of its goods. McDonald recovered a complete indemnity for his liability to Girkaid and Inghams from Bryce and a 40% indemnity from Hudson. He recovered a contribution of 40% from Hudson in respect of his liability in bailment to Bryce. McDonald recovered judgment against Hudson and Bryce in relation to his lost goods. SLE escaped liability entirely.

The appeals 23 McDonald, Bryce and Hudson appealed challenging the judgments they had suffered. 24 Girkaid and Inghams filed a Notice of Contention contending that the primary judge ought to have found that McDonald was liable on the basis that regulation 19 of the Dangerous Goods Regulation 1978 gave Girkaid and Inghams a private right of action against McDonald which he had breached. 25 McDonald filed a Notice of Contention in Bryce's appeal asserting that the primary judge's decision in relation to the first cross-claim should be affirmed on the basis, inter alia, that his Honour should have found that Bryce breached its duty of care to McDonald by failing to warn him of the danger of spontaneous combustion. 26 SLE cross-appealed in each of the appeals complaining of the dismissal of its cross-claims against Bryce, Hudson and McDonald. I could not locate any cross-claim SLE had filed against McDonald which would ground its cross-appeal in this respect. Outline of the issues on appeal 27 There were, essentially, three issues on the appeal. They were, first, whether the primary judge's finding that the fire was caused by the spontaneous combustion of the Brytall/attapulgite mixture could be sustained. Secondly, even if the causation finding stood, whether Girkaid and Inghams had established that McDonald had acted in breach of any statutory or common law duty. Finally, there was a quantum issue going to the value of Girkaid's real estate consequent upon the destruction of the improvements erected upon it. 28 McDonald assumed the principal burden of challenging the primary judge's causation and breach conclusions. As the other parties' liabilities were consequent upon the primary judge's rulings adverse to McDonald, save as to one matter, the consequence of McDonald's appeal succeeding was that judgments on the cross-claims would be reversed. The one exception is Bryce's claim for breach of its bailment contract. Bryce sought to retain its judgment against McDonald on this issue saying it was entitled to do so unless McDonald could prove that Bryce's loss was not caused as a consequence of his failure to use such care and diligence as a prudent man would in relation to his own property: J Sperling Ltd v Bradshaw [1956] 2 All ER 121 at 125.

The facts 29 The core facts were set out by the primary judge in a manner which was not the subject of any complaint on appeal. For convenience I adopt that account (see Girkaid Pty Limited & 1 Ors v McDonald & Ors [2001] NSWSC 1202): "1 This litigation arises out of a fire which, for all practical purposes destroyed an industrial warehouse in Durie Road, Cardiff. The property was owned by Girkaid Pty Limited (Girkaid), the first plaintiff, and leased to two tenants, one of which is Inghams Enterprises Limited (Inghams), the second plaintiff, and the other, Mr Barry Edward McDonald (Mr McDonald) who is the defendant in the actions brought by the plaintiffs and also the cross claimant and cross defendant in other related actions.

2 Mr McDonald operated a transport business, which first started trading in about 1972. The firm focused on short hauls around the Newcastle area and from Newcastle and Sydney. Mr McDonald owned several trucks of varying sizes and used two forklifts in the warehouse. Mr McDonald himself manages and organised the business including, especially, taking responsibility for the warehouse premises, which were leased from Girkaid. Mr McDonald's business was relatively modest, employing only five persons. Amongst the items transported were alcohol products of various kinds, which were kept in the warehouse overnight and distributed the next morning. Chemicals and bagged material were also stored at the warehouse for a relatively short time before transportation to their destinations. Mr McDonald had been using the warehouse since about 1991. One of its customers was Bryce & Company Limited (Bryce) whose products were frequently stored and transported by Mr McDonald. Some of the goods handled by Mr McDonald are "dangerous goods" within the meaning of the Dangerous Goods Act 1975 (the Act) and the Dangerous Goods Regula t ion 1978 (the Regulation) and he possessed the necessary licence.

3 On 29 January 1998, Mr McDonald was telephoned by a lorry driver at about 2am and told that the warehouse was on fire. He quickly drove to the premises, which were blazing fiercely despite the efforts of the fire brigade. The warehouse was so badly affected by the fire that, in due course, it was necessary to demolish the structure to the concrete slab.

4 For obvious reasons, an investigation was conducted as to the cause of the fire. Suspicion focused on the use of a mineral product called attapulgite by Mr McDonald a few days before the fire to soak up a large quantity of tall oil (being held for on transportation on Bryce's account) that had been accidentally spilled. It is the plaintiff's case that the fire was caused by the spontaneous combustion of the mixture of oil and attapulgite (the mixture), which was left in the warehouse by Mr McDonald after the floor had been cleaned. On the other hand, Mr McDonald's case is that the evidence does not establish that the fire was caused in this way and, even if it did, he was not negligent in failing to guard against the risk.

5 It is necessary to briefly explain how the mixture came to be stored in the warehouse. This account comes from the evidence of Mr McDonald, which was not sought to be controverted. Mr McDonald's half of the warehouse is a large area with goods stored more or less around the perimeter. In the Inghams' half was stored additives for its feed mill as well as feed and machinery. On Friday 23 January 1998, a consignment of twenty containers of tall oil, imported and distributed by Bryce, was delivered to the warehouse for further delivery. The product was called Brytall and comprised linoleic and oleic acid as two of its major components. Each container held 900 litres of oil in a plastic bladder. The container itself is roughly octagonal and made of fibreboard with a wooden top and side supports reinforced by steel bands. Each of the containers sat on a wooden pallet. Mr McDonald unloaded the delivery, using a forklift to carry two drums at a time, one on top of the other. After moving six of the containers towards the back of the warehouse without incident, as Mr McDonald entered the factory with the fourth load the top container overbalanced and fell to the floor. Something like two hundred litres of oil remained inside the container. The other container on the pallet was slightly damaged and had a small leak of oil. Although the tops had been dislodged, the containers remained largely undamaged.

6 It is obvious that Mr McDonald needed to clean up the oil and he contacted SLE Industrial Supplies Pty Limited (SLE), with which he had a previous acquaintance, in order to obtain a product with which to do this. Mr McDonald spoke to Mr Emerson, who was one of the directors of SLE working as its sales representative, amongst other things, in which connection he and Mr McDonald had known one another for some time. Indeed, they were friends. Mr McDonald and Mr Emerson gave evidence as to their telephone conversation and what later occurred. As might be expected, their accounts are not identical but, in my view, little turns on this. The substance of their conversation was that Mr McDonald told Mr Emerson that he needed a large quantity of oil absorbent because about 1000 litres of pine oil had been spilled in the warehouse. (Tall oil is extracted from pine trees.) It appears that Mr Emerson agreed to supply some absorbent and arranged for its delivery to the warehouse. Mr McDonald also asked him to pick up an empty container from another firm and hire a pump so that oil in the damaged container could be transferred to an undamaged one. Mr Emerson did this and in due course arrived at the warehouse. In the meantime, a supply of attapulgite had been delivered and Mr McDonald was using it to prevent any further oil spread within the warehouse. Mr Emerson told him that the attapulgite should be spread on top of the oil and, after it was soaked, shovelled up. Mr McDonald placed the tall oil soaked attapulgite (the mixture) into the two drums which had been upset, one of which had 100-200 litres or so of unspilled oil in it, whilst the contents of the other had been transferred to the replacement container. Mr McDonald said that the bulk of the oil was soaked up and placed in the containers by the evening of Friday 23 January, with only a little oil left under some of the pallets that had not been cleaned up by that stage. Twenty-one bags of attapulgite were used in this process overall. He then left for the ensuing long weekend and returned on Tuesday to continued the cleanup, which was completed on the evening of Wednesday 28 January. Pulling or tearing a strip of paper-type material from the top opened the attapulgite bags. This material was gathered up during the process and placed with the mixture in the containers. I infer from Mr McDonald's evidence, although it is not entirely clear about this, that some, if not most, of these strips were put on top of the boxes when the cleanup was finished. By the time the cleanup was completed, one of the drums was about three-quarters full and the other about half full of the mixture. Mr McDonald said that neither of the containers showed any signs of becoming warm. Nor had they given off any fumes or smell. The uncovered containers storing the mixture were left in the warehouse. The positions of these drums, as they were left on the Wednesday night, is indicated by the letters A and B on a diagram supplied to Professor Gray, the plaintiffs' expert and appended to his report. Mr McDonald agreed that the containers were in the positions so marked, although the substance of his evidence was, to my mind, rather indicative than precise on this point. The points marked A and B on the diagram are described in Professor Gray's report as approximate locations that were not measured accurately as conditions in the warehouse, even on the date of his inspection, 16 February 1998, were still extremely difficult and dangerous for movement. Container B contained the bulk of the mixture. In the vicinity of the containers was material described as L7, being an inert, inorganic carbonate. When inspected by Professor Gray, a quantity of this material had fallen onto container B. Professor Gray discovered that the mixture in this container was still self-heating and was very, hot, though the fire had been extinguished two and a half weeks earlier…" 30 Professor Gray gave evidence that in his opinion the fire had been caused by the spontaneous combustion of the tall oil/attapulgite mixture. McDonald and the cross-defendants disputed that conclusion. The primary judge recorded the challenge to Professor Gray's thesis of spontaneous combustion in the following terms: "8 The scientific validity of Professor Gray's conclusion … that the mixture in the container he described as "B" had indeed self-heated was not really in contention, though whether it had done so to the point of ignition spontaneously or because of the heat from the fire otherwise caused was, it was argued, speculative and, in the end, unproved. An examination of container B showed that it had been considerably charred on the inside, more extensively than on the outside, and completely destroyed further up. The other container showed less combustion effects but self-heating had obviously taken place with darkening due to combustion in the interior of the mixture, although there was no significant thermal activity when the inspection occurred. Professor Gray considered that the time lapse between the oil spill and the fire was entirely consistent with the cause of the fire being spontaneous combustion of a body of about the same volume as the Brytall containers were." (emphasis supplied) 31 The following evidence should also be noted. 32 Material safety data sheets ("MSDSs") are prepared by companies such as Bryce and Hudson in accordance with the Worksafe Australia National Code of Practice for the purpose, inter alia, of warning persons dealing with the product of any dangers that might be associated with its use, including fire risk, accidental spillage, and procedures for clean-up (at [30]). 33 McDonald called Mr Underwood to give expert evidence. Mr Underwood had thirty years' experience as a professional engineer, with ten years of "fair involvement" in warehousing, and was an accredited consultant for WorkCover in relation to dangerous goods. According to Mr Underwood the National Code required the following information to be given in an MSDS in relation to spills: · Steps to be taken to minimise a spill or leak. · Methods used to clean up the spill or leak. · Precautions to be observed by the clean-up crew, and · Any special equipment required for clean up. 34 The National Code required the following matters to be dealt with in relation to the fire/explosion hazard of substances: "Fire and explosion hazards under any reasonably foreseeable conditions of use and storage should be defined, and any dangerous decomposition or combustion products listed", and "… any potentially dangerous interaction with other substances should be included." 35 Both EKA and Bryce had produced an MSDS in relation to Pinechem/Brytall. Neither MSDS had been supplied to McDonald. There was no evidence that the EKA MSDS had been supplied to Bryce or that a copy of it was in Australia at the time of the fire. Bryce's MSDS had been prepared by a third party, Kilford & Kilford Pty Ltd, in August 1997. Neither MSDS warned of the risk of Pinechem/Brytall undergoing spontaneous combustion. 36 The Bryce MSDS relevantly stated: "IV SAFE HANDLING INFORMATION Storage and Transport: No special requirements. This product does not have a UN classification. Containers should be kept closed in order to minimise contamination. Keep from extreme heat and open flames, and make sure the material does not come into contact with water or acids. Spills and disposals: In event of a major spill, alert fire brigade and advise of nature & location of spill. Prevent spillage from entering drains or water courses.

Stop leak if safe to do so, and contain spill. Absorb onto vermiculite, sand, sawdust or other absorbent material. Sweep up and shovel or collect recoverable product into labelled containers for recycling or salvage. Recycle containers wherever possible . After spills, wash area, preventing run off from entering drains. If material enters drains, advise emergency services . This material may be suitable for approved incineration or landfill. Dispose of only in compliance with local, state and federal regulations. Launder all contaminated clothing before re-use." (emphasis supplied) 37 The EKA MSDS (Blue 111) relevantly stated: "7. Handling and storage …

Fire and explosion protection: No special measures necessary …

10 Stability and reactivity: … Hazardous Reactions: Product is stable and no hazardous polymerisation will occur . … Conditions to Avoid: Avoid prolonged contact with porous materials." 38 Professor Gray said that polymerisation is part of the process of spontaneous combustion. 39 Attapulgite is a natural mineral which is mined in Western Australia. According to exhibit F, which was a document Hudson had provided to SLE, attapulgite: " Reduces Fire Hazards - It readily absorbs oil and dangerous chemicals, substantially reducing fire risks. Being inorganic and chemically inert, it is non-flammable and safe to use around machinery." 40 There was nothing of any significance in the attapulgite MSDS. Professor Gray said that attapulgite was incombustible. 41 McDonald had been in the transport business since 1972 and in the business of carrying dangerous goods since the early 1980s. He had never heard of the propensity of any kind of oil to undergo spontaneous combustion. McDonald said that he was unaware that Brytall was a dangerous substance because there was no notification on the container to that effect, that he did not know it was classified as dangerous goods, and that the notion of oil spontaneously combusting was completely unknown to him. No one put to McDonald that he should have known Brytall was classified as dangerous goods. No one cross-examined McDonald to suggest he should have known there was a risk of the mixture undergoing spontaneous combustion or that he should have rung Brytall or EKA to determine whether the course he was pursuing in cleaning up the spilt tall oil was safe. 42 McDonald said he did not ring the emergency number shown on the Brytall label because he was not aware that Brytall was a dangerous substance as there was no notification to that effect on the container. He appears to have been fortified in his view that there was no risk associated with the Brytall by the fact that Mr Emerson provided the attapulgite to him in order to clean up the oil spill.

43 Mr Underwood said that McDonald had cleaned up the tall oil appropriately. He also expressed the opinion that on the basis of the information available to McDonald there was no reason for him "to foresee that there would be a risk of any adverse interaction between the Brytall and the attapulgite (clay) either immediately or overtime, should the mixed substances be kept in storage". Mr Underwood said that he did not believe that people involved with the clean-up of goods would have anticipated a problem with using attapulgite to clean up the tall oil. 44 Mr Underwood drew attention to the Standards Australia Handbook, Dangerous Goods - Initial Emergency Response Guide, which provided immediate general information and advice for the handling of dangerous goods in emergency situations. It advised that an emergency response for a spill or a leak for highly flammable (low flashpoint) and flammable liquids was to absorb the spill with earth, sand or other non-combustible material. He said that that "strategy was followed in the Cardiff spillage incident". It might be noted that, according to Professor Gray, Brytall was not in the category of highly flammable or flammable liquids. In other words, Brytall was less dangerous than the goods referred to in the passage Mr Underwood quoted. 45 Mr Bashford, who was the managing director of SLE, had been involved in the business of selling industrial supplies including oil absorbent materials for about fifteen years. He had never heard of the phenomenon of oils combusting spontaneously. He had sold products for mopping up oil spills for about sixteen years. None of those products had, to his recollection, borne any warning about the dangers of spontaneous combustion. No instances in which spontaneous combustion had occurred had come to his attention. Mr Emerson, another director of SLE, who had worked in the area of industrial supplies since about 1980, was not aware of the propensity of organic oil when mixed with attapulgite to undergo spontaneous combustion. 46 Professor Gray said that before he became involved in this case he had no real idea what tall oil was or what it was used for, nor any particular knowledge of its propensity to undergo spontaneous combustion.

Causation (Grounds 1 - 5) 47 McDonald's challenge to the primary judge's conclusion on causation was made in the following grounds of appeal: "1. His Honour erred in finding that the fire was caused by the spontaneous combustion of the mixture of attapulgite and oil ("the mixture"). 2. His Honour ought to have found that the evidence before him, and his primary findings of fact, were capable of supporting an inference only that such spontaneous combustion was a possible, and not a probable, cause of the fire. 3. His Honour erred in finding in the absence of any probative evidence that the chances of the fire resulting from a cause independent of the mixture were very remote. 4. His Honour erred in finding in the absence of any probative evidence that the spontaneous combustion observed after the fire to have occurred in the mixture was more likely to have commenced prior to, and to have caused, the fire than to have occurred subsequent to, and to have been a result of the fire. 5. His Honour erred in adopting the conclusion of the expert witness Professor Gray that it was 'next to impossible' that the spontaneous combustion in container B was caused by heat from the fire (par. 14, 16) without forming an independent judgment of that matter based on the evidence". 48 In order to understand these grounds of appeal it is necessary to consider Professor Gray's evidence in detail.

Professor Gray's evidence 49 Professor Gray was the Emeritus Professor of Chemistry at Macquarie University. He had an impressive curriculum vitae demonstrating a longstanding interest in, and study of, combustion, ignition and explosion theory. His studies included the prevention of spontaneous ignition. He had undertaken consultations involving researching fire and/or explosion issues. No other witness with his expertise was called. 50 Professor Gray expressed the opinion, in essence, that: · the combination of tall oil with attapulgite created a mixture capable of spontaneous combustion; · his observations at the warehouse approximately two weeks after the fire were consistent with the hypothesis that the mixture had spontaneously combusted; · Tall oil had a propensity to undergo spontaneous combustion because of the presence in tall oil of linoleic acid; · Spontaneous ignition of the oil-soaked absorbent material was, with a very high degree of probability, the cause of the fire in question. 51 Mr Finch SC who appeared with Mr Braham for McDonald argued that Professor Gray's theory never rose above the level of possibility. In other words, it did not achieve the standard of probability required to enable Girkaid and Inghams to discharge their burden of proof on the issue of causation. 52 Professor Gray's opinion was based on experience, observations at the site, tests and research. Observations 53 Professor Gray inspected the scene of the fire on 16 February 1998. He set out the observations he made on that occasion in his report of 22 February 1998 (the "first report"). He used a diagram, apparently prepared by McDonald, to understand the distribution of chemicals in the warehouse before the fire occurred. 54 He found that the outside of Container A was "up against solid insulating non-flammable material". He examined the remaining material in Container A by making a clean cut through it. He said that, having regard to the proximity of the container to the insulating non-flammable material, external flame from the fire could not have impinged on the container to cause the combustion pattern he observed. He observed "darkening, due to combustion, in the interior of the attapulgite". Temperature readings in Container A read "close to ambient temperature" which meant, according to Professor Gray, "no significant thermal activity" was taking place. He concluded from his examination of Container A that its contents had undergone "subcritical spontaneous combustion" - a process which "takes place during the run up to fully fledged fire which occurs when the spontaneous combustion activity reaches the outside of the container usually on or close to the top". 55 Preliminary excavations in the area of Container B revealed what Professor Gray described as "easily identifiable visually as oil soaked attapulgite". Further excavation revealed "attapulgite which was showing signs of charring and darkening, an indication of oxidation of oil on the porous material". 56 A temperature reading in the "general vicinity" of Container B showed a temperature of 162ºC (324ºF). Professor Gray expressed the opinion, in his first report, that: "This kind of temperature existing no more than 30 - 35 cms above the ground level, which was awash with wet sludge from the extinguishing water and rain subsequent to the fire, indicated very clearly that chemical exothermic (heat generating) reaction must be taking place even though the fire had been 'extinguished' 2.5 weeks earlier." 57 Container B had been "almost completely destroyed". However, Professor Gray observed charring on container B which, he opined, was "more extensive on the inside … than on the outside". He was in no doubt that the contents of Container B had undergone spontaneous combustion i.e. had burned from the inside out. He observed the contents of container B to be in an "advanced stage of combustion inside a virtually uncharred container wall". There were also "localised variations in colour (and hence combustion degree of advancement)". He said the variations were "characteristic of spontaneous combustion in packed materials of this sort". He said his observations could not "be explained as the effect of a large scale fire occurring over a relatively short period" but were "strongly indicative of a relatively slow combustion process taking place over a period of days or weeks with a limited supply of oxygen". 58 Professor Gray concluded that Container A was "an excellent representation of Container B with a time lag of a day or two" and that "the existence of Container A showing unquestionably the presence of spontaneous combustion is a highly unusual piece of evidence which gives us a picture of what Container B must have been like a day or two before the fire". 59 According to Professor Gray the time taken for spontaneous combustion to reach the outside of a container and thus cause overt fire measured from the time of packing depended upon a variety of factors including the average ambient temperature during the period, the size and shape of the container and the density and packing of the materials. He said that "generally most materials of this type would be expected to ignite after a period of days rather than hours". Accordingly, he expressed the opinion that "the time elapsed between the oil spill and the fire is entirely consistent with the cause being spontaneous combustion of a body (container) of around 1 cubic metre in volume such as the Brytall containers were". 60 Professor Gray said one of the "major components" of tall oil was linoleic acid, "[an] unsaturated fatty acid … renowned for its propensity to undergo spontaneous combustion when distributed on any kind of porous material, in particular in this case attapulgite". He said that linoleic acid was a major component of linseed oil "which itself has been responsible for many major fire losses due to its liability to undergo spontaneous combustion when distributed in such a way as to allow easy access for air (due to porosity)". 61 Professor Gray referred to the Bryce MSDS and noted it did not mention any possible hazard due to spontaneous combustion. Referring to the recommendation that spills should be absorbed onto absorbent material, swept up and collected in labelled containers for recycling or salvage, he said that in cleaning up the spill of Brytall with attapulgite, McDonald had followed the instructions in the Bryce MSDS or at "least behaved in a way entirely consistent with them". He expressed the view that such behaviour had "resulted in spontaneous ignition of the oil-soaked absorbent material which was, with a very high degree of probability, the cause of the fire in question". 62 Professor Gray referred to the statement in the EKA MSDS, "[a]void prolonged contact with porous materials". He said that it referred to the possibility of spontaneous combustion. He opined that if McDonald had followed that instruction "the spontaneous combustion and subsequent fire would not have occurred." 63 Professor Gray concluded: "With an extremely high degree of probability the cause of the fire of January 29th at Macdonald (sic) Transport was spontaneous combustion of "Brytall" oil absorbed on "Attapulgite" absorbent material and the fact that it was left to stand in approximately 1 cubic metre containers for a prolonged period during a period of hot weather. Extremely strong evidence for this is summarised as 1- The physical condition of container "B" and the fact that it was still undergoing spontaneous combustion more than two weeks after the fire. 2- The physical condition of container "A" which was less well insulated and was in the early stages of spontaneous combustion at the time of the fire. 3- The presence of a known spontaneous combustion prone material (linoleic acid) in the oil which was spilled six days before the fire. 4- The absorption of this oil on a highly porous material and subsequent storage in a condition known to be highly favourable to spontaneous combustion of such products. 5- The time period elapsed between the spill and the fire, which is in the region expected for spontaneous combustion to develop in one cubic metre of such material. 6- The warning in the MSDS published by Eka Chemicals with respect to prolonged contact of the oil with porous materials - this indicates that it is highly probable that the propensity for spontaneous combustion under these conditions was well known in the manufacturing industry for this product." (emphasis supplied)

Laboratory tests 64 On 9 March 1998 Professor Gray produced a supplementary report (the "second report") which dealt (inter alia) with a test he undertook with a mixture of Brytall and attapulgite in order to detect the mixture's ability to undergo spontaneous combustion. He said the test was intended as "quantification of the process" as "there is no question at all that [spontaneous combustion] occurs". 65 The test involved packing a container with attapulgite onto which Professor Gray had absorbed a quantity of Brytall, allowing it to stand with the oil absorbed overnight and then placing it in an oven heated to 120ºC. Professor Gray said that the test mixture reached 120ºC within a little over an hour and self-heated another 113º to 223ºC four hours later. The oven temperature was maintained at 120ºC throughout the experiment which was conducted over 30 hours. The temperature of the test mixture did not rise above 223ºC. 66 According to Professor Gray subsequent observations of the mixture revealed "interior darkening" which "indicat[ed] a considerable degree of self-heating above oven temperature". Professor Gray was struck by the similarity between the appearance of the tall oil/attapulgite in the test container after the test and the contents of containers A and B. 67 Professor Gray also undertook gas chromatographic tests of the tall oil which he said in the second report identified it as "largely linoleic acid". 68 Professor Gray referred again to the Bryce and EKA MSDSs in his second report. He opined that the statement "none" in the Bryce MSDS under the heading "Unusual fire and explosion hazards" (sic - the word "unusual" was not in the relevant heading) was "positively misleading when the material contains a significant proportion of linoleic acid". 69 Professor Gray concluded that the matters in his second report reinforced his conclusion that the fire was started by spontaneous combustion taking place in container B. He expressed the opinion that "there is a probability of 80% that this was the fire cause". 70 Professor Gray said that in order to raise the probability closer to 100% it would be necessary to run: "… a complete set of scale tests on a number of containers of various sizes in order to do the standard extrapolation to a 1 cubic metre container such as I observed in the warehouse after the fire. These tests would determine the lowest ambient temperature at which such a large container would undergo spontaneous combustion of sufficient intensity to ignite the cardboard container in which the material was housed. This investigation would be unusual in so far as the porous material on which the oil is dispersed is not itself flammable, i.e. it is simply clay. In standard investigations of eg. linseed oil disbursed on cotton rags, the rags themselves are flammable and ignition tends to occur at the centre of the container where the temperature is the highest. However in this case the highest temperature will occur in a region where there is NO flammable material. It would be necessary for the edge temperature (quite significantly lower than the centre) to be high enough to ignite the cardboard container, thus allowing ingress of fresh oxygen and the appearance of flame on the outside of the container. This kind of investigation has not to my knowledge been carried out." 71 Professor Gray's third report dated 9 September 1999 commented on what I infer were 2 expert reports served by other parties. He noted that the purpose of the test referred to in his second report was: "… purely to show that in practice the combination of materials in question, i.e. tall oil absorbed onto Attapulgite, can self heat significantly and therefore under some conditions self ignite . They were not designed to obtain the critical temperature for the life size drum" (emphasis added). 72 Professor Gray said that tests could be undertaken to determine the ambient temperature at which ignition would be brought about to determine, as I understand his third report, the "critical ambient temperature for the keg" (referring to Container B). He annexed a publication entitled "Spontaneous Ignition Hazards in Stockpiles of Cellulosic Materials: Criteria for Safe Storage" said to demonstrate that testing procedure. Professor Gray was never asked to perform the tests set out in that annexure, nor any tests which would demonstrate under what conditions the self heating reaction in a tall oil/attapulgite mixture would lead to self ignition - although such tests were possible. 73 Professor Gray said: "The (admittedly incomplete) evidence in my report [this appears to be a reference to his second report] is not meant to be obtained as 'a simulation of service conditions'. It is evidence obtained under controlled laboratory conditions which together with the very well understood theory of spontaneous combustion has very strong implications as to what to expect under service conditions . That was as far as my instructions allowed me to go ." (underlining added) 74 In his fourth report dated 19 December 2000, Professor Gray set out the results of an analysis of Brytall which showed that "[t]he polyunsaturated fraction consisted largely of linoleic acid and the unsaturated fraction was largely oleic acid. The concentration of oleic acid was about 150% that of linoleic acid". Mr Faulkner SC who appeared for Girkaid and Inghams agreed that that meant linoleic acid constituted about 6% of the Brytall. The significance of this percentage compared to Professor Gray's earlier statement that the tall oil was "largely linoleic acid" was not explored at the trial. 75 Professor Gray noted that tall oil was not a "well defined chemical compound but a mixture (to some extent dependent on origin)". There was no doubt, according to him, of the "substantial presence of the spontaneous combustion prone compounds oleic and linoleic acids". 76 He then noted that linoleic acid was a major component of linseed oil. He said it was one of the components of linseed oil which was responsible for that oil's propensity "to spontaneously ignite when suitable (sic) disbursed on porous material". Professor Gray then opined that "all references to the spontaneous combustion of linseed oil are indirect references to the proneness of linoleic (and to a lesser extent oleic) acids to spontaneously ignite" (emphasis supplied). 77 Professor Gray attached to his fourth report what he described as a "typical MSDS for linseed oil". That document stated under the heading "Stability and reactivity": "Materials to avoid: porous materials will absorb the oil and can cause spontaneous combustion." 78 The linseed oil MSDS also showed that its constituent ingredients included oleic acid (20 - 30%), linoleic acid (15 - 20%) and linolenic acid (45 - 55%). 79 Professor Gray gave a number of references to literature concerning the spontaneous combustion of linseed oil which, he said, dated back to the nineteenth century. In what I understand to be a reference to a monograph dealing with "Self-heating: evaluating and controlling the hazards" by a P C Bowes, Professor Gray noted that the author referred to the oxidisation rate which is responsible for the self heating and combustion "of linoleic and oleic acids themselves individually" and then said "[l]inoleic acid has been shown to oxidise twelve times more rapidly than oleic acid". 80 Professor Gray said that he had investigated seven incidents involving the spontaneous combustion of "linoleic acid (in combination with other substances)" in the past few years. He gave no details of those investigations. 81 Professor Gray also attached to his fourth report an MSDS and a technical bulletin relating to products containing tall oil fatty acids. Both documents refer to the possibility of spontaneous combustion in relation to materials soaked with the product. The first attachment, which related to a carbon remover, referred to spontaneous combustion in relation to "combustible materials soaked with this product". The second, which related to a zone marking paint, appeared also to refer to combustible materials when it referred to avoiding spontaneous combustion by the safe disposition of "used rags…". Professor Gray identified these documents after undertaking a web search using the words "tall oil spontaneous combustion". 82 Professor Gray suggested that the manufacturers of the products referred to in those two attachments had made "the direct connection with tall oil products and spontaneous combustion … without necessarily identifying components common to linseed and tall oils."

Professor Gray's oral testimony 83 Professor Gray's evidence-in-chief was short and added little to his written reports. His principal oral testimony was given under cross-examination. 84 Professor Gray said he was "virtually certain" that spontaneous combustion occurred in Container B. He accepted, however, that the questionable area was whether that spontaneous combustion caused the fire. 85 Professor Gray said it was possible to design and conduct experiments which so constructed the mixture and the temperature and the time and the volume that one could extrapolate from the test to an actual event to see whether or not the actual event was likely or not to have involved self heating. He said he had not done such an experiment in this case in the absence of instructions. He also said there was no reason, save for instructions, why a large-scale test which would give rise to close to incontrovertible results could not be done. 86 Mr Finch SC cross-examined Professor Gray about the significance of his laboratory test in the following exchange: "[Mr Finch SC]: Q. I thought you had agreed … that the experiment was designed to and did tell us only about the theoretical possibility of a mixture of oil and Attapulgite commencing to self-heat? A. No, not the theoretical possibility. As a result of my experiment it becomes a matter of fact that a mixture can produce. Q. When heated? A. No. Q. Your experiment was heated? A. Yes. Q. And you have done no extrapolation or other process of reasoning to discover whether or not such a mixture or indeed any mixture could involve a self-heating reaction at what we might call ambient temperatures? A. I don't need to. It is a basic law of physical chemistry that that would be so. Q. But it would depend, would it not, on the volume and the mixture? A. No. Q. What would it depend on? A. It depends, as I said, on the basic laws of physical chemistry and later chemical reactions. If you establish that a chemical reaction goes at a particular rate at a particular temperature and produces heat, then at any other temperature that reaction will go at a different rate but it will go at a finite rate provided the ambient temperature is above the absolute zero. Q. There is no doubt, is there, that there are ambient temperatures below which this mixture would not engage in self-heating? A. Are we using self-heating in the literal sense - generating heat. Q. Just generating heat? A. No, there is not. Q. So at any temperature it will in fact generate heat? A. At any finite temperature it will generate heat. The lower the temperature, the smaller the rate - at any lower temperature, the rate of generation of heat will be lower. It will always be greater than zero. Q. At a number of different ambient temperatures, we may see different outcomes, that is whether or not the self-heating progresses to spontaneous combustion or not? A. Yes. Q. Your experiment was not designed to show and did not show whether or not at the ambient temperature in the warehouse this mixture was likely to have proceeded to spontaneous combustion? A. No, it was possible." 87 He returned to the utility of his experiment in a later passage as follows: "[Professor Gray] … the experiment that I described shows that Brytall with Attapulgite is capable of self-heating….Now it follows from basic physical chemical principles that the material which can be shown to be capable of self-heating by chemical reaction can under some conditions, we don't need to define what the conditions are, but there exist conditions in terms of the parameters outlined by counsel yesterday under which that self-heating can cause spontaneous ignition … [I]t follows that my experiment shows that at temperatures higher than 120 degrees there will be significant self-heating and heat generated by this material in the bins. I am not trying to say that my experiment proves that under the conditions in the warehouse spontaneous ignition could have occurred but simply once we have agreed that this material is a self-heater capable of self-heating by chemical reaction, it follows from absolute basic principles of chemistry that there exist conditions under which samples of Attapulgite and tall oil can undergo spontaneous ignition. HIS HONOUR: Q. Even if there is no external source of heat aside from ordinary ambient temperature which might be 120 degrees or so? A. Yes. Q. But absent the fire caused elsewhere - well, your view is indeed that spontaneous ignition caused this fire? A. Yes, yes, that's true …" (emphasis supplied) 88 Professor Gray was asked whether the most likely explanation of what was seen in the containers was that it was induced by the fire which took place in the warehouse. He resisted that proposition by explaining what Mr Finch SC described as the "deflagration wave theory". As explained by Mr Finch SC on appeal that theory, in essence, was that if another fire had enveloped the container in which the mixture had been deposited a deflagration wave would occur which involved the outside skin of the container burning first, setting up a temperature gradient sufficient to burn from the outside into the container and eventually consuming the entirety of the container (the "onion skin effect"). Professor Gray rejected the "other fire" theory, because he did not observe the onion skin effect on either container A or B. 89 Professor Gray explained the deflagration wave theory in the following way (omitting what appear to me to be extraneous passages): "[Professor Gray] … [L]et's take a mental picture of bin B or bin A. The material that was placed in it and scraped off the floor of the warehouse probably the floor, the ground temperature even in Cardiff in January or February are not very high. They are usually around the low 20s or thereabouts so we can assume that the initial temperature of the stuff was around the low 20 mark if that's agreeable. The rate of heat generation by this established thermogenetic reaction would be rather small at that temperature. Now, the fire starts from some other source. It is very intense. There was a lot of oxygen producing material. There was hydrogen peroxide and Calcium hypochlorite. These materials produce pure oxygen. We could have a very hot fire so we picture a bin of reactive material, potentially reactive material, which can generate heat we have already agreed. The exterior of the bin is subjected to 500 degrees centigrade for the sake of argument. That's a reasonable figure I think so in a short time the thin exterior level of the material very quickly will reach 500 degrees centigrade … So the very first thing that happens is that the outer layer is heated very quickly. We are only talking about an infinitesimal thin layer of reactive material. It is suddenly heated from [sic, "from" was probably "to"] 500 degrees to from 20, maybe 25 degrees. … … if we make the assumption that the temperature coefficient is in the same ball park as related reactions then it is a simple calculation to show that the ratio of the heat generation rates in the outer thin annulus at 500 degrees centigrade and the rest of the material which is at 20 degrees centigrade is one billion times approximately, maybe 100 million billion, something like that range, huge, so in a very small fraction of a second all the combustible material in that outer layer will be consumed because the reaction rate is very very rapid. Once you hit it with a very high temperature, you are then locked into very high reaction rates so all the oil in your outer layer is consumed. It is consumed in a way which is going to produce some heat because it is a thermo generating reaction, heat-generating reaction. So the temperature increase resulting from that driven reaction now will be of the order of 300 degrees to completely combust the material in this layer. You are going to get a temperature rise of about 300 because I obtained a temperature rise of 113 as a combustible temperature. … [I]t can be quantified so in the first microsecond of the temperature of the hot fire hitting the outer of the bin, the temperature of the outer sliver becomes 800 degrees centigrade. There is no fuel left there. Everything has gone black so we have a very thin sliver at 800 degrees centigrade and next to it we have got very nearly still 25 degrees or 20 degrees. … One can use the heat conduction equation, given that knowledge which I have just outlined, and one can calculate what would be the approximate temperature in the centre after four hours assuming no chemical reaction. It gets its inflammation purely by heat conduction and the answer is about 20 degrees above so it would get up to about 40 degrees only centigrade, even though it has been exposed externally to 500 degrees for four hours … The other side of that coin which is ultimately what demolishes this alternative theory is that the entry of heat to significantly affect a large body of this material is also very slow if it is purely by heat conduction. … …[T]he reality of this situation is that this is reactive material. It is not simply a burnt cake. It is reactive material and at 800 degrees it is very reactive material. I have just indicated the rate of reaction which is a rate of heat generation itself so at 500 degrees and then the rate of reaction produces heat which heats it to the order of 800 degrees and in so doing consumes all of the oil in that sliver. This reaction thereby produces a very large temperature gradient within the material. In the edge of the material in two directions it produces a very sharp spike of temperature close to the outside of the material so down the outside of the spike. … …There is a large temperature spike generated which has a peak of 800 degrees. Three is a large gradient towards the outside of a factor of 500 so heat would be lost down that and there is an even larger gradient and it can be shown on the inside of the material. This is a very thin spike so as heat is conducted down that spike, down the gradient according to Fourier's law. FINCH: Q. Towards the centre? A. Towards the centre. … Q. Yes? A. There would be reactive material in the next sliver. If I can ask you to imagine this bin being divided into thin sections. … FINCH: Q. Perhaps it is like an onion with different layers of the skin? A. Yes. Q. You have already dealt with the outside onion skin layer? A. Yes.

Q. Now we come to the second layer? A. It suffers the same fate but a little more quickly because on its outside it is now 800 degrees not 500 degrees. It has heated up very quickly. It dumps its heat and is duly consumed and it adds to the temperature, and produces an even steeper gradient on both sides so more heat goes out and more comes in and this goes on through the layers until in fact what happens is the gradient steepens on the way in until there is a balance point reached where the gradient gets to be so steep that the rate of conduction of heat down the gradient, that's into the body of the material, the rate of generation of heat by the fuel and I can call it fuel, in this sliver and the rate of loss of heat down the other end of the gradient towards the outside, they all reach a balance and what is well-known and well understood to occur is that this temperature profile, this spike, assumes a stationary shape but is moving in through the material. It is called a deflagration wave. … Q. Just before you go on, what do each of the onion skin layers look like after this process has happened? Do they look blackened? A. There's no fuel left so they would almost certainly look blackened. … The net result of this is that this wave travels inwards, the combustion wave or deflagration travels inwards from the faces on which the 500 degrees has impinged until they meet other waves coming from other sides. There is no fuel left so that's it. The whole fuel contents of the bin has gone. … [W]hat I am getting at is that where a material of this sort that suffered deflagration waves, these are not spontaneous ignitions. This is a proven deflagration wave caused by the high temperature. The aftermath of that is absolute devastation. There is no fuel left. You can't expose a combustible reactable material to an 800 degree temperature moving in without removing all the fuel. There's nothing left so after this has finished and the fire has been put out it will be stone cold dead on February 15 when I went up to measure it. There is a reason and maybe it is too elementary but I will see. The reason for this is that there is a conservation principle at work here which is conservation of energy which is a basic law of physics. This says that the amount of energy that can be released from a given amount of fuel is fixed. If you release that energy very quickly at a high temperature, you can't maintain that for a very long time. If you release it relatively slowly at a relatively low temperature, you can maintain the reaction for weeks. HIS HONOUR: … And you say therefore, forgive me if I am cutting you off and elaborate if you need to, the only explanation for the fact that there was still ignition seen by you or it was still very hot when seen by you was that it had to be a slower reaction and it therefore could not have been a reaction caused by the fire? A. No, it would have been consumed far too quickly." 90 Mr Finch SC put to Professor Gray that his deflagration wave theory could not be valid because, despite there having been "another fire" (the fire which consumed the warehouse), the onion skin effect was not observed on the containers. Professor Gray gave what the primary judge referred to (at [14]) as "a lengthy explanation, which I need not set out here, but which seemed to me (as a lay person) to be convincing, … [indicating] why [the professor] thought that this was not only unlikely but next to impossible." Professor Gray's response, in short, was that "getting into discussions of the way the fire develops, once it has developed in terms of fuel load and movement of air is extremely precarious and speculative". While he was not prepared to speculate, an explanation was that the fire was not of sufficient intensity to create a destructive deflagration wave in the area of the containers. 91 Professor Gray also sought to explain the reason the onion skin effect was not observed on container B by reference to there having been an intermediate position. He said: "HIS HONOUR: Q. Another question occurs to me: Assume that the fire commenced because of some spontaneous combustion in the mixture. As I understand you to say, that fire would have eventually resulted in temperatures of the order of 500 degrees. FINCH: At the edge of the material, I presume? HIS HONOUR: No. I am saying that you have spontaneous combustion which starts to consume B but moves elsewhere because other parts were burned - Q. Is that right? A. Yes. Q. And so it has other fuel available? A. Yes. Q. Overall then the temperature would or could well, as I understood you to say, reach 500 degrees, perhaps even more? A. Yes. That was a postulation made for the purpose of the illustration. Q. Certainly but that postulation doesn't deal with where the fire started. Providing other material than the fuel in the bins was burned, and we know that occurred, we may see ambient temperatures in the area of 500 degrees. Now whether they are ever 500 degrees in B I understand is a different question? A. Yes. Q. And I understand you to say that you do not believe that in the vicinity of B or for that matter A, the ambient temperature ever reached anything like 500 degrees. Is that right? A. If I could elaborate on that? In a specific fire with normal fuel, of course the actual flame itself will flicker all over the place, but in the actual flame itself you can get temperatures of 500 or 600 or maybe 700 degrees. That would not be in the same place for a long time because at that temperature fuel in that particular spot is consumed quite quickly. Now in this case there is the possibility that part of the time the localised temperatures could have been higher than that because there were oxygen pockets in the warehouse and then you get much higher temperatures than ordinary air so the flame could reach 800 or 900 degrees. HIS HONOUR: Q. Where there is draught? A. Where there is draught and where there is oxygen enrichment from carbon monoxide or where oxygen material is present. So typically temperatures in these sorts of situations would be in the area of 500 to 700 degrees but the average temperature would be somewhere around that figure. Q. Does it follow, however, that at raised temperatures that is, shall we say, about 120 degrees centigrade but short of, shall we say, 500 degrees, in the vicinity of say the B bin, one would still get or one would get an enhanced degree of chemical reaction causing the combustion, so it would now reach the stage of creating what you mentioned? A. Yes. Q. So what you would then be left with is a larger scale than the material in the oven but you would be left with the same phenomenon, burning the material on the inside and relatively unaffected material on the outside so that an acceleration effect caused by the increased temperature of the material would probably increase the ambient temperature? A. I think there is an intermediate situation between fully fledged deflagration wave and the material reacting at a constant temperature with no help from the outside. If you are asking me to comment on an intermediate position where you have got material which may be doing something in the centre and being subjected on the outside to maybe 250 degrees for a short time, that is not enough to start a deflagration wave. FINCH: Q. Not a high enough temperature? A. Well, it is a combination of the two. I am trying to put things in a qualitative way, which is difficult to do, but one thing that could happen is that simply, and I think it is quite likely given the scenario of 250 degrees for 15 minutes, for example, one scenario is that the spontaneous combustion occurs and the outer layer is slightly consumed because it hasn't got the fully fledged deflagration wave and because there is no deflagration wave set up, it is not sufficient and the inside doesn't know anything about what is going on on the outside, so you have completely separate phenomena. … HIS HONOUR: Q. I think I have added enough variations to an already complicated picture but there is an additional complication, isn't there, because we don't know and in fact it is almost certain that the material is not monogamous? [sic, his Honour presumably said "homogenous"] A. No. Q. And certainly we can't know whether it was a mixture similar to the mixture you had in the oven, which is another complication, and we can't talk about rates except of an order of magnitude which you have referred to? A. Yes. 92 I set out a summary of some other passages of Professor Gray's evidence. He accepted: (a) there would be a range of ratios of oil to attapulgite within which spontaneous combustion could take place and, equally, there would be combinations at either end of that range i.e. too dry or too wet where spontaneous ignition would not take place. He had made no analysis of where, along the spectrum of possible ratios, the mixture derived from McDonald's cleanup process fitted.

(b) that his tests to see whether a mixture of attapulgite and oil would self-heat did not attempt to match the composition of the test mixture with the composition of the mixture present in the containers on the day of the fire.

(c) that his laboratory test was designed to show the possibility of tall oil and attapulgite initiating spontaneous combustion.

(d) that it was impossible to test whether the mixture at the warehouse was in a spontaneous combustion range because the composition of the material in the container could not be defined, as the amount of oil that was throughout the container was, or may be, highly variable.

(e) that the presence of a residual 200 litres of oil in one of the containers into which the tall oil/attapulgite mixture was introduced would have important ramifications for the degree to which the mixture could be likely to initiate any spontaneous combustion reaction but that he did not know what the ramifications were of that circumstance.

(f) that he was unable to say whether the charring in Container B had occurred before or during the fire or when the self-heating he observed had commenced.

(g) that the test referred to in his second report was designed to determine whether the "theoretical possibility of self-heating" could occur in a mixture of attapulgite and oil. It was not designed to, and did not, tell whether it was more likely than not that self-heating occurred in the warehouse.

(h) that there was a strong similarity between the visual state of the attapulgite and oil which had undergone "what was undoubtedly spontaneous ignition in my test" and the contents of bin B.

(i) that he did not conduct any test which resulted in the mixture of tall oil and attapulgite igniting. He was not asked to do a "cause investigation". His tests were only designed to show the possibility of the mixture self-heating. (j) that he could not say that the area where Container B was, was the area most affected by fire because he was "not asked to do a detailed investigation". He did not profess to have any opinion on the likely development of the fire.

(k) that McDonald had not done anything contrary to either the Bryce MSDS or the EKA MSDS.

(l) that the statement in his first report that the entry in the EKA MSDS "avoid prolonged contact with porous materials" referred to the possibility of spontaneous combustion if the oil was introduced to porous material was speculation on his part, but that he could not think of any other possible reason.

(m) that the test he carried out on tall oil proved it could self-heat but did not attempt to quantify its propensity to undergo that reaction.

(n) that the literature he had on linseed oil did not assist him in determining the propensity of tall oil to self-heat.

(o) that, speaking of the entry in the EKA MSDS under the heading "Stability and reactivity", "Hazardous reactions: The product is stable and no hazardous polymerisation will occur", the "reaction which is taking place when spontaneous ignition is going on is part polymerisation and part oxidation … so it would be correct to say that the spontaneous ignition reaction involves polymerising".

(p) that linolenic acid which was the primary cause of linseed oil's propensity to spontaneously combust was more reactive than linoleic acid which was present in tall oil. 93 Finally, I note that Professor Gray said that "[f]lash point has got absolutely nothing to do with self ignition".

The primary judge's resolution of the causation issue 94 His Honour approached the resolution of the challenge to Professor Gray's opinion in the following way: "8 The scientific validity of Professor Gray's conclusion that the mixture in the container he described as "B" had indeed self-heated was not really in contention, though whether it had done so to the point of ignition spontaneously or because of the heat from the fire otherwise caused was, it was argued, speculative and, in the end, unproved. An examination of container B showed that it had been considerably charred on the inside, more extensively than on the outside, and completely destroyed further up. The other container showed less combustion effects but self-heating had obviously taken place with darkening due to combustion in the interior of the mixture, although there was no significant thermal activity when the inspection occurred. Professor Gray considered that the time lapse between the oil spill and the fire was entirely consistent with the cause of the fire being spontaneous combustion of a body of about the same volume as the Brytall containers were. 9 It was not contested that two of the main components of Brytall are linoleic and oleic acids. These are unsaturated fatty acids with a propensity to undergo spontaneous combustion when distributed on any kind of porous material such as, in particular, attapulgite. (I use the term "spontaneous combustion" as meaning self-heating to the point of spontaneous ignition.) Professor Gray undertook standard tests to examine the liability of tall oil to undergo spontaneous combustion. This involved packing a container with attapulgite, which had absorbed a particular quantity of Brytall . This was then placed in an oven and raised to an ambient temperature of 120 degrees Centigrade. He said that a considerable degree of self-heating above oven temperature was demonstrated. Whether it was reasonable to infer, however, that the mixture in the containers packed by Mr McDonald achieved a temperature sufficient to cause fire is a somewhat more difficult question to resolve and was the subject of extensive cross-examination. Professor Gray agreed that the degree of the wetness of the mixture of attapulgite and tall oil (to mention but one variable, though probably the most significant) had a significant bearing on its tendency to undergo spontaneous combustion and, of course, he did not suggest that his experiment had replicated the proportions of tall oil and attapulgite in the containers. For obvious reasons, this was impossible. On the other hand, it seems that there was quite a wide range of proportions within which spontaneous ignition could take place. Whether the mixture in the containers, more particularly container B, fell within that range had not been examined by Professor Gray. As I have already mentioned, perhaps as much as 200 litres of oil was left in the broken container to which the mixture was added as the cleanup proceeded. Professor Gray agreed that this might be significant as to the likelihood of spontaneous combustion occurring (on the assumption that it was effectively available to dilute the mixture) but, as I understand his evidence, the true significance of this fact could not be assessed. On the other hand, the overwhelming likelihood is, as I understand Mr McDonald's evidence, the unspilled oil was separated from the mixture by being in its own bladder whilst the mixture was placed in a separate bladder on top of the waste oil and, hence, the waste oil was not absorbed by any of the attapulgite. 10 As a result either of the fire or the steps taken to extinguish it, a large amount of inert, inorganic carbonate (called L7) had fallen on top of container B. A substantial quantity of this product had been stored nearby but when and how it had collapsed onto the container is unknown. Professor Gray said that container B was up against what was, in effect, a wall of insulating material around approximately two-thirds of its circumference and that this material would have inhibited the dissemination of heat from the self-heating process and accordingly encourage the development of higher temperatures. 11 As I have mentioned, the containers were constructed of fibreboard and Professor Gray assumed that it would probably require something of the order of 300 degrees Centigrade to ignite it. However, whether the fibreboard could be or was ignited by contact with the self-heating mixture, Professor Gray was unable to say. Even so, he did see what he described as heavy charring from the inside of the container, which indicated to him that it was likely that the self-heating had got to the stage of ignition which had then destroyed the balance of the container. He inspected the container on 16 February and was unable to say when the charring had occurred, in particular, whether it had occurred before or at the time of the fire. So far as the mechanism itself is concerned, Professor Gray described it as a "hotspot" caused by the self-heating process moving through the material and, when it reaches the surface it gains access to an increased amount of oxygen so that usually overt flame appears within a short time, the oil itself spread on he outside surface of the aggregate being the fuel. There is a convective buoyancy effect, which would tend to cause the hotspot to rise so that if there was no lid on top of the container (as in this case) the oil on top might be expected to burn. Once overt flame has occurred, air is drawn up from the sides and the flame can grow quite quickly and reaches many hundreds of degrees within a short time. I think that it is likely that waste paper material from the used bags of attapulgite was placed on the top of the mixture: it was all rubbish that needed to be disposed of in due course. This, of course, would provide additional fuel. 12 Professor Gray considered that the material in container A, though self-heating, had not sustained the process to the point of ignition but he rejected the suggestion by counsel that the difference between it and container B was that the latter had been exposed for a longer period to a higher temperature because of its proximity to a fire initiated from another source. Professor Gray considered that that was inconsistent with his observations of the contents of container B. Professor Gray also thought it significant that the material in container A had been self-heating but not to the point of spontaneous ignition as a reference or useful comparison with what happened to container A. Indeed, so far as the latter is concerned, he believed that the probability of spontaneous ignition of the material in container B was virtually certain. He accepted, of course, that further conditions had to be present for that ignition to have caused the fire that demolished the warehouse, in particular, that some mode of transfer to transport of the flame itself to other combustible material had to be present. 13 Whilst Professor Gray stated in his report that the probability that this ignition caused the fire was 80%, he did so, in part, by applying what is known as Ockham's razor, the name given to the principle enunciated by William of Ockham, a late English medieval scholastic (c 1285 -c 1349) that "non sunt multiplicanda entia praeter necessitatem" ie., entities are not to be multiplied beyond necessity. This principle is a powerful scientific tool and underlies much of modern science. However, whether it may be usefully invoked for the proof of facts in an action in tort, as was sought to be controverted by Mr Finch of Senior Counsel for Mr McDonald, is not an easy question. Fortunately, it is not one that I need to answer in this case. I will return to this matter later in the judgment. 14 It was suggested to Professor Gray by Mr Finch SC that the spontaneous combustion demonstrated in container B may have been initiated or, at least, developed to the point of ignition by the extreme heat of a fire that had started elsewhere. In a lengthy explanation, which I need not set out here, but which seemed to me (as a lay person) to be convincing, the Professor indicated why he thought that this was not only unlikely but next to impossible. Professor Gray emphatically disagreed with the suggestion that what he saw when he inspected the premises and, in particular, container B, was entirely consistent with there having been a self-heating reaction in that container which commenced during the fire but had not resulted in ignition until sometime later, in the result, demonstrating the particular features that he saw two and a half weeks after the fire. 15 Professor Gray was an impressive witness. His curriculum vitae demonstrates considerable expertise both by way of research and fire and explosion investigation. He has a profound theoretical and practical understanding of the cause and character of spontaneous combustion. I thought he gave his evidence forthrightly and objectively. He was not an advocate for the plaintiffs and gave his evidence in a way that was calculated to assist the Court to consider and determine the issues upon which he had been consulted. 16 The principal argument constructed by the defendants is that, whilst it may well be that spontaneous combustion occurred, the plaintiffs had not established that this occurred prior to and was a cause of the fire. This argument depends on a number of aspects of Professor Gray's evidence but, in particular, his positing of what he described as "an intermediate position" capable of explaining the appearance of the material in container B as the result of self heating, perhaps assisted by rising ambient temperature caused by the fire, but not having reached the point of ignition until after the fire had commenced. I think it is clear on a consideration of the whole of Professor Gray's evidence that he regarded this explanation as being distinctly less successful in explaining the appearance of the material as he saw it than the hypothesis that, indeed, the self-heating of the material had proceeded to the point of ignition prior to and was the cause of the fire. 17 Whilst to some degree this is an over-simplification, I consider that there are five crucial factors which, taken together, point to the probability that spontaneous combustion of the mixture in container B was the cause of the fire. The first of these, of course, is that a mixture of tall oil and attapulgite will self-heat in a mixture such as that in container B to the point of ignition; second, the mixture in container B in fact self heated to the point of ignition; third, there was paper and other similar material, probably soaked to some extent with oil, which was left on top of container B and capable of providing a significant amount of fuel; fourth, the difference between container A and container B strongly suggests that the rising ambient temperature of the fire was not a significant factor in causing self heating in the containers; and, fifth, the chances that the fire resulted from some other independent cause seem to me, as a matter of common sense, to be very remote. There is, of course, no complete description of the contents of the Inghams' premises but their use as described in the evidence does not suggest any likely cause. There was evidence that rats had gnawed vehicle wiring and it was submitted that rats gnawing at the electrical wiring that powered the roller doors and the refrigerator (the only live circuits) had caused some electrical fault that led to the fire. I think that this is most unlikely. There was no evidence that rats had been attracted to this wiring, where the wiring was, whether any short circuit was in proximity to flammable material and no evidence that gave the proposed mechanism any support. I am certainly not prepared to assume that rats gnawing on the live wiring is a possible cause of this fire. To the contrary, I think the theory is most unlikely. The fifth consideration mentioned above is, for obvious reasons, not insignificant, but I reject the submission that, in so considering it, I have reversed the onus of proof. Nor have I had recourse to Ockham's razor. 18 I am therefore of the opinion that the plaintiffs have satisfied the onus of establishing on the balance of probabilities that the fire that destroyed the warehouse and its contents was caused by the spontaneous combustion of the mixture in container B."

Causation: McDonald's submissions 95 Mr Finch SC submitted that Professor Gray made significant concessions which cast doubt on the weight of his opinion. He contended that Professor Gray was only able to initiate self-heating of Brytall in his tests by the application of high temperatures. He submitted that there was no evidence that Brytall or any tall oil had initiated self-heating at room temperatures. He pointed out that Professor Gray's tests were conducted on mixtures of attapulgite and Brytall in an oven raised to a temperature of 120°C, conditions more in keeping with those which existed during the fire than those in the days prior to the fire. 96 McDonald's written submissions argued that the "five crucial factors" identified by the primary judge in par [17] of the judgment were flawed. They contended: "Five factors are identified which taken together are said to lead to the conclusion that spontaneous combustion caused the fire. Each of the first three factors is equally consistent with [McDonald's] hypothesis that the fire caused or contributed to the self-heating. The fourth has no basis, and is unreasoned, and the fifth has no foundation in the evidence and is impermissible speculation. [Girkaid and Inghams], who owned and occupied the adjoining premises, led no evidence to rebut the possibility that the fire commenced in the Inghams warehouse. Nor was any evidence presented to support a conclusion that there were no other possible causes of fire in McDonald's warehouse. [Girkaid and Inghams] adduced no evidence of the likely seat of the fire. Furthermore [McDonald] contends that his Honour's finding on this issue were not reasonably open on the evidence. His Honour failed sufficiently to take into account the important matters that Professor Gray conceded limited the scope of his evidence on this aspect: a. He was unable to say whether the charring in container B had occurred before or during the fire. b. He was unable to say when the self-heating he observed commenced. c. He was not asked to do a "cause investigation".

d. He did not profess to have any opinion on the likely development of the fire. e. Professor Gray was himself only able to initiate self-heating of Brytall in his tests by the application of high temperatures. There was no evidence that Brytall or any tall oil has initiated self-heating at room temperatures. His tests were conducted on mixtures of attapulgite and Brytall in an oven raised to 120ºC, conditions more in keeping with those existing during the fire, than in the days prior to the fire. f. His tests were designed only to show the possibility of the mixture self-heating. The effect of these limitations, when taken with his other evidence was that Professor Gray could do no more than point to spontaneous combustion as a possible cause of the fire. Further, [McDonald] submits that his Honour erred in adopting Professor Gray's conclusions without sufficient reasoned analysis. Of critical importance to his Honour's reasoning on this aspect was the conclusion of Professor Gray which was recorded in the following terms: It was suggested to Professor Gray by Mr Finch SC that the spontaneous combustion demonstrated in container B may have been initiated or, at least, developed to the point of ignition by the extreme heat of a fire which I need not set out here, but which seemed to me (as a lay person) to be convincing, the professor indicated why he thought that this was not only unlikely but next to impossible. [McDonald] contends that the compendious manner in which his Honour has summarised the effect of Professor Gray's evidence in this passage is inaccurate as a summary of his evidence, an impermissible use of the evidence of an expert, and amounts to a failure to give proper reasons. A proper assessment of Professor Gray's evidence on this point called for a consideration of the detail of his evidence, and of the effect of his evidence taken as a whole: Makita (Australia) Pty Ltd v Sprowles (2001) 52 NSWLR 705 at 745. [McDonald's] hypothesis was first put to Professor Gray at between (Black 42 - 45). It was put again on the second day of his evidence at (Black 58) when it resulted in a lengthy discussion about the length of time it would take for the Professor to provide his answer. The answer commenced at (Black 63). This is the "deflagration wave" thesis (Black 69E) an event which he agreed had not actually occurred (Black 72D). [McDonald's] thesis was then put a third time at (Black 73 ff), especially at (Black 74C - E). A reading of these passages reveals, in [McDonald's] submission, that the Professor failed to meet the contentions put to him in a satisfactory manner. He failed to identify a piece of evidence, or process of reasoning, as to why it was more likely that the exothermic reaction commenced or proceeded to ignition, prior to the fire. Important questions remained unanswered. For example, * Given that self-heating is more likely, or more rapid, at higher temperatures, what physical or other evidence suggested that the self heating observed by Professor Gray had not been initiated or accelerated by the fire? and * if the self-heating commenced independently of the fire, as he contended, why did it cease in one of the two bins after the fire? In the circumstances it was not a proper or reasonable reading of his evidence to simply conclude, as his Honour did, that he provided an explanation which "seemed to me … to be convincing". A proper analysis of Professor Gray's evidence would have revealed that his answers did not sufficiently support the thesis that spontaneous combustion of the mixture caused the fire. The Professor could do no more than demonstrate that spontaneous combustion was a possible cause of the fire."

Causation: Girkaid and Inghams' submissions 97 Girkaid and Inghams submitted that the following propositions were not seriously in contest at the trial: "(i) the oil known as Brytall was "dangerous goods" within the provisions of the Dangerous Goods Act , 1975 and the Dangerous Goods Regulations , 1978 (admitted below by the defendant Barry Edward McDonald) ("McDonald"); (ii) the oil was dangerous when exposed to oxygen (atmosphere) via an enlarged surface area because a chemical reaction (polymerisation) occurred; (iii) heat was a product of the chemical reaction; (iv) the chemical reaction could be significantly controlled by sealing any container in which oil soaked goods were stored thus cutting off, or severely restricting, the supply of oxygen; (v) [the] danger would [be] lessened by placing any storage container outside buildings and away from combustible materials."

98 They contended that on the issue of causation the primary factual issues were whether the mixture could spontaneously combust, and whether the stored oil-soaked attapulgite self-heated and caused the fire or whether its heat was caused by the fire. They argued that Professor Gray's evidence was a "strong opinion that the mixture did spontaneously combust and caused the fire". They pointed out that no contrary expert evidence was called, "although Professor Gray's reports revealed that an expert, Mr Gudmann was retained by McDonald (or on behalf of his interests), and Mr Gudmann prepared and served a report to which Professor Gray replied … Mr Gudmann was not called to give evidence". 99 Girkaid and Inghams submitted that the primary judge's finding that the fire was caused by the oil soaked attapulgite which had spontaneously combusted was without appellable error. They contended: "The thrust of the challenge to Professor Gray's evidence was: (i) that Professor Gray's laboratory test was conducted for the limited purpose of proving that oil soaked attapulgite, when exposed to atmosphere, would polymerise and would self heat, and that the test was not continued to a stage of spontaneous combustion; and (ii) that Professor Gray's reasoning was flawed in that, from his observations at the site to which he applied accepted scientific principles, he concluded that the containers had burnt from the inside out, and that it was equally probable that the fire commenced elsewhere in the warehouse and consumed the stored oil soaked attapulgite thus producing the burnt residue which he observed on the site. Professor Gray's observations on site included:

(i) the oil soaked attapulgite in bin B was hot, and still self heating, and was very hot. This was nineteen days after the fire. (It later confirmed his opinion of its ability to polymerise and produce sufficient heat to self combust); and (ii) Bin B had heated from the inside out. In cross examination Professor Gray said that the opposite theory of heat applied externally would have produced the "burnt cake" appearance, that is, burnt on the outside with unburnt materials towards the centre of the mass: (Black, page 41W). He also observed bin A which, when compared to bin B, confirmed his conclusions referred to in sub-para (i) above. There was fuel for the fire. McDonald had placed paper tops of the attapulgite bags in the storage bins on top of the oil soaked attapulgite, and the bins and virtually all the other goods in the premises were stored on wooden pallets on the floor, the bases of which had been soaked in oil. As to the laboratory test, Professor Gray said that it was not scientifically necessary to take the test further once the test had confirmed that the polymerisation of oil soaked attapulgite did produce heat upon exposure to atmosphere. Thereafter, he said, it was merely a matter of application of basic principles of chemistry: "It is a basic law of physical chemistry that that would be so": Black 58J. And, "it follows from absolute basic principles of chemistry" that the mixture had the ability to undergo spontaneous ignition: Black 67Q. That was not challenged".

Causation: consideration 100 Causation is a question of fact to be determined by the application of common sense to the facts of each case: March v E & M H Stramare Pty Limited (1991) 171 CLR 506; Bennett v Minister for Community Welfare (1992) 176 CLR 408. The question of causation is not resolved by philosophical or scientific theories of causation: see Chappel v Hart [1998] HCA 55; (1998) 195 CLR 232 at 242 [23] per McHugh J citing Mason CJ in March v E & M H Stramare Pty Limited (1991) 171 CLR 506 at 509. In that case Mason CJ said: "In philosophy and science, the concept of causation has been developed in the context of explaining phenomena by reference to the relationship between conditions and occurrences. In law, on the other hand, problems of causation arise in the context of ascertaining or apportioning legal responsibility for a given occurrence." 101 A court is not obliged to accept the opinion of an expert merely because there is no other expert opinion adduced on the particular issue. Nor is an expert's statement that something is more probable than not conclusive: Makita (Australia) Pty Limited v Sprowles [2001] NSWCA 305; (2001) 52 NSWLR 705 at [59], [87] and [89] per Heydon JA. 102 However, of course, scientific opinion is accorded significant weight where appropriate - including opinion expressed in language of possibility. In EMI (Australia) Ltd v Bes [1970] 2 NSWR 238 at 242, Herron CJ said: "Medical science may say in individual cases that there is no possible connection between the events and the death, in which case, of course, if the facts stand outside an area in which common experience can be a touchstone, then the judge cannot act as if there were a connection. But if medical science is prepared to say that it is a possible view, then, in my opinion, the judge after examining the lay evidence may decide that it is probable. It is only when medical science denies that there is any such connection that the judge is not entitled in such a case to act on his own intuitive reasoning. It may be, and probably is, the case that medical science will find a possibility not good enough on which to base a scientific deduction, but courts are always concerned to reach a decision on probability and it is no answer, it seems to me that no medical witness states with certainty the very issue which the judge himself has to try." (emphasis supplied) 103 There is no doubt that McDonald's submission that the issue of causation could not be established on the balance of probabilities by evidence which fails to do more than establish a possibility is correct: see Seltsam Pty Limited v McGuiness [2000] NSWCA 29; (2000) 49 NSWLR 262 at [80] ff per Spigelman CJ, at 293 [201] per Stein JA. However that is not to say that evidence which establishes a possibility is irrelevant. In Seltsam, Spigelman CJ pointed out (at [79]): "79 Evidence of possibility, including expert evidence of possibility expressed in opinion form and evidence of possibility from epidemiological research or other statistical indicators, is admissible and must be weighed in the balance with other factors, when determining whether or not, on the balance of probabilities, an inference of causation in a specific case could or should be drawn. Where, however, the whole of the evidence does not rise above the level of possibility, either alone or cumulatively, such an inference is not open to be drawn. …

83 The law in Australia is, in my opinion, as stated by Glass JA in this Court in Fernandez v Tubemakers of Australia Ltd (1975) 2 NSWLR 190 at 197: "The issue of causation involves a question of fact upon which opinion evidence, provided it is expert, is receivable. But a finding of causal connection may be open without any medical evidence at all to support it: Nicolia v Commissioner for Railways (NSW) (1970) 45 ALJR 465, or when the expert evidence does not rise above the opinion that a causal connection is possible: EMI (Australia) Ltd v Bes [1970] 2 NSWR 238; appeal dismissed (1970) 44 ALJR 360N. The evidence will be sufficient if, but only if, the materials offered justify an inference of probable connection. This is the only principle of law. Whether its requirements are met depends upon the evaluation of the evidence. " (emphasis supplied) 104 Spigelman CJ also examined the approach to be taken to determining whether evidence of possibility was capable of supporting a probative inference. He said: "84 It is often difficult to distinguish between permissible inference and conjecture. Characterisation of a reasoning process as one or the other occurs on a continuum in which there is no bright line division. Nevertheless, the distinction exists. 85 Lord Macmillan in Jones v Great Western Railway Co (1930) 47 TLR 39, in the context of stating that a possibility that a negligent act caused injury was not enough, said (at 45): "The dividing line between conjecture and inference is often a very difficult one to draw. A conjecture may be plausible but is of no legal value, for its essence is that it is a mere guess. An inference in the legal sense, on the other hand, is a deduction from the evidence, and if it is a reasonable deduction it may have validity as legal proof. The attribution of an occurrence to a cause is, I take it, always a matter of inference." 86 After referring to this passage, Sir Frederick Jordan in Carr v Baker (1936) 36 SR(NSW) 301 said (at 306): "The existence of a fact may be inferred from other facts when those facts make it reasonably probable that it exists; if they go no further than to show that it is possible that it may exist, then its existence does not go beyond mere conjecture. Conjecture may range from the barely possible to the quite possible." 87 As Lord Wright put it in a frequently cited passage in Caswell v Powell Duffryn Associated Collieries Ltd [1940] AC 152 at 169-170: "Inference must be carefully distinguished from conjecture or speculation. There can be no inference unless there are objective facts from which to infer the other facts which it is sought to establish. In some cases the other facts can be inferred with as much practical certainty, as if they had been actually observed. In other cases the inference does not go beyond reasonable probability. But if there are no positive proved facts from which the inference can be made, the method of inference fails and what is left is mere speculation or conjecture." 88 The test is whether, on the basis of the primary facts, it is reasonable to draw the inference. (See eg Luxton v Vines (1952) 85 CLR 352 at 358). 89 In my opinion, evidence of possibility, including epidemiological studies, should be regarded as circumstantial evidence which may, alone or in combination with other evidence, establish causation in a specific case. 90 Proof on the balance of probabilities, indeed on the beyond reasonable doubt standard, may be established on the basis of circumstantial evidence. As Lord Cairns said in Belhaven and Stenton Peerage [1875] 1 AC 278 at 279: "My Lords in dealing with circumstantial evidence, we have to consider the weight which is to be given to the united force of all the circumstances put together. You may have a ray of light so feeble that by itself it will do little to elucidate a dark corner. But on the other hand, you may have a number of rays, each of them insufficient, but all converging and brought to bear upon the same point, and, when united, producing a body of illumination which will clear away the darkness which you are endeavouring to dispel." 91 Causation, like any other fact can be established by a process of inference which combines primary facts like "strands in a cable" rather than "links in a chain", to use Wigmore's simile. ( Wigmore on Evidence (3rd ed) para 2497, referred to in Shepherd v R (1990) 170 CLR 573 at 579) … 98 The Courts must determine the existence of a causal relationship on the balance of probabilities. However, as is the case with all circumstantial evidence, an inference as to the probabilities may be drawn from a number of pieces of particular evidence, each piece of which does not itself rise above the level of possibility…" (emphasis supplied) 105 Where a question involves a scientific fact, the absence of scientific opinion refuting the scientific theory advanced in support of the case of the party carrying the burden of proof can be significant. Thus in Adelaide Stevedoring Co Ltd v Forst (1940) 64 CLR 538 at 569 Dixon J said: "I think that upon a question of fact of a medical or scientific description, a court can only say that the burden of proof has not been discharged where, upon the evidence, it appears that the present state of knowledge does not admit of an affirmative answer and that competent and trustworthy expert opinion regards an affirmative answer as lacking justification, either as a probable inference or as an accepted hypothesis." 106 In Tubemakers of Australia Ltd v Fernandez (1976) 50 ALJR 720 at 724-725 Mason J (with whose reasons Barwick CJ and Gibbs J agreed) after referring to this passage from Forst said: "The statement of Dixon J in Forst's Case to which I have referred … emphasizes the restrictions upon a court's power to hold that a plaintiff has not discharged the burden of proof. According to this statement, before a court is justified in so holding it must appear not only that the present state of knowledge does not admit of an affirmative answer, but that competent and trustworthy expert opinion regards an affirmative answer as lacking justification, at least as a probable inference ." (emphasis supplied) 107 Spigelman CJ's proposition that the tribunal of fact is entitled to use a scientific opinion as circumstantial evidence was made clear in Fernandez v Tubemakers [1975] 2 NSWLR 190. I have already referred to the extract from Glass JA's judgment which Spigelman CJ regarded as stating the law in Australia. In addition, Mahoney JA (at 199 - 200) accepted that an inference of "actual cause" could be founded upon evidence of "a possible cause of the condition" saying: "…[B]efore the possible cause, the trauma, can in this particular case be inferred to be the cause of the condition, the cause must be related to the condition, not merely temporally, but `sufficiently closely' … the question would be whether the evidence showed the connection between the possible cause and the condition which occurred was sufficiently close to warrant a reasonable mind, faced with the problem of determining the question upon the evidence before it, concluding that the possible was the actual cause." (emphasis suppled) 108 In the High Court in Tubemakers of Australia Ltd v Fernandez Mason J held that it was open to a jury to infer that, on the probabilities, a blow to the plaintiff's hand caused or materially contributed to the occurrence of Dupuytren's contracture in that hand. A specialist gave evidence as to the aetiology of Dupuytren's contracture and said that the respondent's injury "could have" (i.e. it was possible) played a part in the condition of his right hand. Mason J held that the specialist's evidence, combined with the fact that the condition made its appearance shortly after the accident and no alternative cause was established or suggested in evidence, provided a sufficient basis from which the jury could draw an inference of causation favourable to the plaintiff. 109 In X and Y (by her Tutor X) v Pal (1991) 23 NSWLR 26 at 32 - 33, Mahoney JA referred to the conclusion in Tubemakers of Australia Ltd v Fernandez and commented: "But the inference from B to A because, essentially B has followed A, can be drawn only in limited circumstances. Its weight depends upon there being no other acceptable alternative cause available. And the strength of the inference is affected by, for example, the number of times that B has been seen to follow A in the relevant factual context." (emphasis supplied) 110 In Commonwealth v McLean (1996) 41 NSWLR 389 at 410, Handley and Beazley JJA summarised the effect of Adelaide Stevedoring Co Ltd v Forst above at 563 - 564, 569 and Tubemakers of Australia Ltd v Fernandez as being that "a tribunal of fact is entitled to find causation as a matter of commonsense from the sequence of events, although medical science does not suggest an affirmative answer, provided it does not exclude such a finding". 111 McDonald's task in seeking to persuade this Court to reject the primary judge's conclusion on causation was, as Mr Finch SC frankly acknowledged, a difficult one. This is not least because the deference an appeal court accords to the trial judge's advantage (see Fox v Percy [2003] HCA 22; (2003) 214 CLR 118) applies as much to the exercise of the rehearing function in relation to expert as to lay witnesses. 112 Kirby J noted in Shorey v PT Limited [2003] HCA 27; (2003) 77 ALJR 1104 at [32]) that differing views had been expressed on the question whether the judicial authority about disturbing evidence on the basis of assessments of credibility applies, or applies with the same strictness, in the case of expert witnesses. This is true: many of the expressions of a different view are Kirby J's. However, as Mason P (with whom Giles JA and I agreed) said in Forbes v Selleys Pty Limited [2004] NSWCA 149 at [135] - [136], the "present view" of this Court is generally to the contrary of Kirby J's. 113 Kirby J expounded his view that the primary judge enjoyed no forensic advantage which imposed any relevant inhibition on the conduct of review by the Court of Appeal of his Honour's determination of the facts in Ahmedi v Ahmedi (1991) 23 NSWLR 288 at 291 - 292 where he said: "The issue is not one of credibility but of the logical persuasion and internal consistency of the medical opinion stated. An expert opinion is only as persuasive as the facts upon which it is based. Many are the cases in which expert opinions must be rejected because the factual hypotheses upon which they are based are not made out in the evidence: see, generally, Ramsay v Watson (1961) 108 CLR 642; Paric v John Holland (Constructions) Pty Ltd (1985) 59 ALJR 844 at 846; 62 ALR 85 at 87; Lynch v Lynch (1966) 8 FLR 433; 84 WN (Pt 1) (NSW) 315 and Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247 at 255. Just as many opinions have been rejected because the acceptability of the opinion is undermined by reason of the inadequacy of the facts to support it. Courts do not have to accept an expert opinion, simply because voiced by a person with expert qualifications. Courts, and parties before them, are entitled to test the opinion expressed, scrutinising the premises upon which it is based and evaluating its internal logic. The entitlement of a party to have this measure of review is nothing less than an attribute of due process. In some cases, of course, each step in the logical process of reasoning cannot be demonstrated. In some cases, expert knowledge and the state of the scientific art fall short of a capacity to demonstrate conclusively the accuracy of the opinion expressed. But it is certainly open to a court, and to parties before a court to scrutinise the opinion, measuring it by its internal consistency and by the degree to which it appears to be sustained by the expressed factual premises on which it rests." 114 In Ahmedi v Ahmedi at 299-300 Clarke JA (with whom Handley JA agreed) expressed the view that in a case where the trial judge's conclusion depended upon his assessment and evaluation of contradictory expert oral evidence, the principles in Abalos v Australian Postal Commission (1990) 171 CLR 167 imposing restraint on appellate interference applied. He referred to Wilsher v Essex Area Health Authority [1988] AC 1074, where Lord Bridge said (at 1091): "... Where expert witnesses are radically at issue about complex technical questions within their own field and are examined and cross-examined at length about their conflicting theories, I believe that the judge's advantage in seeing them and hearing them is scarcely less important than when he has to resolve some conflict of primary fact between lay witnesses in purely mundane matters." 115 Mr Finch SC's approach to challenging the primary judge's conclusion on causation involved to an extent the approach Kirby P advocated in Ahmedi v Ahmedi. Thus he pointed to the concessions he contended Professor Gray had made as undermining the validity of his opinion. In particular, he submitted that absent positive proof that the tall oil/attapulgite mixture could undergo spontaneous combustion in the conditions which prevailed before the fire, Professor Gray's opinion was not persuasive. 116 However, the evaluation of Professor Gray's evidence, including his explanation of the deflagration wave theory, the intermediate position and his concessions (if that is a proper characterisation of his evidence), was an exercise in which the primary judge did enjoy an advantage. As Gleeson CJ, Gummow and McHugh JJ observed in Shorey v PT Limited above at [6]: "When… an expert concedes under cross-examination that his or her original opinion was formed without knowledge of some material facts, an appreciation of the extent to which the witness accepts that the opinion is to be qualified or discounted may depend upon an assessment of the witness by the trial judge. Sometimes, of course, this will be plain from the transcript of evidence. In other cases, of which the present is an example, the record will be equivocal." 117 Nevertheless this Court must not eschew its re-hearing function. The inference of causation the primary judge drew from the primary facts with the benefit of expert evidence is open to review on a re-hearing in accordance with the principles considered in Warren v Coombes (1979) 142 CLR 531; see Fox v Percy at [25] per Gleeson CJ, Gummow and Kirby JJ, at [87] - [88] per McHugh J. 118 Thus it is plain that the primary judge was entitled to use Professor Gray's opinion about the cause of the fire, even if it did not reach higher than evidence of possibility as McDonald submits, as long as, taken together with other evidence in the case, he could be satisfied that there was a reasonable inference to the civil standard of the balance of probabilities that the fire was caused by the process of spontaneous combustion of the tall oil/attapulgite mixture as Professor Gray opined. His Honour was also entitled to take into account in considering the issue of causation the sequence of events and the absence of any alternative explanation. Before accepting Professor Gray's opinion he was required to consider Professor Gray's evidence in detail and in whole: Makita (Australia) Pty Limited v Sprowles, above, at 745 [89] per Heydon JA. Finally he was required to apply commonsense. 119 The primary judge's consideration of Professor Gray's evidence demonstrates that, contrary to McDonald's submission, his Honour considered that evidence as a whole, including taking into account what McDonald submitted were concessions. 120 McDonald's criticism of the primary judge's "five crucial factors" does not do justice to the entirety of his Honour's consideration of this issue from which it is plain that he also took into account the physical evidence Professor Gray identified at the warehouse, the laboratory test which demonstrated the admitted "possibility" that the tall oil/attapulgite mixture could self-heat, as well as his acceptance when observing Professor Gray deal with the "other fire" hypothesis that his rejection of that proposition was valid. 121 It is plain that his Honour appreciated (see his judgment at [9]) the force of the submission that Professor Gray had not carried out an experiment in which he sought to replicate the ambient conditions in which the tall oil/attapulgite mixture had been stored prior to the fire. The primary judge said that was impossible. There was, in my view, a tension between Professor Gray's evidence on this point. I understood his second report to say such tests had not been done (and I inferred could not be), whereas in his oral evidence he appeared to say it was possible to do such tests but he had not been instructed to undertake that exercise. But even without such a test, Professor Gray said, without challenge, that it was a basic law of physical chemistry and later chemical reactions that "[i]f you establish that a chemical reaction goes at a particular rate at a particular temperature and produces heat, then at any other temperature that reaction will go at a different rate but it will go at a finite rate provided the ambient temperature is above the absolute zero." Professor Gray was entitled to express an expert opinion based on his experience without "having to prove by admissible evidence all the facts on which the opinion is based": Makita (Australia) Pty Limited v Sprowles, above, at 736 [75]. 122 Professor Gray's evidence that linoleic acid was present in sufficient quantities in the tall oil to undergo spontaneous combustion was not challenged. It was information he derived from gas chromatograph tests of the tall oil. He developed his thesis by carrying out the "oven test". The results of that test were not challenged. It established that a mixture of tall oil and attapulgite could self-heat. It was significant, as Professor Gray himself thought, that after the laboratory test the test mixture bore a striking resemblance to his observation of the contents of Containers A and B. In my opinion Professor Gray's evidence demonstrated that the mixture McDonald deposited in Containers A and B between 23 and 28 January 1998 self-heated. Again, I do not understand McDonald to have challenged that proposition. Rather, he submitted that Professor Gray's evidence did not establish that that exothermic reaction had commenced prior to the fire, had led to self-ignition and caused the fire. 123 However, Professor Gray gave evidence, which was again unchallenged, that what he observed in the containers at the warehouse was consistent with a relatively slow combustion process taking place over a period of days or weeks with a limited supply of oxygen and that the time elapsed between the oil spill and the fire was consistent with the cause being spontaneous combustion of a body (container) of around 1 cubic metre in volume such as the Brytall containers were. That was an opinion the primary judge was entitled to accept as a factual foundation for Professor Gray's opinion. 124 It is true that, as McDonald submits, the appearance of Containers A and B was consistent with that reaction having been precipitated by the extreme heat to which the containers were undoubtedly exposed during the fire. However Professor Gray gave an explanation for why he thought it unlikely that that hypothesis was correct. I reject McDonald's submission that the deflagration wave theory or the intermediate position were not explained in a satisfactory manner. His Honour accepted the opinions because they appeared to him "as a lay person" to be convincing. I share his Honour's view. Professor Gray was seeking to explain a scientific phenomenon (the behaviour of fire) which, by definition it seems to me, involves many variables which may cause any given fire to behave differently from another depending, for example, upon the ambient temperature, the fuel available (both its nature and quantity), the prevailing winds, draughts and so on. 125 In addition to Professor Gray's evidence that he had demonstrated that it was possible that tall oil and attapulgite could undergo spontaneous combustion, the primary judge was entitled to take into account, as he did, the fact that no alternative cause of the fire was demonstrated. The primary judge rejected the rather tentatively advanced electric wire-chewing rats theory. That left spontaneous combustion of the mixture as the only possible fire trigger in the warehouse. His Honour was also entitled to take into account the sequence of events: the fact that the fire started within days of the tall oil and attapulgite being combined. As a matter of commonsense the spontaneous combustion of that mixture was the most probable cause of the fire. The primary judge's conclusion that Girkaid and Inghams had discharged their burden of proof on the causation issue should not be disturbed. 126 I would reject grounds 1 - 5.

Breach of duty 127 The primary judge held (at [29]) that regulations 18(e), 19(e) and 19(g) of the Dangerous Goods Regulation 1978 gave Girkaid and Inghams a civil cause of action, applying O'Connor v S P Bray Limited (1937) 56 CLR 464 at 477 - 8 and John Pfeiffer Pty Limited v Canny (1981) 148 CLR 218 at 231. He concluded that McDonald had breached the statutory duty imposed by regulation 18(e) of the Dangerous Goods Regulation 1978. Alternatively, he concluded McDonald had breached his duty of care to Girkaid and Inghams. He did not reach a conclusion on Girkaid and Inghams' case based on the statutory duty said to have been imposed by regulations 19(e) and 19(g). Girkaid and Inghams' Notice of Contention seeks to uphold the conclusion in their favour on breach by recourse to regulations 19(e) and 19(g) if necessary. 128 Mr Finch SC said it was not open to the primary judge to find McDonald had breached a statutory duty imposed by regulation 18(e) because that regulation had not been pleaded against McDonald as founding a civil cause of action. Regulation 18 had been pleaded against McDonald by Bryce in the third cross-claim. That amendment, however, had been allowed on the limited basis that the allegation related to McDonald's failure to obtain the MSDS produced by EKA. 129 Girkaid and Inghams conceded that McDonald's submission was correct. However they argued that they had advanced a case at trial that breach of regulation 18(e) was primary evidence of McDonald's breach of his duty of care and, thus, was relevant to that aspect of their case. 130 McDonald's challenge to the primary judge's conclusion on breach was made in the following grounds of appeal: "6. His Honour erred in construing regulations 18 and 19 of the Dangerous Goods Regulation , 1978 insofar as his Honour made the following findings: (a) that breach of the regulations by the Appellant gives rise to a private cause of action in the First and Second Respondents; (b) that regulation 18 required the Appellant, as the occupier of premises, to be "adequately instructed as to the hazards involved in the handling of" tall oil;

(c) that regulation 18 imposed on the Appellant a duty to ascertain whether there were any hazards associated with the storage of the mixture, or to have or obtain any other knowledge. 7. His Honour erred in finding that the Appellant was negligent, or in breach of any statutory duty, and in particular his Honour erred in finding that the Appellant was negligent for failing to: (a) access and read the Material Safety Data Sheet ("MSDS") issued by EKA Chemicals (Australia) Pty Ltd ("EKA Chemicals"); (b) make inquiry of either Robert Bryce & Company Pty Limited ("Bryce") or EKA Chemicals as to whether it was safe to store the mixture; (c) take any step to acquaint himself with the possible dangers arising from the storage of the mixture, other than to consult the MSDS issued by Bryce; (d) obtain a positive assurance from a person with sufficient expertise that it was safe to store the mixture before doing so. 8. His Honour erred in failing to find that it was a sufficient discharge of the Appellant's duty of care to inquire of, and follow the direction of, Mr Lance Emerson of SLE Industrial Supplies Pty Limited when using the attapulgite and storing the mixture. 9. Alternatively his Honour erred in failing to find that it was sufficient discharge of his duty of care for the Appellant to have acted consistently with the Bryce MSDS in determining whether there was a risk of spontaneous combustion of the mixture. 10. His Honour erred in finding, in the absence of probative evidence, that an inquiry to EKA Chemicals would have disclosed that the storage of the mixture could lead to spontaneous combustion. 11. His Honour erred in failing to find on the basis of the uncontroverted evidence of Mr Underwood that the storage of the mixture by the Appellant was in accordance with reasonable practice, the applicable Australian Standard and the applicable MSDS, and was therefore consistent with his duty of care to the Plaintiffs." The primary judge's resolution of the breach issue 131 The Dangerous Goods Regulation 1978 was repealed in 1999: Dangerous Goods (General) Regulation 1999, s 341(1). There was no suggestion that that repeal affected Girkaid and Inghams' ability to rely upon it: see s 30 Interpretation Act 1987 (NSW). Prior to the repeal, regulations 18 and 19 of the Dangerous Goods Regulation 1978 relevantly provided: "18. Every: … (b) occupier of premises in or on which dangerous goods are kept; … shall take all practicable steps to ensure that every person subject to his direction and control and engaged in the handling of the dangerous goods is: (e) adequately instructed as to the hazards involved in the handling of those goods and precautions to be observed in relation to those goods; …

19. Every: ... (b) occupier of premises in or on which dangerous goods are kept ... shall: (e) take all practicable precautions to prevent the occurrence on the premises of accidents through fire, explosion, leakage of dangerous goods or other causes ... and (g) not do any act in or on the premises that may cause fire ... or any other dangerous occurrence, or that is not reasonably necessary for the purposes of, or properly incidental to, the keeping ... of dangerous goods. " 132 As I noted in the Introduction, Brytall was classified as dangerous goods for the purposes of the Dangerous Goods Regulation 1978 because it had a flash point above 150ºC: Schedule 1, Dangerous Goods Regulation 1978. 133 The primary judge concluded (at [28] - [29]) that regulations 18 and 19 gave rise to a civil cause of action because:

(a) the Dangerous Goods Act 1975 (NSW) and the Dangerous Goods Regulation 1978 imposed "particular duties" upon persons who manufactured and stored dangerous goods;

(b) the purpose of the Dangerous Goods Act 1975 (NSW) and the Dangerous Goods Regulation 1978 was to avert injury to persons who and property that might be at risk by virtue of that manufacture and storage of dangerous goods; and

(c) McDonald's liability to Girkaid and Inghams was correlative to his duty under the Dangerous Goods Act 1975 (NSW) and the Dangerous Goods Regulation 1978. 134 The primary judge then resolved the breach issue in the following way. 135 He accepted that neither McDonald or his employees understood that soaking up the tall oil with attapulgite and depositing the resulting mixture in containers without lids was capable of leading to ignition. Rather, he said the real question was what McDonald "ought to have known". He found (at [28]) that that matter was "prescribed by regulation 18". 136 He noted (at [31]): "In dealing with Mr McDonald's situation it is important to focus on the point that the damage did not arise from cleaning up the spilt tall oil with attapulgite. There was no risk of fire associated with this process. The risk of fire was created by the storage of the mixture in a condition where it was reasonably possible that it would spontaneously ignite and where fuel was available to permit the spread of fire." and held (at [31]) that: "It should have occurred to Mr McDonald that he had in his warehouse a mixture of two products, the chemical properties of which he was unaware, and about which he should make reasonable inquiries. He made no such inquiry. Had he made such inquiries in relation to attapulgite, he would no doubt have been informed that it could safely be used to soak the tall oil. Whether he would have been informed that it was safe to store the mixture is, however, somewhat doubtful." 137 The primary judge concluded that McDonald should have obtained the Bryce MSDS, although he also held that had he done so McDonald might reasonably have inferred that storage of the mixture was not dangerous or hazardous. Despite this, he also held (at [32]) that McDonald should have obtained and referred to the EKA MSDS in order to instruct himself adequately as to the risks associated with the storage of tall oil and its safe clean up and disposal in the event of spillage. Insofar as the EKA MSDS was concerned the primary judge noted that that document said that "prolonged contact with porous materials" should be avoided. Although the EKA MSDS did not elaborate on why such conditions should be avoided, he concluded (at [32]): "… its sense is clear enough. Certainly, at least to a focused mind, it warned of an unspecified danger or risk with the storage of the mixture; but it did not suggest the possibility of spontaneous combustion and, indeed, suggested there was no such risk." 138 The primary judge found that it was appropriate, entirely safe and reasonable to use attapulgite to mop up the spilled oil. He found that the danger was created by the storage of the mixture in containers together with other flammable material (the tops of the attapulgite bags) providing fuel when the point of spontaneous ignition was reached [33]. He held that storage of the mixture pending disposal was clearly a part of the cleanup of the spilled oil. 139 The critical passages in the primary judge's conclusion on breach of the regulation 18(e) statutory duty are (at [33] - [36]): "… EKA's MSDS, at least, warned that prolonged contact between the tall oil and the absorbent should be avoided, and thus indicated that there was more about the mixture, as it were, than met the eye, but there was no warning as to the nature of the danger. Even so, I think it was probable, had Mr McDonald inquired about storage of a mixture of attapulgite and tall oil from EKA, he would have been told of the risk of spontaneous combustion. On the other hand, there was nothing to indicate to Mr McDonald that he was faced with an emergency in respect of which he needed to consult with EKA except that the label itself, by reference an emergency telephone number, pointed to the possibility that a circumstance might arise in which information about the product or dealing with it might be necessary. 34 I consider that reg 18 requires a person in Mr McDonald's position to … at least consider that he or she cannot rely on his or her general knowledge (the equivalent in this situation of ignorance) of the dangerous product being handled or stored as indicating when a dangerous or hazardous situation has arisen. The information about the product and what should be done, say, when it is spilt and what to do with the resultant clean-up material should be acquired before the occasion for its application occurs. Here, it is obvious that spillage of tall oil was possible. It was probable (as Mr McDonald, indeed knew) that some such absorbent as attapulgite might be used to clean it up. The resultant mixture would need to be stored before disposal. To my mind, reg 18 imposed on Mr McDonald the duty to ascertain whether there were any hazards associated with that storage which involved, at least, reference to the MSDS of the distributor (his client) and the manufacturer of a dangerous product and obtaining a specific assurance that what he proposed to do was safe . Mere collection of the spilt tall oil was not, of course, all that he proposed to do. He also proposed to store the resultant mixture until it was disposed of. In short, it is the very point of the dangerous goods regime that those who store them, such as Mr McDonald, cannot take for granted that their common knowledge assumptions about safety can be relied on. Specific information about the particular mode of, say, storage, must be obtained. It matters not, in my view, that tall oil, as here, was prescribed as dangerous for reasons other than the hazard that caused the damage . This point is clear from the regulation itself. 35 It is obvious that Mr McDonald, whatever the duty imposed upon him by virtue of reg 19, was, in the circumstances not "adequately instructed as to the hazards involved in the handling of" tall oil within the meaning of reg 18. Adequate instruction required more than simple reference to the MSDS prepared by Bryce or, for that matter, EKA if those documents did not expressly address the problem about which information was necessary. Whether instruction is adequate is a question to be objectively determined by the hazards involved in the handling of the goods in question. It seems to me that, at least, there should have been an express advertence to the possible circumstance requiring attention, here, clean up of spilled tall oil with an absorbent material and safe storage and disposal of the resultant mixture. Knowledge of chemistry was not necessary -depending on the substances involved - but Mr McDonald was required by reg 18 to be armed with information that directly and unmistakeably dealt with the foreseeable situation. Thus, it was no compliance with reg 18 to wait until something happened that he recognized as an emergency before he instructed himself how to properly respond.

36 The point about the prescribed instruction is that it must be adequate to avoid any real or significant risk to life, limb or property, the protection of which is the fundamental purpose of the legislation. In this case, specific information about whether a mixture of tall oil and a product such as attapulgite kept in containers for days was safe has been demonstrated to be necessary in order to avoid the fire which occurred . That some substances may spontaneously combust is a notorious fact and, as I explain later, the plaintiffs have proved, through the evidence of Professor Gray, that both the fact and the risk that tall oil was capable in circumstances similar to those brought about by Mr McDonald of spontaneous combustion, was sufficiently known in industry (though perhaps not the warehousing industry) to demonstrate that the problem was no mere matter of hypothetical or theoretical interest but was a real and substantial risk which anyone in Mr McDonald's position faced with cleaning up a substantial quantity of tall oil with a product such as attapulgite should have known . It is no answer, as it seems to me, to argue that all that Mr McDonald needed to have done was to consult Bryce and that, as he would not have been usefully informed by its MSDS, his ignorance was excusable or, to put it another way, his instruction was adequate. I do not see how it is possible to be "adequately instructed" by acquiring significantly misleading information. I am far from certain that reg 19(g) does not impose strict liability on a person who does an act that "may cause fire" in premises used for the keeping of dangerous goods. However, it is unnecessary for the purposes of this case to determine this question, since doing an act that might cause fire, in this case, storing the self-combusting mixture of oil and attapulgite, would have been known to Mr McDonald to be dangerous if he had been "adequately instructed" within the meaning of reg 18(e). " (emphasis supplied) 140 In the event he was mistaken as to the application of the Dangerous Goods Regulation 1978 (NSW), the primary judge also dealt with the common law case. He accepted (at [37]) "Professor Gray's evidence that there were a significant number of references in the literature to the propensity of tall oil to self-combust" and "[t]all oil is comprised mainly of linoleic and oleic acids which are unsaturated fatty acids well known for its tendency to undergo spontaneous combustion when disbursed on porous material". 141 He said (at [37]) that: "… any reasonably competent industrial chemist would either be aware of or could easily ascertain information that tall oil had a propensity for spontaneous combustion when absorbed by materials such as attapulgite. This information should have been available from both Bryce's and EKA's MSDS." 142 He held (at [38] - [39]): "38 To my mind, Mr McDonald's duty of care to persons who were foreseeably at risk as a result of fire from dangerous goods stored by him required him to take reasonable steps to inform himself of the nature of the dangerous goods which it was his business to handle and that this included modes of safe disposal in the event of spillage. Whether a reasonable person in Mr McDonald's position has failed in this duty of care, measured purely by reference to the common law, is not easy to determine but, in all the circumstances as I have outlined them, I consider that that Mr McDonald was obliged to acquaint himself with the risks of storage of a mixture of tall oil and attapulgite or similar product before he undertook to do so and that reliance upon merely suggestive and indirect assurances, such as would have been obtained from the MSDS produced by Bryce and EKA would not be sufficient. Only a positive statement in such a document that there was no risk associated with the storage of such a mixture would have sufficed to satisfy Mr McDonald's duty of care . Of course, had he referred to the EKA MSD, he would (or should) have realized that storage of the mixture as he intended was to be avoided. This should, at least, have put him on inquiry. Reference to EKA would probably have revealed the danger of spontaneous combustion and to the conclusion that the mixture should not be stored as he intended. At all events, he should not have stored the mixture unless he had a positive assurance from a person with sufficient expertise that it was safe to do so . 39 However, this matter does not fall to be determined by the common law standard considered in isolation. I have already referred to the provisions of reg 18. I consider that Mr McDonald was required by virtue of that enactment, at the least, to obtain from some suitably qualified person (and Bryce or EKA would fall into this class) reasonable assurance as to the absence of risk of storing the mixture. I do not consider that any person such as Mr McDonald, considering whether it was safe to store the mixture in question, could have reasonably inferred from either the Bryce or the EKA MSDS that there was no risk involved since, on a fair reading, these documents do not suggest any advertence by their authors to the actual situation which Mr McDonald should have assessed . I have said that, so far as Mr McDonald's statutory duty was concerned, the test of adequate information is objective and could not be satisfied by wrong information, even if that came from an apparently appropriately qualified source . But, even if I am mistaken about that, it seems to me that, at least, information about the nature of the hazard must be unambiguous and relate directly to the character of the hazard being considered. Mr Finch SC argued that the implication of the Bryce MSDS, had Mr McDonald sought it, was that there was no risk involved in the storage of the mixture. Whilst the document is open to this interpretation, I think it should be more correctly read by someone in Mr McDonald's situation as not dealing with that situation at all . Assuming that Mr McDonald's next step was to consult the EKA MSDS, that certainly indicated there was some kind of unspecified risk associated with the long term contact between tall oil and porous materials but it, also, did not directly address the problem about which Mr McDonald required information. Further inquiry, was to my mind, necessary and there was a telephone number on the container label which was an obvious source of further information. Having regard to the warning, albeit limited, expressed in the EKA MSDS I think it more likely than not (as has already been mentioned) that Mr McDonald, on such an inquiry, would have been informed of the risk of spontaneous combustion providing he insisted, as I think it was his duty, on obtaining an unambiguous answer to the question of the safety of storage of the mixture . Accordingly, whether the duty to be informed of the risk is directly imposed by reg 18 or by virtue of that regulation informing the common law duty (see Sutherland Shire Council v Heyman (1984) 157 CLR 424 per Mason J at 459.5) does not, in the end, matter very much. Either way, I am satisfied that the appropriate standard of skill and competence to be possessed by a person in Mr McDonald's position, undertaking the storage of dangerous goods, required that he or she be informed of the dangers of storing products such as tall oil soaked into attapulgite and the risk that such a mixture will spontaneously ignite and cause a fire." (emphasis supplied) 143 The primary judge rejected McDonald's case that he had not breached his duty of care because he had acted in accordance with industry practice. He regarded Mr Underwood's evidence that, in substance, the course McDonald took to cleaning up the spill was appropriate, as conflating the issues of cleaning and storage ([40] - [42]). He also rejected Mr Finch SC's submission that as Mr Underwood was himself unaware of the risks of spontaneous combustion arising from the storage of the mixture, it could not be inferred that McDonald should have been aware of those on the basis that Mr Underwood was not in McDonald's situation. He said (at [42]) that the question was not whether McDonald should have had the information but that he should have made inquiries as to the risk of storing the mixture. 144 Accordingly, his Honour held (at [43]) that he was satisfied that Girkaid and Inghams had proved on the balance of probability McDonald was liable to them for losses attributable to the fire because he had breached both his statutory and common law duties. Breach: McDonald's submissions 145 Mr Finch SC submitted that a proper reading of regulations 18 and 19 in light of the Dangerous Goods Act 1975 (NSW) and the Dangerous Goods Regulation 1978 (NSW) as a whole disclosed no legislative intention to create a private cause of action. He pointed to the fact that the Dangerous Goods Regulation 1978 (NSW) provided penalties for breach. He submitted that regulations 18 and 19 supported the power of inspectors to direct compliance with safety standards. He also contended that the syntax of regulation 19(g) was inconsistent with a standard of care imposed for the purpose of creating private rights. He argued that the injunction against doing any act "not reasonably necessary for the purposes of, or properly incidental to, the keeping … of dangerous goods" was directed not to the establishment of a private right but to the directions that might be given by an inspector. 146 Secondly, Mr Finch SC submitted that the approach the primary judge took to the interpretation of regulation 18 as informing the common law duty of care had the effect of erecting a tort of strict liability. He contended that, at the most, regulation 18 required a person in McDonald's position to do what he reasonably ought to know to do or to inform himself adequately. He submitted that "adequately" did not mean making inquiries until a negative answer was received or distrusting the distributor of the product (ie. SLE) that he had bought for the purpose of being used to clean up an oily substance. 147 Mr Finch SC argued that the primary judge's interpretation of regulation 18(e) that it would not have been sufficient for McDonald to have relied upon the assurances he would have obtained from the Bryce and EKA MSDSs had he obtained them but, rather, that he had to in effect mistrust that information and ring another person, set "an extraordinarily high bar" - which was erected at an unreachable height. 148 Mr Finch SC argued that Professor Gray's evidence did not support the primary judge's conclusion (at [36]) that the risk of spontaneous combustion was "sufficiently known in industry" so as to amount to a real and substantial risk of which McDonald should have been aware. He said the finding was at odds with the MSDSs published by those who could be expected to know. Thus, the Bryce MSDS advised that in the event of a spill, the Brytall should be absorbed onto an absorbent material which was the category into which attapulgite fell. The EKA MSDS advised that in the event of spills as much as possible should be collected in "a clean container for (preferable) reuse or disposal". He also drew attention to the statement in the EKA MSDS that there were no materials to avoid and no special measures necessary in relation to fire and explosion protection. 149 He also argued that, in any event, his Honour's finding that the propensity of tall oil to undergo spontaneous combustion was "perhaps not known in the warehousing industry" was significant having regard to the fact that that was the industry in which McDonald worked. 150 Mr Wales SC who appeared for Hudson also contended that this finding was erroneous. He pointed to the evidence from McDonald, Mr Bashford and Mr Emerson that none of them were aware of the phenomenon of spontaneous combustion in connection with oils. He also drew attention to Mr Underwood's evidence that knowledge of the process by which spontaneous combustion took place was a "fairly esoteric reaction". 151 Mr Finch SC criticised the primary judge's conclusion (at [38]) that if McDonald had referred to the EKA MSDS he "would or should have realised that storage of a mixture as he intended was to be avoided" and that he would have been "put … on inquiry". He drew attention to the fact that both Professor Gray and Mr Underwood's evidence was that McDonald had acted consistently with the EKA MSDS. 152 Mr Finch SC also criticised the distinction the primary judge drew between cleaning up the tall oil with attapulgite and subsequently storing the mixture. He submitted that was "an impermissible division of … an indivisible activity". He submitted that the whole point of the MSDSs was to advise that a spill of tall oil should be cleaned up by using absorbent material and collecting the resultant mixture. He said that neither of the MSDSs warned about storage per se. 153 Mr Finch SC also contended that the primary judge's finding (at [36]) that "storing the self-combusting mixture of oil and attapulgite, would have been known to Mr McDonald to be dangerous if he had been "adequately instructed" within the meaning of reg 18(e) was an unrealistic interpretation of that regulation. He argued that a proper understanding of the primary judge's approach was that regulation 18(e) meant that McDonald had to ring EKA even though there was no reason to do so because there was no emergency and there was no other reason to do so and secondly, if, having had no reason to ring EKA he had rung EKA, he would have been advised that it was dangerous to store the mixture of attapulgite and tall oil. He submitted that neither part of the deconstructed proposition could be sustained.

154 Dealing with the first point, Mr Finch SC pointed to what he submitted was an inconsistency in the primary judge's logic. The primary judge found (at [33]) that there was nothing to indicate to McDonald that he was dealing with an emergency. The primary judge drew attention, however, to the fact that the label on the Brytall containers notified a telephone number which could be called for "specialist advice in an emergency only". Mr Finch SC pointed out that despite that telephone number, on the basis of the finding that there was no emergency, there was nothing which should have indicated to McDonald that even as a prudent person required to comply with the Dangerous Goods Regulation 1978, he should ring the manufacturer of Brytall in New Zealand. He submitted that that proposition was reinforced by the fact that SLE had provided the attapulgite to McDonald for the express purpose of mopping up the tall oil and that McDonald was entitled to rely upon SLE's experience in respect of the product it recommended. 155 Dealing with the second point, Mr Finch SC submitted the primary judge's conclusion that if McDonald had made a specific inquiry he would have been informed of the risk of spontaneous combustion, was an impermissible inference. He argued that it was drawing a very long bow to find that if McDonald had called EKA a representative of that firm would have advised him, in effect, to ignore those parts of its MSDS which (as his Honour had found (at [32])) suggested that there was no risk of spontaneous combustion, as well as that part of the MSDS which said that "no hazardous polymerisation will occur" (which it will be recalled was a reference to part of the process of spontaneous combustion), and would have said there was a risk of spontaneous combustion in the steps McDonald proposed to take. 156 Mr Finch SC submitted that the objective facts supported the proposition that if McDonald had rung anybody on the list of potential contacts (Bryce, EKA, SLE) he would have been informed that what he was doing was correct rather than the opposite. 157 In summary, Mr Finch SC submitted that the primary judge's conclusion involved finding that McDonald had a "duty to get an unambiguous answer to a question which he had no occasion to ask".

Breach of duty: Girkaid and Inghams' submissions 158 Girkaid and Inghams submitted that the primary judge's conclusion on breach was correct, whether approached on the basis of breach of a statutory or common law duty. 159 Mr Faulkner SC argued that the Dangerous Goods Regulation 1978 established McDonald's standard of care. Thus, he submitted, regulation 19(g) meant that McDonald should not have taken the steps he took "if it may cause fire". When asked how McDonald was to work that out, he responded that that was a matter for McDonald "but he must do it". 160 Girkaid and Inghams submitted that regulation 19 of the Dangerous Goods Regulation 1978 was designed to avoid injury to persons and to avoid damage to property in circumstances where it was common knowledge that fire in a dangerous goods store may be very dangerous due to resulting chemical explosions, heat intensity, fumes and gases. They argued that regulation 19(g) reflected a legislative intention of "special care and strict liability for breach of regulation 19" and, thus, was not confined to an act which caused fire but was also directed at the risk of fire. 161 Mr Faulkner SC submitted that McDonald would not have been in breach of his statutory duty (as I understood whether under regulation 18(e) or 19) if he had stored the mixture outside his building. (T 103) He argued that McDonald's first breach of the regulation 19 duty was to store the mixture on his premises in unsealed containers. His second breach was failing to remove the containers from the warehouse immediately. He contended that those were acts which "may cause fire" (regulation 19(g)) and, thus, McDonald had acted in breach of statutory duty. 162 Turning to the common law, Mr Faulkner SC submitted that, given that McDonald owed a duty of care to Girkaid and Inghams, it was reasonably foreseeable "as a scientific phenomenon" that storing the mixture in unsealed containers on the premises could result in fire. He drew attention to Professor Gray's evidence that it is a "well known scientific fact that substances containing linoleic acid are capable of spontaneous combustion and ignition when merely mixed with other materials and exposed to atmosphere". 163 Mr Faulkner SC also submitted that the question whether McDonald breached his duty of care was to be determined by reference to the "high responsibility of the chemicals/dangerous goods warehouseman because of the potentially severe danger to the community". He submitted that the community's standard was reflected in the Dangerous Goods Act 1975 and Dangerous Goods Regulation 1978. He contended that that standard was not to be "determined solely or even primarily by reference to the practice followed or supported by a responsible body of opinion in the relevant profession or trade" but was one for the Court to adjudicate, referring to Rogers v Whitaker (1992) 175 CLR 479 at 487. He argued that the standard of care to be observed by a warehouseman of chemicals/dangerous goods was "to make no decision and assumptions as to the storage of goods which have been created in their new physical state by him exposing a chemical to a greater extent of oxidation than would be the case should it merely be lying on the floor at any depth, without ascertaining whether the material in its new state contains any substance (such as linoleic acid) which is capable of spontaneous combustion and ignition when stored in circumstances open to atmosphere". 164 Mr Faulkner SC also submitted that breach of the regulation was primary evidence of breach of McDonald's common law duty referring to Sutherland Shire Council v Heyman (1985) 157 CLR 424 at 459.5 where Mason J (as he then was) said: "The existence of a statutory cause of action, generally based on strict liability, does not exclude liability for a common law duty of care unless the Statute provides otherwise … on this view, the breach of a statutory duty may both (a) itself give rise to a separate cause of action, and (b) be evidence of negligence at common law". 165 Mr Faulkner SC submitted the primary judge was correct in finding that "the risk of fire was created by the storage of the mixture in a condition where it was reasonably possible that it would spontaneously ignite and where fuel was available to permit the spread of fire" (at [31]) and that "the danger was created by the storage of the mixture in containers together with flammable material providing fuel when the point of spontaneous ignition was reached" ([33]). 166 Mr Faulkner SC submitted that the primary judge's finding on common law breach could be supported by the proposition that McDonald was not entitled to make uninformed assumptions but had a duty to ascertain whether there were any hazards associated with that storage. He contended that it had been unnecessary for the primary judge to deal with the question of what McDonald would have been told if he had inquired of Bryce or EKA. He contended that the fact that, had McDonald made the inquiry, he may have received accurate or inaccurate information, was not to point but was rather a red herring raised by McDonald which the primary judge "unfortunately followed". 167 Finally, he submitted that it was sufficient for the primary judge to hold (at [38]) that: "Mr McDonald's duty of care to persons who were foreseeably at risk as a result of fire from dangerous goods stored by him required him to take reasonable steps to inform himself of the nature of the dangerous goods which it was his business to handle and that this included modes of safe disposal in the event of spillage. Whether a reasonable person in Mr McDonald's position has failed in this duty of care, measured purely by reference to the common law, is not easy to determine but, in all the circumstances as I have outlined them, I consider that Mr McDonald was obliged to acquaint himself with the risks of storage of a mixture of tall oil and attapulgite or similar product before he undertook to do so and that reliance upon merely suggestive and indirect assurances, such as would have been obtained from the MSDS produced by Bryce and EKA would not be sufficient."

Statutory duty: consideration 168 In O'Connor v S P Bray Limited (1937) 56 CLR 464 at 477 - 8 Dixon J said: "It is a question of some difficulty whether a civil remedy is given to a person injured in consequence of the breach of that clause [of an Act ]. Such a person may, of course, maintain an action of negligence and rely upon the failure to comply with the statutory regulations as evidence of negligence. But it is a different question whether the enactment itself confers a distinct cause of action. The received doctrine is that when a statute prescribes in the interests of the safety of members of the public or a class of them a course of conduct and does no more than penalise a breach of its provisions, the question whether a private right of action also arises must be determined as a matter of construction. The difficulty is that in such a case the legislature has in fact expressed no intention upon the subject, and an interpretation of the statute, according to ordinary canons of construction, will rarely yield a necessary implication positively giving a civil remedy. As an examination of the decided cases will show, an intention to give, or not to give, a private right has more often than not been ascribed to the legislature as a result of presumptions or by reference to matters governing the policy of the provision rather than the meaning of the instruments . Sometimes it almost appears that a complexion is given to the statute upon very general considerations without either the authority of any general rule of law or the application of any definite rule of construction ... perhaps in the end, a principle of law will be acknowledged as the foundation of the cases. In the absence of a contrary legislative intention, a duty imposed by statute to take measures for the safety of others seems to be regarded as involving a correlative private right, although the sanction is legal, because it protects an interest recognised by the general principles of the common law ... whatever wider rule may ultimately be deduced, I think it may be said that a provision prescribing a specific precaution for the safety of others in a matter where the person upon whom the duty is laid is, under the general law of negligence, bound to exercise due care, the duty will give rise to a correlative private right, unless from the nature of the provision of from the scope of the legislation of which it forms a part a contrary intention appears. The effect of such a provision is to define specifically what must be done in furtherance of the general duty to protect the safety of those affected by the operations carried on." (emphasis supplied) 169 In John Pfeiffer Pty Limited v Canny (1981) 148 CLR 218 at 231 Mason J said, after citing O'Connor v S P Bray Limited: "Ordinarily a duty imposed by statute to take measures for the safety of others involves a correlative private right unless from the nature of the provision or from the scope of the legislation a contrary intention appears... [No] doubt the principle was expressed in these terms so as to exclude the statutory prohibition against driving a motor vehicle without a licence, there having been a reluctance on the part of the court to accept that such a statutory prohibition creates a private right of action." (emphasis supplied) 170 The search for a legislative intention that a statutory provision which is completely silent on the point confers a private cause of action is a somewhat illusory exercise, a point to which Dixon J adverted in the passage from O'Connor v S P Bray Limited set out above. In Byrne v Australian Airlines Limited (1995) 185 CLR 410 at 461 - 462, McHugh J and Gummow J referred to the added difficulty which arises where a private cause of action is said to be created by delegated legislation in the form of regulations made under power conferred on the Executive by statute. In such circumstances, their Honours pointed out, "[I]f the statute did not expressly confer on the Executive a power by regulation to create an action for damages at the suit of any person injured by breach of the substantive provisions of the regulation, it must be difficult to construe the statute and the delegated legislation as impliedly bringing about that result". 171 The proposition to which their Honours referred was one which enjoyed currency prior to the decision in Australian Iron and Steel Pty Limited (1957) 97 CLR 89 in which Kitto J's judgment made it plain (at 98) that it was based on the fallacy that the question whether a private cause of action was created depended on discerning a "disclosure of a positive intention to create such a right." Following that decision, as was pointed out in The Liability of Employers, Glass, McHugh and Douglas (The Law Book Company Limited, second edition at 115), "[a]ctions for breach of duties created by regulation … proliferated". McHugh and Gummow JJ's observation had particular significance in Byrne v Australian Airlines Limited where it was being contended that "the silent statute operates upon an award made by an arbitral body established by the statute". It cannot, however, resolve the outcome of the question whether regulations 18 and 19 created a private cause of action. 172 The long title to the Dangerous Goods Act 1975 describes the Act as one "to consolidate and amend the law relating to explosives and other dangerous substances". 173 It might be accepted, therefore, that the purpose of the Dangerous Goods Act and the Dangerous Goods Regulation was to regulate the keeping, handling and use of dangerous goods in order, inter alia, to reduce the risk of damage by fire or explosion: cf Wright v T I L Services Pty Limited (1956) SR (NSW) 413 at 415 per Owen J construing the Inflammable Liquid Act 1915 and that, therefore, they are directed to the interests of the safety of members of the public or a class of them (O'Connor v S P Bray Limited). 174 However, the question whether a statutory duty confers a correlative private right of action also turns on whether the statute imposes a duty to take "a specific precaution" (O'Connor v S P Bray Limited) or "measures for the safety of others" (John Pfeiffer Pty Limited v Canny). 175 In Storozuk v Commissioner for Railways (1963) 63 SR (NSW) 581 at 593 - 594, Brereton J (with whose reasons Else-Mitchell J agreed) doubted whether regulations which expressed an obligation in terms of a requirement to take "all practicable precautions" founded a civil action. This was because the phrase did not "prescribe or define precisely the means that must be taken …" (referring to Windeyer J in General Constructions Pty Ltd v Peterson (1962) 108 CLR 251 at 257). In Slivak v Lurgi (Australia) Pty Limited [2001] HCA 6; (2001) 205 CLR 304, however, all members of the High Court held, respectively, that a provision of the Occupational Health, Safety & Welfare Act 1986 (SA) which turned on the phrase "ensure so far as reasonably practicable" created a private cause of action (at [27] - [29] per Gleeson CJ, Gummow and Hayne JJ, at [50] per Gaudron J, at [87] per Callinan J). 176 Although the question whether a statute confers a private cause of action ultimately turns on the terms of the particular legislation, Slivak v Lurgi (Australia) Pty Limited is persuasive authority supporting the view that the requirements in regulations 18(e) and 19(e) to take "all practicable steps" and "all practicable precautions" sufficiently identify specific precautions or measures to support the primary judge's conclusion that, taken with the purpose of the dangerous goods legislation, those regulations created a private cause of action. I would, therefore, reject ground 6(a) of McDonald's grounds of appeal insofar as it relates to regulations 18(e) and 19(e). 177 Regulation 19(g) is in a different category. It prescribes the end but not the means. It does not identify any specific precaution or measure which the occupier is to take for the safety of others. It is a blanket prohibition on doing the act in question. It does not tell the occupier what measures must be taken: cf Utah Construction & Engineering Pty Limited v Pataky [1966] AC 629 at 641. In my view, the absence of identification of specific precautions means that regulation 19(g) should not be construed as conferring a correlative private cause of action on Girkaid and Inghams. I would, therefore, uphold ground 6(a) of McDonald's notice of appeal insofar as it related to regulation 19(g). 178 The conclusion that one out of several clauses in the enactment does not create a private right is not inconsistent with a conclusion that other clauses in the same enactment do create such a right: see O'Connor v S P Bray Limited, above, at 479.

Content of the statutory duty: consideration 179 I do not agree, however, with the primary judge's interpretation of the extent of the regulation 18(e) obligation, whether applied in the context of breach of statutory duty or as evidence of breach of McDonald's duty of care. I also do not accept that Girkaid and Inghams' Notice of Contention seeking to sustain the judgment on the basis that McDonald breached a statutory duty imposed by regulation 19(e) is made out. 180 The Dangerous Goods Regulation 1978 (NSW) is clearly in the category of remedial or beneficial provisions enacted for the protection of the public which should be construed to afford "the fullest relief which the fair meaning of its language will allow": Bull v Attorney-General (NSW) [1913] HCA 60; (1913) 17 CLR 370 at 384 per Isaacs J; Commissioner of State Revenue (Vic) v Royal Insurance Australia Ltd (1994) 182 CLR 51 at 98 per Dawson J; Marks v GIO Australia Holdings Ltd [1998] HCA 69; (1998) 196 CLR 494 at 528 per Gummow J; ICI Australia Operations Pty Ltd v WorkCover Authority of New South Wales [2004] NSWCA 55; (2004) 1 DDCR 259 at [349]. 181 Nevertheless, a breach of the Dangerous Goods Regulation 1978 also constitutes an offence: cl 31, Dangerous Goods Regulation. It thus has penal consequences. The approach which should be taken to the interpretation of legislation which is both beneficial and penal was considered in Waugh v Kippen (1986) 160 CLR 156. In their joint judgment the majority, Gibbs CJ, Mason, Wilson and Dawson JJ, referred (at 164) to Gibbs J's statement in Beckwith v The Queen (1976) 135 CLR 569 at 576 that: "The rule formerly accepted, that statutes creating offences are to be strictly construed, has lost much of its importance in modern times. In determining the meaning of a penal statute the ordinary rules of construction must be applied, but if the language of the statute remains ambiguous or doubtful the ambiguity or doubt must be resolved in favour of the subject by refusing to extend the category of criminal offences." 182 Their Honours said (at 164 - 165) that in the event that the principles of interpretation referred to in Bull v Attorney General (NSW) and Beckwith v The Queen came into conflict then: "… [T]he Court must proceed with its primary task of extracting the intention of the legislature from the fair meaning of words by which it has expressed that intention remembering that it is a remedial measure passed for the protection of the worker. It should not be construed so strictly as to deprive the worker of the protection which Parliament intended that he should have … in such a context the strict construction rule is indeed one of last resort. Furthermore, the process of construction must yield for all purposes a definitive statement of the incidence of an obligation imposed on the employer . The legislature cannot speak with a forked tongue. Although the standard of proof applicable to criminal proceedings for a breach of the obligation would differ from that applicable to civil proceedings and the law may provide specific defences by way of answer to a prosecution which have no relevance to civil proceedings … the elements that make up the obligation will be the same in each case." (emphasis supplied) 183 The proposition that where a statute creates both a criminal offence and, properly construed, confers a private right to damages for breach of the statutory duty imposed, "the measure of the duty does not change with the character of the proceedings taken to enforce it" was reiterated by Brennan J in Chugg v Pacific Dunlop Limited (1990) 170 CLR 249 at 252. 184 Waugh v Kippen provides a useful illustration of the application of what I will term the remedial/penal approach to interpretation. It concerned the interpretation of a rule made under the Factories and Shops Act 1960 (Qld) providing that a male employee aged over eighteen should not be "permitted or allowed to lift carry or move by hand any object so heavy as to be likely to cause risk of injury". The majority held (at 165) that the rule did not impose absolute liability. Rather, the words "permitted or allowed" imported the notion of actual or constructive awareness. Thus, the question of liability was to be judged "in the light of what the employer knew or ought to have known of [the] employee's incapacity". 185 Their Honours referred to Sheen v Fields Pty Limited (1984) 58 ALJR 93 where, in considering the interpretation of a rule requiring an employer to provide eye protection where there was a "likelihood of injury to the eyes of an employee", Gibbs CJ (with whom Mason, Wilson and Dawson JJ agreed) said (at pp 95 - 96): "The likelihood of injury must be judged in the light of the circumstances which are known, or which ought to have been known, to the employer on whom the duty is cast. It would be unreasonable to construe the rule as casting an obligation on an employer to protect his employee from the consequences of his own independent decision to adopt a dangerous method of working which is different from the method he was instructed to adopt unless the employer could reasonably be expected to foresee that the employee might act in this way ." (emphasis supplied) 186 Applying that reasoning, the majority held (at 166) that it would be unreasonable to construe the rule under consideration "as casting an obligation on an employer to protect his employee from a risk of injury of which he neither knew nor ought to have known". 187 This approach was consistent with the common law presumption that knowledge (which may include constructive knowledge) is an essential element of criminal offences. In He Kaw Teh v R (1985) 157 CLR 523 at 567, Brennan J (as he then was) explained the rationale of this presumption in the following terms: "The penalties of criminal law cannot provide a deterrent against prohibited conduct to a person who is unable to choose whether to engage in that conduct or not, or he does not know the nature of the conduct he chooses to engage in or he cannot foresee the results which may follow from that conduct (where those results are at least part of the mischief to which the statute is aimed). It requires clear language before it can be said that a statute provides for a person to do or to abstain from doing something at his peril and to make him criminally liable if his conduct turns out to be prohibited because of circumstances that that person did not know or because of results that he could not foresee. However grave the mischief at which a statute is aimed may be the presumption is that the statute does not impose criminal liability without mens rea unless the purpose of the statute is not merely to deter a person from engaging in prohibited conduct but to compel him to take preventive measures to avoid the possibility that, without deliberate conduct on his part, the external elements of the offence might occur. A statute is not so construed unless effective precautions can be taken to avoid the possibility of the occurrence of the external elements of the offence ." (emphasis supplied) 188 The requirement that clear words are required in interpreting penal provisions was emphasised in Murphy v Farmer (1988) 165 CLR 19 at 28, where Deane, Dawson and Gaudron JJ said:

"[T]hose who contend that [a] penalty must be inflicted, must shew that the words of the Act distinctly enact that it shall be incurred under the present circumstances." 189 That statement was applied by Giles JA (with whom James J and Badgery-Parker AJ agreed) in R v Stanley John Miller [2001] NSWCCA 209; (2001) 127 A Crim R 344 at [54] in construing s 73 of the Crimes Act 1900 (NSW). His Honour said it would not be "consonant with this approach so to interpret s 73 that it would have a wide and uncertain reach in place of that given in taking its words in their accepted meanings; nor should a purposive construction be given to it if that would depart from the accepted meanings". (emphasis supplied) 190 Applying these principles to the interpretation of regulations 18(e) and 19(e), it is plain, in my view, that the obligations they imposed were not absolute: Chugg v Dunlop Limited (1990) 170 CLR 249 at 251 per Brennan J; Slivak v Lurgi (Australia) Pty Limited [2001] HCA 6; (2001) 205 CLR 304 at 332 [87] per Callinan J. Rather, the words requiring the occupier to take "all practicable steps" and "all practicable precautions" indicate that regulations 18(e) and 19(e) should be interpreted as imposing a penalty, and a corresponding private cause of action, only in circumstances where the occupier of the relevant premises (in this case McDonald) knew or ought to have known that the course of conduct he was taking involved a risk proscribed by the regulations. 191 Thus, the question whether McDonald took "practicable steps" or "practicable precautions" turned on what he knew or ought to have known or did or ought to have done, having regard to the hazard posed by the Brytall. It would, in my view, be an unreasonable construction of regulation 18(e) to find that it imposed an obligation on McDonald to give instructions in relation to a risk of which he was neither aware nor ought to have been aware. 192 In like vein, regulation 19(e) required McDonald to take precautions he knew or ought to have known were practicable to prevent, in this instance, the dangerous goods catching fire. Again, it would be an unreasonable interpretation of regulation 19(e) to find that it required McDonald to take precautions against a risk of which he was neither aware nor ought to have been aware. 193 A brief survey of authorities which have considered a statutory duty which turned on practicability bears out this conclusion. Thus, in Adsett v K & L Steelfounders and Engineers Ltd [1953] 1 WLR 773; [1953] 2 All ER 320 applied in Brear v British Paints Ltd [1978] 2 NSWLR 253 at 258, it was held that in determining whether a provision requiring "all practicable measures to be taken" has been complied with, regard must be had to the state of knowledge at the material time. 194 Authority also dictates that the question of practicability should be determined in a commonsense manner. All the circumstances of the case, including practice, should be taken into account: Jayne v National Coal Board [1963] 2 All ER 220 at 223 (where Veale J said it was "clearly impracticable to take precautions against a danger which could not be known to be in existence"); Nicholson v Heaven & Earth Gallery Pty Ltd (1994) 126 ALR 233 at 244 per Wilcox J; F D R Pty Ltd v Gilmore; Gilmore v Cecil Bros (1998) 80 IR 411. Something is "practicable" if it is capable of being put into practice or feasible: Uebergang v Australian Wheat Board (1980) 145 CLR 266 at 305 per Stephen and Mason JJ; Pioneer Concrete Services Ltd v Yelnah Pty Ltd (1986) 5 NSWLR 254 at 268 per Young J (as he then was); see also M38/2002 v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCAFC 131; (2003) 199 ALR 290 at [65]. 195 Finally I note that the question whether a step or precaution is "practicable" requires making a value judgment in the light of all the facts: cf Slivak v Lurgi (Australia) Pty Limited, above, at 322 [53] per Gaudron J. 196 Girkaid and Inghams bore the onus of proving that McDonald had breached the statutory duty imposed by regulations 18(e) and 19(e): see Slivak v Lurgi (Australia) Pty Ltd, above, at [92] - [98] per Callinan J with whom judgment on this issue Gleeson CJ, Gummow and Hayne JJ (at [39]) and Gaudron J (at [58]) agreed. 197 This required them to demonstrate both that McDonald knew or ought to have been aware that the course he was going to undertake involved a risk against which practicable steps to ensure adequate instructions were available or precautions to avoid fire should be taken. Thus although, strictly speaking, actions for breach of statutory duty do not invoke the concept of foreseeability, in essence the same notion is imparted by the requirement to determine what McDonald knew or ought to have known. 198 I propose, therefore, to consider this aspect of the case under the general heading of foreseeability. Before I do so, however, I note that Girkaid and Inghams' case on this issue turned on Professor Gray's evidence that the propensity of a mixture of substances containing linoleic acid and other materials to undergo spontaneous combustion was a "well known scientific fact", coupled with the proposition that McDonald ought to have become aware of the risks by making inquiries. That much might be accepted. I reject, however, Mr Faulkner SC's submission that what McDonald would have discovered had he made inquiries was a "red herring". That submission is inconsistent with the proposition that penal consequences will not be imposed upon a person in respect of results which he could not foresee.

Foreseeability 199 In Overseas Tankship (UK) Limited v Miller Steamship Co Pty Limited (The Wagon Mound [No 2]) [1967] 1 AC 617 at 643, Lord Reid, who gave the Advice of the Judicial Committee of the Privy Council, identified a "real risk" as being "… one which would occur to the mind of a reasonable man in the position of the defendant's servant and which he would not brush aside as far-fetched". This approach was adopted in Wyong Shire Council v Shirt (1980) 146 CLR 40 at 47 - 48, where Mason J (with whom Stephen J and Aickin J agreed) said: "In deciding whether there has been a breach of the duty of care the Tribunal of fact must first ask itself whether a reasonable man in the defendant's position would have foreseen that his conduct involved a risk of injury to the plaintiff or to a class of persons including the plaintiff…" (emphasis supplied) 200 In determining whether McDonald should reasonably have foreseen that his conduct might have involved a risk of injury it was not necessary that he should have foreseen the precise risk of injury or how it occurred. It was sufficient if the risk was one of a class of risk which, in a general way, he ought have foreseen: Graham Barclay Oysters Pty Limited v Ryan [2002] HCA 54; (2002) 211 CLR 540 at 578 [87] per McHugh J. 201 It is essential in determining the issue of breach to eschew "the use of increased knowledge or experience embraced in hindsight after the event": Maloney v Commissioner for Railways (NSW) (1978) 52 ALJR 292 at 292, quoted by Gummow J in Rosenberg v Percival [2001] HCA 18; (2001) 205 CLR 434 at 456 [68]. In Rosenberg v Percival at [16], Gleeson CJ warned about the danger of viewing the conduct of parties "through the prism of hindsight", saying: "A foreseeable risk has eventuated, and harm has resulted. The particular risk becomes the focus of attention. But at the time of the allegedly tortious conduct, there may have been no reason to single it out from a number of adverse contingencies, or to attach to it the significance it later assumed. Recent judgments in this Court have drawn attention to the danger of a failure, after the event, to take account of the context, before or at the time of the event, in which a contingency was to be evaluated." 202 Thus, the question of whether McDonald was in breach of any relevant statutory duty or his common law duty of care turned on whether a reasonable holder of a dangerous goods licence in McDonald's position would have foreseen that his conduct in mopping up the tall oil with attapulgite and placing it in containers for several days involved a risk of injury to Girkaid and Inghams. 203 In my view, the risk of injury from the exercise McDonald undertook was not reasonably foreseeable to the holder of a dangerous goods licence in his position. 204 At a very elementary level, this proposition can be tested by reference to the Dangerous Goods Regulation itself. The Regulation did not identify spontaneous combustion as a risk associated with dangerous goods. It might have been thought that if the regulatory authorities were of the view that spontaneous combustion was a risk associated with any class of dangerous goods, that fact would have been stated clearly in the regulations. The risk the regulations required McDonald to guard against in relation to Brytall was the risk that it might ignite if exposed to a naked flame. 205 I have already pointed out that McDonald was unaware that tall oil was classified as dangerous goods. His ignorance would not, of course, be something one would impute to the reasonable holder of a dangerous goods licence. It should be assumed that such a person would have been aware that Brytall was classified as dangerous goods because it had a flash point above 150º C. Accordingly, viewed within the regulatory framework, the question of what McDonald ought to have known would be resolved by saying he ought to have known that if Brytall was exposed to a naked flame it might have burst into flames. 206 It would, however, be an unreasonable interpretation of the regulations to find that they imposed penal consequences on McDonald in respect of a risk which the regulations did not attach to the statutory classification of Brytall. The fact that the Dangerous Goods Regulation 1978 was irrelevant to the risk of spontaneous combustion was made plain by Professor Gray who said that flash point had nothing to do with spontaneous combustion. 207 Accordingly the question was whether McDonald ought to have foreseen the risk of spontaneous combustion of the tall oil/attapulgite mixture, applying common law principles. The primary judge appears to have been influenced by what he said was Professor Gray's evidence that "there were a significant number of references in the literature to the propensity of tall oil to self-combust" (at [37]). However, with respect to the primary judge, Professor Gray's evidence was that there were "voluminous" references to the spontaneous combustion of linseed oil (a substance principally composed of linolenic acid which is a more reactive substance than linoleic acid) and that such references were "indirect references to the proneness of linoleic (and to a lesser extent oleic) acids to spontaneously ignite". Despite his undoubted expertise in the area of combustion issues, the fact was that Professor Gray had never heard of tall oil until this case - let alone of it undergoing spontaneous combustion. The reference in the two annexures to his fourth report to spontaneous combustion related to substances of which tall oil was a component, albeit in unknown quantities. The fact that an internet search using the combination "tall oil spontaneous combustion" turned up only two rather obscure references might be regarded as an indication that, in fact, tall oil was not renowned for its propensity to undergo spontaneous combustion. There was, furthermore, in my view, no evidence that the professed propensity of tall oil to undergo spontaneous combustion was "sufficiently known in industry" let alone in the warehousing industry. The evidence was to the opposite effect. 208 None of the lay witnesses had heard of the risk of an oil undergoing spontaneous combustion. McDonald's evidence was that he had never heard of the propensity of any kind of oil to undergo spontaneous combustion. It was not suggested to him that he ought to have known of that propensity. Similarly, Mr Bashford and Mr Emerson, both of whom had lengthy experience in the area of industrial supplies, had never heard of the phenomenon of oils combusting spontaneously. Mr Underwood accepted that "with hindsight it could be argued that the porosity of the attapulgite was significant in that it may well have increased the physical risk associated with having the oil contact a material with a very high surface area". It certainly does not appear that he was of the view that the propensity of oil to undergo spontaneous combustion was well known to those responsible for dangerous goods. 209 These witnesses all gave evidence with the benefit of hindsight. While not doubting their evidence, which was not challenged, its reliability may be tested by reference to the objective facts at the time of the events under consideration: cf Chappel v Hart [1998] HCA 55; (1998) 195 CLR 232 at 246, footnote 33 per McHugh J. 210 Those objective facts are found in the MSDSs produced in relation to Brytall by Bryce and EKA. Neither of those documents referred to there being any risk associated with mopping up a spill of Brytall with a porous material such as attapulgite and storing it in containers - let alone a risk of spontaneous combustion. 211 The Bryce MSDS recommended that after the spill was absorbed it be shovelled and collected into containers for recycling of salvage. It was implicit in that statement that there was no risk in keeping such a mixture in containers for a period of time. The six or so days the mixture had rested in the containers before the fire started was clearly not unreasonable in that light. 212 The Bryce MSDS also said there were no unusual fire or explosion hazards associated with the product. It might have been thought, therefore, that if one pursued the logic of the reasonable course of inquiry McDonald could be expected to have undertaken, it would have been concluded at that stage. 213 The primary judge found, however, that McDonald should have rung EKA's emergency number. It should be recalled, however, that McDonald said and the primary judge accepted, that the situation with which he was dealing did not amount to an emergency. In such circumstances, it might be thought that a reasonable holder of a dangerous goods licence was not under an obligation to contact an emergency number. 214 Nevertheless, pursuing this course, had McDonald gone so far as to track the chain of information down to the EKA MSDS again he would not, on the primary judge's finding, have been alerted to any risk in the course he was pursuing. Clause 6 of the EKA MSDS would have advised him in the event of a spillage to "collect as much as possible in a clean container". It is true that it warned to "remove all sources of ignition", but it did not identify mixing a spillage with absorbent material in that context as constituting a risk. Clause 7 would have advised him that no special measures were necessary in terms of fire and explosion protection while paragraph 10 would have advised him as to "hazardous reactions", the "product is stable and no hazardous polymerisation will occur". 215 Even if McDonald had obtained that MSDS and read the opaque entry "conditions to avoid - avoid prolonged contact with porous materials", that entry would not, in my view, in light of the assurances in the remainder of the document, have prompted him to make further inquiry. 216 The primary judge's conclusion that if he had telephoned EKA he would have been informed of the risk of spontaneous combustion associated with the course he was about to pursue has an air of unreality. Such advice would, at least, have contradicted the express passages in the MSDS to the effect that there were no fire hazards associated with the product. There was no evidence which justified the primary judge's conclusion. It was, as Mr Finch SC submitted, an impermissible inference.

217 Finally, the fact that on all the evidence McDonald's method of mopping up the Brytall spill was consistent with the practice at the time is powerful, although not decisive, evidence that he did not act negligently: Dovuro Pty Limited v Wilkins [2003] HCA 51; (2003) 77 ALJR 1706 at 1712 [34] per McHugh J; see also Marshall v Lindsey County Council [1935] 1 KB 516 at 540 per Maugham LJ (affirmed [1937] AC 94 cited by Finkelstein J in Dovuro Pty Limited v Wilkins [2000] FCA 1902; (2000) 105 FCR 476 at 507 [112]). 218 I accept that the evidence of practice does not dictate the outcome of the inquiry as to whether or not McDonald complied with the appropriate standard of care. There are some cases, of which Rogers v Whitaker was one, where the question of the standard of care involves no special skill unique to the occupation under consideration, where the Court will be prepared to reach a conclusion that a duty of care has been breached even though it may not accord with evidence given by specialists in the relevant field. However, in my view, that approach is not warranted in this case. The risk which the mixture of tall oil and attapulgite posed was one which, at best, might be known to Professors of Chemistry, but was not, apparently, known to the manufacturers of either tall oil or attapulgite. In such circumstances compliance with both the Bryce and EKA MSDSs was, in my view, evidence that McDonald did not breach his duty of care. 219 The primary judge's conclusion, with respect, appears to have been reached more with the benefit of hindsight than by an assessment of McDonald's position in January 1998 or the objective facts demonstrating that there was no perceived risk of spontaneous combustion in the exercise McDonald undertook. 220 The lack of any evidence that at the time of the events there was perceived to be a risk of spontaneous combustion of Brytall if a spill was cleaned up with a conventional absorbent material such as attapulgite and kept in containers prior to disposal indicates, to my mind, that that risk was not perceived as a real one in the Wyong Shire Council v Shirt sense by the regulatory authorities, the manufacturers or distributor of the product, or by those involved in handling such products and supplying materials for their absorption in the event of spills. 221 The case bears some similarity to Overseas Tankship (UK) Ltd v Morts Dock & Engineering Co Ltd [No 1] [1961] AC 388 ("The Wagon Mound") where the plaintiff's case failed because there was no evidence that the defendant knew or could reasonably have been expected to know that furnace oil was capable of being set alight when spread on water. 222 In my view, Girkaid and Inghams' case on breach failed because there was no evidence that McDonald knew or ought to have known, or ought to have foreseen, the risk of the tall oil/attapulgite mixture undergoing spontaneous combustion. 223 In my view, the primary judge erred in finding that McDonald breached his common law duty of care. 224 Accordingly, I would uphold grounds 7, 8, 9, 10 and 11 of the appeal.

Damages (grounds 12 - 13) 225 McDonald's challenge to the primary judge's conclusion on damages was made in the following grounds of appeal: "12. His Honour erred in finding on the basis of the valuer's evidence that the value of the First Respondent's land after the fire was $130,000.00 when the evidence was that it sold seven months after the fire, in a stable market, for $270,000.00. 13. Alternatively, his Honour erred in not discounting the First Respondent's damages by $140,000.00 being the amount by which it mitigated its loss by reason of the sale in August 1998." 226 Most of the damages were agreed. The only matter of controversy was Girkaid's claim for the value of the warehouse burnt down in the fire. 227 Girkaid called Mr Irving, a registered valuer. The gist of his evidence was that the unimproved value of the land on which the warehouse was situated at January 1998 was $130,000. The freehold market value of the land with tenancies before the fire was, in his opinion, $460,000. Accordingly, he opined that the loss suffered by Girkaid from the loss of the building (leaving aside loss of rent) was $330,000. 228 The defendants challenged Mr Irving's opinion by pointing to the fact that Girkaid had sold the land in August 1998 for $270,000. Mr Irving rejected the proposition that that sale affected his valuation on the basis that he did not know whether there were any special circumstances surrounding it. He had made no enquiries to determine the circumstances of the sale. Girkaid led no evidence as to the circumstances of that sale, nor had it, apparently, briefed Mr Irving with the detail of those circumstances to permit him to assess its significance to the value of the land as at January 1998. 229 The primary judge accepted Mr Irving's valuation (at [60] - [63]) and held that the loss arising from the destruction of the building, exclusive of the loss of rent, was $330,000. 230 McDonald submitted that it was not reasonably open to the primary judge to find that the value of the land after the fire was $130,000. 231 McDonald also submitted that even if Mr Irving's valuation was accepted, to the extent that Girkaid recovered $270,000 by the August 1998 sale, Girkaid had mitigated its loss so that the amount of damages had been overstated by an amount of $140,000. 232 The evidence of valuation was unsatisfactory. Although the valuer was called to establish the value of the loss of the buildings erected on the land, he did not give direct evidence of their value. Rather, he took the view that the appropriate market value of the property was to be assessed by reference to the capitalisation of net rents and, therefore, had regard to what, in his opinion, were comparable sales of properties with such rentals. On this basis, he assessed the market value of the property at $460,000. 233 Mr Irving did not support the unimproved land value as at January 1998 by reference to the sale of identified properties at identified dates. There was no way of assessing, therefore, whether they were truly comparable or, at least, more "comparable" than the August sale of the subject land. In my view it was incumbent upon Girkaid, if it sought to deny the forensic value of the August 1998 sale, to call evidence that there were special circumstances which made it inapplicable. Absent such evidence, the primary judge was entitled to draw an inference adverse to it that such evidence would not have assisted: Commercial Union Assurance Co of Aust Ltd v Ferrcom Pty Ltd (1991) 22 NSWLR 389 at 418 per Handley JA. 234 In such circumstances, I would accept McDonald's submission that it was not reasonably open to the primary judge to accept Mr Irving's evidence that the value of the land post fire was $130,000. Rather, on any view, its actual sale price within some five months of the fire was the best evidence of value. Accordingly, even if I was not of the view that the appeal should succeed on other grounds I would be of the view that Girkaid's damages were overstated by an amount of $140,000. 235 As Bryce and Hudson's liability turned on the primary judge's finding adverse to McDonald, these conclusions dictate that the remaining appeals and cross appeals should be resolved, subject to one matter, in Bryce, Hudson and SLE's favour respectively.

Breach of the bailment contract 236 The exception related to Bryce's claim in bailment against McDonald. Ground 17 in McDonald's Notice of Appeal challenged the primary judge's finding that McDonald was liable to Bryce for its loss in the fire. 237 As I have earlier noted, Bryce sought to retain its judgment against McDonald for breach of its bailment contract on the basis that McDonald would remain liable for Bryce's loss unless he could prove that his negligence was not a contributing factor. 238 As I have found that McDonald was not negligent it follows that he is entitled to succeed on this ground also. 239 I would make the following orders:

(1) Appeal allowed. (2) Judgment entered by Adams J on 8 March 2002 on Girkaid and Inghams' claim set aside. (3) Girkaid and Inghams' claim dismissed. (4) Girkaid and Inghams to pay McDonald's costs of the appeal and of the proceedings below including such costs as McDonald is ordered to pay Bryce, Hudson and SLE.

Hudson's appeal (1) Appeal allowed. (2) Judgments against Hudson set aside. (3) Cross-claims against Hudson dismissed. (4) McDonald to pay Hudson's costs of the appeal and of the proceedings below.

Bryce's appeal (1) Appeal allowed. (2) Orders 4 to 7 inclusive, 12 and 15 of Adams J's orders of 8 March 2002 set aside. (3) McDonald's cross-claim against Bryce dismissed. (4) McDonald to pay Bryce's costs of the appeal and of the proceedings below.

SLE's cross-appeals (1) Cross-appeals dismissed. (2) McDonald to pay SLE's costs of the cross-appeals. 240 YOUNG CJ in EQ: I have read in draft the reasons for judgment of McColl JA and basically agree with them. 241 As to the question of causation I agree with her Honour's closely reasoned conclusion that the primary Judge's finding that Girkaid and Inghams had discharged their burden of proof should not be disturbed. 242 As to breach of duty, I would also agree with her Honour's conclusion. 243 It seems to me that a good check on the result is to ask oneself as a matter of common sense whether a carrier/warehouseman who spills pine oil and takes advice as to how to mop it up and follows that advice and who puts the waste product in two barrels could reasonably foresee, in the absence of any warning, that the mix would spontaneously combust and burn down the factory? 244 The answer is clearly "No". 245 In my view any construction of the regulations with respect to the storage of inflammatory goods which might produce a different result is, with respect, to those who hold that view, too strict and artificial an interpretation. 246 As to damages, I am surprised that anyone considers that they can value a building by taking away the value of the land content from the total value of land and building. The very fact that that particular type of building is on the land may well cause the land value to be greater or less. 247 However, the Judge having been asked by the parties to approach the valuation question in that way, I agree with McColl JA that he fell into error by deducting merely $130,000 from the total package. I agree with the orders proposed by McColl JA.

Last Modified: 09/02/2004 DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.